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

Learn More →

Loss of cholinergic innervation differentially affects eNOS-mediated blood flow, drainage of Aβ and cerebral amyloid angiopathy in the cortex and hippocampus of adult mice

Loss of cholinergic innervation differentially affects eNOS-mediated blood flow, drainage of Aβ... Vascular dysregulation and cholinergic basal forebrain degeneration are both early pathological events in the devel- opment of Alzheimer’s disease (AD). Acetylcholine contributes to localised arterial dilatation and increased cerebral blood flow (CBF) during neurovascular coupling via activation of endothelial nitric oxide synthase (eNOS). Decreased vascular reactivity is suggested to contribute to impaired clearance of β-amyloid (Aβ) along intramural periarterial drainage (IPAD) pathways of the brain, leading to the development of cerebral amyloid angiopathy (CAA). However, the possible relationship between loss of cholinergic innervation, impaired vasoreactivity and reduced clearance of Aβ from the brain has not been previously investigated. In the present study, intracerebroventricular administration of mu-saporin resulted in significant death of cholinergic neurons and fibres in the medial septum, cortex and hip - pocampus of C57BL/6 mice. Arterial spin labelling MRI revealed a loss of CBF response to stimulation of eNOS by the Rho-kinase inhibitor fasudil hydrochloride in the cortex of denervated mice. By contrast, the hippocampus remained responsive to drug treatment, in association with altered eNOS expression. Fasudil hydrochloride significantly increased IPAD in the hippocampus of both control and saporin-treated mice, while increased clearance from the cor- tex was only observed in control animals. Administration of mu-saporin in the TetOAPPSweInd mouse model of AD was associated with a significant and selective increase in Aβ40-positive CAA. These findings support the importance of the interrelationship between cholinergic innervation and vascular function in the aetiology and/or progression of CAA and suggest that combined eNOS/cholinergic therapies may improve the efficiency of Aβ removal from the brain and reduce its deposition as CAA. Keywords: Alzheimer’s disease, Cholinergic, Clearance, eNOS, Vascular reactivity be one of the earliest indicators of the development of Introduction AD [38, 39] and differential perfusion of AD-sensitive Increasing evidence suggests that structural and func- brain areas such as the hippocampus, frontal and tempo- tional alterations of the cerebrovasculature contribute to ral lobes are present in people both with mild cognitive the aetiology and/or progression of Alzheimer’s disease impairment and dementia [2, 18, 34]. (AD). In fact, vascular pathology has been suggested to Cerebral amyloid angiopathy (CAA) is the most com- mon form of cerebrovascular pathology in AD [42] and is characterised by the deposition of β-amyloid (Aβ) pep- *Correspondence: c.hawkes@lancaster.ac.uk Department of Biomedical and Life Sciences, Lancaster University, tides in the walls of cerebral arteries and capillaries [90]. Lancaster, UK While parenchymal plaques are made up predominantly Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, 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 included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Nizari et al. acta neuropathol commun (2021) 9:12 Page 2 of 17 of Aβ42, Aβ40 is more commonly observed in CAA. that stimulate calcium release and binding of calcium- CAA develops topographically, presenting initially in calmodulin to eNOS [25]. ACh activation of eNOS can the occipital lobe, followed by the temporal, frontal and also occur via the insulin-receptor substrate/PI3K/Akt parietal lobes, then in the hippocampus and entorhinal pathway [96] and stimulation of the PI3K/Akt/eNOS cortex at later stages [5, 83, 84]. In addition to causing pathway by the selective Rho- associated, coiled-coil dysfunction and death of mural and endothelial cells, containing protein kinase (ROCK) inhibitor fasudil recent studies suggest that CAA contributes to impaired hydrochloride, has been shown to increase CBF in mice hemodynamic responses in both individuals with AD and and humans [59, 68]. Although multiple downstream people with hereditary CAA [4, 63, 77, 87, 94]. signalling pathways are regulated by ROCK activity [45, A key pathological feature of sporadic CAA is a failure 76], several studies have reported no effect of fasudil −/− of clearance of Aβ from the brain, which is mediated via hydrochloride on cerebral haemodynamics in eNOS enzymatic degradation, uptake in microglia and astro- mice, suggesting that eNOS is the principal NOS iso- cytes and transcytosis across the blood–brain barrier [54, form targeted by fasudil hydrochloride [68, 78]. 55, 93]. Aβ is also removed from the brain along the walls Decreased expression of eNOS has been reported in of the capillaries and arteries via intramural periarterial the occipital cortex in AD, an area of the brain that is drainage (IPAD) and/or glymphatic drainage [7, 36, 56]. hypoperfused in AD [12]. Conversely, eNOS and induc- The IPAD hypothesis of Aβ clearance is based in part ible NOS (iNOS) activity have been shown to be signifi - on experimental observations that nanoparticles, solutes cantly increased in the temporal and frontal cortices of and Aβ injected into the interstitial fluid (ISF) of deep AD patients [20], in association with hyperperfusion of brain structures are transported along and localise to cer- those areas [34]. Several recent studies have reported ebrovascular basement membranes (CVBM) in cortical that endogenous CAA load is increased in eNOS-defi - and leptomeningeal vessels [3, 13, 27–29]. In the mouse cient mice in the absence of alterations in parenchymal brain, this process occurs very rapidly, within 5–10  min Aβ or increased Aβ production [6, 81], suggesting that of injection [8, 13, 27–29]. Since the pattern of distribu- dysfunction of eNOS may also contribute to the aeti- tion of solutes closely mimics that of Aβ accumulation ology of CAA and that this may be related to impair- in CAA and other angiopathies [14, 41, 85], failure of ments in Aβ clearance from the brain. IPAD is a key element of CAA pathology. However, as Loss of cholinergic neurons as an early pathologi- the movement of solutes along CVBMs is counter to the cal feature of AD has been known since the 1980s and direction of blood flow, the driving force that underlies underpins the rationale for the current clinical use of IPAD is still unknown. Recent mathematical modeling AChEIs for the treatment of AD [9, 22, 66]. Two recent suggests that oscillating pulsatile flow generated by the findings from the Alzheimer’s Disease Neuroimaging focal contraction and relaxation of arteries drives IPAD Initiative have reported that vascular dysregulation is [1] and this is supported by recent experimental data an early predictor of the progression to AD and that [3, 64, 88]. Localised arterial dilatation and contraction loss of volume in the basal forebrain precedes patholog- can occur both spontaneously (e.g. vasomotion) and in ical changes in the entorhinal cortex of individuals who response to neuronal activity (e.g. neurovascular cou- went on to develop AD [38, 73]. These findings suggest pling, NVC) and both mechanisms have been shown to that the interplay between loss of cholinergic innerva- be decreased in AD [19, 69, 79]. tion and vascular dysfunction may be important in the Smooth muscle cells that regulate arterial contrac- aetiology of AD. However, although some pathological tion and contribute to the regulation of cerebral blood studies have examined cholinergic loss at the neurovas- flow (CBF) in the cortex and hippocampus receive cular unit under experimental conditions [61] and in innervation from cholinergic neurons that originate AD [60], less has been done to directly investigate the in the basal forebrain. Release of acetylcholine (ACh) functional outcome of perivascular cholinergic dener- via stimulation of the basal forebrain or increasing vation. The aim of this study was to test the hypoth - cholinergic tone using acetylcholinesterase inhibitors esis that loss of cholinergic innervation decreases (AChEIs) has been shown to increase CBF in the cortex CBF and IPAD of Aβ from the cortex and hippocam- and hippocampus [50, 72]. ACh induces vasodilation pus of wildtype mice, leading to increased CAA in the primarily by stimulating the production of nitric oxide TetOAPPSweInd model of AD. (NO) via activation of endothelial nitric oxide synthase (eNOS) [23, 97], although stimulation of neuronal NOS Materials and methods (nNOS)-containing interneurons can also increase CBF Animals [15, 89]. ACh-induced activation of eNOS is medi- C57BL/6 mice were bred at the Open University (OU, ated principally by binding to muscarinic receptors Milton Keynes, UK) and the University of Southampton N izari et al. acta neuropathol commun (2021) 9:12 Page 3 of 17 (Southampton, UK). TetOAPPSweInd mice developed by pre-absorbing purified human Aβ40 peptide with the by Dr Joanna Jankowsky (Baylor College of Medicine, anti-Aβ40 antibody (10:1 molar ratio) alone or in combi- Texas, US) [40] were a generous gift from Dr JoAnne nation with the anti-Aβ42 antibody for 1.5 h at RT before McLaurin (Sunnybrook Research Centre, Toronto, Can- proceeding with tissue incubation and development as ada) and were also bred on a C57BL/6 background. Food described above (Additional file  2: Fig.  1a–d). Photomi- and water were provided ad libitum. All animal work was crographs were obtained using a Nikon Eclipse 80i light approved by the Animal Welfare and Ethics Research microscope (Nikon UK Limited, Surrey, UK) and images Boards (AWERB) at the OU, University of Southampton from the hippocampus and cortex (n = 6 control and and UCL in accordance with Home Office regulations n = 7 saporin) were analysed using Fiji (NIH, Maryland, and project licences (PPL 70/8507 and PPL 30/3095) USA). under the Animals (Scientific Procedures) Act 1986. For single labelling immunofluorescence, sections were washed in 0.01  M PBS, blocked with serum and incu- Mu‑Saporin administration bated overnight at 4  °C with anti-choline acetyltrans- 8–10 week old male C57BL/6 mice and 4-month old male ferase (ChAT; 1:75), anti-p75NTR (1:350), anti-GFAP and female TetOAPPSweInd mice were used for saporin (1:500), anti-Iba1 (1:500) or anti-laminin (1:350). Sec- injections. Mice were anesthetised under isoflurane gas tions were incubated with the appropriate fluorophore- and placed into a stereotaxic frame (Kopf instruments, conjugated secondary antibodies and coverslipped using CA, USA). Analgesia was administered intraperitoneally Mowiol (Sigma, Dorset, UK) containing 0.1% v/v Citif- (Carprieve, 5% w/v, 0.32  ml/kg, Norbrook, Northamp- luor (Citifluor ltd, London, UK). tonshire, UK) and a topical anaesthetic (Cryogesic (ethyl For multiple labelling fluorescent immunohistochem - chloride), Acorus Therapeutics Ltd, Chester, UK) was istry, sections underwent antigen retrieval (Additional applied before making a midline incision. 0.5 µL of mu- file  1: Table  1) and were then incubated overnight at saporin (0.596  µg/µl, Advanced Targeting Systems, CA, 4  °C with either i) anti-ChAT (1:75) and anti-p75NTR USA) or 0.9% sterile saline was injected into the left (1:400), or ii) with anti-NOS (1:200) or anti-eNOS (1:200) and right lateral ventricles (coordinates from Bregma: in combination with anti-GFAP (1:2000) or anti-Iba1 AP = − 0.4  mm, ML = ∓ 1.0  mm, DV = − 2.3  mm) using (1:500). Sections were then incubated for 2 h at RT with a 33 gauge Hamilton syringe. Mice were able to self- the appropriate fluorophore-conjugated secondary anti - administer sugar free jelly containing Carprofen (250  µg bodies, washed in PBS and coverslipped as above. The in 500 µl jelly, Zoetis, London, UK) for 1 week post-sur- specificity of the fluorescently-conjugated secondary gery. Animals were randomly assigned to receive either antibodies was verified by omitting the primary antibod - saline or saporin and all experimenters were blinded to ies (Additional file 2: Fig. 1e–h). treatment until statistical analysis. For all fluorescent imaging, photomicrographs were obtained using a Leica SP5 confocal microscope using Immunohistochemistry the same gain and intensity and maximum projection 45 days after surgery, mice were deeply anesthetised and images were exported to Adobe Photoshop 2020 or Fiji. perfused intracardially with 0.01  M phosphate buffered saline (PBS) followed by 4% paraformaldehyde (PFA). Quantification of immunohistochemistry Brains were post-fixed in 4% PFA overnight, sectioned The density of ChAT staining (neuronal cell bodies or (20 µm thickness) using a cryostat and stored at − 20  °C. fibres) in each brain region was quantified from low mag - Details of primary and secondary antibodies used for nification images by calculating the percentage area cov - immunohistochemistry are listed in Additional file  1: ered by staining using the “Analyze particle” function in Table 1. Fiji (NIH. Maryland, USA). For quantification of staining For enzyme-linked immunohistochemistry, sections in the hippocampus, overlapping images were stitched were washed in 0.01 M PBS, incubated with 3% hydrogen together and values from both the ipsilateral and con- peroxide, rinsed in PBS and treated with 70% formic acid tralateral hemispheres were averaged for each animal. For for 45 s. Sections were then blocked in 15% normal don- quantification of cortical images, six random non-over - key or goat serum (NDS, Sigma-Aldrich, Dorset, UK), lapping images spanning the somatosensory cortex of followed by incubation overnight at 4 °C with anti-Aβ40 the ipsilateral cortex to the somatosensory cortex of the (1:100) or anti-Aβ42 (1:100). The next day, sections were contralateral cortex were captured and averaged per ani- incubated with biotinylated anti-rabbit (1:400) and devel- mal. A single low magnification image/animal was used oped using glucose oxidase enhancement with DAB as to quantify ChAT staining in the medial septum. The per - chromagen (Sigma-Aldrich, Dorset, UK). The specificity centage area containing microglia, astrocytes and blood of the anti-Aβ40 and anti-Aβ42 antibodies was verified vessels was also calculated using the ‘Analyze particle” Nizari et al. acta neuropathol commun (2021) 9:12 Page 4 of 17 function in Fiji. Additionally, for anti-laminin staining, in overdose of sodium pentobarbitone and perfused with order to quantify density by vessel type, a mask was set 0.01  M PBS. Brains were removed, dissected for hip- to select capillaries (0–100  µm ) or large-diameter ves- pocampus and cortex, snap frozen and stored at − 80  °C sels (101  µm -infinity) using images calibrated accord - until use. Tissues were homogenised in RIPA lysis buffer ing to the scale bar. The degree of colocalization between (20 mM Tris pH 8.0, 0.15 M NaCl, 1.27 mM EDTA, 1 ml ChAT and p75NTR and between e/NOS and laminin Igepal, 0.1% SDS, 50 mM NaF, 1.48 mM NaVO contain- was determined from three images/region/animal taken ing 1:100 Protease inhibitor cocktail [Merck Millipore, at × 40 magnification using Pearson’s correlation coeffi - UK]), centrifuged at 10,000  g at 4  °C for 10  min and the cient (PCC, Coloc 2 plugin) in Fiji. For quantification of supernatant was collected. 30 μg or 40 μg of proteins Aβ staining, images from the hippocampus and cortex were separated by gel electrophoresis and membranes taken at × 4 magnification were stitched together, images were then blocked in 8% non-fat milk before incubation were converted to 8 bit greyscale images and the % area with anti-eNOS (1:5000, Cell Signalling Technology, Lon- covered by CAA and plaques were quantified separately don, UK) or anti-nNOS (1:250, Cell Signalling Technol- in Fiji. ogy) overnight at 4  °C. Membranes were then washed in TBST before being incubated in HRP-conjugated Arterial spin labelling MRI anti-rabbbit (1:5000, Fisher Scientific) for 1  h at room Approximately 5  weeks after ICV injection, C57BL/6 temperature and developed using an enhanced chemi- mice that received saline (n = 7) or mu-saporin (n = 7) luminescence kit (GE Healthcare, Little Chalfont, UK). were transported to UCL Centre for Advanced Imag- Membranes were then stripped and re-probed with anti- ing and allowed to acclimate for 1  week before imaging. GAPDH (1:50,000, Sigma-Aldrich) to ensure equal pro- A 9.4 T VNMRS horizontal bore scanner (Agilent Inc., tein loading. Optical density of the bands was quantified Santa Clara, CA, USA) with a 72 mm inner diameter vol- and normalised to GAPDH levels using Fiji. ume coil and 2 channel array head coil (Rapid Biomedi- cal, Columbus, OH, USA) was used for radio frequency Assessment of IPAD transmission and signal detection. Mice were initially 45  days after injection with saline or saporin, mice were anaesthetised under 2% isoflurane in medical air and anesthetised with isoflurane and placed into a stere - maintained under 1.5% during imaging. A rectal probe otaxic frame. For hippocampal injections, 0.5 µL of and a pressure pad (SA Instruments, Stony Brook, NY, 50  µM human Aβ40 HiLyte Fluor 555 (AnaSpec, Cali- USA) were used to measure core temperature and moni- fornia, USA) was injected into the left hippocampus (co- tor respiration throughout the procedure. Heated water ordinates from Bregma: AP = − 1.9  mm, ML = 1.5  mm, tubing and a warm air blower using a feedback system DV = − 1.7 mm, n = 16 control and n = 14 saporin). Mice (SA Instruments, Stony Brook, NY, USA) was used to were perfused with PBS and 4% PFA 5 min post-injection. regulate the temperature of the mice to 37 °C. Following For cortical injections, control (n = 8) and saporin (n = 7) a 5 min acquisition of baseline CBF, mice were adminis- mice were injected with 0.25 µL of 50  µM Aβ40 HiLyte tered 10 mg/kg Fasudil hydrochloride i.p. (Tokyo Chemi- Fluor 555 into the right cortex (co-ordinates from cal Industries, Tokyo, Japan) and re-imaged 10  min Bregma AP = − 2  mm, ML = − 1.5  mm, DV = − 0.5  mm) later for an additional 5  min. At the end of the imaging and mice were perfused 2.5 min later. All injections were experiments, mice were perfusion fixed with 4% PFA carried out at a rate of 0.2 µL/min using a 33 gauge Ham- and their brains collected for immunohistochemistry. A ilton syringe and the injection needle was left in situ for total of 15 brain image slices were acquired with a thick- 2  min to avoid reflux. A separate group of control and ness of 1 mm and an ‘in-plane’ resolution of 0.28 mm per saporin-treated mice (n = 5/group) were administered mouse per experiment. Statistical Parametric Mapping fasudil hydrochloride (10  mg/kg, i.p.) 10  min before (SPM, http://www.fil.ion.ucl.ac.uk/spm/) was applied to intracerebral injections. Tissue sections were processed perfusion-weighted acquired ASL images [91]. Acquired for double-labeling immunohistochemistry as described images were processed using a Matlab (Mathworks, MA, above using anti-laminin (1:350) and anti-α smooth mus- USA) customised script. Regions of interest (cortex and cle actin conjugated to FITC (1:350; Additional file  1: hippocampus) were then manually traced on a single slice Table  1). Brain sections that were ≥ 400 μm away from and quantified using a Matlab script which converted the site of injection were imaged for quantification. The pixel intensity into CBF (ml/100 g/min) [91]. number of capillaries, arteries and veins that contained Aβ40 HiLyte Fluor 555 within each image were counted Western blotting manually and divided by the total area analysed, as 45  days post-injection with saline (n = 7) or mu- described previously [27, 29, 61]. p75-saporin (n = 6), C57BL/6 mice were given an N izari et al. acta neuropathol commun (2021) 9:12 Page 5 of 17 Statistical analysis increase in CBF in the control, but not the saporin group Data were tested for normality using the Shapiro–Wilk compared to baseline (Fig.  2b). In addition, control mice test and the ROUT test was used to identify and remove that were administered fasudil hydrochloride had a sig- statistical outliers. Comparisons between control and nificantly higher CBF compared to saporin-treated mice saporin-treated mice were analysed using two-tailed given the drug (Fig. 2b). These results suggest that dener - Student’s t test or Mann–Whitney U test where data vated hippocampal vessels were still responsive to eNOS were not normally distributed. Analysis of baseline vs stimulation, while cortical vessels were not. stimulated CBF was carried out using paired one-tailed t-test and Wilcoxon matched-pairs signed rank test. Dif- eNOS protein expression in the hippocampus and cortex ferences in NOS activity were analysed using one-way is differentially affected by saporin treatment ANOVA with Sidak post hoc test. Differences in counts To determine whether regional differences in the respon - of Aβ40-positive vessels within each brain region were siveness to fasudil hydrochloride were due to differences analysed using a one-way ANOVA with Sidak post hoc in the levels of NOS expression, cortical and hippocam- analysis or Kruskal–Wallis test with Dunn’s post hoc. In pal tissues were assessed by Western blotting using all cases, significance was set at p < 0.05 and data are dis- eNOS and nNOS-specific antibodies. eNOS expression played as mean ± SEM. was significantly higher in the hippocampus of saporin- treated mice compared to controls (Fig. 3a), while no sta- Results tistically significant differences were observed between mu‑Saporin induces loss of cholinergic neurons and fibres control and saporin mice in the cortex (Fig.  3b). Levels In control mice, immunohistochemistry for ChAT of nNOS did not differ significantly between control and labelled neurons in the medial septum (MS), diago- saporin-treated mice in either the hippocampus or cor- nal band of Broca (DBB) and striatum (Fig.  1a). ChAT- tex (Fig. 3c, d). To determine if NOS expression may have positive fibres in the hippocampus and cortex were also been influenced by possible differences in vessel densi - observed in these animals (Fig.  1b, c). Colocalization ties, the vascular expression of NOS in laminin-positive was noted between ChAT and p75NTR in the majority vessels was quantified in control and saporin tissues of basal forebrain neurons (PCC = 0.68) as well as in fibre (Fig.  3e–h). Quantification of the NOS-to-laminin ratio projections in the hippocampus and cortex (PCC = 0.25 confirmed the significant decrease in NOS expression in and 0.24, respectively) in control animals (Additional the cortex of saporin-treated mice (Fig.  3m). However, file  3: Fig.  2). Administration of mu-saporin induced a the NOS ratio in the hippocampus did not differ signifi - significant loss of ChAT-positive, p75NTR-positive neu - cantly between control and saporin mice (Fig.  3m). We rons in the MS and DBB (Fig.  1d, g, Additional file  3: also observed some NOS expression in glial cells in the Fig.  2), as well as fibres in the hippocampus (Fig.  1e, h, hippocampus of saporin-treated mice. To determine if Additional file  3: Fig. 2) and cortex (Fig.  1f, i, Additional the increased eNOS expression detected by Western blot file  3: Fig.  2), confirming the usefulness of the model to was due to expression in glial cells, hippocampal sections induce significant death of basal forebrain cholinergic were stained with anti-eNOS and anti-GFAP or anti- neurons and their projection fibres. Iba1 (Fig. 3i–l). These results confirmed minimal expres - sion of eNOS in astrocytes, but some colocalization of Cholinergic loss decreases eNOS‑mediated cerebral blood eNOS with Iba1-positive microglia, which was higher in flow in the cortex but not the hippocampus saporin-treated mice, although this did not reach statisti- We have previously found that mu-saporin causes loss of cal significance (p = 0.13, Fig.  3n). These results suggest cholinergic innervation of cerebral blood vessels and that that levels of eNOS are downregulated in the cortex, and this denervation is more pronounced in the cortex than upregulated in the hippocampus of saporin-treated mice the hippocampus [61]. To assess the effect of this loss on and that increased eNOS expression in the hippocampus baseline and evoked CBF, arterial spin labelling MRI was may be due in part to upregulation by microglia. used to image cerebral perfusion in the hippocampus and cortex of control and saporin-treated mice. Baseline CBF Administration of fasudil hydrochloride increases IPAD did not differ between control and saporin mice in either in the hippocampus but not cortex of denervated mice the hippocampus or cortex (Fig.  2a, b). Administration Our previous work has shown that saporin treatment of fasudil hydrochloride caused a significant increase in significantly decreases cholinergic innervation of arterial hippocampal CBF relative to baseline in both control smooth muscle cells in the hippocampus and cortex [61]. and saporin-treated mice (Fig.  2a). The degree of CBF In vitro modelling supports the hypothesis that the local- increase was similar between treatment groups (Fig. 2 a). ised arterial pulsations that regulate CBF [32] also pro- In the cortex, fasudil hydrochloride induced a significant vide the principle driving force for solute clearance from Nizari et al. acta neuropathol commun (2021) 9:12 Page 6 of 17 Fig. 1 Saporin administration kills cholinergic neurons and fibres in wildtype mice. a–f Photomicrographs of ChAT staining in the medial septum and diagonal band of Broca (a and d), hippocampus (b and e) and cortex (c and f) in control (a–c) and mu-saporin treated C57BL/6 mice (d–f). (g‑i), Quantification of % area covered by ChAT-positive neurons in the medial septum (g) and fibres in the hippocampus (h) and cortex (i). Scale bar = 100 μm. n = 5/group, *p < 0.05, **p < 0.01 the brain via IPAD [1]. To determine if loss of choliner- in capillaries and arteries in both the hippocampus and gic innervation altered IPAD, the pattern of distribution cortex (Fig.  4a–f ). Quantification of the number of hip - of human Aβ40-AF555 was evaluated following injection pocampal blood vessels that contained Aβ showed no into the hippocampus or cortex of control and saporin- difference between control and saporin-treated mice treated C57Bl/6 mice under physiologic and stimulated under baseline physiological conditions (Fig. 4c). Admin- conditions. Triple-labelling immunohistochemistry istration of fasudil hydrochloride resulted in significantly demonstrated the presence of Aβ40-AF555 primarily more Aβ-positive blood vessels in both control and N izari et al. acta neuropathol commun (2021) 9:12 Page 7 of 17 These findings confirm that IPAD was not affected by differences in vessel density or glial activation and indicate that clearance of Aβ was stimulated by fasudil hydrochloride and that this responsiveness remains intact in the hippocampus, but not the cortex of dener- vated mice. Loss of cholinergic innervation increases CAA in the hippocampus of TetOAPP mic ‑ e Fig. 2 The hippocampus, but not cortex, of denervated mice To evaluate if cholinergic denervation potentiated Aβ remains responsive to eNOS-stimulated increase in CBF. a and b pathology, 4-month old TetO-APPSweInd mice were Quantification of cerebral blood flow (CBF) in the hippocampus (a) and cortex (b) of control (con) and saporin-treated mice (sap) at administered saline or mu-saporin. Unexpectedly and baseline and 10 min after administration of fasudil hydrochloride in contrast to the observations made in the C57BL/6 (+F), averaged over 5 min. n = 5–7/group, *p < 0.05 mice, no significant differences were noted in the num - ber of ChAT-positive neurons in the medial septum between control and saporin-treated mice (Fig.  5a, d, g). Significantly fewer cholinergic fibres were observed saporin animals compared to baseline (Fig.  4b, c). How- in the hippocampus of saporin vs control mice (Fig.  5b, ever, fasudil treatment did not affect hippocampal vessel e, h), while ChAT fibre density in the cortex was also counts between control vs. saporin mice (Fig. 4c). unaffected by saporin treatment (Fig.  5c, f, i). To deter- Preliminary assessment of IPAD in the cortex using mine if the attenuated effect of saporin in the TetO- the same parameters as those in the hippocampus (e.g. APPSweInd mice was due to endogenous differences in 0.5 μL Aβ40-AF555 + 5  min clearance) revealed a much ChAT and p75NTR expression, fibre appearance and smaller bolus of Aβ at the site of injection and very few density was compared between TetO-APPSweInd mice Aβ-positive vessels were visible at 400 µm away from the and wildtype littermates. The morphology of fibres in injection site compared to the hippocampus (Additional the TetO-APPSweInd appeared dystrophic, with swollen file  4: Fig.  3a and b). Following a series of modifications varicosities and shorter processes than that of cholinergic (Additional file  4: Fig.  3c–e), the injection protocol for fibres in the wildtype mice (Figs.  1b, c, 5b, c). The den - cortical injections was adapted to 0.25 µL Aβ40-AF555 sity of ChAT-positive neurons in the MS was significantly with 2.5  min post-injection time (Fig.  4d–f ), to allow for higher in TetO-APPSweInd mice compared to wildtype sufficient numbers of Aβ40-positive vessels to be counted. animals, although no differences in hippocampal or cor - Similarly to the hippocampus, quantification of corti - tical ChAT fibre density were observed between strains cal vessels that contained Aβ revealed no baseline differ - (Fig.  5j). Analysis of p75NTR expression showed signifi - ences between control and saporin-treated mice (Fig. 4f ). cantly lower receptor expression in the hippocampus of In control animals, administration of fasudil hydrochlo- TetO-APPSweInd mice compared to wildtypes, while no ride resulted in significantly fewer Aβ40-containing ves - differences were observed in the cortex or MS (Fig.  5k). sels (Fig. 4d–f ). Although a similar trend was observed in Additional analysis found that the ratio of p75NTR to saporin animals, the difference was not statistically signif - ChAT expression was significantly lower in the MS and icant (p = 0.08) and no difference was observed between hippocampus of TetO-APPSweInd mice compared to control + fasudil and saporin + fasudil groups (Fig. 4f ). wildtype animals (Fig. 5l). To determine if IPAD may have been influenced by dif - Quantification of Aβ pathology in the hippocampus ferences in vessel number and/or microglia and astrocyte after saporin treatment showed no difference in the per - activation, densities of each were quantified in control centage area covered by Aβ40-positive plaques between and saporin-treated mice. The density of laminin-posi - control and saporin mice (Fig.  6a, b, e). However, Aβ40 tive macrovessels and capillaries did not differ between CAA load was significantly higher in the saporin-treated control and saporin-treated mice in either the cortex or mice (Fig.  6a, b, e). A similar but non-significant pattern hippocampus (Fig.  4g–j), although capillary density was of vascular Aβ42 staining was observed between con- significantly higher in the cortex than the hippocampus trol and saporin mice, while parenchymal Aβ42 was not in both treatment groups (p = 0.003). Similarly, quanti- affected by saporin treatment (Fig.  6a, b, f ). In the cor- fication of Iba1 and GFAP staining revealed no effect of tex, no differences were observed between control and saporin treatment on astrocyte or microglial coverage of saporin-treated mice in the density of Aβ40-positive the hippocampus or cortex (Additional file 4: Fig. 3f–k). Nizari et al. acta neuropathol commun (2021) 9:12 Page 8 of 17 Fig. 3 Regional variation in NOS expression and activity in control and saporin-treated mice. a–d Western blots and quantification of levels of eNOS (a and b) and nNOS (c and d) in the hippocampus (a, c) and cortex (b, d) of control and saporin-treated mice (n = 6–8/group). Molecular weight markers (kDa) are shown on the right hand side. The black line demarcates the original blot (upper) and the same blot re-probed for loading control (lower). e–h Photomicrographs showing the expression of total NOS (green) in laminin-positive vessels (blue) in the hippocampus (e and f) and cortex (g and h) of control (e and g) and saporin-treated mice (f and h). Note the stable expression of NOS in hippocampal vessels of saporin-treated mice, while NOS expression is significantly reduced in cortical vessels of saporin animals. i–l eNOS expression (green) in GFAP-positive astrocytes (blue, i and j) and Iba1-positive microglia (blue, k and l) in the hippocampus of control (i and k) and saporin mice (j and l). Colocalization between eNOS and GFAP or Iba-1 is shown as white-turquoise. m and n Quantification of NOS expression in blood vessels as a ratio to overall vessel density (m) and degree of colocalisation between eNOS and GFAP or Iba1 as measured by the Pearson’s correlation coefficient (n). n = 5/group. Scale bars for f and h = 50 μm; j and l = 20 μm. *p < 0.05 N izari et al. acta neuropathol commun (2021) 9:12 Page 9 of 17 Fig. 4 Administration of fasudil hydrochloride increases IPAD in the hippocampus, but not cortex of denervated mice. a–f Photomicrographs showing the distribution of human Aβ40-AF555 (red) at 5 min post-injection into the hippocampus (a and b) and at 2.5 min after injection into the cortex (d and e) of control mice at baseline (a and d) and after administration of fasudil hydrochloride (b and e). The cerebrovascular basement membrane was labelled with anti-laminin (blue) and smooth muscle cells were identified with anti-α smooth muscle actin (green). Quantification of the total number of Aβ40-containing vessels in the hippocampus (c) and cortex (f) of control (con) and saporin (sap)-treated mice at baseline (n = 14–16 for hippocampus and n = 7–8 for cortex) and after fasudil hydrochloride (+F) (n = 5/group for both regions). g and h Photomicrographs of laminin staining in the hippocampus (g) and cortex (h) of control (upper panels) and saporin-treated mice (lower panels). i and j Quantification of % area covered by laminin in the hippocampus (i) and cortex (j) of control and saporin animals. n = 5–7/group. Scale bars = 100 μm. *p < 0.05, **p < 0.01, ***p < 0.001 plaques or CAA (Fig.  6c, d, g). Likewise, the density of Discussion cortical parenchymal and vascular Aβ42 was unaffected Results from this study suggest that loss of cholinergic by saporin treatment (Fig. 6c, d, h). innervation differentially affects cortical and hippocampal As with the C57BL/6 mice, vessel density between responsiveness to eNOS-stimulated increases in CBF and control and saporin-treated TetO-APPSweInd mice was IPAD in wildtype mice, with hippocampal, but not cortical, similar in both the cortex and hippocampus (Fig.  6i–l). vessels remaining responsive to stimulation. The death of Analysis of GFAP and Iba1 expression revealed no sig- cholinergic nerve fibres resulted in a significant and selec - nificant difference in area coverage between treatment tive increase in Aβ40-positive CAA in the TetO-APPS- groups in either brain region (Additional file  4: Fig. 3l–q). weInd model of AD. These findings support the importance These findings confirm that saporin administration did of the interrelationship between cholinergic innervation not significantly alter vessel density or glial activation in and vascular function in the aetiology and/or progression the TetO-APP mice and support a role for loss of cholin- of CAA and suggest that regional vulnerability or resilience ergic innervation in potentiating CAA pathology. Nizari et al. acta neuropathol commun (2021) 9:12 Page 10 of 17 Fig. 5 Distribution and quantification of cholinergic and p75NTR-positive neurons in control and saporin-treated TetO-APPSweInd mice. a–f Photomicrographs of ChAT staining in the medial septum and Diagonal band of Broca (a and d), hippocampus (b and e) and cortex (c and f) in control (a–c) and mu-saporin treated (d–f) TetO-APPSweInd mice. (g–i), Quantification of % area covered by ChAT-positive neurons in the medial septum (g) and fibres in the hippocampus (h) and cortex (i), n = 4–5/group. j–l Quantification of % area covered by ChAT (j) and p75NTR-positive (k) neurons and fibres and the ratio of ChAT:p75NTR expression (l) in the medial septum (MS), hippocampus (Hippo) and cortex of C57BL/6 and TetO-APPSweInd ( TETAPP) mice, n = 5/group/strain. Scale bar = 100 μm. *p < 0.05, **p < 0.01,* **p < 0.001 to loss of cholinergic dysfunction may contribute to the cognitive impairment [9, 22, 26, 82, 86]. In agreement topographical nature of CAA (Additional file 4 : Fig. 3). with previous studies [58, 61], we found that intracer- Degeneration of cholinergic neurons and shrinkage of ebral administration of mu-saporin, which selectively the basal forebrain are early features of AD and are asso- targets p75NTR-expressing neurons, caused the death ciated with increased Aβ pathology, altered CBF and of ChAT-positive neurons in the MS as well as their fibre N izari et al. acta neuropathol commun (2021) 9:12 Page 11 of 17 Fig. 6 Loss of cholinergic innervation selectively increases Aβ40-positive CAA in the hippocampus of TetO-APP mice. a–h Photomicrographs of hippocampal (a and b) and cortical tissues (c and d) of TetO-APP mice stained with antibodies against human Aβ40 (a–d, left panels) and Aβ42 staining (a–d, right panels). Tissues from control animals are shown in the upper panels and saporin-treated tissues are shown in the lower panels. Arrowheads show plaques and asterisks show CAA-positive vessels. e–h Quantification of % area covered by Aβ40 (e and g) and Aβ42-positive (f and h) plaques and blood vessels in the hippocampus (e and f) and cortex (g and h) of TetO-APPSweInd mice, n = 6–7/group. i–l Photomicrographs of laminin staining in the hippocampus (i) and cortex (j) of control (upper panels) and saporin-treated mice (lower panels). k Histogram showing quantification of laminin density in control and saporin-treated mice. n = 3–5/group. Scale bars = 100 μm. **p < 0.01 projections in the cortex and hippocampus of wildtype Therefore, to evaluate the impact of loss of cholinergic mice. innervation on evoked CBF, we mimicked ACh activa- ACh has a well-known vasodilatory effect in the brain tion of eNOS by using the selective ROCK inhibitor fas- and stimulation of the basal forebrain leads to increased udil hydrochloride, which has been shown to increase cortical CBF [33, 35]. This effect is predominantly CBF by stimulating the PI3K/Akt/eNOS pathway [68, observed during NVC when release of ACh stimulates 76]. Consistent with previous reports [50], loss of cholin- the production of NO via activation of eNOS or indi- ergic innervation in the cortex and hippocampus did not rectly by stimulation of nNOS-containing interneurons affect baseline CBF in either region, supporting a primary [53]. Although ASL MRI can be used to measure NVC in role of ACh on CBF during NVC. However, while admin- the cortex using whisker or forepaw stimulation [49], to istration of fasudil hydrochloride was not able to evoke our knowledge similar methods are not available to stim- a change in CBF in the cortex of saporin-treated mice, ulate NVC in the hippocampus of anesthetised animals. Nizari et al. acta neuropathol commun (2021) 9:12 Page 12 of 17 denervated vessels in the hippocampus remained respon- cortex (0.5  mm from dura) may have flooded the suba - sive to stimulation. rachnoid space, even when using the smaller 0.25 µL vol- Because multiple downstream signalling pathways in ume. Therefore, more detailed in vivo tracer experiments addition to eNOS are regulated by ROCK activity, includ- are needed to clarify rates of IPAD between cortical and ing those relating to smooth muscle contraction [45], we hippocampal regions. However, in agreement with other cannot definitively conclude that the observed effects studies [21, 62], we also found that cerebrovascular den- were due to stimulation of eNOS. However, the find - sity was significantly higher in the cortex compared to ings that levels of eNOS were significantly decreased in the hippocampus. This larger surface area may allow the cortex and increased in the hippocampus of saporin- for solutes contained within the ISF to be more rapidly treated mice, support the hypothesis that loss of cholin- removed from the cortex than from the hippocampus ergic innervation resulted in opposing effects on eNOS under physiological conditions. expression that aligned with the CBF response. Although Administration of fasudil hydrochloride resulted in eNOS is principally expressed by endothelial cells, pre- significantly more vessels with Aβ in the hippocampi of vious studies have reported its expression in neurons, both control and saporin-treated mice, while in the cor- astrocytes and in microglia across various species [16, 75, tex, fewer vessels were found to contain Aβ and this was 92]. Our observation that eNOS was expressed not only observed in control mice only. Although the pattern of in blood vessels but also by microglia in the hippocam- distribution was opposite between the two regions, we pus of saporin-treated mice, suggests that the functional interpret both findings as representing increased IPAD effects of fasudil hydrochloride in the hippocampus may at different rates of clearance. These findings are consist - also be due in part to activation of non-vascular cells. ent with our CBF data and suggest that IPAD is signifi - Although it is not clear why the effect of saporin treat - cantly increased in the presence of vasodilation, which is ment induced an opposite expression of eNOS between in agreement with reports of impaired solute clearance the cortex and hippocampus, endogenous NOS activ- from the brain during hypoperfusion [3, 37]. However, ity in both the nNOS- and eNOS-enriched fractions has the similarity in the number of labeled cortical vessels previously been reported to be higher in the hippocam- between control + fasudil and saporin + fasudil mice pus compared to the cortex [74]. This is supported by suggests that other factors are also contributing to Aβ previous reports showing that changes in CBF in the hip- clearance in denervated mice. Although blood pressure pocampus were more proportional to changes in nNOS was not monitored in the current experiments, previ- activity than in the cortex [52] and that the cortex is more ous studies have shown that fasudil hydrochloride does sensitive than the hippocampus to inhibition of nNOS not alter systolic blood pressure in normotensive rodents activity [43]. or humans [44, 51, 57, 59], suggesting that the observed Previous studies have suggested that contractions of effects were unlikely to be due to changes in peripheral arterial smooth muscle cells are required for drainage blood pressure. In addition, no differences in vessel den - of fluid along cerebral blood vessels [1, 3, 37], although sity or markers of microglia and astrocytes were observed whether this pulsation is sufficient to drive bulk flow of between control and saporin mice in either brain region. ISF and CSF remains controversial [11, 31]. Several stud- Although our findings are consistent with reports of an ies have shown that vasoreactivity in AD is improved association between decreased eNOS expression and following treatment with AChEIs [71]. We hypothesised increased CAA [6, 81], recent work has shown that NVC that there is a direct relationship between vasoreactivity is mediated in part by arteriole caveolae independent of and the efficiency of IPAD and that loss of cholinergic eNOS activation [17]. Further work is required to deter- innervation would impair IPAD of Aβ in a similar pat- mine the factors that regulate Aβ clearance when cholin- tern to that observed for CBF. No differences in IPAD ergic signalling is attenuated. were observed between control and saporin-treated mice Previous studies have reported a relationship in either brain region under baseline physiological con- between basal forebrain degeneration and Aβ pathol- ditions. Our observation that fewer Aβ-positive vessels ogy in the cortex [26] and basal forebrain atrophy has were visible in the cortex after a 5  min diffusion period been suggested to predict cortical Aβ burden [82]. compared to the hippocampus, suggests that IPAD of Aβ Induced loss of cholinergic neurons in rodent mod- may be endogenously faster in the cortex than in the hip- els of AD has also been associated with increased Aβ pocampus. We have previously reported differences in plaque deposition [48, 67], however most studies have the efficiency of IPAD between subcortical brain regions not specifically investigated the effect on vascular Aβ. that are differentially affected by CAA [27]. However, In the present study, administration of mu-saporin in given the relatively small thickness of the mouse cortex TetO-APPSweInd mice resulted in a loss of cholinergic [65], it is possible that the depth of injection into the neurons that was only significant in the hippocampus. N izari et al. acta neuropathol commun (2021) 9:12 Page 13 of 17 The reasons for the discrepancies between the degree animal model that are not present in AD. In addition, of loss between the C57BL/6 and TetO-APPSweInd as age is the major risk factor for both sporadic AD and mice are not clear, but may relate to the dystrophic CAA, additional experiments are needed to determine appearance of cholinergic fibers and decreased whether the effects of cholinergic denervation on CBF p75NTR:ChAT ratio observed in the TetO-APPSweInd and IPAD in the cortex and hippocampus seen here in mice. As binding of the p75NTR by Aβ is known to young adult mice are also observed in aged animals. induce apoptosis [95], it may be that pre-existing Aβ pathology caused damage to cholinergic neurons Conclusions and fibres that induced a downregulation in p75NTR Despite these limitations, findings from this study sup - expression and decreased receptor availability for mu- port a role for loss of cholinergic innervation in the aeti- saporin binding. ology and/or progression of CAA and suggest that this Although unexpected, the difference in sensitivity to may be related to eNOS-mediated vasodynamics that saporin treatment between the cortex and hippocam- contribute to clearance of Aβ from the brain via IPAD pus provided an internal control to study the effect of pathways. Therefore, combined targeting of eNOS and cholinergic loss on Aβ pathology. We found that loss cholinergic signalling/activation may represent a new of cholinergic innervation in the hippocampus was mechanism to improve the efficiency of Aβ removal and associated with a significant increase in Aβ40-positive reduce its deposition as CAA. vessels, consistent with the preferential deposition of Aβ40 in the vasculature in AD [30, 80]. By contrast, Supplementary Information CAA load was not affected in the cortex where cholin - The online version contains supplementary material available at https ://doi. ergic fibre density was not altered by saporin treatment. org/10.1186/s4047 8-020-01108 -z. Parenchymal plaque load did not differ between con - trol and saporin-treated mice in either region. These Additional file 1: Table 1 List of source of primary and secondary anti- findings are consistent with previous studies showing bodies used for immunohistochemistry. significantly more endogenous CAA in the absence of Additional file 2: Fig. 1 a–d Photomicrographs of diffuse parenchymal changes in parenchymal changes or changes in APP plaques identified by the anti-Aβ40 antibody (a) and senile plaques stained by the anti-Aβ42 antibody (b) in TetO-APPSweInd mice. No stain- processing in rabbits administered saporin [10, 70]. ing was observed after pre-absorption of the Aβ40 antibody with Aβ40 These findings are also similar to a study which found peptide (1:10 molar ratio, c). Sections incubated after pre-absorption of that age-related loss of perivascular cholinergic inner- Aβ40 with anti-Aβ40 + anti-Aβ42 (d) showed a similar pattern of staining to that of sections incubated with the anti-Aβ42 antibody alone. e–h No vation in the cortex did not significantly correlate with staining was observed in tissue sections from C57Bl/6 mice incubated increased cortical plaque load in the Tg2576 AD mouse with fluorescently-conjugated secondary antibodies alone. model [46]. However, other studies have reported Additional file 3: Fig. 2 a–f Photomicrographs showing expression of increased plaque load and elevated concentrations of ChAT (green), p75NTR (red) and their colocalization (yellow) in neurons soluble Aβ following saporin-induced cholinergic loss in the medial septum (a and b) and fibers in the hippocampus (c and d) and cortex (e and f) of C57Bl/6 mice. Animals received an intracerebroven- in the APP/PS1 and Tg2576 mouse models [24, 48, tricular injection of either PBS (control, a, c and e) or mu-saporin (b, d and 67]. Many factors may have contributed to these differ - f). Saporin treatment significantly reduced expression of p75NTR, ChAT- ent observations, including the degree of cholinergic positive cell bodies and fibers in the medial septum (b), hippocampus (d) and cortex (f). Images of the hippocampus are composed of individual degeneration, age of the mice and amount of pre-exist- overlapping images stitched together using Fiji. Scale bars: a, b, e, f = 250 ing Aβ pathology before saporin treatment, as well as μm; c and d = 100 μm. the ratio of Aβ40:Aβ42 and progression of pathology Additional file 4: Fig. 3 a and b Photomicrographs showing the distribu- between the different mouse models. Despite these dis - tion of 0.5 μL human Aβ40-AF555 (red) at 400 μm away from the injection site after 5 min post-injection (PI) into the hippocampus (a) and cortex (b) crepancies, our results support a consensus that loss of control mice. The cerebrovascular basement membrane was labelled of cholinergic innervation contributes to increased Aβ with anti-laminin (blue) and smooth muscle cells were identified with pathology. anti-α smooth muscle actin (green). c–e Photomicrographs showing the distribution of Aβ40-AF488 (green) in the cortex of C57BL/6 mice, at 400 In addition to the previously discussed limitations μm away from the injection site. The volume and post-injection (PI) time related to inducing NVC in the hippocampus of anes- is indicated for 3 combinations that were tested to determine the optimal thetised animals and assessment of IPAD ex  vivo, this parameters for quantification of Aβ-positive vessels. f–k Photomicro - graphs and quantification of GFAP (f, g, j) and Iba1 (h, i, k) staining in the study has several other weaknesses. The saporin model hippocampus (f, h) and cortex (g, i) of control (con, upper panels) and induces loss of basal forebrain cholinergic neurons saporin-treated C57Bl/6 mice (sap, lower panels). l–q Photomicrographs in a retrograde manner [47], and over a more rapid and quantification of GFAP (l, m, p) and Iba1 (n, o, q) staining in the hippocampus (l, n) and cortex (m, o) of control (con, upper panels) and timeframe than that observed in AD, which may acti- saporin-treated TetO-APPSweInd mice (sap, lower panels). n = 3–5/group. vate a strong acute inflammatory reaction and/or the Scale bars = 100 μm. development of compensatory mechanisms in the Nizari et al. acta neuropathol commun (2021) 9:12 Page 14 of 17 Abbreviations 4. Arvanitakis Z, Leurgans SE, Wang Z, Wilson RS, Bennett DA, Schneider JA 7-NI: 7-Nitroindazole; Aβ: Beta amyloid; ACh: Acetylcholine; AChEI: Acetylcho- (2011) Cerebral amyloid angiopathy pathology and cognitive domains in linesterase inhibitor; AD: Alzheimer’s disease; APP: Amyloid precursor protein; older persons. Ann Neurol 69:320–327 ASL: Arterial spin labelling; CAA : Cerebral amyloid angiopathy; CBF: Cerebral 5. Attems J, Quass M, Jellinger KA, Lintner F (2007) Topographical distribu- blood flow; CVBM: Cerebrovascular basement membranes; ChAT: Choline tion of cerebral amyloid angiopathy and its effect on cognitive decline acetyltransferase; eNOS: Endothelial nitric oxide synthase; GAPDH: Glyceral- are influenced by Alzheimer disease pathology. J Neurol Sci 257:49–55 dehyde 3-phosphate dehydrogenase; GFAP: Glial acidic fibrillary protein; Iba1: 6. Austin SA, Katusic ZS (2020) Partial loss of endothelial nitric oxide leads Ionized calcium binding adaptor molecule 1; ICV: Intracerebroventricular; to increased cerebrovascular beta amyloid. J Cereb Blood Flow Metab IPAD: Intramural periarterial drainage; iNOS: Inducible nitric oxide synthase; 40:392–403. https ://doi.org/10.1177/02716 78X18 82247 4 ISF: Interstitial fluid; L-NAME: Nitro- l -arginine Methyl Ester Hydrochloride; MRI: 7. Bakker EN, Bacskai BJ, Arbel-Ornath M, Aldea R, Bedussi B, Morris AW, Magnetic resonance imaging; MS: Medial septum; nNOS: Neuronal nitric oxide Weller RO, Carare RO (2016) Lymphatic clearance of the brain: perivas- synthase; NVC: Neurovascular coupling; P75NTR: p75 neurotrophin receptor; cular, paravascular and significance for neurodegenerative diseases. Cell ROCK: Rho- associated, coiled-coil containing protein kinase. Mol Neurobiol 36:181–194. https ://doi.org/10.1007/s1057 1-015-0273-8 8. Ball KK, Cruz NF, Mrak RE, Dienel GA (2010) Trafficking of glucose, lactate, Acknowledgements and amyloid-beta from the inferior colliculus through perivascular routes. The authors wish to thank the staff at the BRU and BRF for their technical J Cereb Blood Flow Metab 30:162–176. https ://doi.org/10.1038/jcbfm support. .2009.206 9. Bartus RT, Dean RL 3rd, Beer B, Lippa AS (1982) The cholinergic hypothesis Authors’ contributions of geriatric memory dysfunction. Science 217:408–414. https ://doi. SN planned and carried out experiments, data analysis and manuscript prepa-org/10.1126/scien ce.70460 51 ration and editing. JW carried out experiments and data analysis relating to 10. Beach TG, Potter PE, Kuo YM, Emmerling MR, Durham RA, Webster SD, MRI experiments. ROC contributed to experimental planning and manuscript Walker DG, Sue LI, Scott S, Layne KJ et al (2000) Cholinergic deafferen- editing. IAR contributed to experimental planning and data analysis. CAH con- tation of the rabbit cortex: a new animal model of Abeta deposition. tributed to experimental planning, data analysis and manuscript preparation Neurosci Lett 283:9–12 and editing. All authors read and approved the final manuscript. 11. Bedussi B, Almasian M, de Vos J, VanBavel E, Bakker EN (2018) Paravas- cular spaces at the brain surface: low resistance pathways for cerebro- Funding spinal fluid flow. J Cereb Blood Flow Metab 38:719–726. https ://doi. This work was supported by funds from Alzheimer’s Research UK.org/10.1177/02716 78X17 73798 4 12. Brown DR, Hunter R, Wyper DJ, Patterson J, Kelly RC, Montaldi D, Availability of data and materials McCullouch J (1996) Longitudinal changes in cognitive function The datasets used and/or analysed during the current study available from the and regional cerebral function in Alzheimer’s disease: a SPECT blood corresponding author on reasonable request. flow study. J Psychiatr Res 30:109–126. https ://doi.org/10.1016/0022- 3956(95)00032 -1 Ethics approval and consent to participate 13. Carare RO, Bernardes-Silva M, Newman TA, Page AM, Nicoll JAR, Perry VH, All animal work was approved by the Animal Welfare and Ethics Research Weller RO (2008) Solutes, but not cells, drain from the brain parenchyma Boards (AWERB) at the OU, University of Southampton and UCL in accord- along basement membranes of capillaries and arteries. Significance for ance with Home Office regulations and project licences (PPL 70/8507 and cerebral amyloid angiopathy and neuroimmunology. Neuropathol Appl PPL 30/3095) under the Animals (Scientific Procedures) Act 1986. Experiments Neurobiol 34:131–144 were reported according to the ARRIVE guidelines. 14. Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO (2013) Review: cer - ebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum Consent for publication of protein elimination failure angiopathies (PEFA) in neurodegenerative Not applicable. disease with a focus on therapy. Neuropathol Appl Neurobiol 39:593– 611. https ://doi.org/10.1111/nan.12042 Competing interests 15. Cauli B, Tong XK, Rancillac A, Serluca N, Lambolez B, Rossier J, Hamel E The authors declare that they have no competing interests. (2004) Cortical GABA interneurons in neurovascular coupling: relays for subcortical vasoactive pathways. J Neurosci 24:8940–8949. https ://doi. Author detailsorg/10.1523/JNEUR OSCI.3065-04.2004 School of Life, Health and Chemical Sciences, Open University, Milton 16. Caviedes A, Varas-Godoy M, Lafourcade C, Sandoval S, Bravo-Alegria 2 3 Keynes, UK. Centre for Advanced Bioimaging, UCL, London, UK. Clinical J, Kaehne T, Massmann A, Figueroa JP, Nualart F, Wyneken U (2017) and Experimental Sciences, University of Southampton, Southampton, UK. Endothelial nitric oxide synthase is present in dendritic spines of neurons Department of Biomedical and Life Sciences, Lancaster University, Lancaster, in primary cultures. Front Cell Neurosci 11:180. https ://doi.org/10.3389/ UK. fncel .2017.00180 17. Chow BW, Nunez V, Kaplan L, Granger AJ, Bistrong K, Zucker HL, Kumar Received: 6 November 2020 Accepted: 15 December 2020 P, Sabatini BL, Gu C (2020) Caveolae in CNS arterioles mediate neuro- vascular coupling. Nature 579:106–110. https ://doi.org/10.1038/s4158 6-020-2026-1 18. Dai W, Lopez OL, Carmichael OT, Becker JT, Kuller LH, Gach HM (2009) Mild cognitive impairment and alzheimer disease: patterns of altered cerebral blood flow at MR imaging. Radiology 250:856–866. https ://doi. References org/10.1148/radio l.25030 80751 1. Aldea R, Weller RO, Wilcock DM, Carare RO, Richardson G (2019) Cerebro- 19. Di Marco LY, Farkas E, Martin C, Venneri A, Frangi AF (2015) Is vasomo- vascular smooth muscle cells as the drivers of intramural periarterial tion in cerebral arteries impaired in alzheimer’s disease? J Alzheimers Dis drainage of the brain. Front Aging Neurosci 11:1. https ://doi.org/10.3389/ 46:35–53. https ://doi.org/10.3233/JAD-14297 6 fnagi .2019.00001 20. Dorheim MA, Tracey WR, Pollock JS, Grammas P (1994) Nitric oxide 2. Alsop DC, Casement M, de Bazelaire C, Fong T, Press DZ (2008) Hip- synthase activity is elevated in brain microvessels in Alzheimer’s disease. pocampal hyperperfusion in Alzheimer’s disease. Neuroimage 42:1267– Biochem Biophys Res Commun 205:659–665. https ://doi.org/10.1006/ 1274. https ://doi.org/10.1016/j.neuro image .2008.06.006 bbrc.1994.2716 3. Arbel-Ornath M, Hudry E, Eikermann-Haerter K, Hou S, Gregory JL, Zhao L, 21. Dorr A, Sahota B, Chinta LV, Brown ME, Lai AY, Ma K, Hawkes CA, McLaurin Betensky RA, Frosch MP, Greenberg SM, Bacskai BJ (2013) Interstitial fluid J, Stefanovic B (2012) Amyloid-beta-dependent compromise of micro- drainage is impaired in ischemic stroke and Alzheimer’s disease mouse vascular structure and function in a model of Alzheimer’s disease. Brain models. Acta Neuropathol 126:353–364. https ://doi.org/10.1007/s0040 135:3039–3050. https ://doi.org/10.1093/brain /aws24 3 1-013-1145-2 N izari et al. acta neuropathol commun (2021) 9:12 Page 15 of 17 22. Francis PT, Palmer AM, Snape M, Wilcock GK (1999) The cholinergic 40. Jankowsky JL, Slunt HH, Gonzales V, Savonenko AV, Wen JC, Jenkins NA, hypothesis of Alzheimer’s disease: a review of progress. J Neurol Neuro- Copeland NG, Younkin LH, Lester HA, Younkin SG et al (2005) Persistent surg Psychiatry 66:137–147 amyloidosis following suppression of Abeta production in a transgenic 23. Furchgott RF, Zawadzki JV (1980) The obligatory role of endothelial cells model of Alzheimer disease. PLoS Med 2:e355. https ://doi.org/10.1371/ in the relaxation of arterial smooth muscle by acetylcholine. Nature journ al.pmed.00203 55 288:373–376. https ://doi.org/10.1038/28837 3a0 41. Jeffrey M, Gonzalez L (2007) Classical sheep transmissible spongiform 24. Gil-Bea FJ, Gerenu G, Aisa B, Kirazov LP, Schliebs R, Ramirez MJ (2012) encephalopathies: pathogenesis, pathological phenotypes and clinical Cholinergic denervation exacerbates amyloid pathology and induces disease. Neuropathol Appl Neurobiol 33:373–394. https ://doi.org/10.111 hippocampal atrophy in Tg2576 mice. Neurobiol Dis 48:439–446. https ://1/j.1365-2990.2007.00868 .x doi.org/10.1016/j.nbd.2012.06.020 42. Jellinger KA, Attems J (2006) Prevalence and impact of cerebrovascular 25. Gratton JP, Fontana J, O’Connor DS, Garcia-Cardena G, McCabe TJ, Sessa pathology in Alzheimer’s disease and parkinsonism. Acta Neurol Scand WC (2000) Reconstitution of an endothelial nitric-oxide synthase (eNOS), 114:38–46 hsp90, and caveolin-1 complex in vitro. Evidence that hsp90 facilitates 43. Kalisch BE, Connop BP, Jhamandas K, Beninger RJ, Boegman RJ (1996) calmodulin stimulated displacement of eNOS from caveolin-1. J Biol Differential action of 7-nitro indazole on rat brain nitric oxide synthase. Chem 275:22268–22272. https ://doi.org/10.1074/jbc.M0016 44200 Neurosci Lett 219:75–78. https ://doi.org/10.1016/s0304 -3940(96)13194 -3 26. Grothe MJ, Ewers M, Krause B, Heinsen H, Teipel SJ, Alzheimer’s Disease 44. Kishi T, Hirooka Y, Masumoto A, Ito K, Kimura Y, Inokuchi K, Tagawa T, Neuroimaging I (2014) Basal forebrain atrophy and cortical amyloid Shimokawa H, Takeshita A, Sunagawa K (2005) Rho-kinase inhibitor deposition in nondemented elderly subjects. Alzheimers Dement improves increased vascular resistance and impaired vasodilation of the 10:S344–353. https ://doi.org/10.1016/j.jalz.2013.09.011 forearm in patients with heart failure. Circulation 111:2741–2747. https :// 27. Hawkes CA, Gatherer M, Sharp MM, Dorr A, Yuen HM, Kalaria R, Weller doi.org/10.1161/CIRCU LATIO NAHA.104.51024 8 RO, Carare RO (2013) Regional differences in the morphological and 45. Kureishi Y, Kobayashi S, Amano M, Kimura K, Kanaide H, Nakano T, Kaibu- functional effects of aging on cerebral basement membranes and chi K, Ito M (1997) Rho-associated kinase directly induces smooth muscle perivascular drainage of amyloid-beta from the mouse brain. Aging Cell contraction through myosin light chain phosphorylation. J Biol Chem 12:224–236. https ://doi.org/10.1111/acel.12045 272:12257–12260. https ://doi.org/10.1074/jbc.272.19.12257 28. Hawkes CA, Hartig W, Kacza J, Schliebs R, Weller RO, Nicoll JA, Carare RO 46. Kuznetsova E, Schliebs R (2013) beta-Amyloid, cholinergic transmis- (2011) Perivascular drainage of solutes is impaired in the ageing mouse sion, and cerebrovascular system: a developmental study in a mouse brain and in the presence of cerebral amyloid angiopathy. Acta Neuro- model of Alzheimer’s disease. Curr Pharm Des 19:6749–6765. https ://doi. pathol 121:431–443. https ://doi.org/10.1007/s0040 1-011-0801-7org/10.2174/13816 12811 31999 90711 29. Hawkes CA, Sullivan PM, Hands S, Weller RO, Nicoll JA, Carare RO (2012) 47. Lapchak PA, Araujo DM, Carswell S, Hefti F (1993) Distribution of [125I] Disruption of arterial perivascular drainage of amyloid-beta from the nerve growth factor in the rat brain following a single intraventricular brains of mice expressing the human APOE epsilon4 allele. PLoS ONE injection: correlation with the topographical distribution of trkA mes- 7:e41636. https ://doi.org/10.1371/journ al.pone.00416 36 senger RNA-expressing cells. Neuroscience 54:445–460. https ://doi. 30. Hernandez-Guillamon M, Delgado P, Penalba A, Rodriguez-Luna D, org/10.1016/0306-4522(93)90265 -h Molina CA, Rovira A, Alvarez-Sabin J, Boada M, Montaner J (2012) 48. Laursen B, Mork A, Plath N, Kristiansen U, Bastlund JF (2013) Cholinergic Plasma beta-amyloid levels in cerebral amyloid angiopathy-associated degeneration is associated with increased plaque deposition and cogni- hemorrhagic stroke. Neurodegener Dis 10:320–323. https ://doi. tive impairment in APPswe/PS1dE9 mice. Behav Brain Res 240:146–152. org/10.1159/00033 3811https ://doi.org/10.1016/j.bbr.2012.11.012 31. Hladky SB, Barrand MA (2014) Mechanisms of fluid movement into, 49. Lecrux C, Hamel E (2016) Neuronal networks and mediators of cortical through and out of the brain: evaluation of the evidence. Fluids Barriers neurovascular coupling responses in normal and altered brain states. CNS 11:26. https ://doi.org/10.1186/2045-8118-11-26 Philos Trans R Soc Lond B Biol Sci. https ://doi.org/10.1098/rstb.2015.0350 32. Hotta H (2016) Neurogenic control of parenchymal arterioles in the 50. Lecrux C, Sandoe CH, Neupane S, Kropf P, Toussay X, Tong XK, Lacalle- cerebral cortex. Prog Brain Res 225:3–39. https ://doi.org/10.1016/ Aurioles M, Shmuel A, Hamel E (2017) Impact of altered cholinergic tones bs.pbr.2016.03.001 on the neurovascular coupling response to whisker stimulation. J Neuro- 33. Hotta H, Uchida S, Kagitani F, Maruyama N (2011) Control of cerebral sci 37:1518–1531. https ://doi.org/10.1523/JNEUR OSCI.1784-16.2016 cortical blood flow by stimulation of basal forebrain cholinergic 51. Liu P, Zhang H, Tang Y, Sheng C, Liu J, Zeng Y (2014) Influence of Rho areas in mice. J Physiol Sci 61:201–209. https ://doi.org/10.1007/s1257 kinase inhibitor Fasudil on late endothelial progenitor cells in peripheral 6-011-0139-x blood of COPD patients with pulmonary artery hypertension. Bosn J 34. Hu WT, Wang Z, Lee VM, Trojanowski JQ, Detre JA, Grossman M (2010) Dis- Basic Med Sci 14:40–44. https ://doi.org/10.17305 /bjbms .2014.2295 tinct cerebral perfusion patterns in FTLD and AD. Neurology 75:881–888. 52. Lourenco CF, Santos RM, Barbosa RM, Cadenas E, Radi R, Laranjinha J https ://doi.org/10.1212/WNL.0b013 e3181 f11e3 5 (2014) Neurovascular coupling in hippocampus is mediated via diffusion 35. Iadecola C, Zhang F (1996) Permissive and obligatory roles of NO in cer- by neuronal-derived nitric oxide. Free Radic Biol Med 73:421–429. https :// ebrovascular responses to hypercapnia and acetylcholine. Am J Physiol doi.org/10.1016/j.freer adbio med.2014.05.021 271:R990–1001. https ://doi.org/10.1152/ajpre gu.1996.271.4.R990 53. Lovick TA, Brown LA, Key BJ (1999) Neurovascular relationships in hip- 36. Iliff JJ, Wang M, Liao Y, Plogg BA, Peng W, Gundersen GA, Benveniste H, pocampal slices: physiological and anatomical studies of mechanisms Vates GE, Deane R, Goldman SA et al (2012) A paravascular pathway underlying flow-metabolism coupling in intraparenchymal microvessels. facilitates CSF flow through the brain parenchyma and the clearance of Neuroscience 92:47–60. https ://doi.org/10.1016/s0306 -4522(98)00737 -4 interstitial solutes, including amyloid beta. Sci Transl Med 4:147ra111. 54. Miners JS, Baig S, Palmer J, Palmer LE, Kehoe PG, Love S (2008) Abeta- https ://doi.org/10.1126/scitr anslm ed.30037 48 degrading enzymes in Alzheimer’s disease. Brain Pathol 18:240–252 37. Iliff JJ, Wang M, Zeppenfeld DM, Venkataraman A, Plog BA, Liao Y, Deane 55. Montagne A, Zhao Z, Zlokovic BV (2017) Alzheimer’s disease: a matter of R, Nedergaard M (2013) Cerebral arterial pulsation drives paravas- blood-brain barrier dysfunction? J Exp Med 214:3151–3169. https ://doi. cular CSF-interstitial fluid exchange in the murine brain. J Neurosci org/10.1084/jem.20171 406 33:18190–18199 56. Morris AW, Carare RO, Schreiber S, Hawkes CA (2014) The cerebrovascular 38. Iturria-Medina Y, Sotero RC, Toussaint PJ, Mateos-Perez JM, Evans AC, basement membrane: role in the clearance of beta-amyloid and cerebral Alzheimer’s Disease Neuroimaging I (2016) Early role of vascular dysregu- amyloid angiopathy. Front Aging Neurosci 6:251. https ://doi.org/10.3389/ lation on late-onset Alzheimer’s disease based on multifactorial data-fnagi .2014.00251 driven analysis. Nat Commun 7:11934. https ://doi.org/10.1038/ncomm 57. Mukai Y, Shimokawa H, Matoba T, Kandabashi T, Satoh S, Hiroki J, Kaibuchi s1193 4 K, Takeshita A (2001) Involvement of Rho-kinase in hypertensive vascular 39. Jack CR Jr, Knopman DS, Jagust WJ, Shaw LM, Aisen PS, Weiner MW, disease: a novel therapeutic target in hypertension. FASEB J 15:1062– Petersen RC, Trojanowski JQ (2010) Hypothetical model of dynamic 1064. https ://doi.org/10.1096/fj.00-0735 fje biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 58. Nag N, Baxter MG, Berger-Sweeney JE (2009) Efficacy of a murine-p75- 9:119–128. https ://doi.org/10.1016/S1474 -4422(09)70299 -6 saporin immunotoxin for selective lesions of basal forebrain cholinergic Nizari et al. acta neuropathol commun (2021) 9:12 Page 16 of 17 neurons in mice. Neurosci Lett 452:247–251. https ://doi.org/10.1016/j. 76. Shi J, Wei L (2013) Rho kinases in cardiovascular physiology and patho- neule t.2009.01.006 physiology: the effect of fasudil. J Cardiovasc Pharmacol 62:341–354. 59. Nagata K, Kondoh Y, Satoh Y, Watahiki Y, Yokoyama E, Yuya H, Hirata Y, https ://doi.org/10.1097/FJC.0b013 e3182 a3718 f Shishido F, Hatazawa J, Kanno I et al (1993) Eec ff ts of fasudil hydrochlo - 77. Shin HK, Jones PB, Garcia-Alloza M, Borrelli L, Greenberg SM, Bacskai BJ, ride on cerebral blood flow in patients with chronic cerebral infarction. Frosch MP, Hyman BT, Moskowitz MA, Ayata C (2007) Age-dependent Clin Neuropharmacol 16:501–510. https ://doi.org/10.1097/00002 826- cerebrovascular dysfunction in a transgenic mouse model of cerebral 19931 2000-00003 amyloid angiopathy. Brain 130:2310–2319 60. Nicolakakis N, Hamel E (2011) Neurovascular function in Alzheimer’s 78. Shin HK, Salomone S, Potts EM, Lee SW, Millican E, Noma K, Huang PL, disease patients and experimental models. J Cereb Blood Flow Metab Boas DA, Liao JK, Moskowitz MA et al (2007) Rho-kinase inhibition acutely 31:1354–1370. https ://doi.org/10.1038/jcbfm .2011.43 augments blood flow in focal cerebral ischemia via endothelial mecha- 61. Nizari S, Carare RO, Romero IA, Hawkes CA (2019) 3D reconstruction of nisms. J Cereb Blood Flow Metab 27:998–1009. https ://doi.org/10.1038/ the neurovascular unit reveals differential loss of cholinergic innervation sj.jcbfm .96004 06 in the cortex and hippocampus of the adult mouse brain. Front Aging 79. Simic G, Babic M, Borovecki F, Hof PR (2014) Early failure of the default- Neurosci 11:172. https ://doi.org/10.3389/fnagi .2019.00172 mode network and the pathogenesis of Alzheimer’s disease. CNS Neuro- 62. Oomen CA, Farkas E, Roman V, van der Beek EM, Luiten PG, Meerlo P sci Ther 20:692–698. https ://doi.org/10.1111/cns.12260 (2009) Resveratrol preserves cerebrovascular density and cognitive func- 80. Suzuki N, Iwatsubo T, Odaka A, Ishibashi Y, Kitada C, Ihara Y (1994) High tion in aging mice. Front Aging Neurosci 1:4. https ://doi.org/10.3389/ tissue content of soluble beta 1-40 is linked to cerebral amyloid angiopa- neuro .24.004.2009 thy. Am J Pathol 145:452–460 63. Paris D, Patel N, DelleDonne A, Quadros A, Smeed R, Mullan M (2004) 81. Tan XL, Xue YQ, Ma T, Wang X, Li JJ, Lan L, Malik KU, McDonald MP, Dopico Impaired angiogenesis in a transgenic mouse model of cerebral amyloi- AM, Liao FF (2015) Partial eNOS deficiency causes spontaneous throm- dosis. Neurosci Lett 366:80–85 botic cerebral infarction, amyloid angiopathy and cognitive impairment. 64. Partyka PP, Godsey GA, Galie JR, Kosciuk MC, Acharya NK, Nagele RG, Mol Neurodegener 10:24. https ://doi.org/10.1186/s1302 4-015-0020-0 Galie PA (2017) Mechanical stress regulates transport in a compliant 3D 82. Teipel S, Heinsen H, Amaro E Jr, Grinberg LT, Krause B, Grothe M, model of the blood-brain barrier. Biomaterials 115:30–39. https ://doi. Alzheimer’s Disease Neuroimaging I (2014) Cholinergic basal forebrain org/10.1016/j.bioma teria ls.2016.11.012 atrophy predicts amyloid burden in Alzheimer’s disease. Neurobiol Aging 65. Paxinos G, Franklin K (2019) Paxinos and Franklin’s the mouse brain in 35:482–491. https ://doi.org/10.1016/j.neuro biola ging.2013.09.029 stereotaxic coordinates, 5th edn. Elsevier, Amsterdam, p 375 83. Thal DR, Griffin WS, de Vos RA, Ghebremedhin E (2008) Cerebral amyloid 66. Perry EK, Morris CM, Court JA, Cheng A, Fairbairn AF, McKeith IG, Irving angiopathy and its relationship to Alzheimer’s disease. Acta Neuropathol D, Brown A, Perry RH (1995) Alteration in nicotine binding sites in Par- 115:599–609 kinson’s disease, Lewy body dementia and Alzheimer’s disease: possible 84. Tian J, Shi J, Bailey K, Mann DM (2003) Negative association between index of early neuropathology. Neuroscience 64:385–395. https ://doi. amyloid plaques and cerebral amyloid angiopathy in Alzheimer’s disease. org/10.1016/0306-4522(94)00410 -7 Neurosci Lett 352:137–140 67. Ramos-Rodriguez JJ, Pacheco-Herrero M, Thyssen D, Murillo-Carretero 85. Tikka S, Mykkanen K, Ruchoux MM, Bergholm R, Junna M, Poyhonen M, MI, Berrocoso E, Spires-Jones TL, Bacskai BJ, Garcia-Alloza M (2013) Rapid Yki-Jarvinen H, Joutel A, Viitanen M, Baumann M et al (2009) Congruence beta-amyloid deposition and cognitive impairment after cholinergic den- between NOTCH3 mutations and GOM in 131 CADASIL patients. Brain ervation in APP/PS1 mice. J Neuropathol Exp Neurol 72:272–285. https :// 132:933–939. https ://doi.org/10.1093/brain /awn36 4 doi.org/10.1097/NEN.0b013 e3182 88a8d d 86. Van Beek AH, Claassen JA (2011) The cerebrovascular role of the cholin- 68. Rikitake Y, Kim HH, Huang Z, Seto M, Yano K, Asano T, Moskowitz MA, ergic neural system in Alzheimer’s disease. Behav Brain Res 221:537–542. Liao JK (2005) Inhibition of Rho kinase (ROCK) leads to increased cerebral https ://doi.org/10.1016/j.bbr.2009.12.047 blood flow and stroke protection. Stroke 36:2251–2257. https ://doi. 87. van Opstal AM, van Rooden S, van Harten T, Ghariq E, Labadie G, Fotiadis org/10.1161/01.STR.00001 81077 .84981 .11 P, Gurol ME, Terwindt GM, Wermer MJH, van Buchem MA et al (2017) Cer- 69. Rivera-Rivera LA, Turski P, Johnson KM, Hoffman C, Berman SE, Kilgas P, ebrovascular function in presymptomatic and symptomatic individuals Rowley HA, Carlsson CM, Johnson SC, Wieben O (2016) 4D flow MRI for with hereditary cerebral amyloid angiopathy: a case-control study. Lancet intracranial hemodynamics assessment in Alzheimer’s disease. J Cereb Neurol 16:115–122. https ://doi.org/10.1016/S1474 -4422(16)30346 -5 Blood Flow Metab 36:1718–1730. https ://doi.org/10.1177/02716 78X15 88. van Veluw SJ, Hou SS, Calvo-Rodriguez M, Arbel-Ornath M, Snyder AC, 61717 1 Frosch MP, Greenberg SM, Bacskai BJ (2020) Vasomotion as a driving force 70. Roher AE, Kuo YM, Potter PE, Emmerling MR, Durham RA, Walker DG, Sue for paravascular clearance in the awake mouse brain. Neuron 105(549– LI, Honer WG, Beach TG (2000) Cortical cholinergic denervation elicits 561):e545. https ://doi.org/10.1016/j.neuro n.2019.10.033 vascular A beta deposition. Ann N Y Acad Sci 903:366–373 89. Vaucher E, Linville D, Hamel E (1997) Cholinergic basal forebrain projec- 71. Rosengarten B, Paulsen S, Burr O, Kaps M (2009) Neurovascular coupling tions to nitric oxide synthase-containing neurons in the rat cerebral in Alzheimer patients: effect of acetylcholine-esterase inhibitors. cortex. Neuroscience 79:827–836. https ://doi.org/10.1016/s0306 Neurobiol Aging 30:1918–1923. https ://doi.org/10.1016/j.neuro biola -4522(97)00033 -x ging.2008.02.017 90. Vinters HV (1987) Cerebral amyloid angiopathy. A critical review. Stroke 72. Sato A, Sato Y, Uchida S (2004) Activation of the intracerebral cholinergic 18:311–324 nerve fibers originating in the basal forebrain increases regional cerebral 91. Wells JA, Holmes HE, O’Callaghan JM, Colgan N, Ismail O, Fisher EM, blood flow in the rat’s cortex and hippocampus. Neurosci Lett 361:90–93. Siow B, Murray TK, Schwarz AJ, O’Neill MJ et al (2015) Increased cerebral https ://doi.org/10.1016/j.neule t.2004.01.004 vascular reactivity in the tau expressing rTg4510 mouse: evidence against 73. Schmitz TW, Nathan Spreng R, Alzheimer’s Disease Neuroimaging I (2016) the role of tau pathology to impair vascular health in Alzheimer’s disease. Basal forebrain degeneration precedes and predicts the cortical spread J Cereb Blood Flow Metab 35:359–362. https ://doi.org/10.1038/jcbfm of Alzheimer’s pathology. Nat Commun 7:13249. https ://doi.org/10.1038/ .2014.224 ncomm s1324 9 92. Wiencken AE, Casagrande VA (1999) Endothelial nitric oxide synthetase 74. Sehba FA, Schwartz AY, Chereshnev I, Bederson JB (2000) Acute decrease (eNOS) in astrocytes: another source of nitric oxide in neocortex. Glia in cerebral nitric oxide levels after subarachnoid hemorrhage. J Cereb 26:280–290 Blood Flow Metab 20:604–611. https ://doi.org/10.1097/00004 647-20000 93. Wilcock DM, Rojiani A, Rosenthal A, Subbarao S, Freeman MJ, Gordon MN, 3000-00018 Morgan D (2004) Passive immunotherapy against Abeta in aged APP- 75. Shafer OT, Chen A, Kumar SM, Muller KJ, Sahley CL (1998) Injury-induced transgenic mice reverses cognitive deficits and depletes parenchymal expression of endothelial nitric oxide synthase by glial and microglial amyloid deposits in spite of increased vascular amyloid and microhemor- cells in the leech central nervous system within minutes after injury. Proc rhage. J Neuroinflamm 1:24 Biol Sci 265:2171–2175. https ://doi.org/10.1098/rspb.1998.0555 N izari et al. acta neuropathol commun (2021) 9:12 Page 17 of 17 94. Williams RJ, Goodyear BG, Peca S, McCreary CR, Frayne R, Smith EE, receptor substrate/Akt signaling pathway in aorta of obese rats. Diabetes Pike GB (2017) Identification of neurovascular changes associated with 56:1014–1024. https ://doi.org/10.2337/db05-1147 cerebral amyloid angiopathy from subject-specific hemodynamic 97. Zhang F, Xu S, Iadecola C (1995) Role of nitric oxide and acetylcholine in response functions. J Cereb Blood Flow Metab 37:3433–3445. https ://doi. neocortical hyperemia elicited by basal forebrain stimulation: evidence org/10.1177/02716 78X17 69105 6 for an involvement of endothelial nitric oxide. Neuroscience 69:1195– 95. Yaar M, Zhai S, Pilch PF, Doyle SM, Eisenhauer PB, Fine RE, Gilchrest BA 1204. https ://doi.org/10.1016/0306-4522(95)00302 -y (1997) Binding of beta-amyloid to the p75 neurotrophin receptor induces apoptosis. A possible mechanism for Alzheimer’s disease. J Clin Invest Publisher’s Note 100:2333–2340. https ://doi.org/10.1172/JCI11 9772 Springer Nature remains neutral with regard to jurisdictional claims in pub- 96. Zecchin HG, Priviero FB, Souza CT, Zecchin KG, Prada PO, Carvalheira JB, lished maps and institutional affiliations. Velloso LA, Antunes E, Saad MJ (2007) Defective insulin and acetylcho- line induction of endothelial cell-nitric oxide synthase through insulin Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals

Loss of cholinergic innervation differentially affects eNOS-mediated blood flow, drainage of Aβ and cerebral amyloid angiopathy in the cortex and hippocampus of adult mice

Loading next page...
 
/lp/springer-journals/loss-of-cholinergic-innervation-differentially-affects-enos-mediated-DA2fODy5au
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
eISSN
2051-5960
DOI
10.1186/s40478-020-01108-z
Publisher site
See Article on Publisher Site

Abstract

Vascular dysregulation and cholinergic basal forebrain degeneration are both early pathological events in the devel- opment of Alzheimer’s disease (AD). Acetylcholine contributes to localised arterial dilatation and increased cerebral blood flow (CBF) during neurovascular coupling via activation of endothelial nitric oxide synthase (eNOS). Decreased vascular reactivity is suggested to contribute to impaired clearance of β-amyloid (Aβ) along intramural periarterial drainage (IPAD) pathways of the brain, leading to the development of cerebral amyloid angiopathy (CAA). However, the possible relationship between loss of cholinergic innervation, impaired vasoreactivity and reduced clearance of Aβ from the brain has not been previously investigated. In the present study, intracerebroventricular administration of mu-saporin resulted in significant death of cholinergic neurons and fibres in the medial septum, cortex and hip - pocampus of C57BL/6 mice. Arterial spin labelling MRI revealed a loss of CBF response to stimulation of eNOS by the Rho-kinase inhibitor fasudil hydrochloride in the cortex of denervated mice. By contrast, the hippocampus remained responsive to drug treatment, in association with altered eNOS expression. Fasudil hydrochloride significantly increased IPAD in the hippocampus of both control and saporin-treated mice, while increased clearance from the cor- tex was only observed in control animals. Administration of mu-saporin in the TetOAPPSweInd mouse model of AD was associated with a significant and selective increase in Aβ40-positive CAA. These findings support the importance of the interrelationship between cholinergic innervation and vascular function in the aetiology and/or progression of CAA and suggest that combined eNOS/cholinergic therapies may improve the efficiency of Aβ removal from the brain and reduce its deposition as CAA. Keywords: Alzheimer’s disease, Cholinergic, Clearance, eNOS, Vascular reactivity be one of the earliest indicators of the development of Introduction AD [38, 39] and differential perfusion of AD-sensitive Increasing evidence suggests that structural and func- brain areas such as the hippocampus, frontal and tempo- tional alterations of the cerebrovasculature contribute to ral lobes are present in people both with mild cognitive the aetiology and/or progression of Alzheimer’s disease impairment and dementia [2, 18, 34]. (AD). In fact, vascular pathology has been suggested to Cerebral amyloid angiopathy (CAA) is the most com- mon form of cerebrovascular pathology in AD [42] and is characterised by the deposition of β-amyloid (Aβ) pep- *Correspondence: c.hawkes@lancaster.ac.uk Department of Biomedical and Life Sciences, Lancaster University, tides in the walls of cerebral arteries and capillaries [90]. Lancaster, UK While parenchymal plaques are made up predominantly Full list of author information is available at the end of the article © The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, 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 included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/publi cdoma in/ zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Nizari et al. acta neuropathol commun (2021) 9:12 Page 2 of 17 of Aβ42, Aβ40 is more commonly observed in CAA. that stimulate calcium release and binding of calcium- CAA develops topographically, presenting initially in calmodulin to eNOS [25]. ACh activation of eNOS can the occipital lobe, followed by the temporal, frontal and also occur via the insulin-receptor substrate/PI3K/Akt parietal lobes, then in the hippocampus and entorhinal pathway [96] and stimulation of the PI3K/Akt/eNOS cortex at later stages [5, 83, 84]. In addition to causing pathway by the selective Rho- associated, coiled-coil dysfunction and death of mural and endothelial cells, containing protein kinase (ROCK) inhibitor fasudil recent studies suggest that CAA contributes to impaired hydrochloride, has been shown to increase CBF in mice hemodynamic responses in both individuals with AD and and humans [59, 68]. Although multiple downstream people with hereditary CAA [4, 63, 77, 87, 94]. signalling pathways are regulated by ROCK activity [45, A key pathological feature of sporadic CAA is a failure 76], several studies have reported no effect of fasudil −/− of clearance of Aβ from the brain, which is mediated via hydrochloride on cerebral haemodynamics in eNOS enzymatic degradation, uptake in microglia and astro- mice, suggesting that eNOS is the principal NOS iso- cytes and transcytosis across the blood–brain barrier [54, form targeted by fasudil hydrochloride [68, 78]. 55, 93]. Aβ is also removed from the brain along the walls Decreased expression of eNOS has been reported in of the capillaries and arteries via intramural periarterial the occipital cortex in AD, an area of the brain that is drainage (IPAD) and/or glymphatic drainage [7, 36, 56]. hypoperfused in AD [12]. Conversely, eNOS and induc- The IPAD hypothesis of Aβ clearance is based in part ible NOS (iNOS) activity have been shown to be signifi - on experimental observations that nanoparticles, solutes cantly increased in the temporal and frontal cortices of and Aβ injected into the interstitial fluid (ISF) of deep AD patients [20], in association with hyperperfusion of brain structures are transported along and localise to cer- those areas [34]. Several recent studies have reported ebrovascular basement membranes (CVBM) in cortical that endogenous CAA load is increased in eNOS-defi - and leptomeningeal vessels [3, 13, 27–29]. In the mouse cient mice in the absence of alterations in parenchymal brain, this process occurs very rapidly, within 5–10  min Aβ or increased Aβ production [6, 81], suggesting that of injection [8, 13, 27–29]. Since the pattern of distribu- dysfunction of eNOS may also contribute to the aeti- tion of solutes closely mimics that of Aβ accumulation ology of CAA and that this may be related to impair- in CAA and other angiopathies [14, 41, 85], failure of ments in Aβ clearance from the brain. IPAD is a key element of CAA pathology. However, as Loss of cholinergic neurons as an early pathologi- the movement of solutes along CVBMs is counter to the cal feature of AD has been known since the 1980s and direction of blood flow, the driving force that underlies underpins the rationale for the current clinical use of IPAD is still unknown. Recent mathematical modeling AChEIs for the treatment of AD [9, 22, 66]. Two recent suggests that oscillating pulsatile flow generated by the findings from the Alzheimer’s Disease Neuroimaging focal contraction and relaxation of arteries drives IPAD Initiative have reported that vascular dysregulation is [1] and this is supported by recent experimental data an early predictor of the progression to AD and that [3, 64, 88]. Localised arterial dilatation and contraction loss of volume in the basal forebrain precedes patholog- can occur both spontaneously (e.g. vasomotion) and in ical changes in the entorhinal cortex of individuals who response to neuronal activity (e.g. neurovascular cou- went on to develop AD [38, 73]. These findings suggest pling, NVC) and both mechanisms have been shown to that the interplay between loss of cholinergic innerva- be decreased in AD [19, 69, 79]. tion and vascular dysfunction may be important in the Smooth muscle cells that regulate arterial contrac- aetiology of AD. However, although some pathological tion and contribute to the regulation of cerebral blood studies have examined cholinergic loss at the neurovas- flow (CBF) in the cortex and hippocampus receive cular unit under experimental conditions [61] and in innervation from cholinergic neurons that originate AD [60], less has been done to directly investigate the in the basal forebrain. Release of acetylcholine (ACh) functional outcome of perivascular cholinergic dener- via stimulation of the basal forebrain or increasing vation. The aim of this study was to test the hypoth - cholinergic tone using acetylcholinesterase inhibitors esis that loss of cholinergic innervation decreases (AChEIs) has been shown to increase CBF in the cortex CBF and IPAD of Aβ from the cortex and hippocam- and hippocampus [50, 72]. ACh induces vasodilation pus of wildtype mice, leading to increased CAA in the primarily by stimulating the production of nitric oxide TetOAPPSweInd model of AD. (NO) via activation of endothelial nitric oxide synthase (eNOS) [23, 97], although stimulation of neuronal NOS Materials and methods (nNOS)-containing interneurons can also increase CBF Animals [15, 89]. ACh-induced activation of eNOS is medi- C57BL/6 mice were bred at the Open University (OU, ated principally by binding to muscarinic receptors Milton Keynes, UK) and the University of Southampton N izari et al. acta neuropathol commun (2021) 9:12 Page 3 of 17 (Southampton, UK). TetOAPPSweInd mice developed by pre-absorbing purified human Aβ40 peptide with the by Dr Joanna Jankowsky (Baylor College of Medicine, anti-Aβ40 antibody (10:1 molar ratio) alone or in combi- Texas, US) [40] were a generous gift from Dr JoAnne nation with the anti-Aβ42 antibody for 1.5 h at RT before McLaurin (Sunnybrook Research Centre, Toronto, Can- proceeding with tissue incubation and development as ada) and were also bred on a C57BL/6 background. Food described above (Additional file  2: Fig.  1a–d). Photomi- and water were provided ad libitum. All animal work was crographs were obtained using a Nikon Eclipse 80i light approved by the Animal Welfare and Ethics Research microscope (Nikon UK Limited, Surrey, UK) and images Boards (AWERB) at the OU, University of Southampton from the hippocampus and cortex (n = 6 control and and UCL in accordance with Home Office regulations n = 7 saporin) were analysed using Fiji (NIH, Maryland, and project licences (PPL 70/8507 and PPL 30/3095) USA). under the Animals (Scientific Procedures) Act 1986. For single labelling immunofluorescence, sections were washed in 0.01  M PBS, blocked with serum and incu- Mu‑Saporin administration bated overnight at 4  °C with anti-choline acetyltrans- 8–10 week old male C57BL/6 mice and 4-month old male ferase (ChAT; 1:75), anti-p75NTR (1:350), anti-GFAP and female TetOAPPSweInd mice were used for saporin (1:500), anti-Iba1 (1:500) or anti-laminin (1:350). Sec- injections. Mice were anesthetised under isoflurane gas tions were incubated with the appropriate fluorophore- and placed into a stereotaxic frame (Kopf instruments, conjugated secondary antibodies and coverslipped using CA, USA). Analgesia was administered intraperitoneally Mowiol (Sigma, Dorset, UK) containing 0.1% v/v Citif- (Carprieve, 5% w/v, 0.32  ml/kg, Norbrook, Northamp- luor (Citifluor ltd, London, UK). tonshire, UK) and a topical anaesthetic (Cryogesic (ethyl For multiple labelling fluorescent immunohistochem - chloride), Acorus Therapeutics Ltd, Chester, UK) was istry, sections underwent antigen retrieval (Additional applied before making a midline incision. 0.5 µL of mu- file  1: Table  1) and were then incubated overnight at saporin (0.596  µg/µl, Advanced Targeting Systems, CA, 4  °C with either i) anti-ChAT (1:75) and anti-p75NTR USA) or 0.9% sterile saline was injected into the left (1:400), or ii) with anti-NOS (1:200) or anti-eNOS (1:200) and right lateral ventricles (coordinates from Bregma: in combination with anti-GFAP (1:2000) or anti-Iba1 AP = − 0.4  mm, ML = ∓ 1.0  mm, DV = − 2.3  mm) using (1:500). Sections were then incubated for 2 h at RT with a 33 gauge Hamilton syringe. Mice were able to self- the appropriate fluorophore-conjugated secondary anti - administer sugar free jelly containing Carprofen (250  µg bodies, washed in PBS and coverslipped as above. The in 500 µl jelly, Zoetis, London, UK) for 1 week post-sur- specificity of the fluorescently-conjugated secondary gery. Animals were randomly assigned to receive either antibodies was verified by omitting the primary antibod - saline or saporin and all experimenters were blinded to ies (Additional file 2: Fig. 1e–h). treatment until statistical analysis. For all fluorescent imaging, photomicrographs were obtained using a Leica SP5 confocal microscope using Immunohistochemistry the same gain and intensity and maximum projection 45 days after surgery, mice were deeply anesthetised and images were exported to Adobe Photoshop 2020 or Fiji. perfused intracardially with 0.01  M phosphate buffered saline (PBS) followed by 4% paraformaldehyde (PFA). Quantification of immunohistochemistry Brains were post-fixed in 4% PFA overnight, sectioned The density of ChAT staining (neuronal cell bodies or (20 µm thickness) using a cryostat and stored at − 20  °C. fibres) in each brain region was quantified from low mag - Details of primary and secondary antibodies used for nification images by calculating the percentage area cov - immunohistochemistry are listed in Additional file  1: ered by staining using the “Analyze particle” function in Table 1. Fiji (NIH. Maryland, USA). For quantification of staining For enzyme-linked immunohistochemistry, sections in the hippocampus, overlapping images were stitched were washed in 0.01 M PBS, incubated with 3% hydrogen together and values from both the ipsilateral and con- peroxide, rinsed in PBS and treated with 70% formic acid tralateral hemispheres were averaged for each animal. For for 45 s. Sections were then blocked in 15% normal don- quantification of cortical images, six random non-over - key or goat serum (NDS, Sigma-Aldrich, Dorset, UK), lapping images spanning the somatosensory cortex of followed by incubation overnight at 4 °C with anti-Aβ40 the ipsilateral cortex to the somatosensory cortex of the (1:100) or anti-Aβ42 (1:100). The next day, sections were contralateral cortex were captured and averaged per ani- incubated with biotinylated anti-rabbit (1:400) and devel- mal. A single low magnification image/animal was used oped using glucose oxidase enhancement with DAB as to quantify ChAT staining in the medial septum. The per - chromagen (Sigma-Aldrich, Dorset, UK). The specificity centage area containing microglia, astrocytes and blood of the anti-Aβ40 and anti-Aβ42 antibodies was verified vessels was also calculated using the ‘Analyze particle” Nizari et al. acta neuropathol commun (2021) 9:12 Page 4 of 17 function in Fiji. Additionally, for anti-laminin staining, in overdose of sodium pentobarbitone and perfused with order to quantify density by vessel type, a mask was set 0.01  M PBS. Brains were removed, dissected for hip- to select capillaries (0–100  µm ) or large-diameter ves- pocampus and cortex, snap frozen and stored at − 80  °C sels (101  µm -infinity) using images calibrated accord - until use. Tissues were homogenised in RIPA lysis buffer ing to the scale bar. The degree of colocalization between (20 mM Tris pH 8.0, 0.15 M NaCl, 1.27 mM EDTA, 1 ml ChAT and p75NTR and between e/NOS and laminin Igepal, 0.1% SDS, 50 mM NaF, 1.48 mM NaVO contain- was determined from three images/region/animal taken ing 1:100 Protease inhibitor cocktail [Merck Millipore, at × 40 magnification using Pearson’s correlation coeffi - UK]), centrifuged at 10,000  g at 4  °C for 10  min and the cient (PCC, Coloc 2 plugin) in Fiji. For quantification of supernatant was collected. 30 μg or 40 μg of proteins Aβ staining, images from the hippocampus and cortex were separated by gel electrophoresis and membranes taken at × 4 magnification were stitched together, images were then blocked in 8% non-fat milk before incubation were converted to 8 bit greyscale images and the % area with anti-eNOS (1:5000, Cell Signalling Technology, Lon- covered by CAA and plaques were quantified separately don, UK) or anti-nNOS (1:250, Cell Signalling Technol- in Fiji. ogy) overnight at 4  °C. Membranes were then washed in TBST before being incubated in HRP-conjugated Arterial spin labelling MRI anti-rabbbit (1:5000, Fisher Scientific) for 1  h at room Approximately 5  weeks after ICV injection, C57BL/6 temperature and developed using an enhanced chemi- mice that received saline (n = 7) or mu-saporin (n = 7) luminescence kit (GE Healthcare, Little Chalfont, UK). were transported to UCL Centre for Advanced Imag- Membranes were then stripped and re-probed with anti- ing and allowed to acclimate for 1  week before imaging. GAPDH (1:50,000, Sigma-Aldrich) to ensure equal pro- A 9.4 T VNMRS horizontal bore scanner (Agilent Inc., tein loading. Optical density of the bands was quantified Santa Clara, CA, USA) with a 72 mm inner diameter vol- and normalised to GAPDH levels using Fiji. ume coil and 2 channel array head coil (Rapid Biomedi- cal, Columbus, OH, USA) was used for radio frequency Assessment of IPAD transmission and signal detection. Mice were initially 45  days after injection with saline or saporin, mice were anaesthetised under 2% isoflurane in medical air and anesthetised with isoflurane and placed into a stere - maintained under 1.5% during imaging. A rectal probe otaxic frame. For hippocampal injections, 0.5 µL of and a pressure pad (SA Instruments, Stony Brook, NY, 50  µM human Aβ40 HiLyte Fluor 555 (AnaSpec, Cali- USA) were used to measure core temperature and moni- fornia, USA) was injected into the left hippocampus (co- tor respiration throughout the procedure. Heated water ordinates from Bregma: AP = − 1.9  mm, ML = 1.5  mm, tubing and a warm air blower using a feedback system DV = − 1.7 mm, n = 16 control and n = 14 saporin). Mice (SA Instruments, Stony Brook, NY, USA) was used to were perfused with PBS and 4% PFA 5 min post-injection. regulate the temperature of the mice to 37 °C. Following For cortical injections, control (n = 8) and saporin (n = 7) a 5 min acquisition of baseline CBF, mice were adminis- mice were injected with 0.25 µL of 50  µM Aβ40 HiLyte tered 10 mg/kg Fasudil hydrochloride i.p. (Tokyo Chemi- Fluor 555 into the right cortex (co-ordinates from cal Industries, Tokyo, Japan) and re-imaged 10  min Bregma AP = − 2  mm, ML = − 1.5  mm, DV = − 0.5  mm) later for an additional 5  min. At the end of the imaging and mice were perfused 2.5 min later. All injections were experiments, mice were perfusion fixed with 4% PFA carried out at a rate of 0.2 µL/min using a 33 gauge Ham- and their brains collected for immunohistochemistry. A ilton syringe and the injection needle was left in situ for total of 15 brain image slices were acquired with a thick- 2  min to avoid reflux. A separate group of control and ness of 1 mm and an ‘in-plane’ resolution of 0.28 mm per saporin-treated mice (n = 5/group) were administered mouse per experiment. Statistical Parametric Mapping fasudil hydrochloride (10  mg/kg, i.p.) 10  min before (SPM, http://www.fil.ion.ucl.ac.uk/spm/) was applied to intracerebral injections. Tissue sections were processed perfusion-weighted acquired ASL images [91]. Acquired for double-labeling immunohistochemistry as described images were processed using a Matlab (Mathworks, MA, above using anti-laminin (1:350) and anti-α smooth mus- USA) customised script. Regions of interest (cortex and cle actin conjugated to FITC (1:350; Additional file  1: hippocampus) were then manually traced on a single slice Table  1). Brain sections that were ≥ 400 μm away from and quantified using a Matlab script which converted the site of injection were imaged for quantification. The pixel intensity into CBF (ml/100 g/min) [91]. number of capillaries, arteries and veins that contained Aβ40 HiLyte Fluor 555 within each image were counted Western blotting manually and divided by the total area analysed, as 45  days post-injection with saline (n = 7) or mu- described previously [27, 29, 61]. p75-saporin (n = 6), C57BL/6 mice were given an N izari et al. acta neuropathol commun (2021) 9:12 Page 5 of 17 Statistical analysis increase in CBF in the control, but not the saporin group Data were tested for normality using the Shapiro–Wilk compared to baseline (Fig.  2b). In addition, control mice test and the ROUT test was used to identify and remove that were administered fasudil hydrochloride had a sig- statistical outliers. Comparisons between control and nificantly higher CBF compared to saporin-treated mice saporin-treated mice were analysed using two-tailed given the drug (Fig. 2b). These results suggest that dener - Student’s t test or Mann–Whitney U test where data vated hippocampal vessels were still responsive to eNOS were not normally distributed. Analysis of baseline vs stimulation, while cortical vessels were not. stimulated CBF was carried out using paired one-tailed t-test and Wilcoxon matched-pairs signed rank test. Dif- eNOS protein expression in the hippocampus and cortex ferences in NOS activity were analysed using one-way is differentially affected by saporin treatment ANOVA with Sidak post hoc test. Differences in counts To determine whether regional differences in the respon - of Aβ40-positive vessels within each brain region were siveness to fasudil hydrochloride were due to differences analysed using a one-way ANOVA with Sidak post hoc in the levels of NOS expression, cortical and hippocam- analysis or Kruskal–Wallis test with Dunn’s post hoc. In pal tissues were assessed by Western blotting using all cases, significance was set at p < 0.05 and data are dis- eNOS and nNOS-specific antibodies. eNOS expression played as mean ± SEM. was significantly higher in the hippocampus of saporin- treated mice compared to controls (Fig. 3a), while no sta- Results tistically significant differences were observed between mu‑Saporin induces loss of cholinergic neurons and fibres control and saporin mice in the cortex (Fig.  3b). Levels In control mice, immunohistochemistry for ChAT of nNOS did not differ significantly between control and labelled neurons in the medial septum (MS), diago- saporin-treated mice in either the hippocampus or cor- nal band of Broca (DBB) and striatum (Fig.  1a). ChAT- tex (Fig. 3c, d). To determine if NOS expression may have positive fibres in the hippocampus and cortex were also been influenced by possible differences in vessel densi - observed in these animals (Fig.  1b, c). Colocalization ties, the vascular expression of NOS in laminin-positive was noted between ChAT and p75NTR in the majority vessels was quantified in control and saporin tissues of basal forebrain neurons (PCC = 0.68) as well as in fibre (Fig.  3e–h). Quantification of the NOS-to-laminin ratio projections in the hippocampus and cortex (PCC = 0.25 confirmed the significant decrease in NOS expression in and 0.24, respectively) in control animals (Additional the cortex of saporin-treated mice (Fig.  3m). However, file  3: Fig.  2). Administration of mu-saporin induced a the NOS ratio in the hippocampus did not differ signifi - significant loss of ChAT-positive, p75NTR-positive neu - cantly between control and saporin mice (Fig.  3m). We rons in the MS and DBB (Fig.  1d, g, Additional file  3: also observed some NOS expression in glial cells in the Fig.  2), as well as fibres in the hippocampus (Fig.  1e, h, hippocampus of saporin-treated mice. To determine if Additional file  3: Fig. 2) and cortex (Fig.  1f, i, Additional the increased eNOS expression detected by Western blot file  3: Fig.  2), confirming the usefulness of the model to was due to expression in glial cells, hippocampal sections induce significant death of basal forebrain cholinergic were stained with anti-eNOS and anti-GFAP or anti- neurons and their projection fibres. Iba1 (Fig. 3i–l). These results confirmed minimal expres - sion of eNOS in astrocytes, but some colocalization of Cholinergic loss decreases eNOS‑mediated cerebral blood eNOS with Iba1-positive microglia, which was higher in flow in the cortex but not the hippocampus saporin-treated mice, although this did not reach statisti- We have previously found that mu-saporin causes loss of cal significance (p = 0.13, Fig.  3n). These results suggest cholinergic innervation of cerebral blood vessels and that that levels of eNOS are downregulated in the cortex, and this denervation is more pronounced in the cortex than upregulated in the hippocampus of saporin-treated mice the hippocampus [61]. To assess the effect of this loss on and that increased eNOS expression in the hippocampus baseline and evoked CBF, arterial spin labelling MRI was may be due in part to upregulation by microglia. used to image cerebral perfusion in the hippocampus and cortex of control and saporin-treated mice. Baseline CBF Administration of fasudil hydrochloride increases IPAD did not differ between control and saporin mice in either in the hippocampus but not cortex of denervated mice the hippocampus or cortex (Fig.  2a, b). Administration Our previous work has shown that saporin treatment of fasudil hydrochloride caused a significant increase in significantly decreases cholinergic innervation of arterial hippocampal CBF relative to baseline in both control smooth muscle cells in the hippocampus and cortex [61]. and saporin-treated mice (Fig.  2a). The degree of CBF In vitro modelling supports the hypothesis that the local- increase was similar between treatment groups (Fig. 2 a). ised arterial pulsations that regulate CBF [32] also pro- In the cortex, fasudil hydrochloride induced a significant vide the principle driving force for solute clearance from Nizari et al. acta neuropathol commun (2021) 9:12 Page 6 of 17 Fig. 1 Saporin administration kills cholinergic neurons and fibres in wildtype mice. a–f Photomicrographs of ChAT staining in the medial septum and diagonal band of Broca (a and d), hippocampus (b and e) and cortex (c and f) in control (a–c) and mu-saporin treated C57BL/6 mice (d–f). (g‑i), Quantification of % area covered by ChAT-positive neurons in the medial septum (g) and fibres in the hippocampus (h) and cortex (i). Scale bar = 100 μm. n = 5/group, *p < 0.05, **p < 0.01 the brain via IPAD [1]. To determine if loss of choliner- in capillaries and arteries in both the hippocampus and gic innervation altered IPAD, the pattern of distribution cortex (Fig.  4a–f ). Quantification of the number of hip - of human Aβ40-AF555 was evaluated following injection pocampal blood vessels that contained Aβ showed no into the hippocampus or cortex of control and saporin- difference between control and saporin-treated mice treated C57Bl/6 mice under physiologic and stimulated under baseline physiological conditions (Fig. 4c). Admin- conditions. Triple-labelling immunohistochemistry istration of fasudil hydrochloride resulted in significantly demonstrated the presence of Aβ40-AF555 primarily more Aβ-positive blood vessels in both control and N izari et al. acta neuropathol commun (2021) 9:12 Page 7 of 17 These findings confirm that IPAD was not affected by differences in vessel density or glial activation and indicate that clearance of Aβ was stimulated by fasudil hydrochloride and that this responsiveness remains intact in the hippocampus, but not the cortex of dener- vated mice. Loss of cholinergic innervation increases CAA in the hippocampus of TetOAPP mic ‑ e Fig. 2 The hippocampus, but not cortex, of denervated mice To evaluate if cholinergic denervation potentiated Aβ remains responsive to eNOS-stimulated increase in CBF. a and b pathology, 4-month old TetO-APPSweInd mice were Quantification of cerebral blood flow (CBF) in the hippocampus (a) and cortex (b) of control (con) and saporin-treated mice (sap) at administered saline or mu-saporin. Unexpectedly and baseline and 10 min after administration of fasudil hydrochloride in contrast to the observations made in the C57BL/6 (+F), averaged over 5 min. n = 5–7/group, *p < 0.05 mice, no significant differences were noted in the num - ber of ChAT-positive neurons in the medial septum between control and saporin-treated mice (Fig.  5a, d, g). Significantly fewer cholinergic fibres were observed saporin animals compared to baseline (Fig.  4b, c). How- in the hippocampus of saporin vs control mice (Fig.  5b, ever, fasudil treatment did not affect hippocampal vessel e, h), while ChAT fibre density in the cortex was also counts between control vs. saporin mice (Fig. 4c). unaffected by saporin treatment (Fig.  5c, f, i). To deter- Preliminary assessment of IPAD in the cortex using mine if the attenuated effect of saporin in the TetO- the same parameters as those in the hippocampus (e.g. APPSweInd mice was due to endogenous differences in 0.5 μL Aβ40-AF555 + 5  min clearance) revealed a much ChAT and p75NTR expression, fibre appearance and smaller bolus of Aβ at the site of injection and very few density was compared between TetO-APPSweInd mice Aβ-positive vessels were visible at 400 µm away from the and wildtype littermates. The morphology of fibres in injection site compared to the hippocampus (Additional the TetO-APPSweInd appeared dystrophic, with swollen file  4: Fig.  3a and b). Following a series of modifications varicosities and shorter processes than that of cholinergic (Additional file  4: Fig.  3c–e), the injection protocol for fibres in the wildtype mice (Figs.  1b, c, 5b, c). The den - cortical injections was adapted to 0.25 µL Aβ40-AF555 sity of ChAT-positive neurons in the MS was significantly with 2.5  min post-injection time (Fig.  4d–f ), to allow for higher in TetO-APPSweInd mice compared to wildtype sufficient numbers of Aβ40-positive vessels to be counted. animals, although no differences in hippocampal or cor - Similarly to the hippocampus, quantification of corti - tical ChAT fibre density were observed between strains cal vessels that contained Aβ revealed no baseline differ - (Fig.  5j). Analysis of p75NTR expression showed signifi - ences between control and saporin-treated mice (Fig. 4f ). cantly lower receptor expression in the hippocampus of In control animals, administration of fasudil hydrochlo- TetO-APPSweInd mice compared to wildtypes, while no ride resulted in significantly fewer Aβ40-containing ves - differences were observed in the cortex or MS (Fig.  5k). sels (Fig. 4d–f ). Although a similar trend was observed in Additional analysis found that the ratio of p75NTR to saporin animals, the difference was not statistically signif - ChAT expression was significantly lower in the MS and icant (p = 0.08) and no difference was observed between hippocampus of TetO-APPSweInd mice compared to control + fasudil and saporin + fasudil groups (Fig. 4f ). wildtype animals (Fig. 5l). To determine if IPAD may have been influenced by dif - Quantification of Aβ pathology in the hippocampus ferences in vessel number and/or microglia and astrocyte after saporin treatment showed no difference in the per - activation, densities of each were quantified in control centage area covered by Aβ40-positive plaques between and saporin-treated mice. The density of laminin-posi - control and saporin mice (Fig.  6a, b, e). However, Aβ40 tive macrovessels and capillaries did not differ between CAA load was significantly higher in the saporin-treated control and saporin-treated mice in either the cortex or mice (Fig.  6a, b, e). A similar but non-significant pattern hippocampus (Fig.  4g–j), although capillary density was of vascular Aβ42 staining was observed between con- significantly higher in the cortex than the hippocampus trol and saporin mice, while parenchymal Aβ42 was not in both treatment groups (p = 0.003). Similarly, quanti- affected by saporin treatment (Fig.  6a, b, f ). In the cor- fication of Iba1 and GFAP staining revealed no effect of tex, no differences were observed between control and saporin treatment on astrocyte or microglial coverage of saporin-treated mice in the density of Aβ40-positive the hippocampus or cortex (Additional file 4: Fig. 3f–k). Nizari et al. acta neuropathol commun (2021) 9:12 Page 8 of 17 Fig. 3 Regional variation in NOS expression and activity in control and saporin-treated mice. a–d Western blots and quantification of levels of eNOS (a and b) and nNOS (c and d) in the hippocampus (a, c) and cortex (b, d) of control and saporin-treated mice (n = 6–8/group). Molecular weight markers (kDa) are shown on the right hand side. The black line demarcates the original blot (upper) and the same blot re-probed for loading control (lower). e–h Photomicrographs showing the expression of total NOS (green) in laminin-positive vessels (blue) in the hippocampus (e and f) and cortex (g and h) of control (e and g) and saporin-treated mice (f and h). Note the stable expression of NOS in hippocampal vessels of saporin-treated mice, while NOS expression is significantly reduced in cortical vessels of saporin animals. i–l eNOS expression (green) in GFAP-positive astrocytes (blue, i and j) and Iba1-positive microglia (blue, k and l) in the hippocampus of control (i and k) and saporin mice (j and l). Colocalization between eNOS and GFAP or Iba-1 is shown as white-turquoise. m and n Quantification of NOS expression in blood vessels as a ratio to overall vessel density (m) and degree of colocalisation between eNOS and GFAP or Iba1 as measured by the Pearson’s correlation coefficient (n). n = 5/group. Scale bars for f and h = 50 μm; j and l = 20 μm. *p < 0.05 N izari et al. acta neuropathol commun (2021) 9:12 Page 9 of 17 Fig. 4 Administration of fasudil hydrochloride increases IPAD in the hippocampus, but not cortex of denervated mice. a–f Photomicrographs showing the distribution of human Aβ40-AF555 (red) at 5 min post-injection into the hippocampus (a and b) and at 2.5 min after injection into the cortex (d and e) of control mice at baseline (a and d) and after administration of fasudil hydrochloride (b and e). The cerebrovascular basement membrane was labelled with anti-laminin (blue) and smooth muscle cells were identified with anti-α smooth muscle actin (green). Quantification of the total number of Aβ40-containing vessels in the hippocampus (c) and cortex (f) of control (con) and saporin (sap)-treated mice at baseline (n = 14–16 for hippocampus and n = 7–8 for cortex) and after fasudil hydrochloride (+F) (n = 5/group for both regions). g and h Photomicrographs of laminin staining in the hippocampus (g) and cortex (h) of control (upper panels) and saporin-treated mice (lower panels). i and j Quantification of % area covered by laminin in the hippocampus (i) and cortex (j) of control and saporin animals. n = 5–7/group. Scale bars = 100 μm. *p < 0.05, **p < 0.01, ***p < 0.001 plaques or CAA (Fig.  6c, d, g). Likewise, the density of Discussion cortical parenchymal and vascular Aβ42 was unaffected Results from this study suggest that loss of cholinergic by saporin treatment (Fig. 6c, d, h). innervation differentially affects cortical and hippocampal As with the C57BL/6 mice, vessel density between responsiveness to eNOS-stimulated increases in CBF and control and saporin-treated TetO-APPSweInd mice was IPAD in wildtype mice, with hippocampal, but not cortical, similar in both the cortex and hippocampus (Fig.  6i–l). vessels remaining responsive to stimulation. The death of Analysis of GFAP and Iba1 expression revealed no sig- cholinergic nerve fibres resulted in a significant and selec - nificant difference in area coverage between treatment tive increase in Aβ40-positive CAA in the TetO-APPS- groups in either brain region (Additional file  4: Fig. 3l–q). weInd model of AD. These findings support the importance These findings confirm that saporin administration did of the interrelationship between cholinergic innervation not significantly alter vessel density or glial activation in and vascular function in the aetiology and/or progression the TetO-APP mice and support a role for loss of cholin- of CAA and suggest that regional vulnerability or resilience ergic innervation in potentiating CAA pathology. Nizari et al. acta neuropathol commun (2021) 9:12 Page 10 of 17 Fig. 5 Distribution and quantification of cholinergic and p75NTR-positive neurons in control and saporin-treated TetO-APPSweInd mice. a–f Photomicrographs of ChAT staining in the medial septum and Diagonal band of Broca (a and d), hippocampus (b and e) and cortex (c and f) in control (a–c) and mu-saporin treated (d–f) TetO-APPSweInd mice. (g–i), Quantification of % area covered by ChAT-positive neurons in the medial septum (g) and fibres in the hippocampus (h) and cortex (i), n = 4–5/group. j–l Quantification of % area covered by ChAT (j) and p75NTR-positive (k) neurons and fibres and the ratio of ChAT:p75NTR expression (l) in the medial septum (MS), hippocampus (Hippo) and cortex of C57BL/6 and TetO-APPSweInd ( TETAPP) mice, n = 5/group/strain. Scale bar = 100 μm. *p < 0.05, **p < 0.01,* **p < 0.001 to loss of cholinergic dysfunction may contribute to the cognitive impairment [9, 22, 26, 82, 86]. In agreement topographical nature of CAA (Additional file 4 : Fig. 3). with previous studies [58, 61], we found that intracer- Degeneration of cholinergic neurons and shrinkage of ebral administration of mu-saporin, which selectively the basal forebrain are early features of AD and are asso- targets p75NTR-expressing neurons, caused the death ciated with increased Aβ pathology, altered CBF and of ChAT-positive neurons in the MS as well as their fibre N izari et al. acta neuropathol commun (2021) 9:12 Page 11 of 17 Fig. 6 Loss of cholinergic innervation selectively increases Aβ40-positive CAA in the hippocampus of TetO-APP mice. a–h Photomicrographs of hippocampal (a and b) and cortical tissues (c and d) of TetO-APP mice stained with antibodies against human Aβ40 (a–d, left panels) and Aβ42 staining (a–d, right panels). Tissues from control animals are shown in the upper panels and saporin-treated tissues are shown in the lower panels. Arrowheads show plaques and asterisks show CAA-positive vessels. e–h Quantification of % area covered by Aβ40 (e and g) and Aβ42-positive (f and h) plaques and blood vessels in the hippocampus (e and f) and cortex (g and h) of TetO-APPSweInd mice, n = 6–7/group. i–l Photomicrographs of laminin staining in the hippocampus (i) and cortex (j) of control (upper panels) and saporin-treated mice (lower panels). k Histogram showing quantification of laminin density in control and saporin-treated mice. n = 3–5/group. Scale bars = 100 μm. **p < 0.01 projections in the cortex and hippocampus of wildtype Therefore, to evaluate the impact of loss of cholinergic mice. innervation on evoked CBF, we mimicked ACh activa- ACh has a well-known vasodilatory effect in the brain tion of eNOS by using the selective ROCK inhibitor fas- and stimulation of the basal forebrain leads to increased udil hydrochloride, which has been shown to increase cortical CBF [33, 35]. This effect is predominantly CBF by stimulating the PI3K/Akt/eNOS pathway [68, observed during NVC when release of ACh stimulates 76]. Consistent with previous reports [50], loss of cholin- the production of NO via activation of eNOS or indi- ergic innervation in the cortex and hippocampus did not rectly by stimulation of nNOS-containing interneurons affect baseline CBF in either region, supporting a primary [53]. Although ASL MRI can be used to measure NVC in role of ACh on CBF during NVC. However, while admin- the cortex using whisker or forepaw stimulation [49], to istration of fasudil hydrochloride was not able to evoke our knowledge similar methods are not available to stim- a change in CBF in the cortex of saporin-treated mice, ulate NVC in the hippocampus of anesthetised animals. Nizari et al. acta neuropathol commun (2021) 9:12 Page 12 of 17 denervated vessels in the hippocampus remained respon- cortex (0.5  mm from dura) may have flooded the suba - sive to stimulation. rachnoid space, even when using the smaller 0.25 µL vol- Because multiple downstream signalling pathways in ume. Therefore, more detailed in vivo tracer experiments addition to eNOS are regulated by ROCK activity, includ- are needed to clarify rates of IPAD between cortical and ing those relating to smooth muscle contraction [45], we hippocampal regions. However, in agreement with other cannot definitively conclude that the observed effects studies [21, 62], we also found that cerebrovascular den- were due to stimulation of eNOS. However, the find - sity was significantly higher in the cortex compared to ings that levels of eNOS were significantly decreased in the hippocampus. This larger surface area may allow the cortex and increased in the hippocampus of saporin- for solutes contained within the ISF to be more rapidly treated mice, support the hypothesis that loss of cholin- removed from the cortex than from the hippocampus ergic innervation resulted in opposing effects on eNOS under physiological conditions. expression that aligned with the CBF response. Although Administration of fasudil hydrochloride resulted in eNOS is principally expressed by endothelial cells, pre- significantly more vessels with Aβ in the hippocampi of vious studies have reported its expression in neurons, both control and saporin-treated mice, while in the cor- astrocytes and in microglia across various species [16, 75, tex, fewer vessels were found to contain Aβ and this was 92]. Our observation that eNOS was expressed not only observed in control mice only. Although the pattern of in blood vessels but also by microglia in the hippocam- distribution was opposite between the two regions, we pus of saporin-treated mice, suggests that the functional interpret both findings as representing increased IPAD effects of fasudil hydrochloride in the hippocampus may at different rates of clearance. These findings are consist - also be due in part to activation of non-vascular cells. ent with our CBF data and suggest that IPAD is signifi - Although it is not clear why the effect of saporin treat - cantly increased in the presence of vasodilation, which is ment induced an opposite expression of eNOS between in agreement with reports of impaired solute clearance the cortex and hippocampus, endogenous NOS activ- from the brain during hypoperfusion [3, 37]. However, ity in both the nNOS- and eNOS-enriched fractions has the similarity in the number of labeled cortical vessels previously been reported to be higher in the hippocam- between control + fasudil and saporin + fasudil mice pus compared to the cortex [74]. This is supported by suggests that other factors are also contributing to Aβ previous reports showing that changes in CBF in the hip- clearance in denervated mice. Although blood pressure pocampus were more proportional to changes in nNOS was not monitored in the current experiments, previ- activity than in the cortex [52] and that the cortex is more ous studies have shown that fasudil hydrochloride does sensitive than the hippocampus to inhibition of nNOS not alter systolic blood pressure in normotensive rodents activity [43]. or humans [44, 51, 57, 59], suggesting that the observed Previous studies have suggested that contractions of effects were unlikely to be due to changes in peripheral arterial smooth muscle cells are required for drainage blood pressure. In addition, no differences in vessel den - of fluid along cerebral blood vessels [1, 3, 37], although sity or markers of microglia and astrocytes were observed whether this pulsation is sufficient to drive bulk flow of between control and saporin mice in either brain region. ISF and CSF remains controversial [11, 31]. Several stud- Although our findings are consistent with reports of an ies have shown that vasoreactivity in AD is improved association between decreased eNOS expression and following treatment with AChEIs [71]. We hypothesised increased CAA [6, 81], recent work has shown that NVC that there is a direct relationship between vasoreactivity is mediated in part by arteriole caveolae independent of and the efficiency of IPAD and that loss of cholinergic eNOS activation [17]. Further work is required to deter- innervation would impair IPAD of Aβ in a similar pat- mine the factors that regulate Aβ clearance when cholin- tern to that observed for CBF. No differences in IPAD ergic signalling is attenuated. were observed between control and saporin-treated mice Previous studies have reported a relationship in either brain region under baseline physiological con- between basal forebrain degeneration and Aβ pathol- ditions. Our observation that fewer Aβ-positive vessels ogy in the cortex [26] and basal forebrain atrophy has were visible in the cortex after a 5  min diffusion period been suggested to predict cortical Aβ burden [82]. compared to the hippocampus, suggests that IPAD of Aβ Induced loss of cholinergic neurons in rodent mod- may be endogenously faster in the cortex than in the hip- els of AD has also been associated with increased Aβ pocampus. We have previously reported differences in plaque deposition [48, 67], however most studies have the efficiency of IPAD between subcortical brain regions not specifically investigated the effect on vascular Aβ. that are differentially affected by CAA [27]. However, In the present study, administration of mu-saporin in given the relatively small thickness of the mouse cortex TetO-APPSweInd mice resulted in a loss of cholinergic [65], it is possible that the depth of injection into the neurons that was only significant in the hippocampus. N izari et al. acta neuropathol commun (2021) 9:12 Page 13 of 17 The reasons for the discrepancies between the degree animal model that are not present in AD. In addition, of loss between the C57BL/6 and TetO-APPSweInd as age is the major risk factor for both sporadic AD and mice are not clear, but may relate to the dystrophic CAA, additional experiments are needed to determine appearance of cholinergic fibers and decreased whether the effects of cholinergic denervation on CBF p75NTR:ChAT ratio observed in the TetO-APPSweInd and IPAD in the cortex and hippocampus seen here in mice. As binding of the p75NTR by Aβ is known to young adult mice are also observed in aged animals. induce apoptosis [95], it may be that pre-existing Aβ pathology caused damage to cholinergic neurons Conclusions and fibres that induced a downregulation in p75NTR Despite these limitations, findings from this study sup - expression and decreased receptor availability for mu- port a role for loss of cholinergic innervation in the aeti- saporin binding. ology and/or progression of CAA and suggest that this Although unexpected, the difference in sensitivity to may be related to eNOS-mediated vasodynamics that saporin treatment between the cortex and hippocam- contribute to clearance of Aβ from the brain via IPAD pus provided an internal control to study the effect of pathways. Therefore, combined targeting of eNOS and cholinergic loss on Aβ pathology. We found that loss cholinergic signalling/activation may represent a new of cholinergic innervation in the hippocampus was mechanism to improve the efficiency of Aβ removal and associated with a significant increase in Aβ40-positive reduce its deposition as CAA. vessels, consistent with the preferential deposition of Aβ40 in the vasculature in AD [30, 80]. By contrast, Supplementary Information CAA load was not affected in the cortex where cholin - The online version contains supplementary material available at https ://doi. ergic fibre density was not altered by saporin treatment. org/10.1186/s4047 8-020-01108 -z. Parenchymal plaque load did not differ between con - trol and saporin-treated mice in either region. These Additional file 1: Table 1 List of source of primary and secondary anti- findings are consistent with previous studies showing bodies used for immunohistochemistry. significantly more endogenous CAA in the absence of Additional file 2: Fig. 1 a–d Photomicrographs of diffuse parenchymal changes in parenchymal changes or changes in APP plaques identified by the anti-Aβ40 antibody (a) and senile plaques stained by the anti-Aβ42 antibody (b) in TetO-APPSweInd mice. No stain- processing in rabbits administered saporin [10, 70]. ing was observed after pre-absorption of the Aβ40 antibody with Aβ40 These findings are also similar to a study which found peptide (1:10 molar ratio, c). Sections incubated after pre-absorption of that age-related loss of perivascular cholinergic inner- Aβ40 with anti-Aβ40 + anti-Aβ42 (d) showed a similar pattern of staining to that of sections incubated with the anti-Aβ42 antibody alone. e–h No vation in the cortex did not significantly correlate with staining was observed in tissue sections from C57Bl/6 mice incubated increased cortical plaque load in the Tg2576 AD mouse with fluorescently-conjugated secondary antibodies alone. model [46]. However, other studies have reported Additional file 3: Fig. 2 a–f Photomicrographs showing expression of increased plaque load and elevated concentrations of ChAT (green), p75NTR (red) and their colocalization (yellow) in neurons soluble Aβ following saporin-induced cholinergic loss in the medial septum (a and b) and fibers in the hippocampus (c and d) and cortex (e and f) of C57Bl/6 mice. Animals received an intracerebroven- in the APP/PS1 and Tg2576 mouse models [24, 48, tricular injection of either PBS (control, a, c and e) or mu-saporin (b, d and 67]. Many factors may have contributed to these differ - f). Saporin treatment significantly reduced expression of p75NTR, ChAT- ent observations, including the degree of cholinergic positive cell bodies and fibers in the medial septum (b), hippocampus (d) and cortex (f). Images of the hippocampus are composed of individual degeneration, age of the mice and amount of pre-exist- overlapping images stitched together using Fiji. Scale bars: a, b, e, f = 250 ing Aβ pathology before saporin treatment, as well as μm; c and d = 100 μm. the ratio of Aβ40:Aβ42 and progression of pathology Additional file 4: Fig. 3 a and b Photomicrographs showing the distribu- between the different mouse models. Despite these dis - tion of 0.5 μL human Aβ40-AF555 (red) at 400 μm away from the injection site after 5 min post-injection (PI) into the hippocampus (a) and cortex (b) crepancies, our results support a consensus that loss of control mice. The cerebrovascular basement membrane was labelled of cholinergic innervation contributes to increased Aβ with anti-laminin (blue) and smooth muscle cells were identified with pathology. anti-α smooth muscle actin (green). c–e Photomicrographs showing the distribution of Aβ40-AF488 (green) in the cortex of C57BL/6 mice, at 400 In addition to the previously discussed limitations μm away from the injection site. The volume and post-injection (PI) time related to inducing NVC in the hippocampus of anes- is indicated for 3 combinations that were tested to determine the optimal thetised animals and assessment of IPAD ex  vivo, this parameters for quantification of Aβ-positive vessels. f–k Photomicro - graphs and quantification of GFAP (f, g, j) and Iba1 (h, i, k) staining in the study has several other weaknesses. The saporin model hippocampus (f, h) and cortex (g, i) of control (con, upper panels) and induces loss of basal forebrain cholinergic neurons saporin-treated C57Bl/6 mice (sap, lower panels). l–q Photomicrographs in a retrograde manner [47], and over a more rapid and quantification of GFAP (l, m, p) and Iba1 (n, o, q) staining in the hippocampus (l, n) and cortex (m, o) of control (con, upper panels) and timeframe than that observed in AD, which may acti- saporin-treated TetO-APPSweInd mice (sap, lower panels). n = 3–5/group. vate a strong acute inflammatory reaction and/or the Scale bars = 100 μm. development of compensatory mechanisms in the Nizari et al. acta neuropathol commun (2021) 9:12 Page 14 of 17 Abbreviations 4. Arvanitakis Z, Leurgans SE, Wang Z, Wilson RS, Bennett DA, Schneider JA 7-NI: 7-Nitroindazole; Aβ: Beta amyloid; ACh: Acetylcholine; AChEI: Acetylcho- (2011) Cerebral amyloid angiopathy pathology and cognitive domains in linesterase inhibitor; AD: Alzheimer’s disease; APP: Amyloid precursor protein; older persons. Ann Neurol 69:320–327 ASL: Arterial spin labelling; CAA : Cerebral amyloid angiopathy; CBF: Cerebral 5. Attems J, Quass M, Jellinger KA, Lintner F (2007) Topographical distribu- blood flow; CVBM: Cerebrovascular basement membranes; ChAT: Choline tion of cerebral amyloid angiopathy and its effect on cognitive decline acetyltransferase; eNOS: Endothelial nitric oxide synthase; GAPDH: Glyceral- are influenced by Alzheimer disease pathology. J Neurol Sci 257:49–55 dehyde 3-phosphate dehydrogenase; GFAP: Glial acidic fibrillary protein; Iba1: 6. Austin SA, Katusic ZS (2020) Partial loss of endothelial nitric oxide leads Ionized calcium binding adaptor molecule 1; ICV: Intracerebroventricular; to increased cerebrovascular beta amyloid. J Cereb Blood Flow Metab IPAD: Intramural periarterial drainage; iNOS: Inducible nitric oxide synthase; 40:392–403. https ://doi.org/10.1177/02716 78X18 82247 4 ISF: Interstitial fluid; L-NAME: Nitro- l -arginine Methyl Ester Hydrochloride; MRI: 7. Bakker EN, Bacskai BJ, Arbel-Ornath M, Aldea R, Bedussi B, Morris AW, Magnetic resonance imaging; MS: Medial septum; nNOS: Neuronal nitric oxide Weller RO, Carare RO (2016) Lymphatic clearance of the brain: perivas- synthase; NVC: Neurovascular coupling; P75NTR: p75 neurotrophin receptor; cular, paravascular and significance for neurodegenerative diseases. Cell ROCK: Rho- associated, coiled-coil containing protein kinase. Mol Neurobiol 36:181–194. https ://doi.org/10.1007/s1057 1-015-0273-8 8. Ball KK, Cruz NF, Mrak RE, Dienel GA (2010) Trafficking of glucose, lactate, Acknowledgements and amyloid-beta from the inferior colliculus through perivascular routes. The authors wish to thank the staff at the BRU and BRF for their technical J Cereb Blood Flow Metab 30:162–176. https ://doi.org/10.1038/jcbfm support. .2009.206 9. Bartus RT, Dean RL 3rd, Beer B, Lippa AS (1982) The cholinergic hypothesis Authors’ contributions of geriatric memory dysfunction. Science 217:408–414. https ://doi. SN planned and carried out experiments, data analysis and manuscript prepa-org/10.1126/scien ce.70460 51 ration and editing. JW carried out experiments and data analysis relating to 10. Beach TG, Potter PE, Kuo YM, Emmerling MR, Durham RA, Webster SD, MRI experiments. ROC contributed to experimental planning and manuscript Walker DG, Sue LI, Scott S, Layne KJ et al (2000) Cholinergic deafferen- editing. IAR contributed to experimental planning and data analysis. CAH con- tation of the rabbit cortex: a new animal model of Abeta deposition. tributed to experimental planning, data analysis and manuscript preparation Neurosci Lett 283:9–12 and editing. All authors read and approved the final manuscript. 11. Bedussi B, Almasian M, de Vos J, VanBavel E, Bakker EN (2018) Paravas- cular spaces at the brain surface: low resistance pathways for cerebro- Funding spinal fluid flow. J Cereb Blood Flow Metab 38:719–726. https ://doi. This work was supported by funds from Alzheimer’s Research UK.org/10.1177/02716 78X17 73798 4 12. Brown DR, Hunter R, Wyper DJ, Patterson J, Kelly RC, Montaldi D, Availability of data and materials McCullouch J (1996) Longitudinal changes in cognitive function The datasets used and/or analysed during the current study available from the and regional cerebral function in Alzheimer’s disease: a SPECT blood corresponding author on reasonable request. flow study. J Psychiatr Res 30:109–126. https ://doi.org/10.1016/0022- 3956(95)00032 -1 Ethics approval and consent to participate 13. Carare RO, Bernardes-Silva M, Newman TA, Page AM, Nicoll JAR, Perry VH, All animal work was approved by the Animal Welfare and Ethics Research Weller RO (2008) Solutes, but not cells, drain from the brain parenchyma Boards (AWERB) at the OU, University of Southampton and UCL in accord- along basement membranes of capillaries and arteries. Significance for ance with Home Office regulations and project licences (PPL 70/8507 and cerebral amyloid angiopathy and neuroimmunology. Neuropathol Appl PPL 30/3095) under the Animals (Scientific Procedures) Act 1986. Experiments Neurobiol 34:131–144 were reported according to the ARRIVE guidelines. 14. Carare RO, Hawkes CA, Jeffrey M, Kalaria RN, Weller RO (2013) Review: cer - ebral amyloid angiopathy, prion angiopathy, CADASIL and the spectrum Consent for publication of protein elimination failure angiopathies (PEFA) in neurodegenerative Not applicable. disease with a focus on therapy. Neuropathol Appl Neurobiol 39:593– 611. https ://doi.org/10.1111/nan.12042 Competing interests 15. Cauli B, Tong XK, Rancillac A, Serluca N, Lambolez B, Rossier J, Hamel E The authors declare that they have no competing interests. (2004) Cortical GABA interneurons in neurovascular coupling: relays for subcortical vasoactive pathways. J Neurosci 24:8940–8949. https ://doi. Author detailsorg/10.1523/JNEUR OSCI.3065-04.2004 School of Life, Health and Chemical Sciences, Open University, Milton 16. Caviedes A, Varas-Godoy M, Lafourcade C, Sandoval S, Bravo-Alegria 2 3 Keynes, UK. Centre for Advanced Bioimaging, UCL, London, UK. Clinical J, Kaehne T, Massmann A, Figueroa JP, Nualart F, Wyneken U (2017) and Experimental Sciences, University of Southampton, Southampton, UK. Endothelial nitric oxide synthase is present in dendritic spines of neurons Department of Biomedical and Life Sciences, Lancaster University, Lancaster, in primary cultures. Front Cell Neurosci 11:180. https ://doi.org/10.3389/ UK. fncel .2017.00180 17. Chow BW, Nunez V, Kaplan L, Granger AJ, Bistrong K, Zucker HL, Kumar Received: 6 November 2020 Accepted: 15 December 2020 P, Sabatini BL, Gu C (2020) Caveolae in CNS arterioles mediate neuro- vascular coupling. Nature 579:106–110. https ://doi.org/10.1038/s4158 6-020-2026-1 18. Dai W, Lopez OL, Carmichael OT, Becker JT, Kuller LH, Gach HM (2009) Mild cognitive impairment and alzheimer disease: patterns of altered cerebral blood flow at MR imaging. Radiology 250:856–866. https ://doi. References org/10.1148/radio l.25030 80751 1. Aldea R, Weller RO, Wilcock DM, Carare RO, Richardson G (2019) Cerebro- 19. Di Marco LY, Farkas E, Martin C, Venneri A, Frangi AF (2015) Is vasomo- vascular smooth muscle cells as the drivers of intramural periarterial tion in cerebral arteries impaired in alzheimer’s disease? J Alzheimers Dis drainage of the brain. Front Aging Neurosci 11:1. https ://doi.org/10.3389/ 46:35–53. https ://doi.org/10.3233/JAD-14297 6 fnagi .2019.00001 20. Dorheim MA, Tracey WR, Pollock JS, Grammas P (1994) Nitric oxide 2. Alsop DC, Casement M, de Bazelaire C, Fong T, Press DZ (2008) Hip- synthase activity is elevated in brain microvessels in Alzheimer’s disease. pocampal hyperperfusion in Alzheimer’s disease. Neuroimage 42:1267– Biochem Biophys Res Commun 205:659–665. https ://doi.org/10.1006/ 1274. https ://doi.org/10.1016/j.neuro image .2008.06.006 bbrc.1994.2716 3. Arbel-Ornath M, Hudry E, Eikermann-Haerter K, Hou S, Gregory JL, Zhao L, 21. Dorr A, Sahota B, Chinta LV, Brown ME, Lai AY, Ma K, Hawkes CA, McLaurin Betensky RA, Frosch MP, Greenberg SM, Bacskai BJ (2013) Interstitial fluid J, Stefanovic B (2012) Amyloid-beta-dependent compromise of micro- drainage is impaired in ischemic stroke and Alzheimer’s disease mouse vascular structure and function in a model of Alzheimer’s disease. Brain models. Acta Neuropathol 126:353–364. https ://doi.org/10.1007/s0040 135:3039–3050. https ://doi.org/10.1093/brain /aws24 3 1-013-1145-2 N izari et al. acta neuropathol commun (2021) 9:12 Page 15 of 17 22. Francis PT, Palmer AM, Snape M, Wilcock GK (1999) The cholinergic 40. Jankowsky JL, Slunt HH, Gonzales V, Savonenko AV, Wen JC, Jenkins NA, hypothesis of Alzheimer’s disease: a review of progress. J Neurol Neuro- Copeland NG, Younkin LH, Lester HA, Younkin SG et al (2005) Persistent surg Psychiatry 66:137–147 amyloidosis following suppression of Abeta production in a transgenic 23. Furchgott RF, Zawadzki JV (1980) The obligatory role of endothelial cells model of Alzheimer disease. PLoS Med 2:e355. https ://doi.org/10.1371/ in the relaxation of arterial smooth muscle by acetylcholine. Nature journ al.pmed.00203 55 288:373–376. https ://doi.org/10.1038/28837 3a0 41. Jeffrey M, Gonzalez L (2007) Classical sheep transmissible spongiform 24. Gil-Bea FJ, Gerenu G, Aisa B, Kirazov LP, Schliebs R, Ramirez MJ (2012) encephalopathies: pathogenesis, pathological phenotypes and clinical Cholinergic denervation exacerbates amyloid pathology and induces disease. Neuropathol Appl Neurobiol 33:373–394. https ://doi.org/10.111 hippocampal atrophy in Tg2576 mice. Neurobiol Dis 48:439–446. https ://1/j.1365-2990.2007.00868 .x doi.org/10.1016/j.nbd.2012.06.020 42. Jellinger KA, Attems J (2006) Prevalence and impact of cerebrovascular 25. Gratton JP, Fontana J, O’Connor DS, Garcia-Cardena G, McCabe TJ, Sessa pathology in Alzheimer’s disease and parkinsonism. Acta Neurol Scand WC (2000) Reconstitution of an endothelial nitric-oxide synthase (eNOS), 114:38–46 hsp90, and caveolin-1 complex in vitro. Evidence that hsp90 facilitates 43. Kalisch BE, Connop BP, Jhamandas K, Beninger RJ, Boegman RJ (1996) calmodulin stimulated displacement of eNOS from caveolin-1. J Biol Differential action of 7-nitro indazole on rat brain nitric oxide synthase. Chem 275:22268–22272. https ://doi.org/10.1074/jbc.M0016 44200 Neurosci Lett 219:75–78. https ://doi.org/10.1016/s0304 -3940(96)13194 -3 26. Grothe MJ, Ewers M, Krause B, Heinsen H, Teipel SJ, Alzheimer’s Disease 44. Kishi T, Hirooka Y, Masumoto A, Ito K, Kimura Y, Inokuchi K, Tagawa T, Neuroimaging I (2014) Basal forebrain atrophy and cortical amyloid Shimokawa H, Takeshita A, Sunagawa K (2005) Rho-kinase inhibitor deposition in nondemented elderly subjects. Alzheimers Dement improves increased vascular resistance and impaired vasodilation of the 10:S344–353. https ://doi.org/10.1016/j.jalz.2013.09.011 forearm in patients with heart failure. Circulation 111:2741–2747. https :// 27. Hawkes CA, Gatherer M, Sharp MM, Dorr A, Yuen HM, Kalaria R, Weller doi.org/10.1161/CIRCU LATIO NAHA.104.51024 8 RO, Carare RO (2013) Regional differences in the morphological and 45. Kureishi Y, Kobayashi S, Amano M, Kimura K, Kanaide H, Nakano T, Kaibu- functional effects of aging on cerebral basement membranes and chi K, Ito M (1997) Rho-associated kinase directly induces smooth muscle perivascular drainage of amyloid-beta from the mouse brain. Aging Cell contraction through myosin light chain phosphorylation. J Biol Chem 12:224–236. https ://doi.org/10.1111/acel.12045 272:12257–12260. https ://doi.org/10.1074/jbc.272.19.12257 28. Hawkes CA, Hartig W, Kacza J, Schliebs R, Weller RO, Nicoll JA, Carare RO 46. Kuznetsova E, Schliebs R (2013) beta-Amyloid, cholinergic transmis- (2011) Perivascular drainage of solutes is impaired in the ageing mouse sion, and cerebrovascular system: a developmental study in a mouse brain and in the presence of cerebral amyloid angiopathy. Acta Neuro- model of Alzheimer’s disease. Curr Pharm Des 19:6749–6765. https ://doi. pathol 121:431–443. https ://doi.org/10.1007/s0040 1-011-0801-7org/10.2174/13816 12811 31999 90711 29. Hawkes CA, Sullivan PM, Hands S, Weller RO, Nicoll JA, Carare RO (2012) 47. Lapchak PA, Araujo DM, Carswell S, Hefti F (1993) Distribution of [125I] Disruption of arterial perivascular drainage of amyloid-beta from the nerve growth factor in the rat brain following a single intraventricular brains of mice expressing the human APOE epsilon4 allele. PLoS ONE injection: correlation with the topographical distribution of trkA mes- 7:e41636. https ://doi.org/10.1371/journ al.pone.00416 36 senger RNA-expressing cells. Neuroscience 54:445–460. https ://doi. 30. Hernandez-Guillamon M, Delgado P, Penalba A, Rodriguez-Luna D, org/10.1016/0306-4522(93)90265 -h Molina CA, Rovira A, Alvarez-Sabin J, Boada M, Montaner J (2012) 48. Laursen B, Mork A, Plath N, Kristiansen U, Bastlund JF (2013) Cholinergic Plasma beta-amyloid levels in cerebral amyloid angiopathy-associated degeneration is associated with increased plaque deposition and cogni- hemorrhagic stroke. Neurodegener Dis 10:320–323. https ://doi. tive impairment in APPswe/PS1dE9 mice. Behav Brain Res 240:146–152. org/10.1159/00033 3811https ://doi.org/10.1016/j.bbr.2012.11.012 31. Hladky SB, Barrand MA (2014) Mechanisms of fluid movement into, 49. Lecrux C, Hamel E (2016) Neuronal networks and mediators of cortical through and out of the brain: evaluation of the evidence. Fluids Barriers neurovascular coupling responses in normal and altered brain states. CNS 11:26. https ://doi.org/10.1186/2045-8118-11-26 Philos Trans R Soc Lond B Biol Sci. https ://doi.org/10.1098/rstb.2015.0350 32. Hotta H (2016) Neurogenic control of parenchymal arterioles in the 50. Lecrux C, Sandoe CH, Neupane S, Kropf P, Toussay X, Tong XK, Lacalle- cerebral cortex. Prog Brain Res 225:3–39. https ://doi.org/10.1016/ Aurioles M, Shmuel A, Hamel E (2017) Impact of altered cholinergic tones bs.pbr.2016.03.001 on the neurovascular coupling response to whisker stimulation. J Neuro- 33. Hotta H, Uchida S, Kagitani F, Maruyama N (2011) Control of cerebral sci 37:1518–1531. https ://doi.org/10.1523/JNEUR OSCI.1784-16.2016 cortical blood flow by stimulation of basal forebrain cholinergic 51. Liu P, Zhang H, Tang Y, Sheng C, Liu J, Zeng Y (2014) Influence of Rho areas in mice. J Physiol Sci 61:201–209. https ://doi.org/10.1007/s1257 kinase inhibitor Fasudil on late endothelial progenitor cells in peripheral 6-011-0139-x blood of COPD patients with pulmonary artery hypertension. Bosn J 34. Hu WT, Wang Z, Lee VM, Trojanowski JQ, Detre JA, Grossman M (2010) Dis- Basic Med Sci 14:40–44. https ://doi.org/10.17305 /bjbms .2014.2295 tinct cerebral perfusion patterns in FTLD and AD. Neurology 75:881–888. 52. Lourenco CF, Santos RM, Barbosa RM, Cadenas E, Radi R, Laranjinha J https ://doi.org/10.1212/WNL.0b013 e3181 f11e3 5 (2014) Neurovascular coupling in hippocampus is mediated via diffusion 35. Iadecola C, Zhang F (1996) Permissive and obligatory roles of NO in cer- by neuronal-derived nitric oxide. Free Radic Biol Med 73:421–429. https :// ebrovascular responses to hypercapnia and acetylcholine. Am J Physiol doi.org/10.1016/j.freer adbio med.2014.05.021 271:R990–1001. https ://doi.org/10.1152/ajpre gu.1996.271.4.R990 53. Lovick TA, Brown LA, Key BJ (1999) Neurovascular relationships in hip- 36. Iliff JJ, Wang M, Liao Y, Plogg BA, Peng W, Gundersen GA, Benveniste H, pocampal slices: physiological and anatomical studies of mechanisms Vates GE, Deane R, Goldman SA et al (2012) A paravascular pathway underlying flow-metabolism coupling in intraparenchymal microvessels. facilitates CSF flow through the brain parenchyma and the clearance of Neuroscience 92:47–60. https ://doi.org/10.1016/s0306 -4522(98)00737 -4 interstitial solutes, including amyloid beta. Sci Transl Med 4:147ra111. 54. Miners JS, Baig S, Palmer J, Palmer LE, Kehoe PG, Love S (2008) Abeta- https ://doi.org/10.1126/scitr anslm ed.30037 48 degrading enzymes in Alzheimer’s disease. Brain Pathol 18:240–252 37. Iliff JJ, Wang M, Zeppenfeld DM, Venkataraman A, Plog BA, Liao Y, Deane 55. Montagne A, Zhao Z, Zlokovic BV (2017) Alzheimer’s disease: a matter of R, Nedergaard M (2013) Cerebral arterial pulsation drives paravas- blood-brain barrier dysfunction? J Exp Med 214:3151–3169. https ://doi. cular CSF-interstitial fluid exchange in the murine brain. J Neurosci org/10.1084/jem.20171 406 33:18190–18199 56. Morris AW, Carare RO, Schreiber S, Hawkes CA (2014) The cerebrovascular 38. Iturria-Medina Y, Sotero RC, Toussaint PJ, Mateos-Perez JM, Evans AC, basement membrane: role in the clearance of beta-amyloid and cerebral Alzheimer’s Disease Neuroimaging I (2016) Early role of vascular dysregu- amyloid angiopathy. Front Aging Neurosci 6:251. https ://doi.org/10.3389/ lation on late-onset Alzheimer’s disease based on multifactorial data-fnagi .2014.00251 driven analysis. Nat Commun 7:11934. https ://doi.org/10.1038/ncomm 57. Mukai Y, Shimokawa H, Matoba T, Kandabashi T, Satoh S, Hiroki J, Kaibuchi s1193 4 K, Takeshita A (2001) Involvement of Rho-kinase in hypertensive vascular 39. Jack CR Jr, Knopman DS, Jagust WJ, Shaw LM, Aisen PS, Weiner MW, disease: a novel therapeutic target in hypertension. FASEB J 15:1062– Petersen RC, Trojanowski JQ (2010) Hypothetical model of dynamic 1064. https ://doi.org/10.1096/fj.00-0735 fje biomarkers of the Alzheimer’s pathological cascade. Lancet Neurol 58. Nag N, Baxter MG, Berger-Sweeney JE (2009) Efficacy of a murine-p75- 9:119–128. https ://doi.org/10.1016/S1474 -4422(09)70299 -6 saporin immunotoxin for selective lesions of basal forebrain cholinergic Nizari et al. acta neuropathol commun (2021) 9:12 Page 16 of 17 neurons in mice. Neurosci Lett 452:247–251. https ://doi.org/10.1016/j. 76. Shi J, Wei L (2013) Rho kinases in cardiovascular physiology and patho- neule t.2009.01.006 physiology: the effect of fasudil. J Cardiovasc Pharmacol 62:341–354. 59. Nagata K, Kondoh Y, Satoh Y, Watahiki Y, Yokoyama E, Yuya H, Hirata Y, https ://doi.org/10.1097/FJC.0b013 e3182 a3718 f Shishido F, Hatazawa J, Kanno I et al (1993) Eec ff ts of fasudil hydrochlo - 77. Shin HK, Jones PB, Garcia-Alloza M, Borrelli L, Greenberg SM, Bacskai BJ, ride on cerebral blood flow in patients with chronic cerebral infarction. Frosch MP, Hyman BT, Moskowitz MA, Ayata C (2007) Age-dependent Clin Neuropharmacol 16:501–510. https ://doi.org/10.1097/00002 826- cerebrovascular dysfunction in a transgenic mouse model of cerebral 19931 2000-00003 amyloid angiopathy. Brain 130:2310–2319 60. Nicolakakis N, Hamel E (2011) Neurovascular function in Alzheimer’s 78. Shin HK, Salomone S, Potts EM, Lee SW, Millican E, Noma K, Huang PL, disease patients and experimental models. J Cereb Blood Flow Metab Boas DA, Liao JK, Moskowitz MA et al (2007) Rho-kinase inhibition acutely 31:1354–1370. https ://doi.org/10.1038/jcbfm .2011.43 augments blood flow in focal cerebral ischemia via endothelial mecha- 61. Nizari S, Carare RO, Romero IA, Hawkes CA (2019) 3D reconstruction of nisms. J Cereb Blood Flow Metab 27:998–1009. https ://doi.org/10.1038/ the neurovascular unit reveals differential loss of cholinergic innervation sj.jcbfm .96004 06 in the cortex and hippocampus of the adult mouse brain. Front Aging 79. Simic G, Babic M, Borovecki F, Hof PR (2014) Early failure of the default- Neurosci 11:172. https ://doi.org/10.3389/fnagi .2019.00172 mode network and the pathogenesis of Alzheimer’s disease. CNS Neuro- 62. Oomen CA, Farkas E, Roman V, van der Beek EM, Luiten PG, Meerlo P sci Ther 20:692–698. https ://doi.org/10.1111/cns.12260 (2009) Resveratrol preserves cerebrovascular density and cognitive func- 80. Suzuki N, Iwatsubo T, Odaka A, Ishibashi Y, Kitada C, Ihara Y (1994) High tion in aging mice. Front Aging Neurosci 1:4. https ://doi.org/10.3389/ tissue content of soluble beta 1-40 is linked to cerebral amyloid angiopa- neuro .24.004.2009 thy. Am J Pathol 145:452–460 63. Paris D, Patel N, DelleDonne A, Quadros A, Smeed R, Mullan M (2004) 81. Tan XL, Xue YQ, Ma T, Wang X, Li JJ, Lan L, Malik KU, McDonald MP, Dopico Impaired angiogenesis in a transgenic mouse model of cerebral amyloi- AM, Liao FF (2015) Partial eNOS deficiency causes spontaneous throm- dosis. Neurosci Lett 366:80–85 botic cerebral infarction, amyloid angiopathy and cognitive impairment. 64. Partyka PP, Godsey GA, Galie JR, Kosciuk MC, Acharya NK, Nagele RG, Mol Neurodegener 10:24. https ://doi.org/10.1186/s1302 4-015-0020-0 Galie PA (2017) Mechanical stress regulates transport in a compliant 3D 82. Teipel S, Heinsen H, Amaro E Jr, Grinberg LT, Krause B, Grothe M, model of the blood-brain barrier. Biomaterials 115:30–39. https ://doi. Alzheimer’s Disease Neuroimaging I (2014) Cholinergic basal forebrain org/10.1016/j.bioma teria ls.2016.11.012 atrophy predicts amyloid burden in Alzheimer’s disease. Neurobiol Aging 65. Paxinos G, Franklin K (2019) Paxinos and Franklin’s the mouse brain in 35:482–491. https ://doi.org/10.1016/j.neuro biola ging.2013.09.029 stereotaxic coordinates, 5th edn. Elsevier, Amsterdam, p 375 83. Thal DR, Griffin WS, de Vos RA, Ghebremedhin E (2008) Cerebral amyloid 66. Perry EK, Morris CM, Court JA, Cheng A, Fairbairn AF, McKeith IG, Irving angiopathy and its relationship to Alzheimer’s disease. Acta Neuropathol D, Brown A, Perry RH (1995) Alteration in nicotine binding sites in Par- 115:599–609 kinson’s disease, Lewy body dementia and Alzheimer’s disease: possible 84. Tian J, Shi J, Bailey K, Mann DM (2003) Negative association between index of early neuropathology. Neuroscience 64:385–395. https ://doi. amyloid plaques and cerebral amyloid angiopathy in Alzheimer’s disease. org/10.1016/0306-4522(94)00410 -7 Neurosci Lett 352:137–140 67. Ramos-Rodriguez JJ, Pacheco-Herrero M, Thyssen D, Murillo-Carretero 85. Tikka S, Mykkanen K, Ruchoux MM, Bergholm R, Junna M, Poyhonen M, MI, Berrocoso E, Spires-Jones TL, Bacskai BJ, Garcia-Alloza M (2013) Rapid Yki-Jarvinen H, Joutel A, Viitanen M, Baumann M et al (2009) Congruence beta-amyloid deposition and cognitive impairment after cholinergic den- between NOTCH3 mutations and GOM in 131 CADASIL patients. Brain ervation in APP/PS1 mice. J Neuropathol Exp Neurol 72:272–285. https :// 132:933–939. https ://doi.org/10.1093/brain /awn36 4 doi.org/10.1097/NEN.0b013 e3182 88a8d d 86. Van Beek AH, Claassen JA (2011) The cerebrovascular role of the cholin- 68. Rikitake Y, Kim HH, Huang Z, Seto M, Yano K, Asano T, Moskowitz MA, ergic neural system in Alzheimer’s disease. Behav Brain Res 221:537–542. Liao JK (2005) Inhibition of Rho kinase (ROCK) leads to increased cerebral https ://doi.org/10.1016/j.bbr.2009.12.047 blood flow and stroke protection. Stroke 36:2251–2257. https ://doi. 87. van Opstal AM, van Rooden S, van Harten T, Ghariq E, Labadie G, Fotiadis org/10.1161/01.STR.00001 81077 .84981 .11 P, Gurol ME, Terwindt GM, Wermer MJH, van Buchem MA et al (2017) Cer- 69. Rivera-Rivera LA, Turski P, Johnson KM, Hoffman C, Berman SE, Kilgas P, ebrovascular function in presymptomatic and symptomatic individuals Rowley HA, Carlsson CM, Johnson SC, Wieben O (2016) 4D flow MRI for with hereditary cerebral amyloid angiopathy: a case-control study. Lancet intracranial hemodynamics assessment in Alzheimer’s disease. J Cereb Neurol 16:115–122. https ://doi.org/10.1016/S1474 -4422(16)30346 -5 Blood Flow Metab 36:1718–1730. https ://doi.org/10.1177/02716 78X15 88. van Veluw SJ, Hou SS, Calvo-Rodriguez M, Arbel-Ornath M, Snyder AC, 61717 1 Frosch MP, Greenberg SM, Bacskai BJ (2020) Vasomotion as a driving force 70. Roher AE, Kuo YM, Potter PE, Emmerling MR, Durham RA, Walker DG, Sue for paravascular clearance in the awake mouse brain. Neuron 105(549– LI, Honer WG, Beach TG (2000) Cortical cholinergic denervation elicits 561):e545. https ://doi.org/10.1016/j.neuro n.2019.10.033 vascular A beta deposition. Ann N Y Acad Sci 903:366–373 89. Vaucher E, Linville D, Hamel E (1997) Cholinergic basal forebrain projec- 71. Rosengarten B, Paulsen S, Burr O, Kaps M (2009) Neurovascular coupling tions to nitric oxide synthase-containing neurons in the rat cerebral in Alzheimer patients: effect of acetylcholine-esterase inhibitors. cortex. Neuroscience 79:827–836. https ://doi.org/10.1016/s0306 Neurobiol Aging 30:1918–1923. https ://doi.org/10.1016/j.neuro biola -4522(97)00033 -x ging.2008.02.017 90. Vinters HV (1987) Cerebral amyloid angiopathy. A critical review. Stroke 72. Sato A, Sato Y, Uchida S (2004) Activation of the intracerebral cholinergic 18:311–324 nerve fibers originating in the basal forebrain increases regional cerebral 91. Wells JA, Holmes HE, O’Callaghan JM, Colgan N, Ismail O, Fisher EM, blood flow in the rat’s cortex and hippocampus. Neurosci Lett 361:90–93. Siow B, Murray TK, Schwarz AJ, O’Neill MJ et al (2015) Increased cerebral https ://doi.org/10.1016/j.neule t.2004.01.004 vascular reactivity in the tau expressing rTg4510 mouse: evidence against 73. Schmitz TW, Nathan Spreng R, Alzheimer’s Disease Neuroimaging I (2016) the role of tau pathology to impair vascular health in Alzheimer’s disease. Basal forebrain degeneration precedes and predicts the cortical spread J Cereb Blood Flow Metab 35:359–362. https ://doi.org/10.1038/jcbfm of Alzheimer’s pathology. Nat Commun 7:13249. https ://doi.org/10.1038/ .2014.224 ncomm s1324 9 92. Wiencken AE, Casagrande VA (1999) Endothelial nitric oxide synthetase 74. Sehba FA, Schwartz AY, Chereshnev I, Bederson JB (2000) Acute decrease (eNOS) in astrocytes: another source of nitric oxide in neocortex. Glia in cerebral nitric oxide levels after subarachnoid hemorrhage. J Cereb 26:280–290 Blood Flow Metab 20:604–611. https ://doi.org/10.1097/00004 647-20000 93. Wilcock DM, Rojiani A, Rosenthal A, Subbarao S, Freeman MJ, Gordon MN, 3000-00018 Morgan D (2004) Passive immunotherapy against Abeta in aged APP- 75. Shafer OT, Chen A, Kumar SM, Muller KJ, Sahley CL (1998) Injury-induced transgenic mice reverses cognitive deficits and depletes parenchymal expression of endothelial nitric oxide synthase by glial and microglial amyloid deposits in spite of increased vascular amyloid and microhemor- cells in the leech central nervous system within minutes after injury. Proc rhage. J Neuroinflamm 1:24 Biol Sci 265:2171–2175. https ://doi.org/10.1098/rspb.1998.0555 N izari et al. acta neuropathol commun (2021) 9:12 Page 17 of 17 94. Williams RJ, Goodyear BG, Peca S, McCreary CR, Frayne R, Smith EE, receptor substrate/Akt signaling pathway in aorta of obese rats. Diabetes Pike GB (2017) Identification of neurovascular changes associated with 56:1014–1024. https ://doi.org/10.2337/db05-1147 cerebral amyloid angiopathy from subject-specific hemodynamic 97. Zhang F, Xu S, Iadecola C (1995) Role of nitric oxide and acetylcholine in response functions. J Cereb Blood Flow Metab 37:3433–3445. https ://doi. neocortical hyperemia elicited by basal forebrain stimulation: evidence org/10.1177/02716 78X17 69105 6 for an involvement of endothelial nitric oxide. Neuroscience 69:1195– 95. Yaar M, Zhai S, Pilch PF, Doyle SM, Eisenhauer PB, Fine RE, Gilchrest BA 1204. https ://doi.org/10.1016/0306-4522(95)00302 -y (1997) Binding of beta-amyloid to the p75 neurotrophin receptor induces apoptosis. A possible mechanism for Alzheimer’s disease. J Clin Invest Publisher’s Note 100:2333–2340. https ://doi.org/10.1172/JCI11 9772 Springer Nature remains neutral with regard to jurisdictional claims in pub- 96. Zecchin HG, Priviero FB, Souza CT, Zecchin KG, Prada PO, Carvalheira JB, lished maps and institutional affiliations. Velloso LA, Antunes E, Saad MJ (2007) Defective insulin and acetylcho- line induction of endothelial cell-nitric oxide synthase through insulin Re Read ady y to to submit y submit your our re researc search h ? Choose BMC and benefit fr ? Choose BMC and benefit from om: : fast, convenient online submission thorough peer review by experienced researchers in your field rapid publication on acceptance support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: Jan 7, 2021

There are no references for this article.