CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority from U.S. Provisional Application No.

61/528,995, filed August 30, 2011 and U.S. Provisional Application No. 61/544,455, filed October 7, 2011 which are herein incorporated by reference in their entirety.

GOVERNMENT FUNDING

[0002] This invention was made with government support under Grant Nos.

R01DK082582 and R01DK082582-S1 awarded by the National Institutes of Health

NIDKK. The government has certain rights in the invention.

FIELD OF THE INVENTION

[0003] Obesity is associated with a state of chronic low-grade inflammation and the present invention establishes that adipose-resident natural killer T (NKT) cells attenuate inflammation in adipose tissue and improves systemic glucose homeostasis in mice at different stages of obesity. Accordingly, the present invention provides methods of treating type-2 diabetes or those at risk for type-2 diabetes using activators of adipose- resident NKT cells. Such activators include particular glycolipids (e.g., a- galactosylceramide and its analogs other than sulfatide analogs) and cytokines that promote M2 macrophage polarization.

[0004] Natural killer T (NKT) cells have been implicated in autoimmunity, microbial infection and cancer and represent an important immunotherapeutic target (Cerundolo et al. (2009) Nat Rev Immunol 9, 28-38). Unlike conventional CD4 ⁺ and CD8 ⁺ T cells, NKT cells recognize and are activated by lipid antigens presented by the MHC class I homologue molecule CD Id on antigen presenting cells such as macrophages and dendritic cells (Bendelac et al. (1995a) Science 268, 863-865; Bendelac (1995b) J Exp Med 182, 2091-2096; Porcelli et al. (1992) Nature 360, 593-597; Beckman et al. (1994) Nature 372, 691-694). Among different types of NKT cells, type 1 or invariant NKT cells are the most abundant and best characterized (Godfrey et al. (2010) Nat Immunol 11, 197-206). The prototypical lipid antigen is the marine sponge-derived a-galactosylceramide (aGalCer) (Kawano et al. (1997) Science 278, 1626-1629), which is not found in mammals, and has been used widely to specifically study type 1 NKT cells in vivo (Bendelac et al. (2007) Annu Rev Immunol 25, 297-336). Upon aGalCer activation, NKT cells secrete large amounts of ¾1 cytokine, IFN-γ, and T _H 2 cytokines, IL-4 and IL-13 (Yoshimoto et al. (1995) Science 270, 1845-1847). aGalCer challenge has been shown to protect against the development of type-1 diabetes (Miyamoto et al. (2001) Nature 413, 531-534; Hong et al. (2001) Nat Med 7, 1052-1056) and autoimmune encephalomyelitis, although some of these findings remain controversial (Bendelac et al. (2007)).

[0005] The ability of NKT cells to secrete both T _H 1 and T _H 2 cytokines upon activation underlies their unique regulatory functions that bridge innate and adaptive immunity (Bendelac et al. (2007); Berzins et al. (2011) Nat Rev Immunol 11, 131-142). It is important to note that secretion of T _H 1 and/or T _H 2 cytokines by NKT cells is context- dependent (Bendelac et al. (2007); Berzins et al. (2011)) as the nature of the lipid antigens, the subsets of NKT cells, and the microenvironment of the tissues may have significant influences on their cytokine profiles (Miyamoto et al. (2001); Bai et al. (2009) Proc Natl AcadSci USA 106, 10254-10259). Indeed, studies have shown that NKT cells may promote or suppress immune processes by skewing adaptive immune responses towards either a T _H 1 or T _H 2 response (Bendelac et al. (2007); Kronenberg (2005) Annu Rev Immunol 23, 877-900). However, whether and how NKT cell activation affects obesity- associated inflammation remains to be characterized.

[0006] Obesity is associated with a state of chronic low-grade inflammation that significantly contributes to the pathogenesis of this disorder and its associated

complications. At late stages of obesity, a variety of immune cells, most notably macrophages (Weisberg et al. (2003) J Clin. Invest. 112, 1796-1808; Xu et al. (2003) J Clin. Invest. 112, 1821-1830), CD8 ⁺ T ( ausch et al. (2008) IntJ Obes (Lond) 32, 451-

463; Nishimura et al. (2009) Nat Med 15, 914-920), mast cells (Liu et al. (2009 Nat Med

Ί 15, 940-945), B cells (Winer et al. (2011) Nat Med 17, 610-617) and myeloid-derived suppressor cells with immunosuppressive functions (Xia et al. (201 1) J Biol Chem 289, 23591-23599) infiltrate adipose tissue during diet-induced obesity, with concurrent downregulation of other immune cells such as regulatory T cells (Treg) (Feuerer et al. (2009) Nat Med 15, 930-939). Some of these immune cells may affect the polarization of macrophages to classical (Ml) or alternative (M2) activation status via directly or indirectly influencing the local T _H I or T _H 2 cytokines in the adipose microenvironment (Olefsky et al. (2010) Annu Rev Physiol 72, 219-246; Donath et al. (2011) Nat Rev Immunol 11, 98-107; Gordon S. et al. (2010) Immunity 32, 593-604). Unlike Ml, M2 macrophages may contribute to improved insulin sensitivity due to their capacity to resolve inflammation and facilitate wound healing (Gordon et al. (2010); Lumeng et al. (2007) J Clin. Invest. 117, 175-184; Odegaard et al. (2008) Cell Metab 7, 496-507; Odegaard et al. (2007 Nature 447, 1116-1120). Bias towards M2 polarization can be promoted by immunomodulatory T _H 2 cytokines such as IL-4 and IL-13. Past studies have identified adipocytes, CD4 ⁺ T cells, and eosinophils in adipose tissue as a potential source of T _H 2 cytokines (Kang et al. (2008) Cell Metab 7, 485-495; Wu et al. (2011) Science 332, 243- 247). Schipper reviews the role of adipose-tissue resident cells in immunometabolism (Schipper et al. (2012) Trends Endocrinol Metab 23, 407-415).

[0007] Further, a study in 2009 reported the presence of NKT cells in adipose tissue, whose abundance seems decreased with obesity (Lynch et al. (2009) Eur J Immunol 39, 1893-1901) and two more recent studies demonstrated the lack of metabolic effect in NKT-deficient CDld' ^~ mice following long-term HFD feeding (Kotas et al. (2011) PLoS ONE 6, e25478; Mantell et al. (2011) PLoS ONE 6, el9831). Despite the seeming negligible role of NKT cells from such loss of function studies, the data reported herein asked whether gain-of-function of NKT cells affects obesity-associated glucose homeostasis and positively demonstrated that aGalCer-mediated activation of NKT cells enhances alternative macrophage polarization in adipose tissue and improves glucose homeostasis in animals at different stages of obesity, providing a new therapeutic target for treating type-2 diabetes, and obesity-associated inflammation. SUMMARY OF THE INVENTION

[0008] Natural killer T (NKT) cells are therapeutic targets in various disease models and under clinical trials for cancer patients. However, their function in obesity and type 2 diabetes heretofor remained unclear. The data gathered in the present invention show that adipose tissues of both mice and humans contain a population of type-1 NKT cells, whose abundance decreases with increased adiposity and insulin resistance. Although loss-of- function of NKT cells had no effect on glucose tolerance in animals with prolonged high- fat diet (HFD) feeding, activation of NKT cells by lipid agonist a-galactosylceramide (aGalCer) enhances alternative macrophage polarization in adipose tissue and improves glucose homeostasis in animals at different stages of obesity. Furthermore, the effect of NKT cells is largely mediated by the IL-4/STAT6 signaling axis in obese adipose tissue.

[0009] According, the present invention is directed to a method for treating an individual having type-2 diabetes or at risk for type-2 diabetes which comprises administering an activator of adipose-resident NKT cells for a time and in an amount to improve glucose tolerance or to decrease insulin resistance in said individual. The activators of the invention include, glycolipids such as a-galactosylceramide and its analogs, with the proviso that said analog is not sulfatide or a sulfatide analog, the cytokines IL-4, IL-10, IL-12 or IL-13 or any other T _H 2 -promoting cytokine or a compound that promotes M2 macrophage polarization.

[0010] In some embodiments, the above method can be used to reduce or ameliorate obesity-associated inflammation.

[0011] Another aspect of the invention relates to an isolated population of mammalian adipose-resident type-1 NKT cells which has at least 50% to about 80% CD4 ^" CD8 ^" type-1 NKT cells and about 20 to 30% CD4 ⁺ CD8 ^" type-1 NKT cells. These cells exhibit the other markers found on type-1 NKT cells (reactive to aGalCer- loaded CDld-tetramer, TCRP ⁺ CD44 ^hi CD69 ⁺ CD25 ^") . In some embodiments, the cells are of murine or human origin. [0012] Still a further aspect of the invention provides an in vitro method to screen for an activator of adipose-resident NKT cells by (a) preparing an isolated population of adipose- resident type-1 NKT; (b) contacting said cells with a candidate activator for a time sufficient to induce said cells to release a TH2 cytokine; and (c) detecting whether said cells have produced increased levels of a TH2 cytokine relative to untreated adipose- resident NKT cells. Methods of detecting gene and protein expression on are well known and include, for example, northern blots and other nucleic acid hybridization techniques, RNA-Seq, various PCR techniques, including q-PCR and RT-PCR, ELISAs, immunoblots and immunohistochemical staining.

[0013] In some embodiments, the method to screen for an activator of adipose-resident NKT cells comprises (a) preparing an isolated population of adipose-resident type-1 NKT; (b) contacting said cells with a candidate activator for a time sufficient to promote M2 macrophage polarization; and (c) detecting whether said cells have increased expression, relative to untreated adipose-resident NKT cells, of an M2 gene selected from any one or more of the group consisting of Arginase 1 (Argl), chitinase 3 -like 3 (Chi313), C-type lectin domain family 7,member a (Clec7a), Programmed cell death 1 ligand 2 (Pdcllg2), Resistin like alpha (Retnla), Interleukin 1 receptor antagonist (Illrn), and Interleukin receptor 4 alpha (I14ra). The foregoing detection techniques, or any other known in the art, can be used.

[0014] Yet another aspect of the invention is drawn to an in vivo method to screen for an activator of adipose-resident NKT cells by (a) administering a candidate activator to one or more mice maintained on a high fat diet (HFD) for at least about 4 days up to about 24 weeks or to one or more control mice; (b) sacrificing said one or more mice from about 4 to about 7 days after administering said activator; (c) preparing white adipose tissue

(WAT) from said mice; and (d) detecting whether (i) said WAT contains increased expression, relative to expression in WAT from control mice, of any one or more of IL4,

IL13, IL12, Argl, Chi3I3, Clec71, Pdcdllg2, Retnla, ILlm and IL4ra, (ii) M2 macrophage polarization increases in said WAT relative to that in WAT from control mice, and/or (iii) whether one or more T _H 2 cytokine are produced. The foregoing detection techniques, or any other known in the art, can be used. The mice used in this method can be wild-type mice or have a genetic predisposition to obesity, such as ob/ob mice.

BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Fig. 1 shows that adipose-resident type I NKT cells increase with 4d HFD and are predominantly CD4 ^" CD8 ^" . Figs. 1A and IB show the analysis of NKT cells in epididymal adipose tissue of 6w-old male lean mice on an LFD (white bars) and after a 4d HFD (black bars). Panels A shows the total number of NKT cells per gram of adipose tissue and Panel B shows the percent of NKT cells in total adipose lymphocytes (n=12 per cohort, with 3 repeats). Fig. 1C provides FACS characterization of cell-surface markers CD4-CD8 of aGalCer-CDld-tetramer-positive NKT cells from indicated tissues. Numbers in the flow histogram indicate the percent of CDld-tetramer-positive NKT cells (n=10-12 per cohort, at least 4 repeats). Values represent mean ± s.e.m. *, <0.05.

[0016] Fig. 2 illustrates that the abundance of adipose NKT cells decreases with HFD feeding in two obese mouse models. Figs. 2 A an 2B show the body and epididymal fat weights, and fasting glucose levels, respectively, of wildtype C57BL/6 male mice fed an HFD for lw to 24w, compared to age-matched male mice on an LFD (n=12-15 per cohort). Fig. 2C provides the percentages of NKT and CD8 ⁺ T lymphocytes in total lymphocytes in adipose tissue during HFD feeding compared to age-matched LFD cohort (n=12 per cohort, 3 repeats). Fig. 2D provides total cell numbers of various T lymphocytes per g adipose tissue during HFD feeding (n=12, 3 repeats). Fig. 2E provides the body weight of adult 36-40w-old ob/ob mice relative to wildtype controls. Figs. 2F and 2G provide the percentages of NKT and CD8 ⁺ T cells in adipose tissue and spleen of ob/ob mice compared to age-matched WT lean animals (n=10 per cohort, 2 repeats), respectively. Values represent mean ± s.e.m. *, <0.05, **, <0.01, and ***, <0.005. In panels A-D, white bars are LFD mice; black bars are HFD mice. In panels E-F, white bars are wildtype mice (WT) and black bars are ob/ob mice. [0017] Fig. 3 shows that aGalCer challenge activates NKT cells in adipose tissue. Fig. 3A is a schematic diagram for three gain-of-function models with 4d-, 8w-, and 24w-HFD feeding. 6w-old mice that have been placed on HFD for 4d, 8w and 24w were injected i.p. with aGalCer or vehicle (veh) on day 0 and day 2. GTT was conducted on day 4 and mice were sacrificed for tissues on day 5. Fig. 3B shows BrdU labeling of CDld-tetramer- positive NKT cells in adipose tissue following aGalCer challenge. Figs. 3C and 3F show flow analysis of NKT cells in adipose tissue of mice on 4d or 8w HFD, respectively, with the indicated treatments. Number refers to the percentage of NKT cells in total CD45 ⁺ lymphocytes in stromal vascular cells (SVC) of adipose tissue. Figs. 3D and 3E show H&E sections of adipose tissue of WT and CDld ^"7" mice after 4d HFD, respectively, with the indicated treatments. Fig. 3G shows H&E sections of adipose tissue of WT mice of 8w HFD for the indicated treatments.

[0018] Fig. 4A graphically depicts the results of a glucose tolerance test (GTT) for WT mice on LFD (diamonds), WT on HFD (circles) and CDld ^"7" mice on HFD (triangles) for 8-weeks (n=8-9 per cohort, 2 repeats), respectively. Figs. 4B-D show the body and epididymal fat weights of mice on LFD (white), HFD with vehicle (grey) or HFD with aGalCer injection (dark) for 4 days, 8weeks or 24 weeks, respectively. Values represent mean ± s.e.m. *, <0.05, **, <0.01, ***, O.005.

[0019] Figs. 5A and 5B graphically quantitates the indicated cell numbers per g adipose tissue for total CD45 ⁺ lymphocytes, NKT cells, CD8 ⁺ T cells, and F4/80 ⁺ CD1 lb ⁺ macrophages of mice that have been on LFD, HFD with vehicle or HFD with aGalCer challenge for 4d or 8w, respectively (n=4 per cohort). Fig. 5C shows a flow analysis of NKT cells in adipose tissue of 14w-old mice that have been on LFD, HFD with vehicle or HFD with aGalCer challenge for 24w (n=3-4 per cohort). Number refers to the percentage of NKT cells in total CD45 ⁺ lymphocytes in SVC of adipose tissue. Values represent mean ± s.e.m.*, P<0.05, **, <0.01, and ***, <0.005. For panels A and B, the bars represent LFD (white), HFD with vehicle (grey) or HFD with aGalCer injection (dark). [0020] Fig. 6 shows that NKT cell activation improves systemic glucose homeostasis in murine diet-induced obesity (DIO). Figs. 6A-6C depicts the results of a GTT of WT mice on LFD (dark line-black), HFD injected with vehicle (light gray line-red) or HFD injected with aGalCer (medium grey line-blue) for different lengths of time. The mice in Fig. 6A were on 4d HFD (n=12-15 per cohort; 3 repeats); in Fig. 6B the mice were on 8w HFD (n=9-10 per cohort: 2 repeats); and in Fig. 6C were on 24w HFD (n=3-4 per HFD cohort and n=10 for LFD). * refer to the P values comparing between HFD + veh and HFD + aGalCer groups. Fig. 6D shows the fasting insulin levels in mice under 8w HFD (n=9- lOper cohort, 2 repeats). Each bar represents LFD (white), HFD with vehicle (grey) or HFD with aGalCer injection (dark). Fig. 6E depicts the results of a GTT of CDld ^"7" mice on LFD (dark line-black), HFD injected with vehicle (light gray line-red) or HFD injected with aGalCer (medium grey line-blue) under 4d HFD (n=9-10 per cohort, 2 repeats). Fig. 6F depicts the results of a GTT of WT mice on LFD with veh (circles) or aGalCer injection (triangles) (n=10 per cohort, 2 repeats) after about 6 weeks on LFD. Values represent mean ± s.e.m. *, <0.05, and ***, <0.005.

[0021] Fig. 7 shows that NKT cell activation increases M2 polarization in adipose tissue of DIO mice. Panel A provides a Q-PC analysis of Ml genes (first three genes) and M2 genes (remaining six genes) in white adipose tissue (WAT) in WT (upper) and CDld ^"7" (lower) mice on 4d HFD (n=10-15 per cohort; 2-3 repeats). Panel B shows a similar Q- PCPv analysis of the same Ml and M2 genes in WAT of 8w HFD WT mice (n=4-5 per cohort, 2 repeats). Figs. 7C and 7D provide a Western blot analysis of Argl expression in WAT of WT and CDld ^"7" cohorts with vehicle or aGalCer treatment (n=3-4 per cohort, 2 repeats). Each lane represents an independent sample in Fig. 7C. Quantitation is shown in Fig. 7D, with the values of the LFD + veh sample set at 1. Values represent mean ± s.e.m. n.s. not significant. *, O.05, **, O.01, and ***, O.005. For panels A, B and C, the bars represent LFD (white), HFD with vehicle (grey) or HFD with aGalCer injection (dark).

[0022] Fig. 8A provides a Q-PCR analysis of adipose tissue of 30w-old mice that have been on LFD (white), HFD with vehicle (grey) or HFD injected with aGalCer (dark) for 24w (n=3-4 per cohort). Fig. 8B shows western blots of Argl expression in the liver of the three cohorts under 4d (upper) or 8w (lower) HFD feeding (n=4-5per cohort, 2 repeats). HSP90 is a loading control.

[0023] Fig. 9 establishes that NKT cell activation increases IL-4 signaling in adipose tissue. Fig. 9A depicts a heat map showing the fold-change of gene expression in WAT of WT or CDld ^"7" mice on 4d HFD with or without aGalCer injection (n=3-4 per cohort). Each lane is one individual. A total of 1,556 differentially-expressed genes with q < 0.001 and fold-change > 2 are shown. Fig. 9B provides the top 10 genes upregulated by aGalCer and their expression changes in IL-4-treated BMDM (GSE25088). n.d., not detected.

Scale: 0-1, blue shades; 1-100, red shades. In Fig. 9A, the columns labeled WT WAT Veh and both CDld-/-_WAT are mixtures of pale blues and reds. For the remaining column, WT WAT aGalCer, the intensities range from light to deep shades of red. In the expanded heat map in Fig. 9B, the intensities in the columns labeled WT WAT Veh and both CDld-/-_WAT are all white/blue shades. For the BMDM CON column, all boxes are white whereas in the IL-4 column, the boxes are red except for the third gene (light blue) and the tenth gene (dark blue).

[0024] Fig. 9C is a Venn diagram showing the overlap of significantly upregulated genes in aGalCer-injected WAT vs. IL-4-treated BMDM datasets. As not all aGalCer- induced genes existed in the IL-4_BMDM dataset, only 1,157 genes, rather than 1,556 genes induced by aGalCer, were included in the Venn diagram. Fig. 9D is a Western blot of Argl expression in WAT of WT and CDld ^"7" cohorts used in Panel A (n=3-4 per cohort, 2 repeats). Each lane represents an independent sample. Fig. 9E shows the quantitative analysis of the blots in Panel D with the values of the WT + veh sample set at 1. Fig. 9F depicts the Q-PC analysis οΐΙ14 mRNA levels in WAT of WT or CD Id ^"7" mice on 4 d HFD with or without aGalCer injection (n=4 per cohort). Values represent mean ± s.e.m. ***, <0.005. For panels E and F, the bars represent vehicle injection (white bar) and aGalCer injection (black bar). [0025] Fig. 10A presents a dendrogram showing the hierarchical clustering of microarray data of individual mice establishes a separation of the WT mice injected with aGalCer from the other groups. Fig. 10B shows the NES for the most significant upregulated pathways identified by GSEA (all with false discovery rate q-value < 0.0003) in WT mice injected with aGalCer. The "T _H 1/T _H 2 differentiation" and "IL-4 pathway" highlighted in pale grey (red). Fig. IOC diagrammatically presents the canonical IL-4 pathway and illustrates the fold induction of each key component by aGalCer in WAT. STAT6 is activated by phosphorylation and not at the transcriptional level. The color scale is shown at the bottom.

[0026] Fig. 11 establishes that the NKT cell activation signal acts through the IL- 4/STAT6 axis in adipose tissue. 6-7w-old mice WT or IL-4 ^"7" mice were placed on either LFD or HFD for 4d with two aGalCer or vehicle injections prior to metabolic phenotyping (n=7-8, 2 repeats). Fig. 1 1 A provides bar graphs showing the weight of body and epididymal adipose tissues on the left and right, respectively. Figs. 1 IB and 11C show the results of a GTT in mice injected with vehicle (circles) or aGalCer (triangles) for WT and IL-4 ^"7" mice, respectively. Fig. 1 ID provides a Q-PC analysis for the same Ml and M2 genes in Fig. 7 from WAT of WT mice (left two bars for each gene) or of IL-4 ^"7" mice (right two bars for each gene). Within each pair, the left bar is injection with vehicle (veh) and the right bar is injection with aGalCer (n=6 per cohort, 2 repeats). Fig. 1 IE shows Western blots for Argl, (pY-) STAT6 and (pY-) STAT3 in WAT of WT and IL-4 ^"7" 4dHFD mice injected with vehicle or aGalCer (n=3-5 per cohort, 2 repeats). In Fig. 1 IF, the ratios of Argl to HSP90 and p-Y STAT3/6 to total STAT3/6 are quantitated. Fig. 11G shows Western blots for p-Y STAT6 proteins in the liver of the WT mice under LFD, 4d HFD with vehicle or 4d HFD with aGalCer injection (n=4-5 per cohort). In Figs. 1 IE and 11G, each lane represents an independent sample. Values represent mean ± s.e.m. *, <0.05, **, O.01, ***, O.005; n.s., not significant. For Figs. 11A and 1 IF, open bars represent 4d HFD mice injected with vehicle; black bars represent 4d HFD mice injected with aGalCer. [0027] Fig. 12A shows H&E sections of adipose tissue from WT mice (left) and IL-4 ^{" "} mice (right) on a 4d HFD after aGalCer injection (n=4-5 mice). Fig. 12B graphically depicts the total number of lymphocytes in adipose tissue of WT or IL-4 ^"7" mice on 4d HFD with or without aGalCer challenge (n=5-6 per cohort, 2 repeats). Fig. 12C provides a Q- PCR analysis for the same Ml and M2 genes in Fig. 7 from WAT of WT, PPARy ^"7" (lower panel), PPAR5 ^"7" (upper panel) mice following 4 day HFD feeding with or without aGalCer injection. Data normalized to "WT+vehicle" whose value was set at 1. Values represent mean ± s.e.m. **, P<0.0\, and ***, <0.005. For all bar graphs, from left to right, the bars represent WT with vehicle injection (white), WT with aGalCer injection (red), IL-4 ^"7" with vehicle injection (green) and IL-4 ^"7" with aGalCer injection (blue).

[0028] Fig. 13 graphically depicts the correlations between adipose NKT cells and various metabolic parameters in humans. NKT cells in visceral adipose tissues of 39 human subjects were analyzed by two oligo sets: Oligos 1-2 measured the Ya24TCR mRNA levels of type 1 NKT cells. As a control, oligo 3 measured the TCR mRNA levels of total T cells. Fig. 13A shows a scatter plot Va24 type 1 NKT mRNA levels versus BMI for Oligo 1 (left) and 2 (right). Fig. 13B shows a scatter plot of total T cell mRNA levels versus BMI. Fig. 13C shows a scatter plot of insulin resistance as measured by HOMA versus type 1 NKT mRNA levels measured with Oligo 1. Fig. 13D shows a scatter plot of glucose levels measured at the 2 h point of an oral GTT versus type 1 NKT mRNA levels measured with Oligo 1.

[0029] Fig. 14A illustrates a cellular model for NKT in Obesity. Fig. 14B depicts a mechanism of lipid activation of NKT cells at the onset and at late stage of obesity.

DETAILED DESCRIPTION OF THE INVENTION 1. Definitions and Abbreviations

[0030] Type-2 diabetes is a chronic metabolic disease in which individuals have insulin resistance and inadequate insulin regulation, preventing proper glucose utilization. Most individuals with type-2 diabetes are obese, and experience hyperglycemia, usually detectable as abnormal or impaired fasting glucose or abnormal or impaired glucose tolerance before diagnosis.

[0031] As used herein "an individual at risk for type-2 diabetes" is an individual who has a set of symptoms and risk factors associated with type-2 diabetes or pre-diabetic conditions leading to type-2 diabetes. Additionally or alternatively, individuals at risk for type-2 diabetes will have a glycated hemoglobin (AIC) level between 5.7-6.4%. Values above that range indicate the presence of diabetes, whereas values below that range are normal. In accordance with well established clinical practice, tests to determine random and fasting glucose levels can also be used to establish whether an individual is at risk for type-2 diabetes.

[0032] Whether an individual has type-2 diabetes or is at risk therefor is determined by standard clinical, medical criteria and diagnosis.

[0033] Adipose tissue, unless qualified in context, includes both visceral fat and subcutaneous fat. It may be composed of white adipose tissue or brown adipose tissue.

[0034] The terms "type-1 NKT" or "invariant NKT" (iNKT) cells are used

interchangeably. These cells are distinct cell types from type-II NKT.

[0035] As used herein, "an activator of adipose-resident NKT cells" includes but is not limited to cytokines and glycolipids, including aGalCer and its analogs (other than sulfatides or a sulfatide analogs) and other compounds (whether small molecule, oligonucleotides, nucleic acid, peptides, protein or otherwise) that stimulate an increase of type-1 NKT in adipose tissue, especially in murine adipose tissue, or promote M2 macrophage polarization. The aGalCer analogs for use in the invention are those in which the galactosyl moiety can be replace by other monosaccharides, e.g., glucose, gulose and N-acetylglucosamine, those in which the lipid chain of the ceramide can be varied in length, and combinations thereof. [0036] Mammalian includes any mammal, with particular preference for murine, human and primate mammals.

[0037] The abbreviations used herein include aGalCer, a-galactosylceramide;

ANOVA, analysis of variance; DIO, diet-induced obesity; GSEA, gene-set enrichment analysis; GTT, glucose tolerance test; HFD, high-fat diet; HOMO, homeostasis model assessment; LFD, low-fat diet; NES, normalized enrichment score; NKT, natural killer T cells; OGTT, oral glucose tolerance test; Q-PC , quantitative PCR; SVC, stromal vascular cells; Veh, vehicle; WAT, white adipose tissue; and WT, wild type.

2. General Overview

[0038] The role of NKT cells in regulating obesity-associated inflammation has heretofore been incomplete. With the discovery and characterization of a new subset of type-I NKT cells in adipose tissue and in diet-induced obesity (DOI), this role is being elucidated. In the present invention, adipose-resident type-1 NKT cells from mice WAT have been characterized and shown to be primarily CD4 ^" CD8 ^" type-1 NKT cells with about 20 to 30% CD4 ⁺ CD8 ^" type-1 NKT cells and sulfatide insensitive, which when activated lead to T _H 2-skewed anti-inflammatory responses and mediate glucose homeostasis. In contrast, spleen and liver type-1 NKT cells are predominantly CD4 ⁺ CD8 ^" type-1 NKT cells, with at most a few percentage of CD4 ^" CD8 ^" type-1 NKT cells, are activated by sulfatide, and when activated lead to T _H 1 -skewed pro-inflammatory response.

[0039] Prior studies and data herein with CD Id -/- mice fed an HFD confirmed loss of NKT cells had no metabolic effect and suggested that the NKT cells were dispensable for glucose homeostasis in loss-of function studies. Further, a study using β2 -microglobulin (P2m)-deficient mice which lack all MHC class I molecules including CD Id concluded that NKT cells infiltrate adipose tissue and that loss of NKT cells reduces inflammation in obesity while activation of NKT cells by one injection of aGalCer had mild effect on glucose tolerance in HFD animals (Ohmura et al. (2010) Arterioscler Thromb Vase Biol 30, 193-199), opposite to the findings in the present invention. However, interpretation of those results is confounded by the fact that the β2ιη knockout mice not only are deficient in NKT but also CD8 ⁺ T cells (Koller et al. (1990) Science 248, 1227-1230; Zijlstra et al. (1990) Nature 344, 742-746). Accordingly, the absence of CD8 ⁺ T cells may account for the reported phenotypes (Nishimura et al. (2009)).

[0040] In contrast, the gain-of-function study of the present invention demonstrated that aGalCer-mediated NKT cell activation promotes M2 macrophage polarization in adipose tissue and exerts a salutary effect on systemic glucose homeostasis through the IL- 4/STAT6 signaling axis. These gain-of-function data are in line with the known function of NKT cells in altering T _H 1/T _H 2 responses in vivo (Bendelac et al. (2007)). The lack-of- effect of NKT cells at late stages of obesity in the absence of stimulation is not surprising given the massive infiltration and expansion of other cells such as CD8 ⁺ T cells and macrophages (Olefsky J.M. et al. (2010) Annu Rev Physiol 72, 219-246; Donath M.Y. et al. (2011) Nat Rev Immunol 11, 98-107) and given the concomitant reduction of NKT cells.

[0041] However, upon activation by a potent stimulus such as aGalCer, NKT cells have significant impact on inflammatory responses in adipose tissue and systemic glucose tolerance in obese animals.

[0042] In line with the role of IL-4 in T _H 2 responses, the data herein suggest the IL-4 is an important mediator of NKT cell function in adipose tissue in the context of obesity and does not exclude a role for other T _H 2 cytokines such as IL-13, which is known to be secreted by NKT cells (Yoshimoto et al. (1995)) as supported by the observation that loss of IL-4 only partially blocked aGalCer-mediated induction of M2 genes in adipose tissue. While IL-4 and IL-13 share a common signaling pathway through the IL-4 receptor a subunit (IL-4 a) and may have diverse functions in vivo, since IL13 is involved in TH2 responses, it may have sufficient anti-inflammatory activity for improving glucose tolerance in obesity.

[0043] The further observation that, unlike their hepatic counterparts, adipose-resident NKT cells have a T _H 2-biasing effect upon aGalCer activation is very intriguing. The difference may be related to the nature of antigen presenting cells, tissue microenvironment and/or the different lineage of NKT cells in adipose tissue (Bendelac et al. (2007)). The data herein showed that unlike CD4 ⁺ CD8 ^" NKT cells in the liver, the majority of NKT cells in adipose tissue are CD4 ^" CD8 ^" , and thus may be associated with specific functional changes. Indeed, earlier studies have shown that CD4 ^" CD8 ^" vs. CD4 ⁺ CD8 ^" NKT cells likely represent functionally separate lineages that may promote different T _H response with distinct capacities to secrete T _H I and T _H 2 cytokines (Bendelac et al. (2007); Lee et al. (2002) J Exp Med 195, 637-641; Benlagha et al. (2002) Science 296, 553-555; Pellicci et al. (2002) J Exp Med 195, 835-844). Alternatively, the tissue-specific NKT effect may reflect a unique microenvironment of adipose tissue in terms of antigen presenting cells. Since adipocytes express CD Id transcripts, they may be able to present lipids to and directly activate NKT cells whereas Kupffer cells in the liver have been shown to be important for aGalCer-mediated NKT cell activation (Schmieg et al. (2005) Proc Natl Acad Sci USA 102, 1127-1132). Finally, as NKT cell activation by bacterial lipid antigens may be toll-like receptor 4-dependent and IL-12-mediated (Brigl et al. (2003) Nat Immunol 4, 1230-1237; Brigl et al. (2011) J Exp Med 208, 1163-1177), it will be interesting to investigate whether different cytokine environments of the liver vs.

adipose tissue may explain the tissue-specific NKT effect upon aGalCer activation.

[0044] The data herein show that aGalCer-mediated NKT activation leads to activation of the IL-4/STAT6 signaling axis in adipose tissue, promoting M2 macrophage polarization. The mechanism of how IL-4 mediates macrophage polarization in adipose tissue remains unclear. It has been shown that STAT6 may modulate the activities of nuclear receptors (PPARy or ΡΡΑΡνβ/δ) (Chawla (2010) Circ Res 106, 1559-1569) or induce the expression of histone H3K27 demethylase Jmjd3 (encoded by Kdm6b gene) (Ishii et al. (2009) Blood 114, 3244-3254; Satoh et al. (2010) Nat Immunol 11, 936-944) to influence M2 macrophage polarization. The data show that M2 marker genes induced by aGalCer are not affected by lack of either PPARy or ΡΡΑΡνβ/δ in myeloid cells, consistent with two recent studies demonstrating that PPARy and PPARp/δ in macrophages or myeloid cells are dispensable for the T _H 2 responses (Szanto et al. (2010) Immunity 33, 699-712; Marathe et al. (2009) J Lipid Res 50, 214-224). The discrepancies may be due to different genetic backgrounds of mice, fatty acids composition of diets, or gut microbiota.

[0045] In line with the animal data, the abundance of type 1 NKT cells in adipose tissue decreases in obese humans. Although this was initially reported by an earlier study (Lynch et al. (2009)), the present data, using a much larger cohort, further demonstrate that abundance of type 1 NKT cells in adipose tissue negatively correlates with BMI, insulin resistance and OGTT glucose levels. The near-perfect correlations between adipose NKT and metabolic parameters suggest that these cells may be involved in the regulation of insulin sensitivity in obese humans. Since aGalCer is not toxic, is well tolerated in humans and often stimulates the expansion of residual NKT cell populations in other disease settings (Berzins et al. (2011); Giaccone et al. (2002) Clin Cancer Res 8, 3702- 3709), one embodiment of the present invention provides a method wherein NKT- activating glycolipids are used in treating type 2 diabetes. Similar methods using NKT lipid agonists can skew T _H 2 bias in adipose tissue and hence delay or ameliorate the development of inflammation and type 2 diabetes in obese patients.

3. Methods of Treatment

[0046] The present invention provides a method for treating an individual having type-2 diabetes or at risk for type-2 diabetes by administering an activator of adipose-resident NKT cells for a time and in an amount to improve glucose tolerance or to decrease insulin resistance in said individual. Such activators can also be used to treat, e.g., reduce or ameliorate, obesity-associate inflammation.

[0047] Activators of adipose-resident NKT cells include glycolipids, ceramides and cytokines which promote M2 macrophage polarization or cause TH2 -biased immune responses that are anti-inflammatory. Particular glycolipids include, but are not limited to a-galactosylceramide (chemical name, (2S,3S,4R)-l-0-(alpha-D-galactosyl)-N- hexacosanoyl-2-amino-l,3,4-octadecanetriol) and its analogs (provided the analogs do not include sulfatide or are sulphated) such as those with galactosyl moiety replaced by another monosaccharides, e.g., glucose, gulose and N-acetylglucosamine or by different length lipid chains in the ceramide moiety. OCH, a synthetic sphingosine truncated derivative of aGalCer (OCH chemical name, tetracosanoic acid 1-galactopyranosyloxy- 3,4-dihydroxydec-2-yl amide), is not effective in vivo.

[0048] Cytokines which can act as activators of adipose-resident NKT cells include, but are not limited to, IL-4, IL-10, IL-12 and IL-13. Any other NKT lipid agonist can be useful to activate adipose-resident NKT cells and can be identified using the screening methods described herein.

[0049] As used herein, "treatment" refers to clinical intervention in an attempt to alter the disease course of the individual or cell being treated, and can be performed either for prophylaxis or during the course of clinical disease.

[0050] As used herein, the terms "therapeutically-effective amount" and "effective amount" are used interchangeably to refer to an amount of a composition of the invention that is sufficient to result in improved glucose tolerance or decreased resistance to insulin. Glucose tolerance and insulin resistance can be assessed by standard assays known in art. Improvements in obesity associated inflammation can be assessed by reduction in one or more measures of inflammation or changes in inflammatory markers that indicate antiinflammatory effects.

[0051] A therapeutically-effective amount can be administered to a patient in one or more doses sufficient to affect glucose tolerance or reduce insulin resistance and thereby reduce the pathological consequences of the type-2 diabetes, or reduce the symptoms thereof. The amelioration or reduction need not be permanent, but may be for a period of time ranging from at least one hour, at least one day, or at least one week or more. The effective amount is generally determined by the physician on a case-by-case basis and is within the skill of one in the art. Several factors are typically taken into account when determining an appropriate dosage to achieve an effective amount. These factors include age, sex and weight of the patient, the severity of the condition, as well as the route of administration, dosage form and regimen and the desired result. [0052] The activators of the invention can be delivered by any convenient route, including oral, parenteral, s.c, i.p, i.v. routes as well as part of a dietary or feeding regimens.

[0053] The optimal dosage, delivery and/or feeding routes and frequency of aGalCer, cytokines, and other T _H -biasing lipid agonists in humans can be readily determined by the skilled artisan using routine methods.

[0054] For example, the daily dosage of the activators can be varied over a wide range from 0.01 to 1,000 mg per adult per day, depending on whether the activator is a glycolipid or cytokine. In general, the compositions contain 0.01, 0.05, 0.1, 0.5, 1.0, 2.5, 5.0, 10.0, 15.0, 25.0, 50.0, 100, 250 and 500 mg of the active ingredient. A medicament in unit dosage form typically contains from about 0.01 mg to about 500 mg of the active ingredient, and preferably from 0.05 mg to about 10 mg when the active ingredient is a glycolipid and from 1 mg to about 100 mg when the active ingredient is a cytokine. An effective amount of the drug is ordinarily supplied at a dosage level from 0.0002 mg/kg to about 20 mg/kg of body weight per day, especially from about 0.001 mg/kg to 0.10 mg/kg of body weight per day. For a glycolipid, the preferred dosage level range from 0.001 mg/kg to about 1 mg/kg of body weight per day and include a range from 0.001 mg/kg to about 0.01 mg/kg of body weight per day. It will be understood, however, that the specific dose level and frequency of dosage for any particular patient may be varied and will depend upon a variety of factors including the activity of the specific compound employed, the metabolic stability, and length of action of that compound, the age, the body weight, general health, sex, diet, mode and time of administration, rate of excretion, drug combination, the severity of the particular condition, and the host undergoing therapy.

[0055] The activators of the invention can be provided, for example, as pharmaceutical compositions for oral, sublingual, subcutaneous, intramuscular, intravenous, transdermal, inhalation, local or rectal administration, with the active principle, alone or in combination with another active principle. Such compositions can be administered in a unit administration form, as a mixture with conventional pharmaceutical supports, to animals and human beings. Suitable unit administration forms comprise oral-route forms such as tablets, gel capsules, powders, granules and oral suspensions or solutions, sublingual and buccal administration forms, aerosols, implants, subcutaneous, transdermal, topical, intraperitoneal, intramuscular, intravenous, subdermal, transdermal, intrathecal and intranasal administration forms and rectal administration forms. Capsules are one preferred embodiment. Compositions for prolonged release (e.g., including sustained or delayed release) can also be administered.

[0056] The activators of the invention can be provided, especially a glycolipid for example, in a consumer food product, including as beverages and powders or liquids and the like for mixing into a consumer food product or beverage.

4. Screening Methods

[0057] The present invention provides in vitro and in vivo methods for screening candidate compounds capable for activating adipose-tissue resident NKT cells or for other medical benefits associated with improving glucose tolerance, reducing insulin resistance and/or reducing obesity-associated inflammation.

[0058] The in vitro methods take advantage of the fact that adipose-resident NKT cells release greater anti-inflammatory mediators into the medium when stimulated with glycolipids, and particularly with aGalCer stimulation. Hence, one such method, a compound is screened for NKT activating activity by preparing an isolated population of adipose-resident type-1 NKT and contacting those cells with a candidate activator for a time and in an amount sufficient to induce increased release of a T _H cytokine relative to untreated adipose-resident NKT cells. Alternately, or in addition, one can monitor for the ability of the candidate activator to promote M2 macrophage polarization relative to controls. Such polarization can be detected, for example, by whether the cells have increased expression, relative to untreated adipose-resident NKT cells, of an M2 gene selected from any one or more of the group consisting of Arginase 1 (Argl), chitinase 3- like 3 (Chi3l3), C-type lectin domain family 7,member a (Clec7a), Programmed cell death 1 ligand 2 (Pdcllg2), Resistin like alpha {Retnla), Interleukin 1 receptor antagonist (ILlrn), and Interleukin receptor 4 alpha (IL4ra).

[0059] To detect in increased gene and protein expression of a T _H 2 cytokine or an M2 gene, a myriad of techniques are available in the art, including but not limited to northern blots and other nucleic acid hybridization techniques, RNA-Seq, various PCR techniques, including q-PCR and RT-PCR, ELISAs, immunoblots and immunohistochemical staining.

[0060] The amounts and times for contacting cells with a potential activator can be readily determined by the skilled artisans and examples thereof are provided in the Example section. For detection, medium or cells can be harvested for analysis.

[0061] This invention further provides methods of screening candidate compounds using a mouse model of obesity. In one embodiment, a method is provided for screening candidate compounds for ability to increase adipose-resident NKT cells, to improve glucose utilization and to skew gene expression to promote M2 macrophage polarization in adipose tissue. Mouse models of obesity include wildtype mice fed HFD diets for from about 4 days to about 24 weeks as well as genetic obesity models such as an ob/ob mouse. The candidate compounds can be fed to the mice or can be delivered by any convenient route, including i.p. (intraperitoneal), s.c. (subcutaneous), oral and the like. Any adipose tissue can be isolated for analysis to detect whether the compound has activated adipose- resident NKT cells. WAT is preferred.

[0062] Hence, one method to screen for an activator of adipose-resident NKT cells comprises administering a candidate activator to one or more mice a high fat diet (HFD) for at least about 4 days up to about 24 weeks or to one or more genetically obese mice. After a period of time, the mice are sacrificed for analysis of WAT. Prior to sacrifice, the mice can be subjected to GTT for assessing NKT activation by glucose tolerance relative to appropriate controls. Once WAT is obtained, one detects whether the WAT contains increased expression, relative to expression in WAT from control mice, of any one or more of IL4, IL13, IL12, Argl, Chi3I3, Clec71, Pdcdllg2, Retnla, Illm and I14ra. Alternatively or in addition, one can analyze the WAT to detect whether M2 macrophage polarization increases in treated WAT relative to untreated control WAT and/or whether one or more T _H 2 cytokine are produced. Methods for detecting such gene and protein expression are known in the art and have been described above.

[0063] The candidate activators for use in the present in vitro and in vivo screening methods can be any type of compound including small molecules, lipids, proteins, peptides, antibodies, siRNA and more and obtained in any manner. The activators can be tested singly (especially in the in vivo methods) or in combinatorial libraries, especially in the in vitro methods adapted for high through put screening. Combinatorial libraries include biological libraries; peptoid libraries (libraries of molecules having the

functionalities of peptides, but with a novel, non-peptide backbone, which are resistant to enzymatic degradation but which nevertheless remain bioactive); spatially addressable parallel solid phase or solution phase libraries; synthetic library methods requiring deconvolution; the One-bead one-compound' library method; and synthetic libraries using affinity chromatography selection.

[0064] The foregoing is considered as illustrative only of the principles of the invention. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation shown and described, and accordingly, all suitable modifications and equivalents may be resorted to, falling within the scope of the invention. All references patents, patent applications or other documents cited are herein incorporated by reference in their entirety.

Example 1

General Materials and Methods

[0065] Mouse models. WT C57/B6 (The Jackson Laboratory #000664), 6N-Lep ^ob ll (ob/ob, #000632), B6A29S6-Cdldl/Cdld2 ^tmlsph /J {CDld' #008881), and B6.129P2- IL4 ^tmlCgn /J (IL-4- , #002253) were purchased from the Jackson Laboratory and bred in our facility. The latter three have been backcrossed in the Jackson Laboratory to the B6 background over 45, 13, and 12 times, respectively. Myeloid cell-specific PPARy ^"7" or PPA 5 ^"7" mice were generated by crossing mice to the

6A29?2-Lyz2 ^tml(cre)If0 /J mice (The Jackson Laboratory #004781) (Kang K. et al. (2008) Cell Metab 7, 485-495) and have been backcrossed to the C57BL/6 background for over 10 times. Mice were housed in microisolators in our brand-new pathogen- free facility with sterile 13% LFD with 13% fat, 67% carbohydrate and 20% protein from Harlan Teklad (#2914) or 60% HFD with 59% fat, 26% carbohydrate and 15% protein from Bio-Serv Inc. (F3282). The composition of each is shown in Table 1.

[0066] Antibodies and reagents for flow cytometry. Fluorochrome- or biotin- conjugated antibodies against CD3 (145-2C11), TCRp (H57-597), CD4 (GK1.5), CD8 (YTS169), F4/80 (BM8), CDl lb (Ml/70), CD45 (30-F11), BrdU-FITC (PRB-1), avidin- PerCP and isotype control antibodies were purchased from BioLegend, UCSF Flow Core Facility or BD Biosciences. aGalCer-loaded CDld-tetramer-PE was generously provided by the NIH Tetramer Facility. Data were analyzed using the CellQuest software (BD Biosciences) and Flowjo (Flowjo.com).

[0067] Quantitation of immune cells in adipose tissue using flow cytometric analysis. Single-cell suspensions of stromal vascular cells (SVC) from adipose tissue were prepared as described (Xia et al. (2011)). SVCs from two fat pads per mouse were diluted in 120-200 μΐ PBS, from which one tenth was used for the subsequent staining. Following incubation with anti-CD 16/CD32 antibody to block Fc receptors, l x lO ⁶ cells were incubated with 20 μΐ of antibodies diluted at optimal concentrations for 20 min at 4°C. Cells were washed three times with PBS and then resuspended in 200 μΐ PBS for analysis using the FACSCalibur Flow Cytometer (BD Biosciences). NKT cells were defined as CD45 ⁺ aGalCer-loaded CD ld-tetramer ⁺ CD3/TCRp ⁺ lymphocytes; CD8 ⁺ T cells were defined as CD8 ⁺ CD45 ⁺ cells; macrophages were as CD45 ⁺ F4/80 ⁺ CD1 lb ⁺ cells. During the run, samples were completely run out and the total number of CD45 ⁺ lymphocytes or immune cells were gated and counted. The total cell number for various immune cells per g adipose tissue was calculated as (% cells in total CD45 ⁺ lymphocytes or immune cells x total CD45 ⁺ lymphocytes or immune cells)/g adipose tissue. [0068] Intracellular flow cytometric analysis. For the Brdu staining, mice were injected (i.p.) with aGalCer (lOOng per g body weight) at day 0, and then with BrdU (Sigma, 0.6 mg/lOg body weight) at 36 h and 12 h prior to sacrifice at day 3. SVC of adipose tissues were purified, labeled with cell surface antibodies and then fixed in cold 70% ethanol at - 20°C overnight. The rest of steps were performed as the regular flow cytometric analysis using BrdU-FITC antibody and DNA dye 7-AAD (Anaspec).

[0069] RNA extraction and quantitative (Q)-PCR. RNA extraction from cells and murine tissues, and Q-PCR were carried out as previously described (36) using Trizol (Invitrogen) for liver, and Trizol plus QIAeasy kit (Qiagen) for adipose tissues with DNase digestion (Roche). Q-PCR data collected on the Roche LightCycler 480 were normalized to ribosomal 132 gene in the corresponding sample. Primer sequences are listed in Table 2.

[0070] H&E histology. Adipose tissues were fixed in 4% formaldehyde, embedded in paraffin and sectioned by the Cornell Histology Core Facility. Pictures were taken using the Axiovert 200M microscope (Zeiss).

[0071] Western blot. Tissues or cells were lysed in Tris-based lysis buffer containing 1% Triton X-100. Normally, 15-30 μg of total lysates unless otherwise indicated were used in a mini SDS-PAGE as described (Sha et al. (2009) Cell Metab 9, 556-564). Antibodies specific for (p-Tyr641) STAT6 and (p-Tyr705) STAT3 (Cell signaling) and Arginase 1 (N- 20, Santa Cruz sc-18351) were used at 1 :500-2000 and for the loading control HSP90 (Santa Cruz) was used at 1 :6000. The secondary antibody goat anti-rabbit IgG HRP (1 : 10,000) was from Bio-Rad. Quantitation of band density was carried out using the ImageLab software of the Bio-Rad ChemiDoc XRS+ system following exposure.

[0072] ELISA. Blood was collected in animals upon 6 h fasting during the day.

Circulating insulin levels were measured using the kit from Millipore per supplier's protocols.

[0073] Statistical analysis. Results are expressed as mean ± s.e.m. Comparisons between groups were made using either unpaired two-tailed Student t-test of the EXCEL software for two-group comparisons or the one- or two-way ANOVA test with the Bonferroni Post-tests of the PRISM software for multi-group comparisons. For human studies, statistical analysis was performed with the SPSS 11.5 statistical software package. P<0.05 was considered as statistically significant.

TABLE 1

Composition of LFD and HFD

Table 2

Q- and T-PC primers

Inflammatory gene analysis (mouse)

SEQ ID

Target genes Sequence F Sequence R NO.

For F,R

TCAGCCGATTTGCTATCTCATA AGTACTTGGGCAGATTGACCTC 1,2

TNF-a (Tnfa)

3,4

Interleukin 4 (114) CATGGGAAAACTCCATGCTT TGGACTCATTCATGGTGCAG

AGACAAAGCCAGAGTCCTTCAG TGCCGAGTAGATCTCAAAGTGA 5,6

Interleukin 6 (116)

TTCTCTGTACCATGACACTCTGC CGTGGAATCTTCCGGCTGTAG 7,8

MW-\ (Mipla)

CTCCAAGCCAAAGTCCTTAGAG AGGAGCTGTCATTAGGGACATC 9,10

Arginase 1 (Argl)

GGCTCAAGGACAACAATTTAGG ACTGTGGAAAAACCGTTGAACT 11,12 chitinase 3 -like 3 (Chi3l3)

C-type lectin domain family TCATTGAAAGCCAAACATCG CCTGGGGAGCTGTATTTCTG 13,14 7,member a (Clecla)

Programmed cell death 1 ligand ACGTGGCCACTTCATGTTTT TCTTGAGGGTTTCCCATCAG 15,16

2 (Pdcllg2)

TATGAACAGATGGGCCTCCT AGCTGGGTTCTCCACCTCTT 17,18

Resistin like alpha (Retnla)

Interleukin 1 receptor antagonist TTGTGCCAAGTCTGGAGATG TTCTCAGAGCGGATGAAGGT 19,20

(Illrn)

Interleukin receptor 4 alpha GAAGCCAGGAGTCAACCAAG ATACAGCGCACCACACTGAC 21,22

(lUra)

Example 2

Characterization of NKT cells in mouse adipose tissue

[0074] CD ld-restricted NKT cells can be classified into two main types with the majority being aGalCer-reactive invariant type I and the rest being variant type II NKT cells.

[0075] Flow cytometric analysis provided an estimate of approximately 15,000 type I NKT cells per gram epididymal adipose tissue in a 6w-old lean mouse, about half of CD4 ⁺ T cells but 4 times higher than CD8 ⁺ T cells (white bars in Fig. 1 A). Further analyses of type I NKT cells in adipose tissue revealed that the proportion and the absolute numbers of type I NKT cells in adipose tissue were increased by approximately 20% and 25%, respectively, with 4d HFD challenge (white vs. black bars of NKT cells in Fig. 1 A-B). This was not seen in other adipose-resident T cells such as CD4 ⁺ , CD8 ⁺ T cells (Fig. 1 A- B). Additionally, adipose-resident type I NKT cells were primarily CD4 ^" CD8 ^" with one quarter being CD4 ⁺ CD8 ^" (Fig. 1C). In contrast, type I NKT cells in liver and spleen consisted primarily of CD4 ⁺ CD8 ^" with a small fraction of double negative CD4 ^" CD8 ^" , and in bone marrow the ratio of CD4 ⁺ CD8 ^" to CD4 ^" CD8 ^" type 1 NKT cells was about 1 : 1 (Fig. 1C). Examination of other activation markers CD44, CD69 and CD25 revealed no differences among type I NKT cells in various tissues as they all were CD44 ^hi CD69 ⁺ CD25 ^" . Taken together, these data suggest that adipose tissues of lean and 4d HFD mice contain CD4 ^" CD 8 ^" type I NKT cells.

Example 3

Abundance of adipose-resident type 1 NKT in obese mice

[0076] To study the impact of obesity on NKT cells in adipose tissue (adipose-resident NKT cells), 6w-old B6 mice were placed on a HFD containing 60% calories derived from fat or a 13% LFD as a control cohort (Example 1). HFD feeding progressively increased body and epididymal fat pad weights as well as fasting glucose levels (Fig. 2A-B).

Interestingly, unlike CD8 ⁺ T cells, abundance of type 1 NKT cells in adipose tissue, in terms of the percentage in total lymphocytes and total cell number per g adipose tissue, gradually decreased with HFD (Fig. 2C-D). [0077] This observation was further confirmed in adult murine ob/ob genetic obesity model: the percent of NKT cells in adipose tissues were significantly reduced in ob/ob mice (Fig. 2E-F). The reduction of NKT in adipose tissue of ob/ob mice was not due to defects in NKT cell development or emigration from the thymus as the level of NKT cells in lymphoid organs such as spleen was not affected by obesity (Fig. 2G). Taken together, these data suggest that HFD progressively decreases the abundance of NKT cells in adipose tissue.

Example 4

aGalCer activates adipose-resident NKT

[0078] To address whether NKT cells can be targeted therapeutically, a gain-of-function approach was used to stimulate NKT cells in vivo with aGalCer (Fig. 3A). Other studies have established that significant expansion of NKT cells was observed from 2-3 d up to 7d after the initial aGalCer injection, a process associated with sustained cytokine production up to 7d (Wilson et al. (2003) Proc Natl Acad Sci USA 100, 10913-10918). To further examine therapeutic effects of NKT cell activation at different stages of obesity, studies were performed in animals that had been on HFD for 4d, 8w and 24w, which represent short-term, long-term and chronic HFD feeding models, respectively.

[0079] Accordingly, 6w-old male mice were fed with either 13% LFD or 60% HFD for indicated period of time from 4d to 24w as described in Example 1. aGalCer (Toronto Research Chemicals) was dissolved in pyridine at 2.5 mg/ml and then diluted 1 :250 in PBS to prepare 10 μg/ml working solution prior to use. In all experiments, mice were injected i.p. with 200 μΐ aGalCer (100 ng aGalCer per g body weight). The vehicle control group was injected with PBS with 0.4%> pyridine. The dosage and frequency of aGalCer injection were the same for all the gain-of-function experiments unless otherwise indicated. HFD mice were injected with aGalCer or vehicle at day 0 and day 2. GTT was conducted on day 4, followed by sacrificing the mice on (day 5) to harvest adipose tissues (or other tissues as indicated), which were processed for immune cell analysis, frozen for Q-PCR or Western blot, or fixed for H&E staining. Flow cytometric analyses of NKT levels and other immune cells were performed on day 5.

[0080] For GTT, mice were fasted for 16-18 h followed by injection of glucose (Sigma) at 1 g/kg body weight. Blood glucose was monitored using One-Touch Ultra Glucometer. Fasting insulin levels were measured following a 6h fast.

[0081] While aGalCer injection had no discernible effect on body and epididymal fat weights (Fig. 4B), it caused massive proliferation of NKT cells in adipose tissue (Fig. 3B) with over a 4-fold increase in percentage (Fig. 3C) and 70-fold increase in total cell number (Fig. 5A) in 4d HFD mice. It also caused milder increases of macrophages and CD8 T cells (Fig. 5A) in 4dHFD mice. In line with these observations, H&E staining of WAT sections revealed a significant increase in the number of cells surrounding adipocytes following aGalCer injection (Fig. 3D), and abolished by CD Id deficiency (Fig. 3E), suggesting that the expansion of immune cells including CD8 ⁺ T cells and

macrophages are NKT-dependent. Similar observations were made in 8w and 24w HFD mice (Figs. 3F-G, 4C-D and 5B-C).

[0082] Metabolically, aGalCer challenge improved systemic glucose tolerance in mice at all stages of HFD (Figs. 6A-C) but had no effect on fasting insulin levels (Fig. 6D). The aGalCer effect on glucose tolerance of HFD mice is NKT cell-dependent as it had no effect in CDl mice (Fig. 6E). Pointing to the importance of HFD in NKT cell effect, aGalCer injection in mice on LFD had no effect on glucose tolerance (Fig. 6F). Thus, NKT cell activation by aGalCer improves systemic glucose tolerance in obese animals. NKT- deficient CDld' ^~ mice fed 8-week HFD exhibited no changes in glucose tolerance as previously observed by others (Fig. 4A; Kotas et al. (2011); Mantell et al. (2011)).

[0083] Accordingly, aGalCer injection activates adipose-resident NKT cells and improves systemic glucose homeostasis in obesity. Example 5

aGalCer induces specific M2 genes and proteins

[0084] To explore possible mechanism underlying the beneficial effect of NKT cell activation, we assessed the inflammatory status of adipose tissue by examining the status of macrophage polarization in the adipose tissue samples obtained in Example 4. aGalCer challenge caused a marked 50-100-fold induction of a subset of M2 genes in adipose tissue in a NKT-dependent manner, including Argl, ChiSIS and Pdcdllg2, and to a much lesser extent, Ml genes in all three HFD models (Figs. 7A-B and 8A). This observation was further supported by the protein level of a key M2 macrophage marker: Argl protein was significantly induced in response to aGalCer in WT WAT but abolished in CDld' ^~ WAT (Fig. 7C-D). Intriguingly, this effect was limited to adipose tissue and was not observed in the liver (Fig. 8B). Hence, it appears that NKT cell activation by aGalCer promotes M2 polarization in adipose tissue in obesity.

Example 6

Global effect of aGalCer in adipose tissue

[0085] Microarray analyses of WAT were performed as described (39) with four groups of mice (n=3-4 per cohort) as follows: WT + veh, WT + aGalCer, CDld ^"7" + veh and CD ld ^"/_ + aGalCer. All mice were on 4d HFD feeding with two aGalCer injections at day 0 and day 2. For gene-set enrichment analysis (GSEA), ranking was based on the normalized enrichment score, which reflected the degree to which a gene set in a certain pathway was overrepresented at the top (upregulated) or bottom (downregulated) of the ranked gene list and was corrected for gene set size. Data have been deposited into the GEO datasets (GSE36032). The array data of IL-4-treated mouse BMDM for 10 days were obtained from the Gene Expression Omnibus (GEO) database with accession number GSE25088 (Szanto A. et al. (2010)).

[0086] Because aGalCer effects on macrophage polarization were similar among all three DIO models, the microarray analysis was performed in adipose tissues harvested from the 4d HFD model, where a relatively higher signal-to-noise ratio was anticipated compared to that of long-term HFD. Indeed, the expression of 1,556 genes was significantly increased by more than 2-fold in an absolute NKT-dependent manner (Fig. 9A). Hierarchical clustering of microarray data of individual mice depicted as dendogram showed clear separation of WT mice injected with aGalCer from the other groups. GSEA indicated pronounced induction of many pathways related to immune functions, including the "T _H 1/T _H 2 differentiation" and the "IL-4 pathway" (Fig. 10B). In comparison with a pre-existing dataset (40), we found that 40% of genes upregulated by IL-4 in macrophages were also induced by aGalCer in WAT (Fig. 9C). Furthermore, most components of the canonical IL-4 signaling pathway were induced by aGalCer in WAT (Fig. IOC). Pointing to a potential role of IL-4, seven out of the top 10 aGalCer-induced genes in WAT were also highly upregulated in IL-4-treated macrophages (40), and many were classical M2 genes such as ChiSIS (138 fold, aGalCer vs. veh in WT WAT), pdcdllg2 (53 fold) and Argl (32 fold) (Fig. 9B).

[0087] The array data was further supported by Q-PC and Western blot analyses of Argl levels of WAT (Fig. 7A and 9D-E). Further as direct support for the elevated IL-4 signaling in WAT, 114 mRNA levels in WAT were highly induced by aGalCer treatment in an NKT-dependent manner (Fig. 9F). Thus, aGalCer-mediated NKT cell activation increases IL-4 signaling in adipose tissue.

Example 7

aGalCer-mediated NKT cell activation in IL-4 ^' mice

[0088] The physiological importance of IL-4 in aGalCer-mediated NKT cell activation was assessed in IL-4 ^~ ' ^~ mice. For this study, 6-7w-old WT or IL-4 ^"7" mice were placed on either LFD or HFD for 4d with two aGalCer or vehicle injections prior to metabolic phenotyping. While both body and adipose weights were comparable among WT and IL- 4 ^~ ' ^~ cohorts following 4-day HFD (Fig. 11A), loss of IL-4 completely abolished the aGalCer effect in glucose tolerance (Fig. 11B-C) and reduced the total number of infiltrating immune cells in adipose tissue following aGalCer injection (Fig. 12A). Indeed, total lymphocytes in WAT of IL-4 ^~ ' ^~ mice were reduced by approximately 40% relative to WT cohorts (Fig. 12B). [0089] Loss of IL-4 markedly attenuated the aGalCer-mediated induction of M2 genes Argl, CM313, and Pdcdllgl (Fig. 1 ID) and Argl protein in WAT (Fig. 11E-F). Moreover, loss of IL-4 completely abolished tyrosine phosphorylation (p-Y) of STAT6, but not p-Y of STAT3, in WAT following aGalCer-mediated NKT activation (Fig.11E-F). Providing further support to the significance of adipose tissue in mediating NKT effect in vivo, hepatic STAT6 was not activated by aGalCer (Fig. 11G).

[0090] Finally, it has been shown that the IL-4/STAT6 effect may be mediated by the activities of nuclear receptors (PPARy or ΡΡΑΡνβ/δ) in macrophages (Chawla (2010)). However, aGalCer-mediated induction of M2 genes was not affected in WAT of myeloid cell-specific PPARy or ^ ^/^-deficient animals when compared to WT control littermates (Fig. 12C). Thus, NKT cells regulate macrophage polarization and exert metabolic control largely through the IL-4/STAT6 axis in obese adipose tissue, in a PPARy- and PPARp/5-independent manner.

Example 8

NKT in Human Obesity

[0091] General Methods. 39 healthy premenopausal adult Chinese women including 25 lean individuals with BMI < 25 Kg/m and 14 overweight/obese subjects with BMI > 25 Kg/m , undergoing elective abdominal surgery for benign, non-infective gynecological were recruited. Pre-operative assessment was carried out within one week of the operation. Anthropometric parameters (body weight, height, waist circumference and blood pressure) were measured, and body composition was determined by bioelectric impedance analysis (Tanita Body Composition Analyzer TBF-410, Japan). All subjects underwent an oral glucose tolerance test (OGTT) with 75 g glucose. Fasting plasma glucose and 2 h glucose levels at OGTT were measured by hexokinase method on a Hitachi 747 analyzer (Roche). Insulin was measured by microparticle enzyme immunoassay (Abbott Laboratories).

HOMA index was calculated to estimate insulin resistance (IR): HOMA-IR = fasting glucose (mM) x fasting insulin (mIU/L)/22.5 (Matthews et al. (1985) Diabetologia 28, 412-419). During the operation, visceral adipose tissues (about 8 cm each) were collected aseptically, and transported immediately to the laboratory for RNA extraction and Q-PCR analysis as described below.

[0092] Q-PCR Analysis of NKT cells in human adipose tissues. Following RNA extraction using the Qiagen mini -RNA purification kit, 1 ^ig of total RNA from each sample was reverse-transcribed. The relative abundance of each gene was determined by Q-PCR on a Prism 7000 sequence detection system (Applied Biosystems), and was normalized against 18S rRNA. Two primer sets were used to detect NKT cell markers in human adipose tissue: one for detecting the splicing event of TCR alpha chain covering around the V-J-C rearrangement region with forward primer at TRA VIO (Va24, Chr. 14) and reverse primer at TRAC (TCR a chain constant region, Chr. 14)—oligo set 1, fragment size 230 bp (Kis et al. (2007) J Leukoc Biol 81, 654-662). The other primer set detected the use of TRAVIO (Va24) -oligo set 2, fragment size 417 bp (Vijayanand et al. (2007) N Engl J Med 356, 1410-1422). A primer set for a common T cell gene TRBC2 (TCR β chain constant 2, chr. 7) (Vijayanand et al. (2007) N EnglJ Med 356, 1410-1422) was included as a control for all T cells (oligo set 3, fragment size 104 bp). Primer sequences are listed in Table 2.

[0093] Results. The abundance of NKT cells in visceral adipose tissues collected from 25 lean and 14 overweight/obese Asian female subjects were analyzed (Table 3; values are mean ± standard error).

[0094] As most CD ld-restricted type 1 human NKT cells express an invariant TCR Va24 (8,41,42), Q-PCR was used to quantitate the levels of Va24-containing mRNA or Va24-associated Va24-Ca splicing event in adipose tissue. Strikingly, Va24 NKT cells in adipose tissue decreased with BMI (Fig. 13 A). Total T cells (CD4, CD8 and NKT cells) exhibited an opposite trend (Fig. 13B). Bivariant correlation analysis further revealed that the levels of Va24 NKT cells in adipose tissue were inversely correlated with insulin resistance as measured by homeostasis model assessment (HOMA) index (Fig. 13C) and glucose levels at the 2 h point of OGTT (Fig. 13D). Thus, the abundance of NKT cells in adipose tissue negatively correlates with BMI, insulin resistance and fasting glucose in humans.

Table 3

Clinical characteristics of lean and overweight/obese human subjects

a BMI: BMI cutoff from lean and overweight/obese is 25.

b Fasting glucose/insulin: glucose or insulin levels after an overnight 16h fast.

c 2h glucose: glucose levels at the end of OGTT, i.e. the 2h time point.

d HOMA-IR = [fasting glucose (mM) x fasting insulin (mIU/L)]/22.5

Example 9

Model for NKT in Obesity

[0095] Fig. 14A shows that adipose tissue possesses predominantly CD4 ^" CD8 ^" type 1 NKT cells with lipid-sensing capability. When activated, these cells promote TH2 responses and reduce inflammation in SVC of adipose tissue via the IL-4/STAT6 signaling axis. NKT cells, and possibly other cells, appear to contribute to the "IL-4" pool in adipose tissue.

[0096] The actions of NKT cells at different stages of obesity are shown in Fig. 14B. At the onset of obesity, NKT cells are activated and thereby enhance M2 polarization via the IL-4/STAT6 signaling axis. In contrast, at the late stages of obesity, NKT cells appear in a "quiescent" or "anergic" state, which may provide a permissive environment for the development of pro-inflammatory responses with massive infiltration of other immune cells such as B cells, macrophages and CD8+ T cells. Hence, loss of NKT cells has a significant impact on inflammation at the onset of obesity, but less so at the late stages of obesity. Conversely, activation of NKT cells at both onset and late stages of obesity enhances TH2 response and attenuates inflammation, improves glucose homeostasis as well as insulin resistance in adipose tissue.