Sparc Is Dispensable for Murine Hematopoiesis, Despite Its Suspected Role in 5q- Myelodysplastic Syndrome

Abstract 4822

Hematopoiesis is a complex process where a limited number of stem cells give rise to all mature blood cells. It involves interplay of several factors, many of which are yet to be identified. In a search for novel regulators of hematopoiesis, we chose to study SPARC (Secreted Protein Acidic and Rich in Cysteine, also known as Osteonection and BM40) because it is downregulated upon hematopoietic differentiation (Bruno et al., Mol Cell Biol, 2004) and might therefore play a role in the regulation of hematopoietic stem cells (HSC). SPARC is a matricellular protein that forms a major component of bone and is ubiquitously expressed in a variety of tissues. It is the founding member of a family of SPARC-like proteins. Several publications have indicated an important role for SPARC in hematopoiesis. In particular – knockdown of SPARC in zebrafish embryos resulted in an altered number of circulating blood cells, and a knockout mouse model showed thrombocytopenia and reduced erythroid colony formation.

We carried out an in depth phenotypic and functional analysis of the hematopoietic system of SPARC knockout mice; using it as a model to gain insight into the role of SPARC in hematopoiesis. These mice are viable and fertile but show severe osteopenia and age-onset cataract at about six months of age. They also show an altered response to tumour growth and wound healing. We used mice (129SVJ background) (Gilmour et al. EMBO, 1998) that were less than six months old. These mice had normal peripheral blood counts and the bone marrow and spleen showed no alterations in morphology or cellularity. A detailed phenotypic analysis of precursors within the bone marrow showed no significant differences in myelo-erythroid precursors as compared to wild types (n=6). Though in vitro, the precursors showed lower ability to form BFU-E (n=5, p=0.048).

In transplantations of lethally irradiated recipient mice, SPARC knockout cells gave rise to multi-lineage long-term reconstitution. Also, when competed with wild type cells, they provided reconstitution as well as their wild type counterparts. When SPARC knockout mice (n=8) were transplanted with wild type cells, there was normal reconstitution, indicating that a SPARC deficient niche can fully support normal hematopoiesis.

We also tested if SPARC deficient mice respond differently to hematopoietic stress. We subjected mice (n=7) to sub lethal dose of irradiation and to experimentally induced anemia (n=7) and followed recovery by analyzing peripheral blood counts. In both SPARC knockouts and wild type mice, the blood counts recovered in a similar fashion.

In conclusion, we find that SPARC is dispensable for murine hematopoiesis. It is possible that there are compensatory mechanisms involving other members of the SPARC family that ultimately lead to normal hematopoiesis in the murine model.

In humans, SPARC maps to the deleted region in 5q MDS and has been reported to be 71 % down regulated in patient samples (Lehmann et al. Leukemia, 2007). It is the most prominent gene that is up regulated in response to lenalidomide, a drug that inhibits the malignant clone (Pellagatti et al. PNAS, 2007). SPARC is thus increasingly speculated to be involved in the pathophysiology of this hematopoetic disease. We analysed the expression levels of SPARC mRNA in the hematopoietic stem/progenitor cell compartment and found high expression levels in the CD34+ fraction of human cord blood cells. In contrast, there is very low level of SPARC expression in all compartments of murine HSCs. Therefore SPARC function may play a more important role in human hematopoiesis than in murine blood cell regulation.