Stem cells play key roles in early normal development (e.g. embryonic stem cells (ESCs)), maintenance of adult organs (e.g. hematopoietic stem cells (HSCs)) and in some cancers (e.g. leukemia stem cells). To what degree these different types of stem cells rely upon shared versus distinct transcriptional programs remains controversial. Sall4 is a zinc finger transcription factor that exists in two distinct splice isoforms, Sall4a (long) and Sall4b (short). Sall4 has been implicated in embryonic, hematopoietic and malignant stem cell transcriptional regulation. Additionally, Sall4 has been proposed as a potential means of ex-vivo hematopoietic stem cell expansion prior to transplantation. Sall4 isoform-specific differences have been described in ESCs, with Sall4b shown to be critical for maintaining ESC “stemness”. Here we investigate the role of Sall4 isoforms in pediatric acute myeloid leukemia (AML) and murine hematopoiesis to unravel shared versus unique transcriptional programs across different stem cell types.

Quantitative real time PCR shows that Sall4b is the predominant Sall4 isoform in murine HSCs and lin-, Sca1+, cKit+ (LSK) cells. Sall4b expression decreases in early lineage-committed progenitors, while Sall4a expression is minimal to absent across murine HSCs and progenitors. Next, we evaluated seven pediatric AML samples and found highly variable Sall4 expression across AML cases. All samples had measurable Sall4a and Sall4b; in 3/7 cases Sall4a and Sall4b expression was similar to that of ESCs, in the other 4 cases Sall4 expression was minimal (<3% of ESCs). To study overexpression of Sall4, we used a murine stem cell retrovirus system to express Sall4a or Sall4b. Bone marrow was harvested from C57/BL6 mice and lineage-committed cells were removed by magnetic column separation. Lineage-negative bone marrow was infected with either empty vector, Sall4a or Sall4b. Transduced bone marrow was then cultured in methylcellulose media to assess colony forming capacity and proliferation in vitro or transplanted in syngeneic mice to assess engraftment and hematopoietic reconstitution in vivo. Sall4a or Sall4b overexpression caused diminished colony forming capacity and cellular proliferation in vitro compared to bone marrow transduced with empty vector (Figure 1). In bone marrow transplant assays, all mice (4/4) transplanted with Sall4b-transduced bone marrow following lethal irradiation succumbed to bone marrow failure within 10 days of transplant. Transplantation of Sall4b-transduced bone marrow into sublethally irradiated mice failed to contribute to hematopoiesis as measured by peripheral blood leukocyte GFP expression (encoded by the viral vector). Together, this data shows that Sall4b-transduced hematopoietic cells fail to engraft and reconstitute hematopoiesis in vivo. We postulated that this phenotype might be mediated through the interaction of Sall4 with Bmi1. Bmi1 is a member of the polycomb complex necessary for normal hematopoiesis, and is known to be bound by Sall4. In preliminary experiments, we have found that overexpression of Sall4 leads to decreased Bmi1 expression at 48 hours post-infection compared to bone marrow infected with empty vector.
Figure 1
Figure 1. Lin- bone marrow expressing Sall4a, Sall4b or empty vector was cultured in methylcellulose; plates were flushed and replated out to three generations. Colony forming units were assessed (A) and viable cells were counted (B) after 7-10 days in culture.
View largeDownload slide

Lin- bone marrow expressing Sall4a, Sall4b or empty vector was cultured in methylcellulose; plates were flushed and replated out to three generations. Colony forming units were assessed (A) and viable cells were counted (B) after 7-10 days in culture.

Figure 1
Figure 1. Lin- bone marrow expressing Sall4a, Sall4b or empty vector was cultured in methylcellulose; plates were flushed and replated out to three generations. Colony forming units were assessed (A) and viable cells were counted (B) after 7-10 days in culture.
View largeDownload slide

Lin- bone marrow expressing Sall4a, Sall4b or empty vector was cultured in methylcellulose; plates were flushed and replated out to three generations. Colony forming units were assessed (A) and viable cells were counted (B) after 7-10 days in culture.

Close modal

In conclusion, our data shows that Sall4b is expressed in murine hematopoietic stem cells and progenitors, suggesting that Sall4b but not Sall4a influences a hematopoietic cell fate. Additionally, Sall4 expression is variable in AML specimens, implicating a potential pathogenic role in some leukemias, while others are Sall4-independent. Lastly, Sall4 overexpression is associated with decreased expression of the critical hematopoietic gene Bmi1. Together this data suggests that hematopoiesis is dependent upon appropriately regulated Sall4 expression with alterations leading to impaired proliferation and self-renewal. These effects on hematopoiesis appear to be mediated at least in part through a dose-dependent effect on Bmi1 expression. Future studies will evaluate other genes targeted by Sall4 in hematopoiesis and leukemia to define Sall4-dependent gene signatures in normal versus malignant hematopoiesis.

Disclosures:

No relevant conflicts of interest to declare.

Topics:
hematopoiesis, protein isoforms, sall4 gene, stem cell factor, enhanced s-cone syndrome, leukemia, transplantation, bone marrow transplantation, cancer, hyperplasia

Author notes

*

Asterisk with author names denotes non-ASH members.

© 2013 by The American Society of Hematology
2013
Sign in via your Institution