Aquaporin-4 inhibition alters cerebral glucose dynamics predominantly in obese animals: an MRI study

Aquaporin-4 inhibition alters cerebral glucose dynamics predominantly in obese animals: an MRI study | 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 Article Aquaporin-4 inhibition alters cerebral glucose dynamics predominantly in obese animals: an MRI study Pablo Tirado-García, Adriana Ferreiro, Raquel González-Alday, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5931441/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 05 May, 2025 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Glucose uptake and metabolism are linked to microvascular blood flow and cellular swelling events, which are altered during obesity and can be quantified using magnetic resonance imaging (MRI). Aquaporin-4 (AQP4), the most abundant water-transporting transmembrane protein in the central nervous system, facilitates glucose transport and metabolism-derived water influx. However, its significance and regulatory capacity remain largely unknown. To better understand these processes, we acquired sequential diffusion tensor and T2*-weighted images of the brains of obese and non-obese mice, both before administering an AQP4 inhibitor and after a subsequent glucose challenge. We then subjected the resulting variables to principal component and linear mixed model analyses to assess the influence of diet, sex, administration of the inhibitor, and brain region on the data. Our findings indicate that AQP4-inhibited mice exhibit MRI values consistent with reduced microvascular blood flow and region-specific inhibition of glucose-induced cell swelling during obesity, highlighting a key role for AQP4 in glucose uptake and metabolism. Additionally, we observed that, prior to any experimental manipulation, obese mice displayed MRI signs of higher cortical blood flow and cerebral cellular anisotropy compared to controls, in agreement with vascular alterations and reactive gliosis processes. Biological sciences/Biophysics Biological sciences/Neuroscience Biological sciences/Biological techniques/Imaging Biological sciences/Computational biology and bioinformatics Biological sciences/Computational biology and bioinformatics/Data processing Biological sciences/Biochemistry Biological sciences/Biochemistry/Neurochemistry Aquaporin-4 Mouse Blood flow Glucose Obesity Swelling TGN-020 Magnetic Resonance Imaging DTI T2* Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 INTRODUCTION Glucose is the main source of energy in the mammalian brain, and is critically important for ATP generation and neurotransmitter synthesis 1 . Its transport and metabolism are the result of a tight balance between microvessels, astrocytes, and neurons, the main components of the neurovascular unit 2 . Obesity, one of the most prevalent diseases of our era, is known to alter such elements and its corresponding equilibrium, including changes in the cerebral vasculature wall, in cerebral blood flow and glucose metabolism 3,4 , increased blood brain-barrier permeability 5 , and the development of astrogliosis and microgliosis 6,7 . Glucose entrance into astrocytes and neurons by their specific transporter channels, as well as the glucose metabolism-associated ion trafficking, involve water co-transport 8 , with water entering the intracellular space mainly by aquaporin-4 ( AQP4 ), the most abundant water channel in the brain 9 . AQP4 is located in astrocytes and ependymal cells, is a key contributor of edema formation and resolution, and it is involved in astrocyte migration and astrogliosis 10 . Alterations in the expression and/or localization of AQP4 lead to a water homeostasis imbalance and have been associated with pathological conditions, including inflammation, stroke or traumatic brain injury 11 . Interestingly, during obesity, AQP4 expression has been reported to be augmented in some brain regions, which has been related to cerebral edema development 12,13 , but decreased in others, potentially due to particular ependymal cells-AQP4 downregulation after a worse glymphatic clearance 14 . During glucose transport and metabolism, water transport through AQP4 is known to yield a certain degree of astrocyte swelling 15–17 (Fig. 1 ), but the extent of the regulatory capacity of such mechanism by AQP4, its influence on glucose metabolism and the corresponding alterations during obesity, remain to be understood. Changes in extracellular cerebral glucose concentration are perceived by specialized glucose-sensing neurons, the glucose-excited (GE) or glucose-inhibited (GI), which alter their firing rate after a glucose stimulus. They have been described in the hypothalamus 18 and in several brain nuclei that are part of the reward system, such as the hippocampus, thalamus and prefrontal cortex. Particularly, on GE neurons, high glucose uptake and the consequent augmented metabolism leads to an increase of the ATP/ADP ratio and the closure of the K ATP channels, increasing neuronal activity, neurotransmitter secretion, and Ca 2+ neuronal influx 19 (Fig. 1 ). This augmented influx into the neurons and its elevated concentration on astrocytes, generates a dilation of the blood vessels 20 . Ionic trafficking in GE neurons may cause, again, astrocyte swelling via increased water entrance through AQP4 transport. Remarkably, glucose-sensing mechanisms are also perturbed during obesity 21 . Magnetic Resonance Imaging (MRI) is a non-invasive imaging technique that provides anatomical and parametric information of structure, function and metabolism 22 . It is broadly used in the clinic as a diagnostic and prognostic tool of numerous brain pathologies such as ischemia, cancer or neurodegeneration 23 . The technique takes advantage of the magnetic properties of the water molecules’ hydrogen atoms, and particularly, diffusion weighted MRI (dMRI) uses the natural Brownian motion of water molecules to obtain information on the anatomical and biological structure of tissues 24 . Depending on the characteristics of such structures, the movement of water molecules can be isotropic, with no preferred direction, or anisotropic, indicating faster diffusion directions, such as in neuronal bundles. Diffusion tensor images (DTI) can measure this anisotropic behavior by quantifying parameters like mean diffusivity (MD, mean diffusion of the water molecules); axial diffusivity (AD, the diffusion along the fastest direction ); radial diffusivity (RD, the diffusion perpendicular to the main direction), and fractional anisotropy (FA, providing information about the extension of the anisotropy of water motion) 25 . Changes in water diffusion have been correlated with changes in AQP4 expression in central nervous system diseases, such as in the early stages of brain edema after hypoxic-ischemic/reperfusion injury, or obesity, which is known to show altered cerebral DTI parameters related to edema formation and neuroinflammation 26 . Diffusion MRI can also be used to tackle cerebral activation by revealing changes in the diffusion properties of water molecules independently to neurovascular coupling. Notably, glucose-induced swelling mechanisms have been reported by dMRI methods in the context of exogenous glucose administration 27 or orexigenic activation 28 . The neuromorphological changes derived from neuronal activity processes are thought to be the cause of the dMRI changes reported during brain activation 29 , but its exact origin, including the potential contribution of activity-derived cellular swelling, time characteristics or sensitivity to monitor normal brain activity, remain to be clarified 30 . T 2 *-weighted imaging (T 2 *WI) is an MRI acquisition able to detect small disturbances in the uniformity of the magnetic field, allowing detection of certain small lesions, and can be used for diagnostic purposes 31 , and detecting blood flow changes 32 . Briefly, the T 2 * relaxation time reflects how quickly the magnetization of water molecules decays due to magnetic field inhomogeneities and magnetic susceptibility differences within tissues. Local changes in the blood oxygen levels are the basis of functional blood oxygen dependent (BOLD) MRI 33 . Indeed, during neuronal activation, T 2 * changes are induced by a change in the ratio deoxyhemoglobin/hemoglobin. Deoxyhemoglobin is a paramagnetic molecule with an iron core that has 4 unpaired electrons per Fe atom, and its presence increases the local magnetic field, and affects T 2 * 34 . Interestingly, previous studies demonstrated brain changes in T 2 *WI images acquired during glucose uptake compared to a control 35 , suggestive of increased microvascular blood flow induced by glucose administration. On these grounds, in this study, we tested the hypothesis that the inhibition of the activity of AQP4 during a glucose challenge would reduce the concomitant swelling process and alter the dMRI effects. Moreover, we hypothesized that the effects on obese animals, where AQP4 expression, glial cells and glucose metabolism are perturbed, would be different. To test that, we acquired three consecutive DTI, and complementary T 2 *WI, to control or high fat diet (HFD) fed C57BL6 animals, before any experimental manipulation (“Basal”), twenty minutes after the administration of saline or 2-(nicotinamide)-1,3,4-thiadiazole ( TGN-020 ), an AQP4 inhibitor which in animal models significantly reduces ischemic brain edema 36 , and immediately after a glucose insult (“Glc short ”), or 30 min post-glucose (“Glc long ”) (Fig. 2 ). We performed two complimentary data analysis of the cerebral MRI parameters. First, we assessed each MRI set separately (basal, Glc short and Glc long ), to unveil the specific effects of diet, sex or treatment ( HFD, sex and TGN-20 effects on T2* and DTI at separated time points ); next, we assessed how MRI the parameters evolved longitudinally in time for each animal, and considered the differences between the experimental groups ( T2* and DTI follow up TGN-20 administration to HFD or CTRL mice ). To do that, we applied principal component analysis (PCA) and subsequent linear mixed-model effects (lme) fitting to the predictor varianles: type of diet, sex, presence of inhibitor (“treatment”) and brain region. With this rationale, we sought to: i) identify potential initial differences of DTI and T 2 *WI derived parameters between obese and non-obese mice, ii) identify sex-related differences in the context of obesity, glucose uptake and AQP4 function, iii) study the specific role of different cerebral regions during obesity, glucose uptake, and TGN-20 inhibition. RESULTS 1. Body weight Male mice of the HFD group had significantly increased body weight (BW) than control animals, both in the “Saline + Glc” and “TGN + Glc” groups (45.1 ± 3.0 Vs 30.3 ± 1.7 for the saline batch, 49.0 ± 3.1 Vs 30.0 ± 1.4 for TGN) (p adj < 0.005). Female HFD animals also depicted higher BW (32.1 ± 5.1 Vs 21.7 ± 0.3 for the saline batch, 30.6 ± 5.4 Vs 21.8 ± 0.4 for TGN) (p adj < 0.05) (anova tests of BW Vs diet, sex and treatment with interaction), although BW differences where not that striking. The groups assigned to different treatment groups showed no significant differences between them. 2. HFD, sex and TGN-20 effects on T2* and DTI at separated time points After performing one independent PCA procedure of the MRI variables per time point - the “Basal”, “Glc short ” and “Glc long ” PCAs -, we achieved a dimensionality reduction, with the first 3 PCA explaining ≥ 90% of the variance (Fig. 3 , A, D, G panels), for each time point, respectively. The component loads of the variables -the composition of the PCA, in terms of the original MRI-, was stable across the three time points, with AD and MD mainly composing PCA 1 , RD and FA PCA 2 , and T 2 * PCA 3 , (Table 1 ). Subsequently, the effects of the predictor variables -diet, sex or treatment- were tested on the three corresponding PCAs. Table 1 Component loads of the variables at times 1, 2 and 3. PCA 1 basal PCA 2 basal PCA 3 basal PCA 1 time 1 PCA 2 time1 PCA 3 time1 PCA 1 time 2 PCA 2 time 2 PCA 3 time 2 MD -0.58 0.29 0.03 -0.67 -0.08 0.01 0.61 0.29 0.10 AD -0.61 -0.16 -0.04 -0.55 0.42 0.00 0.63 -0.21 0.00 RD -0.24 0.77 0.12 -0.49 0.51 0.02 0.26 0.69 0.15 FA -0.48 -0.52 -0.08 -0.10 0.74 0.04 0.39 -0.60 -0.12 T 2 * 0.02 0.15 -0.99 -0.02 0.03 -0.99 -0.06 -0.21 0.98 At basal time -before any administration- we found that the type of diet consumed affected significantly PCA 3 (Fig. 3 C), in a region-dependent manner (significant region:diet , p < 0.05, lme followed by Anova). Particularly, HFD mice showed lower PCA 3 than CTRL mice, in all regions except in the hippocampus, with post-hoc corrected comparisons not reaching statistical relevance. PCA 3 was composed mainly by T 2 * (Fig. 3 B), and the specific analysis of T 2 * as a function of diet and region revealed similar results ( region:diet , p = 0.05), with lower values on the hippocampus of HFD mice, and slightly higher in the rest of the regions, as compared to CTRL, and without reaching robust statistical evidences sub-regionally (Fig. 4 A and B ). In the Glc short measurements, the lme approach revealed a significant sex-dependent diet effect ( diet:sex , p < 0.05) on PCA 2 (Fig. 3 F). Specifically, male HFD animals had higher PCA 2 , as compared to CTRL mice (Fig. 3 F left). PCA 2 was composed by FA and RD (Fig. 3 E), and higher FA and lower RD could be seen in the HFD mice, respectively (Fig. 4 C and D ), with p < 0.05 only for FA (anova test of FA Vs diet, sex and diet:sex ). Additionally, we found that the type of treatment affected PCA 3 , on a region and diet -dependent manner (interaction of region:diet:treatment , p < 0.05), and depending on sexes (treatment:sex, p < 0.05). Particularly, HFD animals with TGN treatment tended to higher PCA 3 , as compared to HFD saline-administered mice, in all regions except on the hippocampus, with adjusted post-hoc contrasts not reaching statistical significance. TGN-treated male animals showed significantly higher PCA 3 values than the saline males (Fig. 3 F right). Since PCA 3 was mainly composed by T 2 *, this was translated into lower values of T 2 * in TGN-treated male animals, as compared to controls (Fig. 4 E) (p < 0.05 sex:treatment , Anova tests, p adj = 0.06 for post-hoc in males) . In the Glc long measurements (30 min after Glc administration) the type of diet was still affecting significantly PCA 2 values (p < 0.05) (Fig. 3 ) , with HFD mice having significantly lower PCA 2 values than CTRL animals. RD and FA were positively and negatively correlated with PCA 2 , respectively (Fig. 3 H), and lower PCA 2 was translated into a lower RD and higher FA on HFD mice ( Fig. 4 F ) , but the independent tests of FA Vs diet, and RD Vs diet did not reach statistical significance. The type of treatment received had a significant effect on PCA 3 , (p < 0.01), with TGN + Glc administered mice presenting significantly lower PCA 3 values, than Saline + Glc mice (Fig. 3 I). Our results also revealed a significant region -dependent diet effect ( region*diet interaction, p < 0.05) indicating that the HFD group had higher or equal PCA 3 values than CTRL mice in all regions but the hippocampus (significantly only in cortex). Since T 2 * and PCA 3 values were positively correlated (Fig. 3 H), lower PCA 3 with TGN indicated lower T 2 ( Fig. 4 G-H ) ( p < 0.05 t-test T 2 * Vs treatment ). 3. T2* and DTI follow up TGN-20 administration to HFD or CTRL mice To quantify the animal-specific changes of the MRI parameters, before and after the administration of either “ saline + glc” or “ TGN + glc” , we built a global MFA that included MRI variables from the three longitudinal time-point measurements. Specifically, we included the subtraction, for each MRI parameter, of the “ Glc short minus Basal” values (Glc short -Basal), “ Glc long minus Glc short ” (Glc long - Glc short ) and “ Glc long minus Basal ” (Glc long - Basal) and performed the MFA analysis on them. Results yielded 4 relevant principal components, and, subsequently, the effects of diet , region , sex or treatment were assessed by a lme approach followed by Anova. After applying such procedure, we observed a common regional pattern, with the hippocampus showing consistently a different tendency, as compared to rest of the regions (data not shown). We thus decided to consider the hippocampus as an independent region and conducted a specific MFA for the hippocampus, and a MFA from the rest of the regions. 3.1- Hippocampal evolution The MFA on the hippocampus resulted in 4 PCAs explaining ≥ 80% of the variance, thus achieving a dimensionality reduction from 12 initial MRI parameters to 4 main components. PCA 1 (30% explained variance) was mostly composed of variables from the time regimes Glc long -Basal and Glc short -Basal, and a PCA 2 (20%) mainly composed by the time regime Glc long -Glc short . PCA 3 contained a mixture of three-time regimes, and PCA 4 was formed by Glc short -Basal and Glc long -Basal. The component loads of the experimental MRI variables on the global PCA, are summarized in Table 2 . Table 2 Hippocampal MFA component loads of the experimental variables. PCA 1 PCA 2 PCA 3 PCA 4 MD Glcshort−B 0.67 -0.32 0.52 0.39 AD Glcshort−B 0.80 -0.40 0.36 -0.10 RD Glcshort−B 0.09 -0.01 0.45 0.88 FA Glcshort−B 0.70 -0.39 0.07 -0.54 T 2 * Glcshort−B -0.14 0.34 0.26 -0.19 MD Glclong−B 0.89 0.37 0.10 0.23 AD Glclong−B 0.77 0.39 0.46 -0.13 RD Glclong−B 0.71 0.22 -0.33 0.56 FA Glclong−B 0.29 0.14 0.75 -0.55 T 2 * Glclong−B -0.43 0.34 0.12 0.01 MD Glclong−Glcshort 0.44 0.79 -0.41 -0.10 AD Glclong−Glcshort -0.03 0.96 0.12 -0.03 RD Glclong−Glcshort 0.65 0.23 -0.69 -0.12 FA Glclong−Glcshort -0.43 0.54 0.67 0.01 T 2 * Glclong−Glcshort -0.43 0.14 -0.08 0.17 On PCA 1, the most contributing variables were (MD 16%, AD 12%, RD 10%) Glclong−B , (AD 13%, FA 10%, MD 8%) Glcshort−B and (RD 10.5%, FA 10%, MD 8%) Glclong−Glcshort , all with a positive contribution. The type of treatment received had a significant effect on PCA 1 (p < 0.05), with the TGN + Glc group showing higher (and positive) PCA 1 values, as compared to Saline + Glc animals, which had negative PCA 1 values (data not shown). Since PCA 1 was composed by changes of MD, AD, RD and FA, positively correlated, negative PCA 1 in the Saline + Glc animals can be translated into decreasing MD, AD, RD and FA from basal to post times, and from basal to pre, but augmenting TGN + Glc group. On PCA 2 , the most contributing variables were (AD 34%, MD 22.5% and FA 10,5%), all from the Glc long -Glc short regime. A significant diet -dependent treatment effect (interaction of diet:treatment , p < 0.001) was revealed. On HFD mice, TGN treatment group showed significantly lower (and negative) PCA 2 values than saline animals, which had positive PCA 2 values (Fig. 4 A). Since PCA 2 represented changes of AD, MD and FA, positively correlated, a positive PCA 2 in the Saline + Glc mice can be translated into augmenting AD, MD and FA values in the time regime from Glc short to Glc long (Glc long – Glc short > 0) but decreasing (Glc long – Glc short < 0) on the TGN + Glc group (Fig. 4 B) (diet:treatment p adj < 0.05 for AD and FA, and subsequent post-hoc padj < 0.05 for treatment effects only on HFD mice). No relevant effects were found on the effects of diet or treatment on PCA 3 and PCA 4 on the hippocampus. 3.2- Cortex, thalamus and hypothalamus evolution MFA on the hippocampus resulted in 4 PCA explaining the target variance with a PCA 1 (31% explained variance) formed by Glc long -Basal and Glc short -Basal, a PCA 2 (23%) mainly composed of variables from the time regime Glc long -Glc short and Glc short -Basal. PCA 3 (18%) was mostly composed of the time regimes Glc long -Basal and Glc short -Basal and PCA 4 (14%) formed by Glc long -Glc short and Glc long -Basal. The specific component loads of the MRI variables on the rest of the regions are summarized in Table 3 . Table 3 Rest of the regions MFA component loads of the experimental variables. PCA 1 PCA 2 PCA 3 PCA 4 MD Glcshort−B 0.29 -0.74 0.52 0.24 AD Glcshort−B 0.73 -0.66 -0.00 0.00 RD Glcshort−B -0.41 -0.31 0.75 0.34 FA Glcshort−B 0.82 -0.35 -0.38 -0.19 T 2 * Glcshort−B 0.08 -0.27 0.17 0.16 MD Glclong−B 0.68 0.16 0.69 0.01 AD Glclong−B 0.92 0.13 0.14 0.32 RD Glclong−B 0.01 0.10 0.90 -0.34 FA Glclong−B 0.78 0.06 -0.35 0.47 T 2 * Glclong−B 0.00 -0.17 0.36 0.02 MD Glclong−Glcshort 0.10 0.90 0.18 -0.21 AD Glclong−Glcshort 0.17 0.90 0.15 0.35 RD Glclong−Glcshort 0.47 0.45 0.13 -0.73 FA Glclong−Glcshort -0.14 0.54 0.07 0.81 T 2 * Glclong−Glcshort -0.07 0.04 0.29 -0.12 On PCA 3 , the most contributing variables were RD and MD (Glc long -Basal 30% and 17.5%, respectively) and RD and MD (Glc short -Basal 20,5% and 10%). Our lme showed a significant sex and region -dependent diet effect (interaction of region*diet*sex , p < 0.05), where the HFD group tended to have lower or equal PCA 3 values. No relevant effects were found on the effects of diet or treatment on PCA 1 , PCA 2 and PCA 4 on the cortex, thalamus and hypothalamus. DISCUSSION In this work, we investigated the role of AQP4 during glucose uptake in obese and non-obese mice by MRI in four brain regions. Specifically, we performed three sequential DTI and T 2 *WI acquisitions to follow the effects of TGN (or its vehicle, saline) administration prior to the glucose insult. Next, having obtained a series of MRI parameters per region and time point, we applied PCA and MFA approaches to the data to reduce dimensionality, and then used linear mixed models on the corresponding main components to infer the potential effects of type of treatment, diet consumed, sex or brain regions, on the MRI variables. Effects of TGN on microvascular blood flow and swelling Our results are consistent with a change in the microvasculature blood flow induced by glucose that is inhibited by TGN, as quantified by lower T 2 * values in TGN mice, as compared to saline animals, (Fig. 3 F, I, 4 E, G, H). Such effects were more pronounced in males, during the Glc short measures, but the sexual dependance disappears in the Glc long . In parallel, our data agrees well with a glucose-induced cell swelling mechanism in the hippocampus of HFD mice that is inhibited by TGN, as detected by higher FA, AD and MD changes from Glc short to Glc long in saline animals, as compared to TGN-treated mice (Fig. 5 ). Additionally, our work reports specific diet effects on brain cellularity, with HFD mice exhibiting lower RD and higher FA than controls, more remarkably in males (Fig. 3 F, 4 C), which is consistent with the presence of gliosis 37 . Notably, this diet effect on diffusivity was still remarkable after the long effect of glucose (Fig. 3 I, 4 F). Considering both the T2* and DTI measures, results are consistent with a TGN reduction of the microvascular blood flow and cerebral activity that is detectable by MRI, highlighting the role of AQP4 during glucose uptake. Indeed, previous studies showed that glucose administration leads to increased cerebral activity and microvasculature blood flow 35 , and that such changes can be detected by MRI as increased markers of cellular swelling and microvascular blood flow, probably through glucose-sensing neurons 38 . Glucose uptake in the CNS is specifically associated with the entry of water into the cells via AQP4, leading to cell swelling and a decrease in extracellular space 15,39 . When glial swelling occurs, the cell shape becomes elongated 40 , resulting in more anisotropic diffusion of the water molecules, and MRI has been used to detect gliosis processes during obesity 41 . Particularly, using DTI, changes in cell shape can be characterized by higher FA 42 , higher AD and MD, and lower RD 43 . TGN-020 is known to inhibit AQP4 channels, preventing the passage of water through them 44 . Such inhibition may cause an alteration in the flow of ions (K + , Ca 2+ , etc), altering astrocyte function and thus impairing the glucose-induced swelling. Our results revealing in vivo the relationship between AQP4 inhibition and impaired changes are consistent with previous studies, which reported ex vivo how the activity-dependent glial swelling was impaired on AQP4 knockout mice 45 , or in vitro how the deletion of AQP4 reduces astrocyte swelling during an oxygen-glucose deprivation challenge 46 . Diet effects Our work reveals that diet affects the T 2 * and DTI values of the mouse brain. For instance, we report region-specific altered T 2 * values on obese mice at basal measurements, as compared to non-obese, suggesting that HFD induced lower values specifically in the hippocampus, on agreement with previous results in humans. Indeed, obesity is associated with both generalized and region-specific reductions in cerebral blood flow, with the hippocampus being a particularly vulnerable area. Previous neuroimaging studies demonstrated that higher body mass index correlated with progressively reduced blood flow across nearly all brain regions, indicating a global decline in perfusion 47 . The hippocampus—critical for memory and linked to Alzheimer’s disease (AD) risk shows pronounced sensitivity, with obesity-related hypoperfusion comparable to aging effects 48 . The experiments here reported also highlight that the swelling-impairment effects of TGN depend on the diet consumed, as found in the results of the MFA, again in the hippocampus (Fig. 5 ). On the Glc short to Glc long period -the interval when glucose is expected to act on the brain- the TGN group had lower FA, MD and AD changes, as compared to the saline animals, only on the HFD subgroup. These results are consistent with the inhibition of a directionally dependent, glucose-induced swelling process that would occur in saline, and the fact that this change was reported only on the HFD group may indicate that AQP4 is more effectively inhibited in this group. Moreover, is consistent with previous experiments that have reported increased AQP4 expression during obesity 12,49 . Sexual dimorphism In this study, we reported some sexual differences, including an increased susceptibility to diet, regarding the effects on brain DTI (Fig. 3 F left ), and to treatment changes on T 2 * (Fig. 3 F right ), in male mice. In animal models of diet-induced obesity, sexual differences are common, with males developing the obese phenotype typically faster and to a higher degree than females, who seem to be more protected against its development 50 . This is reflected, for example, in the BW of the animals, in which diet induces larger increases in the case of males, as reported here. Likewise, brain changes underlying the obese phenotype are sexual-dependent, with male mice being more prone to depict a neuroinflammatory profile 51 . In this sense, our results are consistent with an increased male-specific presence of astrogliosis and microgliosis during HFD. At the same time, our results suggest that TGN inhibition may be more effective under circumstances of increased AQP4 abundance. PCA and MFA In this work, the use of PCA and MFA allowed a reduction in dimensionality, obtaining new principal components that were independent from each other. This implies that the statistical tests performed on the corresponding components were independent 52 , and allowed us to pinpoint those combinations of MRI variables that were affected by predictor variables. Briefly, for example, from the four DTI-derived variables (MD, AD, RD and FA), which are highly correlated between them, the PCA method retained only two relevant components, with MD and AD showing the highest scores on PCA 1 , and RD and FA dominating PCA 2 . The statistical tests on PCAs revealed that only PCA 2 resulted remarkably altered by the explanatory variables, particularly by diet and sex, and we could not find any treatment effects. This led us to explore in detail the significance of the variables sex and diet on FA and RD, and no further tests on MD or AD were performed, and the variable treatment was omitted (Fig. 4 ), thus reducing the number of tests performed and the probability of committing type I errors. Interestingly, in some cases, lower p-values were found on the tests on the created components (PCA 2 ) than on the tests of the MRI variables composing them, suggesting the convenience of the method to find the variable combination that best reveals the underlying effects. In our study, PCA 2 captured the variance related to smaller diffusivities (RD < AD), and to the degree of diffusion anisotropy, and thus to the degree on anisotropy of cells that shape such movements. The consistent negative correlation between FA and RD suggests that RD alterations may be good indicators of potential changes on cellular shape, in agreement with previous literature 53 . T 2 *, on the other hand, was in the three time-points almost exclusively dominating PCA 3 , showing lower correlations with the diffusion behavior of water molecules and potentially reflecting the vascular components. MFA, on the other hand, gave us information on how the time periods, and the variables within, were correlated between them. By definition, MFA builds first separated PCA for each time regime, and after corresponding normalizations, generates a global PCA in which the initial variables are expressed as function of such global PCAs 54 (Table 2 and Table 3 ). In our results, both in the MFA hippocampus , and in rest of the regions analysis, PCA 1 was mainly dominated by variables from the “Glc short - basal” and “Glc long -basal” regimes, with MD and AD showing the higher loadings. On the contrary, the MRI variables with the higher PCA 2 loadings corresponded to the “Glc long -Glc short” interval, with a clear domination of AD, followed by MD. Considering the experimental timeline of TGN/saline and glucose administrations followed here, PCA 1 seems to be expressing potential differences on global diffusivity caused by the TGN administration itself, while PCA 2 may be more related to the variance induced by glucose uptake and metabolism, and its effects on the highest diffusion directions. LIMITATIONS In this work, we used TGN-020 as an inhibitor of AQP4 function, based on evidence of several previous studies 36,55–57 . However, some research has reported the inability of this compound to inhibit AQP4 function 58 , highlighting the need for a clearer understanding of the TGN-020 potential targets before interpreting the results of the experiments in vivo. In our study, based on MRI measurements, TGN administration appeared to effectively inhibit AQP4 channels, since it prevented the glucose administration-induced MRI changes observed in the absence of the inhibitor, the cellular swelling. While our results correlate well with previous studies reporting such swelling inhibition 45,46 future experiments should investigate the mechanistic insights of the MRI-detected processes. There are several limitations of the PCA or MFA approaches carried out with our data. In both cases, the underlying correlation of the initial MRI variables was not that strong (Kaiser-Meyer-Olkin criteria about 0.5), which effectively meant reducing only from 5 to 3 the number of components. On the MFA method, on the other hand, since some animals did not have information on the three time points, due to experimental problems, some data was inevitably lost. Regarding the DTI variables, not all animals were acquired with the same number of directions, as specified in the methods section. Indeed, those animals acquired during the first batch included acquisitions of 6 directions only. This group showed lower data quality, which could add a potential bias in interpretation. It should be noted, however, that the type of study (“batch 1” or “batch 2”) was initially included as a predictor variable in the lme approaches, to test is belonging to either of the groups was affecting significantly the PCA variables, and no relevant effects were reported (data not shown). While our findings offer valuable insights into AQP4 inhibition and glucose metabolism in the context of obesity, it is important to recognize the limitations of using a mouse model to extrapolate results to humans. One significant limitation is the difference in basal metabolic rates between mice and humans, which may affect generalizations related to energy expenditure and glucose regulation. Additionally, while the DIO models replicate key features of human obesity, including glucose intolerance and fat accumulation, the precise metabolic pathways and hormonal responses, particularly those related to glucose sensing and insulin signaling, may differ between the two species. Furthermore, the inflammatory response in mice is often more acute, and the distribution of fat (subcutaneous vs. visceral) may vary, influencing how obesity-related inflammation impacts glucose metabolism and brain function. Although the glucose loading model used in this study helps to mimic postprandial glycemic responses, its exact relevance to human physiology warrants careful consideration. In humans, glucose metabolism and uptake are regulated by a more complex set of hormonal controls. Isoflurane anesthesia can significantly influence cerebral perfusion by inducing vasodilation and altering cerebral blood flow, as reported in previous studies. In our study, we maintained a consistent anesthetic protocol for all animals, ensuring that any observed differences between experimental groups were not confounded by variations in anesthesia depth. While isoflurane anesthesia may influence the absolute values of perfusion-related parameters, our analysis focused on relative differences between conditions (e.g., glucose administration and AQP4 inhibition), which remain interpretable within this controlled setting. Although the findings should be interpreted within the context of an anesthetized model, the use of isoflurane was necessary to ensure stability and immobility during MRI acquisition, which is critical for obtaining high-quality, artifact-free images. While awake imaging could avoid the confounding effects of anesthesia, it introduces other challenges, such as motion artifacts and stress-related physiological changes, which could also confound the interpretation of MRI-derived metrics.Finally, another limitation is that while our study demonstrated the effects of TGN-020 in inhibiting AQP4 and altering brain microvascular blood flow in mice, these results may not directly translate to human neurovascular function. The expression and regulation of AQP4 in human brain tissues might differ, affecting the applicability of our results to human conditions. CONCLUSIONS We can conclude that TGN-020 acts as an AQP4 inhibitor after a glucose insult that results in changes in the neurovascular unit detectable by MRI in vivo, including lower blood flow in several brain regions, as detected by T 2 *WI, and a decrease in glucose-induced cell swelling process detectable by DTI in the hippocampus of obese mice. In addition, we observed in obese mice MRI indicators of altered cortical microvasculature blood flow, potentially related to vascular inflammation, and DTI markers consistent with astro- or microgliosis processes, predominantly on males. MATERIALS AND METHODS 1. Experimental design The mice used in this study were bred and housed in our institutional animal facility at Institute for Biomedical Research Sols-Morreale (Reg. No. Es280790000288). n = 66 adult C57BL6/J mice (32 females) were kept in a room of the animal facility with controlled humidity (47%) and temperature (21–23°C), a 12 h light/day cycle and fed with a standard laboratory food (SAFE DIETS). At 6 weeks of age, 33 animals (18 females) were switched to an HFD (60% fat, 20% carbohydrates, 20% proteins from D12492, Research Diets, New Brunswick, NJ), to induce obesity, and the rest continued with standard diet. After 18 weeks of diet diversification, mice were subjected to MRI under anesthesia provided through a nose mask (isoflurane 1-1.5% in O 2 , 1L/min) during the whole study, to assess the response to a glucose administration stimulus (i.p., 200 µL/25g, 2.08M on a saline solution, 16.65 mmol/kg) preceded by a saline i.p. (100 µL/25g) or TGN-020 (100 µL/25g a 0.24M, 0.96 mmol/kg administration), 20 minutes before the glucose insult 59 . Depending on the diet and the type of administration received, animals were thus assigned to four groups of analysis, i) fed with a standard laboratory control diet and injected first with saline and glc (7 males and 7 females), ii) fed with CTRL and administered with TGN-020 and glc (11 males and 7 females), iii) group fed with HFD and administered with saline and glc (6 males and 7 females), and iv) group fed with HFD and administered with TGN-020 and glc (8 males and 7 females) (Fig. 2 ). Once the MRI experiments finished, animals were euthanized by rapid decapitation, still under the effects of anesthesia administered during the imaging sessions. 2. Compound preparation TGN-020 is water-insoluble; therefore, a sodium salt derivative of the compound (TGN-Na) was prepared to facilitate its subsequent injection into animals using a saline solution. First, 41.20 mg of TGN-020 was placed in a round-bottom flask with 4 ml of Millipore water. Next, 426.24 µL of NaOH (1M) were slowly added, leaving it stirring for 30 minutes at room temperature. Finally, the solution obtained was filtered and then lyophilized to generate the sodium salt (TGN-020-Na) that was injected into the animals (throughout the text, this sodium salt will be referred to as TGN-020). 3. MRI acquisitions Mice were subjected to MRI under anesthesia (isoflurane) using a superconducting magnet with 16 cm diameter and a gradient insert of 360 mT/m (Biospec® 7T, Bruker Biospect, Ettlingen, Germany), in two separated animal cohorts. First, DTI studies (6 directions, b-values = 400 µm 2 /s & 1800 µm 2 /s, field of view = 23 x 23 mm 2 , slice thickness = 1.25 mm, 5 slices) and T 2 *WI to generate T 2 * maps (TR = 300 ms, flip angle = 30, 10 echoes, first TE = 2 ms, inter echo time: 4 ms, 8 averages, and the same geometrical parameters as DTI) were acquired in an initial group of animals (n = 35, n = 19 with DIO, n = 15 females). Subsequently, an additional cohort was investigated (n = 31, n = 14 with DIO, n = 17 females) with DTI (15 directions, b- values = 400 µm 2 /s & 1800 µm 2 /s, field of view = 23mm, slice thickness = 1 mm, 5 slices) and T2*WI with the same conditions of the first batch. In both cases, brain slices were positioned to contain the hypothalamus, thalamus and cortex on the center acquisition slice, while the frontal hippocampus remained located in a contiguous slice. Regions were identified with the help of anatomical atlas. 60 For each mouse, three DTI and T 2 *WI blocks were acquired, at basal (t = t0), 20 min after TGN (or saline) administration and immediately after the glucose insult (t = t1), and 30 min after the glucose administration (t = t2) (Fig. 2 ). 4. Image processing DTI and T 2 *WI of each mouse were processed using Resomapper, a home-made software based on Dipy 61 . Pre-processing algorithms included a noise reduction filter (Patch2self) for DTI and the adaptive soft coefficient matching filter for T 2 *WI, both used to improve image quality and selected based on previous literature 62 . Processing of the DTI images lead to the obtention of MD, AD, RD and FA parametric maps, and by processing the T 2 *WI, T 2 * maps were obtained. Next, ImageJ software (U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/ ) was used to manually select the regions of interest (ROIs), including: the hypothalamus (120 voxels), thalamus (140 voxels), cortex (144 voxels) and the hippocampus (110 voxels)(Fig. 2 ). ROI selection was performed blindly, and corresponding pixel coordinates of each subregion were saved and stored as . txt files. 5. Data filtering and preparation Matlab software (R2010b, MathWorks Inc., Natick, MA) was used to overlay all ROIs on all images of each mouse. Briefly, by automatically reading the coordinates from the Image-J derived . txt files, the information was inferred to use the same coordinates on all diffusion and T 2 * maps. Datasheet documents were then generated to include the information on MD, AD, RD, FA and T 2 * values for each pixel. The information regarding the corresponding subregion, time-point of acquisition (t0, t1 or t2), animal identification, type of diet, sex and type of treatment of each pixel was added in separate columns, resulting in a unique datasheet containing all the MRI information at a pixel level. Using home-made R-scripts, data filtering and restructuring were performed. From the pixel datasheet of MRI generated by Matlab, pixel values were filtered to exclude the potential contamination of the CSF from ventricles, by removing those pixels with MD, AD and RD values > 1300 mm 2 /s or T 2 * values > 30 ms, as previously described 63 . Additionally, pixel outliers from each animal subregion (mean +/- 1.5 - interquartile range) were eliminated. 6. Principal Component Analysis (PCA) Aiming for dimensionality reduction, MRI variables were transformed via PCA (Fig. 2 ). PCA is a mathematical algorithm that reduces the dimensionality of the data while retaining most of the variance of the data. New variables are obtained, called principal components (which are a linear combination of the original variables), whose number is equal to or less than the initial number of variables, which explain most of the variance of the data and are independent of each other 52 . In this study, the number of retained PCA variables was chosen to explain at least ≥ 80% of the variance. Two main PCA approaches were performed, using the prcomp function of R ( https://www.rdocumentation.org/packages ), including the scaling and centering procedures to have unit variance and a shift to zero, respectively, before analysis. In the first assessment, the “ 1.HFD, sex and TGN-20 effects on T2* and DTI at separated time points ” , a PCA for each time (t0, t1 and t2) was performed independently. In the second approach, the “ 2. T 2 * and DTI follow up TGN-20 administration to HFD or CTRL mice ”, a multiple factor analysis (MFA) was calculated. MFA is an extension of PCA that is used to handle data where the same variables (or “data tables”) are measured on different time-sets of observations 64 . Briefly, the procedure is achieved in two steps. First, a PCA of each data-table (or time-observation) is performed. Secondly, all the data tables are concatenated and a generalized PCA is done. With this approach, it is possible to study the time regimes that contribute most to each component derived from the PCA and report the influence of the variables from each time. To do that, we generated the Gl short -basal, Gl long -Gl short and Gl long− basal groups of data by subtracting the corresponding values of MD, AD, RD, FA and T 2 * between the different time-points. MFA was performed using the mfa function of the FactoMiner package ( https://www.rdocumentation.org/packages/FactoMineR/versions/2.9/topics/MFA ). Two additional MFA were performed, either in the region of the hippocampus alone, or using the three remaining regions. In the MFA analysis, those animals that did not show valid DTI images for all time points, could not be included, and the corresponding groups were i) CTRL fed and “saline + glc” (5 males and 5 females), ii) CTRL and “TGN-020 + glc” (7 males and 5 females), iii) HFD and “saline + glc” (5 males and 5 females), and iv) HFD and “TGN-020 + glc” (3 males and 3 females). In all cases, once the relevant PCA or MFA and predictors were identified, and to better understand the biological basis of the findings, we further investigated how the PCA-composing MRI variables were related to such predictors, by graphical representations and exploratory statistical tests (corrected for multiple comparisons). 7. Statistical tests Differences in BW were assessed by anova tests, with diet, sex and treatment as main effects, and their corresponding interactions, using the Anova function of the car package in R. The effects of diet, type of administration (“treatment”), brain region and sex, on the relevant PCA variables was perfomed by linear mixed-effect (lme) fitting 63 , using the lme function of the nlme package 65 , and mouse as a random intercept. An autoregressive correlation between regions was established (AR1 function of the nlme), in which the regions that are closer present more correlation. These lme models included fixed effects such as region , diet , sex and treatment and mouse as a random term. Interactions between different variables were added to explore possible joint effects, including double interactions ( Diet:Region, Diet:Sex, Treatment:Region, Treatment:Diet, Treatment:Sex and Sex:Region ), triple interactions ( Diet:Region:Sex, Diet:Region:Treatment, Region:Sex:Treatment and Sex:Treatment:Diet ) and the quadruple interaction ( Sex:Treatment:Diet:Region ). Next, Anova tests were performed to assess the statistical significance of the main effects and interactions on each PCA. For significant interactions, post-hoc contrasts were performed using the emmeans function ( https://github.com/rvlenth/emmeans ), and corresponding p-values adjusted for multiple comparisons with the Bonferroni method. Declarations ACKNOWLEDGEMENTS We are deeply indebted to Mrs. María José Guillén (CSIC) for excellent technical assistance with animal handling, to María Rodríguez and Mrs. Teresa Navarro (CSIC) and the IIBM Sebastian Cerdán NMR Facility for expert support during MRI acquisitions and granting rapid access to the MR instrumentation. Funding. This work was supported by grants PID2021-122528OB-I00 to PL-L, grant PID2021-126888OA-I00 to B.L., scholarship PRE2022-105662 to A.F. and scholarship PIPF-2022/SAL-GL-25871 to R.G-A. CONTRIBUTIONS Conceptualization, P.L.-L. and B.L.; Methodology, N.A.-R., P.L.-L. and B.L.; Investigation, P.T.-G., R.G.-A. and N.A.-R..; Software, A.F., R.G.-A. and B.L., Formal Analysis, P.T.-G, A.F., and B.L., Writing – Original Draft, P.A.T.-G., and B.L.; Writing – Review & Editing, P.L.-L. and B.L.; Visualization, P.A.T.-G. and B.L., Funding Acquisition, P.L.-L. and B.L., Project Administration, P.L.-L. and B.L., Supervision, N.A-R., P.L-L and B.L. The authors declare no competing interests Ethic statements All animal handling protocols were performed by specialized personnel at our institute’s animal facility (Reg. No. ES280790000188) and approved by the ethical committee of the Institute of Biomedical Research Sols-Morreale, CSIC and the Community of Madrid, complying with national (R.D. 53/2013) and European Community guidelines (2010/63/UE). Ethics approval number PROEX 288/42 (Comunidad de Madrid, dirección general de agricultura, ganadería y alimentación). All methods were carried out in accordance with relevant guidelines and regulations. All methods are reported in accordance with ARRIVE guidelines (https://arriveguidelines.org). Data availability statement The data presented in this study is available on request from the corresponding author due to the need for a formal data sharing agreement. References Mergenthaler, P., Lindauer, U., Dienel, G. A. & Meisel, A. Sugar for the brain: the role of glucose in physiological and pathological brain function. Trends Neurosci 36 , 587–597 (2013). Takahashi, S. Metabolic Contribution and Cerebral Blood Flow Regulation by Astrocytes in the Neurovascular Unit. Cells 11 , (2022). Hwang, J. J. et al. Blunted rise in brain glucose levels during hyperglycemia in adults with obesity and T2DM. JCI Insight 2 , (2017). Huber, G. et al. Telmisartan prevents high-fat diet-induced neurovascular impairments and reduces anxiety-like behavior. J Cereb Blood Flow Metab 41 , 2356–2369 (2021). de Paula, G. C. et al. Hippocampal Function Is Impaired by a Short-Term High-Fat Diet in Mice: Increased Blood-Brain Barrier Permeability and Neuroinflammation as Triggering Events. Front Neurosci 15 , 734158 (2021). Sa, M., Park, M. G. & Lee, C. J. Role of Hypothalamic Reactive Astrocytes in Diet-Induced Obesity. Mol Cells 45 , 65–75 (2022). Valdearcos, M. et al. Microglial Inflammatory Signaling Orchestrates the Hypothalamic Immune Response to Dietary Excess and Mediates Obesity Susceptibility. Cell Metab 27 , 1356 (2018). Koepsell, H. Glucose transporters in brain in health and disease. Pflugers Arch 472 , 1299–1343 (2020). Nagelhus, E. A. & Ottersen, O. P. Physiological roles of aquaporin-4 in brain. Physiol Rev 93 , 1543–1562 (2013). Sun, C. et al. Acutely Inhibiting AQP4 With TGN-020 Improves Functional Outcome by Attenuating Edema and Peri-Infarct Astrogliosis After Cerebral Ischemia. Front Immunol 13 , 870029 (2022). Mader, S. & Brimberg, L. Aquaporin-4 Water Channel in the Brain and Its Implication for Health and Disease. Cells 8 , (2019). Micioni Di Bonaventura, M. V. et al. Brain alterations in high fat diet induced obesity: effects of tart cherry seeds and juice. Nutrients 12 , (2020). Tomassoni, D. et al. Obesity and Age-Related Changes in the Brain of the Zucker Lepr (fa/fa) Rats. Nutrients 12 , (2020). Fu, J.-T. et al. Exploring the reduction in aquaporin-4 and increased expression of ciliary neurotrophic factor with the frontal–striatal gliosis induced by chronic high-fat dietary stress. Journal of Neurochemistry n/a ,. Hertz, L., Peng, L. & Dienel, G. A. Energy metabolism in astrocytes: high rate of oxidative metabolism and spatiotemporal dependence on glycolysis/glycogenolysis. J Cereb Blood Flow Metab 27 , 219–249 (2007). Bak, L. K., Schousboe, A. & Waagepetersen, H. S. The glutamate/GABA-glutamine cycle: aspects of transport, neurotransmitter homeostasis and ammonia transfer. J Neurochem 98 , 641–653 (2006). Reinhard, L., Tidow, H., Clausen, M. J. & Nissen, P. Na(+),K (+)-ATPase as a docking station: protein-protein complexes of the Na(+),K (+)-ATPase. Cell Mol Life Sci 70 , 205–222 (2013). Burdakov, D., Luckman, S. M. & Verkhratsky, A. Glucose-sensing neurons of the hypothalamus. Philos Trans R Soc Lond B Biol Sci 360 , 2227–2235 (2005). Tu, L., Fukuda, M., Tong, Q. & Xu, Y. The ventromedial hypothalamic nucleus: watchdog of whole-body glucose homeostasis. Cell & Bioscience 12 , 71 (2022). Attwell, D. et al. Glial and neuronal control of brain blood flow. Nature 468 , 232–243 (2010). Koekkoek, L. L., Mul, J. D. & la Fleur, S. E. Glucose-Sensing in the Reward System. Frontiers in Neuroscience 11 , (2017). Yousaf, T., Dervenoulas, G. & Politis, M. Advances in MRI Methodology. Int Rev Neurobiol 141 , 31–76 (2018). Živanović, M. et al. The role of magnetic resonance imaging in the diagnosis and prognosis of dementia. Biomol Biomed 23 , 209–224 (2023). Le Bihan, D. et al. Diffusion tensor imaging: concepts and applications. J Magn Reson Imaging 13 , 534–546 (2001). Lope-Piedrafita, S. Chapter 7: Diffusion Tensor Imaging (DTI). in Preclinical MRI: Methods and Protocols 103–116 (2018). Lizarbe, B., Campillo, B., Guadilla, I., López-Larrubia, P. & Cerdán, S. Magnetic resonance assessment of the cerebral alterations associated with obesity development. J Cereb Blood Flow Metab 40 , 2135–2151 (2020). Lizarbe, B. et al. Systemic Glucose Administration Alters Water Diffusion and Microvascular Blood Flow in Mouse Hypothalamic Nuclei - An fMRI Study. Front Neurosci 13 , 921 (2019). Lizarbe, B. et al. Imaging hypothalamic activity using diffusion weighted magnetic resonance imaging in the mouse and human brain. Neuroimage 64 , 448–457 (2013). Nunes, D., Gil, R. & Shemesh, N. A rapid-onset diffusion functional MRI signal reflects neuromorphological coupling dynamics. NeuroImage 231 , 117862 (2021). Roth, B. J. Can MRI Be Used as a Sensor to Record Neural Activity? Sensors (Basel) 23 , (2023). Tang, M. Y., Chen, T. W., Zhang, X. M. & Huang, X. H. GRE T2∗-weighted MRI: principles and clinical applications. Biomed Res Int 2014 , 312142 (2014). Wang, J. et al. Perfusion functional MRI reveals cerebral blood flow pattern under psychological stress. Proceedings of the National Academy of Sciences 102 , 17804–17809 (2005). Ogawa, S., Lee, T. M., Kay, A. R. & Tank, D. W. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 87 , 9868–9872 (1990). Chavhan, G. B., Babyn, P. S., Thomas, B., Shroff, M. M. & Haacke, E. M. Principles, techniques, and applications of T2*-based MR imaging and its special applications. Radiographics 29 , 1433–1449 (2009). Lizarbe, B. et al. Systemic Glucose Administration Alters Water Diffusion and Microvascular Blood Flow in Mouse Hypothalamic Nuclei - An fMRI Study. Front Neurosci 13 , 921 (2019). Igarashi, H., Huber, V. J., Tsujita, M. & Nakada, T. Pretreatment with a novel aquaporin 4 inhibitor, TGN-020, significantly reduces ischemic cerebral edema. Neurol Sci 32 , 113–116 (2011). Taquet, M. et al. Extra-axonal restricted diffusion as an in-vivo marker of reactive microglia. Sci Rep 9 , 13874 (2019). Routh, V. H., McArdle, J. J., Sanders, N. M., Song, Z. & Wang, R. Glucose Sensing Neurons. in Handbook of Neurochemistry and Molecular Neurobiology: Sensory Neurochemistry (eds. Lajtha, A. & Johnson, D. A.) 205–228 (Springer US, New York, NY, 2007). doi:10.1007/978-0-387-30374-1_7. Verkman, A. S., Smith, A. J., Phuan, P.-W., Tradtrantip, L. & Anderson, M. O. The aquaporin-4 water channel as a potential drug target in neurological disorders. Expert Opin Ther Targets 21 , 1161–1170 (2017). Anderson, M. A., Ao, Y. & Sofroniew, M. V. Heterogeneity of reactive astrocytes. Neuroscience Letters 565 , 23–29 (2014). Berkseth, K. E. et al. Hypothalamic gliosis associated with high-fat diet feeding is reversible in mice: a combined immunohistochemical and magnetic resonance imaging study. Endocrinology 155 , 2858–2867 (2014). Metwalli, N. S. et al. Utility of axial and radial diffusivity from diffusion tensor MRI as markers of neurodegeneration in amyotrophic lateral sclerosis. Brain Res 1348 , 156–164 (2010). Messina, R. et al. White matter microstructure abnormalities in pediatric migraine patients. Cephalalgia 35 , 1278–1286 (2015). Oosuka, S. et al. Effects of an Aquaporin 4 Inhibitor, TGN-020, on Murine Diabetic Retina. Int J Mol Sci 21 , (2020). Kitaura, H. et al. Activity-dependent glial swelling is impaired in aquaporin-4 knockout mice. Neuroscience Research 64 , 208–212 (2009). Hermanova, Z. et al. The deletion of AQP4 and TRPV4 affects astrocyte swelling/volume recovery in response to ischemia-mimicking pathologies. Frontiers in Cellular Neuroscience 18 , (2024). Amen, D. G., Wu, J., George, N. & Newberg, A. Patterns of Regional Cerebral Blood Flow as a Function of Obesity in Adults. Journal of Alzheimer’s Disease 77 , 1331–1337 (2020). Tucsek, Z. et al. Aging exacerbates obesity-induced cerebromicrovascular rarefaction, neurovascular uncoupling, and cognitive decline in mice. J Gerontol A Biol Sci Med Sci 69 , 1339–1352 (2014). Tomassoni, D. et al. Obesity and Age-Related Changes in the Brain of the Zucker Lepr (fa/fa) Rats. Nutrients 12 , (2020). Pettersson, U. S., Waldén, T. B., Carlsson, P.-O., Jansson, L. & Phillipson, M. Female Mice are Protected against High-Fat Diet Induced Metabolic Syndrome and Increase the Regulatory T Cell Population in Adipose Tissue. PLOS ONE 7 , 1–10 (2012). Campillo, B. W., Galguera, D., Cerdan, S., López-Larrubia, P. & Lizarbe, B. Short-term high-fat diet alters the mouse brain magnetic resonance imaging parameters consistently with neuroinflammation on males and metabolic rearrangements on females. A pre-clinical study with an optimized selection of linear mixed-effects models. Front Neurosci 16 , 1025108 (2022). Ringnér, M. What is principal component analysis? Nat Biotechnol 26 , 303–304 (2008). Taquet, M. et al. Extra-axonal restricted diffusion as an in-vivo marker of reactive microglia. Scientific Reports 9 , 13874 (2019). Abdi, H. & Valentin, D. Multiple Factor Analysis (MFA). Encyclopedia of Measurement and Statistics (2007). Huber, V. J., Tsujita, M. & Nakada, T. Identification of Aquaporin 4 inhibitors using in vitro and in silico methods. Bioorganic & Medicinal Chemistry 17 , 411–417 (2009). Debacker, C., Djemai, B., Ciobanu, L., Tsurugizawa, T. & Le Bihan, D. Diffusion MRI reveals in vivo and non-invasively changes in astrocyte function induced by an aquaporin-4 inhibitor. PLoS One 15 , e0229702 (2020). Sun, C. et al. Acutely Inhibiting AQP4 With TGN-020 Improves Functional Outcome by Attenuating Edema and Peri-Infarct Astrogliosis After Cerebral Ischemia. Frontiers in Immunology 13 , (2022). Salman, M. M., Kitchen, P., Yool, A. J. & Bill, R. M. Recent breakthroughs and future directions in drugging aquaporins. Trends in Pharmacological Sciences 43 , 30–42 (2022). Debaker, C., Djemai, B., Ciobanu, L., Tsurugizawa, T. & Le Bihan, D. Diffusion MRI reveals in vivo and non-invasively changes in astrocyte function induced by an aquaporin-4 inhibitor. PLOS ONE 15 , e0229702 (2020). Allen Reference Atlas – Mouse Brain [brain atlas]. Garyfallidis, E. et al. Dipy, a library for the analysis of diffusion MRI data. Front Neuroinform 8 , 8 (2014). Schilling, K. G. et al. Denoising of diffusion MRI in the cervical spinal cord - effects of denoising strategy and acquisition on intra-cord contrast, signal modeling, and feature conspicuity. Neuroimage 266 , 119826 (2023). Campillo, B. W., Galguera, D., Cerdan, S., López-Larrubia, P. & Lizarbe, B. Short-term high-fat diet alters the mouse brain magnetic resonance imaging parameters consistently with neuroinflammation on males and metabolic rearrangements on females. A pre-clinical study with an optimized selection of linear mixed-effects models. Front Neurosci 16 , 1025108 (2022). Abdi, H., Williams, L. J. & Valentin, D. Multiple factor analysis: principal component analysis for multitable and multiblock data sets. WIREs Computational Stats 5 , 149–179 (2013). José Pinheiro, Douglas Bates and R Core Team. nlme: Linear and Nonlinear Mixed Effects Models. (2022). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 05 May, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Accepted 22 Apr, 2025 Reviews received at journal 22 Apr, 2025 Reviews received at journal 16 Apr, 2025 Reviewers agreed at journal 15 Apr, 2025 Reviews received at journal 13 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers agreed at journal 11 Apr, 2025 Reviewers invited by journal 11 Apr, 2025 Submission checks completed at journal 11 Apr, 2025 First submitted to journal 01 Apr, 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-5931441","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":442258087,"identity":"cfb00ca2-0ddc-4f32-adf5-542d5d7b6ae8","order_by":0,"name":"Pablo Tirado-García","email":"","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":false,"prefix":"","firstName":"Pablo","middleName":"","lastName":"Tirado-García","suffix":""},{"id":442258088,"identity":"9c3b3420-19c8-4016-b263-22ae2536845b","order_by":1,"name":"Adriana Ferreiro","email":"","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":false,"prefix":"","firstName":"Adriana","middleName":"","lastName":"Ferreiro","suffix":""},{"id":442258089,"identity":"badbb83d-5729-40ec-b12f-235ee3ad07aa","order_by":2,"name":"Raquel González-Alday","email":"","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":false,"prefix":"","firstName":"Raquel","middleName":"","lastName":"González-Alday","suffix":""},{"id":442258090,"identity":"24ab991a-d8c1-48a2-bc31-d66b1fa630f2","order_by":3,"name":"Nuria Arias-Ramos","email":"","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":false,"prefix":"","firstName":"Nuria","middleName":"","lastName":"Arias-Ramos","suffix":""},{"id":442258091,"identity":"d320a549-4202-4c3b-8138-c5e1257e9022","order_by":4,"name":"Blanca Lizarbe","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBACA3bGBiB1gIGfmbGNSC3MUC2SzcRrAVMHGAwOMLARp8Wcmbn5w889d+SMjzO3PWBss4tmYG9/gFeLZTNjg2HPs2fGZocZ2w0Y25JzG3jOGOB32GHGhgSeA4cTtx1mbJNgbGPObZDIIeAXoJaDfw4crt/cDNZSn9sg/xy/w4BaGpuBtiQAgw6k5TDQFgb8DgP6pZlZ5sAzwxkgvyScO57bxpODX4s5e/vjj28O3JHn7z/+7MGHsurcfvbj+B2GChKAmMjYGQWjYBSMglGADwAA53tIkNsThKgAAAAASUVORK5CYII=","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":true,"prefix":"","firstName":"Blanca","middleName":"","lastName":"Lizarbe","suffix":""},{"id":442258092,"identity":"e94b6c24-bfe2-490f-9754-64024b26a724","order_by":5,"name":"Pilar López-Larrubia","email":"","orcid":"","institution":"Instituto de Investigaciones Biomédicas Sols-Morreale, CSIC-UAM","correspondingAuthor":false,"prefix":"","firstName":"Pilar","middleName":"","lastName":"López-Larrubia","suffix":""}],"badges":[],"createdAt":"2025-01-30 15:53:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5931441/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5931441/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-99641-1","type":"published","date":"2025-05-05T15:57:19+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80631468,"identity":"f4b69a0c-f51f-470c-af18-7a6a1877ae7e","added_by":"auto","created_at":"2025-04-15 11:47:46","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1308409,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIon flow between glucose-sensing neurons, astrocytes and the extracellular space. \u003c/strong\u003eGlucose (Glc) enters neurons and astrocytes by GLUT transporters (purple) and is converted into pyruvate (Pyr). Pyr can be either converted into lactate (Lac) or enter the TCA cycle (Krebs cycle). Lactate is released by astrocytes and taken up by neurons through MCT transporters (cyan), where is reconverted into Pyr and can be used as a neuronal substrate for ATP and neurotransmitter production. In glucose-excited (GE) neurons (left panel, blue), high extracellular glucose levels and cellular uptake, changes\u003cstrong\u003e \u003c/strong\u003eATP-ADP ratio, which causes the closure of the K\u003csub\u003eATP\u003c/sub\u003e channels, resulting in plasma membrane depolarization and Ca\u003csup\u003e2+\u003c/sup\u003eentry through voltage-gated channels (green), thereby increasing neuronal activity and neurotransmitter\u003cstrong\u003e \u003c/strong\u003erelease\u003cstrong\u003e. \u003c/strong\u003eThe released glutamate enters astrocytes with Na\u003csup\u003e+ \u003c/sup\u003e(red). To re-establish the concentration, Na\u003csup\u003e+\u003c/sup\u003e leaves astrocytes via the Na\u003csup\u003e+\u003c/sup\u003e/K\u003csup\u003e+\u003c/sup\u003e-ATPase, with counter-transport of K\u003csup\u003e+\u003c/sup\u003e or via the Na\u003csup\u003e+\u003c/sup\u003e/K\u003csup\u003e+/\u003c/sup\u003e2Cl\u003csup\u003e-\u003c/sup\u003e co-transporter (orange). The entry of ions into astrocytes induces water to move intracellularly by AQP4 (blue).\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/a5892e6e22ee5e01d1f9bf0e.png"},{"id":80631481,"identity":"334e5ce1-983c-471d-ba0c-64b7493d51b9","added_by":"auto","created_at":"2025-04-15 11:47:46","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":9006285,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExperimental design and data analysis workflow\u003c/strong\u003e. \u003cstrong\u003eUpper panel\u003c/strong\u003e: Temporal distribution of MRI studies and compound administration. The first set of MRI studies (t0) was acquired before any experimental manipulation. Subsequently, TGN (100 μL/25g a 0.24M, 0.96 mmol/kg administration) or saline (100 μL/25g)\u003cem\u003e \u003c/em\u003ewere injected intraperitoneally (\u003cem\u003ei.p\u003c/em\u003e.). After 20 minutes, glucose was administered (\u003cem\u003ei.p\u003c/em\u003e., 200 μL/25g, 2.08M on a saline solution, 16.65 mmol/kg) and the second set of studies was performed (t1). 30 min later, the last set of acquisitions (t2) was acquired. \u003cstrong\u003eMiddle panel\u003c/strong\u003e: Representative parametric maps of mean diffusivity (MD), axial diffusivity (AD), radial diffusivity (RD), fractional anisotropy (FA) and T\u003csub\u003e2\u003c/sub\u003e*. ROIs investigated are superimposed, including the left and right hippocampus, or right and left cortex, thalamus and hypothalamus. \u003cstrong\u003eLower panel\u003c/strong\u003e: Example of representation of the PCA maps derived from the MRI parameters. After PCA transformation, each pixel and ROIs can be expressed in the new coordinates system.\u0026nbsp;\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/69763853d3b6feffdb717141.png"},{"id":80631489,"identity":"4f3711f5-8169-4b1f-abd1-f7594a0c865b","added_by":"auto","created_at":"2025-04-15 11:47:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6398410,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePCA and lme statistics. A:\u003c/strong\u003e Percentage of explained variances from each principal component (from PCA\u003csub\u003e1\u003c/sub\u003e to PCA\u003csub\u003e5\u003c/sub\u003e), before TGN or saline administration (t0). Note how with 3 components, built as a linear combination of 5 MRI variables, the 100% of the variance is explained. \u003cstrong\u003eB:\u003c/strong\u003e Biplots showing the contributions of the MRI variables to the first and third PCA dimensions (horizontal and vertical axis, respectively), as expressed by the squared cosines (cos2) of the coordinates. See how PCA\u003csub\u003e1\u003c/sub\u003e has contributions from AD, MD, FA and RD (to a lesser extent) and that PCA\u003csub\u003e3\u003c/sub\u003e is composed mainly by (-T\u003csub\u003e2\u003c/sub\u003e*), while \u003cstrong\u003eC:\u003c/strong\u003e Results of linear mixed model statistics on the PCA\u003csub\u003e3\u003c/sub\u003e, and corresponding PCA\u003csub\u003e3 \u003c/sub\u003emean values per animal and region on basal time, for control (green) or HFD (brown) mice, including animals form both sexes and treatment group. \u003cstrong\u003eD\u003c/strong\u003e: Percentage of explained variances adding each dimension of the PCA at Glc\u003csub\u003eshort\u003c/sub\u003e. Again, more than 90% of the variance is explained by only 3 variables. \u003cstrong\u003eE:\u003c/strong\u003e Biplots of the contributions of the \u0026nbsp;Glc\u003csub\u003eshort \u003c/sub\u003eMRI variables to PCA\u003csub\u003e2\u003c/sub\u003e and PCA\u003csub\u003e3\u003c/sub\u003e (horizontal and vertical axis, respectively), as expressed by the squared cosines (cos\u003csup\u003e2\u003c/sup\u003e) of the coordinates. Note how PCA\u003csub\u003e2\u003c/sub\u003e has positive contributions from FA and AD, and negative additions from RD and MD, while PCA\u003csub\u003e3\u003c/sub\u003e is still mostly T\u003csub\u003e2\u003c/sub\u003e*. \u003cstrong\u003eF\u003c/strong\u003e: Values of PCA\u003csub\u003e2\u003c/sub\u003e (left) and PCA\u003csub\u003e3\u003c/sub\u003e (right), as a function of the predictor variables that have statistically significant effects. For PCA\u003csub\u003e2\u003c/sub\u003e, values for males and females fed with control (green) or HFD (brown) on time 1 were averaged between areas. Only the interaction \u003cem\u003ediet:sex\u003c/em\u003e was significant (p \u0026lt; 0.05) and \u003cem\u003epost-hoc\u003c/em\u003e tests showed increased PCA\u003csub\u003e2\u003c/sub\u003e on males (p\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.001). PCA\u003csub\u003e3\u003c/sub\u003e is shown as mean regional values for \u003cem\u003evehicle + Glc\u003c/em\u003e (red) or \u003cem\u003eTGN-020 + Glc\u003c/em\u003e (blue) or at \u0026nbsp;Glc\u003csub\u003eshort\u003c/sub\u003e, for males and females. The significant interaction \u003cem\u003esex:treatment\u003c/em\u003e (p \u0026lt; 0.01) on PCA\u003csub\u003e3\u003c/sub\u003e showed significant \u003cem\u003epost-hoc\u003c/em\u003e tests increases with treatment only on males (p \u0026lt; 0.05) (right panel). \u003cstrong\u003eG:\u003c/strong\u003e Percentage of explained variances adding each dimension, with three dimensions explaining again more than 90% of the variance.\u003cstrong\u003e H:\u003c/strong\u003e Biplots showing the contributions of the MRI variables adquired at the last time point (Glc\u003csub\u003elong\u003c/sub\u003e), to the PCA\u003csub\u003e2\u003c/sub\u003e and PCA\u003csub\u003e3\u003c/sub\u003e components (horizontal and vertical axis, respectively), as expressed by the squared cosines (cos2) of the coordinates. In this case, RD contributes positively to PCA\u003csub\u003e2\u003c/sub\u003e, and FA and AD negatively, while PCA\u003csub\u003e3\u003c/sub\u003e is mainly composed by T2*. \u003cstrong\u003eI:\u003c/strong\u003e PCA\u003csub\u003e3\u003c/sub\u003e mean values per animal at t\u003csub\u003e2\u003c/sub\u003e, for \u003cem\u003evehicle + Glc\u003c/em\u003e (red) or \u003cem\u003eTGN-020 + Glc\u003c/em\u003e (blue), with a significant decrease in PCA\u003csub\u003e3\u003c/sub\u003e on the \u003cem\u003eTGN-020 + Glc \u003c/em\u003eanimals (p \u0026lt; 0.01).\u003c/p\u003e","description":"","filename":"Fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/52142f92fd958bb1e37728d8.png"},{"id":80632922,"identity":"374cd63c-a420-4a26-9b71-cf8e3b87f475","added_by":"auto","created_at":"2025-04-15 11:55:46","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":6570756,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMRI changes with diet, treatment and sex: A: \u003c/strong\u003eT\u003csub\u003e2\u003c/sub\u003e* values for CTRL (green dots) and HFD (brown dots) mice, superimposed to a boxplot representation, for every region assessed. The effect of diet was significant on T\u003csub\u003e2\u003c/sub\u003e* (p \u0026lt; 0.05). \u003cstrong\u003eB\u003c/strong\u003e: parametric maps of T\u003csub\u003e2\u003c/sub\u003e* values in the mouse brain, for a CTRL (top) and HFD (bottom) animals. Note the lower values of T2* in the HFD mouse. \u003cstrong\u003e\u0026nbsp;C: \u003c/strong\u003eFA and RD values for CTRL (green dots) and HFD (brown dots) mice (average values of four regions) during the Glc\u003csub\u003eshort \u003c/sub\u003emeasurements, superimposed to a boxplot representation. The effect of diet was significant on FA (p\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.005). \u003cstrong\u003eD\u003c/strong\u003e: Parametric maps of FA values in the mouse brain, for a representative CTRL (top) and a HFD (bottom) animal at the Glc\u003csub\u003eshort \u003c/sub\u003etime point. Note the higher FA values in the HFD mouse. \u003cstrong\u003eE:\u003c/strong\u003e T\u003csub\u003e2\u003c/sub\u003e* values for saline+Glc (red dots) and TGN+Glc (blue dots) mice, superimposed to a boxplot representation, for males and females separately. The effect of the interaction sex:treatment was significant on T\u003csub\u003e2\u003c/sub\u003e* (p \u0026lt; 0.05). \u003cstrong\u003eF\u003c/strong\u003e: FA and RD values for CTRL (green dots) and HFD (brown dots) mice (average values of four regions), superimposed to a boxplot representation, during the Glc\u003csub\u003elong \u003c/sub\u003emeasurements. \u003cstrong\u003eG: \u003c/strong\u003eT\u003csub\u003e2\u003c/sub\u003e* values for saline+Glc (red dots) and TGN+Glc (blue dots) mice, superimposed to a boxplot representation, for\u003cstrong\u003e \u003c/strong\u003eGlc\u003csub\u003elong\u003c/sub\u003e.\u003cstrong\u003e \u003c/strong\u003eThe effect of treatment was significant (p \u0026lt; 0.05). \u003cstrong\u003eH\u003c/strong\u003e: Parametric maps of T\u003csub\u003e2\u003c/sub\u003e* values in the mouse brain, for a representative animal from the saline + glc group (top) and an animal from the TGN + glc batch HFD (bottom) animal, at the Glcs\u003csub\u003elong\u003c/sub\u003e time point. See the lower FA values in the TGN mouse.\u003c/p\u003e","description":"","filename":"Fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/c0e8e2f6a9a81e647ed74668.png"},{"id":80631485,"identity":"5385ed03-a509-44c5-8711-6e4f8c807d33","added_by":"auto","created_at":"2025-04-15 11:47:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1572253,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMFA Hippocampal time changes during a glucose insult with or without AQP4 inhibition. A: \u003c/strong\u003eHippocampal\u003cstrong\u003e \u003c/strong\u003ePCA\u003csub\u003e2\u003c/sub\u003e\u003cstrong\u003e \u003c/strong\u003emean values of CTRL and HFD mice for the \u003cem\u003evehicle + Glc\u003c/em\u003e (red) or \u003cem\u003eTGN-020 + Glc\u003c/em\u003e (blue) groups, as found with MFA, where the interaction \u003cem\u003ediet:treament\u003c/em\u003e showed significant effects on PCA\u003csub\u003e2\u003c/sub\u003e. Post-hoc tests depicted relevant differences only on obese animals (p \u0026lt;0.01). \u003cstrong\u003eB:\u003c/strong\u003e Main MRI variables underlying hippocampal PCA\u003csub\u003e2\u003c/sub\u003e, depicting the experimental AD, MD and FA changes between \u0026nbsp;Glc\u003csub\u003eshort\u003c/sub\u003e and \u0026nbsp;Glc\u003csub\u003elong\u003c/sub\u003e, particularly values from the “long” subtracted by the “short”.\u0026nbsp;\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/db4f99bffe1ec3ea1f5a3a42.png"},{"id":82537480,"identity":"2d17696a-6b24-4277-9887-409ea5e52461","added_by":"auto","created_at":"2025-05-12 16:07:14","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":25147985,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5931441/v1/6a817362-d0f6-4a45-8506-d0b8b4f8398a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Aquaporin-4 inhibition alters cerebral glucose dynamics predominantly in obese animals: an MRI study","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eGlucose is the main source of energy in the mammalian brain, and is critically important for ATP generation and neurotransmitter synthesis\u003csup\u003e1\u003c/sup\u003e. Its transport and metabolism are the result of a tight balance between microvessels, astrocytes, and neurons, the main components of the neurovascular unit\u003csup\u003e2\u003c/sup\u003e. Obesity, one of the most prevalent diseases of our era, is known to alter such elements and its corresponding equilibrium, including changes in the cerebral vasculature wall, in cerebral blood flow and glucose metabolism\u003csup\u003e3,4\u003c/sup\u003e, increased blood brain-barrier permeability \u003csup\u003e5\u003c/sup\u003e, and the development of astrogliosis and microgliosis\u003csup\u003e6,7\u003c/sup\u003e. Glucose entrance into astrocytes and neurons by their specific transporter channels, as well as the glucose metabolism-associated ion trafficking, involve water co-transport\u003csup\u003e8\u003c/sup\u003e, with water entering the intracellular space mainly by aquaporin-4 (\u003cb\u003eAQP4\u003c/b\u003e), the most abundant water channel in the brain\u003csup\u003e9\u003c/sup\u003e. AQP4 is located in astrocytes and ependymal cells, is a key contributor of edema formation and resolution, and it is involved in astrocyte migration and astrogliosis\u003csup\u003e10\u003c/sup\u003e. Alterations in the expression and/or localization of AQP4 lead to a water homeostasis imbalance and have been associated with pathological conditions, including inflammation, stroke or traumatic brain injury\u003csup\u003e11\u003c/sup\u003e. Interestingly, during obesity, AQP4 expression has been reported to be augmented in some brain regions, which has been related to cerebral edema development\u003csup\u003e12,13\u003c/sup\u003e, but decreased in others, potentially due to particular ependymal cells-AQP4 downregulation after a worse glymphatic clearance\u003csup\u003e14\u003c/sup\u003e. During glucose transport and metabolism, water transport through AQP4 is known to yield a certain degree of astrocyte \u003cb\u003eswelling\u003c/b\u003e\u003csup\u003e15\u0026ndash;17\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), but the extent of the regulatory capacity of such mechanism by AQP4, its influence on glucose metabolism and the corresponding alterations during obesity, remain to be understood.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eChanges in extracellular cerebral glucose concentration are perceived by specialized glucose-sensing neurons, the glucose-excited (GE) or glucose-inhibited (GI), which alter their firing rate after a glucose stimulus. They have been described in the hypothalamus\u003csup\u003e18\u003c/sup\u003e and in several brain nuclei that are part of the reward system, such as the hippocampus, thalamus and prefrontal cortex. Particularly, on GE neurons, high glucose uptake and the consequent augmented metabolism leads to an increase of the ATP/ADP ratio and the closure of the K\u003csub\u003eATP\u003c/sub\u003e channels, increasing neuronal activity, neurotransmitter secretion, and Ca\u003csup\u003e2+\u003c/sup\u003e neuronal influx\u003csup\u003e19\u003c/sup\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). This augmented influx into the neurons and its elevated concentration on astrocytes, generates a dilation of the blood vessels\u003csup\u003e20\u003c/sup\u003e. Ionic trafficking in GE neurons may cause, again, astrocyte swelling via increased water entrance through AQP4 transport. Remarkably, glucose-sensing mechanisms are also perturbed during obesity\u003csup\u003e21\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eMagnetic Resonance Imaging (MRI) is a non-invasive imaging technique that provides anatomical and parametric information of structure, function and metabolism\u003csup\u003e22\u003c/sup\u003e. It is broadly used in the clinic as a diagnostic and prognostic tool of numerous brain pathologies such as ischemia, cancer or neurodegeneration\u003csup\u003e23\u003c/sup\u003e. The technique takes advantage of the magnetic properties of the water molecules\u0026rsquo; hydrogen atoms, and particularly, diffusion weighted MRI (dMRI) uses the natural Brownian motion of water molecules to obtain information on the anatomical and biological structure of tissues\u003csup\u003e24\u003c/sup\u003e. Depending on the characteristics of such structures, the movement of water molecules can be isotropic, with no preferred direction, or anisotropic, indicating faster diffusion directions, such as in neuronal bundles. Diffusion tensor images (DTI) can measure this anisotropic behavior by quantifying parameters like mean diffusivity (MD, mean diffusion of the water molecules); axial diffusivity (AD, the diffusion along the fastest direction ); radial diffusivity (RD, the diffusion perpendicular to the main direction), and fractional anisotropy (FA, providing information about the extension of the anisotropy of water motion)\u003csup\u003e25\u003c/sup\u003e. Changes in water diffusion have been correlated with changes in AQP4 expression in central nervous system diseases, such as in the early stages of brain edema after hypoxic-ischemic/reperfusion injury, or obesity, which is known to show altered cerebral DTI parameters related to edema formation and neuroinflammation\u003csup\u003e26\u003c/sup\u003e. Diffusion MRI can also be used to tackle cerebral activation by revealing changes in the diffusion properties of water molecules independently to neurovascular coupling. Notably, glucose-induced swelling mechanisms have been reported by dMRI methods in the context of exogenous glucose administration\u003csup\u003e27\u003c/sup\u003e or orexigenic activation\u003csup\u003e28\u003c/sup\u003e. The neuromorphological changes derived from neuronal activity processes are thought to be the cause of the dMRI changes reported during brain activation\u003csup\u003e29\u003c/sup\u003e, but its exact origin, including the potential contribution of activity-derived cellular swelling, time characteristics or sensitivity to monitor normal brain activity, remain to be clarified\u003csup\u003e30\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e*-weighted imaging (T\u003csub\u003e2\u003c/sub\u003e*WI) is an MRI acquisition able to detect small disturbances in the uniformity of the magnetic field, allowing detection of certain small lesions, and can be used for diagnostic purposes\u003csup\u003e31\u003c/sup\u003e, and detecting blood flow changes\u003csup\u003e32\u003c/sup\u003e. Briefly, the T\u003csub\u003e2\u003c/sub\u003e* relaxation time reflects how quickly the magnetization of water molecules decays due to magnetic field inhomogeneities and magnetic susceptibility differences within tissues. Local changes in the blood oxygen levels are the basis of functional blood oxygen dependent (BOLD) MRI\u003csup\u003e33\u003c/sup\u003e. Indeed, during neuronal activation, T\u003csub\u003e2\u003c/sub\u003e* changes are induced by a change in the ratio deoxyhemoglobin/hemoglobin. Deoxyhemoglobin is a paramagnetic molecule with an iron core that has 4 unpaired electrons per Fe atom, and its presence increases the local magnetic field, and affects T\u003csub\u003e2\u003c/sub\u003e*\u003csup\u003e34\u003c/sup\u003e. Interestingly, previous studies demonstrated brain changes in T\u003csub\u003e2\u003c/sub\u003e*WI images acquired during glucose uptake compared to a control\u003csup\u003e35\u003c/sup\u003e, suggestive of increased microvascular blood flow induced by glucose administration.\u003c/p\u003e \u003cp\u003eOn these grounds, in this study, we tested the hypothesis that the inhibition of the activity of AQP4 during a glucose challenge would reduce the concomitant swelling process and alter the dMRI effects. Moreover, we hypothesized that the effects on obese animals, where AQP4 expression, glial cells and glucose metabolism are perturbed, would be different. To test that, we acquired three consecutive DTI, and complementary T\u003csub\u003e2\u003c/sub\u003e*WI, to control or high fat diet (HFD) fed C57BL6 animals, before any experimental manipulation (\u0026ldquo;Basal\u0026rdquo;), twenty minutes after the administration of saline or 2-(nicotinamide)-1,3,4-thiadiazole (\u003cb\u003eTGN-020\u003c/b\u003e), an AQP4 inhibitor which in animal models significantly reduces ischemic brain edema\u003csup\u003e36\u003c/sup\u003e, and immediately after a glucose insult (\u0026ldquo;Glc\u003csub\u003eshort\u003c/sub\u003e\u0026rdquo;), or 30 min post-glucose (\u0026ldquo;Glc\u003csub\u003elong\u003c/sub\u003e\u0026rdquo;) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). We performed two complimentary data analysis of the cerebral MRI parameters. First, we assessed each MRI set separately (basal, Glc\u003csub\u003eshort\u003c/sub\u003e and Glc\u003csub\u003elong\u003c/sub\u003e), to unveil the specific effects of diet, sex or treatment (\u003cem\u003eHFD, sex and TGN-20 effects on T2* and DTI at separated time points\u003c/em\u003e); next, we assessed how MRI the parameters evolved longitudinally in time for each animal, and considered the differences between the experimental groups (\u003cem\u003eT2* and DTI follow up TGN-20 administration to HFD or CTRL mice\u003c/em\u003e). To do that, we applied principal component analysis (PCA) and subsequent linear mixed-model effects (lme) fitting to the predictor varianles: type of diet, sex, presence of inhibitor (\u0026ldquo;treatment\u0026rdquo;) and brain region. With this rationale, we sought to: i) identify potential initial differences of DTI and T\u003csub\u003e2\u003c/sub\u003e*WI derived parameters between obese and non-obese mice, ii) identify sex-related differences in the context of obesity, glucose uptake and AQP4 function, iii) study the specific role of different cerebral regions during obesity, glucose uptake, and TGN-20 inhibition.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e1. Body weight\u003c/h2\u003e \u003cp\u003eMale mice of the HFD group had significantly increased body weight (BW) than control animals, both in the \u0026ldquo;Saline\u0026thinsp;+\u0026thinsp;Glc\u0026rdquo; and \u0026ldquo;TGN\u0026thinsp;+\u0026thinsp;Glc\u0026rdquo; groups (45.1\u0026thinsp;\u0026plusmn;\u0026thinsp;3.0 Vs 30.3\u0026thinsp;\u0026plusmn;\u0026thinsp;1.7 for the saline batch, 49.0\u0026thinsp;\u0026plusmn;\u0026thinsp;3.1 Vs 30.0\u0026thinsp;\u0026plusmn;\u0026thinsp;1.4 for TGN) (p\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.005). Female HFD animals also depicted higher BW (32.1\u0026thinsp;\u0026plusmn;\u0026thinsp;5.1 Vs 21.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3 for the saline batch, 30.6\u0026thinsp;\u0026plusmn;\u0026thinsp;5.4 Vs 21.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4 for TGN) (p\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.05) (anova tests of BW Vs diet, sex and treatment with interaction), although BW differences where not that striking. The groups assigned to different treatment groups showed no significant differences between them.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003e2. HFD, sex and TGN-20 effects on T2* and DTI at separated time points\u003c/h3\u003e\n\u003cp\u003eAfter performing one independent PCA procedure of the MRI variables per time point - the \u0026ldquo;Basal\u0026rdquo;, \u0026ldquo;Glc\u003csub\u003eshort\u003c/sub\u003e\u0026rdquo; and \u0026ldquo;Glc\u003csub\u003elong\u003c/sub\u003e\u0026rdquo; PCAs -, we achieved a dimensionality reduction, with the first 3 PCA explaining\u0026thinsp;\u0026ge;\u0026thinsp;90% of the variance (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, A, D, G panels), for each time point, respectively. The component loads of the variables -the composition of the PCA, in terms of the original MRI-, was stable across the three time points, with AD and MD mainly composing PCA\u003csub\u003e1\u003c/sub\u003e, RD and FA PCA\u003csub\u003e2\u003c/sub\u003e, and T\u003csub\u003e2\u003c/sub\u003e* PCA\u003csub\u003e3\u003c/sub\u003e, (Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Subsequently, the effects of the predictor variables -diet, sex or treatment- were tested on the three corresponding PCAs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComponent loads of the variables at times 1, 2 and 3.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"10\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePCA\u003csub\u003e1\u003c/sub\u003e basal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePCA\u003csub\u003e2\u003c/sub\u003e basal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePCA\u003csub\u003e3\u003c/sub\u003e basal\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePCA\u003csub\u003e1\u003c/sub\u003e time 1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePCA\u003csub\u003e2\u003c/sub\u003e time1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003ePCA\u003csub\u003e3\u003c/sub\u003e time1\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003ePCA\u003csub\u003e1\u003c/sub\u003e time 2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003ePCA\u003csub\u003e2\u003c/sub\u003e time 2\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c10\"\u003e \u003cp\u003ePCA\u003csub\u003e3\u003c/sub\u003e time 2\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e-0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.61\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.55\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e-0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.49\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.51\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e-0.60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e-0.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e-0.99\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e-0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e-0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e \u003cp\u003e0.98\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAt basal time -before any administration- we found that the type of diet consumed affected significantly PCA\u003csub\u003e3\u003c/sub\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC), in a region-dependent manner (significant \u003cem\u003eregion:diet\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, lme followed by Anova). Particularly, HFD mice showed lower PCA\u003csub\u003e3\u003c/sub\u003e than CTRL mice, in all regions except in the hippocampus, with post-hoc corrected comparisons not reaching statistical relevance. PCA\u003csub\u003e3\u003c/sub\u003e was composed mainly by T\u003csub\u003e2\u003c/sub\u003e* (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB), and the specific analysis of \u003cb\u003eT\u003c/b\u003e\u003csub\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e*\u003c/b\u003e as a function of diet and region revealed similar results (\u003cem\u003eregion:diet\u003c/em\u003e, p\u0026thinsp;=\u0026thinsp;0.05), with lower values on the hippocampus of HFD mice, and slightly higher in the rest of the regions, as compared to CTRL, and without reaching robust statistical evidences sub-regionally (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA and \u003cb\u003eB\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIn the Glc\u003csub\u003eshort\u003c/sub\u003e measurements, the lme approach revealed a significant sex-dependent diet effect (\u003cem\u003ediet:sex\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) on PCA\u003csub\u003e2\u003c/sub\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). Specifically, male HFD animals had higher PCA\u003csub\u003e2\u003c/sub\u003e, as compared to CTRL mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF left). PCA\u003csub\u003e2\u003c/sub\u003e was composed by FA and RD (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), and higher FA and lower RD could be seen in the HFD mice, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC \u003cb\u003eand D\u003c/b\u003e), with p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 only for FA (anova test of FA Vs diet, sex and \u003cem\u003ediet:sex\u003c/em\u003e). Additionally, we found that the type of treatment affected PCA\u003csub\u003e3\u003c/sub\u003e, on a \u003cem\u003eregion\u003c/em\u003e and \u003cem\u003ediet\u003c/em\u003e-dependent manner (interaction of \u003cem\u003eregion:diet:treatment\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), and depending on sexes (treatment:sex, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Particularly, HFD animals with TGN treatment tended to higher PCA\u003csub\u003e3\u003c/sub\u003e, as compared to HFD saline-administered mice, in all regions except on the hippocampus, with adjusted post-hoc contrasts not reaching statistical significance. TGN-treated male animals showed significantly higher PCA\u003csub\u003e3\u003c/sub\u003e values than the saline males (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF right). Since PCA\u003csub\u003e3\u003c/sub\u003e was mainly composed by T\u003csub\u003e2\u003c/sub\u003e*, this was translated into lower values of T\u003csub\u003e2\u003c/sub\u003e* in TGN-treated male animals, as compared to controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 \u003cem\u003esex:treatment\u003c/em\u003e, Anova tests, p\u003csub\u003eadj\u003c/sub\u003e = 0.06 for post-hoc in males) .\u003c/p\u003e \u003cp\u003eIn the Glc\u003csub\u003elong\u003c/sub\u003e measurements (30 min after Glc administration) the type of diet was still affecting significantly PCA\u003csub\u003e2\u003c/sub\u003e values (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e, with HFD mice having significantly lower PCA\u003csub\u003e2\u003c/sub\u003e values than CTRL animals. RD and FA were positively and negatively correlated with PCA\u003csub\u003e2\u003c/sub\u003e, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), and lower PCA\u003csub\u003e2\u003c/sub\u003e was translated into a lower RD and higher FA on HFD mice \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF\u003cb\u003e)\u003c/b\u003e, but the independent tests of FA Vs diet, and RD Vs diet did not reach statistical significance. The type of treatment received had a significant effect on PCA\u003csub\u003e3\u003c/sub\u003e, (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01), with TGN\u0026thinsp;+\u0026thinsp;Glc administered mice presenting significantly lower PCA\u003csub\u003e3\u003c/sub\u003e values, than Saline\u0026thinsp;+\u0026thinsp;Glc mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI). Our results also revealed a significant \u003cem\u003eregion\u003c/em\u003e-dependent \u003cem\u003ediet\u003c/em\u003e effect (\u003cem\u003eregion*diet\u003c/em\u003e interaction, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) indicating that the HFD group had higher or equal PCA\u003csub\u003e3\u003c/sub\u003e values than CTRL mice in all regions but the hippocampus (significantly only in cortex). Since T\u003csub\u003e2\u003c/sub\u003e* and PCA\u003csub\u003e3\u003c/sub\u003e values were positively correlated (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), lower PCA\u003csub\u003e3\u003c/sub\u003e with \u003cb\u003eTGN\u003c/b\u003e indicated lower T\u003csub\u003e2\u003c/sub\u003e \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG-H\u003cb\u003e) (\u003c/b\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.05 t-test T\u003csub\u003e2\u003c/sub\u003e* Vs treatment\u003cb\u003e).\u003c/b\u003e\u003c/p\u003e\n\u003ch3\u003e3. T2* and DTI follow up TGN-20 administration to HFD or CTRL mice\u003c/h3\u003e\n\u003cp\u003eTo quantify the animal-specific changes of the MRI parameters, before and after the administration of either \u0026ldquo;\u003cem\u003esaline\u0026thinsp;+\u0026thinsp;glc\u0026rdquo;\u003c/em\u003e or \u0026ldquo;\u003cem\u003eTGN\u0026thinsp;+\u0026thinsp;glc\u0026rdquo;\u003c/em\u003e, we built a global MFA that included MRI variables from the three longitudinal time-point measurements. Specifically, we included the subtraction, for each MRI parameter, of the \u0026ldquo;\u003cem\u003eGlc\u003c/em\u003e\u003csub\u003e\u003cem\u003eshort\u003c/em\u003e\u003c/sub\u003e \u003cem\u003eminus Basal\u0026rdquo;\u003c/em\u003e values (Glc\u003csub\u003eshort\u003c/sub\u003e-Basal), \u0026ldquo;\u003cem\u003eGlc\u003c/em\u003e\u003csub\u003e\u003cem\u003elong\u003c/em\u003e\u003c/sub\u003e \u003cem\u003eminus Glc\u003c/em\u003e\u003csub\u003e\u003cem\u003eshort\u003c/em\u003e\u003c/sub\u003e\u0026rdquo; (Glc\u003csub\u003elong\u003c/sub\u003e- Glc\u003csub\u003eshort\u003c/sub\u003e) and \u0026ldquo;\u003cem\u003eGlc\u003c/em\u003e\u003csub\u003e\u003cem\u003elong\u003c/em\u003e\u003c/sub\u003e \u003cem\u003eminus Basal\u003c/em\u003e\u0026rdquo; (Glc\u003csub\u003elong\u003c/sub\u003e- Basal) and performed the MFA analysis on them. Results yielded 4 relevant principal components, and, subsequently, the effects of \u003cem\u003ediet\u003c/em\u003e, \u003cem\u003eregion\u003c/em\u003e, \u003cem\u003esex\u003c/em\u003e or \u003cem\u003etreatment\u003c/em\u003e were assessed by a lme approach followed by Anova. After applying such procedure, we observed a common regional pattern, with the hippocampus showing consistently a different tendency, as compared to rest of the regions (data not shown). We thus decided to consider the hippocampus as an independent region and conducted a specific MFA for the hippocampus, and a MFA from the rest of the regions.\u003c/p\u003e\n\u003ch3\u003e3.1- Hippocampal evolution\u003c/h3\u003e\n\u003cp\u003eThe MFA on the hippocampus resulted in 4 PCAs explaining\u0026thinsp;\u0026ge;\u0026thinsp;80% of the variance, thus achieving a dimensionality reduction from 12 initial MRI parameters to 4 main components. PCA\u003csub\u003e1\u003c/sub\u003e (30% explained variance) was mostly composed of variables from the time regimes Glc\u003csub\u003elong\u003c/sub\u003e-Basal and Glc\u003csub\u003eshort\u003c/sub\u003e-Basal, and a PCA\u003csub\u003e2\u003c/sub\u003e (20%) mainly composed by the time regime Glc\u003csub\u003elong\u003c/sub\u003e-Glc\u003csub\u003eshort\u003c/sub\u003e. PCA\u003csub\u003e3\u003c/sub\u003e contained a mixture of three-time regimes, and PCA\u003csub\u003e4\u003c/sub\u003e was formed by Glc\u003csub\u003eshort\u003c/sub\u003e-Basal and Glc\u003csub\u003elong\u003c/sub\u003e-Basal. The component loads of the experimental MRI variables on the global PCA, are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHippocampal MFA component loads of the experimental variables.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePCA\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePCA\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePCA\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePCA\u003csub\u003e4\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.70\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.54\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.37\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.71\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.55\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.10\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.03\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.03\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.43\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eOn PCA\u003csub\u003e1,\u003c/sub\u003e the most contributing variables were (MD 16%, AD 12%, RD 10%) \u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e, (AD 13%, FA 10%, MD 8%)\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e and (RD 10.5%, FA 10%, MD 8%)\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e, all with a positive contribution. The type of treatment received had a significant effect on PCA\u003csub\u003e1\u003c/sub\u003e (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), with the TGN\u0026thinsp;+\u0026thinsp;Glc group showing higher (and positive) PCA\u003csub\u003e1\u003c/sub\u003e values, as compared to Saline\u0026thinsp;+\u0026thinsp;Glc animals, which had negative PCA\u003csub\u003e1\u003c/sub\u003e values (data not shown). Since PCA\u003csub\u003e1\u003c/sub\u003e was composed by changes of MD, AD, RD and FA, positively correlated, negative PCA\u003csub\u003e1\u003c/sub\u003e in the Saline\u0026thinsp;+\u0026thinsp;Glc animals can be translated into decreasing MD, AD, RD and FA from basal to post times, and from basal to pre, but augmenting TGN\u0026thinsp;+\u0026thinsp;Glc group.\u003c/p\u003e \u003cp\u003eOn PCA\u003csub\u003e2\u003c/sub\u003e, the most contributing variables were (AD 34%, MD 22.5% and FA 10,5%), all from the Glc\u003csub\u003elong\u003c/sub\u003e-Glc\u003csub\u003eshort\u003c/sub\u003e regime. A significant \u003cem\u003ediet\u003c/em\u003e-dependent treatment effect (interaction of \u003cem\u003ediet:treatment\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) was revealed. On HFD mice, TGN treatment group showed significantly lower (and negative) PCA\u003csub\u003e2\u003c/sub\u003e values than saline animals, which had positive PCA\u003csub\u003e2\u003c/sub\u003e values (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Since PCA\u003csub\u003e2\u003c/sub\u003e represented changes of AD, MD and FA, positively correlated, a positive \u003cb\u003ePCA\u003c/b\u003e\u003csub\u003e\u003cb\u003e2\u003c/b\u003e\u003c/sub\u003e in the Saline\u0026thinsp;+\u0026thinsp;Glc mice can be translated into augmenting AD, MD and FA values in the time regime from Glc\u003csub\u003eshort\u003c/sub\u003e to Glc\u003csub\u003elong\u003c/sub\u003e (Glc\u003csub\u003elong\u003c/sub\u003e \u0026ndash; Glc\u003csub\u003eshort\u003c/sub\u003e \u0026gt; 0) but decreasing (Glc\u003csub\u003elong\u003c/sub\u003e \u0026ndash; Glc\u003csub\u003eshort\u003c/sub\u003e \u0026lt; 0) on the TGN\u0026thinsp;+\u0026thinsp;Glc group (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB) (diet:treatment p\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.05 for AD and FA, and subsequent post-hoc padj\u0026thinsp;\u0026lt;\u0026thinsp;0.05 for treatment effects only on HFD mice).\u003c/p\u003e \u003cp\u003eNo relevant effects were found on the effects of diet or treatment on PCA\u003csub\u003e3\u003c/sub\u003e and PCA\u003csub\u003e4\u003c/sub\u003e on the hippocampus.\u003c/p\u003e\n\u003ch3\u003e3.2- Cortex, thalamus and hypothalamus evolution\u003c/h3\u003e\n\u003cp\u003eMFA on the hippocampus resulted in 4 PCA explaining the target variance with a PCA\u003csub\u003e1\u003c/sub\u003e (31% explained variance) formed by Glc\u003csub\u003elong\u003c/sub\u003e-Basal and Glc\u003csub\u003eshort\u003c/sub\u003e-Basal, a PCA\u003csub\u003e2\u003c/sub\u003e (23%) mainly composed of variables from the time regime Glc\u003csub\u003elong\u003c/sub\u003e-Glc\u003csub\u003eshort\u003c/sub\u003e and Glc\u003csub\u003eshort\u003c/sub\u003e-Basal. PCA\u003csub\u003e3\u003c/sub\u003e (18%) was mostly composed of the time regimes Glc\u003csub\u003elong\u003c/sub\u003e-Basal and Glc\u003csub\u003eshort\u003c/sub\u003e-Basal and PCA\u003csub\u003e4\u003c/sub\u003e (14%) formed by Glc\u003csub\u003elong\u003c/sub\u003e-Glc\u003csub\u003eshort\u003c/sub\u003e and Glc\u003csub\u003elong\u003c/sub\u003e-Basal. The specific component loads of the MRI variables on the \u003cem\u003erest of the regions\u003c/em\u003e are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cem\u003eRest of the regions\u003c/em\u003e MFA component loads of the experimental variables.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePCA\u003csub\u003e1\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePCA\u003csub\u003e2\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePCA\u003csub\u003e3\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePCA\u003csub\u003e4\u003c/sub\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.52\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.24\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.73\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.66\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.34\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.82\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlcshort\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.08\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.69\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.32\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.34\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.78\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e-0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.47\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlclong\u0026minus;B\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e-0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRD\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.47\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFA\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.81\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT\u003csub\u003e2\u003c/sub\u003e\u003csup\u003e*\u003c/sup\u003e\u003csub\u003eGlclong\u0026minus;Glcshort\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e-0.07\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.04\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e-0.12\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eOn PCA\u003csub\u003e3\u003c/sub\u003e, the most contributing variables were RD and MD (Glc\u003csub\u003elong\u003c/sub\u003e-Basal 30% and 17.5%, respectively) and RD and MD (Glc\u003csub\u003eshort\u003c/sub\u003e-Basal 20,5% and 10%). Our lme showed a significant \u003cem\u003esex\u003c/em\u003e and \u003cem\u003eregion\u003c/em\u003e-dependent \u003cem\u003ediet\u003c/em\u003e effect (interaction of \u003cem\u003eregion*diet*sex\u003c/em\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), where the HFD group tended to have lower or equal PCA\u003csub\u003e3\u003c/sub\u003e values.\u003c/p\u003e \u003cp\u003eNo relevant effects were found on the effects of diet or treatment on PCA\u003csub\u003e1\u003c/sub\u003e, PCA\u003csub\u003e2\u003c/sub\u003e and PCA\u003csub\u003e4\u003c/sub\u003e on the cortex, thalamus and hypothalamus.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eIn this work, we investigated the role of AQP4 during glucose uptake in obese and non-obese mice by MRI in four brain regions. Specifically, we performed three sequential DTI and T\u003csub\u003e2\u003c/sub\u003e*WI acquisitions to follow the effects of TGN (or its vehicle, saline) administration prior to the glucose insult. Next, having obtained a series of MRI parameters per region and time point, we applied PCA and MFA approaches to the data to reduce dimensionality, and then used linear mixed models on the corresponding main components to infer the potential effects of type of treatment, diet consumed, sex or brain regions, on the MRI variables.\u003c/p\u003e\n\u003ch3\u003eEffects of TGN on microvascular blood flow and swelling\u003c/h3\u003e\n\u003cp\u003eOur results are consistent with a change in the microvasculature blood flow induced by glucose that is inhibited by TGN, as quantified by lower T\u003csub\u003e2\u003c/sub\u003e* values in TGN mice, as compared to saline animals, (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, I, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE, G, H). Such effects were more pronounced in males, during the Glc\u003csub\u003eshort\u003c/sub\u003e measures, but the sexual dependance disappears in the Glc\u003csub\u003elong\u003c/sub\u003e. In parallel, our data agrees well with a glucose-induced cell swelling mechanism in the hippocampus of HFD mice that is inhibited by TGN, as detected by higher FA, AD and MD changes from Glc\u003csub\u003eshort\u003c/sub\u003e to Glc\u003csub\u003elong\u003c/sub\u003e in saline animals, as compared to TGN-treated mice (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Additionally, our work reports specific diet effects on brain cellularity, with HFD mice exhibiting lower RD and higher FA than controls, more remarkably in males (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC), which is consistent with the presence of gliosis\u003csup\u003e37\u003c/sup\u003e. Notably, this diet effect on diffusivity was still remarkable after the long effect of glucose (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eConsidering both the T2* and DTI measures, results are consistent with a TGN reduction of the microvascular blood flow and cerebral activity that is detectable by MRI, highlighting the role of AQP4 during glucose uptake. Indeed, previous studies showed that glucose administration leads to increased cerebral activity and microvasculature blood flow\u003csup\u003e35\u003c/sup\u003e, and that such changes can be detected by MRI as increased markers of cellular swelling and microvascular blood flow, probably through glucose-sensing neurons\u003csup\u003e38\u003c/sup\u003e. Glucose uptake in the CNS is specifically associated with the entry of water into the cells via AQP4, leading to cell swelling and a decrease in extracellular space\u003csup\u003e15,39\u003c/sup\u003e. When glial swelling occurs, the cell shape becomes elongated\u003csup\u003e40\u003c/sup\u003e, resulting in more anisotropic diffusion of the water molecules, and MRI has been used to detect gliosis processes during obesity\u003csup\u003e41\u003c/sup\u003e. Particularly, using DTI, changes in cell shape can be characterized by higher FA\u003csup\u003e42\u003c/sup\u003e, higher AD and MD, and lower RD\u003csup\u003e43\u003c/sup\u003e. TGN-020 is known to inhibit AQP4 channels, preventing the passage of water through them\u003csup\u003e44\u003c/sup\u003e. Such inhibition may cause an alteration in the flow of ions (K\u003csup\u003e+\u003c/sup\u003e, Ca\u003csup\u003e2+\u003c/sup\u003e, etc), altering astrocyte function and thus impairing the glucose-induced swelling. Our results revealing \u003cem\u003ein vivo\u003c/em\u003e the relationship between AQP4 inhibition and impaired changes are consistent with previous studies, which reported \u003cem\u003eex vivo\u003c/em\u003e how the activity-dependent glial swelling was impaired on AQP4 knockout mice\u003csup\u003e45\u003c/sup\u003e, or \u003cem\u003ein vitro\u003c/em\u003e how the deletion of AQP4 reduces astrocyte swelling during an oxygen-glucose deprivation challenge\u003csup\u003e46\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eDiet effects\u003c/h3\u003e\n\u003cp\u003eOur work reveals that diet affects the T\u003csub\u003e2\u003c/sub\u003e* and DTI values of the mouse brain. For instance, we report region-specific altered T\u003csub\u003e2\u003c/sub\u003e* values on obese mice at basal measurements, as compared to non-obese, suggesting that HFD induced lower values specifically in the hippocampus, on agreement with previous results in humans. Indeed, obesity is associated with both generalized and region-specific reductions in cerebral blood flow, with the hippocampus being a particularly vulnerable area. Previous neuroimaging studies demonstrated that higher body mass index correlated with progressively reduced blood flow across nearly all brain regions, indicating a global decline in perfusion\u003csup\u003e47\u003c/sup\u003e. The hippocampus\u0026mdash;critical for memory and linked to Alzheimer\u0026rsquo;s disease (AD) risk shows pronounced sensitivity, with obesity-related hypoperfusion comparable to aging effects\u003csup\u003e48\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe experiments here reported also highlight that the swelling-impairment effects of TGN depend on the diet consumed, as found in the results of the MFA, again in the hippocampus (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). On the Glc\u003csub\u003eshort\u003c/sub\u003e to Glc\u003csub\u003elong\u003c/sub\u003e period -the interval when glucose is expected to act on the brain- the TGN group had lower FA, MD and AD changes, as compared to the saline animals, only on the HFD subgroup. These results are consistent with the inhibition of a directionally dependent, glucose-induced swelling process that would occur in saline, and the fact that this change was reported only on the HFD group may indicate that AQP4 is more effectively inhibited in this group. Moreover, is consistent with previous experiments that have reported increased AQP4 expression during obesity\u003csup\u003e12,49\u003c/sup\u003e.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSexual dimorphism\u003c/h2\u003e \u003cp\u003eIn this study, we reported some sexual differences, including an increased susceptibility to diet, regarding the effects on brain DTI (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF \u003cb\u003eleft\u003c/b\u003e), and to treatment changes on T\u003csub\u003e2\u003c/sub\u003e* (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF \u003cb\u003eright\u003c/b\u003e), in male mice. In animal models of diet-induced obesity, sexual differences are common, with males developing the obese phenotype typically faster and to a higher degree than females, who seem to be more protected against its development\u003csup\u003e50\u003c/sup\u003e. This is reflected, for example, in the BW of the animals, in which diet induces larger increases in the case of males, as reported here. Likewise, brain changes underlying the obese phenotype are sexual-dependent, with male mice being more prone to depict a neuroinflammatory profile\u003csup\u003e51\u003c/sup\u003e. In this sense, our results are consistent with an increased male-specific presence of astrogliosis and microgliosis during HFD. At the same time, our results suggest that TGN inhibition may be more effective under circumstances of increased AQP4 abundance.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003ePCA and MFA\u003c/h2\u003e \u003cp\u003eIn this work, the use of PCA and MFA allowed a reduction in dimensionality, obtaining new principal components that were independent from each other. This implies that the statistical tests performed on the corresponding components were independent\u003csup\u003e52\u003c/sup\u003e, and allowed us to pinpoint those combinations of MRI variables that were affected by predictor variables. Briefly, for example, from the four DTI-derived variables (MD, AD, RD and FA), which are highly correlated between them, the PCA method retained only two relevant components, with MD and AD showing the highest scores on PCA\u003csub\u003e1\u003c/sub\u003e, and RD and FA dominating PCA\u003csub\u003e2\u003c/sub\u003e. The statistical tests on PCAs revealed that only PCA\u003csub\u003e2\u003c/sub\u003e resulted remarkably altered by the explanatory variables, particularly by diet and sex, and we could not find any treatment effects. This led us to explore in detail the significance of the variables sex and diet on FA and RD, and no further tests on MD or AD were performed, and the variable treatment was omitted (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), thus reducing the number of tests performed and the probability of committing type I errors. Interestingly, in some cases, lower p-values were found on the tests on the created components (PCA\u003csub\u003e2\u003c/sub\u003e) than on the tests of the MRI variables composing them, suggesting the convenience of the method to find the variable combination that best reveals the underlying effects. In our study, PCA\u003csub\u003e2\u003c/sub\u003e captured the variance related to smaller diffusivities (RD\u0026thinsp;\u0026lt;\u0026thinsp;AD), and to the degree of diffusion anisotropy, and thus to the degree on anisotropy of cells that shape such movements. The consistent negative correlation between FA and RD suggests that RD alterations may be good indicators of potential changes on cellular shape, in agreement with previous literature\u003csup\u003e53\u003c/sup\u003e. T\u003csub\u003e2\u003c/sub\u003e*, on the other hand, was in the three time-points almost exclusively dominating PCA\u003csub\u003e3\u003c/sub\u003e, showing lower correlations with the diffusion behavior of water molecules and potentially reflecting the vascular components.\u003c/p\u003e \u003cp\u003eMFA, on the other hand, gave us information on how the time periods, and the variables within, were correlated between them. By definition, MFA builds first separated PCA for each time regime, and after corresponding normalizations, generates a global PCA in which the initial variables are expressed as function of such global PCAs\u003csup\u003e54\u003c/sup\u003e (Table \u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Table \u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In our results, both in the \u003cem\u003eMFA hippocampus\u003c/em\u003e, and in \u003cem\u003erest of the regions\u003c/em\u003e analysis, PCA\u003csub\u003e1\u003c/sub\u003e was mainly dominated by variables from the \u0026ldquo;Glc\u003csub\u003eshort\u003c/sub\u003e\u003cem\u003e-\u003c/em\u003ebasal\u0026rdquo; and \u0026ldquo;Glc\u003csub\u003elong\u003c/sub\u003e-basal\u0026rdquo; regimes, with MD and AD showing the higher loadings. On the contrary, the MRI variables with the higher PCA\u003csub\u003e2\u003c/sub\u003e loadings corresponded to the \u0026ldquo;Glc\u003csub\u003elong\u003c/sub\u003e-Glc\u003csub\u003eshort\u0026rdquo;\u003c/sub\u003e interval, with a clear domination of AD, followed by MD. Considering the experimental timeline of TGN/saline and glucose administrations followed here, PCA\u003csub\u003e1\u003c/sub\u003e seems to be expressing potential differences on global diffusivity caused by the TGN administration itself, while PCA\u003csub\u003e2\u003c/sub\u003e may be more related to the variance induced by glucose uptake and metabolism, and its effects on the highest diffusion directions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eLIMITATIONS\u003c/h2\u003e \u003cp\u003eIn this work, we used TGN-020 as an inhibitor of AQP4 function, based on evidence of several previous studies\u003csup\u003e36,55\u0026ndash;57\u003c/sup\u003e. However, some research has reported the inability of this compound to inhibit AQP4 function\u003csup\u003e58\u003c/sup\u003e, highlighting the need for a clearer understanding of the TGN-020 potential targets before interpreting the results of the experiments in vivo. In our study, based on MRI measurements, TGN administration appeared to effectively inhibit AQP4 channels, since it prevented the glucose administration-induced MRI changes observed in the absence of the inhibitor, the cellular swelling. While our results correlate well with previous studies reporting such swelling inhibition\u003csup\u003e45,46\u003c/sup\u003e future experiments should investigate the mechanistic insights of the MRI-detected processes.\u003c/p\u003e \u003cp\u003eThere are several limitations of the PCA or MFA approaches carried out with our data. In both cases, the underlying correlation of the initial MRI variables was not that strong (Kaiser-Meyer-Olkin criteria about 0.5), which effectively meant reducing only from 5 to 3 the number of components. On the MFA method, on the other hand, since some animals did not have information on the three time points, due to experimental problems, some data was inevitably lost.\u003c/p\u003e \u003cp\u003eRegarding the DTI variables, not all animals were acquired with the same number of directions, as specified in the methods section. Indeed, those animals acquired during the first batch included acquisitions of 6 directions only. This group showed lower data quality, which could add a potential bias in interpretation. It should be noted, however, that the type of study (\u0026ldquo;batch 1\u0026rdquo; or \u0026ldquo;batch 2\u0026rdquo;) was initially included as a predictor variable in the lme approaches, to test is belonging to either of the groups was affecting significantly the PCA variables, and no relevant effects were reported (data not shown).\u003c/p\u003e \u003cp\u003eWhile our findings offer valuable insights into AQP4 inhibition and glucose metabolism in the context of obesity, it is important to recognize the limitations of using a mouse model to extrapolate results to humans. One significant limitation is the difference in basal metabolic rates between mice and humans, which may affect generalizations related to energy expenditure and glucose regulation. Additionally, while the DIO models replicate key features of human obesity, including glucose intolerance and fat accumulation, the precise metabolic pathways and hormonal responses, particularly those related to glucose sensing and insulin signaling, may differ between the two species. Furthermore, the inflammatory response in mice is often more acute, and the distribution of fat (subcutaneous vs. visceral) may vary, influencing how obesity-related inflammation impacts glucose metabolism and brain function. Although the glucose loading model used in this study helps to mimic postprandial glycemic responses, its exact relevance to human physiology warrants careful consideration. In humans, glucose metabolism and uptake are regulated by a more complex set of hormonal controls.\u003c/p\u003e \u003cp\u003eIsoflurane anesthesia can significantly influence cerebral perfusion by inducing vasodilation and altering cerebral blood flow, as reported in previous studies. In our study, we maintained a consistent anesthetic protocol for all animals, ensuring that any observed differences between experimental groups were not confounded by variations in anesthesia depth. While isoflurane anesthesia may influence the absolute values of perfusion-related parameters, our analysis focused on relative differences between conditions (e.g., glucose administration and AQP4 inhibition), which remain interpretable within this controlled setting.\u003c/p\u003e \u003cp\u003eAlthough the findings should be interpreted within the context of an anesthetized model, the use of isoflurane was necessary to ensure stability and immobility during MRI acquisition, which is critical for obtaining high-quality, artifact-free images. While awake imaging could avoid the confounding effects of anesthesia, it introduces other challenges, such as motion artifacts and stress-related physiological changes, which could also confound the interpretation of MRI-derived metrics.Finally, another limitation is that while our study demonstrated the effects of TGN-020 in inhibiting AQP4 and altering brain microvascular blood flow in mice, these results may not directly translate to human neurovascular function. The expression and regulation of AQP4 in human brain tissues might differ, affecting the applicability of our results to human conditions.\u003c/p\u003e \u003c/div\u003e"},{"header":"CONCLUSIONS","content":"\u003cp\u003eWe can conclude that TGN-020 acts as an AQP4 inhibitor after a glucose insult that results in changes in the neurovascular unit detectable by MRI in vivo, including lower blood flow in several brain regions, as detected by T\u003csub\u003e2\u003c/sub\u003e*WI, and a decrease in glucose-induced cell swelling process detectable by DTI in the hippocampus of obese mice. In addition, we observed in obese mice MRI indicators of altered cortical microvasculature blood flow, potentially related to vascular inflammation, and DTI markers consistent with astro- or microgliosis processes, predominantly on males.\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e1. Experimental design\u003c/h2\u003e \u003cp\u003eThe mice used in this study were bred and housed in our institutional animal facility at Institute for Biomedical Research Sols-Morreale (Reg. No. Es280790000288). n\u0026thinsp;=\u0026thinsp;66 adult C57BL6/J mice (32 females) were kept in a room of the animal facility with controlled humidity (47%) and temperature (21\u0026ndash;23\u0026deg;C), a 12 h light/day cycle and fed with a standard laboratory food (SAFE DIETS). At 6 weeks of age, 33 animals (18 females) were switched to an HFD (60% fat, 20% carbohydrates, 20% proteins from D12492, Research Diets, New Brunswick, NJ), to induce obesity, and the rest continued with standard diet. After 18 weeks of diet diversification, mice were subjected to MRI under anesthesia provided through a nose mask (isoflurane 1-1.5% in O\u003csub\u003e2\u003c/sub\u003e, 1L/min) during the whole study, to assess the response to a glucose administration stimulus (i.p., 200 \u0026micro;L/25g, 2.08M on a saline solution, 16.65 mmol/kg) preceded by a saline i.p. (100 \u0026micro;L/25g) or TGN-020 (100 \u0026micro;L/25g a 0.24M, 0.96 mmol/kg administration), 20 minutes before the glucose insult\u003csup\u003e59\u003c/sup\u003e. Depending on the diet and the type of administration received, animals were thus assigned to four groups of analysis, i) fed with a standard laboratory control diet and injected first with saline and glc (7 males and 7 females), ii) fed with CTRL and administered with TGN-020 and glc (11 males and 7 females), iii) group fed with HFD and administered with saline and glc (6 males and 7 females), and iv) group fed with HFD and administered with TGN-020 and glc (8 males and 7 females) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Once the MRI experiments finished, animals were euthanized by rapid decapitation, still under the effects of anesthesia administered during the imaging sessions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e2. Compound preparation\u003c/h2\u003e \u003cp\u003eTGN-020 is water-insoluble; therefore, a sodium salt derivative of the compound (TGN-Na) was prepared to facilitate its subsequent injection into animals using a saline solution. First, 41.20 mg of TGN-020 was placed in a round-bottom flask with 4 ml of Millipore water. Next, 426.24 \u0026micro;L of NaOH (1M) were slowly added, leaving it stirring for 30 minutes at room temperature. Finally, the solution obtained was filtered and then lyophilized to generate the sodium salt (TGN-020-Na) that was injected into the animals (throughout the text, this sodium salt will be referred to as TGN-020).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3. MRI acquisitions\u003c/h2\u003e \u003cp\u003eMice were subjected to MRI under anesthesia (isoflurane) using a superconducting magnet with 16 cm diameter and a gradient insert of 360 mT/m (Biospec\u0026reg; 7T, Bruker Biospect, Ettlingen, Germany), in two separated animal cohorts. First, DTI studies (6 directions, b-values\u0026thinsp;=\u0026thinsp;400 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e/s \u0026amp; 1800 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e/s, field of view\u0026thinsp;=\u0026thinsp;23 x 23 mm\u003csup\u003e2\u003c/sup\u003e, slice thickness\u0026thinsp;=\u0026thinsp;1.25 mm, 5 slices) and T\u003csub\u003e2\u003c/sub\u003e*WI to generate T\u003csub\u003e2\u003c/sub\u003e* maps (TR\u0026thinsp;=\u0026thinsp;300 ms, flip angle\u0026thinsp;=\u0026thinsp;30, 10 echoes, first TE\u0026thinsp;=\u0026thinsp;2 ms, inter echo time: 4 ms, 8 averages, and the same geometrical parameters as DTI) were acquired in an initial group of animals (n\u0026thinsp;=\u0026thinsp;35, n\u0026thinsp;=\u0026thinsp;19 with DIO, n\u0026thinsp;=\u0026thinsp;15 females). Subsequently, an additional cohort was investigated (n\u0026thinsp;=\u0026thinsp;31, n\u0026thinsp;=\u0026thinsp;14 with DIO, n\u0026thinsp;=\u0026thinsp;17 females) with DTI (15 directions, b- values\u0026thinsp;=\u0026thinsp;400 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e/s \u0026amp; 1800 \u0026micro;m\u003csup\u003e2\u003c/sup\u003e/s, field of view\u0026thinsp;=\u0026thinsp;23mm, slice thickness\u0026thinsp;=\u0026thinsp;1 mm, 5 slices) and T2*WI with the same conditions of the first batch. In both cases, brain slices were positioned to contain the hypothalamus, thalamus and cortex on the center acquisition slice, while the frontal hippocampus remained located in a contiguous slice. Regions were identified with the help of anatomical atlas.\u003csup\u003e60\u003c/sup\u003e For each mouse, three DTI and T\u003csub\u003e2\u003c/sub\u003e*WI blocks were acquired, at basal (t\u0026thinsp;=\u0026thinsp;t0), 20 min after TGN (or saline) administration and immediately after the glucose insult (t\u0026thinsp;=\u0026thinsp;t1), and 30 min after the glucose administration (t\u0026thinsp;=\u0026thinsp;t2) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4. Image processing\u003c/h2\u003e \u003cp\u003eDTI and T\u003csub\u003e2\u003c/sub\u003e*WI of each mouse were processed using Resomapper, a home-made software based on Dipy\u003csup\u003e61\u003c/sup\u003e. Pre-processing algorithms included a noise reduction filter (Patch2self) for DTI and the \u003cem\u003eadaptive soft coefficient matching\u003c/em\u003e filter for T\u003csub\u003e2\u003c/sub\u003e*WI, both used to improve image quality and selected based on previous literature\u003csup\u003e62\u003c/sup\u003e. Processing of the DTI images lead to the obtention of MD, AD, RD and FA parametric maps, and by processing the T\u003csub\u003e2\u003c/sub\u003e*WI, T\u003csub\u003e2\u003c/sub\u003e* maps were obtained. Next, ImageJ software (U. S. National Institutes of Health, Bethesda, Maryland, USA, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://imagej.nih.gov/ij/\u003c/span\u003e\u003cspan address=\"https://imagej.nih.gov/ij/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) was used to manually select the regions of interest (ROIs), including: the hypothalamus (120 voxels), thalamus (140 voxels), cortex (144 voxels) and the hippocampus (110 voxels)(Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). ROI selection was performed blindly, and corresponding pixel coordinates of each subregion were saved and stored as .\u003cem\u003etxt\u003c/em\u003e files.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e5. Data filtering and preparation\u003c/h2\u003e \u003cp\u003e \u003cb\u003eMatlab\u003c/b\u003e software (R2010b, MathWorks Inc., Natick, MA) was used to overlay all ROIs on all images of each mouse. Briefly, by automatically reading the coordinates from the Image-J derived .\u003cem\u003etxt\u003c/em\u003e files, the information was inferred to use the same coordinates on all diffusion and T\u003csub\u003e2\u003c/sub\u003e* maps. Datasheet documents were then generated to include the information on MD, AD, RD, FA and T\u003csub\u003e2\u003c/sub\u003e* values for each pixel. The information regarding the corresponding subregion, time-point of acquisition (t0, t1 or t2), animal identification, type of diet, sex and type of treatment of each pixel was added in separate columns, resulting in a unique datasheet containing all the MRI information at a pixel level.\u003c/p\u003e \u003cp\u003eUsing home-made R-scripts, data filtering and restructuring were performed. From the pixel datasheet of MRI generated by Matlab, pixel values were filtered to exclude the potential contamination of the CSF from ventricles, by removing those pixels with MD, AD and RD values\u0026thinsp;\u0026gt;\u0026thinsp;1300 mm\u003csup\u003e2\u003c/sup\u003e/s or T\u003csub\u003e2\u003c/sub\u003e* values\u0026thinsp;\u0026gt;\u0026thinsp;30 ms, as previously described\u003csup\u003e63\u003c/sup\u003e. Additionally, pixel outliers from each animal subregion (mean +/- 1.5 - interquartile range) were eliminated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e6. Principal Component Analysis (PCA)\u003c/h2\u003e \u003cp\u003eAiming for dimensionality reduction, MRI variables were transformed via PCA (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). PCA is a mathematical algorithm that reduces the dimensionality of the data while retaining most of the variance of the data. New variables are obtained, called principal components (which are a linear combination of the original variables), whose number is equal to or less than the initial number of variables, which explain most of the variance of the data and are independent of each other\u003csup\u003e52\u003c/sup\u003e. In this study, the number of retained PCA variables was chosen to explain at least\u0026thinsp;\u0026ge;\u0026thinsp;80% of the variance.\u003c/p\u003e \u003cp\u003eTwo main PCA approaches were performed, using the \u003cb\u003eprcomp\u003c/b\u003e function of R (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.rdocumentation.org/packages\u003c/span\u003e\u003cspan address=\"https://www.rdocumentation.org/packages\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), including the scaling and centering procedures to have unit variance and a shift to zero, respectively, before analysis. In the first assessment, the \u0026ldquo;\u003cem\u003e1.HFD, sex and TGN-20 effects on T2* and DTI at separated time points\u003c/em\u003e\u003cb\u003e\u0026rdquo;\u003c/b\u003e, a PCA for each time (t0, t1 and t2) was performed independently. In the second approach, the \u0026ldquo;\u003cem\u003e2. T\u003c/em\u003e\u003csub\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sub\u003e\u003cem\u003e* and DTI follow up TGN-20 administration to HFD or CTRL mice\u003c/em\u003e\u0026rdquo;, a multiple factor analysis (MFA) was calculated. MFA is an extension of PCA that is used to handle data where the same variables (or \u0026ldquo;data tables\u0026rdquo;) are measured on different time-sets of observations\u003csup\u003e64\u003c/sup\u003e. Briefly, the procedure is achieved in two steps. First, a PCA of each data-table (or time-observation) is performed. Secondly, all the data tables are concatenated and a generalized PCA is done. With this approach, it is possible to study the time regimes that contribute most to each component derived from the PCA and report the influence of the variables from each time. To do that, we generated the Gl\u003csub\u003eshort\u003c/sub\u003e-basal, Gl\u003csub\u003elong\u003c/sub\u003e-Gl\u003csub\u003eshort\u003c/sub\u003e and Gl\u003csub\u003elong\u0026minus;\u003c/sub\u003ebasal groups of data by subtracting the corresponding values of MD, AD, RD, FA and T\u003csub\u003e2\u003c/sub\u003e* between the different time-points. MFA was performed using the mfa function of the FactoMiner package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.rdocumentation.org/packages/FactoMineR/versions/2.9/topics/MFA\u003c/span\u003e\u003cspan address=\"https://www.rdocumentation.org/packages/FactoMineR/versions/2.9/topics/MFA\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Two additional MFA were performed, either in the region of the hippocampus alone, or using the three remaining regions. In the MFA analysis, those animals that did not show valid DTI images for all time points, could not be included, and the corresponding groups were i) CTRL fed and \u0026ldquo;saline\u0026thinsp;+\u0026thinsp;glc\u0026rdquo; (5 males and 5 females), ii) CTRL and \u0026ldquo;TGN-020\u0026thinsp;+\u0026thinsp;glc\u0026rdquo; (7 males and 5 females), iii) HFD and \u0026ldquo;saline\u0026thinsp;+\u0026thinsp;glc\u0026rdquo; (5 males and 5 females), and iv) HFD and \u0026ldquo;TGN-020\u0026thinsp;+\u0026thinsp;glc\u0026rdquo; (3 males and 3 females).\u003c/p\u003e \u003cp\u003eIn all cases, once the relevant PCA or MFA and predictors were identified, and to better understand the biological basis of the findings, we further investigated how the PCA-composing MRI variables were related to such predictors, by graphical representations and exploratory statistical tests (corrected for multiple comparisons).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003e7. Statistical tests\u003c/h2\u003e \u003cp\u003eDifferences in BW were assessed by anova tests, with diet, sex and treatment as main effects, and their corresponding interactions, using the Anova function of the car package in R. The effects of diet, type of administration (\u0026ldquo;treatment\u0026rdquo;), brain region and sex, on the relevant PCA variables was perfomed by linear mixed-effect (lme) fitting\u003csup\u003e63\u003c/sup\u003e, using the lme function of the nlme package\u003csup\u003e65\u003c/sup\u003e, and mouse as a random intercept. An autoregressive correlation between regions was established (AR1 function of the nlme), in which the regions that are closer present more correlation. These lme models included fixed effects such as \u003cem\u003eregion\u003c/em\u003e, \u003cem\u003ediet\u003c/em\u003e, \u003cem\u003esex\u003c/em\u003e and \u003cem\u003etreatment\u003c/em\u003e and mouse as a random term. Interactions between different variables were added to explore possible joint effects, including double interactions (\u003cem\u003eDiet:Region, Diet:Sex, Treatment:Region, Treatment:Diet, Treatment:Sex\u003c/em\u003e and \u003cem\u003eSex:Region\u003c/em\u003e), triple interactions (\u003cem\u003eDiet:Region:Sex, Diet:Region:Treatment, Region:Sex:Treatment\u003c/em\u003e and \u003cem\u003eSex:Treatment:Diet\u003c/em\u003e) and the quadruple interaction (\u003cem\u003eSex:Treatment:Diet:Region\u003c/em\u003e). Next, Anova tests were performed to assess the statistical significance of the main effects and interactions on each PCA. For significant interactions, post-hoc contrasts were performed using the \u003cb\u003eemmeans\u003c/b\u003e function (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/rvlenth/emmeans\u003c/span\u003e\u003cspan address=\"https://github.com/rvlenth/emmeans\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), and corresponding p-values adjusted for multiple comparisons with the Bonferroni method.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are deeply indebted to Mrs. Mar\u0026iacute;a Jos\u0026eacute; Guill\u0026eacute;n (CSIC) for excellent technical assistance with animal handling, to Mar\u0026iacute;a Rodr\u0026iacute;guez and Mrs. Teresa Navarro (CSIC) and the IIBM Sebastian Cerd\u0026aacute;n NMR Facility for expert support during MRI acquisitions and granting rapid access to the MR instrumentation.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eFunding.\u003c/strong\u003e This work was supported by grants PID2021-122528OB-I00 to PL-L, grant PID2021-126888OA-I00 to B.L., scholarship PRE2022-105662 to A.F. and scholarship PIPF-2022/SAL-GL-25871 to R.G-A.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eCONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization, P.L.-L. and B.L.; Methodology, N.A.-R., P.L.-L. and B.L.; Investigation,\u003c/p\u003e\n\u003cp\u003eP.T.-G., R.G.-A. and N.A.-R..; Software, A.F., R.G.-A. and B.L., Formal Analysis, P.T.-G, A.F., and B.L., Writing \u0026ndash; Original Draft, P.A.T.-G., and B.L.; Writing \u0026ndash; Review \u0026amp; Editing, P.L.-L. and B.L.; Visualization, P.A.T.-G. and B.L., Funding Acquisition, P.L.-L. and B.L., Project Administration, P.L.-L. and B.L., Supervision, N.A-R., P.L-L and B.L.\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eEthic statements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll animal handling protocols were performed by specialized personnel at our institute\u0026rsquo;s animal facility (Reg. No. ES280790000188) and approved by the ethical committee of the Institute of Biomedical Research Sols-Morreale, CSIC and the Community of Madrid, complying with national (R.D. 53/2013) and European Community guidelines (2010/63/UE). Ethics approval number PROEX 288/42 (Comunidad de Madrid, direcci\u0026oacute;n general de agricultura, ganader\u0026iacute;a y alimentaci\u0026oacute;n). \u0026nbsp;All methods were carried out in accordance with relevant guidelines and regulations. All methods are reported in accordance with ARRIVE guidelines (https://arriveguidelines.org).\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data presented in this study is available on request from the corresponding author due to the need for a formal data sharing agreement.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMergenthaler, P., Lindauer, U., Dienel, G. A. \u0026amp; Meisel, A. Sugar for the brain: the role of glucose in physiological and pathological brain function. \u003cem\u003eTrends Neurosci\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 587\u0026ndash;597 (2013).\u003c/li\u003e\n\u003cli\u003eTakahashi, S. Metabolic Contribution and Cerebral Blood Flow Regulation by Astrocytes in the Neurovascular Unit. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eHwang, J. J. \u003cem\u003eet al.\u003c/em\u003e Blunted rise in brain glucose levels during hyperglycemia in adults with obesity and T2DM. \u003cem\u003eJCI Insight\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eHuber, G. \u003cem\u003eet al.\u003c/em\u003e Telmisartan prevents high-fat diet-induced neurovascular impairments and reduces anxiety-like behavior. \u003cem\u003eJ Cereb Blood Flow Metab\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 2356\u0026ndash;2369 (2021).\u003c/li\u003e\n\u003cli\u003ede Paula, G. C. \u003cem\u003eet al.\u003c/em\u003e Hippocampal Function Is Impaired by a Short-Term High-Fat Diet in Mice: Increased Blood-Brain Barrier Permeability and Neuroinflammation as Triggering Events. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 734158 (2021).\u003c/li\u003e\n\u003cli\u003eSa, M., Park, M. G. \u0026amp; Lee, C. J. Role of Hypothalamic Reactive Astrocytes in Diet-Induced Obesity. \u003cem\u003eMol Cells\u003c/em\u003e \u003cstrong\u003e45\u003c/strong\u003e, 65\u0026ndash;75 (2022).\u003c/li\u003e\n\u003cli\u003eValdearcos, M. \u003cem\u003eet al.\u003c/em\u003e Microglial Inflammatory Signaling Orchestrates the Hypothalamic Immune Response to Dietary Excess and Mediates Obesity Susceptibility. \u003cem\u003eCell Metab\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1356 (2018).\u003c/li\u003e\n\u003cli\u003eKoepsell, H. Glucose transporters in brain in health and disease. \u003cem\u003ePflugers Arch\u003c/em\u003e \u003cstrong\u003e472\u003c/strong\u003e, 1299\u0026ndash;1343 (2020).\u003c/li\u003e\n\u003cli\u003eNagelhus, E. A. \u0026amp; Ottersen, O. P. Physiological roles of aquaporin-4 in brain. \u003cem\u003ePhysiol Rev\u003c/em\u003e \u003cstrong\u003e93\u003c/strong\u003e, 1543\u0026ndash;1562 (2013).\u003c/li\u003e\n\u003cli\u003eSun, C. \u003cem\u003eet al.\u003c/em\u003e Acutely Inhibiting AQP4 With TGN-020 Improves Functional Outcome by Attenuating Edema and Peri-Infarct Astrogliosis After Cerebral Ischemia. \u003cem\u003eFront Immunol\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 870029 (2022).\u003c/li\u003e\n\u003cli\u003eMader, S. \u0026amp; Brimberg, L. Aquaporin-4 Water Channel in the Brain and Its Implication for Health and Disease. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, (2019).\u003c/li\u003e\n\u003cli\u003eMicioni Di Bonaventura, M. V. \u003cem\u003eet al.\u003c/em\u003e Brain alterations in high fat diet induced obesity: effects of tart cherry seeds and juice. \u003cem\u003eNutrients\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eTomassoni, D. \u003cem\u003eet al.\u003c/em\u003e Obesity and Age-Related Changes in the Brain of the Zucker Lepr (fa/fa) Rats. \u003cem\u003eNutrients\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eFu, J.-T. \u003cem\u003eet al.\u003c/em\u003e Exploring the reduction in aquaporin-4 and increased expression of ciliary neurotrophic factor with the frontal\u0026ndash;striatal gliosis induced by chronic high-fat dietary stress. \u003cem\u003eJournal of Neurochemistry\u003c/em\u003e \u003cstrong\u003en/a\u003c/strong\u003e,.\u003c/li\u003e\n\u003cli\u003eHertz, L., Peng, L. \u0026amp; Dienel, G. A. Energy metabolism in astrocytes: high rate of oxidative metabolism and spatiotemporal dependence on glycolysis/glycogenolysis. \u003cem\u003eJ Cereb Blood Flow Metab\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 219\u0026ndash;249 (2007).\u003c/li\u003e\n\u003cli\u003eBak, L. K., Schousboe, A. \u0026amp; Waagepetersen, H. S. The glutamate/GABA-glutamine cycle: aspects of transport, neurotransmitter homeostasis and ammonia transfer. \u003cem\u003eJ Neurochem\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e, 641\u0026ndash;653 (2006).\u003c/li\u003e\n\u003cli\u003eReinhard, L., Tidow, H., Clausen, M. J. \u0026amp; Nissen, P. Na(+),K (+)-ATPase as a docking station: protein-protein complexes of the Na(+),K (+)-ATPase. \u003cem\u003eCell Mol Life Sci\u003c/em\u003e \u003cstrong\u003e70\u003c/strong\u003e, 205\u0026ndash;222 (2013).\u003c/li\u003e\n\u003cli\u003eBurdakov, D., Luckman, S. M. \u0026amp; Verkhratsky, A. Glucose-sensing neurons of the hypothalamus. \u003cem\u003ePhilos Trans R Soc Lond B Biol Sci\u003c/em\u003e \u003cstrong\u003e360\u003c/strong\u003e, 2227\u0026ndash;2235 (2005).\u003c/li\u003e\n\u003cli\u003eTu, L., Fukuda, M., Tong, Q. \u0026amp; Xu, Y. The ventromedial hypothalamic nucleus: watchdog of whole-body glucose homeostasis. \u003cem\u003eCell \u0026amp; Bioscience\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 71 (2022).\u003c/li\u003e\n\u003cli\u003eAttwell, D. \u003cem\u003eet al.\u003c/em\u003e Glial and neuronal control of brain blood flow. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e468\u003c/strong\u003e, 232\u0026ndash;243 (2010).\u003c/li\u003e\n\u003cli\u003eKoekkoek, L. L., Mul, J. D. \u0026amp; la Fleur, S. E. Glucose-Sensing in the Reward System. \u003cem\u003eFrontiers in Neuroscience\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eYousaf, T., Dervenoulas, G. \u0026amp; Politis, M. Advances in MRI Methodology. \u003cem\u003eInt Rev Neurobiol\u003c/em\u003e \u003cstrong\u003e141\u003c/strong\u003e, 31\u0026ndash;76 (2018).\u003c/li\u003e\n\u003cli\u003eŽivanović, M. \u003cem\u003eet al.\u003c/em\u003e The role of magnetic resonance imaging in the diagnosis and prognosis of dementia. \u003cem\u003eBiomol Biomed\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 209\u0026ndash;224 (2023).\u003c/li\u003e\n\u003cli\u003eLe Bihan, D. \u003cem\u003eet al.\u003c/em\u003e Diffusion tensor imaging: concepts and applications. \u003cem\u003eJ Magn Reson Imaging\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 534\u0026ndash;546 (2001).\u003c/li\u003e\n\u003cli\u003eLope-Piedrafita, S. Chapter 7: Diffusion Tensor Imaging (DTI). in \u003cem\u003ePreclinical MRI: Methods and Protocols\u003c/em\u003e 103\u0026ndash;116 (2018).\u003c/li\u003e\n\u003cli\u003eLizarbe, B., Campillo, B., Guadilla, I., L\u0026oacute;pez-Larrubia, P. \u0026amp; Cerd\u0026aacute;n, S. Magnetic resonance assessment of the cerebral alterations associated with obesity development. \u003cem\u003eJ Cereb Blood Flow Metab\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 2135\u0026ndash;2151 (2020).\u003c/li\u003e\n\u003cli\u003eLizarbe, B. \u003cem\u003eet al.\u003c/em\u003e Systemic Glucose Administration Alters Water Diffusion and Microvascular Blood Flow in Mouse Hypothalamic Nuclei - An fMRI Study. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 921 (2019).\u003c/li\u003e\n\u003cli\u003eLizarbe, B. \u003cem\u003eet al.\u003c/em\u003e Imaging hypothalamic activity using diffusion weighted magnetic resonance imaging in the mouse and human brain. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 448\u0026ndash;457 (2013).\u003c/li\u003e\n\u003cli\u003eNunes, D., Gil, R. \u0026amp; Shemesh, N. A rapid-onset diffusion functional MRI signal reflects neuromorphological coupling dynamics. \u003cem\u003eNeuroImage\u003c/em\u003e \u003cstrong\u003e231\u003c/strong\u003e, 117862 (2021).\u003c/li\u003e\n\u003cli\u003eRoth, B. J. Can MRI Be Used as a Sensor to Record Neural Activity? \u003cem\u003eSensors (Basel)\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eTang, M. Y., Chen, T. W., Zhang, X. M. \u0026amp; Huang, X. H. GRE T2\u0026lowast;-weighted MRI: principles and clinical applications. \u003cem\u003eBiomed Res Int\u003c/em\u003e \u003cstrong\u003e2014\u003c/strong\u003e, 312142 (2014).\u003c/li\u003e\n\u003cli\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e Perfusion functional MRI reveals cerebral blood flow pattern under psychological stress. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 17804\u0026ndash;17809 (2005).\u003c/li\u003e\n\u003cli\u003eOgawa, S., Lee, T. M., Kay, A. R. \u0026amp; Tank, D. W. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e87\u003c/strong\u003e, 9868\u0026ndash;9872 (1990).\u003c/li\u003e\n\u003cli\u003eChavhan, G. B., Babyn, P. S., Thomas, B., Shroff, M. M. \u0026amp; Haacke, E. M. Principles, techniques, and applications of T2*-based MR imaging and its special applications. \u003cem\u003eRadiographics\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 1433\u0026ndash;1449 (2009).\u003c/li\u003e\n\u003cli\u003eLizarbe, B. \u003cem\u003eet al.\u003c/em\u003e Systemic Glucose Administration Alters Water Diffusion and Microvascular Blood Flow in Mouse Hypothalamic Nuclei - An fMRI Study. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 921 (2019).\u003c/li\u003e\n\u003cli\u003eIgarashi, H., Huber, V. J., Tsujita, M. \u0026amp; Nakada, T. Pretreatment with a novel aquaporin 4 inhibitor, TGN-020, significantly reduces ischemic cerebral edema. \u003cem\u003eNeurol Sci\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 113\u0026ndash;116 (2011).\u003c/li\u003e\n\u003cli\u003eTaquet, M. \u003cem\u003eet al.\u003c/em\u003e Extra-axonal restricted diffusion as an in-vivo marker of reactive microglia. \u003cem\u003eSci Rep\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 13874 (2019).\u003c/li\u003e\n\u003cli\u003eRouth, V. H., McArdle, J. J., Sanders, N. M., Song, Z. \u0026amp; Wang, R. Glucose Sensing Neurons. in \u003cem\u003eHandbook of Neurochemistry and Molecular Neurobiology: Sensory Neurochemistry\u003c/em\u003e (eds. Lajtha, A. \u0026amp; Johnson, D. A.) 205\u0026ndash;228 (Springer US, New York, NY, 2007). doi:10.1007/978-0-387-30374-1_7.\u003c/li\u003e\n\u003cli\u003eVerkman, A. S., Smith, A. J., Phuan, P.-W., Tradtrantip, L. \u0026amp; Anderson, M. O. The aquaporin-4 water channel as a potential drug target in neurological disorders. \u003cem\u003eExpert Opin Ther Targets\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 1161\u0026ndash;1170 (2017).\u003c/li\u003e\n\u003cli\u003eAnderson, M. A., Ao, Y. \u0026amp; Sofroniew, M. V. Heterogeneity of reactive astrocytes. \u003cem\u003eNeuroscience Letters\u003c/em\u003e \u003cstrong\u003e565\u003c/strong\u003e, 23\u0026ndash;29 (2014).\u003c/li\u003e\n\u003cli\u003eBerkseth, K. E. \u003cem\u003eet al.\u003c/em\u003e Hypothalamic gliosis associated with high-fat diet feeding is reversible in mice: a combined immunohistochemical and magnetic resonance imaging study. \u003cem\u003eEndocrinology\u003c/em\u003e \u003cstrong\u003e155\u003c/strong\u003e, 2858\u0026ndash;2867 (2014).\u003c/li\u003e\n\u003cli\u003eMetwalli, N. S. \u003cem\u003eet al.\u003c/em\u003e Utility of axial and radial diffusivity from diffusion tensor MRI as markers of neurodegeneration in amyotrophic lateral sclerosis. \u003cem\u003eBrain Res\u003c/em\u003e \u003cstrong\u003e1348\u003c/strong\u003e, 156\u0026ndash;164 (2010).\u003c/li\u003e\n\u003cli\u003eMessina, R. \u003cem\u003eet al.\u003c/em\u003e White matter microstructure abnormalities in pediatric migraine patients. \u003cem\u003eCephalalgia\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 1278\u0026ndash;1286 (2015).\u003c/li\u003e\n\u003cli\u003eOosuka, S. \u003cem\u003eet al.\u003c/em\u003e Effects of an Aquaporin 4 Inhibitor, TGN-020, on Murine Diabetic Retina. \u003cem\u003eInt J Mol Sci\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eKitaura, H. \u003cem\u003eet al.\u003c/em\u003e Activity-dependent glial swelling is impaired in aquaporin-4 knockout mice. \u003cem\u003eNeuroscience Research\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 208\u0026ndash;212 (2009).\u003c/li\u003e\n\u003cli\u003eHermanova, Z. \u003cem\u003eet al.\u003c/em\u003e The deletion of AQP4 and TRPV4 affects astrocyte swelling/volume recovery in response to ischemia-mimicking pathologies. \u003cem\u003eFrontiers in Cellular Neuroscience\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eAmen, D. G., Wu, J., George, N. \u0026amp; Newberg, A. Patterns of Regional Cerebral Blood Flow as a Function of Obesity in Adults. \u003cem\u003eJournal of Alzheimer\u0026rsquo;s Disease\u003c/em\u003e \u003cstrong\u003e77\u003c/strong\u003e, 1331\u0026ndash;1337 (2020).\u003c/li\u003e\n\u003cli\u003eTucsek, Z. \u003cem\u003eet al.\u003c/em\u003e Aging exacerbates obesity-induced cerebromicrovascular rarefaction, neurovascular uncoupling, and cognitive decline in mice. \u003cem\u003eJ Gerontol A Biol Sci Med Sci\u003c/em\u003e \u003cstrong\u003e69\u003c/strong\u003e, 1339\u0026ndash;1352 (2014).\u003c/li\u003e\n\u003cli\u003eTomassoni, D. \u003cem\u003eet al.\u003c/em\u003e Obesity and Age-Related Changes in the Brain of the Zucker Lepr (fa/fa) Rats. \u003cem\u003eNutrients\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003ePettersson, U. S., Wald\u0026eacute;n, T. B., Carlsson, P.-O., Jansson, L. \u0026amp; Phillipson, M. Female Mice are Protected against High-Fat Diet Induced Metabolic Syndrome and Increase the Regulatory T Cell Population in Adipose Tissue. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 1\u0026ndash;10 (2012).\u003c/li\u003e\n\u003cli\u003eCampillo, B. W., Galguera, D., Cerdan, S., L\u0026oacute;pez-Larrubia, P. \u0026amp; Lizarbe, B. Short-term high-fat diet alters the mouse brain magnetic resonance imaging parameters consistently with neuroinflammation on males and metabolic rearrangements on females. A pre-clinical study with an optimized selection of linear mixed-effects models. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 1025108 (2022).\u003c/li\u003e\n\u003cli\u003eRingn\u0026eacute;r, M. What is principal component analysis? \u003cem\u003eNat Biotechnol\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 303\u0026ndash;304 (2008).\u003c/li\u003e\n\u003cli\u003eTaquet, M. \u003cem\u003eet al.\u003c/em\u003e Extra-axonal restricted diffusion as an in-vivo marker of reactive microglia. \u003cem\u003eScientific Reports\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 13874 (2019).\u003c/li\u003e\n\u003cli\u003eAbdi, H. \u0026amp; Valentin, D. Multiple Factor Analysis (MFA). \u003cem\u003eEncyclopedia of Measurement and Statistics\u003c/em\u003e (2007).\u003c/li\u003e\n\u003cli\u003eHuber, V. J., Tsujita, M. \u0026amp; Nakada, T. Identification of Aquaporin 4 inhibitors using in vitro and in silico methods. \u003cem\u003eBioorganic \u0026amp; Medicinal Chemistry\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 411\u0026ndash;417 (2009).\u003c/li\u003e\n\u003cli\u003eDebacker, C., Djemai, B., Ciobanu, L., Tsurugizawa, T. \u0026amp; Le Bihan, D. Diffusion MRI reveals in vivo and non-invasively changes in astrocyte function induced by an aquaporin-4 inhibitor. \u003cem\u003ePLoS One\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e0229702 (2020).\u003c/li\u003e\n\u003cli\u003eSun, C. \u003cem\u003eet al.\u003c/em\u003e Acutely Inhibiting AQP4 With TGN-020 Improves Functional Outcome by Attenuating Edema and Peri-Infarct Astrogliosis After Cerebral Ischemia. \u003cem\u003eFrontiers in Immunology\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eSalman, M. M., Kitchen, P., Yool, A. J. \u0026amp; Bill, R. M. Recent breakthroughs and future directions in drugging aquaporins. \u003cem\u003eTrends in Pharmacological Sciences\u003c/em\u003e \u003cstrong\u003e43\u003c/strong\u003e, 30\u0026ndash;42 (2022).\u003c/li\u003e\n\u003cli\u003eDebaker, C., Djemai, B., Ciobanu, L., Tsurugizawa, T. \u0026amp; Le Bihan, D. Diffusion MRI reveals in vivo and non-invasively changes in astrocyte function induced by an aquaporin-4 inhibitor. \u003cem\u003ePLOS ONE\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e0229702 (2020).\u003c/li\u003e\n\u003cli\u003eAllen Reference Atlas \u0026ndash; Mouse Brain [brain atlas].\u003c/li\u003e\n\u003cli\u003eGaryfallidis, E. \u003cem\u003eet al.\u003c/em\u003e Dipy, a library for the analysis of diffusion MRI data. \u003cem\u003eFront Neuroinform\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 8 (2014).\u003c/li\u003e\n\u003cli\u003eSchilling, K. G. \u003cem\u003eet al.\u003c/em\u003e Denoising of diffusion MRI in the cervical spinal cord - effects of denoising strategy and acquisition on intra-cord contrast, signal modeling, and feature conspicuity. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cstrong\u003e266\u003c/strong\u003e, 119826 (2023).\u003c/li\u003e\n\u003cli\u003eCampillo, B. W., Galguera, D., Cerdan, S., L\u0026oacute;pez-Larrubia, P. \u0026amp; Lizarbe, B. Short-term high-fat diet alters the mouse brain magnetic resonance imaging parameters consistently with neuroinflammation on males and metabolic rearrangements on females. A pre-clinical study with an optimized selection of linear mixed-effects models. \u003cem\u003eFront Neurosci\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 1025108 (2022).\u003c/li\u003e\n\u003cli\u003eAbdi, H., Williams, L. J. \u0026amp; Valentin, D. Multiple factor analysis: principal component analysis for multitable and multiblock data sets. \u003cem\u003eWIREs Computational Stats\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 149\u0026ndash;179 (2013).\u003c/li\u003e\n\u003cli\u003eJos\u0026eacute; Pinheiro, Douglas Bates and R Core Team. nlme: Linear and Nonlinear Mixed Effects Models. (2022).\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Aquaporin-4, Mouse, Blood flow, Glucose, Obesity, Swelling, TGN-020, Magnetic Resonance Imaging, DTI, T2*","lastPublishedDoi":"10.21203/rs.3.rs-5931441/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5931441/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGlucose uptake and metabolism are linked to microvascular blood flow and cellular swelling events, which are altered during obesity and can be quantified using magnetic resonance imaging (MRI). Aquaporin-4 (AQP4), the most abundant water-transporting transmembrane protein in the central nervous system, facilitates glucose transport and metabolism-derived water influx. However, its significance and regulatory capacity remain largely unknown. To better understand these processes, we acquired sequential diffusion tensor and T2*-weighted images of the brains of obese and non-obese mice, both before administering an AQP4 inhibitor and after a subsequent glucose challenge. We then subjected the resulting variables to principal component and linear mixed model analyses to assess the influence of diet, sex, administration of the inhibitor, and brain region on the data. Our findings indicate that AQP4-inhibited mice exhibit MRI values consistent with reduced microvascular blood flow and region-specific inhibition of glucose-induced cell swelling during obesity, highlighting a key role for AQP4 in glucose uptake and metabolism. Additionally, we observed that, prior to any experimental manipulation, obese mice displayed MRI signs of higher cortical blood flow and cerebral cellular anisotropy compared to controls, in agreement with vascular alterations and reactive gliosis processes.\u003c/p\u003e","manuscriptTitle":"Aquaporin-4 inhibition alters cerebral glucose dynamics predominantly in obese animals: an MRI study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-15 11:47:41","doi":"10.21203/rs.3.rs-5931441/v1","editorialEvents":[{"type":"communityComments","content":1},{"type":"decision","content":"Accepted","date":"2025-04-22T05:05:48+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-22T04:12:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-16T07:24:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"97168531100540967448185287162022035446","date":"2025-04-15T07:54:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-13T12:31:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"305462968859510682978357224321735659288","date":"2025-04-11T22:07:13+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"122017825481633073090818508141502404186","date":"2025-04-11T21:10:15+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-11T20:05:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-11T04:09:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-04-01T20:09:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b4c340a0-4a32-41f4-8efd-ead90a5fe15c","owner":[],"postedDate":"April 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":47085321,"name":"Biological sciences/Biophysics"},{"id":47085322,"name":"Biological sciences/Neuroscience"},{"id":47085323,"name":"Biological sciences/Biological techniques/Imaging"},{"id":47085324,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":47085325,"name":"Biological sciences/Computational biology and bioinformatics/Data processing"},{"id":47085326,"name":"Biological sciences/Biochemistry"},{"id":47085327,"name":"Biological sciences/Biochemistry/Neurochemistry"}],"tags":[],"updatedAt":"2025-05-12T16:00:37+00:00","versionOfRecord":{"articleIdentity":"rs-5931441","link":"https://doi.org/10.1038/s41598-025-99641-1","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-05-05 15:57:19","publishedOnDateReadable":"May 5th, 2025"},"versionCreatedAt":"2025-04-15 11:47:41","video":"","vorDoi":"10.1038/s41598-025-99641-1","vorDoiUrl":"https://doi.org/10.1038/s41598-025-99641-1","workflowStages":[]},"version":"v1","identity":"rs-5931441","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5931441","identity":"rs-5931441","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}