Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

Sonia Mazzitelli; Fabia Filipello; Marco Rasile; Eliana Lauranzano; Chiara Starvaggi-Cucuzza; Matteo Tamborini; Davide Pozzi; Isabella Barajon; Toni Giorgino; Antonino Natalello; Michela Matteoli

doi:10.1186/s40478-016-0381-9

Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

Mazzitelli, Sonia; Filipello, Fabia; Rasile, Marco; Lauranzano, Eliana; Starvaggi-Cucuzza, Chiara; Tamborini, Matteo; Pozzi, Davide; Barajon, Isabella; Giorgino, Toni; Natalello, Antonino; Matteoli, Michela 2016-10-10 00:00:00 Substantial data indicate that amyloid-β (Aβ), the major component of senile plaques, plays a central role in Alzheimer’s Disease and indeed the assembly of naturally occurring amyloid peptides into cytotoxic aggregates is linked to the disease pathogenesis. Although Aβ42 is a highly aggregating form of Aβ, the co-occurrence of shorter Aβ peptides might affect the aggregation potential of the Aβ pool. In this study we aimed to assess whether the structural behavior of human Aβ42 peptide inside the brain is influenced by the concomitant presence of N-terminal fragments produced by the proteolytic activity of glial cells. We show that the occurrence of the human C-terminal truncated 1–24 Aβ fragment impairs Aβ42 clearance through blood brain barrier and promotes the formation of Aβ42 aggregates even in the healthy brain. By showing that Aβ1-24 has seeding properties for aggregate formation in intracranially injected wild type mice, our study provide the proof-of-concept that peptides produced upon Aβ42 cleavage by activated glial cells may cause phenotypic defects even in the absence of genetic mutations associated with Alzheimer’s Disease, possibly contributing to the development of the sporadic form of the pathology. Keywords: Amyloid-β,Alzheimer’s disease, Microglia, Proteolytic activity, Aβ24 Introduction shown to represent an important factor in initializing Aβ Alzheimer’s disease (AD) is a protein misfolding pathology, fibrillogenesis and toxicity [7], indicating that the presence caused by accumulation of abnormally folded Aβ and tau of different Aβ forms may affect the development of AD polypeptides, which form amyloid plaques and neurofibril- in vivo. lary tangles in the brain of affected individuals. Aβ Consistently, while small amounts of Aβ-containing aggregates have been linked with learning and memory brain extracts, deriving from either AD patient or AD deficits in both human and mouse models of the disease, transgenic mouse, induce β-amyloidosis and glial activa- making Aβ deposition a target for prevention and treat- tion once intracranially injected in pre-depositing AD ment [1–3]. In the last years, a lot of effort has been transgenic mice [8–11], the chronic infusion of soluble, focused on the identification of the processes leading to Aβ synthetic Aβ42 peptides into wild type (wt) rodent aggregation. Evidence have indicated that, although Aβ42 is brains does not result in amyloid deposition [9]. The a highly aggregating form of Aβ [4, 5], the co-occurrence of finding that Aβ42 alone fails to show seeding properties Aβ peptides with different length can affect the neurotoxic in the healthy brain and does not trigger pathogenetic and aggregation potential of the Aβ pool (reviewed in [6]). pathways indicates the occurrence of efficient clearance As an example, changes in the ratio of Aβ40/42 has been mechanisms and suggests that brain-specific cofactors, specifically present in pathological conditions, are needed for effective seeding [9]. * Correspondence: m.matteoli@in.cnr.it; Although aggregation properties of full length Aβ42 michela.matteoli@humanitasresearch.it have been deeply explored both in vitro and in vivo, Equal contributors IRCCS Humanitas, via Manzoni 56, 20089 Rozzano, Italy much less is known about the in vivo aggregating prop- IN-CNR, via Vanvitelli 32, 20129 Milano, Italy erties of shorter Aβ fragments. This aspect may be Full list of author information is available at the end of the article © 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 2 of 19 particularly relevant, considering that N-terminal fragments detectable by the amyloidogenic dye Congo red both 2.5 of different length are largely produced by the proteo- and 6 months after injection (Fig. 1c and f, g). Consist- lytic activity of glial cells during the development of ently, the highest number of Congo red positive aggre- AD [12]. Indeed, several proteases including neprilysin gates was detected in the brains of mice injected with [13], insulin-degrading enzyme [14], endothelin-converting H-Aβ42/H-Aβ24 mix (Fig. 1f and g). enzyme [15], angiotensin-converting enzyme [16] and As a further confirmation for the formation of amyloi- matrix metalloproteinase-9 (MMP9) [17–20] have been dogenic aggregates, brain sections from mice i.c. injected shown to degrade soluble Aβ in vitro, acting at specific and incubated for 2.5 months were stained with thiofla- cleavage sites and generating characteristic Aβ fragments. vin T (ThT), a benzothiazole dye that exhibits enhanced These proteolytic activities, therefore, are critical in deter- fluorescence upon binding to amyloid fibrils [27, 28]. mining the quantitative and qualitative pattern of cerebral ThT-positive aggregates were detected in wt mice brain Aβ levels ([21]; reviewed in [22]). MMP9 in particular, injected with H-Aβ24 or with H-Aβ42/H-Aβ24 mix which generates different C-terminal truncated Aβ frag- (Fig. 2a and c); both ThT- (Fig. 2a) and 6E10- (Fig. 2b) ments, including 1–16, 1–20, 1–23, 1–30, 1–33 and 1–34 positive aggregates were surrounded by microglia, as re- [19, 23], is thought to play relevant roles in different patho- vealed by Iba1 staining (Fig. 2a and b, red staining). logical contexts, as suggested by the observation that its ex- Again, no ThT positive aggregates were detected in pression can be stimulated by diverse insults, including Aβ vehicle-injected wt mice, whereas only ThT-positive itself [24], and is up-regulated in glia cells adjacent to amyl- blood vessels, which are not surrounded by microglia, oid deposits [25]. were visible in wt mice injected with H-Aβ42 (Fig. 2a, Since Aβ N-terminal fragments of varying lengths are arrows). The recruitment of microglia around aggregates expected to exhibit different physico-chemical properties was reminiscent of the plaques present in APP/PS1 trans- which may result in different aggregation behaviors, as genic mice brain (Fig. 2b, right). Consistent with data also indicated by modeling of aggregation determinants reported in AD patients and mice models [29–33], a with bioinformatics methods [26], we aimed to investigate significant increase in the levels of tumor necrosis factor- whether dynamics of interaction and structural behavior alpha (TNF-alpha) was detected by ELISA in the serum of human Aβ42 peptide inside the brain are influenced by and brains of mice i.c. injected with H-Aβ24 and of H- the concomitant presence of C-terminal truncated frag- Aβ42/H-Aβ24 mix, and much less with H-Aβ42 alone ments. We took advantage of the commercially available (Fig. 2d and e). These data indicate that H-Aβ24 and, even synthetic human Aβ1-24 peptide (referred to as Aβ24), a more prominently, H-Aβ42/H-Aβ24 mix induce an C-terminal truncated Aβ fragment overlapping with inflammatory reaction in the brain of injected mice. MMP9 cleavage products (residues 1–20 and 1–23, [20]) Interestingly, 6 months after i.c. injection, few scattered and coincides with a turn region between β-sheets in 6E10-positive spots also start to become detectable in the recently-resolved fibrillar structures. Our results indicate non-injected side (Fig. 3a and b), suggesting a possible that the presence of Aβ24 in intracranially injected wild spreading of misfolded/aggregating Aβ. Also, amyloid ag- type mice impairs Aβ42 clearance and promotes forma- gregates were detectable at the hippocampal level in the tion of Aβ42 aggregates even in the healthy brain. injected hemisphere (Fig. 3c). Results Behavioral deficits in wt mice injected with Aβ24 peptide Synthetic Aβ24 fragments promote aggregates formation To investigate whether the formation of Aβ aggregates in wt mice brain and the increase in TNF-alpha were accompanied by the Three month-old wt mice were intracranially (i.c.) occurrence of cognitive defects [34–36], mice were ana- injected with either the oligomeric form of the single H- lysed using 3 different behavioral tests: the open field, a Aβ42 peptide or with an equimolar mixture of oligo- recognized paradigm for assessing motor activity and meric H-Aβ42 and H-Aβ24 peptides, and the brains anxiety-like behaviors in response to a novel environ- were examined after 2.5 or 6 months (see cartoon, ment [37–39]; the sociality task, which also unveils anx- Fig. 1h). As previously described [9], H-Aβ42 injected in ious and aggressive behaviors; and the novel object the brain of wt mice and examined 6 months later did recognition (NOR), that, monitoring the time spent by not cause Aβ deposition. H-Aβ24, and even more po- mice to explore a novel object, enables the assessment tently H-Aβ42/H-Aβ24 mix, induced the formation of of possible declines in learning and memory. Consist- aggregates, detected with the Aβ N-terminal specific ently with literature data [40], 6 months old APP/PS1 antibody 6E10, followed by HRP (Fig. 1a and e). 6E10- mice displayed significant hyperactivity and anxiety com- positive aggregates were already detectable 2.5 months pared to the wt littermates, as shown by both the open after injection of H-Aβ24 or H-Aβ42/H-Aβ24 mix in the field and sociality tasks (Fig. 4a–f). Interestingly, wt mice mice brains (Fig. 1b and d). Aggregates were also injected with the mix of the two peptides displayed Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 3 of 19 Fig. 1 (See legend on next page.) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 4 of 19 (See figure on previous page.) Fig. 1 Intracranial injection of H-Aβ24 and H-Aβ42/H-Aβ24 mix in wt mice induces amyloid aggregate deposition. a 6E10 DAB staining of hippocampal brain sections 6 months after vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix injection. Sagittal plane sections are shown on the left; middle and right panels show enlargements of the dorsal cerebral cortex. 6E10-immunopositive aggregates are visible on a cresyl violet-luxol fast blue counterstaining. Scale bars: left 1 mm, middle 100 μm, right 50 μm. b 6E10 DAB staining of brain sections 2.5 months after vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix injection. Scale bar: 50 μm. c Congo red staining of brain sections 2.5 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. On the right, representative images of aggregates detected 6 months after H-Aβ42/H-Aβ24 mix injection are shown. Top: bright field images; bottom: epifluorescence images -FITC filter. Scale bar: 50 μm. d and e Quantification of 6E10 DAB positive plaques 2.5 or 6 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H- Aβ42/H-Aβ24 mix. 6 brain sections were analyzed for each mice (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (****P < 0.0001; ***P < 0.001). f and g Quantification of Congo red-positive plaques area (μm ) per brain section at 2.5 or 6 month after injection. 10 sections of 50 μm thickness per slice were analyzed for each mouse brain (N = 6 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01). h Cartoon depicting the experimental scheme Fig. 2 Intracranial injection of H-Aβ24 and H-Aβ42/H-Aβ24 mix in wt mice induces microglia recruitment and TNF-alpha production. a Immunofluorescence staining of ThT-positive aggregates (green) surrounded by Iba1 positive microglia cells (red) in wt mice brain slice 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Arrows in H-Aβ42 panel indicate Aβ accumulation in vessels. Scale bar: 10 μm. b IF staining of 6E10 positive aggregates (blue) surrounded by Iba1 positive microglia cells (red) in wt mice brains 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. IF staining of 6E10 positive plaques and surrounding microglia are shown in a brain section of 8 months old APP/PS1 transgenic mouse for comparison. Scale bar: 10 μm. c Quantification of ThT-positive plaques in sections of wt mice brains 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Aβ detected in correspondence of blood vessels (H-Aβ42 left panel, arrows) was excluded from the analysis. 6 sections were analyzed for each mouse brain (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (*** P < 0.001). d and e TNF-alpha levels measured by ELISA in brain homogenates d and serum e of wt mice 2.5 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (*** P <0.001; ** P <0. 01) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 5 of 19 Fig. 3 Intracranial injection of H-Aβ42/H-Aβ24 mix in wt mice induces spreading of misfolded Aβ. a 6E10 DAB staining of hippocampal brain sections 6 months after H-Aβ42/H-Aβ24 mix injection. Sagittal plane sections are shown on the left; middle and right panels show enlargements of cerebral cortex; the injected hemisphere is referred as ipsilateral (top panel) and the non-injected one as contralateral (bottom panel). 6E10- immunopositive aggregates are visible on a cresyl violet-luxol fast blue counterstaining. Scale bars: left 1 mm, middle 100 μm, right 50 μm. b Quantification of 6E10 DAB positive plaques in the contralateral hemisphere 6 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/ H-Aβ24 mix. 6 brain sections were analyzed for each mouse (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**** P < 0.0001, *** P < 0.001). c Representative images of 6E10 DAB positive aggregates at the hippocampal level 6 months after Aβ injection. Scale bar: left and middle 100 μm, right 50 μm abnormal behavioral phenotypes very similarly to age occurrence of cognitive defects comparable to APP/PS1 matched APP/PS1 mice, both in terms of enhanced mice. motor activity, as indicated by the total distance traveled (Fig. 4a), and in terms of increased anxiety levels, as re- Predicted physico-chemical properties of Aβ24 aggregation vealed by the longer time spent in the periphery (Fig. 4b) and cross-aggregation and the shorter time spent in the center (Fig. 4c) of the We next aimed at assessing the molecular basis of aggre- arena. This result was also confirmed by the sociality gate formation in the brains injected with H-Aβ24 or task in which APP/PS1 and wt Aβ-injected mice had a with the H-Aβ42/H-Aβ24 mix. In order to understand higher number of contact and spent less time with the whether H-Aβ24 is endowed with different propensity to second animal (Fig. 4d–f). In the NOR test, APP/PS1 form aggregates relative to H-Aβ42, we modeled the de- mice as well as wt mice injected with the mix of the two terminants of aggregation with bioinformatics methods, peptides spent less time than wt littermates exploring using different algorithms which predict the aggregation the novel object at the 1 h delay test and more time ex- propensity on the basis of sequences [26]. First, the ploring the novel object at the 24 h delay test, indicating AMYLPRED2 consensus analysis identified two distinct a significant defect in learning and memory (Fig. 4g and amyloidogenic regions in the full Aβ42 peptide, respect- h). Although not displaying brain aggregates, some be- ively encompassing residues 15–22 and 29–42. Of the havioral alterations were detected in mice intracranially two regions, the first was predicted in Aβ24 as well injected with oligomeric H-Aβ42, consistent with litera- (Table 1). This is an indication that Aβ24 maintains ture data [41, 42]. Conversely, no differences were fibrillogenic potential in isolation [43], although the noticed between the not injected and vehicle injected wt resulting fibrillar structure will likely be different due to mice, showing that the injection per se was not respon- the lack of second aggregation-prone region. sible for the observed behavior defects. These data Second, the PASTA 2.0 algorithm by Walsh et al. [44] indicate that injection of H-Aβ24 and, even more prom- predicted the strongest self-aggregating segment in inently, H-Aβ42/H-Aβ24 mix into wt brain results in the Aβ42 to be residues 31–41 (at about−10.6 kcal/mol, Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 6 of 19 Fig. 4 Intracranial injection of H-Aβ induces behavioral defects in wt mice. a–c Mice hyperlocomotion, assessed by Open Field test. Left panel a total distance (cm) travelled by mice in 30 min using 5 min time bin representation. b, c panels: total distance travelled in 30 min in the periphery or in the center of the arena using 5 min time bin representation. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test. Asterisks are referred to the vehicle vs H-Aβ42/H-Aβ24 mix condition. No significant differences are present between APP/PS1 tested mice and the H- Aβ42/H-Aβ24 mix condition. (****P < 0.0001; ***P < 0.001; **P <0.01). d–f Social interaction, assessed using the social free test. Graphic representation of the number of contacts/min between the two mice in the same arena. H-Aβ42 d,H-Aβ24 e or H-Aβ42/H-Aβ24 mix f are compared to vehicle injected mice and APP/PS1 mice. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01; *P < 0.05). g and h Assessment of learning and memory by NOR. g Number of contacts with the new and old object either after 1 h and 24 h of delay. h Number of contacts with the new object both after 1 h and 24 h of delay. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test. Asterisks are referred to the control wt vs APP/PS1 mice when specified or to the vehicle vs other conditions (****P < 0.0001; **P <0.01; *P < 0.05). A total of 33 mice (6 mice foreachexperimentalgroup plus 6APP/PS1and 3 wt mice6monthsold) wereanalysed (Fig. 5a), thus estimating Aβ24 to be both a weaker self- Table 1 Amyloigenic regions in Aβ24 and Aβ42 peptides predicted binder (by approximately 5 kcal/mol) and to have a by the 11 indicated methods, and AMYLPRED2-derived consensus comparatively more marked, although still not domin- score ant, tendency towards the self-assembly in the antiparal- Method Aβ24 Aβ42 lel arrangement (Fig. 5b). It is worth noting that the AGGRESCAN 17–22 17–22, 30–42 algorithm also predicts cross-aggregation propensities AmyloidMutants 4–12, 15–23 14–22, 34–42 between segments in the 1–24 and 25–42 ranges, the strongest pair being at−4.84 kcal/mol, which is in the Amyloidogenic Pattern 16–21 16–21 same order of magnitude of the self-aggregation propen- Average Packing Density 16–21 16–21, 32–37 sities in the 1–24 region (Fig. 5c). Beta-strand contiguity 15–20 15–20, 29–41 Hexapeptide Conf. Energy 16–22 16–22, 29–42 Aβ24 displays a low fibrillar aggregation propensity and NetCSSP 1–23 1–23, 28–37 supports the formation of intermolecular β-sheets Pafig 7–24 7–42 The in vitro aggregation of Aβ samples was monitored by ThT assay [45] (Fig. 6a–c). ThT fluorescence emission was SecStr 15–20 15–20 recorded at different incubation times at 37 °C in physio- TANGO 17–21 17–21, 29–41 logical buffer. Under our experimental conditions, and in WALTZ 15–24 15–23, 28–42 agreement with previously reported data [46], oligomeric AMYLPRED2 (CONSENSUS5) 15–22 15–22, 29–42 H-Aβ42 rapidly formed ThT-positive aggregates already Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 7 of 19 Fig. 5 Bioinformatics analysis of Aβ42, Aβ24 and cross-aggregating regions and their stabilities. Regions and energies are predicted by Walsh’sPASTA 2.0 algorithm; lower energies indicate higher fibrillation propensity. a Self-aggregation propensities for the full length Aβ42 protein; b self-aggregation propensities for the Aβ24 fragment. c Cross-aggregation between the Aβ24 fragment and the C-terminal region (aa 25 to 42) of Aβ42 Fig. 6 In vitro aggregation of Aβ peptides studied by ThT fluorescence and infrared spectroscopy. a ThetimecourseofaggregationofAβ42, of Aβ24, and of the equimolar mixture of the two peptides incubated at 37 °C in PBS was monitored by ThT fluorescence with excitation and emission wavelengths at 450 nm and 485 nm, respectively. b ThT fluorescence emission spectra (excitation at 450 nm) of samples as in a) reported for selected incubation times. c Fluorescence emission spectra with excitation at 270 nm of Aβ42 and of Aβ24 incubated at 37 °C in PBS for 96 h. d ATR-FTIR spectra of Aβ42, of Aβ24, and H-Aβ42/H-Aβ24 mix incubated at 37 °C in PBS for different times, as indicated. Spectra are reported after Fourier self deconvolution (see Materials and Methods) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 8 of 19 after 24 h, while oligomeric H-Aβ24 displayed very low electrical resistance (TEER) values reached from day 7 ThT fluorescence up to 96 h of incubation (Fig. 6a and b), after plating (Fig. 7f). The in vitro BBB model was used to indicating a much lower fibrillation capability than H- investigate the passage of FAM-labeled or 488-conjugated Aβ42. Further, we recorded the emission fluorescence H-Aβ42 across the endothelial cell monolayer. While the spectra of the samples incubated for 96 h at 37 °C after addition of fluorescently labeled H-Aβ42 to the apical side tyrosine excitation at 270 nm, which is also able to of cell inserts resulted in effective BBB crossing, fluores- produce ThT fluorescence [47]. The tyrosine emission at cently labeled scrambled H-Aβ42 was not efficiently trans- around 301 nm was detected for both H-Aβ24 and H- cytosed (Fig. 7g) thus confirming the reliability and the Aβ42 samples, but only the latter showed the ThT emis- selectivity of the model as described in literature [50]. sion typical of amyloid fibrils at around 485 nm (Fig. 6c). Notably, the concomitant presence of H-Aβ24 at the An intermediate behavior was observed for the equimolar apical side of the cells significantly reduced fluorescently- mixture of oligomeric H-Aβ42/H-Aβ24 peptides (Fig. 6a labeled H-Aβ42 transfer (Fig. 7g, h). No effect of H-Aβ24 and b). The secondary structures of the peptide assemblies on the diffusion of fluorescently labeled scrambled H- were also investigated by Fourier transform infrared Aβ42 was detected (Fig. 7g, h). Similar results were ob- (FTIR) spectroscopy in attenuated total reflection (ATR) tained using different Aβ42 concentrations (1 μMor 100 [48, 49] (Fig. 6d). The FTIR spectra of the H-Aβ24 sam- nM Aβ42, Fig. 7g and h). Consistently, H-Aβ42 apical-to- ples incubated at 37 °C for 20 min and for 6 days were basolateral passage, detected by dot blot analysis of characterized by a broad Amide I band, due to the C = O medium recovered from the basolateral compartment, re- stretching vibrations of the peptide bond, with two main vealed a decreased amount of H-Aβ42 when the latter was −1 −1 peaks at ~1695 cm and ~1628 cm , assigned to the for- pre-incubated with H-Aβ24 at the apical side (Fig. 7i). mation of intermolecular β-sheet structures. In compari- These data indicatethatH-Aβ24 presence results in H- son to H-Aβ24, H-Aβ42 at 20 min displayed a higher Aβ42 retention, thus reducing its efflux through the BBB −1 intensity of the ~1628 cm peak. During incubation at and therefore preventing an efficient mechanism of Aβ42 37 °C of H-Aβ42, this component increased in intensity clearance. −1 while that at ~1695 cm decreased (Fig. 6d). These spec- Given that H-Aβ24 retains H-Aβ42, thus reducing its tral changes have been already observed in the fibrillogen- clearance through the BBB, lower H-Aβ42 levels in the esis of Aβ peptides and of other proteins, and assigned to circulation are expected. For this reason, H-Aβ42 levels the conformational conversion towards fibrillar structures were quantified by ELISA assay in serum samples of wt with a parallel orientation of the intermolecular β-sheets mice 1 week after injection of amyloid species into the [48, 49]. An intermediate behavior was observed for the brain. As expected, increased Aβ42 peripheral levels equimolar H-Aβ42/H-Aβ24 mix (Fig. 6d). Therefore, the were detected in Aβ42 i.c. injected mice (Fig. 7j). Not- spectroscopic analyses indicated a low fibrillization pro- ably, a significant reduction of H-Aβ42 peripheral levels pensity for H-Aβ24 and confirmed the higher antiparallel was detected when H-Aβ24 and H-Aβ42 were injected content of this peptide, as predicted by bioinformatics together into the brain at equimolar amounts (4pmol of analysis (Fig. 5c). These data, together with the cross- H-Aβ42 or 2pmol of H-Aβ24 plus 2pmol of H-Aβ42) aggregation potential between H-Aβ1-24 and H-Aβ25-42 (Fig 7k). In order to exclude that the lower Aβ42 plasma (Fig. 5c), suggest that H-Aβ24 may promote aggregate for- content, observed after the i.c. injection of H-Aβ24 and mation involving intermolecular β-sheet interactions, H-Aβ42, could result from the reduced amount of Aβ42 which possibly retain H-Aβ42 through cross-aggregation injected (2pmol vs 4 pmol), the experiment was repeated between segments in the 1–24 and 25–42 ranges. upon injection of the same amount of Aβ42, either in as- sociation or not with Aβ24 (8pmol of H-Aβ42 or 8pmol Synthetic Aβ24 fragment impairs Aβ42 clearance in a of H-Aβ24 plus 8pmol of H-Aβ42). Reduced Aβ42 per- blood brain barrier model ipheral levels were detected also in this case when H- To directly investigate whether oligomeric H-Aβ24 forms Aβ24 was co-injected together with H-Aβ42 (Fig. 7l). aggregates retaining Aβ42 and impairing the clearance of Aβ42 through the blood brain barrier (BBB), we used an Injected H-Aβ24 aggregates with endogenously produced in vitro BBB transwell model, formed by (brain) endothe- mouse Aβ42 lial bEnd.3 cells cultivated in the abluminal compartment Considering that H-Aβ24 would reduce Aβ42 clearance of cell culture inserts until a post-confluent monolayer thus causing an increase in the levels of brain Aβ42, we had grown (Fig. 7a). Confirmation of the morphological hypothesized that deposits observed in mice injected and functional properties of the endothelial cell monolayer with only H-Aβ24 could derive from a co-aggregation of (Fig. 7b) were obtained by immunostaining for the tight the injected peptide and endogenously produced mouse junction proteins claudin5 (Fig. 7c) and ZO1 (Fig. 7d), Aβ42, retained in the brain. Consistently, immunoblot connexin43 (CX43) (Fig. 7e) and by the transendothelial analysis of brain homogenate fractions using Aβ-specific Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 9 of 19 Fig. 7 Aβ24 fragment diminishes Aβ42 clearance through the BBB. a Graphical representation of the in vitro BBB model composed by a monolayer of brain endothelial cells seeded and cultured on an inverted matrix-coated porous membrane, allowing an apical compartment (donor-“brain”-side) physically separated from the basolateral chamber (receiving-“blood”-side). b Representative images of brain endothelial cells tightly wedged together and c expressing cell type-specific tight junctional proteins claudin-5 (green) d and ZO-1 (green), and e the gap junction protein Connexin-43 (green). Nuclei counter-stained with DAPI (blue). Scale bars: 20 μm. f Barrier properties monitored in terms of gradual increase in transendothelial electrical resistance (TEER) during cell monolayer’s formation over time. g and h Apical-to-basolateral exchange across endothelial monolayer of fluorescent Aβ42 or scramble Aβ42 control peptide, 1 μM g) or 100 nM h), over 120 min in presence or absence of Aβ24 at equimolar concentration. Quantification of unidirectional trans- endothelial Aβ42 transport by fluorescence spectrophotometry n ≥ 3 experiments; statistical analysis was performed by One way Anova, followed by Bonferroni’s post hoc test for multiple comparisons (**P < 0.01; ***P < 0.001). i) Dot blot analysis of medium collected in the abluminal compartment 120 min after brain endothelial cell monolayers exposure to Aβ42 or Aβ42/Aβ24 mix. Histograms represent the densitometry quantification upon staining with anti-6E10 antibody. Results are expressed as mean values of triplicates in each experimental group ± SE. Values were normalized on control; statistical analysis was performed by unpaired T test (**P < 0.01). j Aβ42 (pg/ml) absolute values detected by ELISA in the serum of mice 1 week after H-Aβ42 intracranical injection. k Aβ42 serum level measured 1 week after the injection of vehicle, H-Aβ42 and H-Aβ42/H-Aβ24 mix (2pmol of H-Aβ24 plus 2pmol of H-Aβ42) or l H-Aβ42/H-Aβ24 mix (8pmol of H-Aβ24 plus 8pmol of H-Aβ42). N = 4 to 6 animals per experimental group. Values are normalized on vehicle. Statistical analysis was performed by unpaired T test (*P < 0.05) monoclonal antibody m3.2 (which recognizesresidues10– homeostatic conditions [52–55]. However, on the other 15 of murine Aβ [51] (Fig. 8; supernatant, top panel and hand, microglia might be detrimental exacerbating Aβ pellet, bottom panel) showed the presence of m3.2 positive deposition and causing neuronal damage through the bands (squares) in wt mice injected with H-Aβ24 and, at a production of amyloidogenic, truncated forms of Aβ lesser extent in mice injected with H-Aβ24/H-Aβ42 mix. [12, 54]. To obtain the proof-of-concept that C- This result corroborates our hypothesis by which H-Aβ24 terminal truncated fragments could be produced by injected in the brain interacts with endogenously produced microglia, we treated primary microglia cultures with mouse Aβ42. As a specificity control, brain homogenates of 488-conjugated H-Aβ42 and in line with literature, we mice lacking the amyloid precursor protein APP (App−/−) observed that already after 3 h of treatment Aβ was were negative for the m3.2 antibody (Fig. 8). efficiently internalized by the cells (Fig. 9a). We took advantage of different antibodies, which recognize dis- Microglial MMP9 is responsible for Aβ42 degradation and tinct portions of the protein to assess the formation of for the production of a C-terminal truncated Aβ fragment H-Aβ42-derived Aβ forms (Fig. 9b). Double immuno- In AD patients and in transgenic models of the disease, fluorescence using antibodies against the C-terminal microglial activation in response to Aβ is followed by Aβ (anti-Aβ42) and the N-terminal (6E10) regions of Aβ42 internalization via phagocytosis, in an attempt of restoring revealed that, 24 h after its internalization, an antigen Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 10 of 19 right). Notably, ELISA for the C-terminal domain of the protein revealed that the modifications in the nature of the extracellular amyloid fragments were prevented by microglia incubation with protease inhibitors (Fig. 9e), indicating that microglia promote the metabolic process- ing of Aβ42, thus favoring the production of fragments which include the N-terminal, but not the C-terminal domain of the Aβ peptide. The lack of changes in the amount of Aβ42 in the medium of microglia, as detected with the antibody 4G8 which recognizes aa 17–24 (Fig. 9d), suggests that the truncated fragment contains the N-terminal part of the protein up to at least aa 23– 24, residues which are known to be cleaved by MMP9 [20]. Consistently with the involvement of MMP9 in Aβ42 processing, the production of the C-terminal trun- cated fragment was inhibited in microglia cells lines (N9) exposed to specific MMP9 siRNA to reduce the en- zyme expression (Fig. 9f). The same results were ob- tained by dot blot staining. Indeed, the exposure of primary microglia to protease inhibitors (Fig. 9g and h), the knocking-down of MMP9 in N9 cells (Fig. 9g and i) −/− as well as the use of primary microglia from MMP9 mice (Fig. 9g and j) prevented the reduction in the amount of the C-terminal fragment in the medium of cells exposed to Aβ42 for 24 h. Staining with 6E10 anti- body revealed the same amount of N-terminal fragments Fig. 8 Production of murine-Aβ in H-Aβ24 and H-Aβ24/H-Aβ42 injected mice. Tris-Tricine SDS–polyacrylamide gel electrophoresis (PAGE) followed in the extracellular media of cells under the different ex- by immunoblotting using murine-Aβ-specific monoclonal antibody m3.2 perimental conditions (Fig 9g, right). To demonstrate of brain homogenate factions (supernatant, top panel and pellet, bottom that the N-terminal fragments deriving from microglia panel) of mice injected with vehicle, H-Aβ24, H-Aβ42 or H-Aβ24/H-Aβ42 proteolytic activity are responsible for a diminished mix and analyzed after 4 months. Asterisks indicate the presence of 6KDa Aβ42 clearance in vitro, we performed the same experi- band positive to m3.2 antibody only in the pellet fraction of mice injected with H-Aβ24. Actin was used as loading control ment as in Fig. 7g, by incubating the BBB model with microglia conditioned medium (previously treated with unlabeled Aβ42 for 24 h) in the presence or absence of becomes detectable inside microglia, which is select- protease inhibitors. As expected, adding microglia condi- ively recognized by the 6E10 (N-ter) but not by the tioned medium significantly reduced 488-conjugated H- anti-Aβ42 (C-ter) (Fig. 9b arrows). Aβ42 clearance across brain endothelial cell monolayers. In line with the formation of a truncated Aβ42 form, The presence of protease inhibitors in microglia medium dot blot analysis of the extracellular medium of micro- increased fluorescently labelled H-Aβ42 passage through glia exposed to H-Aβ42 revealed, after 24 h, a reduction the BBB (Fig. 9k). Overall, these results suggest that of extracellular Aβ42, when stained by the C-terminal microglia play a central role in the production of C- anti-Aβ42 (directed against the aa 38–42) but not by the terminal truncated fragments by the activity of extracel- 6E10 antibody, directed against the N-terminal domain lular proteases. This phenomenon could be responsible of Aβ42 (aa 1–16) (Fig. 9c). We then quantified Aβ42 for an enhanced Aβ42 deposition and seeding in the extracellular concentrations over time using an Aβ42 healthy brain. ELISA kit based on a capture antibody directed against the C-terminal domain of the protein and a detection Discussion and conclusions antibody against aa 11–28. Consistently, we observed a Although the significance of amyloid deposits for the significant decrease of Aβ42 in the microglia medium, pathogenesis of AD is still under debate [56–58], the obser- starting from 12 h after treatment (Fig 9d, left). On the vation that, in brain, harmful proteins show high propensity contrary, an ELISA kit based on the capture antibody to aggregate indicates that formation of deposits is import- 6E10 revealed the lack of changes in the amount of ant in the pathogenesis of brain disorders. While Aβ- Aβ42 in the medium of microglia (detection antibody containing brain extracts from AD patient or transgenic 4G8 against the central part of Aβ42, aa 17–24, Fig. 9d, mouse model have been found to induce Aβ deposition in Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 11 of 19 ab DAPI,anti-Aβ42,6E10 c d 4000 4000 H-Aβ42 ** 30 3000 3000 anti-Aβ42 20 2000 20 2000 6E10 10 1000 10 1000 0 0 0 0 hours hours hours hours ef g H-Aβ42 ctrl ctrl siRNA +prot.inhib. control 4000 +MMP9 siRNA +prot.inh. *** ** CTRLsiRNA MMP9siRNA MMP9 -/- hours hours ctrl siRNA -/- h +prot.inhib. i j MMP9 k +MMP9 siRNA ** 1.5 20 *** 1.0 0.5 0 0 0 microglia microglia +prot.inhib. hours hours hours Fig. 9 Microglia promote the formation of a C-terminal lacking amyloid fragment in vitro. a 3D reconstruction by Imaris software of microglia cells (Iba1 positive, red) treated or not for 3 h with 1 μMAβ42–488 (green). Scale bar: 4 μm. b Bright field images of microglia cells stained with 6E10 (red) or anti-Aβ42 (green) antibodies, before (left panel) or after (right panel) 24 h incubation with H-Aβ42. Arrows indicate 6E10 positive puncta not co-localizing with anti-Aβ42 positive domains. Scale bar: 10 μm. On top: schematic representation of H-Aβ42 sequence, showing the binding sites for the different antibodies. c Dot blot analysis of the extracellular medium collected 6 and 24 h after microglia exposure to H-Aβ42. Histograms represent the densitometry quantification upon staining with anti-Aβ42 C-terminal antibody (left histogram) or 6E10 N-terminal antibody (right histogram). Intensity values are shown. N = 5 independent experiments, statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01, *P < 0.05). d Quantification of Aβ42 extracellular levels in microglia cultures exposed to H-Aβ42 for the indicated time points. Two different ELISA kits based on a capture antibody against the C-terminal (left) or N-terminal (right) domains of the protein were used. N = 5 independent experiments. Statistical analysis was performed by One way Anova, Bonferroni’s post hoc test for multiple comparisons (*** P < 0.001). e ELISA of C-terminal-containing Aβ42 in the extracellular medium of microglia exposed to H-Aβ42 in both control conditions and in the presence of protease inhibitors. N = 3 independent experiments. Statistical analysis was performed by two way Anova, Bonferroni’s post hoc test for multiple comparisons. f ELISA of C-terminal-containing Aβ42 in the extracellular medium of N9 cells exposed to siRNAcontrol or MMP9siRNA. Statistical analysis was performed by two way Anova, Bonferroni’s post hoc test for multiple comparisons. g Representative dot blots of the extracellular medium from H-Aβ42-treated microglia in control conditions, in the −/− presence of protease inhibitors, from siRNAcontrol or MMP9siRNA or from MMP9 microglia. Blots are immunostained with anti-Aβ42 or 6E10 antibodies. h-j Dot blot quantification of the extracellular medium collected from microglia in control conditions or in the presence of protease inhibitors h), from −/− siRNAcontrol or MMP9siRNA i)or from MMP9 microglia j) after 6 and 24 h of Aβ42 treatment. k) Apical-to-basolateral exchange across the BBB of fluorescent Aβ42 over 120 min upon pre-incubation with Aβ42-treated microglia conditioned medium in the presence or absence of protease inhibitors. Quantification of unidirectional trans-endothelial Aβ42 transport by fluorescence spectrophotometry of N = 2 independent experiments; statistical analysis was performed by One way Anova, Bonferroni’s post hoc test for multiple comparisons (**P < 0.01) the healthy brain [8–11, 59], literature evidences clearly in- brain. Here we show that the injection of a C-terminal dicate that intracranial injection of a single form of syn- truncated synthetic Aβ peptide (Aβ24), which may result thetic Aβ (Aβ42) does not induce plaque formation in wt from microglial MMP9 proteolytic activity, has seeding NT NT NT 6 0 24 1 NT NT NT 0h 6h 24h NT NT 0h NT 6h 24h 0h 6h 24h NT C-ter-H-Aβ42 (pg/ml) anti-Aβ42 (densitometry) anti-Ab42( densitometry) anti-Aβ42 (densitometry) C-ter-H-Aβ42 (pg/ml) 6E10 (densitometry) anti-Ab42(densitometry) anti-H-Aβ42 C-ter-H-Aβ42 (pg/ml) Papp (f old change) N-ter-H-Aβ42 (pg/ml) 6E10 Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 12 of 19 properties for aggregate formation in intracranially injected and, thereby, increasing Aβ concentration, may foster dis- wild type mice. Our results thus provide a direct demon- ease progression [64]. stration of the concept that biologically relevant mixtures We demonstrate that Aβ24, when concomitantly of Aβ forms may result in vivo in more complex aggrega- present with Aβ42 at the abluminal side of the endothe- tion dynamics than those predicted by in vitro studies. Not- lium, slows amyloid clearance through BBB. This is ably, amyloid deposits are sporadically detectable also in likely due to the fact that H-Aβ24 aggregates, which dis- thenon-injectedhemisphereand in thehippocampus, thus play higher antiparallel character, retain Aβ42, thus excluding they may represent post-injectional sprouting or impairing its clearance through the BBB. Although we leftover after cortical injection and brain lesioning. cannot exclude that Aβ42 levels measured in the serum Although it is known that the number of amyloid plaques of injected mice may reflect other routes of clearance does not necessarily correlate with cognitive impairments besides BBB, our results suggest that the same mechan- ([60–62]; reviewed in [63]), the presence of amyloid de- ism may operate also in vivo. In particular, impaired posits at the hippocampal level may explain the occurrence clearance might overcome the critical concentration of of the behavioral defects observed in injected in healthy proteins initiating the aggregation process and causing mice. the formation of ThT, Congo-red and 6E10-positive It is now established that protein aggregation takes aggregates detected in wt brains upon injection of the place, biophysically, once a critical concentration of H-Aβ24/H-Aβ42 mix. Hence, H-Aβ24, and possibly proteins has been overcome [64]. The lag phase is re- other C-terminal truncated fragments, may function as duced by the presence of “seeds” [64], which enhance scaffold proteins to favor both human (in the case of H- fibrils formation. An efficient process of clearance is re- Aβ24/H-Aβ42 mix injection) and mouse Aβ42 (in the quired in order to prevent the increase in concentration case of H-Aβ24 injection) recruitment, fibrillation and of “seeds”, which may in turn initiate the aggregation deposition. Interestingly, Aβ24 has been detected as a process [64]. Consistently, Aβ clearance rates were major amyloid component in leptomeninges of patients foundtobeimpairedin ADpatientscomparedto cog- affected by HCHWA-D (hereditary cerebral hemorrhage nitively normal controls, while there were no differ- with amyloidosis, Dutch type). These data indicate that ences in Aβ production rates [65]. Although no specific the Aβ24 fragment is in fact formed also in human evidence on the clearance of oligomeric forms is cur- brain, possibly generated by carboxyl-terminal limited rently available, oligomeric Aβ intermediates have been proteolysis [74]. Although future studies will be required found to alter proteasomal clearance [66]. to clarify what type of Aβ is deposited, our results are in Several mechanisms for Aβ clearance have been iden- line with the work of Schlenzig and colleagues which tified, including drainage via the BBB, which is mostly showed that N-terminally truncated and pyroglutamate- mediated by the low-density lipoprotein receptor related modified amyloid beta peptides are less soluble than full- protein-1 (LRP1). LRP1 is localized on the abluminal length peptides, increasing aggregation propensity and side of the brain capillary endothelium and mediates Aβ seeding of amyloid peptides [64, 75]. Notably, wt mice transport across the BBB in the direction of brain to injected with H-Aβ24 alone or with H-Aβ24/H-Aβ42 mix, blood [67–69]. When Aβ binds to LRP1 at the brain side display, besides Aβ protein deposition, behavioral defects of the BBB, a process of transcytosis starts, which medi- similar to those of age matched APP/PS1 mice. ates rapid Aβ clearance. Notably, LRP1 expression is re- One may want to consider whether structural informa- duced during aging and in AD as well as in patients with tion available for Aβ peptides could provide a possible the Dutch-type of cerebrovascular β-amyloidosis. The rationalization of the effect of cleavage at residue 24. transcytosis process is very efficient and, indeed, human Aβ The structural arrangement of Aβ fibrils and, even more injected into different brain regions of wt mice is rapidly re- importantly, oligomers and intermediates is still a matter covered in the plasma [70]. Consistently, no plaques nor ag- of debate; structural determinations are made difficult, gregates are formed in wt brain upon injection of Aβ42 [8, among other factors, by the presence of extensive poly- 9]. Interestingly, post-translationally modified forms of Aβ morphisms [76]. For fibrillar aggregates, numerous stud- are cleared less efficiently from the brain, like in the case of ies point towards a β-sandwich motif, i.e. two sheets N-terminal truncated and pyroglutamate-modified Aβ and with strands oriented perpendicular to the long axis of phosphorylated Aβ [71, 72]. Notably, Aβ peptides with the fibril; β-sandwiches are in turn arranged in higher higher β-sheet content are cleared less efficiently from order structures, i.e. with two-fold or three-fold sym- brain, due to a low-affinity LRP/Aβ interaction, mediating metry around the fibril axis [77]. A common feature of brain accumulation of amyloid. Based on the view that in- such models is that the two sheets comprise the two sufficient clearance of Aβ plays an essential role in the separate aggregation-prone regions identified above; pathogenesis of AD [73], one may speculate that even low therefore, the elimination of the C-terminal region in amounts of Aβ forms impairing physiological Aβ clearance Aβ24 would delete one of the sheets, and hence be Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 13 of 19 incompatible with the commonly assumed Aβ42 ar- KLVFFA has been crystallized by Eisenberg and collab- rangement (Fig. 10). orators in the antiparallel configuration [79]. These Although the consensus is weaker for what concerns data are consistent with previous observations that the the structure of oligomeric aggregates and other pos- Aβ24 fragment is present in both monomeric and ag- sible kinetic intermediates, recently proposed models gregated forms [55]. on the basis of NMR data [78] again foresee the pres- Aβ24 represents a prototypical example of a C- ence of cross-β contacts between the two extended terminal truncated fragment generated by Aβ42 prote- amyloidogenic regions identified above. Based on our olysis. Indeed, in line with the concept that Aβ42 is in- results and the fact that fibrillar and oligomeric aggrega- ternalized and metabolized by microglial cells [80–82], tion coexist in a “competitive equilibrium”,one may we provide the evidence that C-terminal truncated frag- hypothesize that the presence of Aβ24 shifts the competi- ments can be produced by microglia in a protease- tive balance towards oligomerization, and away from fib- dependent manner. These fragments may share similar rillation. Structurally, this is consistent with the following properties to those of Aβ24. Among those produced by data: first, although there is no conclusive structure for MMP9 (1–16, 1–20, 1–23, 1–30, 1–33 and 1–34, [19]), Aβ42, NMR-derived models (Fig. 10) have a consensus in one could argue that only Aβ16 should have a different attributing the steric zipper (the dry spine of a parallel-β character than the other fragments, because it lacks the fibril) to the 25–42 C-terminal region of Aβ42, which is of 16–20 region which is known to be amyloidogenic and course absent in Aβ24; second, the two algorithms provid- predicted as such by Walsh’s algorithm (among others) ing a prediction for the parallel/antiparallel character of fi- for all other proteolytic products. One would therefore brils (Walsh’sand Tartaglia’s) indicated that Aβ24 still has argue that, in the 1–16 case, both hypothetical self- aggregation potential, a relatively larger antiparallel β aggregation and cross-aggregation pathways would be propensity and that there is cross-aggregation potential abolished (Fig. 10). Conversely, C-terminal truncated between Aβ1-24 and Aβ25-42. Finally, the 16–21 KLVFFA fragments from the cleavage site at aa residue 20 up to region is predicted as the fragment’smost amyloidogenic; the 34 may possibly share similar properties to synthetic ab PDB: 2LMN PDB: 2LMP cd PDB: 2MXU PDB: 2MXU Fig. 10 Two-fold a and three-fold b symmetric structural models reported for Aβ40 by Tycko et al. [94]; and c the NMR model by Ishii et al. [89]. Peptide residues 9 to 24 shown in solid orange; residues 25 to 40 are transparent; residues 1–8 were not resolved. d same as c), with the 16–21 KLVFFA region highlighted; KLVFFA has been crystallized by Eisenberg et al. in the antiparallel configuration [79] Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 14 of 19 Aβ24. The in vivo relevance of fragments with length in small aliquots at−80 °C. Immediately prior to use, ali- possibly exceeding the aa 34 is more questionable, given quots were quickly resuspended in 50 mM NaPi, that fragments produced by microglia lack the C-terminal 100 mM NaCl pH 7.4 buffer at a concentration of region, as recognized by the antibody against aa 34–42. 40 μM, strongly vortexed, sonicated for 30 s and left at The role of microglia in the production of fragments room temperature for 20 min before being further di- which could in principle favor plaque formation, like luted for in vivo and in vitro experiments (the concen- Aβ24, adds an additional detrimental role to microglia tration used for each experiment is specified in the in AD. Indeed, an inverse correlation exists between figure legend). For Aβ clearance experiments, Aβ42–488 microglia activation and neurodegeneration [83] or cog- (HILyte, AnaspeC) and Aβ42-FAM (Anaspect) were nitive impairment [81, 84], indicating the occurrence of solubilised in NH OH and stored in small aliquots at a vicious cycle based on Aβ deposition, inflammation, −80 °C. neuronal damage and cognitive decline. Our evidence demonstrates that microglia may produce, through their Antibodies metabolic activity, C-terminal truncated Aβ forms which Antibodies used for immunoblot (western/dot blot), immu- in turn could initiate amyloid aggregation and cause noprecipitation, and immunofluorescence were as follows: phenotypic defects even in the absence of genetic muta- monoclonal antibody 6E10 (1:2000; Covance), which recog- tions associated with AD. Here we provide new evidence nizesresidues1–16 of human Aβ; rabbit anti-human beta for the contribution of microglia activation to the devel- amyloid 1–42 (1:1000; Alpha Diagnostic International) opment of the sporadic form of AD. which recognizes C-terminal 6 aa peptide from human beta 1–42; 4G8 antibody (1:2000; Covance) directed against the Materials and methods central part of Aβ42 (aa 17–24); m3.2 (kindly provided by Mice Prof. Paul Matthews) recognizes residues 10–15 of murine Mice were housed in the SPF animal facility of Humani- Aβ; rabbit anti-Iba1 antibody (1:500; Wako); ZO-1 (1:1000, tas Clinical and Research Center in individually venti- clone R40.76, Millipore); Connexin-43 (1:400; C6219, lated cages. Procedures involving animals handling and Sigma); Claudin-5 (1:800; ABT45, Millipore). Secondary care conformed to protocols approved by the Humanitas antibodies (1:200; Alexa Fluor®-conjugated, Molecular Clinical and Research Center (Rozzano, Milan, Italy) in Probes). compliance with national (4D.L. N.116, G.U., suppl. 40, 18-2-1992) and international law and policies (EEC Prediction of aggregation propensities Council Directive 2010/63/EU, OJ L 276/33, 22-09-2010; We used the AMYLPRED2 meta-predictor [87] to com- National Institutes of Health Guide for the Care and Use pare the aggregation profiles of the sequences of full of Laboratory Animals, US National Research Council, Aβ42 peptide with respect to the truncated form includ- 2011). The study was approved by the Italian Ministry of ing Aβ residues 1–24 (henceforth Aβ24). The meta- Health (approval n. 6/2014). All the experimental proce- predictor identifies putative amyloigenic regions on the dures followed the guidelines established by the Italian basis of the consensus between 11 methods considering Council on Animal Care and were approved by the Ital- a range of physico-chemical properties. Further analysis ian Government decree No. 27/2010. All efforts were was carried out with the algorithms PASTA 2.0 [44] and made to minimize the number of subjects used and their PAGE/ABSOLUTERATE [88] in order to obtain quanti- suffering. Mice were housed in cages with free access to tative predictions of putative aggregation propensities, food and water at 22 °C and with a 12-h alternating rates, and the fibrillar’s beta-strand parallel versus anti- light/dark cycle. Double transgenic APPswe/PSEN1dE9 parallel character. Putative three-dimensional fibrillar (APP/PS1) mice were purchased from Jackson Labora- arrangements were obtained from the PDB database en- −/− tory [85]. C57BL/6 J-App (App ) mice were provided tries 2LMN, 2LMP, 2MXU [89]. by Hertie Institute and Deutsches Zentrum für Neurode- generative Erkrankungen (DZNE), Tübingen, Germany. Thioflavin T (ThT) assays −/− BALB/c Mmp9 < tm1Tvu > (MMP9 ) P1-P3 pups were Thioflavin T (ThT) dye was purchased from Sigma provided by Istituto Nazionale dei Tumori, Milan, Italy. Aldrich. For the ThT assays, 200 μLof Aβ42 (at 8 μM concentration), of Aβ24 (at 8 μM concentration), or of Preparation of synthetic Aβ peptides the equimolar mixture of the two peptides (at 4 μM con- Synthetic human Aβ1–42 and Aβ1–24 were purchased centration each) were incubated in 50 mM NaPi, from Bachem and prepared as previously described [86] 100 mM NaCl pH 7.4 buffer at 37 °C with ThT at 10 μM to obtain oligomeric Aβ forms. Briefly, lyophilized Aβ concentration. At different incubation times, the fluores- 1–42 and 1–24 were dissolved in dimethyl sulfoxide cence emission spectra of the samples were collected (DMSO, Sigma) to a concentration of 2 mM and stored after excitation at 450 nm [45] or at 270 nm [47] by the Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 15 of 19 Cary Eclipse Spectrofluorimeter (Varian Australia Pty divided into a peripheral and a central region using Ltd, Mulgrave VIC, Australia). Quartz cuvettes of 1 cm ViewPoint video tracking software. The central quad- path length were employed. For time course experi- rants are collectively referred to as the center zone and ments, the samples were kept at 37 °C and analyzed at peripheral quadrants are collectively referred to as the each time point. peripheral zone. Data were collected continually for 30 min and the distance traveled (cm), velocity (cm/s), ATR-FTIR spectroscopy and the distance traveled in the center zone versus the For the ATR-FTIR measurements, 2 μLof Aβ42 (at peripheral zone were all recorded and scored automatic- 100 μM concentration), of Aβ24 (at 100 μM concentra- ally. In the open field task ambulatory movements (val- tion), or of the equimolar mixture of the two peptides ued as distance traveled and movement speeds) and (at 50 μM concentration each) were deposed on the sin- anxiety-like behaviors (as the as distance traveled in the gle reflection diamond element of the ATR device center zone versus the peripheral zone) can be assessed (Quest, Specac, UK). Spectra were recorded after solvent in response to a novel environment. evaporation to allow the formation of an hydrated film as previously described [48]. FTIR measurements were Social interaction task performed using the Varian 670-IR spectrometer (Varian The social interaction test was used to measure how mice Australia Pty Ltd, Mulgrave VIC, Australia) under the respond to a social partner during a 5-min test following following conditions: 1000 scan coadditions, 25 kHz of −1 isolation housing. Since isolation housing potentiates ex- scan speed, 2 cm of spectral resolution, triangular pression of innate territorial defensive responses, in this apodization, and a nitrogen-cooled Mercury Cadmium test we evaluated the reaction of mice to the presence of a Telluride detector. Fourier self deconvolution was ob- −1 new animal in free conditions in the same arena. For this tained with a full width at half height of 13.33 cm and test, mice were placed for 30 min alone into the open field a resolution enhancement factor K = 1.5 [48] using the arena to familiarize themselves with the new environment. Resolutions-Pro software (Varian Australia Pty Ltd, Mul- After this time a new female was introduced and the rec- grave VIC, Australia). ord started. We used the ViewPoint system to count the number of time the tested mouse made contact with the Intracranial injections new females. The decision to use females was made to Stereotaxic intracranial injections in mice brains were avoid any occurring of an aggressive behaviour simply made under a mixture of ketamine (100 mg/kg) and caused by a normal intermale instinct. xylazine (10 mg/kg) anesthesia. After surgical exposure of dura mater, both the bregma and the skull surface served as the stereotaxic zero points. Using a Hamilton syringe, 4 μLofH-Aβ or vehicle (vehicle consists of Novel object recognition (NOR) physiological buffer−50 mM NaPi, 100 mM NaCl The test apparatus consisted of an open field box measur- pH 7.4−) were injected into the neocortex (AP 1 mm, ing 50 cm × 25 cm and all sessions were video-recorded. ML 2 mm, DV−2 mm) with a speed of 1.5 μL/min, the The first day the animal was allowed to explore the empty needle was kept in place for an additional minute be- field arena for a 10-min time period (habituation session) fore it was slowly drawn out. For each experiment 4 before being exposed to a 10-min period of familiarization pmol of H-Aβ42, 4 pmol of H-Aβ24 and 2 pmol of H- session in the presence of identical objects (A/A). This Aβ42 plus 2 pmol of H-Aβ24 for the mix condition familiarization session was followed by 1 h and 24 h delays were injected, as described in [70, 90]. Different condi- during which the animals were returned to their home tionswereusedfor specific experimentsasspecified in cages. After the delay the animals performed 10-min of test the text. After suturing the incision, mice were session (A/B) in which one object was kept as during the maintained on a warm pad until recovery from the familiarization session (A) and another was changed (B). anesthesia, then returned to their cages. All procedures The objects were made of hard plastic and had previously were conducted in accordance with institutional guide- been counterbalanced to control for any object preference lines for the care and use of experimental animals. bias. The total amount of time spent with each object was recorded and scored using fully automated ViewPoint video Behavioral tests tracking software. The time spent around each object was Open field defined as the time in which the animal directed its nose to Mice were placed in a multi-unit open field maze (View- theobjectatadistance <2.0 cm and/or by the animal Point instruments) with field chamber (25 cm long and touching the object with its nose. Data are shown as the 25 cm wide), and activity was recorded using ViewPoint total amount of time that animals spend exploring the video tracking software. Each quadrant was digitally novel object during both the 1 and 24 h delay. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 16 of 19 Microglia cell culture Human Beta-Amyloid (1–24) (Bachem) were pre- Primary microglia were obtained from mixed cultures pre- incubated for 20 min in FluoroBrite™ DMEM (Gibco) pared from the cerebral cortex of mice at the postnatal medium at 37 °C, and added to the apical (brain) side cor- day 1 (P1-3). Microglia cells were isolated by shaking responding to the donor compartment. The basolateral flasks for 45 min at 230 rpm at day 10 after plating. Cells (blood) side of the transwells represented the receiver side were then seeded on poly-L-ornithine (Sigma) pre-coated and was exposed to medium alone. Alternatively, primary wells at the density of 1,5×10 cell/mL in DMEM contain- microglial cells were incubated for 24 h at 37 °C in Fluoro- ing 20 % heat-inactivated fetal bovine serum (FBS) and in- Brite™ DMEM medium containing Human Beta-Amyloid cubated at 37 ° C in a humidified atmosphere of 5 % CO2 (1–42) (Bachem) in the presence or absence of protease and 95 % air. Where indicated, cells were pre-treated for inhibitors (Roche). Microglia conditioned medium was 45 min with protease inhibitors and EDTA 2.5 mM recovered and spun for 2 min at max speed. 1 μMHiLyte (Roche) which were directly added to the cell culture FluorTM 488-labeled Beta-Amyloid (1–42) was pre- medium. N9 cells were maintained in IDMEM (Gibco incubated for 30 min at 37 °C with microglia conditioned Laboratories, USA), supplemented with 10 % FBS. For medium, and added to the apical compartment. Following siRNA transfection, N9 cells were plated at a concentra- 120 min of incubation at 37 °C with slow mixing, samples tion of at 1×10 cells/mL into either 96 or 24 multiwell were collected from the upper and lower chambers to as- plate. Specific siRNAs were diluted at a final concentra- sess the movement of fluorescently-labelled Aβ42 across tion of 20 nM siRNA. Cells were used within 48–72 h the bEnd.3 monolayer (apical to basolateral). The level of after transfection. When not differently indicated, for fluorescence in the media collected was measured for in vitro experiments microglia were treated with H-Aβ42 488-Aβ42 (λ = 503 nm and λ =528 nm) or for FAM- ex em 400 nM. Aβ42 (λ = 492 nm and λ =518 nm) using a Synergy™ ex em H4 Hybrid Multi-Mode Microplate Reader (BioTek). Rela- Endothelial cells culture and BBB model tive fluorescence units were converted to concentration Mouse brain endothelial cells (bEnd.3) [BEND3] (ATCC® according to prepared standard, and were corrected for CRL2299™) were used as a representative BBB model. background fluorescence. The amount of fluorescently- bEnd.3 cells were cultured at 37 °C, 5 % CO /saturated labelled Aβ42 was calculated as apparent permeability humidity in DMEM supplemented with 10 % heat- (P ) coefficient [ref. Zhao Z 2015 Nat Neurosci; Keaney app inactivated fetal bovine serum (FBS), 1 % penicillin- J 2015 Sci Adv] and expressed as fold change of control. streptomycin. Briefly, Aβ42 volume cleared (ΔV ) was calculated using the equation ΔV =C ×V /C ,where C and c lower lower upper upper Transendothelial cell electrical resistance (TEER) assay C are fluorescently-labeled Aβ42 concentrations in lower bEnd.3 cells were seeded at a concentration of 30000 the donor and receiving compartments, respectively, and cells/cm onto geltrex-coated (thin layer; Gibco) trans- V is the volume on the basolateral side. The volume lower well inserts (polycarbonate, 12 mm diameter, 3 μm pore cleared (ΔV ) was plotted against assay time. Permeability size; Costar) until a monolayer was established. TEER coefficients (P) were calculated by dividing against the was assessed using a Voltohmeter (Millicell Electrical surface area of the filter (1.12 cm ). Resistance System, Millipore). Background resistance from cell-free matrix-coated transwells was subtracted from recorded values to determine absolute TEER values Immunocytochemistry and cells imaging and corrected for the area covered by the cell mono- Cells were fixed for 15 min at room temperature in a layer. TEER was measured once a day to monitor cell 4 % (w/v) PFA, 4 % (w/v) sucrose, 20 mM NaOH confluence and development of tight junctions. Change and 5 mM MgCl in PBS, pH 7.4. Cells were perme- in absolute TEER from T for each individual transwell abilized and blocked for 30–60 min at room was recorded over time and then averaged for each day temperature in 15 % (w/v) goat serum, 0.3 % (v/v) before treatment. Triton X-100, 450 mM NaCl, 20 mM phosphate buffer, pH 7.4 and incubated at 4 °C overnight with Aβ42 permeability of the endothelial barrier primary antibodies diluted in blocking buffer. Cover- To assess Aβ42 exchange across the BBB model, bEnd.3 slips were mounted onto slides with PBS containing cells were cultured upside-down on a transwells system as 70 % glycerol and 1 μM DAPI. Representative images described above, until steady-state TEER had been were taken using a confocal microscope Fluoview reached. 1 μM FAM-labeled Human Beta-Amyloid (1–42) FV1000 Olympus IX81 (Center Valley, PA, USA) with an or FAM-labeled scrambled Beta-Amyloid (1–42) or oil immersion objective (×40 or × 60 × 1.4 NA Plan- HiLyte Fluor™ 488-labeled Beta-Amyloid (1–42) (Anaspec Apochromat; Olympus) using laser excitation at 405, 488 Peptide, Eurogentec) in the presence or absence of or 594 nm, and processed using Fiji [91]. Alternatively Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 17 of 19 bEnd.3 cells were stained with Diff Quick (Dade Behering, Histological and immunohistochemical analyses BioMap) and acquired with inverted microscope Olympus Mice were anesthetized and perfused with 0.9 % saline, IX53 (Center Valley, PA, USA). followed by 4 % PFA. The 30 μm cryosections of brain were blocked in PBS containing 10 % goat serum and Brain homogenate preparation and western blot analysis 0.1 % Triton X-100 for 1 h at room temperature before Animals were anesthetized and perfused with PBS before being incubated overnight at 4 °C with primary anti- brains were removed, weighted and homogeneted in a bodies. The following day, the slides were rinsed in PBS mild hypotonic buffer (50 mM Tris pH 8, 150 mM NaCl, and incubated at room temperature for 1 h with second- 5 mM EDTA supplemented with phosphatase and prote- ary antibody. The slides were processed using the ABC ase inhibitor, EDTA-free, Roche). Supernatants were then detection kit (Vector Laboratories). The presence of the centrifuged for 1 h at 3000 × g to pellet insoluble material, antigens was revealed using the DAB (diaminobenzidine) including insoluble Aβ species. Samples were either frozen (brown) peroxidase substrate kit (Vector Laboratories). on dry ice or LDS sample buffer (Invitrogen) containg 5 % Immunofluorescence and ThT staining were performed β-mercaptoethanol was immediately added. Pellets were on brain sections. Brain slices were washed three times directly resuspended in 20 μl of LDS sample buffer. 10 μl in PBS and incubated for 1 h at room temperature in sample were loaded onto a Bolt 4–12 % Bis-Tris Plus Gels blocking solution (3 % BSA). Subsequently, the slides (Thermo Fisher) for Western Blot evaluation. Samples were washed 3 times and incubated overnight with spe- were separated using MES buffer and transferred onto a cific antibodies. The next day, sections were washed in 0.22 μM nitrocellulose membrane (Bio-Rad). Membranes PBS and incubated for 2 h at room temperature with were washed in TBS-Tween (150 mM NaCl, 50 mM Tris specific fluorochrome-conjugated secondary antibodies and 0.1 % (v/v) Tween-20) and incubated for 45 min at diluted in 3 % BSA in the dark. ThT solution was pre- room temperature in blocking solution (5 % milk or 2.5 % pared as described above. Final solution was added to serum bovine albumin in TBST). Membranes were subse- free floating slides for 1 min and very quickly washed quently probed overnight at 4 °C with primary antibodies with 80 % methanol followed by 3 washes with distilled diluted in TBS-T buffer. Membranes were washed exten- water. Images were acquired by Virtual Slides micro- sively and incubated for 45 min at room temperature with scope (VS120, Olympus). For quantification, both dif- horseradish peroxidase-conjugated (HRP) secondary anti- fuse-plaques and dense-core plaques were considered, as body diluted in TBST buffer. Antibody-specific signals described in [92]. Small dots and deposits at the slice were detected using enhanced chemiluminescence re- edge were not counted. The entire surface of the slice agents (Clarity Western ECL substrate, Bio-Rad). was examined. For congo red histological staining, slices were air-dried on glass overnight, then stained with a Dot blot analysis 0.2 % congo red solution according to [93]. Specimens Aliquots of supernatant samples (200 μL) were loaded on were acquired by transmittance polarized light micros- nitrocellulose membrane Trans-Blot Transfer Medium copy, FITCH filtered epifluorescence. For quantification, (0.22 μm, Bio-Rad), by vacuum deposition on the Bio-Dot bright field images were analyzed by color segmentation SF blotting apparatus (Bio-Rad). Serial dilution curves of plugin-ImageJ software (NIH, Bethesda, MD). The entire Aβ42 synthetic protein were preliminarily run to obtain area of deposits was considered. non-saturating condition of immunodetection. DB dots images were analysed by Image Lab™ software (Bio-Rad). Statistical analysis Statistical analysis was performed with PRISM software ELISA (Graph-Pad Software, San Diego, CA, USA). Data are Aβ42 levels were determined using specific ELISA kit expressed as mean ± SEM. Comparisons between two (Amyloid-beta (x-42) ELISA IBL, International) following groups were performed using Student’s t test or by non- manufacturer’s instructions. Briefly, 100 μL of sample was parametric two-tailed Mann–Whitney U test. For the added into the pre-coated plate and was incubated over- comparison of more than two groups, two-way ANOVA night at 4 °C. After washing each well of the pre-coated followed by Bonferroni’s post hoc test was used. Differ- plate, 100 μL of labeled antibody solution was added and ences were considered significant at *P < 0.05, **P < 0.01, the mixture was incubated for 1 h at 4 °C in the dark. ***P < 0.001. After washing, chromogen was added and the mixture Funding was incubated for 30 min at room temperature in the This research has been supported by Cariplo 2015–0594 to MM, Italian Ministry dark. After the addition of stop solution, the resulting of Health GR-2011-02347377 to EL and MM, Cariplo 2014–0655 to IB and MM. TG would like to acknowledge prof. Amedeo Caflisch (Univ. of Zurich) for color was assayed at 450 nm using a microplate absorb- providing the PAGE software. TG also acknowledges support from the CNR-STM ance reader (Synergy H4 Synergy™ H4 Microplate Reader, 2013 mobility scheme. SM has been supported by Fondazione Veronesi. MT has BioTek). been supported by Fondazione Vollaro. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 18 of 19 Availability of data and materials 16. Hu J, et al. Angiotensin-converting enzyme degrades Alzheimer amyloid Not applicable. beta-peptide (A beta); retards A beta aggregation, deposition, fibril formation; and inhibits cytotoxicity. J Biol Chem. 2001;276(51):47863–8. 17. Backstrom JR, et al. Matrix metalloproteinase-9 (MMP-9) is synthesized in Authors’ contributions neurons of the human hippocampus and is capable of degrading the MM, SM, FF and MR designed the experiments. SM, FF and MR performed amyloid-beta peptide (1–40). J Neurosci. 1996;16(24):7910–9. analyses and most of the experiments. MM, FF and MR wrote the manuscript. 18. Zhang Q, et al. Metabolite-initiated protein misfolding may trigger EL set the blood brain barrier model and performed the experiment with Alzheimer’s disease. Proc Natl Acad Sci U S A. 2004;101(14):4752–7. endothelial cells. CS performed ELISA. MT performed intractranical injections in 19. Yan P, et al. Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro mice. DP and IB contributed data analysis and interpretation. TG performed and compact plaques in situ. J Biol Chem. 2006;281(34):24566–74. bioinformatics analyses. AN performed Tht and FTIR to test amyloid 20. Zhang R, et al. Interprotofilament interactions between Alzheimer’s Abeta1- aggregation propensity. All authors read and approved the final manuscript. 42 peptides in amyloid fibrils revealed by cryoEM. Proc Natl Acad Sci U S A. 2009;106(12):4653–8. 21. Frautschy SA, et al. Protease inhibitor coinfusion with amyloid beta-protein Competing interests results in enhanced deposition and toxicity in rat brain. J Neurosci. 1998; The authors declare that they have no competing interests. 18(20):8311–21. 22. Saido T, Leissring MA. Proteolytic degradation of amyloid beta-protein. Cold Consent for publication Spring Harb Perspect Med. 2012;2(6):a006379. Not applicable. 23. Hernandez-Guillamon M, et al. Sequential Amyloid-beta Degradation by the Matrix Metalloproteases MMP-2 and MMP-9. J Biol Chem. 2015;290(24):15078–91. Author details 24. Deb S, Gottschall PE. Increased production of matrix metalloproteinases in 1 2 IRCCS Humanitas, via Manzoni 56, 20089 Rozzano, Italy. Hertie Institute and enriched astrocyte and mixed hippocampal cultures treated with beta- Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), amyloid peptides. J Neurochem. 1996;66(4):1641–7. Otfried-Müller-Straße 27, 72076 Tübingen, Germany. Humanitas University, 25. Yin KJ, et al. Matrix metalloproteinases expressed by astrocytes mediate via Manzoni 56, 20089 Rozzano, Italy. IN-CNR, Corso Stati Uniti 4, 35126 extracellular amyloid-beta peptide catabolism. J Neurosci. 2006;26(43):10939–48. Padova, Italy. Department of Biotechnology and Biosciences, University of 26. Belli M, Ramazzotti M, Chiti F. Prediction of amyloid aggregation in vivo. Milano-Bicocca, Piazza della Scienza 2, 20126 Milano, Italy. IN-CNR, via EMBO Rep. 2011;12(7):657–63. Vanvitelli 32, 20129 Milano, Italy. 27. Biancalana M, Koide S. Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochim Biophys Acta. 2010;1804(7):1405–12. Received: 21 September 2016 Accepted: 24 September 2016 28. Khurana R, et al. Mechanism of thioflavin T binding to amyloid fibrils. J Struct Biol. 2005;151(3):229–38. 29. Alvarez A, et al. Serum TNF-alpha levels are increased and correlate negatively with free IGF-I in Alzheimer disease. Neurobiol Aging. 2007;28(4):533–6. References 30. Fillit H, et al. Elevated circulating tumor necrosis factor levels in Alzheimer’s 1. Mattson MP. Pathways towards and away from Alzheimer’s disease. Nature. disease. Neurosci Lett. 1991;129(2):318–20. 2004;430(7000):631–9. 31. Ma SL, et al. Association between tumor necrosis factor-alpha promoter 2. Scheuner D, et al. Secreted amyloid beta-protein similar to that in the senile polymorphism and Alzheimer’s disease. Neurology. 2004;62(2):307–9. plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 32. Janelsins MC, et al. Early correlation of microglial activation with enhanced and APP mutations linked to familial Alzheimer’s disease. Nat Med. 1996; tumor necrosis factor-alpha and monocyte chemoattractant protein-1 2(8):864–70. expression specifically within the entorhinal cortex of triple transgenic 3. Selkoe DJ. Aging, amyloid, and Alzheimer’s disease: a perspective in honor Alzheimer’s disease mice. J Neuroinflammation. 2005;2:23. of Carl Cotman. Neurochem Res. 2003;28(11):1705–13. 33. Sheng JG, et al. Lipopolysaccharide-induced-neuroinflammation increases 4. Bitan G, et al. Amyloid beta-protein (Abeta) assembly: Abeta 40 and Abeta intracellular accumulation of amyloid precursor protein and amyloid beta 42 oligomerize through distinct pathways. Proc Natl Acad Sci U S A. 2003; peptide in APPswe transgenic mice. Neurobiol Dis. 2003;14(1):133–45. 100(1):330–5. 34. Terry RD, et al. Physical basis of cognitive alterations in Alzheimer’s disease: 5. Yan Y, Wang C. Abeta42 is more rigid than Abeta40 at the C terminus: synapse loss is the major correlate of cognitive impairment. Ann Neurol. implications for Abeta aggregation and toxicity. J Mol Biol. 2006;364(5):853–62. 1991;30(4):572–80. 6. Hubin E, et al. Transient dynamics of Abeta contribute to toxicity in 35. DeKosky ST, Scheff SW. Synapse loss in frontal cortex biopsies in Alzheimer’s Alzheimer’s disease. Cell Mol Life Sci. 2014;71(18):3507–21. disease: correlation with cognitive severity. Ann Neurol. 1990;27(5):457–64. 7. Kuperstein I, et al. Neurotoxicity of Alzheimer’s disease Abeta peptides is 36. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer’s disease: progress induced by small changes in the Abeta42 to Abeta40 ratio. EMBO J. 2010; and problems on the road to therapeutics. Science. 2002;297(5580):353–6. 29(19):3408–20. 37. Prut L, Belzung C. The open field as a paradigm to measure the effects of 8. Kane MD, et al. Evidence for seeding of beta-amyloid by intracerebral drugs on anxiety-like behaviors: a review. Eur J Pharmacol. 2003;463(1–3):3–33. infusion of Alzheimer brain extracts in beta-amyloid precursor protein- 38. Rodriguiz RM, Wetsel WC. Assessments of Cognitive Deficits in Mutant Mice. transgenic mice. J Neurosci. 2000;20(10):3606–11. In: Levin ED, Buccafusco JJ, editors. Animal Models of Cognitive Impairment. 9. Meyer-Luehmann M, et al. Exogenous induction of cerebral beta-amyloidogenesis Boca Raton (FL): CRC Press/Taylor & Francis; 2006. is governed by agent and host. Science. 2006;313(5794):1781–4. 39. Kliethermes CL, Cronise K, Crabbe JC. Anxiety-like behavior in mice in two 10. Eisele YS, et al. Induction of cerebral beta-amyloidosis: intracerebral versus apparatuses during withdrawal from chronic ethanol vapor inhalation. systemic Abeta inoculation. Proc Natl Acad Sci U S A. 2009;106(31):12926–31. Alcohol Clin Exp Res. 2004;28(7):1012–9. 11. Watts JC, et al. Bioluminescence imaging of Abeta deposition in bigenic 40. Cheng D, et al. Novel behavioural characteristics of the APP (Swe)/ mouse models of Alzheimer’s disease. Proc Natl Acad Sci U S A. 2011;108(6): PS1DeltaE9 transgenic mouse model of Alzheimer’s disease. Behav Brain 2528–33. Res. 2013;245:120–7. 12. Oberstein TJ, et al. Astrocytes and microglia but not neurons preferentially 41. Chambon C, et al. Behavioural and cellular effects of exogenous amyloid- generate N-terminally truncated Abeta peptides. Neurobiol Dis. 2015;73:24–35. beta peptides in rodents. Behav Brain Res. 2011;225(2):623–41. 13. Howell S, Nalbantoglu J, Crine P. Neutral endopeptidase can hydrolyze beta-amyloid (1–40) but shows no effect on beta-amyloid precursor protein 42. Faucher P, et al. Hippocampal Injections of Oligomeric Amyloid beta- metabolism. Peptides. 1995;16(4):647–52. peptide (1–42) Induce Selective Working Memory Deficits and Long-lasting 14. Kurochkin IV, Goto S. Alzheimer’s beta-amyloid peptide specifically interacts with Alterations of ERK Signaling Pathway. Front Aging Neurosci. 2015;7:245. and is degraded by insulin degrading enzyme. FEBS Lett. 1994;345(1):33–7. 43. Mikros E, et al. High-Resolution NMR Spectroscopy of the beta-Amyloid (1–28) 15. Eckman EA, Reed DK, Eckman CB. Degradation of the Alzheimer’s amyloid Fibril Typical for Alzheimer’s Disease. Angew Chem Int Ed Engl. 2001;40(19):3603–5. beta peptide by endothelin-converting enzyme. J Biol Chem. 2001;276(27): 44. Walsh I, et al. PASTA 2.0: an improved server for protein aggregation 24540–8. prediction. Nucleic Acids Res. 2014;42:W301–7. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 19 of 19 45. LeVine 3rd H. Thioflavine T interaction with synthetic Alzheimer’s disease 73. Bates KA, et al. Clearance mechanisms of Alzheimer’s amyloid-beta peptide: beta-amyloid peptides: detection of amyloid aggregation in solution. implications for therapeutic design and diagnostic tests. Mol Psychiatry. Protein Sci. 1993;2(3):404–10. 2009;14(5):469–86. 46. Huang SH, et al. Effect of C-terminal residues of Abeta on copper binding 74. Castano EM, et al. The length of amyloid-beta in hereditary cerebral affinity, structural conversion and aggregation. PLoS One. 2014;9(3), e90385. hemorrhage with amyloidosis, Dutch type. Implications for the role of amyloid- beta 1–42 in Alzheimer’s disease. J Biol Chem. 1996;271(50):32185–91. 47. Librizzi F, et al. alpha-Casein Inhibits Insulin Amyloid Formation by Preventing the 75. Schlenzig D, et al. Pyroglutamate formation influences solubility and Onset of Secondary Nucleation Processes. J Phys Chem Lett. 2014;5(17):3043–8. amyloidogenicity of amyloid peptides. Biochemistry. 2009;48(29):7072–8. 48. Cerf E, et al. Antiparallel beta-sheet: a signature structure of the oligomeric 76. Nasica-Labouze J, et al. Amyloid beta Protein and Alzheimer’s Disease: amyloid beta-peptide. Biochem J. 2009;421(3):415–23. When Computer Simulations Complement Experimental Studies. Chem Rev. 49. Sarroukh R, et al. ATR-FTIR: a “rejuvenated” tool to investigate amyloid 2015;115(9):3518–63. proteins. Biochim Biophys Acta. 2013;1828(10):2328–38. 77. Luhrs T, et al. 3D structure of Alzheimer’s amyloid-beta (1–42) fibrils. Proc 50. Mancini S, et al. The hunt for brain Abeta oligomers by peripherally Natl Acad Sci U S A. 2005;102(48):17342–7. circulating multi-functional nanoparticles: Potential therapeutic approach for 78. Ahmed M, et al. Structural conversion of neurotoxic amyloid-beta (1–42) Alzheimer disease. Nanomedicine. 2016;12(1):43–52. oligomers to fibrils. Nat Struct Mol Biol. 2010;17(5):561–7. 51. Wesson DW, et al. Chronic anti-murine Abeta immunization preserves odor 79. Colletier JP, et al. Molecular basis for amyloid-beta polymorphism. Proc Natl guided behaviors in an Alzheimer’s beta-amyloidosis model. Behav Brain Acad Sci U S A. 2011;108(41):16938–43. Res. 2013;237:96–102. 80. Joshi P, et al. Microglia convert aggregated amyloid-beta into neurotoxic 52. Heneka MT, Golenbock DT, Latz E. Innate immunity in Alzheimer’s disease. forms through the shedding of microvesicles. Cell Death Differ. 2014;21(4): Nat Immunol. 2015;16(3):229–36. 582–93. 53. Wyss-Coray T, Rogers J. Inflammation in Alzheimer disease-a brief review of 81. Edison P, et al. Microglia, amyloid, and cognition in Alzheimer’s disease: An the basic science and clinical literature. Cold Spring Harb Perspect Med. [11C] (R) PK11195-PET and [11C] PIB-PET study. Neurobiol Dis. 2008;32(3):412–9. 2012;2(1):a006346. 82. Griciuc A, et al. Alzheimer’s disease risk gene CD33 inhibits microglial 54. Meda L, et al. Activation of microglial cells by beta-amyloid protein and uptake of amyloid beta. Neuron. 2013;78(4):631–43. interferon-gamma. Nature. 1995;374(6523):647–50. 83. Serrano-Pozo A, et al. Reactive glia not only associates with plaques but also 55. Bateman DA, Chakrabartty A. Cell surface binding and internalization of abeta parallels tangles in Alzheimer’s disease. Am J Pathol. 2011;179(3):1373–84. modulated by degree of aggregation. Int J Alzheimers Dis. 2011;2011:962352. 84. Okello A, et al. Microglial activation and amyloid deposition in mild 56. Giannakopoulos P, et al. Tangle and neuron numbers, but not amyloid load, cognitive impairment: a PET study. Neurology. 2009;72(1):56–62. predict cognitive status in Alzheimer’s disease. Neurology. 2003;60(9):1495–500. 85. Jankowsky JL, et al. Mutant presenilins specifically elevate the levels of the 57. Elman JA, et al. Neural compensation in older people with brain amyloid- 42 residue beta-amyloid peptide in vivo: evidence for augmentation of a beta deposition. Nat Neurosci. 2014;17(10):1316–8. 42-specific gamma secretase. Hum Mol Genet. 2004;13(2):159–70. 58. Jack Jr CR, et al. Age-specific population frequencies of cerebral beta- 86. Amijee H, et al. The N-methylated peptide SEN304 powerfully inhibits Abeta amyloidosis and neurodegeneration among people with normal cognitive (1–42) toxicity by perturbing oligomer formation. Biochemistry. 2012;51(42): function aged 50–89 years: a cross-sectional study. Lancet Neurol. 2014; 8338–52. 13(10):997–1005. 87. Tsolis AC, et al. A consensus method for the prediction of ‘aggregation- 59. Duran-Aniotz C, et al. Brains from non-Alzheimer’s individuals containing prone’ peptides in globular proteins. PLoS One. 2013;8(1), e54175. amyloid deposits accelerate Abeta deposition in vivo. Acta Neuropathol 88. Tartaglia GG, et al. Prediction of aggregation rate and aggregation-prone Commun. 2013;1:76. segments in polypeptide sequences. Protein Sci. 2005;14(10):2723–34. 60. McLean CA, et al. Soluble pool of Abeta amyloid as a determinant of severity 89. Xiao Y, et al. Abeta (1–42) fibril structure illuminates self-recognition and of neurodegeneration in Alzheimer’s disease. Ann Neurol. 1999;46(6):860–6. replication of amyloid in Alzheimer’s disease. Nat Struct Mol Biol. 2015;22(6): 61. Price JL, et al. Neuropathology of nondemented aging: presumptive evidence 499–505. for preclinical Alzheimer disease. Neurobiol Aging. 2009;30(7):1026–36. 90. Zhao Z, et al. Central role for PICALM in amyloid-beta blood–brain barrier 62. Games D, et al. Alzheimer-type neuropathology in transgenic mice transcytosis and clearance. Nat Neurosci. 2015;18(7):978–87. overexpressing V717F beta-amyloid precursor protein. Nature. 1995; 91. Schindelin J, et al. Fiji: an open-source platform for biological-image 373(6514):523–7. analysis. Nat Methods. 2012;9(7):676–82. 63. Karran E, Mercken M, De Strooper B. The amyloid cascade hypothesis for 92. Serrano-Pozo A, et al. Neuropathological alterations in Alzheimer disease. Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Cold Spring Harb Perspect Med. 2011;1(1):a006189. Rev Drug Discov. 2011;10(9):698–712. 93. Wilcock DM, Gordon MN, Morgan D. Quantification of cerebral amyloid 64. Thal DR. Clearance of amyloid beta-protein and its role in the spreading of angiopathy and parenchymal amyloid plaques with Congo red Alzheimer’s disease pathology. Front Aging Neurosci. 2015;7:25. histochemical stain. Nat Protoc. 2006;1(3):1591–5. 65. Mawuenyega KG, et al. Decreased clearance of CNS beta-amyloid in 94. Paravastu AK, et al. Molecular structural basis for polymorphism in Alzheimer’s disease. Science. 2010;330(6012):1774. Alzheimer’s beta-amyloid fibrils. Proc Natl Acad Sci U S A. 2008;105(47): 66. Cecarini V, et al. Crosstalk between the ubiquitin-proteasome system and 18349–54. autophagy in a human cellular model of Alzheimer’s disease. Biochim Biophys Acta. 2012;1822(11):1741–51. 67. Shibata M, et al. Clearance of Alzheimer’s amyloid-ss (1–40) peptide from brain by LDL receptor-related protein-1 at the blood–brain barrier. J Clin Invest. 2000;106(12):1489–99. Submit your next manuscript to BioMed Central 68. Deane R, et al. LRP/amyloid beta-peptide interaction mediates differential brain efflux of Abeta isoforms. Neuron. 2004;43(3):333–44. and we will help you at every step: 69. Deane R, et al. Clearance of amyloid-beta peptide across the blood–brain • We accept pre-submission inquiries barrier: implication for therapies in Alzheimer’s disease. CNS Neurol Disord Drug Targets. 2009;8(1):16–30. � Our selector tool helps you to ﬁnd the most relevant journal 70. Cho SM, et al. Correlations of amyloid-beta concentrations between CSF � We provide round the clock customer support and plasma in acute Alzheimer mouse model. Sci Rep. 2014;4:6777. � Convenient online submission 71. Russo C, et al. Pyroglutamate-modified amyloid beta-peptides– � Thorough peer review AbetaN3(pE)–strongly affect cultured neuron and astrocyte survival. J Neurochem. 2002;82(6):1480–9. � Inclusion in PubMed and all major indexing services 72. Kumar S, et al. Phosphorylation of amyloid-beta peptide at serine 8 � Maximum visibility for your research attenuates its clearance via insulin-degrading and angiotensin-converting enzymes. J Biol Chem. 2012;287(11):8641–51. Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals http://www.deepdyve.com/lp/springer-journals/amyloid-1-24-c-terminal-truncated-fragment-promotes-amyloid-1-42-F81dsRMkzt

Loading next page...

References (109)

J. Price, D. McKeel, V. Buckles, C. Roe, C. Xiong, M. Grundman, L. Hansen, R. Petersen, J. Parisi, D. Dickson, Charles Smith, D. Davis, F. Schmitt, W. Markesbery, J. Kaye, R. Kurlan, C. Hulette, B. Kurland, R. Higdon, W. Kukull, J. Morris (2009)
Neuropathology of nondemented aging: Presumptive evidence for preclinical Alzheimer disease
Neurobiology of Aging, 30
D. Selkoe (2003)
Aging, Amyloid, and Alzheimer's Disease: A Perspective in Honor of Carl Cotman
Neurochemical Research, 28
D. Thal (2015)
Clearance of amyloid β-protein and its role in the spreading of Alzheimer's disease pathology
Frontiers in Aging Neuroscience, 7
M. Kane, W. Lipinski, M. Callahan, F. Bian, R. Durham, R. Schwarz, A. Roher, L. Walker (2000)
Evidence for seeding of beta -amyloid by intracerebral infusion of Alzheimer brain extracts in beta -amyloid precursor protein-transgenic mice.
The Journal of neuroscience : the official journal of the Society for Neuroscience, 20 10
R. Deane, Zhenhua Wu, A. Sagare, Judianne Davis, S. Yan, Katie Hamm, Feng Xu, Margaret Parisi, Barbra LaRue, H. Hu, P. Spijkers, Huang Guo, Xiaomei Song, P. Lenting, W. Nostrand, B. Zlokovic (2004)
LRP/Amyloid β-Peptide Interaction Mediates Differential Brain Efflux of Aβ Isoforms
Neuron, 43
Jianguo Hu, A. Igarashi, M. Kamata, H. Nakagawa (2001)
Angiotensin-converting Enzyme Degrades Alzheimer Amyloid β-Peptide (Aβ); Retards Aβ Aggregation, Deposition, Fibril Formation; and Inhibits Cytotoxicity*
The Journal of Biological Chemistry, 276
David Bateman, A. Chakrabartty (2011)
Cell Surface Binding and Internalization of Aβ Modulated by Degree of Aggregation
International Journal of Alzheimer's Disease, 2011
C. Russo, E. Violani, S. Salis, V. Venezia, V. Dolcini, G. Damonte, U. Benatti, C. D'Arrigo, E. Patrone, P. Carlo, G. Schettini (2002)
Pyroglutamate‐modified amyloid β‐peptides – AβN3(pE) – strongly affect cultured neuron and astrocyte survival
Journal of Neurochemistry, 82
E. Castaño, F. Prelli, C. Soto, R. Beavis, E. Matsubara, M. Shoji, B. Frangione (1996)
The length of amyloid-beta in hereditary cerebral hemorrhage with amyloidosis, Dutch type. Implications for the role of amyloid-beta 1-42 in Alzheimer's disease.
The Journal of biological chemistry, 271 50
A. Okello, P. Edison, H. Archer, F. Turkheimer, J. Kennedy, R. Bullock, Z. Walker, A. Kennedy, Nick Fox, M. Rossor, D. Brooks (2009)
Microglial activation and amyloid deposition in mild cognitive impairment
Neurology, 72
P. Giannakopoulos, F. Herrmann, T. Bussiere, C. Bouras, E. Kovari, D. Perl, J. Morrison, G. Gold, P. Hof (2003)
Tangle and neuron numbers, but not amyloid load, predict cognitive status in Alzheimer’s disease
Neurology, 60
A. Griciuc, A. Serrano‐Pozo, A. Parrado, Andrea Lesinski, Caroline Asselin, K. Mullin, B. Hooli, Se Choi, B. Hyman, R. Tanzi (2013)
Alzheimer’s Disease Risk Gene CD33 Inhibits Microglial Uptake of Amyloid Beta
Neuron, 78
Ping Yan, Xiaoyan Hu, Haowei Song, Kejie Yin, R. Bateman, J. Cirrito, Qingli Xiao, F. Hsu, J. Turk, Jan Xu, C. Hsu, D. Holtzman, J. Lee (2006)
Matrix Metalloproteinase-9 Degrades Amyloid-β Fibrils in Vitro and Compact Plaques in Situ*
Journal of Biological Chemistry, 281
Levin Ed, Buccafusco Jj (2006)
Assessments of Cognitive Deficits in Mutant Mice -- Animal Models of Cognitive Impairment
M. Mattson (2004)
Pathways towards and away from Alzheimer's disease
Nature, 430
(2011)
Molecular basis for amyloid-beta polymorphism
K. Mawuenyega, Wendy Sigurdson, Vitaliy Ovod, L. Munsell, Tom Kasten, J. Morris, K. Yarasheski, R. Bateman (2010)
Decreased Clearance of CNS β-Amyloid in Alzheimer’s Disease
Science, 330
J. Elman, Hwamee Oh, Cindee Madison, S. Baker, J. Vogel, Shawn Marks, Sam Crowley, J. O’Neil, W. Jagust (2014)
Neural compensation in older people with brain β-amyloid deposition
Nature neuroscience, 17
M. Hernández-Guillamon, S. Mawhirt, S. Blais, J. Montaner, T. Neubert, A. Rostagno, J. Ghiso (2015)
Sequential Amyloid-β Degradation by the Matrix Metalloproteases MMP-2 and MMP-9*
The Journal of Biological Chemistry, 290
E. Eckman, D. Reed, C. Eckman (2001)
Degradation of the Alzheimer's Amyloid β Peptide by Endothelin-converting Enzyme*
The Journal of Biological Chemistry, 276
P. Joshi, E. Turola, A. Ruiz, A. Bergami, D. Libera, L. Benussi, Paola Giussani, G. Magnani, G. Comi, Giuseppe Legname, R. Ghidoni, Roberto Furlan, M. Matteoli, C. Verderio (2013)
Microglia convert aggregated amyloid-β into neurotoxic forms through the shedding of microvesicles
Cell Death and Differentiation, 21
G. Tartaglia, A. Cavalli, R. Pellarin, A. Caflisch (2005)
Prediction of aggregation rate and aggregation‐prone segments in polypeptide sequences
Protein Science, 14
Yvonne Eisele, T. Bolmont, M. Heikenwalder, Franziska Langer, L. Jacobson, Zheng-xin Yan, K. Roth, A. Aguzzi, M. Staufenbiel, L. Walker, M. Jucker (2009)
Induction of cerebral β-amyloidosis: Intracerebral versus systemic Aβ inoculation
Proceedings of the National Academy of Sciences, 106
Masayoshi Shibata, S. Yamada, S. Kumar, M. Calero, J. Bading, B. Frangione, D. Holtzman, C. Miller, D. Strickland, J. Ghiso, B. Zlokovic (2000)
Clearance of Alzheimer's amyloid-ss(1-40) peptide from brain by LDL receptor-related protein-1 at the blood-brain barrier.
The Journal of clinical investigation, 106 12
J. Jankowsky, Daniel Fadale, Jeffrey Anderson, Guilian Xu, V. Gonzales, N. Jenkins, N. Copeland, Michael Lee, L. Younkin, S. Wagner, S. Younkin, D. Borchelt (2004)
Mutant presenilins specifically elevate the levels of the 42 residue beta-amyloid peptide in vivo: evidence for augmentation of a 42-specific gamma secretase.
Human molecular genetics, 13 2
J. Watts, K. Giles, Sunny Grillo, Azucena Lemus, S. DeArmond, S. Prusiner (2015)
Correction for Watts et al., Bioluminescence imaging of Aβ deposition in bigenic mouse models of Alzheimer's disease
Proceedings of the National Academy of Sciences, 112
L. Meda, M. Cassatella, G. Szendrei, L. Otvos, P. Baron, M. Villalba, D. Ferrari, F. Rossi (1995)
Activation of microglial cells by beta-amyloid protein and interferon-gamma.
Nature, 374 6523
H. Amijee, C. Bate, Alun Williams, J. Virdee, R. Jeggo, D. Spanswick, D. Scopes, J. Treherne, S. Mazzitelli, Ross Chawner, C. Eyers, A. Doig (2012)
The N-methylated peptide SEN304 powerfully inhibits Aβ(1-42) toxicity by perturbing oligomer formation.
Biochemistry, 51 42
A. Serrano‐Pozo, Matthew Mielke, T. Gómez-Isla, R. Betensky, J. Growdon, M. Frosch, B. Hyman (2011)
Reactive glia not only associates with plaques but also parallels tangles in Alzheimer's disease.
The American journal of pathology, 179 3
David Cheng, W. Logge, Jac Low, B. Garner, T. Karl (2013)
Novel behavioural characteristics of the APP Swe /PS1ΔE9 transgenic mouse model of Alzheimer's disease
Behavioural Brain Research, 245
S. DeKosky, S. Scheff (1990)
Synapse loss in frontal cortex biopsies in Alzheimer's disease: Correlation with cognitive severity
Annals of Neurology, 27
Shu-Hsiang Huang, S. Ke, Ta-Hsin Lin, H. Huang, Yi-Cheng Chen (2014)
Effect of C-Terminal Residues of Aβ on Copper Binding Affinity, Structural Conversion and Aggregation
PLoS ONE, 9
Yiling Xiao, B. Ma, Dan McElheny, S. Parthasarathy, Fei Long, M. Hoshi, R. Nussinov, Y. Ishii (2015)
Aβ(1–42) Fibril Structure Illuminates Self-recognition and Replication of Amyloid in Alzheimer’s
Nature structural & molecular biology, 22
J. Watts, K. Giles, Sunny Grillo, Azucena Lemus, S. DeArmond, S. Prusiner (2011)
Bioluminescence imaging of Aβ deposition in bigenic mouse models of Alzheimer's disease
Proceedings of the National Academy of Sciences, 108
Rabia Sarroukh, E. Goormaghtigh, J. Ruysschaert, V. Raussens (2013)
ATR-FTIR: a "rejuvenated" tool to investigate amyloid proteins.
Biochimica et biophysica acta, 1828 10
R. Zhang, X. Hu, H. Khant, S. Ludtke, W. Chiu, M. Schmid, C. Frieden, J. Lee (2009)
Interprotofilament interactions between Alzheimer's Aβ1–42 peptides in amyloid fibrils revealed by cryoEM
Proceedings of the National Academy of Sciences, 106
C. Russo, E. Violani, S. Salis, V. Venezia, V. Dolcini, G. Damonte, U. Benatti, C. D'Arrigo, E. Patrone, P. Carlo, G. Schettini (2002)
Pyroglutamate-modified amyloid beta-peptides--AbetaN3(pE)--strongly affect cultured neuron and astrocyte survival.
Journal of neurochemistry, 82 6
Melanie Meyer-Luehmann, Janaky Coomaraswamy, T. Bolmont, S. Kaeser, C. Schaefer, E. Kilger, Anton Neuenschwander, D. Abramowski, P. Frey, A. Jaton, J. Vigouret, P. Paganetti, D. Walsh, P. Mathews, J. Ghiso, M. Staufenbiel, L. Walker, M. Jucker (2006)
Exogenous Induction of Cerebral ß-Amyloidogenesis Is Governed by Agent and Host
Science, 313
Soo-Min Cho, Hyunjin Kim, Sejin Lee, H. Kim, Woong Kim, Tae Kim, Dong Kim, Youngsoo Kim (2014)
Correlations of amyloid-β concentrations between CSF and plasma in acute Alzheimer mouse model
Scientific Reports, 4
R. Rodriguiz, W. Wetsel (2006)
Assessments of Cognitive Deficits in Mutant Mice
Zhen Zhao, A. Sagare, Qingyi Ma, Matthew Halliday, Pan Kong, Kassandra Kisler, E. Winkler, Anita Ramanathan, T. Kanekiyo, G. Bu, Nelly Owens, Sanket Rege, Gabriel Si, Ashim Ahuja, Donghui Zhu, C. Miller, J. Schneider, Manami Maeda, Takahiro Maeda, Tohru Sugawara, J. Ichida, B. Zlokovic (2015)
Central role for PICALM in amyloid–β blood–brain barrier transcytosis and clearance
Nature neuroscience, 18
K. Yin, J. Cirrito, Ping Yan, Xiaoyan Hu, Qingli Xiao, Xiaoou Pan, R. Bateman, Haowei Song, F. Hsu, J. Turk, Jan Xu, C. Hsu, J. Mills, D. Holtzman, J. Lee (2006)
Matrix Metalloproteinases Expressed by Astrocytes Mediate Extracellular Amyloid-β Peptide Catabolism
The Journal of Neuroscience, 26
M. Kane, W. Lipinski, M. Callahan, F. Bian, R. Durham, R. Schwarz, A. Roher, L. Walker (2000)
Evidence for Seeding of β-Amyloid by Intracerebral Infusion of Alzheimer Brain Extracts in β-Amyloid Precursor Protein-Transgenic Mice
The Journal of Neuroscience, 20
Catriona McLean, Catriona McLean, R. Cherny, R. Cherny, Fiona Fraser, Fiona Fraser, S. Fuller, S. Fuller, Margaret Smith, Konrad Vbeyreuther, A. Bush, A. Bush, A. Bush, C. Masters, Colin Masters (1999)
Soluble pool of Aβ amyloid as a determinant of severity of neurodegeneration in Alzheimer's disease
Annals of Neurology, 46
R. Deane, R. Bell, A. Sagare, B. Zlokovic (2009)
Clearance of amyloid-beta peptide across the blood-brain barrier: implication for therapies in Alzheimer's disease.
CNS & neurological disorders drug targets, 8 1
D. Scheuner, C. Eckman, C. Eckman, M. Jensen, X. Song, M. Citron, N. Suzuki, T. Bird, J. Hardy, M. Hutton, W. Kukull, Eric Larson, E. Levy-Lahad, M. Viitanen, E. Peskind, P. Poorkaj, G. Schellenberg, R. Tanzi, W. Wasco, L. Lannfelt, D. Selkoe, S. Younkin (1996)
Secreted amyloid β–protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease
Nature Medicine, 2
J. Jankowsky, J. Jankowsky, Daniel Fadale, Jeffrey Anderson, Guilian Xu, V. Gonzales, N. Jenkins, N. Copeland, Michael Lee, L. Younkin, S. Wagner, S. Younkin, D. Borchelt (2004)
Mutant presenilins specifically elevate the levels of the 42 residue β-amyloid peptide in vivo: evidence for augmentation of a 42-specific γ secretase
Human Molecular Genetics, 13
R. Deane, R. Bell, A. Sagare, B. Zlokovic (2009)
Clearance of amyloid-β peptide across the blood-brain barrier: Implication for therapies in Alzheimer’s disease
Cns & Neurological Disorders-drug Targets, 8
S. Deb, P. Gottschall (1996)
Increased Production of Matrix Metalloproteinases in Enriched Astrocyte and Mixed Hippocampal Cultures Treated with β‐Amyloid Peptides
Journal of Neurochemistry, 66
I. Kurochkin, S. Goto (1994)
Alzheimer's β‐amyloid peptide specifically interacts with and is degraded by insulin degrading enzyme
FEBS Letters, 345
Antonios Tsolis, N. Papandreou, V. Iconomidou, S. Hamodrakas (2013)
A Consensus Method for the Prediction of ‘Aggregation-Prone’ Peptides in Globular Proteins
PLoS ONE, 8
J. Schindelin, Ignacio Arganda-Carreras, E. Frise, V. Kaynig, M. Longair, T. Pietzsch, S. Preibisch, C. Rueden, S. Saalfeld, Benjamin Schmid, J. Tinevez, D. White, V. Hartenstein, K. Eliceiri, P. Tomančák, A. Cardona (2012)
Fiji: an open-source platform for biological-image analysis
Nature Methods, 9
Ian Walsh, F. Seno, Silvio Tosatto, A. Trovato (2014)
PASTA 2.0: an improved server for protein aggregation prediction
Nucleic Acids Research, 42
M. Biancalana, S. Koide (2010)
Molecular mechanism of Thioflavin-T binding to amyloid fibrils.
Biochimica et biophysica acta, 1804 7
D. Games, David Adams, Ree Alessandrini, R. Barbour, Patricia Borthelette, C. Blackwell, Tony Carr, J. Clemens, T. Donaldson, F. Gillespie, T. Guido, S. Hagopian, K. Johnson-wood, K. Khan, Mike Lee, P. Leibowitz, I. Lieberburg, S. Little, E. Masliah, L. McConlogue, M. Montoya-Zavala, L. Mucke, L. Paganini, E. Penniman, M. Power, D. Schenk, P. Seubert, B. Snyder, F. Soriano, H. Tan, J. Vitale, S. Wadsworth, B. Wolozin, Jun Zhao (1995)
Alzheimer-type neuropathology in transgenic mice overexpressing V717F β-amyloid precursor protein
Nature, 373
Yilin Yan, Chunyu Wang (2006)
Aβ42 is More Rigid than Aβ40 at the C Terminus: Implications for Aβ Aggregation and Toxicity
Journal of Molecular Biology, 364
S. Ma, N. Tang, L. Lam, H. Chiu (2004)
Association between tumor necrosis factor-alpha promoter polymorphism and Alzheimer's disease.
Neurology, 62 2
H. Fillit, W. Ding, L. Buée, J. Kalman, L. Altstiel, B. Lawlor, G. Wolf-Klein (1991)
Elevated circulating tumor necrosis factor levels in Alzheimer's disease
Neuroscience Letters, 129
J. Backstrom, G. Lim, M. Cullen, Z. Tökés (1996)
Matrix Metalloproteinase-9 (MMP-9) Is Synthesized in Neurons of the Human Hippocampus and Is Capable of Degrading the Amyloid-β Peptide (1–40)
The Journal of Neuroscience, 16
Thorsten Lührs, C. Ritter, M. Adrian, Dominique Riek-Loher, B. Bohrmann, H. Döbeli, D. Schubert, R. Riek (2005)
3D structure of Alzheimer's amyloid-beta(1-42) fibrils.
Proceedings of the National Academy of Sciences of the United States of America, 102 48
D. Wesson, José Morales‐Corraliza, Matthew Mazzella, D. Wilson, P. Mathews (2013)
Chronic anti-murine Aβ immunization preserves odor guided behaviors in an Alzheimer's β-amyloidosis model
Behavioural Brain Research, 237
T. Wyss-Coray, J. Rogers (2012)
Inflammation in Alzheimer disease-a brief review of the basic science and clinical literature.
Cold Spring Harbor perspectives in medicine, 2 1
A. Paravastu, R. Leapman, Wai‐Ming Yau, R. Tycko (2008)
Molecular structural basis for polymorphism in Alzheimer's β-amyloid fibrils
Proceedings of the National Academy of Sciences, 105
M. Janelsins, M. Mastrangelo, S. Oddo, F. LaFerla, H. Federoff, W. Bowers (2005)
Early correlation of microglial activation with enhanced tumor necrosis factor-alpha and monocyte chemoattractant protein-1 expression specifically within the entorhinal cortex of triple transgenic Alzheimer's disease mice
Journal of Neuroinflammation, 2
P. Edison, H. Archer, A. Gerhard, R. Hinz, N. Pavese, F. Turkheimer, A. Hammers, Yen Tai, Nick Fox, A. Kennedy, M. Rossor, D. Brooks (2008)
Microglia, amyloid, and cognition in Alzheimer's disease: An [11C](R)PK11195-PET and [11C]PIB-PET study
Neurobiology of Disease, 32
I. Kuperstein, K. Broersen, I. Benilova, J. Rozenski, W. Jonckheere, Maja Debulpaep, A. Vandersteen, I. Segers-Nolten, K. Werf, V. Subramaniam, D. Braeken, G. Callewaert, C. Bartic, R. D'Hooge, I. Martins, F. Rousseau, J. Schymkowitz, B. Strooper (2010)
Neurotoxicity of Alzheimer's disease Aβ peptides is induced by small changes in the Aβ42 to Aβ40 ratio
The EMBO Journal, 29
Yilin Yan, Chunyu Wang (2006)
Abeta42 is more rigid than Abeta40 at the C terminus: implications for Abeta aggregation and toxicity.
Journal of molecular biology, 364 5
C. Chambon, N. Wegener, A. Gravius, W. Danysz (2011)
Behavioural and cellular effects of exogenous amyloid-β peptides in rodents
Behavioural Brain Research, 225
Qinghai Zhang, E. Powers, J. Nieva, Mary Huff, Maria Dendle, J. Bieschke, C. Glabe, A. Eschenmoser, P. Wentworth, R. Lerner, J. Kelly (2004)
Metabolite-initiated protein misfolding may trigger Alzheimer's disease.
Proceedings of the National Academy of Sciences of the United States of America, 101 14
D. Schlenzig, S. Manhart, Yeliz Cinar, M. Kleinschmidt, G. Hause, D. Willbold, S. Funke, S. Schilling, H. Demuth (2009)
Pyroglutamate formation influences solubility and amyloidogenicity of amyloid peptides.
Biochemistry, 48 29
S. Howell, J. Nalbantoglu, P. Crine (1995)
Neutral endopeptidase can hydrolyze beta-amyloid(1-40) but shows no effect on beta-amyloid precursor protein metabolism.
Peptides, 16 4
R. Deane, Zhenhua Wu, A. Sagare, Judianne Davis, S. Yan, Katie Hamm, Feng Xu, Margaret Parisi, Barbra LaRue, H. Hu, P. Spijkers, Huang Guo, Xiaomei Song, P. Lenting, W. Nostrand, B. Zlokovic (2004)
LRP/amyloid beta-peptide interaction mediates differential brain efflux of Abeta isoforms.
Neuron, 43 3
L. Prut, C. Belzung (2003)
The open field as a paradigm to measure the effects of drugs on anxiety-like behaviors: a review.
European journal of pharmacology, 463 1-3
E. Hubin, E. Hubin, N. Nuland, K. Broersen, Kris Pauwels (2014)
Transient dynamics of Aβ contribute to toxicity in Alzheimer’s disease
Cellular and Molecular Life Sciences, 71
M. Heneka, D. Golenbock, E. Latz (2015)
Innate immunity in Alzheimer's disease
Nature Immunology, 16
Emilie Cerf, Rabia Sarroukh, Shiori Tamamizu‐Kato, L. Breydo, Sylvie Derclaye, Y. Dufrêne, V. Narayanaswami, E. Goormaghtigh, J. Ruysschaert, V. Raussens (2009)
Antiparallel beta-sheet: a signature structure of the oligomeric amyloid beta-peptide.
The Biochemical journal, 421 3
J. Hardy, D. Selkoe (2009)
The Amyloid Hypothesis of Alzheimer ’ s Disease : Progress and Problems on the Road to Therapeutics
RM Rodriguiz, WC Wetsel (2006)
Animal Models of Cognitive Impairment
E. Karran, M. Mercken, B. Strooper (2011)
The amyloid cascade hypothesis for Alzheimer's disease: an appraisal for the development of therapeutics
Nature Reviews Drug Discovery, 10
D. Scheuner, C. Eckman, M. Jensen, X. Song, M. Citron, N. Suzuki, T. Bird, J. Hardy, M. Hutton, W. Kukull, E. Larson, E. Levy-Lahad, M. Viitanen, E. Peskind, P. Poorkaj, G. Schellenberg, R. Tanzi, W. Wasco, L. Lannfelt, D. Selkoe, S. Younkin (1996)
Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease.
Nature medicine, 2 8
H. Levine (1993)
Thioflavine T interaction with synthetic Alzheimer's disease β‐amyloid peptides: Detection of amyloid aggregation in solution
Protein Science, 2
A. Alvarez, R. Cacabelos, Carolina Sanpedro, M. García-Fantini, M. Aleixandre (2007)
Serum TNF-alpha levels are increased and correlate negatively with free IGF-I in Alzheimer disease
Neurobiology of Aging, 28
T. Oberstein, P. Spitzer, H. Klafki, Philipp Linning, Florian Neff, H. Knölker, P. Lewczuk, J. Wiltfang, J. Kornhuber, J. Maler (2015)
Astrocytes and microglia but not neurons preferentially generate N-terminally truncated Aβ peptides
Neurobiology of Disease, 73
C. Kliethermes, K. Cronise, J. Crabbe (2004)
Anxiety-like behavior in mice in two apparatuses during withdrawal from chronic ethanol vapor inhalation.
Alcoholism, clinical and experimental research, 28 7
Thorsten Lührs, C. Ritter, M. Adrian, Dominique Riek-Loher, B. Bohrmann, H. Döbeli, D. Schubert, R. Riek (2005)
3D structure of Alzheimer's amyloid-β(1–42) fibrils
Proceedings of the National Academy of Sciences of the United States of America, 102
Mahiuddin Ahmed, Judianne Davis, Darryl Aucoin, Takeshi Sato, S. Ahuja, S. Aimoto, J. Elliott, W. Nostrand, Steven Smith (2010)
Structural conversion of neurotoxic amyloid-β(1–42) oligomers to fibrils
Nature structural & molecular biology, 17
S. Frautschy, David Horn, J. Sigel, M. Harris-White, John Mendoza, Fusheng Yang, T. Saido, G. Cole (1998)
Protease Inhibitor Coinfusion with Amyloid β-Protein Results in Enhanced Deposition and Toxicity in Rat Brain
The Journal of Neuroscience, 18
P. Faucher, N. Mons, J. Micheau, C. Louis, D. Béracochéa (2016)
Hippocampal Injections of Oligomeric Amyloid β-peptide (1–42) Induce Selective Working Memory Deficits and Long-lasting Alterations of ERK Signaling Pathway
Frontiers in Aging Neuroscience, 7
R. Khurana, Chris Coleman, C. Ionescu-Zanetti, S. Carter, V. Krishna, R. Grover, R. Roy, Shashi Singh (2005)
Mechanism of thioflavin T binding to amyloid fibrils.
Journal of structural biology, 151 3
G. Bitan, M. Kirkitadze, A. Lomakin, Sabrina Vollers, G. Benedek, D. Teplow (2002)
Amyloid β-protein (Aβ) assembly: Aβ40 and Aβ42 oligomerize through distinct pathways
Proceedings of the National Academy of Sciences of the United States of America, 100
E. Mikros, D. Benaki, E. Humpfer, M. Spraul, S. Loukas, C. Stassinopoulou, M. Pelecanou (2001)
High-Resolution NMR Spectroscopy of the beta-Amyloid(1-28) Fibril Typical for Alzheimer's Disease.
Angewandte Chemie, 40 19
T. Saido, M. Leissring (2012)
Proteolytic degradation of amyloid β-protein.
Cold Spring Harbor perspectives in medicine, 2 6
J. Sheng, S. Bora, Guilian Xu, D. Borchelt, D. Price, V. Koliatsos (2003)
Lipopolysaccharide-induced-neuroinflammation increases intracellular accumulation of amyloid precursor protein and amyloid β peptide in APPswe transgenic mice
Neurobiology of Disease, 14
F. Librizzi, R. Carrotta, Dario Spigolon, D. Bulone, P. Biagio (2014)
α-Casein Inhibits Insulin Amyloid Formation by Preventing the Onset of Secondary Nucleation Processes.
The journal of physical chemistry letters, 5 17
M. Belli, M. Ramazzotti, F. Chiti (2011)
Prediction of amyloid aggregation in vivo
EMBO reports, 12
K. Bates, G. Verdile, G. Verdile, G. Verdile, Qian Li, Qian Li, D. Ames, P. Hudson, C. Masters, C. Masters, R. Martins, R. Martins, R. Martins (2009)
Clearance mechanisms of Alzheimer's amyloid-β peptide: implications for therapeutic design and diagnostic tests
Molecular Psychiatry, 14
Sathish Kumar, Sandesh Singh, Désirée Hinze, Michaele Josten, H. Sahl, M. Siepmann, J. Walter (2012)
Phosphorylation of Amyloid-β Peptide at Serine 8 Attenuates Its Clearance via Insulin-degrading and Angiotensin-converting Enzymes*
The Journal of Biological Chemistry, 287
R. Terry, E. Masliah, D. Salmon, N. Butters, R. DeTeresa, Robert Hill, L. Hansen, R. Katzman (1991)
Physical basis of cognitive alterations in alzheimer's disease: Synapse loss is the major correlate of cognitive impairment
Annals of Neurology, 30
S. Mancini, Stefania Minniti, Maria Gregori, G. Sancini, A. Cagnotto, P. Couraud, L. Ordóñez-Gutiérrez, F. Wandosell, M. Salmona, F. Re (2016)
The hunt for brain Aβ oligomers by peripherally circulating multi-functional nanoparticles: Potential therapeutic approach for Alzheimer disease.
Nanomedicine : nanotechnology, biology, and medicine, 12 1
D. Wilcock, M. Gordon, D. Morgan (2006)
Quantification of cerebral amyloid angiopathy and parenchymal amyloid plaques with Congo red histochemical stain
Nature Protocols, 1
A. Serrano‐Pozo, M. Frosch, E. Masliah, B. Hyman (2011)
Neuropathological alterations in Alzheimer disease.
Cold Spring Harbor perspectives in medicine, 1 1
E. Castaño, F. Prelli, C. Soto, R. Beavis, E. Matsubara, M. Shoji, B. Frangione (1996)
The length of Amyloid-β in Hereditary Cerebral Hemorrhage with Amyloidosis, Dutch Type
The Journal of Biological Chemistry, 271
Jessica Nasica-Labouze, P. Nguyen, F. Sterpone, O. Berthoumieu, Nicolae-Viorel Buchete, S. Côté, A. Simone, A. Doig, P. Faller, Angel Garcia, A. Laio, M. Li, S. Melchionna, N. Mousseau, Y. Mu, A. Paravastu, S. Pasquali, David Rosenman, B. Strodel, B. Tarus, J. Viles, Tong Zhang, Chunyu Wang, P. Derreumaux (2015)
Amyloid β Protein and Alzheimer's Disease: When Computer Simulations Complement Experimental Studies.
Chemical reviews, 115 9
C. Durán-Aniotz, R. Morales, I. Moreno-Gonzalez, P. Hu, C. Soto (2013)
Brains from non-Alzheimer’s individuals containing amyloid deposits accelerate Aβ deposition in vivo
Acta Neuropathologica Communications, 1
S. Ma, N. Tang, L. Lam, H. Chiu (2004)
Association between tumor necrosis factor-α promoter polymorphism and Alzheimer’s disease
Neurology, 62
E. Mikros, D. Benaki, E. Humpfer, M. Spraul, S. Loukas, C. Stassinopoulou, M. Pelecanou (2001)
High‐Resolution NMR Spectroscopy of the β‐Amyloid(1–28) Fibril Typical for Alzheimer's Disease
Angewandte Chemie, 40
C. Jack, H. Wiste, S. Weigand, W. Rocca, D. Knopman, M. Mielke, V. Lowe, M. Senjem, J. Gunter, G. Preboske, V. Pankratz, P. Vemuri, R. Petersen (2014)
Age-specific population frequencies of cerebral β-amyloidosis and neurodegeneration among people with normal cognitive function aged 50–89 years: a cross-sectional study
The Lancet Neurology, 13
V. Cecarini, L. Bonfili, M. Cuccioloni, M. Mozzicafreddo, G. Rossi, L. Buizza, D. Uberti, M. Angeletti, A. Eleuteri (2012)
Crosstalk between the ubiquitin-proteasome system and autophagy in a human cellular model of Alzheimer's disease.
Biochimica et biophysica acta, 1822 11
S. Howell, J. Nalbantoglu, P. Crine (1995)
Neutral endopeptidase can hydrolyze β-amyloid(1–40) but shows no effect on β-amyloid precursor protein metabolism
Peptides, 16

Publisher: Springer Journals
Copyright: Copyright © 2016 by The Author(s).
Subject: Biomedicine; Neurosciences; Pathology; Neurology
eISSN: 2051-5960
DOI: 10.1186/s40478-016-0381-9
pmid: 27724899
Publisher site: See Article on Publisher Site

Abstract

Substantial data indicate that amyloid-β (Aβ), the major component of senile plaques, plays a central role in Alzheimer’s Disease and indeed the assembly of naturally occurring amyloid peptides into cytotoxic aggregates is linked to the disease pathogenesis. Although Aβ42 is a highly aggregating form of Aβ, the co-occurrence of shorter Aβ peptides might affect the aggregation potential of the Aβ pool. In this study we aimed to assess whether the structural behavior of human Aβ42 peptide inside the brain is influenced by the concomitant presence of N-terminal fragments produced by the proteolytic activity of glial cells. We show that the occurrence of the human C-terminal truncated 1–24 Aβ fragment impairs Aβ42 clearance through blood brain barrier and promotes the formation of Aβ42 aggregates even in the healthy brain. By showing that Aβ1-24 has seeding properties for aggregate formation in intracranially injected wild type mice, our study provide the proof-of-concept that peptides produced upon Aβ42 cleavage by activated glial cells may cause phenotypic defects even in the absence of genetic mutations associated with Alzheimer’s Disease, possibly contributing to the development of the sporadic form of the pathology. Keywords: Amyloid-β,Alzheimer’s disease, Microglia, Proteolytic activity, Aβ24 Introduction shown to represent an important factor in initializing Aβ Alzheimer’s disease (AD) is a protein misfolding pathology, fibrillogenesis and toxicity [7], indicating that the presence caused by accumulation of abnormally folded Aβ and tau of different Aβ forms may affect the development of AD polypeptides, which form amyloid plaques and neurofibril- in vivo. lary tangles in the brain of affected individuals. Aβ Consistently, while small amounts of Aβ-containing aggregates have been linked with learning and memory brain extracts, deriving from either AD patient or AD deficits in both human and mouse models of the disease, transgenic mouse, induce β-amyloidosis and glial activa- making Aβ deposition a target for prevention and treat- tion once intracranially injected in pre-depositing AD ment [1–3]. In the last years, a lot of effort has been transgenic mice [8–11], the chronic infusion of soluble, focused on the identification of the processes leading to Aβ synthetic Aβ42 peptides into wild type (wt) rodent aggregation. Evidence have indicated that, although Aβ42 is brains does not result in amyloid deposition [9]. The a highly aggregating form of Aβ [4, 5], the co-occurrence of finding that Aβ42 alone fails to show seeding properties Aβ peptides with different length can affect the neurotoxic in the healthy brain and does not trigger pathogenetic and aggregation potential of the Aβ pool (reviewed in [6]). pathways indicates the occurrence of efficient clearance As an example, changes in the ratio of Aβ40/42 has been mechanisms and suggests that brain-specific cofactors, specifically present in pathological conditions, are needed for effective seeding [9]. * Correspondence: m.matteoli@in.cnr.it; Although aggregation properties of full length Aβ42 michela.matteoli@humanitasresearch.it have been deeply explored both in vitro and in vivo, Equal contributors IRCCS Humanitas, via Manzoni 56, 20089 Rozzano, Italy much less is known about the in vivo aggregating prop- IN-CNR, via Vanvitelli 32, 20129 Milano, Italy erties of shorter Aβ fragments. This aspect may be Full list of author information is available at the end of the article © 2016 The Author(s). Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 2 of 19 particularly relevant, considering that N-terminal fragments detectable by the amyloidogenic dye Congo red both 2.5 of different length are largely produced by the proteo- and 6 months after injection (Fig. 1c and f, g). Consist- lytic activity of glial cells during the development of ently, the highest number of Congo red positive aggre- AD [12]. Indeed, several proteases including neprilysin gates was detected in the brains of mice injected with [13], insulin-degrading enzyme [14], endothelin-converting H-Aβ42/H-Aβ24 mix (Fig. 1f and g). enzyme [15], angiotensin-converting enzyme [16] and As a further confirmation for the formation of amyloi- matrix metalloproteinase-9 (MMP9) [17–20] have been dogenic aggregates, brain sections from mice i.c. injected shown to degrade soluble Aβ in vitro, acting at specific and incubated for 2.5 months were stained with thiofla- cleavage sites and generating characteristic Aβ fragments. vin T (ThT), a benzothiazole dye that exhibits enhanced These proteolytic activities, therefore, are critical in deter- fluorescence upon binding to amyloid fibrils [27, 28]. mining the quantitative and qualitative pattern of cerebral ThT-positive aggregates were detected in wt mice brain Aβ levels ([21]; reviewed in [22]). MMP9 in particular, injected with H-Aβ24 or with H-Aβ42/H-Aβ24 mix which generates different C-terminal truncated Aβ frag- (Fig. 2a and c); both ThT- (Fig. 2a) and 6E10- (Fig. 2b) ments, including 1–16, 1–20, 1–23, 1–30, 1–33 and 1–34 positive aggregates were surrounded by microglia, as re- [19, 23], is thought to play relevant roles in different patho- vealed by Iba1 staining (Fig. 2a and b, red staining). logical contexts, as suggested by the observation that its ex- Again, no ThT positive aggregates were detected in pression can be stimulated by diverse insults, including Aβ vehicle-injected wt mice, whereas only ThT-positive itself [24], and is up-regulated in glia cells adjacent to amyl- blood vessels, which are not surrounded by microglia, oid deposits [25]. were visible in wt mice injected with H-Aβ42 (Fig. 2a, Since Aβ N-terminal fragments of varying lengths are arrows). The recruitment of microglia around aggregates expected to exhibit different physico-chemical properties was reminiscent of the plaques present in APP/PS1 trans- which may result in different aggregation behaviors, as genic mice brain (Fig. 2b, right). Consistent with data also indicated by modeling of aggregation determinants reported in AD patients and mice models [29–33], a with bioinformatics methods [26], we aimed to investigate significant increase in the levels of tumor necrosis factor- whether dynamics of interaction and structural behavior alpha (TNF-alpha) was detected by ELISA in the serum of human Aβ42 peptide inside the brain are influenced by and brains of mice i.c. injected with H-Aβ24 and of H- the concomitant presence of C-terminal truncated frag- Aβ42/H-Aβ24 mix, and much less with H-Aβ42 alone ments. We took advantage of the commercially available (Fig. 2d and e). These data indicate that H-Aβ24 and, even synthetic human Aβ1-24 peptide (referred to as Aβ24), a more prominently, H-Aβ42/H-Aβ24 mix induce an C-terminal truncated Aβ fragment overlapping with inflammatory reaction in the brain of injected mice. MMP9 cleavage products (residues 1–20 and 1–23, [20]) Interestingly, 6 months after i.c. injection, few scattered and coincides with a turn region between β-sheets in 6E10-positive spots also start to become detectable in the recently-resolved fibrillar structures. Our results indicate non-injected side (Fig. 3a and b), suggesting a possible that the presence of Aβ24 in intracranially injected wild spreading of misfolded/aggregating Aβ. Also, amyloid ag- type mice impairs Aβ42 clearance and promotes forma- gregates were detectable at the hippocampal level in the tion of Aβ42 aggregates even in the healthy brain. injected hemisphere (Fig. 3c). Results Behavioral deficits in wt mice injected with Aβ24 peptide Synthetic Aβ24 fragments promote aggregates formation To investigate whether the formation of Aβ aggregates in wt mice brain and the increase in TNF-alpha were accompanied by the Three month-old wt mice were intracranially (i.c.) occurrence of cognitive defects [34–36], mice were ana- injected with either the oligomeric form of the single H- lysed using 3 different behavioral tests: the open field, a Aβ42 peptide or with an equimolar mixture of oligo- recognized paradigm for assessing motor activity and meric H-Aβ42 and H-Aβ24 peptides, and the brains anxiety-like behaviors in response to a novel environ- were examined after 2.5 or 6 months (see cartoon, ment [37–39]; the sociality task, which also unveils anx- Fig. 1h). As previously described [9], H-Aβ42 injected in ious and aggressive behaviors; and the novel object the brain of wt mice and examined 6 months later did recognition (NOR), that, monitoring the time spent by not cause Aβ deposition. H-Aβ24, and even more po- mice to explore a novel object, enables the assessment tently H-Aβ42/H-Aβ24 mix, induced the formation of of possible declines in learning and memory. Consist- aggregates, detected with the Aβ N-terminal specific ently with literature data [40], 6 months old APP/PS1 antibody 6E10, followed by HRP (Fig. 1a and e). 6E10- mice displayed significant hyperactivity and anxiety com- positive aggregates were already detectable 2.5 months pared to the wt littermates, as shown by both the open after injection of H-Aβ24 or H-Aβ42/H-Aβ24 mix in the field and sociality tasks (Fig. 4a–f). Interestingly, wt mice mice brains (Fig. 1b and d). Aggregates were also injected with the mix of the two peptides displayed Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 3 of 19 Fig. 1 (See legend on next page.) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 4 of 19 (See figure on previous page.) Fig. 1 Intracranial injection of H-Aβ24 and H-Aβ42/H-Aβ24 mix in wt mice induces amyloid aggregate deposition. a 6E10 DAB staining of hippocampal brain sections 6 months after vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix injection. Sagittal plane sections are shown on the left; middle and right panels show enlargements of the dorsal cerebral cortex. 6E10-immunopositive aggregates are visible on a cresyl violet-luxol fast blue counterstaining. Scale bars: left 1 mm, middle 100 μm, right 50 μm. b 6E10 DAB staining of brain sections 2.5 months after vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix injection. Scale bar: 50 μm. c Congo red staining of brain sections 2.5 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. On the right, representative images of aggregates detected 6 months after H-Aβ42/H-Aβ24 mix injection are shown. Top: bright field images; bottom: epifluorescence images -FITC filter. Scale bar: 50 μm. d and e Quantification of 6E10 DAB positive plaques 2.5 or 6 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H- Aβ42/H-Aβ24 mix. 6 brain sections were analyzed for each mice (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (****P < 0.0001; ***P < 0.001). f and g Quantification of Congo red-positive plaques area (μm ) per brain section at 2.5 or 6 month after injection. 10 sections of 50 μm thickness per slice were analyzed for each mouse brain (N = 6 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01). h Cartoon depicting the experimental scheme Fig. 2 Intracranial injection of H-Aβ24 and H-Aβ42/H-Aβ24 mix in wt mice induces microglia recruitment and TNF-alpha production. a Immunofluorescence staining of ThT-positive aggregates (green) surrounded by Iba1 positive microglia cells (red) in wt mice brain slice 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Arrows in H-Aβ42 panel indicate Aβ accumulation in vessels. Scale bar: 10 μm. b IF staining of 6E10 positive aggregates (blue) surrounded by Iba1 positive microglia cells (red) in wt mice brains 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. IF staining of 6E10 positive plaques and surrounding microglia are shown in a brain section of 8 months old APP/PS1 transgenic mouse for comparison. Scale bar: 10 μm. c Quantification of ThT-positive plaques in sections of wt mice brains 2.5 months after injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Aβ detected in correspondence of blood vessels (H-Aβ42 left panel, arrows) was excluded from the analysis. 6 sections were analyzed for each mouse brain (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (*** P < 0.001). d and e TNF-alpha levels measured by ELISA in brain homogenates d and serum e of wt mice 2.5 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/H-Aβ24 mix. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (*** P <0.001; ** P <0. 01) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 5 of 19 Fig. 3 Intracranial injection of H-Aβ42/H-Aβ24 mix in wt mice induces spreading of misfolded Aβ. a 6E10 DAB staining of hippocampal brain sections 6 months after H-Aβ42/H-Aβ24 mix injection. Sagittal plane sections are shown on the left; middle and right panels show enlargements of cerebral cortex; the injected hemisphere is referred as ipsilateral (top panel) and the non-injected one as contralateral (bottom panel). 6E10- immunopositive aggregates are visible on a cresyl violet-luxol fast blue counterstaining. Scale bars: left 1 mm, middle 100 μm, right 50 μm. b Quantification of 6E10 DAB positive plaques in the contralateral hemisphere 6 months after the injection of vehicle, H-Aβ42, H-Aβ24 or H-Aβ42/ H-Aβ24 mix. 6 brain sections were analyzed for each mouse (N = 3 mice for each group). Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**** P < 0.0001, *** P < 0.001). c Representative images of 6E10 DAB positive aggregates at the hippocampal level 6 months after Aβ injection. Scale bar: left and middle 100 μm, right 50 μm abnormal behavioral phenotypes very similarly to age occurrence of cognitive defects comparable to APP/PS1 matched APP/PS1 mice, both in terms of enhanced mice. motor activity, as indicated by the total distance traveled (Fig. 4a), and in terms of increased anxiety levels, as re- Predicted physico-chemical properties of Aβ24 aggregation vealed by the longer time spent in the periphery (Fig. 4b) and cross-aggregation and the shorter time spent in the center (Fig. 4c) of the We next aimed at assessing the molecular basis of aggre- arena. This result was also confirmed by the sociality gate formation in the brains injected with H-Aβ24 or task in which APP/PS1 and wt Aβ-injected mice had a with the H-Aβ42/H-Aβ24 mix. In order to understand higher number of contact and spent less time with the whether H-Aβ24 is endowed with different propensity to second animal (Fig. 4d–f). In the NOR test, APP/PS1 form aggregates relative to H-Aβ42, we modeled the de- mice as well as wt mice injected with the mix of the two terminants of aggregation with bioinformatics methods, peptides spent less time than wt littermates exploring using different algorithms which predict the aggregation the novel object at the 1 h delay test and more time ex- propensity on the basis of sequences [26]. First, the ploring the novel object at the 24 h delay test, indicating AMYLPRED2 consensus analysis identified two distinct a significant defect in learning and memory (Fig. 4g and amyloidogenic regions in the full Aβ42 peptide, respect- h). Although not displaying brain aggregates, some be- ively encompassing residues 15–22 and 29–42. Of the havioral alterations were detected in mice intracranially two regions, the first was predicted in Aβ24 as well injected with oligomeric H-Aβ42, consistent with litera- (Table 1). This is an indication that Aβ24 maintains ture data [41, 42]. Conversely, no differences were fibrillogenic potential in isolation [43], although the noticed between the not injected and vehicle injected wt resulting fibrillar structure will likely be different due to mice, showing that the injection per se was not respon- the lack of second aggregation-prone region. sible for the observed behavior defects. These data Second, the PASTA 2.0 algorithm by Walsh et al. [44] indicate that injection of H-Aβ24 and, even more prom- predicted the strongest self-aggregating segment in inently, H-Aβ42/H-Aβ24 mix into wt brain results in the Aβ42 to be residues 31–41 (at about−10.6 kcal/mol, Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 6 of 19 Fig. 4 Intracranial injection of H-Aβ induces behavioral defects in wt mice. a–c Mice hyperlocomotion, assessed by Open Field test. Left panel a total distance (cm) travelled by mice in 30 min using 5 min time bin representation. b, c panels: total distance travelled in 30 min in the periphery or in the center of the arena using 5 min time bin representation. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test. Asterisks are referred to the vehicle vs H-Aβ42/H-Aβ24 mix condition. No significant differences are present between APP/PS1 tested mice and the H- Aβ42/H-Aβ24 mix condition. (****P < 0.0001; ***P < 0.001; **P <0.01). d–f Social interaction, assessed using the social free test. Graphic representation of the number of contacts/min between the two mice in the same arena. H-Aβ42 d,H-Aβ24 e or H-Aβ42/H-Aβ24 mix f are compared to vehicle injected mice and APP/PS1 mice. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01; *P < 0.05). g and h Assessment of learning and memory by NOR. g Number of contacts with the new and old object either after 1 h and 24 h of delay. h Number of contacts with the new object both after 1 h and 24 h of delay. Statistical analysis was performed by One way Anova, Bonferroni multiple comparison test. Asterisks are referred to the control wt vs APP/PS1 mice when specified or to the vehicle vs other conditions (****P < 0.0001; **P <0.01; *P < 0.05). A total of 33 mice (6 mice foreachexperimentalgroup plus 6APP/PS1and 3 wt mice6monthsold) wereanalysed (Fig. 5a), thus estimating Aβ24 to be both a weaker self- Table 1 Amyloigenic regions in Aβ24 and Aβ42 peptides predicted binder (by approximately 5 kcal/mol) and to have a by the 11 indicated methods, and AMYLPRED2-derived consensus comparatively more marked, although still not domin- score ant, tendency towards the self-assembly in the antiparal- Method Aβ24 Aβ42 lel arrangement (Fig. 5b). It is worth noting that the AGGRESCAN 17–22 17–22, 30–42 algorithm also predicts cross-aggregation propensities AmyloidMutants 4–12, 15–23 14–22, 34–42 between segments in the 1–24 and 25–42 ranges, the strongest pair being at−4.84 kcal/mol, which is in the Amyloidogenic Pattern 16–21 16–21 same order of magnitude of the self-aggregation propen- Average Packing Density 16–21 16–21, 32–37 sities in the 1–24 region (Fig. 5c). Beta-strand contiguity 15–20 15–20, 29–41 Hexapeptide Conf. Energy 16–22 16–22, 29–42 Aβ24 displays a low fibrillar aggregation propensity and NetCSSP 1–23 1–23, 28–37 supports the formation of intermolecular β-sheets Pafig 7–24 7–42 The in vitro aggregation of Aβ samples was monitored by ThT assay [45] (Fig. 6a–c). ThT fluorescence emission was SecStr 15–20 15–20 recorded at different incubation times at 37 °C in physio- TANGO 17–21 17–21, 29–41 logical buffer. Under our experimental conditions, and in WALTZ 15–24 15–23, 28–42 agreement with previously reported data [46], oligomeric AMYLPRED2 (CONSENSUS5) 15–22 15–22, 29–42 H-Aβ42 rapidly formed ThT-positive aggregates already Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 7 of 19 Fig. 5 Bioinformatics analysis of Aβ42, Aβ24 and cross-aggregating regions and their stabilities. Regions and energies are predicted by Walsh’sPASTA 2.0 algorithm; lower energies indicate higher fibrillation propensity. a Self-aggregation propensities for the full length Aβ42 protein; b self-aggregation propensities for the Aβ24 fragment. c Cross-aggregation between the Aβ24 fragment and the C-terminal region (aa 25 to 42) of Aβ42 Fig. 6 In vitro aggregation of Aβ peptides studied by ThT fluorescence and infrared spectroscopy. a ThetimecourseofaggregationofAβ42, of Aβ24, and of the equimolar mixture of the two peptides incubated at 37 °C in PBS was monitored by ThT fluorescence with excitation and emission wavelengths at 450 nm and 485 nm, respectively. b ThT fluorescence emission spectra (excitation at 450 nm) of samples as in a) reported for selected incubation times. c Fluorescence emission spectra with excitation at 270 nm of Aβ42 and of Aβ24 incubated at 37 °C in PBS for 96 h. d ATR-FTIR spectra of Aβ42, of Aβ24, and H-Aβ42/H-Aβ24 mix incubated at 37 °C in PBS for different times, as indicated. Spectra are reported after Fourier self deconvolution (see Materials and Methods) Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 8 of 19 after 24 h, while oligomeric H-Aβ24 displayed very low electrical resistance (TEER) values reached from day 7 ThT fluorescence up to 96 h of incubation (Fig. 6a and b), after plating (Fig. 7f). The in vitro BBB model was used to indicating a much lower fibrillation capability than H- investigate the passage of FAM-labeled or 488-conjugated Aβ42. Further, we recorded the emission fluorescence H-Aβ42 across the endothelial cell monolayer. While the spectra of the samples incubated for 96 h at 37 °C after addition of fluorescently labeled H-Aβ42 to the apical side tyrosine excitation at 270 nm, which is also able to of cell inserts resulted in effective BBB crossing, fluores- produce ThT fluorescence [47]. The tyrosine emission at cently labeled scrambled H-Aβ42 was not efficiently trans- around 301 nm was detected for both H-Aβ24 and H- cytosed (Fig. 7g) thus confirming the reliability and the Aβ42 samples, but only the latter showed the ThT emis- selectivity of the model as described in literature [50]. sion typical of amyloid fibrils at around 485 nm (Fig. 6c). Notably, the concomitant presence of H-Aβ24 at the An intermediate behavior was observed for the equimolar apical side of the cells significantly reduced fluorescently- mixture of oligomeric H-Aβ42/H-Aβ24 peptides (Fig. 6a labeled H-Aβ42 transfer (Fig. 7g, h). No effect of H-Aβ24 and b). The secondary structures of the peptide assemblies on the diffusion of fluorescently labeled scrambled H- were also investigated by Fourier transform infrared Aβ42 was detected (Fig. 7g, h). Similar results were ob- (FTIR) spectroscopy in attenuated total reflection (ATR) tained using different Aβ42 concentrations (1 μMor 100 [48, 49] (Fig. 6d). The FTIR spectra of the H-Aβ24 sam- nM Aβ42, Fig. 7g and h). Consistently, H-Aβ42 apical-to- ples incubated at 37 °C for 20 min and for 6 days were basolateral passage, detected by dot blot analysis of characterized by a broad Amide I band, due to the C = O medium recovered from the basolateral compartment, re- stretching vibrations of the peptide bond, with two main vealed a decreased amount of H-Aβ42 when the latter was −1 −1 peaks at ~1695 cm and ~1628 cm , assigned to the for- pre-incubated with H-Aβ24 at the apical side (Fig. 7i). mation of intermolecular β-sheet structures. In compari- These data indicatethatH-Aβ24 presence results in H- son to H-Aβ24, H-Aβ42 at 20 min displayed a higher Aβ42 retention, thus reducing its efflux through the BBB −1 intensity of the ~1628 cm peak. During incubation at and therefore preventing an efficient mechanism of Aβ42 37 °C of H-Aβ42, this component increased in intensity clearance. −1 while that at ~1695 cm decreased (Fig. 6d). These spec- Given that H-Aβ24 retains H-Aβ42, thus reducing its tral changes have been already observed in the fibrillogen- clearance through the BBB, lower H-Aβ42 levels in the esis of Aβ peptides and of other proteins, and assigned to circulation are expected. For this reason, H-Aβ42 levels the conformational conversion towards fibrillar structures were quantified by ELISA assay in serum samples of wt with a parallel orientation of the intermolecular β-sheets mice 1 week after injection of amyloid species into the [48, 49]. An intermediate behavior was observed for the brain. As expected, increased Aβ42 peripheral levels equimolar H-Aβ42/H-Aβ24 mix (Fig. 6d). Therefore, the were detected in Aβ42 i.c. injected mice (Fig. 7j). Not- spectroscopic analyses indicated a low fibrillization pro- ably, a significant reduction of H-Aβ42 peripheral levels pensity for H-Aβ24 and confirmed the higher antiparallel was detected when H-Aβ24 and H-Aβ42 were injected content of this peptide, as predicted by bioinformatics together into the brain at equimolar amounts (4pmol of analysis (Fig. 5c). These data, together with the cross- H-Aβ42 or 2pmol of H-Aβ24 plus 2pmol of H-Aβ42) aggregation potential between H-Aβ1-24 and H-Aβ25-42 (Fig 7k). In order to exclude that the lower Aβ42 plasma (Fig. 5c), suggest that H-Aβ24 may promote aggregate for- content, observed after the i.c. injection of H-Aβ24 and mation involving intermolecular β-sheet interactions, H-Aβ42, could result from the reduced amount of Aβ42 which possibly retain H-Aβ42 through cross-aggregation injected (2pmol vs 4 pmol), the experiment was repeated between segments in the 1–24 and 25–42 ranges. upon injection of the same amount of Aβ42, either in as- sociation or not with Aβ24 (8pmol of H-Aβ42 or 8pmol Synthetic Aβ24 fragment impairs Aβ42 clearance in a of H-Aβ24 plus 8pmol of H-Aβ42). Reduced Aβ42 per- blood brain barrier model ipheral levels were detected also in this case when H- To directly investigate whether oligomeric H-Aβ24 forms Aβ24 was co-injected together with H-Aβ42 (Fig. 7l). aggregates retaining Aβ42 and impairing the clearance of Aβ42 through the blood brain barrier (BBB), we used an Injected H-Aβ24 aggregates with endogenously produced in vitro BBB transwell model, formed by (brain) endothe- mouse Aβ42 lial bEnd.3 cells cultivated in the abluminal compartment Considering that H-Aβ24 would reduce Aβ42 clearance of cell culture inserts until a post-confluent monolayer thus causing an increase in the levels of brain Aβ42, we had grown (Fig. 7a). Confirmation of the morphological hypothesized that deposits observed in mice injected and functional properties of the endothelial cell monolayer with only H-Aβ24 could derive from a co-aggregation of (Fig. 7b) were obtained by immunostaining for the tight the injected peptide and endogenously produced mouse junction proteins claudin5 (Fig. 7c) and ZO1 (Fig. 7d), Aβ42, retained in the brain. Consistently, immunoblot connexin43 (CX43) (Fig. 7e) and by the transendothelial analysis of brain homogenate fractions using Aβ-specific Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 9 of 19 Fig. 7 Aβ24 fragment diminishes Aβ42 clearance through the BBB. a Graphical representation of the in vitro BBB model composed by a monolayer of brain endothelial cells seeded and cultured on an inverted matrix-coated porous membrane, allowing an apical compartment (donor-“brain”-side) physically separated from the basolateral chamber (receiving-“blood”-side). b Representative images of brain endothelial cells tightly wedged together and c expressing cell type-specific tight junctional proteins claudin-5 (green) d and ZO-1 (green), and e the gap junction protein Connexin-43 (green). Nuclei counter-stained with DAPI (blue). Scale bars: 20 μm. f Barrier properties monitored in terms of gradual increase in transendothelial electrical resistance (TEER) during cell monolayer’s formation over time. g and h Apical-to-basolateral exchange across endothelial monolayer of fluorescent Aβ42 or scramble Aβ42 control peptide, 1 μM g) or 100 nM h), over 120 min in presence or absence of Aβ24 at equimolar concentration. Quantification of unidirectional trans- endothelial Aβ42 transport by fluorescence spectrophotometry n ≥ 3 experiments; statistical analysis was performed by One way Anova, followed by Bonferroni’s post hoc test for multiple comparisons (**P < 0.01; ***P < 0.001). i) Dot blot analysis of medium collected in the abluminal compartment 120 min after brain endothelial cell monolayers exposure to Aβ42 or Aβ42/Aβ24 mix. Histograms represent the densitometry quantification upon staining with anti-6E10 antibody. Results are expressed as mean values of triplicates in each experimental group ± SE. Values were normalized on control; statistical analysis was performed by unpaired T test (**P < 0.01). j Aβ42 (pg/ml) absolute values detected by ELISA in the serum of mice 1 week after H-Aβ42 intracranical injection. k Aβ42 serum level measured 1 week after the injection of vehicle, H-Aβ42 and H-Aβ42/H-Aβ24 mix (2pmol of H-Aβ24 plus 2pmol of H-Aβ42) or l H-Aβ42/H-Aβ24 mix (8pmol of H-Aβ24 plus 8pmol of H-Aβ42). N = 4 to 6 animals per experimental group. Values are normalized on vehicle. Statistical analysis was performed by unpaired T test (*P < 0.05) monoclonal antibody m3.2 (which recognizesresidues10– homeostatic conditions [52–55]. However, on the other 15 of murine Aβ [51] (Fig. 8; supernatant, top panel and hand, microglia might be detrimental exacerbating Aβ pellet, bottom panel) showed the presence of m3.2 positive deposition and causing neuronal damage through the bands (squares) in wt mice injected with H-Aβ24 and, at a production of amyloidogenic, truncated forms of Aβ lesser extent in mice injected with H-Aβ24/H-Aβ42 mix. [12, 54]. To obtain the proof-of-concept that C- This result corroborates our hypothesis by which H-Aβ24 terminal truncated fragments could be produced by injected in the brain interacts with endogenously produced microglia, we treated primary microglia cultures with mouse Aβ42. As a specificity control, brain homogenates of 488-conjugated H-Aβ42 and in line with literature, we mice lacking the amyloid precursor protein APP (App−/−) observed that already after 3 h of treatment Aβ was were negative for the m3.2 antibody (Fig. 8). efficiently internalized by the cells (Fig. 9a). We took advantage of different antibodies, which recognize dis- Microglial MMP9 is responsible for Aβ42 degradation and tinct portions of the protein to assess the formation of for the production of a C-terminal truncated Aβ fragment H-Aβ42-derived Aβ forms (Fig. 9b). Double immuno- In AD patients and in transgenic models of the disease, fluorescence using antibodies against the C-terminal microglial activation in response to Aβ is followed by Aβ (anti-Aβ42) and the N-terminal (6E10) regions of Aβ42 internalization via phagocytosis, in an attempt of restoring revealed that, 24 h after its internalization, an antigen Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 10 of 19 right). Notably, ELISA for the C-terminal domain of the protein revealed that the modifications in the nature of the extracellular amyloid fragments were prevented by microglia incubation with protease inhibitors (Fig. 9e), indicating that microglia promote the metabolic process- ing of Aβ42, thus favoring the production of fragments which include the N-terminal, but not the C-terminal domain of the Aβ peptide. The lack of changes in the amount of Aβ42 in the medium of microglia, as detected with the antibody 4G8 which recognizes aa 17–24 (Fig. 9d), suggests that the truncated fragment contains the N-terminal part of the protein up to at least aa 23– 24, residues which are known to be cleaved by MMP9 [20]. Consistently with the involvement of MMP9 in Aβ42 processing, the production of the C-terminal trun- cated fragment was inhibited in microglia cells lines (N9) exposed to specific MMP9 siRNA to reduce the en- zyme expression (Fig. 9f). The same results were ob- tained by dot blot staining. Indeed, the exposure of primary microglia to protease inhibitors (Fig. 9g and h), the knocking-down of MMP9 in N9 cells (Fig. 9g and i) −/− as well as the use of primary microglia from MMP9 mice (Fig. 9g and j) prevented the reduction in the amount of the C-terminal fragment in the medium of cells exposed to Aβ42 for 24 h. Staining with 6E10 anti- body revealed the same amount of N-terminal fragments Fig. 8 Production of murine-Aβ in H-Aβ24 and H-Aβ24/H-Aβ42 injected mice. Tris-Tricine SDS–polyacrylamide gel electrophoresis (PAGE) followed in the extracellular media of cells under the different ex- by immunoblotting using murine-Aβ-specific monoclonal antibody m3.2 perimental conditions (Fig 9g, right). To demonstrate of brain homogenate factions (supernatant, top panel and pellet, bottom that the N-terminal fragments deriving from microglia panel) of mice injected with vehicle, H-Aβ24, H-Aβ42 or H-Aβ24/H-Aβ42 proteolytic activity are responsible for a diminished mix and analyzed after 4 months. Asterisks indicate the presence of 6KDa Aβ42 clearance in vitro, we performed the same experi- band positive to m3.2 antibody only in the pellet fraction of mice injected with H-Aβ24. Actin was used as loading control ment as in Fig. 7g, by incubating the BBB model with microglia conditioned medium (previously treated with unlabeled Aβ42 for 24 h) in the presence or absence of becomes detectable inside microglia, which is select- protease inhibitors. As expected, adding microglia condi- ively recognized by the 6E10 (N-ter) but not by the tioned medium significantly reduced 488-conjugated H- anti-Aβ42 (C-ter) (Fig. 9b arrows). Aβ42 clearance across brain endothelial cell monolayers. In line with the formation of a truncated Aβ42 form, The presence of protease inhibitors in microglia medium dot blot analysis of the extracellular medium of micro- increased fluorescently labelled H-Aβ42 passage through glia exposed to H-Aβ42 revealed, after 24 h, a reduction the BBB (Fig. 9k). Overall, these results suggest that of extracellular Aβ42, when stained by the C-terminal microglia play a central role in the production of C- anti-Aβ42 (directed against the aa 38–42) but not by the terminal truncated fragments by the activity of extracel- 6E10 antibody, directed against the N-terminal domain lular proteases. This phenomenon could be responsible of Aβ42 (aa 1–16) (Fig. 9c). We then quantified Aβ42 for an enhanced Aβ42 deposition and seeding in the extracellular concentrations over time using an Aβ42 healthy brain. ELISA kit based on a capture antibody directed against the C-terminal domain of the protein and a detection Discussion and conclusions antibody against aa 11–28. Consistently, we observed a Although the significance of amyloid deposits for the significant decrease of Aβ42 in the microglia medium, pathogenesis of AD is still under debate [56–58], the obser- starting from 12 h after treatment (Fig 9d, left). On the vation that, in brain, harmful proteins show high propensity contrary, an ELISA kit based on the capture antibody to aggregate indicates that formation of deposits is import- 6E10 revealed the lack of changes in the amount of ant in the pathogenesis of brain disorders. While Aβ- Aβ42 in the medium of microglia (detection antibody containing brain extracts from AD patient or transgenic 4G8 against the central part of Aβ42, aa 17–24, Fig. 9d, mouse model have been found to induce Aβ deposition in Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 11 of 19 ab DAPI,anti-Aβ42,6E10 c d 4000 4000 H-Aβ42 ** 30 3000 3000 anti-Aβ42 20 2000 20 2000 6E10 10 1000 10 1000 0 0 0 0 hours hours hours hours ef g H-Aβ42 ctrl ctrl siRNA +prot.inhib. control 4000 +MMP9 siRNA +prot.inh. *** ** CTRLsiRNA MMP9siRNA MMP9 -/- hours hours ctrl siRNA -/- h +prot.inhib. i j MMP9 k +MMP9 siRNA ** 1.5 20 *** 1.0 0.5 0 0 0 microglia microglia +prot.inhib. hours hours hours Fig. 9 Microglia promote the formation of a C-terminal lacking amyloid fragment in vitro. a 3D reconstruction by Imaris software of microglia cells (Iba1 positive, red) treated or not for 3 h with 1 μMAβ42–488 (green). Scale bar: 4 μm. b Bright field images of microglia cells stained with 6E10 (red) or anti-Aβ42 (green) antibodies, before (left panel) or after (right panel) 24 h incubation with H-Aβ42. Arrows indicate 6E10 positive puncta not co-localizing with anti-Aβ42 positive domains. Scale bar: 10 μm. On top: schematic representation of H-Aβ42 sequence, showing the binding sites for the different antibodies. c Dot blot analysis of the extracellular medium collected 6 and 24 h after microglia exposure to H-Aβ42. Histograms represent the densitometry quantification upon staining with anti-Aβ42 C-terminal antibody (left histogram) or 6E10 N-terminal antibody (right histogram). Intensity values are shown. N = 5 independent experiments, statistical analysis was performed by One way Anova, Bonferroni multiple comparison test (**P < 0.01, *P < 0.05). d Quantification of Aβ42 extracellular levels in microglia cultures exposed to H-Aβ42 for the indicated time points. Two different ELISA kits based on a capture antibody against the C-terminal (left) or N-terminal (right) domains of the protein were used. N = 5 independent experiments. Statistical analysis was performed by One way Anova, Bonferroni’s post hoc test for multiple comparisons (*** P < 0.001). e ELISA of C-terminal-containing Aβ42 in the extracellular medium of microglia exposed to H-Aβ42 in both control conditions and in the presence of protease inhibitors. N = 3 independent experiments. Statistical analysis was performed by two way Anova, Bonferroni’s post hoc test for multiple comparisons. f ELISA of C-terminal-containing Aβ42 in the extracellular medium of N9 cells exposed to siRNAcontrol or MMP9siRNA. Statistical analysis was performed by two way Anova, Bonferroni’s post hoc test for multiple comparisons. g Representative dot blots of the extracellular medium from H-Aβ42-treated microglia in control conditions, in the −/− presence of protease inhibitors, from siRNAcontrol or MMP9siRNA or from MMP9 microglia. Blots are immunostained with anti-Aβ42 or 6E10 antibodies. h-j Dot blot quantification of the extracellular medium collected from microglia in control conditions or in the presence of protease inhibitors h), from −/− siRNAcontrol or MMP9siRNA i)or from MMP9 microglia j) after 6 and 24 h of Aβ42 treatment. k) Apical-to-basolateral exchange across the BBB of fluorescent Aβ42 over 120 min upon pre-incubation with Aβ42-treated microglia conditioned medium in the presence or absence of protease inhibitors. Quantification of unidirectional trans-endothelial Aβ42 transport by fluorescence spectrophotometry of N = 2 independent experiments; statistical analysis was performed by One way Anova, Bonferroni’s post hoc test for multiple comparisons (**P < 0.01) the healthy brain [8–11, 59], literature evidences clearly in- brain. Here we show that the injection of a C-terminal dicate that intracranial injection of a single form of syn- truncated synthetic Aβ peptide (Aβ24), which may result thetic Aβ (Aβ42) does not induce plaque formation in wt from microglial MMP9 proteolytic activity, has seeding NT NT NT 6 0 24 1 NT NT NT 0h 6h 24h NT NT 0h NT 6h 24h 0h 6h 24h NT C-ter-H-Aβ42 (pg/ml) anti-Aβ42 (densitometry) anti-Ab42( densitometry) anti-Aβ42 (densitometry) C-ter-H-Aβ42 (pg/ml) 6E10 (densitometry) anti-Ab42(densitometry) anti-H-Aβ42 C-ter-H-Aβ42 (pg/ml) Papp (f old change) N-ter-H-Aβ42 (pg/ml) 6E10 Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 12 of 19 properties for aggregate formation in intracranially injected and, thereby, increasing Aβ concentration, may foster dis- wild type mice. Our results thus provide a direct demon- ease progression [64]. stration of the concept that biologically relevant mixtures We demonstrate that Aβ24, when concomitantly of Aβ forms may result in vivo in more complex aggrega- present with Aβ42 at the abluminal side of the endothe- tion dynamics than those predicted by in vitro studies. Not- lium, slows amyloid clearance through BBB. This is ably, amyloid deposits are sporadically detectable also in likely due to the fact that H-Aβ24 aggregates, which dis- thenon-injectedhemisphereand in thehippocampus, thus play higher antiparallel character, retain Aβ42, thus excluding they may represent post-injectional sprouting or impairing its clearance through the BBB. Although we leftover after cortical injection and brain lesioning. cannot exclude that Aβ42 levels measured in the serum Although it is known that the number of amyloid plaques of injected mice may reflect other routes of clearance does not necessarily correlate with cognitive impairments besides BBB, our results suggest that the same mechan- ([60–62]; reviewed in [63]), the presence of amyloid de- ism may operate also in vivo. In particular, impaired posits at the hippocampal level may explain the occurrence clearance might overcome the critical concentration of of the behavioral defects observed in injected in healthy proteins initiating the aggregation process and causing mice. the formation of ThT, Congo-red and 6E10-positive It is now established that protein aggregation takes aggregates detected in wt brains upon injection of the place, biophysically, once a critical concentration of H-Aβ24/H-Aβ42 mix. Hence, H-Aβ24, and possibly proteins has been overcome [64]. The lag phase is re- other C-terminal truncated fragments, may function as duced by the presence of “seeds” [64], which enhance scaffold proteins to favor both human (in the case of H- fibrils formation. An efficient process of clearance is re- Aβ24/H-Aβ42 mix injection) and mouse Aβ42 (in the quired in order to prevent the increase in concentration case of H-Aβ24 injection) recruitment, fibrillation and of “seeds”, which may in turn initiate the aggregation deposition. Interestingly, Aβ24 has been detected as a process [64]. Consistently, Aβ clearance rates were major amyloid component in leptomeninges of patients foundtobeimpairedin ADpatientscomparedto cog- affected by HCHWA-D (hereditary cerebral hemorrhage nitively normal controls, while there were no differ- with amyloidosis, Dutch type). These data indicate that ences in Aβ production rates [65]. Although no specific the Aβ24 fragment is in fact formed also in human evidence on the clearance of oligomeric forms is cur- brain, possibly generated by carboxyl-terminal limited rently available, oligomeric Aβ intermediates have been proteolysis [74]. Although future studies will be required found to alter proteasomal clearance [66]. to clarify what type of Aβ is deposited, our results are in Several mechanisms for Aβ clearance have been iden- line with the work of Schlenzig and colleagues which tified, including drainage via the BBB, which is mostly showed that N-terminally truncated and pyroglutamate- mediated by the low-density lipoprotein receptor related modified amyloid beta peptides are less soluble than full- protein-1 (LRP1). LRP1 is localized on the abluminal length peptides, increasing aggregation propensity and side of the brain capillary endothelium and mediates Aβ seeding of amyloid peptides [64, 75]. Notably, wt mice transport across the BBB in the direction of brain to injected with H-Aβ24 alone or with H-Aβ24/H-Aβ42 mix, blood [67–69]. When Aβ binds to LRP1 at the brain side display, besides Aβ protein deposition, behavioral defects of the BBB, a process of transcytosis starts, which medi- similar to those of age matched APP/PS1 mice. ates rapid Aβ clearance. Notably, LRP1 expression is re- One may want to consider whether structural informa- duced during aging and in AD as well as in patients with tion available for Aβ peptides could provide a possible the Dutch-type of cerebrovascular β-amyloidosis. The rationalization of the effect of cleavage at residue 24. transcytosis process is very efficient and, indeed, human Aβ The structural arrangement of Aβ fibrils and, even more injected into different brain regions of wt mice is rapidly re- importantly, oligomers and intermediates is still a matter covered in the plasma [70]. Consistently, no plaques nor ag- of debate; structural determinations are made difficult, gregates are formed in wt brain upon injection of Aβ42 [8, among other factors, by the presence of extensive poly- 9]. Interestingly, post-translationally modified forms of Aβ morphisms [76]. For fibrillar aggregates, numerous stud- are cleared less efficiently from the brain, like in the case of ies point towards a β-sandwich motif, i.e. two sheets N-terminal truncated and pyroglutamate-modified Aβ and with strands oriented perpendicular to the long axis of phosphorylated Aβ [71, 72]. Notably, Aβ peptides with the fibril; β-sandwiches are in turn arranged in higher higher β-sheet content are cleared less efficiently from order structures, i.e. with two-fold or three-fold sym- brain, due to a low-affinity LRP/Aβ interaction, mediating metry around the fibril axis [77]. A common feature of brain accumulation of amyloid. Based on the view that in- such models is that the two sheets comprise the two sufficient clearance of Aβ plays an essential role in the separate aggregation-prone regions identified above; pathogenesis of AD [73], one may speculate that even low therefore, the elimination of the C-terminal region in amounts of Aβ forms impairing physiological Aβ clearance Aβ24 would delete one of the sheets, and hence be Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 13 of 19 incompatible with the commonly assumed Aβ42 ar- KLVFFA has been crystallized by Eisenberg and collab- rangement (Fig. 10). orators in the antiparallel configuration [79]. These Although the consensus is weaker for what concerns data are consistent with previous observations that the the structure of oligomeric aggregates and other pos- Aβ24 fragment is present in both monomeric and ag- sible kinetic intermediates, recently proposed models gregated forms [55]. on the basis of NMR data [78] again foresee the pres- Aβ24 represents a prototypical example of a C- ence of cross-β contacts between the two extended terminal truncated fragment generated by Aβ42 prote- amyloidogenic regions identified above. Based on our olysis. Indeed, in line with the concept that Aβ42 is in- results and the fact that fibrillar and oligomeric aggrega- ternalized and metabolized by microglial cells [80–82], tion coexist in a “competitive equilibrium”,one may we provide the evidence that C-terminal truncated frag- hypothesize that the presence of Aβ24 shifts the competi- ments can be produced by microglia in a protease- tive balance towards oligomerization, and away from fib- dependent manner. These fragments may share similar rillation. Structurally, this is consistent with the following properties to those of Aβ24. Among those produced by data: first, although there is no conclusive structure for MMP9 (1–16, 1–20, 1–23, 1–30, 1–33 and 1–34, [19]), Aβ42, NMR-derived models (Fig. 10) have a consensus in one could argue that only Aβ16 should have a different attributing the steric zipper (the dry spine of a parallel-β character than the other fragments, because it lacks the fibril) to the 25–42 C-terminal region of Aβ42, which is of 16–20 region which is known to be amyloidogenic and course absent in Aβ24; second, the two algorithms provid- predicted as such by Walsh’s algorithm (among others) ing a prediction for the parallel/antiparallel character of fi- for all other proteolytic products. One would therefore brils (Walsh’sand Tartaglia’s) indicated that Aβ24 still has argue that, in the 1–16 case, both hypothetical self- aggregation potential, a relatively larger antiparallel β aggregation and cross-aggregation pathways would be propensity and that there is cross-aggregation potential abolished (Fig. 10). Conversely, C-terminal truncated between Aβ1-24 and Aβ25-42. Finally, the 16–21 KLVFFA fragments from the cleavage site at aa residue 20 up to region is predicted as the fragment’smost amyloidogenic; the 34 may possibly share similar properties to synthetic ab PDB: 2LMN PDB: 2LMP cd PDB: 2MXU PDB: 2MXU Fig. 10 Two-fold a and three-fold b symmetric structural models reported for Aβ40 by Tycko et al. [94]; and c the NMR model by Ishii et al. [89]. Peptide residues 9 to 24 shown in solid orange; residues 25 to 40 are transparent; residues 1–8 were not resolved. d same as c), with the 16–21 KLVFFA region highlighted; KLVFFA has been crystallized by Eisenberg et al. in the antiparallel configuration [79] Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 14 of 19 Aβ24. The in vivo relevance of fragments with length in small aliquots at−80 °C. Immediately prior to use, ali- possibly exceeding the aa 34 is more questionable, given quots were quickly resuspended in 50 mM NaPi, that fragments produced by microglia lack the C-terminal 100 mM NaCl pH 7.4 buffer at a concentration of region, as recognized by the antibody against aa 34–42. 40 μM, strongly vortexed, sonicated for 30 s and left at The role of microglia in the production of fragments room temperature for 20 min before being further di- which could in principle favor plaque formation, like luted for in vivo and in vitro experiments (the concen- Aβ24, adds an additional detrimental role to microglia tration used for each experiment is specified in the in AD. Indeed, an inverse correlation exists between figure legend). For Aβ clearance experiments, Aβ42–488 microglia activation and neurodegeneration [83] or cog- (HILyte, AnaspeC) and Aβ42-FAM (Anaspect) were nitive impairment [81, 84], indicating the occurrence of solubilised in NH OH and stored in small aliquots at a vicious cycle based on Aβ deposition, inflammation, −80 °C. neuronal damage and cognitive decline. Our evidence demonstrates that microglia may produce, through their Antibodies metabolic activity, C-terminal truncated Aβ forms which Antibodies used for immunoblot (western/dot blot), immu- in turn could initiate amyloid aggregation and cause noprecipitation, and immunofluorescence were as follows: phenotypic defects even in the absence of genetic muta- monoclonal antibody 6E10 (1:2000; Covance), which recog- tions associated with AD. Here we provide new evidence nizesresidues1–16 of human Aβ; rabbit anti-human beta for the contribution of microglia activation to the devel- amyloid 1–42 (1:1000; Alpha Diagnostic International) opment of the sporadic form of AD. which recognizes C-terminal 6 aa peptide from human beta 1–42; 4G8 antibody (1:2000; Covance) directed against the Materials and methods central part of Aβ42 (aa 17–24); m3.2 (kindly provided by Mice Prof. Paul Matthews) recognizes residues 10–15 of murine Mice were housed in the SPF animal facility of Humani- Aβ; rabbit anti-Iba1 antibody (1:500; Wako); ZO-1 (1:1000, tas Clinical and Research Center in individually venti- clone R40.76, Millipore); Connexin-43 (1:400; C6219, lated cages. Procedures involving animals handling and Sigma); Claudin-5 (1:800; ABT45, Millipore). Secondary care conformed to protocols approved by the Humanitas antibodies (1:200; Alexa Fluor®-conjugated, Molecular Clinical and Research Center (Rozzano, Milan, Italy) in Probes). compliance with national (4D.L. N.116, G.U., suppl. 40, 18-2-1992) and international law and policies (EEC Prediction of aggregation propensities Council Directive 2010/63/EU, OJ L 276/33, 22-09-2010; We used the AMYLPRED2 meta-predictor [87] to com- National Institutes of Health Guide for the Care and Use pare the aggregation profiles of the sequences of full of Laboratory Animals, US National Research Council, Aβ42 peptide with respect to the truncated form includ- 2011). The study was approved by the Italian Ministry of ing Aβ residues 1–24 (henceforth Aβ24). The meta- Health (approval n. 6/2014). All the experimental proce- predictor identifies putative amyloigenic regions on the dures followed the guidelines established by the Italian basis of the consensus between 11 methods considering Council on Animal Care and were approved by the Ital- a range of physico-chemical properties. Further analysis ian Government decree No. 27/2010. All efforts were was carried out with the algorithms PASTA 2.0 [44] and made to minimize the number of subjects used and their PAGE/ABSOLUTERATE [88] in order to obtain quanti- suffering. Mice were housed in cages with free access to tative predictions of putative aggregation propensities, food and water at 22 °C and with a 12-h alternating rates, and the fibrillar’s beta-strand parallel versus anti- light/dark cycle. Double transgenic APPswe/PSEN1dE9 parallel character. Putative three-dimensional fibrillar (APP/PS1) mice were purchased from Jackson Labora- arrangements were obtained from the PDB database en- −/− tory [85]. C57BL/6 J-App (App ) mice were provided tries 2LMN, 2LMP, 2MXU [89]. by Hertie Institute and Deutsches Zentrum für Neurode- generative Erkrankungen (DZNE), Tübingen, Germany. Thioflavin T (ThT) assays −/− BALB/c Mmp9 < tm1Tvu > (MMP9 ) P1-P3 pups were Thioflavin T (ThT) dye was purchased from Sigma provided by Istituto Nazionale dei Tumori, Milan, Italy. Aldrich. For the ThT assays, 200 μLof Aβ42 (at 8 μM concentration), of Aβ24 (at 8 μM concentration), or of Preparation of synthetic Aβ peptides the equimolar mixture of the two peptides (at 4 μM con- Synthetic human Aβ1–42 and Aβ1–24 were purchased centration each) were incubated in 50 mM NaPi, from Bachem and prepared as previously described [86] 100 mM NaCl pH 7.4 buffer at 37 °C with ThT at 10 μM to obtain oligomeric Aβ forms. Briefly, lyophilized Aβ concentration. At different incubation times, the fluores- 1–42 and 1–24 were dissolved in dimethyl sulfoxide cence emission spectra of the samples were collected (DMSO, Sigma) to a concentration of 2 mM and stored after excitation at 450 nm [45] or at 270 nm [47] by the Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 15 of 19 Cary Eclipse Spectrofluorimeter (Varian Australia Pty divided into a peripheral and a central region using Ltd, Mulgrave VIC, Australia). Quartz cuvettes of 1 cm ViewPoint video tracking software. The central quad- path length were employed. For time course experi- rants are collectively referred to as the center zone and ments, the samples were kept at 37 °C and analyzed at peripheral quadrants are collectively referred to as the each time point. peripheral zone. Data were collected continually for 30 min and the distance traveled (cm), velocity (cm/s), ATR-FTIR spectroscopy and the distance traveled in the center zone versus the For the ATR-FTIR measurements, 2 μLof Aβ42 (at peripheral zone were all recorded and scored automatic- 100 μM concentration), of Aβ24 (at 100 μM concentra- ally. In the open field task ambulatory movements (val- tion), or of the equimolar mixture of the two peptides ued as distance traveled and movement speeds) and (at 50 μM concentration each) were deposed on the sin- anxiety-like behaviors (as the as distance traveled in the gle reflection diamond element of the ATR device center zone versus the peripheral zone) can be assessed (Quest, Specac, UK). Spectra were recorded after solvent in response to a novel environment. evaporation to allow the formation of an hydrated film as previously described [48]. FTIR measurements were Social interaction task performed using the Varian 670-IR spectrometer (Varian The social interaction test was used to measure how mice Australia Pty Ltd, Mulgrave VIC, Australia) under the respond to a social partner during a 5-min test following following conditions: 1000 scan coadditions, 25 kHz of −1 isolation housing. Since isolation housing potentiates ex- scan speed, 2 cm of spectral resolution, triangular pression of innate territorial defensive responses, in this apodization, and a nitrogen-cooled Mercury Cadmium test we evaluated the reaction of mice to the presence of a Telluride detector. Fourier self deconvolution was ob- −1 new animal in free conditions in the same arena. For this tained with a full width at half height of 13.33 cm and test, mice were placed for 30 min alone into the open field a resolution enhancement factor K = 1.5 [48] using the arena to familiarize themselves with the new environment. Resolutions-Pro software (Varian Australia Pty Ltd, Mul- After this time a new female was introduced and the rec- grave VIC, Australia). ord started. We used the ViewPoint system to count the number of time the tested mouse made contact with the Intracranial injections new females. The decision to use females was made to Stereotaxic intracranial injections in mice brains were avoid any occurring of an aggressive behaviour simply made under a mixture of ketamine (100 mg/kg) and caused by a normal intermale instinct. xylazine (10 mg/kg) anesthesia. After surgical exposure of dura mater, both the bregma and the skull surface served as the stereotaxic zero points. Using a Hamilton syringe, 4 μLofH-Aβ or vehicle (vehicle consists of Novel object recognition (NOR) physiological buffer−50 mM NaPi, 100 mM NaCl The test apparatus consisted of an open field box measur- pH 7.4−) were injected into the neocortex (AP 1 mm, ing 50 cm × 25 cm and all sessions were video-recorded. ML 2 mm, DV−2 mm) with a speed of 1.5 μL/min, the The first day the animal was allowed to explore the empty needle was kept in place for an additional minute be- field arena for a 10-min time period (habituation session) fore it was slowly drawn out. For each experiment 4 before being exposed to a 10-min period of familiarization pmol of H-Aβ42, 4 pmol of H-Aβ24 and 2 pmol of H- session in the presence of identical objects (A/A). This Aβ42 plus 2 pmol of H-Aβ24 for the mix condition familiarization session was followed by 1 h and 24 h delays were injected, as described in [70, 90]. Different condi- during which the animals were returned to their home tionswereusedfor specific experimentsasspecified in cages. After the delay the animals performed 10-min of test the text. After suturing the incision, mice were session (A/B) in which one object was kept as during the maintained on a warm pad until recovery from the familiarization session (A) and another was changed (B). anesthesia, then returned to their cages. All procedures The objects were made of hard plastic and had previously were conducted in accordance with institutional guide- been counterbalanced to control for any object preference lines for the care and use of experimental animals. bias. The total amount of time spent with each object was recorded and scored using fully automated ViewPoint video Behavioral tests tracking software. The time spent around each object was Open field defined as the time in which the animal directed its nose to Mice were placed in a multi-unit open field maze (View- theobjectatadistance <2.0 cm and/or by the animal Point instruments) with field chamber (25 cm long and touching the object with its nose. Data are shown as the 25 cm wide), and activity was recorded using ViewPoint total amount of time that animals spend exploring the video tracking software. Each quadrant was digitally novel object during both the 1 and 24 h delay. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 16 of 19 Microglia cell culture Human Beta-Amyloid (1–24) (Bachem) were pre- Primary microglia were obtained from mixed cultures pre- incubated for 20 min in FluoroBrite™ DMEM (Gibco) pared from the cerebral cortex of mice at the postnatal medium at 37 °C, and added to the apical (brain) side cor- day 1 (P1-3). Microglia cells were isolated by shaking responding to the donor compartment. The basolateral flasks for 45 min at 230 rpm at day 10 after plating. Cells (blood) side of the transwells represented the receiver side were then seeded on poly-L-ornithine (Sigma) pre-coated and was exposed to medium alone. Alternatively, primary wells at the density of 1,5×10 cell/mL in DMEM contain- microglial cells were incubated for 24 h at 37 °C in Fluoro- ing 20 % heat-inactivated fetal bovine serum (FBS) and in- Brite™ DMEM medium containing Human Beta-Amyloid cubated at 37 ° C in a humidified atmosphere of 5 % CO2 (1–42) (Bachem) in the presence or absence of protease and 95 % air. Where indicated, cells were pre-treated for inhibitors (Roche). Microglia conditioned medium was 45 min with protease inhibitors and EDTA 2.5 mM recovered and spun for 2 min at max speed. 1 μMHiLyte (Roche) which were directly added to the cell culture FluorTM 488-labeled Beta-Amyloid (1–42) was pre- medium. N9 cells were maintained in IDMEM (Gibco incubated for 30 min at 37 °C with microglia conditioned Laboratories, USA), supplemented with 10 % FBS. For medium, and added to the apical compartment. Following siRNA transfection, N9 cells were plated at a concentra- 120 min of incubation at 37 °C with slow mixing, samples tion of at 1×10 cells/mL into either 96 or 24 multiwell were collected from the upper and lower chambers to as- plate. Specific siRNAs were diluted at a final concentra- sess the movement of fluorescently-labelled Aβ42 across tion of 20 nM siRNA. Cells were used within 48–72 h the bEnd.3 monolayer (apical to basolateral). The level of after transfection. When not differently indicated, for fluorescence in the media collected was measured for in vitro experiments microglia were treated with H-Aβ42 488-Aβ42 (λ = 503 nm and λ =528 nm) or for FAM- ex em 400 nM. Aβ42 (λ = 492 nm and λ =518 nm) using a Synergy™ ex em H4 Hybrid Multi-Mode Microplate Reader (BioTek). Rela- Endothelial cells culture and BBB model tive fluorescence units were converted to concentration Mouse brain endothelial cells (bEnd.3) [BEND3] (ATCC® according to prepared standard, and were corrected for CRL2299™) were used as a representative BBB model. background fluorescence. The amount of fluorescently- bEnd.3 cells were cultured at 37 °C, 5 % CO /saturated labelled Aβ42 was calculated as apparent permeability humidity in DMEM supplemented with 10 % heat- (P ) coefficient [ref. Zhao Z 2015 Nat Neurosci; Keaney app inactivated fetal bovine serum (FBS), 1 % penicillin- J 2015 Sci Adv] and expressed as fold change of control. streptomycin. Briefly, Aβ42 volume cleared (ΔV ) was calculated using the equation ΔV =C ×V /C ,where C and c lower lower upper upper Transendothelial cell electrical resistance (TEER) assay C are fluorescently-labeled Aβ42 concentrations in lower bEnd.3 cells were seeded at a concentration of 30000 the donor and receiving compartments, respectively, and cells/cm onto geltrex-coated (thin layer; Gibco) trans- V is the volume on the basolateral side. The volume lower well inserts (polycarbonate, 12 mm diameter, 3 μm pore cleared (ΔV ) was plotted against assay time. Permeability size; Costar) until a monolayer was established. TEER coefficients (P) were calculated by dividing against the was assessed using a Voltohmeter (Millicell Electrical surface area of the filter (1.12 cm ). Resistance System, Millipore). Background resistance from cell-free matrix-coated transwells was subtracted from recorded values to determine absolute TEER values Immunocytochemistry and cells imaging and corrected for the area covered by the cell mono- Cells were fixed for 15 min at room temperature in a layer. TEER was measured once a day to monitor cell 4 % (w/v) PFA, 4 % (w/v) sucrose, 20 mM NaOH confluence and development of tight junctions. Change and 5 mM MgCl in PBS, pH 7.4. Cells were perme- in absolute TEER from T for each individual transwell abilized and blocked for 30–60 min at room was recorded over time and then averaged for each day temperature in 15 % (w/v) goat serum, 0.3 % (v/v) before treatment. Triton X-100, 450 mM NaCl, 20 mM phosphate buffer, pH 7.4 and incubated at 4 °C overnight with Aβ42 permeability of the endothelial barrier primary antibodies diluted in blocking buffer. Cover- To assess Aβ42 exchange across the BBB model, bEnd.3 slips were mounted onto slides with PBS containing cells were cultured upside-down on a transwells system as 70 % glycerol and 1 μM DAPI. Representative images described above, until steady-state TEER had been were taken using a confocal microscope Fluoview reached. 1 μM FAM-labeled Human Beta-Amyloid (1–42) FV1000 Olympus IX81 (Center Valley, PA, USA) with an or FAM-labeled scrambled Beta-Amyloid (1–42) or oil immersion objective (×40 or × 60 × 1.4 NA Plan- HiLyte Fluor™ 488-labeled Beta-Amyloid (1–42) (Anaspec Apochromat; Olympus) using laser excitation at 405, 488 Peptide, Eurogentec) in the presence or absence of or 594 nm, and processed using Fiji [91]. Alternatively Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 17 of 19 bEnd.3 cells were stained with Diff Quick (Dade Behering, Histological and immunohistochemical analyses BioMap) and acquired with inverted microscope Olympus Mice were anesthetized and perfused with 0.9 % saline, IX53 (Center Valley, PA, USA). followed by 4 % PFA. The 30 μm cryosections of brain were blocked in PBS containing 10 % goat serum and Brain homogenate preparation and western blot analysis 0.1 % Triton X-100 for 1 h at room temperature before Animals were anesthetized and perfused with PBS before being incubated overnight at 4 °C with primary anti- brains were removed, weighted and homogeneted in a bodies. The following day, the slides were rinsed in PBS mild hypotonic buffer (50 mM Tris pH 8, 150 mM NaCl, and incubated at room temperature for 1 h with second- 5 mM EDTA supplemented with phosphatase and prote- ary antibody. The slides were processed using the ABC ase inhibitor, EDTA-free, Roche). Supernatants were then detection kit (Vector Laboratories). The presence of the centrifuged for 1 h at 3000 × g to pellet insoluble material, antigens was revealed using the DAB (diaminobenzidine) including insoluble Aβ species. Samples were either frozen (brown) peroxidase substrate kit (Vector Laboratories). on dry ice or LDS sample buffer (Invitrogen) containg 5 % Immunofluorescence and ThT staining were performed β-mercaptoethanol was immediately added. Pellets were on brain sections. Brain slices were washed three times directly resuspended in 20 μl of LDS sample buffer. 10 μl in PBS and incubated for 1 h at room temperature in sample were loaded onto a Bolt 4–12 % Bis-Tris Plus Gels blocking solution (3 % BSA). Subsequently, the slides (Thermo Fisher) for Western Blot evaluation. Samples were washed 3 times and incubated overnight with spe- were separated using MES buffer and transferred onto a cific antibodies. The next day, sections were washed in 0.22 μM nitrocellulose membrane (Bio-Rad). Membranes PBS and incubated for 2 h at room temperature with were washed in TBS-Tween (150 mM NaCl, 50 mM Tris specific fluorochrome-conjugated secondary antibodies and 0.1 % (v/v) Tween-20) and incubated for 45 min at diluted in 3 % BSA in the dark. ThT solution was pre- room temperature in blocking solution (5 % milk or 2.5 % pared as described above. Final solution was added to serum bovine albumin in TBST). Membranes were subse- free floating slides for 1 min and very quickly washed quently probed overnight at 4 °C with primary antibodies with 80 % methanol followed by 3 washes with distilled diluted in TBS-T buffer. Membranes were washed exten- water. Images were acquired by Virtual Slides micro- sively and incubated for 45 min at room temperature with scope (VS120, Olympus). For quantification, both dif- horseradish peroxidase-conjugated (HRP) secondary anti- fuse-plaques and dense-core plaques were considered, as body diluted in TBST buffer. Antibody-specific signals described in [92]. Small dots and deposits at the slice were detected using enhanced chemiluminescence re- edge were not counted. The entire surface of the slice agents (Clarity Western ECL substrate, Bio-Rad). was examined. For congo red histological staining, slices were air-dried on glass overnight, then stained with a Dot blot analysis 0.2 % congo red solution according to [93]. Specimens Aliquots of supernatant samples (200 μL) were loaded on were acquired by transmittance polarized light micros- nitrocellulose membrane Trans-Blot Transfer Medium copy, FITCH filtered epifluorescence. For quantification, (0.22 μm, Bio-Rad), by vacuum deposition on the Bio-Dot bright field images were analyzed by color segmentation SF blotting apparatus (Bio-Rad). Serial dilution curves of plugin-ImageJ software (NIH, Bethesda, MD). The entire Aβ42 synthetic protein were preliminarily run to obtain area of deposits was considered. non-saturating condition of immunodetection. DB dots images were analysed by Image Lab™ software (Bio-Rad). Statistical analysis Statistical analysis was performed with PRISM software ELISA (Graph-Pad Software, San Diego, CA, USA). Data are Aβ42 levels were determined using specific ELISA kit expressed as mean ± SEM. Comparisons between two (Amyloid-beta (x-42) ELISA IBL, International) following groups were performed using Student’s t test or by non- manufacturer’s instructions. Briefly, 100 μL of sample was parametric two-tailed Mann–Whitney U test. For the added into the pre-coated plate and was incubated over- comparison of more than two groups, two-way ANOVA night at 4 °C. After washing each well of the pre-coated followed by Bonferroni’s post hoc test was used. Differ- plate, 100 μL of labeled antibody solution was added and ences were considered significant at *P < 0.05, **P < 0.01, the mixture was incubated for 1 h at 4 °C in the dark. ***P < 0.001. After washing, chromogen was added and the mixture Funding was incubated for 30 min at room temperature in the This research has been supported by Cariplo 2015–0594 to MM, Italian Ministry dark. After the addition of stop solution, the resulting of Health GR-2011-02347377 to EL and MM, Cariplo 2014–0655 to IB and MM. TG would like to acknowledge prof. Amedeo Caflisch (Univ. of Zurich) for color was assayed at 450 nm using a microplate absorb- providing the PAGE software. TG also acknowledges support from the CNR-STM ance reader (Synergy H4 Synergy™ H4 Microplate Reader, 2013 mobility scheme. SM has been supported by Fondazione Veronesi. MT has BioTek). been supported by Fondazione Vollaro. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 18 of 19 Availability of data and materials 16. Hu J, et al. Angiotensin-converting enzyme degrades Alzheimer amyloid Not applicable. beta-peptide (A beta); retards A beta aggregation, deposition, fibril formation; and inhibits cytotoxicity. J Biol Chem. 2001;276(51):47863–8. 17. Backstrom JR, et al. Matrix metalloproteinase-9 (MMP-9) is synthesized in Authors’ contributions neurons of the human hippocampus and is capable of degrading the MM, SM, FF and MR designed the experiments. SM, FF and MR performed amyloid-beta peptide (1–40). J Neurosci. 1996;16(24):7910–9. analyses and most of the experiments. MM, FF and MR wrote the manuscript. 18. Zhang Q, et al. Metabolite-initiated protein misfolding may trigger EL set the blood brain barrier model and performed the experiment with Alzheimer’s disease. Proc Natl Acad Sci U S A. 2004;101(14):4752–7. endothelial cells. CS performed ELISA. MT performed intractranical injections in 19. Yan P, et al. Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro mice. DP and IB contributed data analysis and interpretation. TG performed and compact plaques in situ. J Biol Chem. 2006;281(34):24566–74. bioinformatics analyses. AN performed Tht and FTIR to test amyloid 20. Zhang R, et al. Interprotofilament interactions between Alzheimer’s Abeta1- aggregation propensity. All authors read and approved the final manuscript. 42 peptides in amyloid fibrils revealed by cryoEM. Proc Natl Acad Sci U S A. 2009;106(12):4653–8. 21. Frautschy SA, et al. Protease inhibitor coinfusion with amyloid beta-protein Competing interests results in enhanced deposition and toxicity in rat brain. J Neurosci. 1998; The authors declare that they have no competing interests. 18(20):8311–21. 22. Saido T, Leissring MA. Proteolytic degradation of amyloid beta-protein. Cold Consent for publication Spring Harb Perspect Med. 2012;2(6):a006379. Not applicable. 23. Hernandez-Guillamon M, et al. Sequential Amyloid-beta Degradation by the Matrix Metalloproteases MMP-2 and MMP-9. J Biol Chem. 2015;290(24):15078–91. Author details 24. Deb S, Gottschall PE. Increased production of matrix metalloproteinases in 1 2 IRCCS Humanitas, via Manzoni 56, 20089 Rozzano, Italy. Hertie Institute and enriched astrocyte and mixed hippocampal cultures treated with beta- Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), amyloid peptides. J Neurochem. 1996;66(4):1641–7. Otfried-Müller-Straße 27, 72076 Tübingen, Germany. Humanitas University, 25. Yin KJ, et al. Matrix metalloproteinases expressed by astrocytes mediate via Manzoni 56, 20089 Rozzano, Italy. IN-CNR, Corso Stati Uniti 4, 35126 extracellular amyloid-beta peptide catabolism. J Neurosci. 2006;26(43):10939–48. Padova, Italy. Department of Biotechnology and Biosciences, University of 26. Belli M, Ramazzotti M, Chiti F. Prediction of amyloid aggregation in vivo. Milano-Bicocca, Piazza della Scienza 2, 20126 Milano, Italy. IN-CNR, via EMBO Rep. 2011;12(7):657–63. Vanvitelli 32, 20129 Milano, Italy. 27. Biancalana M, Koide S. Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochim Biophys Acta. 2010;1804(7):1405–12. Received: 21 September 2016 Accepted: 24 September 2016 28. Khurana R, et al. Mechanism of thioflavin T binding to amyloid fibrils. J Struct Biol. 2005;151(3):229–38. 29. Alvarez A, et al. Serum TNF-alpha levels are increased and correlate negatively with free IGF-I in Alzheimer disease. Neurobiol Aging. 2007;28(4):533–6. References 30. Fillit H, et al. Elevated circulating tumor necrosis factor levels in Alzheimer’s 1. Mattson MP. Pathways towards and away from Alzheimer’s disease. Nature. disease. Neurosci Lett. 1991;129(2):318–20. 2004;430(7000):631–9. 31. Ma SL, et al. Association between tumor necrosis factor-alpha promoter 2. Scheuner D, et al. Secreted amyloid beta-protein similar to that in the senile polymorphism and Alzheimer’s disease. Neurology. 2004;62(2):307–9. plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 32. Janelsins MC, et al. Early correlation of microglial activation with enhanced and APP mutations linked to familial Alzheimer’s disease. Nat Med. 1996; tumor necrosis factor-alpha and monocyte chemoattractant protein-1 2(8):864–70. expression specifically within the entorhinal cortex of triple transgenic 3. Selkoe DJ. Aging, amyloid, and Alzheimer’s disease: a perspective in honor Alzheimer’s disease mice. J Neuroinflammation. 2005;2:23. of Carl Cotman. Neurochem Res. 2003;28(11):1705–13. 33. Sheng JG, et al. Lipopolysaccharide-induced-neuroinflammation increases 4. Bitan G, et al. Amyloid beta-protein (Abeta) assembly: Abeta 40 and Abeta intracellular accumulation of amyloid precursor protein and amyloid beta 42 oligomerize through distinct pathways. Proc Natl Acad Sci U S A. 2003; peptide in APPswe transgenic mice. Neurobiol Dis. 2003;14(1):133–45. 100(1):330–5. 34. Terry RD, et al. Physical basis of cognitive alterations in Alzheimer’s disease: 5. Yan Y, Wang C. Abeta42 is more rigid than Abeta40 at the C terminus: synapse loss is the major correlate of cognitive impairment. Ann Neurol. implications for Abeta aggregation and toxicity. J Mol Biol. 2006;364(5):853–62. 1991;30(4):572–80. 6. Hubin E, et al. Transient dynamics of Abeta contribute to toxicity in 35. DeKosky ST, Scheff SW. Synapse loss in frontal cortex biopsies in Alzheimer’s Alzheimer’s disease. Cell Mol Life Sci. 2014;71(18):3507–21. disease: correlation with cognitive severity. Ann Neurol. 1990;27(5):457–64. 7. Kuperstein I, et al. Neurotoxicity of Alzheimer’s disease Abeta peptides is 36. Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer’s disease: progress induced by small changes in the Abeta42 to Abeta40 ratio. EMBO J. 2010; and problems on the road to therapeutics. Science. 2002;297(5580):353–6. 29(19):3408–20. 37. Prut L, Belzung C. The open field as a paradigm to measure the effects of 8. Kane MD, et al. Evidence for seeding of beta-amyloid by intracerebral drugs on anxiety-like behaviors: a review. Eur J Pharmacol. 2003;463(1–3):3–33. infusion of Alzheimer brain extracts in beta-amyloid precursor protein- 38. Rodriguiz RM, Wetsel WC. Assessments of Cognitive Deficits in Mutant Mice. transgenic mice. J Neurosci. 2000;20(10):3606–11. In: Levin ED, Buccafusco JJ, editors. Animal Models of Cognitive Impairment. 9. Meyer-Luehmann M, et al. Exogenous induction of cerebral beta-amyloidogenesis Boca Raton (FL): CRC Press/Taylor & Francis; 2006. is governed by agent and host. Science. 2006;313(5794):1781–4. 39. Kliethermes CL, Cronise K, Crabbe JC. Anxiety-like behavior in mice in two 10. Eisele YS, et al. Induction of cerebral beta-amyloidosis: intracerebral versus apparatuses during withdrawal from chronic ethanol vapor inhalation. systemic Abeta inoculation. Proc Natl Acad Sci U S A. 2009;106(31):12926–31. Alcohol Clin Exp Res. 2004;28(7):1012–9. 11. Watts JC, et al. Bioluminescence imaging of Abeta deposition in bigenic 40. Cheng D, et al. Novel behavioural characteristics of the APP (Swe)/ mouse models of Alzheimer’s disease. Proc Natl Acad Sci U S A. 2011;108(6): PS1DeltaE9 transgenic mouse model of Alzheimer’s disease. Behav Brain 2528–33. Res. 2013;245:120–7. 12. Oberstein TJ, et al. Astrocytes and microglia but not neurons preferentially 41. Chambon C, et al. Behavioural and cellular effects of exogenous amyloid- generate N-terminally truncated Abeta peptides. Neurobiol Dis. 2015;73:24–35. beta peptides in rodents. Behav Brain Res. 2011;225(2):623–41. 13. Howell S, Nalbantoglu J, Crine P. Neutral endopeptidase can hydrolyze beta-amyloid (1–40) but shows no effect on beta-amyloid precursor protein 42. Faucher P, et al. Hippocampal Injections of Oligomeric Amyloid beta- metabolism. Peptides. 1995;16(4):647–52. peptide (1–42) Induce Selective Working Memory Deficits and Long-lasting 14. Kurochkin IV, Goto S. Alzheimer’s beta-amyloid peptide specifically interacts with Alterations of ERK Signaling Pathway. Front Aging Neurosci. 2015;7:245. and is degraded by insulin degrading enzyme. FEBS Lett. 1994;345(1):33–7. 43. Mikros E, et al. High-Resolution NMR Spectroscopy of the beta-Amyloid (1–28) 15. Eckman EA, Reed DK, Eckman CB. Degradation of the Alzheimer’s amyloid Fibril Typical for Alzheimer’s Disease. Angew Chem Int Ed Engl. 2001;40(19):3603–5. beta peptide by endothelin-converting enzyme. J Biol Chem. 2001;276(27): 44. Walsh I, et al. PASTA 2.0: an improved server for protein aggregation 24540–8. prediction. Nucleic Acids Res. 2014;42:W301–7. Mazzitelli et al. Acta Neuropathologica Communications (2016) 4:110 Page 19 of 19 45. LeVine 3rd H. Thioflavine T interaction with synthetic Alzheimer’s disease 73. Bates KA, et al. Clearance mechanisms of Alzheimer’s amyloid-beta peptide: beta-amyloid peptides: detection of amyloid aggregation in solution. implications for therapeutic design and diagnostic tests. Mol Psychiatry. Protein Sci. 1993;2(3):404–10. 2009;14(5):469–86. 46. Huang SH, et al. Effect of C-terminal residues of Abeta on copper binding 74. Castano EM, et al. The length of amyloid-beta in hereditary cerebral affinity, structural conversion and aggregation. PLoS One. 2014;9(3), e90385. hemorrhage with amyloidosis, Dutch type. Implications for the role of amyloid- beta 1–42 in Alzheimer’s disease. J Biol Chem. 1996;271(50):32185–91. 47. Librizzi F, et al. alpha-Casein Inhibits Insulin Amyloid Formation by Preventing the 75. Schlenzig D, et al. Pyroglutamate formation influences solubility and Onset of Secondary Nucleation Processes. J Phys Chem Lett. 2014;5(17):3043–8. amyloidogenicity of amyloid peptides. Biochemistry. 2009;48(29):7072–8. 48. Cerf E, et al. Antiparallel beta-sheet: a signature structure of the oligomeric 76. Nasica-Labouze J, et al. Amyloid beta Protein and Alzheimer’s Disease: amyloid beta-peptide. Biochem J. 2009;421(3):415–23. When Computer Simulations Complement Experimental Studies. Chem Rev. 49. Sarroukh R, et al. ATR-FTIR: a “rejuvenated” tool to investigate amyloid 2015;115(9):3518–63. proteins. Biochim Biophys Acta. 2013;1828(10):2328–38. 77. Luhrs T, et al. 3D structure of Alzheimer’s amyloid-beta (1–42) fibrils. Proc 50. Mancini S, et al. The hunt for brain Abeta oligomers by peripherally Natl Acad Sci U S A. 2005;102(48):17342–7. circulating multi-functional nanoparticles: Potential therapeutic approach for 78. Ahmed M, et al. Structural conversion of neurotoxic amyloid-beta (1–42) Alzheimer disease. Nanomedicine. 2016;12(1):43–52. oligomers to fibrils. Nat Struct Mol Biol. 2010;17(5):561–7. 51. Wesson DW, et al. Chronic anti-murine Abeta immunization preserves odor 79. Colletier JP, et al. Molecular basis for amyloid-beta polymorphism. Proc Natl guided behaviors in an Alzheimer’s beta-amyloidosis model. Behav Brain Acad Sci U S A. 2011;108(41):16938–43. Res. 2013;237:96–102. 80. Joshi P, et al. Microglia convert aggregated amyloid-beta into neurotoxic 52. Heneka MT, Golenbock DT, Latz E. Innate immunity in Alzheimer’s disease. forms through the shedding of microvesicles. Cell Death Differ. 2014;21(4): Nat Immunol. 2015;16(3):229–36. 582–93. 53. Wyss-Coray T, Rogers J. Inflammation in Alzheimer disease-a brief review of 81. Edison P, et al. Microglia, amyloid, and cognition in Alzheimer’s disease: An the basic science and clinical literature. Cold Spring Harb Perspect Med. [11C] (R) PK11195-PET and [11C] PIB-PET study. Neurobiol Dis. 2008;32(3):412–9. 2012;2(1):a006346. 82. Griciuc A, et al. Alzheimer’s disease risk gene CD33 inhibits microglial 54. Meda L, et al. Activation of microglial cells by beta-amyloid protein and uptake of amyloid beta. Neuron. 2013;78(4):631–43. interferon-gamma. Nature. 1995;374(6523):647–50. 83. Serrano-Pozo A, et al. Reactive glia not only associates with plaques but also 55. Bateman DA, Chakrabartty A. Cell surface binding and internalization of abeta parallels tangles in Alzheimer’s disease. Am J Pathol. 2011;179(3):1373–84. modulated by degree of aggregation. Int J Alzheimers Dis. 2011;2011:962352. 84. Okello A, et al. Microglial activation and amyloid deposition in mild 56. Giannakopoulos P, et al. Tangle and neuron numbers, but not amyloid load, cognitive impairment: a PET study. Neurology. 2009;72(1):56–62. predict cognitive status in Alzheimer’s disease. Neurology. 2003;60(9):1495–500. 85. Jankowsky JL, et al. Mutant presenilins specifically elevate the levels of the 57. Elman JA, et al. Neural compensation in older people with brain amyloid- 42 residue beta-amyloid peptide in vivo: evidence for augmentation of a beta deposition. Nat Neurosci. 2014;17(10):1316–8. 42-specific gamma secretase. Hum Mol Genet. 2004;13(2):159–70. 58. Jack Jr CR, et al. Age-specific population frequencies of cerebral beta- 86. Amijee H, et al. The N-methylated peptide SEN304 powerfully inhibits Abeta amyloidosis and neurodegeneration among people with normal cognitive (1–42) toxicity by perturbing oligomer formation. Biochemistry. 2012;51(42): function aged 50–89 years: a cross-sectional study. Lancet Neurol. 2014; 8338–52. 13(10):997–1005. 87. Tsolis AC, et al. A consensus method for the prediction of ‘aggregation- 59. Duran-Aniotz C, et al. Brains from non-Alzheimer’s individuals containing prone’ peptides in globular proteins. PLoS One. 2013;8(1), e54175. amyloid deposits accelerate Abeta deposition in vivo. Acta Neuropathol 88. Tartaglia GG, et al. Prediction of aggregation rate and aggregation-prone Commun. 2013;1:76. segments in polypeptide sequences. Protein Sci. 2005;14(10):2723–34. 60. McLean CA, et al. Soluble pool of Abeta amyloid as a determinant of severity 89. Xiao Y, et al. Abeta (1–42) fibril structure illuminates self-recognition and of neurodegeneration in Alzheimer’s disease. Ann Neurol. 1999;46(6):860–6. replication of amyloid in Alzheimer’s disease. Nat Struct Mol Biol. 2015;22(6): 61. Price JL, et al. Neuropathology of nondemented aging: presumptive evidence 499–505. for preclinical Alzheimer disease. Neurobiol Aging. 2009;30(7):1026–36. 90. Zhao Z, et al. Central role for PICALM in amyloid-beta blood–brain barrier 62. Games D, et al. Alzheimer-type neuropathology in transgenic mice transcytosis and clearance. Nat Neurosci. 2015;18(7):978–87. overexpressing V717F beta-amyloid precursor protein. Nature. 1995; 91. Schindelin J, et al. Fiji: an open-source platform for biological-image 373(6514):523–7. analysis. Nat Methods. 2012;9(7):676–82. 63. Karran E, Mercken M, De Strooper B. The amyloid cascade hypothesis for 92. Serrano-Pozo A, et al. Neuropathological alterations in Alzheimer disease. Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Cold Spring Harb Perspect Med. 2011;1(1):a006189. Rev Drug Discov. 2011;10(9):698–712. 93. Wilcock DM, Gordon MN, Morgan D. Quantification of cerebral amyloid 64. Thal DR. Clearance of amyloid beta-protein and its role in the spreading of angiopathy and parenchymal amyloid plaques with Congo red Alzheimer’s disease pathology. Front Aging Neurosci. 2015;7:25. histochemical stain. Nat Protoc. 2006;1(3):1591–5. 65. Mawuenyega KG, et al. Decreased clearance of CNS beta-amyloid in 94. Paravastu AK, et al. Molecular structural basis for polymorphism in Alzheimer’s disease. Science. 2010;330(6012):1774. Alzheimer’s beta-amyloid fibrils. Proc Natl Acad Sci U S A. 2008;105(47): 66. Cecarini V, et al. Crosstalk between the ubiquitin-proteasome system and 18349–54. autophagy in a human cellular model of Alzheimer’s disease. Biochim Biophys Acta. 2012;1822(11):1741–51. 67. Shibata M, et al. Clearance of Alzheimer’s amyloid-ss (1–40) peptide from brain by LDL receptor-related protein-1 at the blood–brain barrier. J Clin Invest. 2000;106(12):1489–99. Submit your next manuscript to BioMed Central 68. Deane R, et al. LRP/amyloid beta-peptide interaction mediates differential brain efflux of Abeta isoforms. Neuron. 2004;43(3):333–44. and we will help you at every step: 69. Deane R, et al. Clearance of amyloid-beta peptide across the blood–brain • We accept pre-submission inquiries barrier: implication for therapies in Alzheimer’s disease. CNS Neurol Disord Drug Targets. 2009;8(1):16–30. � Our selector tool helps you to ﬁnd the most relevant journal 70. Cho SM, et al. Correlations of amyloid-beta concentrations between CSF � We provide round the clock customer support and plasma in acute Alzheimer mouse model. Sci Rep. 2014;4:6777. � Convenient online submission 71. Russo C, et al. Pyroglutamate-modified amyloid beta-peptides– � Thorough peer review AbetaN3(pE)–strongly affect cultured neuron and astrocyte survival. J Neurochem. 2002;82(6):1480–9. � Inclusion in PubMed and all major indexing services 72. Kumar S, et al. Phosphorylation of amyloid-beta peptide at serine 8 � Maximum visibility for your research attenuates its clearance via insulin-degrading and angiotensin-converting enzymes. J Biol Chem. 2012;287(11):8641–51. Submit your manuscript at www.biomedcentral.com/submit

Journal

Acta Neuropathologica Communications – Springer Journals

Published: Oct 10, 2016

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

Learn More →

Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

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

Learn More →

Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

Amyloid-β 1–24 C-terminal truncated fragment promotes amyloid-β 1–42 aggregate formation in the healthy brain

References (109)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies