Pulmonary Stromal Cells Expressing CC-Chemokine Receptor 5 Promote Metastasis Via Erythroid Differentiation Regulator 1.

The protein erythroid differentiation regulator 1 (ERDR-1) is a stromal-derived factor that was previously found to induce hemoglobin synthesis in erythroleukmia cells and proliferation in hematopoietic progenitors (Dormer et al. Cytokine. 2004). However, previous work had not suggested a role for ERDR-1 in either local or distant tumor growth. We have previously shown that stromal cells, and in particular fibrocytes, which are CD45+ monocyte-derived stromal cell precursors, promote metastasis and this effect is partly dependent on expression of CC-chemokine receptor 5 (CCR5). Here, we sought to determine a mechanism for this finding.

We used Affymetrix Mouse Arrrays to identify genes that were differentially expressed in lung homogenates from wild type (WT) and CCR5 ^−/− mice 6, 24 and 48 hours after the intravenous (i.v.) injection of B16-F10 tumor cells. After identification of ERDR-1 as a candidate gene, we evaluated WT and CCR5 ^−/− primary stromal cells for expression of ERDR-1. We also meausred ERDR-1 expression in mouse embryonic fibroblasts (MEFs), and knocked-down the gene in these cells using 2 different shRNA sequences specific for ERDR-1. A third line of MEFs expressing a scrambled (“mock”) shRNA sequence was generated as a control for non-specific effects. To model the effect of stromal ERDR-1 on metastasis, we transferred ERDR-1 knockdown or mock knockdown MEFs into CCR5 ^−/− or WT mice 48 hours prior to injection of B16-F10 cells. Lung nodules were counted 14 days after tumor injection. We also evaluated the effect of ERDR-1 knockdown on in vitro proliferation and survival.

ERDR-1 was overexpressed in the lungs of wild type (WT) compared to CCR5 ^−/− mice by 3.7, 2.7, and 3.5 fold at 6, 24, and 48 hours after tumor injection, respectively. Semi quantitative end-point PCR confirmed this observation in all mice tested (n = 30). Real Time RT-PCR revealed that the in vitro expression of ERDR-1 was higher in primary stromal cells from WT compared to CCR5 ^−/− mice by 5.2 fold (3.3 to 11.2, p < 0.001, results pooled from 3 separate primary cultures). This difference was most pronounced in the CD45+ (fibrocyte) population. There was a substantial increase in the expression of ERDR-1 mRNA and protein in MEFs compared to primary WT stromal cells (15.86 +/− 6.9 fold greater expression by Real Time RT-PCR, p < 0.001) and this correlated with a greater number of metastases in the lungs of mice receiving MEFs compared to WT primary stromal cells prior to B16-F10 injection. In CCR5 ^−/− mice receiving ERDR-1 knockdown MEF lines prior to tumor injection we found a 37% and 50% decrease in the number of metastases compared to mice receiving the mock knockdown MEFs ( p = 0.06 and p < 0.05, respectively). When MEF lines transduced with 43.1% lower efficiency were used, mice receiving the knockdown lines had a 24% and 26% reduction in the number of metastases compared to mice receiving the mock knockdown MEFs (p < 0.05 for both groups). Conversely, when ERDR-1 knockdown lines were transferred to WT mice prior to tumor injection, there was no significant reduction in metastases compared to the group receiving mock knockdown MEFs. This was similar to our previous finding that numbers of metastases were similar for WT mice receiving either CCR5 ^−/− or WT stromal cells prior to tumor cells. These results indicated that endogenous ERDR-1 expression in WT but not CCR5 ^−/− mice was sufficient to compensate for lower ERDR-1 expression in transferred stromal cells. Interestingly, we also found that knockdown of ERDR-1 in MEFs led to enhanced apoptosis in vitro (1.7 and 2.8 –fold increase in Annexin-V+ cells) compared to mock knockdown MEFs, suggesting that ERDR-1 acts as a stromal cell survival factor.

The relative inability of primary stromal cells from CCR5 ^−/− mice to promote metastatic tumor growth in the lung was found to correlate with decreased expression of ERDR-1. Knocking down ERDR-1 substantially decreased the number of lung metastasis induced by WT MEFs. Thus, the production of ERDR-1 by stromal cells appears to be important for the CCR5-dependent promotion of metastatic tumor growth.

No relevant conflicts of interest to declare.