Loss of brain insulin production impairs learning and memory in female mice

Loss of brain insulin production impairs learning and memory in female mice | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Loss of brain insulin production impairs learning and memory in female mice Stella Baehring, Timothy P. O’Leary, Haoning Howard Cen, Danae Holenka, and 13 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7826673/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Diabetes is a metabolic disorder characterized by dysfunctional insulin release and action, and it is a risk factor for Alzheimer’s disease, the most common form of dementia. Alterations in brain insulin signalling and metabolism have been linked with Alzheimer’s disease in multiple studies. It has been previously shown that the ancestral insulin gene, Ins2 , is transcribed locally within the brain. Here we demonstrate that Ins2 mRNA is higher in females than males, and modulated by diet. Moreover, we demonstrate that the Ins2 protein is found in the hippocampus, a brain region with established roles in learning and memory. To specifically determine how insulin produced locally in the brain influences hippocampal function, specifically learning and memory, we used mice with germline Ins2 knockout ( Ins2 −/− ) and the normal complement of wildtype Ins1 alleles. Compensation from the Ins1 gene ensured normal glucose tolerance, normal insulin sensitivity, normal fasting insulin, and normal body weight under these conditions. We assessed visuo-spatial learning and memory in male and female Ins2 −/− and wild-type littermate control mice using the Morris water maze. Learning and memory performance of female Ins2 −/− mice was significantly impaired relative to wild-type mice, whereas the performance of male Ins2 −/− and wild-type mice did not differ. We profiled isolated hippocampi from female Ins2 −/− and littermate control mice using RNA sequencing to provide an unbiased analysis of gene expression differences that underlie these behavioural changes. Cyclin D1 ( Ccnd1 ) was significantly reduced in Ins2 −/− mice, prompting us to examine adult neurogenesis using exogenous mitotic marker EdU and the immature neuronal marker doublecortin (DCX). Collectively, our data demonstrate female-specific roles for brain-derived Ins2 on learning and memory function in mice. Central insulin production learning and memory Alzheimer’s disease hippocampus Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Diabetes is a significant risk factor for dementia ( 1 ) and people with type 2 diabetes can present with cognitive impairment, specifically in learning and memory, similarly to early onset Alzheimer’s disease ( 2 ). People with Alzheimer’s disease often present with hyperglycemia and insulin dysfunction in peripheral organs, and may also have ‘insulin resistance’ in the brain ( 3 ). Post-mortem brain tissue pointed to impaired insulin signalling in the cerebral cortex and hippocampus of Alzheimer’s disease samples compared to control ( 4 ). Studies in humans and animals show that insulin plays a multifactorial role in brain function such as enhancing synaptic plasticity, dendritic spine formation, and increasing neurotransmitter turnover ( 5 , 6 ). Insulin also influences the clearance of the amyloid β peptide and phosphorylation of tau, which are hallmarks of Alzheimer’s disease ( 7 ). Insulin insensitivity in the brain is associated with reduced insulin receptor levels, binding affinity, and disrupts insulin signalling ( 8 ). Insulin signalling is altered in the brains of Alzheimer’s disease patients and alterations in insulin receptor substrates (IRS-1 and IRS-2) is an early pathological characteristic ( 9 ). A recent study found that the loss of insulin signaling in astrocytes exacerbates Alzheimer-like pathology including amyloid plaque accumulation ( 10 ). Astrocytes are crucial in regulating important neuronal functions via insulin, such as dopaminergic signaling and cholesterol synthesis, pointing to the importance of insulin availability in local brain regions ( 11 ). Interestingly, insulin in the brain is produced by neurons, but not glial cells ( 12 , 13 ). Together, these separate lines of evidence suggest that impairment of brain insulin action may underly a component of the pathology of Alzheimer’s disease. Since insulin action on the brain plays a significant role in cognitive functions in humans and various pre-clinical model systems ( 14 – 16 ), some have proposed administering insulin directly to the brain as a treatment bypassing metabolic and brain barriers that decrease bioavailability ( 17 ). Indeed, augmenting insulin signalling tone via intranasal administration of insulin can improve cognitive function in Alzheimer’s disease patients ( 14 , 17 ). Lifestyle interventions that increase insulin sensitivity, such as diet and exercise, improve learning and memory, and are being used as preventative and therapeutic approaches for Alzheimer’s disease ( 6 , 18 , 19 ). Collectively evidence points to a potential role for brain-directed insulin and insulin signalling on cognition. Most studies investigating the effects of insulin on the brain were done under the assumption that insulin in the brain is of pancreatic origin. Although peripheral insulin crosses the blood brain barrier via receptor-mediated transport ( 20 ), accumulating evidence shows that insulin is also produced locally in the brain ( 21 ). Insulin expression has been measured in several brain regions, including cerebral cortex, choroid plexus, cerebellum, and the hippocampus ( 22 – 27 ). The hippocampus is also the main region involved in memory and learning, as well as the first to be affected by neural deterioration in Alzheimer’s disease. In mice, the specificity of these observations has been confirmed with mice lacking insulin 2 ( Ins2 ) gene (human homolog)( 25 , 26 ), which survive due to the presence of an additional, pancreas-specific insulin coding gene insulin 1 ( Ins1 )( 28 ). Notably, we have previously shown that Ins1 does not upregulate its gene expression in the brain to compensate for Ins2 loss ( 25 , 26 ). Insulin synthesis has been reported in neuronal progenitors of the hippocampus and olfactory bulbs ( 12 , 23 ), key sites of adulthood neurogenesis ( 29 , 30 ). Studies on neurogenesis showed that activation of insulin in the neurogenic niche stimulates neural stem cell survival, proliferation, and differentiation ( 31 , 32 ), suggesting that brain insulin and neurogenesis could be linked. Collectively, these data and others ( 25 , 26 ) clearly demonstrate that a small amount of insulin can be synthesized in the brain and that its production occurs in regions highly relevant for learning and memory. However, little is known about the environmental factors that influence insulin production in the brain. Most importantly, whether brain-derived insulin directly influences learning, memory, and neurogenesis has yet to be examined. In this study, we used Ins2 knockout mice to further confirm protein insulin synthesis in the hippocampus and tested how the loss of Ins2 influences learning and memory in both sexes. We determined that Ins2 expression in the mouse hippocampus is modulated by diet and biological sex. We found that female mice lacking Ins2 have deficits in learning and memory. These data provide the first loss-of-function evidence that Ins2 plays a role in hippocampus-controlled behaviour. Materials and Methods Experimental animals Experiments charactering Ins2 expression used five-month-old C57BL/6J mice (Strain 000664). All other experiments used male and female Ins 2 +/+ (wild-type) and Ins2 − /− (knockout) littermates between 3–18 months of age with a background that is primarily C57BL/6J. Ins2 − /− mice were originally made by Duville et al ( 33 ) and have the LacZ/neomycin cassette inserted into the Ins2 locus. Mice were bred in-house, by crossing male and female Ins2 +/− mice. Mice were housed in same-sex groups of 2–4, in cages with a plastic dome and nesting material for enrichment. In some cases, mice were singly housed due to aggression or barbering, but effort was taken to ensure a similar number of singly housed mice were present for each genotype and sex. Except where indicated, food and water were available ad libitum . Mice received a chow diet (total calories = 4.68 kcal/g; 25.3% calories from fat, 19.8% calories from protein, 54.9% calories from carbohydrate; Catalog #5015 Lab Diets, Richmond, IN) or a high fat diet (total calories = 5.56 kcal/g; 58.0% calories from fat, 16.4% calories from protein, 25.5% calories from carbohydrate; Catalog #D12330 Open-Source Diets/Research Diets, New Brunswick, NJ). Mice were housed on a reverse 12 h light dark cycle (lights off at 7:00 am), and all experiments were completed during the dark phase of the light cycle. Transport of mice from the home-cage to behavioral tests was completed with a 500 ml plastic container to reduce handling stress. Some studies used mice with GFP knocked into the endogenous Ins2 gene locus, which have been described and phenotyped elsewhere ( 34 ). All experimental procedures were approved by the UBC animal care committee and adhered to Canadian Council on Animal Care guidelines. Immunohistochemistry and immunoblotting for C peptide, β-gal and pro-insulin Mice were perfused under isoflurane anesthesia with phosphate-buffered saline (PBS), followed by ice-cold 4% paraformaldehyde in PBS. Brains were post-fixed in paraformaldehyde for 4 hrs, and then cryoprotected with 30% sucrose in PBS for 48 hours. Brains were embedded in OCT, frozen on dry-ice and stored at -80ºC, before sectioning at 40µm on a cryostat. Immunostaining was completed using rabbit anti–C-peptide (1:300; Cell Signaling Technology, 4593), rabbit anti–β-gal (1:100; Thermo Fisher Scientific, A-11132) and mouse anti-proinsulin (1:100; R&D Systems, Bio-Techne, MAB13361) primary anti-bodies. Donkey anti-rabbit Cy3 (1:200; Jackson ImmunoResearch, 711-165-152) and donkey anti-mouse Alexa488 (1:200; Jackson ImmunoResearch, 715-545-150) were used as secondary antibodies, followed by nuclear staining with Hoechst 33258 (1ug/ml, Invitrogen, Thermo Fisher Scientific). Sections were mounted onto slides, cover-slipped with VECTASHIELD mounting medium (Vector Laboratories), and imaged using a LSM 800 confocal microscope (Carl Zeiss). Immunoblots of proinsulin were conducted as in our previous publication ( 26 ). Briefly, mice were perfused with 10 mL of ice-cold phosphate-buffered saline before collecting the samples. To extract protein from the pancreas and hippocampus, samples were dissolved in lysis buffers as described previously ( 26 , 35 ). The whole hippocampi of 5 Ins2 +/+ and 5 Ins2 −/− mice were lysed and separated on 15% SDS-PAGE and blotted onto polyvinylidene difluoride membranes (Millipore, IPVH00010) for 30 minutes at 16 V in the transfer buffer (25 mM Tris base, pH 7.4, 192 mM glycine, 10% methanol). The membranes were blocked with 5% skim milk for 1 hour and then incubated with primary antibody against proinsulin (1:1000; Cell Signaling Technology, 8138) or GAPDH (1:10,000; Cell Signaling Technology, 2118) at 4°C overnight. After extensive washing in Tris-buffered saline with 0.1% Tween-20, the membranes were incubated with horseradish peroxidase-conjugated anti-mouse (1:3000; Cell Signaling Technology, 7076S) or anti-rabbit secondary antibody (1:10,000; Thermo Scientific, NCI1460KR) and the bands were visualized using ECL solutions (Thermo Scientific, NCI4080KR; Advansta, K-12045-D50) according to the manufacturer’s instructions. RNA extraction, cDNA synthesis and quantitative polymerase chain reaction Brains were rapidly removed from mice following decapitation after C02 euthanasia. Dissection of the brain occurred on ice, wherein the hippocampus, cerebellum, cerebral cortex, olfactory bulbs, hypothalamus and midbrain (including all residual tissue) were isolated. All brain tissue was then snap frozen on dry ice and stored at -80 ° C. Animals were euthanized in an alternating sequence based on sex to mitigate circadian effects on gene transcription. Dissected brain tissue was homogenized in Trizol (Thermo Fisher) to protect against RNA degradation. Chloroform (50ul) was then added, and the tissue and shaken for 20 seconds before a 15-minute incubation at room temperature. Samples were centrifuged (14000rpm, 15 minutes at 4 ° C) and the aqueous phase was transferred to 70% ethanol. RNA extraction was performed using the Qiagen RNasy minikit (cat #74108 Thermo Fisher), while DNA was synthesized using the qScript cDNA Synthesis Kit (cat#101414-100, Quanta Bioscience). Ins2 mRNA was measured using quantitative polymerase chain reaction (qPCR). The cDNA was subjected to Taqman real-time PCR (StepOnePlus, Applied Biosystems). The following primers were used; Ins2 Taqman reverse 280◊259: GAT CTA CAA TGC CAC GCT TCT G, Ins2 Taqman probe 224◊207: CCT GCT CCC GGG CCT CCA. PCR conditions were 2 min at 50°C, 10 min at 95°C and 40 cycles 15-sec cycles at 95°C and lastly 1 min at 60°C. Samples were normalized to the housekeeping gene beta actin by converting raw values to delta values. The reverse log of delta values were used for all analyses. Water and cDNA from Ins2 − /− tissue were used as negative controls. Metabolic characterization Age-related metabolic phenotyping included measurement of body weight, fasting blood glucose (FBG), glucose tolerance tests (GTT), insulin tolerance tests (ITT) and glucose stimulated insulin secretion (GSIS). For FBG and GTT, mice were fasted for four hours in clean cages prior to glucose measurement. For GTT and ITT, baseline levels of glucose were obtained 15 min prior to intra-peritoneal injections of either glucose (GTT; 4mg/kg) or insulin (ITT; 0.75U/kg, or 1.5U/kg) in PBS. Blood glucose was then measured at 15, 30, 60, 90 and 120 min after injections using OneTouch glucometers and measurement strips. For GSIS, the legs of mice were shaved to facilitate blood collection from the lateral saphenous vein. Mice were fasted for 4 hours, and baseline blood samples were obtained, followed an intra-peritoneal injection of glucose (4mg/kg). Blood samples were then obtained at 15 and 30min post injection. All blood samples were kept on ice, before being centrifuged and serum collected. Insulin levels were then quantified using an enzyme linked immunosorbent assay, based on the manufacturer’s instructions (ALPCO Diagnostics, Salem, NH). Morris water maze The Morris water maze consisted of a white circular pool (110 cm diameter) filled with water to a depth of 16 cm. The water was 23 ° C and made opaque with the addition of non-toxic white tempera paint (Schola, 2002737). A circular escape platform (11.5 cm diameter) was placed in the pool and was 0.5cm below the water surface. The water maze was placed in a diffusely lit room with many extra-maze cues, including large geometric posters adhered to the walls, a desk with a computer, a door, and the geometric layout of the room. The experimenter also served as an extra-maze cue and stood in a similar position for each trial. Mice were placed individually into holding cages lined with paper towel during training. The Anymaze computer tracking system was used to record the movement of mice, and to obtain measures of learning and memory performance. Mice first completed acquisition training which consisted of 6 days of training with 4 trials per day. The escape platform was located in the same location (NW quadrant) across trials. For each trial, mice were placed into the pool at one of four release locations (pseudo-randomly determined) and were given a maximum of 60 sec to reach the escape platform. After reaching the platform, mice remained on the platform for 5–10 seconds before being returned to the holding cage. If mice did not reach the platform in 60 sec, they were guided to the platform by the experimenter. Mice were tested in squads of 4–6, and the inter-trial interval ranged from 2–8 minutes. Performance of mice was measured using latency to locate the escape platform, distance travelled to reach the escape platform, cumulative search error and swim-speed. Cumulative search error was obtained using custom software and consisted of the total distance from the escape platform, summed across all recorded positions of the mouse (5 hz). A correction factor is applied, where the CSE value expected for the optimal swim path (determined by start location and swim speed) is subtracted from the total CSE. The day following acquisition training, mice completed a 60 sec probe trial without the escape platform to measure memory. The number of times mice crossed over the location of the escape platform (annulus crossings) and respective locations in other quadrants of the pool were recorded. The day following the acquisition probe, mice completed an additional day of training (acquisition re-training) to determine the extent of extinction occurred in the probe trial, and to re-establish baseline levels of learning performance. Mice then completed reversal training to measure behavioral flexibility and the learning of a new spatial location of the escape platform. During reversal training the escape platform was moved to the opposite side of the pool. Mice completed 6 days of training with 4 trials per day, using the same procedure as acquisition training. Following reversal training, mice completed a 60-sec probe trial to measure memory for the new escape platform location. Anxiety-like behaviour Mice were tested on the open-field, light-dark box and elevated plus maze to measure locomotor activity and anxiety-like behaviour. All trials were 10 min in length and the test apparatuses were cleaned with 70% ethanol between trials. After trials, mice were placed in a clean holding cage, while the remaining cagemates completed their trials. During testing mice were brought in groups of 7–9 mice into a dark and quiet anteroom. Mice habituated to the anteroom for at least 25 min before behavioral testing. Testing was completed in a different room than the Morris water maze training. Tests were completed with 7-13- day intervals, to reduce potential cross-test habituation. Movement of mice in the open-field, light-dark box and elevated plus maze were recorded with the Ethovision (Noldus) video tracking system. Species typical behaviors were measured with the Boris (Friad and Gamba, 2016) event scoring software, using video recordings from trials. The open field consisted of a box (70 x 70 cm) made of transparent Plexiglas, with 23 cm high walls. For each trial, mice were placed into the one of four corners of the open field (pseudo-randomly determined). Locomotor activity was measured with distance travelled, while anxiety-like behaviour was assessed with entries into the center of the open-field (9th of area). Species typical behaviors were also recorded including number of rears, grooming duration, and freezing duration. The light-dark box was constructed from Plexiglas and was divided into two chambers that were either brightly (light zone) or dimly lit (dark zone). A single light placed above the light zone provided lighting in the light-dark box and testing room. For each trial, mice were placed in the light zone facing the opening to the dark zone. The same behavioural measures recorded in the open field were also recorded in the light-dark box, except that time in the light chamber was used as a measure of anxiety-like behaviour. The elevated plus maze consisted of two pairs of arms, each attached to a central square. One pair of arms had opaque walls around the edge (closed arms), while the other pair of arms had no walls (open arms). A small rim (0.5 cm) was present around the edge of the open arms to prevent mice from falling. A single light provided dim lighting for the elevated plus maze and testing room. For each trial, mice were placed onto the central hub of the elevated plus maze, facing a closed arm. The same behavioral measurements that were recorded in the open field were also recorded in the elevated plus maze, except that time in the open arms was used as measure of anxiety-like behaviour. The number of times mice dipped their heads over the edge of the open arms (head-dips) was also recorded as a species typical measure of anxiety-like behaviour. Bulk hippocampal RNA sequencing Removal and dissection of brains were completed as described above for qPCR. Animals were euthanized in an alternating sequence based on genotype to prevent potential circadian effects on gene transcription. Body weight and un-fasted blood glucose (measured the from the neck after decapitation) were also measured to ensure no pre-existing differences between Ins2 +/+ and Ins2 −/− mice at the time of euthanasia. Sample quality control was performed using the Agilent 2100 Bioanalyzer. Qualifying samples were then prepped following the standard protocol for the NEBnext Ultra ii Stranded mRNA (New England Biolabs). Sequencing was performed on the Illumina NextSeq 500 with Paired End 43bp × 43bp reads. Sequencing data was demultiplexed using Illumina's bcl2fastq2. De-multiplexed read sequences were then aligned to the Mus musculous (PAR-masked)/mm10 reference sequence using STAR aligner. Quality control led to one outlier sample (genotype -/-) being removed. Hierarchical clustering and principal component analysis (PCA) identified 3 additional outliers (1 Ins2 +/+ and 2 Ins2 ) that were removed. Genes with expression below 0.5 counts per million were dropped from the analysis. Differential expression analysis associated with genotype was performed using limma-trend ( 36 ) after quantile normalization or using DESeq2 ( 37 ), which had consistent results. The DESeq2 results were used in the figures and Gene Set Enrichment Analysis (GSEA). GSEA was performed with clusterProfiler R package using Gene Ontology (Biological Process) knowledgebase (FDR < 0.0001) ( 38 ), and simplify() function was used to remove redundant GO terms ( https://guangchuangyu.github.io/2015/10/use-simplify-to-remove-redundancy-of-enriched-go-terms/ ). Quantification of adult hippocampal neurogenesis Twenty-month-old Ins2 +/+ and Ins2 −/− mice were given EdU in drinking water (0.25 mg/ml) for four weeks. The EdU water was freshly prepared every week and was given to mice in 120 ml bottles. Body weight and water consumption were recorded every week to ensure consumption of EdU water. Mice with given Edu-water for 4 weeks, after which they were transcardially perfused with PBS (0.1M) followed by 4% paraformaldehyde. Brains were extracted, post-fixed in 4% paraformaldehyde overnight, and then transferred to a 30% sucrose solution for cryoprotection. Brains were sliced into 40 µm coronal sections using a cryostat at -20°C. Sections were collected in series of 6 through the rostral-caudal extent of the hippocampus and stored in anti-freeze solution (ethylene glycol, glycerol, and 0.1 M PBS) at -20°C until immunostaining. To measure adult neurogenesis a series of sections were stained for both EdU and doublecortin (DCX). Staining for EdU was completed with the Click-iT Plus EdU Alexa Fluor 647 Imaging Kit (Thermo Fisher Scientific). Staining for DCX was completed via Immunohistochemistry using a Rabbit anti-DCX IgG polyclonal antibody (cat# 326 003 Thermo Scientific) and a Donkey anti-Rabbit IgG H + L Alexa Fluor 488 secondary antibody (cat# A-21206 Invitrogen). We used vertical sections of intestinal villi from EdU treated animals as positive controls for EdU labelling. Stained sections were imaged with a Zeiss LSM 880 confocal microscope. Dorsal and ventral regions of the dentate gyrus were imaged using a 20X objective. Image stacks through the section were obtained, stitched together and maximum intensity projections were then used to count EdU and DCX-positive cells. The experimenter was blind to all conditions during quantification, and six sections were analyzed per animal. The sampling area within the dentate gyrus was also obtained, and cell counts normalized to sampling area. Statistical Analysis For data using WT C57BL/6J mice, data were analyzed with sex as a factor when possible. Subsequent data for experiments using Ins2 +/+ and Ins2 −/− mice were analyzed separately for each sex and age to increase statistical power to detect differences due to genotype. For tolerance tests and glucose stimulated insulin secretion, mixed design ANOVAs (2 x 6 or 2 x 3, respectively) were used with genotype as a between subject factor and sampling time as a within subject factor. For acquisition and reversal training in the MWM, analyses were completed separately for days 1–3 and days 4–6, which reflect initial learning and near-asymptotic performance respectively. Measures of learning performance were analyzed with mixed design ANOVAs (2 x 3), using genotype as a between subject factor and day of training as within subject factor. All other analyses were completed using one-way between subject ANOVAs comparing genotypes. Post-hoc tests were completed using unpaired t-tests, with Bonferonni corrections applied independently for each statistical test. Results Ins2 is produced in the brain and its expression is sexually dimorphic and modulated by diet We previously identified Ins2 mRNA and insulin protein in the mouse brain (25), but did not show hippocampal C-peptide and proinsulin immunoreactivity that was specific (i.e. absent in Ins2 null brains). Here, we showed that labelling of C-peptide, a byproduct and marker of insulin synthesis, can be observed within the hippocampus of Ins2 +/+ mice, particularly within the dentate gyrus-CA3 pathway ( Figure 1A) . This was confirmed with immunoblots showing a small but clear Ins2 -specific band for proinsulin in hippocampus lysates, as well as the expected prominent band in pancreas ( Figure 1B ). As expected, no insulin protein was found in hippocampi from Ins 2 -/- because the non-ancestral Ins1 gene is not robustly expressed in the mouse brain (25). Complementary immunostaining in the hippocampus of wildtype Ins2 +/+ mice revealed clear labelling of proinsulin within the CA3 and choroid plexus, and this staining was largely absent in Ins2 -/- mice ( Figure 1C ). Importantly, in Ins2 +/- mice, proinsulin labelling colocalized with b-gal labelling, a proxy for Ins2 gene activity and mRNA expression. We were also able to visualize green fluorescence in cultured hippocampal neurons from mice with GFP knocked into the endogenous Ins2 locus, which we have previously shown correlates with Ins2 mRNA, pre-mRNA, and proteins levels (34) ( Figure 1D ). Together, these data provide further strong evidence that the brain can locally produce insulin. We next measured Ins2 mRNA levels across six brain regions (hippocampus, cerebellum, cerebral cortex, olfactory bulbs, hypothalamus, all residual tissue was classified as midbrain) in male and female wildtype C57BL/6J mice. Brain Ins2 expression was highest in the hippocampus and cerebellum of both sexes ( Figure 2A ). Ins2 mRNA was consistently higher in females than males (F(1,14) = 7.58, p < 0.05), and clear differences were observed across brain regions (F(5,70) = 18.42, p< 0.0001). Sex differences in Ins2 expression were more prominent within certain brain regions (interaction, (F(5,70) = 2.67, p < 0.05)), where females showed higher expression of Ins2 than males in the hippocampus and cerebellum (post-hoc, p < 0.005) ( Figure 2A ). Because the expression of Ins2 and levels of insulin in the pancreas are modulated by diet, so we next tested whether a similar regulation occurs in brain. These independent experiments confirmed higher expression of Ins2 in the hippocampus of female mice relative to other brain regions (i.e. cerebral cortex) and relative to males (Figure 2B,C) . In female mice, Ins2 expression was reduced after being fed a high-fat diet when compared to a control diet, whereas diet did not affect Ins2 expression in male mice ( Figure 2B,C ). These data demonstrate that Ins2 mRNA levels differ between brain regions, sexes, and dietary contexts. Ins1 is sufficient to preserve normal peripheral metabolism in the absence of Ins2 Before using Ins2 -/- mice to examine the physiological roles of Ins2 in the brain, it was critical to confirm that this was independent of significant whole-body metabolic effects. Previous studies have shown that mice lacking Ins2 alone for life have relatively normal metabolic physiology (25), supporting previous observations that Ins1 can compensate for loss of Ins2 (39). Previous work only focused on males, so we next characterized metabolism in littermate male and female Ins2 -/- mice side-by-side to confirm that Ins1 can compensate for loss of Ins2 in both sexes. In addition, we examined mice from 3-16 months of age to determine if compensation by Ins1 can be maintained during aging. For both males and females, body weight of Ins2 -/- mice did not differ from Ins2 +/+ mice at any age (Figure 3A,B) . Fasted (4 hours) blood glucose levels were largely similar between Ins2 -/- and Ins2 +/+ mice and differed only in males at the oldest age (16-18 months) tested ( Figure 3C,D) . Ins2 -/- and Ins2 +/+ mice did not differ in glucose tolerance tests (Figure 3E-H) nor in insulin tolerance tests at either 8 or 16-18 months of age (Figure 3I-L) . Ins2 -/- and Ins2 +/+ mice also showed similar glucose stimulated insulin secretion ( Figure 3M-P ), although there was a trend for higher insulin in female Ins2 +/+ mice compared with Ins2 -/- mice at the oldest age measured ( Figure 3N ). Together, these data suggest that I ns1 can produce a near-complete compensation for the absence of Ins2 in both male and female mice on this normal chow diet, and this compensation can be maintained through aging up to 16-18 months. Loss of Ins2 produces visuo-spatial learning and memory impairments in female, but not male mice The hippocampus is critical for a variety of different cognitive processes, and in rodents is perhaps most well-known for its role in visuo-spatial navigation and memory (40). Given Ins2 expression is prominent within the hippocampus relative to other neuronal populations, we next examined whether hippocampal-dependent behaviour was affected by Ins2 -/- ablation in the Morris water maze, a commonly used and well-validated test of visuo-spatial learning and memory in rodents (41). Given that the sexual dimorphism in hippocampal Ins2 expression, we compared aged female ( Figure 4A-F ) and male ( Figure 4G-L ) littermates of both genotypes. During acquisition training, female Ins2 -/- and Ins2 +/+ mice did not differ on measures of learning including latency ( Figure 4A ), distance ( Figure 4D ), or cumulative search error ( Figure 4B ). During the acquisition probe memory test, however, female Ins2 -/- completed fewer crossings over the location of the escape platform (annulus crossings) than Ins2 +/+ mice (Figure 4C) . We then provided an additional day of acquisition training (re-training, RT) to reduce potential extinction effects, and to provide an additional indirect measure of memory by assessing retention of learning performance from the last day of acquisition training (2 days prior). On the re-training day, female Ins2 -/- performed worse than Ins2 +/+ , mice on all measures of learning performance ( Figure 4A,B,D ). During the first 3 days of reversal training, female Ins2 -/- and Ins2 +/+ did not differ on measures of learning performance; over the last 3 days of reversal training, however, female Ins2 -/- mice performed worse than Ins2 +/+ mice on all measures of learning performance ( Figure 4A,B,D ). During the reversal probe memory test, Ins2 -/- mice completed fewer crossings over the escape platform location than Ins2 +/+ mice (Figure 4D) . We also measured swim speed of mice throughout acquisition and reversal training, which can reflect motivation to find the escape platform. Although, female Ins2 -/- mice swam quicker than Ins2 +/+ mice during the first 3 days of acquisition training, this difference was not present during the last 3 days of acquisition training, or at any point during reversal training (Figure 4E) . Given that differences in learning performance were not present in acquisition training, we can conclude that differences in swim-speed do not account for the impaired learning and memory observed in Ins2 -/- mice. Taken together, these results indicate impaired learning and memory for the reversal escape platform location in Ins2 -/- mice. We also assessed learning and memory in male Ins2 -/- and Ins2 +/+ mice in the Morris water maze but did not find any differences in learning performance during either acquisition or reversal training (Figure 4G,H,J,K). We also did not observe differences in memory performance during acquisition or reversal memory probe tests (Figure 4I,L) . Thus, unlike female mice, loss of Ins2 -/- in male mice is not sufficient to disrupt visuo-spatial learning or memory. Loss of Ins2 does not influence anxiety-like behaviour or locomotor activity. The function of the hippocampus is known to differ along its dorsal-ventral axis, with the dorsal hippocampus being involved in visuo-spatial navigation and memory, and the ventral hippocampus being involved more so in anxiety-related behaviour (42, 43). Therefore, we examined anxiety-like behaviour in Ins2 -/- mice, to provide a more complete picture for the role of Ins2 on hippocampal function. Ins2 -/- and Ins2 +/+ mice were tested on three commonly used tests of anxiety-like behaviour: the open-field, light-dark box, and elevated-plus maze. All three tests were effective in producing marked avoidance of anxiogenic areas, with mice spending less time in anxiogenic regions than expected by chance (Supplemental Figure 1) . For both males and females, however, Ins2 -/- and Ins2 +/+ mice did not differ in anxiety-related behaviour on the OF, LDB or EPM. We also extended our analysis to include species-typical measures of exploration and anxiety-like behaviour (i.e. grooming, rearing, freezing, etc.), but did not find any differences between Ins2 -/- and Ins2 +/+ mice (Supplemental Figure 1) . Each of the OF, LDB and EPM can also be used to assess locomotor activity within novel environments, however, we did not observe any differences between Ins2 -/- and Ins2 +/+ mice in locomotor activity in either males or females (Supplemental Figure 1) . Thus, the effects of Ins2 knockout in female mice appear to be specific to learning and memory in the Morris water maze. Transcriptomics of hippocampi from aged female Ins2 -/- mice identifies cyclin D1 as downregulated We conducted bulk hippocampal RNA-seq in aged female Ins2 -/- mice to identify potential molecular mechanisms that could underlie the defects in learning and memory. We started with 11 samples in each group, but three samples were deemed outliers based on hierarchical clustering and PCA and were removed from the analysis ( Figure 5A,B ). We confirmed the absence of Ins2 mRNA in female Ins2 -/- mice (Figure 5C ). Then, we identified 6 downregulated and 13 upregulated genes in Ins2 -/- mice ( Figure 5D ). Among them, the cell cycle regulatory gene Ccnd1 and the stem cell regulatory long non-coding RNA Gm26793 (44) were the most significantly downregulated and upregulated genes, respectively. It should be noted that these genes were robustly altered even when the outlier samples were included. Gene set enrichment analysis showed that the (semi-redundant) gene ontology terms “aerobic respiration” and “oxidative phosphorylation” were the most significantly upregulated pathways ( Figure 5E ). Insulin is known to be a regulator of oxidative phosphorylation in many tissues, including the brain (45). The gene ontology term “epithelium development” was the only significantly downregulated pathway ( Figure 5E ). Together, these relatively well-powered transcriptomic studies point to potential molecular mechanisms underlying the effects of insulin signalling the in female rodent hippocampus. Neurogenesis in Ins2 -/- mice. Given that Ins2 -/- mice had reduced Ccnd1 mRNA, a gene with known roles in neurogenesis, we next examined if hippocampal neurogenesis was altered in aged Ins2 -/- mice using EdU, a thymidine analogue that incorporates into proliferating cells, and by identifying immature neurons via doublecortin. We first confirmed the robustness of our staining by confirming Edu positive cells in intestinal villi, a region with high levels of cell proliferation ( Supplemental Figure 2A) . In female brains, the densities of Edu positive cells and DCX positive cells did not differ between Ins2 -/- and Ins2 +/+ mice, in either the dorsal or ventral dentate gyrus ( Supplemental Figure 2B-D ). In male brains, Ins2 -/- mice had a larger density of EdU positive cells in the dorsal (F(1,10) = 5.57, p<0.05)( Supplemental Figure 2C ), but not ventral hippocampus ( Figure 2D ). One could speculate that this may represent compensation that may prevent male Ins2 -/- mice from exhibiting defects in learning and memory. The density of DCX cells did not differ between Ins2 -/- and Ins2 +/+ male mice in either the dorsal or ventral hippocampus ( Supplemental Figure 2E,F ). Together, these data suggest that adult neurogenesis, measured at this late stage of life, is not a major factor in differences in learning and memory in female Ins2 -/- mice. Discussion The objectives of this study were to further elucidate the expression of insulin in the hippocampus and to determine the role of brain-produced insulin on hippocampus-dependent learning and memory. Using Ins2 knockout mice devoid of hippocampal pro-insulin, but with limited effects on peripheral metabolism, we defined a female-specific role for brain-insulin production in learning and memory. Given that Ins2 is expressed in both the brain and pancreas, it was important to assess the effect of Ins2 knockout on peripheral metabolism, which could theoretically affect brain function indirectly. In-depth metabolic characterization of Ins2 −/− and Ins2 +/+ mice showed that they were similar in most measures of metabolism, including body weight, glucose tolerance and insulin tolerance tests. Together these results suggest that the changes in Ins2 expression and behavioral impairments observed in Ins2 −/− mice are likely not due to peripheral metabolic dysfunction. The relatively few metabolic changes in Ins2 −/− mice indicates that the Ins1 allele can produce a near-complete compensation for the loss of Ins2 in the context of the diet and housing conditions used in our study, and that this compensation can be maintained through aging. The mechanism for this compensation is largely unexplored but is likely supported by elevated Ins1 expression within the pancreas and pancreatic β-cell hyperplasia ( 39 ). This is likely due to inter-allelic redundancy and compensation within the pancreas, where both alleles contribute to insulin production, and may be especially important under periods of metabolic stress. Indeed, we found that metabolic changes in Ins2 −/− mice were present only at 16–18 months, when the accumulated stressors associated with aging were expected to be the largest. However, given that Ins2 is also expressed in the brain, we cannot rule out that loss of brain-produced insulin also contributed to the few metabolic differences observed in Ins2 −/− mice. To the best of our knowledge, this is the first time cognitive behaviour has been examined in the context of genetic insulin reduction. Our study was also unique in that we deleted only Ins2 , resulting in brain-selective insulin knockout. To contextualize our knockout model, we extended on our previous work that provided conclusive evidence, using knockout controls, that insulin is produce locally in the brain, and more prominently in the hippocampus relative to other brain regions ( 25 ). Other work has examined Ins2 expression and insulin synthesis across different brain regions in a plethora of animal models ( 23 , 24 , 26 , 46 – 48 ). Studies that include hippocampal regions consistently report elevated levels of insulin expression relative to other brain regions suggesting a role for local Ins2 production in hippocampal function ( 12 , 25 ). This is further supported by in vitro studies reporting Ins2 expression and insulin synthesis within cultured primary hippocampal neurons ( 49 , 50 ). Importantly, we confirmed that Ins2 protein is found in the hippocampus of wildtype but not Ins2 knockout mice. Moreover, we show that female mice show higher Ins2 expression than males in both the hippocampus and cerebellum. There are currently very few studies that examine sex differences in Ins2 expression within the brain, and to the best of our knowledge, this is the first study to show that Ins2 expression within the hippocampus is sexually dimorphic. It had been proposed that insulin within the brain can be modulated by environmental factors including diet ( 51 ), and here we show that environmental effects on Ins2 brain expression are also sexual dimorphic. Women are at higher risk to develop early onset Alzheimer’s disease ( 52 ). Consistent with this, we found evidence for a female-specific role of brain-derived insulin in hippocampal-dependent learning and memory, particularly in spatial-visual tasks. It has been suggested that male rodents have better spatial abilities like spatial working memory and orientation, whereas females are better in location memory ( 53 ). In our experiments, there were not significant differences in learning or memory measures between male and female wildtype mice. When compared to other studies, it should be noted that our experiments were conducted on older mice, which is more relevant to Alzheimer’s disease. The ability to perform tasks of learning can be influenced by non-cognitive factors, thus potentially confounding measures of learning and memory ( 54 ). For example, performance on the Morris water maze can be influenced by differences in stress reactivity, motivation to locate the escape platform, impaired visual ability and impaired swimming ability due to motor dysfunction ( 55 ). The similar anxiety-like behavioural profiles in female Ins2 −/− and Ins2 +/+ mice suggests that our reported differences in visuo-spatial learning and memory are not likely due to pre-existing differences in stress reactivity within the water maze. This is further supported by the lack of differences in swim-speed observed in the water maze indicating unimpaired swimming ability and similar motivation to locate the escape platform. Therefore, these results provide additional confidence that differences in learning and memory on the water maze in female Ins2 −/− mice were not due to potential non-cognitive factors influencing performance. The functional role of the hippocampus has been proposed to differ along the dorso-ventral axis, with the dorsal hippocampus involved moreso in cognitive function, and the ventral hippocampus involved in stress and emotionality. Given that female Ins2 −/− mice showed impaired learning and memory, we also examined if differences extended to anxiety-like behaviour. We did not observe any differences in anxiety-like behaviour in either male or female Ins2 −/− mice. We should note, however, that over-expressing Ins2 within the choroid plexus is sufficient to reduce anxiety-like behaviour, but only when mice are given a high-fat diet ( 27 ). This may indicate a role for Ins2 in regulating other brain behaviours, including anxiety, specifically during metabolic challenges. The Ins2 knockout mouse model in this study is most appropriate to directly test the effects of endogenous insulin of the brain to date. Silencing the Ins2 gene, while keeping Ins1 intact, allows to specifically examine the causal effects of insulin synthesis in the brain independent of peripheral insulin. However, our model does not come without its own limitations. The Ins2 gene is also, in addition to the pancreas and brain, expressed in the thymus. The thymus is most known for its role in immunity and insulin deletion in the thymus can induce autoimmune diabetes ( 56 ). Thus, with a global Ins2 knockout, it cannot be entirely excluded that the results of this study are exclusively due to brain-insulin loss but may indirectly relate to immune dysfunction. Generating tissue-specific Ins2 knockout could be used to circumvent this limitation, but our efforts to do so have been unsuccessful. Another limitation of our study was that the loss of Ins2 was lifelong. Future studies could employ a floxed Ins2 allele and brain-specific Cre viral vectors or drug-inducible transgenes in mice to remove Ins2 in the adult brain, in a region-specific manner. Other methods of acutely testing for local brain insulin ablated insulin-expressing neurons with the beta-cell toxin streptozotocin ( 57 – 59 ). However, toxic drugs may also affect other areas of the brain, and it would be difficult to say whether effects on behavior are due to brain-derived insulin loss or generalized toxicity. Our work also does not identify the cellular target(s) of local insulin production, not their molecular mechanisms. Insulin receptors have been shown to mediate important roles of insulin in many neuronal cell populations in the brain, including the hippocampus ( 60 – 62 ). Insulin also acts directly though insulin receptors to modulate the activity of non-neuronal brain cells including astrocytes ( 63 – 65 ). Indeed, inducible loss of insulin action via insulin receptor and IGF1 receptor deletion in astrocytes significantly worsened multiple Alzheimer’s pathologies in a mouse model ( 10 ). In a recent study, cerebrovascular insulin receptors were also shown to be deficient in Alzheimer’s disease ( 66 ). Clearly, more research will be required to fully elucidate local insulin-insulin receptor trophic circuits across multiple cell types in the brain. In conclusion, we confirmed that Ins2 is produced in the hippocampus and discovered a significant sex difference in mRNA levels. This study used a genetic model of Ins2 deletion to assess the influence of brain insulin production, in the absence of major metabolic disturbances. We also showed for the first time that insulin produced in the brain plays an important role in spatial navigation in female mice. Declarations Declaration of Interests: The authors declare no competing interests. Funding: Work was supported by a Brain Canada Multi-Investigator grant to J.D.J., S.X.B., and P.P. Acknowledgments: We thank many colleagues for helpful discussions. Data availability: RNA sequencing data are online. All other data are available on request. Author Contributions: S.B. designed studies, performed experiments, analyzed/interpreted data, and co-wrote the manuscript. T.O. designed studies, performed experiments, analyzed/interpreted data, and co-wrote the manuscript. H.H.C. analyzed bioinformatic data and helped write the manuscript. D.H. performed experiments, analyzed/interpreted data. K.K. performed experiments. S.A. performed experiments. D.S. performed experiments. M.B. performed experiments and analyzed bioinformatic data. H.L. performed experiments. A.E.M. performed experiments, analyzed/interpreted data. H.M. performed experiments, analyzed/interpreted data. M.M.P. performed experiments, analyzed/interpreted data S.S. performed experiments, analyzed/interpreted data. P.P. supported bioinformatic studies, interpreted data, and edited the manuscript S.X.B. designed studies, interpreted data, and edited the manuscript E-K. K. analyzed data and provided a key reagent. J.D.J. conceived the project, designed studies, interpreted data, co-wrote the manuscript, and is the guarantor of this work. References Hölscher, C. (2011) Diabetes as a risk factor for Alzheimer's disease: insulin signalling impairment in the brain as an alternative model of Alzheimer's disease. Portland Press Ltd. Ryan, C. M., and Geckle, M. (2000) Why is learning and memory dysfunction in Type 2 diabetes limited to older adults? Diabetes/metabolism research and reviews 16 , 308-315 De Felice, F. G., Lourenco, M. V., and Ferreira, S. T. (2014) How does brain insulin resistance develop in Alzheimer's disease? Alzheimer's & Dementia 10 , S26-S32 Talbot, K., Wang, H.-Y., Kazi, H., Han, L.-Y., Bakshi, K. P., Stucky, A., Fuino, R. L., Kawaguchi, K. R., Samoyedny, A. J., Wilson, R. S., Arvanitakis, Z., Schneider, J. A., Wolf, B. A., Bennett, D. A., Trojanowski, J. Q., and Arnold, S. E. (2012) Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. Journal of Clinical Investigation 122 , 1316-1338 Forner, S., Baglietto-Vargas, D., Martini, A. C., Trujillo-Estrada, L., and LaFerla, F. M. (2017) Synaptic Impairment in Alzheimer's Disease: A Dysregulated Symphony. Trends Neurosci 40 , 347-357 Kellar, D., and Craft, S. (2020) Brain insulin resistance in Alzheimer's disease and related disorders: mechanisms and therapeutic approaches. Lancet Neurol 19 , 758-766 Laws, S. M., Gaskin, S., Woodfield, A., Srikanth, V., Bruce, D., Fraser, P. E., Porter, T., Newsholme, P., Wijesekara, N., and Burnham, S. (2017) Insulin resistance is associated with reductions in specific cognitive domains and increases in CSF tau in cognitively normal adults. Scientific reports 7 , 1-11 Ma, L., Wang, J., and Li, Y. (2015) Insulin resistance and cognitive dysfunction. Clinica chimica acta 444 , 18-23 Talbot, K., Wang, H.-Y., Kazi, H., Han, L.-Y., Bakshi, K. P., Stucky, A., Fuino, R. L., Kawaguchi, K. R., Samoyedny, A. J., and Wilson, R. S. (2012) Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. The Journal of clinical investigation 122 , 1316-1338 Chen, W., Huang, Q., Lazdon, E. K., Gomes, A., Wong, M., Stephens, E., Royal, T. G., Frenkel, D., Cai, W., and Kahn, C. R. (2023) Loss of insulin signaling in astrocytes exacerbates Alzheimer-like phenotypes in a 5xFAD mouse model. Proceedings of the National Academy of Sciences 120 , e2220684120 Cai, W., Xue, C., Sakaguchi, M., Konishi, M., Shirazian, A., Ferris, H. A., Li, M. E., Yu, R., Kleinridders, A., and Pothos, E. N. (2018) Insulin regulates astrocyte gliotransmission and modulates behavior. The Journal of clinical investigation 128 , 2914-2926 Kuwabara, T., Kagalwala, M. N., Onuma, Y., Ito, Y., Warashina, M., Terashima, K., Sanosaka, T., Nakashima, K., Gage, F. H., and Asashima, M. (2011) Insulin biosynthesis in neuronal progenitors derived from adult hippocampus and the olfactory bulb. EMBO Mol Med 3 , 742-754 Devaskar, S. U., Giddings, S. J., Rajakumar, P. A., Carnaghi, L. R., Menon, R. K., and Zahm, D. S. (1994) Insulin gene expression and insulin synthesis in mammalian neuronal cells. J Biol Chem 269 , 8445-8454 Freiherr, J., Hallschmid, M., Frey, W. H., Brünner, Y. F., Chapman, C. D., Hölscher, C., Craft, S., De Felice, F. G., and Benedict, C. (2013) Intranasal insulin as a treatment for Alzheimer’s disease: a review of basic research and clinical evidence. CNS drugs 27 , 505-514 Zhang, H., Hao, Y., Manor, B., Novak, P., Milberg, W., Zhang, J., Fang, J., and Novak, V. (2015) Intranasal insulin enhanced resting-state functional connectivity of hippocampal regions in type 2 diabetes. Diabetes 64 , 1025-1034 Li, X., Yang, Y., Bai, X., Wang, X., Tan, H., Chen, Y., Zhu, Y., Liu, Q., Wu, M. N., and Li, Y. (2024) A brain-derived insulin signal encodes protein satiety for nutrient-specific feeding inhibition. Cell Rep 43 , 114282 Born, J., Lange, T., Kern, W., McGregor, G. P., Bickel, U., and Fehm, H. L. (2002) Sniffing neuropeptides: a transnasal approach to the human brain. Nature Neuroscience 5 , 514-516 Ko, Y., and Chye, S. M. (2020) Lifestyle intervention to prevent Alzheimer’s disease. Reviews in the Neurosciences 31 , 817-824 Braak, H., and Braak, E. (1995) Staging of Alzheimer's disease-related neurofibrillary changes. Neurobiology of aging 16 , 271-278 Gray, S. M., Aylor, K. W., and Barrett, E. J. (2017) Unravelling the regulation of insulin transport across the brain endothelial cell. Diabetologia 60 , 1512-1521 Dakic, T., Jevdjovic, T., Lakic, I., Ruzicic, A., Jasnic, N., Djurasevic, S., Djordjevic, J., and Vujovic, P. (2023) The Expression of Insulin in the Central Nervous System: What Have We Learned So Far? International Journal of Molecular Sciences 24 , 6586 Molnar, G., Farago, N., Kocsis, A. K., Rozsa, M., Lovas, S., Boldog, E., Baldi, R., Csajbok, E., Gardi, J., Puskas, L. G., and Tamas, G. (2014) GABAergic neurogliaform cells represent local sources of insulin in the cerebral cortex. J Neurosci 34 , 1133-1137 Nemoto, T., Toyoshima-Aoyama, F., Yanagita, T., Maruta, T., Fujita, H., Koshida, T., Yonaha, T., Wada, A., Sawaguchi, A., and Murakami, M. (2014) New insights concerning insulin synthesis and its secretion in rat hippocampus and cerebral cortex: amyloid-beta1-42-induced reduction of proinsulin level via glycogen synthase kinase-3beta. Cell Signal 26 , 253-259 Mazucanti, C. H., Liu, Q. R., Lang, D., Huang, N., O'Connell, J. F., Camandola, S., and Egan, J. M. (2019) Release of insulin produced by the choroid plexis is regulated by serotonergic signaling. JCI Insight 4 Mehran, A. E., Templeman, N. M., Brigidi, G. S., Lim, G. E., Chu, K. Y., Hu, X., Botezelli, J. D., Asadi, A., Hoffman, B. G., Kieffer, T. J., Bamji, S. X., Clee, S. M., and Johnson, J. D. (2012) Hyperinsulinemia drives diet-induced obesity independently of brain insulin production. Cell Metab 16 , 723-737 Lee, J., Kim, K., Cho, J. H., Bae, J. Y., O'Leary, T. P., Johnson, J. D., Bae, Y. C., and Kim, E. K. (2020) Insulin synthesized in the paraventricular nucleus of the hypothalamus regulates pituitary growth hormone production. JCI Insight 5 Mazucanti, C. H., Kennedy, V., Jr., Premathilake, H. U., Doyle, M. E., Tian, J., Liu, Q. R., O'Connell, J., Camandola, S., and Egan, J. M. (2023) AAV5-mediated manipulation of insulin expression in choroid plexus has long-term metabolic and behavioral consequences. Cell Rep 42 , 112903 Duvillie, B., Cordonnier, N., Deltour, L., Dandoy-Dron, F., Itier, J. M., Monthioux, E., Jami, J., Joshi, R. L., and Bucchini, D. (1997) Phenotypic alterations in insulin-deficient mutant mice. Proc Natl Acad Sci U S A 94 , 5137-5140 Berdugo-Vega, G., Dhingra, S., and Calegari, F. (2023) Sharpening the blades of the dentate gyrus: how adult-born neurons differentially modulate diverse aspects of hippocampal learning and memory. Embo J 42 , e113524 Hitoshi, S., Mandyam, C. D., and Shetty, A. K. (2023) Editorial: Advances in adult neurogenesis. Front Neurosci 17 , 1307844 Brooker, G. J., Kalloniatis, M., Russo, V. C., Murphy, M., Werther, G. A., and Bartlett, P. F. (2000) Endogenous IGF‐1 regulates the neuronal differentiation of adult stem cells. Journal of neuroscience research 59 , 332-341 Åberg, M. A., Åberg, N. D., Palmer, T. D., Alborn, A.-M., Carlsson-Skwirut, C., Bang, P., Rosengren, L. E., Olsson, T., Gage, F. H., and Eriksson, P. S. (2003) IGF-I has a direct proliferative effect in adult hippocampal progenitor cells. Molecular and Cellular Neuroscience 24 , 23-40 Duvillié, B., Cordonnier, N., Deltour, L., Dandoy-Dron, F., Itier, J.-M., Monthioux, E., Jami, J., Joshi, R. L., and Bucchini, D. (1997) Phenotypic alterations in insulin-deficient mutant mice. Proceedings of the National Academy of Sciences 94 , 5137-5140 Chu, C. M. J., Modi, H., Ellis, C., Krentz, N. A. J., Skovso, S., Zhao, Y. B., Cen, H., Noursadeghi, N., Panzhinskiy, E., Hu, X., Dionne, D. A., Xia, Y. H., Xuan, S., Huising, M. O., Kieffer, T. J., Lynn, F. C., and Johnson, J. D. (2022) Dynamic Ins2 Gene Activity Defines beta-Cell Maturity States. Diabetes 71 , 2612-2631 Kim, S., Kim, N., Park, S., Jeon, Y., Lee, J., Yoo, S. J., Lee, J. W., Moon, C., Yu, S. W., and Kim, E. K. (2020) Tanycytic TSPO inhibition induces lipophagy to regulate lipid metabolism and improve energy balance. Autophagy 16 , 1200-1220 Ritchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., and Smyth, G. K. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 43 , e47-e47 Love, M. I., Huber, W., and Anders, S. (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15 , 550 Yu, G., Wang, L. G., Han, Y., and He, Q. Y. (2012) clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 16 , 284-287 Leroux, L., Desbois, P., Lamotte, L., Duvillie, B., Cordonnier, N., Jackerott, M., Jami, J., Bucchini, D., and Joshi, R. L. (2001) Compensatory responses in mice carrying a null mutation for Ins1 or Ins2. Diabetes 50 Suppl 1 , S150-153 Vorhees, C. V., and Williams, M. T. (2014) Assessing spatial learning and memory in rodents. Ilar j 55 , 310-332 D'Hooge, R., and De Deyn, P. P. (2001) Applications of the Morris water maze in the study of learning and memory. Brain Res Brain Res Rev 36 , 60-90 Bannerman, D. M., Grubb, M., Deacon, R. M., Yee, B. K., Feldon, J., and Rawlins, J. N. (2003) Ventral hippocampal lesions affect anxiety but not spatial learning. Behav Brain Res 139 , 197-213 Fanselow, M. S., and Dong, H. W. (2010) Are the dorsal and ventral hippocampus functionally distinct structures? Neuron 65 , 7-19 Liu, Z., Wan, X., Chen, J., Ma, Y., Fu, Y., Chen, Y., Wen, M., Yang, Y., Qian, Y., Zhang, Y., Zhu, D., Li, J., Jing, N., and Yang, X. (2025) Genomic locus of lncRNA-Gm26793 forms an inter-chromosomal interaction with Cubn to ensure proper stem cell differentiation in vitro and in vivo. Cell Discov 11 , 53 Milstein, J. L., and Ferris, H. A. (2021) The brain as an insulin-sensitive metabolic organ. Mol Metab 52 , 101234 Young III, W. S. (1986) Periventricular hypothalamic cells in the rat brain contain insulin mRNA. Neuropeptides 8 , 93-97 Havrankova, J., Schmechel, D., Roth, J., and Brownstein, M. (1978) Identification of insulin in rat brain. Proceedings of the National Academy of Sciences 75 , 5737-5741 Csajbok, E. A., and Tamas, G. (2016) Cerebral cortex: a target and source of insulin? Diabetologia 59 , 1609-1615 Kuwabara, T., Kagalwala, M. N., Onuma, Y., Ito, Y., Warashina, M., Terashima, K., Sanosaka, T., Nakashima, K., Gage, F. H., and Asashima, M. (2011) Insulin biosynthesis in neuronal progenitors derived from adult hippocampus and the olfactory bulb. EMBO Molecular Medicine 3 , 742-754 Nemoto, T., Toyoshima-Aoyama, F., Yanagita, T., Maruta, T., Fujita, H., Koshida, T., Yonaha, T., Wada, A., Sawaguchi, A., and Murakami, M. (2014) New insights concerning insulin synthesis and its secretion in rat hippocampus and cerebral cortex: Amyloid-β1–42-induced reduction of proinsulin level via glycogen synthase kinase-3β. Cellular Signalling 26 , 253-259 de la Monte, S. M., Tong, M., Bowling, N., and Moskal, P. (2011) si-RNA inhibition of brain insulin or insulin-like growth factor receptors causes developmental cerebellar abnormalities: relevance to fetal alcohol spectrum disorder. Mol Brain 4 , 13 Neu, S. C., Pa, J., Kukull, W., Beekly, D., Kuzma, A., Gangadharan, P., Wang, L.-S., Romero, K., Arneric, S. P., and Redolfi, A. (2017) Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. JAMA neurology 74 , 1178-1189 Fleischer, A. W., and Frick, K. M. (2023) New perspectives on sex differences in learning and memory. Trends in Endocrinology & Metabolism 34 , 526-538 Cahill, L., McGaugh, J. L., and Weinberger, N. M. (2001) The neurobiology of learning and memory: some reminders to remember. Trends Neurosci 24 , 578-581 O'Leary, T. P., and Brown, R. E. (2022) Visuo-spatial learning and memory impairments in the 5xFAD mouse model of Alzheimer's disease: Effects of age, sex, albinism, and motor impairments. Genes Brain Behav 21 , e12794 Fan, Y., Rudert, W. A., Grupillo, M., He, J., Sisino, G., and Trucco, M. (2009) Thymus-specific deletion of insulin induces autoimmune diabetes. The EMBO Journal 28 , 2812-2824 Grünblatt, E., Salkovic‐Petrisic, M., Osmanovic, J., Riederer, P., and Hoyer, S. (2007) Brain insulin system dysfunction in streptozotocin intracerebroventricularly treated rats generates hyperphosphorylated tau protein. Journal of neurochemistry 101 , 757-770 Plaschke, K., Kopitz, J., Siegelin, M., Schliebs, R., Salkovic-Petrisic, M., Riederer, P., and Hoyer, S. (2010) Insulin-resistant brain state after intracerebroventricular streptozotocin injection exacerbates Alzheimer-like changes in Tg2576 AβPP-overexpressing mice. Journal of Alzheimer's disease 19 , 691-704 Lester-Coll, N., Rivera, E. J., Soscia, S. J., Doiron, K., Wands, J. R., and de la Monte, S. M. (2006) Intracerebral streptozotocin model of type 3 diabetes: relevance to sporadic Alzheimer's disease. Journal of Alzheimer's Disease 9 , 13-33 Chen, W., Cai, W., Hoover, B., and Kahn, C. R. (2022) Insulin action in the brain: cell types, circuits, and diseases. Trends Neurosci 45 , 384-400 Dixon-Salazar, T. J., Fourgeaud, L., Tyler, C. M., Poole, J. R., Park, J. J., and Boulanger, L. M. (2014) MHC class I limits hippocampal synapse density by inhibiting neuronal insulin receptor signaling. J Neurosci 34 , 11844-11856 Soto, M., Cai, W., Konishi, M., and Kahn, C. R. (2019) Insulin signaling in the hippocampus and amygdala regulates metabolism and neurobehavior. Proceedings of the National Academy of Sciences 116 , 6379-6384 Cai, W., Xue, C., Sakaguchi, M., Konishi, M., Shirazian, A., Ferris, H. A., Li, M. E., Yu, R., Kleinridders, A., Pothos, E. N., and Kahn, C. R. (2018) Insulin regulates astrocyte gliotransmission and modulates behavior. J Clin Invest 128 , 2914-2926 Taib, B., Deme, P., Gupta, S., Yoo, S. W., Khuder, S. S., Hoke, A., Li, Z., Ahima, R. S., and Haughey, N. J. (2025) Insulin acts on astrocytes to shift their substrate preference to fatty acids. iScience 28 , 111642 Fernandez, A. M., Martinez-Rachadell, L., Navarrete, M., Pose-Utrilla, J., Davila, J. C., Pignatelli, J., Diaz-Pacheco, S., Guerra-Cantera, S., Viedma-Moreno, E., Palenzuela, R., Ruiz de Martin Esteban, S., Mostany, R., Garcia-Caceres, C., Tschöp, M., Iglesias, T., de Ceballos, M. L., Gutierrez, A., and Torres Aleman, I. (2022) Insulin regulates neurovascular coupling through astrocytes. Proc Natl Acad Sci U S A 119 , e2204527119 Leclerc, M., Bourassa, P., Tremblay, C., Caron, V., Sugère, C., Emond, V., Bennett, D. A., and Calon, F. (2022) Cerebrovascular insulin receptors are defective in Alzheimer’s disease. Brain 146 , 75-90 Additional Declarations No competing interests reported. Supplementary Files Ins2KOBrainSupplementalFigs1and22025V11submitted.pptx Supplemental Figure 1. Ins2 -/- and Ins2 +/+ mice do not differ in anxiety-like behaviour or locomotor activity. (A-C) Anxiety-like behaviour in male and female, Ins2 -/- and Ins2 +/+ mice (time in anxiogenic regions) on each of the open-field, light-dark box and elevated-plus maze. (D-F) Locomotor activity in male and female, Ins2 -/- and Ins2 +/+ mice (time in anxiogenic regions) on each of the open-field, light-dark box and elevated-plus maze. (G-I) Measures of species-typical behaviours (rears, grooming, freezing) in the open-field. (J-L) Measures of species-typical behaviours (rears, grooming, freezing) in the light-dark box. (M-P) Measures of species-typical behaviours (rears, grooming, freezing, open-arm, head-dips) in the elevated plus maze. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 10 Nov, 2025 Reviews received at journal 31 Oct, 2025 Reviewers agreed at journal 22 Oct, 2025 Reviews received at journal 14 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers agreed at journal 01 Oct, 2025 Reviewers invited by journal 24 Sep, 2025 Editor assigned by journal 23 Sep, 2025 Submission checks completed at journal 23 Sep, 2025 First submitted to journal 18 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7826673","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":533540927,"identity":"2debf995-fa8e-4294-ac24-a24a673abef0","order_by":0,"name":"Stella Baehring","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Stella","middleName":"","lastName":"Baehring","suffix":""},{"id":533540928,"identity":"0effe324-fc42-4376-bc55-9514b60591c6","order_by":1,"name":"Timothy P. O’Leary","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Timothy","middleName":"P.","lastName":"O’Leary","suffix":""},{"id":533540929,"identity":"cf37b9a3-8beb-4115-95b0-8aacdcfa1a3f","order_by":2,"name":"Haoning Howard Cen","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Haoning","middleName":"Howard","lastName":"Cen","suffix":""},{"id":533540930,"identity":"db087358-7e4a-49be-a028-96a3980b8d1b","order_by":3,"name":"Danae Holenka","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Danae","middleName":"","lastName":"Holenka","suffix":""},{"id":533540931,"identity":"c82e9d01-4a4b-49dd-8455-58d5b208b943","order_by":4,"name":"Kyungchan Kim","email":"","orcid":"","institution":"Daegu Gyeongbuk Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Kyungchan","middleName":"","lastName":"Kim","suffix":""},{"id":533540932,"identity":"9aeb3d40-743a-49df-a269-634691d85ca4","order_by":5,"name":"Sunday Ajayi","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Sunday","middleName":"","lastName":"Ajayi","suffix":""},{"id":533540933,"identity":"3458e747-3012-48e9-9511-a3b13d9e6981","order_by":6,"name":"Daniel Semenov","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Semenov","suffix":""},{"id":533540934,"identity":"8939c7f4-70ca-49bc-8671-ff4c09cef807","order_by":7,"name":"Manuel Belmadani","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Manuel","middleName":"","lastName":"Belmadani","suffix":""},{"id":533540935,"identity":"ff84361f-f34d-40d6-9f47-cfbcb746421b","order_by":8,"name":"Hong Li","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Li","suffix":""},{"id":533540936,"identity":"174beca0-94d3-44d7-942b-262b960052b7","order_by":9,"name":"Arya E. Mehran","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Arya","middleName":"E.","lastName":"Mehran","suffix":""},{"id":533540937,"identity":"5731f32a-4d04-4035-9dfa-0404e57f96a6","order_by":10,"name":"Honey Modi","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Honey","middleName":"","lastName":"Modi","suffix":""},{"id":533540938,"identity":"cfd61d00-efa9-475d-9152-b9e84e23c3ed","order_by":11,"name":"Melissa M. Page","email":"","orcid":"","institution":"Université catholique de Louvain","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"M.","lastName":"Page","suffix":""},{"id":533540939,"identity":"388795e7-6a63-4ee0-a40a-18c9a31468ae","order_by":12,"name":"Søs Skovsø","email":"","orcid":"","institution":"Valkyrie Life Sciences","correspondingAuthor":false,"prefix":"","firstName":"Søs","middleName":"","lastName":"Skovsø","suffix":""},{"id":533540940,"identity":"da2f11f0-2aed-4e57-89ad-f1395900ada4","order_by":13,"name":"Paul Pavlidis","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Paul","middleName":"","lastName":"Pavlidis","suffix":""},{"id":533540941,"identity":"cfdd2781-321e-4a3d-bcbf-0cd479c48b01","order_by":14,"name":"Shernaz X. Bamji","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Shernaz","middleName":"X.","lastName":"Bamji","suffix":""},{"id":533540942,"identity":"a911250a-6c0d-48c7-be1a-76c5f450712a","order_by":15,"name":"Eun-Kyoung Kim","email":"","orcid":"","institution":"Daegu Gyeongbuk Institute of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Eun-Kyoung","middleName":"","lastName":"Kim","suffix":""},{"id":533540943,"identity":"8b82c7ee-380e-4649-8c75-cf359afe6623","order_by":16,"name":"James D. Johnson","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIie3PIQvCQBTA8XcIWk7FNjnQr7AxWPHLaLri9QXDQNjK1KrfYiJoVQ5WHGZBgys2g0XOIo6bQYRtRsH7l3sH94N7ACrVD1Zx5FEC5GA5tUArIOiTmN8TgJT0nELibXuxgA6UvFEYC/tIg8PwBGKQQ3y2MDHQZNhRE0dnFhxDHflh3sfYkgBwQNO+RZDLWbDvJ/8s55DJZXUXL9K8u5zqkjyySWPKlsniKdGqLu9KgtxsQrTLgmCdYuRHFsERN2bJLpvROJvU6mx+FXanZXi+1RQ2b9cOw/gkbtkkTQdsOG/3dRGQtb96pVKpVH/ZE+igT7lu8pUnAAAAAElFTkSuQmCC","orcid":"","institution":"University of British Columbia","correspondingAuthor":true,"prefix":"","firstName":"James","middleName":"D.","lastName":"Johnson","suffix":""}],"badges":[],"createdAt":"2025-10-10 11:53:54","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7826673/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7826673/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94453813,"identity":"fa4a76b4-005e-477c-adc4-78e7090a71a0","added_by":"auto","created_at":"2025-10-27 14:42:56","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":160409,"visible":true,"origin":"","legend":"","description":"","filename":"Ins1KOBrainmanuscript2025v11Biorxivsubmitted2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/99ee88c790908a305c97880f.docx"},{"id":94453385,"identity":"c4a8194f-268d-4521-956a-637c63624506","added_by":"auto","created_at":"2025-10-27 14:42:31","extension":"json","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":16585,"visible":true,"origin":"","legend":"","description":"","filename":"1f7cc541c48b498e92890872bec608d7.json","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/4be558b69d429116ee4f60d5.json"},{"id":94452993,"identity":"9ebcc429-e5bb-4534-861d-a2c4bbd903f2","added_by":"auto","created_at":"2025-10-27 14:42:08","extension":"pptx","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3755374,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainSupplementalFigs1and22025V11submitted.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/21f172d415eccf4ddad378bf.pptx"},{"id":94454201,"identity":"e7f01537-54d9-4b7c-bc76-c88d2d625d8b","added_by":"auto","created_at":"2025-10-27 14:43:16","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":198882,"visible":true,"origin":"","legend":"","description":"","filename":"1f7cc541c48b498e92890872bec608d71enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/d4f0034ef6ad4297b3427c44.xml"},{"id":94453867,"identity":"1bbb4083-407e-404c-8900-700527bea183","added_by":"auto","created_at":"2025-10-27 14:42:59","extension":"pptx","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5126754,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainMainFig2025V11submittedFig1.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/9e957c35f5ea50d15f2421ac.pptx"},{"id":94453293,"identity":"ccda21d3-dd45-4106-a9e4-0fd597d9f650","added_by":"auto","created_at":"2025-10-27 14:42:26","extension":"pptx","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1075374,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainMainFig2025V11submittedFig2.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/7674a9bc726ada311c5aca27.pptx"},{"id":94454400,"identity":"18896346-ea54-4f33-9f93-b131fb5e9abe","added_by":"auto","created_at":"2025-10-27 14:43:27","extension":"pptx","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":495125,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainMainFig2025V11submittedFig3.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/b8c1edfd1377db7799a8b1be.pptx"},{"id":94453681,"identity":"5b4edbd4-2506-4924-bef6-60c88969d200","added_by":"auto","created_at":"2025-10-27 14:42:47","extension":"pptx","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2327970,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainMainFig2025V11submittedFig4.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/6b03c7a5c799d7e0ff0c633d.pptx"},{"id":94453984,"identity":"a108288f-92e6-4aa8-b7bf-c7d41b3c471d","added_by":"auto","created_at":"2025-10-27 14:43:03","extension":"pptx","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":800732,"visible":true,"origin":"","legend":"","description":"","filename":"Ins2KOBrainMainFig2025V11submittedFig5.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/b289c3073966bc8bd3d02344.pptx"},{"id":94453933,"identity":"185f25da-212f-4543-ae8f-e95de545e8a1","added_by":"auto","created_at":"2025-10-27 14:43:01","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":196129,"visible":true,"origin":"","legend":"","description":"","filename":"1f7cc541c48b498e92890872bec608d71structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/cc94b5adf2fac27bb5705b06.xml"},{"id":94453396,"identity":"78c1c370-d05a-4589-8616-fee2afcb401d","added_by":"auto","created_at":"2025-10-27 14:42:32","extension":"html","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":215975,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/81e8ee831797f30801711686.html"},{"id":94454276,"identity":"b5341b47-b805-4709-ada2-041eb8aab31e","added_by":"auto","created_at":"2025-10-27 14:43:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1587451,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpecific\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e Ins2 \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eprotein immunoreactivity in the mouse brain. (A) \u003c/strong\u003eImmunoreactivity for C-peptide, a product of insulin biosynthesis, is detected in the hilus, stratum lucidum (SL), and CA3 regions of hippocampus. C-peptide immunoreactive cells were observed in the granule cell layer of the dentate gyrus, as indicated by arrows. \u003cstrong\u003e(B) \u003c/strong\u003eImmunoblot analysis of proinsulin expression in the pancreas and hippocampus from WT and \u003cem\u003eIns2\u003c/em\u003e KO mice.\u003cstrong\u003e (C) \u003c/strong\u003eProinsulin-positive cells were detected and co-localized with β-gal in the CA3 regions. β-galactosidase (β-gal) staining was detected in \u003cem\u003eIns2\u003c/em\u003e knockout and heterozygous mice but not wildtype (WT). Pro-insulin was detected in WT and β-gal heterozygous mice but not knockouts.\u003cstrong\u003e (D) \u003c/strong\u003eGFP fluorescence in hippocampal neuron cultures of from \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003eGFP/GFP\u003c/sup\u003e knock-in, replacement mice.\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/40ec3925db76be4f228ebabb.png"},{"id":94453377,"identity":"6540e268-b206-4457-81da-3273a086ae49","added_by":"auto","created_at":"2025-10-27 14:42:31","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":58034,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e expression in the brain is sexually dimorphic and is modulated by diet in female mice.\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e(A)\u003c/strong\u003e \u003cem\u003eIns2\u003c/em\u003e gene expression in male and female mice across multiple brain regions, including the hippocampus, cerebellum, cerebral cortex, olfactory bulb, hypothalamus and midbrain (all remaining tissue). \u003cem\u003eIns2 \u003c/em\u003eexpression was normalized to beta-actin expression within each region. \u003cstrong\u003e(B,C)\u003c/strong\u003e \u003cem\u003eIns2\u003c/em\u003e expression in the hippocampus and cerebral cortex of male and female mice fed high-fat or control chow diets. **=p\u0026lt;.005, ***=p\u0026lt;.0001.\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/ef27fc69718426338e895f92.png"},{"id":94454044,"identity":"234c3a7d-254e-4ca4-ac87-8dd5d276b7c3","added_by":"auto","created_at":"2025-10-27 14:43:06","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":210050,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMetabolic profile of young and old \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003e knockout mice.\u003c/strong\u003e \u003cstrong\u003e(A,B) \u003c/strong\u003eBody weight of male and female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003emice did not differ from \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice at any of the ages measured. \u003cstrong\u003e(C,D) \u003c/strong\u003eFasted (4 hours) blood glucose levels were largely similar between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice and differed only in males at the oldest age (16-18 months) measured. \u003cstrong\u003e(E-L) \u003c/strong\u003e\u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice did not differ in either glucose tolerance or insulin tolerance tests. \u003cstrong\u003e(M-P) \u003c/strong\u003e\u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice were also similar in glucose stimulated insulin secretion, although there was a trend for higher insulin in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e than \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003emice in old age (16-18 months).\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/554e7c9c03545ad356194a8b.png"},{"id":94454190,"identity":"e189e9c1-657a-43c0-b9c3-70cbc3864003","added_by":"auto","created_at":"2025-10-27 14:43:14","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":266807,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVisuo-spatial learning and memory is impaired in female \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cstrong\u003e-/-\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e, but not in male \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cstrong\u003e-/-\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e mice.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e Latency to locate the escape platform in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice during acquisition (A1-6), re-training (RT) and reversal training (R1-6). Performance for the first 3 days and last 3 days of acquisition and reversal training were analyzed separately (indicated by horizontal dashed line). \u003cstrong\u003e(B) \u003c/strong\u003eCumulative search error for female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/-\u003c/em\u003e\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/+\u003c/em\u003e\u003c/sup\u003e mice on the same study days. \u003cstrong\u003e(C) \u003c/strong\u003eThe number of platform annulus crossings completed by female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice during the acquisition probe trial. \u003cstrong\u003e(D)\u003c/strong\u003e Distance travelled by female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/-\u003c/em\u003e\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/+\u003c/em\u003e\u003c/sup\u003e mice to reach the escape platform on each study day. \u003cstrong\u003e(E) \u003c/strong\u003eSwim-speed for female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/-\u003c/em\u003e\u003c/sup\u003e\u003csup\u003e \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/+\u003c/em\u003e\u003c/sup\u003e mice over acquisition and reversal training. \u003cstrong\u003e(F) \u003c/strong\u003ePlatform annulus crossings during the reversal probe trial.\u003cstrong\u003e (G-L)\u003c/strong\u003e Show the same measurements as above, but for male mice. *=p\u0026lt;.05\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/8a78265473846778bd2d608a.png"},{"id":94454264,"identity":"6f11e0b6-c358-4c16-a859-f45abd7eeb43","added_by":"auto","created_at":"2025-10-27 14:43:22","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":567430,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBulk mRNA sequencing and cluster analysis of isolated \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003csup\u003e\u003cstrong\u003e/- \u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eand\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e Ins2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cstrong\u003e+/+\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003e hippocampi from female mice. (A)\u003c/strong\u003e Cluster dendrogram and \u003cstrong\u003e(B) \u003c/strong\u003ePCA showing the 3 outlier samples. \u003cstrong\u003e(C)\u003c/strong\u003e Bulk mRNA sequencing shows no \u003cem\u003eIns2\u003c/em\u003e mRNA for\u003cem\u003e Ins2 \u003c/em\u003eknockouts \u003cstrong\u003e(D)\u003c/strong\u003e Differentially expressed genes ranked from most significantly downregulated to upregulated (top to bottom). \u003cstrong\u003e(E)\u003c/strong\u003e Gene set enrichment analysis using Gene Ontology - Biological Process (FDR\u0026lt; 0.0001).\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/82eac7bd411318691dfd7c87.png"},{"id":94453388,"identity":"84128cae-8ccd-4521-9b47-441a44234fb0","added_by":"auto","created_at":"2025-10-27 14:42:32","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":5713,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEffects of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2 \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eloss on adult hippocampal neurogenesis in aged male and female mice. (A)\u003c/strong\u003e Specific labeling of EdU in the intestine, our positive control for proliferating cells in aged animals. \u003cstrong\u003e(B)\u003c/strong\u003e Co-labeling of EdU and DCX markers in the dentate gyrus of the hippocampus. \u003cstrong\u003e(C,D)\u003c/strong\u003e Quantification of the density of EdU\u003csup\u003e+\u003c/sup\u003e cells in the dorsal and ventral dentate gyrus in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e-/- \u003c/em\u003e\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e+/+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e \u003c/em\u003emice. \u003cstrong\u003e(E,F)\u003c/strong\u003e Quantification of the density of DCX\u003csup\u003e+\u003c/sup\u003e cells in the dorsal and ventral dentate gyrus.*=p\u0026lt;.05\u003c/p\u003e","description":"","filename":"placeholderimage.png","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/2c13abeb67e606dcc62b149a.png"},{"id":94468633,"identity":"cba2d593-3e69-4f8a-bdd6-56134c44dcc1","added_by":"auto","created_at":"2025-10-27 15:25:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3896641,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/588e4956-e369-4126-b94a-539eef3735ea.pdf"},{"id":94453818,"identity":"e2e8e5d7-f0d6-4f01-8b86-a4569af23c66","added_by":"auto","created_at":"2025-10-27 14:42:57","extension":"pptx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3755374,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSupplemental Figure 1. \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e-/-\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003csup\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eand \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eIns2\u003c/strong\u003e\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u003cstrong\u003e+/+\u003c/strong\u003e\u003c/em\u003e\u003c/sup\u003e\u003csup\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003emice\u003c/strong\u003e\u003csup\u003e \u003c/sup\u003e\u003cstrong\u003edo not differ in anxiety-like behaviour or locomotor activity\u003c/strong\u003e. \u003cstrong\u003e(A-C)\u003c/strong\u003e Anxiety-like behaviour in male and female, \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+ \u003c/sup\u003emice (time in anxiogenic regions) on each of the open-field, light-dark box and elevated-plus maze. \u003cstrong\u003e(D-F)\u003c/strong\u003e Locomotor activity in male and female, \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/- \u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+ \u003c/sup\u003emice (time in anxiogenic regions) on each of the open-field, light-dark box and elevated-plus maze.\u003cstrong\u003e (G-I)\u003c/strong\u003e Measures of species-typical behaviours (rears, grooming, freezing) in the open-field. \u003cstrong\u003e(J-L)\u003c/strong\u003e Measures of species-typical behaviours (rears, grooming, freezing) in the light-dark box. \u003cstrong\u003e(M-P)\u003c/strong\u003e Measures of species-typical behaviours (rears, grooming, freezing, open-arm, head-dips) in the elevated plus maze.\u003c/p\u003e","description":"","filename":"Ins2KOBrainSupplementalFigs1and22025V11submitted.pptx","url":"https://assets-eu.researchsquare.com/files/rs-7826673/v1/5d2266c5c68a92aa67c22ec7.pptx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Loss of brain insulin production impairs learning and memory in female mice","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDiabetes is a significant risk factor for dementia (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) and people with type 2 diabetes can present with cognitive impairment, specifically in learning and memory, similarly to early onset Alzheimer\u0026rsquo;s disease (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). People with Alzheimer\u0026rsquo;s disease often present with hyperglycemia and insulin dysfunction in peripheral organs, and may also have \u0026lsquo;insulin resistance\u0026rsquo; in the brain (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Post-mortem brain tissue pointed to impaired insulin signalling in the cerebral cortex and hippocampus of Alzheimer\u0026rsquo;s disease samples compared to control (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Studies in humans and animals show that insulin plays a multifactorial role in brain function such as enhancing synaptic plasticity, dendritic spine formation, and increasing neurotransmitter turnover (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Insulin also influences the clearance of the amyloid β peptide and phosphorylation of tau, which are hallmarks of Alzheimer\u0026rsquo;s disease (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). Insulin insensitivity in the brain is associated with reduced insulin receptor levels, binding affinity, and disrupts insulin signalling (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e). Insulin signalling is altered in the brains of Alzheimer\u0026rsquo;s disease patients and alterations in insulin receptor substrates (IRS-1 and IRS-2) is an early pathological characteristic (\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). A recent study found that the loss of insulin signaling in astrocytes exacerbates Alzheimer-like pathology including amyloid plaque accumulation (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Astrocytes are crucial in regulating important neuronal functions via insulin, such as dopaminergic signaling and cholesterol synthesis, pointing to the importance of insulin availability in local brain regions (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). Interestingly, insulin in the brain is produced by neurons, but not glial cells (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). Together, these separate lines of evidence suggest that impairment of brain insulin action may underly a component of the pathology of Alzheimer\u0026rsquo;s disease. Since insulin action on the brain plays a significant role in cognitive functions in humans and various pre-clinical model systems (\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e), some have proposed administering insulin directly to the brain as a treatment bypassing metabolic and brain barriers that decrease bioavailability (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Indeed, augmenting insulin signalling tone via intranasal administration of insulin can improve cognitive function in Alzheimer\u0026rsquo;s disease patients (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Lifestyle interventions that increase insulin sensitivity, such as diet and exercise, improve learning and memory, and are being used as preventative and therapeutic approaches for Alzheimer\u0026rsquo;s disease (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e). Collectively evidence points to a potential role for brain-directed insulin and insulin signalling on cognition.\u003c/p\u003e\u003cp\u003eMost studies investigating the effects of insulin on the brain were done under the assumption that insulin in the brain is of pancreatic origin. Although peripheral insulin crosses the blood brain barrier via receptor-mediated transport (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e), accumulating evidence shows that insulin is also produced locally in the brain (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Insulin expression has been measured in several brain regions, including cerebral cortex, choroid plexus, cerebellum, and the hippocampus (\u003cspan additionalcitationids=\"CR23 CR24 CR25 CR26\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). The hippocampus is also the main region involved in memory and learning, as well as the first to be affected by neural deterioration in Alzheimer\u0026rsquo;s disease. In mice, the specificity of these observations has been confirmed with mice lacking insulin 2 (\u003cem\u003eIns2\u003c/em\u003e) gene (human homolog)(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e), which survive due to the presence of an additional, pancreas-specific insulin coding gene insulin 1 (\u003cem\u003eIns1\u003c/em\u003e)(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e). Notably, we have previously shown that \u003cem\u003eIns1\u003c/em\u003e does not upregulate its gene expression in the brain to compensate for \u003cem\u003eIns2\u003c/em\u003e loss (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Insulin synthesis has been reported in neuronal progenitors of the hippocampus and olfactory bulbs (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e), key sites of adulthood neurogenesis (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e). Studies on neurogenesis showed that activation of insulin in the neurogenic niche stimulates neural stem cell survival, proliferation, and differentiation (\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e), suggesting that brain insulin and neurogenesis could be linked. Collectively, these data and others (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e) clearly demonstrate that a small amount of insulin can be synthesized in the brain and that its production occurs in regions highly relevant for learning and memory. However, little is known about the environmental factors that influence insulin production in the brain. Most importantly, whether brain-derived insulin directly influences learning, memory, and neurogenesis has yet to be examined.\u003c/p\u003e\u003cp\u003eIn this study, we used \u003cem\u003eIns2\u003c/em\u003e knockout mice to further confirm protein insulin synthesis in the hippocampus and tested how the loss of \u003cem\u003eIns2\u003c/em\u003e influences learning and memory in both sexes. We determined that \u003cem\u003eIns2\u003c/em\u003e expression in the mouse hippocampus is modulated by diet and biological sex. We found that female mice lacking \u003cem\u003eIns2\u003c/em\u003e have deficits in learning and memory. These data provide the first loss-of-function evidence that \u003cem\u003eIns2\u003c/em\u003e plays a role in hippocampus-controlled behaviour.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eExperimental animals\u003c/h2\u003e\u003cp\u003eExperiments charactering \u003cem\u003eIns2\u003c/em\u003e expression used five-month-old C57BL/6J mice (Strain 000664). All other experiments used male and female \u003cem\u003eIns\u003c/em\u003e2\u003csup\u003e+/+\u003c/sup\u003e (wild-type) and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u003c/em\u003e/\u0026minus;\u003c/sup\u003e (knockout) littermates between 3\u0026ndash;18 months of age with a background that is primarily C57BL/6J. \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u003c/em\u003e/\u0026minus;\u003c/sup\u003e mice were originally made by Duville et al (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) and have the LacZ/neomycin cassette inserted into the \u003cem\u003eIns2\u003c/em\u003e locus. Mice were bred in-house, by crossing male and female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/\u0026minus;\u003c/sup\u003e mice. Mice were housed in same-sex groups of 2\u0026ndash;4, in cages with a plastic dome and nesting material for enrichment. In some cases, mice were singly housed due to aggression or barbering, but effort was taken to ensure a similar number of singly housed mice were present for each genotype and sex. Except where indicated, food and water were available \u003cem\u003ead libitum\u003c/em\u003e. Mice received a chow diet (total calories\u0026thinsp;=\u0026thinsp;4.68 kcal/g; 25.3% calories from fat, 19.8% calories from protein, 54.9% calories from carbohydrate; Catalog #5015 Lab Diets, Richmond, IN) or a high fat diet (total calories\u0026thinsp;=\u0026thinsp;5.56 kcal/g; 58.0% calories from fat, 16.4% calories from protein, 25.5% calories from carbohydrate; Catalog #D12330 Open-Source Diets/Research Diets, New Brunswick, NJ). Mice were housed on a reverse 12 h light dark cycle (lights off at 7:00 am), and all experiments were completed during the dark phase of the light cycle. Transport of mice from the home-cage to behavioral tests was completed with a 500 ml plastic container to reduce handling stress. Some studies used mice with GFP knocked into the endogenous \u003cem\u003eIns2\u003c/em\u003e gene locus, which have been described and phenotyped elsewhere (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). All experimental procedures were approved by the UBC animal care committee and adhered to Canadian Council on Animal Care guidelines.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eImmunohistochemistry and immunoblotting for C peptide, β-gal and pro-insulin\u003c/h3\u003e\n\u003cp\u003eMice were perfused under isoflurane anesthesia with phosphate-buffered saline (PBS), followed by ice-cold 4% paraformaldehyde in PBS. Brains were post-fixed in paraformaldehyde for 4 hrs, and then cryoprotected with 30% sucrose in PBS for 48 hours. Brains were embedded in OCT, frozen on dry-ice and stored at -80\u0026ordm;C, before sectioning at 40\u0026micro;m on a cryostat. Immunostaining was completed using rabbit anti\u0026ndash;C-peptide (1:300; Cell Signaling Technology, 4593), rabbit anti\u0026ndash;β-gal (1:100; Thermo Fisher Scientific, A-11132) and mouse anti-proinsulin (1:100; R\u0026amp;D Systems, Bio-Techne, MAB13361) primary anti-bodies. Donkey anti-rabbit Cy3 (1:200; Jackson ImmunoResearch, 711-165-152) and donkey anti-mouse Alexa488 (1:200; Jackson ImmunoResearch, 715-545-150) were used as secondary antibodies, followed by nuclear staining with Hoechst 33258 (1ug/ml, Invitrogen, Thermo Fisher Scientific). Sections were mounted onto slides, cover-slipped with VECTASHIELD mounting medium (Vector Laboratories), and imaged using a LSM 800 confocal microscope (Carl Zeiss).\u003c/p\u003e\u003cp\u003eImmunoblots of proinsulin were conducted as in our previous publication (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e). Briefly, mice were perfused with 10 mL of ice-cold phosphate-buffered saline before collecting the samples. To extract protein from the pancreas and hippocampus, samples were dissolved in lysis buffers as described previously (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e). The whole hippocampi of 5 \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e and 5 \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice were lysed and separated on 15% SDS-PAGE and blotted onto polyvinylidene difluoride membranes (Millipore, IPVH00010) for 30 minutes at 16 V in the transfer buffer (25 mM Tris base, pH 7.4, 192 mM glycine, 10% methanol). The membranes were blocked with 5% skim milk for 1 hour and then incubated with primary antibody against proinsulin (1:1000; Cell Signaling Technology, 8138) or GAPDH (1:10,000; Cell Signaling Technology, 2118) at 4\u0026deg;C overnight. After extensive washing in Tris-buffered saline with 0.1% Tween-20, the membranes were incubated with horseradish peroxidase-conjugated anti-mouse (1:3000; Cell Signaling Technology, 7076S) or anti-rabbit secondary antibody (1:10,000; Thermo Scientific, NCI1460KR) and the bands were visualized using ECL solutions (Thermo Scientific, NCI4080KR; Advansta, K-12045-D50) according to the manufacturer\u0026rsquo;s instructions.\u003c/p\u003e\n\u003ch3\u003eRNA extraction, cDNA synthesis and quantitative polymerase chain reaction\u003c/h3\u003e\n\u003cp\u003eBrains were rapidly removed from mice following decapitation after C02 euthanasia. Dissection of the brain occurred on ice, wherein the hippocampus, cerebellum, cerebral cortex, olfactory bulbs, hypothalamus and midbrain (including all residual tissue) were isolated. All brain tissue was then snap frozen on dry ice and stored at -80\u003cb\u003e\u0026deg;\u003c/b\u003eC. Animals were euthanized in an alternating sequence based on sex to mitigate circadian effects on gene transcription. Dissected brain tissue was homogenized in Trizol (Thermo Fisher) to protect against RNA degradation. Chloroform (50ul) was then added, and the tissue and shaken for 20 seconds before a 15-minute incubation at room temperature. Samples were centrifuged (14000rpm, 15 minutes at 4\u003cb\u003e\u0026deg;\u003c/b\u003eC) and the aqueous phase was transferred to 70% ethanol. RNA extraction was performed using the Qiagen RNasy minikit (cat #74108 Thermo Fisher), while DNA was synthesized using the qScript cDNA Synthesis Kit (cat#101414-100, Quanta Bioscience).\u003c/p\u003e\u003cp\u003e\u003cem\u003eIns2\u003c/em\u003e mRNA was measured using quantitative polymerase chain reaction (qPCR). The cDNA was subjected to Taqman real-time PCR (StepOnePlus, Applied Biosystems). The following primers were used; \u003cem\u003eIns2\u003c/em\u003e Taqman reverse 280\u0026loz;259: GAT CTA CAA TGC CAC GCT TCT G, \u003cem\u003eIns2\u003c/em\u003e Taqman probe 224\u0026loz;207: CCT GCT CCC GGG CCT CCA. PCR conditions were 2 min at 50\u0026deg;C, 10 min at 95\u0026deg;C and 40 cycles 15-sec cycles at 95\u0026deg;C and lastly 1 min at 60\u0026deg;C. Samples were normalized to the housekeeping gene beta actin by converting raw values to delta values. The reverse log of delta values were used for all analyses. Water and cDNA from \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u003cem\u003e\u0026minus;\u003c/em\u003e/\u0026minus;\u003c/sup\u003e tissue were used as negative controls.\u003c/p\u003e\n\u003ch3\u003eMetabolic characterization\u003c/h3\u003e\n\u003cp\u003eAge-related metabolic phenotyping included measurement of body weight, fasting blood glucose (FBG), glucose tolerance tests (GTT), insulin tolerance tests (ITT) and glucose stimulated insulin secretion (GSIS). For FBG and GTT, mice were fasted for four hours in clean cages prior to glucose measurement. For GTT and ITT, baseline levels of glucose were obtained 15 min prior to intra-peritoneal injections of either glucose (GTT; 4mg/kg) or insulin (ITT; 0.75U/kg, or 1.5U/kg) in PBS. Blood glucose was then measured at 15, 30, 60, 90 and 120 min after injections using OneTouch glucometers and measurement strips. For GSIS, the legs of mice were shaved to facilitate blood collection from the lateral saphenous vein. Mice were fasted for 4 hours, and baseline blood samples were obtained, followed an intra-peritoneal injection of glucose (4mg/kg). Blood samples were then obtained at 15 and 30min post injection. All blood samples were kept on ice, before being centrifuged and serum collected. Insulin levels were then quantified using an enzyme linked immunosorbent assay, based on the manufacturer\u0026rsquo;s instructions (ALPCO Diagnostics, Salem, NH).\u003c/p\u003e\n\u003ch3\u003eMorris water maze\u003c/h3\u003e\n\u003cp\u003eThe Morris water maze consisted of a white circular pool (110 cm diameter) filled with water to a depth of 16 cm. The water was 23\u003cb\u003e\u0026deg;\u003c/b\u003eC and made opaque with the addition of non-toxic white tempera paint (Schola, 2002737). A circular escape platform (11.5 cm diameter) was placed in the pool and was 0.5cm below the water surface. The water maze was placed in a diffusely lit room with many extra-maze cues, including large geometric posters adhered to the walls, a desk with a computer, a door, and the geometric layout of the room. The experimenter also served as an extra-maze cue and stood in a similar position for each trial. Mice were placed individually into holding cages lined with paper towel during training. The Anymaze computer tracking system was used to record the movement of mice, and to obtain measures of learning and memory performance.\u003c/p\u003e\u003cp\u003eMice first completed acquisition training which consisted of 6 days of training with 4 trials per day. The escape platform was located in the same location (NW quadrant) across trials. For each trial, mice were placed into the pool at one of four release locations (pseudo-randomly determined) and were given a maximum of 60 sec to reach the escape platform. After reaching the platform, mice remained on the platform for 5\u0026ndash;10 seconds before being returned to the holding cage. If mice did not reach the platform in 60 sec, they were guided to the platform by the experimenter. Mice were tested in squads of 4\u0026ndash;6, and the inter-trial interval ranged from 2\u0026ndash;8 minutes. Performance of mice was measured using latency to locate the escape platform, distance travelled to reach the escape platform, cumulative search error and swim-speed. Cumulative search error was obtained using custom software and consisted of the total distance from the escape platform, summed across all recorded positions of the mouse (5 hz). A correction factor is applied, where the CSE value expected for the optimal swim path (determined by start location and swim speed) is subtracted from the total CSE. The day following acquisition training, mice completed a 60 sec probe trial without the escape platform to measure memory. The number of times mice crossed over the location of the escape platform (annulus crossings) and respective locations in other quadrants of the pool were recorded. The day following the acquisition probe, mice completed an additional day of training (acquisition re-training) to determine the extent of extinction occurred in the probe trial, and to re-establish baseline levels of learning performance. Mice then completed reversal training to measure behavioral flexibility and the learning of a new spatial location of the escape platform. During reversal training the escape platform was moved to the opposite side of the pool. Mice completed 6 days of training with 4 trials per day, using the same procedure as acquisition training. Following reversal training, mice completed a 60-sec probe trial to measure memory for the new escape platform location.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eAnxiety-like behaviour\u003c/h2\u003e\u003cp\u003eMice were tested on the open-field, light-dark box and elevated plus maze to measure locomotor activity and anxiety-like behaviour. All trials were 10 min in length and the test apparatuses were cleaned with 70% ethanol between trials. After trials, mice were placed in a clean holding cage, while the remaining cagemates completed their trials. During testing mice were brought in groups of 7\u0026ndash;9 mice into a dark and quiet anteroom. Mice habituated to the anteroom for at least 25 min before behavioral testing. Testing was completed in a different room than the Morris water maze training. Tests were completed with 7-13- day intervals, to reduce potential cross-test habituation. Movement of mice in the open-field, light-dark box and elevated plus maze were recorded with the Ethovision (Noldus) video tracking system. Species typical behaviors were measured with the Boris (Friad and Gamba, 2016) event scoring software, using video recordings from trials.\u003c/p\u003e\u003cp\u003eThe open field consisted of a box (70 x 70 cm) made of transparent Plexiglas, with 23 cm high walls. For each trial, mice were placed into the one of four corners of the open field (pseudo-randomly determined). Locomotor activity was measured with distance travelled, while anxiety-like behaviour was assessed with entries into the center of the open-field (9th of area). Species typical behaviors were also recorded including number of rears, grooming duration, and freezing duration.\u003c/p\u003e\u003cp\u003eThe light-dark box was constructed from Plexiglas and was divided into two chambers that were either brightly (light zone) or dimly lit (dark zone). A single light placed above the light zone provided lighting in the light-dark box and testing room. For each trial, mice were placed in the light zone facing the opening to the dark zone. The same behavioural measures recorded in the open field were also recorded in the light-dark box, except that time in the light chamber was used as a measure of anxiety-like behaviour.\u003c/p\u003e\u003cp\u003eThe elevated plus maze consisted of two pairs of arms, each attached to a central square. One pair of arms had opaque walls around the edge (closed arms), while the other pair of arms had no walls (open arms). A small rim (0.5 cm) was present around the edge of the open arms to prevent mice from falling. A single light provided dim lighting for the elevated plus maze and testing room. For each trial, mice were placed onto the central hub of the elevated plus maze, facing a closed arm. The same behavioral measurements that were recorded in the open field were also recorded in the elevated plus maze, except that time in the open arms was used as measure of anxiety-like behaviour. The number of times mice dipped their heads over the edge of the open arms (head-dips) was also recorded as a species typical measure of anxiety-like behaviour.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eBulk hippocampal RNA sequencing\u003c/h3\u003e\n\u003cp\u003eRemoval and dissection of brains were completed as described above for qPCR. Animals were euthanized in an alternating sequence based on genotype to prevent potential circadian effects on gene transcription. Body weight and un-fasted blood glucose (measured the from the neck after decapitation) were also measured to ensure no pre-existing differences between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice at the time of euthanasia. Sample quality control was performed using the Agilent 2100 Bioanalyzer. Qualifying samples were then prepped following the standard protocol for the NEBnext Ultra ii Stranded mRNA (New England Biolabs). Sequencing was performed on the Illumina NextSeq 500 with Paired End 43bp \u0026times; 43bp reads. Sequencing data was demultiplexed using Illumina's bcl2fastq2. De-multiplexed read sequences were then aligned to the \u003cem\u003eMus musculous\u003c/em\u003e (PAR-masked)/mm10 reference sequence using STAR aligner. Quality control led to one outlier sample (genotype -/-) being removed. Hierarchical clustering and principal component analysis (PCA) identified 3 additional outliers (1 \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e and 2 \u003cem\u003eIns2\u003c/em\u003e) that were removed. Genes with expression below 0.5 counts per million were dropped from the analysis. Differential expression analysis associated with genotype was performed using limma-trend (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e) after quantile normalization or using DESeq2 (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e), which had consistent results. The DESeq2 results were used in the figures and Gene Set Enrichment Analysis (GSEA). GSEA was performed with clusterProfiler R package using Gene Ontology (Biological Process) knowledgebase (FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e), and simplify() function was used to remove redundant GO terms (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://guangchuangyu.github.io/2015/10/use-simplify-to-remove-redundancy-of-enriched-go-terms/\u003c/span\u003e\u003cspan address=\"https://guangchuangyu.github.io/2015/10/use-simplify-to-remove-redundancy-of-enriched-go-terms/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eQuantification of adult hippocampal neurogenesis\u003c/h3\u003e\n\u003cp\u003eTwenty-month-old \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice were given EdU in drinking water (0.25 mg/ml) for four weeks. The EdU water was freshly prepared every week and was given to mice in 120 ml bottles. Body weight and water consumption were recorded every week to ensure consumption of EdU water. Mice with given Edu-water for 4 weeks, after which they were transcardially perfused with PBS (0.1M) followed by 4% paraformaldehyde. Brains were extracted, post-fixed in 4% paraformaldehyde overnight, and then transferred to a 30% sucrose solution for cryoprotection. Brains were sliced into 40 \u0026micro;m coronal sections using a cryostat at -20\u0026deg;C. Sections were collected in series of 6 through the rostral-caudal extent of the hippocampus and stored in anti-freeze solution (ethylene glycol, glycerol, and 0.1 M PBS) at -20\u0026deg;C until immunostaining.\u003c/p\u003e\u003cp\u003eTo measure adult neurogenesis a series of sections were stained for both EdU and doublecortin (DCX). Staining for EdU was completed with the Click-iT Plus EdU Alexa Fluor 647 Imaging Kit (Thermo Fisher Scientific). Staining for DCX was completed via Immunohistochemistry using a Rabbit anti-DCX IgG polyclonal antibody (cat# 326 003 Thermo Scientific) and a Donkey anti-Rabbit IgG H\u0026thinsp;+\u0026thinsp;L Alexa Fluor 488 secondary antibody (cat# A-21206 Invitrogen). We used vertical sections of intestinal villi from EdU treated animals as positive controls for EdU labelling.\u003c/p\u003e\u003cp\u003eStained sections were imaged with a Zeiss LSM 880 confocal microscope. Dorsal and ventral regions of the dentate gyrus were imaged using a 20X objective. Image stacks through the section were obtained, stitched together and maximum intensity projections were then used to count EdU and DCX-positive cells. The experimenter was blind to all conditions during quantification, and six sections were analyzed per animal. The sampling area within the dentate gyrus was also obtained, and cell counts normalized to sampling area.\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eStatistical Analysis\u003c/h2\u003e\u003cp\u003eFor data using WT C57BL/6J mice, data were analyzed with sex as a factor when possible. Subsequent data for experiments using \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice were analyzed separately for each sex and age to increase statistical power to detect differences due to genotype. For tolerance tests and glucose stimulated insulin secretion, mixed design ANOVAs (2 x 6 or 2 x 3, respectively) were used with genotype as a between subject factor and sampling time as a within subject factor. For acquisition and reversal training in the MWM, analyses were completed separately for days 1\u0026ndash;3 and days 4\u0026ndash;6, which reflect initial learning and near-asymptotic performance respectively. Measures of learning performance were analyzed with mixed design ANOVAs (2 x 3), using genotype as a between subject factor and day of training as within subject factor. All other analyses were completed using one-way between subject ANOVAs comparing genotypes. Post-hoc tests were completed using unpaired t-tests, with Bonferonni corrections applied independently for each statistical test.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cem\u003eIns2 is produced in the brain and its expression is sexually dimorphic and modulated by diet\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe previously identified \u003cem\u003eIns2\u003c/em\u003e mRNA and insulin protein in the mouse brain (25), but did not show hippocampal C-peptide and proinsulin immunoreactivity that was specific (i.e. absent in \u003cem\u003eIns2\u003c/em\u003e null brains). Here, we showed that labelling of C-peptide, a byproduct and marker of insulin synthesis, can be observed within the hippocampus of \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice, particularly within the dentate gyrus-CA3 pathway (\u003cstrong\u003eFigure 1A)\u003c/strong\u003e. This was confirmed with immunoblots showing a small but clear \u003cem\u003eIns2\u003c/em\u003e-specific band for proinsulin in hippocampus lysates, as well as the expected prominent band in pancreas (\u003cstrong\u003eFigure 1B\u003c/strong\u003e). As expected, no insulin protein was found in hippocampi from \u003cem\u003eIns\u003c/em\u003e2\u003csup\u003e-/-\u003c/sup\u003e because the non-ancestral \u003cem\u003eIns1\u003c/em\u003e gene is not robustly expressed in the mouse brain (25). Complementary immunostaining in the hippocampus of wildtype \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice revealed clear labelling of proinsulin within the CA3 and choroid plexus, and this staining was largely absent in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003emice (\u003cstrong\u003eFigure 1C\u003c/strong\u003e). Importantly, in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/-\u003c/sup\u003e mice, proinsulin labelling colocalized with\u0026nbsp;b-gal labelling, a proxy for \u003cem\u003eIns2\u003c/em\u003e gene activity and mRNA expression. We were also able to visualize green fluorescence in cultured hippocampal neurons from mice with GFP knocked into the endogenous \u003cem\u003eIns2\u003c/em\u003e locus, which we have previously shown correlates with \u003cem\u003eIns2\u003c/em\u003e mRNA, pre-mRNA, and proteins levels\u0026nbsp;(34)\u0026nbsp;(\u003cstrong\u003eFigure 1D\u003c/strong\u003e). Together, these data provide further strong evidence that the brain can locally produce insulin.\u003c/p\u003e\n\u003cp\u003eWe next measured \u003cem\u003eIns2\u003c/em\u003e mRNA levels across six brain regions (hippocampus, cerebellum, cerebral cortex, olfactory bulbs, hypothalamus, all residual tissue was classified as midbrain) in male and female wildtype C57BL/6J mice. Brain \u003cem\u003eIns2\u003c/em\u003e expression was highest in the hippocampus and cerebellum of both sexes (\u003cstrong\u003eFigure 2A\u003c/strong\u003e). \u003cem\u003eIns2\u003c/em\u003e mRNA was consistently higher in females than males (F(1,14) = 7.58, p \u0026lt; 0.05), and clear differences were observed across brain regions (F(5,70) = 18.42, p\u0026lt; 0.0001). Sex differences in \u003cem\u003eIns2\u003c/em\u003e expression were more prominent within certain brain regions (interaction, (F(5,70) = 2.67, p \u0026lt; 0.05)), where females showed higher expression of \u003cem\u003eIns2\u003c/em\u003e than males in the hippocampus and cerebellum (post-hoc, p \u0026lt; 0.005) (\u003cstrong\u003eFigure 2A\u003c/strong\u003e). Because the expression of \u003cem\u003eIns2\u0026nbsp;\u003c/em\u003eand levels of insulin in the pancreas are modulated by diet, so we next tested whether a similar regulation occurs in brain. These independent experiments confirmed higher expression of \u003cem\u003eIns2\u003c/em\u003e in the hippocampus of female mice relative to other brain regions (i.e. cerebral cortex) and relative to males \u003cstrong\u003e(Figure 2B,C)\u003c/strong\u003e. In female mice, \u003cem\u003eIns2\u0026nbsp;\u003c/em\u003eexpression was reduced after being fed a high-fat diet when compared to a control diet, whereas diet did not affect \u003cem\u003eIns2\u003c/em\u003e expression in male mice (\u003cstrong\u003eFigure 2B,C\u003c/strong\u003e). These data demonstrate that \u003cem\u003eIns2\u003c/em\u003e mRNA levels differ between brain regions, sexes, and dietary contexts.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eIns1 is sufficient to preserve normal peripheral metabolism in the absence of Ins2\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBefore using \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice to examine the physiological roles of Ins2 in the brain, it was critical to confirm that this was independent of significant whole-body metabolic effects. Previous studies have shown that mice lacking \u003cem\u003eIns2\u0026nbsp;\u003c/em\u003ealone for life have relatively normal metabolic physiology (25), supporting previous observations that \u003cem\u003eIns1\u003c/em\u003e can compensate for loss of \u003cem\u003eIns2\u0026nbsp;\u003c/em\u003e(39). Previous work only focused on males, so we next characterized metabolism in littermate male and female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice side-by-side to confirm that \u003cem\u003eIns1\u003c/em\u003e can compensate for loss of \u003cem\u003eIns2 in\u0026nbsp;\u003c/em\u003eboth sexes. In addition, we examined mice from 3-16 months of age to determine if compensation by \u003cem\u003eIns1\u003c/em\u003e can be maintained during aging. For both males and females, body weight of \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice did not differ from \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice at any age \u003cstrong\u003e(Figure 3A,B)\u003c/strong\u003e. Fasted (4 hours) blood glucose levels were largely similar between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice and differed only in males at the oldest age (16-18 months) tested (\u003cstrong\u003eFigure 3C,D)\u003c/strong\u003e. \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand\u003cem\u003e\u0026nbsp;Ins2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice did not differ in glucose tolerance tests \u003cstrong\u003e(Figure 3E-H)\u003c/strong\u003e nor in insulin tolerance tests at either 8 or 16-18 months of age \u003cstrong\u003e(Figure 3I-L)\u003c/strong\u003e. \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice also showed similar glucose stimulated insulin secretion (\u003cstrong\u003eFigure 3M-P\u003c/strong\u003e), although there was a trend for higher insulin in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice compared with \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice at the oldest age measured (\u003cstrong\u003eFigure 3N\u003c/strong\u003e). Together, these data suggest that \u003cem\u003eI\u003c/em\u003e\u003cem\u003ens1\u003c/em\u003e can produce a near-complete compensation for the absence of \u003cem\u003eIns2\u003c/em\u003e in both male and female mice on this normal chow diet, and this compensation can be maintained through aging up to 16-18 months.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLoss of Ins2 produces visuo-spatial learning and memory impairments in\u003c/em\u003e \u003cem\u003efemale, but not male mice\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe hippocampus is critical for a variety of different cognitive processes, and in rodents is perhaps most well-known for its role in visuo-spatial navigation and memory (40). Given \u003cem\u003eIns2\u003c/em\u003e expression is prominent within the hippocampus relative to other neuronal populations, we next examined whether hippocampal-dependent behaviour was affected by \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e ablation in the Morris water maze, a commonly used and well-validated test of visuo-spatial learning and memory in rodents (41). Given that the sexual dimorphism in hippocampal \u003cem\u003eIns2\u003c/em\u003e expression, we compared aged female (\u003cstrong\u003eFigure 4A-F\u003c/strong\u003e) and male (\u003cstrong\u003eFigure 4G-L\u003c/strong\u003e) littermates of both genotypes. During acquisition training, female\u003cem\u003e\u0026nbsp;Ins2\u003csup\u003e-/-\u003c/sup\u003e\u0026nbsp;\u003c/em\u003eand \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003emice did not differ on measures of learning including latency (\u003cstrong\u003eFigure 4A\u003c/strong\u003e), distance (\u003cstrong\u003eFigure 4D\u003c/strong\u003e), or cumulative search error (\u003cstrong\u003eFigure 4B\u003c/strong\u003e). During the acquisition probe memory test, however, female \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u0026nbsp;\u003c/em\u003ecompleted fewer crossings over the location of the escape platform (annulus crossings) than \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003emice \u003cstrong\u003e(Figure 4C)\u003c/strong\u003e. We then provided an additional day of acquisition training (re-training, RT) to reduce potential extinction effects, and to provide an additional indirect measure of memory by assessing retention of learning performance from the last day of acquisition training (2 days prior). On the re-training day, female \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e performed worse than \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003e, mice on all measures of learning performance (\u003cstrong\u003eFigure 4A,B,D\u003c/strong\u003e). During the first 3 days of reversal training, female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003edid not differ on measures of learning performance; over the last 3 days of reversal training, however, female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice performed worse than \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003e mice on all measures of learning performance (\u003cstrong\u003eFigure 4A,B,D\u003c/strong\u003e). During the reversal probe memory test, \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003emice completed fewer crossings over the escape platform location than \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u0026nbsp;\u003c/em\u003emice \u003cstrong\u003e(Figure 4D)\u003c/strong\u003e. We also measured swim speed of mice throughout acquisition and reversal training, which can reflect motivation to find the escape platform. Although, female \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003emice swam quicker than \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003e mice during the first 3 days of acquisition training, this difference was not present during the last 3 days of acquisition training, or at any point during reversal training \u003cstrong\u003e(Figure 4E)\u003c/strong\u003e. Given that differences in learning performance were not present in acquisition training, we can conclude that differences in swim-speed do not account for the impaired learning and memory observed in \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003emice. Taken together, these results indicate impaired learning and memory for the reversal escape platform location in \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003emice.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe also assessed learning and memory in male \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e and \u003cem\u003eIns2\u003csup\u003e+/+\u003c/sup\u003e\u003c/em\u003e mice in the Morris water maze but did not find any differences in learning performance during either acquisition or reversal training \u003cstrong\u003e(Figure 4G,H,J,K).\u003c/strong\u003e We also did not observe differences in memory performance during acquisition or reversal memory probe tests \u003cstrong\u003e(Figure 4I,L)\u003c/strong\u003e. Thus, unlike female mice, loss of \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e in male mice is not sufficient to disrupt visuo-spatial learning or memory.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eLoss of Ins2 does not influence anxiety-like behaviour or locomotor activity.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe function of the hippocampus is known to differ along its dorsal-ventral axis, with the dorsal hippocampus being involved in visuo-spatial navigation and memory, and the ventral hippocampus being involved more so in anxiety-related behaviour (42, 43). Therefore, we examined anxiety-like behaviour in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003emice, to provide a more complete picture for the role of \u003cem\u003eIns2\u003c/em\u003e on hippocampal function. \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice were tested on three commonly used tests of anxiety-like behaviour: the open-field, light-dark box, and elevated-plus maze. All three tests were effective in producing marked avoidance of anxiogenic areas, with mice spending less time in anxiogenic regions than expected by chance \u003cstrong\u003e(Supplemental Figure 1)\u003c/strong\u003e. For both males and females, however, \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u0026nbsp;\u003c/sup\u003emice did not differ in anxiety-related behaviour on the OF, LDB or EPM. We also extended our analysis to include species-typical measures of exploration and anxiety-like behaviour (i.e. grooming, rearing, freezing, etc.), but did not find any differences between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice \u003cstrong\u003e(Supplemental Figure 1)\u003c/strong\u003e. Each of the OF, LDB and EPM can also be used to assess locomotor activity within novel environments, however, we did not observe any differences between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u0026nbsp;\u003c/sup\u003eand \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice in locomotor activity in either males or females \u003cstrong\u003e(Supplemental Figure 1)\u003c/strong\u003e. Thus, the effects of \u003cem\u003eIns2\u003c/em\u003e knockout in female mice appear to be specific to learning and memory in the Morris water maze.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eTranscriptomics of hippocampi from aged female Ins2\u003csup\u003e-/-\u003c/sup\u003e mice identifies cyclin D1 as downregulated\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted bulk hippocampal RNA-seq in aged female \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e mice to identify potential molecular mechanisms that could underlie the defects in learning and memory. We started with 11 samples in each group, but three samples were deemed outliers based on hierarchical clustering and PCA and were removed from the analysis (\u003cstrong\u003eFigure 5A,B\u003c/strong\u003e). We confirmed the absence of \u003cem\u003eIns2\u003c/em\u003e mRNA in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice \u003cstrong\u003e(Figure 5C\u003c/strong\u003e). Then, we identified 6 downregulated and 13 upregulated genes in \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e mice\u0026nbsp;\u003c/em\u003e(\u003cstrong\u003eFigure 5D\u003c/strong\u003e). Among them, the cell cycle regulatory gene \u003cem\u003eCcnd1\u003c/em\u003e and the stem cell regulatory long non-coding RNA Gm26793 (44) were the most significantly downregulated and upregulated genes, respectively. It should be noted that these genes were robustly altered even when the outlier samples were included. Gene set enrichment analysis showed that the (semi-redundant) gene ontology terms “aerobic respiration” and “oxidative phosphorylation” were the most significantly upregulated pathways (\u003cstrong\u003eFigure 5E\u003c/strong\u003e). Insulin is known to be a regulator of oxidative phosphorylation in many tissues, including the brain (45). The gene ontology term “epithelium development” was the only significantly downregulated pathway (\u003cstrong\u003eFigure 5E\u003c/strong\u003e). Together, these relatively well-powered transcriptomic studies point to potential molecular mechanisms underlying the effects of insulin signalling the in female rodent hippocampus.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNeurogenesis in Ins2\u003csup\u003e-/-\u003c/sup\u003e mice.\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eGiven that \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice had reduced \u003cem\u003eCcnd1\u003c/em\u003e mRNA, a gene with known roles in neurogenesis, we next examined if hippocampal neurogenesis was altered in aged \u003cem\u003eIns2\u003csup\u003e-/-\u003c/sup\u003e\u003c/em\u003e mice using EdU, a thymidine analogue that incorporates into proliferating cells, and by identifying immature neurons via doublecortin. We first confirmed the robustness of our staining by confirming Edu positive cells in intestinal villi, a region with high levels of cell proliferation (\u003cstrong\u003eSupplemental\u003c/strong\u003e \u003cstrong\u003eFigure 2A)\u003c/strong\u003e. In female brains, the densities of Edu positive cells and DCX positive cells did not differ between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice, in either the dorsal or ventral dentate gyrus (\u003cstrong\u003eSupplemental\u003c/strong\u003e \u003cstrong\u003eFigure 2B-D\u003c/strong\u003e). In male brains, \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice had a larger density of EdU positive cells in the dorsal (F(1,10) = 5.57, p\u0026lt;0.05)(\u003cstrong\u003e\u0026nbsp;Supplemental\u003c/strong\u003e \u003cstrong\u003eFigure 2C\u003c/strong\u003e), but not ventral hippocampus (\u003cstrong\u003eFigure 2D\u003c/strong\u003e). One could speculate that this may represent compensation that may prevent male \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice from exhibiting defects in learning and memory. The density of DCX cells did not differ between \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e and Ins2\u003csup\u003e+/+\u003c/sup\u003e male mice in either the dorsal or ventral hippocampus (\u003cstrong\u003eSupplemental\u003c/strong\u003e \u003cstrong\u003eFigure 2E,F\u003c/strong\u003e). Together, these data suggest that adult neurogenesis, measured at this late stage of life, is not a major factor in differences in learning and memory in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e-/-\u003c/sup\u003e mice.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe objectives of this study were to further elucidate the expression of insulin in the hippocampus and to determine the role of brain-produced insulin on hippocampus-dependent learning and memory. Using \u003cem\u003eIns2\u003c/em\u003e knockout mice devoid of hippocampal pro-insulin, but with limited effects on peripheral metabolism, we defined a female-specific role for brain-insulin production in learning and memory.\u003c/p\u003e\u003cp\u003eGiven that \u003cem\u003eIns2\u003c/em\u003e is expressed in both the brain and pancreas, it was important to assess the effect of \u003cem\u003eIns2\u003c/em\u003e knockout on peripheral metabolism, which could theoretically affect brain function indirectly. In-depth metabolic characterization of \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice showed that they were similar in most measures of metabolism, including body weight, glucose tolerance and insulin tolerance tests. Together these results suggest that the changes in \u003cem\u003eIns2\u003c/em\u003e expression and behavioral impairments observed in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice are likely not due to peripheral metabolic dysfunction. The relatively few metabolic changes in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice indicates that the \u003cem\u003eIns1\u003c/em\u003e allele can produce a near-complete compensation for the loss of \u003cem\u003eIns2\u003c/em\u003e in the context of the diet and housing conditions used in our study, and that this compensation can be maintained through aging. The mechanism for this compensation is largely unexplored but is likely supported by elevated \u003cem\u003eIns1\u003c/em\u003e expression within the pancreas and pancreatic β-cell hyperplasia (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). This is likely due to inter-allelic redundancy and compensation within the pancreas, where both alleles contribute to insulin production, and may be especially important under periods of metabolic stress. Indeed, we found that metabolic changes in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice were present only at 16\u0026ndash;18 months, when the accumulated stressors associated with aging were expected to be the largest. However, given that \u003cem\u003eIns2\u003c/em\u003e is also expressed in the brain, we cannot rule out that loss of brain-produced insulin also contributed to the few metabolic differences observed in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice.\u003c/p\u003e\u003cp\u003eTo the best of our knowledge, this is the first time cognitive behaviour has been examined in the context of genetic insulin reduction. Our study was also unique in that we deleted only \u003cem\u003eIns2\u003c/em\u003e, resulting in brain-selective insulin knockout. To contextualize our knockout model, we extended on our previous work that provided conclusive evidence, using knockout controls, that insulin is produce locally in the brain, and more prominently in the hippocampus relative to other brain regions (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). Other work has examined \u003cem\u003eIns2\u003c/em\u003e expression and insulin synthesis across different brain regions in a plethora of animal models (\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e). Studies that include hippocampal regions consistently report elevated levels of insulin expression relative to other brain regions suggesting a role for local \u003cem\u003eIns2\u003c/em\u003e production in hippocampal function (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e). This is further supported by \u003cem\u003ein vitro\u003c/em\u003e studies reporting \u003cem\u003eIns2\u003c/em\u003e expression and insulin synthesis within cultured primary hippocampal neurons (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e). Importantly, we confirmed that Ins2 protein is found in the hippocampus of wildtype but not \u003cem\u003eIns2\u003c/em\u003e knockout mice. Moreover, we show that female mice show higher \u003cem\u003eIns2\u003c/em\u003e expression than males in both the hippocampus and cerebellum. There are currently very few studies that examine sex differences in \u003cem\u003eIns2\u003c/em\u003e expression within the brain, and to the best of our knowledge, this is the first study to show that \u003cem\u003eIns2\u003c/em\u003e expression within the hippocampus is sexually dimorphic. It had been proposed that insulin within the brain can be modulated by environmental factors including diet (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e), and here we show that environmental effects on \u003cem\u003eIns2\u003c/em\u003e brain expression are also sexual dimorphic.\u003c/p\u003e\u003cp\u003eWomen are at higher risk to develop early onset Alzheimer\u0026rsquo;s disease (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e). Consistent with this, we found evidence for a female-specific role of brain-derived insulin in hippocampal-dependent learning and memory, particularly in spatial-visual tasks. It has been suggested that male rodents have better spatial abilities like spatial working memory and orientation, whereas females are better in location memory (\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e). In our experiments, there were not significant differences in learning or memory measures between male and female wildtype mice. When compared to other studies, it should be noted that our experiments were conducted on older mice, which is more relevant to Alzheimer\u0026rsquo;s disease.\u003c/p\u003e\u003cp\u003eThe ability to perform tasks of learning can be influenced by non-cognitive factors, thus potentially confounding measures of learning and memory (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e). For example, performance on the Morris water maze can be influenced by differences in stress reactivity, motivation to locate the escape platform, impaired visual ability and impaired swimming ability due to motor dysfunction (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e). The similar anxiety-like behavioural profiles in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e and \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e+/+\u003c/sup\u003e mice suggests that our reported differences in visuo-spatial learning and memory are not likely due to pre-existing differences in stress reactivity within the water maze. This is further supported by the lack of differences in swim-speed observed in the water maze indicating unimpaired swimming ability and similar motivation to locate the escape platform. Therefore, these results provide additional confidence that differences in learning and memory on the water maze in female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice were not due to potential non-cognitive factors influencing performance.\u003c/p\u003e\u003cp\u003eThe functional role of the hippocampus has been proposed to differ along the dorso-ventral axis, with the dorsal hippocampus involved moreso in cognitive function, and the ventral hippocampus involved in stress and emotionality. Given that female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice showed impaired learning and memory, we also examined if differences extended to anxiety-like behaviour. We did not observe any differences in anxiety-like behaviour in either male or female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice. We should note, however, that over-expressing \u003cem\u003eIns2\u003c/em\u003e within the choroid plexus is sufficient to reduce anxiety-like behaviour, but only when mice are given a high-fat diet (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e). This may indicate a role for \u003cem\u003eIns2\u003c/em\u003e in regulating other brain behaviours, including anxiety, specifically during metabolic challenges.\u003c/p\u003e\u003cp\u003eThe \u003cem\u003eIns2\u003c/em\u003e knockout mouse model in this study is most appropriate to directly test the effects of endogenous insulin of the brain to date. Silencing the \u003cem\u003eIns2\u003c/em\u003e gene, while keeping \u003cem\u003eIns1\u003c/em\u003e intact, allows to specifically examine the causal effects of insulin synthesis in the brain independent of peripheral insulin. However, our model does not come without its own limitations. The \u003cem\u003eIns2\u003c/em\u003e gene is also, in addition to the pancreas and brain, expressed in the thymus. The thymus is most known for its role in immunity and insulin deletion in the thymus can induce autoimmune diabetes (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e). Thus, with a global \u003cem\u003eIns2\u003c/em\u003e knockout, it cannot be entirely excluded that the results of this study are exclusively due to brain-insulin loss but may indirectly relate to immune dysfunction. Generating tissue-specific \u003cem\u003eIns2\u003c/em\u003e knockout could be used to circumvent this limitation, but our efforts to do so have been unsuccessful. Another limitation of our study was that the loss of \u003cem\u003eIns2\u003c/em\u003e was lifelong. Future studies could employ a floxed \u003cem\u003eIns2\u003c/em\u003e allele and brain-specific Cre viral vectors or drug-inducible transgenes in mice to remove \u003cem\u003eIns2\u003c/em\u003e in the adult brain, in a region-specific manner. Other methods of acutely testing for local brain insulin ablated insulin-expressing neurons with the beta-cell toxin streptozotocin (\u003cspan additionalcitationids=\"CR58\" citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e). However, toxic drugs may also affect other areas of the brain, and it would be difficult to say whether effects on behavior are due to brain-derived insulin loss or generalized toxicity. Our work also does not identify the cellular target(s) of local insulin production, not their molecular mechanisms. Insulin receptors have been shown to mediate important roles of insulin in many neuronal cell populations in the brain, including the hippocampus (\u003cspan additionalcitationids=\"CR61\" citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e). Insulin also acts directly though insulin receptors to modulate the activity of non-neuronal brain cells including astrocytes (\u003cspan additionalcitationids=\"CR64\" citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e). Indeed, inducible loss of insulin action via insulin receptor and IGF1 receptor deletion in astrocytes significantly worsened multiple Alzheimer\u0026rsquo;s pathologies in a mouse model (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). In a recent study, cerebrovascular insulin receptors were also shown to be deficient in Alzheimer\u0026rsquo;s disease (\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e). Clearly, more research will be required to fully elucidate local insulin-insulin receptor trophic circuits across multiple cell types in the brain.\u003c/p\u003e\u003cp\u003eIn conclusion, we confirmed that \u003cem\u003eIns2\u003c/em\u003e is produced in the hippocampus and discovered a significant sex difference in mRNA levels. This study used a genetic model of \u003cem\u003eIns2\u003c/em\u003e deletion to assess the influence of brain insulin production, in the absence of major metabolic disturbances. We also showed for the first time that insulin produced in the brain plays an important role in spatial navigation in female mice.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eDeclaration of Interests: \u003c/strong\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e Work\u0026nbsp;was supported by a Brain Canada Multi-Investigator grant to J.D.J., S.X.B., and P.P.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments: \u003c/strong\u003eWe thank many colleagues for helpful discussions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability:\u0026nbsp;\u003c/strong\u003eRNA sequencing data are online. All other data are available on request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eS.B. designed studies, performed experiments, analyzed/interpreted data, and co-wrote the manuscript.\u003c/p\u003e\n\u003cp\u003eT.O. designed studies, performed experiments, analyzed/interpreted data, and co-wrote the manuscript.\u003c/p\u003e\n\u003cp\u003eH.H.C. analyzed bioinformatic data and helped write the manuscript.\u003c/p\u003e\n\u003cp\u003eD.H. performed experiments, analyzed/interpreted data.\u003c/p\u003e\n\u003cp\u003eK.K. performed experiments.\u003c/p\u003e\n\u003cp\u003eS.A. performed experiments.\u003c/p\u003e\n\u003cp\u003eD.S. performed experiments.\u003c/p\u003e\n\u003cp\u003eM.B. performed experiments and analyzed bioinformatic data.\u003c/p\u003e\n\u003cp\u003eH.L. performed experiments.\u003c/p\u003e\n\u003cp\u003eA.E.M. performed experiments, analyzed/interpreted data.\u003c/p\u003e\n\u003cp\u003eH.M. performed experiments, analyzed/interpreted data.\u003c/p\u003e\n\u003cp\u003eM.M.P. performed experiments, analyzed/interpreted data\u003c/p\u003e\n\u003cp\u003eS.S. performed experiments, analyzed/interpreted data.\u003c/p\u003e\n\u003cp\u003eP.P. supported bioinformatic studies, interpreted data, and edited the manuscript\u003c/p\u003e\n\u003cp\u003eS.X.B. designed studies, interpreted data, and edited the manuscript\u003c/p\u003e\n\u003cp\u003eE-K. K. analyzed data and provided a key reagent.\u003c/p\u003e\n\u003cp\u003eJ.D.J. conceived the project, designed studies, interpreted data, co-wrote the manuscript, and is the guarantor of this work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eH\u0026ouml;lscher, C. (2011) Diabetes as a risk factor for Alzheimer\u0026apos;s disease: insulin signalling impairment in the brain as an alternative model of Alzheimer\u0026apos;s disease. Portland Press Ltd.\u003c/li\u003e\n\u003cli\u003eRyan, C. M., and Geckle, M. (2000) Why is learning and memory dysfunction in Type 2 diabetes limited to older adults? \u003cem\u003eDiabetes/metabolism research and reviews\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 308-315\u003c/li\u003e\n\u003cli\u003eDe Felice, F. G., Lourenco, M. V., and Ferreira, S. T. (2014) How does brain insulin resistance develop in Alzheimer\u0026apos;s disease? \u003cem\u003eAlzheimer\u0026apos;s \u0026amp; Dementia\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, S26-S32\u003c/li\u003e\n\u003cli\u003eTalbot, K., Wang, H.-Y., Kazi, H., Han, L.-Y., Bakshi, K. P., Stucky, A., Fuino, R. L., Kawaguchi, K. R., Samoyedny, A. J., Wilson, R. S., Arvanitakis, Z., Schneider, J. A., Wolf, B. A., Bennett, D. A., Trojanowski, J. Q., and Arnold, S. E. (2012) Demonstrated brain insulin resistance in Alzheimer\u0026rsquo;s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. \u003cem\u003eJournal of Clinical Investigation\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, 1316-1338\u003c/li\u003e\n\u003cli\u003eForner, S., Baglietto-Vargas, D., Martini, A. C., Trujillo-Estrada, L., and LaFerla, F. M. (2017) Synaptic Impairment in Alzheimer\u0026apos;s Disease: A Dysregulated Symphony. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 347-357\u003c/li\u003e\n\u003cli\u003eKellar, D., and Craft, S. (2020) Brain insulin resistance in Alzheimer\u0026apos;s disease and related disorders: mechanisms and therapeutic approaches. \u003cem\u003eLancet Neurol\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 758-766\u003c/li\u003e\n\u003cli\u003eLaws, S. M., Gaskin, S., Woodfield, A., Srikanth, V., Bruce, D., Fraser, P. E., Porter, T., Newsholme, P., Wijesekara, N., and Burnham, S. (2017) Insulin resistance is associated with reductions in specific cognitive domains and increases in CSF tau in cognitively normal adults. \u003cem\u003eScientific reports\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1-11\u003c/li\u003e\n\u003cli\u003eMa, L., Wang, J., and Li, Y. (2015) Insulin resistance and cognitive dysfunction. \u003cem\u003eClinica chimica acta\u003c/em\u003e \u003cstrong\u003e444\u003c/strong\u003e, 18-23\u003c/li\u003e\n\u003cli\u003eTalbot, K., Wang, H.-Y., Kazi, H., Han, L.-Y., Bakshi, K. P., Stucky, A., Fuino, R. L., Kawaguchi, K. R., Samoyedny, A. J., and Wilson, R. S. (2012) Demonstrated brain insulin resistance in Alzheimer\u0026rsquo;s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. \u003cem\u003eThe Journal of clinical investigation\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, 1316-1338\u003c/li\u003e\n\u003cli\u003eChen, W., Huang, Q., Lazdon, E. K., Gomes, A., Wong, M., Stephens, E., Royal, T. G., Frenkel, D., Cai, W., and Kahn, C. R. (2023) Loss of insulin signaling in astrocytes exacerbates Alzheimer-like phenotypes in a 5xFAD mouse model. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, e2220684120\u003c/li\u003e\n\u003cli\u003eCai, W., Xue, C., Sakaguchi, M., Konishi, M., Shirazian, A., Ferris, H. A., Li, M. E., Yu, R., Kleinridders, A., and Pothos, E. N. (2018) Insulin regulates astrocyte gliotransmission and modulates behavior. \u003cem\u003eThe Journal of clinical investigation\u003c/em\u003e \u003cstrong\u003e128\u003c/strong\u003e, 2914-2926\u003c/li\u003e\n\u003cli\u003eKuwabara, T., Kagalwala, M. N., Onuma, Y., Ito, Y., Warashina, M., Terashima, K., Sanosaka, T., Nakashima, K., Gage, F. H., and Asashima, M. (2011) Insulin biosynthesis in neuronal progenitors derived from adult hippocampus and the olfactory bulb. \u003cem\u003eEMBO Mol Med\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 742-754\u003c/li\u003e\n\u003cli\u003eDevaskar, S. U., Giddings, S. J., Rajakumar, P. A., Carnaghi, L. R., Menon, R. K., and Zahm, D. S. (1994) Insulin gene expression and insulin synthesis in mammalian neuronal cells. \u003cem\u003eJ Biol Chem\u003c/em\u003e \u003cstrong\u003e269\u003c/strong\u003e, 8445-8454\u003c/li\u003e\n\u003cli\u003eFreiherr, J., Hallschmid, M., Frey, W. H., Br\u0026uuml;nner, Y. F., Chapman, C. D., H\u0026ouml;lscher, C., Craft, S., De Felice, F. G., and Benedict, C. (2013) Intranasal insulin as a treatment for Alzheimer\u0026rsquo;s disease: a review of basic research and clinical evidence. \u003cem\u003eCNS drugs\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 505-514\u003c/li\u003e\n\u003cli\u003eZhang, H., Hao, Y., Manor, B., Novak, P., Milberg, W., Zhang, J., Fang, J., and Novak, V. (2015) Intranasal insulin enhanced resting-state functional connectivity of hippocampal regions in type 2 diabetes. \u003cem\u003eDiabetes\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 1025-1034\u003c/li\u003e\n\u003cli\u003eLi, X., Yang, Y., Bai, X., Wang, X., Tan, H., Chen, Y., Zhu, Y., Liu, Q., Wu, M. N., and Li, Y. (2024) A brain-derived insulin signal encodes protein satiety for nutrient-specific feeding inhibition. \u003cem\u003eCell Rep\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 114282\u003c/li\u003e\n\u003cli\u003eBorn, J., Lange, T., Kern, W., McGregor, G. P., Bickel, U., and Fehm, H. L. (2002) Sniffing neuropeptides: a transnasal approach to the human brain. \u003cem\u003eNature Neuroscience\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 514-516\u003c/li\u003e\n\u003cli\u003eKo, Y., and Chye, S. M. (2020) Lifestyle intervention to prevent Alzheimer\u0026rsquo;s disease. \u003cem\u003eReviews in the Neurosciences\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 817-824\u003c/li\u003e\n\u003cli\u003eBraak, H., and Braak, E. (1995) Staging of Alzheimer\u0026apos;s disease-related neurofibrillary changes. \u003cem\u003eNeurobiology of aging\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 271-278\u003c/li\u003e\n\u003cli\u003eGray, S. M., Aylor, K. W., and Barrett, E. J. (2017) Unravelling the regulation of insulin transport across the brain endothelial cell. \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 1512-1521\u003c/li\u003e\n\u003cli\u003eDakic, T., Jevdjovic, T., Lakic, I., Ruzicic, A., Jasnic, N., Djurasevic, S., Djordjevic, J., and Vujovic, P. (2023) The Expression of Insulin in the Central Nervous System: What Have We Learned So Far? \u003cem\u003eInternational Journal of Molecular Sciences\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 6586\u003c/li\u003e\n\u003cli\u003eMolnar, G., Farago, N., Kocsis, A. K., Rozsa, M., Lovas, S., Boldog, E., Baldi, R., Csajbok, E., Gardi, J., Puskas, L. G., and Tamas, G. (2014) GABAergic neurogliaform cells represent local sources of insulin in the cerebral cortex. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 1133-1137\u003c/li\u003e\n\u003cli\u003eNemoto, T., Toyoshima-Aoyama, F., Yanagita, T., Maruta, T., Fujita, H., Koshida, T., Yonaha, T., Wada, A., Sawaguchi, A., and Murakami, M. (2014) New insights concerning insulin synthesis and its secretion in rat hippocampus and cerebral cortex: amyloid-beta1-42-induced reduction of proinsulin level via glycogen synthase kinase-3beta. \u003cem\u003eCell Signal\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 253-259\u003c/li\u003e\n\u003cli\u003eMazucanti, C. H., Liu, Q. R., Lang, D., Huang, N., O\u0026apos;Connell, J. F., Camandola, S., and Egan, J. M. (2019) Release of insulin produced by the choroid plexis is regulated by serotonergic signaling. \u003cem\u003eJCI Insight\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eMehran, A. E., Templeman, N. M., Brigidi, G. S., Lim, G. E., Chu, K. Y., Hu, X., Botezelli, J. D., Asadi, A., Hoffman, B. G., Kieffer, T. J., Bamji, S. X., Clee, S. M., and Johnson, J. D. (2012) Hyperinsulinemia drives diet-induced obesity independently of brain insulin production. \u003cem\u003eCell Metab\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 723-737\u003c/li\u003e\n\u003cli\u003eLee, J., Kim, K., Cho, J. H., Bae, J. Y., O\u0026apos;Leary, T. P., Johnson, J. D., Bae, Y. C., and Kim, E. K. (2020) Insulin synthesized in the paraventricular nucleus of the hypothalamus regulates pituitary growth hormone production. \u003cem\u003eJCI Insight\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e\u003c/li\u003e\n\u003cli\u003eMazucanti, C. H., Kennedy, V., Jr., Premathilake, H. U., Doyle, M. E., Tian, J., Liu, Q. R., O\u0026apos;Connell, J., Camandola, S., and Egan, J. M. (2023) AAV5-mediated manipulation of insulin expression in choroid plexus has long-term metabolic and behavioral consequences. \u003cem\u003eCell Rep\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 112903\u003c/li\u003e\n\u003cli\u003eDuvillie, B., Cordonnier, N., Deltour, L., Dandoy-Dron, F., Itier, J. M., Monthioux, E., Jami, J., Joshi, R. L., and Bucchini, D. (1997) Phenotypic alterations in insulin-deficient mutant mice. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 5137-5140\u003c/li\u003e\n\u003cli\u003eBerdugo-Vega, G., Dhingra, S., and Calegari, F. (2023) Sharpening the blades of the dentate gyrus: how adult-born neurons differentially modulate diverse aspects of hippocampal learning and memory. \u003cem\u003eEmbo J\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, e113524\u003c/li\u003e\n\u003cli\u003eHitoshi, S., Mandyam, C. D., and Shetty, A. K. (2023) Editorial: Advances in adult neurogenesis. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 1307844\u003c/li\u003e\n\u003cli\u003eBrooker, G. J., Kalloniatis, M., Russo, V. C., Murphy, M., Werther, G. A., and Bartlett, P. F. (2000) Endogenous IGF‐1 regulates the neuronal differentiation of adult stem cells. \u003cem\u003eJournal of neuroscience research\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 332-341\u003c/li\u003e\n\u003cli\u003e\u0026Aring;berg, M. A., \u0026Aring;berg, N. D., Palmer, T. D., Alborn, A.-M., Carlsson-Skwirut, C., Bang, P., Rosengren, L. E., Olsson, T., Gage, F. H., and Eriksson, P. S. (2003) IGF-I has a direct proliferative effect in adult hippocampal progenitor cells. \u003cem\u003eMolecular and Cellular Neuroscience\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 23-40\u003c/li\u003e\n\u003cli\u003eDuvilli\u0026eacute;, B., Cordonnier, N., Deltour, L., Dandoy-Dron, F., Itier, J.-M., Monthioux, E., Jami, J., Joshi, R. L., and Bucchini, D. (1997) Phenotypic alterations in insulin-deficient\u0026thinsp;mutant\u0026thinsp;mice. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e94\u003c/strong\u003e, 5137-5140\u003c/li\u003e\n\u003cli\u003eChu, C. M. J., Modi, H., Ellis, C., Krentz, N. A. J., Skovso, S., Zhao, Y. B., Cen, H., Noursadeghi, N., Panzhinskiy, E., Hu, X., Dionne, D. A., Xia, Y. H., Xuan, S., Huising, M. O., Kieffer, T. J., Lynn, F. C., and Johnson, J. D. (2022) Dynamic Ins2 Gene Activity Defines beta-Cell Maturity States. \u003cem\u003eDiabetes\u003c/em\u003e \u003cstrong\u003e71\u003c/strong\u003e, 2612-2631\u003c/li\u003e\n\u003cli\u003eKim, S., Kim, N., Park, S., Jeon, Y., Lee, J., Yoo, S. J., Lee, J. W., Moon, C., Yu, S. W., and Kim, E. K. (2020) Tanycytic TSPO inhibition induces lipophagy to regulate lipid metabolism and improve energy balance. \u003cem\u003eAutophagy\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 1200-1220\u003c/li\u003e\n\u003cli\u003eRitchie, M. E., Phipson, B., Wu, D., Hu, Y., Law, C. W., Shi, W., and Smyth, G. K. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. \u003cem\u003eNucleic Acids Research\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, e47-e47\u003c/li\u003e\n\u003cli\u003eLove, M. I., Huber, W., and Anders, S. (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 550\u003c/li\u003e\n\u003cli\u003eYu, G., Wang, L. G., Han, Y., and He, Q. Y. (2012) clusterProfiler: an R package for comparing biological themes among gene clusters. \u003cem\u003eOMICS\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 284-287\u003c/li\u003e\n\u003cli\u003eLeroux, L., Desbois, P., Lamotte, L., Duvillie, B., Cordonnier, N., Jackerott, M., Jami, J., Bucchini, D., and Joshi, R. L. (2001) Compensatory responses in mice carrying a null mutation for Ins1 or Ins2. \u003cem\u003eDiabetes\u003c/em\u003e \u003cstrong\u003e50 Suppl 1\u003c/strong\u003e, S150-153\u003c/li\u003e\n\u003cli\u003eVorhees, C. V., and Williams, M. T. (2014) Assessing spatial learning and memory in rodents. \u003cem\u003eIlar j\u003c/em\u003e \u003cstrong\u003e55\u003c/strong\u003e, 310-332\u003c/li\u003e\n\u003cli\u003eD\u0026apos;Hooge, R., and De Deyn, P. P. (2001) Applications of the Morris water maze in the study of learning and memory. \u003cem\u003eBrain Res Brain Res Rev\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 60-90\u003c/li\u003e\n\u003cli\u003eBannerman, D. M., Grubb, M., Deacon, R. M., Yee, B. K., Feldon, J., and Rawlins, J. N. (2003) Ventral hippocampal lesions affect anxiety but not spatial learning. \u003cem\u003eBehav Brain Res\u003c/em\u003e \u003cstrong\u003e139\u003c/strong\u003e, 197-213\u003c/li\u003e\n\u003cli\u003eFanselow, M. S., and Dong, H. W. (2010) Are the dorsal and ventral hippocampus functionally distinct structures? \u003cem\u003eNeuron\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 7-19\u003c/li\u003e\n\u003cli\u003eLiu, Z., Wan, X., Chen, J., Ma, Y., Fu, Y., Chen, Y., Wen, M., Yang, Y., Qian, Y., Zhang, Y., Zhu, D., Li, J., Jing, N., and Yang, X. (2025) Genomic locus of lncRNA-Gm26793 forms an inter-chromosomal interaction with Cubn to ensure proper stem cell differentiation in vitro and in vivo. \u003cem\u003eCell Discov\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 53\u003c/li\u003e\n\u003cli\u003eMilstein, J. L., and Ferris, H. A. (2021) The brain as an insulin-sensitive metabolic organ. \u003cem\u003eMol Metab\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 101234\u003c/li\u003e\n\u003cli\u003eYoung III, W. S. (1986) Periventricular hypothalamic cells in the rat brain contain insulin mRNA. \u003cem\u003eNeuropeptides\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 93-97\u003c/li\u003e\n\u003cli\u003eHavrankova, J., Schmechel, D., Roth, J., and Brownstein, M. (1978) Identification of insulin in rat brain. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 5737-5741\u003c/li\u003e\n\u003cli\u003eCsajbok, E. A., and Tamas, G. (2016) Cerebral cortex: a target and source of insulin? \u003cem\u003eDiabetologia\u003c/em\u003e \u003cstrong\u003e59\u003c/strong\u003e, 1609-1615\u003c/li\u003e\n\u003cli\u003eKuwabara, T., Kagalwala, M. N., Onuma, Y., Ito, Y., Warashina, M., Terashima, K., Sanosaka, T., Nakashima, K., Gage, F. H., and Asashima, M. (2011) Insulin biosynthesis in neuronal progenitors derived from adult hippocampus and the olfactory bulb. \u003cem\u003eEMBO Molecular Medicine\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 742-754\u003c/li\u003e\n\u003cli\u003eNemoto, T., Toyoshima-Aoyama, F., Yanagita, T., Maruta, T., Fujita, H., Koshida, T., Yonaha, T., Wada, A., Sawaguchi, A., and Murakami, M. (2014) New insights concerning insulin synthesis and its secretion in rat hippocampus and cerebral cortex: Amyloid-\u0026beta;1\u0026ndash;42-induced reduction of proinsulin level via glycogen synthase kinase-3\u0026beta;. \u003cem\u003eCellular Signalling\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 253-259\u003c/li\u003e\n\u003cli\u003ede la Monte, S. M., Tong, M., Bowling, N., and Moskal, P. (2011) si-RNA inhibition of brain insulin or insulin-like growth factor receptors causes developmental cerebellar abnormalities: relevance to fetal alcohol spectrum disorder. \u003cem\u003eMol Brain\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 13\u003c/li\u003e\n\u003cli\u003eNeu, S. C., Pa, J., Kukull, W., Beekly, D., Kuzma, A., Gangadharan, P., Wang, L.-S., Romero, K., Arneric, S. P., and Redolfi, A. (2017) Apolipoprotein E genotype and sex risk factors for Alzheimer disease: a meta-analysis. \u003cem\u003eJAMA neurology\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 1178-1189\u003c/li\u003e\n\u003cli\u003eFleischer, A. W., and Frick, K. M. (2023) New perspectives on sex differences in learning and memory. \u003cem\u003eTrends in Endocrinology \u0026amp; Metabolism\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 526-538\u003c/li\u003e\n\u003cli\u003eCahill, L., McGaugh, J. L., and Weinberger, N. M. (2001) The neurobiology of learning and memory: some reminders to remember. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 578-581\u003c/li\u003e\n\u003cli\u003eO\u0026apos;Leary, T. P., and Brown, R. E. (2022) Visuo-spatial learning and memory impairments in the 5xFAD mouse model of Alzheimer\u0026apos;s disease: Effects of age, sex, albinism, and motor impairments. \u003cem\u003eGenes Brain Behav\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, e12794\u003c/li\u003e\n\u003cli\u003eFan, Y., Rudert, W. A., Grupillo, M., He, J., Sisino, G., and Trucco, M. (2009) Thymus-specific deletion of insulin induces autoimmune diabetes. \u003cem\u003eThe EMBO Journal\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 2812-2824\u003c/li\u003e\n\u003cli\u003eGr\u0026uuml;nblatt, E., Salkovic‐Petrisic, M., Osmanovic, J., Riederer, P., and Hoyer, S. (2007) Brain insulin system dysfunction in streptozotocin intracerebroventricularly treated rats generates hyperphosphorylated tau protein. \u003cem\u003eJournal of neurochemistry\u003c/em\u003e \u003cstrong\u003e101\u003c/strong\u003e, 757-770\u003c/li\u003e\n\u003cli\u003ePlaschke, K., Kopitz, J., Siegelin, M., Schliebs, R., Salkovic-Petrisic, M., Riederer, P., and Hoyer, S. (2010) Insulin-resistant brain state after intracerebroventricular streptozotocin injection exacerbates Alzheimer-like changes in Tg2576 A\u0026beta;PP-overexpressing mice. \u003cem\u003eJournal of Alzheimer\u0026apos;s disease\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 691-704\u003c/li\u003e\n\u003cli\u003eLester-Coll, N., Rivera, E. J., Soscia, S. J., Doiron, K., Wands, J. R., and de la Monte, S. M. (2006) Intracerebral streptozotocin model of type 3 diabetes: relevance to sporadic Alzheimer\u0026apos;s disease. \u003cem\u003eJournal of Alzheimer\u0026apos;s Disease\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 13-33\u003c/li\u003e\n\u003cli\u003eChen, W., Cai, W., Hoover, B., and Kahn, C. R. (2022) Insulin action in the brain: cell types, circuits, and diseases. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 384-400\u003c/li\u003e\n\u003cli\u003eDixon-Salazar, T. J., Fourgeaud, L., Tyler, C. M., Poole, J. R., Park, J. J., and Boulanger, L. M. (2014) MHC class I limits hippocampal synapse density by inhibiting neuronal insulin receptor signaling. \u003cem\u003eJ Neurosci\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 11844-11856\u003c/li\u003e\n\u003cli\u003eSoto, M., Cai, W., Konishi, M., and Kahn, C. R. (2019) Insulin signaling in the hippocampus and amygdala regulates metabolism and neurobehavior. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e116\u003c/strong\u003e, 6379-6384\u003c/li\u003e\n\u003cli\u003eCai, W., Xue, C., Sakaguchi, M., Konishi, M., Shirazian, A., Ferris, H. A., Li, M. E., Yu, R., Kleinridders, A., Pothos, E. N., and Kahn, C. R. (2018) Insulin regulates astrocyte gliotransmission and modulates behavior. \u003cem\u003eJ Clin Invest\u003c/em\u003e \u003cstrong\u003e128\u003c/strong\u003e, 2914-2926\u003c/li\u003e\n\u003cli\u003eTaib, B., Deme, P., Gupta, S., Yoo, S. W., Khuder, S. S., Hoke, A., Li, Z., Ahima, R. S., and Haughey, N. J. (2025) Insulin acts on astrocytes to shift their substrate preference to fatty acids. \u003cem\u003eiScience\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 111642\u003c/li\u003e\n\u003cli\u003eFernandez, A. M., Martinez-Rachadell, L., Navarrete, M., Pose-Utrilla, J., Davila, J. C., Pignatelli, J., Diaz-Pacheco, S., Guerra-Cantera, S., Viedma-Moreno, E., Palenzuela, R., Ruiz de Martin Esteban, S., Mostany, R., Garcia-Caceres, C., Tsch\u0026ouml;p, M., Iglesias, T., de Ceballos, M. L., Gutierrez, A., and Torres Aleman, I. (2022) Insulin regulates neurovascular coupling through astrocytes. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e119\u003c/strong\u003e, e2204527119\u003c/li\u003e\n\u003cli\u003eLeclerc, M., Bourassa, P., Tremblay, C., Caron, V., Sug\u0026egrave;re, C., Emond, V., Bennett, D. A., and Calon, F. (2022) Cerebrovascular insulin receptors are defective in Alzheimer\u0026rsquo;s disease. \u003cem\u003eBrain\u003c/em\u003e \u003cstrong\u003e146\u003c/strong\u003e, 75-90\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"metabologia","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Metabologia](https://link.springer.com/journal/44357)","snPcode":"44357","submissionUrl":"https://submission.springernature.com/new-submission/44357/3?","title":"Metabologia","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Central insulin production, learning and memory, Alzheimer’s disease, hippocampus","lastPublishedDoi":"10.21203/rs.3.rs-7826673/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7826673/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiabetes is a metabolic disorder characterized by dysfunctional insulin release and action, and it is a risk factor for Alzheimer\u0026rsquo;s disease, the most common form of dementia. Alterations in brain insulin signalling and metabolism have been linked with Alzheimer\u0026rsquo;s disease in multiple studies. It has been previously shown that the ancestral insulin gene, \u003cem\u003eIns2\u003c/em\u003e, is transcribed locally within the brain. Here we demonstrate that \u003cem\u003eIns2\u003c/em\u003e mRNA is higher in females than males, and modulated by diet. Moreover, we demonstrate that the Ins2 protein is found in the hippocampus, a brain region with established roles in learning and memory. To specifically determine how insulin produced locally in the brain influences hippocampal function, specifically learning and memory, we used mice with germline \u003cem\u003eIns2\u003c/em\u003e knockout (\u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e) and the normal complement of wildtype \u003cem\u003eIns1\u003c/em\u003e alleles. Compensation from the \u003cem\u003eIns1\u003c/em\u003e gene ensured normal glucose tolerance, normal insulin sensitivity, normal fasting insulin, and normal body weight under these conditions. We assessed visuo-spatial learning and memory in male and female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e and wild-type littermate control mice using the Morris water maze. Learning and memory performance of female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice was significantly impaired relative to wild-type mice, whereas the performance of male \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e and wild-type mice did not differ. We profiled isolated hippocampi from female \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e and littermate control mice using RNA sequencing to provide an unbiased analysis of gene expression differences that underlie these behavioural changes. Cyclin D1 (\u003cem\u003eCcnd1\u003c/em\u003e) was significantly reduced in \u003cem\u003eIns2\u003c/em\u003e\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e mice, prompting us to examine adult neurogenesis using exogenous mitotic marker EdU and the immature neuronal marker doublecortin (DCX). Collectively, our data demonstrate female-specific roles for brain-derived \u003cem\u003eIns2\u003c/em\u003e on learning and memory function in mice.\u003c/p\u003e","manuscriptTitle":"Loss of brain insulin production impairs learning and memory in female mice","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-27 11:37:18","doi":"10.21203/rs.3.rs-7826673/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-10T12:42:29+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-31T18:11:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"89136842761622845112553666839815673032","date":"2025-10-22T17:13:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-14T09:07:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81996134003783918443403607061146766107","date":"2025-10-02T13:38:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"208241493534845958329196454631791628062","date":"2025-10-01T13:36:26+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-24T11:39:15+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-23T08:10:24+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-23T08:09:55+00:00","index":"","fulltext":""},{"type":"submitted","content":"Metabologia","date":"2025-09-18T10:42:19+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"metabologia","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Metabologia](https://link.springer.com/journal/44357)","snPcode":"44357","submissionUrl":"https://submission.springernature.com/new-submission/44357/3?","title":"Metabologia","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Open","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"66c9aa24-31ee-4793-a348-1f0295d281a4","owner":[],"postedDate":"October 27th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-12T10:56:37+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-27 11:37:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7826673","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7826673","identity":"rs-7826673","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}