The biology of leptin, as outlined above, was elucidated by studying leptin deficiency and examining the effects of leptin repletion on low-leptin states in animal models and humans. These leptin-deficient states include not only genetic leptin deficiency (very rare in humans) and caloric restriction (including anorexia), but also other conditions associated with diminished fat stores (including lipodystrophy syndromes and untreated insulin deficiency) (2
). Administration of exogenous leptin mitigates the increased appetite, decreased energy expenditure and neuroendocrine dysfunction associated with each of these states of leptin insufficiency. Leptin also attenuates the hyperglycemia caused by uncontrolled insulin-deficient diabetes and lipodystrophy syndromes (12
The ability of exogenous leptin to reduce food intake/body weight (and to modulate metabolism and endocrine function) in the face of replete adipose stores, especially in obese individuals, has proven to be more limited, however (41
). Indeed, commensurate with their large adipose mass, obese individuals exhibit elevated circulating leptin concentrations relative to lean subjects (6
), and these elevated leptin concentrations (obviously) fail to return body adiposity to within the normal range. In rare instances, genetic mutations may abrogate LEPR function (here, circulating leptin is increased to concentrations well above those observed in “normal” obese humans of similar fat mass) (43
). These instances are similar to classical hormone resistance/insensitivity syndromes (growth hormone insensitivity, Type A insulin resistance, etc.), in which genetic alterations of the hormone receptor prevent hormone action (44
In the vast majority of human obesity however, Lepr
is unperturbed, and obesity and any diminished responsiveness to leptin must result from other mechanisms. The poor efficacy of endogenous leptin to promote leanness in obese subjects has given rise to the notion of functional “leptin resistance,” which is based loosely upon and often compared to the concept of insulin resistance in type 2 diabetes - where diminished cellular and metabolic insulin responsiveness coexist with the hypersecretion of insulin (46
). Implicit in the concept of leptin resistance is the idea that processes that promote and/or result from obesity impair leptin action, thereby facilitating the occurrence of obesity and attenuating the potential efficacy of therapy with exogenous leptin. Understanding mechanisms that may underlie “leptin resistance” is thus crucial both for determining the causes of obesity and identifying potential mechanisms that can be targeted for therapy.
While the goal of this NIH conference was to generate a universal, quantifiable definition of leptin resistance, the common application of “leptin resistance” to label both the presence of hyperleptinemia in obesity and the failure of exogenous leptin administration to provide therapeutic benefit suggests the difficulty of identifying such a universal definition. Regarding the use of “leptin resistance” to label hyperleptinemia in obesity: Due to their increased adipose tissue mass, virtually all obese humans and animal models (except some animal strains and extraordinarily rare patients with genetic leptin deficiency) display elevated circulating leptin concentrations relative to lean controls (1
). Additionally, processes other than those attendant to leptin action and/or LEPR-B signaling per se
likely contribute to the determination of adiposity and etiology of obesity in each person- including the function and/or anatomic variation of numerous neural circuits: While some of these circuits may be directly modulated and/or developmentally programmed by leptin, many are not (47
) (). Furthermore, the finding that even very obese individuals exhibit changes in hunger and energy expenditure upon moderate weight loss, and that these changes are blunted by the administration of leptin, suggests that the elevated levels of leptin in obese individuals may be functionally relevant (54
). Thus, defining leptin resistance as hyperleptinemia in obesity merely serves to note the elevated concentration of a single hormone whose production is expected to be increased in this condition, alterations in the action of which may or may not be primary to the genesis of obesity, and where levels of the hormone retain some physiologic relevance. Thus, we do not favor referring to the coexistence of hyperleptinemia in obesity as “leptin resistance.”
When focusing on the potential therapeutic utility of leptin, the limited ability of exogenous leptin to promote desired outcomes represents the central issue, and it is this aspect of the leptin resistance concept that we will focus on in this review. Even restricting use of “leptin resistance” to states poorly responsive to exogenous leptin, the term covers a multitude of situations, however. It refers, for example, to assays of leptin action as diverse as the induction of STAT3 phosphorylation in the brain (in animal models) and other tissues, to the attenuation of feeding and the restraint of body weight and adiposity (in either animals or humans). These may be assayed following acute or long-term treatment with leptin doses ranging from the approximately physiologic to the heroically pharmacologic. Practically speaking, therefore, “leptin resistance” is a term so broadly-applied and context-dependent that there can be no universal, quantifiable, clinically useful definition of “leptin resistance.”
Studies in animal models have suggested a number of basic mechanisms that may underlie attenuated responsiveness to leptin. Many of these represent processes engaged during or promoted by overnutrition and obesity, including changes in circulating leptin-binding proteins, reduced transport of leptin across the blood-brain barrier and/or the provocation of processes that diminish cellular LEPR-B signaling (inflammation, ER stress, feedback inhibition, etc.) (25
). Alterations in the development of leptin-regulated neurons and other components of the circuitry that controls leptin action could also blunt leptin action throughout life (50
). While differing in their specifics, these potential explanations for the decreased efficacy of leptin all postulate that nutritional alterations or obesity (or increased ambient leptin concentrations themselves (60
)) impair leptin action. Indeed, interference with many of the cellular mechanisms that attenuate LEPR-B signaling improves leptin action in cells and animal models, revealing that these mechanisms decrease leptin action in vivo
, as well as suggesting the potential utility of these processes as points of therapeutic intervention (26
). Thus, while it is not possible to precisely define leptin resistance in a universal, precise, and quantifiable manner, it is clearly useful to identify and understand mechanisms that may attenuate leptin action in vivo
The “normal” response to exogenous leptin, against which leptin resistance is often defined (especially in rodents), most often results from the attainment of circulating leptin concentrations thousands of times higher than physiological (63
), and the anorectic response to these doses is modest and subject to relatively rapid tachyphylaxis. During evolution, undernutrition presumably represented a greater threat to survival than did overnutrition, with the result that the defense against starvation (low leptin) produces stronger responses than does the defense against nutritional surfeit (high leptin). A similar line of reasoning suggests that the efficacy of leptin may be near maximal at the concentrations found in most obese people at baseline, and that the addition of exogenous leptin may, therefore, raise circulating leptin concentrations without substantially increasing leptin action. While a related hypothesis suggests that obesity was also selected against by factors including the likelihood of predation (64
), the ability of palatable calorically-dense diets to promote obesity in most animals and the failure of elevated leptin levels to reduce body weight in obese animals suggests that leptin may not play a major role at this end of the spectrum. The possibility that leptin action may be near its physiological maximum in states of obesity and the notion that obesity attenuates leptin action are not mutually exclusive- both mechanisms may contribute.
In human subjects, it is generally not possible to examine the molecular mechanisms associated with or underlying “leptin resistance” experimentally; rather, the state must be operationally defined as decreased responsiveness to exogenous leptin by certain criteria. As noted, however, there is no standard for the definition of human “leptin resistance.” Is leptin resistance defined by the response to a high or low dose of leptin, given once for a short time or chronically over weeks to months, in patients at baseline or following some amount of weight reduction? What is the attenuated response by which we define leptin resistance: decreased food intake, weight loss, alterations in blood glucose or lipids, hepatic triglycerides, immune function, etc.?
Since the efficacy of leptin for the control of many metabolic parameters (e.g., weight, glucose, lipids, hepatic steatosis, etc.) is likely to be of interest, leptin responsiveness must be defined in terms of the response of specific parameters, individually, to leptin treatment.
Leptin sensitivity, not defining “leptin resistance,” is what really matters
It is more practical in humans to define sensitivity rather than resistance to leptin's actions. While clinical “leptin resistance” may be captured in a general way as poor responsiveness to exogenous leptin, for the reasons outlined above, it is not possible to define clinical leptin resistance in a precise manner that can be assessed with a single, universal assay. A pragmatic approach to leptin resistance and therapeutic leptin action thus focuses not on defining clinical leptin resistance in a universal manner, but rather on assessing leptin sensitivity: Which individuals are likely to respond to leptin and/or can be sensitized to exogenous leptin? While this may seem, on the surface, to be a semantic argument, it acknowledges that there can be no universal definition of leptin resistance and that defining who we can effectively treat and the parameters of effective treatment represent the crucial issues.
How to assess leptin sensitivity effectively in the clinical setting is not currently clear. The available information suggests that, in general, leptin sensitivity is greatest in those with low adiposity and low endogenous circulating leptin in the non-weight reduced state (65
). Body adiposity and leptin concentration are unlikely to represent the only predictors for leptin sensitivity, however, as each individual is likely to exhibit an idiosyncratic response to overnutrition or obesity (i.e., in terms of genetic differences or ER stress or other responses that might limit leptin action). Indeed, the sexual dimorphism in circulating leptin concentrations suggest underlying differences in leptin production and/or action by sex (6
). Similarly, circulating free leptin concentrations are determined not only by leptin production, but by its clearance and by levels of circulating soluble LEPR (59
). Furthermore, adiponectin and other circulating factors may predict aspects of the therapeutic efficacy of leptin (66
). Thus, it will be crucial to assess and report measures of leptin responsiveness in large samples of individuals in the context of a variety of parameters that may affect leptin sensitivity (e.g., age, sex, BMI, adiposity, fat distribution, circulating levels of leptin and other factors, etc.). Also, leptin is manufactured in multiple forms, each with different pharmacodynamics parameters and potentially distinct determinants of efficacy.
In addition to the modulation of body weight and adiposity, the examination of other potentially useful outcomes of leptin therapy (e.g., glucose homeostasis, lipid metabolism, hepatic lipid content, etc.) will be important. Furthermore, as the relationship between potential measures of acute leptin action (e.g., 24-hour food intake, energy expenditure, leptin-stimulated STAT3 phosphorylation in peripheral blood monocytes, or brain imaging) and the potential therapeutic chronic effects of leptin are unclear, it would be useful to examine both acute and long-term responsiveness of individuals to determine the potential predictive value of acute studies.
Can Leptin Sensitivity be Increased?
Some investigators have suggested that pharmacologic interruption of the cellular mechanisms apparently attenuating LEPR-B signaling could increase leptin sensitivity (55
). Indeed, early attempts have met with preliminary success in animal models. Furthermore, the finding that some compounds (e.g., the amylin derivative, pramlintide) augment leptin action in some individuals suggests that it may be possible to increase the sensitivity of some individuals to therapeutic leptin administration (67
). Whether such a result is a consequence of the initial weight loss induced by a non-leptin drug, by direct impact of such a drug on leptin receptor signaling, or mediated by other phenomena remains unclear. Certainly however, combinatorial approaches appear to hold some promise for clinical leptin therapy. As above for responsiveness to leptin alone, it will be important to determine traits and/or acute assays and endpoints that predict the responsiveness of patients to many types of potential leptin-sensitizing therapies.