miRNA expression was not due to difference as both GS, maGS and ES cells, used in this study, were of male origin and contained XY sex chromosomes. We also evaluated whether expression of imprinted miRNAs in mouse GS and maGS cells alters during their in vitro differentiation. As expected, stem cell, and GS cell marker genes were silenced during differentiation of GS and maGS cells but was not complete in ES cells. During differentiation, expression of imprinted miRNAs also underwent changes but varied both with the differentiation stage and the miRNA. Given that the EBs contain a mixed population of cells from three germ layers and that the expression of individual imprinted miRNA may vary with the cell/tissue types wherein they may have different functions, the differential expression pattern of imprinted miRNAs in differentiating GS, maGS and ES cells may reflect their differential ability to differentiate into various cell types. Interestingly, although a clear pattern was not evident, changes in the expression pattern of imprinted miRNA during in vitro differentiation of maGS cells were apparently more similar to those of ES cells and clearly differed from differentiating GS cells. This observation is similar to those of Zovoilis et al., who found that maGS cells resembled ES cells, but the expression of certain miRNAs such as miR-290 cluster was retained during their in vitro differentiation. We also observed that, similar to previous reports on several miRNAs, mature miRNA originating from both 39 and 59 arms of the miR-296 accumulated as sister pairs in undifferentiated testis-derived germline stem cells. However, their expression pattern differed among the differentiating cells of the three groups and, the EBs generated from GS cells resembled those of ES cells for the expression pattern of miR-296- 3p. Since the phenomenon of miRNA strand selection for the functional stability occurs in a tissuedependent manner, the differences in the expression of miR-296-3p and miR-296-5p among the EBs of the three groups probably reflects the different proportion of cells of three germlayers in them. It was also observed that, differentiating EBs generated from GS cells had significantly high level of miR-127 and miR-127-5p, which might suggest their possible role during in vitro differentiation of SSC. Our result is similar to a previous study which showed high expression of miR-127 in testicular samples. The miR-127 was preferentially expressed in immature mouse testes that principally contained mitotically active spermatogonia, meiosis I spermatocyte and round spermatid, but remained at SAR131675 structure medium level in purified pachytene spermatocyte and round spermatid and declined in adult testis. However, the mechanism by which miR27 affects the SSC biology remains unclear. It is likely that miR127 may act by down-regulating.