Abstract
Abstract 36
Chronic myeloid leukemia (CML) is caused by the BCR-ABL oncogene. CML patients lack expression of IRF-8 - an interferon-regulated transcription factor that has been shown to exert tumor suppressor functions. IRF-8 is also critical for the development of a rare dendritic cell population, so called plasmocytoid dendritic cells (PDC). PDC are quantitatively significantly reduced or absent in the peripheral blood of first diagnosis CML patients. PDC are also the major producers of IFN-alpha (IFNa) in man. IFNa is a cytokine that has significant therapeutic efficiency in the treatment of CML patients.
We here wished to experimentally test, whether BCR-ABL expression and loss of IRF-8 may be causally linked to a reduction of PDC in murine CML and whether there could be any functional relevance for PDC loss in CML development or treatment.
PDC counts were studied from peripheral blood samples of primary CML patients at diagnosis, at the time of remission or from healthy donors. PDC function was assessed in vitro by treatment of magnetic bead-enriched PDC with Toll-like receptor 9-specific oligos (ODN 2216) and subsequent assessment of the intracellular IFNa expression in stimulated PDC. A supposed link between BCR-ABL expression, IRF-8 repression and loss of PDC counts was studied in vivo using a murine CML transduction-transplantation model (C57/Bl6 mice, 7Gy sub-lethal irradiation for conditioning). Multiparameter flow cytometry and cell sorting were performed to analyze and enrich, BCR-ABL-positive (GFP+) hematopoietic subpopulations and PDC in order to then quantitate their IRF-8 and BCR-ABL transcript level by RT-PCR. In order to also test the functional relevance of PDC during CML leukemogenesis, CML mice were injected intravenously, weekly from day +5 after transplantation with in vitro generated PDC. Mice were simultaneously also s.c.-injected weekly with ODN 2216 to stimulate IFNα secretion in adoptively transferred PDC in vivo.
As previously reported, newly diagnosed CML patients displayed a significantly reduced PDC count when compared to healthy donors (p<0.001). Upon remission induction with imatinib, PDC counts restored partially, but to a much lesser extend in patients successfully treated with IFNa therapy. Importantly, albeit significantly reduced in number, BCR-ABL-positive first diagnosis CML PDC seem to be functionally intact: CML and healthy donor PDC produced comparable amounts of IFNa in response to Toll-like receptor 9 -specific CpG ODN 2216 stimulation. This suggested that BCR-ABL may compromise PDC function by quantitative rather than qualitative dysregulation.
CML mice developed a fatal, BCR-ABL-positive myeloproliferation within 13 to 29 days with 88% penetrance. Compared to control mice (n=8), CML mice (n=14) showed a 7-fold and 3-fold reduction of the frequency of B220+mPDCA-1+ PDC in bone marrow and spleen, respectively. This was associated with a statistically significant (4-fold) suppression of IRF8 mRNA expression in sorted BCR-ABL(GFP)-positive PDC relative to BCR-ABL-negative PDC from the same mice (n=3) or from control transplantations (n=5). By RT-PCR, there was a trend also for lower IRF8 expression in CML progenitor cells (Lin− c-Kit+ Sca-1- GFP+), but not in the stem cell enriching population (Lin− c-Kit+ Sca-1+ GFP+). This implied that IRF8 expression is lost during BCR-ABL-induced leukemogenesis in more mature compartments, which supposedly include PDC precursors.
Intriguingly, a once weekly adoptive transfer of in vitro generated (to > 30% enriched) PDC for three successive weeks combined with a once weekly subcutaneous injection of CpG ODN 2216 for three weeks was sufficient to almost double survival of CML mice.
Using a murine model of CML, we provide first experimental evidence that BCR-ABL induced myeloproliferation is causally linked to a quantitative suppression of PDC, and that this is associated with a BCR-ABL-mediated suppression of IRF8 transcription. Since adoptively transferred PDC were capable of counteracting murine CML development, BCR-ABL may facilitate leukemogenesis in part by obstructing PDC maturation. PDC could thus be a novel immunological effector cell population that exerts and/or integrates anti-leukemic immune responses in CML.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.