**3. Discussion**

Here, we show several effects of KITD816V on murine placental development. These findings comprise aberrant placental structure and increased differentiation into P-TGC subtypes, whereas other TGC subtypes such as SpA-TGCs, S-TGC, and C-TGCs remain underrepresented. Since TGC variety is essential for placental development, diminished levels of SpA-TGCs may lead to impaired establishment of blood flow at the implantation site by not properly formed blood vessels, lack of dilating spiral arteries, and insufficient secretion of vasodilators and other angiogenic factors [31,37]. C-TGCs contribute to formation of the vessels delivering maternal blood to the labyrinth [39]. Loss of these highly specialized cell types affects invasion, exchange of nutrients and gases, as well as hormone secretion, resulting in decreased functionality of the placenta. Moreover, proliferation in KITD816V placentas ceases at E10.5 when KI-67-positive cells cannot be detected anymore. At that timepoint, the embryo already shows severe growth retardation. Although we cannot exclude additive embryonic effects due to KITD816V expression, we suspect this phenotype to result from malnourishment due to placental failure.

While in other cell types activation of KIT resulted in induction of proliferation and impairment of differentiation [40], trophoblast and placenta development are hallmarked by reduced proliferation in combination with increased and skewed differentiation. In cells of the hematopoietic system, it had previously been shown that KIT signaling results in upregulation of c-Myc, c-Myb, and Gata2 [27,36]. Interestingly, in trophoblast cells, Gata2 binds to and transactivates expression of Pl1. Further, Gata2 level was also shown to be correlated with Plf expression [41,42]. In this study, KITD816V-TSC showed a significant increase in *Gata2* expression upon differentiation. Therefore, we speculate that, also in trophoblast cells, KIT signaling induces Gata2 which in turn transactivates Pl1 and Plf. This would lead to a rapid induction of differentiation, which is accompanied by exiting from the cell cycle. It also explains the high number of PL1+/PLF<sup>+</sup> cells observed in KITD816V placentas. In the context of premature differentiation, it is interesting to note that we were able to generate self-renewing TSC lines. The established TSC did not show any aberrant overall growth parameters and ease of handling. However, the fact that, at day 0 of differentiation, P-ERK and P-AKT levels were comparable to day 3 of regular differentiation together with the already upregulated markers for TGC-differentiation strongly suggests that the growth conditions (FGF4, Heparin, and CM) are able to override the KIT-mediated signals leading to differentiation. This helps to explain the phenotype observed in KITD816V-placentas. During pre- and early post-implantation development, trophectoderm and ectoplacental cone cells rely on FGF4-induced signaling cascades leading to expression of trophectoderm and TSC master regulators Cdx2, Tfap2c, and Elf5 [43–45]. This protects the cells from differentiation inducing the effect of KIT signaling. Post implantation, when the different layers of the placenta are laid down, FGF signaling and expression of the marker genes decline. Since the KIT signaling cascade is already in place and active (as hallmarked by upregulation of *Pl1* and *Plf* at day 0 of TSC differentiation), a premature differentiation is induced, resulting in a smaller and disorganized placenta.

Interestingly, the placental alterations reported here are phenocopied in mouse models of Suppressor of Cytokine Signaling (SOCS) 3 deficiency as well as administration of retinoic acid (RA) to the mother. Administration of RA was demonstrated to result in loss of proliferation, differentiation to TGCs, and a reduced spongiotrophoblast layer [46]. Also, TPBPA levels were reduced whereas PL1 levels were increased [46]. Interestingly, in another study, RA resulted in an increase of invading TGCs [38]. In our model, chronic KIT activation results in significantly more invasive cells. Of note, RA was shown to promote KIT expression and translation in spermatogonia [47–49]. Hence, we speculate that, in trophoblast, KIT acts downstream of RA. In KITD816V mice, RA signaling is active, independent of RA presence (Figure 6).

In Leukemia Inhibitory Factor (LIF)/JAK/STAT3 signaling, Leukemia Inhibitory Factor (LIF) binds to its receptor and Janus Kinases (JAK) are activated. JAK then phosphorylates Signal Transducers and Activators of Transcription (STAT) 3 in the cytoplasm. Activated STAT3 induces the expression of SOCS3, which then represses LIF in a negative feedback loop [50]. Augmentation in LIF levels due to SOCS3 deficiency increases TGC differentiation [51]. SOCS3 deficiency was also reported to result in erythrocytosis and reduced spongiotrophoblast layer. Further, constitutively active STAT3 was observed [52]. We find increased levels of phosphorylated STAT3 in KITD816V-TSC upon differentiation for 9 days in comparison to WT-TSC. STAT3 also plays a role in cell migration and invasion [53], both of which we demonstrate to be affected in the KITD816V mouse model. Taken together, we conclude that KIT is also involved in LIF/JAK/STAT3 signaling (Figure 6). *Int. J. Mol. Sci.* **2020**, *21*, x 12 of 18 reported to result in erythrocytosis and reduced spongiotrophoblast layer. Further, constitutively active STAT3 was observed [52]. We find increased levels of phosphorylated STAT3 in KITD816V-TSC upon differentiation for 9 days in comparison to WT-TSC. STAT3 also plays a role in cell migration and invasion [53], both of which we demonstrate to be affected in the KITD816V mouse model. Taken together, we conclude that KIT is also involved in LIF/JAK/STAT3 signaling (Figure 6). Here, we demonstrate that KITD816V placentas are severely affected by constitutively active KIT signaling. The results show that, in placenta development, KIT signaling is required for the fine-tuned induction of differentiation. Moreover, we speculate that KIT signaling is essential for P-TGC differentiation since this is the major TGC-type observed in the studies.

**Figure 6.** Proposed mechanism of KITD816V signaling in TSC: Schematic summary depicting possible signaling mechanisms that lead to KITD816V placental phenotype observed in this study. Independent of SCF binding, the KITD816V receptor is active. KIT inhibits SOCS3, thereby interrupting the negative feedback loop of Leukemia Inhibitory Factor (LIF) signaling resulting in accumulation of P-STAT3. Further, KIT signaling induces the expression of GATA2, which then transactivates PL1 and PLF. Independent of retinoic acid (RA) presence, KIT impinges on RA-controlled gene expression. Taken together, chronic KITD816V signaling results in severely diminished placental proliferation and reduction of spongiotrophoblast cells but increase in differentiation towards P-TGC and elevated invasion. In the presence of FGF4, Fibroblast Growth Factor Receptor (FGFR) signaling can override KIT-mediated signals. The black arrows indicate relations found in this study, whereas the gray and dashed arrows present hypothetical interactions, some of which have only been demonstrated in other cell types and T bars indicate repression. **Figure 6.** Proposed mechanism of KITD816V signaling in TSC: Schematic summary depicting possible signaling mechanisms that lead to KITD816V placental phenotype observed in this study. Independent of SCF binding, the KITD816V receptor is active. KIT inhibits SOCS3, thereby interrupting the negative feedback loop of Leukemia Inhibitory Factor (LIF) signaling resulting in accumulation of P-STAT3. Further, KIT signaling induces the expression of GATA2, which then transactivates PL1 and PLF. Independent of retinoic acid (RA) presence, KIT impinges on RA-controlled gene expression. Taken together, chronic KITD816V signaling results in severely diminished placental proliferation and reduction of spongiotrophoblast cells but increase in differentiation towards P-TGC and elevated invasion. In the presence of FGF4, Fibroblast Growth Factor Receptor (FGFR) signaling can override KIT-mediated signals. The black arrows indicate relations found in this study, whereas the gray and dashed arrows present hypothetical interactions, some of which have only been demonstrated in other cell types and T bars indicate repression.

Here, we demonstrate that KITD816V placentas are severely affected by constitutively active KIT signaling. The results show that, in placenta development, KIT signaling is required for the fine-tuned induction of differentiation. Moreover, we speculate that KIT signaling is essential for P-TGC differentiation since this is the major TGC-type observed in the studies.
