**3. Discussion**

The pluripotency factors LIN28A and LIN28B inhibit the maturation of *let-7* miRNAs [39,62]. Recently, we showed that both Lin28A and LIN28B were significantly decreased and levels of *let-7* miRNAs (*let-7a, let-7b, let-7c, let-7d, let-7e, let-7f, let-7g, let-7i*) were significantly increased in term human

placentas from IUGR pregnancies compared to control pregnancies [18]. We further demonstrated that double knockout of LIN28A and LIN28B in immortalized first trimester human trophoblast (ACH-3P) cells resulted in more robust increase in *let-7* miRNAs compared to knockout of either LIN28A or LIN28B. Similarly, double knockin of LIN28A and LIN28B in Sw.71 cells led to more robust decrease in *let-7* miRNAs compared to knockin of either LIN28A or LIN28B [18]. In this study, we show that RNA interference (RNAi) of LIN28A or LIN28B in sheep TE in vivo resulted in significant increase in *let-7* miRNAs (*let-7a, let-7b, let-7c, let-7d, let-7e, let-7f, let-7g, let-7i*). Moreover, the conceptus elongation was significantly reduced after RNAi of LIN28A or LIN28B in TE.

Although animal models with global gene knockout or knockdown have been extensively and successfully used in many studies, it is difficult to exclude the effect of global gene manipulation while focusing on mechanisms involved in one tissue type or organ. Incubating the day 9-hatched sheep blastocyst with shRNA expressing lentiviral particles caused viral infection of only trophoblast cells while all other cells including the inner cell mass were spared of lentiviral infection [61]. Hence, significant reduction in LIN28A or LIN28B in day 16 TE conceptus was restricted to the trophoblast cells only. Conceptus elongation in sheep was due to rapid proliferation of trophoblast cells [7–9]; therefore, a significant reduction in conceptus length at day 16 after trophectoderm-specific LIN28A or LIN28B knockdown indicates reduced proliferation of trophoblast cells.

LIN28 is a part of a complex genetic pathway known to regulate multiple downstream targets [64]. Inhibition of biogenesis of mature *let-7* miRNAs is one of the main pathways through which LIN28 regulates the expression of its downstream targets [32,65]. *Let-7* miRNAs reduce expression of many genes by degrading their mRNAs or inhibiting translation [39]. Our results show that RNAi of LIN28A or LIN28B and resultant increase in *let-7* miRNAs in day 16 TE significantly reduced IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B, and c-MYC. Previous studies have shown the role of IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B, and c-MYC in cell proliferation [17,38–40,43,45,46,48,51], suggesting that reduced proliferation of trophoblast cells after LIN28A or LIN28B knockdown in TE is due to reduced expression of proliferation-associated genes (Figure 12).

**Figure 12.** Graphical abstract. (**A**) In control day 9 TE, *let-7* miRNAs are low because of high LIN28A and LIN28B. Because of low levels of let-7 miRNAs, IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B, and c-MYC are higher leading to increased proliferation of trophoblast cells and hence conceptus elongation. (**B**) In LIN28A or LIN28B KD day 9 TE, *let-7* miRNAs are higher. Elevated let-7 miRNAs target *IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B,* and *c-MYC*, leading to reduced expression of these genes, leading to reduced proliferation of trophoblast cells and, hence, reduced conceptus elongation.

To further investigate the role of the LIN28-*let-7* miRNA axis in sheep trophoblast cells in vitro, OTR cells were generated from day 16 TE. Surprisingly, both LIN28A and LIN28B were depleted and the *let-7* miRNAs were significantly higher in OTR cells compared to day 16 TE. The senescence of OTR cells after only a few passages along with high *let-7* miRNAs suggests that these cells are a differentiated phenotype of trophoblast cells compared to day 16 TE. Furthermore, the expression of proliferation-associated genes including IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B, and c-MYC was also significantly reduced in OTR cells compared to day 16 TE. To overcome the OTR cell senescence, we generated immortalized ovine trophoblast (iOTR) cells by expressing hTERT in these cells. The OTR cells were differentiated and depleted LIN28 and immortalization with hTERT expression did not change LIN28 expression in iOTR cells. Immortalized human first trimester trophoblast cells (Sw.71 cells) were also generated by expressing hTERT in first trimester human trophoblast cells at passage 3 [66]. Sw.71 cells also had depleted LIN28A and LIN28B and high *let-7* miRNAs compared to other trophoblast-derived cell lines such as ACH-3P cells [18]. The iOTR cells generated in this study were similar to Sw.71 cells in terms of LIN28 and *let-7* miRNAs expression, which may be because both cell lines were generated by immortalizing the passaged trophoblast cells. The contrasting levels of LIN28, *let-7* miRNAs, and *let-7* miRNA target genes between day 16 TE and iOTR cells should be taken in consideration if using these cells for further studies.

To generate iOTR cells that were more similar to day 16 TE, we overexpressed LIN28A or LIN28B in iOTR cells. Overexpression of LIN28A or LIN28B in iOTR cells led to a significant decrease in *let-7* miRNAs and significant increase in expression of IGF2BP1, IGF2BP2, IGF2BP3, HMGA1, ARID3B, and c-MYC compared to control. The proliferation of both AKI and BKI iOTR cells was significantly increased compared to control; however, there was no change in cell invasion. We recently showed that knockout of LIN28A and LIN28B in ACH-3P cells reduced cell proliferation but did not affect cell invasion [19]. Both ACH-3P and iOTR cells have low invasion; therefore, it would be hard to see a further reduction in invasion. Moreover, cell proliferation and invasion are two distinct and critical processes during placental development. Reduced trophoblast cell proliferation without reduced invasion can be sufficient to cause impaired conceptus elongation and attachment. These results suggest that the LIN28-*let-7* miRNA axis plays a role in proliferation of immortalized trophoblast cells by regulating the expression of genes associated with cell proliferation. We suggest that the iOTR cells overexpressing LIN28A and LIN28B would be a better choice to study molecular mechanisms in ovine trophoblast cells compared to iOTR cells with depleted LIN28A and LIN28B.

To our knowledge, this is the first in vivo study defining the role of LIN28-*let*-7 miRNA axis in early placental development by trophoblast-specific RNAi of LIN28A or LIN28B. Due to a wide range of *let-7* miRNAs target genes, as well as the ability of LIN28 to directly bind the mRNA of different genes, LIN28 knockdown might be affecting many different genetic pathways in trophoblast cells, which are yet to be explored. Knockdown of LIN28A or LIN28B and high *let-7* miRNAs in TE led to reduced conceptus elongation, which can result in impaired placentation, fetal growth restriction, loss of pregnancy, and reduced fertility in domestic ruminants. We recently showed that term human placentas from IUGR pregnancies have low LIN28 and high *let-7* miRNAs [18]. MicroRNAs can be easily measured in tissue biopsies, blood, and other biological samples. Based on our studies, we suggest that low LIN28 or high *let-7* miRNAs in placenta could be detected in blood and be used as potential biomarkers for intrauterine growth restriction.
