Abstract

Abstract 2334

Poster Board II-311

The outcome for patients with Chronic Lymphocytic Leukemia (CLL) is highly variable and correlates with clinical parameters such as disease stage, as well as various biological markers that reflect the genomic, phenotypic and functional properties of the tumor and its microenvironment. CD38, Cd49d and ZAP-70 are three such markers that are associated with a poor clinical outcome and which have previously been shown to play a role in migration. Progressive disease is invariably associated with tissue invasion. In the present study we investigated the possibility that the markers CD38, CD49d and ZAP-70 might be physically associated as well as functionally linked. Using 4-color flow cytometry we measured the level of expression of CD38 and CD49d by CD19+/CD5+ CLL cells in patients with CLL presenting to our routine diagnostic laboratory. A correlation was observed between both the percentage positivity and mean fluorescent intensity of CD38 and CD49d expression (r2=0.42, p=<0.0001, n=234; r2=0.28, p=<0.0001, n=61 respectively). Expression levels were not always linked, with some patients globally expressing a high percentage of either CD38 or CD49d positive cells with a low percentage of the other. However, within individual cells, a strong correlation was apparent, so that cells expressing a higher CD38 Mean Fluorescence Intensity (MFI) also had a high CD49d MFI and vice-versa. Since previous studies have shown that CD49d, CD44 and matrix metalloproteinase-9 (MMP-9) are physically associated in the CLL cell membrane and are involved in tissue invasion, we investigated the possibility that CD38 and ZAP-70 might also form part of this complex. CLL cell lysates were immunoprecipitated using anti-CD38 and then western blotted for CD49d, CD44, MMP-9 and ZAP-70. In 9/9 CD38+ cases, these immunoprecipitates contained both CD49d and CD44, whilst control immunoprecipitates on the same lysates and performed with an irrelevant antibody did not. In addition, the CD38 immunoprecipitate from 5/5 patients tested also contained MMP-9 and in 4/5 cases ZAP-70. CD5 was not detected in any of the samples, further confirming the specificity of these findings. The spatial relationship between CD38 and CD49d was next investigated by immunofluorescence confocal microscopy. CLL cells from 5 different patients known to express both CD38 and CD49d were adhered to poly-L-lysine coated slides, stained by indirect immunofluorescence and analyzed using confocal microscopy. Pixel counts showed that a mean of 64% of CD38 co-localised with CD49d (SEM 5.1%) and a mean of 72% of CD49d with CD38 (SEM 4.8%). These results thus show that the membrane of some CLL cells contains a macromolocular complex comprising CD38, CD49d, MMP9, CD44 and ZAP70, which is likely to play a role in mediating migration across the endothelial barrier into the tissues. Our findings also provide a possible explanation for the correlation between the level of expression of these molecules and adverse outcome since, as noted above, infiltration of the tissues is generally observed in patients with progressive disease. As normal B-lymphocytes do not contain the CD44/CD49d/MMP-9 complex and express neither CD38 nor ZAP-70, this structure would also seem to be an ideal target for therapies aimed at inhibiting tissue infiltration and disease progression in CLL.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.