• Open Access

Leptin receptor is involved in STAT3 activation in human colorectal adenoma

Authors


To whom correspondence should be addressed. E-mail: nakajima-tky@umin.ac.jp

Abstract

The possible role of leptin in colorectal tumors has been investigated in previous studies; however, to date, the conclusions remain under debate. Therefore, we investigated the serum leptin levels in colorectal adenoma patients. In addition, expression of the leptin receptor, and the leptin receptor-mediated signaling pathways were investigated in biopsy specimens collected from human patients with colorectal adenoma. No significant difference in the mean serum leptin level was observed between the colorectal adenoma patients and the control subjects; however, increased expression and activation of the leptin receptor, as indicated by findings such as the phosphorylation of Tyr 1141, was observed in the colorectal adenoma tissues. In addition, activation of the JAK/STAT signaling pathway mediated by the leptin receptor and increased transcriptional regulation of downstream target molecules were observed in colorectal adenomas compared with the non-adenoma tissues. These results indicate STAT3-mediated leptin receptor signaling pathways may be activated in human colorectal adenomas. (Cancer Sci 2011; 102: 367–372)

Colorectal cancer is a major cause of mortality and morbidity worldwide;(1) however, the mechanism of colorectal carcinogenesis remains unclear. Recently, the existence of an association between obesity or metabolic abnormalities and an elevated risk of colorectal cancer was reported.(2,3) Adipose tissue was reported to be not only an energy storage organ, but also an active endocrine organ that secretes important adipocytokines such as adiponectin, leptin, tumor necrosis factor-α (TNF-α), free fatty acid and resistin.(4,5)

Leptin is a 167-amino acid peptide that plays a central role in the hypothalamus in relation to mammalian feeding behavior and energy expenditure.(6) Plasma leptin levels have been reported to be strongly correlated with the body mass index (BMI) in humans(7–9) and also to be elevated in obese subjects. Leptin exerts its activity thorough its specific membrane receptor, the leptin receptor (ObR), belonging to the class 1 cytokine receptor family.(5) Two isoforms, the long and short variants of ObR, namely, ObRL and ObRS, have been identified, and only the long isoform of ObR has been shown to have full signaling potential, with the short isoform showing diminished or abolished capacity for signaling.(5,10)

Several studies have reported the association between serum leptin levels and the presence of several cancers such as prostate(11–13) and breast(14,15) cancer. Similarly, previous studies have also shown an association between serum leptin levels and the presence of colorectal cancer.(16–23) However, the results of these previous studies are contradictory and difficult to interpret. While some studies have shown a decrease in the serum leptin levels in colorectal cancer patients,(16–20) others have reported elevated serum leptin levels in colorectal cancer patients.(21–23) Thus, the association between leptin and the presence of colorectal cancer has not yet been clarified. In addition, previous studies have shown that leptin stimulated cell proliferation in several types of carcinoma cell lines in vitro.(24–26) However, the molecular mechanisms underlying the promotion of human colorectal carcinogenesis by leptin remain unclear.

In the present study we investigated the association between plasma leptin levels/leptin receptor-mediated signaling and the development of colorectal adenoma.

Materials and Methods

Study population.  One hundred and forty-four patients who underwent endoscopic mucosal resection for colorectal adenoma between June 2006 and April 2009 at Yokohama City University Hospital, and 64 control subjects who were detected to have no colorectal polyps on colonoscopy were recruited for this study. The exclusion criteria were subjects with colorectal carcinoma, familial adenomatous polyposis, inflammatory bowel disease, radiation colitis or any malignant disease, and also subjects with a previous history of colectomy, gastrectomy or colorectal polypectomy. Written informed consent was obtained from all subjects prior to their participation in the study. The study protocol was approved by the Yokohama City University Hospital Ethics Committee.

Collection and analysis of blood samples for determination of the leptin levels.  Blood samples were obtained in the morning on the day of colonoscopy after the subjects had fasted overnight. Serum leptin levels were measured by enzyme-linked immunosorbent assay of human leptin (SRL Co., Tokyo, Japan).

Immunohistochemical analyses.  The expressions of ObR and phospho-STAT3 (p-STAT3) were investigated in the colorectal adenoma and normal colorectal tissues. A total of 61 adenoma tissue samples were obtained endoscopically from the study subjects. Formalin-fixed and paraffin-embedded samples were deparaffinized and rehydrated. The sections were incubated with antibodies for ObR (1:50; Santa Cruz Biotechnology, Santa Cruz, CA, USA) and p-STAT3 (Tyr 705) (1:50; Cell Signaling Technology, Danvers, MA, USA) as the primary antibodies, using an LSAB2 kit (Dako Cytomation, Carpinteria, CA, USA). They were then incubated with biotinylated immunoglobulin as the secondary antibody and treated with peroxidase-conjugated streptavidin. The antibody complex was visualized with 3,3′-diaminobenzidine, tetrahydrochloride (Dojindo Laboratories, Kumamoto, Japan). The expressions of ObR and p-STAT3 were analyzed by light microscopy in 10 different fields of each section, and the mean percentage of adenoma cells that showed positive staining was scored by two pathologists. The ObR and p-STAT3 expressions were classified into two categories depending on the percentage of cells showing positive staining: negative, 0–15% of all the tumor cells showing positive staining; and positive, >15% of all tumor cells showing positive staining, as previously described.(27)

Western blot analysis.  Twenty-five colorectal adenoma patients were randomly selected, and biopsy samples obtained from the colorectal adenomas and normal areas were isolated. The extracted protein was separated using sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) and the separated proteins were transferred to a polyvinylidene difluoride (PVDF) membrane (Amersham, London, UK). The membranes were probed with primary antibodies specific for phospho-ObR (p-ObR) (Tyr 1141), p-ObR (Tyr 985), ObR (Santa Cruz Biotechnology), phospho-JAK2 (p-JAK2), JAK2, p-STAT3 (Tyr 705) STAT3 (Cell Signaling Technology) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Trevigen, Gaithersburg, MD, USA). Horseradish-peroxidase-conjugated secondary antibodies and the electrochemi-luminescence (ECL) detection kit (Amersham) were used for the detection of specific proteins.

Real-time RT-PCR.  Twelve colorectal adenoma patients were randomly selected and biopsy samples of the adenoma and adjacent normal tissues obtained from the colorectal adenoma and normal areas were isolated. Total RNA from the colorectal adenoma and normal tissue biopsy specimens was extracted using the RNeasy mini kit (Qiagen, Hilden, Germany). For the real-time reverse-transcriptase polymerase chain reaction, total RNA was reverse-transcribed into cDNA and amplified using the real-time quantitative polymerase chain reaction using the Step One Plus Real Time PCR System (Applied Biosystems, Foster City, CA, USA). Probes and primer pairs specific for ObRL, ObRS, BclX, c-Myc, cyclin D1, cdc2, cyclin B1, VEGF and 18S were purchased from Applied Biosystems. The concentrations of the target genes were determined using the competitive computed tomography method and the values were normalized to the internal control.

Statistical analysis.  Statistical analyses were performed using the Mann–Whitney U-test and chi-square test. All analyses were performed using the Stat View software (SAS Institute, Cary, NC, USA). P < 0.05 was regarded as denoting statistical significance.

Results

Serum leptin levels and colorectal tumors.  The clinical characteristics of the colorectal adenoma patients and control subjects without colorectal polyps are shown in Table 1. No significant difference in the mean serum leptin level was observed between the two groups. There were also no significant differences in age, BMI or other obesity-related factors between the two groups. A good correlation was observed between the BMI and serum leptin levels (R = 0.533, P < 0.01) (Fig. 1a), as previously reported.(7–9) We also investigated the differences in the serum leptin levels depending on the tumor size and pathological grade; however, no correlations were observed (Fig. 1b,c).

Table 1.   Characteristics of the study patients
 NormalAdenomaP value
  1. Data are shown as mean ± standard deviation. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05. **P < 0.01. BMI, body mass index; VFA, visceral fat area; FBS, fasting blood sugar.

N 64144 
Age (years)62.1 ± 13.864.7 ± 10.20.12
Sex (M/F) 33/31100/440.18
Waist circumference (cm)84.4 ± 10.386.5 ± 10.50.28
BMI (kg/m2)22.7 ± 3.523.3 ± 3.20.20
VFA (cm2)75.7 ± 50.893.5 ± 53.50.08
FBS (mg/dL)112.6 ± 26.9108.7 ± 31.10.44
HbA1c (%)5.7 ± 1.25.6 ± 1.00.43
Leptin (ng/mL)5.6 ± 4.35.4 ± 4.20.70
Figure 1.

 Correlation between serum leptin levels and body mass index (BMI), tumor pathology and size. (a) Correlation between serum leptin levels and BMI. Each point represents each individual patient (P < 0.01, R = 0.533). (b) Correlation between serum leptin levels and pathological grade (mild, moderate and severe atypia). Each column represents the mean ± SEM from 23 to 65 patients. (c) Correlation between serum leptin levels and tumor size. Each point represents each individual patient (= 0.664, R = 0.037).

Leptin receptor, ObR, expression in the colorectal adenoma and normal colorectal tissues.  To examine the ObR expression in colorectal adenoma and normal colorectal tissues, immunohistochemical staining and gene expression analyses were performed. ObR was clearly expressed in the cytoplasm of the colorectal adenoma gland cells, but only slightly in the normal colorectal gland cells in the vicinity of the adenomas (Fig. 2a–f). The frequency of detection of ObR in the colorectal adenomas was 67.2% (41/61). For the present study, no isoform-specific antibodies for ObRL and ObRS were available. Therefore, we conducted gene expression analyses specific for ObRL and ObRS. The mRNA expression levels of ObRL and ObRS in the colorectal adenomas and normal colorectal tissues were investigated. The mRNA expression level of ObRL was significantly higher in the colorectal adenomas than in the normal colorectal tissues. In contrast, the expression of ObRS was slightly but not significantly higher in the colorectal adenomas than in the normal colorectal tissues (Fig. 2g,h). Furthermore, western blot analysis was performed to analyze the phosphorylation level of the cytoplastic domain of ObR to investigate the signaling pathway of the leptin receptor. Western blot analysis showed significant increase of ObR expression in the colorectal adenomas than in the normal colorectal tissues (Fig. 3a). Moreover, the Tyr 1141 phosphorylation level of ObR that is required for leptin-induced activation of STAT3(28) was significantly higher in the colorectal adenomas than that in the normal colorectal tissues. In contrast, no difference was observed in the Tyr 985 phosphorylation level of ObR that is required for activation of the extracellular-signal-regulated kinase (ERK) signaling pathway (Fig. 3b,c).(29) These results suggest that the phosphorylation of ObRL in adenomas might activate the JAK/STAT signaling pathway.

Figure 2.

 Immunohistochemical staining for ObR and mRNA level of the leptin receptor. (a) Normal colorectal tissue. (b,c) Colorectal adenoma tissues. (d–f) magnified view of (a–c), respectively. The red arrowhead represents a normal gland and the red arrow points to an adenoma gland. The relative mRNA expressions of (g) ObRL and (h) ObRS in colorectal adenoma and normal colorectal tissues were expressed as the ratio relative to the expression of 18S. Each column represents the mean ± SEM from 12 patients. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05. **P < 0.01. ObR, leptin receptor; ObRL, leptin receptor long variant; ObRS, leptin receptor short variant.

Figure 3.

 Western blot analysis for leptin receptor (ObR) and phosphorylated ObR (Tyr 1141 or Tyr 985). (a) ObR, (b) Tyr 1141-phosphorylated and (c) Tyr 985-phosphorylated ObR. Left panels: representative western blot images for ObR, Tyr 1141-phosphorylated and Tyr 985-phosphorylated ObR. Lanes 1, 2 and 3, normal colorectal tissues; lanes 4, 5 and 6, colorectal adenoma tissues. Right panels: ratios of ObR, Tyr 1141-phosphorylated and Tyr 985-phosphorylated ObR expressions to the expression level of GAPDH are shown. Each column represents the mean with the SEM from 25 patients. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05. **P < 0.01.

Phosphorylated STAT3 in colorectal adenoma.  To investigate the activation of STAT3, immunohistochemical staining and western blot analysis of STAT3 phosphorylation (p-STAT3) status were performed. Expression of p-STAT3 was predominantly observed in the nuclei of the adenoma gland cells, but only faint expression was observed in the normal gland cells (Fig. 4). The percentage of cells showing positive staining for p-STAT3 in the examined tissue specimens of colorectal adenoma was 49.1% (30/61). The expression level of p-STAT3 was significantly higher in ObR-positive adenomas than in ObR-negative adenomas, as shown in Table 2. The western blot analysis showed that the levels of p-JAK2 and p-STAT3 were significantly higher in the adenomas than in the normal colorectal tissues (Fig. 5). In addition, the mRNA levels of the genes encoded by STAT3 were analyzed by real-time RT-PCR. The expression levels of the apoptosis-suppressing protein BclX, the late G1 to G1/S phase proteins cyclin D1 and c-Myc, the G2/M phase proteins cdc2 and cyclin B1(30) and the genes encoding the angiogenesis protein VEGF(31) were significantly higher in the adenomas than in the normal colorectal tissues (Fig. 6). These results suggest that the JAK/STAT signaling pathway is activated in colorectal adenomas. As shown in Table 3, the mRNA expression levels of BclX, c-Myc, cdc2 and cyclinB1 were significantly higher in ObR-positive colorectal adenomas than in ObR-negative colorectal adenomas, as evaluated by immunohistochemistry.

Figure 4.

 Immunohistochemical staining for phosphorylated STAT3 in colorectal tissues. (a) Normal colorectal tissue. (b,c) Colorectal adenoma tissues. (d–f) Magnified view of (a–c), respectively.

Table 2.   Correlation between the expressions of ObR and p-STAT3 in colorectal adenomas by immunohistochemical analysis
 ObR-positive adenomaObR-negative adenomaP value
  1. Data are shown as the percentage and number of phospho-STAT3 (p-STAT3)-positive colorectal adenoma samples in ObR-positive and ObR-negative colorectal adenoma. Statistical analysis was performed using the chi-square test. *P < 0.05. **P < 0.01. ObR, leptin receptor.

p-STAT3-positive adenoma58.8% (24/41)30% (6/20)<0.05*
Figure 5.

 Western blot analysis for phosphorylated JAK2 and STAT3. (a) Phosphorylated JAK2 and (b) phosphorylated STAT3 in normal colorectal and adenoma tissues. Left panels: representative western blot images for phosphorylated and total levels of JAK2 and STAT3. Lanes 1, 2 and 3, normal colorectal tissues; lanes 4, 5 and 6, colorectal adenoma tissues, respectively. Right panels: ratios of the phosphorylated protein levels compared with the total protein level. Each column represents the mean with the SEM from 25 patients. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05. **P < 0.01.

Figure 6.

 The expression of downstream genes encoded by STAT3 transcriptional regulation in normal colorectal and adenoma tissues. The relative mRNA expressions of (a) BclX, (b) cyclinD1, (c) c-Myc, (d) cdc2, (e) cyclin B1 and (f) VEGF in colorectal adenoma and normal colorectal tissues were expressed as the ratios relative to the expression of 18S. Each column represents the mean with the SEM from 12 patients. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05, **P < 0.01.

Table 3.   The expression of downstream genes encoded by STAT3 transcriptional regulation in ObR-positive adenomas and ObR-negative adenomas
 ObR-positive adenoma (n = 7)ObR-negative adenoma (n = 5)P value
  1. Data are shown as mean ± standard deviation. Statistical analysis was performed using the Mann–Whitney U-test. *P < 0.05. **P < 0.01. ObR, leptin receptor; VEGF, vascular endothelial growth factor.

BclX0.34 ± 0.090.24 ± 0.07<0.05*
cdc20.50 ± 0.120.24 ± 0.11<0.01**
Cyclin D10.96 ± 0.170.76 ± 0.320.21
Cyclin B10.58 ± 0.170.28 ± 0.07<0.01**
c-Myc0.57 ± 0.230.21 ± 0.09<0.01**
VEGF0.29 ± 0.060.28 ± 0.080.46

Discussion

Although recent studies have shown an association between serum leptin levels and the presence/absence of colorectal adenoma, the relationship still remains controversial.(16–23) Serum leptin levels have been shown to be strongly correlated with the BMI.(7–9) Therefore, as bodyweight might influence this correlation, the bodyweight differences should be carefully analyzed while interpreting the above correlation. Patients with cachexia were included in the cancer group in several studies.(18,19) We therefore suspect that this might have influenced the results and caused the conflicts in the results of the previous studies. To reduce the differences in the serum leptin levels caused by the effect of cancer on bodyweight, we investigated the serum leptin levels in patients with colorectal adenoma, which is regarded as a precancerous lesion,(32) not associated with bodyweight loss. Also in the present study, we observed a significant correlation between serum leptin levels and the BMI, consistent with previous reports;(7–9) we consider that this reflects the high reliability of our data. We observed no statistically significant differences in the serum leptin levels between patients with colorectal adenoma and normal control subjects in the present study. This result suggests the possibility of colorectal adenoma being associated with leptin receptor expression or activation, but not with serum leptin concentrations, assuming that leptin plays some role in colorectal adenoma growth. To elucidate this hypothesis, we investigated the expression and signal transduction mediated by leptin receptors in colorectal adenomas.

Several earlier studies have demonstrated, by immunohistochemical analysis, the expression of ObR in colorectal cancer and normal colorectal tissues.(33) In addition, recent studies have also confirmed the expression of ObR in colorectal adenomas and cancers.(34) However, none of the previous studies considered the expression of ObR isofoms, namely, ObRL and ObRS, in the colorectal tissues. In the present study, we showed, by immunohistochemical analysis, that ObR was clearly expressed in colorectal adenomas, but only weakly expressed in normal colorectal tissues. In addition to the immunohistochemical data, colorectal adenomas were also found to show significantly higher expression levels of the gene for ObRL, but not for ObRS, than normal colorectal tissues. These results suggest that the expression of ObRL rather than ObRS might be important for the downstream signal transduction in colorectal adenomas. Therefore, we investigated the ObRL-mediated signaling pathways in colorectal adenomas. It is known that phosphorylation of Tyr 1141 of ObRL by leptin activates the JAK/STAT signaling pathway.(28) We demonstrated significantly increased phosphorylation of Tyr 1141, but not Tyr 985, in colorectal adenomas than in the normal colorectal tissues. Taken together, these results suggest that induction of ObRL gene expression in colorectal adenomas might augment phosphorylation of Tyr 1141 of ObRL by leptin, which might result in activation of the JAK/STAT signaling pathway. In fact, we showed enhanced activation of the JAK/STAT signaling pathway and higher gene expressions downstream of the STAT3 signaling pathway in colorectal adenomas than in the normal colorectal tissues. Although we could not show direct evidence of this signaling in human colorectal adenomas, our results provided evidence to suggest that leptin-mediated STAT3 signaling through activation of ObRL in colorectal adenoma might control the expression of genes involved in the cell cycle and apoptosis. Further investigations are required to clarify the growth mechanism of colorectal adenoma.

In conclusion, STAT3-mediated leptin signaling through the activation of ObRL in colorectal adenoma directly controls the expressions of genes involved in the cell cycle and apoptosis, resulting in the growth of adenoma cells.

Acknowledgments

We thank Machiko Hiraga for her technical assistance. This work was supported in part by a Grant-in-Aid for research on the Third-Term Comprehensive Control Research for Cancer from the Ministry of Health, Labour and Welfare, Japan to A. N., a grant from the National Institute of Biomedical Innovation (NBIO) to A. N., a grant from the Ministry of Education, Culture, Sports, Science and Technology, Japan (KIBAN-B) to A. N., and the grant program, “Collaborative Development of Innovative Seed” from the Japan Science and Technology Agency (JST).

Disclosure Statement

The authors declare no conflict of interest.

Ancillary