Function of Pumilio Genes in Human Embryonic Stem Cells and Their Effect in Stemness and Cardiomyogenesis

Posttranscriptional regulation plays a fundamental role in the biology of embryonic stem cells (ESCs). Many studies have demonstrated that multiple mRNAs are coregulated by one or more RNA binding proteins (RBPs) that orchestrate the expression of these molecules. A family of RBPs, known as PUF (Pumilio-FBF), is highly conserved among species and has been associated with the undifferentiated and differentiated states of different cell lines. In humans, two homologs of the PUF family have been found: Pumilio 1 (PUM1) and Pumilio 2 (PUM2). To understand the role of these proteins in human ESCs (hESCs), we first demonstrated the influence of the silencing of PUM1 and PUM2 on pluripotency genes. OCT4 and NANOG mRNA levels decreased significantly with the knockdown of Pumilio, suggesting that PUMILIO proteins play a role in the maintenance of pluripotency in hESCs. Furthermore, we observed that the hESCs silenced for PUM1 and 2 exhibited an improvement in efficiency of in vitro cardiomyogenic differentiation. Using in silico analysis, we identified mRNA targets of PUM1 and PUM2 expressed during cardiomyogenesis. With the reduction of PUM1 and 2, these target mRNAs would be active and could be involved in the progression of cardiomyogenesis.


INTRODUCTION
Human embryonic stem cells (hESCs) are pluripotent cells derived from the 42 inner cell mass of the blastocyst that have potential for differentiation into three germ 43 layers (1-3). In an undifferentiated state, hESCs are characterized by the expression of 44 stemness factors such as OCT4 (POU5F1), SOX2 and NANOG (4). These three 45 transcription factors, which are positively regulated, are responsible for pluripotency 46 maintenance and contribute to the repression of lineage-specific genes (reviewed by 5). 47 When hESCs are stimulated to initiate the differentiation process, expression of genes 48 associated with pluripotency is negatively regulated and genes associated with the germ 49 layer begin to be positively regulated (6).

50
A complex network of gene expression underlies the molecular signaling that 51 will give rise to the adult heart. Cardiomyogenic differentiation is a highly regulated 52 process that depends on the fine regulation of gene expression (7). In vitro 53 cardiomyogenic differentiation of hESCs mimics embryonic development and can be 54 used as a model for cardiac development studies per se and as a model for research 55 ranging from tissue electrophysiology to drug screening (reviewed by 8).  The expression of PUM1 and PUM2 has been observed in hESCs and several 69 human fetal and adult tissues, indicating a possible participation in the maintenance of 70 germ cells (11,12). Furthermore, in mammals, the disruption of PUM proteins promotes 71 defective germline phenotypes (18,19 and 10 ng/ml human βFGF. The cells were passaged every 3-4 days by enzymatic 101 dissociation using 0.25% trypsin/EDTA. Cardiomyogenic differentiation assays were 102 conducted using an embryoid body (EB) protocol adapted from previously described 103 (31,32) or a monolayer protocol previously reported (33).

104
Regarding EB cardiac differentiation protocol, briefly, 7x10 5 cells/well were 105 plated onto Growth Factor Reduced Matrigel ® Matrix (Corning) 6-well coated dishes.    hours. Then, the medium was replaced with supplemented DMEM, as described above.

141
After 48 and 72 hours, the medium was collected and centrifuged twice at 141000 x g. 142 The cell pellet was resuspended in 1X PBS and stored at -80 ºC.   Table S1). We generated standard curves for The immunofluorescence protocol followed as previously described (7). Briefly, 179 monolayer cultures we fixed with paraformaldehyde 4%, rinsed with PBS, followed by   Statistical analysis was performed using GraphPad Prism 7 software. The data 208 sets are expressed as the means ± standard deviation. According to data sets were used 209 unpaired Student's t-test or one-way ANOVA followed by Tukey post hoc test.

210
Differences with p<0.05 were considered statistically significant.

235
To understand the role of PUM1 and PUM2 in hESC maintenance or during 236 cardiomyogenic differentiation, we silenced their expression using short hairpin RNAs. 237 We produced lentiviral particles containing shRNA that recognize PUM1, PUM2 and a  percentage of cTnT+ cells not changed between the different treatments ( Figure 3C).

272
These results demonstrated that when PUM was silenced, hESCs followed EBs cardiac 273 differentiation efficiently, with no statistically significant changes. 274 We performed a monolayer cardiomyogenic differentiation protocol, as 275 previously described (33) ( Figure 4A). In this protocol we transfected hESCs with shSc   Table S2).   (22,41), and a compensatory regulation mechanism has been observed when one 335 of these genes is silenced by increasing the expression of the other (43). We evaluated 336 the levels of PUM1 and PUM2 mRNAs after silencing these genes individually, and we 11 337 did not observe this compensation, at least at the mRNA level. We hypothesized that 338 due to their high similarity, the shRNA used to knockdown one transcript impacted the 339 stability of the other transcript, at least in this cell type. 340 We observed that the PUM1 and PUM2 silencing altered the mRNA levels of