The efficacy of low doses of STZ in modeling T2D in WD-fed mice is not well understood, thus we aimed to compare the glucose phenotype of WD and WD + STZ mice. The mice were fed with WD for 26 weeks and received two injections of STZ (75 and 50 mg/kg body weight) or vehicle 3 days apart after 7 weeks of WD feeding (Figure 1A). After 26 weeks of WD feeding, WD + STZ mice shown significantly higher glucose levels during the glucose tolerance test (GTT) than WD- and NC-fed mice (Figures 1B, C). Throughout the experiment, WD + STZ mice consistently displayed higher fasting blood glucose levels after STZ injection compared to NC-fed mice, which was sustained until the end of the study (Figure 1D). As expected, WD-fed mice also demonstrated higher fasting blood glucose levels compared to NC-fed mice, as previously reported (Figure 1D) (Gavini et al., 2018; Elshareif et al., 2022). At week 14, WD + STZ mice had higher fasting glucose levels than WD-fed mice, and by week 26, fasting glucose levels were trending higher than those in WD-fed mice (Figure 1D).

To investigate if the observed differences between the WD and WD + STZ groups were associated with changes in circulating insulin levels, serum insulin was measured after a 4-h fast. After 26 weeks, WD + STZ mice shown significantly lower fasting serum insulin levels compared to WD-fed mice, with a level of 2.8 ng/mL (Figure 1E). This level was comparable to the insulin levels in the NC group at 2.3 ng/mL, while WD-fed mice displayed the highest circulating concentration of 5.5 ng/mL post-fast (Figure 1E). Furthermore, during the glucose challenge at week 26, where tail blood samples were obtained 30 min post-glucose injection, WD + STZ mice demonstrated lower serum insulin levels compared to NC and WD-fed mice (Figure 1F). This suggested that the two paradigms model different stages of T2D, with WD representing prediabetes with pronounced peripheral insulin resistance and WD + STZ representing later stages of T2D with the additional characteristic of impaired insulin secretion. This hypothesis was confirmed using calculated Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) scores. The WD-fed mice had the highest average index score of 60, significantly higher than the average NC score of 25 (Figure 1G). The WD + STZ mice had an average index score of 45, indicating an intermediate value between WD-fed and NC-fed mice (Figure 1G). Our findings indicate that WD and WD + STZ mice exhibit peripheral insulin resistance based on the HOMA-IR scores compared to NC group.

To further confirm this observation, we conducted in vivo testing to assess the response to insulin in muscle in the control, WD, and WD + STZ groups. Insulin signaling was evaluated by immunoblotting for the Ser473 residue of phosphorylated protein kinase B (pAKT) and total AKT in the soleus muscle after insulin injection (0.75 U/kg body weight). Consistent with expectations, NC-fed mice injected with insulin demonstrated an increase in the pAKT/AKT ratio compared to the vehicle (Figures 1H, I). In contrast, WD-fed mice and WD + STZ mice did not show a significant change in the protein ratio of pAKT/AKT when injected with insulin compared to the vehicle (Figures 1H, I). Moreover, there was a significant difference in ratios between insulin-injected WD-fed mice and NC-fed mice (Figures 1H, I), while no significant difference was observed between the NC and WD + STZ insulin groups (Figures 1H, I).

These findings suggest that WD mice exhibit insulin resistance while maintaining normal insulin secretion, a more characteristic model of earlier stages of T2D. On the other hand, the combined use of STZ and WD appears to produce a more complex diabetic phenotype characterized by a combination of impaired insulin secretion and peripheral insulin resistance.

Throughout the paradigm, we performed echocardiograms on the mice to measure indices of cardiac dysfunction between WD-fed mice and WD + STZ mice. At the end of the paradigm, WD-fed mice had higher heart rates, with an average of 505 beats per minute (BPM), compared to the average NC heart rate of 445 BPM (Figure 2A). The WD + STZ mice had trending higher heart rates than the NC-fed mice at an average of 480 BPM (Figure 2A). There was no evident change in ejection fraction between groups (Figure 2B). This was clear throughout the paradigm (Supplementary Figure S1A). However, the average global longitudinal strain (GLS) in WD + STZ mice was −16%, which was significantly reduced compared to NC and WD-fed mice (Figure 2C). This altered GLS was evident at week 10 compared to that in WD-fed mice and at week 14 compared to both WD- and NC-fed mice (Supplementary Figure S1B). No changes in left ventricle posterior wall diameter (LVPWd) were detected in WD and WD + STZ groups compared to NC-fed mice (Supplementary Figure S1C).

Given that both the WD and WD + STZ groups had characteristics of early cardiac dysfunction, we were interested in whether this phenotype was accompanied by changes in cardiac autonomic tone that can manifest in prediabetic and obese humans (Williams et al., 2019). Using the echocardiogram data, we determined whether the WD and WD + STZ groups developed changes in heart rate variability (HRV), a common indicator in cardiac autonomic neuropathy (Yadav et al., 2017; Williams et al., 2019). The mean RR interval, standard deviation of normal RR intervals (SDNN), and root mean square of successive differences between normal heartbeats (RMSSD) were calculated. At the end of the paradigm, the average mean RR interval was 120 and 125 ms for WD and WD + STZ groups, respectively, compared to the NC mean RR interval of 137 ms (Figure 2D). The average SDNN was 0.77 and 0.6 ms for WD and WD + STZ groups respectively, which was significantly lower than the NC SDNN average of 1.71 ms (Figure 2E). This observation was comparable to what we saw with the RMSSD, which was 0.81 ms for WD and 0.87 ms for WD + STZ mice compared to 2.11 ms average from the NC-fed mice (Figure 2F).

We hypothesized that the features of cardiac dysfunction and decreases in HRV indices may be associated with loss of autonomic innervation. To determine whether there was a loss in cardiac sympathetic tone, we stained perfused heart tissue with tyrosine hydroxylase (TH), a rate limiting enzyme for catecholamine synthesis (Figure 2G) (Qin et al., 2021). We quantified the projection length of TH + fibers, where both WD and WD + STZ cardiac tissues had the same decrease in projection length compared to NC (Figure 2H). Immunoblots for cardiac TH levels yielded comparable results, where WD and WD + STZ groups had an average of 10% and 15% lower protein levels, respectively, compared to NC (Figures 2I, J). We also observed a decrease in protein levels for the parasympathetic marker choline acetyltransferase (CHAT), from cardiac tissue of both WD and WD + STZ groups (Figures 2K, L) (Qin et al., 2021). These data suggest that both models display the same characteristics in heart dysfunction potentially due to cardiac autonomic neuropathy.

In the context of diabetic neuropathy-associated somatosensory dysfunction, we conducted a longitudinal comparison of mechanical and thermal sensitivity behaviors in mice subjected to a WD or a combination of WD and STZ (WD + STZ). Conversely, both WD and WD + STZ mice consistently exhibited lower mechanical allodynia thresholds compared to NC mice throughout the entire paradigm (Supplementary Figure S2A). At weeks 10 and 14 of the paradigm, both WD and WD + STZ mice exhibited lower thermal latency than NC-fed mice (Supplementary Figure S2B). However, by the end of the paradigm, neither the WD nor WD + STZ mice demonstrated significant thermal hyperalgesia.

To investigate whether this observed hypersensitivity phenotype was accompanied by neuronal injury in the dorsal root ganglia (DRG), potentially resulting from the effects of the WD and STZ, we performed quantitative polymerase chain reaction (qPCR) targeting the activating transcription factor 3 (Atf3) mRNA levels in the lumbar DRG, as previously described (Nascimento et al., 2011). Remarkably, we found that both WD and WD + STZ mice exhibited a similar increase in Atf3 mRNA levels, ranging between 20% and 30% higher than those observed in NC DRG (Supplementary Figure S2C). This data shows that STZ treatment did not worsen the neuropathy condition.

We sought to compare further the metabolic phenotype of WD-fed mice together with WD-fed mice treated with STZ. After 5 weeks of feeding, WD and WD + STZ mice were significantly heavier than their NC counterparts (Figure 3A). The mass of WD and WD + STZ mice comprised approximately 40% fat mass (Figure 3B) and 50% lean mass (Figure 3C) compared to the NC-fed mice composed of 20% fat mass (Figure 3B) and 80% lean mass (Figure 3C).

We then sought to determine whether STZ further impaired the metabolic energy expenditure profile of WD-fed mice. At week 21 of the paradigm, mice were subjected to indirect calorimetry for 72 h (Figure 3D). Both the WD and WD + STZ mice displayed elevated rates of energy expenditure during both the day and night, resulting in a 25% increase compared to the NC-fed mice (Figure 3E). To assess the mice’s ability to regulate their body temperature by modulating energy expenditure, we subjected them to a 72-h indirect calorimetry at 24.5°C, followed by a 4-h cold challenge (Figure 3F). While both the WD and WD + STZ mice exhibited higher energy expenditure at ambient temperature compared to the NC group, their response to cold exposure was blunted (Figures 3F, G).

In summary, both the WD and WD + STZ groups maintained similar body composition and energy expenditure profiles at both normal temperatures and during the cold challenge.

During indirect calorimetry experiments, we calculated the respiratory exchange ratio (RER) to determine the substrate type utilized by mice for energy production (Figure 4A). On average, the RER for NC-fed was 0.90, whereas the RER was lower for mice on a WD and WD + STZ at 0.85 and 0.84, respectively (Figure 4B). The RER of NC mice was highest during nighttime, averaging at 0.95, while the WD and WD + STZ mice maintained significantly lower ratios of 0.85 (Figure 4B). During the daytime, all groups exhibited the lowest RER values, with NC-fed mice recording 0.86 (Figure 4B). Both WD and WD + STZ mice had significantly lower RER values than NC, with rates of 0.84 and 0.81, respectively (Figure 4B). These findings indicate that WD and WD + STZ mice primarily utilize fat as a fuel source, whereas NC-fed mice predominantly rely on carbohydrates, especially during nighttime when they are most active. Notably, despite having similar body weights, food intake analysis revealed that the WD group consumed significantly more food than the WD + STZ and NC groups overall, with higher daytime food intake (Figure 4C). Furthermore, both the WD and WD + STZ groups exhibited lower water intake compared to the NC group, particularly during nighttime (Figure 4D).

All studies were conducted in accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and the approval of the Loyola University Chicago Institutional Animal Care and Use Committee. All methods were conducted in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines, with the experimenter blinded to treatments. Male C57BL/6J (#000664) mice were obtained from Jackson Laboratory (Maine, United States). Mice were housed under a 12:12 h light/dark cycle and were 7 weeks old at the beginning of the paradigm. Mice received either NC (Teklad LM-485) or WD (TD88137; 42%kcal from fat, 34% sucrose by weight, and 0.2% cholesterol) for 26 weeks.

Body weights were measured weekly throughout the paradigm. At week 21, 24 h prior to indirect calorimetry, measurements of lean, fluid, and fat mass were performed using Minispec LF50 nuclear magnetic resonance (NMR) analyzer (Bruker Corporation, Billerica, MA).

Mice were fasted in the morning (4 h) and given an i.p. injection of glucose (1 g/kg BW). Fasting glucose measurements were obtained before injection as performed previously (Gavini et al., 2018; Elshareif et al., 2022). Blood glucose levels were measured using an AlphaTrak glucometer (Fisher Scientific, United States) before and after injection at various time points.

Mice were evaluated for mechanical allodynia using von Frey filaments and acclimated in testing chambers 1h before the experiment. 6 calibrated filaments (0.16; 0.4; 1; 2; 4; 6; 8 g) (North Coast Medical, California, United States) were applied for 1 s with 6 stimulations as described previously (Gavini et al., 2018; Elshareif et al., 2022).

Mice were assessed for thermal hyperalgesia using the Hargreaves Method Plantar Test Apparatus (IITC Life Science, California, United States). Mice were acclimated for 1 h in the testing chambers and then stimulated 6 times each for a maximum of 20 s to avoid tissue injury.

Tail blood was collected after 4 h fast and 30 min post-glucose injection (1 g/kg BW) in independent experiments. Samples were spun for 10 min at 2,000 g and serum were collected. Serum insulin levels were quantified using an Ultra-Sensitive Mouse Insulin ELISA Kit (Crystal Chem, Illinois, United States) following the manufacturer’s instructions. HOMA-IR scores were calculated using the following equation: [fasting glucose (mg/dL) * fasting insulin (mU/L)]/405.

Proteins were extracted from frozen hearts and soleus muscle explants. Tissues were submerged in ice-cold RIPA buffer (ThermoFisher, #89900) containing protease and phosphatase inhibitors (ThermoFisher, #A32959). Tissues were powdered and homogenized using a bullet blender bead lysis kit (Next Advance). The samples were spun at 800 g for 2 min at 4°C and supernatant was collected and centrifuged at 12,000 g for 10 min at 4°C. Protein concentrations were determined using a Pierce BCA Protein Assay Kit (ThermoFisher, #23225). Proteins were separated using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on 4%–20% gradient gels (Bio-Rad, #4561094) and then transferred to polyvinylidene fluoride (PVDF) membranes (ThermoFisher, #IB24002) using the iBlot 2 transfer system (ThermoFisher). Protein expression was measured by chemiluminescence using the ChemiDoc imaging system (Bio-Rad). Protein expression was quantified using Ponceau S (ThermoFisher, #A40000279). The following primary antibodies were used: Tyrosine hydroxylase, 1:1000 (Millipore Ab152), AKT, 1:1000 (Cell Signal 9272s), pAKT, 1:1000 (Cell Signal 9271s), ChAT, (1:2000, ab178850).

Mice were housed in individual cages within a TSE PhenoMaster climate chamber maintained at 24.5°C and 40% humidity. Water and food were provided ad libitum. Oxygen consumption, carbon dioxide production, food intake, and water intake were measured using a TSE PhenoMaster system (TSE Systems, Chesterfield, MO, United States). Calorimetry and food intake data were analyzed using the CalR web-based analysis tool (version 1.3), and energy expenditure data were normalized to lean body mass.

For cold exposure experiments, climate chamber temperature was reduced to 4°C, and indirect calorimetry measurements were performed for 4 h after which mice were removed from the cages and returned to regular housing conditions.

Echocardiography was performed throughout the paradigm using Visual Sonics Vevo 3100 system equipped with MX550D transducer (Visual Sonics). Mice were anesthetized with isoflurane and measurements were performed using short-axis M-mode scans. Mice were monitored to ensure that they recovered from anesthesia without difficulty. Measurements were used to quantify HRV parameters using Kubios HRV Scientific.

Fresh hearts were kept in 10% formalin solution for 24 h at 4°C then washed and transferred to a 30% sucrose solution. Fixed samples were frozen sectioned for staining. Slides were washed with PBS and blocked for 1 h at ambient temperature in PBS containing 3%BSA and 0.3% TritonX-100. Slides were incubated with primary antibody Tyrosine hydroxylase, 1:1000 (Millipore Ab152) overnight at 4°C. The next day after washing with PBS, slides were incubated with secondary antibody for 1 h at ambient temperature. After washing, slides were co-stained with DAPI and mounted for imaging.

RNA extraction was performed using the Acturus Picopure RNA Isolation kit (ThermoFisher, KIT0204) following the manufacturer’s instructions. 400 ng RNA from the lumbar DRG was converted to cDNA using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, California, United States). Quantitative PCR was performed using cDNA diluted 1:10, SYBR green (Roche, #04913850001), and 10 µM forward and reverse primers for ATF3 (F: CGA​AGA​CTG​GAG​CAA​AAT​G, R: AGG​TTA​GCA​AAA​TCC​TCA​AAT​AC) and β-actin (F: ACCTTCTAC AATGAGCTGCG, R: CTG​GAT​GGC​TAC​GTA​CAT​GC). The mRNA levels were calculated relative to β-actin as the housekeeping gene using the ΔΔCt method.

Statistical analyses were performed using GraphPad Prism 9. All data are shown as mean ± SEM. Body weight and AKT Western blot analysis were measured using repeated measures two-way analysis of variance (ANOVA) and two-way ANOVA, respectively. All other comparisons between treatments were performed using one-way analysis of variance (ANOVA). Fisher’s LSD was used for all post hoc analyses. Statistical significance was set at p < 0.05.