Costello P, Sargent M, Maurice D, Esnault C, Foster K, Anjos-Afonso F et al. MRTF-SRF signaling is required for seeding of HSC/Ps in bone marrow during development. Blood. 2015;125(8):1244-1255.

Prepared by: Allan Dong & Satvik Mareedu, Fall 2015.

LAY SUMMARY

The hematopoietic system constantly creates all of your blood cells over the course of your life. These blood cells all arise from hematopoietic stem cells (HSCs), which themselves reside in specialized stem-cell niches (a term for the specific microenvironment in the bone marrow that allows HSCs to thrive and renew). These cells do not start in these niches, though, in the early days of the fetus, these HSCs are initially generated in the regions of the fetus that would eventually become the torso and the yolk and placenta. They collect in the fetal liver, and finally colonize the bone marrow just prior to birth. Needless to say, if the HSCs do not become established in the bone marrow, the fetus will not survive.

Complex chemokine interactions ensure that the HSCs get to where they need to go and do what they need to do. One of the most important chemokines is serum response factor (SRF), which integrates cytoskeletal gene expression with extra cellular signals. It integrates with two families of signal regulated cofactors: MAP-kinase regulated ternary complex factors (TCFs: SAP-w, Elk1, and Net) and G-actin-regulated myocardin-related transcription factors (MRTFs: MRTF-A and MRTF-B). Fetal liver cells lacking all TCFs can effectively start hematopoiesis and lacking either MRTF-A or MRTF-B can have viable hematopoiesis, however, lacking both MRTFs resulted in death while in the embryo stage.

The experiments involve replacing the natural occurring genes coding for the chemokine with specific variations of them using vav-iCre - a tool used in genetic modification to insert genes into specific places. The mice are bred together, and the resulting embryos demonstrate normal Mendelian ratios of the specific chemokine gene variations that the experiment desires. Through examination of the resulting mice, it is possible to determine what happened to the HSCs. Further experiments involve irradiating adult mice in order to destroy their existing bone marrow, then injecting into the blood stream a new sample of fetal liver stem cells that were genetically modified to express or knock out certain genes. This was done in order to determine whether that gene combination was not found in the bone marrow because it was the chemokine to help the HSCs to target the bone marrow or because it was the chemokine to help the HSCs settle in to colonize the bone marrow.

While these experiments are highly suggestive of a possible method to recolonize sick or defective bone marrow with healthy colonies, it must be cautioned that these results have yet to be proven in human models. Also, all of our findings were done so on a short timescale and thus requires retesting on a longer timescale to ensure no significant long-term side effects.

SCIENTIFIC SUMMARY

Hematopoietic stem cells (HSCs) are created in the aorta-gonad-mesenphros region as well as the yolk and placenta and collect the fetal liver. Through a complex interaction of chemokines, cytoskeletal gene expression, and extracellular signals, these HSCs migrate from the fetal liver and colonize the bone marrow. The chemokine interactions are particularly important in ensuring that the HSCs properly constitute into a viable hematopoietic system.

The serum response factor (SRF) transcription factor integrates cytoskeletal gene expression and extra cellular signals using two families of signal-regulated cofactors: MAP-kinase-regulated ternary complex factors (TCFs: SAP-1, Elk1, and NET) and G-actin regulated myocardin-related transcription factors (MRTFs: MRTF-A and MRTF-B). Prior studies demonstrated that fetal liver cells lacking in all 3 TCFs an effectively reconstitute hematopoiesis, indicating that TCF-SRF signaling may not be critical in hematopoiesis system development. Using the vav-iCre gene insertion technique with SRF allele Srf f/f, MRTFA allele Mrtfa -/-, and MRTFB allele Mrtfb fl/+, the researchers bred mice embryos that displayed the chemokine phenotypes they desired. Further experiments involved modifying fetal liver cells with a non-functional Srf allele and injecting them into the tail vein of an total-body irradiated and acid-watered normal mouse and using modified LSK cells for transmigration assays

The results show that SRF is not required for fetal liver hematopoiesis but is required to graft durable colonies of HSCs because of defective adhesion and fails to establish viable hematopoiesis in the bone marrow. Fetal HSC/P homing requires SRF but the disparate effects between injecting SrfKO into irradiated and non-irradiated mice suggests that this effect can be partially mitigated by increasing chemokine signaling or vascular permeability. Using a transwell assay, the researchers found that the chemotactic response SRF signaling is required for chemotactic response of HSC/Ps and that SrfKO HSC/P cells fail to migrate in response to SDF-1 signals. Analyzing LSK transcription with RNA-seq, the researchers discovered that multiple cytoskeletal genes are Srf-dependent. Then the researchers  reperformed the previous experiments but to discover the effects of a lack of MRTF-A or MRTF-B instead of SRF.  They discovered that while lacking either Mrtfa of Mrtfb could still result in at least some colony formation and barely any effect in chemotactic responses, but lacking both would result in practically no colony formation and substantial defective chemotactic response. Overall, these results show that SRF signaling is absolutely critical for establishing hematopoietic function and that the two MRTF genes function redundantly but some MRTF signaling is also critical for establishing hematopoietic function.

While these experiments are highly suggestive of possible methods to recolonize sick or defective bone marrow with healthy colonies, it must be cautioned that these results have yet to be proven in human models. Further research is needed to confirm the same findings in human models, with longer followup times in order to ensure no significant long-term side effects.