Angiotensin-converting enzyme (ACE) is a validated risk locus for developing late-onset Alzheimer's disease (AD). Although ACE1 expression and enzyme activity correlate with AD diagnosis, the mechanism by which this occurs is unclear. As a cell membrane-bound and shed peptidase, ACE1 is most commonly known to produce angiotensin II (Ang II), which has been linked to AD pathogenesis but also has been shown to cleave toxic Aβ42 to Aβ40, further complicating its role in AD. Previous work from our group characterized a rare ACE coding variant discovered through whole-genome and whole-exome sequencing of late-onset AD families: ACE rs4980 (R1279Q mutation) increases neuronal ACE1 and subsequent signaling through the central renin-angiotensin system (RAS), inducing age-associated hippocampal neurodegeneration. In this work, we report on two additional ACE variants associated with increased risk for developing AD: rs3730043 (T916M) and rs142947404 (N1036K). These variants were selected to investigate their effect on ACE1 protein processing and function in SH-SY5Y stable cell lines. In these cell lines, ACE1 protein trafficking to the cell surface was unaltered. Interestingly, however, both T916M and N1036K mutant cell lines resulted in increased ACE1 catalytic activity. Consequently, both mutant cell lines produced elevated levels of Ang II, a known mediator of neurodegeneration. This study provides further evidence for the role of ACE1 in AD and warrants continued research on this topic.

Angiotensin-converting enzyme ( ACE ) is a validated risk locus for developing late-onset Alzheimer’s disease (LOAD). ACE1 controls blood pressure through the renin-angiotensin system (RAS), but it is also present and acts locally in the brain. Hypertension is associated with an increased risk for developing AD, and people taking select RAS-targeting therapeutics have reduced incidence of AD. The ACE variant rs4980 (R1284Q murine mutation) was discovered in LOAD families through WGS. Our group previously showed that ACE1 R1284Q caused age-associated hippocampal neurodegeneration and gliosis in mutant knock-in (KI) mice, which was more aggressive in females. Importantly, these phenotypes were rescued by treatment with anti-hypertensive drugs. Our previous study showed that ACE1 R1284Q caused neuron death in mice, however the mechanism is still unknown. This work aims to identify vulnerable hippocampal cell populations and pathways which might clarify the mechanism of ACE1 R1284Q-mediated neurodegeneration. Single nuclei were extracted from 30mg of flash-frozen hippocampi from 6-, and 12-month-old R1284Q ACE +/+ (WT) and ACE KI/KI (KI) and processed using 10X Genomics 3’ Dual-Index chemistry. Libraries were sequenced using the NovaSeq 6000. Preprocessing, quality control, and integration was performed on reads before subjecting them to cluster, differential expression, and pathway analyses. Together, the samples were represented by 143,000 nuclei in 23 clusters encompassing all neuronal and glial populations in the hippocampus. We identified the expression of every RAS component, excluding renin, in the hippocampus. Notably, inhibitory and excitatory neurons had the most differentially expressed genes (DEGs) in 12-month KI samples compared to controls. Additionally, 12-month KI females had upregulated microglial C1q genes compared to males. We found expression of RAS genes in the hippocampus, which to our knowledge has not yet been characterized by single-nucleus RNA sequencing. Furthermore, inhibitory neuron transcriptomes are the most affected in 12-month KI mice, suggesting altered neuron communication that may lead to neuron loss. Additionally, we identified sex-specific differences in 12-month KI female microglia, which may explain the more aggressive gliosis seen in female KIs compared to males. Future directions include performing ingenuity pathway analysis to identify upstream regulators of DEGs and experimental perturbations to test our hypotheses.

Accumulating evidence implicates the gut microbiome (GMB) in the pathogenesis and progression of Alzheimer's disease (AD). We recently showed that the GMB regulates reactive astrocytosis and Aβ plaque accumulation in a male APPPS1-21 AD mouse model. Yet, the mechanism(s) by which GMB perturbation alters reactive astrocytosis in a manner that reduces Aβ deposition remain unknown. Here, we performed metabolomics on plasma from mice treated with antibiotics (ABX) and identified a significant increase in plasma propionate, a gut-derived short-chain fatty acid, only in male mice. Administration of sodium propionate reduced reactive astrocytosis and Aβ plaques in APPPS1-21 mice, phenocopying the ABX-induced phenotype. Astrocyte-specific RNA-Seq on ABX- and propionate-treated mice showed reduced expression of proinflammatory and increased expression of neurotrophic genes. Next, we performed flow cytometry experiments, in which we found that ABX and propionate decreased peripheral RAR-related orphan receptor-γ+ (Rorγt+) CD4+ (Th17) cells and IL-17 secretion, which positively correlated with reactive astrocytosis. Last, using an IL-17 mAb to deplete IL-17, we found that propionate reduced reactive astrocytosis and Aβ plaques in an IL-17-dependent manner. Together, these results suggest that gut-derived propionate regulates reactive astrocytosis and Aβ amyloidosis by decreasing peripheral Th17 cells and IL-17 release. Thus, propionate treatment or strategies boosting propionate production may represent novel therapeutic strategies for the treatment of AD.

Angiotensin I converting enzyme (ACE1) maintains blood pressure homeostasis by converting angiotensin I into angiotensin II in the renin-angiotensin system (RAS). ACE1 is expressed in the brain, where an intrinsic RAS regulates complex cognitive functions including learning and memory. ACE1 has been implicated in neurodegenerative disorders including Alzheimer's disease and Parkinson's disease, but the mechanisms remain incompletely understood. Here, we performed single-nucleus RNA sequencing to characterize the expression of RAS genes in the hippocampus and discovered that Ace is mostly expressed in CA1 region excitatory neurons. To gain a deeper understanding of the function of neuronal ACE1, we generated ACE1 conditional knockout (cKO) mice lacking ACE1 expression specifically in hippocampal and cortical excitatory neurons. ACE1 cKO mice exhibited hippocampus-dependent memory impairment in the Morris water maze, y-maze, and fear conditioning tests. Total ACE1 level was significantly reduced in the cortex and hippocampus of ACE1 cKO mice showing that excitatory neurons are the predominant cell type expressing ACE1 in the forebrain. Despite similar reductions in total ACE1 level in both the hippocampus and cortex, the RAS pathway was dysregulated in the hippocampus only. Importantly, ACE1 cKO mice exhibited age-related capillary loss selectively in the hippocampus. Here, we show selective vulnerability of the hippocampal microvasculature and RAS pathway to neuronal ACE1 knockout. Our results provide important insights into the function of ACE1 in the brain and demonstrate a connection between neuronal ACE1 and cerebrovascular function in the hippocampus.

Angiotensin I converting enzyme (ACE1) maintains blood pressure homeostasis by converting angiotensin I (angI) into angiotensin II (angII) in the renin-angiotensin system (RAS). ACE1 is expressed in the brain, where an intrinsic RAS regulates complex cognitive functions including learning and memory. ACE1 has been implicated in neurodegenerative disorders including Alzheimer’s disease (AD) and Parkinson’s disease (PD), but the mechanisms remain incompletely understood. Here, we performed single-nucleus RNA sequencing to characterize the expression RAS genes in the hippocampus and discovered that Ace is mostly expressed in CA region excitatory neurons. To gain a deeper understanding of the function of neuronal ACE1, we generated ACE1 conditional knockout (cKO) mice lacking ACE1 expression specifically in hippocampal and cortical excitatory neurons. Interestingly, ACE1 cKO mice exhibited hippocampus-dependent memory impairment in the Morris water maze, y-maze, and fear conditioning tests, but exhibited normal motor skills in rotarod. Total ACE1 level was significantly reduced in the cortex and hippocampus of ACE1 cKO mice showing that excitatory neurons are the predominant cell type expressing ACE1 in the forebrain. Despite similar reductions in total ACE1 level in both the hippocampus and cortex, the RAS pathway was dysregulated in the hippocampus only. Importantly, ACE cKO mice exhibited exacerbated age-related capillary loss selectively in the hippocampus. Here, we show selective vulnerability of the hippocampal microvasculature and RAS pathway to neuronal ACE1 knockout. Our results provide important insights into the function of ACE1 in the brain and demonstrate a connection between neuronal ACE and cerebrovascular function in the hippocampus.### Competing Interest StatementThe authors have declared no competing interest.* ACE1 : angiotensin I converting enzyme angI : angiotensin I angII : angiotensin II RAS : renin-angiotensin system AD : Alzheimer’s disease cKO : conditional knockout AT1R : angII type 1 receptor ARBs : AT1R blockers ACEis : ACE1 inhibitors CNS : central nervous system snRNA-seq : single nucleus RNA-sequencing RER : respiratory exchange ratio PV : parvalbumin NPY : neuropeptide Y NVC : neurovascular coupling CBF : cerebral blood flow BBB : blood brain barrier SMCs : smooth muscle cells

Current scientific research is driven by the ability to manipulate gene expression by utilizing the Cre/loxP system in transgenic mouse models. However, artifacts in Cre-driver mouse lines that introduce undesired effects and confound results are increasingly being reported. Here, we show aberrant neuroinflammation and synaptic changes in two widely used Cre-driver mouse models. Neuroinflammation in CaMKIIα-iCre mice was characterized by the activation and proliferation of microglia and astrocytes in synaptic layers of the hippocampus. Increased GFAP and Iba1 levels were observed in hippocampal brain regions of 4-, 8- and 22-month-old CaMKIIα-iCre mice compared to WT littermates. Synaptic changes in NMDAR, AMPAR, PSD95 and phosphorylated CaMKIIα became apparent in 8-month-old CaMKIIα-iCre mice but were not observed in 4-month-old CaMKIIα-iCre mice. Synaptophysin and synaptoporin were unchanged in CaMKIIα-iCre compared to WT mice, suggesting that synaptic alterations may occur in excitatory postsynaptic regions in which iCre is predominantly expressed. Finally, hippocampal volume was reduced in 22-month-old CaMKIIα-iCre mice compared to WT mice. We tested the brains of mice of additional common Cre-driver mouse models for neuroinflammation; the nestin-Cre mouse model showed synaptic changes and astrocytosis marked by increased GFAP+ astrocytes in cortical and hippocampal regions, while the original CaMKIIα-Cre T29-1 strain was comparable to WT mice. The mechanisms underlying abnormal neuroinflammation in nestin-Cre and CaMKIIα-iCre are unknown but may be associated with high levels of Cre expression. Our findings are critical to the scientific community and demonstrate that the correct Cre-driver controls must be included in all studies using these mice.

Background Amyloid plaque deposition and axonal degeneration are early events in AD pathogenesis. Aβ disrupts microtubules in presynaptic dystrophic neurites, resulting in the accumulation of impaired endolysosomal and autophagic organelles transporting β-site amyloid precursor protein cleaving enzyme (BACE1). Consequently, dystrophic neurites generate Aβ42 and significantly contribute to plaque deposition. Farnesyltransferase inhibitors (FTIs) have recently been investigated for repositioning toward the treatment of neurodegenerative disorders and block the action of farnesyltransferase (FTase) to catalyze farnesylation, a post-translational modification that regulates proteins involved in lysosome function and microtubule stability. In postmortem AD brains, FTase and its downstream signaling are upregulated. However, the impact of FTIs on amyloid pathology and dystrophic neurites is unknown. Methods We tested the effects of the FTIs LNK-754 and lonafarnib in the 5XFAD mouse model of amyloid pathology. Results In 2-month-old 5XFAD mice treated chronically for 3 months, LNK-754 reduced amyloid plaque burden, tau hyperphosphorylation, and attenuated the accumulation of BACE1 and LAMP1 in dystrophic neurites. In 5-month-old 5XFAD mice treated acutely for 3 weeks, LNK-754 reduced dystrophic neurite size and LysoTracker-Green accumulation in the absence of effects on Aβ deposits. Acute treatment with LNK-754 improved memory and learning deficits in hAPP/PS1 amyloid mice. In contrast to LNK-754, lonafarnib treatment was less effective at reducing plaques, tau hyperphosphorylation and dystrophic neurites, which could have resulted from reduced potency against FTase compared to LNK-754. We investigated the effects of FTIs on axonal trafficking of endolysosomal organelles and found that lonafarnib and LNK-754 enhanced retrograde axonal transport in primary neurons, indicating FTIs could support the maturation of axonal late endosomes into lysosomes. Furthermore, FTI treatment increased levels of LAMP1 in mouse primary neurons and in the brains of 5XFAD mice, demonstrating that FTIs stimulated the biogenesis of endolysosomal organelles. Conclusions We show new data to suggest that LNK-754 promoted the axonal trafficking and function of endolysosomal compartments, which we hypothesize decreased axonal dystrophy, reduced BACE1 accumulation and inhibited amyloid deposition in 5XFAD mice. Our results agree with previous work identifying FTase as a therapeutic target for treating proteinopathies and could have important therapeutic implications in treating AD.