High-level expression of the T-cell chemokines CCL3 and CCL4 by chronic lymphocytic leukemia B cells in nurselike cell cocultures and after BCR stimulation

Abstract

In lymphatic tissues, chronic lymphocytic leukemia (CLL) cells are interspersed with CD68(+) nurselike cells (NLCs), T cells, and other stromal cells that constitute the leukemia microenvironment. However, the mechanism regulating colocalization of CLL and these accessory cells are largely unknown. To dissect the molecular cross talk between CLL and NLCs, we profiled the gene expression of CD19-purified CLL cells before and after coculture with NLCs. NLC coculture induced high-level expression of B-cell maturation antigen and 2 chemoattractants (CCL3, CCL4) by CLL cells. CCL3/CCL4 induction in NLC cocultures correlated with ZAP-70 expression by CLL cells. High CCL3/CCL4 protein levels were found in CLL cocultures with NLCs, and CCL3/CCL4 induction was abrogated by R406, a Syk inhibitor, suggesting that NLCs induce these chemokines via B-cell receptor (BCR) activation. BCR triggering also caused robust CCL3/CCL4 protein secretion by CLL cells. High CCL3 and CCL4 plasma levels in CLL patients suggest that this pathway plays a role in vivo. These studies reveal a novel mechanism of cross talk between CLL cells and their microenvironment, namely, the secretion of 2 T-cell chemokines in response to NLC coculture and BCR stimulation. Through these chemokines, CLL cells can recruit accessory cells and thereby actively create a supportive microenvironment.

Figure 1

DNA microarray analysis identifies a homogeneous gene expression response of CLL cells after coculture with NLCs. (A) These heatmaps depict the top 10 genes that are up-regulated (top 10 rows of the graph) or down-regulated (bottom 10 rows) in 9 different CLL cell samples before and after 14 days of coculture with NLCs (14d NLC). The changes in gene expression are depicted for each gene relative to its expression level immediately after Ficoll isolation and CD19 purification and before initiation of NLC cultures (black squares). Shades of red and green indicate up- or down-regulation, respectively, of a given gene according to the color scheme displayed below the heatmaps. Displayed are the expression responses in 9 different CLL patient samples, as indicated on the top horizontal axis. (B) To illustrate the correlation between ZAP-70 expression and CCL3 and CCL4 induction, ZAP-70 gene expression in individual CLL cases was classified as high or low, as determined by microarray analysis and correlated with CCL3 and CCL4 expression before and after NLC coculture. In CLL cases with high ZAP-70 gene expression (cases 1, 3, 6, 7, and 9), NLC coculture uniformly induced a robust expression of CCL3 and CCL4, whereas in cases with low ZAP-70 gene expression the response was more heterogeneous (no induction of CCL3 and CCL4 in case 2; intermediate induction in cases 4, 5, and 8).

Figure 2

Induction of CCL3 and CCL4 protein expression in CLL-NLC cocultures. CLL PBMCs were placed in long-term cultures with outgrowth of NLCs, as described in “RNA isolation from CLL B cells before and after NLC coculture.” To determine the kinetics of induction of CCL3 and CCL4 protein expression, we analyzed supernatant samples at the time points indicated on the horizontal axis by ELISA. (A) The different symbols that are connected by the lines represent the mean CCL3 and CCL4 concentrations in 20 different CLL patients' samples at the indicated time points. In accordance with the microarray data, we found higher CCL4 protein expression compared with CCL3. (B,C) The individual protein expression data for CCL3 and CCL4 in the 20 CLL cases are displayed in dot plot diagrams. ZAP-70⁺ cases (•) can be distinguished from ZAP-70⁻ cases (○).

Figure 3

CCL3 and CCL4 protein expression in ZAP-70⁺ and ZAP-70⁻ CLL cases. Displayed are the mean and SEM for CCL3 (A) and CCL4 (B) protein concentrations in supernatants from ZAP-70⁺ CLL samples (, n = 10) or ZAP-70⁻ CLL samples (, n = 10). In ZAP-70⁺ CLL samples, we found higher levels of both CCL3 and CCL4 at all time points. However, these differences did not reach statistical significance.

Figure 4

BCR engagement protects CLL cells from undergoing apoptosis. Presented are contour maps of CLL B cells from one patient defining the relative green (DiOC₆) and red (PI) fluorescence intensities of CLL cells on the horizontal and vertical axes, respectively. The viability was determined after 24 hours of culture in medium alone (control) or medium supplemented with anti-IgM mAbs at the concentrations indicated above each of the contour maps. This stain allows us to gate on the vital cell population (DiOC₆^bright, PI^exclusion), apoptotic cells (DiOC₆^dim, PI^exclusion), and the dead cells (DiOC₆^dim, PI^positive). The percentage of each cell population is indicated next to each of these gates. In this case, anti-IgM stimulation protected approximately 20% of the cells from undergoing apoptosis, resulting in a viable population of 56.5% at 14 μg/mL anti-IgM and 55.1% at 42 μg/mL anti-IgM. The bar diagram on the lower right displays an increased viability of CLL cells after culture with anti-IgM in 4 different CLL samples. CLL cells were cultured in medium alone (control) or medium supplemented with anti-IgM mAbs at the indicated concentrations, and the viability was assessed after 24 and 48 hours by staining with DiOC₆ and PI. Displayed are the means plus or minus SD of 4 different patient samples.

Figure 5

Induction of CCL3 and CCL4 protein expression by CLL cells after BCR engagement: anti-IgM dose-response. CLL cells were placed in culture medium supplemented with different concentrations of anti-IgM, as indicated on the vertical axis. CLL cells cultured with medium alone were used as controls. After 24 hours, supernatants were removed and assayed for CCL3 (A) and CCL4 (B) protein expression by ELISA. Displayed are the mean (± SD, n = 3) CCL3 and CCL4 concentrations, as indicated on the horizontal axis. Whereas anti-IgM induced a dose-dependent induction of CCL3 and CCL4, CLL cells cultured in medium alone did not express significant amounts of CCL3 and CCL4.

Figure 6

Time course of CCL3 and CCL4 protein expression by CLL cells after BCR engagement. CLL cells were placed in culture medium and stimulated with 15 μg/mL anti-IgM. Supernatants were removed at the time points indicated on the horizontal axis and assayed for CCL3 (A) and CCL4 (B) protein expression by ELISA. The different symbols that are connected by the lines represent the CCL3 and CCL4 concentrations in 4 different CLL samples at the indicated time points.

Figure 7

The Syk inhibitor R406 abrogates induction of CCL3 and CCL4 secretion by CLL cells in cocultures with NLCs. The bars represent the mean (±SEM) CCL3 and CCL4 concentration in CLL cell supernatants from 4 different patients incubated for 24 or 48 hours with NLCs in the presence or absence of the Syk inhibitor R406, as indicated on the horizontal axis. Treatment with R406 significantly inhibited the induction of CCL3 (A) and CCL4 (B) secretion by the CLL cells, as indicated by the asterisks, with P values less than .05 at 24 hours (*) and 48 hours (**). No significant CCL3 or CCL4 concentrations were detected in supernatants of NLCs alone (data not shown).

Figure 8

CCL3 and CCL4 protein levels in blood plasma from CLL patients and healthy volunteers. This bar diagram displays the mean (± SEM) concentrations of CCL3 and CCL4 in 67 plasma samples from CLL patients and normal donors (n = 9), as displayed on the horizontal axis. CLL plasma displayed higher levels of both CCL 3 and CCL4 compared with normal blood plasma, suggesting that secretion of these chemokines by CLL cells occurs in vivo.