In Vivo Labeling of Newly Synthesized DNA Suggests That the CD38⁺ Fraction Is Enriched in Proliferating Cells within a Clone of Chronic Lymphocytic Leukemia B Cells.

B-cell chronic lymphocytic leukemia (B-CLL) has been considered for years as a slowly progressive invasion of phenotypically homogeneous, monoclonal B cells that divide rarely and accumulate because of a primary apoptotic defect. Recent studies are clarifying that the leukemic clone can be comprised of subpopulations of cells expressing different surface markers suggestive of intraclonal diversity. Indeed, it is now widely accepted that activation signals (i.e., antigenic stimulation, interaction with accessory cells, cytokines) can push B-CLL cells into the cell cycle or rescue them from apoptosis. Unanswered questions are whether all or only a subset of cells in the clone participates in these proliferative events and to what extent those cells that have divided can be identified.

We have tried to answer this by labeling those cells that divide in vivo with the non-radioactive stable isotope deuterium (²H) in the form of deuterated water. Nine B-CLL patients with leukemic clones comprising 1.7 – 97.8% CD38⁺ cells drank daily, for 12 weeks, an aliquot of ²H²O. CD19⁺CD5⁺CD38⁺ and CD19⁺CD5⁺CD38⁻ cells were isolated from PBMC drawn after 8 weeks (labeling period) and 24 weeks (washout period) and the incorporation of ²H into equal amounts of genomic DNA was measured by gas chromatography/mass spectrometry.

At week 8, the mean ²H enrichment in the CD38⁺ population was higher than in the CD38⁻ fraction (0.862 ± 0.24 vs. 0.445 ± 0.13; P = 0.034). Enhanced ²H enrichment in the CD38⁺ population was detected in every patient with CD38⁺/CD38⁻ ratios ranging from 1.15 to 5.58. In 4 patients, the CD38⁺/CD38⁻ ratio was above 2.0. In contrast, at week 24, 12 weeks after the cessation of ²H²O intake, ²H enrichment in DNA in the CD38⁺ vs. the CD38⁻ population had changed. In 7 of 9 patients, the CD38⁺/CD38⁻ enhancement ratios diminished, remaining essentially constant for the other 2 patients. The difference of CD38⁺/CD38⁻ mean ratios at week 8 and week 24 was significant (2.254 ± 0.12 vs. 1.124 ± 0.13; P = 0.035). Moreover, the incorporation of ²H into CD38⁺ cells fell in 6 of 9 patients, with reciprocal increases in label incorporation in the CD38⁻ cells for 8 of 9 patients. Thus, the differences in ²H incorporation into DNA at weeks 8 and 24 suggest either that the life spans of the two subpopulations differ (i.e., CD38⁺ cells may be shorter lived) or that expression of CD38 may be a dynamic process in vivo changing from positive to negative in individual cells over time after replication.

These in vivo findings indicate that, within the same B-CLL patient, the leukemic subpopulation that expresses CD38 contains more cells that have recently undergone cell division than the CD38⁻ subpopulation. This may relate to the clinical correlation of CD38 expression with poor clinical outcome. The ²H²O labeling technique provides a safe in vivo method to further fractionate the proliferative leukemic compartment as a means to better understand the pathophysiology of B-CLL and possibly provide an important target for therapy.