REST is Required for Nuclear Reprogramming 1 REST is Required for Nuclear Reprogramming by Inhibiting TGFβ Signaling Pathway
Qing-Ran Kong,Bing-Teng Xie,Heng Zhang,Jing-Yu Li,Tian-Qing Huang,Ren-Yue Wei,Zhong-Hua Liu
2016-01-01
Abstract:Differentiated cells can be reprogrammed by transcription factors, and these factors that are responsible for successful reprogramming need to be further identified. Here, we show that the neuronal repressor REST is rich in porcine oocytes and requires for nuclear transfer (NT)-mediated reprogramming through inhibiting TGFβ signaling pathway. REST was dramatically degraded after oocyte activation, but the residual REST was incorporated into the transferred donor nuclei during reprogramming in NT embryos. Inhibition of REST function in oocytes compromised the development of NT embryos but not that of IVF and PA embryos. Bioinformation analysis of putative targets of REST indicated that REST might function on reprogramming in NT embryos by inhibiting TGFβ pathway. Further results showed that the developmental failure of REST-inhibited NT embryos could be rescued by treament of SB431542, an inhibitor of TGF-β pathway. Thus, REST is a newly discovered transcription factor that is required for NT-mediated nuclear reprogramming. Embryonic cells differentiate into all three germ layers of the body as development progresses. Once differentiated, the reversion of the differentiated state to pluripotency is strictly limited in normal development. However, experimentally the differentiated state can be returned to the pluripotent state by transcription factors (1,2). Despite numerous attempts, the factors responsible for successful nuclear reprogramming still need to elucidate. Transcription factors maintaining the pluripotency of embryonic stem cells (ESCs) are called pluripotent factors, and they have an important role in nuclear reprogramming, such as Oct4, Sox2 and Nanog (3,4). Thus, we can identify and characterize reprogramming factors by screening the pluripotent factors. The repressor element 1 (RE1)-silencing transcription factor (REST), as a zinc finger protein, binds 21-bp RE1 sites and functions as a key negative regulator of neurogenesis, so it is also called neuron-restrictive silencer element (NRSE) (5). Recently, REST has http://www.jbc.org/cgi/doi/10.1074/jbc.M116.743849 The latest version is at JBC Papers in Press. Published on November 7, 2016 as Manuscript M116.743849 Copyright 2016 by The American Society for Biochemistry and Molecular Biology, Inc. by gest on N ovem er 2, 2016 hp://w w w .jb.org/ D ow nladed from REST is Required for Nuclear Reprogramming 2 been reported to induce gene expression by recruiting TET3 to the DNA for directed 5hmC generation and NSD3-mediated H3K36 trimethylation in neurons (6). Furthermore, REST has different roles in different cellular contexts, such as oncogenic and tumour-supressor functions and haematopoietic and cardiac differentiation (7,8). In 2008, REST has been proved to maintain self-renewal and pluripotency of mouse ESCs through suppression of microRNAs and believed to be a major pluripotent factor (9,10). However, it has not been elaborated in nuclear reprogramming. Here, we provide evidence that REST plays a unique role in NT-mediated reprogramming as a supressor of TGFβ signaling pathway in pig. RESULTS Expression pattern of REST We first investigated the expression of REST in porcine oocytes, NT and parthenogenetic activation (PA) embryos by real-time PCR and Western blot. Porcine fetal fibroblasts (PFFs) were used as donor cells to construct NT embryos, and REST was observed in the cells by Western blot. Large amounts of REST mRNA and protein were stored in oocytes. After activation, REST mRNA was significantly decreased in NT and PA embryos (p < 0.001) and maintained at a low level from the four-cell to blastocyst stages (Figure 1A), and REST protein was also degraded in oneand two-cell NT and PA embryos (Figure 1B). We performed immunofluorescence analysis to locate REST protein in oocytes and embryos. REST was dispersed in the MII oocyte cytoplasm (Figure 2A and A’, n=17), and was incorporated into transferred donor nuclei in NT embryos when the nuclei were condensed at 2 h post-NT (Figure 2B and B’, n=15) and decondensed at 6 h post-NT (Figure 2C and C’, n=10). In donor cells, REST was also incorporated into the nuclei (Figure 2D and D’). To confirm whether maternal REST could be incorporated into the transferred donor nuclei, hREST-GFP mRNA was injected into oocytes at least 2 h before NT. In control, GFP mRNA was injected and the GFP signals were dispersed in the embryos at 2 h (Figure 2E and E’, n=10) and 6 h (Figure 2G and G’, n=12) post-NT. But in the hREST-GFP mRNA injected embryos GFP signals were obviously observed in the transferred donor nuclei at 2 h (Figure 2F and F’, n=20) and 6 h (Figure 2H and H’, n=16) post-NT. In in vitro fertilization (IVF) and PA embryos, maternal REST was not incorporated into the nuclei when they were condensed (Figure 2I and I’, n=12; K and K’, n=14) and incorporated with the nuclei when they were decondensed (Figure 2J and J’, n=14; L and L ’ , n=17). These results demonstrate that maternal REST is incorporated into transferred donor nuclei, suggesting it may function in the process of NT-mediated nuclear reprogramming. Inhibition of REST in NT embryos To test the role of REST in nuclear reprogramming, its function was inhibited by injection of anti-REST antibody into MII oocytes at least 2 h before NT, PA, IVF, and intracytoplasmic sperm injection (ICSI). The successful injection of the antibody used here was verified by immunofluorescence analysis (Figure 3A and A’, n=15; B and B’, n=15). In addition, no or weak signal was detected by immunostaining of the anti-REST antibody injected oocytes at 2 h post-injection (Figure 3C and C’, n=16), indicating that the injected antibody had been degraded and would not affect donor cell-derived and zygotic REST. by gest on N ovem er 2, 2016 hp://w w w .jb.org/ D ow nladed from REST is Required for Nuclear Reprogramming 3 And, effectively match of anti-REST antibody to the porcine REST among whole oocyte proteins was verified by Western blot (Figure 3D). Then the in vitro developmental competency of porcine NT embryos, regarded as a stringent test of reprogramming efficiency, was examined. The rates of cleavage and cell numbers of blastocyst showed no significant difference among NT embryos with no injection (Con-NT), anti-REST antibody injection (anti-REST-NT), and IgG (IgG-NT) injection, but the proportion of anti-REST-NT embryos that developed to blastocysts was significantly lower than that of IgG-NT and Con-NT embryos (7.27 versus 20.8 and 21.53%, respectively; p < 0.05; Table 1). And more anti-REST-NT embryos arrested at the twoor four-cell stage in comparison with IgG-NT and Con-NT embryos (54.39 versus 30.33 and 28.09%, respectively; p < 0.05; Table 1). To further confirm the results, REST specific locked nucleic acid (REST-LNA) was injected into oocytes at 33h of in vitro maturation (IVM). Q-PCR and Western blot showed REST mRNA and protein were effectively reduced in oocytes at 42h of IVM by REST-LNA injection (p < 0.001; Figure 3E and F). Consistent with anti-REST antibody, the proportion of NT embryos developed to the blastocyst stage was significantly decreased in REST-LNA injection group (3.06 versus 17.46 and 15.37%; p < 0.05; Table 2). We also overexpressed REST by injection of hREST-GFP mRNA into oocytes, and it had no significant effect on development of NT embryos. In contrast to that in NT embryos, injections of anti-REST antibody and hREST-GFP mRNA did not affect embryonic development of IVF, ICSI and PA embryos (Table 1). Therefore, we suggest that REST is required for successful nuclear reprogramming. Inhibition of REST upregulates TGFβ signaling pathway in NT embryos REST binds RE1 sites to repress gene expression throughout the body (11). To determine how maternal REST regulates NT-mediated nuclear reprogramming, we searched REST targeted genes in pig. A consensus RE1 based on the sequences of 32 known RE1 elements NT(T/C)AG(A/C)(A/G)CCNN(A/G)G(A/C)( G/S)AG was used to screen porcine UCSC genome sequence database (susScr3) by using a PERL script (11). The number of putative RE1s identified in the porcine genome was 1,662, and there are 324 genes that have RE1s within 10kb from their transcriptional start site (TSS; Supplementary Table S1). Pathway analysis showed that REST putative targeted genes widely involved in TGFβ signaling pathway (p < 0.001, FDR = 0.1642; Figure 4A). We therefore decided to examine expression of several key genes of TGFβ pathway in NT embryos. By RT-PCR analysis, the expressions of TGFβR1, ACVR2A, ACVR2B, Smad2 and Smad3 were not detected in oocytes and high in donor cell PFFs. In one-cell NT embryos, the expressions of these genes were at relatively low levels but dramatically upregulated after REST deficiency (Figure 4B). Consistent with RT-PCR results, Q-PCR showed that REST inhibition significantly enhanced the expressions in NT embryos from the one to four-cell stages (Figure 4C; p < 0.001). Moreover, we observed increase of Smad3 and phosphorylated Smad3 (Smad3-p) expressions in two-cell REST-deficient NT embryos by Western blot and immunofluorescence analysis (Figure 4D and E). The results show that inhibition of REST upregulates TGFβ signaling pathway in NT by gest on N ovem er 2, 2016 hp://w w w .jb.org/ D ow nladed from REST is Required for Nuclear Reprogramming 4 embryos. Upregulation of TGFβ pathway by inhibition of REST is involved in reprogramming failure To figure out whether upregulation of TGFβ pathway in REST-deficient NT embryos leads to the failure of NT-mediated nuclear reprogramming, a specific TGFβ pathway inhibitor, SB431542, was used. We found 0.1 μM SB431542 treatment for 12 h post-activation had no negative effect on development of PA embryos (Table 3), so NT embryos were treated as the method. SB431542 treatment could dramatically decrease Smad3 and Smad3-p in normal and REST-deficient NT embryos detected by Western blot and immunofluorescence analysis (Figure 4D and E), indicating TGFβ pathwa