You are here

Article Text

Article menu
Colon
Original article
Salutary effects of adiponectin on colon cancer: in vivo and in vitro studies in mice
  1. Hyun-Seuk Moon1,
  2. Xiaowen Liu1,
  3. Jutta M Nagel1,
  4. John P Chamberland1,
  5. Kalliope N Diakopoulos1,
  6. Mary T Brinkoetter1,
  7. Maria Hatziapostolou2,3,
  8. Yan Wu4,
  9. Simon C Robson4,
  10. Dimitrios Iliopoulos2,3,
  11. Christos S Mantzoros1,5
  1. 1Division of Endocrinology, Diabetes and Metabolism, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA
  2. 2Department of Cancer Immunology & AIDS, Dana-Farber Cancer Institute, Boston, Massachusetts, USA
  3. 3Department of Microbiology & Immunobiology, Harvard Medical School, Boston, Massachusetts, USA
  4. 4Department of Medicine, Gastroenterology and Transplantation Institute, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA
  5. 5Section of Endocrinology, Boston VA Healthcare System, Harvard Medical School, Boston, Massachusetts, USA
  1. Correspondence to Dr Christos S Mantzoros, Professor of Medicine, Harvard Medical School, JP9B52A, 150 S. Huntington Ave, Boston, MA 02130, USA; cmantzor{at}bidmc.harvard.edu

Abstract

Background Obesity and a high-fat diet are associated with the risk and progression of colon cancer. Low adiponectin levels may play an important role in the development of colon and other obesity-related malignancies. No previous studies have directly investigated the mechanistic effects of adiponectin on colon cancer in the settings of obesity, a high-fat diet and/or adiponectin deficiency.

Objective To investigate the effects of adiponectin on the growth of colorectal cancer in adiponectin-deficient or wild-type-C57BL/6 mice fed a low-fat or high-fat diet.

Results Mice fed a high-fat-diet gained more weight and had larger tumours than mice fed a low-fat-diet. Adiponectin administration suppressed implanted tumour growth, causing larger central necrotic areas. Adiponectin treatment also suppressed angiogenesis assessed by CD31 staining and VEGFb and VEGFd mRNA expression in tumours obtained from mice fed a high-fat-diet and from adiponectin-deficient mice. Adiponectin treatment decreased serum insulin levels in mice on a high-fat-diet and increased serum-interleukin (IL)-12 levels in adiponectin-deficient mice. In vitro, it was found that adiponectin directly controls malignant potential (cell proliferation, adhesion, invasion and colony formation) and regulates metabolic (AMPK/S6), inflammatory (STAT3/VEGF) and cell cycle (p21/p27/p53/cyclins) signalling pathways in both mouse MCA38 and human HT29, HCT116 and LoVo colon cancer cell lines in a LKB1-dependent way.

Conclusion These new mechanistic and pathophysiology studies provide evidence for an important role of adiponectin in colon cancer. The data indicate that adiponectin or analogues might be useful agents in the management or chemoprevention of colon cancer.

  • Adiponectin
  • colon cancer
  • adiponectin-insufficient heterozygous mice
  • high-fat diet
  • metabolic/inflammatory/cell cycle signalling pathways
  • LKB1
  • endocrine hormones
  • colon carcinogenesis
  • pancreatic cancer

https://doi.org/10.1136/gutjnl-2012-302092

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Significance of this study

    What is already known about this subject?

    • It has been proposed that adiponectin has an important role in attenuating the development of obesity-related malignancy.

    • The plasma levels of adiponectin are decreased in most animal models of obesity, and in obese human subjects, particularly those with visceral obesity.

    • In epidemiology studies low adiponectin levels are correlated with an increased risk for several obesity-associated malignancies, including endometrial, breast, prostate and colon cancer.

    What are the new findings?

    • Exogenous administration of recombinant adiponectin suppresses tumour growth of implanted colon cancer cells in mice and the effects are more pronounced in conditions of adiponectin deficiency, such as Western diet-induced obesity and metabolic syndrome.

    • Decreased colon tumour growth in adiponectin-treated mice is associated with improvement of insulin resistance in response. This effect is more pronounced in insulin-resistant obese mice, indicating a permissive role of adiponectin in conditions of relative adiponectin deficiency.

    • Adiponectin directly controls malignant potential (cell proliferation, adhesion, invasion and colony formation) properties and regulates metabolic (AMPK/S6), inflammatory (STAT3/VEGF) and cell cycle (p21/p27/p53/cyclins) signalling pathways of both mouse and human colon cancer cell lines in an LKB1-dependent way.

    How might it impact on clinical practice in the foreseeable future?

    • Our new pathophysiology and mechanistic studies using this mouse model of obesity and metabolic dysfunction, which is closest to that induced by a Western diet in humans, provide evidence for a causal role of adiponectin in colon cancer. Our data indicate that adiponectin might be a useful agent in the management and/or chemoprevention of colon cancer.

    Introduction

    Several epidemiology studies have linked the risk of colorectal cancer with obesity.1–4 Potential links between obesity and subsequent colon cancer risk include metabolic dysfunction, dyslipidaemia and hyperinsulinaemia, in the context of insulin resistance.5 ,6 In addition, elevated levels of various growth factors and cytokines play a role in the low-grade, chronic inflammatory state seen in obesity and in colon cancer.7 ,8 More recently, it has been proposed that abnormal adiponectin secretion, associated with intra-abdominal obesity, has variable roles at different stages of obesity-induced cancer.9–11

    Adiponectin is an adipocyte-derived, insulin-sensitising hormone. The circulating levels of this hormone are decreased owing to obesity,11 ,12 owing to genetic polymorphisms of the adiponectin gene,13 and/or exposure to unhealthy diet and lack of exercise—that is, conditions leading to insulin resistance.5 ,14 ,15 In epidemiology studies low adiponectin levels are correlated with an increased risk for several obesity-associated malignancies, including endometrial, breast, prostate and especially colon cancer,10 both cross-sectionally and prospectively.16 It has been shown that adiponectin suppresses angiogenesis by inducing endothelial cell apoptosis and also inhibits mammary tumourigenesis in nude mice.17 By contrast, no previous studies have examined the mechanisms underlying the effects of adiponectin on colon cancer by using an animal model close to humans with obesity and insulin resistance—that is, diet-induced obese mice.18

    We designed in vitro and in vivo studies to investigate the potential anticancer effect of adiponectin in colon malignancy. More specifically, we randomised wild-type (WT) C57BL/6 and heterozygous (Het) mice to be fed either a high-fat (HF) diet, modelling the Western diet that leads to diet-induced obesity in humans,18 or a low-fat (LF) diet. We then investigated whether a low physiological dose of adiponectin has an anti-tumour role against the implanted colon cancer cell growth in either genotype and/or diet groups. Further, we studied whether the mechanisms underlying a potential adiponectin effect might include altered angiogenesis, insulin resistance or changed immune responsiveness. Finally, we investigated whether LKB1, a tumour suppressor gene,19 is required for adiponectin-regulated metabolic, inflammatory and proliferative properties in mouse and human colon cancer cell lines in vitro.

    Materials and methods

    In vivo and in vitro experiments were performed as described in detail in the online supplementary methods.

    Results

    In vivo animal studies

    An HF diet affects body weight, hormonal and inflammatory profile in an adiponectin genotype-dependent manner

    Mice were randomly assigned to eight groups using a standard 2×2×2 factorial design. The experimental design and timeline are shown in figure 1A. After 5 months on an HF diet, mice reached a significantly higher body weight than mice on an LF diet (diet, p<0.0001 at baseline) (figure 1B and supplementary data 1). Het mice deficient in adiponectin also gained significantly more body weight than WT mice (genotype, p=0.025 at baseline) (figure 1B and supplementary data 1). As expected, adiponectin levels were significantly lower in Het mice than in their WT littermates (62% decrease, p<0.0001 at baseline). Also, WT mice maintained on an HF diet had significantly higher adiponectin levels than WT mice on an LF diet, whereas Het mice had similar adiponectin levels in both HF and LF groups (diet×genotype interaction, p=0.01 at baseline). Leptin, glucose, insulin levels and homoeostatic model assessments of insulin resistance (HOMA-IR) were significantly higher in mice maintained on an HF diet than in the mice on an LF diet after adjusting for body weight (all p<0.01 at baseline). Interferon (IFN)γ, interleukin (IL)-2, IL-4, IL-5, IL-10 and tumour necrosis factor (TNF)α serum cytokine levels were not significantly different among the groups before initiation of adiponectin treatment (data not shown). However, serum IL-1β and total IL-12 were significantly lower in Het mice than in WT mice (35% decrease, p=0.005 and 29% decrease, p=0.0001, respectively, at baseline). Also, mice maintained on an HF diet had significantly higher serum keratinocyte-derived chemokine (KC) levels than mice on an LF diet (180% increase, p=0.002).

    Figure 1
    Figure 1

    Adiponectin treatment inhibits colorectal cancer growth and causes extensive central necrosis. (A) Experiment time line; (B) body weight; (C) final tumour weight at autopsy; (D) haematoxylin and eosin staining of representative tumours and (E) corresponding total necrotic areas per gram tumour weight of wild type (WT) and heterozygous (Het) mice fed either a high fat (HF) or a low fat (LF) diet with adiponectin (APN) or placebo (Pla) treatment in eight groups (HF+WT+APN, HF+WT+Pla, LF+WT+APN, LF+WT+Pla, HF+Het+APN, HF+Het+Pla, LF+Het+APN, LF+Het+Pla) were measured as describe in detail in the ‘Supplementary methods’ section. General linear models were used to analyse the main effects of diets, genotype, adiponectin treatment and their interactions at sacrifice. Individual differences between the treatment groups were identified by one-way analysis of variance followed by the protected least significant differences technique. Means with different letters are significantly different, p<0.05. Values are means ± SE; n=3–4/group.

    Adiponectin treatment does not affect body weight, hormonal profile and inflammatory profile at the time of tumour implantation

    Mean body weight, serum concentration of leptin, insulin, glucose and HOMA-IR were not significantly different between mice treated with adiponectin and mice treated with placebo for 1 week—that is, at the time of tumour implantation (data not shown). Similarly, there were no significant changes in serum concentrations of the inflammation markers IFNγ, IL-4, IL-5, IL-10, KC and TNFα between mice treated with adiponectin and mice treated with placebo for 1 week before tumour implantation (data not shown). Serum concentrations of total IL-12 for Het mice were significantly higher in adiponectin-treated mice (1.17±0.05 ng/ml) than in placebo-treated mice (0.96±0.06 ng/ml), whereas for WT mice, total IL-12 concentrations were not changed in adiponectin-treated mice compared with placebo-treated mice (treatment×genotype interaction, p=0.0005), indicating a permissive role of adiponectin to correct IL-12 deficiency only in adiponectin-deficient mice which had lower levels to start with.

    Somatometric measurements, hormonal profile, inflammation profile and colorectal cancer growth outcomes 21 days after tumour implantation

    Adiponectin treatment has no effect on body weight and visceral fat

    Twenty-one days after tumour implantation, mice maintained on an HF diet still had significantly higher body weight than mice on an LF diet (p=0.004, figure 1B and supplementary data 1). Accordingly, mice maintained on an HF diet had significantly more visceral fat than mice on an LF diet (p<0.0001, supplementary data 1 and 2A). Although the body weight differences between WT and Het mice became non-significant, Het mice had significantly more visceral fat than WT mice (p=0.001).

    Adiponectin treatment inhibits colorectal cancer growth

    An HF diet led to significantly larger tumours in mice compared with a an LF diet (230% increase, p<0.0001, figure 1C and supplementary data 1). The final tumour weight was significantly reduced by ∼23% (p=0.02) in adiponectin-treated mice compared with placebo-treated mice (figure 1C and supplementary data 1). Haematoxylin and eosin staining (figure 1D and supplementary data 1) showed significantly larger central necrotic areas despite smaller tumour sizes in adiponectin-treated mice (31.8±5.0%) than in placebo-treated mice (23.3±5.0%; p=0.04; figure 1E and supplementary data 1).

    Adiponectin treatment reduces adiponectin receptor 1 (AdipoR1) and receptor 2 (AdipoR2) mRNA expression

    Expression of AdipoR1 and AdipoR2 mRNA in tumour tissue was assessed by RT-qPCR analysis. AdipoR1 mRNA was significantly more abundant than AdipoR2 mRNA in all groups (45.6±1.2-fold, p<0.0001). There were no significant differences in AdipoR1 mRNA expression among all groups (data not shown). In contrast, Het mice and mice maintained on an HF diet had significantly higher AdipoR2 mRNA expression levels than WT mice (p=0.04) and mice on an LF diet (p<0.0001), respectively. In addition, AdipoR2, but not AdipoR1, mRNA expression was significantly decreased in adiponectin-treated groups compared with placebo groups (p=0.0003, data not shown).

    Adiponectin treatment alters proliferative pathways—cleaved capsase-3 staining of apoptotic cells showed that adiponectin treatment significantly increased cell apoptosis levels in both WT and Het mice (p<0.0001, figures 2A, 2B and 2C and supplementary data 1). Ki67 staining, a cellular marker for proliferation, showed that there was a significant interaction between treatment, diet and genotype (p=0.007, treatment×diet×genotype interaction, supplementary data 1 and 3). In groups on an HF diet, mice treated with adiponectin showed significantly fewer Ki67-labelled cells (30.4±2.2%) than placebo-treated mice (38.6±2.0%, p=0.004). In groups on an LF diet, only adiponectin-treated Het mice, but not WT mice had significantly fewer Ki67-labelled cells than placebo-treated mice (p=0.0012, treatment×genotype interaction). We also found that the tumours of adiponectin-treated Het mice exhibited increased AMPK and decreased S6 activation in comparison with placebo-treated mice, suggesting that the antiproliferative effect of adiponectin is mediated by the AMPK–S6 axis (supplementary data 1, 4A and 4B).

    Figure 2
    Figure 2

    Adiponectin treatment increases apoptosis in implanted colon cancer cells. (A) Immunohistochemical staining of cleaved caspase-3 in viable tumour areas of representative tumours; (B) the corresponding average number of apoptotic cells per section; (C) the immunohistochemical staining of cleaved caspase-3 in whole sections of representative tumours of wild type (WT) and heterozygous (Het) mice fed either a high fat (HF) or a low fat (LF) diet with adiponectin (APN) or placebo (Pla) treatment in eight groups (HF+WT+APN, HF+WT+Pla, LF+WT+APN, LF+WT+Pla, HF+Het+APN, HF+Het+Pla, LF+Het+APN, LF+Het+Pla) were measured as describe in detail in the ‘Supplementary Methods section’. At least 10 sections of each tumour were analysed. Values are means ± SE; n=4–8 tumour/group. General linear models were used to analyse the main effects of diets, genotype, adiponectin treatment and their interactions. Individual differences between the treatment groups were identified by one-way analysis of variance followed by the protected least significant differences technique. Means with different letters are significantly different, p<0.05.

    Adiponectin treatment reduces the expression of angiogenic proteins

    Angiogenesis in tumour tissue was assessed by anti-CD31 antibody staining in available tumours from each treatment group (figure 3A and B and supplementary data 1). Tumours from mice on an HF diet had extensive central necrosis, and tumour vascularisation was only evaluated in vital tumour tissue. Overall, we found that tumours from mice treated with adiponectin had significantly fewer dense microvessel areas (6.2±0.3%) than mice treated with placebo (7.9±0.3%, p<0.0001). This difference was more pronounced in mice on an LF diet than those on an HF diet (p=0.007, treatment×diet interaction). Interestingly, we also found that tumours from Het mice had significantly more microvessel density (7.7±0.3%) than WT mice (6.5±0.2%, p=0.0003).

    Figure 3
    Figure 3

    Adiponectin treatment decreases angiogenesis in implanted colon cancer cells. (A) Immunohistochemical staining of CD31 of representative tumours; (B) the corresponding average level of microvessel area (% of tumour area; (C) vascular endothelial growth factor (VEGF)b mRNA and (D) VEGFd mRNA expression of wild type (WT) and heterozygous (Het) mice fed either a high fat (HF) or a low fat (LF) diet with adiponectin (APN) or placebo (Pla) treatment in eight groups (HF+WT+APN, HF+WT+Pla, LF+WT+APN, LF+WT+Pla, HF+Het+APN, HF+Het+Pla, LF+Het+APN, LF+Het+Pla) were measured as describe in detail in the ‘Supplementary methods’ section. General linear models were used to analyse the main effects of diets, genotype, adiponectin treatment and their interactions. Individual differences between the treatment groups were identified by one-way analysis of variance followed by the protected least significant differences technique. At least 10 sections of each tumour were analysed. Means with different letters are significantly different, p<0.05. Values are means±SE; n=4–8 tumour/group.

    Angiogenesis was also measured in tumour tissue by VEGFa, VEGFb, VEGFc and VEGFd mRNA expression. Overall, VEGFb and VEGFd mRNA expression was downregulated in adiponectin-treated mice in comparison with placebo-treated mice (p=0.01 and p=0.01, respectively) (figure 3C and D and supplementary data 1). VEGFb and VEGFd mRNA expression was also upregulated by an HF diet compared with an LF diet (p=0.01 and p=0.04, respectively). VEGFa and VEGFc mRNA expression was not significantly affected by diet, genotype and/or adiponectin/placebo treatment.

    Adiponectin treatment improves insulin resistance and increases levels of anti-inflammatory cytokines

    After 21 days of tumour implantation, adiponectin-treated mice showed a significantly greater decrease of insulin levels than placebo-treated mice in mice fed an HF diet, but not in those fed an LF diet (treatment×diet, p=0.04; supplementary data 1 and 5A). Similar trends were also found with respect to HOMA-IR (trt×diet, p=0.03; supplementary data 1 and 5B). Adiponectin treatment did not significantly affect serum concentrations of glucose and leptin (supplementary data 1, 5C and 5D). In addition, tumour weights were positively correlated with total body weight, visceral fat (supplementary data 1 and 2B), serum insulin levels (supplementary data 1, 2C and 2D) and HOMA (supplementary data 1, 2E and 2F), irrespective of diet, genotype and treatment. Serum concentrations of total IL-12 were increased over time in adiponectin-treated Het, but not WT, mice in comparison with placebo-treated mice (p=0.004, supplementary data 1 and 6D). Overall, WT mice had significantly higher total IL-12 levels than Het mice at all time points (supplementary data 1, 6A, 6B and 6C). The other serum inflammation markers, IFNγ, IL-1β, IL-2, IL-4, IL-5, IL-10, KC and TNFα were not affected by adiponectin after 21 days of tumour implantation (data not shown). Interestingly, serum concentrations of IL-1β (p<0.0001), IL-2 (p<0.0001), IL-4 (p=0.06), IL-10 (p=0.003) and KC (p<0.001) were significantly increased in all groups, irrespective of adiponectin treatment, after 21 days of tumour implantation. Mice on an HF diet also exhibited significantly higher levels of KC than mice on an LF diet (time*diet interaction, p<0.0001), but adiponectin had no effect on KC levels.

    In vitro cell culture studies

    Adiponectin decreases cell proliferation, colony formation, adhesion and invasion of colon cancer cell lines

    We observed that cell proliferation was decreased by ∼40 SD% at high physiological/pharmacological doses—that is, 20–50 μg/ml of adiponectin when compared with control in all colon cancer cell lines (supplementary data 7). Treatment with adiponectin in low physiological concentrations ranging from 5 to 10 μg/ml had no effect on cell proliferation in human HT29 and HCT116 colon cancer cell lines. By contrast, 10 μg/ml of adiponectin administration decreased cell proliferation by ∼15 SD% in human LoVo and mouse MCA38 colon cancer cell lines. Next, cell adhesion assays were performed using fibronectin as an adhesion substrate. We observed that adiponectin-treated colon cancer cell lines had significantly decreased adhesion activity in comparison with control (supplementary data 8A). We performed Matrigel invasion assays as an in vitro model system for metastasis, showing that adiponectin treatment effectively inhibits invasion of colon cancer cell lines (supplementary data 8B). We also examined the growth-inhibitory effects of adiponectin on colon cancer cell lines using a colony formation assay and found that adiponectin treatment significantly decreases colony number in comparison with control (supplementary data 8C).

    Antiproliferative effects of adiponectin are mediated by AMPK activation of colon cancer cell lines

    To understand the potential molecular mechanisms underlying the antiproliferative effects of adiponectin, we evaluated the signalling events induced by adiponectin in colon cancer cell lines. Similar to our in vivo adiponectin signalling data, we found that adiponectin increases phosphorylation of AMPK within 30 min in human HT29 (supplementary data 9A), and within 15 min in mouse MCA38, colon cancer cell lines (supplementary data 9B). By contrast, adiponectin-stimulated AMPK phosphorylation was abolished by AMPK siRNA administration in both cell lines (supplementary data 9C).

    Depletion of LKB1 blocks adiponectin-modulated AMK–S6 signalling pathways of colon cancer cell lines

    As shown in supplementary data 10, adiponectin increased phosphorylation of the tumour suppressor LKB1 in human and mouse MCA38 colon cancer cell lines. We also found that adiponectin-stimulated LKB1 activation was abolished by LKB1 siRNA administration (data not shown). Hence, we next sought to determine the biological importance of depleting LKB1 before studying the effects of adiponectin on intracellular signalling pathways. Adiponectin-increased AMPK and -reduced S6 activation were blocked by LKB1 siRNA administration in all colon cancer cell lines (figure 4A and supplementary data 11 and 12), suggesting that LKB1 is required for adiponectin-mediated modulation of the AMPK–S6 axis in human and mouse colon cancer cell lines.

    Figure 4
    Figure 4

    Depletion of LKB1 blocks adiponectin-mediated AMPK-S6 signalling pathways and malignant potential in human and mouse colon cancer cell lines. The cells were cultured as described in detail in the ‘Supplementary methods’ section. (A) The cells were transfected with LKB1 siRNA for 5 h as described in detail in the ‘Supplementary methods’ section and were then stimulated with adiponectin (20 μg/ml) for 30 min. (B, C) The cells were incubated with adiponectin (20μg/ml) for 24 h and cell viability, adhesion, invasion and colony formation were then measured as described in detail in the ‘Supplementary methods’ section. All data were analysed using one-way analysis of variance followed by a post hoc test for multiple comparisons. Values are means (n=3) ± SD. Means with different letters are significantly different, p<0.05.

    Depletion of LKB1 abrogates adiponectin-mediated inhibition of cell proliferation, colony formation, adhesion and invasion of colon cancer cell lines

    Similar to the results shown in supplementary data 3, adiponectin treatment efficiently inhibited cell proliferation of HT29 (figure 4B and supplementary data 13), MCA38 (figure 4C and supplementary data 14), HCT116 (supplementary data 15A) and LoVo (supplementary data 15B) colon cancer cell lines. By contrast, all these effects were blocked by LKB1 siRNA administration. Also, despite minor differences in the modulation of adiponectin-regulated cell proliferation activity, we did not observe major differences in the magnitude of LKB1-mediated adiponectin-reduced cell proliferation activity in response to adiponectin administration in human and mouse colon cancer cell lines. In addition to examining the effect of adiponectin on cell proliferation, we examined whether LKB1 is required for adiponectin-mediated inhibition of cell adhesion and invasion of colon cancer cells. Similar to the results shown in supplementary data 5A and 5B, adiponectin inhibited adhesion and invasion of colon cancer cell lines (figure 4B and C and supplementary data 15A and 15B). However, these effects were reversed by LKB1 siRNA administration. We also found that the adiponectin-mediated reduction in the number and size of colonies in human and mouse colon cancer cell lines is abolished by LKB1 siRNA administration in long-term colony formation assays (figure 4B and C and supplementary data 15A and 15B). These results collectively show that adiponectin-induced LKB1 activation is indeed a crucial component of the signalling machinery used by adiponectin in modulating cell proliferation, colony formation, cell adhesion and/or invasion of colon cancer cell lines. Moreover, this effect is specific since adiponectin-mediated cell proliferation, colony formation, adhesion and invasion were not influenced by siRNA administration in comparison with control (data not shown).

    Adiponectin suppresses STAT3–VEGF signalling of colon cancer cell lines

    High doses of adiponectin inhibited STAT3 phosphorylation in all colon cancer cell lines (figure 5A and supplementary data 16A). Next, we assessed the significance of STAT3 inhibition in colon cancer cell viability. We have found that STAT3 downregulation by two different siRNAs inhibits colon cancer cell viability (figure 5B and supplementary data 16B). In order to further study the role of adiponectin in the STAT3 signalling pathway, we tested VEGF mRNA expression levels, by real-time PCR analysis. Adiponectin treatment inhibited VEGF mRNA expression levels (figure 5C and supplementary data 16C). In addition, suppression of VEGF expression levels by two different siRNAs blocked colon cancer cell viability (figure 5D and supplementary data 16D). Furthermore, inhibition of LKB1 by siRNA blocked the suppressive effects of adiponectin on STAT3 phosphorylation (figure 5E) and VEGF mRNA expression (figure 5F) levels in all colon cancer cell lines. These data suggest that adiponectin-regulated STAT3–VEGF signalling pathways are mediated by LKB1 and also that the STAT3–VEGF pathway is essential for colon cancer growth.

    Figure 5
    Figure 5

    Depletion of LKB1 blocks adiponectin-mediated STAT3-VEGF signalling pathways in human and mouse colon cancer cells lines. The cells were cultured as described in detail in the ‘Supplementary methods’ section. (A) The cells were incubated with adiponectin at the indicated concentration for 24 h and STAT3 phosphorylation was then measured by ELISA, as described in detail in the ‘Supplementary methods’ section. (B) The cells were transfected with two different siRNAs against STAT3 and cell viability was then measured by MTT assay, as described in detail in the ‘Supplementary methods’ section. (C) The cells were incubated with adiponectin at the indicated concentration for 24 h and VEGF mRNA expression levels were then measured by real-time PCR analysis as described in detail in the ‘Supplementary methods’ section. (D) The cells were transfected with two different siRNAs against VEGF and cell viability was then measured by MTT assay as described in detail in the ‘Supplementary methods’ section. (E, F) The cells were transfected with LKB1 siRNA as described in detail in the ‘Supplementary methods’ section and were then incubated with adiponectin (50 μg/ml) for 24 h. All data were analysed using one-way analysis of variance followed by a post hoc test for multiple comparisons. Values are means (n=3) ± SD. Means with different letters are significantly different, p<0.05.

    Depletion of LKB1 abrogates adiponectin-mediated expression of cell cycle regulatory genes of colon cancer cell lines

    Adiponectin treatment caused sharp reductions in cyclin E2 expression in human HT29 (figure 6A and supplementary data 17) and mouse MCA38 (figure 6B and supplementary data 18) colon cancer cell lines. However, during the same time of adiponectin treatment the expression of cyclin D1 was not changed in both colon cancer cell lines. Also, we observed that adiponectin increases expression of p53, p21 and p27 in human HT29 and mouse MCA38 colon cancer cell lines. Importantly, the effects of adiponectin on expression of tumour suppressor and cell cycle regulatory genes were largely abolished after administration of LKB1 siRNA (figure 6A and B, supplementary data 17 and 18). Also, despite minor differences in the magnitude of signalling activation in the different colon cell lines studied, similar results were also seen in HCT116 (supplementary data 19A) and LoVo (supplementary data 19B) colon cancer cell lines.

    Figure 6
    Figure 6

    Depletion of LKB1 abrogates adiponectin-mediated expression of tumour suppressor and cell cycle regulatory genes in human and mouse colon cancer cell lines. The cells were cultured as described in detail in the ‘Supplementary methods’ section. All cells were transfected with LKB1 siRNA as described in detail in the ‘Supplementary methods’ section. (A, B) The cells were stimulated with adiponectin (20 μg/ml) for 24 h and western blotting was then performed as described in detail in the ‘Supplementary methods’ section. All data were analysed using one-way analysis of variance followed by a post hoc test for multiple comparisons. Values are means (n=3) ± SD. Means with different letters are significantly different, p<0.05.

    Depletion of LKB1 abrogates adiponectin-induced apoptosis of colon cancer cell lines

    To check whether adiponectin might have a proapoptotic effect on human and mouse colon cancer cell lines, we performed TUNEL assays and showed that the population of dead cells was increased by adiponectin administration in comparison with control (supplementary data 20). By contrast, proapoptotic effects of adiponectin on human and mouse colon cancer cell lines were abolished by LKB1 siRNA administration.

    Discussion

    We have examined how adiponectin suppresses colon cancer tumour growth by influencing obesity, insulin resistance, immune responses and the regulation of angiogenesis and intracellular signalling pathways. We used a genetically adiponectin-deficient animal model (Het) fed either an HF or an LF diet and treated with either adiponectin or placebo to investigate the underlying mechanisms of how genetic predisposition to low adiponectin and/or exposure to an HF diet that induces obesity11 ,16 may promote colon cancer cell growth.

    We found that exogenous administration of a physiological dose of adiponectin limits transplanted colon cancer cell growth and promotes central necrosis in vivo. Adiponectin treatment directly suppressed colon cancer tumour growth in HF- and LF-fed Het mice in comparison with placebo-treated mice while adiponectin treatment had a lesser effect on growth of LF-fed WT mice which do not exhibit hypo-adiponectinaemia. We also found that adiponectin-treated mice have significantly larger central necrotic areas despite smaller tumour sizes than placebo-treated mice. These data were consistent with the hypothesis that for adiponectin-deficient subjects an adiponectin supplement might have a more pronounced effect, and suggest a permissive role for adiponectin.

    We then focused on potential mechanisms underlying the effect of adiponectin and found that adiponectin increased the number of apoptotic cells (cleaved caspase-3), reduced the expression of angiogenic factors (CD31, VEGFb and VEGFd), improved insulin resistance in HF-fed mice significantly and increased antiangiogenesis cytokines (IL-12) in Het mice fed either an HF or an LF diet. All the above metabolic and non-metabolic effects of adiponectin might have contributed to its anti-tumour effects (supplementary data 21).

    One of the main mechanisms thought to influence the increased risk of colon cancer in the obese population is hyperinsulinaemia and increased levels of insulin-like growth factors in conjunction with insulin resistance.6 Three weeks after tumour implantation and continuing adiponectin treatment, the tumour weight was positively correlated with the body weight, visceral fat, serum insulin levels and HOMA-IR in both WT and Het mice, indicating that the higher the degree of obesity and insulin resistance the larger the tumour size. Improvement of insulin resistance in response to adiponectin supplementation in low physiological doses was only found in HF-fed mice, a mouse model of diet-induced obesity and insulin resistance reminiscent of human obesity.20 Our results suggest that the decreased colon tumour growth in adiponectin-treated mice is associated with improvement of insulin resistance and that such responses are more pronounced in insulin-resistant obese mice, indicating a permissive role of adiponectin in states of relative adiponectin deficiency.

    IL-12 is a 70 kDa heterodimer protein, composed of p35 and p40 subunits and considered to be a potent immunostimulatory cytokine. It has many anti-tumour effects, including antiproliferation,21–24 antiangiogenesis,24 ,25 activation of NK cells,26 maturation of activated CD4 T lymphocytes to Th1 cells27 and activation of CD8 cytotoxic T lymphocytes.28 We found that adiponectin treatment significantly increased serum concentrations of IL-12 in Het mice but not WT mice, suggesting that the indirect anti-tumour effect of adiponectin can work through regulation of immune responses in a permissive role—that is, in the adiponectin-deficient state accompanied by lower IL-12, whereas the contribution of adiponectin may not be as important as in adiponectin-repletion state with normal IL-12.

    We then confirmed the direct effect of adiponectin on decreasing proliferation by Ki67 antibody staining, increasing central necrotic areas and significantly suppressing microvessel growth in tumours as shown by anti-CD31 antibody staining. This endothelial cell marker functions as an adhesion receptor molecule, playing a key role in leucocyte trafficking across the endothelial layer.29 ,30 Interestingly, we found that tumours from adiponectin-insufficient Het mice had significantly more CD31-stained microvessel area than WT mice. Furthermore, we confirmed in our study that VEGFd mRNA expression, another marker of angiogenesis, was downregulated by adiponectin treatment in comparison with placebo treatment, and upregulated by an HF diet in comparison with an LF diet. VEGFd is upregulated in colorectal cancer,31 and its expression correlates with lymph node metastasis in colorectal carcinomas,32 representing thus a potential anticancer and antimetastasis target.33 These results suggest that adiponectin has both an anticancer and an antimetastasis effect through regulating the expression of angiogenic proteins.

    A number of immunosuppressive factors produced by tumour cells in a STAT3-dependent manner are angiogenic factors, including VEGF,34–36 and a role of constitutively activated STAT3 in tumour cells in promoting tumour angiogenesis and metastasis has been well documented.37 In agreement with these previous studies, we found that adiponectin treatment significantly downregulates mRNA levels of VEGF in vivo and in vitro. In addition, we found that suppression of VEGF expression levels by siRNA administration blocks colon cancer cell viability in vitro. Furthermore, inhibition of LKB1 by siRNA blocked the suppressive effects of adiponectin on STAT3 phosphorylation and VEGF mRNA expression levels in all colon cancer cell lines, suggesting that adiponectin-regulated STAT3–VEGF signalling pathways are mediated by LKB1 and also that the STAT3–VEGF pathway is essential for colon cancer growth.

    Activation of AMPK causes cell cycle arrest in certain cancer cell lines such as hepatoma HepG2, prostate carcinoma PC-3 and breast cancer MCF-7 by upregulating tumour suppressor genes such as p53, p21 and p27.38–40 Also, it has been shown that adiponectin mediates expression of cell cycle regulatory proteins and activates AMPK signalling in colon cancer cell lines.41 Similar to these reports, we observed that the suppressive effect of adiponectin on colon cancer cell proliferation was mediated by activation of AMPK and/or decreasing phosphorylation of S6. We also found that adiponectin increases expression of the cell cycle regulatory gene, cyclin E2, which helps to drive the progression from the G1 to the S phase,42 in human and mouse colon cancer cell lines. Moreover, in a manner comparable to a previous study on adiponectin signalling in human endometrial cancer cell lines,43 we found that adiponectin increases activation of the tumour suppressor gene LKB1 in human and mouse colon cancer cell lines. Hence, it is possible to speculate that the increased expression of the tumour suppressor gene LKB1 promoted by adiponectin would further contribute to the control of proliferation and regulate cell cycle transition from the G1 to the S phase in colon cancer cells.

    Tumour suppressor effects of LKB1 are due to the ability to activate the master metabolic regulator AMPK.44 The genetic depletion of LKB1 in mouse embryonic fibroblasts results in a loss of AMPK activation following energy stresses that raise cellular AMP.45 We found that adiponectin-mediated AMPK activation, as shown by inhibition of S6 phosphorylation and expression of cell cycle regulators, is abolished by LKB1 siRNA administration. Also, we found for the first time that gene knockdown of LKB1 impedes the adiponectin-induced effects on cellular activities, including cell proliferation, colony formation, adhesion and invasion in human and mouse colon cancer cell lines. Our in vitro studies indicate that LKB1 is important in the adiponectin-mediated AMPK–S6 axis and that the increased activation of LKB1 in response to adiponectin treatment inhibits cell proliferation, colony formation, adhesion and invasion properties by regulating tumour suppressor and cell cycle regulatory genes in colon cancer cell lines. Future studies are needed to prove the role of LKB1 in vivo.

    In summary, we found for the first time that exogenous administration of recombinant adiponectin suppresses tumour growth of implanted colon cancer cells in mice. This effect is more pronounced in states of adiponectin deficiency, such as Western diet-induced obesity and metabolic dysfunction. Also, we note that adiponectin directly controls the malignant potential of cells (proliferation, adhesion, invasion and colony formation) and regulates metabolic (AMPK–S6), inflammatory (IL12, STAT3–VEGF) and cell cycle (p21/p27/p53/cyclins) signalling pathways of both mouse and human colon cancer cell lines in an LKB1-dependent way.

    These new mechanistic studies, using the mouse model of obesity and metabolic dysfunction that is closest to the obesity and metabolic syndrome induced by Western diet in humans, provide evidence for a causal role of adiponectin in colon cancer. We suggest that adiponectin might prove to be a useful agent in the management or chemoprevention of colon cancer.

    Acknowledgments

    The authors thank Dr Young-Bum Kim, Division of Endocrinology, Diabetes and Metabolism, Beth Israel Deaconess Medical Center, Harvard Medical School, for technical assistance in the signalling study.

    References

    1. ↵
      1. Jemal A,
      2. Siegel R,
      3. Ward E,
      4. et al
      . Cancer statistics. CA Cancer J Clin 2008;58:71–96.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Calle EE,
      2. Kaaks R
      . Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer 2004;4:579–91.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Gunter MJ,
      2. Leitzmann MF
      . Obesity and colorectal cancer: epidemiology, mechanisms and candidate genes. J Nutr Biochem 2006;17:145–56.
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Chao A,
      2. Thun MJ,
      3. Connell CJ,
      4. et al
      . Meat consumption and risk of colorectal cancer. JAMA 2005;293:172–82.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Ealey KN,
      2. Archer MC
      . Elevated circulating adiponectin and elevated insulin sensitivity in adiponectin transgenic mice are not associated with reduced susceptibility to colon carcinogenesis. Int J Cancer 2009;124:2226–30.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. LeRoith D,
      2. Novosyadlyy R,
      3. Gallagher EJ,
      4. et al
      . Obesity and type 2 diabetes are associated with an increased risk of developing cancer and a worse prognosis; epidemiological and mechanistic evidence. Exp Clin Endocrinol Diabetes 2008;116:(Suppl 1):S4–6.
      OpenUrlCrossRefPubMed
    7. ↵
      1. Lukas M
      . Inflammatory bowel disease as a risk factor for colorectal cancer. Dig Dis 2010;28:619–24.
      OpenUrlCrossRefPubMed
    8. ↵
      1. Fenton JI,
      2. Birmingham JM
      . Adipokine regulation of colon cancer: adiponectin attenuates interleukin-6-induced colon carcinoma cell proliferation via STAT-3. Mol Carcinog 2010;49:700–9.
      OpenUrlPubMed
    9. ↵
      1. Birmingham JM,
      2. Busik JV,
      3. Hansen-Smith FM,
      4. et al
      . Novel mechanism for obesity-induced colon cancer progression. Carcinogenesis 2009;30:690–7.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      1. Kelesidis I,
      2. Kelesidis T,
      3. Mantzoros CS
      . Adiponectin and cancer: a systematic review. Br J Cancer 2006;94:1221–5.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Ziemke F,
      2. Mantzoros CS
      . Adiponectin in insulin resistance: lessons from translational research. Am J Clin Nutr 2010;91:258S–61S.
      OpenUrlAbstract/FREE Full Text
    12. ↵
      1. Mantzoros CS,
      2. Li T,
      3. Manson JE,
      4. et al
      . Circulating adiponectin levels are associated with better glycemic control, more favorable lipid profile and reduced inflammation in women with type 2 diabetes. J Clin Endocrinol Metab 2005;90:4542–8.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Kaklamani VG,
      2. Wisinski KB,
      3. Sadim M,
      4. et al
      . Variants of the adiponectin (ADIPOQ) and adiponectin receptor 1 (ADIPOR1) genes and colorectal cancer risk. JAMA 2008;300:1523–31.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Hu E,
      2. Liang P,
      3. Spiegelman BM
      . AdipoQ is a novel adipose-specific gene dysregulated in obesity. J Biol Chem 1996;271:10697–703.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Arita Y,
      2. Kihara S,
      3. Ouchi N,
      4. et al
      . Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 1999;257:79–83.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Wei EK,
      2. Giovannucci E,
      3. Fuchs CS,
      4. et al
      . Low plasma adiponectin levels and risk of colorectal cancer in men: a prospective study. J Natl Cancer Inst 2005;97:1688–94.
      OpenUrlAbstract/FREE Full Text
    17. ↵
      1. Wang Y,
      2. Lam JB,
      3. Lam KS,
      4. et al
      . Adiponectin modulates the glycogen synthase kinase-3beta/beta-catenin signaling pathway and attenuates mammary tumorigenesis of MDA-MB-231 cells in nude mice. Cancer Res 2006;66:11462–70.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Pierroz DD,
      2. Ziotopoulou M,
      3. Ungsunan L,
      4. et al
      . Effects of acute and chronic administration of the melanocortin agonist MTII in mice with diet-induced obesity. Diabetes 2002;51:1337–45.
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. Taliaferro-Smith L,
      2. Nagalingam A,
      3. Zhong D,
      4. et al
      . LKB1 is required for adiponectin-mediated modulation of AMPK-S6K axis and inhibition of migration and invasion of breast cancer cells. Oncogene 2009;28:2621–33.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Hecker PA,
      2. O'Shea KM,
      3. Galvao TF,
      4. et al
      . Role of adiponectin in the development of high fat diet-induced metabolic abnormalities in mice. Horm Metab Res 2011;43:100–5.
      OpenUrlCrossRefPubMed
    21. ↵
      1. Sugiyama M,
      2. Takahashi H,
      3. Hosono K,
      4. et al
      . Adiponectin inhibits colorectal cancer cell growth through the AMPK/mTOR pathway. Int J Oncol 2009;34:339–44.
      OpenUrlPubMed
    22. ↵
      1. Yamamoto K,
      2. Kiyohara T,
      3. Murayama Y,
      4. et al
      . Production of adiponectin, an anti-inflammatory protein, in mesenteric adipose tissue in Crohn's disease. Gut 2005;54:789–96.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Arsenescu V,
      2. Narasimhan ML,
      3. Halide T,
      4. et al
      . Adiponectin and Plant-derived Mammalian adiponectin Homolog Exert a Protective effect in murine Colitis. Dig Dis Sci 2011;56:2818–32.
      OpenUrlCrossRefPubMed
    24. ↵
      1. Wigginton JM,
      2. Gruys E,
      3. Geiselhart L,
      4. et al
      . IFN-gamma and Fas/FasL are required for the antitumor and antiangiogenic effects of IL-12/pulse IL-2 therapy. J Clin Invest 2001;108:51–62.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Lee JC,
      2. Kim DC,
      3. Gee MS,
      4. et al
      . Interleukin-12 inhibits angiogenesis and growth of transplanted but not in situ mouse mammary tumor virus-induced mammary carcinomas. Cancer Res 2002;62:747–55.
      OpenUrlAbstract/FREE Full Text
    26. ↵
      1. Kodama T,
      2. Takeda K,
      3. Shimozato O,
      4. et al
      . Perforin-dependent NK cell cytotoxicity is sufficient for anti-metastatic effect of IL-12. Eur J Immunol 1999;29:1390–6.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Zilocchi C,
      2. Stoppacciaro A,
      3. Chiodoni C,
      4. et al
      . Interferon gamma-independent rejection of interleukin 12-transduced carcinoma cells requires CD4+ T cells and Granulocyte/Macrophage colony-stimulating factor. J Exp Med 1998;188:133–43.
      OpenUrlAbstract/FREE Full Text
    28. ↵
      1. Brunda MJ,
      2. Luistro L,
      3. Warrier RR,
      4. et al
      . Antitumor and antimetastatic activity of interleukin 12 against murine tumors. J Exp Med 1993;178:1223–30.
      OpenUrlAbstract/FREE Full Text
    29. ↵
      1. Wu Y,
      2. Madri J
      . Insights into monocyte-driven osteoclastogenesis and its link with hematopoiesis: regulatory roles of PECAM-1 (CD31) and SHP-1. Crit Rev Immunol 2010;30:423–33.
      OpenUrlPubMed
    30. ↵
      1. Shi C,
      2. Velazquez P,
      3. Hohl TM,
      4. et al
      . Monocyte trafficking to hepatic sites of bacterial infection is chemokine independent and directed by focal intercellular adhesion molecule-1 expression. J Immunol 2010;184:6266–74.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. White JD,
      2. Hewett PW,
      3. Kosuge D,
      4. et al
      . Vascular endothelial growth factor-D expression is an independent prognostic marker for survival in colorectal carcinoma. Cancer Res 2002;62:1669–75.
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Funaki H,
      2. Nishimura G,
      3. Harada S,
      4. et al
      . Expression of vascular endothelial growth factor D is associated with lymph node metastasis in human colorectal carcinoma. Oncology 2003;64:416–22.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Roskoski R Jr
      . Vascular endothelial growth factor (VEGF) signaling in tumor progression. Crit Rev Oncol Hematol 2007;62:179–213.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Yu H,
      2. Kortylewski M,
      3. Pardoll D
      . Crosstalk between cancer and immune cells: role of STAT3 in the tumour microenvironment. Nat Rev Immunol 2007;7:41–51.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Sumimoto H,
      2. Imabayashi F,
      3. Iwata T,
      4. et al
      . The BRAF-MAPK signaling pathway is essential for cancer-immune evasion in human melanoma cells. J Exp Med 2006;203:1651–6.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Melani C,
      2. Chiodoni C,
      3. Forni G,
      4. et al
      . Myeloid cell expansion elicited by the progression of spontaneous mammary carcinomas in c-erbB-2 transgenic BALB/c mice suppresses immune reactivity. Blood 2003;102:2138–45.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Haura EB,
      2. Turkson J,
      3. Jove R
      . Mechanisms of disease: Insights into the emerging role of signal transducers and activators of transcription in cancer. Nat Clin Pract Oncol 2005;2:315–24.
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      1. Rattan R,
      2. Giri S,
      3. Singh AK,
      4. et al
      . 5-Aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside inhibits cancer cell proliferation in vitro and in vivo via AMP-activated protein kinase. J Biol Chem 2005;280:39582–93.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Xiang X,
      2. Saha AK,
      3. Wen R,
      4. et al
      . AMP-activated protein kinase activators can inhibit the growth of prostate cancer cells by multiple mechanisms. Biochem Biophys Res Commun 2004;321:161–7.
      OpenUrlCrossRefPubMedWeb of Science
    40. ↵
      1. Imamura K,
      2. Ogura T,
      3. Kishimoto A,
      4. et al
      . Cell cycle regulation via p53 phosphorylation by a 5'-AMP activated protein kinase activator, 5-aminoimidazole- 4-carboxamide-1-beta-D-ribofuranoside, in a human hepatocellular carcinoma cell line. Biochem Biophys Res Commun 2011;287:562–7.
      OpenUrl
    41. ↵
      1. Kim AY,
      2. Lee YS,
      3. Kim KH,
      4. et al
      . Adiponectin represses colon cancer cell proliferation via AdipoR1- and -R2-mediated AMPK activation. Mol Endocrinol 2010;24:1441–52.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Oshita T,
      2. Shigemasa K,
      3. Nagai N,
      4. et al
      . p27, cyclin E and CDK2 expression in normal and cancerous endometrium. Int J Oncol 2002;21:737–43.
      OpenUrlPubMedWeb of Science
    43. ↵
      1. Moon HS,
      2. Chamberland JP,
      3. Aronis K,
      4. et al
      . Direct role of adiponectin and adiponectin receptors in endometrial cancer: in vitro and ex vivo studies in humans. Mol Cancer Ther 2011;10:2234–43.
      OpenUrlAbstract/FREE Full Text
    44. ↵
      1. Hardie DG
      . New roles for the LKB1-->AMPK pathway. Curr Opin Cell Biol 2005;17:167–73.
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      1. Hardie DG
      . The AMP-activated protein kinase pathway–new players upstream and downstream. J Cell Sci 2004;117:5479–87.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • H-SM, XL, JMN and JPC contributed equally to this work.

    • Funding This study was supported by a joint California Walnut Commission and AICR grant. The Mantzoros Laboratory is supported by DK79929-02, DK058785-08 and DK081913-02 grants from the National Institute of Diabetes and Digestive and Kidney Diseases. The Robson laboratory acknowledges support from NIH via HL076540 and HL094400. The Mantzoros Laboratory is also supported by a discretionary grant from Beth Israel Deaconess Medical Center and a VA Merit award (grant 10684957).

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    Linked Articles