University of California, San Francisco, Metabolism Section, Department of Veterans Affairs Medical Center, San Francisco, California 94121
Address all correspondence and requests for reprints to: Carl Grunfeld, M.D., Ph.D., Metabolism Section (111F), Department of Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, California 94121. E-mail: . grunfld{at}itsa.ucsf.edu
In the last few years, we have developed an increasing understanding of how immunosuppression occurs during malnutrition. The focus has shifted from nutrients to hormones, in particular leptin. Curiously, our understanding of malnutrition was fed by research in obesity.
Leptin deficiency is well known to cause obesity through increased eating and decreased energy expenditure in ob/ob mice (1). However, ob/ob mice were also found to have defects in their immune system. They have atrophy of the thymus, spleen, and lymph nodes. The atrophy is predominantly due to decreased lymphocytes, especially T-lymphocytes (2, 3, 4, 5). There is an increase in circulating monocytes, but their phagocytosis is impaired. Cells from ob/ob mice show a defective mixed lymphocyte reaction, with poor generation of interferon-, a proinflammatory cytokine, and increased IL-4, an anti-inflammatory cytokine (6). As a consequence of the T-cell deficit, ob/ob mice are more susceptible to lipopolysaccharide (LPS)-induced death (7) and less susceptible to T-cell mediated toxicity in vivo (5). Most important, with leptin treatment these defects are rapidly restored. Leptin treatment in vitro of cells from the ob/ob mice reverses the defects in the mixed lymphocyte reaction, leading to vigorous secretion of interferon-
and blunting of IL-4 secretion (6). Leptin treatment decreases susceptibility of ob/ob mice to LPS-induced death (7) and allows T-cell induction of cytokines in vivo (5). Although these experimental models relate to toxicities, they reflect an abnormal immune response that means an inability to clear infection.
The link between the leptin deficiency and starvation came with the realization that the hormonal profile of ob/ob mice resembled that of starvation (8) and that starvation is also a state of leptin deficiency. It had been known that ob/ob mice had decreased T3, decreased gonadal hormones, and increased cortisol. Leptin treatment rapidly reverses these changes in ob/ob mice (9, 10). Of course, decreased T3, decreased gonadal hormones, and increased cortisol are characteristic of the euhormone-sick syndromes that occur in illness and starvation. It is of note that leptin levels fall rapidly with starvation, even before adipose tissue depletion (11). In a key study (12) that has been confirmed (13), leptin administration was also shown to restore the hormonal changes of starvation toward normal despite the loss of body weight, lower glucose levels, and increased ketones that accompany starvation. Thus, although lack of nutrients was the key initiator, the hormonal trigger was the decrease in leptin and leptin treatment that could override the lack of nutrients. These data changed the view of leptin from a satiety factor or obesity hormone to a hormone whose deficiency triggers the necessary adaptation to fasting and whose presence signals repletion and assures that the adaptation program is silenced.
Given the parallels between the immunological changes in the ob/ob mice and in starvation, it was logical to determine whether leptin played a role in the immunosuppression of starvation, because starved animals show decreases in lymphocyte number and spleen and thymus weight similar to that found in ob/ob mice (4, 6, 13, 14). In experiments that paralleled those of the endocrine system, it was shown that administration of leptin reversed the decreases in lymphocyte number and spleen and thymus weight (4, 6, 13). Starved mice cannot develop delayed hypersensitivity, and leptin restores that response (6). Starvation induces an unbalanced cytokine response to LPS with increased susceptibility to LPS-induced death, and leptin administration restores that response toward normal (13). As with the euhormone-sick hormonal responses, immunological restoration occurs despite the loss of body weight, lower glucose levels, and increased ketones that accompany starvation (13).
These studies in mice raise interesting questions about what happens in humans. There are limited published data in humans with the ob/ob or db/db genotype. In one family with a leptin mutation, 7 of 11 children with the phenotypic syndrome died of infections in childhood (15). Other studies of the immune system in humans with genetic leptin deficiency are underway and should be followed closely.
Similar to what was found with cells from ob/ob mice, leptin treatment of human lymphocytes during a mixed lymphocyte reaction in vitro enhances interferon- production and blunts IL-4 production (6).
Thus, it is logical to ask whether leptin may play a role in the immunosuppression of malnutrition in humans. In this issue of JCEM, data are presented that raise this possibility. Palacio et al. (16) studied children with protein-calorie malnutrition, who had slightly decreased fat and significantly lower leptin levels. On admission, there was no correlation between body fat and leptin, reflecting the acute suppressive effects of malnutrition. With feeding, leptin levels rapidly returned to normal, with restoration of the correlation to body fat before restoration of normal fat content. Likewise with feeding, the ability of the infants polymorphonuclear white cells to make interferon- and TNF-
increased, whereas production of IL-4 decreased. These changes are similar, but not identical with what is found with leptin treatment of ob/ob and starved mice.
The data of Palacio et al. (16) raise the question as to whether leptin may help in restoring the human immune response in leptin deficiency. Although malnutrition is best treated with nourishment, there are other leptin-deficient states in humans who are well nourished. In addition to humans with genetic leptin deficiency, leptin is decreased in humans with severe lipodystrophy, be it congenital, acquired, or HIV-associated (17, 18, 19). The data of Palacio et al. (16) suggest that the immune system should be examined in those diseases, and if deficient, the effect of leptin on the immune system should be studied. Trials of leptin for the metabolic abnormalities of lipodystrophy have begun (20). Looking at the immune response in these trials would add another important aspect of leptin biology.
Finally, it should be recognized that in most of these studies leptin did not fully restore the changes in the endocrine and immune systems. There are likely other hormone signals yet to be discovered that may play important roles in these changes.
Footnotes
The work from the laboratory cited herein was supported in part by the Research Service of the Department of Veterans Affairs and the NIH (DK-49448).
Abbreviations: LPS, Lipopolysaccharide.
Received May 13, 2002.
Accepted May 16, 2002.
References