Role of alternative splicing in hematopoietic stem cells during development

With the advent and increasing resolution of genome-wide sequencing modalities, alternative splicing (AS) has become an emerging area of interest in normal development and disease (1). Splicing of mRNA transcripts results in the production of a multitude of alternative isoforms from a single genomic locus, dramatically enhancing the diversity of the transcriptome and proteome (2). AS may affect RNA nuclear export and transcript stability, thereby regulating transcript levels. Distinct splicing isoforms may produce functionally equivalent, divergent, or inactive mature proteins (3).

Recently, attention towards characterizing the impact of splicing events during hematopoiesis has intensified, driven in part by the discovery of recurrent splicing factor mutations in myelodysplastic syndromes (MDS), clonal disorders of the hematopoietic stem cell (HSC) (4). The finding that mutations in one of 4 splicing factors occur in nearly 50% of patients has resulted in numerous studies characterizing the roles of AS events during both normal hematopoiesis and disease. By mapping whole-transcriptome splicing in murine HSCs from several high-quality datasets, Goldstein et al. found that the majority (248/322) of HSC-specific genes, including HoxA9, Meis1, Prdm16, and Hlf are expressed as multiple isoforms with similar or divergent functions (5). Previously, Bowman et al. had identified an abundance of unspliced transcripts mapping to DNA binding and RNA processing factors in HSCs, regulated upon HSC activation (6). Similarly, Wong et al. identified a program of widespread intron retention and consequent non-sense mediated decay (NMD) of these transcripts during granulocytic differentiation as a rapid mechanism of post-transcriptional regulation that was conserved between mice and humans (7). Chen et al. compiled a comprehensive “atlas” of transcriptional diversity during human hematopoiesis by sequencing isolated human HSCs and their progeny, mapping gene expression, AS, and novel splice site utilization throughout the hematopoietic hierarchy (8). Of the approximately 2,000 genes with differential splice site selection between cell types as many as sixty percent were predicted to produce functional protein changes or result in premature stop codons that would slate the transcript for NMD (8). Komeno et al. demonstrated that AS of Runx1 affecting exon 6 resulting in a RUNX1a ortholog unique to human, determines the size of the HSC compartment (9). In MDS and several other cancers AS is co-opted as evident by recurrent splicing factor mutations (4,10-14). While full understanding about how these mutations cause disease is still unclear, key splice events contributing to the disease have been identified. As an example, point mutations in the serine-arginine rich splicing factor 2 (SRSF2), result in inclusion of a poison exon in EZH2 thereby mimicking pathogenic loss of function mutations in EZH2, mutually exclusive with SRSF2 mutations (15).

A recent publication by Cesana et al. from the Daley laboratory reports an interesting link between splicing and post-transcriptional regulation in the developmental adaptation of HSC function (16). During ontogeny HSCs lose their proliferative potential and alter their lineage output to adjust to the organism’s needs. The investigators isolated early HSCs (CD34⁺ CD38⁻ CD90⁺ CD45RA⁻) and CD34⁺ CD38⁺ progenitors from fetal livers (FL), cord blood (CB) and bone marrow (BM) and performed deep RNA-sequencing (RNA-seq) and miRNA profiling. Numerous transcripts were differentially expressed, originally thought of as “universal” markers of stemness, with only a small subset uniformly expressed across HSC populations. Among the uniformly expressed transcripts, a few showed differential regulation of their isoforms between FL, CB, and BM HSCs. HMGA2 caught the investigators’ eye. HMGA2 is expressed in two specific isoforms, a full-length (HMGA2-L) and a shorter isoform (HMGA2-S). While HMGA2-L predominates in FL HSCs, HMGA2-S is more abundant in CB HSCs. Interestingly, differential splicing results in selection of an alternative and much shorter 3' UTR in HMGA2-S than in HMGA2-L. Upon further scrutiny it is evident that the HMGA2-S 3' UTR lacks most conserved miRNA binding sites, especially for the let-7 family of miRNAs. Functional relevance is experimentally confirmed: while the HMGA2-L isoform is repressed by the let-7 miRNA family, the HMGA2-S isoform is unaffected. Incidentally, HMGA2-S preference co-occurs with down-regulation of LIN28B in CB HSC. LIN28B is well known for its importance in stem cell maintenance and its role in inhibiting let-7 biogenesis (17,18). This suggests, that by escaping let-7 mediated suppression, HMGA2-S may maintain a stem cell program in CB HSCs despite down-regulation of LIN28B and up-regulation of let-7 miRNA family members. Indeed, HMGA2-S and HMGA2-L are functionally equivalent in driving HSC stemness. Frequently, alternative 3' UTR choices are achieved through alternative poly-adenylation through selection of proximal or distal polyadenylation sites within the 3' UTR sequence [reviewed in (19)]. In the case of HMGA2, the alternative 3' UTR is the result of AS of the otherwise skipped alternative terminal exon 4. To determine how this AS event may be achieved, the authors screened for potential splicing regulators and experimentally identified CLK3 kinase as the most likely candidate. CLK3 belongs to the CDC-like kinases, which regulate RNA binding protein and splicing factor activity via their phosphorylation (20). In a next step therefore, the investigators identified global splicing changes induced by knockdown of CLK3 and computationally identified RNA binding proteins potentially regulated by CLK3 through extraction of most common binding motifs in differentially spliced exons. This data pointed towards SRSF1 as the CLK3 phosphorylation target responsible for HMGA2 differential splicing. The HMGA2-L but not the HMGA2-S isoform contains SRSF1 binding motifs and the HMGA2-L alternative exons are indeed regulated by SRSF1. Interestingly, the two HMGA2 isoforms are functionally indistinguishable. The CLK3-SRSF1 mediated AS solely seems to serve preservation of HMGA2 expression and maintenance of a HMGA2-orchestrated HSC-specific program in an increasingly let-7 dominated cellular context. Interestingly, adult CD34⁺ hematopoietic stem and progenitor cells but not FL or CB HSCs regain proliferative and repopulating potential upon overexpression of HMGA2-S or CLK3, suggesting that the CLK3-SRSF1-HMGA2 axis may contribute to the developmentally defined HSC-specific program inherent to FL and CB. In this context, it is of particular interest that SRSF1 is overexpressed in breast cancer (21,22) highlighting the close link between stemness and cancer.

Several questions remain. It would be of particular interest whether the CLK3-SRSF1-HMGA2 axis entirely explains the FL and CB stem cell phenotype and if therapeutic modulation of this axis could thus reverse the aging stem cell phenotype. While CLK3 and SRSF1 emerged as the investigators’ top targets, it is well known that CLKs can phosphorylate more than one SR protein and that SR proteins cooperate and compete with each other and other splicing factors, such as members of the HNRNP family [reviewed in (23)]. It will thus be of importance to understand to what extent the overall cellular context modifies the CLK3-SRSF1-HMGA2 axis.

Understanding the broader role of the CLK3-SRSF1 module in HSC function and ontogeny may open up novel therapeutic avenues. With the recent FDA approval of the first therapeutic splicing regulator Nusinersen in spinal muscular atrophy (24), transient modulation of the SR protein mediated splicing program in HSC may be feasible and allow HSC expansion in vitro for the purpose of stem cell transplantation or other therapeutic needs.

Acknowledgements

Funding: Dr. Halene is funded by the NIH/NIDDK R01DK102792, the State of Connecticut under the Regenerative Medicine Research Fund, the Leukemia and Lymphoma Society, the Department of Defense, and the Edward P. Evans and Frederick Deluca Foundations.

Footnote

Conflicts of Interest: The authors have no conflicts of interest to declare.

References

1. Pan Q, Shai O, Lee LJ, et al. Deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. Nat Genet 2008;40:1413-5. [Crossref] [PubMed]
2. Nilsen TW, Graveley BR. Expansion of the eukaryotic proteome by alternative splicing. Nature 2010;463:457-63. [Crossref] [PubMed]
3. Baralle FE, Giudice J. Alternative splicing as a regulator of development and tissue identity. Nat Rev Mol Cell Biol 2017;18:437-51. [Crossref] [PubMed]
4. Yoshida K, Sanada M, Shiraishi Y, et al. Frequent pathway mutations of splicing machinery in myelodysplasia. Nature 2011;478:64-9. [Crossref] [PubMed]
5. Goldstein O, Meyer K, Greenshpan Y, et al. Mapping Whole-Transcriptome Splicing in Mouse Hematopoietic Stem Cells. Stem Cell Reports 2017;8:163-76. [Crossref] [PubMed]
6. Bowman TV, McCooey AJ, Merchant AA, et al. Differential mRNA processing in hematopoietic stem cells. Stem Cells 2006;24:662-70. [Crossref] [PubMed]
7. Wong JJ, Ritchie W, Ebner OA, et al. Orchestrated intron retention regulates normal granulocyte differentiation. Cell 2013;154:583-95. [Crossref] [PubMed]
8. Chen L, Kostadima M, Martens JHA, et al. Transcriptional diversity during lineage commitment of human blood progenitors. Science 2014;345. [Crossref] [PubMed]
9. Komeno Y, Yan M, Matsuura S, et al. Runx1 exon 6-related alternative splicing isoforms differentially regulate hematopoiesis in mice. Blood 2014;123:3760-9. [Crossref] [PubMed]
10. Graubert TA, Shen D, Ding L, et al. Recurrent mutations in the U2AF1 splicing factor in myelodysplastic syndromes. Nat Genet 2011;44:53-7. [Crossref] [PubMed]
11. Furney SJ, Pedersen M, Gentien D, et al. SF3B1 mutations are associated with alternative splicing in uveal melanoma. Cancer Discov 2013;3:1122-9. [Crossref] [PubMed]
12. Wang L, Lawrence MS, Wan Y, et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N Engl J Med 2011;365:2497-506. [Crossref] [PubMed]
13. Papaemmanuil E, Cazzola M, Boultwood J, et al. Somatic SF3B1 Mutation in Myelodysplasia with Ring Sideroblasts. N Engl J Med 2011;365:1384-95. [Crossref] [PubMed]
14. Liang Y, Tebaldi T, Rejeski K, et al. SRSF2 mutations drive oncogenesis by activating a global program of aberrant alternative splicing in hematopoietic cells. Leukemia 2018. [Epub ahead of print]. [Crossref] [PubMed]
15. Kim E, Ilagan JO, Liang Y, et al. SRSF2 Mutations Contribute to Myelodysplasia by Mutant-Specific Effects on Exon Recognition. Cancer Cell 2015;27:617-30. [Crossref] [PubMed]
16. Cesana M, Guo MH, Cacchiarelli D, et al. A CLK3-HMGA2 Alternative Splicing Axis Impacts Human Hematopoietic Stem Cell Molecular Identity throughout Development. Cell Stem Cell 2018;22:575-588.e577. [Crossref] [PubMed]
17. Shyh-Chang N, Daley GQ. Lin28: primal regulator of growth and metabolism in stem cells. Cell Stem Cell 2013;12:395-406. [Crossref] [PubMed]
18. Viswanathan SR, Daley GQ, Gregory RI. Selective blockade of microRNA processing by Lin28. Science 2008;320:97-100. [Crossref] [PubMed]
19. Tian B, Manley JL. Alternative polyadenylation of mRNA precursors. Nat Rev Mol Cell Biol 2017;18:18-30. [Crossref] [PubMed]
20. Colwill K, Pawson T, Andrews B, et al. The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO J 1996;15:265-75. [PubMed]
21. Anczuków O, Akerman M, Clery A, et al. SRSF1-Regulated Alternative Splicing in Breast Cancer. Mol Cell 2015;60:105-17. [Crossref] [PubMed]
22. Anczuków O, Rosenberg AZ, Akerman M, et al. The splicing factor SRSF1 regulates apoptosis and proliferation to promote mammary epithelial cell transformation. Nat Struct Mol Biol 2012;19:220-8. [Crossref] [PubMed]
23. Long JC, Caceres JF. The SR protein family of splicing factors: master regulators of gene expression. Biochem J 2009;417:15-27. [Crossref] [PubMed]
24. Mercuri E, Darras BT, Chiriboga CA, et al. Nusinersen versus Sham Control in Later-Onset Spinal Muscular Atrophy. N Engl J Med 2018;378:625-35. [Crossref] [PubMed]