Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

Lara López-González; Margaret Martínez-de-la-Torre; Luis Puelles

doi:10.1007/s00429-022-02601-y

Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

López-González, Lara; Martínez-de-la-Torre, Margaret; Puelles, Luis 2023-03-01 00:00:00 The ventromedial hypothalamic nucleus (VMH) is one of the most distinctive hypothalamic tuberal structures, subject of numerous classic and modern functional studies. Commonly, the adult VMH has been divided in several portions, attending to differences in cell aggregation, cell type, connectivity, and function. Consensus VMH partitions in the literature com- prise the dorsomedial (VMHdm), and ventrolateral (VMHvl) subnuclei, which are separated by an intermediate or central (VMHc) population (topographic names based on the columnar axis). However, some recent transcriptome analyses have identified a higher number of different cell types in the VMH, suggesting additional subdivisions, as well as the possibility of separate origins. We offer a topologic and genoarchitectonic developmental study of the mouse VMH complex using the prosomeric axis as a reference. We analyzed genes labeling specific VMH subpopulations, with particular focus upon the Nkx2.2 transcription factor, a marker of the alar-basal boundary territory of the prosencephalon, from where some cells seem to migrate dorsoventrally into VMH. We also identified separate neuroepithelial origins of a Nr2f1-positive subpopulation, and a new Six3-positive component, as well as subtle differences in origin of Nr5a1 positive versus Nkx2.2-positive cell populations entering dorsoventrally the VMH. Several of these migrating cell types are born in the dorsal tuberal domain and translocate ventralwards to reach the intermediate tuberal domain, where the adult VMH mass is located in the adult. This work provides a more detailed area map on the intrinsic organization of the postmigratory VMH complex, helpful for deeper functional studies of this basal hypothalamic entity. Keywords Ventromedial hypothalamic nucleus · Tuberal hypothalamus · Acroterminal · Tangential migration · Prosomeric model · Nkx2.2 Abbreviations DMH-T Terminal part of the dorsomedial hypotha- ABasM Medial (acroterminal) part of the anterobasal lamic nucleus nucleusDV Dorsoventral dimension ABasW Wing (terminal) part of the anterobasal FP Floor plate nucleus hp1 Hypothalamo-telencephalic prosomere 1 A/B Alar/basal limit hp2 Hypothalamo-telencephalic prosomere 2 AP Anteroposterior dimensionHyA Hypothalamo-amygdalar corridor Arc Arcuate nucleusIF Immunofluorescence AT Acroterminal regionIHC Immunohistochemistry Di Diencephalon ISH In situ hybridization DMH-P Peduncular part of the dorsomedial hypotha-M Mamillary area lamic nucleusME Medial eminence NHy Neurohypophysis OCh Optic chiasma * Luis Puelles p2 Prosomere 2 puelles@um.es p3 Prosomere 3 1PHy Peduncular hypothalamus University of Murcia, IMIB-Arrixaca Institute of Biomedical PM Perimamillary area Research, El Palmar, 30120 Murcia, Spain Vol.:(0123456789) 1 3 Brain Structure and Function PoA Preoptic area variety of much studied physiologic functions, which include PPa Peduncular paraventricular area metabolic regulation (Hetherington and Ranson 1942; PPaV Peduncular ventral paraventricular nucleus Frohman et al. 1974; Elmquist et al. 1999; Dhillon et al. PRM Periretromamillary area 2006; Kim et al. 2011; Meek et al. 2016), and reproductive PSPa Peduncular subparaventricular area (Pfaff and Sakuma 1979a, 1979b; Correa et al. 2015; Hashi- PTh Prethalamus kawa et al. 2017; Lewis et al. 2022), or aggressive behaviors RM Retromamillary area (Yang et al. 2013; Wang et al. 2015; Kennedy et al. 2020; RTuD Dorsal retrotuberal area Hashikawa et al. 2017; Lewis et al. 2022). There is a recent RTuI Intermediate retrotuberal area review by Khodai and Luckman (2021). Conventionally, this RTuV Ventral retrotuberal area structure is subdivided in 2–3 parts regarding its aggregation SCh Suprachiasmatic nucleus patterns, neuronal morphology, molecular phenotype, and Subl Subliminal band birth dating, normally described topographically as dorso- Tel Telencephalon medial, central, and ventrolateral VMH formations as seen Th Thalamus in coronal sections interpreted as cross-sections within the THy Terminal hypothalamus columnar model (Gurdjian 1927; Krieg 1932; Shimada and TPa Terminal paraventricular area Nakamura 1973; Altman and Bayer 1978, 1986; McClel- TSPa Terminal subparaventricular area lan 2006; Kim et al. 2019; van Veen et al. 2020). Introduc- TuD Dorsal tuberal area tion of the updated prosomeric model of the hypothalamus TuI Intermediate tuberal area by Puelles et al. (2012) provisionally did not change these TuV Ventral tuberal area terms, to avoid confusion, but they are clearly incongru- TSbO Tuberal suboptic nucleus ent with the differently oriented prosomeric forebrain axis VMH Ventromedial hypothalamic nucleus (e.g., the ‘dorsomedial’ part would be described rather as VMHdc Ventromedial hypothalamic nucleus, dorso- ‘caudomedial’, and the ‘ventrolateral’ part as ‘rostrolateral’). caudal part Though this conventional schema is habitually visualized in VMHdrl Ventromedial hypothalamic nucleus rostrolat- a single coronal section plane (i.e., is bidimensional), Van eral dorsal part Houten and Brawer (1978), also thinking in columnar terms, VMHdrm Ventromedial hypothalamic nucleus rostrome- contemplated in addition the ‘anteroposterior’ dimension dial dorsal part (equivalent to the dorsoventral axis in prosomeric terms). VMHil Ventromedial hypothalamic nucleus, interme- They identified “anterior”, “middle”, and “posterior” differ - dio-lateral part ences within the dorsomedial and ventrolateral VMH parti- VMHim Ventromedial hypothalamic nucleus, interme- tions studied in coronal sections which correspond to the dio-medial part prosomeric dorsoventral topologic differences presented in VMHvi Ventromedial hypothalamic nucleus, ventro- the present report (see also Table 3). intermediate part As regards connections, the VMH nucleus sensu lato VMHvl Ventromedial hypothalamic nucleus, ventro- is connected reciprocally to the amygdala, septum, preop- lateral part tic area, paraventricular and anterior (subparaventricular) VMHvm Ventromedial hypothalamic nucleus, ventro- hypothalamus, contralateral VMH, tuberal, dorsomedial, medial part ventral and dorsal premamillary, medial mamillary and ret- romamillary nuclei, prethalamic zona incerta, paraventricu- lar and parataenial thalamic nuclei, tegmental ventral area, Introduction periaqueductal gray and raphe nuclei (Saper et al. 1976; Canteras et al. 1994; Saper and Lowell 2014; Shimogawa The classic hypothalamic ventromedial nucleus (VMH) is et al. 2015). Many of these areas are differentially inner - one of the largest structures in the entire hypothalamus. vated by the diverse VMH subdivisions. For instance, fore- It was identified within the tuberal area as the “principal brain structures regulating the steroid hormonal signaling nucleus” by Ramón y Cajal (1911). This nucleus contains system, including the medial preoptic, tuberal and ventral largely glutamatergic neurons (Ziegler et al. 2002; Puelles premamillary nuclei, receive inputs mainly from the classic et al. 2012) that form several cell masses aggregated in an ventrolateral VMH subdivision (Canteras et al. 1994). More- ovoid block of the tuberal medial hypothalamic stratum sur- over, the different VMH parts seem associated to different rounded and delimited by a shell of afferent amygdalar inputs functions. Dorsomedial VMH and central/core VMH were (Krieg 1932; Heimer and Nauta 1969). The shell shows non- linked to metabolic circuitry (Kim et al. 2011; Meek et al. glutamatergic cell types, in part migrated from the overly- 2016), whereas ventrolateral VMH controls reproductive ing alar plate (Diaz et al. 2015). The VMH is involved in a 1 3 Brain Structure and Function and aggressive behavior (Lee et al. 2014; Lin et al. 2011; the acroterminal region of the hypothalamus (Puelles et al. Hashikawa et al. 2017; Lewis et al. 2022). 2012). These antecedents, taken jointly with some molecular In this work we have examined the origin of molecu- developmental aspects reviewed in Puelles et al. (2012), sug- larly defined cells populating different VMH subdivisions gest that the VMH subdivisions may relate to subtly different based on the detailed topologic developmental map of dif- origins of these neuronal subpopulations, with some relevant ferentially specified progenitor domains available within the progenitor domains possibly lying dorsally or rostrally to updated prosomeric model (Puelles et al. 2012; Puelles and the place where the VMH nucleus develops (descriptors Rubenstein 2015; Puelles 2018, 2019). Our material includes according to the columnar paradigm; see Fig. 1). The global sagittal, horizontal, and transversal sections oriented accord- cytoarchitectonic boundary that delimits the VMH complex ing to the ‘natural’ prosomeric axis (defined as parallel to may be due largely to the VMH shell plexus formed around the alar-basal boundary, the floorplate, and the precociously it. Most studies addressing the development of the VMH underlying notochord; Fig. 1). We complemented these data have used the columnar model of the brain as a morphologic with some tracing experiments in organotypic cultures of reference (Herrick 1910, 1933; Kuhlenbeck 1973; Alvarez- embryonic hypothalamus to demonstrate the reality of dors- Bolado et al. 1995), and essentially employed coronal sec- oventral migratory displacements predicted by Puelles et al. tions through the nucleus, which was assumed to develop in (2012). We indeed found that different VMH subpopula- its adult position. tions are born in different basal tuberal histogenetic pro- Altman and Bayer (1986), in their autoradiographic neu- genitor areas, either coinciding with the final VMH locus rogenetic study of the rat hypothalamus, identified the vent- (only radial migration involved) or placed dorsal and/or rolateral VMH as the earliest generated region, followed by rostral to the VMH ventricular zone, thus implying signifi- the dorsomedial VMH (called ‘dorsalis’ by these authors), cant short-range tangential migrations. Some components whereas their ‘VMH pars basalis’ (possibly referring to the seem to originate in the overlying alar plate. In particular, local periventricular stratum) is the last to become postmi- we provide additional evidence for the tangential dorsoven- totic, in a sequence ranging between E13 and E17. However, tral migration of the Nkx2.2-expressing VMH cell popula- these authors did not identify precisely the neuroepithelial tions, complementing the material previously commented origin of the VMH subdivisions and apparently assumed by Puelles et al. (2012). a local radial origin for all of them. A previous autoradio- graphic study of Shimada and Nakamura (1973) reported the birthdate interval for VMH neurons in the mouse Material and methods between E10-E14, but only vaguely ascribed their origin to the underlying neuroepithelium. Ulterior studies described Allen atlas brain database some radial migratory cell movements in cultured coronal VMH slices (Dellovade et al. 2000, 2001; McClellan et al. We selected several significative gene expression images 2006, 2008). Only Puelles et al. (2012; pp 285–287) seem from the Allen Developing Mouse Brain Atlas (https:// to have considered the possibility of tangential migrations devel oping mouse. br ain- map. org/). Whereas we analyzed being involved in the development of this nuclear complex. all mouse material available in this database for each gene Some molecular markers specific of different VMH selected, we chose images from E13.5, E15.5, E18.5, P1 regions have been reported (Kurrasch et al. 2007). Recently and P4 to build the figures. Some of then combine Allen cell line reporter studies (for Shh, Gli, Neurog2, Ascl1) fol- Atlas material join with our lab material. We interpreted lowed partially some VMH subpopulations and identified the Atlas coronal section planes as horizontal (i.e., parallel the positions they occupy in the adult VMH (Corman et al. to the hypothalamic alar-basal boundary shown in Fig. 1b 2018; Aslanpour et al. 2020a, b). Nevertheless, confusion (A/B limit). persists, unfortunately, since the widely used coronal section plane is usually understood within the columnar model as Animals demonstrating transversal relationships. In contrast, conven- tional coronal sections are roughly horizontal in the proso- We studied mouse specimens from several stages of devel- meric model due to the latter’s different axial references opment processed for in situ hybridization or immunohis- (e.g., the alar-basal boundary and the floor plate) ending in tochemistry techniques: E12.5 (n = 2), E13.5 (n = 1), E14.5 (n = 8), E16.5 (n = 5), E18.5 (n = 5), adults (n = 2). We used 1 3 Brain Structure and Function E12.5 mice for migration assays (n = 6, see section 2.5). The (PBS) at 4 °C. After washing, they were embedded in 4% morning in which a vaginal plug was detected was consid- agarose in PBS for sectioning. Vibratome sections were ered as E0.5 in all embryos. Pregnant females were sacri- obtained 100 μm-thick for ISH or ISH followed by DAB- ficed by cervical dislocation after inhalation of isofluorane, immunohistochemistry, or 50 mμ-thick for immunoreactions. and then the embryos were extracted. Embryonic brains were dissected out after anesthesia on ice followed by decapita- Immunohistochemistry tion. For adult animals, after standard sodium pentobarbital anesthesia, the mice were perfused with 4% paraformalde- We performed free floating immunostaining of vibratome hyde. The brains were dissected out and fixed overnight in sections. For immunofluorescence reaction, sections were 4% paraformaldehyde in pH 7.4 phosphate-buffered saline washed in PBS-T (PBS-0.3% Triton X-100), blocked (3% 1 3 Brain Structure and Function ◂Fig. 1 Comparative schemata of the hypothalamus representing: a, b intermediate area, TuI tuberal intermediate area, NHy neurohypophy- the prosomeric model and c the columnar model. A panoramic view of sis, ME medial eminence, DMH-P peduncular part of the dorsomedial the position of the hypothalamus in the forebrain is shown in (a). a, b hypothalamic nucleus, DMH-T terminal part of the dorsomedial hypotha- The prosomeric hypothalamus is divided into peduncular and termi- lamic nucleus, VMH ventromedial hypothalamic nucleus, RTuD (main or nal portions (PHy; THy) corresponding to hypothalamo-telencephalic subl.), retrotuberal dorsal area (main or subliminal); TuD (main or subl.), prosomeres hp1 and hp2. In these schemata the yellow line indicates tuberal dorsal area (main or subliminal); AT acroterminal area, A/B limit, the dorsal longitudinal limit between hypothalamus and telencepha- alar/basal limit; PSPa peduncular subparaventricular area, TSPa terminal lon, whereas the retromamillary (RM) and mamillary (M) areas contact subparaventricular area, OCh optic chiasma, PPa peduncular paraven- ventrally the hypothalamic floorplate (FP), a primary axial longitudinal tricular area, TPa terminal paraventricular area landmark induced by the notochord (not shown). The diencephalic and telencephalic roofplate (RP), another primary longitudinal landmark, ends rostrally over the preoptic region (PoA). An intermediate light blue longitudinal line identified as A/B in a , b represents the postulated alar/ BSA in PBS-T, 1–3 h), and incubated in the primary anti- basal boundary, parallel to both FP and RP; it is held to result from dor- body solution (diluted in 3% BSA in PBS-T, 48 h, 4 ℃). soventral patterning antagonism between ventralizing floorplate signals Following incubation and several PBT washes, the sections and dorsalizing roofplate signals (it is thus a secondary axial landmark shared by all brain parts and supported by many gene expression pat- were incubated 2 h with the respective u fl orochrome-labeled terns). The rostromedian aspect of the entire hypothalamo-preoptic secondary antibodies, either Alexa 488 donkey anti-rabbit region represents the singular acroterminal domain (AT in b), which or donkey Alexa 594 anti-mouse (ThermoFisher; 1:200, marks from floor to roof the topologic rostral end of the neural tube. 2 h). For DAB-immunohistochemistry, vibratome sections These consistent and causally fundamented axial landmarks justify the spatial orientations provided [R(ostral), C(audal), D(orsal), V(entral)], were washed in PBS, and then treated with 0.1% hydro- such that the diencephalon proper lies caudal to the hypothalamus and gen peroxide in PBS for 30 min, in the dark and at room the telencephalic vesicle is a dorsal outgrowth of the hypothalamus. The temperature, to inactivate endogenous peroxidase activity. eye evaginates out of the alar AT domain (see optic chiasma –OCh– in After standard PBS-T washes, and the blocking step (3% b). b This more detailed schema displays hypothalamic structure divided in two transverse neuromeres (hp1/PHy; hp2/THy) extending dorsal- BSA in PBS-T, 1–3 h), the floating sections were incubated ward from the floorplate (FP) up to the hypothalamo-telencephalic limit with the primary antibody for 48 h at 4 ℃. After PBS-T (HTL, marked by a thick yellow line). A light blue line tagged ‘A/B washes we applied a biotinylated goat anti-rabbit or anti- limit’ marks the longitudinal alar-basal limit which separates the alar mouse secondary antibody (1:200, 2 h at room temperature; and basal hypothalamic progenitor areas. There is a definite AP and DV pattern held to be causally significant. The progenitor domains identi- Vector Laboratories, Burlingame, CA, United States), fol- fied on the basis of differential expression of transcription factors and lowed by a streptavidin/horseradish peroxidase (HRP) com- other molecular markers (Puelles et al. 2012) are thus organized rostro- plex (1:200, 2 h; Vectastain-ABC kit; Vector Laboratories, caudally (relative to the diencephalon, hp1/PHy and hp2/THy) and dor- Burlingame, CA, United States). Histochemical detection of soventrally (relative to FP, A/B, HTL, and the telencephalic RP). The rostromedian section of the neural tube where right and left brain halves the peroxidase activity was carried out using 0.03% diam- meet is identified as the acroterminal sector (AT). There is a dorsoventral inobenzidine (DAB) and 0.005% H O Primary antibod- 2 2. pattern of seven longitudinal progenitor areas (2 alar and 5 basal), each ies were used as follows: mouse anti-Nkx2.2 (1.50; DSHB, one with its peduncular and terminal part (see list of Abbreviations for Ref. 745A5-s), rabbit anti-Couptf1/Nr2f1 (1.200; Abcam, the respective areal names). We add some secondary subdivisions such as the liminal PSPa/TSPa alar band and the subliminal RTu/Tu basal Ref. ab96846), rabbit anti-Nkx2.1 (1:200; Sigma Aldrich, band, these being concepts used in the text. The ventromedial nucleus Ref. SAB3500757), rabbit anti-Isl1 (1:200; Abcam, Ref. (VMH) is placed in its adult position with its neighbor, the dorsomedial ab20670), rabbit anti- Otp (1: 200; F. Vaccarino), rabbit anti nucleus complex (VMH; DMH-P; DMH-T; within the green RTuI/TuI TH (1:200; Bio-Techne R&D Systems, Ref. NB300-109). domain). c According to both classical and recent studies the columnar model does not separate PHy from THy and defines its axial or longitu- dinal dimension as roughly parallel to the prosomeric AT domain, some- In situ hybridization how implying an extension of the brainstem and midbrain axis across the diencephalon and the hypothalamus into a telencephalic end (not We used the restriction enzymes and polymerases suitable stipulated precisely). In general, the entire columnar hypothalamus is conceived recently as a floor and basal region of the diencephalon, sub- for specific riboprobe synthesis in the presence of digoxi- divided by transverse planes into preoptic, anterior, intermediate (tuberal) genin- 11-UTP. The hybridization protocol used was accord- and posterior (mamillary) regions (Swanson 2012); note the prosomeric ing to Shimamura et al. (1994). Mouse cDNA probes used AT domain is conceived rather as prechordal floor. This model classi- for in situ hybridization were Nxk2.2 and Nr5a1 (J.R. Ruben- fies the whole prosomeric alar hypothalamus as an Anterior hypotha- lamic region (containing both paraventricular and anterior hypothalamic stein), Otp (A. Simeone), Six3 (P. Bovolenta), and Satb2 (our nuclei). The prosomeric basal hypothalamus corresponds to the tuberal own lab, NCBI accession number NM_001358580). and mamillary parts of the columnar hypothalamus. The spatial orienta- tions in this model are rotated 90 degrees relative to the prosomeric ones Organotypic cultures (compare b with c) R rostral, C caudal, D dorsal, V ventral, Di dienceph- alon, Tel telencephalon, hp1 hypothalamic prosomere 1, hp2 hypotha- lamic prosomere 2, FP floor plate, PoA preoptic area, PHy peduncular Brains of embryos dissected from skin and other append- hypothalamus, THy terminal hypothalamus, RM retromamillary area, M ages at E12.5 were collected in artificial cerebrospinal mamillary area, PRM periretromamillary area, PM perimamillary area, fluid (ACSF) solution at pH 7.4 containing: 4 mM KCl, RTuV retrotuberal ventral area, TuV tuberal ventral area, RTuI retrotuberal 1 3 Brain Structure and Function 1.5 CaCl2, 0.75 mM MgCl2, 129 mM NaCl, and 10 mM (Puelles et al. 2012; Puelles 2019), which momentarily D-glucose. We dissected the tissue with dissection tweezers forces translation of both unconciliable terminologies (see discarding meninges and telencephalic vesicles, and opened our Table 3) under the assumption that causal explana- the neural tube along the midline. We placed separately the tions will only emerge from the prosomeric notions. The two brain halves upon membrane culture inserts (Millicell prosomeric hypothalamus consists of two prosomeres, Millipore, 0.4 mm, PICM0RG50) within small Petri dishes, hp2 and hp1, which represent its structure transversal to with the ventricular surface up, contacting the air, and the the axis of the forebrain (Fig. 1; the prosomeric axis is pial surface touching across the membrane a substrate of co-defined by the notochord, the floorplate, the alar-basal MEM-supplemented medium (1% Penicilin/Streptomycine, boundary and the roofplate as mutually parallel longitudi- 0.065% glucose, 0.5% glutamine, and 1% inactivated fetal nal reference landmarks lying at different dorsoventral lev - bovine serum). Explants were acclimatized for 1 h (37 ℃, els; Puelles 2018; Amat et al. 2022; the optic tract may be 5% CO2), and subsequently marked through the ventricular taken as another such reference; Puelles 2022). These two surface with a CMFDA-coated tungsten particle borne on units are both hypothalamo-telencephalic in spatial range a sharpened tungsten needle (Alifragis et al. 2002; López and contain respectively the terminal hypothalamus which González et al. 2021), testing diverse labeling loci along extends into the non-evaginated preoptic telencephalic the estimated alar-basal boundary of the hypothalamus (i.e., subpallium (THy in hp2; Fig. 1a, b) and the peduncular varying the dorsoventral position relative to this limit, and hypothalamus that expands into the rest of the telencepha- changing also the anteroposterior position along the THy, lon (PHy in hp1; Fig. 1a, b; the telencephalon is strictly including its rostromedian acroterminal domain). After two a hypothalamic bilateral dorsal evagination, being thus days in culture conditions (37 ℃, 5% CO2), the explants epihypothalamic and the Hypo-Thalamus is a wrong term, were fixed with cold paraformaldehyde 4% in PBS for because this brain region lies rostral rather than ventral to 10 min. To check the position of the CMFDA particle, as the thalamus and diencephalon proper). THy also includes well as the labeled cells in the mantle layer, all explants in its acroterminal alar part the eye vesicle and eye stalk were processed for immunofluorescence with the mouse region. PHy is continuous dorsally through the interven- anti-Nkx2.2 antibody (1.50; DSHB, Ref. 745A5-s). tricular foramen with the whole evaginated telencephalic vesicle (Puelles et al. 2012; Puelles and Rubenstein 2015). Image analysis The right and left hypothalamic sides meet rostrally along the acroterminal area (AT) of THy, which is a singularly We scanned the ISH and ISH/IHC images at high resolu- differentiated median transverse territory displaying dors- tion with the Aperio ImageScope software (Leica Biosys- oventral alar and basal structural specializations (terminal tems). Immunofluorescent (IF) images were obtained from lamina, optic chiasma, anterobasal area, arcuate/median the sectioned brains and the whole-mounted explants using eminence area, infundibulum and neurohypophysis, tub- a confocal SP8 Leica microscope. Individual optic sections eromamillary, and mamillary subregions; Puelles et al. were 3 µm apart, and image stacks of various Z sizes were 2012; Puelles and Rubenstein 2015). The hypothalamic generated according to the structures of interest. All figures dorsoventral zonal pattern (alar and basal subdomains) is from the Allen Developing Mouse Brain Atlas (https://de vel continuous caudalwards with the equivalent diencephalic oping mouse. brain- map. org/) were 180º rotated (nose at the pattern (Puelles et al. 2012; Puelles and Rubenstein 2015; right side) to have the same orientation than in our own Díaz et al. 2015; Ferran et al. 2015; López-González et al. lab images. Figures were constructed using ImageJ, Adobe 2021), with shared tagmatic properties down to the isthmo- Photoshop and Adobe Illustrator software. mesencephalic boundary (Puelles 2018). The evaginated telencephalon as well as the essentially acroterminal eye are thus held to be alar derivatives of the hypothalamus, Results evaginated within hp1 and hp2, respectively. Hypothalamic Shh expression induced by the underlying notochord ini- Background introduction to the prosomeric tially extends to its floor and basal plates (limited by the hypothalamus alar-basal boundary) but is later lost at the acroterminal infundibular part of the intermediate tuberal basal plate The neurodevelopmental field is performing gradually a area (TuI), a secondary effect due to adenohypophysial paradigm change from the old columnar model (which has Tbx3 signals (Trowe et al. 2013). Regionalization is dem- proven to have difficulties assimilating and explaining gene onstrated further through gene expression patterns charac- expression patterns) and the modern prosomeric model. teristic of the mentioned structural divisions (e.g., Ntn1, Given their fundamental discrepancy on the length axis of Lmx1b, Lmx1a, Foxa1 are markers present in the hypo- the forebrain, there is a terminological problem associated thalamic floor plate; Six3, Fgf10, Fgf8, Dlk1 appear in the 1 3 Brain Structure and Function acroterminal area; Nkx2.2 is a linear longitudinal marker, subdomains, called dorsal, intermediate, and ventral (e.g., expressed in a band along the forebrain alar-basal limit and TuD, TuI, TuV within hp2; similar for the RTu subdivi- overlapping slightly both alar and basal domains; the alar sions within hp1; Puelles et al. 2012). In the present report overlap domain is known as the ‘liminal band’, meaning we are concerned essentially with the TuD and TuI hp2 the rim of the alar plate, whereas the basal overlap domain basal longitudinal zones, which relate to the VMH nucleus. is known as the ‘subliminal band’; Puelles et al. 2012; It should be noted that TuD is itself divided into the dorsal Fig. 1a, b). Detailed genoarchitectonic dorsoventral pattern Nkx2.2-positive subliminal band (the area of basal overlap in the hypothalamus allows to distinguish four alar and five of the Nkx2.2 band) and the underlying Nkx2.2 negative basal domains that extend longitudinally across both THy ‘main part’ of TuD (Puelles et al. 2012; their Fig. 8.14B). (hp2) and PHy (hp1) (see Fig. 1a, b, its legend, and the abbreviations list). The paraventricular (Pa) and subpara- Borders of the VMH ventricular (SPa) areas are distinct superposed alar plate domains, where Pa has three dorsoventral subdivisions (not The VMH nucleus occupies a large dorsal area within the relevant here). There is a major tuberal/retrotuberal basal TuI region of the terminal hypothalamus, caudal to the longitudinal complex (Tu/RTu) across hp2 and hp1 and acroterminal arcuate nucleus domain (Arc) and dorsal to under the alar-basal boundary; it lies dorsal to underlying the terminal part of the dorsomedial nucleus (DM-T); the parallel perimamillary/periretromamillary (PM/PRM) and overlying TuD region contains the wings of the anteroba- mamillary/retromamillary (M/RM) longitudinal areal com- sal nucleus, whose rostromedian fused part is acroterminal plexes. The large Tu/RTu subdivides dorsoventrally into 3 (ABasM; ABasW; Fig. 1b; Puelles et al. 2012). The sharp Fig. 2 The tuberal/retrotuberal Isl1 marker delimits negatively the while restricted to the VMH primordium, does not fill completely all whole VMH primordium early during development. This panel shows its volume; note particularly a ventral (VMHvm) triangular sector of sagittal sections of the basal hypothalamus at E14.5 a–c and E16.5 the VMH which is also delimited by Isl1 cells, but devoid of Nkx2.2 d, f labeled with double immunofluorescence for Nkx2.2, and Isl1; cells. Scale bars represent 200 µm. TuI tuberal intermediate, RTuD the c, f panels compare the separated red channel a, d and green retrotuberal dorsal, TuD tuberal dorsal, VMHim medial-intermediate channel b, e images in adjacent sections. The basal areas surround- VMH subnucleus; VPa ventral paraventricular nucleus, PSPa pedun- ing (delimiting) the VMH primordium, including the TuI and TuD cular subparaventricular area, VMHvm ventral-medial VMH subnu- areas, are marked by intense Isl1 labeling (green), with few Isl1 cells cleus, Arc arcuate nucleus, VMHdrm dorsal-rostromedial VMH sub- dispersed within the primordium. In contrast, Nkx2.2 labeling (red), nucleus 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 3 Adult distribution of Nkx2.2-positive cells in the VMH. a–g and lateral dorsorostral subnuclei (VMHdrm, VMHdrl; Caudorostral series of transverse sections orthogonal to the proso- Fig. 3b–e, h). We also subdivided the intermediate VMH meric forebrain axis taken through the adult terminal hypothalamus part (classic ‘central’ part) into medial and lateral-interme- to examine systematically the VMH nucleus, and double-reacted for diate subnuclei (VMHim, VMHil; Fig. 3b–h). Finally, the Nkx2.2 ISH (blue) and TH IHC (brown). The Nkx2.2-positive cells are distributed differentially over diverse identified VMH subdivi- large ventral VMH region (old ‘ventrolateral part’) was sub- sions. a Nkx2.2 signal is absent at the caudalmost dorsal VMH level, divided into three parts, the medial, intermediate, and lateral where we distinguish the VMHdc. (b–f) Progressing rostralward in ventral subnuclei (VMHvm, VMHvi, VMHvl; Fig. 3c–h; the series, Nkx2.2 signal is strongly expressed laterally at VMHdrl note the equivalences with the older terminology obviously and VMHil, and extends into VMHvi; there is also weaker expres- sion at the VMHdrm and VMHim. In contrast, VMHvm, and VMHvl are not exact; see Table 3). are largely devoid of Nkx2.2 signal. g The rostral-most level of the VMH is also largely Nkx2.2 negative and can probably be interpreted Distribution of gene expression as VMHim and VMHvm. h Sagittal section of the adult medial hypo- during development thalamus marked with Nkx2.1 IHC in brown and Otp ISH in blue (from Puelles et al. 2012). Red lines represent the section levels in a–g; see also caudal/C and rostral/R orientations. Scale bars represent We searched genes expressed in mouse TuD and TuI at 200 µm. PoA preoptic area, RPa retroparaventricular area, AH ante- E13.5 and E18.5 to characterize the expected differential rior hypothalamic nucleus; A/B, alar/basal limit, VMHdc dorsocau- molecular profile of the VMH, using the AGEA tool of the dal subnucleus of the ventromedial hypothalamic nucleus, DMH-T terminal part of the dorsomedial hypothalamic nucleus, VPM ventral Allen Developing Mouse Brain Atlas database (https://de vel premamillary nucleus, DPM dorsal premamillary nucleus, M mamil- opingm ouse.b rain-m ap.o rg). Tables 1, 2 classify 22 relevant lary area, VMHdrm dorsal-rostromedial subnucleus of the ventrome- genes according to their characteristic expression domains at dial hypothalamic nucleus, VMHdrl dorsal-rostrolateral subnucleus of E13.5 and E18.5, respectively. Representative markers are the ventromedial hypothalamic nucleus, VMHim medial-intermediate subnucleus of the ventromedial hypothalamic nucleus, VMHil lateral- shown in Fig. 5, with the respective tridimensional signal intermediate VMH subnucleu, SCh suprachiasmatic nucleus, VMHvm distribution presented in medio-laterally ordered sagittal sec- ventral-medial VMH subnucleus, VMHvl ventral-lateral VMH sub- tion planes taken from the Allen Developing Mouse Brain nucleus, VMHvi ventral-intermediate VMH subnucleus, ac anterior Atlas. Although there clearly are areas of overlap between comisure, DMH-P peduncular part of the dorsomedial hypothalamic nucleus, PRM periretromamillary area, Th thalamus, PTh prethala- the diverse markers, a dorsoventral, rostrocaudal and medi- mus olateral pattern can be distinguished within the prosomeric model that underpins the detailed topographic subdivision boundary of the VMH proper can be assessed (apart from we propose (VMHdc/VMHdrm/VMHdrl; VMHim/VMHil; other possibilities) by its lack of Isl1 signal, in contrast with VMHvm/VMHvi/VMHvl). the Isl1-positive rest of the tuberal region (Fig. 2). Impor- It is possible to cluster the markers studied at E13.5 into tantly, some VMH neuronal subpopulations seem to have five positionally distinct subgroups (alar, only TuD, only non-TuI origins. They share early molecular markers with TuI, TuD + TuI, only acroterminal; Table 1). These results cell populations of the overlying TuD region (either its sub- can be compared with the respective VMH expression pat- liminal or main parts), or, alternatively, of neighboring TuD/ tern at E18.5 (Table 2). TuI acroterminal areas, or even of the overlying alar TSPa domain, where some VMH cells apparently originate, as 1) The Vax1 gene appears expressed selectively at E13.5 we will illustrate below. Puelles et al. (2012) previously along the longitudinal subparaventricular alar domain suggested a dorsoventral migration of cells expressing the (both THy and PHy; Fig. 4w, x; note this domain cor- Nkx2.2 marker into the VMH primordium. responds to the ‘hypothalamic diagonal’ of Shimogori Our material corroborates the already previously recog- et al. 2010 but is conceived here as strictly longitudinal; nized inner heterogeneity of VMH. Dorsomedial, central, check Fig. 1); at E18.6 a Vax1-positive subpopulation and ventrolateral VMH parts are conventionally described was identified within VMHdrm/VMHdrl (Fig. 5k). in the literature, but we prefer to characterize them as dor- 2) 4 genes -Sema3a, Nkx2.2, Fbxw7, and Tcf7l2- were sal, intermediate, and ventral regions, respectively (consist- selectively expressed at E13.5 at the TuD zone (implying ently with the prosomeric axial dimension; see Fig. 1). We separately either the main or subliminal parts of TuD, or found that a slightly more detailed subdivision was needed encompassing both of them, as well as the correspond- to describe fully the observable molecular diversity. We sub- ing TuD acroterminal portion; Fig. 4a, b, e, f, m, n). The divided the dorsal part of VMH into a dorsocaudal element same markers subsequently label at E18.5 mainly our (VMHdc; this is the old ‘dorsomedial’ portion; Fig. 3a, h; dorsal VMH subdivisions (VMHdc/VMHdrm/VMH- see Table 3) and two novel components called by us medial 1 3 Brain Structure and Function Table 1 Expression of 22 SPa (Alar) Subl.TuD Main TuD TuI TuD-AT TuI-AT gene markers at the terminal neurogenic regions of the Vax1 + hypothalamus near the VMH Sema3a + primordium at E13.5 Nkx2.2 + + + Fbxw7 + + Tcf7l2 + + Nr2f1 + Lmo4 + Satb2 + Nr5a1 + + + Adcyap1 + + + Robo2 + + + Calb1 + + Slc17a6 + + Cnr1 + + + Enc1 + + + Sox14 + + + Bcl11a + + + Dner + + + Chl1 + + Mapt + + + + Nkx2.1 + + + + Six3 + + SPa subparaventricular area, Subl.TuD subliminal tuberal dorsal area, Main TuD main tuberal dorsal area, TuI tuberal intermediate area, TuD-AT acroterminal section of the tuberal dorsal area, TuI-AT acroterminal section of the tuberal intermediate area drl) but also extend variously into the intermediate ones in the VMHdc, VMHdrm, and VMHvl, with some vari- (VMHim/VMHil) (Fig. 5b, f). able extra locations. 3) Three genes -Nr2f1, Lmo4, Satb2- appeared expressed 5) Six3 was initially expressed at E13.5 at the acroterminal exclusively at the TuI domain at E13.5 or earlier tuberal region (as well as in alar acroterminal regions; (Fig. 4k, l, s, t); these labeled subsequently our interme- Fig. 4i, j; note this gene labels the prospective acroter- diate and ventral VMH subdivisions at E18.5 (VMHim/ minal domain already from neural plate stages onwards; il, VMHvm/vl; Tables 2; Fig. 5e, i). Two of these mark- Lagutin et al. 2003), with some extension into TuI. Dis- ers included also the VMHdc subdivision (old dorsome- tinct Six3-labeling was displayed later at E18.5 by the dial part). novel VMHvi subnucleus (Fig. 5d; note this distinct 4) 12 genes in Table 1 are expressed both within TuD subnucleus was never described before within the clas- and TuI domains at E13.5 (Fig. 4c, d, g, h, q, r, o, p, sic ‘ventrolateral’ sector). These Six3-positive cells are u, v). They can we regrouped according to the location interpreted to migrate from the acroterminal TuD area at E18.5 of the corresponding labeled cell populations and/or the arcuate TuI acroterminal area, which also (Fig. 5a, c, g, h, j): Nr5a1 and Robo2 have a similar expresses initially Six3. E18.5 pattern, labeling only the VMHdc, VMHdrm, and VMHim subnuclei. Calb1 and Slc17a6 share labeling within VMHdrm, VMHim, and VMHil at E18.5 (though There clearly exist areas of overlap between most of the Slc17a6 signal also extends into VMHdc). Several genes markers studied, indicating that most VMH subdivisions -Enc1, Bcl11a, Dner, Chl1- share a strong presence at are subtly heterogeneous in molecular profile, with regional the VMHvm and VMHvl subdivisions, though their sig- variations. This is consistent with the molecular diversity nals variously spread also into other neighboring subnu- observed in transcriptomic studies (see Discussion). The clei. The genes Cnr1, Mapt, and Nkx2.1 share expression VMH nevertheless shows on the whole significant partial 1 3 Brain Structure and Function Table 2 Expression of 22 gene VMHdc VMHdrm VMHdrl VMHim VMHil VMHvm VMHvl VMHvi markers within the different VM subnuclei at E18.5 Vax1 + + Sema3a + + Nkx2.2 + + + + + Fbxw7 + + Tcf7l2 + Nr2f1 + + + + Lmo4 + + + Satb2 + + + + Nr5a1 + + + Adcyap1 + + + + Robo2 + + + Calb1 + + + Slc17a6 + + + + Cnr1 + + + + Enc1 + + + Sox14 + + + + Bcl11a + + + + Dner + + + + + Chl1 + + + + Mapt + + + + + Nkx2.1 + + + + + + Six3 + + + VMHdc dorsocaudal VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHdrl dorsal- rostrolateral VMH subnucleus, VMHim medial-intermediate VMH subnucleus, VMHil lateral-intermediate VMH subnucleus, VMHvm ventral-medial VMH subnucleus, VMHvi ventral-intermediate VMH subnu- cleus, VMHvl ventral-lateral VMH subnucleus dorsoventral, mediolateral, and rostrocaudal sorting of its associated to the TuI progenitor area—cases of Satb2, molecularly distinct subpopulations. These may be ten- Nr2f1, and Nkx2.1- imply a priori a radial migration pattern, tatively classified according to their apparent positional whereas origins associated to the overlying alar TSPa or to neuroepithelial origins, tangential versus radial migration the TuD—cases of Vax1, Nkx2.2, Nr5a, Tcfl2, and Sox14- routes, and differential molecular profile. The classic schema suggest tangential dorsoventral displacements. At least in of three VMH parts (dorsomedial, central, ventrolateral) is the case of Six3 it is possible to study separately a restricted too simple to account for the level of heterogeneity observed acroterminal origin, which would involve a priori a rostral (e.g., it completely misses the Six3-positive VMHvi subdivi- source and a caudally oriented tangential displacement. sion as well as the dorsorostral VMH subdivisions described here and our medial–lateral distinctions; Table 3). Vax1 Development of ventromedial nucleus Vax1 is a typical early selective marker of the alar sub- subpopulations marked with selected genes paraventricular area, which extends over the terminal and peduncular hypothalamic domains (PSPa and TSPa) and To analyze in more detail the variant development of VMH includes at its ventral rim the liminal alar Nkx2.2-positive subpopulations according to their progenitor origins, we zone (Puelles et al. 2012; in blue in Fig. 1a, b). Initially the selected for follow-up a group of genes expressed in dis- expression of Vax1 ends strictly at the alar-basal boundary. tinct progenitor areas at E13.5 whose derivatives constitute Vax1-labeled cells start to move into the subjacent basal characteristic VMH subpopulations at E18.5: Vax1, Nkx2.2, tuberal region at E13.5 (Fig. 6a, b). Two days later, this Nr5a1, Satb2, Tcfl2, Sox14, Nr2f1, Nkx2.1, and Six3. Origins marker delineates rostrodorsal parts of the VMH primordium 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 4 Differential dorsoventral expression of VMH markers at larger ventral part of the TuD/RTuD area as ‘the main TuD/ E13.5 in two neighboring sections for each marker (identified at left). RTu’ (Fig. 1b). These distinctions are not contemplated in This panel illustrates various sagittal ISH images of markers listed in the columnar tradition (Table 3). our Tables 1, 2 taken from E13.5 mice material at the Allen Devel- Results on zebrafish embryos indicate that the earliest oping Mouse Brain Atlas. The dashed black line identifies the alar- basal boundary (identifying D to the right and V to the left; R to the appearance of Nkx2.2 signal occurs at neural plate stages bottom). TuI, main TuD and subliminal TuD are delimited by solid (Hauptmann and Gerster 2000; Hauptmann et al. 2002). In lines. Other hypothalamic domains are not identified, being outside E12.5 mouse embryos Nkx2.2 is already expressed longitu- the scope of this analysis. The observed patterns vary in the degree in dinally along the alar-basal boundary of the midbrain, dien- which TuI versus TuD, or both, show expression at this stage. Scale bar in k represents 200 µm. RTuI retrotuberal intermediate area, TuI cephalon, and hypothalamus (Fig. 7a). At E13.5, Nkx2.2- tuberal intermediate area, main or subl. RTuD main or subliminal ret- expressing cells start to migrate ventralwards (mainly within rotuberal dorsal area, main or subl. TuD main or subliminal tuberal THy) and the corresponding signal begins to protrude dorsal area ventralwards under the terminal subliminal band strictly expanding within the basal tuberal region; this represents the (Fig. 6c, d; note this result changes the classic simpler con- first sign of the migration of Nkx2.2 cells into the VMH pri- cept of the dorsal region of VMH, contemplating only the mordium (primVMH; Fig. 7b). This process is clearly more caudally placed ‘dorsomedial VMH part’; obviously, our advanced at E14.5 (primVMH; Figs. 7b, c, 8a–c). Cross- ‘dorsorostral’ descriptor refers to the prosomeric axis and sections at E14.5 illustrate that Nkx2.2 signal is present at columnar authors may want to use a different descriptor). the ventricular and mantle zones of the subliminal and main At postnatal stages the tuberal expression of Vax1 coincides parts of the TuD, as well as in the underlying VMH nuclear mainly with the VMHdrm/VMHdrl subnucleus with some primordium (only mantle zone), while the TuI ventricular dispersion into VMHim/VMHil (Figs. 5k; 6d). zone deep to VMH remains negative (primVMH; Fig. 7d–g). At E18.5 there is already a definitive distribution pattern of Nkx2.2 Nkx2.2 signal within VMH (Fig. 8d–g). The dorsoventral tubero-tuberal VMH migration stops within TuI at some dis- Nkx2.2 is well known as an early longitudinal marker of the tance of the perimamillary band, leaving a substantial space alar-basal boundary along the whole forebrain tagma (mid- for the terminal part of the DMH nucleus, as well as for brain, diencephalon, and hypothalamus; note this is a basic the separately tangentially migrated VPM nucleus (López- token of the updated extended forebrain concept held within González et al. 2021). Interestingly, at E13.5-E14.5 there is the prosomeric model; Puelles et al. 2012; Puelles and also a previously unidentified shorter parallel dorsoventral Rubenstein 2015; Puelles 2018; Amat et al. 2022). This lin- Nkx2.2 migration coming out of the peduncular retrotuberal eal signal also appears bordering the spike of Shh expression subliminal band that extends ventralwards into RTu, where that marks the transverse alar zona limitans intrathalamica it stops in contact with the Otp-positive periretromamillary (alar p3/p2 boundary), also known as the mid-diencephalic band; this parallel migration is distinctly separated from the organizer (Puelles and Martínez 2013). This gene (as well VMH and may contribute Nkx2.2-expressing neurons to the as others such as Nkx2.9, Ptc, etc.) is apparently selectively dorsomedial nucleus (Fig. 7b, c). upregulated by particularly high local concentrations of We performed some in vitro fluorescent labeling experi- SHH signal diffusing dorsalward from the underlying floor ments on organotypic cultures of embryonic hypothalamus and basal plate Shh expression domain or from the related to visualize the ventralward tubero-tuberal migration into ZLI core domain which obeys a different enhancer (Briscoe VMH originated from the primary Nkx2.2-expressing band et al. 1999; Puelles and Martínez 2013; Nishi et al. 2015; across the alar-basal boundary of THy. A small particle of Andreu-Cervera et al. 2018). In the hypothalamus, Nkx2.2 CMFDA (see Methods) was placed at the TuD in E12.5 signal appears early on as a thin longitudinal band that over- explants and further analysis was done at E14.5, at which laps the lineal boundary between the alar and basal plates stage the VMH is already fairly well formed, though the and stops as it reaches the acroterminal border (Fig. 7a). migration is not complete yet. All explants were treated The mixed alar-basal expression led to the concepts of alar for Nkx2.2 immunofluorescence to check the position of liminal and basal subliminal subdivisions of the Nkx2.2 band the CMFDA particle, and in all cases (n = 6) a stream of (Puelles et al. 2012; ‘liminal’ refers to the classic notion of CMFDA-labeled cells overlapped with the Nkx2.2-positive limen lamina alaris, or rim of the alar plate; see red and blue cells advancing into TuI, showing a comparable disposition bands in Fig. 1a, b). The subliminal subdomain thus forms (n = 4 in Fig. 8h–k). Some Nkx2.2/CMFDA-positive migrat- the upper rim of the TuD/RTuD basal tuberal hypothalamic ing cells were observed relative to the labeling sites placed progenitor area, whereas the liminal area is a subdomain at across the width of the Nkx2.2-positive band (white arrows the ventral rim of the subparaventricular hypothalamic area in Fig. 8h’–k’). This reveals that some dorsoventrally migrat- across THy and PHy. We will refer to the non-subliminal ing cells do not express Nkx2.2, which possibly includes 1 3 Brain Structure and Function 1 3 Brain Structure and Function Fig. 5 Differential dorsoventral expression of VMH markers at to illustrate the relevant five mediolateral section levels available. E18.5 or P4. This Figure (in two parts) presents for each marker We tentatively identified the VMH subdivisions described in the text (identified at left) a mediolateral series of five parallel sagittal ISH (VMHdc, VMHdrm, VMHdrl, VMHim, VMHil, VMHvm, VMHvi, images of E18.5 (or P4, indicated) mice embryos from the Allen and VMHvl). Scale bars represent 200 µm. VMHim medial-inter- Developing Mouse Brain Atlas illustrating the expression pat- mediate VMH subnucleus, VMHdc dorsocaudal VMH subnucleus, tern of the markers identified in our Tables 1, 2 across VMH. The VMHdrm dorsal-rostromedial VMH subnucleus, VMHvi ventral- expression patterns are compared to the outlines of Nkx2.2 (red out- intermediate VMH subnucleus, VMHil lateral-intermediate VMH line), Nr5a1 (black outline), and Six3 (green outline) expression. It subnucleus, VMHdrl dorsal-rostrolateral VMH subnucleus seemed informative towards 3D assessment of relative topography the Nr5a1-expressing cells (see next section) and/or passing VMH primordium (Fig. 8k, k’). This result is consistent with Vax1 cells coming from the overlying alar subparaventricu- descriptive images in which Nkx2.2-expressing cells seem lar area (previous section). In one experiment in which the to originate throughout the TuD/RTuD boundary, appearing CMFDA particle was placed at the TuD/RTuD limit where it in transient continuity with the VMH primordium (Figs. 7b, crosses the orthogonal THy/PHy border migrated CMFDA- 8k). Nkx2.2 double-labeled cells were found also within the 1 3 Brain Structure and Function 1 3 Table 3 Terminological equivalence of the main nuclei and histogenetic areas contemplated in this report according to the nomenclature of the columnar and prosomeric forebrain models Columnar term Abbr Columnar histoge- Prosomeric term Nuclear Abbr Prosomeric histoge- Areal Abbr Prosomere Hypothalamic part Population origin netic area netic area related literature Supramamillary SUMm Mamillary Retromamillary RMM Retromamillary area RM hp1 PHy nucleus medial part nucleus medial part Supramamillary SUMl Mamillary Retromamillary RML Retromamillary area RM hp1 PHy nucleus lateral part nucleus lateral part Medial mamillary MM Mamillary Medial mamillary MM Mamillary area M hp2 THy nucleus Lateral mamillary LM Mamillary Lateral mamillary LM Mamillary area M hp2 THy nucleus nucleus Posterior hypothala- PHr Mamillary Periretromamillary PRM Periretromamillary PRM hp1 PHy band area mus, hypothalamic part Posterior hypothala- PHc Prerubral tegmentum Prerubral tegmentum p3Tg + p2Tg p3 and p2 basal plate p3B, p2B p3 + p2 Dienc Puelles et al. (2012) mus, diencephalic part Tuberomamillary TM Mamillary Perimamillary PM Perimamillary area PM hp2 THy nucleus + Dorsal band + Dorsal pre- DPM Retrotuberal ventral premamillary mamillary nucleus area and Tuberal nucleus ventral area Tuberomamillary his- TM Mamillary Ventral retrotuberal RTuV + TuV Ventral retrotuberal RTuV + TuV hp1 + hp2 PHy + THy taminergic neurons and tuberal areas and tuberal areas Ventral premamillary PMv Mamillary Ventral premamillary VPMc Migrated tangentially RM hp2 THy López-González et al. nucleus nucleus, core from retromamil- (2021) lary área (located Ventral premamillary VPMsh TuI) nucleus, shell Dorsomedial hypo- DMHa Tuberal Dorsomedial hypo- DMHc/s-P Intermediate retrotu- RTuI hp1 PHy Díaz et al. (2015) thalamic nucleus, thalamic nucleus, beral area anterior part peduncular part (core/shell) Dorsomedial hypo- DMHp/v Tuberal Dorsomedial hypo- DMHc-P Intermediate tuberal TuI Hp2 THy Díaz et al. (2015) thalamic nucleus, thalamic nucleus area posterior/ventral terminal part (core/ part shell) Brain Structure and Function 1 3 Table 3 (continued) Columnar term Abbr Columnar histoge- Prosomeric term Nuclear Abbr Prosomeric histoge- Areal Abbr Prosomere Hypothalamic part Population origin netic area netic area related literature Subthalamic nucleus STN Mamillary Subthalamic nucleus STh Migrated dorsalward RM hp1 PHy Skidmore et al. (2008) from retromamil- lary area Parasubthalamic PSTN Mamillary Parasubthalamic PSTh Mifrated dorsalward RM hp1 PHy nucleus nucleus from retromamil- lary area Arcuate hypothalamic ARH Tuberal Arcuate nucleus Arc Intermediate tuberal TuI hp2 THy (acroterminal) nucleus área (acroterminal part) Ventromedial hypo- VMHdm Tuberal Ventromedial hypo- VMHdc Intermediate tuberal TuI hp2 THy Present results thalamic nucleus, thalamic nucleus, area dorsomedial part dorsocaudal part Ventromedial hypo- VMHa Tuberal Ventromedial hypo- VMHdrm Dorsal tuberal TuD + SPa hp2 THy Present results thalamic nucleus, thalamic nucleus, área + Alar subpara- anterior part rostromedial dorsal ventricular area part Ventromedial hypo- VMHdrl thalamic nucleus rostrolateral dorsal part Ventromedial hypo- VMHc Tuberal Ventromedial hypo- VMHim Dorsal tuberal area TuD hp2 THy Present results thalamic nucleus, thalamic nucleus, central part (core) intermedio-medial part Ventromedial hypo- VMHil thalamic nucleus, intermedio-lateral part Ventromedial hypo- VMHvl Tuberal Ventromedial hypo- VMHvm Intermediate tuberal TuI hp2 THy Present results thalamic nucleus, thalamic nucleus, area ventrolateral part ventromedial part Ventromedial hypo- VMHvi thalamic nucleus, ventro-intermediate part Ventromedial hypo- VMHvl thalamic nucleus, ventrolateral part For the columnar terms (italic) we used the terminology, schemata, and table of Hahn et al (2019), whereas for the prosomeric terms (roman) we used Puelles et al (2012) and VMH subdivision concepts introduced in the present report as reference Brain Structure and Function Fig. 6 Apparent alar origin of Vax1-expressing VMH cells. a, b plays Vax1 transcripts mainly at its VMHdrm/VMHdrl subdivisions Sagittal images illustrating two Vax1 ISH-reacted sections in E13.5 (shown at P4). Scale bars represent 200 µm. RTuI retrotuberal inter- mice from the Allen Developing Mouse Brain Atlas. The black mediate area, TuI tuberal intermediate area, main or subl. RTuD, dashed line identifies the alar-basal boundary (alar to the right; basal main or subliminal retrotuberal dorsal area; main or subl. TuD main to the left, as in Fig. 1b). The TuI, main TuD and subliminal TuD or subliminal tuberal dorsal area, PSPa peduncular subparaventricular progenitor domains are delimited by solid lines. At b there appears area, TSPa tuberal subparaventricular area, primVMH primordium of Vax1 signal penetrating the underlying TuD area. c At E15.5, the area the ventromedial hypothalamic nucleus, VMHdrm dorsal-rostrome- showing displaced Vax1 labeling extends ventralwards into a dorsal dial VMH subnucleus part of the VMH primordium. d The final VMH configuration dis- Nr5a1 it), including the associated acroterminal TuD domain, fully devoid of Nkx2.2 signal (Fig. 9a, d, g, k). At E13.5, Nr5a1- A comparison of the relatively complementary expression of positive cells are present at the deepest stratum of the local (TuD) mantle, extending into TuI (Fig. 9d–f, k–n). These Nkx2.2 and Nr5a1 within VMH illustrates the basic organi- zation we propose for the ventromedial primordium. Nr5a1 cells thereafter invade the deep region of the underlying VMH primordium (Fig. 9k–n). The migrating deep stream (previously known as steroidogenic factor 1, SF1), has been widely analyzed developmentally and functionally, and is of Nr5a1 cells thus seems covered by the more superficially migrating separate stream of Nkx2.2 cells (Fig. 9a–f, h–j, sometimes considered a selective transient marker of the entire VMH nucleus (Ikeda et al. 1995; Shinoda et al. 1995; l–n). The migration of Nr5a1 cells originating at the Nr5a1- positive acroterminal TuI area is oriented ventrocaudally, Dellovade et al. 2000; Tran et al. 2003; Davis et al. 2004; Dhillon et al. 2006; Kim et al. 2011, 2019; Büdefeld et al. apparently incorporating into the main TuD migration into VMH. Once the Nr5a1 cells reach the deep part of VMH 2012; Cheung et al. 2013; see Discussion). However, this contrasts with our present data, given that most Nkx2.2- they spread out radially and tangentially within the nucleus (Fig. 9l–n). At E16.5 Nr5a1 cells are visible mainly medially positive VMH cells seem to be Nr5a1 negative. Nkx2.2 has received relatively less attention so far as a VMH marker at the VMHdrm, and VMHim subdivisions, with a medio- laterally decreasing density gradient partially overlapping (Kurrasch et al. 2007; Puelles et al. 2012; Corman et al. 2018). with Nkx2.2 cells in an inverse lateromedial gradiental distri- bution. At this stage we found intermixed Nr5a1 and Nkx2.2 In contrast to Nkx2.2, the gene Nr5a1 is excluded early on from the narrow subliminal TuD band but is strongly populations at the VMHdrm and VMHim, whereas Nr5a1 cells are absent at the VMHil subnucleus where Nkx2.2 cells expressed at the underlying intrinsically Nkx2.2 negative main TuD area (though migrating Nkx2.2 cells pass through are massively present (Fig. 9o–q). 1 3 Brain Structure and Function Fig. 7 Evolution of Nkx2.2 and Otp signals relative to the emer- d–g), and this process later expands rostralwards into THy, in parallel gence of the VMH primordium between E12.5 and E14.5. All pan- to the ventralward migratory phenomena leading to the tuberal VMH; els show double Nkx2.2 ISH/Otp IHC in sagittal a–c or oblique d–g the intercalated subparaventricular alar area (PSPa) is crossed sub- sections from E12.5, E13.5 and E14.5 embryos (levels of section in ventricularly by the dorsally migrating cells, but essentially remains d–g indicated in c). The dash black line always marks the alar-basal unlabelled, excepting some liminal expression (PPa; TSPa; prim- boundary and roughly parallel solid black lines delimit the sublimi- VMH; a–c). Our oblique sections through this area intersect the nal and main TuD areas in (a–c). The curved red line in a–c iden- VMH primordium and the PPaV at four levels, providing images tifies the hypotalamo-telencephalic boundary dorsal to the Otp-pos- consistent with a migratory interpretation. Note that the VMH Nkx2.2 itive paraventricular area (PPa; brown). b The blue Nkx2.2 signal cell population does not represent the entire VMH population. Scale previously related only to the subliminal TuD (a) now clearly starts bars represent 200 µm. ZLi zona limitans, Main or subl. TuD main or to extend ventralwards into the TuI territory at E13.5 (primVMH). c subliminal tuberal dorsal area, PPa peduncular paraventricular area, This process progresses considerably by E14.5 (primVMH). Note that PPaV ventral peduncular paraventricular nucleus, Tel telencephalon, a less important ventral expansion of Nkx2.2 cells is observed like- PoA preoptic area, RM retromamillary area, M mamillary area, PRM wise at the RTuD domain, which comes to contact the Otp-positive periretromamillary area, PM perimamillary area, TuI tuberal interme- periretromamillary band (PRM). d–g The original early Nkx2.2-pos- diate area, primVMH primordium of the ventromedial hypothalamic itive subliminal band (blue) shows a dorsalward migration into the nucleus, PTh prethalamus, HyA hypothalamo-amygdalar corridor, ventral part of the Otp-positive alar paraventricular area (PPaV; a–c; PSPa, peduncular subparaventricular area, A/B limit alar/basal limit 1 3 Brain Structure and Function Fig. 8 A Nkx2.2-expressing population partially connected with the fication images from the dashed square areas in (h–k). White arrows TuD reaches the TuI domain to form part of the VMH primordium. in h’,j’,k’ indicate cells that correspond to double-labeled CMFDA- a–g Sagittal sections in medio–lateral order comparing the VMH Nkx2.2 elements. The white arrows in a, i indicate the direction of primordium labeled with anti-Nkx2.2 antibody in red fluorescence Nkx2.2 or CMFDA label extension at higher magnification in the at migratory stage E14.5 (a–c) and postmigratory stage E18.5 (d–g). respective framed areas. Scale bars in c, h represent 200 µm. Scale The approximate contour of the whole VMH nucleus was traced with bar in h’ represents 100 µm. RTuI retrotuberal intermediate area, TuI a white line in f. At E18.5 the specific subdivisions of the VMH that tuberal intermediate area, RTuD retrotuberal dorsal area, TuD tuberal are invaded by this cell type can be identified tentatively (VMHdrm, dorsal area, primVMH primordium of the ventromedial hypothalamic VMHim, VMHil, VMHvi). h–k Representative sagittal images of nucleus, PPaV ventral peduncular paraventricular nucleus, VMHdrm our tracing experiments on organotypic cultures from E12.5 mice dorsal-rostromedial subnucleus of the ventromedial hypothalamic (48 h culture conditions, E12.5-E14.5), marked with the anti-Nkx2.2 nucleus, VMHim medial-intermediate VMH subnucleus, VMHvi ven- antibody (red), and illustrating ventralwards cell migration when tral-intermediate VMH subnucleus, VMHil lateral-intermediate VMH a CMFDA particle (black) was placed at the upper part of TuI (h), subnucleus the TuD (i, j), or the RTuD-TuD boundary (k). (h’–k’) Higher magni- 1 3 Brain Structure and Function Tcf7l2/Satb2/Sox14 overlap between them (Fig. 12e, e’). At E14.5 the VMH primordium contains both examined populations; Nkx2.2 At E13.5 the Tcf7l2 gene labels the TuD acroterminal man- cells are largely restricted to its dorsal part, whereas Nr2f1 tle, extending somewhat into the neighboring rostral TuI. cells appear mainly in its ventral part, with a small overlap This Tcf7l2-TuI pattern is partially complementary to that visible in sagittal sections (Fig. 12f). Using horizontal sec- of Satb2, whose expression occurs mainly at the periven- tions oriented parallel to the prosomeric alar-basal boundary tricular stratum of TuI (Fig. 10a–h). At E15.5, the primor- (see Fig. 1), the ventricular zone observed in the dorsalmost dium of the VMHdrm subnucleus is selectively marked by sections through basal hypothalamus (next to the alar-basal Tcf7l2, whereas the VMHdc primordium and the ventral boundary) contains Nkx2.2-positive elements, but no Nr2f1 parts of VMH are marked with Satb2 (Fig. 10i–p). At late cells (Fig. 12g–g’). On the contrary, at levels through the TuI developmental stages, the rounded VMHdrm is perfectly the ventricular zone shows Nr2f1 expression but no Nkx2.2 delineated by Tcf7l2, as is the VMHdc portion by Satb2 signal, though migrated Nkx2.2 cells are present in the man- (Figs. 10q–x, 11a–f). The latter gene also marks the three tle layer (Fig. 12h–h’). We also compared at E16.5 the Nr2f1 VMHvm, VMHvi and VMHvl divisions, some of which VMH subpopulation with Nkx2.2 or Nr5a1-marked cells. are also recognizable with Sox14, another TuI-TuD early Although a degree of overlap exists between any of these marker, which appears expressed at the VMHdrl, but not at populations, there are still indications of a partially differen- the VMHdrm companion, and similarly at the VMHvm, but tial dorsoventral distribution, with Nr5a1 and Nkx2.2 cells not at the VMHvl (Fig. 11g–i). occupying more importantly the VMHdrm subnucleus, and Nr2f1 cells preferentially appearing in ventral parts of VMH Nr2f1 where the other two markers are absent (Fig. 13a–d). The Nr2f1 marker (previously known as Couptf1) apparently Nkx2.1 labels a VMH subpopulation arising exclusively within TuI. This gene is expressed throughout the TuI/RTuI domains Nkx2.1 is widely expressed in the hypothalamic basal plate of the basal hypothalamus (excepting the corresponding at early stages (starting at neural plate stages; Shimamura acroterminal area) at E11.5 (not shown; see Allen Develop- et al. 1995; Qiu et al. 1998) but is absent at the retromamil- ing Mouse Brain Atlas) and E13.5 (Fig. 12a, b). The TuD/ lary area (RM) and its migrated VPM and STh derivatives RTuD area where early expression of Nkx2.2 and Nr5a1 (Puelles et al. 2012; López-González et al. 2021); there is was observed is devoid of Nr2f1 signal (Fig. 12a, b). At also a thin longitudinal band of Nkx2.1-positive cells ven- E15.5, Nr2f1 signal appears both within the dorsomedial trally within the SPa alar area; this may correspond to its hypothalamic nucleus (DMH; hp1 + hp2) and the VMH liminal subdomain and its origin remains uncertain (van (hp2; Fig. 12c, d). The latter formation shows two kinds of den Akker et al. 2008; Puelles et al. 2012; LP, unpublished labeling: an abundant Nr2f1 signal that marks particularly observations). At E13.5, Nkx2.1 is strongly expressed in the ventral parts of the VMH primordium, and dispersed the ventricular and mantle zones of the TuD and TuI pro- Nr2f1 neurons within the prospective intermediate part of genitor domains (Fig. 14a–c). Moreover, Nkx2.1 marks also VMH (red and blue asterisks in Fig. 12b, d). Subsequently the whole basal acroterminal territory from E11.5 onwards this marker appears strongly expressed at the VMHdc, and (Allen Developing Mouse Brain Atlas data; Fig. 14a; Puelles ventral portions of VMH. It also shows weaker expression at et al. 2012, their Figs. 8.9D; 8.10D). At E15.5, the VMH the transitional limits of the VMHim, VMHil, and VMHdrm primordium contains deep Nkx2.1-positive neurons in its subdivisions (Figs. 5i, 13a–d). ventral and dorsal parts (not so at the medial-intermediate We compared with double immunofluorescence the dis- part), and separate superficial cells are observed ventrally tribution of Nkx2.2- versus Nr2f1-positive cells. At E12.5, (Fig. 14k–m). At E18.5 the Nkx2.1 signal largely occupies the longitudinal Nkx2.2-positive hypothalamic progenitor the VMHdc, VMHdrm (with slight extension into VMHdrl), territory appears in sagittal sections dorsal to the longitu- VMHil VMHvm, and VMHvl subnuclei (Fig. 14n–p). dinal domain marked by Nr2f1, with only a small area of 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 9 Nkx2.2 and Nr5a1 reveal different migrated VMH popula- Six3 tions. a–n Nkx2.2 and Nr5a1 ISH reactions are compared in medi- olateral sagittal (a–f) and dorsoventral horizontal (g–n) sections of This is a particularly distinct and previously undescribed E13.5 mouse embryos from the Allen Developing Mouse Brain Atlas partial marker of the VMH nucleus, where it identifies a (horizontal relative to the prosomeric axis), illustrating the postulated zones of origin of the populations expressing selectively one or the new subdivision. Early on, Six3 is strongly expressed at the other of these markers and their topographic relationship with the acroterminal parts of TuD and TuI, the latter representing ventrally displaced and expanding VMH primordium, where both the primordium of the arcuate nucleus, median eminence populations remain spatially distinct. Sections (b, c, g, h) also show and infundibulum (Fig. 15a). In contrast, the neighboring Nkx2.2-expressing cells that migrate dorsalward into peduncular and terminal parts of the ventral paraventricular subarea (migPaV). subliminal and main TuD domains are Six3-negative, as The black dash line in b, e marks the prosomeric horizontal section well as the incipient VMH primordium (Fig. 15b). At E16.5, plane illustrated at various levels in g–n. The section in g is slightly though, strong Six3 signal can be identified at the acrotermi- oblique and shows at the left side alar (peduncular and terminal) nal arcuate and anterobasal nuclear primordia (basal plate) PSPa and TSPa regions bordering ventrally the liminal ventricular zone, whereas at the right side we see the transition into the underly- and the suprachiasmatic primordium (alar plate) while the ing basal subliminal ventricular zone; note none of these sides shows VMH primordium shows a novel strongly Six3-positive significant mantle layer labeling of VMH nature, which only starts to ovoid subpopulation within its ventral-intermediate part appear in the underlying sections h–j and l–n, always restricted to the (VMHvi; Fig. 15c, d). Subsequently, at E18.5, Six3 contin- terminal hypothalamus in what respects the VMH primordium. Note in h–j and l–n that at E13.5 the two cell types intermix somewhat ues labeling strongly the well delimited, ovoid mass seen at dorsal levels through VMH h, l, whereas they stay separate more previously (VMHvi); this is found intercalated between the ventrally j–n). o–q Combined blue Nr5a1 ISH and brown Nkx2.2 molecularly different VMHvm and VMHvl subdivisions immunohistochemical sagittal images at E16.5 show that intermixing (VMHvi; 15e–j). A weaker and more dispersed expression increases significantly by this stage. A blue dash contour was traced around the Nr5a1-positive VMH population, and a brown dash con- of Six3 is detected as well at the VMHim and VMHdrm tour surrounds the Nkx2.2-positive VMH population, allowing this subdivisions, whose labeled cells are less well delimited comparison. Scale bars represent 200 µm. TuD tuberal dorsal area, from the arcuate population, in contrast with the VMHvi migPaV migration of the ventral paraventricular cells, primVMH pri- (Fig. 15h–j). mordium of the ventromedial hypothalamic nucleus, AT acroterminal area, RTuI, retrotuberal intermediate area, TuI tuberal intermediate area, RTuD retrotuberal dorsal area, VMHdc dorsocaudal VMH sub- nucleus, VMHim medial-intermediate VMH subnucleus, VMHdrm Discussion dorsal-rostromedial VMH subnucleus, VMHil lateral-intermediate VMH subnucleus, VMHdrl dorsal-rostrolateral VMH subnucleus The VMH nucleus is one of the larger hypothalamic struc- tures (called ‘principal hypothalamic nucleus’ by Cajal). A comparison of Nkx2.1 and Nkx2.2 distribution at E14.5 Classically it has been widely defined as presenting dor - corroborates the predominancy of Nkx2.1 at ventral loci of somedial, central/core, and ventrolateral cytoarchitectonic the VMH, whereas Nkx2.2 is placed mainly at intermediate subdivisions (VMHdm, VMHc, VMHvl). Note the corre- levels, with some extension into the dorsocaudal subnucleus sponding topographic descriptor terms refer to the columnar (Fig. 14e–g). In our horizontal sections (similar to conven- forebrain axis postulated as running into the telencephalon tional columnar coronal sections but interpreted along the (the prosomeric dorsal direction). The VMH massively axial dimension), we observe Nkx2.1 and Nkx2.2 cells in the contains glutamatergic neurons (Puelles et al. 2012; their ventricular stratum at TuD levels (Fig. 14h–i’’’). The pres- Figs. 17A–C; 20A–C; 22, 23A,B). However, recent scR- ence of Nkx2.1-positive cells increased in the ventricular NAseq transcriptomic studies found an unexpected diversity layer at TuI levels, whereas Nkx2.2 cells are totally absent of neuronal molecular profiles in the VMH (Kim et al. 2019 in this stratum (Fig. 14j, j’). At these sections the VMH pri- found 12 neuronal clusters in the core portion of the VMH mordium in its entirety is revealed complementarily labeled and 17 neuronal clusters in the ventrolateral VMH part, by the Nkx2.2 and Nkx2.1 markers (Fig. 14j, j’, j’’). whereas van Veen et al. 2020 described six VMH neuronal 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 10 Comparisons of Satb2 and Tcf7l2 patterns reveal a rostrocau- (which is implicitly interpreted as a transversal section and dal subdivision in the dorsal sector of VMH. Satb2 and Tcf7l2 ISH does not show the more peripheral parts of the tuberal area) images from the Allen Developing Mouse Brain Atlas are shown and sagittal or true transversal sections of the VMH (see in either prosomeric horizontal sections at three dorsoventral lev- our Fig. 3a–g) are seldom examined. This leads to practi- els at E13.5 (a–c, e–g), E15.5 (i–k, m–o), and P1 (q–s, u–w) or in corresponding sagittal sections (d, h, l, p, t, x). The Satb2 marker cal invisibility of our highly relevant TuD and acroterminal is expressed initially only within TuI, where a dense periventricu- domains, the main sources of tangentially migrated VMH lar stratum of positive cells is visible at E13.5 (a, b; note additional cells. VMH-unrelated expression appears also at the acroterminal TuD area We have addressed this issue consistently with previous in a, d). At E15.5 the dorsalmost VMH Satb2 cells concentrate at the VMHdc subdivision (i, l), while other cells of this type have invaded prosomeric analysis, correlating diverse molecularly charac- massively the ventral parts of the VMH primordium (VMHvm; terized VMH neuronal populations to the set of molecularly VMHvl; j, k). At P1 the Satb2 subpopulation persists majoritarily defined progenitor domains previously reported in the area at the VMHdc and the ventral complex (q–t). In contrast, the Tcf712 of interest (Puelles et al. 2012; review in Diaz and Puelles marker initially appears expressed selectively at the acroterminal TuD area (e–h); later, Tcf7l2 cells migrate ventralwards into the rostral part 2020; see also Morales-Delgado et al. 2011, 2014). We of the VMH primordium (m–p), invading mainly the VMHdrm sub- were able to map early emergence of some distinct VMH division, while other acroterminal TuD cells seem to invade the TuL cell types in specific surrounding domains of the embry - nucleus (u–x) (e–h for E13.5; m–p for E15.5; q–x for P1). Scale bars onic hypothalamus, and then traced them via intermediate represent 200 µm. AT acroterminal domain, TuD tuberal dorsal area, TuI tuberal intermediate area, VMHdc dorsocaudal VMH subnucleus, stages to their ulterior topography within the VMH complex. VMHvl ventral-lateral VMH subnucleus, VMHvm ventral-medial These tracings suggested or were consistent with tangential VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, or radial migration patterns according to the characteristic VMH ventromedial hypothalamic nucleus, PM perimamillary area, spatiotemporal transitions deployed in each case. Neurons TuL tuberal lateral nucleus that form the VMH proper (leaving aside the even more het- erogeneous surrounding shell of the nucleus) were shown to types). Affinati et al. (2021) identified 24 VMH clusters originate mainly either in the TuD or TuI progenitor domains roughly categorized into 6 main groups. Comparison of within basal THy, sometimes involving also or exclusively diverse gene markers mapped in diverse section planes in the corresponding TuD or TuI acroterminal subregions (see the developing VMH already led Puelles et al. (2012) to the the novel prosomeric concept of the acroterminal domain conclusion that VMH was heterogenous and might not be as the rostromedian end of the THy in Puelles et al. 2012; originated as a whole within the intermediate tuberal area Puelles and Rubenstein 2015; Ferran et al. 2015; Puelles (TuI), the subdivision of the prosomeric basal hypothalamus 2018; Diaz and Puelles, 2020). A further contingent of VMH in whose terminal part (THy) the VMH lies in the adult. cells apparently originates from the overlying alar terminal Evidence was presented suggesting the contribution to VMH subparaventricular area (TSPa), otherwise previously identi- at least of Nkx2.2- and Pdyn-expressing cells, which appar- fied as a source of various dorsoventral peptidergic neuronal ently originated from the overlying TuD progenitor domain migrations from alar progenitor domains into the subjacent (Puelles et al. 2012; their Figs. 26A–N). Unfortunately, this tuberal/retrotuberal basal plate (Diaz et al. 2015). Since the well-documented notion has been generally disregarded in adult VMH apparently lies strictly within TuI, the TuD, the subsequent literature. Accrued reports on loss of func- acroterminal, and TSPa origins necessarily imply dorsoven- tion of selected VMH gene markers (Kurrasch et al. 2007; tral or rostrocaudal tangential migrations of the correspond- Cheung et al. 2013; Lu et al. 2013; Corman et al. 2018; ing derivatives finally found inside the VMH. The existence Aslanpour et al. 2020a, b) did not contemplate the proso- of dorsoventral tangential cell translocations coming from meric AP and DV subdivisions of the tuberal hypothalamic the TSPa or TuD areas was verified experimentally. region and the phenotypes observed were interpreted under We also described the approximate final distribution of the traditional (simpler) columnar assumptions (e.g., see the specific markers we traced developmentally regarding Altman and Bayer 1978, 1986). These include the undocu- different subdivisions or subnuclei identified in the E18.5 mented assumption that the whole VMH cell population is VMH. Due to the prosomeric approach, implying use of produced locally in the tuberal area (i.e., without any tangen- diverse axial references and positional landmarks (e.g., tial migration; further comments on this below). Columnar the longitudinal floor plate and the alar-basal boundary; reports on the VMH frequently concentrate descriptions on or the transversal acroterminal domain and the intrahy- a sole coronal section level midways through the nucleus pothalamic boundary) and a related map of molecularly 1 3 Brain Structure and Function Fig. 11 Differentially labeled populations of the VMH at late devel- coincidences at lateral levels (VMHdrl, VMHil, VMHvl). Scale bars opmental stages. Satb2 a–c, Tcf712 d–f, and Sox14 g–i ISH medi- represent 200 µm. VMHvm ventral-medial VMH subnucleus, VMHim olateral sagittal sections of E18.5 or P4 mice. These tree markers medial-intermediate VMH subnucleus, VMHdc dorsocaudal VMH are differentially expressed in diverse VMH subnuclei: Satb2 labels subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHvi distinctly the VMHdc subunit and the ventral area (VMHvm, VMHvi ventral-intermediate VMH subnucleus, VPM ventral premamillary and VMHvl), Tcf7l2 is selective for the VMHdrm-VMHdrl subdo- nucleus, VMHvl ventral-lateral VMH subnucleus, VMHil lateral- mains, and Sox14 seems to coincide with Satb2 at deep and interme- intermediate VMH subnucleus, VMHdrl dorsal-rostrolateral VMH diate sagittal sections (signal at VMHvm; g, h), but differs drastically subnucleus, DMH-P peduncular part of the dorsomedial hypotha- at lateral levels i; comparison with Tcf7l2 shows a complementary lamic nucleus, DMH-T terminal part of the dorsomedial hypotha- pattern at medial and intermediate levels (VMHdrm; g, h) and some lamic nucleus, VMHvm ventral-medial VMH subnucleus defined progenitor domains (Puelles et al. 2012; Puelles was more detailed than was usual heretofore, inspiring and and Rubenstein 2015; Ferran et al. 2015; Diaz et al. 2015, justifying the proposed terminological changes (Fig. 16; 2020; López-González et al. 2021; Fig. 1a,b) in the context Table 3). VMH molecular regionalization along the non- of 22 marker genes, our analysis of inner VMH structure arbitrary dorsoventral, rostrocaudal, and mediolateral 1 3 Brain Structure and Function (radial) dimensions was reexamined, leading to an expanded but progressive steps in that direction are absolutely neces- subdivision map of the VMH complex. This contains sary to the progress of neuroscience. minimally 8 parts that are consistent with the existence of The first origin of Nr5a1- and Nkx2.2-positive neurons combinations of several neuronal populations that show that eventually populate the VMHdc, VMHdrm, VMHdrl, differential molecular features, showing in their postmigra- VMHim. and VMHil subdivisions was traced to the early tory configuration various degrees of partial overlap. The TuD area (differentially at its main ventral subdivision and conventional simpler tripartite schema was approximately its overlying subliminal Nkx2.2-positive part); these two conserved by terminological distinction of dorsal, inter- TuD populations migrate respectively in deep versus super- mediate, and ventral VMH regions, subdivided as follows: ficial mantle cell streams, and later show a medial to lateral our dorsal VMH region includes not only the conventional (radial) stratification (Nkx2.2 cells born at the subliminal ‘dorsomedial’ part, whose topologic position is more pre- TuD migrate superficially and finally lie lateral/superficial cisely described as ‘dorsocaudal’ (VMHdc), but also a pre- to the population of Nr5a1 cells born at the underlying main viously undescribed ‘dorsorostral’ part, which divides into TuD and migrating in a deeper radial level). Both parallel medial and lateral dorsorostral subcomponents (VMHdrm, cell populations massively migrate ventralwards (tangen- VMHdrl); our intermediate VMH region roughly corre- tially) as observed at E12.5-E14.5, thus forming transient sponds to the conventional ‘central or core’ part, but divides separate deep and superficial streams that spread out within distinctly into medial and lateral-intermediate subunits the upper dorsal and intermediate parts of the VMH pri- (VMHim, VMHil); finally, our prosomeric ventral VMH mordium once they reach the TuI. Additional molecularly region includes the conventional ‘ventrolateral’ part and distinct VMH cell populations were traced instead to local displays distinct medial, intermediate, and lateral subunits origins within the TuI ventricular zone, which underlies (VMHvm; VMHvi; VMHvl). Our VMHdc indeed happens radially the VMH primordium. These comprise at least cell to be ‘relatively dorsal’ in the prosomeric VMH (i.e., close types expressing Nr2f1, Lmo4, Satb2 (Tables 1, 2). In some to the prosomeric alar-basal boundary of the tuberal area, cases, a given molecular subtype may be produced at both like its VMHdrm/VMHdrl companions), and thus merits the the TuD and TuI domains (e.g., Nkx2.1, Sox14). ‘dorsal’ descriptor. The latter was previously widely used in The novel dorsal-rostral VMH subdivisions (VMHdrm, reference to the seemingly obsolete columnar axis (check its VMHdrl; first described in the present report, though previ- inconsistence with the primarily longitudinal Nkx2.2-posi- ously implied by some reported observations) apparently tive band, commented in Puelles and Rubenstein 2015, and derive mainly from the acroterminal part of the TuD pro- note the wrong columnar assumption that coronal sections genitor region (represented, e.g., by Tcf7l2, Sox14, and some through the hypothalamus show the dorsoventral dimension, rostral Nr5a1 cells); the migration into VMH of these com- though they demonstrably show the anteroposterior one by ponents thus implies an oblique dorso-rostrocaudal direc- passing caudalwards into the diencephalon and midbrain; tional vector. Fig. 1c). The ‘medial’ descriptor in this term was changed to The Six3-expressing cells of the VMHvi also arise within ‘caudal’ (VMHdm = VMHdc) because this term is not suf- the acroterminal basal domain; it is open to discussion ficiently selective; all dorsal, intermediate, and ventral parts whether they come from the TuI or TuD acroterminal sub- of the VMH have ‘medial’ portions (i.e., parts closer to the areas, or from both. We suspect that the TuI acroterminal periventricular stratum); this includes also the dorsal-rostral subarea is principally related to the production of arcuate part of VMH that has a distinct medial half (VMHdrm). We nucleus neuronal types (in caudal continuity with the DM reproduced the resulting structural schema in axially true complex), where mainly GABAergic neurons are produced; transversal, sagittal, and horizontal section planes in Fig. 16, this would suggest that the Six3 cells of the VMH (also first and we tried to clarify the new terminology for the readers described in the present report) may arise instead within the in Table 3. The New Neuromorphology (Nieuwenhuys and molecularly distinct acroterminal TuD (site of the proso- Puelles 2016) made possible by modern molecular and trans- meric ‘anterobasal nucleus’, or classical ‘retrochiasmatic genic experimental data claims the need to clarify various area’). newly emerging concepts by carefully rationalized terminol- Finally, Vax1-expressing VMH neurons originate clearly ogy changes. It is expected that full incorporation into usage within the overlying subparaventricular alar progeni- of the new terms and concepts requires generational change, tor area (TSPa), where Dlx family genes are also strongly expressed and GABAergic cells are thus expected to arise. 1 3 Brain Structure and Function This observation of some potentially GABAergic neurons Both TuI and TuD are progenitor areas for VMH cells expressing Vax1 and maybe Dlx genes within VMH (mainly at the VMHdrm and VMHdrl subnuclei) is unexpected Topographically, the development of VMH seems linked and needs to be checked in appropriate material (see, for to the basal region of the terminal hypothalamus (THy), instance, Puelles et al. 2012; their Figs. 8–18 and 8–15C, namely to its tuberal territory (Tu). Nothing similar to D, which seem to support this conclusion). the VMH develops within the cognate RTu area of the 1 3 Brain Structure and Function ◂Fig. 12 Nr2f1 expression illustrates a significant local TuI contribu- underlined by Puelles et al. 2012 but was not noticed when tion to the VMH primordium. a–d Nr2f1 ISH sagittal images at E13.5 these columnar names were introduced, probably because and E15.5 from the Allen Developing Mouse Brain Atlas. Solid lines it was assumed that the terminal DMH component was as in a, b delimit the longitudinal (subliminal) TuD/RTuD, (main) TuD/ a whole caudal to the VMH, which is maybe slightly less RTuD and TuI/RTuI progenitor basal domains. Red asterisks in b, d show a strong Nr2f1 signal at the ventral part of VMH primor- incongruent in columnar terms, but is in any case equally dium and its more mature derivative. Blue asterisks in b, d identify false in prosomeric terms; see also Puelles 2019). In contrast dorsal areas of relatively weaker Nr2f1 signal at the VMH primor- to the VMH that is restricted to TuI, the DMH complex of dium. (e–h’) Double Nkx2.2 (red) and Nr2f1 (green) immunofluo- RTuI is bineuromeric (having terminal/hp2 and peduncular/ rescence in sagittal (e–f) and horizontal (g–h’) sections of E12.5 and E14.5 embryos. The dash lines in f indicate the horizontal section hp1 moieties). This DMH longitudinal continuity passing planes at g and h. At TuD/RTuD levels (just dorsal to VMH) the ven- finally under the VMH (in the prosomeric schema) already tricular zone and adjacent mantle zone are Nkx2.2-positive (vz; mz; suggests that something extraordinary linked only to THy g–g’), whereas more ventrally through the VMH primordium (within and hp2 affects the development of VMH. Given that the TuI) the ventricular zone is only Nr2f1-positive and the mantle zone has a mixture of green cells (particularly at deep levels), with red and DMH formation is largely GABAergic and essentially yellow cells (vz; mz; h–h’). Scale bars in a, e, g represent 200 µm. locally produced (with minor immigrated glutamatergic core Scale bar in g’ represents 100 µm. Scale bar in h’ represents 50 µm. nuclei; Puelles et al. 2012), it might be easily deduced that primVMH primordium of the ventromedial hypothalamic nucleus, perhaps the massively glutamatergic VMH subarea of TuI RTuI retrotuberal intermediate area, TuI tuberal intermediate area, RTuD retrotuberal dorsal area, TuD tuberal dorsal area, PPaV pedun- is not really a part of TuI, but a variant subdomain of termi- cular paraventricular ventral nucleus, DMH dorsomedial hypotha- nal Tu intercalated between TuD and TuI which may have lamic nucleus, VMHim medial- intermediate VMH subnucleus, vz unique properties. Based on the identification within TuD ventricular zone, mz mantle zone, VMHil lateral-intermediate VMH of the earliest expression of the Nkx2.2 and Nr5a1 markers subnucleus, OCh optic chiasma which later characterize some positionally distinct types of VMH cells, Puelles et al. (2012) estimated that at least some peduncular hypothalamus (PHy) (in prosomeric theory, VMH neurons possibly originated via selective dorsoven- THy and PHy belong to different prosomeres -hp2 and hp1, tral tangential migration from the terminal TuD domain. At respectively- and Tu/RTu have therein analogous topologi- that time, it was unclear whether any VMH neurons arise cally basal positions under the alar-basal boundary; they primarily at TuI level. Puelles et al. (2012) rather inclined are thus presumed to result both from similar ventralizing to the assumption that all VMH subpopulations might be versus dorsalizing antagonism occurring in two molecularly migrated from TuD. distinct neuromeres, thus the observed differences; the exist- We reexamined this issue with more abundant molecular ence of important typological neuronal differences between data and some experimental testing of the previously pre- these two hypothalamic areas is both unsuspected and inex- dicted dorsoventral migration, corroborating its existence, plicable within the obsolete now more than 100 years-old but reaching the conclusion that the mouse VMH appar- columnar frame of thought). The unique developing VMH ently contains both local and exogenous neuronal popula- structure is accordingly one of the many differential features tions. Some neurons with differential molecular profiles can that can be explained by the postulate of the dual hp1 and be traced to origins within either TuI or TuD, separately, hp2 hypothalamo-telencephalic prosomeres (Puelles et al. or jointly. Moreover, additional VMH subpopulations were 2012; Ferran et al. 2015) but has no explanation within the found that seem to originate specifically in the basal AT area alternative columnar system. The ample tuberal domain lying dorsal to the Arc area, and some neurons migrate into of THy can be divided dorsoventrally (according to the VMH coming from the overlying alar subparaventricular prosomeric axis) on the basis of molecular and structural area. This supports the idea that patterning of the tuberal features into dorsal (TuD), intermediate (TuI) and ventral area occurs differentially at PHy compared to THy, intro- (TuV) progenitor domains. All three of them extend ros- ducing the unique and complex VMH areal phenomenon tralwards into the corresponding parts of the rostromedian found dorsal to the terminal end of the DM ‘column’. acroterminal area (AT; Puelles et al. 2012). TuI is the topo- The contextual background of any neural developmen- graphic site of both the terminal part of DMH and the VMH, tal analysis includes data on local neurogenesis. Report- with the terminologically unacceptable oddity that the ter- edly, VMH neurogenesis occurs in mice between E9.5 and minal DMH component clearly lies ventral to VMH (this E14.5, with a peak at E11.5-E12.5 (Shimada and Nakamura topologic incongruency of the columnar names was already 1973; Aslanpour et al. 2020b; similar rat data in Ifft 1972 1 3 Brain Structure and Function Fig. 13 Nkx2.2, Nr5a1 and Nr2f1 signals represent partially overlapping subpopulations of the VMH primordium. a, b Double immunofluorescence for Nkx2.2 (red signal; pink dash line) and Nr2f1 (green signal and dash line) in VMH sagittal sections at two section levels at E16.5, showing partial overlap. c, d Combined Nr5a1 ISH (dark bluish signal; blue dash line) and Nr2f1 immunohisto- chemical reaction (brown signal and dash line) at two different section levels. There is also only a partial overlap of these two populations. Scale bars repre- sent 200 µm. VMHvm ventral- medial VMH subnucleus, VMHim medial- intermediate VMH subnucleus, VMHdrm dorsal-rostromedial VMH sub- nucleus, VMHdc dorsocaudal VMH subnucleus or Altman and Bayer 1986). Interpreting their data in terms The TuD area is divided into dorsal subliminal and ven- of the columnar model, Aslanpour et al. (2020b) identified tral main portions; the migrating Nkx2.2 cells are thought a “rostrocaudal” gradient in VMH birthdates, which we to arise at the subliminal part, so that they must cross the translate topologically to the prosomeric model as a dors- underlying main part to reach the VMH. The main TuD oventral gradient. Presumably TuD neurogenesis precedes domain is where Nkx2.1-positive Nr5a1-expressing cells that of TuI, which is consistent with reported wholemount are produced. The subliminal TuD portion reportedly does rat results using AChE as an early differentiation marker not express either Nkx2.1 or Nr5a1. The molecular pair (Puelles et al. 2015; Amat et al. 2022). Accordingly, we Slit/Robo involved in regulation of neuronal migration and deduce that our dorsal tuberal Nr5a1 and Nkx2.2 VMH cells axonal guidance is expressed at the VMH primordia. The would be born earlier (E10.5) than the ventral ones (E11.5), transcription factors Nkx2.1/Isl1/Otp regulate Robo2 expres- due to their origin in the more precocious TuD area. sion. Moreover Slit1–/– and Slit2–/– mice show increased cell density at the VMH subventricular zone (Romanov Dorsally originated Nkx2.2 cells invade et al. 2020). It would be interesting to analyze whether the the intermediate VMH via tangential migration Slit/Robo signaling pathway affects the migrating Nkx2.2 population of VMH. Descriptive analysis of gene patterns led Puelles et al. (2012) The transcription factor Nr5a1 (SF1) plays a role in to suggest the migration en masse into VMH of Nkx2.2 ele- VMH development regarding its neuronal identity (Tran ments born at the subliminal dorsal tuberal area that under- et al. 2003), cell distribution (Ikeda et al. 1995; Shinoda lies the alar/basal limit. We confirmed this migratory process et al.1995; Davis et al. 2004) and connectivity (Tran et al. with our own descriptive and experimental data. We collat- 2003). This factor was widely used in functional studies of erally noted that less numerous ventrally migrating Nkx2.2 the VMH nucleus (Büdefeld et al. 2012; Hashikawa et al. cells from the subliminal dorsal retrotuberal area seem to 2017; Kennedy et al. 2020; Lewis et al. 2022). In contrast, invade the area of the peduncular dorsomedial hypothalamic there is little information about the role of the Nkx2.2 tran- nucleus, though they are not clearly detectable at advanced scription factor in VMH development (Kurrasch et al. 2007; embryonic stages (Puelles et al. 2012; Kim et al. 2021). We Puelles et al. 2012). In this study we have compared the also observed migration of CMFDA-labeled Nkx2.2 cells developmental distribution of both, Nr5a1 and Nkx2.2, not- into the VMH primordium when the TuD area under the ing that, irrespective of partial overlaps, they largely occupy alar-basal border was marked. different sectors of the VMH. The literature often wrongly assumes that Nr5a1 is a general transient marker of the 1 3 Brain Structure and Function VMH, but that is certainly not the case, according to our of the VMH. This is the case of Nr2f1 cells identified even- data. Nr5a1 cells occupy a larger part of VMH than Nkx2.2 tually at the ventral VMH tier (VMHvm, VMHvl), though ones, but both are restricted to given VMH subnuclei (Nr5a1 some of them overlap with tangentially migrated Nr5a1 and mainly at VMHdc, VMHdrm and VMHim; Nkx2.2 is found Nkx2.2 elements in VMHim/il, respectively. Nr2f1 cells also in a mediolateral gradient along the intermediate VMH appear separately at the VMHdc subnucleus (=conventional tier, being stronger at its lateral part, in contrast with the VMHdm subdivision). Similarly, the Satb2 marker identifies medially predominant Nr5a1 subpopulation). This pattern another radially migrated local cell population found within is thus explained by our interpretation that both populations the ventral VMH tier (VMHvm, VMHvl), which is differen- are exogenous relative to the VMH area and each invades tially absent from the intermediate tier. These observations it differentially according to different origins, routes, and suggest that also the radially migrated components of VMH final distributions (presumably affected by their differential may have subtly diverse topographic origins and postmigra- reaction to the local adhesivity or attractor scenario found tion distributions. within the VMH). The Nr5a1 pattern across AT/main TuD/TuI-VMHdc/ Varied neuronal identities within the VMH VMHdrm/VMHim is reproduced by other markers (e.g., Cnr1, Slc17a6, Robo2), whereas the characteristic Nkx2.2 Lu et al. (2013) showed evidence suggesting that the early pattern across subliminal-TuD-VMHil/VMHdrl is rather transcription factor Rax has a role in conferring a Nr5a1- unique. expressing and glutamatergic cell fate to cells in the central VMH. Rax is an early gene expressed at the TuI and TuD TuI origin of some VMH neurons (our interpretation of Allen Developing Mouse Brain Atlas material at E11.5 and E13.5), overlapping the Shh-positive Radial migration has been assumed conventionally as the basal plate territory, while a large part of the rostral acroter- only mechanism laying down the VMH primordium in minal domain (particularly the TuI one) loses secondarily the the tuberal mantle. Such cells would move from the local Shh signal. We noted that the prospective VMH area within ventricular zone into the neighboring mantle, presumably TuI retains up to E13.5 a central patch of low level Shh sig- guided by outside-in stratification rather that radial migra- nal. Lu et al. (2013) observed that when Rax was floxed tion proper (McClellan et al. 2006). Misplaced cells con- under control of the Shh promotor, a central region of Nr5a1 taining estrogen receptors identified in Nr5a1 null mice cells within VMH lost this signal (among others related to or in GABA R1 knockout mice have been interpreted to the glutamatergic phenotype- see their Figs. 7M–P; 7A’, represent failed radial migration (Dellovade et al. 2000; F’, B’) and came to be occupied by GABAergic neurons McClellan et al. 2008). Moreover, agonists and antagonists (their Fig. 7G’). Remarkably, surrounding dorsocaudal and of GABA and GABA receptors acting in vitro on coronal ventral Nr5a1 cells of the VMH remained unaffected. The A B hypothalamus slices (these are actually horizontal relative authors illustrated the tuberal expression of Shh at E10.5 to the hypothalamic basal plate) were reported to change the in a true transversal section (note the eye stalks and tel- speed and orientation of radially migrating neurons (Dello- encephalic vesicles lying above the hypothalamus) but did vade et al. 2001; McClellan et al. 2008). In all these cases not identify this area as the TuD subregion (a longitudinal the authors did not consider the possibility that processes band extending under the alar-basal boundary); this is the involving tangential migration might be affected instead or dorsal basal domain that selectively retains strong primary simultaneously. Shh signal normally (Andreu-Cervera et al. 2019; Puelles The conventional ‘ventrolateral’ VMH subregion, which et al. 2012). If this curious phenotype change occurring in apparently is majoritarily colonized by radially migrated the central or intermediate part of VMH depends on flox- neurons, is well known for the selective mapping of estro- ing that selectively occurred at the longitudinal Shh-positive gen receptors (e.g., Cheung et al. 2013; Krause and Ingra- TuD domain, the whole terminal TuD area producing tan- ham, 2017; Ma et al. 2021). Some authors described also gentially migrating Nkx2.2 and Nr5a1 VMH neurons would Nkx2.1 expression as highly restricted to ‘ventrolateral’ be expected to have been affected. It is thus difficult to see VMH (Tran et al. 2003; Davis et al. 2004; Cheung et al. why some parts of VMH developed normally. It is actually 2013). Our data indicate that early Nkx2.1 expression dis- somewhat unclear in this report whether Rax is expressed tributes widely throughout the tuberal region. At E18.5 we strongly enough at the level of the TuD band, to respond to still observe expression of this marker at the tuberal perive- the local Shh floxing effect (see in this regard the results of ntricular stratum deep to the VMH, as well as in neurons Orquera et al. 2016, suggesting that there is little Rax expres- of some VMH subdivisions, namely the VMHdc, VMHvm, sion within TuD). and VMHvl subnuclei. Other radially migrated local popula- A more detailed study of the Allen coronal series for tions studied by us are mainly found at the ventral portion E11.5 and E13.5 indicated that ventral to the strongly 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 14 Widespread basal expression of Nkx2.1 hypothalamus Aslanpour et al (2020a) suggested that Ascl1, a bHLH marker but selectivity at the VMH primordium. a–c Medio-lateral transcription factor expressed early on in the ventricular E13.5 sagittal sections from the Allen Developing Mouse Brain stratum of the TuD and TuI domains, likewise switches on Atlas with Nkx2.1 ISH show widespread basal signal of this hypo- differentiation programs that give rise preferentially to some thalamus marker (check also Fig. 3h in the adult). e–g Comparison of Nkx2.2 (red) and Nkx2.1 (green) immunofluorescence in medi- VMH neurons (e.g., our intermediate or dorsocaudal VMH olateral sagittal sections at E14.5. The dashed contour of the VMH populations). primordium is indicated in (e). The labeled cell populations are Our results indicate that VMH neurons originate from mostly topographically distinct, though a few double-labeled cells are diverse progenitor areas, and occupy characteristic, partly observed dorsocaudally (yellow in f, g). (h–j) Dorsoventral horizon- tal sections of E14.5 embryos reacted for Nkx2.2 (red) and Nkx2.1 overlapping parts of the VMH. Hypothetic areas of over- (green) immunofluorescence (section levels marked in f). At level h lap between the diverse progenitor domains might explain there is a poor Nkx2.1 signal (h’); this starts to appear stronger at the occasional visualization of double-labeled VMH cells ventricular and periventricular zone i levels (i’, i’’; i’’ and i’’’), and that express different combinations of transcription fac- expands into the VMH mantle at j levels (j’, j”). k–m Medio-lateral sagittal sections at E15.5 from the Allen Developing Mouse Brain tors (e.g., Nkx2.2-Nr5a1, Nkx2.2-Nkx2.1, Nkx2.2-Nr2f1), Atlas reacted for Nkx2.1 ISH, showing preferent reaction at VMHim, besides cells that express Nkx2.2 without Nr5a1, Nkx2.1 or VMHvm, and VMHvl subdivisions. n–p Comparison of double Nr2f1. Similarly, Corman et al. (2020) identified Shh cells Nkx2.2 (red) and Nkx2.1 (green) immunofluorescence at mediolat- and Shh-responsive (Gli1) lineage cells as contributing to eral sagittal section levels from E18.5 embryos. Note double-labeled yellow cells at VMHim (o). Scale bars in a, e. k, n represent 200 µm. the VMH population; they studied several VMH markers Scale bar in i’ represents 100 µm. Scale bar in i’’’ represents 50 µm. and quantified that, approximately, a half of Nr5a1 cells, a AT acroterminal area, TuI tuberal intermediate area, TuD tuberal dor- half of Nkx2.1 cells and 20% of Nkx2.2 cells belong to a Shh sal area, VMHvm ventral-medial VMH subnucleus, VMHim medial- lineage occupying a rostral VMH subregion, whereas Gli1 intermediate VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHvl ventral-lateral VMH subnucleus, primVMH pri- lineage cells represent a proportion of caudal VMH cells, mordium of the ventromedial hypothalamic nucleus, VMHdc dorso- with relatively less Nr5a1 and Nkx2.1 elements, and more caudal VMH subnucleus, VMHil lateral-intermediate VMH subnu- Nkx2.2 neurons. These results agree with ours as regards cleus, VMHdrl, dorsal-rostrolateral VMH subnucleus the existence of molecularly diverse domains of origin that produce qualitatively different VMH subpopulations. Shh-positive TuD band and wholly inside the Rax-positive Studies examining loss of function of single transcription domain there also appears a patch of weaker Shh expres- factors, or a morphogen such as Shh, have in common that, sion which can be ascribed to the upper TuI area (VMH while a discrete part of the VMH population is affected, in primordium), whose radial derivatives might contribute none of them the whole nucleus is disrupted. Consistently selectively to the central VMH portion that results devoid with our present results, this suggests that there are several of Nr5a1 cells and abnormally populated by GABAergic neighboring but different sites of origin of VMH cells, appar - neurons (Lu et al. 2013). We therefore consider it likely that ently with subtle variations in their transcriptomic set. Differ - the observed restriction to central VMH of the Rax lack of ent subpopulations arising outside the radial position of the function phenotype may be explained as due to an effect on VMH (upper TuI) converge via parallel tangential migrations the local patch of weak TuI Shh signal. Lu et al. (2013) did into the nuclear primordium. Once there, they partly disperse not show images of the TuD area above the VMH at later and intermix with other VMH cell types, without achieving stages, which might have indicated whether the Nr5a1 cells a fully homogeneous distribution. This eventually results in that normally migrate dorsoventrally into VMH may have the characteristic cellular typological profile of parts of the failed to do so, displaying instead abnormal positions or sig- nucleus (Fig. 16; VMHdc, VMHdrm, VMHdrl, VMHim, nals of cell death. Moreover, when Rax was deleted under VMHil, VMHvm, VMHvi and VMHvl), with different neu- the action of the Six3 promoter (a gene primarily overlap- rons, transcriptomic profiles, connections, and functions. We ping Nkx2.1 expression only at the acroterminal Arc area), think that birthdating results, as well as most analyses of lack Nkx2.1 was also dramatically diminished in the VMH pri- of function phenotypes, are so far non illuminating, because mordium (Lu et al., 2013). These results imply that some they were planned and interpreted assuming a single origin VMH cells are produced radially at the TuI area (under Shh of the heterogeneous VMH neurons. control) and others derive tangentially from the acrotermi- Indeed, recent single-cell transcriptomic studies have nal TuI area (under Six3 control; it should be investigated increased the ability to identify different VMH cell types. whether this bears particularly on our novel Six3-positive The available data on the hypothalamus already has made VMHvi subnucleus). Both processes may require indepen- us realize that the number of distinct cell types was being dently additional Rax function. It would be interesting to conventionally vastly underestimated in the hypothalamus. examine similarly whether Rax affects the Nkx2.2-positive In one of these studies, six principal VMH clusters were population of the VMH. identified in postnatal mice (P10), each of them represented by a particular transcription factor: Tac1, Rprm, Pdyn, Sst, 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 15 The early acroterminal marker Six3 (see a) later reveals a dis- corresponds to our novel Six3-positive VMHvi subdivision. tinct ventral VMH subnucleus (d–f, i, j). Six3 ISH images from Allen In this way, Hashikawa et al. (2017) identified a functional Developing Mouse Brain Atlas at E13.5 (a, b) and E16.5 (c, d) illus- division of the ‘ventrolateral’ part regarding with differen- trate that apart of Six3 cells developing within the acroterminal Arc tial mating and fighting behaviors. These data support the and SCh nuclei, a small compact ovoid mass appears labeled within the neighboring VMHvi subdivision as development advances. Paral- split of the classical ‘ventrolateral’ subdivision into the three lel red lines in d represent the somewhat oblique section planes of subnuclei; VMHvm, VMHvi, and VMHvl that we propose sections e–g, which interest caudally the negative thalamus (Th) and in our present VMH model. Six3-positive prethalamus (PTh), apart of the alar Otp-immunoposi- Affinati et al. (2021) similarly identified 24 molecularly tive paraventricular nucleus (Pa) and hypothalamo-amygdalar corri- dor (HyA). e–g Oblique serial sections showing combined Six3 ISH differentiable cell clusters in the VMH, which they were able (blue) and Otp IHC (brown) expression in an E18.5 embryo, identify- to categorize into 6 groups using majoritarily shared gene ing the novel VMHvi subnucleus and additional more disperse simi- markers. According to these authors, Dlk1 and Esr1 express- lar cells in VMHim and VMHdrm. The black curved line represents ing neurons are present at ‘ventrolateral’ classic division, the hypothalamo-diencephalic boundary, also shown in black in (c, d; note here the counterstain is TH, marking the A13 cell group). Lepr expressing cells are located into the ‘core’, and Foxp1- h–j Six3 ISH caudorostral serial sections from an E18.5 embryo cells map at their “anterior and lateral” to VMH core popu- transversal to the basal hypothalamus (the optic chiasma –OCh– and lation, possibly corresponding to our VMHdrl subnucleus. suprachiasmatic nuclei –SCh– appear dorsally; TSbO, tuberal subop- It seems probable, nevertheless, that even our more tic nucleus). The contour of the VMH complex is delineated and the intermediate ventral position of the novel Six3-positive VMHvi sub- detailed VMH subdivision model is surpassed by these division is evident. Scale bars represent 200 µm. RM retromamillary single-cell transcriptomic data, in the sense that all detected area, M mamillary area, AT acroterminal area, NHy neurohypophysis, clusters cannot be ascribed uniquely to our specific VMH TuI tuberal intermediate area, main or subl. TuD main or subliminal subdivisions. This means that our present minimally subdi- tuberal dorsal area, PTh prethalamus, PHy peduncular hypothalamus, THy terminal hypothalamus, OCh optic chiasma, VMHvm ventral- vided model probably also is inadequate in the long run to medial VMH subnucleus, VMHim medial-intermediate VMH subnu- fully explain the real VMH populational complexity. Our cleus, VMHdrm dorsal-rostromedial VMH subnucleus, Arc arcuate present attempt to introduce the idea of alternative progeni- nucleus, VMHvi ventral-intermediate VMH subnucleus, A13 A13 tor sources (and converging tangential migrations) for sev- dopaminergic cell population, PTh prethalamus, SCh suprachiasmatic nucleus, Th thalamus, PPa peduncular paraventricular area, VMHdc eral components of the whole VMH population thus cannot dorsocaudal VMH subnucleus, VMHil lateral-intermediate VMH be definitive and just points the way for future more detailed subnucleus, VMHvl ventral-lateral VMH subnucleus, VPM ventral studies (e.g., each of our postulated progenitor domains may premamillary nucleus, DPM dorsal premamillary nucleus, OT optic be subdivided). Interestingly, we also reached similar con- tract, TSbO tuberal suboptic nucleus clusions in two other previous studies, namely in the analysis of multiple progenitor domains (and mixed tangential migra- Hpcal1, Galanin (Veen et al. 2020). Tac1 appears in the tions) contributing different neuron types to the prepontine VMHdrm and VMHdrl subdivisions. In contrast, Pdyn is interpeduncular complex (Lorente-Cánovas et al. 2012), as present at the intermediate VMH subdivisions, which makes well as in our study of the tangentially migrated basal hypo- plausible a phenotypic correspondence with the Nkx2.2 cell thalamic ventral premamillary nucleus (López-González phenotype, though further studies would be needed. et al. 2021). A more resolutive single-cell transcriptomic analysis identified 29 neuronal types within VMH, 17 of them appar - ently restricted to the conventional ‘ventrolateral’ VMH Conclusion (Kim et al. 2019). These results show that the conventional columnar schema does not explain the existence of 29 dif- In this study we have addressed the identification and ferent glutamatergic cell types within its rather simplistic description of some molecularly identified VMH cell popula- (tripartite) VMH subdivision concept. After reinterpretation tions during development using immunofluorescence, in situ of such data within our present prosomeric VMH model we hybridization, and data extracted from the Allen Developing first note that the conventional ‘ventrolateral VMH’ only Mouse Brain Atlas. In our detailed descriptive analysis of grossly corresponds to our ventral VMH region, whose dem- these populations, we found important differences as regards onstrated subdivision into at least VMHvm, VMHvi, and their rostro-caudal and dorsoventral distribution throughout VMHvl parts was not known to columnar-based authors. the VMH nucleus, which also partly change over time, sug- Accordingly, it would be interesting to re-map the given set gesting in some cases the existence of tangential migration of 17 ‘ventrolateral’ clusters relative to our three ventral phenomena. These differences have not been described VMH subdivisions. Remarkably, Kim et al. (2019) identified before due to the widespread use of a (single) coronal sec- a Six3-expressing cell cluster and they mapped it specifi- tion plane in the analysis of this nucleus, which does not help cally to the ‘ventrolateral’ VMH region (their Satb2_Six3 the comprehensive interpretation of this territory, particu- cluster disclosed by SMART-Seq); this entity most probably larly when misinterpreted as cross-sections under columnar 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 16 Schemata illustrating our updated VMH model with color- tuberal region, forming part of the intermediate part of VMH coded molecularly-typified populations. The schemata of successive as suggested by Puelles et al. (2012). VMH sections are arranged into horizontal a–e, sagittal f–j and trans- versal k–q series (see inset drawings above illustrating the section Acknowledgements We thank the Allen Institute for Brain Science for plane as a black line relative to the bent length axis of the prosomeric public availability of the markers analyzed (Website: 2013 Allen Insti- model, represented by the alar-basal boundary in red). The molecu- tute for Brain Science. Allen Developing Mouse Brain Atlas. http:// larly characterized cell groups mapped are represented as color-coded developing mouse. br ain-map. or g). Infrastructure support was provided small circles. Note that these circles do not represent individual cells. by the University of Murcia and IMIB-Arrixaca Institute of Murcia. Black dash lines in a–e and k–q separate the mediolateral halves of the VMH. Solid black lines in f–q delimitate the dorsal, intermedi- Author contributions LLG: conceptualized and designed the study, ate and ventral parts of the VMH. Red dash lines in f–j show the performed most of the experiments, made image composition and rough position of horizontal sections a–e, and red dash lines in k–q wrote the first draft of the manuscript. MMT: performed part of the show the positions of the sagittal section levels f–j in the transversal experiments. LP: conceptualized and designed the study, wrote the n fi al schemata. VMHdrl dorsal-rostrolateral VMH subnucleus, VMHdrm manuscript, and provided funding acquisition. dorsal-rostromedial VMH subnucleus, VMHdc dorsocaudal VMH subnucleus, VMHil, lateral-intermediate VMH subnucleus, VMHim Funding Open Access funding provided thanks to the CRUE-CSIC medial-intermediate VMH subnucleus, VMHvl ventral-lateral VMH agreement with Springer Nature. This work was supported by a Spanish subnucleus, VMHvi ventral-intermediate VMH subnucleus, VMHvm Ministry of Economy and Competitiveness grant, BFU2014-57516P ventral-medial VMH subnucleus, 3 V third ventricle (with European Community FEDER support), and a Seneca Founda- tion (Autonomous Community of Murcia) Excellency Research grant, reference: 19904/ GERM/15 (to L.P.), project name: Genoarchitectonic tradition. Many authors working on the VMH seem to be Brain Development and Applications to Neurodegenrative Diseases and Cancer, by Seneca Foundation (5672 Fundación Séneca). unaware that this nucleus is a basal plate derivative of only one of the two hypothalamic neuromeres (terminal hypo- Data availability The original contributions presented in the study are thalamus; Puelles et al. 2012), and its possible patterning included in the article, further inquiries can be directed to the corre- relationships with the hypothalamic floor plate and the alar- sponding author. Images from the Allen Developing Mouse Brain Atlas (https://de veloping mouse. br ain-map. or g/) were employed in this work. basal boundary (García-Calero et al. 2008; Andreu-Cervera et al. 2019), or the acroterminal (prechordal) area of influ- Declarations ence, remain likewise unfocused, because these notions are disregarded by the obsolete columnar model. Using the Conflict of interest The authors declare that they have no conflict of interpretive conceptual apparatus offered by the prosomeric interest. model, we showed that there exists a clear correspondence Ethical approval The animal study was reviewed and approved by between several distinct progenitor areas (e.g., alar terminal Directive 2010/63/EU, Royal Decree 1201/2005 and 53/2013 Law SPa, and basal main TuD, subliminal TuD, TuI, TuD-AT, 32/107 University of Murcia Committee (No. A13170406). TuI-AT) where different VMH cell types are born and ulte- Open Access This article is licensed under a Creative Commons Attri- riorly transferred via convergent tangential or radial migra- bution 4.0 International License, which permits use, sharing, adapta- tion to the developing VMH nucleus. This provides a new tion, distribution and reproduction in any medium or format, as long example of the utility of the prosomeric model to explain as you give appropriate credit to the original author(s) and the source, the organization of complex brain structures. Given that the provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are new single-cell transcriptomic tools are manifestly sensible included in the article's Creative Commons licence, unless indicated to the morphologic models used to map the characteristic otherwise in a credit line to the material. If material is not included in clusters they detect, novel alternative interpretive systems the article's Creative Commons licence and your intended use is not need to be taken into consideration and tested as regards permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a their possible advantages. A comprehensive and understand- copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . able molecularly defined cell type census over development and consequent further study of the respective mature cell populations can be significantly improved by searching opti- References mal structural and developmental models. We focused a good part of our analysis on the Nkx2.2 Affinati AH, Sabatini PV, True C, Tomlinson AJ, Kirigiti M, Linds- transcription factor because it is related in origin to the alar/ ley SR, Li C, Olson DP, Kievit P, Myers MG, Rupp AC (2021) basal hypothalamic boundary, a not very well understood Cross-species analysis defines the conservation of anatomically segregated VMH neuron populations. Elife. https:// doi. org/ 10. limit (some expert voices define it arbitrarily as an ‘oblique’ 7554/ eLife. 69065 landmark; e.g., Shimogori et al. 2010), and also because Alifragis P, Parnavelas JG, Nadarajah B (2002) A novel method of it is a forgotten gene marker in most studies of VMH. We labeling and characterizing migrating neurons in the developing confirmed in organotypic cultures the related dorsoventral central nervous system. Exp Neurol 174(2):259–265. https:// doi. org/ 10. 1006/ exnr. 2002. 7870 tangential migration that transfers a distinct Nkx2.2 cell pop- ulation from the dorsal tuberal region to the intermediate 1 3 Brain Structure and Function Altman J, Bayer SA (1978) Development of the diencephalon in the the ventromedial nucleus of the hypothalamus. J Neurobiol rat. I. Autoradiographic study of the time of origin and settling 49(4):264–276. https:// doi. org/ 10. 1002/ neu. 10011 patterns of neurons of the hypothalamus. J Comp Neurol 182(4 Pt Dhillon H, Zigman JM, Ye C, Lee CE, McGovern RA, Tang V, Kenny 2):945–971. https:// doi. org/ 10. 1002/ cne. 90182 0511 CD, Christiansen LM, White RD, Edelstein EA, Coppari R, Altman J, Bayer SA (1986) The development of the rat hypothalamus. Balthasar N, Cowley MA, Chua S Jr, Elmquist JK, Lowell BB Adv Anat Embryol Cell Biol 100:1–178 (2006) Leptin directly activates SF1 neurons in the VMHH, and Alvarez-Bolado G, Rosenfeld MG, Swanson LW (1995) Model of fore- this action by leptin is required for normal body-weight homeo- brain regionalization based on spatiotemporal patterns of POU-III stasis. Neuron 49(2):191–203. https:// doi. org/ 10. 1016/j. neuron. homeobox gene expression, birthdates, and morphological fea-2005. 12. 021 tures. J Comp Neurol 355:237–295. https:// doi. org/ 10. 1002/ cne. Díaz C, Morales-Delgado N, Puelles L (2015) Ontogenesis of pepti- 90355 0207 dergic neurons within the genoarchitectonic map of the mouse Amat JA, Martínez-de-la-Torre M, Trujillo CM, Fernández B, Puelles hypothalamus. Front Neuroanat. https:// doi. org/ 10. 3389/ fnana. L (2022) Neurogenetic heterochrony in chick, lizard, and rat 2014. 00162 mapped with wholemount acetylcholinesterase and the proso- Diaz C, Puelles L (2020) Developmental genes and malformations in meric model. Brain Behav Evol 97(1–2):48–82. https:// doi. org/ the hypothalamus. Front Neuroanat 14:607111. https://doi. or g/10. 10. 1159/ 00052 42163389/ fnana. 2020. 607111 Andreu-Cervera A, Anselme I, Karam A, Laclef C, Catala M, Schnei- Elmquist JK, Elias CF, Saper CB (1999) From lesions to leptin: der-Maunoury S (2019) The ciliopathy gene Ftm/Rpgrip1l con- hypothalamic control of food intake and body weight. Neuron trols mouse forebrain patterning via region-specific modulation 22(2):221–232. https:// doi. org/ 10. 1016/ s0896- 6273(00) 81084-3 of Hedgehog/Gli Signaling. J Neurosci 39:2398–2415. https://doi. Ferran JL, Puelles L, Rubenstein JL (2015) Molecular codes defin- org/ 10. 1523/ JNEUR OSCI. 2199- 18. 2019 ing rostrocaudal domains in the embryonic mouse hypothalamus. Aslanpour S, Rosin JM, Balakrishnan A, Klenin N, Blot F, Gradwohl Front Neuroanat 9:46. https:// doi. org/ 10. 3389/ fnana. 2015. 00046 G, Schuurmans C, Kurrasch DM (2020a) Ascl1 is required to Frohman LA, Bernardis LL, Stachura ME (1974) Factors modifying specify a subset of ventromedial hypothalamic neurons. Develop- plasma insulin and glucose responses to ventromedial hypotha- ment. https:// doi. org/ 10. 1242/ dev. 180067 lamic stimulation. Metabolism 23(11):1047–1056. https://doi. or g/ Aslanpour S, Han S, Schuurmans C, Kurrasch DM (2020b) Neurog2 10. 1016/ 0026- 0495(74) 90071-7 acts as a classical proneural gene in the ventromedial hypothala- García-Calero E, Fernández-Garre P, Martínez S, Puelles L (2008) mus and is required for the early phase of neurogenesis. J Neu- Early mammillary pouch specification in the course of prechordal rosci 40(18):3549–3563. https:// doi. org/ 10. 1523/ JNEUR OSCI. ventralization of the forebrain tegmentum. Dev Biol 320:366–377. 2610- 19. 2020https:// doi. org/ 10. 1016/j. ydbio. 2008. 05. 545 Briscoe J, Sussel L, Serup P, Hartigan-O’Connor D, Jessell TM, Gurdjian ES (1927) The diencephalon of the albino rat. Studies on the Rubenstein JL, Ericson J (1999) Homeobox gene Nkx2.2 and brain of the rat. J Comp Neurol 43:1–144 specification of neuronal identity by graded sonic hedgehog sig- Hahn JD, Sporns O, Watts AG, Swanson LW (2019) Macroscale intrin- nalling. Nature 398:622–627. https:// doi. org/ 10. 1038/ 19315 sic network architecture of the hypothalamus. Proc Natl Acad Sci Büdefeld T, Tobet SA, Majdic G (2012) Steroidogenic factor 1 and the U S A 116:8018–8027. https://doi. or g/10. 1073/ pnas. 18194 48116 central nervous system. J Neuroendocrinol 24(1):225–235. https:// Hashikawa K, Hashikawa Y, Tremblay R, Zhang J, Feng JE, Sabol A, doi. org/ 10. 1111/j. 1365- 2826. 2011. 02174.x Piper WT, Lee H, Rudy B, Lin D (2017) Esr1+ cells in the ven- Canteras NS, Simerly RB, Swanson LW (1994) Organization of projec- tromedial hypothalamus control female aggression. Nat Neurosci tions from the ventromedial nucleus of the hypothalamus: a Pha- 20:1580–1590. https:// doi. org/ 10. 1038/ nn. 4644 seolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol Hauptmann G, Gerster T (2000) Regulatory gene expression patterns 348(1):41–79. https:// doi. org/ 10. 1002/ cne. 90348 0103 reveal transverse and longitudinal subdivisions of the embryonic Cheung CC, Kurrasch DM, Liang JK, Ingraham HA (2013) Genetic zebrafish forebrain. Mech Dev 91(1–2):105–118. https:// doi. org/ labeling of steroidogenic factor-1 (SF-1) neurons in mice reveals 10. 1016/ s0925- 4773(99) 00277-4 ventromedial nucleus of the hypothalamus (VMHH) circuitry Hauptmann G, Söll I, Gerster T (2002) The early embryonic zebrafish beginning at neurogenesis and development of a separate non- forebrain is subdivided into molecularly distinct transverse and SF-1 neuronal cluster in the ventrolateral VMHH. J Comp Neurol longitudinal domains. Brain Res Bull 57:371–375 521:1268–1288. https:// doi. org/ 10. 1002/ cne. 23226 Heimer L, Nauta WJ (1969) The hypothalamic distribution of the stria Corman TS, Bergendahl SE, Epstein DJ (2018) Distinct temporal terminalis in the rat. Brain Res 13:284–297 requirements for Sonic hedgehog signaling in development of Hetherington AW, Ranson SW (1942) The relation of various hypotha- the tuberal hypothalamus. Development. https://doi. or g/10. 1242/ lamic lesions to adiposity in the rat. J Comp Neurol 76:475–499 dev. 167379 Ifft JD (1972) An autoradiographic study of the time of final division Correa SM, Newstrom DW, Warne JP, Flandin P, Cheung CC, Lin- of neurons in rat hypothalamic nuclei. J Comp Neurol. https://doi. Moore AT, Pierce AA, Xu AW, Rubenstein JL, Ingraham HA org/ 10. 1002/ cne. 90144 0204 (2015) An estrogen-responsive module in the ventromedial hypo- Ikeda Y, Luo X, Abbud R, Nilson JH, Parker KL (1995) The nuclear thalamus selectively drives sex-specific activity in females. Cell receptor steroidogenic factor 1 is essential for the formation of the Rep 10(1):62–74. https:// doi. org/ 10. 1016/j. celrep. 2014. 12. 011 ventromedial hypothalamic nucleus. Mol Endocrinol 9(4):478– Davis AM, Seney ML, Stallings NR, Zhao L, Parker KL, Tobet SA 486. https:// doi. org/ 10. 1210/ mend.9. 4. 76590 91 (2004) Loss of steroidogenic factor 1 alters cellular topography in Kennedy A, Kunwar PS, Li LY, Stagkourakis S, Wagenaar DA, Ander- the mouse ventromedial nucleus of the hypothalamus. J Neurobiol son DJ (2020) Stimulus-specific hypothalamic encoding of a per - 60:424–436. https:// doi. org/ 10. 1002/ neu. 20030 sistent defensive state. Nature 586(7831):730–734. https://d oi.o rg/ Dellovade TL, Young M, Ross EP, Henderson R, Caron K, Parker K, 10. 1038/ s41586- 020- 2728-4 Tobet SA (2000) Disruption of the gene encoding SF-1 alters the Khodai T, Luckman SM (2021) Ventromedial nucleus of the hypothala- distribution of hypothalamic neuronal phenotypes. J Comp Neurol mus neurons under the magnifying glass. Endocrinology. https:// 423(4):579–589doi. org/ 10. 1210/ endocr/ bqab1 41 Kim KW, Sohn JW, Kohno D, Xu Y, Williams K, Elmquist JK (2011) Dellovade TL, Davis AM, Ferguson C, Sieghart W, Homanics SF-1 in the ventral medial hypothalamic nucleus: a key regulator GE, Tobet SA (2001) GABA influences the development of 1 3 Brain Structure and Function of homeostasis. Mol Cell Endocrinol 336(1–2):219–223. https:// Meek TH, Nelson JT, Matsen ME, Dorfman MD, Guyenet SJ, Damian doi. org/ 10. 1016/j. mce. 2010. 11. 019 V, Allison MB, Scarlett JM, Nguyen HT, Thaler JP, Olson DP, Kim DW, Yao Z, Graybuck LT, Kim TK, Nguyen TN, Smith KA, Fong Myers MG Jr, Schwartz MW, Morton GJ (2016) Functional iden- O, Yi L, Koulena N, Pierson N, Shah S, Lo L, Pool AH, Oka Y, tification of a neurocircuit regulating blood glucose. Proc Natl Pachter L, Cai L, Tasic B, Zeng H, Anderson DJ (2019) Multi- Acad Sci U S A 113(14):E2073–E2082. https:// doi. org/ 10. 1073/ modal analysis of cell types in a hypothalamic node controlling pnas. 15211 60113 social behavior. Cell 179(3):713–728. https:// doi. org/ 10. 1016/j. Morales-Delgado N, Merchan P, Bardet SM, Ferrán JL, Puelles L, Díaz cell. 2019. 09. 020 C (2011) Topography of somatostatin gene expression relative Kim DW, Liu K, Wang ZQ, Zhang YS, Bathini A, Brown MP, Lin SH, to molecular progenitor domains during ontogeny of the mouse Washington PW, Sun C, Lindtner S, Lee B, Wang H, Shimogori hypothalamus. Front Neuroanat 5:10. https:// doi. org/ 10. 3389/ T, Rubenstein JLR, Blackshaw S (2021) Gene regulatory networks fnana. 2011. 00010 controlling differentiation, survival, and diversification of hypo- Morales-Delgado N, Castro-Robles B, Ferrán JL, Martinez-de-la-Torre thalamic Lhx6-expressing GABAergic neurons. Commun Biol M, Puelles L, Díaz C (2014) Regionalized differentiation of CRH, 4(1):95. https:// doi. org/ 10. 1038/ s42003- 020- 01616-7 TRH, and GHRH peptidergic neurons in the mouse hypothala- Krause WC, Ingraham HA (2017) Origins and functions of the ven- mus. Brain Struct Funct 219:1083–1111. https://doi. or g/10. 1007/ trolateral VMHH: a complex neuronal cluster orchestrating sex s00429- 013- 0554-2 differences in metabolism and behavior. Adv Exp Med Biol Nieuwenhuys R, Puelles L (2016) Towards a new neuromor- 1043:199–213. https:// doi. org/ 10. 1007/ 978-3- 319- 70178-3_ 10 phology. Switz, Springer Int Publ. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / Krieg WJS (1932) The hypothalamus of the albino rat. J Comp Neurol 978-3- 319- 25693-1 55:19–89 Nishi Y, Zhang X, Jeong J, Peterson KA, Vedenko A, Bulyk ML, Hide Kurrasch DM, Cheung CC, Lee FY, Tran PV, Hata K, Ingraham HA WA, McMahon AP (2015) Development 142:3286–3293. https:// (2007) The neonatal ventromedial hypothalamus transcriptome doi. org/ 10. 1242/ dev. 124636 reveals novel markers with spatially distinct patterning. J Neurosci Orquera DP, Nasif S, Low MJ, Rubinstein M, de Souza FSJ (2016) 50:13624–13634. https://doi. or g/10. 1523/ JNEUR OSCI. 2858- 07. Essential function of the transcription factor Rax in the early pat- 2007 terning of the mammalian hypothalamus. Dev Biol 416:212–224. Lagutin OV, Zhu CC, Kobayashi D, Topczewski J, Shimamura K, https:// doi. org/ 10. 1016/j. ydbio. 2016. 05. 021 Puelles L, Russell HR, McKinnon PJ, Solnica-Krezel L, Oliver G Pfaff DW, Sakuma Y (1979a) Deficit in the lordosis reflex of female (2003) Six3 repression of Wnt signaling in the anterior neuroec- rats caused by lesions in the ventromedial nucleus of the hypo- toderm is essential for vertebrate forebrain development. Genes thalamus. J Physiol 288:203–210 Dev 17:368–379. https:// doi. org/ 10. 1101/ gad. 10594 03 Pfaff DW, Sakuma Y (1979b) Facilitation of the lordosis reflex of Lee H, Kim DW, Remedios R, Anthony TE, Chang A, Madisen L, Zeng female rats fromthe ventromedial nucleus of the hypothalamus. J H, Anderson DJ (2014) Scalable control of mounting and attack Physiol 288:189–202 by Esr1+ neurons in the ventromedial hypothalamus. Nature Puelles L (2018) Developmental studies of avian brain organization. 509(7502):627–632. https:// doi. org/ 10. 1038/ natur e13169 Int J Dev Biol 62:207–224. https://doi. or g/10. 1387/ i jdb.17027 9LP Lewis JE, Woodward OR, Nuzzaci D, Smith CA, Adriaenssens AE, Puelles L (2019) Survey of midbrain, diencephalon, and hypothalamus Billing L, Brighton C, Phillips BU, Tadross JA, Kinston SJ, Ciab- neuroanatomic terms whose prosomeric definition conflicts with atti E, Göttgens B, Tripodi M, Hornigold D, Baker D, Gribble FM, columnar tradition. Front Neuroanat 13:20. htt ps: // do i. org/ 10 . Reimann F (2022) Relaxin/insulin-like family peptide receptor 4 3389/ fnana. 2019. 00020 (Rxfp4) expressing hypothalamic neurons modulate food intake Puelles L (2022) Prosomeric classification of retinorecipient centers: a and preference in mice. Mol Metab. https:// doi. org/ 10. 1016/j. new causal scenario. Brain Struct Funct 227:1171–1193. https:// molmet. 2022. 101604doi. org/ 10. 1007/ s00429- 022- 02461-6 Lin D, Boyle MP, Dollar P, Lee H, Lein ES, Perona P, Anderson DJ Puelles L, Martinez S (2013) Patterning of the Diencephalon. John LR (2011) Functional identification of an aggression locus in the Rubenstein, Pasko Rakic, patterning and cell type specification mouse hypothalamus. Nature 470(7333):221–226. https:// doi. in the developing CNS and PNS 8. Academic Press, USA, pp org/ 10. 1038/ natur e09736 151–172. https:// doi. org/ 10. 1016/ B978-0- 12- 397265- 1. 00048-4 López-González L, Alonso A, García-Calero E, de Puelles E, Puelles Puelles L, Rubenstein JL (2015) A new scenario of hypothalamic L (2021) Tangential intrahypothalamic migration of the mouse organization: rationale of new hypotheses introduced in the ventral premamillary nucleus and Fgf8 signaling. Front Cell Dev updated prosomeric model. Front Neuroanat 9:27. ht tps:/ / doi. Biol. https:// doi. org/ 10. 3389/ fcell. 2021. 676121org/ 10. 3389/ fnana. 2015. 00027 Lu F, Kar D, Gruenig N, Zhang ZW, Cousins N, Rodgers HM, Swin- Puelles L, Martinez-de-la-Torre M, Bardet S, Rubenstein JLR (2012) dell EC, Jamrich M, Schuurmans C, Mathers PH, Kurrasch DM Hypothalamus. The Mouse Nervous System, Watson C, Paxi- (2013) Rax is a selector gene for mediobasal hypothalamic cell nos G, Puelles L San Diego. Academic Press/Elsevier, CA, pp types. J Neurosci 33:259–272. https:// doi. org/ 10. 1523/ JNEUR 221–312 OSCI. 0913- 12. 2013 Qiu M, Shimamura K, Sussel L, Chen S, Rubenstein JL (1998) Con- Ma T, Wong SZH, Lee B, Ming GL, Song H (2021) Decoding neuronal trol of anteroposterior and dorsoventral domains of Nkx-6.1 gene composition and ontogeny of individual hypothalamic nuclei. expression relative to other Nkx genes during vertebrate CNS Neuron 109(7):1150–1167. https:// doi. or g/ 10. 1016/j. neur on. development. Mech Dev 72:77–88. https://doi. or g/10. 1016/ s0925- 2021. 01. 0264773(98) 00018-5 McClellan KM, Parker KL, Tobet S (2006) Development of the ven- Ramón y Cajal S (1911) Histologie du Système Nerveux de l’homme tromedial nucleus of the hypothalamus. Front Neuroendocrinol et des Vertébrés. Vol. 2. Re-edit. of 1954 Madrid: CSIC 27(2):193–209. https:// doi. org/ 10. 1016/j. yfrne. 2006. 02. 002 Romanov RA, Tretiakov EO, Kastriti ME, Zupancic M, Häring M, McClellan KM, Calver AR, Tobet SA (2008) GABAB receptors role in Korchynska S, Popadin K, Benevento M, Rebernik P, Lallemend cell migration and positioning within the ventromedial nucleus of F, Nishimori K, Clotman F, Andrews WD, Parnavelas JG, Farlik the hypothalamus. Neuroscience 151(4):1119–1131. https:// doi. M, Bock C, Adameyko I, Hökfelt T, Keimpema E, Harkany T org/ 10. 1016/j. neuro scien ce. 2007. 11. 048 (2020) Molecular design of hypothalamus development. Nature 582(7811):246–252. https:// doi. org/ 10. 1038/ s41586- 020- 2266-0 1 3 Brain Structure and Function Saper CB, Lowell BB (2014) The hypothalamus. Curr Biol. https://doi. Tran PV, Lee MB, Marín O, Xu B, Jones KR, Reichardt LF, Ruben- org/ 10. 1016/j. cub. 2014. 10. 023 stein JR, Ingraham HA (2003) Requirement of the orphan nuclear Saper CB, Swanson LW, Cowan WM (1976) The efferent connec- receptor SF-1 in terminal differentiation of ventromedial hypotha- tions of the ventromedial nucleus of the hypothalamus of the rat. lamic neurons. Mol Cell Neurosci 4:441–453. https:// doi. org/ 10. J Comp Neurol 169(4):409–442. https :// d oi. o r g/ 10 . 100 2/ c ne. 1016/ s1044- 7431(03) 00027-7 90169 0403 Trowe MO, Zhao L, Weiss AC, Christoffels V, Epstein DJ, Kispert Shimada M, Nakamura T (1973) Time of neuron origin in mouse hypo- A (2013) Inhibition of Sox2-dependent activation of Shh in the thalamic nuclei. Exp Neurol 41:163–173. https://doi. or g/10. 1016/ ventral diencephalon by Tbx3 is required for formation of the 0014- 4886(73) 90187-8 neurohypophysis. Development 140(11):2299–2309. https:// doi. Shimamura K, Hirano S, McMahon AP, Takeichi M (1994) Wnt-org/ 10. 1242/ dev. 094524 1-dependent regulation of local E-cadherin and alpha N-catenin van den Akker WM, Brox A, Puelles L, Durston AJ, Medina L (2008) expression in the embryonic mouse brain. Development Comparative functional analysis provides evidence for a crucial 120:2225–2234 role for the homeobox gene Nkx2.1/Titf-1 in forebrain evolution. Shimamura K, Hartigan DJ, Martinez S, Puelles L, Rubenstein JL J Comp Neurol 506:211–223. https:// doi. org/ 10. 1002/ cne. 21542 (1995) Longitudinal organization of the anterior neural plate and Van Houten M, Brawer JR (1978) Cytology of neurons of the hypotha- neural tube. Development 121:3923–3933 lamic ventromedial nucleus in the adult male rat. J Comp Neurol Shimogawa Y, Sakuma Y, Yamanouchi K (2015) Efferent and afferent 178(1):89–116. https:// doi. org/ 10. 1002/ cne. 90178 0106 connections of the ventromedial hypothalamic nucleus determined van Veen JE, Kammel LG, Bunda PC, Shum M, Reid MS, Massa MG, by neural tracer analysis: implications for lordosis regulation in Arneson D, Park JW, Zhang Z, Joseph AM, Hrncir H, Liesa M, female rats. Neurosci Res 91:19–33. https:// doi. org/ 10. 1016/j. Arnold AP, Yang X, Correa SM (2020) Hypothalamic estrogen neures. 2014. 10. 016 receptor alpha establishes a sexually dimorphic regulatory node Shimogori T, Lee DA, Miranda-Angulo A, Yang Y, Wang H, Jiang L, of energy expenditure. Nat Metab 2(4):351–363. https:// doi. org/ Yoshida AC, Kataoka A, Mashiko H, Avetisyan M, Qi L, Qian 10. 1038/ s42255- 020- 0189-6 J, Blackshaw S (2010) A genomic atlas of mouse hypothalamic Wang L, Chen IZ, Lin D (2015) Collateral pathways from the ven- development. Nat Neurosci 13:767–775. https://doi. or g/10. 1038/ tromedial hypothalamus mediate defensive behaviors. Neuron nn. 2545 85(6):1344–1358. https:// doi. org/ 10. 1016/j. neuron. 2014. 12. 025 Shinoda K, Lei H, Yoshii H, Nomura M, Nagano M, Shiba H, Sasaki H, Yang CF, Chiang MC, Gray DC, Prabhakaran M, Alvarado M, Juntti Osawa Y, Ninomiya Y, Niwa O (1995) Developmental defects of SA, Unger EK, Wells JA, Shah NM (2013) Sexually dimorphic the ventromedial hypothalamic nucleus and pituitary gonadotroph neurons in the ventromedial hypothalamus govern mating in both in the Ftz-F1 disrupted mice. Dev Dyn 204(1):22–29. https://doi. sexes and aggression in males. Cell 153(4):896–909. https:// doi. org/ 10. 1002/ aja. 10020 40104org/ 10. 1016/j. cell. 2013. 04. 017 Skidmore JM, Cramer JD, Martin JF, Martin DM (2008) Cre fate map- Ziegler DR, Cullinan WE, Herman JP (2002) Distribution of vesicular ping reveals lineage specific defects in neuronal migration with glutamate transporter mRNA in rat hypothalamus. J Comp Neurol loss of Pitx2 function in the developing mouse hypothalamus and 448(3):217–229. https:// doi. org/ 10. 1002/ cne. 10257 subthalamic nucleus. Mol Cell Neurosci 37:696–707. https://doi. org/ 10. 1016/j. mcn. 2007. 12. 015 Publisher's Note Springer Nature remains neutral with regard to Swanson LW (2012) Brain architecture. Oxford University Press, New jurisdictional claims in published maps and institutional affiliations. York 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Brain Structure and Function Springer Journals http://www.deepdyve.com/lp/springer-journals/populational-heterogeneity-and-partial-migratory-origin-of-the-ifmkwxvjgx

Loading next page...

References (91)

K. McClellan, K. Parker, S. Tobet (2006)
Development of the ventromedial nucleus of the hypothalamus
Frontiers in Neuroendocrinology, 27
A. Hetherington, S. Ranson (1942)
The relation of various hypothalamic lesions to adiposity in the rat
Journal of Comparative Neurology, 76
C. Díaz, N. Morales-Delgado, L. Puelles (2015)
Ontogenesis of peptidergic neurons within the genoarchitectonic map of the mouse hypothalamus
Frontiers in Neuroanatomy, 8
Tanya Corman, Solsire Bergendahl, D. Epstein (2018)
Distinct temporal requirements for Sonic hedgehog signaling in development of the tuberal hypothalamus
Development, 145
D. Kurrasch, C. Cheung, F. Lee, P. Tran, K. Hata, H. Ingraham (2007)
The Neonatal Ventromedial Hypothalamus Transcriptome Reveals Novel Markers with Spatially Distinct Patterning
The Journal of Neuroscience, 27
J. Ferran, L. Puelles, J. Rubenstein (2015)
Molecular codes defining rostrocaudal domains in the embryonic mouse hypothalamus
Frontiers in Neuroanatomy, 9
J. Elmquist, C. Elias, C. Saper (1999)
From Lesions to Leptin Hypothalamic Control of Food Intake and Body Weight
Neuron, 22
Koichi Hashikawa, Y. Hashikawa, R. Tremblay, Jiaxing Zhang, James Feng, Alexander Sabol, Walter Piper, Hyosang Lee, B. Rudy, Dayu Lin (2017)
Esr1+ cells in the ventromedial hypothalamus control female aggression
Nature neuroscience, 20
K. McClellan, A. Calver, S. Tobet (2008)
GABAB receptors role in cell migration and positioning within the ventromedial nucleus of the hypothalamus
Neuroscience, 151
W. Akker, Aurora Brox, L. Puelles, A. Durston, L. Medina (2008)
Comparative functional analysis provides evidence for a crucial role for the homeobox gene Nkx2.1/Titf‐1 in forebrain evolution
Journal of Comparative Neurology, 506
Li Wang, Irene Chen, Dayu Lin (2015)
Collateral Pathways from the Ventromedial Hypothalamus Mediate Defensive Behaviors
Neuron, 85
G. Hauptmann, Iris Söll, T. Gerster (2002)
The early embryonic zebrafish forebrain is subdivided into molecularly distinct transverse and longitudinal domains
Brain Research Bulletin, 57
Dayu Lin, Maureen Boyle, P. Dollár, Hyosang Lee, P. Perona, E. Lein, D. Anderson (2010)
Functional identification of an aggression locus in the mouse hypothalamus
Nature, 470
Y. Ikeda, Xunrong Luo, R. Abbud, J. Nilson, K. Parker (1995)
The nuclear receptor steroidogenic factor 1 is essential for the formation of the ventromedial hypothalamic nucleus.
Molecular endocrinology, 9 4
D. Pfaff, Y. Sakuma (1979)
Deficit in the lordosis reflex of female rats caused by lesions in the ventromedial nucleus of the hypothalamus.
The Journal of Physiology, 288
J. Altman, S. Bayer (1978)
Development of the diencephalon in the rat. I. Autoradiographic study of the time of origin and settling patterns of neurons of the hypothalamus
Journal of Comparative Neurology, 182
Hyosang Lee, Dong-Wook Kim, Ryan Remedios, T. Anthony, Angela Chang, L. Madisen, Hongkui Zeng, D. Anderson (2014)
Scalable Control of Mounting and Attack by ESR1+ Neurons in the Ventromedial Hypothalamus
Nature, 509
J. Altman, S. Bayer (1986)
The development of the rat hypothalamus.
Advances in anatomy, embryology, and cell biology, 100
Shaghayegh Aslanpour, Jessica Rosin, Anjali Balakrishnan, N. Klenin, Florence Blot, G. Gradwohl, C. Schuurmans, D. Kurrasch (2020)
Ascl1 is required to specify a subset of ventromedial hypothalamic neurons
Development, 147
Jchn Ifft (1972)
An autoradiographic study of the time of final division of neurons in rat hypothalamic nuclei
Journal of Comparative Neurology, 144
N. Morales-Delgado, P. Merchán, S. Bardet, J. Ferran, L. Puelles, C. Díaz (2010)
Topography of Somatostatin Gene Expression Relative to Molecular Progenitor Domains during Ontogeny of the Mouse Hypothalamus
Frontiers in Neuroanatomy, 5
T. Shimogori, Daniel Lee, Ana Miranda-Angulo, Yanqin Yang, Hong Wang, Lizhi Jiang, Aya Yoshida, A. Kataoka, Hiromi Mashiko, Marina Avetisyan, Lixin Qi, Jiang Qian, S. Blackshaw (2010)
A genomic atlas of mouse hypothalamic development
Nature Neuroscience, 13
Dong-Wook Kim, Zizhen Yao, Lucas Graybuck, Tae Kim, T. Nguyen, Kimberly Smith, Olivia Fong, Lynn Yi, N. Koulena, N. Pierson, Sheel Shah, Liching Lo, Allan-Hermann Pool, Yuki Oka, L. Pachter, L. Cai, Bosiljka Tasic, Hongkui Zeng, D. Anderson (2019)
Multimodal Analysis of Cell Types in a Hypothalamic Node Controlling Social Behavior
Cell, 179
L. Puelles (2022)
Prosomeric classification of retinorecipient centers: a new causal scenario
Brain Structure and Function, 227
T. Dellovade, M. Young, E. Ross, R. Henderson, K. Caron, K. Parker, S. Tobet (2000)
Disruption of the gene encoding SF-1 alters the distribution of hypothalamic neuronal phenotypes.
The Journal of comparative neurology, 423 4
H. Donkelaar, L. Vasung (2015)
Development of the Diencephalon
T. Budefeld, Stuart Tobet, Gregor Majdic, Gregor Majdic (2012)
Steroidogenic Factor 1 and the Central Nervous System
Journal of Neuroendocrinology, 24
E. Gurdjian (1927)
The diencephalon of the albino rat . Studies on the brain of the rat. No. 2
Journal of Comparative Neurology, 43
TL Dellovade, M Young, EP Ross, R Henderson, K Caron, K Parker, SA Tobet (2000)
Disruption of the gene encoding SF-1 alters the distribution of hypothalamic neuronal phenotypes
J Comp Neurol, 423
J. Amat, M. Martı́nez-de-la-Torre, C. Trujillo, B. Fernández, L. Puelles (2022)
Neurogenetic Heterochrony in Chick, Lizard, and Rat Mapped with Wholemount Acetylcholinesterase and the Prosomeric Model
Brain, Behavior and Evolution, 97
L. Puelles (2019)
Survey of Midbrain, Diencephalon, and Hypothalamus Neuroanatomic Terms Whose Prosomeric Definition Conflicts With Columnar Tradition
Frontiers in Neuroanatomy, 13
N. Morales-Delgado, B. Castro-Robles, J. Ferran, M. Martı́nez-de-la-Torre, L. Puelles, C. Díaz (2013)
Regionalized differentiation of CRH, TRH, and GHRH peptidergic neurons in the mouse hypothalamus
Brain Structure & Function, 219
Tansi Khodai, S. Luckman (2021)
Ventromedial Nucleus of the Hypothalamus Neurons Under the Magnifying Glass
Endocrinology, 162
Stephanie Correa, D. Newstrom, J. Warne, P. Flandin, C. Cheung, Alexander Lin-Moore, Andrew Pierce, Allison Xu, J. Rubenstein, H. Ingraham (2015)
An estrogen-responsive module in the ventromedial hypothalamus selectively drives sex-specific activity in females.
Cell reports, 10 1
Cindy Yang, Michael Chiang, Daniel Gray, M. Prabhakaran, Maricruz Alvarado, Scott Juntti, Elizabeth Unger, J. Wells, N. Shah (2013)
Sexually Dimorphic Neurons in the Ventromedial Hypothalamus Govern Mating in Both Sexes and Aggression in Males
Cell, 153
Tong Ma, S. Wong, Bora Lee, G. Ming, Hongjun Song (2021)
Decoding neuronal composition and ontogeny of individual hypothalamic nuclei
Neuron, 109
D. Kim, Kai Liu, Zoe Wang, Yi Zhang, Abhijith Bathini, Matthew Brown, S. Lin, Parris Washington, Changyu Sun, S. Lindtner, Bora Lee, Hong Wang, T. Shimogori, J. Rubenstein, S. Blackshaw (2020)
Gene regulatory networks controlling differentiation, survival, and diversification of hypothalamic Lhx6-expressing GABAergic neurons
Communications Biology, 4
N. Canteras, R. Simerly, L. Swanson (1994)
Organization of projections from the ventromedial nucleus of the hypothalamus: A Phaseolus vulgaris‐Leucoagglutinin study in the rat
Journal of Comparative Neurology, 348
C. Saper, L. Swanson, W. Cowan (1976)
The efferent connections of the ventromedial nucleus of the hypothalamus of the rat ,
Journal of Comparative Neurology, 169
Mark-Oliver Trowe, Li Zhao, Anna-Carina Weiss, V. Christoffels, D. Epstein, A. Kispert (2013)
Inhibition of Sox2-dependent activation of Shh in the ventral diencephalon by Tbx3 is required for formation of the neurohypophysis
Development, 140
G. Hauptmann, T. Gerster (2000)
Regulatory gene expression patterns reveal transverse and longitudinal subdivisions of the embryonic zebrafish forebrain
Mechanisms of Development, 91
C. Saper, B. Lowell (2014)
The hypothalamus
Current Biology, 24
A. Kennedy, Prabhat Kunwar, Ling-yun Li, Stefanos Stagkourakis, D. Wagenaar, D. Anderson (2020)
Stimulus-specific hypothalamic encoding of a persistent defensive state
Nature, 586
C. Cheung, D. Kurrasch, Jenna Liang, H. Ingraham (2013)
Genetic labeling of steroidogenic factor‐1 (SF‐1) neurons in mice reveals ventromedial nucleus of the hypothalamus (VMH) circuitry beginning at neurogenesis and development of a separate non‐SF‐1 neuronal cluster in the ventrolateral VMH
Journal of Comparative Neurology, 521
D. Ziegler, W. Cullinan, J. Herman (2002)
Distribution of vesicular glutamate transporter mRNA in rat hypothalamus
Journal of Comparative Neurology, 448
Lara López-González, Antonia Alonso, E. Garcia-Calero, Eduardo Puelles, L. Puelles (2021)
Tangential Intrahypothalamic Migration of the Mouse Ventral Premamillary Nucleus and Fgf8 Signaling
Frontiers in Cell and Developmental Biology, 9
Alison Affinati, Paul Sabatini, C. True, Abigail Tomlinson, M. Kirigiti, Sarah Lindsley, Chien Li, D. Olson, P. Kievit, M. Myers, Alan Rupp (2021)
Cross-Species Analysis Defines the Conservation of Anatomically-Segregated VMH Neuron Populations
bioRxiv
K. Shimamura, Shinji Hirano, Andy McMahon, M. Takeichi (1994)
Wnt-1-dependent regulation of local E-cadherin and alpha N-catenin expression in the embryonic mouse brain.
Development, 120 8
Y. Nishi, Xiaoxiao Zhang, Jieun Jeong, Kevin Peterson, Anastasia Vedenko, M. Bulyk, Winston Hide, A. McMahon (2015)
A direct fate exclusion mechanism by Sonic hedgehog-regulated transcriptional repressors
Development, 142
L. Puelles, J. Rubenstein (2015)
A new scenario of hypothalamic organization: rationale of new hypotheses introduced in the updated prosomeric model
Frontiers in Neuroanatomy, 9
Fuqu Lu, Deepon Kar, Nicole Gruenig, Zi Zhang, Nicole Cousins, Helen Rodgers, E. Swindell, M. Jamrich, C. Schuurmans, P. Mathers, D. Kurrasch (2013)
Rax Is a Selector Gene for Mediobasal Hypothalamic Cell Types
The Journal of Neuroscience, 33
J. Lewis, Orla Woodward, Danaé Nuzzaci, Christopher Smith, A. Adriaenssens, Lawrence Billing, Cheryl Brighton, Benjamin Phillips, J. Tadross, S. Kinston, E. Ciabatti, B. Göttgens, Marco Tripodi, D. Hornigold, D. Baker, F. Gribble, F. Reimann (2022)
Relaxin/insulin-like family peptide receptor 4 (Rxfp4) expressing hypothalamic neurons modulate food intake and preference in mice
Molecular Metabolism, 66
C. Díaz, L. Puelles (2020)
Developmental Genes and Malformations in the Hypothalamus
Frontiers in Neuroanatomy, 14
J. Hahn, O. Sporns, A. Watts, L. Swanson (2019)
Macroscale intrinsic network architecture of the hypothalamus
Proceedings of the National Academy of Sciences, 116
W. Krieg (1932)
The hypothalamus of the albino rat
Journal of Comparative Neurology, 55
AH Affinati, PV Sabatini, C True, AJ Tomlinson, M Kirigiti, SR Lindsley, C Li, DP Olson, P Kievit, MG Myers, AC Rupp (2021)
Cross-species analysis defines the conservation of anatomically segregated VMH neuron populations
Elife
Y. Shimogawa, Y. Sakuma, K. Yamanouchi (2015)
Efferent and afferent connections of the ventromedial hypothalamic nucleus determined by neural tracer analysis: Implications for lordosis regulation in female rats
Neuroscience Research, 91
J. Veen, Laura Kammel, Patricia Bunda, Michael Shum, M. Reid, M. Massa, Douglas Arneson, J. Park, Zhi Zhang, Alexia Joseph, Haley Hrncir, M. Liesa, A. Arnold, Xia Yang, Stephanie Correa (2020)
Hypothalamic estrogen receptor alpha establishes a sexually dimorphic regulatory node of energy expenditure
Nature metabolism, 2
W. Krause, H. Ingraham (2017)
Origins and Functions of the Ventrolateral VMH: A Complex Neuronal Cluster Orchestrating Sex Differences in Metabolism and Behavior.
Advances in experimental medicine and biology, 1043
Ki Kim, Jong-Woo Sohn, D. Kohno, Yong Xu, K. Williams, J. Elmquist (2011)
SF-1 in the ventral medial hypothalamic nucleus: A key regulator of homeostasis
Molecular and Cellular Endocrinology, 336
J. Briscoe, L. Sussel, P. Serup, D. Hartigan-O’Connor, T. Jessell, J. Rubenstein, J. Ericson (1999)
Homeobox gene Nkx2.2 and specification of neuronal identity by graded Sonic hedgehog signalling
Nature, 398
A. Andreu-Cervera, I. Anselme, A. Karam, C. Laclef, M. Catala, S. Schneider-Maunoury (2018)
The Ciliopathy Gene Ftm/Rpgrip1l Controls Mouse Forebrain Patterning via Region-Specific Modulation of Hedgehog/Gli Signaling
The Journal of Neuroscience, 39
T. Dellovade, A. Davis, C. Ferguson, W. Sieghart, G. Homanics, S. Tobet (2001)
GABA influences the development of the ventromedial nucleus of the hypothalamus.
Journal of neurobiology, 49 4
(2012)
Hypothalamus. The Mouse Nervous System
Daniela Orquera, Sofia Nasif, M. Low, M. Rubinstein, F. Souza (2016)
Essential function of the transcription factor Rax in the early patterning of the mammalian hypothalamus.
Developmental biology, 416 1
H. Dhillon, J. Zigman, C. Ye, Charlotte Lee, R. McGovern, V. Tang, Christopher Kenny, Lauryn Christiansen, Ryan White, E. Edelstein, R. Coppari, N. Balthasar, M. Cowley, S. Chua, J. Elmquist, B. Lowell (2006)
Leptin Directly Activates SF1 Neurons in the VMH, and This Action by Leptin Is Required for Normal Body-Weight Homeostasis
Neuron, 49
E. Garcia-Calero, Pedro Fernández-Garre, S. Martinez, L. Puelles (2008)
Early mammillary pouch specification in the course of prechordal ventralization of the forebrain tegmentum.
Developmental biology, 320 2
L Puelles, S Martinez (2013)
Patterning of the Diencephalon
John LR Rubenstein, Pasko Rakic, patterning and cell type specification in the developing CNS and PNS 8
K. Shimamura, Dennis Hartigan, Salvador Martínez, L. Puelles, J. Rubenstein (1995)
Longitudinal organization of the anterior neural plate and neural tube.
Development, 121 12
(2013)
Pasko Rakic, patterning and cell type specification in the developing CNS and PNS 8
G. Alvarez-Bolado, M. Rosenfeld, L. Swanson (1995)
Model of forebrain regionalization based on spatiotemporal patterns of POU‐III homeobox gene expression, birthdates, and morphological features
Journal of Comparative Neurology, 355
L Puelles, M Martinez-de-la-Torre, S Bardet, JLR Rubenstein (2012)
Hypothalamus
The Mouse Nervous System, Watson C, Paxinos G, Puelles L San Diego
Mark Houten, J. Brawer (1978)
Cytology of neurons of the hypothalamic ventromedial nucleus in the adult male rat
Journal of Comparative Neurology, 178
J. Skidmore, J. Cramer, James Martin, Donna Martin (2008)
Cre fate mapping reveals lineage specific defects in neuronal migration with loss of Pitx2 function in the developing mouse hypothalamus and subthalamic nucleus
Molecular and Cellular Neuroscience, 37
DW Kim, K Liu, ZQ Wang, YS Zhang, A Bathini, MP Brown, SH Lin, PW Washington, C Sun, S Lindtner, B Lee, H Wang, T Shimogori, JLR Rubenstein, S Blackshaw (2021)
Gene regulatory networks controlling differentiation, survival, and diversification of hypothalamic Lhx6-expressing GABAergic neurons
Commun Biol, 4
J Altman, SA Bayer (1986)
The development of the rat hypothalamus
Adv Anat Embryol Cell Biol, 100
O. Lagutin, C. Zhu, D. Kobayashi, J. Topczewski, K. Shimamura, L. Puelles, Helen Russell, P. Mckinnon, L. Solnica-Krezel, G. Oliver (2003)
Six3 repression of Wnt signaling in the anterior neuroectoderm is essential for vertebrate forebrain development.
Genes & development, 17 3
R. Romanov, Evgenii Tretiakov, M. Kastriti, Maja Zupančič, M. Häring, S. Korchynska, K. Popadin, M. Benevento, P. Rebernik, F. Lallemend, K. Nishimori, F. Clotman, W. Andrews, J. Parnavelas, M. Farlik, C. Bock, I. Adameyko, T. Hökfelt, E. Keimpema, T. Harkany (2020)
Molecular design of hypothalamus development
Nature, 582
Thomas Meek, Jarrell Nelson, Miles Matsen, M. Dorfman, Stephan Guyenet, Vincent Damian, Margaret Allison, J. Scarlett, Hong Nguyen, J. Thaler, D. Olson, M. Myers, M. Schwartz, G. Morton (2016)
Functional identification of a neurocircuit regulating blood glucose
Proceedings of the National Academy of Sciences, 113
D. Pfaff, Y. Sakuma (1979)
Facilitation of the lordosis reflex of female rats from the ventromedial nucleus of the hypothalamus.
The Journal of Physiology, 288
Shaghayegh Aslanpour, Sisu Han, C. Schuurmans, D. Kurrasch (2020)
Neurog2 Acts as a Classical Proneural Gene in the Ventromedial Hypothalamus and Is Required for the Early Phase of Neurogenesis
The Journal of Neuroscience, 40
R. Nieuwenhuys, L. Puelles (2015)
Towards a New Neuromorphology
P. Alifragis, J. Parnavelas, B. Nadarajah (2002)
A Novel Method of Labeling and Characterizing Migrating Neurons in the Developing Central Nervous System
Experimental Neurology, 174
L. Frohman, L. Frohman, L. Bernardis, L. Bernardis, M. Stachura, M. Stachura (1974)
Factors modifying plasma insulin and glucose responses to ventromedial hypothalamic stimulation.
Metabolism: clinical and experimental, 23 11
Lennart Heimer, Walle Nauta (1969)
The hypothalamic distribution of the stria terminalis in the rat.
Brain research, 13 2
L. Puelles (2018)
Developmental studies of avian brain organization.
The International journal of developmental biology, 62 1-2-3
A. Davis, M. Seney, N. Stallings, Liping Zhao, K. Parker, S. Tobet (2004)
Loss of steroidogenic factor 1 alters cellular topography in the mouse ventromedial nucleus of the hypothalamus.
Journal of neurobiology, 60 4
M. Shimada, Tsuneo Nakamura (1973)
Time of neuron origin in mouse hypothalamic nuclei.
Experimental neurology, 41 1
M. Qiu, K. Shimamura, L. Sussel, Sandy Chen, J. Rubenstein (1998)
Control of anteroposterior and dorsoventral domains of Nkx-6.1 gene expression relative to other Nkx genes during vertebrate CNS development
Mechanisms of Development, 72
P. Tran, Martin Lee, O. Marín, Baoji Xu, K. Jones, L. Reichardt, John Rubenstein, H. Ingraham (2003)
Requirement of the orphan nuclear receptor SF-1 in terminal differentiation of ventromedial hypothalamic neurons
Molecular and Cellular Neuroscience, 22
K. Shinoda, Hong Lei, H. Yoshii, M. Nomura, M. Nagano, Hiroharu Shiba, H. Sasaki, Y. Osawa, Y. Ninomiya, O. Niwa, K. Morohashi, E. Li (1995)
Developmental defects of the ventromedial hypothalamic nucleus and pituitary gonadotroph in the Ftz‐F1 disrupted mice
Developmental Dynamics, 204

Publisher: Springer Journals
Copyright: Copyright © The Author(s) 2023
ISSN: 1863-2653
eISSN: 1863-2661
DOI: 10.1007/s00429-022-02601-y
Publisher site: See Article on Publisher Site

Abstract

The ventromedial hypothalamic nucleus (VMH) is one of the most distinctive hypothalamic tuberal structures, subject of numerous classic and modern functional studies. Commonly, the adult VMH has been divided in several portions, attending to differences in cell aggregation, cell type, connectivity, and function. Consensus VMH partitions in the literature com- prise the dorsomedial (VMHdm), and ventrolateral (VMHvl) subnuclei, which are separated by an intermediate or central (VMHc) population (topographic names based on the columnar axis). However, some recent transcriptome analyses have identified a higher number of different cell types in the VMH, suggesting additional subdivisions, as well as the possibility of separate origins. We offer a topologic and genoarchitectonic developmental study of the mouse VMH complex using the prosomeric axis as a reference. We analyzed genes labeling specific VMH subpopulations, with particular focus upon the Nkx2.2 transcription factor, a marker of the alar-basal boundary territory of the prosencephalon, from where some cells seem to migrate dorsoventrally into VMH. We also identified separate neuroepithelial origins of a Nr2f1-positive subpopulation, and a new Six3-positive component, as well as subtle differences in origin of Nr5a1 positive versus Nkx2.2-positive cell populations entering dorsoventrally the VMH. Several of these migrating cell types are born in the dorsal tuberal domain and translocate ventralwards to reach the intermediate tuberal domain, where the adult VMH mass is located in the adult. This work provides a more detailed area map on the intrinsic organization of the postmigratory VMH complex, helpful for deeper functional studies of this basal hypothalamic entity. Keywords Ventromedial hypothalamic nucleus · Tuberal hypothalamus · Acroterminal · Tangential migration · Prosomeric model · Nkx2.2 Abbreviations DMH-T Terminal part of the dorsomedial hypotha- ABasM Medial (acroterminal) part of the anterobasal lamic nucleus nucleusDV Dorsoventral dimension ABasW Wing (terminal) part of the anterobasal FP Floor plate nucleus hp1 Hypothalamo-telencephalic prosomere 1 A/B Alar/basal limit hp2 Hypothalamo-telencephalic prosomere 2 AP Anteroposterior dimensionHyA Hypothalamo-amygdalar corridor Arc Arcuate nucleusIF Immunofluorescence AT Acroterminal regionIHC Immunohistochemistry Di Diencephalon ISH In situ hybridization DMH-P Peduncular part of the dorsomedial hypotha-M Mamillary area lamic nucleusME Medial eminence NHy Neurohypophysis OCh Optic chiasma * Luis Puelles p2 Prosomere 2 puelles@um.es p3 Prosomere 3 1PHy Peduncular hypothalamus University of Murcia, IMIB-Arrixaca Institute of Biomedical PM Perimamillary area Research, El Palmar, 30120 Murcia, Spain Vol.:(0123456789) 1 3 Brain Structure and Function PoA Preoptic area variety of much studied physiologic functions, which include PPa Peduncular paraventricular area metabolic regulation (Hetherington and Ranson 1942; PPaV Peduncular ventral paraventricular nucleus Frohman et al. 1974; Elmquist et al. 1999; Dhillon et al. PRM Periretromamillary area 2006; Kim et al. 2011; Meek et al. 2016), and reproductive PSPa Peduncular subparaventricular area (Pfaff and Sakuma 1979a, 1979b; Correa et al. 2015; Hashi- PTh Prethalamus kawa et al. 2017; Lewis et al. 2022), or aggressive behaviors RM Retromamillary area (Yang et al. 2013; Wang et al. 2015; Kennedy et al. 2020; RTuD Dorsal retrotuberal area Hashikawa et al. 2017; Lewis et al. 2022). There is a recent RTuI Intermediate retrotuberal area review by Khodai and Luckman (2021). Conventionally, this RTuV Ventral retrotuberal area structure is subdivided in 2–3 parts regarding its aggregation SCh Suprachiasmatic nucleus patterns, neuronal morphology, molecular phenotype, and Subl Subliminal band birth dating, normally described topographically as dorso- Tel Telencephalon medial, central, and ventrolateral VMH formations as seen Th Thalamus in coronal sections interpreted as cross-sections within the THy Terminal hypothalamus columnar model (Gurdjian 1927; Krieg 1932; Shimada and TPa Terminal paraventricular area Nakamura 1973; Altman and Bayer 1978, 1986; McClel- TSPa Terminal subparaventricular area lan 2006; Kim et al. 2019; van Veen et al. 2020). Introduc- TuD Dorsal tuberal area tion of the updated prosomeric model of the hypothalamus TuI Intermediate tuberal area by Puelles et al. (2012) provisionally did not change these TuV Ventral tuberal area terms, to avoid confusion, but they are clearly incongru- TSbO Tuberal suboptic nucleus ent with the differently oriented prosomeric forebrain axis VMH Ventromedial hypothalamic nucleus (e.g., the ‘dorsomedial’ part would be described rather as VMHdc Ventromedial hypothalamic nucleus, dorso- ‘caudomedial’, and the ‘ventrolateral’ part as ‘rostrolateral’). caudal part Though this conventional schema is habitually visualized in VMHdrl Ventromedial hypothalamic nucleus rostrolat- a single coronal section plane (i.e., is bidimensional), Van eral dorsal part Houten and Brawer (1978), also thinking in columnar terms, VMHdrm Ventromedial hypothalamic nucleus rostrome- contemplated in addition the ‘anteroposterior’ dimension dial dorsal part (equivalent to the dorsoventral axis in prosomeric terms). VMHil Ventromedial hypothalamic nucleus, interme- They identified “anterior”, “middle”, and “posterior” differ - dio-lateral part ences within the dorsomedial and ventrolateral VMH parti- VMHim Ventromedial hypothalamic nucleus, interme- tions studied in coronal sections which correspond to the dio-medial part prosomeric dorsoventral topologic differences presented in VMHvi Ventromedial hypothalamic nucleus, ventro- the present report (see also Table 3). intermediate part As regards connections, the VMH nucleus sensu lato VMHvl Ventromedial hypothalamic nucleus, ventro- is connected reciprocally to the amygdala, septum, preop- lateral part tic area, paraventricular and anterior (subparaventricular) VMHvm Ventromedial hypothalamic nucleus, ventro- hypothalamus, contralateral VMH, tuberal, dorsomedial, medial part ventral and dorsal premamillary, medial mamillary and ret- romamillary nuclei, prethalamic zona incerta, paraventricu- lar and parataenial thalamic nuclei, tegmental ventral area, Introduction periaqueductal gray and raphe nuclei (Saper et al. 1976; Canteras et al. 1994; Saper and Lowell 2014; Shimogawa The classic hypothalamic ventromedial nucleus (VMH) is et al. 2015). Many of these areas are differentially inner - one of the largest structures in the entire hypothalamus. vated by the diverse VMH subdivisions. For instance, fore- It was identified within the tuberal area as the “principal brain structures regulating the steroid hormonal signaling nucleus” by Ramón y Cajal (1911). This nucleus contains system, including the medial preoptic, tuberal and ventral largely glutamatergic neurons (Ziegler et al. 2002; Puelles premamillary nuclei, receive inputs mainly from the classic et al. 2012) that form several cell masses aggregated in an ventrolateral VMH subdivision (Canteras et al. 1994). More- ovoid block of the tuberal medial hypothalamic stratum sur- over, the different VMH parts seem associated to different rounded and delimited by a shell of afferent amygdalar inputs functions. Dorsomedial VMH and central/core VMH were (Krieg 1932; Heimer and Nauta 1969). The shell shows non- linked to metabolic circuitry (Kim et al. 2011; Meek et al. glutamatergic cell types, in part migrated from the overly- 2016), whereas ventrolateral VMH controls reproductive ing alar plate (Diaz et al. 2015). The VMH is involved in a 1 3 Brain Structure and Function and aggressive behavior (Lee et al. 2014; Lin et al. 2011; the acroterminal region of the hypothalamus (Puelles et al. Hashikawa et al. 2017; Lewis et al. 2022). 2012). These antecedents, taken jointly with some molecular In this work we have examined the origin of molecu- developmental aspects reviewed in Puelles et al. (2012), sug- larly defined cells populating different VMH subdivisions gest that the VMH subdivisions may relate to subtly different based on the detailed topologic developmental map of dif- origins of these neuronal subpopulations, with some relevant ferentially specified progenitor domains available within the progenitor domains possibly lying dorsally or rostrally to updated prosomeric model (Puelles et al. 2012; Puelles and the place where the VMH nucleus develops (descriptors Rubenstein 2015; Puelles 2018, 2019). Our material includes according to the columnar paradigm; see Fig. 1). The global sagittal, horizontal, and transversal sections oriented accord- cytoarchitectonic boundary that delimits the VMH complex ing to the ‘natural’ prosomeric axis (defined as parallel to may be due largely to the VMH shell plexus formed around the alar-basal boundary, the floorplate, and the precociously it. Most studies addressing the development of the VMH underlying notochord; Fig. 1). We complemented these data have used the columnar model of the brain as a morphologic with some tracing experiments in organotypic cultures of reference (Herrick 1910, 1933; Kuhlenbeck 1973; Alvarez- embryonic hypothalamus to demonstrate the reality of dors- Bolado et al. 1995), and essentially employed coronal sec- oventral migratory displacements predicted by Puelles et al. tions through the nucleus, which was assumed to develop in (2012). We indeed found that different VMH subpopula- its adult position. tions are born in different basal tuberal histogenetic pro- Altman and Bayer (1986), in their autoradiographic neu- genitor areas, either coinciding with the final VMH locus rogenetic study of the rat hypothalamus, identified the vent- (only radial migration involved) or placed dorsal and/or rolateral VMH as the earliest generated region, followed by rostral to the VMH ventricular zone, thus implying signifi- the dorsomedial VMH (called ‘dorsalis’ by these authors), cant short-range tangential migrations. Some components whereas their ‘VMH pars basalis’ (possibly referring to the seem to originate in the overlying alar plate. In particular, local periventricular stratum) is the last to become postmi- we provide additional evidence for the tangential dorsoven- totic, in a sequence ranging between E13 and E17. However, tral migration of the Nkx2.2-expressing VMH cell popula- these authors did not identify precisely the neuroepithelial tions, complementing the material previously commented origin of the VMH subdivisions and apparently assumed by Puelles et al. (2012). a local radial origin for all of them. A previous autoradio- graphic study of Shimada and Nakamura (1973) reported the birthdate interval for VMH neurons in the mouse Material and methods between E10-E14, but only vaguely ascribed their origin to the underlying neuroepithelium. Ulterior studies described Allen atlas brain database some radial migratory cell movements in cultured coronal VMH slices (Dellovade et al. 2000, 2001; McClellan et al. We selected several significative gene expression images 2006, 2008). Only Puelles et al. (2012; pp 285–287) seem from the Allen Developing Mouse Brain Atlas (https:// to have considered the possibility of tangential migrations devel oping mouse. br ain- map. org/). Whereas we analyzed being involved in the development of this nuclear complex. all mouse material available in this database for each gene Some molecular markers specific of different VMH selected, we chose images from E13.5, E15.5, E18.5, P1 regions have been reported (Kurrasch et al. 2007). Recently and P4 to build the figures. Some of then combine Allen cell line reporter studies (for Shh, Gli, Neurog2, Ascl1) fol- Atlas material join with our lab material. We interpreted lowed partially some VMH subpopulations and identified the Atlas coronal section planes as horizontal (i.e., parallel the positions they occupy in the adult VMH (Corman et al. to the hypothalamic alar-basal boundary shown in Fig. 1b 2018; Aslanpour et al. 2020a, b). Nevertheless, confusion (A/B limit). persists, unfortunately, since the widely used coronal section plane is usually understood within the columnar model as Animals demonstrating transversal relationships. In contrast, conven- tional coronal sections are roughly horizontal in the proso- We studied mouse specimens from several stages of devel- meric model due to the latter’s different axial references opment processed for in situ hybridization or immunohis- (e.g., the alar-basal boundary and the floor plate) ending in tochemistry techniques: E12.5 (n = 2), E13.5 (n = 1), E14.5 (n = 8), E16.5 (n = 5), E18.5 (n = 5), adults (n = 2). We used 1 3 Brain Structure and Function E12.5 mice for migration assays (n = 6, see section 2.5). The (PBS) at 4 °C. After washing, they were embedded in 4% morning in which a vaginal plug was detected was consid- agarose in PBS for sectioning. Vibratome sections were ered as E0.5 in all embryos. Pregnant females were sacri- obtained 100 μm-thick for ISH or ISH followed by DAB- ficed by cervical dislocation after inhalation of isofluorane, immunohistochemistry, or 50 mμ-thick for immunoreactions. and then the embryos were extracted. Embryonic brains were dissected out after anesthesia on ice followed by decapita- Immunohistochemistry tion. For adult animals, after standard sodium pentobarbital anesthesia, the mice were perfused with 4% paraformalde- We performed free floating immunostaining of vibratome hyde. The brains were dissected out and fixed overnight in sections. For immunofluorescence reaction, sections were 4% paraformaldehyde in pH 7.4 phosphate-buffered saline washed in PBS-T (PBS-0.3% Triton X-100), blocked (3% 1 3 Brain Structure and Function ◂Fig. 1 Comparative schemata of the hypothalamus representing: a, b intermediate area, TuI tuberal intermediate area, NHy neurohypophy- the prosomeric model and c the columnar model. A panoramic view of sis, ME medial eminence, DMH-P peduncular part of the dorsomedial the position of the hypothalamus in the forebrain is shown in (a). a, b hypothalamic nucleus, DMH-T terminal part of the dorsomedial hypotha- The prosomeric hypothalamus is divided into peduncular and termi- lamic nucleus, VMH ventromedial hypothalamic nucleus, RTuD (main or nal portions (PHy; THy) corresponding to hypothalamo-telencephalic subl.), retrotuberal dorsal area (main or subliminal); TuD (main or subl.), prosomeres hp1 and hp2. In these schemata the yellow line indicates tuberal dorsal area (main or subliminal); AT acroterminal area, A/B limit, the dorsal longitudinal limit between hypothalamus and telencepha- alar/basal limit; PSPa peduncular subparaventricular area, TSPa terminal lon, whereas the retromamillary (RM) and mamillary (M) areas contact subparaventricular area, OCh optic chiasma, PPa peduncular paraven- ventrally the hypothalamic floorplate (FP), a primary axial longitudinal tricular area, TPa terminal paraventricular area landmark induced by the notochord (not shown). The diencephalic and telencephalic roofplate (RP), another primary longitudinal landmark, ends rostrally over the preoptic region (PoA). An intermediate light blue longitudinal line identified as A/B in a , b represents the postulated alar/ BSA in PBS-T, 1–3 h), and incubated in the primary anti- basal boundary, parallel to both FP and RP; it is held to result from dor- body solution (diluted in 3% BSA in PBS-T, 48 h, 4 ℃). soventral patterning antagonism between ventralizing floorplate signals Following incubation and several PBT washes, the sections and dorsalizing roofplate signals (it is thus a secondary axial landmark shared by all brain parts and supported by many gene expression pat- were incubated 2 h with the respective u fl orochrome-labeled terns). The rostromedian aspect of the entire hypothalamo-preoptic secondary antibodies, either Alexa 488 donkey anti-rabbit region represents the singular acroterminal domain (AT in b), which or donkey Alexa 594 anti-mouse (ThermoFisher; 1:200, marks from floor to roof the topologic rostral end of the neural tube. 2 h). For DAB-immunohistochemistry, vibratome sections These consistent and causally fundamented axial landmarks justify the spatial orientations provided [R(ostral), C(audal), D(orsal), V(entral)], were washed in PBS, and then treated with 0.1% hydro- such that the diencephalon proper lies caudal to the hypothalamus and gen peroxide in PBS for 30 min, in the dark and at room the telencephalic vesicle is a dorsal outgrowth of the hypothalamus. The temperature, to inactivate endogenous peroxidase activity. eye evaginates out of the alar AT domain (see optic chiasma –OCh– in After standard PBS-T washes, and the blocking step (3% b). b This more detailed schema displays hypothalamic structure divided in two transverse neuromeres (hp1/PHy; hp2/THy) extending dorsal- BSA in PBS-T, 1–3 h), the floating sections were incubated ward from the floorplate (FP) up to the hypothalamo-telencephalic limit with the primary antibody for 48 h at 4 ℃. After PBS-T (HTL, marked by a thick yellow line). A light blue line tagged ‘A/B washes we applied a biotinylated goat anti-rabbit or anti- limit’ marks the longitudinal alar-basal limit which separates the alar mouse secondary antibody (1:200, 2 h at room temperature; and basal hypothalamic progenitor areas. There is a definite AP and DV pattern held to be causally significant. The progenitor domains identi- Vector Laboratories, Burlingame, CA, United States), fol- fied on the basis of differential expression of transcription factors and lowed by a streptavidin/horseradish peroxidase (HRP) com- other molecular markers (Puelles et al. 2012) are thus organized rostro- plex (1:200, 2 h; Vectastain-ABC kit; Vector Laboratories, caudally (relative to the diencephalon, hp1/PHy and hp2/THy) and dor- Burlingame, CA, United States). Histochemical detection of soventrally (relative to FP, A/B, HTL, and the telencephalic RP). The rostromedian section of the neural tube where right and left brain halves the peroxidase activity was carried out using 0.03% diam- meet is identified as the acroterminal sector (AT). There is a dorsoventral inobenzidine (DAB) and 0.005% H O Primary antibod- 2 2. pattern of seven longitudinal progenitor areas (2 alar and 5 basal), each ies were used as follows: mouse anti-Nkx2.2 (1.50; DSHB, one with its peduncular and terminal part (see list of Abbreviations for Ref. 745A5-s), rabbit anti-Couptf1/Nr2f1 (1.200; Abcam, the respective areal names). We add some secondary subdivisions such as the liminal PSPa/TSPa alar band and the subliminal RTu/Tu basal Ref. ab96846), rabbit anti-Nkx2.1 (1:200; Sigma Aldrich, band, these being concepts used in the text. The ventromedial nucleus Ref. SAB3500757), rabbit anti-Isl1 (1:200; Abcam, Ref. (VMH) is placed in its adult position with its neighbor, the dorsomedial ab20670), rabbit anti- Otp (1: 200; F. Vaccarino), rabbit anti nucleus complex (VMH; DMH-P; DMH-T; within the green RTuI/TuI TH (1:200; Bio-Techne R&D Systems, Ref. NB300-109). domain). c According to both classical and recent studies the columnar model does not separate PHy from THy and defines its axial or longitu- dinal dimension as roughly parallel to the prosomeric AT domain, some- In situ hybridization how implying an extension of the brainstem and midbrain axis across the diencephalon and the hypothalamus into a telencephalic end (not We used the restriction enzymes and polymerases suitable stipulated precisely). In general, the entire columnar hypothalamus is conceived recently as a floor and basal region of the diencephalon, sub- for specific riboprobe synthesis in the presence of digoxi- divided by transverse planes into preoptic, anterior, intermediate (tuberal) genin- 11-UTP. The hybridization protocol used was accord- and posterior (mamillary) regions (Swanson 2012); note the prosomeric ing to Shimamura et al. (1994). Mouse cDNA probes used AT domain is conceived rather as prechordal floor. This model classi- for in situ hybridization were Nxk2.2 and Nr5a1 (J.R. Ruben- fies the whole prosomeric alar hypothalamus as an Anterior hypotha- lamic region (containing both paraventricular and anterior hypothalamic stein), Otp (A. Simeone), Six3 (P. Bovolenta), and Satb2 (our nuclei). The prosomeric basal hypothalamus corresponds to the tuberal own lab, NCBI accession number NM_001358580). and mamillary parts of the columnar hypothalamus. The spatial orienta- tions in this model are rotated 90 degrees relative to the prosomeric ones Organotypic cultures (compare b with c) R rostral, C caudal, D dorsal, V ventral, Di dienceph- alon, Tel telencephalon, hp1 hypothalamic prosomere 1, hp2 hypotha- lamic prosomere 2, FP floor plate, PoA preoptic area, PHy peduncular Brains of embryos dissected from skin and other append- hypothalamus, THy terminal hypothalamus, RM retromamillary area, M ages at E12.5 were collected in artificial cerebrospinal mamillary area, PRM periretromamillary area, PM perimamillary area, fluid (ACSF) solution at pH 7.4 containing: 4 mM KCl, RTuV retrotuberal ventral area, TuV tuberal ventral area, RTuI retrotuberal 1 3 Brain Structure and Function 1.5 CaCl2, 0.75 mM MgCl2, 129 mM NaCl, and 10 mM (Puelles et al. 2012; Puelles 2019), which momentarily D-glucose. We dissected the tissue with dissection tweezers forces translation of both unconciliable terminologies (see discarding meninges and telencephalic vesicles, and opened our Table 3) under the assumption that causal explana- the neural tube along the midline. We placed separately the tions will only emerge from the prosomeric notions. The two brain halves upon membrane culture inserts (Millicell prosomeric hypothalamus consists of two prosomeres, Millipore, 0.4 mm, PICM0RG50) within small Petri dishes, hp2 and hp1, which represent its structure transversal to with the ventricular surface up, contacting the air, and the the axis of the forebrain (Fig. 1; the prosomeric axis is pial surface touching across the membrane a substrate of co-defined by the notochord, the floorplate, the alar-basal MEM-supplemented medium (1% Penicilin/Streptomycine, boundary and the roofplate as mutually parallel longitudi- 0.065% glucose, 0.5% glutamine, and 1% inactivated fetal nal reference landmarks lying at different dorsoventral lev - bovine serum). Explants were acclimatized for 1 h (37 ℃, els; Puelles 2018; Amat et al. 2022; the optic tract may be 5% CO2), and subsequently marked through the ventricular taken as another such reference; Puelles 2022). These two surface with a CMFDA-coated tungsten particle borne on units are both hypothalamo-telencephalic in spatial range a sharpened tungsten needle (Alifragis et al. 2002; López and contain respectively the terminal hypothalamus which González et al. 2021), testing diverse labeling loci along extends into the non-evaginated preoptic telencephalic the estimated alar-basal boundary of the hypothalamus (i.e., subpallium (THy in hp2; Fig. 1a, b) and the peduncular varying the dorsoventral position relative to this limit, and hypothalamus that expands into the rest of the telencepha- changing also the anteroposterior position along the THy, lon (PHy in hp1; Fig. 1a, b; the telencephalon is strictly including its rostromedian acroterminal domain). After two a hypothalamic bilateral dorsal evagination, being thus days in culture conditions (37 ℃, 5% CO2), the explants epihypothalamic and the Hypo-Thalamus is a wrong term, were fixed with cold paraformaldehyde 4% in PBS for because this brain region lies rostral rather than ventral to 10 min. To check the position of the CMFDA particle, as the thalamus and diencephalon proper). THy also includes well as the labeled cells in the mantle layer, all explants in its acroterminal alar part the eye vesicle and eye stalk were processed for immunofluorescence with the mouse region. PHy is continuous dorsally through the interven- anti-Nkx2.2 antibody (1.50; DSHB, Ref. 745A5-s). tricular foramen with the whole evaginated telencephalic vesicle (Puelles et al. 2012; Puelles and Rubenstein 2015). Image analysis The right and left hypothalamic sides meet rostrally along the acroterminal area (AT) of THy, which is a singularly We scanned the ISH and ISH/IHC images at high resolu- differentiated median transverse territory displaying dors- tion with the Aperio ImageScope software (Leica Biosys- oventral alar and basal structural specializations (terminal tems). Immunofluorescent (IF) images were obtained from lamina, optic chiasma, anterobasal area, arcuate/median the sectioned brains and the whole-mounted explants using eminence area, infundibulum and neurohypophysis, tub- a confocal SP8 Leica microscope. Individual optic sections eromamillary, and mamillary subregions; Puelles et al. were 3 µm apart, and image stacks of various Z sizes were 2012; Puelles and Rubenstein 2015). The hypothalamic generated according to the structures of interest. All figures dorsoventral zonal pattern (alar and basal subdomains) is from the Allen Developing Mouse Brain Atlas (https://de vel continuous caudalwards with the equivalent diencephalic oping mouse. brain- map. org/) were 180º rotated (nose at the pattern (Puelles et al. 2012; Puelles and Rubenstein 2015; right side) to have the same orientation than in our own Díaz et al. 2015; Ferran et al. 2015; López-González et al. lab images. Figures were constructed using ImageJ, Adobe 2021), with shared tagmatic properties down to the isthmo- Photoshop and Adobe Illustrator software. mesencephalic boundary (Puelles 2018). The evaginated telencephalon as well as the essentially acroterminal eye are thus held to be alar derivatives of the hypothalamus, Results evaginated within hp1 and hp2, respectively. Hypothalamic Shh expression induced by the underlying notochord ini- Background introduction to the prosomeric tially extends to its floor and basal plates (limited by the hypothalamus alar-basal boundary) but is later lost at the acroterminal infundibular part of the intermediate tuberal basal plate The neurodevelopmental field is performing gradually a area (TuI), a secondary effect due to adenohypophysial paradigm change from the old columnar model (which has Tbx3 signals (Trowe et al. 2013). Regionalization is dem- proven to have difficulties assimilating and explaining gene onstrated further through gene expression patterns charac- expression patterns) and the modern prosomeric model. teristic of the mentioned structural divisions (e.g., Ntn1, Given their fundamental discrepancy on the length axis of Lmx1b, Lmx1a, Foxa1 are markers present in the hypo- the forebrain, there is a terminological problem associated thalamic floor plate; Six3, Fgf10, Fgf8, Dlk1 appear in the 1 3 Brain Structure and Function acroterminal area; Nkx2.2 is a linear longitudinal marker, subdomains, called dorsal, intermediate, and ventral (e.g., expressed in a band along the forebrain alar-basal limit and TuD, TuI, TuV within hp2; similar for the RTu subdivi- overlapping slightly both alar and basal domains; the alar sions within hp1; Puelles et al. 2012). In the present report overlap domain is known as the ‘liminal band’, meaning we are concerned essentially with the TuD and TuI hp2 the rim of the alar plate, whereas the basal overlap domain basal longitudinal zones, which relate to the VMH nucleus. is known as the ‘subliminal band’; Puelles et al. 2012; It should be noted that TuD is itself divided into the dorsal Fig. 1a, b). Detailed genoarchitectonic dorsoventral pattern Nkx2.2-positive subliminal band (the area of basal overlap in the hypothalamus allows to distinguish four alar and five of the Nkx2.2 band) and the underlying Nkx2.2 negative basal domains that extend longitudinally across both THy ‘main part’ of TuD (Puelles et al. 2012; their Fig. 8.14B). (hp2) and PHy (hp1) (see Fig. 1a, b, its legend, and the abbreviations list). The paraventricular (Pa) and subpara- Borders of the VMH ventricular (SPa) areas are distinct superposed alar plate domains, where Pa has three dorsoventral subdivisions (not The VMH nucleus occupies a large dorsal area within the relevant here). There is a major tuberal/retrotuberal basal TuI region of the terminal hypothalamus, caudal to the longitudinal complex (Tu/RTu) across hp2 and hp1 and acroterminal arcuate nucleus domain (Arc) and dorsal to under the alar-basal boundary; it lies dorsal to underlying the terminal part of the dorsomedial nucleus (DM-T); the parallel perimamillary/periretromamillary (PM/PRM) and overlying TuD region contains the wings of the anteroba- mamillary/retromamillary (M/RM) longitudinal areal com- sal nucleus, whose rostromedian fused part is acroterminal plexes. The large Tu/RTu subdivides dorsoventrally into 3 (ABasM; ABasW; Fig. 1b; Puelles et al. 2012). The sharp Fig. 2 The tuberal/retrotuberal Isl1 marker delimits negatively the while restricted to the VMH primordium, does not fill completely all whole VMH primordium early during development. This panel shows its volume; note particularly a ventral (VMHvm) triangular sector of sagittal sections of the basal hypothalamus at E14.5 a–c and E16.5 the VMH which is also delimited by Isl1 cells, but devoid of Nkx2.2 d, f labeled with double immunofluorescence for Nkx2.2, and Isl1; cells. Scale bars represent 200 µm. TuI tuberal intermediate, RTuD the c, f panels compare the separated red channel a, d and green retrotuberal dorsal, TuD tuberal dorsal, VMHim medial-intermediate channel b, e images in adjacent sections. The basal areas surround- VMH subnucleus; VPa ventral paraventricular nucleus, PSPa pedun- ing (delimiting) the VMH primordium, including the TuI and TuD cular subparaventricular area, VMHvm ventral-medial VMH subnu- areas, are marked by intense Isl1 labeling (green), with few Isl1 cells cleus, Arc arcuate nucleus, VMHdrm dorsal-rostromedial VMH sub- dispersed within the primordium. In contrast, Nkx2.2 labeling (red), nucleus 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 3 Adult distribution of Nkx2.2-positive cells in the VMH. a–g and lateral dorsorostral subnuclei (VMHdrm, VMHdrl; Caudorostral series of transverse sections orthogonal to the proso- Fig. 3b–e, h). We also subdivided the intermediate VMH meric forebrain axis taken through the adult terminal hypothalamus part (classic ‘central’ part) into medial and lateral-interme- to examine systematically the VMH nucleus, and double-reacted for diate subnuclei (VMHim, VMHil; Fig. 3b–h). Finally, the Nkx2.2 ISH (blue) and TH IHC (brown). The Nkx2.2-positive cells are distributed differentially over diverse identified VMH subdivi- large ventral VMH region (old ‘ventrolateral part’) was sub- sions. a Nkx2.2 signal is absent at the caudalmost dorsal VMH level, divided into three parts, the medial, intermediate, and lateral where we distinguish the VMHdc. (b–f) Progressing rostralward in ventral subnuclei (VMHvm, VMHvi, VMHvl; Fig. 3c–h; the series, Nkx2.2 signal is strongly expressed laterally at VMHdrl note the equivalences with the older terminology obviously and VMHil, and extends into VMHvi; there is also weaker expres- sion at the VMHdrm and VMHim. In contrast, VMHvm, and VMHvl are not exact; see Table 3). are largely devoid of Nkx2.2 signal. g The rostral-most level of the VMH is also largely Nkx2.2 negative and can probably be interpreted Distribution of gene expression as VMHim and VMHvm. h Sagittal section of the adult medial hypo- during development thalamus marked with Nkx2.1 IHC in brown and Otp ISH in blue (from Puelles et al. 2012). Red lines represent the section levels in a–g; see also caudal/C and rostral/R orientations. Scale bars represent We searched genes expressed in mouse TuD and TuI at 200 µm. PoA preoptic area, RPa retroparaventricular area, AH ante- E13.5 and E18.5 to characterize the expected differential rior hypothalamic nucleus; A/B, alar/basal limit, VMHdc dorsocau- molecular profile of the VMH, using the AGEA tool of the dal subnucleus of the ventromedial hypothalamic nucleus, DMH-T terminal part of the dorsomedial hypothalamic nucleus, VPM ventral Allen Developing Mouse Brain Atlas database (https://de vel premamillary nucleus, DPM dorsal premamillary nucleus, M mamil- opingm ouse.b rain-m ap.o rg). Tables 1, 2 classify 22 relevant lary area, VMHdrm dorsal-rostromedial subnucleus of the ventrome- genes according to their characteristic expression domains at dial hypothalamic nucleus, VMHdrl dorsal-rostrolateral subnucleus of E13.5 and E18.5, respectively. Representative markers are the ventromedial hypothalamic nucleus, VMHim medial-intermediate subnucleus of the ventromedial hypothalamic nucleus, VMHil lateral- shown in Fig. 5, with the respective tridimensional signal intermediate VMH subnucleu, SCh suprachiasmatic nucleus, VMHvm distribution presented in medio-laterally ordered sagittal sec- ventral-medial VMH subnucleus, VMHvl ventral-lateral VMH sub- tion planes taken from the Allen Developing Mouse Brain nucleus, VMHvi ventral-intermediate VMH subnucleus, ac anterior Atlas. Although there clearly are areas of overlap between comisure, DMH-P peduncular part of the dorsomedial hypothalamic nucleus, PRM periretromamillary area, Th thalamus, PTh prethala- the diverse markers, a dorsoventral, rostrocaudal and medi- mus olateral pattern can be distinguished within the prosomeric model that underpins the detailed topographic subdivision boundary of the VMH proper can be assessed (apart from we propose (VMHdc/VMHdrm/VMHdrl; VMHim/VMHil; other possibilities) by its lack of Isl1 signal, in contrast with VMHvm/VMHvi/VMHvl). the Isl1-positive rest of the tuberal region (Fig. 2). Impor- It is possible to cluster the markers studied at E13.5 into tantly, some VMH neuronal subpopulations seem to have five positionally distinct subgroups (alar, only TuD, only non-TuI origins. They share early molecular markers with TuI, TuD + TuI, only acroterminal; Table 1). These results cell populations of the overlying TuD region (either its sub- can be compared with the respective VMH expression pat- liminal or main parts), or, alternatively, of neighboring TuD/ tern at E18.5 (Table 2). TuI acroterminal areas, or even of the overlying alar TSPa domain, where some VMH cells apparently originate, as 1) The Vax1 gene appears expressed selectively at E13.5 we will illustrate below. Puelles et al. (2012) previously along the longitudinal subparaventricular alar domain suggested a dorsoventral migration of cells expressing the (both THy and PHy; Fig. 4w, x; note this domain cor- Nkx2.2 marker into the VMH primordium. responds to the ‘hypothalamic diagonal’ of Shimogori Our material corroborates the already previously recog- et al. 2010 but is conceived here as strictly longitudinal; nized inner heterogeneity of VMH. Dorsomedial, central, check Fig. 1); at E18.6 a Vax1-positive subpopulation and ventrolateral VMH parts are conventionally described was identified within VMHdrm/VMHdrl (Fig. 5k). in the literature, but we prefer to characterize them as dor- 2) 4 genes -Sema3a, Nkx2.2, Fbxw7, and Tcf7l2- were sal, intermediate, and ventral regions, respectively (consist- selectively expressed at E13.5 at the TuD zone (implying ently with the prosomeric axial dimension; see Fig. 1). We separately either the main or subliminal parts of TuD, or found that a slightly more detailed subdivision was needed encompassing both of them, as well as the correspond- to describe fully the observable molecular diversity. We sub- ing TuD acroterminal portion; Fig. 4a, b, e, f, m, n). The divided the dorsal part of VMH into a dorsocaudal element same markers subsequently label at E18.5 mainly our (VMHdc; this is the old ‘dorsomedial’ portion; Fig. 3a, h; dorsal VMH subdivisions (VMHdc/VMHdrm/VMH- see Table 3) and two novel components called by us medial 1 3 Brain Structure and Function Table 1 Expression of 22 SPa (Alar) Subl.TuD Main TuD TuI TuD-AT TuI-AT gene markers at the terminal neurogenic regions of the Vax1 + hypothalamus near the VMH Sema3a + primordium at E13.5 Nkx2.2 + + + Fbxw7 + + Tcf7l2 + + Nr2f1 + Lmo4 + Satb2 + Nr5a1 + + + Adcyap1 + + + Robo2 + + + Calb1 + + Slc17a6 + + Cnr1 + + + Enc1 + + + Sox14 + + + Bcl11a + + + Dner + + + Chl1 + + Mapt + + + + Nkx2.1 + + + + Six3 + + SPa subparaventricular area, Subl.TuD subliminal tuberal dorsal area, Main TuD main tuberal dorsal area, TuI tuberal intermediate area, TuD-AT acroterminal section of the tuberal dorsal area, TuI-AT acroterminal section of the tuberal intermediate area drl) but also extend variously into the intermediate ones in the VMHdc, VMHdrm, and VMHvl, with some vari- (VMHim/VMHil) (Fig. 5b, f). able extra locations. 3) Three genes -Nr2f1, Lmo4, Satb2- appeared expressed 5) Six3 was initially expressed at E13.5 at the acroterminal exclusively at the TuI domain at E13.5 or earlier tuberal region (as well as in alar acroterminal regions; (Fig. 4k, l, s, t); these labeled subsequently our interme- Fig. 4i, j; note this gene labels the prospective acroter- diate and ventral VMH subdivisions at E18.5 (VMHim/ minal domain already from neural plate stages onwards; il, VMHvm/vl; Tables 2; Fig. 5e, i). Two of these mark- Lagutin et al. 2003), with some extension into TuI. Dis- ers included also the VMHdc subdivision (old dorsome- tinct Six3-labeling was displayed later at E18.5 by the dial part). novel VMHvi subnucleus (Fig. 5d; note this distinct 4) 12 genes in Table 1 are expressed both within TuD subnucleus was never described before within the clas- and TuI domains at E13.5 (Fig. 4c, d, g, h, q, r, o, p, sic ‘ventrolateral’ sector). These Six3-positive cells are u, v). They can we regrouped according to the location interpreted to migrate from the acroterminal TuD area at E18.5 of the corresponding labeled cell populations and/or the arcuate TuI acroterminal area, which also (Fig. 5a, c, g, h, j): Nr5a1 and Robo2 have a similar expresses initially Six3. E18.5 pattern, labeling only the VMHdc, VMHdrm, and VMHim subnuclei. Calb1 and Slc17a6 share labeling within VMHdrm, VMHim, and VMHil at E18.5 (though There clearly exist areas of overlap between most of the Slc17a6 signal also extends into VMHdc). Several genes markers studied, indicating that most VMH subdivisions -Enc1, Bcl11a, Dner, Chl1- share a strong presence at are subtly heterogeneous in molecular profile, with regional the VMHvm and VMHvl subdivisions, though their sig- variations. This is consistent with the molecular diversity nals variously spread also into other neighboring subnu- observed in transcriptomic studies (see Discussion). The clei. The genes Cnr1, Mapt, and Nkx2.1 share expression VMH nevertheless shows on the whole significant partial 1 3 Brain Structure and Function Table 2 Expression of 22 gene VMHdc VMHdrm VMHdrl VMHim VMHil VMHvm VMHvl VMHvi markers within the different VM subnuclei at E18.5 Vax1 + + Sema3a + + Nkx2.2 + + + + + Fbxw7 + + Tcf7l2 + Nr2f1 + + + + Lmo4 + + + Satb2 + + + + Nr5a1 + + + Adcyap1 + + + + Robo2 + + + Calb1 + + + Slc17a6 + + + + Cnr1 + + + + Enc1 + + + Sox14 + + + + Bcl11a + + + + Dner + + + + + Chl1 + + + + Mapt + + + + + Nkx2.1 + + + + + + Six3 + + + VMHdc dorsocaudal VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHdrl dorsal- rostrolateral VMH subnucleus, VMHim medial-intermediate VMH subnucleus, VMHil lateral-intermediate VMH subnucleus, VMHvm ventral-medial VMH subnucleus, VMHvi ventral-intermediate VMH subnu- cleus, VMHvl ventral-lateral VMH subnucleus dorsoventral, mediolateral, and rostrocaudal sorting of its associated to the TuI progenitor area—cases of Satb2, molecularly distinct subpopulations. These may be ten- Nr2f1, and Nkx2.1- imply a priori a radial migration pattern, tatively classified according to their apparent positional whereas origins associated to the overlying alar TSPa or to neuroepithelial origins, tangential versus radial migration the TuD—cases of Vax1, Nkx2.2, Nr5a, Tcfl2, and Sox14- routes, and differential molecular profile. The classic schema suggest tangential dorsoventral displacements. At least in of three VMH parts (dorsomedial, central, ventrolateral) is the case of Six3 it is possible to study separately a restricted too simple to account for the level of heterogeneity observed acroterminal origin, which would involve a priori a rostral (e.g., it completely misses the Six3-positive VMHvi subdivi- source and a caudally oriented tangential displacement. sion as well as the dorsorostral VMH subdivisions described here and our medial–lateral distinctions; Table 3). Vax1 Development of ventromedial nucleus Vax1 is a typical early selective marker of the alar sub- subpopulations marked with selected genes paraventricular area, which extends over the terminal and peduncular hypothalamic domains (PSPa and TSPa) and To analyze in more detail the variant development of VMH includes at its ventral rim the liminal alar Nkx2.2-positive subpopulations according to their progenitor origins, we zone (Puelles et al. 2012; in blue in Fig. 1a, b). Initially the selected for follow-up a group of genes expressed in dis- expression of Vax1 ends strictly at the alar-basal boundary. tinct progenitor areas at E13.5 whose derivatives constitute Vax1-labeled cells start to move into the subjacent basal characteristic VMH subpopulations at E18.5: Vax1, Nkx2.2, tuberal region at E13.5 (Fig. 6a, b). Two days later, this Nr5a1, Satb2, Tcfl2, Sox14, Nr2f1, Nkx2.1, and Six3. Origins marker delineates rostrodorsal parts of the VMH primordium 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 4 Differential dorsoventral expression of VMH markers at larger ventral part of the TuD/RTuD area as ‘the main TuD/ E13.5 in two neighboring sections for each marker (identified at left). RTu’ (Fig. 1b). These distinctions are not contemplated in This panel illustrates various sagittal ISH images of markers listed in the columnar tradition (Table 3). our Tables 1, 2 taken from E13.5 mice material at the Allen Devel- Results on zebrafish embryos indicate that the earliest oping Mouse Brain Atlas. The dashed black line identifies the alar- basal boundary (identifying D to the right and V to the left; R to the appearance of Nkx2.2 signal occurs at neural plate stages bottom). TuI, main TuD and subliminal TuD are delimited by solid (Hauptmann and Gerster 2000; Hauptmann et al. 2002). In lines. Other hypothalamic domains are not identified, being outside E12.5 mouse embryos Nkx2.2 is already expressed longitu- the scope of this analysis. The observed patterns vary in the degree in dinally along the alar-basal boundary of the midbrain, dien- which TuI versus TuD, or both, show expression at this stage. Scale bar in k represents 200 µm. RTuI retrotuberal intermediate area, TuI cephalon, and hypothalamus (Fig. 7a). At E13.5, Nkx2.2- tuberal intermediate area, main or subl. RTuD main or subliminal ret- expressing cells start to migrate ventralwards (mainly within rotuberal dorsal area, main or subl. TuD main or subliminal tuberal THy) and the corresponding signal begins to protrude dorsal area ventralwards under the terminal subliminal band strictly expanding within the basal tuberal region; this represents the (Fig. 6c, d; note this result changes the classic simpler con- first sign of the migration of Nkx2.2 cells into the VMH pri- cept of the dorsal region of VMH, contemplating only the mordium (primVMH; Fig. 7b). This process is clearly more caudally placed ‘dorsomedial VMH part’; obviously, our advanced at E14.5 (primVMH; Figs. 7b, c, 8a–c). Cross- ‘dorsorostral’ descriptor refers to the prosomeric axis and sections at E14.5 illustrate that Nkx2.2 signal is present at columnar authors may want to use a different descriptor). the ventricular and mantle zones of the subliminal and main At postnatal stages the tuberal expression of Vax1 coincides parts of the TuD, as well as in the underlying VMH nuclear mainly with the VMHdrm/VMHdrl subnucleus with some primordium (only mantle zone), while the TuI ventricular dispersion into VMHim/VMHil (Figs. 5k; 6d). zone deep to VMH remains negative (primVMH; Fig. 7d–g). At E18.5 there is already a definitive distribution pattern of Nkx2.2 Nkx2.2 signal within VMH (Fig. 8d–g). The dorsoventral tubero-tuberal VMH migration stops within TuI at some dis- Nkx2.2 is well known as an early longitudinal marker of the tance of the perimamillary band, leaving a substantial space alar-basal boundary along the whole forebrain tagma (mid- for the terminal part of the DMH nucleus, as well as for brain, diencephalon, and hypothalamus; note this is a basic the separately tangentially migrated VPM nucleus (López- token of the updated extended forebrain concept held within González et al. 2021). Interestingly, at E13.5-E14.5 there is the prosomeric model; Puelles et al. 2012; Puelles and also a previously unidentified shorter parallel dorsoventral Rubenstein 2015; Puelles 2018; Amat et al. 2022). This lin- Nkx2.2 migration coming out of the peduncular retrotuberal eal signal also appears bordering the spike of Shh expression subliminal band that extends ventralwards into RTu, where that marks the transverse alar zona limitans intrathalamica it stops in contact with the Otp-positive periretromamillary (alar p3/p2 boundary), also known as the mid-diencephalic band; this parallel migration is distinctly separated from the organizer (Puelles and Martínez 2013). This gene (as well VMH and may contribute Nkx2.2-expressing neurons to the as others such as Nkx2.9, Ptc, etc.) is apparently selectively dorsomedial nucleus (Fig. 7b, c). upregulated by particularly high local concentrations of We performed some in vitro fluorescent labeling experi- SHH signal diffusing dorsalward from the underlying floor ments on organotypic cultures of embryonic hypothalamus and basal plate Shh expression domain or from the related to visualize the ventralward tubero-tuberal migration into ZLI core domain which obeys a different enhancer (Briscoe VMH originated from the primary Nkx2.2-expressing band et al. 1999; Puelles and Martínez 2013; Nishi et al. 2015; across the alar-basal boundary of THy. A small particle of Andreu-Cervera et al. 2018). In the hypothalamus, Nkx2.2 CMFDA (see Methods) was placed at the TuD in E12.5 signal appears early on as a thin longitudinal band that over- explants and further analysis was done at E14.5, at which laps the lineal boundary between the alar and basal plates stage the VMH is already fairly well formed, though the and stops as it reaches the acroterminal border (Fig. 7a). migration is not complete yet. All explants were treated The mixed alar-basal expression led to the concepts of alar for Nkx2.2 immunofluorescence to check the position of liminal and basal subliminal subdivisions of the Nkx2.2 band the CMFDA particle, and in all cases (n = 6) a stream of (Puelles et al. 2012; ‘liminal’ refers to the classic notion of CMFDA-labeled cells overlapped with the Nkx2.2-positive limen lamina alaris, or rim of the alar plate; see red and blue cells advancing into TuI, showing a comparable disposition bands in Fig. 1a, b). The subliminal subdomain thus forms (n = 4 in Fig. 8h–k). Some Nkx2.2/CMFDA-positive migrat- the upper rim of the TuD/RTuD basal tuberal hypothalamic ing cells were observed relative to the labeling sites placed progenitor area, whereas the liminal area is a subdomain at across the width of the Nkx2.2-positive band (white arrows the ventral rim of the subparaventricular hypothalamic area in Fig. 8h’–k’). This reveals that some dorsoventrally migrat- across THy and PHy. We will refer to the non-subliminal ing cells do not express Nkx2.2, which possibly includes 1 3 Brain Structure and Function 1 3 Brain Structure and Function Fig. 5 Differential dorsoventral expression of VMH markers at to illustrate the relevant five mediolateral section levels available. E18.5 or P4. This Figure (in two parts) presents for each marker We tentatively identified the VMH subdivisions described in the text (identified at left) a mediolateral series of five parallel sagittal ISH (VMHdc, VMHdrm, VMHdrl, VMHim, VMHil, VMHvm, VMHvi, images of E18.5 (or P4, indicated) mice embryos from the Allen and VMHvl). Scale bars represent 200 µm. VMHim medial-inter- Developing Mouse Brain Atlas illustrating the expression pat- mediate VMH subnucleus, VMHdc dorsocaudal VMH subnucleus, tern of the markers identified in our Tables 1, 2 across VMH. The VMHdrm dorsal-rostromedial VMH subnucleus, VMHvi ventral- expression patterns are compared to the outlines of Nkx2.2 (red out- intermediate VMH subnucleus, VMHil lateral-intermediate VMH line), Nr5a1 (black outline), and Six3 (green outline) expression. It subnucleus, VMHdrl dorsal-rostrolateral VMH subnucleus seemed informative towards 3D assessment of relative topography the Nr5a1-expressing cells (see next section) and/or passing VMH primordium (Fig. 8k, k’). This result is consistent with Vax1 cells coming from the overlying alar subparaventricu- descriptive images in which Nkx2.2-expressing cells seem lar area (previous section). In one experiment in which the to originate throughout the TuD/RTuD boundary, appearing CMFDA particle was placed at the TuD/RTuD limit where it in transient continuity with the VMH primordium (Figs. 7b, crosses the orthogonal THy/PHy border migrated CMFDA- 8k). Nkx2.2 double-labeled cells were found also within the 1 3 Brain Structure and Function 1 3 Table 3 Terminological equivalence of the main nuclei and histogenetic areas contemplated in this report according to the nomenclature of the columnar and prosomeric forebrain models Columnar term Abbr Columnar histoge- Prosomeric term Nuclear Abbr Prosomeric histoge- Areal Abbr Prosomere Hypothalamic part Population origin netic area netic area related literature Supramamillary SUMm Mamillary Retromamillary RMM Retromamillary area RM hp1 PHy nucleus medial part nucleus medial part Supramamillary SUMl Mamillary Retromamillary RML Retromamillary area RM hp1 PHy nucleus lateral part nucleus lateral part Medial mamillary MM Mamillary Medial mamillary MM Mamillary area M hp2 THy nucleus Lateral mamillary LM Mamillary Lateral mamillary LM Mamillary area M hp2 THy nucleus nucleus Posterior hypothala- PHr Mamillary Periretromamillary PRM Periretromamillary PRM hp1 PHy band area mus, hypothalamic part Posterior hypothala- PHc Prerubral tegmentum Prerubral tegmentum p3Tg + p2Tg p3 and p2 basal plate p3B, p2B p3 + p2 Dienc Puelles et al. (2012) mus, diencephalic part Tuberomamillary TM Mamillary Perimamillary PM Perimamillary area PM hp2 THy nucleus + Dorsal band + Dorsal pre- DPM Retrotuberal ventral premamillary mamillary nucleus area and Tuberal nucleus ventral area Tuberomamillary his- TM Mamillary Ventral retrotuberal RTuV + TuV Ventral retrotuberal RTuV + TuV hp1 + hp2 PHy + THy taminergic neurons and tuberal areas and tuberal areas Ventral premamillary PMv Mamillary Ventral premamillary VPMc Migrated tangentially RM hp2 THy López-González et al. nucleus nucleus, core from retromamil- (2021) lary área (located Ventral premamillary VPMsh TuI) nucleus, shell Dorsomedial hypo- DMHa Tuberal Dorsomedial hypo- DMHc/s-P Intermediate retrotu- RTuI hp1 PHy Díaz et al. (2015) thalamic nucleus, thalamic nucleus, beral area anterior part peduncular part (core/shell) Dorsomedial hypo- DMHp/v Tuberal Dorsomedial hypo- DMHc-P Intermediate tuberal TuI Hp2 THy Díaz et al. (2015) thalamic nucleus, thalamic nucleus area posterior/ventral terminal part (core/ part shell) Brain Structure and Function 1 3 Table 3 (continued) Columnar term Abbr Columnar histoge- Prosomeric term Nuclear Abbr Prosomeric histoge- Areal Abbr Prosomere Hypothalamic part Population origin netic area netic area related literature Subthalamic nucleus STN Mamillary Subthalamic nucleus STh Migrated dorsalward RM hp1 PHy Skidmore et al. (2008) from retromamil- lary area Parasubthalamic PSTN Mamillary Parasubthalamic PSTh Mifrated dorsalward RM hp1 PHy nucleus nucleus from retromamil- lary area Arcuate hypothalamic ARH Tuberal Arcuate nucleus Arc Intermediate tuberal TuI hp2 THy (acroterminal) nucleus área (acroterminal part) Ventromedial hypo- VMHdm Tuberal Ventromedial hypo- VMHdc Intermediate tuberal TuI hp2 THy Present results thalamic nucleus, thalamic nucleus, area dorsomedial part dorsocaudal part Ventromedial hypo- VMHa Tuberal Ventromedial hypo- VMHdrm Dorsal tuberal TuD + SPa hp2 THy Present results thalamic nucleus, thalamic nucleus, área + Alar subpara- anterior part rostromedial dorsal ventricular area part Ventromedial hypo- VMHdrl thalamic nucleus rostrolateral dorsal part Ventromedial hypo- VMHc Tuberal Ventromedial hypo- VMHim Dorsal tuberal area TuD hp2 THy Present results thalamic nucleus, thalamic nucleus, central part (core) intermedio-medial part Ventromedial hypo- VMHil thalamic nucleus, intermedio-lateral part Ventromedial hypo- VMHvl Tuberal Ventromedial hypo- VMHvm Intermediate tuberal TuI hp2 THy Present results thalamic nucleus, thalamic nucleus, area ventrolateral part ventromedial part Ventromedial hypo- VMHvi thalamic nucleus, ventro-intermediate part Ventromedial hypo- VMHvl thalamic nucleus, ventrolateral part For the columnar terms (italic) we used the terminology, schemata, and table of Hahn et al (2019), whereas for the prosomeric terms (roman) we used Puelles et al (2012) and VMH subdivision concepts introduced in the present report as reference Brain Structure and Function Fig. 6 Apparent alar origin of Vax1-expressing VMH cells. a, b plays Vax1 transcripts mainly at its VMHdrm/VMHdrl subdivisions Sagittal images illustrating two Vax1 ISH-reacted sections in E13.5 (shown at P4). Scale bars represent 200 µm. RTuI retrotuberal inter- mice from the Allen Developing Mouse Brain Atlas. The black mediate area, TuI tuberal intermediate area, main or subl. RTuD, dashed line identifies the alar-basal boundary (alar to the right; basal main or subliminal retrotuberal dorsal area; main or subl. TuD main to the left, as in Fig. 1b). The TuI, main TuD and subliminal TuD or subliminal tuberal dorsal area, PSPa peduncular subparaventricular progenitor domains are delimited by solid lines. At b there appears area, TSPa tuberal subparaventricular area, primVMH primordium of Vax1 signal penetrating the underlying TuD area. c At E15.5, the area the ventromedial hypothalamic nucleus, VMHdrm dorsal-rostrome- showing displaced Vax1 labeling extends ventralwards into a dorsal dial VMH subnucleus part of the VMH primordium. d The final VMH configuration dis- Nr5a1 it), including the associated acroterminal TuD domain, fully devoid of Nkx2.2 signal (Fig. 9a, d, g, k). At E13.5, Nr5a1- A comparison of the relatively complementary expression of positive cells are present at the deepest stratum of the local (TuD) mantle, extending into TuI (Fig. 9d–f, k–n). These Nkx2.2 and Nr5a1 within VMH illustrates the basic organi- zation we propose for the ventromedial primordium. Nr5a1 cells thereafter invade the deep region of the underlying VMH primordium (Fig. 9k–n). The migrating deep stream (previously known as steroidogenic factor 1, SF1), has been widely analyzed developmentally and functionally, and is of Nr5a1 cells thus seems covered by the more superficially migrating separate stream of Nkx2.2 cells (Fig. 9a–f, h–j, sometimes considered a selective transient marker of the entire VMH nucleus (Ikeda et al. 1995; Shinoda et al. 1995; l–n). The migration of Nr5a1 cells originating at the Nr5a1- positive acroterminal TuI area is oriented ventrocaudally, Dellovade et al. 2000; Tran et al. 2003; Davis et al. 2004; Dhillon et al. 2006; Kim et al. 2011, 2019; Büdefeld et al. apparently incorporating into the main TuD migration into VMH. Once the Nr5a1 cells reach the deep part of VMH 2012; Cheung et al. 2013; see Discussion). However, this contrasts with our present data, given that most Nkx2.2- they spread out radially and tangentially within the nucleus (Fig. 9l–n). At E16.5 Nr5a1 cells are visible mainly medially positive VMH cells seem to be Nr5a1 negative. Nkx2.2 has received relatively less attention so far as a VMH marker at the VMHdrm, and VMHim subdivisions, with a medio- laterally decreasing density gradient partially overlapping (Kurrasch et al. 2007; Puelles et al. 2012; Corman et al. 2018). with Nkx2.2 cells in an inverse lateromedial gradiental distri- bution. At this stage we found intermixed Nr5a1 and Nkx2.2 In contrast to Nkx2.2, the gene Nr5a1 is excluded early on from the narrow subliminal TuD band but is strongly populations at the VMHdrm and VMHim, whereas Nr5a1 cells are absent at the VMHil subnucleus where Nkx2.2 cells expressed at the underlying intrinsically Nkx2.2 negative main TuD area (though migrating Nkx2.2 cells pass through are massively present (Fig. 9o–q). 1 3 Brain Structure and Function Fig. 7 Evolution of Nkx2.2 and Otp signals relative to the emer- d–g), and this process later expands rostralwards into THy, in parallel gence of the VMH primordium between E12.5 and E14.5. All pan- to the ventralward migratory phenomena leading to the tuberal VMH; els show double Nkx2.2 ISH/Otp IHC in sagittal a–c or oblique d–g the intercalated subparaventricular alar area (PSPa) is crossed sub- sections from E12.5, E13.5 and E14.5 embryos (levels of section in ventricularly by the dorsally migrating cells, but essentially remains d–g indicated in c). The dash black line always marks the alar-basal unlabelled, excepting some liminal expression (PPa; TSPa; prim- boundary and roughly parallel solid black lines delimit the sublimi- VMH; a–c). Our oblique sections through this area intersect the nal and main TuD areas in (a–c). The curved red line in a–c iden- VMH primordium and the PPaV at four levels, providing images tifies the hypotalamo-telencephalic boundary dorsal to the Otp-pos- consistent with a migratory interpretation. Note that the VMH Nkx2.2 itive paraventricular area (PPa; brown). b The blue Nkx2.2 signal cell population does not represent the entire VMH population. Scale previously related only to the subliminal TuD (a) now clearly starts bars represent 200 µm. ZLi zona limitans, Main or subl. TuD main or to extend ventralwards into the TuI territory at E13.5 (primVMH). c subliminal tuberal dorsal area, PPa peduncular paraventricular area, This process progresses considerably by E14.5 (primVMH). Note that PPaV ventral peduncular paraventricular nucleus, Tel telencephalon, a less important ventral expansion of Nkx2.2 cells is observed like- PoA preoptic area, RM retromamillary area, M mamillary area, PRM wise at the RTuD domain, which comes to contact the Otp-positive periretromamillary area, PM perimamillary area, TuI tuberal interme- periretromamillary band (PRM). d–g The original early Nkx2.2-pos- diate area, primVMH primordium of the ventromedial hypothalamic itive subliminal band (blue) shows a dorsalward migration into the nucleus, PTh prethalamus, HyA hypothalamo-amygdalar corridor, ventral part of the Otp-positive alar paraventricular area (PPaV; a–c; PSPa, peduncular subparaventricular area, A/B limit alar/basal limit 1 3 Brain Structure and Function Fig. 8 A Nkx2.2-expressing population partially connected with the fication images from the dashed square areas in (h–k). White arrows TuD reaches the TuI domain to form part of the VMH primordium. in h’,j’,k’ indicate cells that correspond to double-labeled CMFDA- a–g Sagittal sections in medio–lateral order comparing the VMH Nkx2.2 elements. The white arrows in a, i indicate the direction of primordium labeled with anti-Nkx2.2 antibody in red fluorescence Nkx2.2 or CMFDA label extension at higher magnification in the at migratory stage E14.5 (a–c) and postmigratory stage E18.5 (d–g). respective framed areas. Scale bars in c, h represent 200 µm. Scale The approximate contour of the whole VMH nucleus was traced with bar in h’ represents 100 µm. RTuI retrotuberal intermediate area, TuI a white line in f. At E18.5 the specific subdivisions of the VMH that tuberal intermediate area, RTuD retrotuberal dorsal area, TuD tuberal are invaded by this cell type can be identified tentatively (VMHdrm, dorsal area, primVMH primordium of the ventromedial hypothalamic VMHim, VMHil, VMHvi). h–k Representative sagittal images of nucleus, PPaV ventral peduncular paraventricular nucleus, VMHdrm our tracing experiments on organotypic cultures from E12.5 mice dorsal-rostromedial subnucleus of the ventromedial hypothalamic (48 h culture conditions, E12.5-E14.5), marked with the anti-Nkx2.2 nucleus, VMHim medial-intermediate VMH subnucleus, VMHvi ven- antibody (red), and illustrating ventralwards cell migration when tral-intermediate VMH subnucleus, VMHil lateral-intermediate VMH a CMFDA particle (black) was placed at the upper part of TuI (h), subnucleus the TuD (i, j), or the RTuD-TuD boundary (k). (h’–k’) Higher magni- 1 3 Brain Structure and Function Tcf7l2/Satb2/Sox14 overlap between them (Fig. 12e, e’). At E14.5 the VMH primordium contains both examined populations; Nkx2.2 At E13.5 the Tcf7l2 gene labels the TuD acroterminal man- cells are largely restricted to its dorsal part, whereas Nr2f1 tle, extending somewhat into the neighboring rostral TuI. cells appear mainly in its ventral part, with a small overlap This Tcf7l2-TuI pattern is partially complementary to that visible in sagittal sections (Fig. 12f). Using horizontal sec- of Satb2, whose expression occurs mainly at the periven- tions oriented parallel to the prosomeric alar-basal boundary tricular stratum of TuI (Fig. 10a–h). At E15.5, the primor- (see Fig. 1), the ventricular zone observed in the dorsalmost dium of the VMHdrm subnucleus is selectively marked by sections through basal hypothalamus (next to the alar-basal Tcf7l2, whereas the VMHdc primordium and the ventral boundary) contains Nkx2.2-positive elements, but no Nr2f1 parts of VMH are marked with Satb2 (Fig. 10i–p). At late cells (Fig. 12g–g’). On the contrary, at levels through the TuI developmental stages, the rounded VMHdrm is perfectly the ventricular zone shows Nr2f1 expression but no Nkx2.2 delineated by Tcf7l2, as is the VMHdc portion by Satb2 signal, though migrated Nkx2.2 cells are present in the man- (Figs. 10q–x, 11a–f). The latter gene also marks the three tle layer (Fig. 12h–h’). We also compared at E16.5 the Nr2f1 VMHvm, VMHvi and VMHvl divisions, some of which VMH subpopulation with Nkx2.2 or Nr5a1-marked cells. are also recognizable with Sox14, another TuI-TuD early Although a degree of overlap exists between any of these marker, which appears expressed at the VMHdrl, but not at populations, there are still indications of a partially differen- the VMHdrm companion, and similarly at the VMHvm, but tial dorsoventral distribution, with Nr5a1 and Nkx2.2 cells not at the VMHvl (Fig. 11g–i). occupying more importantly the VMHdrm subnucleus, and Nr2f1 cells preferentially appearing in ventral parts of VMH Nr2f1 where the other two markers are absent (Fig. 13a–d). The Nr2f1 marker (previously known as Couptf1) apparently Nkx2.1 labels a VMH subpopulation arising exclusively within TuI. This gene is expressed throughout the TuI/RTuI domains Nkx2.1 is widely expressed in the hypothalamic basal plate of the basal hypothalamus (excepting the corresponding at early stages (starting at neural plate stages; Shimamura acroterminal area) at E11.5 (not shown; see Allen Develop- et al. 1995; Qiu et al. 1998) but is absent at the retromamil- ing Mouse Brain Atlas) and E13.5 (Fig. 12a, b). The TuD/ lary area (RM) and its migrated VPM and STh derivatives RTuD area where early expression of Nkx2.2 and Nr5a1 (Puelles et al. 2012; López-González et al. 2021); there is was observed is devoid of Nr2f1 signal (Fig. 12a, b). At also a thin longitudinal band of Nkx2.1-positive cells ven- E15.5, Nr2f1 signal appears both within the dorsomedial trally within the SPa alar area; this may correspond to its hypothalamic nucleus (DMH; hp1 + hp2) and the VMH liminal subdomain and its origin remains uncertain (van (hp2; Fig. 12c, d). The latter formation shows two kinds of den Akker et al. 2008; Puelles et al. 2012; LP, unpublished labeling: an abundant Nr2f1 signal that marks particularly observations). At E13.5, Nkx2.1 is strongly expressed in the ventral parts of the VMH primordium, and dispersed the ventricular and mantle zones of the TuD and TuI pro- Nr2f1 neurons within the prospective intermediate part of genitor domains (Fig. 14a–c). Moreover, Nkx2.1 marks also VMH (red and blue asterisks in Fig. 12b, d). Subsequently the whole basal acroterminal territory from E11.5 onwards this marker appears strongly expressed at the VMHdc, and (Allen Developing Mouse Brain Atlas data; Fig. 14a; Puelles ventral portions of VMH. It also shows weaker expression at et al. 2012, their Figs. 8.9D; 8.10D). At E15.5, the VMH the transitional limits of the VMHim, VMHil, and VMHdrm primordium contains deep Nkx2.1-positive neurons in its subdivisions (Figs. 5i, 13a–d). ventral and dorsal parts (not so at the medial-intermediate We compared with double immunofluorescence the dis- part), and separate superficial cells are observed ventrally tribution of Nkx2.2- versus Nr2f1-positive cells. At E12.5, (Fig. 14k–m). At E18.5 the Nkx2.1 signal largely occupies the longitudinal Nkx2.2-positive hypothalamic progenitor the VMHdc, VMHdrm (with slight extension into VMHdrl), territory appears in sagittal sections dorsal to the longitu- VMHil VMHvm, and VMHvl subnuclei (Fig. 14n–p). dinal domain marked by Nr2f1, with only a small area of 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 9 Nkx2.2 and Nr5a1 reveal different migrated VMH popula- Six3 tions. a–n Nkx2.2 and Nr5a1 ISH reactions are compared in medi- olateral sagittal (a–f) and dorsoventral horizontal (g–n) sections of This is a particularly distinct and previously undescribed E13.5 mouse embryos from the Allen Developing Mouse Brain Atlas partial marker of the VMH nucleus, where it identifies a (horizontal relative to the prosomeric axis), illustrating the postulated zones of origin of the populations expressing selectively one or the new subdivision. Early on, Six3 is strongly expressed at the other of these markers and their topographic relationship with the acroterminal parts of TuD and TuI, the latter representing ventrally displaced and expanding VMH primordium, where both the primordium of the arcuate nucleus, median eminence populations remain spatially distinct. Sections (b, c, g, h) also show and infundibulum (Fig. 15a). In contrast, the neighboring Nkx2.2-expressing cells that migrate dorsalward into peduncular and terminal parts of the ventral paraventricular subarea (migPaV). subliminal and main TuD domains are Six3-negative, as The black dash line in b, e marks the prosomeric horizontal section well as the incipient VMH primordium (Fig. 15b). At E16.5, plane illustrated at various levels in g–n. The section in g is slightly though, strong Six3 signal can be identified at the acrotermi- oblique and shows at the left side alar (peduncular and terminal) nal arcuate and anterobasal nuclear primordia (basal plate) PSPa and TSPa regions bordering ventrally the liminal ventricular zone, whereas at the right side we see the transition into the underly- and the suprachiasmatic primordium (alar plate) while the ing basal subliminal ventricular zone; note none of these sides shows VMH primordium shows a novel strongly Six3-positive significant mantle layer labeling of VMH nature, which only starts to ovoid subpopulation within its ventral-intermediate part appear in the underlying sections h–j and l–n, always restricted to the (VMHvi; Fig. 15c, d). Subsequently, at E18.5, Six3 contin- terminal hypothalamus in what respects the VMH primordium. Note in h–j and l–n that at E13.5 the two cell types intermix somewhat ues labeling strongly the well delimited, ovoid mass seen at dorsal levels through VMH h, l, whereas they stay separate more previously (VMHvi); this is found intercalated between the ventrally j–n). o–q Combined blue Nr5a1 ISH and brown Nkx2.2 molecularly different VMHvm and VMHvl subdivisions immunohistochemical sagittal images at E16.5 show that intermixing (VMHvi; 15e–j). A weaker and more dispersed expression increases significantly by this stage. A blue dash contour was traced around the Nr5a1-positive VMH population, and a brown dash con- of Six3 is detected as well at the VMHim and VMHdrm tour surrounds the Nkx2.2-positive VMH population, allowing this subdivisions, whose labeled cells are less well delimited comparison. Scale bars represent 200 µm. TuD tuberal dorsal area, from the arcuate population, in contrast with the VMHvi migPaV migration of the ventral paraventricular cells, primVMH pri- (Fig. 15h–j). mordium of the ventromedial hypothalamic nucleus, AT acroterminal area, RTuI, retrotuberal intermediate area, TuI tuberal intermediate area, RTuD retrotuberal dorsal area, VMHdc dorsocaudal VMH sub- nucleus, VMHim medial-intermediate VMH subnucleus, VMHdrm Discussion dorsal-rostromedial VMH subnucleus, VMHil lateral-intermediate VMH subnucleus, VMHdrl dorsal-rostrolateral VMH subnucleus The VMH nucleus is one of the larger hypothalamic struc- tures (called ‘principal hypothalamic nucleus’ by Cajal). A comparison of Nkx2.1 and Nkx2.2 distribution at E14.5 Classically it has been widely defined as presenting dor - corroborates the predominancy of Nkx2.1 at ventral loci of somedial, central/core, and ventrolateral cytoarchitectonic the VMH, whereas Nkx2.2 is placed mainly at intermediate subdivisions (VMHdm, VMHc, VMHvl). Note the corre- levels, with some extension into the dorsocaudal subnucleus sponding topographic descriptor terms refer to the columnar (Fig. 14e–g). In our horizontal sections (similar to conven- forebrain axis postulated as running into the telencephalon tional columnar coronal sections but interpreted along the (the prosomeric dorsal direction). The VMH massively axial dimension), we observe Nkx2.1 and Nkx2.2 cells in the contains glutamatergic neurons (Puelles et al. 2012; their ventricular stratum at TuD levels (Fig. 14h–i’’’). The pres- Figs. 17A–C; 20A–C; 22, 23A,B). However, recent scR- ence of Nkx2.1-positive cells increased in the ventricular NAseq transcriptomic studies found an unexpected diversity layer at TuI levels, whereas Nkx2.2 cells are totally absent of neuronal molecular profiles in the VMH (Kim et al. 2019 in this stratum (Fig. 14j, j’). At these sections the VMH pri- found 12 neuronal clusters in the core portion of the VMH mordium in its entirety is revealed complementarily labeled and 17 neuronal clusters in the ventrolateral VMH part, by the Nkx2.2 and Nkx2.1 markers (Fig. 14j, j’, j’’). whereas van Veen et al. 2020 described six VMH neuronal 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 10 Comparisons of Satb2 and Tcf7l2 patterns reveal a rostrocau- (which is implicitly interpreted as a transversal section and dal subdivision in the dorsal sector of VMH. Satb2 and Tcf7l2 ISH does not show the more peripheral parts of the tuberal area) images from the Allen Developing Mouse Brain Atlas are shown and sagittal or true transversal sections of the VMH (see in either prosomeric horizontal sections at three dorsoventral lev- our Fig. 3a–g) are seldom examined. This leads to practi- els at E13.5 (a–c, e–g), E15.5 (i–k, m–o), and P1 (q–s, u–w) or in corresponding sagittal sections (d, h, l, p, t, x). The Satb2 marker cal invisibility of our highly relevant TuD and acroterminal is expressed initially only within TuI, where a dense periventricu- domains, the main sources of tangentially migrated VMH lar stratum of positive cells is visible at E13.5 (a, b; note additional cells. VMH-unrelated expression appears also at the acroterminal TuD area We have addressed this issue consistently with previous in a, d). At E15.5 the dorsalmost VMH Satb2 cells concentrate at the VMHdc subdivision (i, l), while other cells of this type have invaded prosomeric analysis, correlating diverse molecularly charac- massively the ventral parts of the VMH primordium (VMHvm; terized VMH neuronal populations to the set of molecularly VMHvl; j, k). At P1 the Satb2 subpopulation persists majoritarily defined progenitor domains previously reported in the area at the VMHdc and the ventral complex (q–t). In contrast, the Tcf712 of interest (Puelles et al. 2012; review in Diaz and Puelles marker initially appears expressed selectively at the acroterminal TuD area (e–h); later, Tcf7l2 cells migrate ventralwards into the rostral part 2020; see also Morales-Delgado et al. 2011, 2014). We of the VMH primordium (m–p), invading mainly the VMHdrm sub- were able to map early emergence of some distinct VMH division, while other acroterminal TuD cells seem to invade the TuL cell types in specific surrounding domains of the embry - nucleus (u–x) (e–h for E13.5; m–p for E15.5; q–x for P1). Scale bars onic hypothalamus, and then traced them via intermediate represent 200 µm. AT acroterminal domain, TuD tuberal dorsal area, TuI tuberal intermediate area, VMHdc dorsocaudal VMH subnucleus, stages to their ulterior topography within the VMH complex. VMHvl ventral-lateral VMH subnucleus, VMHvm ventral-medial These tracings suggested or were consistent with tangential VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, or radial migration patterns according to the characteristic VMH ventromedial hypothalamic nucleus, PM perimamillary area, spatiotemporal transitions deployed in each case. Neurons TuL tuberal lateral nucleus that form the VMH proper (leaving aside the even more het- erogeneous surrounding shell of the nucleus) were shown to types). Affinati et al. (2021) identified 24 VMH clusters originate mainly either in the TuD or TuI progenitor domains roughly categorized into 6 main groups. Comparison of within basal THy, sometimes involving also or exclusively diverse gene markers mapped in diverse section planes in the corresponding TuD or TuI acroterminal subregions (see the developing VMH already led Puelles et al. (2012) to the the novel prosomeric concept of the acroterminal domain conclusion that VMH was heterogenous and might not be as the rostromedian end of the THy in Puelles et al. 2012; originated as a whole within the intermediate tuberal area Puelles and Rubenstein 2015; Ferran et al. 2015; Puelles (TuI), the subdivision of the prosomeric basal hypothalamus 2018; Diaz and Puelles, 2020). A further contingent of VMH in whose terminal part (THy) the VMH lies in the adult. cells apparently originates from the overlying alar terminal Evidence was presented suggesting the contribution to VMH subparaventricular area (TSPa), otherwise previously identi- at least of Nkx2.2- and Pdyn-expressing cells, which appar- fied as a source of various dorsoventral peptidergic neuronal ently originated from the overlying TuD progenitor domain migrations from alar progenitor domains into the subjacent (Puelles et al. 2012; their Figs. 26A–N). Unfortunately, this tuberal/retrotuberal basal plate (Diaz et al. 2015). Since the well-documented notion has been generally disregarded in adult VMH apparently lies strictly within TuI, the TuD, the subsequent literature. Accrued reports on loss of func- acroterminal, and TSPa origins necessarily imply dorsoven- tion of selected VMH gene markers (Kurrasch et al. 2007; tral or rostrocaudal tangential migrations of the correspond- Cheung et al. 2013; Lu et al. 2013; Corman et al. 2018; ing derivatives finally found inside the VMH. The existence Aslanpour et al. 2020a, b) did not contemplate the proso- of dorsoventral tangential cell translocations coming from meric AP and DV subdivisions of the tuberal hypothalamic the TSPa or TuD areas was verified experimentally. region and the phenotypes observed were interpreted under We also described the approximate final distribution of the traditional (simpler) columnar assumptions (e.g., see the specific markers we traced developmentally regarding Altman and Bayer 1978, 1986). These include the undocu- different subdivisions or subnuclei identified in the E18.5 mented assumption that the whole VMH cell population is VMH. Due to the prosomeric approach, implying use of produced locally in the tuberal area (i.e., without any tangen- diverse axial references and positional landmarks (e.g., tial migration; further comments on this below). Columnar the longitudinal floor plate and the alar-basal boundary; reports on the VMH frequently concentrate descriptions on or the transversal acroterminal domain and the intrahy- a sole coronal section level midways through the nucleus pothalamic boundary) and a related map of molecularly 1 3 Brain Structure and Function Fig. 11 Differentially labeled populations of the VMH at late devel- coincidences at lateral levels (VMHdrl, VMHil, VMHvl). Scale bars opmental stages. Satb2 a–c, Tcf712 d–f, and Sox14 g–i ISH medi- represent 200 µm. VMHvm ventral-medial VMH subnucleus, VMHim olateral sagittal sections of E18.5 or P4 mice. These tree markers medial-intermediate VMH subnucleus, VMHdc dorsocaudal VMH are differentially expressed in diverse VMH subnuclei: Satb2 labels subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHvi distinctly the VMHdc subunit and the ventral area (VMHvm, VMHvi ventral-intermediate VMH subnucleus, VPM ventral premamillary and VMHvl), Tcf7l2 is selective for the VMHdrm-VMHdrl subdo- nucleus, VMHvl ventral-lateral VMH subnucleus, VMHil lateral- mains, and Sox14 seems to coincide with Satb2 at deep and interme- intermediate VMH subnucleus, VMHdrl dorsal-rostrolateral VMH diate sagittal sections (signal at VMHvm; g, h), but differs drastically subnucleus, DMH-P peduncular part of the dorsomedial hypotha- at lateral levels i; comparison with Tcf7l2 shows a complementary lamic nucleus, DMH-T terminal part of the dorsomedial hypotha- pattern at medial and intermediate levels (VMHdrm; g, h) and some lamic nucleus, VMHvm ventral-medial VMH subnucleus defined progenitor domains (Puelles et al. 2012; Puelles was more detailed than was usual heretofore, inspiring and and Rubenstein 2015; Ferran et al. 2015; Diaz et al. 2015, justifying the proposed terminological changes (Fig. 16; 2020; López-González et al. 2021; Fig. 1a,b) in the context Table 3). VMH molecular regionalization along the non- of 22 marker genes, our analysis of inner VMH structure arbitrary dorsoventral, rostrocaudal, and mediolateral 1 3 Brain Structure and Function (radial) dimensions was reexamined, leading to an expanded but progressive steps in that direction are absolutely neces- subdivision map of the VMH complex. This contains sary to the progress of neuroscience. minimally 8 parts that are consistent with the existence of The first origin of Nr5a1- and Nkx2.2-positive neurons combinations of several neuronal populations that show that eventually populate the VMHdc, VMHdrm, VMHdrl, differential molecular features, showing in their postmigra- VMHim. and VMHil subdivisions was traced to the early tory configuration various degrees of partial overlap. The TuD area (differentially at its main ventral subdivision and conventional simpler tripartite schema was approximately its overlying subliminal Nkx2.2-positive part); these two conserved by terminological distinction of dorsal, inter- TuD populations migrate respectively in deep versus super- mediate, and ventral VMH regions, subdivided as follows: ficial mantle cell streams, and later show a medial to lateral our dorsal VMH region includes not only the conventional (radial) stratification (Nkx2.2 cells born at the subliminal ‘dorsomedial’ part, whose topologic position is more pre- TuD migrate superficially and finally lie lateral/superficial cisely described as ‘dorsocaudal’ (VMHdc), but also a pre- to the population of Nr5a1 cells born at the underlying main viously undescribed ‘dorsorostral’ part, which divides into TuD and migrating in a deeper radial level). Both parallel medial and lateral dorsorostral subcomponents (VMHdrm, cell populations massively migrate ventralwards (tangen- VMHdrl); our intermediate VMH region roughly corre- tially) as observed at E12.5-E14.5, thus forming transient sponds to the conventional ‘central or core’ part, but divides separate deep and superficial streams that spread out within distinctly into medial and lateral-intermediate subunits the upper dorsal and intermediate parts of the VMH pri- (VMHim, VMHil); finally, our prosomeric ventral VMH mordium once they reach the TuI. Additional molecularly region includes the conventional ‘ventrolateral’ part and distinct VMH cell populations were traced instead to local displays distinct medial, intermediate, and lateral subunits origins within the TuI ventricular zone, which underlies (VMHvm; VMHvi; VMHvl). Our VMHdc indeed happens radially the VMH primordium. These comprise at least cell to be ‘relatively dorsal’ in the prosomeric VMH (i.e., close types expressing Nr2f1, Lmo4, Satb2 (Tables 1, 2). In some to the prosomeric alar-basal boundary of the tuberal area, cases, a given molecular subtype may be produced at both like its VMHdrm/VMHdrl companions), and thus merits the the TuD and TuI domains (e.g., Nkx2.1, Sox14). ‘dorsal’ descriptor. The latter was previously widely used in The novel dorsal-rostral VMH subdivisions (VMHdrm, reference to the seemingly obsolete columnar axis (check its VMHdrl; first described in the present report, though previ- inconsistence with the primarily longitudinal Nkx2.2-posi- ously implied by some reported observations) apparently tive band, commented in Puelles and Rubenstein 2015, and derive mainly from the acroterminal part of the TuD pro- note the wrong columnar assumption that coronal sections genitor region (represented, e.g., by Tcf7l2, Sox14, and some through the hypothalamus show the dorsoventral dimension, rostral Nr5a1 cells); the migration into VMH of these com- though they demonstrably show the anteroposterior one by ponents thus implies an oblique dorso-rostrocaudal direc- passing caudalwards into the diencephalon and midbrain; tional vector. Fig. 1c). The ‘medial’ descriptor in this term was changed to The Six3-expressing cells of the VMHvi also arise within ‘caudal’ (VMHdm = VMHdc) because this term is not suf- the acroterminal basal domain; it is open to discussion ficiently selective; all dorsal, intermediate, and ventral parts whether they come from the TuI or TuD acroterminal sub- of the VMH have ‘medial’ portions (i.e., parts closer to the areas, or from both. We suspect that the TuI acroterminal periventricular stratum); this includes also the dorsal-rostral subarea is principally related to the production of arcuate part of VMH that has a distinct medial half (VMHdrm). We nucleus neuronal types (in caudal continuity with the DM reproduced the resulting structural schema in axially true complex), where mainly GABAergic neurons are produced; transversal, sagittal, and horizontal section planes in Fig. 16, this would suggest that the Six3 cells of the VMH (also first and we tried to clarify the new terminology for the readers described in the present report) may arise instead within the in Table 3. The New Neuromorphology (Nieuwenhuys and molecularly distinct acroterminal TuD (site of the proso- Puelles 2016) made possible by modern molecular and trans- meric ‘anterobasal nucleus’, or classical ‘retrochiasmatic genic experimental data claims the need to clarify various area’). newly emerging concepts by carefully rationalized terminol- Finally, Vax1-expressing VMH neurons originate clearly ogy changes. It is expected that full incorporation into usage within the overlying subparaventricular alar progeni- of the new terms and concepts requires generational change, tor area (TSPa), where Dlx family genes are also strongly expressed and GABAergic cells are thus expected to arise. 1 3 Brain Structure and Function This observation of some potentially GABAergic neurons Both TuI and TuD are progenitor areas for VMH cells expressing Vax1 and maybe Dlx genes within VMH (mainly at the VMHdrm and VMHdrl subnuclei) is unexpected Topographically, the development of VMH seems linked and needs to be checked in appropriate material (see, for to the basal region of the terminal hypothalamus (THy), instance, Puelles et al. 2012; their Figs. 8–18 and 8–15C, namely to its tuberal territory (Tu). Nothing similar to D, which seem to support this conclusion). the VMH develops within the cognate RTu area of the 1 3 Brain Structure and Function ◂Fig. 12 Nr2f1 expression illustrates a significant local TuI contribu- underlined by Puelles et al. 2012 but was not noticed when tion to the VMH primordium. a–d Nr2f1 ISH sagittal images at E13.5 these columnar names were introduced, probably because and E15.5 from the Allen Developing Mouse Brain Atlas. Solid lines it was assumed that the terminal DMH component was as in a, b delimit the longitudinal (subliminal) TuD/RTuD, (main) TuD/ a whole caudal to the VMH, which is maybe slightly less RTuD and TuI/RTuI progenitor basal domains. Red asterisks in b, d show a strong Nr2f1 signal at the ventral part of VMH primor- incongruent in columnar terms, but is in any case equally dium and its more mature derivative. Blue asterisks in b, d identify false in prosomeric terms; see also Puelles 2019). In contrast dorsal areas of relatively weaker Nr2f1 signal at the VMH primor- to the VMH that is restricted to TuI, the DMH complex of dium. (e–h’) Double Nkx2.2 (red) and Nr2f1 (green) immunofluo- RTuI is bineuromeric (having terminal/hp2 and peduncular/ rescence in sagittal (e–f) and horizontal (g–h’) sections of E12.5 and E14.5 embryos. The dash lines in f indicate the horizontal section hp1 moieties). This DMH longitudinal continuity passing planes at g and h. At TuD/RTuD levels (just dorsal to VMH) the ven- finally under the VMH (in the prosomeric schema) already tricular zone and adjacent mantle zone are Nkx2.2-positive (vz; mz; suggests that something extraordinary linked only to THy g–g’), whereas more ventrally through the VMH primordium (within and hp2 affects the development of VMH. Given that the TuI) the ventricular zone is only Nr2f1-positive and the mantle zone has a mixture of green cells (particularly at deep levels), with red and DMH formation is largely GABAergic and essentially yellow cells (vz; mz; h–h’). Scale bars in a, e, g represent 200 µm. locally produced (with minor immigrated glutamatergic core Scale bar in g’ represents 100 µm. Scale bar in h’ represents 50 µm. nuclei; Puelles et al. 2012), it might be easily deduced that primVMH primordium of the ventromedial hypothalamic nucleus, perhaps the massively glutamatergic VMH subarea of TuI RTuI retrotuberal intermediate area, TuI tuberal intermediate area, RTuD retrotuberal dorsal area, TuD tuberal dorsal area, PPaV pedun- is not really a part of TuI, but a variant subdomain of termi- cular paraventricular ventral nucleus, DMH dorsomedial hypotha- nal Tu intercalated between TuD and TuI which may have lamic nucleus, VMHim medial- intermediate VMH subnucleus, vz unique properties. Based on the identification within TuD ventricular zone, mz mantle zone, VMHil lateral-intermediate VMH of the earliest expression of the Nkx2.2 and Nr5a1 markers subnucleus, OCh optic chiasma which later characterize some positionally distinct types of VMH cells, Puelles et al. (2012) estimated that at least some peduncular hypothalamus (PHy) (in prosomeric theory, VMH neurons possibly originated via selective dorsoven- THy and PHy belong to different prosomeres -hp2 and hp1, tral tangential migration from the terminal TuD domain. At respectively- and Tu/RTu have therein analogous topologi- that time, it was unclear whether any VMH neurons arise cally basal positions under the alar-basal boundary; they primarily at TuI level. Puelles et al. (2012) rather inclined are thus presumed to result both from similar ventralizing to the assumption that all VMH subpopulations might be versus dorsalizing antagonism occurring in two molecularly migrated from TuD. distinct neuromeres, thus the observed differences; the exist- We reexamined this issue with more abundant molecular ence of important typological neuronal differences between data and some experimental testing of the previously pre- these two hypothalamic areas is both unsuspected and inex- dicted dorsoventral migration, corroborating its existence, plicable within the obsolete now more than 100 years-old but reaching the conclusion that the mouse VMH appar- columnar frame of thought). The unique developing VMH ently contains both local and exogenous neuronal popula- structure is accordingly one of the many differential features tions. Some neurons with differential molecular profiles can that can be explained by the postulate of the dual hp1 and be traced to origins within either TuI or TuD, separately, hp2 hypothalamo-telencephalic prosomeres (Puelles et al. or jointly. Moreover, additional VMH subpopulations were 2012; Ferran et al. 2015) but has no explanation within the found that seem to originate specifically in the basal AT area alternative columnar system. The ample tuberal domain lying dorsal to the Arc area, and some neurons migrate into of THy can be divided dorsoventrally (according to the VMH coming from the overlying alar subparaventricular prosomeric axis) on the basis of molecular and structural area. This supports the idea that patterning of the tuberal features into dorsal (TuD), intermediate (TuI) and ventral area occurs differentially at PHy compared to THy, intro- (TuV) progenitor domains. All three of them extend ros- ducing the unique and complex VMH areal phenomenon tralwards into the corresponding parts of the rostromedian found dorsal to the terminal end of the DM ‘column’. acroterminal area (AT; Puelles et al. 2012). TuI is the topo- The contextual background of any neural developmen- graphic site of both the terminal part of DMH and the VMH, tal analysis includes data on local neurogenesis. Report- with the terminologically unacceptable oddity that the ter- edly, VMH neurogenesis occurs in mice between E9.5 and minal DMH component clearly lies ventral to VMH (this E14.5, with a peak at E11.5-E12.5 (Shimada and Nakamura topologic incongruency of the columnar names was already 1973; Aslanpour et al. 2020b; similar rat data in Ifft 1972 1 3 Brain Structure and Function Fig. 13 Nkx2.2, Nr5a1 and Nr2f1 signals represent partially overlapping subpopulations of the VMH primordium. a, b Double immunofluorescence for Nkx2.2 (red signal; pink dash line) and Nr2f1 (green signal and dash line) in VMH sagittal sections at two section levels at E16.5, showing partial overlap. c, d Combined Nr5a1 ISH (dark bluish signal; blue dash line) and Nr2f1 immunohisto- chemical reaction (brown signal and dash line) at two different section levels. There is also only a partial overlap of these two populations. Scale bars repre- sent 200 µm. VMHvm ventral- medial VMH subnucleus, VMHim medial- intermediate VMH subnucleus, VMHdrm dorsal-rostromedial VMH sub- nucleus, VMHdc dorsocaudal VMH subnucleus or Altman and Bayer 1986). Interpreting their data in terms The TuD area is divided into dorsal subliminal and ven- of the columnar model, Aslanpour et al. (2020b) identified tral main portions; the migrating Nkx2.2 cells are thought a “rostrocaudal” gradient in VMH birthdates, which we to arise at the subliminal part, so that they must cross the translate topologically to the prosomeric model as a dors- underlying main part to reach the VMH. The main TuD oventral gradient. Presumably TuD neurogenesis precedes domain is where Nkx2.1-positive Nr5a1-expressing cells that of TuI, which is consistent with reported wholemount are produced. The subliminal TuD portion reportedly does rat results using AChE as an early differentiation marker not express either Nkx2.1 or Nr5a1. The molecular pair (Puelles et al. 2015; Amat et al. 2022). Accordingly, we Slit/Robo involved in regulation of neuronal migration and deduce that our dorsal tuberal Nr5a1 and Nkx2.2 VMH cells axonal guidance is expressed at the VMH primordia. The would be born earlier (E10.5) than the ventral ones (E11.5), transcription factors Nkx2.1/Isl1/Otp regulate Robo2 expres- due to their origin in the more precocious TuD area. sion. Moreover Slit1–/– and Slit2–/– mice show increased cell density at the VMH subventricular zone (Romanov Dorsally originated Nkx2.2 cells invade et al. 2020). It would be interesting to analyze whether the the intermediate VMH via tangential migration Slit/Robo signaling pathway affects the migrating Nkx2.2 population of VMH. Descriptive analysis of gene patterns led Puelles et al. (2012) The transcription factor Nr5a1 (SF1) plays a role in to suggest the migration en masse into VMH of Nkx2.2 ele- VMH development regarding its neuronal identity (Tran ments born at the subliminal dorsal tuberal area that under- et al. 2003), cell distribution (Ikeda et al. 1995; Shinoda lies the alar/basal limit. We confirmed this migratory process et al.1995; Davis et al. 2004) and connectivity (Tran et al. with our own descriptive and experimental data. We collat- 2003). This factor was widely used in functional studies of erally noted that less numerous ventrally migrating Nkx2.2 the VMH nucleus (Büdefeld et al. 2012; Hashikawa et al. cells from the subliminal dorsal retrotuberal area seem to 2017; Kennedy et al. 2020; Lewis et al. 2022). In contrast, invade the area of the peduncular dorsomedial hypothalamic there is little information about the role of the Nkx2.2 tran- nucleus, though they are not clearly detectable at advanced scription factor in VMH development (Kurrasch et al. 2007; embryonic stages (Puelles et al. 2012; Kim et al. 2021). We Puelles et al. 2012). In this study we have compared the also observed migration of CMFDA-labeled Nkx2.2 cells developmental distribution of both, Nr5a1 and Nkx2.2, not- into the VMH primordium when the TuD area under the ing that, irrespective of partial overlaps, they largely occupy alar-basal border was marked. different sectors of the VMH. The literature often wrongly assumes that Nr5a1 is a general transient marker of the 1 3 Brain Structure and Function VMH, but that is certainly not the case, according to our of the VMH. This is the case of Nr2f1 cells identified even- data. Nr5a1 cells occupy a larger part of VMH than Nkx2.2 tually at the ventral VMH tier (VMHvm, VMHvl), though ones, but both are restricted to given VMH subnuclei (Nr5a1 some of them overlap with tangentially migrated Nr5a1 and mainly at VMHdc, VMHdrm and VMHim; Nkx2.2 is found Nkx2.2 elements in VMHim/il, respectively. Nr2f1 cells also in a mediolateral gradient along the intermediate VMH appear separately at the VMHdc subnucleus (=conventional tier, being stronger at its lateral part, in contrast with the VMHdm subdivision). Similarly, the Satb2 marker identifies medially predominant Nr5a1 subpopulation). This pattern another radially migrated local cell population found within is thus explained by our interpretation that both populations the ventral VMH tier (VMHvm, VMHvl), which is differen- are exogenous relative to the VMH area and each invades tially absent from the intermediate tier. These observations it differentially according to different origins, routes, and suggest that also the radially migrated components of VMH final distributions (presumably affected by their differential may have subtly diverse topographic origins and postmigra- reaction to the local adhesivity or attractor scenario found tion distributions. within the VMH). The Nr5a1 pattern across AT/main TuD/TuI-VMHdc/ Varied neuronal identities within the VMH VMHdrm/VMHim is reproduced by other markers (e.g., Cnr1, Slc17a6, Robo2), whereas the characteristic Nkx2.2 Lu et al. (2013) showed evidence suggesting that the early pattern across subliminal-TuD-VMHil/VMHdrl is rather transcription factor Rax has a role in conferring a Nr5a1- unique. expressing and glutamatergic cell fate to cells in the central VMH. Rax is an early gene expressed at the TuI and TuD TuI origin of some VMH neurons (our interpretation of Allen Developing Mouse Brain Atlas material at E11.5 and E13.5), overlapping the Shh-positive Radial migration has been assumed conventionally as the basal plate territory, while a large part of the rostral acroter- only mechanism laying down the VMH primordium in minal domain (particularly the TuI one) loses secondarily the the tuberal mantle. Such cells would move from the local Shh signal. We noted that the prospective VMH area within ventricular zone into the neighboring mantle, presumably TuI retains up to E13.5 a central patch of low level Shh sig- guided by outside-in stratification rather that radial migra- nal. Lu et al. (2013) observed that when Rax was floxed tion proper (McClellan et al. 2006). Misplaced cells con- under control of the Shh promotor, a central region of Nr5a1 taining estrogen receptors identified in Nr5a1 null mice cells within VMH lost this signal (among others related to or in GABA R1 knockout mice have been interpreted to the glutamatergic phenotype- see their Figs. 7M–P; 7A’, represent failed radial migration (Dellovade et al. 2000; F’, B’) and came to be occupied by GABAergic neurons McClellan et al. 2008). Moreover, agonists and antagonists (their Fig. 7G’). Remarkably, surrounding dorsocaudal and of GABA and GABA receptors acting in vitro on coronal ventral Nr5a1 cells of the VMH remained unaffected. The A B hypothalamus slices (these are actually horizontal relative authors illustrated the tuberal expression of Shh at E10.5 to the hypothalamic basal plate) were reported to change the in a true transversal section (note the eye stalks and tel- speed and orientation of radially migrating neurons (Dello- encephalic vesicles lying above the hypothalamus) but did vade et al. 2001; McClellan et al. 2008). In all these cases not identify this area as the TuD subregion (a longitudinal the authors did not consider the possibility that processes band extending under the alar-basal boundary); this is the involving tangential migration might be affected instead or dorsal basal domain that selectively retains strong primary simultaneously. Shh signal normally (Andreu-Cervera et al. 2019; Puelles The conventional ‘ventrolateral’ VMH subregion, which et al. 2012). If this curious phenotype change occurring in apparently is majoritarily colonized by radially migrated the central or intermediate part of VMH depends on flox- neurons, is well known for the selective mapping of estro- ing that selectively occurred at the longitudinal Shh-positive gen receptors (e.g., Cheung et al. 2013; Krause and Ingra- TuD domain, the whole terminal TuD area producing tan- ham, 2017; Ma et al. 2021). Some authors described also gentially migrating Nkx2.2 and Nr5a1 VMH neurons would Nkx2.1 expression as highly restricted to ‘ventrolateral’ be expected to have been affected. It is thus difficult to see VMH (Tran et al. 2003; Davis et al. 2004; Cheung et al. why some parts of VMH developed normally. It is actually 2013). Our data indicate that early Nkx2.1 expression dis- somewhat unclear in this report whether Rax is expressed tributes widely throughout the tuberal region. At E18.5 we strongly enough at the level of the TuD band, to respond to still observe expression of this marker at the tuberal perive- the local Shh floxing effect (see in this regard the results of ntricular stratum deep to the VMH, as well as in neurons Orquera et al. 2016, suggesting that there is little Rax expres- of some VMH subdivisions, namely the VMHdc, VMHvm, sion within TuD). and VMHvl subnuclei. Other radially migrated local popula- A more detailed study of the Allen coronal series for tions studied by us are mainly found at the ventral portion E11.5 and E13.5 indicated that ventral to the strongly 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 14 Widespread basal expression of Nkx2.1 hypothalamus Aslanpour et al (2020a) suggested that Ascl1, a bHLH marker but selectivity at the VMH primordium. a–c Medio-lateral transcription factor expressed early on in the ventricular E13.5 sagittal sections from the Allen Developing Mouse Brain stratum of the TuD and TuI domains, likewise switches on Atlas with Nkx2.1 ISH show widespread basal signal of this hypo- differentiation programs that give rise preferentially to some thalamus marker (check also Fig. 3h in the adult). e–g Comparison of Nkx2.2 (red) and Nkx2.1 (green) immunofluorescence in medi- VMH neurons (e.g., our intermediate or dorsocaudal VMH olateral sagittal sections at E14.5. The dashed contour of the VMH populations). primordium is indicated in (e). The labeled cell populations are Our results indicate that VMH neurons originate from mostly topographically distinct, though a few double-labeled cells are diverse progenitor areas, and occupy characteristic, partly observed dorsocaudally (yellow in f, g). (h–j) Dorsoventral horizon- tal sections of E14.5 embryos reacted for Nkx2.2 (red) and Nkx2.1 overlapping parts of the VMH. Hypothetic areas of over- (green) immunofluorescence (section levels marked in f). At level h lap between the diverse progenitor domains might explain there is a poor Nkx2.1 signal (h’); this starts to appear stronger at the occasional visualization of double-labeled VMH cells ventricular and periventricular zone i levels (i’, i’’; i’’ and i’’’), and that express different combinations of transcription fac- expands into the VMH mantle at j levels (j’, j”). k–m Medio-lateral sagittal sections at E15.5 from the Allen Developing Mouse Brain tors (e.g., Nkx2.2-Nr5a1, Nkx2.2-Nkx2.1, Nkx2.2-Nr2f1), Atlas reacted for Nkx2.1 ISH, showing preferent reaction at VMHim, besides cells that express Nkx2.2 without Nr5a1, Nkx2.1 or VMHvm, and VMHvl subdivisions. n–p Comparison of double Nr2f1. Similarly, Corman et al. (2020) identified Shh cells Nkx2.2 (red) and Nkx2.1 (green) immunofluorescence at mediolat- and Shh-responsive (Gli1) lineage cells as contributing to eral sagittal section levels from E18.5 embryos. Note double-labeled yellow cells at VMHim (o). Scale bars in a, e. k, n represent 200 µm. the VMH population; they studied several VMH markers Scale bar in i’ represents 100 µm. Scale bar in i’’’ represents 50 µm. and quantified that, approximately, a half of Nr5a1 cells, a AT acroterminal area, TuI tuberal intermediate area, TuD tuberal dor- half of Nkx2.1 cells and 20% of Nkx2.2 cells belong to a Shh sal area, VMHvm ventral-medial VMH subnucleus, VMHim medial- lineage occupying a rostral VMH subregion, whereas Gli1 intermediate VMH subnucleus, VMHdrm dorsal-rostromedial VMH subnucleus, VMHvl ventral-lateral VMH subnucleus, primVMH pri- lineage cells represent a proportion of caudal VMH cells, mordium of the ventromedial hypothalamic nucleus, VMHdc dorso- with relatively less Nr5a1 and Nkx2.1 elements, and more caudal VMH subnucleus, VMHil lateral-intermediate VMH subnu- Nkx2.2 neurons. These results agree with ours as regards cleus, VMHdrl, dorsal-rostrolateral VMH subnucleus the existence of molecularly diverse domains of origin that produce qualitatively different VMH subpopulations. Shh-positive TuD band and wholly inside the Rax-positive Studies examining loss of function of single transcription domain there also appears a patch of weaker Shh expres- factors, or a morphogen such as Shh, have in common that, sion which can be ascribed to the upper TuI area (VMH while a discrete part of the VMH population is affected, in primordium), whose radial derivatives might contribute none of them the whole nucleus is disrupted. Consistently selectively to the central VMH portion that results devoid with our present results, this suggests that there are several of Nr5a1 cells and abnormally populated by GABAergic neighboring but different sites of origin of VMH cells, appar - neurons (Lu et al. 2013). We therefore consider it likely that ently with subtle variations in their transcriptomic set. Differ - the observed restriction to central VMH of the Rax lack of ent subpopulations arising outside the radial position of the function phenotype may be explained as due to an effect on VMH (upper TuI) converge via parallel tangential migrations the local patch of weak TuI Shh signal. Lu et al. (2013) did into the nuclear primordium. Once there, they partly disperse not show images of the TuD area above the VMH at later and intermix with other VMH cell types, without achieving stages, which might have indicated whether the Nr5a1 cells a fully homogeneous distribution. This eventually results in that normally migrate dorsoventrally into VMH may have the characteristic cellular typological profile of parts of the failed to do so, displaying instead abnormal positions or sig- nucleus (Fig. 16; VMHdc, VMHdrm, VMHdrl, VMHim, nals of cell death. Moreover, when Rax was deleted under VMHil, VMHvm, VMHvi and VMHvl), with different neu- the action of the Six3 promoter (a gene primarily overlap- rons, transcriptomic profiles, connections, and functions. We ping Nkx2.1 expression only at the acroterminal Arc area), think that birthdating results, as well as most analyses of lack Nkx2.1 was also dramatically diminished in the VMH pri- of function phenotypes, are so far non illuminating, because mordium (Lu et al., 2013). These results imply that some they were planned and interpreted assuming a single origin VMH cells are produced radially at the TuI area (under Shh of the heterogeneous VMH neurons. control) and others derive tangentially from the acrotermi- Indeed, recent single-cell transcriptomic studies have nal TuI area (under Six3 control; it should be investigated increased the ability to identify different VMH cell types. whether this bears particularly on our novel Six3-positive The available data on the hypothalamus already has made VMHvi subnucleus). Both processes may require indepen- us realize that the number of distinct cell types was being dently additional Rax function. It would be interesting to conventionally vastly underestimated in the hypothalamus. examine similarly whether Rax affects the Nkx2.2-positive In one of these studies, six principal VMH clusters were population of the VMH. identified in postnatal mice (P10), each of them represented by a particular transcription factor: Tac1, Rprm, Pdyn, Sst, 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 15 The early acroterminal marker Six3 (see a) later reveals a dis- corresponds to our novel Six3-positive VMHvi subdivision. tinct ventral VMH subnucleus (d–f, i, j). Six3 ISH images from Allen In this way, Hashikawa et al. (2017) identified a functional Developing Mouse Brain Atlas at E13.5 (a, b) and E16.5 (c, d) illus- division of the ‘ventrolateral’ part regarding with differen- trate that apart of Six3 cells developing within the acroterminal Arc tial mating and fighting behaviors. These data support the and SCh nuclei, a small compact ovoid mass appears labeled within the neighboring VMHvi subdivision as development advances. Paral- split of the classical ‘ventrolateral’ subdivision into the three lel red lines in d represent the somewhat oblique section planes of subnuclei; VMHvm, VMHvi, and VMHvl that we propose sections e–g, which interest caudally the negative thalamus (Th) and in our present VMH model. Six3-positive prethalamus (PTh), apart of the alar Otp-immunoposi- Affinati et al. (2021) similarly identified 24 molecularly tive paraventricular nucleus (Pa) and hypothalamo-amygdalar corri- dor (HyA). e–g Oblique serial sections showing combined Six3 ISH differentiable cell clusters in the VMH, which they were able (blue) and Otp IHC (brown) expression in an E18.5 embryo, identify- to categorize into 6 groups using majoritarily shared gene ing the novel VMHvi subnucleus and additional more disperse simi- markers. According to these authors, Dlk1 and Esr1 express- lar cells in VMHim and VMHdrm. The black curved line represents ing neurons are present at ‘ventrolateral’ classic division, the hypothalamo-diencephalic boundary, also shown in black in (c, d; note here the counterstain is TH, marking the A13 cell group). Lepr expressing cells are located into the ‘core’, and Foxp1- h–j Six3 ISH caudorostral serial sections from an E18.5 embryo cells map at their “anterior and lateral” to VMH core popu- transversal to the basal hypothalamus (the optic chiasma –OCh– and lation, possibly corresponding to our VMHdrl subnucleus. suprachiasmatic nuclei –SCh– appear dorsally; TSbO, tuberal subop- It seems probable, nevertheless, that even our more tic nucleus). The contour of the VMH complex is delineated and the intermediate ventral position of the novel Six3-positive VMHvi sub- detailed VMH subdivision model is surpassed by these division is evident. Scale bars represent 200 µm. RM retromamillary single-cell transcriptomic data, in the sense that all detected area, M mamillary area, AT acroterminal area, NHy neurohypophysis, clusters cannot be ascribed uniquely to our specific VMH TuI tuberal intermediate area, main or subl. TuD main or subliminal subdivisions. This means that our present minimally subdi- tuberal dorsal area, PTh prethalamus, PHy peduncular hypothalamus, THy terminal hypothalamus, OCh optic chiasma, VMHvm ventral- vided model probably also is inadequate in the long run to medial VMH subnucleus, VMHim medial-intermediate VMH subnu- fully explain the real VMH populational complexity. Our cleus, VMHdrm dorsal-rostromedial VMH subnucleus, Arc arcuate present attempt to introduce the idea of alternative progeni- nucleus, VMHvi ventral-intermediate VMH subnucleus, A13 A13 tor sources (and converging tangential migrations) for sev- dopaminergic cell population, PTh prethalamus, SCh suprachiasmatic nucleus, Th thalamus, PPa peduncular paraventricular area, VMHdc eral components of the whole VMH population thus cannot dorsocaudal VMH subnucleus, VMHil lateral-intermediate VMH be definitive and just points the way for future more detailed subnucleus, VMHvl ventral-lateral VMH subnucleus, VPM ventral studies (e.g., each of our postulated progenitor domains may premamillary nucleus, DPM dorsal premamillary nucleus, OT optic be subdivided). Interestingly, we also reached similar con- tract, TSbO tuberal suboptic nucleus clusions in two other previous studies, namely in the analysis of multiple progenitor domains (and mixed tangential migra- Hpcal1, Galanin (Veen et al. 2020). Tac1 appears in the tions) contributing different neuron types to the prepontine VMHdrm and VMHdrl subdivisions. In contrast, Pdyn is interpeduncular complex (Lorente-Cánovas et al. 2012), as present at the intermediate VMH subdivisions, which makes well as in our study of the tangentially migrated basal hypo- plausible a phenotypic correspondence with the Nkx2.2 cell thalamic ventral premamillary nucleus (López-González phenotype, though further studies would be needed. et al. 2021). A more resolutive single-cell transcriptomic analysis identified 29 neuronal types within VMH, 17 of them appar - ently restricted to the conventional ‘ventrolateral’ VMH Conclusion (Kim et al. 2019). These results show that the conventional columnar schema does not explain the existence of 29 dif- In this study we have addressed the identification and ferent glutamatergic cell types within its rather simplistic description of some molecularly identified VMH cell popula- (tripartite) VMH subdivision concept. After reinterpretation tions during development using immunofluorescence, in situ of such data within our present prosomeric VMH model we hybridization, and data extracted from the Allen Developing first note that the conventional ‘ventrolateral VMH’ only Mouse Brain Atlas. In our detailed descriptive analysis of grossly corresponds to our ventral VMH region, whose dem- these populations, we found important differences as regards onstrated subdivision into at least VMHvm, VMHvi, and their rostro-caudal and dorsoventral distribution throughout VMHvl parts was not known to columnar-based authors. the VMH nucleus, which also partly change over time, sug- Accordingly, it would be interesting to re-map the given set gesting in some cases the existence of tangential migration of 17 ‘ventrolateral’ clusters relative to our three ventral phenomena. These differences have not been described VMH subdivisions. Remarkably, Kim et al. (2019) identified before due to the widespread use of a (single) coronal sec- a Six3-expressing cell cluster and they mapped it specifi- tion plane in the analysis of this nucleus, which does not help cally to the ‘ventrolateral’ VMH region (their Satb2_Six3 the comprehensive interpretation of this territory, particu- cluster disclosed by SMART-Seq); this entity most probably larly when misinterpreted as cross-sections under columnar 1 3 Brain Structure and Function 1 3 Brain Structure and Function ◂Fig. 16 Schemata illustrating our updated VMH model with color- tuberal region, forming part of the intermediate part of VMH coded molecularly-typified populations. The schemata of successive as suggested by Puelles et al. (2012). VMH sections are arranged into horizontal a–e, sagittal f–j and trans- versal k–q series (see inset drawings above illustrating the section Acknowledgements We thank the Allen Institute for Brain Science for plane as a black line relative to the bent length axis of the prosomeric public availability of the markers analyzed (Website: 2013 Allen Insti- model, represented by the alar-basal boundary in red). The molecu- tute for Brain Science. Allen Developing Mouse Brain Atlas. http:// larly characterized cell groups mapped are represented as color-coded developing mouse. br ain-map. or g). Infrastructure support was provided small circles. Note that these circles do not represent individual cells. by the University of Murcia and IMIB-Arrixaca Institute of Murcia. Black dash lines in a–e and k–q separate the mediolateral halves of the VMH. Solid black lines in f–q delimitate the dorsal, intermedi- Author contributions LLG: conceptualized and designed the study, ate and ventral parts of the VMH. Red dash lines in f–j show the performed most of the experiments, made image composition and rough position of horizontal sections a–e, and red dash lines in k–q wrote the first draft of the manuscript. MMT: performed part of the show the positions of the sagittal section levels f–j in the transversal experiments. LP: conceptualized and designed the study, wrote the n fi al schemata. VMHdrl dorsal-rostrolateral VMH subnucleus, VMHdrm manuscript, and provided funding acquisition. dorsal-rostromedial VMH subnucleus, VMHdc dorsocaudal VMH subnucleus, VMHil, lateral-intermediate VMH subnucleus, VMHim Funding Open Access funding provided thanks to the CRUE-CSIC medial-intermediate VMH subnucleus, VMHvl ventral-lateral VMH agreement with Springer Nature. This work was supported by a Spanish subnucleus, VMHvi ventral-intermediate VMH subnucleus, VMHvm Ministry of Economy and Competitiveness grant, BFU2014-57516P ventral-medial VMH subnucleus, 3 V third ventricle (with European Community FEDER support), and a Seneca Founda- tion (Autonomous Community of Murcia) Excellency Research grant, reference: 19904/ GERM/15 (to L.P.), project name: Genoarchitectonic tradition. Many authors working on the VMH seem to be Brain Development and Applications to Neurodegenrative Diseases and Cancer, by Seneca Foundation (5672 Fundación Séneca). unaware that this nucleus is a basal plate derivative of only one of the two hypothalamic neuromeres (terminal hypo- Data availability The original contributions presented in the study are thalamus; Puelles et al. 2012), and its possible patterning included in the article, further inquiries can be directed to the corre- relationships with the hypothalamic floor plate and the alar- sponding author. Images from the Allen Developing Mouse Brain Atlas (https://de veloping mouse. br ain-map. or g/) were employed in this work. basal boundary (García-Calero et al. 2008; Andreu-Cervera et al. 2019), or the acroterminal (prechordal) area of influ- Declarations ence, remain likewise unfocused, because these notions are disregarded by the obsolete columnar model. Using the Conflict of interest The authors declare that they have no conflict of interpretive conceptual apparatus offered by the prosomeric interest. model, we showed that there exists a clear correspondence Ethical approval The animal study was reviewed and approved by between several distinct progenitor areas (e.g., alar terminal Directive 2010/63/EU, Royal Decree 1201/2005 and 53/2013 Law SPa, and basal main TuD, subliminal TuD, TuI, TuD-AT, 32/107 University of Murcia Committee (No. A13170406). TuI-AT) where different VMH cell types are born and ulte- Open Access This article is licensed under a Creative Commons Attri- riorly transferred via convergent tangential or radial migra- bution 4.0 International License, which permits use, sharing, adapta- tion to the developing VMH nucleus. This provides a new tion, distribution and reproduction in any medium or format, as long example of the utility of the prosomeric model to explain as you give appropriate credit to the original author(s) and the source, the organization of complex brain structures. Given that the provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are new single-cell transcriptomic tools are manifestly sensible included in the article's Creative Commons licence, unless indicated to the morphologic models used to map the characteristic otherwise in a credit line to the material. If material is not included in clusters they detect, novel alternative interpretive systems the article's Creative Commons licence and your intended use is not need to be taken into consideration and tested as regards permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a their possible advantages. A comprehensive and understand- copy of this licence, visit http://cr eativ ecommons. or g/licen ses/ b y/4.0/ . able molecularly defined cell type census over development and consequent further study of the respective mature cell populations can be significantly improved by searching opti- References mal structural and developmental models. We focused a good part of our analysis on the Nkx2.2 Affinati AH, Sabatini PV, True C, Tomlinson AJ, Kirigiti M, Linds- transcription factor because it is related in origin to the alar/ ley SR, Li C, Olson DP, Kievit P, Myers MG, Rupp AC (2021) basal hypothalamic boundary, a not very well understood Cross-species analysis defines the conservation of anatomically segregated VMH neuron populations. Elife. https:// doi. org/ 10. limit (some expert voices define it arbitrarily as an ‘oblique’ 7554/ eLife. 69065 landmark; e.g., Shimogori et al. 2010), and also because Alifragis P, Parnavelas JG, Nadarajah B (2002) A novel method of it is a forgotten gene marker in most studies of VMH. We labeling and characterizing migrating neurons in the developing confirmed in organotypic cultures the related dorsoventral central nervous system. Exp Neurol 174(2):259–265. https:// doi. org/ 10. 1006/ exnr. 2002. 7870 tangential migration that transfers a distinct Nkx2.2 cell pop- ulation from the dorsal tuberal region to the intermediate 1 3 Brain Structure and Function Altman J, Bayer SA (1978) Development of the diencephalon in the the ventromedial nucleus of the hypothalamus. J Neurobiol rat. I. Autoradiographic study of the time of origin and settling 49(4):264–276. https:// doi. org/ 10. 1002/ neu. 10011 patterns of neurons of the hypothalamus. J Comp Neurol 182(4 Pt Dhillon H, Zigman JM, Ye C, Lee CE, McGovern RA, Tang V, Kenny 2):945–971. https:// doi. org/ 10. 1002/ cne. 90182 0511 CD, Christiansen LM, White RD, Edelstein EA, Coppari R, Altman J, Bayer SA (1986) The development of the rat hypothalamus. Balthasar N, Cowley MA, Chua S Jr, Elmquist JK, Lowell BB Adv Anat Embryol Cell Biol 100:1–178 (2006) Leptin directly activates SF1 neurons in the VMHH, and Alvarez-Bolado G, Rosenfeld MG, Swanson LW (1995) Model of fore- this action by leptin is required for normal body-weight homeo- brain regionalization based on spatiotemporal patterns of POU-III stasis. Neuron 49(2):191–203. https:// doi. org/ 10. 1016/j. neuron. homeobox gene expression, birthdates, and morphological fea-2005. 12. 021 tures. J Comp Neurol 355:237–295. https:// doi. org/ 10. 1002/ cne. Díaz C, Morales-Delgado N, Puelles L (2015) Ontogenesis of pepti- 90355 0207 dergic neurons within the genoarchitectonic map of the mouse Amat JA, Martínez-de-la-Torre M, Trujillo CM, Fernández B, Puelles hypothalamus. Front Neuroanat. https:// doi. org/ 10. 3389/ fnana. L (2022) Neurogenetic heterochrony in chick, lizard, and rat 2014. 00162 mapped with wholemount acetylcholinesterase and the proso- Diaz C, Puelles L (2020) Developmental genes and malformations in meric model. Brain Behav Evol 97(1–2):48–82. https:// doi. org/ the hypothalamus. Front Neuroanat 14:607111. https://doi. or g/10. 10. 1159/ 00052 42163389/ fnana. 2020. 607111 Andreu-Cervera A, Anselme I, Karam A, Laclef C, Catala M, Schnei- Elmquist JK, Elias CF, Saper CB (1999) From lesions to leptin: der-Maunoury S (2019) The ciliopathy gene Ftm/Rpgrip1l con- hypothalamic control of food intake and body weight. Neuron trols mouse forebrain patterning via region-specific modulation 22(2):221–232. https:// doi. org/ 10. 1016/ s0896- 6273(00) 81084-3 of Hedgehog/Gli Signaling. J Neurosci 39:2398–2415. https://doi. Ferran JL, Puelles L, Rubenstein JL (2015) Molecular codes defin- org/ 10. 1523/ JNEUR OSCI. 2199- 18. 2019 ing rostrocaudal domains in the embryonic mouse hypothalamus. Aslanpour S, Rosin JM, Balakrishnan A, Klenin N, Blot F, Gradwohl Front Neuroanat 9:46. https:// doi. org/ 10. 3389/ fnana. 2015. 00046 G, Schuurmans C, Kurrasch DM (2020a) Ascl1 is required to Frohman LA, Bernardis LL, Stachura ME (1974) Factors modifying specify a subset of ventromedial hypothalamic neurons. Develop- plasma insulin and glucose responses to ventromedial hypotha- ment. https:// doi. org/ 10. 1242/ dev. 180067 lamic stimulation. Metabolism 23(11):1047–1056. https://doi. or g/ Aslanpour S, Han S, Schuurmans C, Kurrasch DM (2020b) Neurog2 10. 1016/ 0026- 0495(74) 90071-7 acts as a classical proneural gene in the ventromedial hypothala- García-Calero E, Fernández-Garre P, Martínez S, Puelles L (2008) mus and is required for the early phase of neurogenesis. J Neu- Early mammillary pouch specification in the course of prechordal rosci 40(18):3549–3563. https:// doi. org/ 10. 1523/ JNEUR OSCI. ventralization of the forebrain tegmentum. Dev Biol 320:366–377. 2610- 19. 2020https:// doi. org/ 10. 1016/j. ydbio. 2008. 05. 545 Briscoe J, Sussel L, Serup P, Hartigan-O’Connor D, Jessell TM, Gurdjian ES (1927) The diencephalon of the albino rat. Studies on the Rubenstein JL, Ericson J (1999) Homeobox gene Nkx2.2 and brain of the rat. J Comp Neurol 43:1–144 specification of neuronal identity by graded sonic hedgehog sig- Hahn JD, Sporns O, Watts AG, Swanson LW (2019) Macroscale intrin- nalling. Nature 398:622–627. https:// doi. org/ 10. 1038/ 19315 sic network architecture of the hypothalamus. Proc Natl Acad Sci Büdefeld T, Tobet SA, Majdic G (2012) Steroidogenic factor 1 and the U S A 116:8018–8027. https://doi. or g/10. 1073/ pnas. 18194 48116 central nervous system. J Neuroendocrinol 24(1):225–235. https:// Hashikawa K, Hashikawa Y, Tremblay R, Zhang J, Feng JE, Sabol A, doi. org/ 10. 1111/j. 1365- 2826. 2011. 02174.x Piper WT, Lee H, Rudy B, Lin D (2017) Esr1+ cells in the ven- Canteras NS, Simerly RB, Swanson LW (1994) Organization of projec- tromedial hypothalamus control female aggression. Nat Neurosci tions from the ventromedial nucleus of the hypothalamus: a Pha- 20:1580–1590. https:// doi. org/ 10. 1038/ nn. 4644 seolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol Hauptmann G, Gerster T (2000) Regulatory gene expression patterns 348(1):41–79. https:// doi. org/ 10. 1002/ cne. 90348 0103 reveal transverse and longitudinal subdivisions of the embryonic Cheung CC, Kurrasch DM, Liang JK, Ingraham HA (2013) Genetic zebrafish forebrain. Mech Dev 91(1–2):105–118. https:// doi. org/ labeling of steroidogenic factor-1 (SF-1) neurons in mice reveals 10. 1016/ s0925- 4773(99) 00277-4 ventromedial nucleus of the hypothalamus (VMHH) circuitry Hauptmann G, Söll I, Gerster T (2002) The early embryonic zebrafish beginning at neurogenesis and development of a separate non- forebrain is subdivided into molecularly distinct transverse and SF-1 neuronal cluster in the ventrolateral VMHH. J Comp Neurol longitudinal domains. Brain Res Bull 57:371–375 521:1268–1288. https:// doi. org/ 10. 1002/ cne. 23226 Heimer L, Nauta WJ (1969) The hypothalamic distribution of the stria Corman TS, Bergendahl SE, Epstein DJ (2018) Distinct temporal terminalis in the rat. Brain Res 13:284–297 requirements for Sonic hedgehog signaling in development of Hetherington AW, Ranson SW (1942) The relation of various hypotha- the tuberal hypothalamus. Development. https://doi. or g/10. 1242/ lamic lesions to adiposity in the rat. J Comp Neurol 76:475–499 dev. 167379 Ifft JD (1972) An autoradiographic study of the time of final division Correa SM, Newstrom DW, Warne JP, Flandin P, Cheung CC, Lin- of neurons in rat hypothalamic nuclei. J Comp Neurol. https://doi. Moore AT, Pierce AA, Xu AW, Rubenstein JL, Ingraham HA org/ 10. 1002/ cne. 90144 0204 (2015) An estrogen-responsive module in the ventromedial hypo- Ikeda Y, Luo X, Abbud R, Nilson JH, Parker KL (1995) The nuclear thalamus selectively drives sex-specific activity in females. Cell receptor steroidogenic factor 1 is essential for the formation of the Rep 10(1):62–74. https:// doi. org/ 10. 1016/j. celrep. 2014. 12. 011 ventromedial hypothalamic nucleus. Mol Endocrinol 9(4):478– Davis AM, Seney ML, Stallings NR, Zhao L, Parker KL, Tobet SA 486. https:// doi. org/ 10. 1210/ mend.9. 4. 76590 91 (2004) Loss of steroidogenic factor 1 alters cellular topography in Kennedy A, Kunwar PS, Li LY, Stagkourakis S, Wagenaar DA, Ander- the mouse ventromedial nucleus of the hypothalamus. J Neurobiol son DJ (2020) Stimulus-specific hypothalamic encoding of a per - 60:424–436. https:// doi. org/ 10. 1002/ neu. 20030 sistent defensive state. Nature 586(7831):730–734. https://d oi.o rg/ Dellovade TL, Young M, Ross EP, Henderson R, Caron K, Parker K, 10. 1038/ s41586- 020- 2728-4 Tobet SA (2000) Disruption of the gene encoding SF-1 alters the Khodai T, Luckman SM (2021) Ventromedial nucleus of the hypothala- distribution of hypothalamic neuronal phenotypes. J Comp Neurol mus neurons under the magnifying glass. Endocrinology. https:// 423(4):579–589doi. org/ 10. 1210/ endocr/ bqab1 41 Kim KW, Sohn JW, Kohno D, Xu Y, Williams K, Elmquist JK (2011) Dellovade TL, Davis AM, Ferguson C, Sieghart W, Homanics SF-1 in the ventral medial hypothalamic nucleus: a key regulator GE, Tobet SA (2001) GABA influences the development of 1 3 Brain Structure and Function of homeostasis. Mol Cell Endocrinol 336(1–2):219–223. https:// Meek TH, Nelson JT, Matsen ME, Dorfman MD, Guyenet SJ, Damian doi. org/ 10. 1016/j. mce. 2010. 11. 019 V, Allison MB, Scarlett JM, Nguyen HT, Thaler JP, Olson DP, Kim DW, Yao Z, Graybuck LT, Kim TK, Nguyen TN, Smith KA, Fong Myers MG Jr, Schwartz MW, Morton GJ (2016) Functional iden- O, Yi L, Koulena N, Pierson N, Shah S, Lo L, Pool AH, Oka Y, tification of a neurocircuit regulating blood glucose. Proc Natl Pachter L, Cai L, Tasic B, Zeng H, Anderson DJ (2019) Multi- Acad Sci U S A 113(14):E2073–E2082. https:// doi. org/ 10. 1073/ modal analysis of cell types in a hypothalamic node controlling pnas. 15211 60113 social behavior. Cell 179(3):713–728. https:// doi. org/ 10. 1016/j. Morales-Delgado N, Merchan P, Bardet SM, Ferrán JL, Puelles L, Díaz cell. 2019. 09. 020 C (2011) Topography of somatostatin gene expression relative Kim DW, Liu K, Wang ZQ, Zhang YS, Bathini A, Brown MP, Lin SH, to molecular progenitor domains during ontogeny of the mouse Washington PW, Sun C, Lindtner S, Lee B, Wang H, Shimogori hypothalamus. Front Neuroanat 5:10. https:// doi. org/ 10. 3389/ T, Rubenstein JLR, Blackshaw S (2021) Gene regulatory networks fnana. 2011. 00010 controlling differentiation, survival, and diversification of hypo- Morales-Delgado N, Castro-Robles B, Ferrán JL, Martinez-de-la-Torre thalamic Lhx6-expressing GABAergic neurons. Commun Biol M, Puelles L, Díaz C (2014) Regionalized differentiation of CRH, 4(1):95. https:// doi. org/ 10. 1038/ s42003- 020- 01616-7 TRH, and GHRH peptidergic neurons in the mouse hypothala- Krause WC, Ingraham HA (2017) Origins and functions of the ven- mus. Brain Struct Funct 219:1083–1111. https://doi. or g/10. 1007/ trolateral VMHH: a complex neuronal cluster orchestrating sex s00429- 013- 0554-2 differences in metabolism and behavior. Adv Exp Med Biol Nieuwenhuys R, Puelles L (2016) Towards a new neuromor- 1043:199–213. https:// doi. org/ 10. 1007/ 978-3- 319- 70178-3_ 10 phology. Switz, Springer Int Publ. h t t p s : / / d o i . o r g / 1 0 . 1 0 0 7 / Krieg WJS (1932) The hypothalamus of the albino rat. J Comp Neurol 978-3- 319- 25693-1 55:19–89 Nishi Y, Zhang X, Jeong J, Peterson KA, Vedenko A, Bulyk ML, Hide Kurrasch DM, Cheung CC, Lee FY, Tran PV, Hata K, Ingraham HA WA, McMahon AP (2015) Development 142:3286–3293. https:// (2007) The neonatal ventromedial hypothalamus transcriptome doi. org/ 10. 1242/ dev. 124636 reveals novel markers with spatially distinct patterning. J Neurosci Orquera DP, Nasif S, Low MJ, Rubinstein M, de Souza FSJ (2016) 50:13624–13634. https://doi. or g/10. 1523/ JNEUR OSCI. 2858- 07. Essential function of the transcription factor Rax in the early pat- 2007 terning of the mammalian hypothalamus. Dev Biol 416:212–224. Lagutin OV, Zhu CC, Kobayashi D, Topczewski J, Shimamura K, https:// doi. org/ 10. 1016/j. ydbio. 2016. 05. 021 Puelles L, Russell HR, McKinnon PJ, Solnica-Krezel L, Oliver G Pfaff DW, Sakuma Y (1979a) Deficit in the lordosis reflex of female (2003) Six3 repression of Wnt signaling in the anterior neuroec- rats caused by lesions in the ventromedial nucleus of the hypo- toderm is essential for vertebrate forebrain development. Genes thalamus. J Physiol 288:203–210 Dev 17:368–379. https:// doi. org/ 10. 1101/ gad. 10594 03 Pfaff DW, Sakuma Y (1979b) Facilitation of the lordosis reflex of Lee H, Kim DW, Remedios R, Anthony TE, Chang A, Madisen L, Zeng female rats fromthe ventromedial nucleus of the hypothalamus. J H, Anderson DJ (2014) Scalable control of mounting and attack Physiol 288:189–202 by Esr1+ neurons in the ventromedial hypothalamus. Nature Puelles L (2018) Developmental studies of avian brain organization. 509(7502):627–632. https:// doi. org/ 10. 1038/ natur e13169 Int J Dev Biol 62:207–224. https://doi. or g/10. 1387/ i jdb.17027 9LP Lewis JE, Woodward OR, Nuzzaci D, Smith CA, Adriaenssens AE, Puelles L (2019) Survey of midbrain, diencephalon, and hypothalamus Billing L, Brighton C, Phillips BU, Tadross JA, Kinston SJ, Ciab- neuroanatomic terms whose prosomeric definition conflicts with atti E, Göttgens B, Tripodi M, Hornigold D, Baker D, Gribble FM, columnar tradition. Front Neuroanat 13:20. htt ps: // do i. org/ 10 . Reimann F (2022) Relaxin/insulin-like family peptide receptor 4 3389/ fnana. 2019. 00020 (Rxfp4) expressing hypothalamic neurons modulate food intake Puelles L (2022) Prosomeric classification of retinorecipient centers: a and preference in mice. Mol Metab. https:// doi. org/ 10. 1016/j. new causal scenario. Brain Struct Funct 227:1171–1193. https:// molmet. 2022. 101604doi. org/ 10. 1007/ s00429- 022- 02461-6 Lin D, Boyle MP, Dollar P, Lee H, Lein ES, Perona P, Anderson DJ Puelles L, Martinez S (2013) Patterning of the Diencephalon. John LR (2011) Functional identification of an aggression locus in the Rubenstein, Pasko Rakic, patterning and cell type specification mouse hypothalamus. Nature 470(7333):221–226. https:// doi. in the developing CNS and PNS 8. Academic Press, USA, pp org/ 10. 1038/ natur e09736 151–172. https:// doi. org/ 10. 1016/ B978-0- 12- 397265- 1. 00048-4 López-González L, Alonso A, García-Calero E, de Puelles E, Puelles Puelles L, Rubenstein JL (2015) A new scenario of hypothalamic L (2021) Tangential intrahypothalamic migration of the mouse organization: rationale of new hypotheses introduced in the ventral premamillary nucleus and Fgf8 signaling. Front Cell Dev updated prosomeric model. Front Neuroanat 9:27. ht tps:/ / doi. Biol. https:// doi. org/ 10. 3389/ fcell. 2021. 676121org/ 10. 3389/ fnana. 2015. 00027 Lu F, Kar D, Gruenig N, Zhang ZW, Cousins N, Rodgers HM, Swin- Puelles L, Martinez-de-la-Torre M, Bardet S, Rubenstein JLR (2012) dell EC, Jamrich M, Schuurmans C, Mathers PH, Kurrasch DM Hypothalamus. The Mouse Nervous System, Watson C, Paxi- (2013) Rax is a selector gene for mediobasal hypothalamic cell nos G, Puelles L San Diego. Academic Press/Elsevier, CA, pp types. J Neurosci 33:259–272. https:// doi. org/ 10. 1523/ JNEUR 221–312 OSCI. 0913- 12. 2013 Qiu M, Shimamura K, Sussel L, Chen S, Rubenstein JL (1998) Con- Ma T, Wong SZH, Lee B, Ming GL, Song H (2021) Decoding neuronal trol of anteroposterior and dorsoventral domains of Nkx-6.1 gene composition and ontogeny of individual hypothalamic nuclei. expression relative to other Nkx genes during vertebrate CNS Neuron 109(7):1150–1167. https:// doi. or g/ 10. 1016/j. neur on. development. Mech Dev 72:77–88. https://doi. or g/10. 1016/ s0925- 2021. 01. 0264773(98) 00018-5 McClellan KM, Parker KL, Tobet S (2006) Development of the ven- Ramón y Cajal S (1911) Histologie du Système Nerveux de l’homme tromedial nucleus of the hypothalamus. Front Neuroendocrinol et des Vertébrés. Vol. 2. Re-edit. of 1954 Madrid: CSIC 27(2):193–209. https:// doi. org/ 10. 1016/j. yfrne. 2006. 02. 002 Romanov RA, Tretiakov EO, Kastriti ME, Zupancic M, Häring M, McClellan KM, Calver AR, Tobet SA (2008) GABAB receptors role in Korchynska S, Popadin K, Benevento M, Rebernik P, Lallemend cell migration and positioning within the ventromedial nucleus of F, Nishimori K, Clotman F, Andrews WD, Parnavelas JG, Farlik the hypothalamus. Neuroscience 151(4):1119–1131. https:// doi. M, Bock C, Adameyko I, Hökfelt T, Keimpema E, Harkany T org/ 10. 1016/j. neuro scien ce. 2007. 11. 048 (2020) Molecular design of hypothalamus development. Nature 582(7811):246–252. https:// doi. org/ 10. 1038/ s41586- 020- 2266-0 1 3 Brain Structure and Function Saper CB, Lowell BB (2014) The hypothalamus. Curr Biol. https://doi. Tran PV, Lee MB, Marín O, Xu B, Jones KR, Reichardt LF, Ruben- org/ 10. 1016/j. cub. 2014. 10. 023 stein JR, Ingraham HA (2003) Requirement of the orphan nuclear Saper CB, Swanson LW, Cowan WM (1976) The efferent connec- receptor SF-1 in terminal differentiation of ventromedial hypotha- tions of the ventromedial nucleus of the hypothalamus of the rat. lamic neurons. Mol Cell Neurosci 4:441–453. https:// doi. org/ 10. J Comp Neurol 169(4):409–442. https :// d oi. o r g/ 10 . 100 2/ c ne. 1016/ s1044- 7431(03) 00027-7 90169 0403 Trowe MO, Zhao L, Weiss AC, Christoffels V, Epstein DJ, Kispert Shimada M, Nakamura T (1973) Time of neuron origin in mouse hypo- A (2013) Inhibition of Sox2-dependent activation of Shh in the thalamic nuclei. Exp Neurol 41:163–173. https://doi. or g/10. 1016/ ventral diencephalon by Tbx3 is required for formation of the 0014- 4886(73) 90187-8 neurohypophysis. Development 140(11):2299–2309. https:// doi. Shimamura K, Hirano S, McMahon AP, Takeichi M (1994) Wnt-org/ 10. 1242/ dev. 094524 1-dependent regulation of local E-cadherin and alpha N-catenin van den Akker WM, Brox A, Puelles L, Durston AJ, Medina L (2008) expression in the embryonic mouse brain. Development Comparative functional analysis provides evidence for a crucial 120:2225–2234 role for the homeobox gene Nkx2.1/Titf-1 in forebrain evolution. Shimamura K, Hartigan DJ, Martinez S, Puelles L, Rubenstein JL J Comp Neurol 506:211–223. https:// doi. org/ 10. 1002/ cne. 21542 (1995) Longitudinal organization of the anterior neural plate and Van Houten M, Brawer JR (1978) Cytology of neurons of the hypotha- neural tube. Development 121:3923–3933 lamic ventromedial nucleus in the adult male rat. J Comp Neurol Shimogawa Y, Sakuma Y, Yamanouchi K (2015) Efferent and afferent 178(1):89–116. https:// doi. org/ 10. 1002/ cne. 90178 0106 connections of the ventromedial hypothalamic nucleus determined van Veen JE, Kammel LG, Bunda PC, Shum M, Reid MS, Massa MG, by neural tracer analysis: implications for lordosis regulation in Arneson D, Park JW, Zhang Z, Joseph AM, Hrncir H, Liesa M, female rats. Neurosci Res 91:19–33. https:// doi. org/ 10. 1016/j. Arnold AP, Yang X, Correa SM (2020) Hypothalamic estrogen neures. 2014. 10. 016 receptor alpha establishes a sexually dimorphic regulatory node Shimogori T, Lee DA, Miranda-Angulo A, Yang Y, Wang H, Jiang L, of energy expenditure. Nat Metab 2(4):351–363. https:// doi. org/ Yoshida AC, Kataoka A, Mashiko H, Avetisyan M, Qi L, Qian 10. 1038/ s42255- 020- 0189-6 J, Blackshaw S (2010) A genomic atlas of mouse hypothalamic Wang L, Chen IZ, Lin D (2015) Collateral pathways from the ven- development. Nat Neurosci 13:767–775. https://doi. or g/10. 1038/ tromedial hypothalamus mediate defensive behaviors. Neuron nn. 2545 85(6):1344–1358. https:// doi. org/ 10. 1016/j. neuron. 2014. 12. 025 Shinoda K, Lei H, Yoshii H, Nomura M, Nagano M, Shiba H, Sasaki H, Yang CF, Chiang MC, Gray DC, Prabhakaran M, Alvarado M, Juntti Osawa Y, Ninomiya Y, Niwa O (1995) Developmental defects of SA, Unger EK, Wells JA, Shah NM (2013) Sexually dimorphic the ventromedial hypothalamic nucleus and pituitary gonadotroph neurons in the ventromedial hypothalamus govern mating in both in the Ftz-F1 disrupted mice. Dev Dyn 204(1):22–29. https://doi. sexes and aggression in males. Cell 153(4):896–909. https:// doi. org/ 10. 1002/ aja. 10020 40104org/ 10. 1016/j. cell. 2013. 04. 017 Skidmore JM, Cramer JD, Martin JF, Martin DM (2008) Cre fate map- Ziegler DR, Cullinan WE, Herman JP (2002) Distribution of vesicular ping reveals lineage specific defects in neuronal migration with glutamate transporter mRNA in rat hypothalamus. J Comp Neurol loss of Pitx2 function in the developing mouse hypothalamus and 448(3):217–229. https:// doi. org/ 10. 1002/ cne. 10257 subthalamic nucleus. Mol Cell Neurosci 37:696–707. https://doi. org/ 10. 1016/j. mcn. 2007. 12. 015 Publisher's Note Springer Nature remains neutral with regard to Swanson LW (2012) Brain architecture. Oxford University Press, New jurisdictional claims in published maps and institutional affiliations. York 1 3

Journal

Brain Structure and Function – Springer Journals

Published: Mar 1, 2023

Keywords: Ventromedial hypothalamic nucleus; Tuberal hypothalamus; Acroterminal; Tangential migration; Prosomeric model; Nkx2.2

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

Populational heterogeneity and partial migratory origin of the ventromedial hypothalamic nucleus: genoarchitectonic analysis in the mouse

References (91)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies