**6. Gene Regulation by LIN28-***let-7* **miRNA Axis in Trophoblast Cells**

Due to the profound role of *let-7* miRNAs as differentiation-inducing miRNAs, the focus of our lab is to investigate the role of LIN28-*let-7* miRNA axis in trophoblast cells. Both LIN28A and LIN28B are highly expressed in human placenta and are localized to trophoblast cells [11,130–132]. High throughput genotyping array reveals that LIN28B is paternally imprinted in human placenta [133,134]. Using single cell transcriptome profiling, Liu et al. identified 14 different cell types in human placenta and showed that paternally imprinted LIN28B has high expression in CTBs, EVTs and STB, whereas it has no to low expression in mesenchymal cells, macrophages, and blood cells in placenta [135]. They further showed that LIN28B expression in week 24 EVTs was lower compared to week 8 EVTs [135], suggesting that expression of LIN28B in trophoblast cells reduces as the pregnancy progresses. LIN28B is the main paralogue of LIN28 in human placenta and *LIN28B* mRNA is 1300-fold higher compared to *LIN28A* mRNA in term human placental tissue [14]. Immunohistochemical analysis of term human placenta shows that LIN28B expression in CTBs and STB is higher compared to placental decidual cells [14]. In 2013, Gu el al. compared the expression of miRNAs between first and third trimester human placentas [136]. They reported that along with many other miRNAs, *let-7a, let-7c, let-7d, let-7f, let-7g*, and *let-7i* are upregulated in third trimester compared to first trimester human placenta [136]. We measured *LIN28A* and *LIN28B* mRNA in first trimester (11 week) vs. term human placentas and found that *LIN28A* mRNA was nearly 700-fold higher and *LIN28B* mRNA was nearly 300-fold higher in first trimester compared to term human placenta (Figure 3). Based on these results, we suggest that increased expression of *let-7* miRNAs in term human placentas, reported by Gu et al., is due to reduced expression of *LIN28A* and *LIN28B*. Low LIN28 and higher level of *let-7* miRNAs in term

placenta compared to first trimester placenta suggest that the proliferation rate of trophoblast cells is higher during the first trimester and decreases with advancement in gestational age. As LIN28-*let-7* miRNA axis regulates expression of several genes, it would not be surprising to see a difference in gene expression in first trimester vs. third trimester human placenta. *Int. J. Mol. Sci.* **2019**, *20*, x FOR PEER REVIEW 8 of 20

**Figure 3.** mRNA was extracted from first trimester (week 11) and term human placentas and *LIN28A* and *LIN28B* mRNA levels were measured using real-time RT-PCR, where \* *p* < 0.05.

**Figure 3.** mRNA was extracted from first trimester (week 11) and term human placentas and *LIN28A* and *LIN28B* mRNA levels were measured using real-time RT-PCR, where \* *p* < 0.05. In IUGR pregnancies, the size of placenta is significantly smaller compared to normal pregnancies [137], which suggests the role of reduced trophoblast proliferation in etiology of IUGR. In a recently published study, we showed that term human placentas from IUGR pregnancies have low LIN28A and LIN28B, and high *let-7* miRNAs compared to term human placentas from normal pregnancies [11]. Canfield et al. reported that term human placentas from preeclamptic pregnancies have reduced LIN28B but no change in LIN28A compared to normal term placentas [14]. They further demonstrated that in first trimester human placenta, LIN28B is higher in extravillous cytotrophoblasts compared to villous trophoblast cells, indicating their role in trophoblast cell In IUGR pregnancies, the size of placenta is significantly smaller compared to normal pregnancies [137], which suggests the role of reduced trophoblast proliferation in etiology of IUGR. In a recently published study, we showed that term human placentas from IUGR pregnancies have low LIN28A and LIN28B, and high *let-7* miRNAs compared to term human placentas from normal pregnancies [11]. Canfield et al. reported that term human placentas from preeclamptic pregnancies have reduced LIN28B but no change in LIN28A compared to normal term placentas [14]. They further demonstrated that in first trimester human placenta, LIN28B is higher in extravillous cytotrophoblasts compared to villous trophoblast cells, indicating their role in trophoblast cell invasion [14]. Low LIN28 and high *let-7* miRNAs during the first trimester of pregnancy can lead to reduced trophoblast proliferation and invasion leading to pregnancy-related disorders.

invasion [14]. Low LIN28 and high *let-7* miRNAs during the first trimester of pregnancy can lead to reduced trophoblast proliferation and invasion leading to pregnancy-related disorders. Due to the limitation that humans cannot be used as experimental models, most studies investigating molecular mechanisms involved in human placental development are conducted using placental cell lines. Commonly used human trophoblast-derived cell lines include BeWo, ACH-3P, JEG3, JAR, Sw.71, and HTR8/SVneo. LIN28A knockdown in immortalized first trimester human trophoblast (ACH-3P) cells drives these cells towards syncytial differentiation and increases the expression of syncytiotrophoblast markers including *hCG, LGALS13*, and *ERVW-1* [132]*.* Moreover, knockdown of LIN28A increases the expression of *let-7* miRNAs including *let-7a, let-7c, let-7d, let-7e, let-7g*, and *let-7i* [132], suggesting that differentiation of cells might be due to increased levels of *let-7*  miRNAs*.* Overexpression of LIN28B in HTR8 cells increases cell proliferation, invasion, and migration, whereas knockdown of LIN28B in JEG3 cells reduces cell proliferation [14]. In a recently published study, we further investigated the correlation between LIN28 and *let-7* miRNAs in trophoblast cells using first trimester human trophoblast-derived ACH-3P and Sw.71 cells. ACH-3P cells were generated by fusing first trimester human trophoblast cells with human choriocarcinoma cells, whereas Sw.71 cells were generated by overexpressing human telomerase reverse transcriptase (h-TERT) in first trimester human trophoblast cells [138,139]. These cell lines have contrasting levels of LIN28 and *let-7* miRNAs [11]. ACH-3P cells have high expression of LIN28A and LIN28B whereas these proteins are not detectable in Sw.71 cells [11]. The expression of all *let-7* miRNAs is 50–500-fold higher in Sw.71 cells compared to ACH-3P cells, potentially due to depleted LIN28A and LIN28B in Sw.71 cells which are major suppressors of *let-7* miRNAs [11]. The contrasting levels of LIN28 and *let-7* miRNAs between ACH-3P and Sw.71 cells are potentially due to the difference of methodology used to generate these cell lines. LIN28A knockout in ACH-3P cells increases *let-7a, let-7b, let-7c, let-7d*, and *let-7e*, whereas LIN28B knockout in ACH-3P cells increases *let-7a, let-7b, let-7c, let-7d, let-7e*, Due to the limitation that humans cannot be used as experimental models, most studies investigating molecular mechanisms involved in human placental development are conducted using placental cell lines. Commonly used human trophoblast-derived cell lines include BeWo, ACH-3P, JEG3, JAR, Sw.71, and HTR8/SVneo. LIN28A knockdown in immortalized first trimester human trophoblast (ACH-3P) cells drives these cells towards syncytial differentiation and increases the expression of syncytiotrophoblast markers including *hCG, LGALS13*, and *ERVW-1* [132]. Moreover, knockdown of LIN28A increases the expression of *let-7* miRNAs including *let-7a, let-7c, let-7d, let-7e, let-7g*, and *let-7i* [132], suggesting that differentiation of cells might be due to increased levels of *let-7* miRNAs. Overexpression of LIN28B in HTR8 cells increases cell proliferation, invasion, and migration, whereas knockdown of LIN28B in JEG3 cells reduces cell proliferation [14]. In a recently published study, we further investigated the correlation between LIN28 and *let-7* miRNAs in trophoblast cells using first trimester human trophoblast-derived ACH-3P and Sw.71 cells. ACH-3P cells were generated by fusing first trimester human trophoblast cells with human choriocarcinoma cells, whereas Sw.71 cells were generated by overexpressing human telomerase reverse transcriptase (h-TERT) in first trimester human trophoblast cells [138,139]. These cell lines have contrasting levels of LIN28 and *let-7* miRNAs [11]. ACH-3P cells have high expression of LIN28A and LIN28B whereas these proteins are not detectable in Sw.71 cells [11]. The expression of all *let-7* miRNAs is 50–500-fold higher in Sw.71 cells compared to ACH-3P cells, potentially due to depleted LIN28A and LIN28B in Sw.71 cells which are major suppressors of *let-7* miRNAs [11]. The contrasting levels of LIN28 and *let-7* miRNAs between ACH-3P and Sw.71 cells are potentially due to the difference of methodology used to generate these cell lines. LIN28A knockout in ACH-3P cells increases *let-7a, let-7b, let-7c, let-7d*, and *let-7e*, whereas LIN28B knockout in ACH-3P cells increases *let-7a, let-7b, let-7c, let-7d, let-7e*, and *let-7i* [11]. According to another study, knockdown of LIN28B in ACH-3P cells increases *let-7c, let-7d, let-7e, let-7f*, and *let-7i* [140]. Double knockout of LIN28A and LIN28B in ACH-3P

overexpression causes reduction in all *let-7* miRNAs*.* However, overexpression of both LIN28A and

and *let-7i* [11]*.* According to another study, knockdown of LIN28B in ACH-3P cells increases *let-7c,* 

cells results in increased expression of all *let-7* miRNAs compared to knockout of either LIN28A or LIN28B [11]. Similarly, LIN28A overexpression in Sw.71 cells decreases *let-7d* and *let-7i*, whereas LIN28B overexpression causes reduction in all *let-7* miRNAs. However, overexpression of both LIN28A and LIN28B in Sw.71 cells results in decreased expression of all *let-7* miRNAs compared to overexpression of either LIN28A or LIN28B [11]. These results suggest that LIN28A and LIN28B work in coordination to suppress *let-7* miRNAs and manipulating one paralogue of LIN28 in human trophoblast cells might not induce a similar phenotype compared to if both paralogues are changed.

The majority of *let-7*-regulated genes are associated with cell proliferation, migration, and invasion—processes which are crucial during early human placental development. We recently demonstrated that double knockout of LIN28A and LIN28B in ACH-3P cells increases in *let-7* miRNAs and leads to reduction in expression of proliferation-associated genes including high-mobility group AT-hook 1 (*HMGA1*), MYC protooncogene (*c-MYC*), vascular endothelial growth factor A (*VEGF-A*), and Wnt family member 1 (*WNT1*). LIN28A/B knockout reduces trophoblast cell proliferation and drives them towards differentiating to a syncytiotrophoblast [11,130]. Similarly, double knockin of LIN28A/B in Sw.71 cells leads to reduction in *let-7* miRNAs and increases the expression of *HMGA1*, *c-MYC*, *VEGF-A*, and *WNT1* [11]. Other than its role in cell proliferation, VEGF-A is required at all steps of angiogenesis during placental development [21]. Reduced expression of VEGF-A due to high *let-7* miRNAs can lead to serious pregnancy complications due to impaired angiogenesis in placenta.

Several studies have demonstrated that *let-7* miRNAs bind the 30 -UTR of HMGA2 and reduce its expression in cancer cells [102,141,142]. However, in a recent study, we found a different mechanism of HMGA2 regulation in human trophoblast cells. Double knockout of LIN28A/B in ACH-3P cells increases *let-7* miRNAs but does not change HMGA2 expression [130]. Along with increased *let-7* miRNAs, LIN28A/B double knockout also increases miR-182. The exact mechanism behind increases in miR-182 in LIN28A/B double knockout ACH-3P cells is not clear. We further showed that HMGA2 expression in trophoblast cells is regulated by a transcription-repressing complex comprised of breast cancer susceptibility gene 1 (BRCA1), CtBP-interacting protein (CtIP), and zinc finger protein 350 (ZNF350). This complex, also called BRCA1 repressor complex, binds the promoter region of HMGA2 and inhibits its transcription [130]. In LIN28A/B double knockout ACH-3P cells, high miR-182 targets BRCA1 leading to inhibition of BRCA1 repressor complex and hence increases HMGA2 expression [130]. Therefore, the expected decrease in HMGA2 due to high *let-7* miRNAs is rescued by inhibition of BRCA1 repressor complex. These findings indicate that all genetic pathways demonstrated in the cancer cells might not be applicable in trophoblast cells. It further suggests that rapid proliferation of trophoblast cells during early placental development is more precisely regulated compared to cancer cells.

Although in vitro studies demonstrate the vital role of LIN28-*let-7* miRNA axis in trophoblast function, its role in placental development in vivo is not well understood. Using sheep as an experimental model, we investigated the role of LIN28-*let-7* miRNA axis in trophoblast proliferation in vivo. In sheep, the hatched blastocyst undergoes a phase of trophectoderm elongation before attachment to the uterine epithelium. The conceptus elongation is accomplished by rapid proliferation of trophoblast cells. Trophoblast proliferation is a critical process in early placental development both in humans and sheep. Trophoblast specific knockdown of LIN28A or LIN28B in sheep leads to reduced conceptus elongation due to reduced proliferation of trophoblast cells [143], suggesting that both LIN28A and LIN28B are equally important in early placental development. Knockdown of LIN28A and LIN28B leads to an increase in *let-7* miRNAs and decrease in expression of proliferation-associated genes including insulin like growth factor 2 mRNA binding proteins (IGF2BP1-3), high mobility group AT-hook 1 (*HMGA1*), AT-rich interaction domain 3B (ARID3B), and MYC protooncogene (*c-MYC*) [143]. Additionally, overexpression of LIN28A or LIN28B in immortalized ovine trophoblast cells (iOTR) reduces *let-7* miRNAs, increases the expression of proliferation associated genes, and increases cell proliferation [143]. These findings further strengthen the data from in vitro studies about the role of LIN28-*let-7* miRNA axis in proliferation of human trophoblast cells.
