Weight management in individuals with obesity is a first-line lifestyle intervention to prevent worsening of insulin resistance, progression to type 2 diabetes (T2D), and adverse cardiovascular disease outcome. However, persistent weight reduction is often hard to maintain, with <6% of subjects able to achieve sustained weight loss during a 15-year follow-up (1). Current evidence suggests that repeated weight loss and subsequent gain, a phenomenon known as weight cycling or “yo-yo dieting,” exacerbates the risk for T2D and cardiovascular disease (2). The full spectrum of its adverse effects and the underlying mechanisms, however, remain poorly understood.
In this issue of Diabetes, Winn et al. (3) reveal a surprising consequence of yo-yo dieting on inducing β-cell maladaptation, leading to impaired insulin secretion mismatched for the insulin-resistant peripheral tissue. This work builds upon the initial discovery of weight cycling–induced worsening of peripheral insulin resistance compared with that of similarly obese mice (4). In an insulin-resistant state, β-cells increase insulin secretion to maintain normoglycemia and compensate for the diminished peripheral insulin action. Persistence of insulin resistance leads to eventual decompensation due to β-cell dysfunction, dedifferentiation, and death (5,6). The most intriguing finding of the current study lies in the islet insulin secretion defects upon weight cycling despite impaired peripheral insulin sensitivity comparable to the persistently obese state. This metabolic conundrum suggests that β-cell–intrinsic defects are uniquely induced by weight cycling and exacerbate decompensation (Fig. 1). These defects do not involve diminished glucose responsiveness but instead entail impaired cAMP-dependent secretion and reduced insulin content. Transcriptomic analysis of islets from obese and weight-cycled mice revealed an upregulated replication-related program and downregulation of genes involved in protein localization, β-cell identity, and functional maturity, indicative of a functionally immature β-cell state primed to replicate. However, no corresponding differences in replication were observed. As attempted replication under conditions of stress can trigger cell death (7,8), β-cells subjected to weight cycling may not be survival competent. This notion is supported by a decrease in islet size and area along with reduced β-cell number. Intriguingly, the authors noted induction of Ins1 and Ins2, a potential compensatory response to attenuated β-cell functional mass. As cell cycle genes are involved in DNA repair, differential gene expression in this pathway may reflect an ongoing DNA damage response. Direct assessment of β-cell DNA damage and cell death preceding and during the weight cycling regimen will clarify the mechanisms that underly β-cell decompensation triggered by nutrient fluctuations.
What, then, are the signals that drive β-cell maladaptation during weight cycling? Body weight fluctuations evoke nutrient fluxes in metabolic tissues along with altered appetite control in the central nervous system (9), which raises the possibility of interorgan cross talk subserving β-cell defects. Adipose tissue bears the brunt of weight cycling as a major site of lipid storage and mobilization, and its heightened proinflammatory T-cell response contributes to impaired glucose tolerance (4). While the authors found that the cause of weight cycling–induced worsening of glucose tolerance largely lies in β-cell insulin secretion defects, they noted reduced glucose uptake in adipose tissue and muscle. This constellation of metabolic abnormalities highlights the complexity of tissue responses to repeated insults of nutritional overload and deprivation. Weight loss and regain leads to lipolysis and fatty acid re-esterification in adipose tissue (10). Nutritional stress due to lipid overload and hyperglycemia induces β-cell adaptation during weight gain, while subsequent weight loss and the accompanying onslaught of lipolytic mediators could trigger a maladaptive response (11). Metabolome and secretome profiling across multiple tissues at distinct stages of weight cycling will identify the signaling molecules contributing to β-cell maladaptation, either directly or through tissue cross talk. Given the regulatory importance of diurnal nutrient cycles and cell-intrinsic metabolic oscillations in β-cell function (12,13), it is important to define the mechanisms by which β-cells adapt to repeated nutrient variations. Furthermore, in light of the balance between mTORC1 and AMPK activity in β-cell metabolic adaptation (14,15), these nutrient-sensing mechanisms may modulate the adaptive and maladaptive responses to drastic nutrient fluctuations induced by weight cycling.
The adaptation of cellular response to changing nutrient cues at the transcriptional level involves the epigenome. Epigenetic dysregulation has been identified as a major player underlying β-cell defects as well as the dysfunction of peripheral insulin-sensitive tissues in diabetes (16–18). Chronic exposure to a particular metabolic state (such as hyperglycemia) can induce irreversible epigenetic and functional changes that persist even after cessation of the exposure (19). This phenomenon of “metabolic memory” could be pertinent to adverse outcomes associated with weight cycling. Notably, a recent study suggested that the timing of dietary intervention is critical in shaping the β-cell response. A dietary switch to induce weight loss prevented obesity-related β-cell epigenetic programming and functional defects if incorporated prior to the onset of β-cell decompensation (20). Mapping the epigenomic changes in β-cells and peripheral tissues may shed light on the epigenetic program and the relevant temporal window that fosters metabolic memory in response to weight cycling.
Accumulating evidence suggests the existence of an autoregulatory adaptive mechanism or “famine reaction” that predisposes to obesity following a period of starvation (21). The current study highlights the underappreciated adverse impact of yo-yo dieting on islet physiology that exacerbates diabetes risk. To leverage these findings for safe and effective weight management, future studies are needed to dissect the underlying mechanisms and define the impact of the extent, duration, and frequency of weight cycling on cardiometabolic outcomes. These new findings also caution against extreme interventions in patients with obesity with inevitable weight cycling that may hasten rather than prevent diabetes. With the future studies it will spur, this investigation provides a basis for close monitoring of islet function in subjects undergoing weight interventions to minimize the metabolic risk.
See accompanying article, p. 2313.
Funding. K.M. is supported by a grant from the National Institutes of Health (R01DK112794) and an Arthur Riggs Diabetes and Metabolism Institute Pilot Project. S.D. is supported by a grant from the National Institutes of Health (R01DK120523), a Human Islet Research Network New Investigator Award (subaward to S.D. via UC4DK104162), an Arthur Riggs Diabetes and Metabolism Institute Pilot Project, and start-up support from City of Hope.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.