Although chronic lymphocytic leukemia (CLL) is an accumulative disorder of CD5+ B cells, a proliferative fraction exists and the extent of this proliferation correlates with clinical course. We previously demonstrated heterogeneous in vivo labeling kinetics of cells in 3 clonal fractions sorted based on reciprocal densities of chemokine (C-X-C motif) receptor 4 (CXCR4) and CD5 from CLL cases recruited on an in vivo labeling study. In that study, the CXCR4dimCD5br fraction contained more 2H-labeled DNA and hence recently divided cells (proliferative fraction) than the CXCR4brCD5dim(resting) fraction. Here, we have analyzed multiple CXCR4/CD5 based subclonal fractions and characterized their phenotypic differences and ability to respond to signals via the BCR and CXCR4 or a combination of the two.
PBMCs from a cohort of 50 untreated IgM+ CLL cases were used for this study. Subclonal fractions marked based on differences in relative densities of CXCR4 and CD5 by CLL cells are referred to as CXCR4br, CXCR4int or CXCR4dim and CD5br, CD5int or CD5dim. Sub-populations adjacent to each other on flow cytometric plots differed at least 5 fold in the densities of both CXCR4 and CD5. When subcategorizing CLL clones based on these markers, 9 subfractions were identified. To permit comparisons between these fractions, each was defined as containing only 2-5% of the total clonal CD19+CD5+population while ensuring no overlap in CXCR4 or CD5 densities among each fraction.
Interestingly, certain phenotypic and signaling features consistently clustered with distinct subfractions of the CLL clone. Subfraction analyses confirmed an enrichment of Ki-67+ cells in the CXCR4dimCD5br proliferative fraction of every CLL case. In addition, this subfraction showed the highest percentage of cells expressing smIgM and smIgD (p<0.01). Relative density (RD) of expression was deduced as a ratio of mean fluorescence intensity (MFI) of a molecule expressed by any subfraction compared to that expressed by CXCR4intCD5int cells within the same clone. When considering all cases the CXCR4dim CD5dim cells showed the highest RD of smIgM. However, among U-CLL cases the CXCR4dim CD5br cells expressed the highest RD for smIgM, and the CXCR4dim CD5dim subfraction showed the highest RD of smIgM expression in M-CLLs. Overall, both CXCR4dim CD5br and CXCR4int CD5br showed the highest RD for IgD expression, whereas among U-CLL and M-CLL cases the CXCR4int CD5br and CXCR4dim CD5int subfractions showed the highest RD of smIgD expression respectively. The CXCR4brCD5br subfraction was characterized by the highest % and RD of CD79b, the BCR signaling molecule, and CD22 and CTLA-4, negative regulators of signaling. However, Siglec-10, another negative regulator of BCR signaling, was most highly expressed both in % and RD by the CXCR4dim CD5br subfraction in all CLL cases. Finally, the highest percentages of cells expressing CD21, CD38, CD43, CD69, CD180 associated best with the 3 CD5br subfractions. Of note, CD38 was expressed at highest RD in the CXCR4dim CD5dim in M-CLL cases but in the CXCR4dim CD5brsubfraction in the U-CLL cases.
When analyzing phosphorylation of Akt, Erk, p38MAPK and Syk by phosphoflow among all cases, the CXCR4brCD5br fraction exhibited the highest constitutive levels. Constitutive levels of phospho-Btk and - Syk were significantly higher in the CXCR4int CD5br subfraction in M-CLL than in U-CLL cases (p<0.05). Anti-BCR mAbs induced significant increases in phospho-p38MAPK and -Akt in the CXCR4br CD5br subfraction when comparing U-CLL to M-CLL cells. CXCL12 induced significant increases in phospho- Akt and -STAT-5 in CXCR4int CD5br subfraction in U-CLL cases (p<0.01), whereas a combination of signals via BCR and CXCR4 induced increases in phospho-Erk in the CXCR4int CD5br and CXCR4br CD5brpopulations in M-CLL cases.
Together, these findings highlight the impact of the molecules expressed by the different subfractions on their ability to relay/inhibit signals elicited via the BCR or CXCR4. They also provide a general frame of reference to further understand functional post-replicative intraclonal heterogeneity of CLL clones and help define aggressive subfractions of the CLL clone as target populations for future therapies.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.