Activation of CCR9/CCL25 in cutaneous melanoma mediates preferential metastasis to the small intestine

Abstract

Purpose: Specific chemokines and their respective receptors have been implicated in distant tumor cell metastasis. Cutaneous melanoma has a distinct pattern of metastasis, preferentially targeting the submucosa of the small intestine. However, the underlying pathogenic mechanism remains unknown. Migration of CCR9(+) lymphocytes to the small intestine is known to occur in response to the chemoattractant effects of CCL25 (thymus-expressed chemokine). The integrin heterodimers alphabeta are also known to be important mediators of cellular adhesion. We hypothesize that the mechanism of small intestinal metastasis by melanoma is via the CCR9-CCL25 axis and specific integrins.

Experimental design: Quantitative reverse transcription-PCR, flow cytometry, and immunohistochemistry were used to assess melanoma tumors for CCR9 and CCL25. Integrin expression was assessed using flow cytometry. CCR9 expression by quantitative reverse transcription-PCR was assessed in primary (n = 23) and metastatic (n = 198) melanomas, and melanoma lines derived from small intestinal metastases (n = 23).

Results: We showed CCR9 expression in 88 of 102 paraffin-embedded metastatic melanomas from the small intestine, 8 of 8 melanoma lines derived from metastases in the small intestine, and 0 of 96 metastatic melanomas from other sites. In vitro migration and invasion studies done on CCR9(+) melanoma lines showed migration in response to CCL25 that was inhibited by anti-CCR9 antibody or by short interfering RNA CCR9. Flow cytometric analysis confirmed CCR9 expression by melanomas to the small intestine and showed concomitant alpha(4)beta(1) integrin expression.

Conclusions: Our findings show that functionally active CCR9 on melanoma cells facilitates metastasis to the small intestine. The CCR9-CCL25 axis may explain the high incidence of melanoma metastasis to this specific location.

Fig. 1

CR9 expression in melanoma lines. A, CCR9 expression in melanoma lines derived from small intestinal metastases. MOLT 4 cell line was used as a positive control for CCR9 expression. B, CCR9 expression was absent in melanoma lines derived from melanoma metastases to other visceral organs.ME-21andMOLT 4 are shown as positive controls for CCR9 expression. Columns; mean bars, SD.

Fig. 2

Flow cytometric analysis of CCR9 and integrin expression by melanoma cells. Flow cytometric analysis of CCR9, α₄, and β₁ integrin molecule expression (open histograms) versus their respective isotype controls (shaded histograms) on single-cell suspensions derived from cryopreserved small intestinal melanoma metastases (SI Met ; data shown is one representative sample of four tested), and two cell lines derived from a small intestinal metastasis (ME-17 and ME-21). ME-54, a cell line derived from a liver metastasis, was used as a negative control for CCR9 expression; MOLT-4 was used as a positive control. Low-level expression of CCR9, and higher levels of α₄ and β₁ integrin expression were seen on tumor cells in the single-cell suspension as well as the two tumor cell lines derived from small intestinal metastases.

Fig. 3

CCR9 expression in PEATs from metastatic melanoma. CCR9 mRNA expression by melanoma metastases to the small intestine (A), and other visceral organs (B), assessed by qRT-PCR. Normal kidney tissue, negative control (103 in A). Normal duodenum tissue, positive control (104 in A). Blank spaces are negative results. MOLT 4 cell line used as a positive control (not shown: CCR9/GAPDH mRNA value, 3.5 × 10⁻³).

Fig. 4

CCL25 expression in PEATs from small intestinal melanoma metastases. CCL25 mRNA expression by melanoma metastases to the small intestine assessed by qRT-PCR. Normal duodenum tissue, positive control (103). Normal kidney tissue, negative control (104).

Fig. 5

Representative immunohistochemistry staining for CCR9. Representative immunohistochemistry staining using anti-CCR9 antibody for CCR9 expression in melanoma small intestinal metastases specimens demonstrating strong immunoreactivity (A and B). Representative staining of negative control tissues is shown in C.

Fig. 6

CCR9(+) melanoma cell migration and invasive responses to CCL25. A, cell migration of two representative melanoma lines ME-17 and ME-87 that were CCR9(+). Stimulation with CCL25 (100 ng/mL) significantly increased the number of migrating (ME-17 and ME-87) cells (both P < 0.001) as determined by the migration assay. No treatment; CCL25 treatment. CCR9(−) melanoma lines showed no response to CCL25. B, cell Matrigel matrix invasion assay of two representative melanoma lines (ME-17 and ME-87) not treated and treated with CCL25, respectively (P < 0.002). Addition of the anti-CCR9 antibody (1 µg/mL) resulted in a significant decrease in the number of cells that invade across the Matrigel matrix in response to CCL25 (P < 0.004). CCL25 + anti-CCR9 antibody. C, qRT-PCR analysis of representative CCR9(+) melanoma lines was done after CCR9 siRNA and control siRNA transfection. After siRNA treatment, a significant decrease in CCR9 expression is seen in ME-21 (P = 0.002) cells and ME-17 (P = 0.004) cells: ME-21 cell line; ME-17 cell line. D, cell migration assay of two representative melanoma lines after CCR9 siRNA transfection after treatment with CCL25; Y-axis, number of cells per high-power field. Controls include CCL25 treatment alone, with laminin siRNA or scrambled siRNA. There was a significant decrease in CCR9 siRNA-transfected cells migrating in response to CCL25 (P < 0.004 and P < 0.01, respectively): ME-21 cell line; ME-17 cell line.