One proposed mechanism of vascular injury associated with diabetes is through impairment in the regulation of the transient receptor potential cation channel (TRPC). Under physiologically healthy conditions, the precapillary arterial wall responds to high pressure by upregulating TRPC, triggering reactive vasoconstriction. However, continuous stimulation ultimately may result in poor flow autoregulation to the variation of the stimulant (141). Both aging and deficiencies in insulin-like growth factor-1 (IGF-1) in the presence of hypertension have been shown to impair upregulation of TRPC within the retina (142, 143). The loss of these protective autoregulatory processes leaves the microcirculation vulnerable to the damage caused by the variation of pressures within the microvasculature resulting in increased blood-brain barrier leakage and neuroinflammation (109, 144). There is also a gradual accumulation within the microvasculature of multiple molecular fractures of the intima and internal elastic lamina. The accumulation of such patches eventually transforms the elastic lamina from a fully elastic, homogeneous structure to a stiffened, friable wall, resulting in a reduction in vascular compliance. The resulting array of structural degenerative events in the vessel wall, including death of endothelial cells, basal membrane thickening, and atrophy of the smooth muscle cells (145, 146) can all lead to rupture (microbleeds), microinfarcts, and loss of tight junctions with reduced blood brain-barrier integrity. In the retinal capillaries of the diabetic, leukocyte drag and adhesion occur due to leukocyte elevated stiffness and the upregulation of endothelial derived adhesion molecules (ICAMS and VCAMS) that adhere with the leukocyte surface integrins resulting in the adherence of the leukocyte to the endothelium and degradation producing focal microvascular leakage and occlusion (147–149). The resultant increased permeability of the blood-brain barrier can lead to interstitial accumulation of cytokine proteins with further worsening of inflammatory/oxidative stress events that damage the gray and white matter parenchyma. Pericytes and oligodendrocytes are extremely vulnerable to this ischemic and toxic insult (38) resulting, as well, in the impairment of the tissue repair mechanisms (150).

Diabetes mellitus, therefore, is now recognized to involve a complex, systemic, autoimmune, inflammatory disorder that causes focal microvascular occlusions and alterations of the blood-retinal and blood-brain barriers that occur in “pre-diabetic” as well as diabetic individuals (1, 3, 86). In the Diabetes Prevention Program (DPP), 7.9% of subjects with impaired glucose tolerance had retinopathy (151), similar to the 8.1% prevalence of retinopathy observed among individuals with prediabetes in the Gutenberg Health Study (152) with the variability of the glycemia a significant recognized risk factor for DR development and progression (153). HbA1c has been the primary criteria differentiating the diagnosis of prediabetes from diabetes, but it must be stressed that there are many well-characterized “pitfalls” of this including other systemic illness and hematological disorders that disrupt the reliability of HbA1c as an integrated measure of mean plasma glucose. Although DR or its progression is recognized to be related to HbA1c. However, studies across the globe have observed that the risk of many of the associated co-morbidities are the same in diabetics and prediabetics and affect all age groups indicating these are complicated interactions (4). Across the striae of hyperglycemia definitions, the increasing requirement for insulin, brought about by the progressive insulin resistance occurring in peripheral fat together with the pancreas Beta cell loss (both a product of immunologic reactions) is associated in itself, with aggravated risk of vascular induced retinal as well as cognitive dysfunction (3).

As discussed above, in the diabetic or prediabetic patient, there is chronic systemic low-grade inflammation (86) which is reflected by high levels of serum cytokines such as tumor necrosis factor-alpha (TNF-α), C reactive protein (CRP), and the interlukins, IL-6, iL-18, IL-1B and the receptor antagonist, C5a (154) along with granulocyte and monocyte elevations (140, 155). While this inflammatory state is thought to be the mechanism by which metabolic disorders, such as diabetes. are associated with the development of small vessel associated organ failure, there is significant variability of the cytokines that indicates a complex process (156–158) and which, therefore, will require more complicated design trials to define which will predict the development and progression (or regression) of the neuropathic lesions.

The loss within the retina of the ganglion cell layer neurons with hyperglycemia appears preceded by earlier loss of the microvascular pericytes and then endothelia, as discussed above, that results in loss of reactivity, with early vasodilation that is observed clinically along with early abnormal permeability of the endothelial barriers (42, 55). Chronic, variable hyperglycemia has also been suggested to cause an increase in release of glutamate with loss of neuroprotective factors due to oxidative stress with the accumulation of waste products of glycolysis that is thought to trigger the ganglion cell apoptosis (66). However, we must acknowledge that all of the supporting structures of the neurons, including the Müller cells (macroglia), microglia, and infiltrating monocytes have all been defined as critical actors in this neuroinflammatory process. Activated Müller cells mediate both protective and detrimental effects upon the ganglion cell layer neurons of the retina through a variety of receptors that result in the expression of multiple factors that affect neuronal survival including GFAP, vimentin (159, 160) and growth factors, such as brain-derived neurotrophic factor (BDNF) and platelet-derived growth factor-B (PDGF-B) (161). They also manifest a protective role by absorbing glutamate, reducing the cytotoxic effects of the extracellular glutamate levels (162), providing protection for neuronal growth, survival, and synaptic plasticity while reducing the oxidative stress (64, 163). Diabetes not only reduces the Muller uptake of extracellular glutamate (164) but also reduces the activity of the enzyme, glutamine synthetase, hindering the ability to convert the excess glutamate to inactive glutamine (165) or to oxidize the glutamate to a-ketoglutarate (162). Activated Müller cells, therefore serve not only as rapid sensors of neuronal damage to initiate repair and neural regeneration (166, 167), but they may also initiate a deleterious inflammatory process with the release of proinflammatory cytokines, such as activating transcription factor 4 (results in release of ICAM-1 and vascular endothelial growth factor (VEGF) (75, 168) as well as TNFα, interleukin-1 (IL-1), and other cytokines known to exacerbate the apoptosis of adjacent neurons (65, 161, 169). This dual action, therefore, is certainly a double-edged sword and should be taken into account in the design of therapeutic approaches addressed to abrogate Muller glial activation.

Microglia within the retina and brain are the principal immune effector cells, constantly surveying their environment in preparation to react to insult or injury. Normally phagocytic microglia should clear damaged myelin and allow neuronal repair to proceed. However, a chronic, pro-inflammatory state, such as that induced with hyperglycemia, alters the response, preventing remyelination (170). In human diabetic retinas, activated microglia are often observed to be associated with the vasculature, leading to the term “microglial perivasculitis”, and resulting in the release of numerous cytokines, including IL-6, IL-1B, TNFa, and monocyte chemoattractant protein-1(MCP-1) with several studies indicating these cytokines contribute to the microvascular permeability and occlusive complications as well as progressive neuronal apoptosis (64, 171). It should be noted, however, that activated microglia may also demonstrate a neuro-supportive function, similar to Mueller cells producing anti-inflammatory and neurotrophic factors, including IGF-1, BDNF, and GDNF (glial cell derived neurotrophic factor), among others (172–174).

In the brain, activated microglia and their supportive activities observed in periventricular WMHI’s (136), interestingly, have not been noted in the subcortical WMHI’s (121). Similarly, in the retina a variance has been demonstrated in the reactivity to the ischemia wherein the deeper portion of the microvascular complex, serving the middle nuclear neuronal layer, appears to provide a different reaction to the focal ischemia than within the more superficial ganglion cell layer. Ultimately, the detection of microglial activation may have value in early disease diagnosis, as modulation of microglial responses appears to perhaps provide the ability to alter disease progression. While receiving less attention than microglia, nonetheless mast cells within the brain represent an important source of immune signaling. Mast cells occur within the dura and meninges, and have demonstrated the ability to promote blood-brain barrier breakdown, edema, neutrophil infiltration, and hemorrhage in animal models of focal cerebral ischemia (175, 176). Yet another factor, synthesized, stored, and released by mast cells, and implicated in retinal and brain barrier breakdown is the angiogenic, vascular endothelial growth factor (VEGF) (175, 177) the treatment of which is discussed below.

Hyperglycemia and glucolipotoxicity-induced oxidative stress also drive intracellular changes that result in upregulation of proinflammatory mediators (86, 178, 179). These are generally produced by external toxins, nutrients, etc. that transduce within the cells to cause DNA methylation and histone modification leading to altered gene expression (178, 180). The mediators also cause alterations of microRNAs (miRNA), small, single-stranded non-coding RNA molecules that regulate post-transcriptional gene expression; such alterations may occur even as early as within the fetus of a pregnant diabetic and may accumulate throughout life with recurrent stress-induced influences (86, 181, 182). Such hyperglycemia induced dysregulation occurs within multiple tissues resulting in the established complications of diabetes, in particular endothelial dysfunction associated with retinopathy (183). In particular, miRNA-126, which occurs only within microvascular endothelia, seems to be the miRNA most linked to pathways in the development of both type 1 and 2 diabetes and the tissue complications; since miRNA-126 has been shown to play an important role in maintaining endothelial cell homeostasis and vascular integrity (184, 185), it has been suggested that this unique plasma miRNA might become a valuable tool to predict the micro- and macro-vascular complications. In addition, clinical series have demonstrated that elevated serum levels of multiple other miRNA’s, notably -7,-122, and -210, antedate the development of type 2 diabetes and have also been found significantly associated with the microvascular complications (183), suggesting the possibility they may represent potential targets for evaluation of their sensitivity for predicting and monitoring microvascular injury and repair.

Growth differentiation factor-15 (GDF-15), also known as macrophage inhibitory cytokine-1, has been demonstrated enhanced as a response to tissue ischemia (186). Investigations have shown a positive association between GDF-15 serum concentrations and diabetes risk of retinopathy with higher levels associated in proportion with the grade level of retinopathy (187). However, the mechanisms involved in the pathogenesis remain unclear, although possible explanations have suggested GDF-15 is involved in oxidative stress and endothelial dysfunction (187, 188) as well as in the inflammatory and immune processes (187, 189).

Glucagon-like peptide-1 (GLP-1) receptor agonists (e.g. Liraglutide, Semaglutide, Dulaglutide, Extendin-4) have been demonstrated to pass the blood-brain barrier wherein they affect the neural tissue through the basic anti-apoptotic, antioxidant, and neurotrophic effects of GLP-1 receptor activation (195). This results in activation of both the protein kinase A (PKA) and phosphatylinositol-3-kinase (PI-3K) pathways, the results of which influence insulin activation in adipose tissue and also result in the transcription of genes responsible for antioxidant, anti-apoptotic, neurotropic, and anti-inflammatory effects (195–197). GLP-1 receptor activation results in reduction of the accumulation of intracellular reactive oxygen species (ROS) within microglia with increased production of NO (improving vascular autoregulation) as well as increased levels of antioxidant glutathione peroxidase and superoxide dismutase-1 (196). Furthermore GLP-1, via its effect of decreasing caspase 3 & 7 activity, also inhibits microglial production of TNFa, IL-1B, and IL-6, resulting in reduced insulin resistance systemically and neurally (198, 199). While glucose utilization, in general within neurons, occurs primarily via insulin-independent glucose transporter-3 (GLUT-3), there is known to be considerable variation within the brain. The forebrain, cerebral cortex, and hippocampus show co-expression of the insulin-dependent glucose transporter-4 and within these areas GLP-1 receptor activation appears to decrease the insulin resistance and perhaps even reverse it. In animal models GLP-1 receptor activation has been demonstrated to improve cognitive functions by increasing glucose utilization (200). In clinical studies, however the results have been mixed, although the studies have remained small with varying individual traits that may cloud the recognition of efficacy (196, 201). The observed effects upon ischemic injury demonstrated within the studies appear predominantly within the penumbra of infarcts, and therefore future studies should be undertaken as the examination methods and understanding improve. Moreover, the new glucagon-like peptide-1 receptor (GLP-1R) combinations, developed together with the gastric inhibitory polypeptide (GIP) and glucagon agonists, also need well-planned clinical trials (196, 202).

The biguanides (e.g. Metformin) in animal studies have demonstrated glucose-reducing effects through activation of AMP activated protein kinase (AMPK), with reduction in clinical studies of proinflammatory cytokines due to inhibition of the activation of macrophage nuclear factor kappa B (κ) that results in reduction in IL-1B, IL-6, and TNFa (140). However, the clinical studies have been mixed in the demonstration of such inflammatory marker reductions (140). Some clinical studies have demonstrated reductions in the deposition of advanced glycosylated end products within vascular smooth muscle (203, 204), with the UKPDS demonstrating reduction in the risk of stroke. However, at best the evidence is conflicting, with similar conflicting results also of the alpha-glucosidase inhibitors in which modestly reduced CRP levels are associated with some improvements demonstrated in coronary flow, but no data with regard to neurovascular effects (140). Therefore, to understand better what effect the bioguanides may have upon the neuropathology requires further clinical testing (140).

The thiazolidinedione class of antidiabetic drugs are known to activate the peroxisome proliferator-activated receptor γ (PPARγ). Following many types of acute and chronic neural injury, PPARγ serves as a master gatekeeper of cytoprotective responses to stress, improving the chances of cellular survival and recovery of homeostatic equilibrium (205). In the retinal ganglion cells PPARγ activation has been demonstrated, in animal studies, to result in neuroprotection against the glutamate-induced cytotoxicity (206). In the acute injury phase PPARγ inhibits the NFκB pathway to mitigate inflammation and stimulates the Nrf2/ARE axis to reduce oxidative stress. During the chronic phase of acute brain injuries, PPARγ activation culminates in the repair of gray and white matter, with preservation of the blood-brain barrier, reconstruction of the neurovascular unit, resolution of inflammation, and with apparent long-term functional recovery (205). Rosiglitazone has demonstrated reduced inflammatory markers and superoxide anion production with inhibition of ubiquitin-proteasome activity in atherosclerotic plaques of T2DM patients (140), while Pioglitazone systemically has demonstrated reduced adipose tissue macrophage activity resulting in decreased pro-inflammatory cytokines (IL-6, IL-1B) in neutrophils, macrophages and dendritic cells of peripheral organs (140).

Regarding the clinical neurologic effects of the sulfonylureas, information does not appear supportive of a beneficial effect. The UK General Practice Research Database revealed that, compared with metformin monotherapy, treatment with the sulfonylureas resulted in a significant 24% greater stroke occurrence (207, 208), while the Veterans Administration Data Base demonstrated an 11% increased incidence (209). Analysis of a multitude of additional, randomized controlled trials has also revealed an increased stroke incidence compared with metformin, DPP-4 inhibitors, GLP-1 receptor agonists, glitazones or with just insulin (210). The Glinide group of drugs exhibit mechanisms similar to that of the sulfonylureas; Mitiglinide has demonstrated some reduction in oxidative stress levels and inflammatory markers IL-6, IL-18, and TNF-a in T2DM, but studies of clinical neurologic effects are lacking (140).

Dipeptidyl peptidase (DPP-4) inhibitors such as Sitagliptin have been demonstrated to reduce CRP, TNF-a, TLR-2, TLR-4, IKKB, and CCR-2 systemically with improvement in inflammation and endothelial function, independent of the hypoglycemic activity, in T2DM patients with associated coronary artery disease (140). The DPP-4 inhibitors, however, unlike the GLP-1 activators, cannot pass the BBB or BRB, but interestingly their effect within the central nervous system appears to be through associated increased endogenous GLP-1 levels (140). To date, no randomized prospective studies have been conducted on a sufficient number of humans to study the neurocognitive or retinal effects.

SGLT-2 inhibitors (SGLT-2i) or “Flozins” (Empagliflozin, Canagliflozin, Dapagliflozin, Ertugliflozin, Sotagliflozin)) represent a new class of systemic, oral treatment of hyperglycemia. Plasma glucose is lowered by inhibiting the reabsorption of glucose within the kidneys. Clinical studies have demonstrated that SGLT2i lead to a decrease in insulin secretion, enhanced beta-cell function and insulin sensitivity, with a demonstrated decrease of renin-angiotensin and sympathetic vasomotor system activity that has provided a demonstrated additive advantage for retinopathy (211).

The SGLT-2i are lipid-soluble with varying ability to penetrate the BRB and BBB. In addition they have variable affinity for the SGLT1 receptors as well, demonstrating protection against ischemia/reperfusion brain damage (211). SGLT2i show an anti-inflammatory and anti-atherosclerotic effect with inhibition of AchE, and as well, mitigate oxidative stress, exerting a protective effect on the neurovascular unit, blood-brain barrier, pericytes, astrocytes, microglia, and oligodendrocytes (211). Among the group, Sotagliflozin demonstrates the greatest affinity to SGLT1 receptors, and therefore is often called a “dual SGLT1/SGLT2 inhibitor”; however, it is the newest Flozin and has as yet not been used in diabetic patients on a large scale (211). Empaglifozin has the highest SGLT2 selectivity and greatest brain/serum ratio and has been demonstrated to significantly increase the level of cerebral BDNF, which modulates neurotransmission and ensures growth, survival, and plasticity of neurons (212). SGLT2i’s also in animal studies also demonstrate enhanced synaptogenesis and angiogenesis, con- tributing to neuroplasticity (213) a mechanism essential in neurorehabilitation.

The effects of SGLT1 versus SGLT2 inhibition in neuroprotection varies among the drugs evaluated and studies (214), as the sites of the receptors vary considerably within the brain parenchymal locations and demonstrate varied functional outcomes. Certainly, the efficacy requires further investigation and delineation, which is of increasing clinical relevance due to the availability of SGLT1/2 dual inhibitors, such as sotagliflozin, Furthermore there is now in vivo evidence indicating that SGLT2 may have an influence on the composition of the cerebrospinal fluid (CSF), whose role in the pathology of neurodegenerative disorders provides a new direction for research (215).

In conclusion, patients with insulin-resistant type2 diabetes as well as pre-diabetes are at increased risk and have inferior outcomes than non-insulin resistant or non-hyperglycemic subjects. Glycemic control achieved with the use of sulfonylureas or insulin was reported to increase the incidence of stroke in older observational studies but not in recent prospective, randomized, controlled trials. It is likely that an imbalance in the unmeasured clinical factors is responsible for the results in these observational studies. Insulin use itself may represent a more potent marker for insulin-resistance, and hence a better predictor of risk for stroke or progressive small vessel ischemic disease than HbA1c. The use of metformin, a-glucosidase inhibitors and DPP-4 inhibitors have thus far in studies demonstrated conflicting clinical evidence, as noted above while the utilization of pioglitazone and long acting GLP-1 receptor agonists have been associated with a decreased incidence of ischemic cerebral vascular disease. Once dementia due to small vessel disease has developed in diabetic patients, it appears not to be reversed by tighter glycemia control, but rather may be aggravated even more by the variability of the serum glucose levels and hypoinsulinemia as have been demonstrated within the retina (194). This has raised concern within some the few clinical studies of SGLT2i; although less likely to cause hypoglycaemia than insulin, they have been reported to be associated with increased hypoglycaemia and secondary neurologic AE’s when used in combination with metformin and sulfonylureas (216). Interestingly, intranasal insulin, has been proposed as an alternative adjunct therapy to improve cognitive function and memory (in doses that do not cause hypoglycemia), and certainly should be evaluated.

Increased oxidative stress is known to play a central role in the pathogenesis of many diseases such as atherosclerosis, vascular inflammation, and endothelial dysfunction and is thought to play a role in the progression of prediabetes to diabetes (140) as well as in the onset and progression of the neurovascular complications.

Nicotinamide adenine dinucleotide phosphate (NADPH) oxidases appear to play a central role in redox-signaling cascades, under normal physiologic conditions producing reactive oxygen species (ROS) primarily for functions of innate immunity and the production of necessary hormones (226). Because of their overarching role in ROS generation in the diabetes state, NADPH oxidases appear as an attractive target for future therapeutic strategies. Animal studies have demonstrated Resveratrol’s ability to suppress neuroinflammation by inhibiting NADPH oxidase and attenuating NF-κB-induced expression of iNOS (inducible nitric oxide synthase), COX-2 (cyclo-oxygenase-2, an inducible inflammatory enzyme), and sPLA2 (secretory phospholipase A2) (227–229). Resveratrol also has demonstrated some restoration of the microRNA functions discussed above, thus making it appear as an attractive treatment, although human studies of diabetics are currently lacking. The polyamine oxidase pathways, including a key enzyme, spermidine oxidase, also have been demonstrated in animal and human studies to be associated with lipid peroxidation and retinal neurovascular injury from oxidative stress with the production of toxic byproducts, such as acrolein, a highly reactive unsaturated aldehyde that is also found as a contaminant in food, air, and water. Vitreous samples in human studies (of FDP-lysine, a biomarker of acrolein, formed when acrolein conjugates with lysine residues) have demonstrated correlations with retinopathy but only with the more severe forms (230). Scavengers of such toxins appear to include a number of nitrogen (amino)-containing, current oral medications including hydralazine, carnosine, aminoguanidine, pyridoxamine, edaravone, and phenelzine. A potent acrolein scavenger, 2-hydrazino-4,6-dimethylpyrimidine (2-HDP), was observed to significantly decrease the presence of FDP-lysine in the retina of diabetic rats with reduced activation of microglia and inflammation markers and with attenuation of Müller cell gliosis, improved oxidative stress markers, and improved visual function (231). Certainly clinical studies focused on the impact of acrolein-scavenging agents are needed. While defense against oxidative stress is thought to be a major event involved in the inhibition of the harmful effects upon Müller cells and neurons, compounds with both antioxidant activity and such amino toxin-scavenging capacity would appear to have a greater impact (230).

Coenzyme Q10 (CoQ10) is an endogenous antioxidant within the inner membrane of the mitochondria of the endothelia and smooth muscle of the microvasculature. Recent animal studies have provided evidence for the potential pleiotropic and anti-inflammatory role of CoQ10 by demonstrating its ability to inhibit induced interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and nuclear factor-κB (NF-κB) expression (232, 233).

Reduced levels of CoQ10 in the aging and those with DM have been reported (234). CoQ10 supplementation influences the cholinergic system and in clinical studies appears to protect cholinergic neurons in patients with Alzheimer’s disease, as well as improving cardiovascular function (235). In addition, CoQ10 appears to reduce hyperglycemia (236–238), thereby protecting against endothelial abnormalities and dysfunction in DM (237) while reducing oxidative stress (237, 239). In a study of middle-aged, diabetic rats, evaluated cognitive function and learning memory were improved. However, in clinical studies of CoZ10 have suggested supplementation may improve learning and memory deficits induced by diabetes (240). However in reviewing patient evaluations, variations in the results were suggested to be because of penetration variability within various brain regions and pathways resulting in different types of learning, memory, and cognitive processing which would cause significant variation in the CNS function (240). Therefore, the CNS functional evaluation of CoQ10 supplementation remains to be adequately evaluated by clinical studies as extensively discussed above for the many systemic therapies.

The hyperglycemia induced increase in intracellular reactive oxygen species that leads to the secondary microvascular endothelia and pericyte alterations with leukocyte adhesion resulting in microvascular occlusion and endothelial injury, insulin resistance, protein/macromolecule glycation, and inhibition of NO synthesis. The NO inhibition, in turn, alters the microvascular reactivity required for the normal autoregulatory control of blood flow. In human diabetic retinas after administration of oral pentoxifylline an interesting phenomenon was observed of improvement in the capillary WBC velocity (148), although the mechanism, whether through improved WBC deformability or from reduced endothelial production of ICAM’s and VCAM’s, is unknown as well as the effects on the BRB and microvascular occlusion. Pentoxifylline has demonstrated a beneficial effect overall of reduced systemic inflammatory markers (241), but the clinical effects within the retina and brain remain to be investigated.

In the diabetic or prediabetic patient, as discussed above, there is pervasive, chronic systemic low-grade inflammation which is reflected by elevated levels of serum cytokines such as TNF-α, CRP, IL-6, iL-18, IL-1B receptor antagonist, and C5a (140, 154, 242) along with elevated serum WBC’s (granulocytes, monocytes) and elevated ICAM-1 and VCAM-1 but with a reduction in the protective effects of brain-derived neurotrophic factor (BDNF) (155). While this systemic inflammatory state is noted to be associated with the development and progression of retinopathy (243), of cerebral WMHI’s (244) and small vessel disease heart failure (140), the mechanisms remain to be defined as, indeed, the etiology is complex (245). The dysfunction of the neurovascular unit in diabetes includes several intracellular signaling cascades resulting in proinflammatory cascades that can either be helpful, neutral or detrimental to cell survival. Since there is now a recognized mutual relationship between systemic inflammation and several metabolic reactions that occur within the retina and CNS, interest is developing to improve the effects of immunomodulatory agents upon the disease process of the microvasculature (140, 246). For example, IL-1β has been demonstrated to induce pericyte apoptosis via NF-κB activation with increased endothelial permeability (84, 171) and occlusive degeneration of retinal capillaries in DR (84, 247). Therefore, limiting the IL-1β-triggered inflammatory processes, perhaps by blocking its receptor could offer a valid therapeutic approach.

A new method of potentially limiting the end results of systemic inflammation is through modification of the semaphorin molecular signaling cascades. The semaphorins are extracellular signaling molecules with recent animal study evidence indicating the role in regulating adipogenesis, adipose inflammation, and diabetic complications including microvascular permeability in multiple organs (248). Within the five semaphorin classes in vertebrates, the Class-3 family (Sema3A-3G) appear the most pertinent. Sema3E has been identified in a dietary mouse model as a regulator of adipose tissue macrophage accumulation in obesity with inflammation that contributes to systemic insulin resistance (249). Systemic antibody inhibition to Sema3E markedly reduced the tissue inflammation and improved insulin resistance, suggesting that a Sema3E peptide vaccine may provide a therapeutic potential for obesity as well as modifying diabetic complications in general.

In addition, several semaphorins have been noted to be involved directly in the retinopathy complications. Sema3A is induced in ischemic retinal ganglion cells in response to IL-1 after vascular injury in animal models (250) and, promoting endothelial apoptosis (251), appears to prevent revascularization, early in the course when the neurons are still salvageable (100). In the same model Sema3A also appears to promote apoptosis in retinal neurons (252), contributing to the retinal progressive neurodegeneration. A Sema3A-neutralizing antibody alleviated both the increased permeability and neurodegeneration while also facilitating normal revascularization of the inner retina (253) suggesting a potential therapeutic course for early staged retinopathy. However, as yet no clinical studies are begun. In the same model Sema3E, also produced by ischemic neurons, appeared to normalize revascularization and suppresses extra retinal vascular outgrowth (254). Sema3E expression is reduced in the vitreous of DM patients with more advanced retinopathy, suggesting suppression (254). Stimulation of Sema3E may offer an additional, adjunctive therapy if such protein mediators are able to sufficiently pass the BRB and BBB.

Toll-like receptors (TLRs) are a class of pattern recognition receptors on macrophages and microglia within neural tissue that normally recognize a diverse set of pathogen-associated molecules not present in the normal host. When activated, TLRs signal downstream pathways that are also known to activate the NF-κB and IL-1β generation paths, which, with the amplification by astrocytes, result in the generation of pro-inflammatory cytokines and chemokines. TLRs have been found that are also capable of responding to endogenously derived molecules, such as components released from necrotic cells or molecules that may be formed as a consequence of other pathologic mechanisms. More specifically TLR-4 polymorphisms have been identified associated with Type 2 DM raising the question as to whether these receptors additionally contribute to the inflammatory processes associated with the resultant neurodegeneration (245, 255). Targeting these molecules, therefore, could reveal a number of molecular and pharmacologic potentials to modulate the resultant inflammation. As an example, minocycline, due to its ability to readily penetrate the CNS and demonstrate anti-inflammatory properties, has appeared well-suited for such retinal and CNS disorders (27, 256). Minocycline blocks microglial activation in response to a variety of inflammatory stimuli by inhibiting TLR-2 and TLR4 signaling and several MAP kinases. In a rat model of diabetic retinopathy, minocycline treatment repressed the release of pro-inflammatory cytokines IL-1β, and TNF-α with a concomitant reduction in caspase-3 mediated apoptosis (257). In patients with diabetic macula edema oral minocycline has been demonstrated to reduce abnormal vascular permeability with a modest improvement in visual acuity and OCT defined tissue edema (258). More recently another antibiotic, azithromycin (also with immunomodulatory capabilities) has been noted to block RGC death in a retinal ischemia/reperfusion mouse model by modifying the inflammatory state (259). Taken together, these findings underscore the importance of further testing to define if such toll receptor blocking agents may be utilized as microglia-directed immunotherapy in human disease.

Advanced glycation end products (AGE’s) are proteins or lipids that are formed through non-enzymatic glycation as a result of being exposed to hyperglycemic conditions and are intensively related to the microvascular and macrovascular complications (22, 140). AGEs bind to and activate the AGE receptor (RAGE) on the surface of microglia, astrocytes, vascular endothelial cells, and neurons, initiating a resultant proinflammatory response. This acts via heterodimerization with TLR-4 to stimulate microglia in the production of pro-IL-1B, pro-IL-18, and NLRP-3 (260). In this context, the measurement of soluble RAGE has been postulated as a valuable biomarker of these molecular events (261). Blocking the interaction with RAGE has been demonstrated in animal models to reduce the production of the resultant proinflammatory mediators (262), suggesting potential for human therapeutic intervention. However, while the circulating peripheral proinflammatory markers have been associated with CNS WMHIs (136, 244) there is considerable variability associated with retinopathy (156), requiring further investigation.

Extensive research has been conducted studying the events resulting in neuron death that occur with an acute stroke and how the pathways may be used as potential therapeutic targets for treating the occurrence of small as well as large vessel chronic ischemia. Activation of the mitochondrial permeability transition pore (MPTP) has been associated with an increased permeability associated with stroke that results in neuronal death. Cyclosporine A, a widely used immunosuppressant, has been recently shown to possess neuroprotective properties through its ability to block the MPTP. This has recently stimulated research on its use to reduce the severity of cell death with stroke or small vessel ischemic conditions (263).

Histone acetylation is increased within the retina when exposed to elevated glucose due to a decreased activity of histone deacetylase (HDAC) that has been demonstrated to result in an increased expression of pro-inflammatory cytokines from Müller cells (264). The favorable effects of both Valproate and Memantine on the BRB and BBB are attributed to the suppression of NF-κB activity by hyperacetylation of HDAC (220). This facilitates its transcription and release from astrocytes. Memantine appears to exert additional anti-inflammatory effects by also regulating the microglial activation (265).

The activation of the microglial PPARγ., as noted above, ameliorates the disruption of the neural injury after an ischemic insult through activation of the NFκB pathway restoring the local levels of claudin-5 (266) that also assist in restoration of the BBB due to an apparent anti-inflammatory action induced of macrophages and microglia (205, 242, 267). Various compounds with different therapeutic mechanisms appear to offer protection against neuroinflammation and progressive neurodegeneration through PPARγ activation (e.g. the anti-dyslipidemic, and the hypoglycemic thiazolidinedione agents), that, certainly require further investigation (220, 268). There also have been demonstrated anti-angiogenic effects within the retina. While trials of Fenofibrate (a PPARα agonist) have not shown beneficial effects upon lipid exudates present within the retina, a significant reduction in progression of diabetic retinopathy has been demonstrated (269, 270). Synthesized compounds also have recently been developed that act as dual PPARa & PPARγ agonists with apparent improved glucose and lipid metabolism. While Muraglitazar was discontinued during phase 3 trials due to adverse vascular events, Tesaglitazar and Aleglitazar appear promising along with new pan-PPAR agonists (271).

The Muller cell release of a multiplicity of inflammatory enzymes in response to the hyperglycemic and ischemic stimuli of the endoplasmic reticulum has been demonstrated due to the activation and release of transcription factor 4 (ATF4). This is blocked in animal studies by the local or systemic administration of the small molecule, 4-phenylbutyrate (PBA) (272). It is notable that PBA has been used many years for the treatment of patients with sickle-cell disease, thalassemia, and cystic fibrosis (272), and, as it is a small molecule that penetrates the blood-retinal and blood-brain barrier, its potential effects upon diabetes neuronal complications appears to warrant future investigation (272).

Targeting activation of microglia (175, 273–275) and mast cells (175, 276, 277) has gained increasing traction as a potential therapeutic avenue for the treatment of neuralvascular degeneration due to inflammation induced small vessel disease. The interleukins are the prototypic inflammatory cytokines, which often operate in concert with other inflammatory enzymes such as TNF-α. Antagonists to Il-1 have been shown to reduce levels of inflammation as well as improve the glycemic control in T2DM patients via increased B-cell recovered insulin secretion (278). Since the retinal and CNS neuroinflammatory lesion progression in the past has been associated with the adipose inflammatory resistance to insulin, perhaps IL-6 or IL-1 receptor antagonists may provide interesting future treatments. Tocilizumab, a humanized monoclonal antibody against the IL-6 receptor, blocking activation, has been shown to improve insulin sensitivity in rheumatoid arthritis (RA) patients (279). With regard to TNF-α antagonists, many phase IV studies have demonstrated that TNF-α blockers (antibodies) consistently decrease levels of the inflammatory cytokine, CRP, and have demonstrated protective effects against cardiovascular events in RA patients (140). TNF-α is certainly recognized as a primary modulator of the pathology encountered in a number of neurologic disorders and has gathered significant attention in the recent past, although the serious side effects of commercial TNF-α inhibitors have resulted in concerns regarding their efficacy, and treatments such as monoclonal antibody solutions most likely will not be considered for administration to diabetics en-mass. However, there is increasing effort to extract the TNF-α inhibitor/blocker effects within a host of natural products, phytochemicals, and nutraceuticals that may prove to represent alternative treatments. Many natural phytochemicals have been demonstrated to inhibit inflammation with concomitant amelioration of oxidative stress and, together with lesser side effects, have been suggested they be offered as alternative treatments for a number of inherited neurodegenerations (e.g. Alzheimer’s and Parkinson’s) (280), and as such, may also warrant consideration for neurodegenerative complications of DM, although the effect upon the microvasculature is unknown.

Growth differentiation factor-15 (GDF-15) belongs to the transforming growth factor-β family and is also known as “macrophage inhibitory cytokine-1”. Its expression is enhanced in response to tissue ischemia (186). Investigations have demonstrated a positive association between GDF-15 serum concentrations and diabetes risk of retinopathy with higher levels associated in proportion with the retinopathy severity grade level (187). However, the mechanisms in the pathogenesis are not clear, although possible explanations have suggested that GDF-15 inhibits the oxidative stress and endothelial dysfunction (188, 281) with involvement in the inflammatory processes as well (187). Further research is anticipated (140).

During the inflammatory process, as discussed above, the endothelial excitation is a primary factor that results, not only in increased vascular permeability and eventually endothelial apoptosis, but also the secretion of intercellular adhesion molecules that adhere to leukocyte integrins, resulting in the complex VLA-4, the primary surface reactant on the leukocyte that adheres the cell to the endothelial derived ICAM-1 and VCAM-1. As discussed above the adhesion of the leukocyte results in microvascular occlusion with breakdown of the leukocyte resulting in endothelial leakage (147). The humanized anti-VLA-4 mAb (natalizumab) has demonstrated clinical activity in a number of autoimmune diseases (282), and its investigation in diabetes, especially for early retinopathy, appears warranted, although the administration of monoclonal antibodies remains a severe barrier for mass treatment. Similarly, the activation of the protein kinase C (PKC) pathway appears integrated in the genesis as well as progression of diabetic retinopathy. The PKC-B inhibitor, Ruboxistourin is the most researched PKC inhibitor in cellular, animal and human studies, demonstrating normalization of retinal blood flow in diabetic patients (283). The PKC DRS studies determined the drug to significantly delay the occurrence of visual impairment but could not reduce the progression of the more severe NPDR stages to proliferative retinopathy or alter the severity of significant macular edema (52, 284). Similar to other singular pathway therapies, further trials are required to examine potential combinations that will more adequately address the complexities of the interactions.

Targeting microRNA dysregulation, which results in a number of the aspects of the diabetic microvascular injury (but in particular endothelial dysfunction) certainly renders this as a target for monitoring and warrants further investigation for possible intervention (183, 285). Human prospective studies are necessary to evaluate the effects within the microvasculature, perhaps initiating with the retina, with evaluation of the microvascular density, flow, leakage with then the evaluation of the resultant neural cellular structural abnormalities and function pursued as they become more easily defined. With regard to the regulation of this signaling in the brain, insulin and insulin-like growth factor-1 (IGF-1), activating PI3K-Akt signaling that protects the BBB properties of the endothelial cells through the phosphorylation of GSK-3β, has been demonstrated to increase the expression of multiple microRNAs that influence the function and integrity of the endothelial barrier (286). Upregulation of serum IGF-1 concentration has been associated with acceleration of DR (287); however, the Randomized Clinical Trials (RCTS) resulted in conflicting results of the inhibition by Octreotide on progression of DR, although this was predominantly oriented toward the more advanced retinopathy forms (288). Further investigation certainly appears warranted for intervention during the earlier DR stages, when the processes are reversible as discussed above.

Because of the multiplicity of pathways involved in the development and progression of the neurovascular injury, as discussed above, Chalke and Kale have suggested that a combination approach appears worthy to consider and have identified a number of potential medications (289). Citicoline, an intermediate in the synthesis of phosphatidylcholine, has been used to treat CNS degenerations such as Alzheimer’s and Parkinson’s with demonstrated promotion of repair and growth of cell membranes with neurotransmitter improvements. The neuroprotective action appears via interference with the deposition of B-amyloid resulting in improved functional outcomes (290). In addition, it has been evaluated in animal studies as a neuroprotective agent in the treatment of glaucoma (291, 292) wherein it appears to reduce neuronal degeneration by reducing caspace 3 production, and to reduce reactive gliosis by decreasing GFAP levels with some anti-inflammatory effects through the prevention of the upregulation of NF-kB. Chalke and Kale, in their review (289) suggest combining Citicoline with Resveratrol demonstrating the ability as discussed above to suppress Caspase 3 and caspase 8, with neuroinflammation inhibition of NADPH oxidase and the attenuation of NF-κB-induced expression of iNOS, COX-2, and sPLA2 (152, 229, 293) as well as restoring of mitochondrial SIRT-1. Combination drug studies have demonstrated stroke penumbra reduction in animals (289), but clinical studies have yet to begin.

Chalke and Kale have also suggested combining Duloxetine with N-Acetyl Cysteine. Duloxetine is a serotonin and norepinephrine inhibitor that demonstrates anti-oxidative and anti-inflammatory properties, as well as the ability to modulate the expression of angiogenic and neurotrophic factors that assist in attenuating the DR neuronal and vascular effects in animal studies (294). N-acetyl cysteine, as discussed above, is an antioxidant, scavenging reactive oxidative species with restoration of superoxide dismutase and inhibition of NF-kB, reducing inflammation and pericyte loss. Therefore, the combination, through adjunctive pathways may be beneficial to early DR in preventing progression, but as discussed, further clinical studies are necessary.

Chalke and Kane also considered the combination of CD5-2 with an angiopoietin-2 inhibitor. CD5-2 is an oligonucleotide that increases vascular endothelial cadherin, restoring pericyte integrity of the endothelial cell barrier, as well as attenuating VEGF signaling in the resultant inflammation. The angiopoietin-2 inhibitor interacts with the TIE-2 tyrosine kinase receptors improving endothelial health and adhesion. The combination is suggested as a potential improved alternative treatment for the progressive breakdown of the BRB, wherein Anti-VEGF agents combined with steroids, although reducing edema, have demonstrated little functional testing improvement (289).