In Vitro Propagation of Mesenchymal Stromal Cells From Marrow Aspirates of Patients with Chronic Lymphocytic Leukemia Is Dependent Upon Physiologic Oxygen Tension

Abstract 2839

The tumor microenvironment may play an important role in the growth and/or survival of leukemia cells of patients (pts) with chronic lymphocytic leukemia (CLL). Studies on the interaction of CLL cells with the microenvironment have been facilitated by our capacity to culture accessory cells in vitro. However, the conditions for culturing such cells in ambient oxygen(O₂) at 21% are different than those present in lymphoid tissues which have O₂ concentrations ranging from 1–7%. The difference between in vivo and in vitro O₂ tensions might influence the biology of leukemia accessory cells.

To examine this, we studied the effect(s) of O₂ tension on our ability to propagate mesenchymal stromal cells (MSCs) from marrow aspirates of pts with CLL. Equal numbers of fresh or viably frozen marrow mononuclear cells were seeded in DMEM with 10% FBS into separate flasks for culture at 37° C in incubators at atmospheric O₂ (Atmos-O₂) or at 5% O₂ (physiologic; Phys-O₂), both with 5% CO₂. The cells were monitored for viability and growth over time. We found that only Phys-O₂ tension allowed for the generation and long-term expansion of MSCs. Out of the 6 pts tested, 3 developed virtually no MSCs (<10 cells), and 3 generated less than 6×10⁴ MSCs (ranging from 2±1×10⁴ to 6±1×10⁴) in Atmos-O₂ after 47±7 days in vitro. In contrast, high numbers of MSCs developed in Phys-O₂ for all 6 pts, ranging from 45±8×10⁴ to 80×10⁴ cells, resulting in highly significant differences in yields between the 2 culture conditions (p<0.01). Moreover, the MSCs generated in Phys-O₂ continued to proliferate over time, whereas MSCs in Atmos-O₂ did not. Under Phys-O₂, MSCs were successfully expanded from marrow aspirates of 16 out of 18 CLL pts. The morphology and phenotype of the MSCs generated were similar to that of healthy MSCs, expressing CD29, CD44, CD105 and D7-FIB, and lacking expression of CD14, CD31, CD34, or CD45.

We next examined whether the differences in cell yields between the two culture conditions could be the result of compromised MSC proliferation in Atmos-O₂. To address this, MSCs generated in Phys-O₂ were seeded into separate flasks and exposed to Atmos-O₂ or Phys-O₂ and proliferation was monitored by BrdU incorporation and viable cell counts. We found that MSCs seeded in Atmos-O₂ proliferated significantly less well than MSCs in Phys-O₂ (n=3). However, MSC viability was not significantly affected by the change in O₂ tension, suggesting that replicative senescence could be induced in MSCs exposed to Atmos-O₂. To test this hypothesis, MSCs generated in Phys-O₂ were seeded separately under Atmos-O₂ or Phys-O₂ and stained for the senescence-associated beta-galactosidase (SA-B-Gal) marker. We found a significant increase in the fraction SA-B-gal+ MSCs exposed to Atmos-O₂ compared to Phys-O₂ (70±18% vs. 13±5%; p<0.0001). MSC morphology in Atmos-O₂ was also consistent with senescence, marked by a wide-spread cytoplasm and enlarged nucleus. The cell cycle regulator p16INK4 also was distinctively induced in MSCs exposed to Atmos-O₂ compared to Phys-O₂ (n=2), consistent with its role in inhibiting cell cycle progression and mediating senescence.

We next ask if a disruption of the redox balance plays a role in MSC biology modulated by O₂, using the free radical scavenger beta-mercaptoethanol (BME). MSCs generated in Phys-O₂ were seeded separately under Phys-O₂ or Atmos-O₂ +/− BME. We observed that MSC proliferation in Atmos-O₂ was restored by the addition of BME to the levels observed in Phys- O₂ using BrdU incorporation (n=3). However, in a similar context, BME did not restore MSC expansion in Atmos-O₂ measured by viable cell counts, or in the generation of MSCs in Atmos-O2 when added at culture initiation (n=2), suggesting that redox balance disruption is not the main mechanism by which high O₂ tension affects MSC biology.

We finally evaluated whether co-cultures of CLL cells and MSCs in Phys-O₂ and Atmos-O₂ tensions equally support CLL cell survival. We found that CLL cell survival was significantly enhanced when co-cultured in Phys-O₂ compared to Atmos-O₂ after 17 days (87±15% vs 44±17% viable cells; p<0.0001). These results suggest that Phys-O₂ tension is not only critical to generate MSCs in vitro, but it also has a profound impact on the biology of these accessory cells, which in turn affects the survival of the leukemic cells. Studies conducted under Phys-O₂ tension might further our understanding of the mechanisms governing CLL cell survival in vivo.