One Step from Prediabetes to Diabetes: Hypothalamic Inflammation?

Type 2 diabetes (T2D) is a devastating disease that arises from prediabetic conditions including glucose intolerance, insulin resistance, hyperlipidemia, and obesity, which, together with a few others, collectively is termed metabolic syndrome. Due to the absence of severe hyperglycemia and its chronic complications, prediabetes offers an early therapeutic window to reverse glucose and insulin abnormalities to prevent the development of T2D. To this end, understanding the pathogenic mechanisms underlying the progression from prediabetes to diabetes represents an area of keen research. In this issue, Burgos-Ramos et al. (1) provide evidence demonstrating that the development of diabetes from prediabetes in a mouse model may involve a central nervous system mechanism in relation with hypothalamic inflammation (Fig. 1).

Fig. 1.

Hypothalamic inflammation links central insulin resistance to diabetes. Hypothalamic regulation of feeding, body weight, and glucose homeostasis is mediated by multiple signaling pathways including insulin singling. The hypothalamic insulin signaling cascade is directed by activation of insulin receptor, IRS proteins, and downstream phosphatidylinositol-3-OH kinase (PI3K), which control AKT and FOXO to regulate gene expression of neuropeptides such as proopiomelanocortin (POMC) and neuropeptide Y (NPY). In IRS2-deficient mice, some of them can develop diabetes, but others present only prediabetic syndrome, likely due to the differential compensatory effects by IRS1. The compromised compensation of hypothalamic IRS1 signaling in diabetic IRS2-deficient mice is related to the certain pattern and perhaps increased magnitude of hypothalamic inflammation. Such hypothalamic changes in IRS2-deficient mice align with recent literature showing the central mechanisms of obesity and T2D involve IκB kinase β (IKKβ)/NF-κB- and JNK-mediated hypothalamic inflammation, a central pathogenic event caused by overnutrition-induced endoplasmic reticulum (ER) stress, oxidative stress, autophagy defects, and cytokine activation. AA, Amino acids; AP1, activator protein-1; FFA, free fatty acid; FOXO, forkhead box O; PIP2, phosphatidylinositol 4,5-bisphosphate; PTEN, phosphate and tensin homolog.

This study from Burgos-Ramos et al. (1) was based on mice that are genetically deficient of insulin receptor substrate 2 (IRS2), an animal model that resembles T2D (2, 3). IRS proteins are intracellular docking molecules that bind to activated insulin receptors in response to insulin and undergo rapid activation through tyrosine phosphorylation by insulin receptors. Activated IRS proteins bind SH2 domain-containing proteins such as phosphatidylinositol-3-OH kinase to relay intracellular signaling of insulin stimulation. IRS2 protein was originally cloned based on sequence resemblance to IRS1 (4). IRS2 is widely expressed throughout the body including the arcuate nucleus and paraventricular nucleus of the hypothalamus (5), two critical hypothalamic regions that regulate whole-body energy and glucose homeostasis (6–12). IRS2 was found to play nonredundant roles with IRS1 in regulating peripheral glucose metabolism, β-cell function, and energy homeostasis (2, 3, 5, 13). IRS2-deficient mice often develop diabetic symptoms including hyperglycemia and β-cell dysfunction (2, 3, 5, 13). Interestingly, magnitude of glucose disorders in IRS2-deficient mice can be variable, with certain congenic strains only presenting prediabetic changes (14). This phenotypic divergence can offer an experimental tool to study pathogenic factors that may differentially modulate central insulin signaling pathway leading to two different outcomes: prediabetes or overt diabetes.

T2D and related metabolic syndrome belong to systemic endocrine problems, and recent research has interestingly pointed to a pathogenic root in brain immune dysregulation (15–28). Systemic glucose homeostasis, which is a physiological state of balanced glucose disposal and production, is achieved through the coordinated actions of multiple organs including the liver, pancreas, adipose tissue, skeletal muscle, and the brain as well. The hypothalamus in the brain exerts essential roles in this process by sensing circulating metabolic signals (e.g. insulin and leptin) and instructing various neuroendocrine and neural pathways to control peripheral glucose metabolism (9–11). However, under pathological conditions such as overnutrition and obesity, hypothalamic neurons are interrupted due to the induction of metabolic inflammation, mediated by proinflammatory pathways such as IκB kinase β (IKKβ) and downstream nuclear factor-κB (NF-κB) (15–21) or c-Jun N-terminal kinase (JNK) (22–24) (Fig. 1). Although many probably still remain unidentified, hypothalamic regulatory changes caused by inflammation/stress essentially include the reduction of insulin and leptin signaling in the hypothalamus (15–27, 29). Among these findings, hypothalamic inflammation was shown to not only cause feeding and body weight changes but also employ body weight-independent manners to cause systemic glucose intolerance. The molecular mechanisms can involve changes in hypothalamic insulin signaling, because hypothalamic action of insulin is indeed crucial for control of body weight and also body weight-independent peripheral glucose metabolism. In the study by Burgos-Ramos et al. (1), IRS2-deficient mice were divided into two phenotypic groups, one having diabetes and the other only prediabetes. By profiling the compensatory reaction of IRS1 signaling and the relationship with hypothalamic inflammation, the study suggested that differential activation of hypothalamic inflammatory pathways seems to be a pathogenic factor for variable sensitivities of hypothalamic insulin signaling and correlated onset of diseases (Fig. 1).

In the first set of studies, the authors comparatively profiled feeding, body weight, and glucose homeostasis between diabetic and prediabetic IRS2-deficient mice. Both groups have elevated blood glucose and insulin levels, indicative of a common systemic insulin resistance. In prediabetic IRS2-deficient mice, their body weight significantly increased and hyperleptinemia developed even when maintained on a normal chow, a finding consistent with the obesogenic action of hypothalamic insulin resistance. However, these mice did not develop diabetes, which might be related to compensatory action of IRS1 in the form of increased IRS1 protein level and enhanced IRS1-induced Janus kinase 2 and Akt activation. Compared with prediabetic IRS2-deficient mice, diabetic IRS2-deficient mice displayed overeating and impaired energy expenditure, but body weight decreased due to chronic malnutrition, all of which belong to diabetic symptoms. The development of diabetes in these mice correlated with the observation that IRS1 did not respond to compensate for the genetic deficiency of IRS2. Altogether, when IRS1 could provide certain compensation, IRS2-deficient mice were only prediabetic, characterized by overweight/obesity and systemic insulin resistance. In contrast, when IRS2 was deficient but without compensatory effects of IRS1, mice developed diabetes. Overall, levels of compensation from hypothalamic IRS1 signaling may underlie the differential diabetic phenotypes among IRS2-deficient mice.

Subsequently, the authors found that the development of diabetes in IRS2-deficient mice was related to a more complete induction of hypothalamic inflammation, and on the other hand, only a partial induction of hypothalamic inflammation was detected in prediabetic IRS2-deficient mice. Regarding the potential underlying pathways, both NF-κB and JNK were significantly activated in diabetic IRS2-deficient mice, agreeing with recent findings showing that NF-κB- and JNK-mediated hypothalamic inflammation cause peripheral insulin resistance and systemic glucose intolerance (20, 22–24). Furthermore, several classical NF-κB products were examined, such as TNF-α, suppressor of cytokine signaling 3 (SOCS3), and IL-6. Results showed that TNF-α mRNA increased in the hypothalamus of the diabetic group, SOCS3 mRNA increased in the hypothalamus of the prediabetic group, and IL-6 mRNA increased in both groups. Thus, a certain format of hypothalamic inflammation, presumably due to activation of both NF-κB and JNK, was associated with the absent compensatory action of hypothalamic IRS1 signaling in the diabetic IRS2-deficient mice. This finding aligns with the established role of these inflammatory pathways in inhibiting insulin signaling (15, 30, 31), and the effects of inhibiting the hypothalamic NF-κB or JNK pathway in counteracting glucose metabolic disorders in peripheral tissues (17–23, 25). On the other hand, despite these understandings, it remains to be experimentally proved whether the pattern of hypothalamic inflammation revealed in this study could indeed mediate the transition from prediabetes to diabetes in the group of IRS2-deficient mice. Further research is needed to examine the predicted causal role of hypothalamic inflammation in the development of diabetes in this genetic model and perhaps other models as well.

In summary, taking advantage of the divergent metabolic phenotypes of IRS2-deficient mice, this study suggests that central insulin resistance can cause both prediabetic syndrome and diabetes, and different formats of hypothalamic inflammation could represent a pathogenic factor that determines prediabetic vs. diabetic outcomes. It can also suggest that proinflammatory NF-κB and JNK pathways are significantly involved in this potential hypothalamic mechanism, an interesting concept that will anticipate experimental assessments in near-future research.