Cyclophilin A enhances cell proliferation and tumor growth of liver fluke-associated cholangiocarcinoma

Abstract

Background: Cyclophilin A (CypA) expression is associated with malignant phenotypes in many cancers. However, the role and mechanisms of CypA in liver fluke-associated cholangiocarcinoma (CCA) are not presently known. In this study, we investigated the expression of CypA in CCA tumor tissues and CCA cell lines as well as regulation mechanisms of CypA in tumor growth using CCA cell lines.

Methods: CypA expression was determined by real time RT-PCR, Western blot or immunohistochemistry. CypA silence or overexpression in CCA cells was achieved using gene delivery techniques. Cell proliferation was assessed using MTS assay or Ki-67 staining. The effect of silencing CypA on CCA tumor growth was determined in nude mice. The effect of CypA knockdown on ERK1/2 activation was assessed by Western blot.

Results: CypA was upregulated in 68% of CCA tumor tissues. Silencing CypA significantly suppressed cell proliferation in several CCA cell lines. Likewise, inhibition of CypA peptidyl-prolyl cis-trans isomerase (PPIase) activity using cyclosporin A (CsA) decreased cell proliferation. In contrast, overexpression of CypA resulted in 30% to 35% increases in proliferation of CCA cell lines. Interestingly, neither silence nor overexpression of CypA affected cell proliferation of a non-tumor human cholangiocyte cell line, MMNK1. Suppression of CypA expression attenuated ERK1/2 activity in CCA M139 cells by using both transient and stable knockdown methods. In the in vivo study, there was a 43% reduction in weight of tumors derived from CypA-silenced CCA cell lines compared with control vector CCA tumors in mice; these tumors with stable CypA silencing showed a reduced cell proliferation.

Conclusions: CypA is upregulated in majority of CCA patients' tissues and confers a significant growth advantage in CCA cells. Suppression of CypA expression decreases proliferation of CCA cell lines in vitro and reduces tumor growth in the nude mouse model. Inhibition of CypA activity also reduces CCA cell proliferation. The ERK1/2 pathway may be involved in the CypA-mediated CCA cell proliferation. Thus, CypA may represent an important new therapeutic target for liver fluke-associated CCA.

Figure 1

Overexpression of CypA in CCA tissue specimens. Immunohistochemistry of CypA expression in cancerous and cancer-adjacent normal bile duct was examined. (A) CypA immunostaining in one representative paired CCA case. Stronger staining for CypA (dark brown) is seen in tumor tissue compared with adjacent normal bile duct lining cells (arrows). (B) CypA immunoreactivity index in normal bile duct versus CCA for 57 paired CCA cases. The majority of the paired tissue specimens (39/57, 68%) demonstrated higher CypA expression in malignant areas than in adjacent normal bile duct areas. (C) The overall median of the CypA immunoreactivity index is significantly higher in CCA samples compared to the adjacent normal bile duct tissue (P < 0.05; n = 57).

Figure 2

Endogenous expression of CypA in CCA cell lines correlates with cell proliferation rates. CypA expression was examined in 6 CCA cell lines and in a non-tumor human cholangiocyte cell line, MMNK1. When compared to MMNK1, three CCA cell lines (M055, KKU100, and M156) had relatively low CypA expression levels, and 3 cell lines (M139, M213, and M214) demonstrated relatively high CypA expression at both (A) mRNA and (B) protein levels as determined by real-time PCR and Western blot, respectively. (C) MTS cell proliferation assay demonstrates that low-CypA cells demonstrated lower proliferation rates than the high-CypA cells. MTS reading of each sample at the day 5 culture was normalized to that of its initial cell seeding at day 0, generating the fold increase.

Figure 3

Stable silence of CypA in M139 cells results in reduced cell proliferation. Stable cell lines expressing shCypA (CypA silence, M139-shCypA) or control shRNA (empty vector, M139-shV) were generated. CypA silence was confirmed by (A) real-time PCR using β-actin as an internal control; and (B) Western blot. (C) Stable CypA silence led to reduced cell proliferation in vitro. Cell proliferation rates are shown as the fold increase, normalized to the rate at the time of seeding. (D) CsA treatment inhibits cell proliferation.

Figure 4

Stable silence of CypA reduces wound healing ability, and thus migration ability of M139 cells in vitro. To measure cell migration potential, wound healing assay was carried out in M139-shV and M139-shCypA cells using RPMI 1640 medium supplemented with 0%, 2%, and 5% FBS. The wound width measured at 24 hours for each experiment was normalized to that measured at initial wound (0 hour), generating the relative migration distance. (A) Representative images taken from the same field at 0 and 24 hour. (B) Average relative wound width (relative migration distance) was compared among groups. Bar graphs represent means ± SD; n = 4, ** denotes P < 0.05 versus the control.

Figure 5

Effects of transient silence and forced overexpression of CypA on cell proliferation of CCA cell lines and a non-tumor cholangiocyte cell line. (A) CypA mRNA levels after transient silence of CypA in M139, M213, M214 CCA cell lines and MMNK1. (B) Transient silence of CypA reduces cell proliferation of CCA cells, but not MMNK1 cells in vitro. (C) CypA mRNA levels after transient overexpression of CypA for 48 hours. (D) Transient overexpression of CypA in CCA cell lines showed higher rate of cell proliferation compared to the control whereas MMNK1 cell proliferation was not affected by CypA overexpressed. Cell proliferation rate is shown as the fold increase compared to initial seeding. Data represent means ± SD of three independent experiments. ** denotes P < 0.05 versus the control.

Figure 6

CypA depletion attenuates ERK1/2 activity in M139 cells in vitro. ERK1/2 phosphorylation in CypA knockdown cells was determined by Western blot analysis. (A) Transient knockdown CypA decreased pERK1/2:ERK1/2 ratio in a time dependent fashion, and CypA protein level is positively correlated with ERK1/2 activity (phosphorylation). M139 cells (2 × 10⁵ cells) were seeded in 6-well plate for 24 hours before being transfected with 100 pmole/mL siRNA against CypA or negative control siRNA for 6 hours before replacing with complete medium. Lipofectamine™ 2000 was used as a transfection reagent. Total proteins were collected at 24, 48, and 72 hours after transfection. (B) Stable knockdown CypA cells had lower ERK activity (phosphorylation) than the control cells. Protein samples were collected from M139-shV and M139-shCypA cells grown in complete medium for 24 hours. All samples were subjected to Western blot analysis for CypA, ERK1/2 and pERK1/2 detection, and b-actin was used as a loading control. Intensity of the immunoreactive bands was measured by ImageQuant TL software (GE Healthcare). Protein level of each sample was normalized to b-actin prior to comparison between samples. The pERK1/2:ERK1/2 ratio value reflects ERK1/2 activity.

Figure 7

Stable silence of CypA results in decreased tumor growth and cell proliferation in a nude mouse model of subcutaneous (s.c.) xenograft. (A) Tumor weight (in grams). Insets: a representative mouse picture and s.c. tumor mass from each group. (B) Tumor tissues from M139-shCypA inoculated mice had significant lower levels of CypA-mRNA than those from M139-shV mice. The stability of CypA silencing in tumor tissues derived from M139-shCypA cells was confirmed by real-time PCR. (C) Immunohistochemical staining for CypA and Ki-67. (D) Quantification of Ki-67 immunohistochemical staining. Three tumors per group were analyzed and Ki-67 positive nuclei from four random high power (40×) fields were counted for each tumor tissue. Lower frequency of Ki-67 positive cells is shown in the tissue from M139-shCypA mice compared with that of Ki-67 positive cells from M139-shV mice (** P < 0.05).

Figure 8

A schematic representation of a proposed mechanism involved in CypA-mediated cell proliferation in CCA. CypA is upregulated in CCA cells, and several carcinogenic factors such as HIF-1a and p53 may regulate CypA expression. Unusual high CypA protein level may lead to sustained activation of ERK1/2 and NF-kB through direct or indirect mechanisms, which in turn activate cell survival, proliferation, and tumor progression. CypA may be secreted from the cells and act as an autocrine/paracrine growth factor via interacting with its membrane receptor CD147. Using specific inhibitor against CypA or against CypA/CD147 binding may inhibit CCA cell proliferation and tumor progression.