Beneficial effects of empagliflozin and liraglutide on the cerebral microcirculation of diabetic rats

This study aimed to evaluate the effects of the antidiabetics liraglutide, a GLP‐1 analog, and empagliflozin, an SGLT‐2 inhibitor, on the brain microcirculation of diabetic rats.


| INTRODUC TI ON
Type 2 diabetes mellitus (DM) is a highly prevalent chronic disease.
Global estimates confirm the enormous burden of diabetes in adults, especially in developing countries, which will experience the most significant increase in cases over the following years. 1,2In addition, it has been estimated that almost half of all people living with diabetes are undiagnosed. 1crovascular disorders are frequent chronic complications in patients with DM.The main drivers of diabetes-related microvascular dysfunction are hyperglycemia, obesity, insulin resistance, and hypertension. 3In addition, there is growing evidence that microvascular dysfunction is one of the key underlying mechanisms of brain disorders associated with diabetes. 46][7] Hyperglycemia is a risk factor for dementia, even among persons without diabetes or in prediabetes, indicating that brain pathology may start before the onset of diabetes. 8,9Hyperglycemic conditions cause inflammation and functional alterations of cerebral endothelial cells, 10 increasing the risk of cerebrovascular complications during diabetes.

Antidiabetic drugs such as GLP-1 receptor agonists and SGLT2
inhibitors have beneficial effects on glycemic control and body weight by distinct mechanisms, and both decreased cardiovascular risk in clinical trials. 11,12The GLP-1 receptor agonist liraglutide has been approved to treat obesity in patients without diabetes, 13 while the SGLT2 inhibitor empagliflozin is a well-tolerated antihyperglycemic agent approved for treating adults with type 2 diabetes, which also reduces the renal and cardiovascular risk. 14though the main guidelines recommend these medications for DM treatment, their effects on brain microcirculation during diabetes are unknown.
6][17][18][19] We hypothesized that the two drug classes could protect the cerebral microcirculation of diabetic subjects, and their combination may have additive effects on metabolic and microcirculatory alterations of DM.Therefore, the present study aimed to compare the effects of liraglutide and empagliflozin treatments alone and in combination on brain microcirculation using a rat model of DM.

| Animals
Sixty male Wistar rats weighing 200-210 g, bred, and maintained in the animal facility of Iguaçu University, were used in this study.
All the animals received food and water ad libitum under a 12 h light-dark cycle (lights on from 7:00 a.m. to 7:00 p.m.) and controlled temperature (25 ± 1°C).All the experimental procedures

| Experimental model
Animals were fed with commercial food (Purina®) and filtered water ad libitum until 42 days of age.A control group (n = 12) remained with the commercial diet, while the rest of the rats received a HFD.
The HFD composition followed the suggestion from the American Institute of Nutrition (AIN93-M).The normal diet contains 14% protein, 77% carbohydrate, and 9% lipids, and the HFD contains 14% protein, 33% carbohydrate, 53% lipids, and 0.5% NaCl.HFD was manufactured by Prag Soluções.DM was experimentally induced by combining a HFD and a single dose of streptozotocin (35 mg/kg), injected after 2 weeks on HFD.Rats remained on a HFD for another 6 weeks (Figure 1).After euthanasia, we collected the pancreas for histochemistry and morphological assessment of the islet's size.We did not observe significant changes in the size of the islets with this experimental protocol, indicating that pancreatic islets and insulin production were preserved (data not shown).

| Drug treatments
One week after the streptozotocin injection, DM rats were randomly separated into four experimental groups (n = 12 per group): DM treated with vehicle, DM treated with liraglutide (DM + Lira), DM treated with empagliflozin (DM + Empa), and DM treated with empagliflozin plus liraglutide (DM + Empa+Lira).Liraglutide (Victoza® Novo Nordisk) was administered subcutaneously at 100 μg/kg/day, and empagliflozin (Jardiance® Boehringer Ingelheim) was administered at 10 mg/kg/day orally.The DM group, and the control group on the regular diet, received a 0.9% saline solution (vehicle) by subcutaneous injection and by gavage simultaneously.Drugs were administered at the beginning of the circadian cycle's light phase daily for 5 weeks.

| Glucose tolerance test
The oral glucose tolerance test (OGTT) was performed following 12 h of fasting before drug administration on the last day of drug treatment to determine insulin sensitivity changes.Blood samples were collected from the tail vein before and 20, 40, 60, and 120 minutes after oral glucose administration (2 g/kg of a 50% glucose solution).Blood glucose concentrations were determined using the Accu-Chek Glucose Meter (Roche).The incremental area under the curve (AUC) of the glucose tolerance graphs was calculated as the integrated area under the curve above the basal value (time 0) over the 120-min sampling period.

| Postprandial triglycerides
After 12 h of fasting following the end of the treatment regime, blood samples were collected from the tail vein before and 2 h after lipid overload (2 mL/kg of olive oil administered orally).Blood triglycerides were quantified using the Accutrend® Plus device (Roche Diagnostics).The same rats were used for the measurements of blood triglycerides and glucose.

| Intravital microscopy
Rats were anesthetized with a combination of ketamine (100 mg/ kg i.p.) and xylazine (10 mg/kg, i.p.) (Cristália).To access the cerebral microcirculation, the rat was fixed in a stereotaxic apparatus, and a cranial window was created over the left parietal bone (1-5 mm lateral, between the coronal and lambdoid sutures) using a high-speed drill to expose the brain's microvascular surface, as previously described. 20This window enabled visualization of in vivo labeled leukocytes after intravenous injection, via the penian vein, of 0.3 mg/kg rhodamine 6G.Fluorescent leukocytes were visualized by intravital microscopy on postcapillary venules via epi-illumination at 510-550 nm using a 590 nm emission filter.The leukocyte-endothelial interactions were evaluated by counting the number of leukocytes adhering to the venular wall (100 μm long) over 30 s and were expressed as the number of cells/min/100 μm.Leukocyte rolling was defined as the movement of white blood cells on the endothelial surface of the vessel at a speed lower than the circulating red blood cells and was expressed as the number of cells/min using ZEN blue software (Zeiss).These parameters were determined in brain surface venules with diameters ranging from 50 to 120 cm.During the analysis process, adherent cells are identified and defined as those that do not move or detach from the endothelium within 30 s' video.Rolling cells, on the other hand, are those that contact the vessel wall, but do not remain adhered during the determined time. 21

| Brain immunohistochemistry
At the end of intravital microscopy analyses, rats were euthanized using an overdose of pentobarbital.The rats were intracardially perfused with a saline solution followed by 4% paraformaldehyde.The brains were quickly excised and serially sectioned at 40 μm.The sections were washed with PBS and incubated with 10% NGS diluted in PBS with 0.3% Triton X-100 for 90 min.They were then incubated with biotinylated isolectin B4 (IB4; 1:100; Vector Laboratories) overnight at 4°C, washed with PBS, and incubated with streptavidin-Cy3 (1:400) for 2 h.Slices were imaged by fluorescence microscopy, and the images were acquired with a 10x ocular and 10x objective microscope (Zeiss-AXIO SCOPE A1), producing a final magnification of 100×.

| Quantitative analysis of structural capillary density
Fluorescent images of brain sections labeled with IB4 were analyzed using the AngioTool software (https://ccrod.cancer.gov/confluence/ displ ay/ROB2/Downl oads) that allows the quantification of vascular networks in immunofluorescence-stained images, the use of which has already been described for the analysis of murine brain and retinal angiogenesis. 22Images for quantitative analysis were acquired with the fluorescence microscope using a 10x objective and a final magnification of 100×.The image was immediately segmented and skeletonized, and the analysis of the vasculature was performed automatically.For the analysis of the cerebral microvascular angioarchitecture, the AngioTool program was used to determine the total length of capillaries in the cerebral cortex of animals.

| Statistical analysis
The results are expressed as the mean ± standard error of the mean (SEM) for each group.Statistical analyses were performed in GraphPad Prism version 8.2.1 (GraphPad Software Inc.).Statistical comparisons of more than two groups were performed using analysis F I G U R E 1 Experimental design and timeline.DM was experimentally induced in male Wistar rats with a HFD for 8 weeks combined with one injection of streptozotocin (35 mg/kg).Treatments with liraglutide (100 μg/kg s.c.) and empagliflozin (10 mg/kg, oral) were administered for 5 weeks.Biochemical measures and intravital microscopy were undertaken after the treatment period.HFD, high-fat diet; OGTT, oral glucose tolerance test; STZ, streptozotocin; TG, triglycerides.
of variance (ANOVA), followed by Tukey's multiple comparisons post hoc tests as described in the figure legends.Differences with p ≤ .05were considered statistically significant.

| Effects of liraglutide and empagliflozin on the body mass of diabetic rats
Body weight was evaluated at the beginning of the experiment when animals were placed on the respective diets and every week throughout the experimental period in all groups.The treatments started in the third week.By the fourth week, DM rats had significantly increased body mass compared to the control while the DM + Empa+Lira rats had significantly lower body weight compared to DM rats.After 5 weeks, the DM + Lira and DM + Empa+Lira groups had significantly reduced body weight compared to DM rats and were not significantly different from the control.Throughout the experimental period, there were no significant differences between the DM and DM + Empa groups (Figure 2A).The combination of empagliflozin and liraglutide reduced DM rats' body weight more effectively than the other treatments.Relative weight gain was greater in DM rats (24%) compared to controls (11%), and significantly reduced in by liraglutide (8,7%)   and by the combined treatment (4,6%) compared to DM + Empa (23%) and to DM (Figure 2B).

| Effects of liraglutide and empagliflozin on the metabolic parameters of DM rats
Glucose tolerance and blood triglycerides were evaluated after the 5-week treatment of DM rats with liraglutide, empagliflozin, or the combination of both.Non-treated DM rats presented hyperglycemia indicating increased glucose intolerance (Figure 3).Empagliflozin was more effective than liraglutide in improving the glucose tolerance of DM rats.The combination therapy also significantly improved the glucose tolerance of DM rats to levels similar to those seen in rats that received the empagliflozin monotherapy (Figure 3A,B).
Furthermore, the combination therapy was more effective than the monotherapies in improving the fasting glycemia of DM rats, being the only treatment protocol that significantly altered this parameter (Figure 3C).
Postprandial triglyceridemic was significantly increased in DM rats compared to control rats.The treatment with liraglutide, however, significantly attenuated postprandial hypertriglyceridemia, while treatment with empagliflozin or combination therapy showed only a trend to decrease postprandial blood triglycerides (Figure 4).

| Effects of the liraglutide and empagliflozin on the brain microcirculation of DM rats
Intravital assessments of leukocyte-endothelium interaction revealed that DM rats had a marked increase in the number of rolling

Relative weight gain (%)
and firmly adhered leukocytes compared to the control group.
Treatment of DM rats with liraglutide or empagliflozin reduced leukocyte-endothelium interactions to levels equal to those of the control group (Figure 5A).The combination therapy had similar effects to the monotherapies on leukocyte adhesion (Figure 5B) and rolling (Figure 5C) in the endothelial surface of brain vessels in DM rats.
The cerebrovascular structural capillary density evaluated by IB4 immunofluorescence-stained images showed that empagliflozin significantly improved the structural capillary density of DM rat brains, as determined by increased total vessel length.
Liraglutide had a modest effect on capillary density.The empagliflozin monotherapy was the most effective in preventing capillary rarefaction in DM rats.The combination therapy with liraglutide and empagliflozin did not promote any additional beneficial effects on vascular rarefaction compared to the empagliflozin monotherapy (Figure 6).

| DISCUSS ION
Microvascular dysfunction is a frequent complication of diabetes.
Non-controlled DM is a risk factor for stroke, dementia, and depression, and brain microvascular dysfunction is a critical underlying mechanism. 4Antidiabetic drugs such as SGLT2 inhibitors and GLP-1 DM rats.In contrast, empagliflozin was more effective in reducing hyperglycemia.Liraglutide increases insulin and decreases glucagon secretion, which contributes to controlling glycemia.This drug also stimulates satiety leading to a reduction in food intake and weight loss by mechanisms that involve central regulation of feeding behavior. 23Empagliflozin lowers blood glucose by increasing urinary glucose excretion.Body weight loss with SGLT2 inhibitors may be associated with adipose tissue reduction, lean tissue reduction, or water loss. 24hesion and rolling of leukocytes on the endothelium of brain venules were increased in diabetic animals, indicating cerebral microvascular inflammation.Compared to liraglutide, empagliflozintreated DM rats showed a better reduction in capillary rarefaction associated with a more significant reduction in leukocyte adhesion in cerebral venules.Treatments with liraglutide and empagliflozin alone were able to reduce the adhesion of leukocytes which is an extremely important inflammatory parameter of the microcirculation since leukocytes adhesion endothelium allow the formation of platelet plugs, microthrombus, and reduction of cerebral perfusion. 25,26In fact, only empagliflozin significantly increased capillary density, but the association of drugs ensured that capillary density remained at very high levels, with a trend of statistical significance with Alzheimer's disease. 19In experimental models of DM, SGLT2 inhibitors were shown to improve coronary microvascular function and contractile performance of genetically obese (ob/ob) mice. 25other study showed that diabetic obese mice exhibited ultrastructural abnormalities in the neurovascular units characterized by mural endothelial cell tight and adherent junction decrease, pericyte loss, basement membrane thickening, astrocyte activation with detachment and retraction from mural cells, microglia cell activation with aberrant and oligodendrocyte myelin splitting, disarray, and axonal collapse.In this study, the authors showed that empagliflozin ameliorated the neurovascular ultrastructure of diabetic mice. 27positive association of diabetes-related microvascular disease with cardiovascular outcomes has been clinically observed. 3,4tative mechanisms to explain this association include glucose toxicity and the generation of advanced glycation end products that may deposit in endothelial cells, forming crosslinks between molecules in the basement membranes of the extracellular matrix. 28other potential mechanism is endothelial dysfunction triggering smooth muscle cell proliferation, migration, and matrix secretion, contributing to atherosclerosis progression in diabetes. 3Chronic hyperglycemia and glucose toxicity can be especially damaging to cell types more vulnerable to elevated plasma glucose levels, such as pancreatic β-cells, neurons, and endothelial cells, that equilibrate their intracellular glucose level to their extracellular environment. 29e results of our study indicate that empagliflozin was superior to liraglutide in preventing brain microvascular dysfunctions associated with DM.In DM patients, the risk of microvascular complications is strongly associated with hyperglycemia. 30Therefore, long-term treatment with empagliflozin may be beneficial in protecting the brain microcirculation of diabetic patients, especially those with cerebrovascular risk factors or a family history of neurological disorders.
Nevertheless, obesity and hyperlipidemia are well-known risk factors for vascular diseases, such as atherosclerosis, which contribute to microcirculation pathology in diabetes. 31In addition, obesity is intimately associated with the pathogenesis of insulin resistance due to the formation of metabolic products derived from lipids, adipokine imbalance, and pro-inflammatory cytokines, which interfere with insulin signaling. 32Therefore, obese diabetic patients, accompanied or not by dyslipidemia, may take advantage of the association of empagliflozin with liraglutide to prevent cerebral microvascular complications.
In conclusion, our results suggest that empagliflozin and liraglutide, whether administered separately or together, exhibited beneficial effects on cerebral microcirculation in DM rats by reducing leukocyte adhesion.This highlights the critical role these therapies can play in mitigating microvascular complications in individuals with diabetes and overweight or obesity, underlining the significance of combination therapy in protecting cerebral microvascular health.
were performed in strict accordance with the guidelines of the National Council for the Control of Animal Experimentation (CONCEA) and approved by the Ethics Committee on Animal Use (CEUA) of Iguaçu University, under the license number PEBIO/ UNIG N o 010/2017.

F I G U R E 3
agonists reduce body weight and blood glucose by distinct mechanisms, both decreasing cardiovascular risks.However, the effect of these medications on the brain microcirculation of diabetic subjects is still unknown.The present study demonstrated that antidiabetic treatments with empagliflozin and liraglutide protected the microcirculation of the cerebral cortex of DM rats by mechanisms potentially related to the metabolic effects of improving glucose tolerance and hypertriglyceridemia and by reducing body weight.When considering the effects of the monotherapies, liraglutide was superior to empagliflozin in reducing body weight and postprandial hyperlipidemia in Empagliflozin and liraglutide combination improved hyperglycemia of diabetic rats.(A) Oral glucose tolerance test (OGTT) after the treatments and (B) area under the curve (AUC) of OGTT graph.Data are shown as mean ± SEM.Statistical analysis was performed by two-way repeated measures ANOVA with Tukey's multiple comparisons tests (**p < .01,***p < .001,****p < .0001DM vs. Control, DM + Empa, and DM + Empa+Lira; #p < .05,##p < .01DM vs. DM + Lira).(C) Fasting blood glucose was measured before the OGTT.Statistical analysis was performed by one-way ANOVA with Tukey's multiple comparisons tests (*p < .05 vs. Control; #p < .05 vs. DM).

(F I G U R E 4
not reached apparently due to a greater dispersion of data).The combination of SGLT2 inhibitors and GLP-1 receptor agonists has already been clinically evaluated in DM patients, but its effect on brain microcirculation was still unknown.Clinical studies have shown that this combination therapy significantly reduces F I G U R E 5 Effects of liraglutide and empagliflozin treatments on leukocyte adhesion and rolling in the brain microcirculation of diabetic rats.(A) Representative fluorescence intravital microscopy images of brain microcirculation of control rats, DM rats that received vehicle or were treated with empagliflozin (DM + Empa), liraglutide (DM + Lira), or the combination of both drugs (DM + Empa+Lira).Venules with adherent and rolling leukocytes in the brain microcirculation of rats.(B) Quantification of leukocyte adhesion to vessel walls.(C) Quantification of leukocyte rolling on vessel walls after intravenous rhodamine 6G injection.Data are shown as mean ± SEM.Statistical analysis was performed by one-way ANOVA and Tukey's multiple comparisons tests (*p < .05 vs. Control; # p < .05 and ##p < .01 vs. DM).Effects of liraglutide and empagliflozin on blood triglyceride levels of diabetic rats.Postprandial blood triglycerides were measured 2 h after a lipid overload.Data are shown as mean ± SEM.Statistical analysis was performed by one-way ANOVA and Tukey's multiple comparisons tests (**p < .01DM vs. control, *p < .05DM vs. DM + Lira).body weight, and systolic blood pressure compared with either drug as monotherapy. 11Furthermore, empagliflozin has been demonstrated to reduce vascular damage and cognitive impairment in a murine model of type 2 diabetes mixed

F I G U R E 6
Empagliflozin treatment protected the brain microcirculation of diabetic rats.Total vessel length and lacunarity were determined by AngioTool analysis of IB4-labeled images of brain slices of the cortex from control rats and type 2 diabetic rats treated with vehicle (DM), treated with empagliflozin (DM + Empa), liraglutide (DM + Lira), or the combination of both drugs (DM + Empa+Lira).Scale bar 50 μm.Graphical representation of the structural capillary density of the cerebral cortex.Data are shown as the mean ± SEM.N = 5 per group.Statistical analysis was performed by one-way ANOVA and Tukey's multiple comparisons tests (***p < .001 vs. Control; #p < .05 vs. DM).