Cutting Edge: Fasting-Induced Hypoleptinemia Expands Functional Regulatory T Cells in Systemic Lupus Erythematosus

Fasting and/or caloric restriction are known to alleviate the manifestations of autoimmune diseases through mechanisms that remain mostly elusive (1). For example, caloric restriction prolongs life in (NZB × NZF)F₁ (NZB/W) mice (2) that spontaneously develop systemic lupus erythematosus (SLE), an autoimmune disease characterized by the breakdown of tolerance to multiple nuclear Ags and an uncontrolled activation of self-reactive lymphocytes that cause inflammation and tissue damage. Although it has been shown that in NZB/W mice the reduction of caloric intake associates with a decrease in lymphoproliferation, Ab production, and secretion of proinflammatory mediators (3), it remains unclear what drives the observed beneficial clinical, immunological, and biochemical effects (4).

During fasting, the levels of the circulating adipokine leptin are dramatically reduced (5). Leptin has all of the characteristics of a proinflammatory cytokine (6) and links nutritional status with neuroendocrine functions and immune responses. Leptin promotes Th1 cell differentiation and the development of autoimmune responses in several animal disease models (7), and it can act as a negative factor for the expansion of CD4⁺ regulatory T cells (T_Reg), a subset of T cells with an important role in the maintenance of peripheral immune tolerance to self-Ags (8).

In an attempt to connect these observations, we investigated the effects of fasting in NZB/W mice and found that the reduction in circulating levels of leptin caused by starvation directly promoted the expansion of functional T_Reg. These findings can help to explain some of the beneficial effects of fasting in SLE.

C57BL6/J (B6) wild-type (WT) mice, leptin-deficient B6^ob/ob (ob/ob), leptin receptor-deficient B6^db/db (db/db), and NZB/W mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and maintained at the University of California Los Angeles. Mice were treated in accordance with institutional guidelines under approved protocols. All experiments were conducted in age-matched female mice that were divided into three groups. One control group had ad libitum access to food and received i.p. injections of 0.2 ml PBS at 9 am and 6 pm daily for 2 d. The other two groups of mice were deprived of food for 48 h and received i.p. injections of either 0.2 ml PBS or recombinant leptin (R&D Systems, Minneapolis, MN) dissolved in PBS at a dose of 1 μg/g body weight twice daily (also at 9 am and 6 pm). All mice had continuous access to water.

After blood drawing, erythrocytes were removed using red cell lysing buffer (Sigma-Aldrich, St. Louis, MO), and PBMC were used for flow cytometry. For splenocytes, single-cell suspensions were prepared by passing cells through a cell strainer before red cell lysis and resuspension in HL-1 medium (BioWhittaker, Walkersville, MD). CD4⁺CD25⁺ and CD4⁺CD25⁻ T cells were isolated via magnetic bead separation with Miltenyi Biotec kits using an AutoMACS System (Miltenyi Biotec, Auburn, CA) and found >95% pure by flow cytometry analysis.

Phenotypic analyses were performed with combinations of fluorochrome-conjugated Ab using standard techniques. After Fc blocking, fluorochrome-conjugated anti-mouse Ab to CD4 and CD25 (eBioscience, San Diego, CA) or isotype control Ab were used for staining prior to acquisition on an FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and subsequent analysis using FlowJo software (Tree Star, Ashland, OR). For intracellular staining, cells were first stained for the expression of cell-surface markers and then fixed, permeabilized, and stained using the Foxp3 staining kit (eBioscience) according to the manufacturer’s instructions.

After cell isolation, CD4⁺CD25⁻ effector T cells (T_Eff) were labeled with CFSE by incubation in 0.5 μM CFSE at 37°C for 10 min. Samples were then placed in HL-1/5% FCS and incubated in ice for 5 min, then washed two times. T_Reg/T_Eff (ratio 1:1 or 1:4) mixtures or control samples consisting of T_Eff alone were plated in 96-well round-bottom plates (Corning Life Sciences, Lowell, MA). Dynabeads mouse anti-CD3/CD28 beads were added to a final ratio of 0.5 beads/cell. Samples were incubated in a 37°C incubator for 72 h. Proliferation of CFSE-labeled T_Eff was evaluated by flow cytometry and expressed as FlowJo software-calculated division index (the average number of cell divisions). Supernatants were collected for ELISA detection of IFN-γ in the cocultures. Percent suppression was calculated using proliferation measurements and the following formula: percent suppression = (1 − T_Eff in T_Reg:T_Eff/T_Eff alone) × 100.

Concentration of leptin and IFN-γ were determined by commercial ELISA kits (R&D Systems) according to the manufacturer’s instructions.

Analyses were performed using Prism 4 software (GraphPad, San Diego, CA). Nonparametric testing among three groups was performed by Kruskal-Wallis ANOVA. Comparisons among three groups were performed by Dunn’s multiple comparison test. The p values <0.05 were considered statistically significant.

The effects of fasting were compared among WT mice, leptin-deficient ob/ob mice, leptin receptor-deficient db/db mice, and lupus prone NZB/W mice. Control animals were fed ad libitum or starved 48 h (treated or not with leptin). As expected, starvation led to a reduction in body weight in all groups of mice when compared with ad libitum-fed animals (Table I). Starvation also associated with a significant reduction in the number of splenocytes and T_Eff in starved WT, ob/ob, and NZB/W mice (Supplemental Table I).

It is well established that fasting associates with the reduction of circulating leptin levels (5). Accordingly, WT and NZB/W mice starved for 48 h had hypoleptinemia when compared with ad libitum-fed animals, whereas no changes in leptin levels were observed in starved ob/ob mice [which cannot produce functional leptin because of a mutation in the leptin gene (9)] and db/db mice [which carry a mutation in the leptin receptor (10)] (Table I). The administration of leptin to starved mice restored the circulating levels of this adipokine to normal concentrations in leptin-sufficient mice (Table I), and spleen cellularity returned to values similar to those seen in ad libitum-fed controls in starved mice that received leptin (Supplemental Table I).

Because we reported that leptin can constrain the expansion of T_Reg in vivo (8), we investigated whether the reduction in circulating levels of leptin induced by fasting could affect the T_Reg numbers. It was found that the frequency of T_Reg in starved WT and NZB/W lupus mice was significantly increased as compared with ad libitum-fed mice (Fig. 1). Of interest, leptin administration to ob/ob mice, which had no change of leptin levels after starvation (Table I), associated with a reduction in the number of T_Reg (Fig. 1C, 1D), suggesting that leptin per se has inhibitory effects on the expansion of T_Reg in vivo. Also, db/db mice (which are hyperleptinemic but unresponsive to leptin) showed no change in T_Reg frequency after starvation, and leptin replacement did not change this finding (Fig. 1E, 1F). The T_Reg that had expanded after starvation had similar CD25 expression (which can be a marker of immune cell activation) in the different groups of animals (Supplemental Table I), and the frequency of CD4⁺CD25⁻ cells (T_Eff) in PBMC was reduced by fasting and restored by leptin replacement in WT, ob/ob, and NZB/W mice (Supplemental Table I).

Taken together, these results indicate that fasting associates with an increase in T_Reg frequency when the leptin pathway is intact (in WT and NZB/W mice), and the T_Reg frequency remains unaltered when fasting cannot alter leptinemia due to an impairment of the leptin/leptin receptor axis (such as in ob/ob and in db/db mice). These events depend on leptin because in WT and NZB/W mice the T_Reg frequency returned to numbers comparable to those found in ad libitum-fed mice following leptin replacement (Fig. 1).

To also address whether the observed changes in T_Reg frequency associated or not with a normal function of the suppressor cells, in vitro suppression assays were performed. The results showed that the changes in T_Reg frequency under the different conditions (ad libitum, fast, and fast + leptin) did not associate with differences in the suppressive capacity of T_Reg on T_Eff (Fig. 2, Supplemental Table II), indicating that T_Reg expanded by fasting were functionally comparable among groups.

Altogether, these results widen earlier observations that showed that caloric restriction ameliorated SLE manifestations (2, 11, 12). The observation that functional T_Reg could expand during fasting in lupus mice is relevant to the disease pathogenesis because the frequency of T_Reg is lower in NZB/W mice as compared with WT mice (13, 14), and this aspect has been linked to an inability of the dysfunctional immune system in SLE to control disease course and progression (15).

The link between nutritional status and immune regulation suggested by recent reports that showed that the mammalian target of rapamycin pathway could influence the activity of T_Reg through leptin (16) identifies in the current study a new possibility to modulate T_Reg activity in SLE via leptin-based intervention [also considering that leptin is elevated in SLE patients (17)]. In particular, the frequency of T_Reg could be tuned for therapeutic purposes in SLE not only through fasting but also, more specifically, using neutralizing Abs to leptin to promote the expansion in vivo of functional T_Reg in the disease.

The authors have no financial conflicts of interest.