S-allyl Cysteine and Cannabidiol are Equally Effective as Metformin in Preserving Neurovascular Integrity, Retinal Structure, and Cognitive Function in db/db Type 2 Diabetic Mice

doi:10.21203/rs.3.rs-3995875/v1

Download PDF

Article

S-allyl Cysteine and Cannabidiol are Equally Effective as Metformin in Preserving Neurovascular Integrity, Retinal Structure, and Cognitive Function in db/db Type 2 Diabetic Mice

https://doi.org/10.21203/rs.3.rs-3995875/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Type 2 diabetes mellitus (T2D) is associated with central nervous system (CNS) alterations marked by neurovascular, inflammatory, and oxidative damage, resulting in cognitive dysfunction and retinal neurodegeneration. This study explored the therapeutic potential of naturally derived S-allyl-cysteine (SAC) and cannabidiol (CBD) in LepR db/db mice, targeting blood-brain/retinal barrier (BBB/BRB) leakage, glial activation, and DNA oxidative damage to alleviate memory deficits and retinal neurodegeneration.

Diabetic mice underwent 23 weeks of dietary treatments: diabetic db/db controls, SAC, CBD, SAC + CBD, and metformin, with nondiabetic db/+ mice as negative controls. Memory function was assessed using novel object recognition and passive avoidance tests, while retinal thickness was measured via in vivo OCT imaging. Immunofluorescence imaging quantified neurovascular leakage (IgG extravasation), glial activation (GFAP and Iba1) and DNA oxidation (8OHdG) in brain sections and retinal whole-mounts.

In diabetic db/db mice with hallmark metabolic dysregulation, CNS alterations included BBB/BRB leakage, glial activation, and retinal neurodegeneration of the ganglion cell complex (GCC), correlating with memory impairments.

SAC and CBD exhibited significant therapeutic effects against CNS pathophysiologies, attenuating glial activation and synergistically preventing BBB leakage. In the retina, these compounds attenuated BRB leakage risk and reduced glial-mediated neuroinflammation. SAC and CBD preserved GCC thickness and improved memory functions, proving to be as effective as or superior to metformin against diabetic-associated CNS pathophysiology.

Health sciences/Diseases/Endocrine system and metabolic diseases

Health sciences/Diseases/Neurological disorders

Cannabidiol

Cognitive dysfunction

Glial activation

Neurovascular

S-allyl-cysteine

Type 2 diabetes mellitus

Cognitive dysfunction is recognised as a prevalent complication and comorbidity of type 2 diabetes mellitus (T2D) [1] and accumulating evidence supports the role of cerebrovascular dysfunction, chronic neurovascular inflammation and oxidative stress as causal risk factors [2, 3]. Recent studies on the retina suggest a microvascular axis for cognitive decline in T2D, with a significant association between retinal capillary integrity, retinal neurodegeneration and deficits in memory functions [4–6]. Clinical neuroimaging studies have linked T2D-associated cognitive dysfunction to BBB disruptions and atrophy in cognition-related brain regions [7], [8]. Animal studies support these findings, demonstrating that neurovascular breakdown and glial-mediated neuroinflammation contribute to learning and memory deficits [9], [10]. Our previous studies on diet-induced insulin resistance have indicated that vascular pathology precedes cognitive impairment [11], suggesting that therapeutics targeting microangiopathy and glial overactivation may improve cognitive outcomes.

The deterioration of the neurovascular unit (NVU) observed in T2D involves the impairment of both the blood-brain barrier (BBB) and blood-retinal barrier (BRB) [12]. Emerging data indicate that astrogliosis and chronic activation of resident microglia facilitate the breakdown in NVU functionality [8, 9]. The state of reactive gliosis, mediated by diabetic metabolic dysregulation, leads to over-production of reactive oxygen species (ROS) and a concurrent reduction in antioxidant capacity. These factors contribute to cellular oxidative stress, which manifests in DNA damage and ultimately, neuronal cell loss [10, 11]. These findings underscore the need for interventions aimed at positively modulating glial cell activation and preserving neurovascular integrity. Such approaches may mitigate the risk of cognitive decline and ocular degenerative pathology in the context of T2D.

Naturally derived biomolecules boasting anti-inflammatory and antioxidant properties offer a promising, safe and cost-effective approach to the treatment of metabolic [12, 13], vascular [14, 15] and neurodegenerative [16, 17] disorders. Two such compounds, s-allyl-cysteine (SAC) derived from aged garlic (Allium sativum) and cannabidiol (CBD) from the Cannabis sativa plant, have garnered attention as neuroprotective agents with therapeutic potential in T2D complications within neural tissues.

SAC, an organosulfur molecule and a potent ROS scavenger, is reported to improve spatial memory recognition and mitigate oxidative stress in senescence-accelerated SAMP8 and SAMP10 mice [22] and in a murine model of type 1 diabetes mellitus (T1D) [23]. Moreover, SAC exhibits neurotrophic effects in inflammatory-induced cognitive dysfunction, characterised by attenuation of astrogliosis and a reduction in microglial immunoreactivity [24]. Clinical studies that administer SAC as a garlic supplement have also shown its potential to slow retinal neurodegeneration and enhance visual acuity in T2D [25]. However, the mechanisms underlying the beneficial effects of SAC in T2D have yet to be directly explored.

CBD, the major non-psychotropic phytocannabinoid, showcases pleiotropic cellular and physiological effects in murine models of T1D, mild traumatic brain injury, and virus-induced demyelination [22–24]. CBD consistently exhibits the capacity to reduce proinflammatory cytokines, attenuate glial cell activation, mitigate BBB hyperpermeability, and inhibit neurodegeneration [29]. Furthermore, exogenous provision of CBD has conferred protection of cognitive function in models of Alzheimer’s disease, in part through the modulation of the endocannabinoid system [30]. Consequently, it is imperative to examine the physiological response to exogenous CBD in the context of cognitive performance in T2D.

This study was designed to investigate the mechanisms underlying the supposed protective effects of SAC and CBD in maintaining integrity of the NVU within the brain and retina, with the aim of preventing cognitive dysfunction in a robust model of T2D. Furthermore, this investigation explores the potential synergistic therapeutic effects of combined SAC + CBD treatment, which remains a novel area of exploration. Using the widely adopted leptin receptor-deficient db/db mouse model of T2D, we hypothesise that SAC and CBD, administered individually or in combination, may ameliorate BBB and BRB hyperpermeability through the modulation of microglial and astrocytic immunoreactivity, thereby mitigating the burden of neuroinflammation and oxidative stress. Provision of SAC and/or CBD may preserve memory function and retinal neuronal structure in T2D.

2.1 SAC and CBD attenuate markers of metabolic dysregulation linked to diabetes

This study investigated the potential metabolic effects of SAC and CBD in db/db mice with T2D phenotype (Fig. 1). Diabetic db/db control mice showed significant hyperglycaemia, hyperinsulinemia and hypertriglyceridemia (Fig. 1A-C). Surrogate markers of insulin sensitivity demonstrated resistance in both glucose (HOMA-IR) and lipid (TyG) metabolism contexts (Fig. 1D-E). Additionally, a significant increase in body weight was observed in the db/db control mice in comparison to nondiabetic db/+ controls (Fig. 1F).

Dietary intervention with SAC markedly reduced diabetic insulin and triglyceride levels, maintaining the TyG measure of insulin sensitivity at levels comparable to nondiabetic db/+ controls. There was no significant impact on body weight in SAC treated db/db mice. The positive metabolic effects of SAC were found to be equipotent, or better than those observed db/db mice treated with metformin.

Treatment with CBD in db/db mice led to a paradoxical increase in plasma glucose that reached statistical significance. However, all treatment groups indicated a reversed trend in plasma glucose levels after the 23-week intervention. The therapeutic effects of CBD on metabolic status were primarily mediated by insulin, with a significant reduction in plasma insulin levels and suggestive improvements in HOMA-IR insulin resistance, although not the TyG score per se. The CBD-associated reduction in plasma triglyceride concentration was non-significant and comparatively less in magnitude than that of other treatment groups. Notably, CBD monotherapy showed the only significant attenuation of obesity in db/db mice.

Surprisingly, treatment with SAC + CBD combination in db/db mice demonstrated no synergistic improvements in the T2D metabolic phenotype compared to monotherapy interventions. SAC + CBD treatment resulted in comparable levels of plasma glucose but was less effective in lowering insulin levels relative to the single therapy groups. Administration of SAC + CBD normalised plasma triglycerides with similar efficacy to SAC treatment alone. However, no notable improvements were observed in markers of insulin sensitivity as determined by HOMA-IR, or TyG in SAC + CBD-treated mice. Furthermore, single therapy groups appeared to outperform the SAC + CBD combinatory group in improving insulin sensitivity, with the difference in effects approaching significance. Treatment with SAC + CBD combination showed no therapeutic benefits for body weight in obese db/db mice.

The oral hypoglycaemic metformin showed weak therapeutic effects against the diabetic metabolic phenotypes. Dietary metformin for 23 weeks did not positively regulate glucose homeostasis, but rather significantly exacerbated plasma glucose concertation in db/db mice. Treatment with metformin significantly reduced plasma insulin and triglyceride levels; however, it did not confer positive effects in their respective measures of insulin sensitivity, HOMA-IR and TyG index. Body weight also remained elevated in metformin-treated db/db mice compared to nondiabetic db/+ controls.

2.2 SAC and CBD mitigate diabetic neuroinflammation and oxidative stress, and synergistically protect BBB integrity

The ameliorative potential of SAC and CBD against BBB leakage, glial activation and oxidative damage were evaluated through immunofluorescence neuroimaging of the hippocampal formation (HPF) and cortex (CTX) (Fig. 2).

BBB permeability, assessed by the immunoreactivity of extravasated IgG within the brain parenchyma (Figs. 2A-C), was over threefold greater in db/db controls compared to nondiabetic db/+ controls. The significant increase in BBB leakage was noted in both the CTX and HPF regions (Fig. 2D). Treatment with SAC and CBD monotherapy suggested some improvement in BBB integrity in the HPF, with an average reduction in parenchymal IgG by 50%, approaching statistical significance. In contrast to metabolic measures, SAC + CBD combination therapy demonstrated synergistic BBB protection against IgG extravasation. Significant reductions in IgG abundance were evident in both brain regions, with a more pronounced effect in the HPF. Metformin treatment also significantly reduced IgG abundance in the brain.

Astrogliosis was determined as GFAP immunoreactivity (Fig. 2E). Diabetic db/db controls showed a significant increase in GFAP–astrocyte reactivity, consistent with the observed BBB breakdown. Dietary SAC and particularly CBD essentially abolished astrocyte reactivity in db/db mice, with GFAP abundance markedly decreased even compared to nondiabetic db/+ controls. Metformin significantly decreased GFAP immunoreactivity in db/db mice, but the observed improvement was modest compared to naturally derived agents, with significant differences in efficacy observed compared to CBD-containing formulations.

The pathological increase in BBB leakage and astrogliosis was also mirrored by the increase in Iba1-microglial activation, as the principal measure of neuroinflammation (Fig. 2F). The elevated Iba1 immunointensity in db/db mice was normalised by SAC and CBD monotherapies and combination treatment. Metformin also significantly suppressed Iba1-microglial immunointensity.

Oxidative damage to mitochondrial and nuclear DNA, an established pathophysiology of T2D, was evaluated through 8OHdG immunointensity (Fig. 2G) and showed minimal increase in db/db mice compared to nondiabetic db/+ mice. However, CBD- containing formulations nevertheless displayed substantial reductions in 8OHdG staining. In contrast, the SAC and metformin interventions showed no apparent effect on DNA oxidation. Representative images of glial activation and DNA oxidation are shown in Fig. 2H-J.

2.3 Natural compounds prevented diabetes-associated thinning in the inner retina

Retinal OCT imaging was utilised for the in vivo assessment of thickness of the total neuroretina and relevant sublayers at the optic nerve head, as indicated in the representative images (Figs. 3A-C). Although the thickness of the total neuroretina was comparable between diabetic db/db and nondiabetic db/+ controls, there was a significant increase in thickness following a 23-week SAC + CBD and metformin intervention (Fig. 3D).

The diabetes-associated thinning of the GCC was evident. However, db/db mice treated with SAC and CBD, alone and in combination, exhibited GCC preservation with equal effectiveness as metformin treated mice (Fig. 3E).

The INL and ONL, as dense cellular and synaptic regions of the retina, were examined for diabetic thinning. The INL did not show significant differences in db/db mice compared to nondiabetic controls. However, among the interventions tested, the SAC + CBD combination displayed a marked increase in INL thickness (Fig. 3F). In contrast, the ONL was significantly thicker in db/db controls relative to nondiabetic controls (Fig. 3G). Notably, db/db mice that received either SAC + CBD combination or metformin indicated significant increases in ONL thickness.

2.4 Diabetic astrogliosis and neuroinflammation in the retina were prevented with natural compounds and metformin.

Following in vivo OCT imaging, retinal wholemounts were assessed for neurovascular and glial pathophysiology (Figs. 4A-C). The data confirm parallel mechanisms of neurodegenerative thinning as observed in diabetic cerebral pathology.

Consistent with the assessment of BBB leakage, retinal extravasation of IgG was detected in 8% of db/db mice, while no such no leakage was observed in nondiabetic db/+ control mice (Fig. 4D). Approximately 40% of BRB leakage events occurred in db/db controls. Among the dietary intervention groups, SAC-treated mice accounted for 40% of the BRB leakage events; however, there was a tenfold decrease in the immunointensity of extravascular IgG in the retina parenchyma. SAC + CBD-treated mice represented 20% of BRB leakage events and demonstrated a further tenfold decrease in IgG immunostaining. Treatment with CBD monotherapy or metformin showed total protection of BRB with no observable IgG extravasation.

In alignment with the BRB breach, corresponding changes in GFAP-astrogliosis and Iba1-microglial neuroinflammation were observed in the db/db retina (Fig. 4E). Astrogliosis was increased in db/db mice, and all treatment interventions showed significant reductions in GFAP immunointensity. Although no statistical differences were observed among the intervention groups, CBD-containing therapies suggested greater efficacy in attenuating astrogliosis.

The abundance of Iba1 immunointensity more than doubled in db/db control mice compared to nondiabetic controls (Fig. 4F). However, the provision of SAC and CBD monotherapy and combination treatments completely prevented the increase in Iba1-microglial activation in db/db retinas. Notably, dietary intervention with natural compounds demonstrated equipotent efficacy to metformin in suppressing microglia Iba1 abundance in the retina.

2.5 SAC and CBD improve memory impairment with greater efficacy than metformin

To determine whether the observed improvements in neurovascular integrity and glial cell activation conferred benefits in cognitive function, we conducted two functional tests of memory (Fig. 5).

In the NOR test for short-term memory, lower preference index scores reflect a failure to identify and explore the novel object. The NOR test revealed significant decrease in preference index in db/db mice compared to nondiabetic db/+ controls (Fig. 5A). In line with the beneficial effects of SAC and CBD on the neuroinflammatory milieu, 23 weeks of treatment resulted in complete preservation of short-term memory. While SAC + CBD prevented the decrease in the preference index, no further additive effects were realised with combination therapy. Interestingly, metformin was less effective in supporting performance in the NOR test compared to other treatment groups, especially SAC or CBD monotherapy.

The PA test for long-term memory showed a diabetes-associated impairment consistent with findings from the NOR challenge (Fig. 5B). There was a significant decrease in the latency time recorded for db/db mice compared to db/+ controls, indicating an inadequate fear memory response. After the 23-week intervention, SAC conferred significant improvement in latency, while CBD presented non-significant improvements compared to untreated diabetic db/db control mice. Treatment with SAC + CBD combination showed a lesser therapeutic effect than the single therapy groups. Similarly, metformin displayed minimal efficacy on PA latency, resembling the SAC + CBD combination group.

2.6 Diabetic metabolic status is associated with pathological changes in the brain and retina

Pearson’s correlation coefficient analysis (Fig. 6) revealed significant associations between metabolic status and CNS pathology in diabetes. There was a significant association between elevated plasma glucose and Iba1- microglial activation in the brain. The most consistent finding was a moderate to strong association between poor insulin regulation and NVU injury, reflected in significant correlation coefficients for BBB leakage, GFAP-astrogliosis and Iba1-microglial activation. The HOMA-IR measure of insulin resistance was also significantly associated with retinal astrogliosis. There were no significant associations between plasma metabolic parameters and cerebral oxidative stress marker, 8OHdG or retinal Iba1- microglia staining, however the data indicated trends of weak to moderate positive and negative correlations, respectively. Furthermore, plasma triglycerides showed no significant associations with markers of neuroglial and vascular injury in the CNS.

There were reversed correlations between metabolic parameters and memory performance; trends deviated from insulin-centric associations and varied according to the memory test employed. Preference index in the NOR test of short-term memory was positively correlated with disrupted metabolic status; most notably with plasma glucose and insulin resistance relative to lipid metabolism indicated by TyG index. Conversely, the PA test of long-term memory showed weak to strong negative associations, whereby a decrease in latency was significantly correlated elevated plasma glucose and TyG index.

2.7 BBB leakage, with associated neuroinflammation and oxidative stress, collectively serve as pathological correlates of diabetic memory impairment

Pearson’s correlation coefficient analyses (Fig. 7) demonstrated strong associations between cerebral pathophysiological markers and memory impairment in the db/db model of T2D. BBB dysfunction was significantly associated with neuroinflammation, astrogliosis, and oxidative stress (Fig. 7A); memory impairment directly correlates with BBB leakage and glial activation (Fig. 7B). The correlations for extravasated IgG immunointensity in Fig. 7A were positive and strong for GFAP-positive astrocytes, moderate for Iba1-positive microglial activation, and weakest for the 8OHdG marker of DNA oxidation.

The follow-up Pearson’s correlation analysis in Fig. 7B indicated weak to strong negative associations between short- and long-term memory function and various immunofluorescent markers of cerebral pathology. Generally, the correlations were comparable across short- and long-term memory tests. A decreased preference index was moderately associated with cerebral microglial activation and astrogliosis. No correlation was identified between short term memory function and 8OHdG immunoreactivity. However, there was a strong association between short-term memory impairment and IgG extravasation in the brain.

Long-term memory impairment, signified by decreased latency in the PA test, showed significant associations that were strong for BBB leakage, moderate for Iba1-microglial, respectively, and weak for 8OHdG immunointensity. The association between long-term memory impairment and GFAP-astrocyte reactivity was moderate but did not reach significance.

This study aimed to investigate the individual and combined effects of naturally derived compounds, SAC and CBD, in preserving cognitive function by preventing BBB/BRB disruption, glial cell activation and oxidative DNA damage in the established db/db mouse model of T2D.

Our metabolic measures in db/db mice were consistent with moderate T2D, including hyperglycaemia, hyperinsulinemia, and hypertriglyceridemia, along with insulin resistance and obesity. These moderately diabetic db/db mice displayed impaired short-term episodic memory, as evidenced by a decrease in preference index in the NOR test. Additionally, the mice showed reduced latency time in the PA test, indicating dysfunction in long-term memory formation mediated by the hippocampus and amygdala in response to aversive stimuli [42]. Importantly, our data indicate that dietary intervention with SAC and/or CBD improved short-term memory function, with SAC also preserving long-term memory.

The exact mechanisms by which SAC/CBD prevented cognitive decline in moderately diabetic db/db mice are not fully understood. In the db/db mice, either SAC, CBD, or their combination effectively reduced hyperinsulinemia to a similar extent as the anti-diabetic drug, metformin. These findings align with recent evidence indicating that SAC and CBD enhance insulin receptor sensitivity, lipid metabolism and cellular glucose uptake [43], [44]. Notably, SAC demonstrated efficacy in improving hypertriglyceridemia in db/db mice, in line with established lipid-modulating effects of cysteine compounds [16]. Additionally, CBD treatment resulted in a reduction in body weight, supporting recent clinical findings on the potential use of phytocannabinoids in obesity treatment [45]. However, it is crucial to note that the overall benefits of SAC and CBD, whether administered alone or in combination, against metabolic dysregulation were modest, leaving measures of insulin resistance elevated in db/db mice. Similarly, metformin exhibited limited restorative effects on diabetic metabolic status in this study, failing to significantly impact hyperglycaemia and insulin resistance as demonstrated in other studies [46]. These data suggest that the therapeutic effects of SAC and CBD in the CNS may be potentiated through mechanisms independent of glucose and insulin regulation.

Instead, the benefits of SAC and CBD on cognitive performance in db/db mice may attribute to its neurovascular protective anti-inflammatory/-oxidative properties. Indeed, the cognitive-preserving actions of SAC and CBD monotherapies have been reported in multiple disease models, involving vascular protection and amelioration of chronic reactive gliosis and oxidative stress in the CNS [47], [48].

Furthermore, the db/db mice exhibited significant cerebral abnormalities such as BBB leakage, reactive gliosis, and oxidative DNA damage. Retinal changes we observed in db/db mice included reduced GCC thickness, increased BRB leakage risk, and significant glial activation, all associated with impaired memory functions. These findings align with existing evidence linking cerebral and retinal microangiopathy, neuroinflammation, and oxidative stress to cognitive dysfunction and neural degeneration in the context of T2D [6], [12]. Interestingly, Corem et al., [49] reported transient BBB disruption in normoglycemic db/db mice, which disappeared upon diabetes onset. Emerging evidence suggests that both transient and prolonged BBB disruptions involve astrocyte reactivity and microglial activation in T2D and other neurodegenerative disorders [50]. Additionally, activated glia contribute to oxidative stress, leading to DNA mutations and cellular damage that exacerbates neuroinflammation [51].

Remarkably, treatment with natural agents conferred substantial neuroprotective effects in diabetic db/db mice. SAC and CBD, either alone or in combination, demonstrated potent anti-inflammatory effects, reducing GFAP-astrogliosis and Iba1-microglial activation. The SAC + CBD combination treatment notably restored BBB integrity more effectively than monotherapies. These findings are supported by previous studies, with SAC shown to decrease oxidative DNA damage in cerebral ischemia [52] and ameliorate reactive gliosis in the streptozotocin (STZ) rat model of diabetes [23]. CBD has been reported to attenuate BBB leakage and microglial activation in a viral multiple sclerosis model [27] and modulate astrocyte activity through the endocannabinoid system, suggesting potential therapeutic benefits for various neurological disorders [53].

Due to physiological similarities, we recently reported that the microvascular dysfunction and neuroinflammatory cascades in the diabetes brain are closely mirrored in the retina, leading to thinning of GCC cell layer [6]. Immunofluorescence analysis in this study showed an increased risk of focal BRB leakage, accompanied by GFAP-astrogliosis and Iba1-microglial activation in diabetic db/db mice. Evidence from T2D models suggests that BRB leakage and glial-mediated neuroinflammation contribute to retinal ganglion cell death [13], [54]. Importantly, our study provided novel evidence that SAC and CBD, whether alone or combined, provide neuroprotection against GCC thinning comparable to metformin. This preservation of retinal thickness may be due to reduced BRB leakage risk and severity, as well as complete protection against abnormal glial activation in the retina.

Although research on SAC and CBD in ocular diseases is limited, SAC has demonstrated potential to suppress GFAP-astrocyte immunoexpression in kainate-induced toxicity [55]. Similarly, CBD has been reported to reduce the expression of inflammatory cytokines and pro-adhesion molecules, which promote vascular permeability and cell death [28]. These reports suggest that SAC and CBD promote retinal health through modulation of glial cell activation.

Mechanistic research supports the effects of SAC and CBD on neurocognition. SAC's antioxidant actions involving nuclear factor (erythroid-derived 2)-like 2 (Nrf2) have been shown to preserve spatial learning and memory functions, along with immunomodulatory effects that down-regulate pro-inflammatory nuclear factor kappa B (NF-κB) cascades and reduce gliosis in the hippocampus [24], [56]. Similarly, CBD demonstrated benefits in improving episodic memory and social learning in models of Alzheimer’s disease, attributed to its modulation of various pathways involved neuroprotection, including the NF-κB/Nrf2 pathway [57]. CBD is also shown to activate nuclear receptors such as peroxisome proliferator-activated receptor gamma (PPARγ), which downregulates the expression of inflammatory genes in glia cells [58]. Overall, the neuroprotective effects of SAC and CBD observed in this study are supported by their anti-inflammatory and antioxidative actions, which may help mitigate memory impairment in T2D.

In contrast to SAC/CBD, metformin showed limited protection against glial-mediated neuroinflammation, despite a protective effect on the BBB. The modest improvements in memory function observed in metformin-treated db/db mice in this study may be due to its limited efficacy in neuroprotection. Contrary to our findings, De Oliveira et al., [59] reported substantial therapeutic benefits of metformin in STZ-induced T1D mice, including mitigation of spatial memory decline, astrocyte and microglial activation, and reduced levels of pro-leakage vascular endothelial growth factor (VEGF). Likewise, Pilipenko et al., [60] documented improved cognitive function and reduced reactive gliosis in toxin-induced Alzheimer’s disease models. However, recent findings suggest minimal cognitive benefits with metformin treatment [61], and earlier studies even indicate that metformin may exacerbate neuroinflammation [62]. These discrepancies warrant further investigation and suggest that while metformin is an effective anti-diabetic medication, it may not be the optimal therapeutic choice for addressing CNS complications.

Clinical evidence highlights the impact of metabolic dysregulation on neurovascular injury and cognitive deficits [63]. Our study re-affirmed the association between CNS pathology and the metabolic profile of T2D, particularly disruptions in insulin regulation. The central hypothesis of our study proposed that SAC and CBD could preserve cognitive function, in part, by ameliorating the underlying cerebral pathophysiology. To verify this hypothesis, we investigated the relationship between BBB leakage and makers of glial activation and DNA oxidation. Additionally, we investigated the links between memory performance and cerebral pathophysiologies in experimental T2D.

Our study revealed significant correlations between BBB leakage and GFAP-astrogliosis, as well as Iba1-microglial activation, which are consistent with existing literature [64]. Under diabetic conditions, reactive astrocytes exhibit reduced functionality [65], leading to retraction of astrocytic end-feet from microvessels [66]. Similarly, pathologically active microglia promote vascular permeability by producing inflammatory cytokines and engaging in abnormal phagocytosis of neurovascular elements [67]. Our correlation findings emphasise the role of reactive gliosis in the pathogenesis of BBB leakage in T2D. Unsurprisingly, impairments in short-term and long-term hippocampal memory functions were associated with BBB leakage, as well as glial activation. Therefore, the cognitive benefits of SAC and CBD formulations may be linked to their actions on neurovascular and glial components.

While our study provides valuable insights, certain aspects warrant further investigation. For instance, we observed significant changes in the total neuroretina and nuclear sub-layers (INL and ONL) following treatment with SAC and/or CBD and metformin, despite the absence of apparent diabetes-related changes. Research on the diabetic changes in INL and ONL thicknesses is inconclusive, and these sub-layers may not be prime targets of diabetic neurodegeneration [68], [69]. Additionally, our ex vivo assessments of the retina revealed a lower incidence of BRB leakage compared to previous studies in the same model [70]. These disparities may be explained by variations in immunofluorescence markers of BRB leakage and the significant differences in experimental timelines.

It is important to acknowledge the limitations of our study. The use of a single experimental time-point precludes the temporal characterisation of CNS injury in T2D. Furthermore, the assessment of short-term memory using the NOR test was hindered by reduced locomotor activity in diabetic mice, especially in the SAC and CBD monotherapy groups, and these results should be interpreted cautiously. Future studies should consider a broader panel of cognitive tests and their suitability for the db/db obese phenotype. Regarding ex vivo measures, the assessment of neuroinflammation could have been strengthened by additional examination of proinflammatory cytokines and/or other secretary products of microglia and reactive astrocytes. Similarly, the retina could be assessed for additional measures of neurodegeneration, such as expression of TUNEL or apoptotic-positive cells.

The phenotype of experimental and clinical T2D includes hypertension, systemic inflammation, and peripheral vascular abnormalities. While these measures extend beyond the scope of our study, they require investigation to elucidate the broader effects of SAC and CBD in therapeutic management of T2D.

In conclusion, our study enhances the understanding of T2D-related complications in the CNS. The findings demonstrate that SAC and CBD are neuroprotective, with the capacity to prevent neurovascular lesions, glial activation, neurodegeneration, and oxidative injury. These effects contribute to preserving memory function and support the use of naturally derived agents in diabetes therapy.

4.1 Ethics statement

The experimental procedures conducted in this study were approved by the Curtin Animal Ethics Committee (approval no. ARE 2018-19). This research was carried out in full compliance with the Australian Code for the Care and Use of Animals for Scientific Purposes (8th Edition, 2021 updated version), and in adherence to Curtin University’s strict ethical standards and the ARRIVE guidelines for reporting experiments involving animals. Efforts were made to minimise animal suffering and to reduce the number of animals used, consistent with ensuring adequate statistical power. Our data analysis procedures strictly followed ethical standards, and the findings are reported with full transparency, including a clear acknowledgment of any limitations.

4.2 Animals and treatment intervention

Male C57BLK/6J mice bearing spontaneous monogenetic mutations in the leptin receptor (db/db), along with their heterozygote db/+ counterparts, were acquired from the Jackson Laboratory (California, USA) and maintained at the Animal Resource Centre (Perth, Western Australia). At 4 weeks, mice were transferred to the Animal Facility at Curtin University (Perth, Western Australia. Following 7 d of acclimatisation, animals were randomised to dietary intervention group for 23 weeks, which corresponded to a 28-week experimental endpoint, respectively. The groups were: 1) nondiabetic db/+ controls, 2) diabetic db/db controls, 3) SAC, 4) CBD, 5) SAC + CBD, and 6) metformin. Treatments were incorporated into AIN93M-base chow and thoroughly mixed to homogeneity. To ensure stability, diets were prepared under low-light conditions, cold-pressed into pellets, and packed in 1.5 kg air-sealed bags (Specialty Feeds; Perth, Western Australia). Stock diets were stored in the dark at 4°C.

Nondiabetic db/+ controls and diabetic db/db controls were fed standard chow. Additional diabetic and nondiabetic controls were maintained on standard chow until a 14-week experimental timepoint, and these animals were used for correlation analyses. For the treatment groups, diets were formulated based on previous evidence of food intake (5–6 g per day) and average body weight of 48g for db/db mice for the duration of experiments [27, 28]. SAC (50 mg/kg; 0.04% w/w) was purchased in powder form from ChemScene (New Jersey, USA) and CBD (75 mg/kg; 0.06% w/w) was prepared as a > 99% extract and then solubilised at a concentration of 14.5% in Miglyol 812N (medium chain triglyceride carrier oil), and generously provided by Zelira Therapeutics, (Perth, Western Australia). A combination of SAC + CBD (0.04% and 0.06% w/w, respectively) was prepared in the same manner as monotherapy groups, and metformin hydrochloride (200 mg/kg; 0.16% w/w) was obtained in tablet form with vet-approval (Perth, Western Australia). The dosages of these biomolecules and metformin have been previously published with demonstrated therapeutic effects [20, 29–32]. Diets were measured and replenished weekly, and animals had ad libitum access to food and water. Body weight was tracked weekly and immediately prior to euthanasia. Animals were housed in groups within a temperature-controlled laboratory on a 12 h light/dark cycle.

4.3 Cognitive Assessment

4.3.1 Novel Object Recognition

The novel object recognition (NOR) test was performed to assess non-spatial recognition memory via cortical-hippocampal neural circuitry as outlined [37] with slight modifications. The test took place in a dedicated behavioral testing room under dim lighting, and a video tracking system was used for automated analysis via the HVS Image 2014 software (HVS Image, UK). On the first day of the procedure, mice were allowed a 10 min habituation period in an empty experimental arena (dimensions: 45 cm x 45 cm x 40 cm). On the second day, mice were placed into to the arena, which had two identical cubic objects positioned diagonally 10 cm from the arena wall. After a 10-minute familiarisation period, the mice were returned to their home cage. The familiarisation threshold was 10 s of active exploration per object. Failure to meet the cut-off resulted in exclusion from the NOR test. To assess short-term memory, one of the previously encountered objects was randomly replaced with a novel, curved object that was distinctly different and unfamiliar to the mice. Following a 2 h delay, mice were placed back into the arena for 5 min of object exploration referred to as the ‘test phase’. Exploration was regarded as sniffing or rearing towards an object. A 70% (v/v) ethanol solution was used following each experiment to eliminate odour cues. The Preference Index (PI) was quantified as the proportion of time spent exploring the novel object to the total exploration time for both objects during the 5 min test period, as shown with the formula: Preference index (PI) = Novel object[s]/ (Novel object[s] + familiar object[s] x 100. Mice were not tested for 24 h after the NOR test.

4.3.2 Passive Avoidance

The step-through Passive Avoidance test was used to evaluate long-term memory function, using a previously described protocol from our lab [38] with minor adjustments (Ugo Basile, Italy). On the first day, mice were placed in the bright compartment of the apparatus, with a light intensity of 1000 lux. After a delay of 30 s, the door to the dark chamber was opened and upon entry, mice received a 2 s, 0.3 mA shock to establish an aversive memory response. The total experimental duration was 300 s. Post shock, the animals were kept in the dark compartment for an additional 10 seconds before being transferred back to their home cages. Mice that did not enter the darker chamber within the 300 s experimental window were returned to their home cages and subsequently excluded from the assessment. On the second day, after a 24 h interval, mice were reintroduced to the brightly lit chamber under identical experimental conditions. However, the experiment was extended to 345 s, and mice that retreated into the dark chamber were not given a shock. Latency was calculated as the it took the mice to enter the dark chamber on the second ‘test’ day minus the time it took on the first ‘training’ day.

4.4 Optical Coherence Tomography

Retinal imaging was carried out according to a previously published protocol [39] with minor amendments. Mice were anaesthetised using ketamine/xylazine (90 mg/kg ketamine, 12.5 mg/kg xylazine). Eye drops of 1% tropicamide and lignocaine (Akorn, Inc.) to induce pupil dilation. Retinal imaging was performed on the Bioptigen optical coherence tomography (OCT) system (Bioptigen, Morrisville, NC USA). Volume scans were captured from the optic nerve head (ONH) to the peripheral area of the retinal at the posterior pole of the eye. Cross-sectional B-scan images were stored for subsequent analyses. Measurements were taken of retinal thickness in the following areas: 1) total neuroretina, 2) ganglion cell complex (inclusive of the nerve fibre layer, retinal ganglion cell layer and inner plexiform layer), and 3) inner and outer nuclear layers. Mean thicknesses around the ONH were averaged from 6 measurement points in nasal-temporal regions using FIJI software (ImageJ, US) [40].

4.5 Euthanasia and sample collection

Immediately after retinal OCT imaging, blood was collected from deeply anaesthetised mice via cardiac puncture. Mice were then euthanised by cervical dislocation. The brain was removed from cranium and bisected, with the left hemisphere snap-frozen in liquid nitrogen and stored at -80°C and the right hemisphere fixed in 4% paraformaldehyde overnight then cryopreserved in 20% sucrose for 3 d and frozen using isopentane/dry ice for storage at -80°C. The right eye was also removed from the head and fixed and cryopreserved as indicated for the above, followed by storage in 0.1% sodium azide at 4°C. Brains were embedded in Tissue-Tek Optimal Cutting Temperature compound and coronal sections at 20 µm with cortex and hippocampal formation evident were prepared onto Poly-Lysine coated slides (Tranjan) using the Leica CM1520 Cryostat.

4.6 Plasma measures for glucose, triglycerides, and insulin resistance

Blood samples were collected in EDTA-coated syringes and subsequently transferred to 1.5 ml Eppendorf tubes for centrifuge at 4°C for 10 min at 4000 rpm. The resultant plasma supernatant was aliquoted and preserved at − 80°C for metabolic analyses. Non-fasting plasma glucose levels were determined at an optimised dilution ratio of 1:100 using a commercially available colorimetric assay kit (Abcam). In addition, insulin levels were evaluated using the Ultrasensitive Mouse Insulin ELISA kit (Mercodia), while plasma triglycerides were quantified using a colorimetric assay (Randox). All plasma measures were analysed in accordance with the manufacturers' instructions.

Insulin resistance was calculated using Homeostasis model assessment–insulin resistance (HOMA-IR) with the HOMA Calculator version 2.2.3 (Diabetes Trials Unit), and Triglyceride-glucose (TyG) index according to formula described below:

ln [triglycerides (mg/dL) × glucose (mg/dL)/2]

4.7 Brain immunofluorescence

The BBB integrity was assessed using a validated protocol for measuring extravasated immunoglobulin G (IgG) in brain parenchyme [41]. Brain sections were co-stained with laminin α4 to indicate the capillary basement membrane and enhance microvascular boundary delineation. To summarise, left hemisphere 20 µm coronal sections were fixed with 4% paraformaldehyde for 10 min at 20°C. To prevent non-specific binding, sections were blocked using 10% donkey serum (ThermoFisher) for 30 min at 20°C. The sections were washed in 0.1 M phosphate buffer saline (PBS) and incubated with goat anti-laminin α4 (1:200, RnD Systems) in an antibody signal enhancer solution overnight at 4°C. Sections were washed in PBS and incubated with donkey anti-goat IgG AlexaFluor 555 followed by goat anti-mouse IgG AlexaFluor 647 mixed in PBS, with each secondary incubation lasting 2 h at 20°C.

Neuroinflammation represented as reactive gliosis and oxidative DNA damage were evaluated as previously reported [38]. In brief, the right hemisphere 20 µm fixed cryosections of the right hemisphere were rehydrated in PBS, and non-specific binding sites were blocked in 10% donkey serum (ThermoFisher) for 30 min at 20°C. Sections were incubated overnight at 4°C with either rabbit anti-ionized calcium-binding protein 1 (1:200, Iba1; Novachem) for microglia or a combination of goat anti-glial fibrillary acidic protein (1:500, GFAP; Abcam) for astrocytes and mouse anti-8-Hydroxyguanosine (1:500, 8-OHdG (15A3); Abcam) for DNA oxidation. Thereafter, sections were incubated with either goat anti-rabbit IgG AlexaFluor 488 (1:500, ThermoFisher) or a combination of donkey anti-goat IgG AlexaFluor 555 (1:1000, ThermoFisher) and 1:500 donkey anti-Mouse IgG AlexaFluor 647 (ThermoFisher) for 2 h at 20°C. DAPI was used as a nuclei counterstain.

The cortex and hippocampal formation were captured at 20x magnification (imaging specifications include a OrcaFlash 4.0 camera; Colibri 7 LED-modules 385, 475, 555 and 630nm with penta-band filter set 112HE) on the Zeiss Axioscan Z1 slide scanner (Carl Zeiss, Germany). Images were processed offline using the deep learning features on Zeiss Zen ZEN Blue 3.3 Desk software. To quantitate IgG leakage, the Intellesis trainable segmentation module with machine learning automation capabilities was used based on a protocol developed in-house [41]. Semi-quantitative analysis of protein immunoexpression was determined based on voxel intensity per volume for each region of interest.

4.8 Retina immunofluorescence

Retina immunofluorescence staining and imaging were carried out according to an optimised protocol developed in-house. Briefly, retinas were removed from the eye cup and transferred to 96 well-plates containing PBS. They were incubated with Tris-EDTA Buffer (10mM, pH 6.0), overnight at 37°C, then blocked and permeabilised in a buffer solution made up of 0.2% Tween20, 2% Triton X-100, 0.2% bovine serum albumin in PBS for 1 h at 20°C with gentle shaking. Retinas were washed in PBS and co-stained for microglia (1:500, Iba1; Novachem) and astrocytes (1:500, GFAP; Abcam) (1:500), overnight at 4°C with gentle shaking. On the second day of the protocol, retinas were incubated with donkey anti-goat IgG AlexaFluor 647 (1:1000, ThermoFisher) for 2 h at 20°C with gentle shaking. Retinas were washed with PBS and incubated with goat anti- mouse IgG AlexaFluor 488 (as the primary IgG stain to measure IgG extravasation) and goat anti-rabbit IgG AlexaFluor 546 (1:500, ThermoFisher) under identical conditions. Retinas were mounted onto Poly-Lysine coated slides (Tranjan) with the vitreous side facing upwards.

Retinal wholemounts were imaged at 20x magnification using the Dragonfly Confocal Microscope with the Sona-4.2B-6 sCMOS camera (Andor, Oxford Instruments). Fluorescence micrographs were captured with the following settings for emission wavelengths: IgG – 521/38 nm, 42.5 ms exposure, 24% laser intensity; Iba1–594/43 nm, 42.5ms exposure, 24% laser intensity; GFAP – 685/47 nm, 42.5ms exposure, 18% laser intensity. De-identified retinal images were collected as .ims files and batch converted to maximum projection.tiff files using FIJI software (ImageJ, US). Semi-quantitative analyses of GFAP and Iba1 immunointensities were conducted using ZEN Blue 3.3 Desk software. The presence of extravascular IgG was first assessed in Imaris software (Oxford Instruments) with thorough examination the 40 µm z-stack range for each image, and later corroborated on the max projection .tiff files using ZEN Blue 3.3 Desk software. The boundaries of IgG leakage in the tissue parenchyma were determined manually by two researchers blinded to animal experimental groups.

4.9 Statistical Analysis

Statistical analyses were performed in GraphPad Prism 9 (U.S.A). Data are represented as mean ± SEM and significance was considered at P < 0.05. The D’Agostino–Pearson omnibus was used to verify data normality. For normally distributed data, parametric one-way analysis of variance (ANOVA) followed by Fisher’s least significant difference (LSD) post-hoc test was employed for pairwise comparisons. Correlations were examined using Pearson’s coefficient was used to analyse two associations: 1) between BBB integrity and cerebral inflammatory/oxidative markers, and 2) between cognitive outcomes and cerebral pathophysiological measures.

8OHdG, 8-Hydroxyguanosine; BBB, blood brain barrier; BRB, blood retinal barrier; CBD, cannabidiol; CNS, central nervous system; CTX, cortex; GCC, ganglion cell complex; GFAP, glial fibrillary acidic protein; HPF, hippocampal formation; HOMA-IR, Homeostasis model assessment–insulin resistance; IgG, immunoglobulin G; Iba1, ionized calcium-binding protein 1; INL, inner nuclear layer; ONH, optic nerve head; NVU, neurovascular unit; ONL, outer nuclear layer; OCT, optical coherence tomography; SAC, S-allyl-cysteine; TyG, Triglyceride-glucose; T2D, Type 2 diabetes mellitus

Additional information

The study was partially financially supported by Zelira Therapeutics Ltd. This does not alter our adherence to Scientific Reports policies on sharing data and materials. The authors declare no other competing interests.

Funding

The project and some authors were financially supported by the National Health and Medical Research Council of Australia (GNT1135590). The Andor Dragonfly microscope suite and the Zeiss Axioscan Z.1 digital slide scanner located at Curtin Health and innovation Research Institute were funded by the Australian Research Council under grant LE200100122.

Acknowledgments

The authors would like to acknowledge the facilities and technical assistance of Building 300, Curtin Animal Facility, and the Bioresources Facility (North) at Harry Perkins Institute of Medical Research, Australia for their support with animal maintenance. The authors also acknowledge the Microscopy and Histology Suite and technical officers at Curtin Health and Innovation Research Institute for training and expertise in tissue preparation and the use of microscope instruments. The protocol for preparation of retinal wholemounts was based on previous work by Professor Jennifer Rodger and Ms. Carole Bartlett from the University of Western Australia. Moreover, May Majimbi gratefully received support from an Australian Government Research Training Program (RTP) scholarship.

Author contributions

MM: Contributed to the study design, conducted comprehensive research, provided support in OCT imaging, performed data analysis and interpretation, and authored and edited the manuscript. JM: Contributed to the study design and results interpretation and provided primary manuscript edits. VL: Contributed to study design, provided expertise in animal maintenance, and contributed to manuscript edits. SM: Specialised in animal anaesthesia, collected OCT images, and participated in manuscript editing. MN: Assisted with sample collection, optimised immunofluorescence staining and analysis. EB: Contributed to study design, conducted preliminary literature review, and participated in manuscript editing. AS, AM, and HS: Equally contributed to manuscript editing. FC: Provided consultation and advice regarding the suitability of OCT imaging. RT: Provided oversight of the project, contributed to study design, assisted with data interpretation, and participated in manuscript editing. All authors approved the submitted version.

Data availability statement

The data supporting this study are available upon request. Please contact Ryu Takechi at [email protected] for access to the data.

V. Srikanth, A. J. Sinclair, F. Hill-Briggs, C. Moran, and G. J. Biessels, “Type 2 diabetes and cognitive dysfunction—towards effective management of both comorbidities,” Lancet Diabetes Endocrinol., vol. 8, no. 6, pp. 535–545, Jun. 2020, doi: 10.1016/S2213-8587(20)30118-2.
M. Bogush, N. A. Heldt, and Y. Persidsky, “Blood Brain Barrier Injury in Diabetes: Unrecognized Effects on Brain and Cognition,” J. Neuroimmune Pharmacol., vol. 12, no. 4, p. 593, Dec. 2017, doi: 10.1007/S11481-017-9752-7.
P. Van Dyken and B. Lacoste, “Impact of Metabolic Syndrome on Neuroinflammation and the Blood–Brain Barrier,” Front. Neurosci., vol. 12, Dec. 2018, doi: 10.3389/FNINS.2018.00930.
R. S. Allen et al., “Retinal Deficits Precede Cognitive and Motor Deficits in a Rat Model of Type II Diabetes,” Invest. Ophthalmol. Vis. Sci., vol. 60, no. 1, pp. 123–133, Jan. 2019, doi: 10.1167/IOVS.18-25110.
C. E. Hugenschmidt et al., “The cross-sectional and longitudinal associations of diabetic retinopathy with cognitive function and brain MRI findings: the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial,” Diabetes Care, vol. 37, no. 12, pp. 3244–3252, Dec. 2014, doi: 10.2337/DC14-0502.
M. Majimbi et al., “In vivo retinal imaging is associated with cognitive decline, blood-brain barrier disruption and neuroinflammation in Type 2 diabetic mice,” Front. Endocrinol. (Lausanne)., vol. 14, no. October, p. 1224418, 2023, doi: 10.3389/fendo.2023.1224418.
C. Moran et al., “Brain atrophy in type 2 diabetes: Regional distribution and influence on cognition,” Diabetes Care, vol. 36, no. 12, pp. 4036–4042, Dec. 2013, doi: 10.2337/DC13-0143/-/DC1.
J. M. Starr, J. M. Wardlaw, K. Ferguson, A. MacLullich, I. J. Deary, and I. Marshall, “Increased blood-brain barrier permeability in type II diabetes demonstrated by gadolinium magnetic resonance imaging,” J. Neurol. Neurosurg. Psychiatry, vol. 74, no. 1, pp. 70–76, 2003, doi: 10.1136/jnnp.74.1.70.
L. J. Min et al., “Peroxisome proliferator-activated receptor-γ activation with angiotensin ii type 1 receptor blockade is pivotal for the prevention of blood-brain barrier impairment and cognitive decline in type 2 diabetic mice,” Hypertension, vol. 59, no. 5, pp. 1079–1088, 2012, doi: 10.1161/HYPERTENSIONAHA.112.192401.
J. C. L. Mamo et al., “Probucol prevents blood-brain barrier dysfunction and cognitive decline in mice maintained on pro-diabetic diet,” Diabetes Vasc. Dis. Res., vol. 16, no. 1, pp. 87–97, Jan. 2019, doi: 10.1177/1479164118795274.
R. Takechi et al., “Blood-brain barrier dysfunction precedes cognitive decline and neurodegeneration in diabetic insulin resistant mouse model: An implication for causal link,” Front. Aging Neurosci., vol. 9, no. DEC, pp. 1–12, Dec. 2017, doi: 10.3389/fnagi.2017.00399.
S. H. Sinclair, E. Miller, K. S. Talekar, and S. S. Schwartz, “Diabetes mellitus associated neurovascular lesions in the retina and brain: A review,” Front. Ophthalmol., vol. 2, Oct. 2022, doi: 10.3389/fopht.2022.1012804.
S. Hachana, M. Pouliot, R. Couture, and E. Vaucher, “Diabetes-Induced Inflammation and Vascular Alterations in the Goto–Kakizaki Rat Retina,” https://doi.org/10.1080/02713683.2020.1712730, vol. 45, no. 8, pp. 965–974, Aug. 2020, doi: 10.1080/02713683.2020.1712730.
S. Yang, J. Zhang, and L. Chen, “The cells involved in the pathological process of diabetic retinopathy,” Biomed. Pharmacother., vol. 132, p. 110818, Dec. 2020, doi: 10.1016/J.BIOPHA.2020.110818.
J. H. Yun, S. W. Park, Y. J. Park, C. H. Cho, and J. H. Kim, “Angiopoietin 2 induces astrocyte apoptosis via αvβ5-integrin signaling in diabetic retinopathy,” Cell Death Dis. 2016 72, vol. 7, no. 2, pp. e2101–e2101, Feb. 2016, doi: 10.1038/cddis.2015.347.
B. Zhai et al., “Hypoglycemic and hypolipidemic effect of S-allyl-cysteine sulfoxide (alliin) in DIO mice,” Sci. Rep., vol. 8, no. 1, Dec. 2018, doi: 10.1038/S41598-018-21421-X.
J. Gorelick et al., “THC and CBD affect metabolic syndrome parameters including microbiome in mice fed high fat-cholesterol diet,” J. Cannabis Res., vol. 4, no. 1, p. 27, Dec. 2022, doi: 10.1186/S42238-022-00137-W.
C. P. Stanley, W. H. Hind, and S. E. O’Sullivan, “Is the cardiovascular system a therapeutic target for cannabidiol?,” Br. J. Clin. Pharmacol., vol. 75, no. 2, p. 313, Feb. 2013, doi: 10.1111/J.1365-2125.2012.04351.X.
J. N. Syu et al., “S-allylcysteine Improves Blood Flow Recovery and Prevents Ischemic Injury by Augmenting Neovasculogenesis,” Cell Transplant., vol. 26, no. 10, pp. 1636–1647, Oct. 2017, doi: 10.1177/0963689717724792/ASSET/IMAGES/LARGE/10.1177_0963689717724792-FIG6.JPEG.
H. Shi et al., “S-allyl cysteine activates the Nrf2-dependent antioxidant response and protects neurons against ischemic injury in vitro and in vivo,” J. Neurochem., vol. 133, no. 2, pp. 298–308, Apr. 2015, doi: 10.1111/JNC.12986.
B. R. Lins et al., “Cannabinoids in traumatic brain injury and related neuropathologies: preclinical and clinical research on endogenous, plant-derived, and synthetic compounds,” J. Neuroinflammation, vol. 20, no. 1, p. 77, 2023, doi: 10.1186/s12974-023-02734-9.
N. Nishiyama, T. Moriguchi, N. Morihara, and H. Saito, “Ameliorative Effect of S-Allylcysteine, a Major Thioallyl Constituent in Aged Garlic Extract, on Learning Deficits in Senescence-Accelerated Mice,” J. Nutr., vol. 131, no. 3, pp. 1093S-1095S, Mar. 2001, doi: 10.1093/JN/131.3.1093S.
T. Baluchnejadmojarad, Z. Kiasalari, S. Afshin-Majd, Z. Ghasemi, and M. Roghani, “S-allyl cysteine ameliorates cognitive deficits in streptozotocin-diabetic rats via suppression of oxidative stress, inflammation, and acetylcholinesterase,” Eur. J. Pharmacol., vol. 794, pp. 69–76, 2017, doi: 10.1016/J.EJPHAR.2016.11.033.
M. Zarezadeh, T. Baluchnejadmojarad, Z. Kiasalari, S. Afshin-Majd, and M. Roghani, “Garlic active constituent s-allyl cysteine protects against lipopolysaccharide-induced cognitive deficits in the rat: Possible involved mechanisms,” Eur. J. Pharmacol., vol. 795, pp. 13–21, 2017, doi: 10.1016/J.EJPHAR.2016.11.051.
M. Afarid, E. Sadeghi, M. Johari, E. Namvar, and F. Sanie-Jahromi, “Evaluation of the Effect of Garlic Tablet as a Complementary Treatment for Patients with Diabetic Retinopathy,” J. Diabetes Res., 2022, doi: 10.1155/2022/6620661.
H. Jiang et al., “Effects of cannabinoid (CBD) on blood brain barrier permeability after brain injury in rats,” Brain Res., vol. 1768, Oct. 2021, doi: 10.1016/J.BRAINRES.2021.147586.
M. Mecha, A. Feliú, P. M. Iñigo, L. Mestre, F. J. Carrillo-Salinas, and C. Guaza, “Cannabidiol provides long-lasting protection against the deleterious effects of inflammation in a viral model of multiple sclerosis: a role for A2A receptors,” Neurobiol. Dis., vol. 59, pp. 141–150, Nov. 2013, doi: 10.1016/J.NBD.2013.06.016.
A. B. El-Remessy, M. Al-Shabrawey, Y. Khalifa, N. T. Tsai, R. B. Caldwell, and G. I. Liou, “Neuroprotective and blood-retinal barrier-preserving effects of cannabidiol in experimental diabetes.,” Am. J. Pathol., vol. 168, no. 1, pp. 235–244, Jan. 2006, doi: 10.2353/AJPATH.2006.050500.
E. Brook et al., “Blood-brain barrier disturbances in diabetes-associated dementia: Therapeutic potential for cannabinoids,” Pharmacol. Res., vol. 141, pp. 291–297, Mar. 2019, doi: 10.1016/J.PHRS.2019.01.009.
Y. Xiong and C. S. Lim, “Understanding the Modulatory Effects of Cannabidiol on Alzheimer’s Disease,” Brain Sci., vol. 11, no. 9, Sep. 2021, doi: 10.3390/BRAINSCI11091211.
Y. Kondo et al., “Leprdb/db Mice with Senescence Marker Protein-30 Knockout (Leprdb/dbSmp30Y/–) Exhibit Increases in Small Dense-LDL and Severe Fatty Liver Despite Being Fed a Standard Diet,” PLoS One, vol. 8, no. 6, p. e65698, Jun. 2013, doi: 10.1371/JOURNAL.PONE.0065698.
Y. Teng, D. Li, P. Guruvaiah, N. Xu, and Z. Xie, “Dietary Supplement of Large Yellow Tea Ameliorates Metabolic Syndrome and Attenuates Hepatic Steatosis in db/db Mice,” Nutrients, vol. 10, no. 1, Jan. 2018, doi: 10.3390/NU10010075.
H. Zeinali, T. Baluchnejadmojarad, S. Fallah, M. Sedighi, N. Moradi, and M. Roghani, “S-allyl cysteine improves clinical and neuropathological features of experimental autoimmune encephalomyelitis in C57BL/6 mice,” Biomed. Pharmacother., vol. 97, pp. 557–563, Jan. 2018, doi: 10.1016/J.BIOPHA.2017.10.155.
W. H. De Oliveira et al., “Effects of metformin on inflammation and short-term memory in streptozotocin-induced diabetic mice,” Brain Res., vol. 1644, pp. 149–160, 2016, doi: 10.1016/j.brainres.2016.05.013.
J. T. Dearborn et al., “Effects of chronic cannabidiol in a mouse model of naturally occurring neuroinflammation, neurodegeneration, and spontaneous seizures,” Sci. Rep., vol. 12, no. 1, Dec. 2022, doi: 10.1038/s41598-022-15134-5.
K. Iffland and F. Grotenhermen, “An Update on Safety and Side Effects of Cannabidiol: A Review of Clinical Data and Relevant Animal Studies,” Cannabis Cannabinoid Res., 2017, doi: 10.1089/can.2016.0034.
I. Piątkowska-Chmiel, M. Gawrońska-Grzywacz, Ł. Popiołek, M. Herbet, and J. Dudka, “The novel adamantane derivatives as potential mediators of inflammation and neural plasticity in diabetes mice with cognitive impairment,” Sci. Reports 2022 121, vol. 12, no. 1, pp. 1–18, Apr. 2022, doi: 10.1038/s41598-022-10187-y.
V. Lam et al., “Synthesis of human amyloid restricted to liver results in an Alzheimer disease–like neurodegenerative phenotype,” PLOS Biol., vol. 19, no. 9, p. e3001358, Sep. 2021, doi: 10.1371/JOURNAL.PBIO.3001358.
S. McLenachan, D. Zhang, E. Hao, L. Zhang, S. C. Chen, and F. K. Chen, “Human limbal neurospheres prevent photoreceptor cell death in a rat model of retinal degeneration,” Clin. Experiment. Ophthalmol., vol. 45, no. 6, pp. 613–624, Aug. 2017, doi: 10.1111/CEO.12940.
J. Schindelin et al., “Fiji: an open-source platform for biological-image analysis,” Nat. Methods 2012 97, vol. 9, no. 7, pp. 676–682, Jun. 2012, doi: 10.1038/nmeth.2019.
M. Nesbit, J. C. Mamo, M. Majimbi, V. Lam, and R. Takechi, “Automated Quantitative Analysis of ex vivo Blood-Brain Barrier Permeability Using Intellesis Machine-Learning,” Front. Neurosci., vol. 15, no. April, pp. 1–8, 2021, doi: 10.3389/fnins.2021.617221.
M. Hashimoto, T. Nakai, T. Masutani, K. Unno, and Y. Akao, “Improvement of Learning and Memory in Senescence-Accelerated Mice by S-Allylcysteine in Mature Garlic Extract,” Nutrients, vol. 12, no. 6, p. 1834, Jun. 2020, doi: 10.3390/nu12061834.
F. da S. Rodrigues et al., “Cannabidiol treatment improves metabolic profile and decreases hypothalamic inflammation caused by maternal obesity.,” Front. Nutr., vol. 10, p. 1150189, 2023, doi: 10.3389/fnut.2023.1150189.
S. Takemura et al., “S Allyl cysteine improves nonalcoholic fatty liver disease in type 2 diabetes Otsuka Long Evans Tokushima Fatty rats via regulation of hepatic lipogenesis and glucose metabolism,” vol. 53, no. 2, pp. 94–101, 2013, doi: 10.3164/jcbn.13.
P. Bielawiec, E. Harasim-Symbor, and A. Chabowski, “Phytocannabinoids: Useful Drugs for the Treatment of Obesity? Special Focus on Cannabidiol,” Front. Endocrinol. (Lausanne)., vol. 11, p. 114, Mar. 2020, doi: 10.3389/FENDO.2020.00114.
M. Foretz, B. Guigas, L. Bertrand, M. Pollak, and B. Viollet, “Metformin: From mechanisms of action to therapies,” Cell Metab., vol. 20, no. 6, pp. 953–966, 2014, doi: 10.1016/j.cmet.2014.09.018.
X. M. Rodríguez Mesa, A. F. Moreno Vergara, L. A. Contreras Bolaños, N. Guevara Moriones, A. L. Mejiá Piñeros, and S. P. Santander González, “Therapeutic Prospects of Cannabinoids in the Immunomodulation of Prevalent Autoimmune Diseases,” Cannabis cannabinoid Res., vol. 6, no. 3, pp. 196–210, Jun. 2021, doi: 10.1089/CAN.2020.0183.
Y. Kosuge, “Neuroprotective mechanisms of S-allyl-L-cysteine in neurological disease (Review),” Exp. Ther. Med., pp. 1565–1569, 2019, doi: 10.3892/etm.2019.8391.
N. Corem, S. Anzi, S. Gelb, and A. Ben-Zvi, “Leptin receptor deficiency induces early, transient and hyperglycaemia-independent blood-brain barrier dysfunction,” Sci. Rep., vol. 9, no. 1, Dec. 2019, doi: 10.1038/s41598-019-39230-1.
C. Yan et al., “Dysfunction of the neurovascular unit in diabetes-related neurodegeneration,” Biomed. Pharmacother., vol. 131, p. 110656, 2020, doi: 10.1016/j.biopha.2020.110656.
K. Luc, A. Schramm-Luc, T. J. Guzik, and T. P. Mikolajczyk, “Oxidative stress and inflammatory markers in prediabetes and diabetes,” Journal of Physiology and Pharmacology, vol. 70, no. 6. Polish Physiological Society, Dec. 01, 2019, doi: 10.26402/jpp.2019.6.01.
A. L. Colín-González et al., “Aged Garlic Extract Attenuates Cerebral Damage and Cyclooxygenase-2 Induction after Ischemia and Reperfusion in Rats,” Plant Foods Hum. Nutr., vol. 66, no. 4, pp. 348–354, Nov. 2011, doi: 10.1007/S11130-011-0251-3/METRICS.
E. Kozela, A. Juknat, and Z. Vogel, “Modulation of Astrocyte Activity by Cannabidiol, a Nonpsychoactive Cannabinoid,” Int. J. Mol. Sci., vol. 18, no. 8, Aug. 2017, doi: 10.3390/IJMS18081669.
N. K. Acharya et al., “Retinal pathology is associated with increased blood-retina barrier permeability in a diabetic and hypercholesterolaemic pig model: Beneficial effects of the LpPLA2 inhibitor Darapladib,” Diabetes Vasc. Dis. Res., vol. 14, no. 3, pp. 200–213, May 2017, doi: 10.1177/1479164116683149.
H. M. Chao, I. L. Chen, and J. H. Liu, “S-allyl L-cysteine protects the retina against kainate excitotoxicity in the rat,” Am. J. Chin. Med., vol. 42, no. 3, pp. 693–708, 2014, doi: 10.1142/S0192415X14500451.
H. Javed et al., “S-allyl cysteine attenuates oxidative stress associated cognitive impairment and neurodegeneration in mouse model of streptozotocin-induced experimental dementia of Alzheimer’s type,” Brain Res., vol. 1389, pp. 133–142, May 2011, doi: 10.1016/J.BRAINRES.2011.02.072.
S. A. Farr, M. N. Goodland, M. L. Niehoff, and B. J. Young, “Cannabidiol (CBD) Improves Cognition and Decreases Anxiety in the SAMP8 Mouse Model of Alzheimer’s,” FASEB J., vol. 34, no. S1, p. 1, Apr. 2020, doi: https://doi.org/10.1096/fasebj.2020.34.s1.09741.
J. Peng, M. Fan, C. An, F. Ni, W. Huang, and J. Luo, “A narrative review of molecular mechanism and therapeutic effect of cannabidiol (CBD),” Basic Clin. Pharmacol. Toxicol., vol. 130, no. 4, pp. 439–456, Apr. 2022, doi: 10.1111/BCPT.13710.
M. O. da Silva et al., “Early neurovascular retinal changes detected by swept-source OCT in type 2 diabetes and association with diabetic kidney disease,” Int. J. Retin. Vitr., vol. 7, no. 1, pp. 1–9, Dec. 2021, doi: 10.1186/S40942-021-00347-Z/FIGURES/2.
V. Pilipenko, K. Narbute, J. Pupure, I. K. Langrate, R. Muceniece, and V. Kluša, “Neuroprotective potential of antihyperglycemic drug metformin in streptozocin-induced rat model of sporadic Alzheimer’s disease,” Eur. J. Pharmacol., vol. 881, p. 173290, Aug. 2020, doi: 10.1016/J.EJPHAR.2020.173290.
F. Ping, N. Jiang, and Y. Li, “Association between metformin and neurodegenerative diseases of observational studies: Systematic review and meta-analysis,” BMJ Open Diabetes Res. Care, vol. 8, no. 1, Jul. 2020, doi: 10.1136/bmjdrc-2020-001370.
P. Picone et al., “Metformin increases APP expression and processing via oxidative stress, mitochondrial dysfunction and NF-κB activation: Use of insulin to attenuate metformin’s effect,” Biochim. Biophys. Acta, vol. 1853, no. 5, pp. 1046–1059, May 2015, doi: 10.1016/J.BBAMCR.2015.01.017.
Z. H. Li, Y. Y. Jiang, C. Y. Long, Q. Peng, and R. S. Yue, “The gut microbiota-astrocyte axis: Implications for type 2 diabetic cognitive dysfunction,” CNS Neurosci. Ther., vol. 29, no. September 2022, pp. 59–73, 2023, doi: 10.1111/cns.14077.
J. H. Zhang, X. Y. Zhang, Y. Q. Sun, R. H. Lv, M. Chen, and M. Li, “Metformin use is associated with a reduced risk of cognitive impairment in adults with diabetes mellitus: A systematic review and meta-analysis,” Front. Neurosci., vol. 16, Aug. 2022, doi: 10.3389/FNINS.2022.984559/FULL.
C. Iadecola, “The Neurovascular Unit Coming of Age: A Journey through Neurovascular Coupling in Health and Disease,” Neuron, vol. 96, no. 1, pp. 17–42, 2017, doi: 10.1016/j.neuron.2017.07.030.
M. R. Hayden, “Hypothesis: Astrocyte foot processes detachment from the neurovascular unit in female diabetic mice may impair modulation of information processing—six degrees of separation,” Brain Sci., vol. 9, no. 4, Apr. 2019, doi: 10.3390/brainsci9040083.
M. Vargas-Soria, M. García-Alloza, and M. Corraliza-Gómez, “Effects of diabetes on microglial physiology: a systematic review of in vitro, preclinical and clinical studies,” J. Neuroinflammation, vol. 20, no. 1, Dec. 2023, doi: 10.1186/S12974-023-02740-X.
F. Ishibashi and M. Tavakoli, “Thinning of Macular Neuroretinal Layers Contributes to Sleep Disorder in Patients With Type 2 Diabetes Without Clinical Evidences of Neuropathy and Retinopathy,” Front. Endocrinol. (Lausanne)., vol. 11, p. 69, Feb. 2020, doi: 10.3389/FENDO.2020.00069/BIBTEX.
S. Vujosevic and E. Midena, “Retinal Layers Changes in Human Preclinical and Early Clinical Diabetic Retinopathy Support Early Retinal Neuronal and Müller Cells Alterations,” J. Diabetes Res., vol. 2013, 2013, doi: 10.1155/2013/905058.
A. K. H. Cheung et al., “Aldose Reductase Deficiency Prevents Diabetes-Induced Blood-Retinal Barrier Breakdown, Apoptosis, and Glial Reactivation in the Retina of db/db Mice,” 2005.

No competing interests reported.

Download PDF

Editor invited by journal
20 Mar, 2024
Submission checks completed at journal
18 Mar, 2024
First submitted to journal
28 Feb, 2024

You are reading this latest preprint version

S-allyl Cysteine and Cannabidiol are Equally Effective as Metformin in Preserving Neurovascular Integrity, Retinal Structure, and Cognitive Function in db/db Type 2 Diabetic Mice

Status:

Version 1

Abstract

Figures

1 Introduction

2 Results

2.1 SAC and CBD attenuate markers of metabolic dysregulation linked to diabetes

2.2 SAC and CBD mitigate diabetic neuroinflammation and oxidative stress, and synergistically protect BBB integrity

2.3 Natural compounds prevented diabetes-associated thinning in the inner retina

2.4 Diabetic astrogliosis and neuroinflammation in the retina were prevented with natural compounds and metformin.

2.5 SAC and CBD improve memory impairment with greater efficacy than metformin

2.6 Diabetic metabolic status is associated with pathological changes in the brain and retina

3 Discussion

4 Materials and methods

4.1 Ethics statement

4.2 Animals and treatment intervention

4.3 Cognitive Assessment

4.3.1 Novel Object Recognition

4.3.2 Passive Avoidance

4.4 Optical Coherence Tomography

4.5 Euthanasia and sample collection

4.6 Plasma measures for glucose, triglycerides, and insulin resistance

4.7 Brain immunofluorescence

4.8 Retina immunofluorescence

4.9 Statistical Analysis

Abbreviations

Declarations

Additional information

Funding

Acknowledgments

Author contributions

Data availability statement

References

Additional Declarations

Status:

Version 1