Implication of metastasis suppressor gene, Kiss-1 and its receptor Kiss-1R in colorectal cancer

Colorectal cancer tissues (n = 94) and normal background tissues (n = 80) were collected immediately after surgery with approval by the South East Wales Local Research Ethics Committee (ref 05/WSE03/92) and patients’ consent. The tissue samples were stored in a deep freezer (−80°C) until further use. Patients were routinely followed after surgery and the median follow up period was 120 months. Tumour tissues and normal tissues (>10 cm away from the tumor margin) were obtained with confirmation by a pathologist.

HT115 and HRT18 cancer cell lines were obtained from the European Collection of Animal Cell Culture (ECACC, Salisbury, England, UK).

ERK inhibitorII (FR180204) (Calbiochem, Germany), Kisspeptin-10 (No.2570) (Tocris Bioscience, Bristol, UK) and Kisspeptin-234 (Tocris Bioscience, Bristol, UK). Kisspeptin-10, a short peptide of 10 amino acids, is proteolytically processed from Kiss-1 [13]. Kisspeptin-234 is a Kisspeptin-10/Kiss-1R antagonist, which belongs to Kisspeptin-10 analog. Antibodies to Kiss-1 (SC-101246) and Kiss-1R (SC-48220) were purchased from Santa Cruz Biotechnologies Inc., (Santa Cruz, CA, USA).

RNA extraction kits, reverse transcription kits and RT-PCR Mix were purchased from Promega (WI, USA) and Bio-Rad (CA, USA). Conventional PCR primers were designed using Beacon Designer (Palo Alto, CA, USA) and synthesized by Invitrogen (Paisley, Scotland, UK). Following the manufacturer’s protocol, total RNA was isolated using TRI reagent (Sigma). The concentration of RNA was detected by a UV spectrophotometer at 260 and 280 nm. cDNA samples were synthesized and the final reaction volume was 20μl. Primers used for RT-PCR are given in Table 1, and GAPDH was used as the house keeping control.

Real-time quantitative PCR, based on the Amplifluor™ technology, was used to quantify the level of mRNA expression of Kiss-1 and Kiss-1R from the cDNA samples of coloretal tissues and cells, mentioned above. All colorectal cDNA samples were synchronously examined for Kiss-1 and Kiss-1R along with a set of internal controls. Q-PCR primers (Table 1) were designed using Beacon Design software (PREMIER Biosoft, Palo Alto, CA). Real-time PCR was carried out using an IcyclerIQ™ (Bio-Rad, Hemel Hempstead, UK) with the following cycling conditions: 94°C for 5 min, 80–90 cycles of: 94°C for 10 sec, 55°C for 35 sec and 72°C for 20 sec.

Frozen sections of colorectal tumours and adjacent background tissues were sectioned at a thickness of 6 μm using a cryostat . The samples were mounted onto Super Frost Plus microscope slides (Fisher, UK). The samples were fixed in a mixture of 50% acetone and 50% methanol and then air-dried. After rehydration and blocking with 5% horse serum solution, the sections were probed with the appropriate primary antibody and secondary antibodies. Following the instructions, the avidin-biotin complex (Vector Laboratories, Burlingame, CA, USA) was applied before staining with diaminobenzidine chromogen. Nuclei were counterstained in Mayer’s haematoxylin. Sections from fresh frozen human placenta were used as positive controls.

Hammerhead ribozymes targeting Kiss-1 and Kiss-1R were designed using Zuker’s mRNA Fold programme [14] based on the secondary structure of Kiss-1 and Kiss-1R mRNA (Figure 1A and B, respectively). The ribozymes were synthesized using touchdown PCR and cloned into the pEF6/V5-His TOPO TA expression plasmid vector (Invitrogen, Paisley, UK) according to the protocol provided. Ribozyme transgenes and empty plasmids were transfected into the two colorectal cell lines HT115 and HRT18 respectively, utilizing an Easyjet Plus electroporator (EquiBio, Kent, United Kingdom). Following selection using blasticidin, verified transfectants which lost the expression of Kiss-1 and Kiss-1R were used in subsequent experiments. During these experiments, cDNA generated from human placenta was used as a positive control.

The ECIS system (Ztheta, Applied Biophysics Inc., USA) was used to quantify cell migration as previously reported by Jiang et al.[15]. HT115 cells were prepared for six repeats per group and seeded at 40,000 cells per well in 200 μl of DMEM medium alone or medium supplemented with 200 nM ERK small inhibitor, 300 nM Kisspeptin-10 and 300 nM Kisspeptin-234.

Cells were counted and 1×10⁶ cells were seeded to a tissue culture flask. After an overnight incubation and 4 hours treatment with appropriate peptide receptor or inhibitor (300 nM Kisspeptin-10, 300 nM Kisspeptin-234 or 200 nM ERK inhibitor), the medium was collected. This method is well established within the laboratory, and has previously been published [16].

The relationship between the expression of Kiss-1 and Kiss-1R and tumour grade, TNM staging and nodal status was respectively analyzed using Mann–Whitney U test (Tables 2 and 3). Quantitative data of IHC, in vitro assays, ECIS assay and zymography assay was analyzed using the Student’s test, Kruskal-Wallis and chi-squared test, where appropriate. Survival analysis and multivariate analysis were carried out using the SPSS20 software package. Differences were considered to be statistically significant at p < 0.05.

Kiss-1 and Kiss-1R expression was analysed in colorectal cancer tissues and adjacent normal tissues using real time PCR (Tables 2 and 3). Decreased levels of Kiss-1 transcript were seen in Dukes B and C tumours compared with Dukes A carcinomas (p < 0.05). The expression level of Kiss-1 decreased as TNM stage progressed, and statistical analysis revealed significant links between TNM stage I and stage III&IV. A significantly decreased expression of Kiss-1R was observed in tumour tissues compared with normal background tissues. Interestingly, Kiss-1R expression in patients undergoing chemo-radio therapy was considerably higher in comparison to that in patients without therapy.

A significant difference was also observed in the expression levels between living patients and those who had died. The average copy numbers of Kiss-1 and Kiss-1R transcripts in Dukes B were then employed as the respective thresholds for the survival analysis. Kaplan-Meier survival analysis demonstrated that patients with a low expression level of Kiss-1 appeared to have similar overall survival and disease free survival compared to patients with high expression of Kiss-1 (p > 0.05). In contrast to Kiss-1, the expression pattern of Kiss-1R revealed that high levels of Kiss-1R transcript are associated with both a poor overall survival (Figure 2C, p = 0.0011) and poor disease free survival (Figure 2D, p = 0.0033). Multivariate analyses have further demonstrated that T-stage and Kiss-1R are independent prognostic factors (p = 0.025 and p = 0.003, respectively) for colorectal related death. Furthermore, TNM staging and Kiss-1R (p = 0.03 and p = 0.012, respectively) are independent prognostic factors for colorectal cancer related incidence (death, recurrence and metastasis) (Table 4).

Immunochemical staining was carried out on a portion of paired normal and tumour tissues (n = 23 pairs). Intense expression of Kiss-1 and Kiss-1R was primarily observed in the cytoplasm of epithelial cells in adjacent normal tissues compared with cancer cells (p value < 0.05), as shown in Figure 2A and B.

The expression of Kiss-1 and Kiss-1R in both HT115 and HRT18 cells transfected with corresponding ribozyme transgenes was examined using conventional PCR, real time PCR and Western blot (Figures 1 and 3A).

Analysis of the in vitro function assays revealed cells with Kiss-1 knockdown had significantly increased invasiveness (p = 0.015) and migration (p = 0.0094) compared HT115 pEF cells (Figure 3B and C), but knockdown of Kiss-1 and Kiss-1R did not demonstrate a significant change on cell growth and adhesion of HT115 and HRT18 cells (p > 0.05).

HT115 Kiss-1 knockdown cells treated with Kisspeptin-10 had a significant decrease in motility compared to the control cells (Figure 4A), but no significant difference was observed between control and Kiss-1R knockdown cells. In contrast, neither Kiss-1 knockdown or Kiss-1R knockdown cells treated with Kisspeptin-234 showed a difference in comparison to the control groups (no treatment) (p > 0.05) (Figure 4B). The adherence of Kiss-1 knockdown cells treated with ERK inhibitor significantly decreased compared with that of the control group (no treatment) after wounding (p = 0.0397) (Figure 4C). A similar effect was also observed in the comparison between the control group (no treatment) and cells treated with ERK inhibitor + Kisspeptin-234 (Figure 4E). In contrast to the groups treated with ERK inhibitor or ERK inhibitor together with Kisspeptin-234, there were more striking links between the Kiss-1 knockdown cells treated with ERK inhibitor + Kisspeptin-10 and its control (no treatment) (p = 0.0164) (Figure 4D).

MMP-9 expression levels were evaluated in three pairs of HT115 cells (pEF and Kiss-1 knockdown), which were treated with three different treatments (no treatment, ERK inhibitor, ERK inhibitor and Kisspeptin-10) (Figure 5). The weakest expression of MMP-9 was observed in the cells treated with ERK inhibitor and Kisspeptin-10 (Figure 5A). We further explored the influence of Kiss-1 knockdown together with treatment of Kisspeptin-10 and 234 on MMP-9 and MMP-2 activity, together with the inhibition of the ERK pathway in the HT115 cell line. Cells were divided into three treatment groups, namely no treatment (control group), Kisspeptin-10, and Kisspeptin-234 respectively. For each group, two pairs of HT115 pEF and HT115 Kiss-1 knockdown cells were treated with or without ERK inhibitor, namely + ERK inhibitor and control respectively (Figure 5B). Figure 5C shows expression of Kiss-1 and treatment of the cells with the inhibitors had a greater impact on the enzyme activity of MMP-9 than on that of MMP-2. Hence, the difference in MMP-9 expression in HT115 cells was chosen to analyze further. Figure 5D shows the difference in the enzyme activity of MMP-9 in HT115 pEF (left) and Kiss-1 knockdown (right) cells of all three groups (no treatment, Kisspeptin-10 and Kisspeptin-234). The enzyme activity of MMP-9 in HT115 pEF cells treated with ERK inhibitor was significantly inhibited compared with that of controls in all three groups. Similarly, with HT115 pEF cells, the enzyme activity of MMP-9 in HT115 Kiss-1 knockdown cells treated with ERK inhibitor brought about a strong inhibition in comparison to the controls in the groups of no treatment (p = 0.00017) and Kisspetin-10 (0.001544) (Figure 5D). Furthermore, the results show that both HT115 cells treated with Kisspeptin-10 exhibited a decreased expression of MMP-9 in comparison to the cells treated with serum free medium and Kisspeptin-234.