Abstract
RLIM/Rnf12 is an E3 ubiquitin ligase that has originally been identified as a transcriptional cofactor associated with LIM domain transcription factors. Indeed, this protein modulates transcriptional activities and multiprotein complexes recruited by several classes of transcription factors thereby enhancing or repressing transcription. Around 10 years ago, RLIM/Rnf12 has been identified as a major regulator for the process of X chromosome inactivation (XCI), the transcriptional silencing of one of the two X chromosomes in female mice and ESCs. However, the precise roles of RLIM during XCI have been controversial. Here, we discuss the cellular and developmental functions of RLIM as an E3 ubiquitin ligase and its roles during XCI in conjunction with its target protein Rex1.
Introduction
X dosage compensation in female mice occurs early during embryogenesis in two waves. An early, imprinted form of XCI (iXCI), which silences exclusively the paternally inherited X (Xp), directly follows zygotic genome activation (ZGA) at the end of the 2-cell stage. While this pattern of XCI is maintained in extraembryonic tissues including trophoblast and primitive endoderm, epiblast cells which give rise to the embryo proper reactivate the Xp (XCR) and undergo a random form of XCI (rXCI) around implantation (). The long non-coding (lnc) RNA Xist plays crucial roles during both forms of XCI and paints the X from which it is expressed (; ). Initiation of Xist transcription is considered the onset of XCI and an early phase of continued Xist expression is required for the maintenance of the XCI state (). Prior to upregulation of Xist during rXCI, which is mostly investigated in female ESC models, both X chromosomes transiently move into spatial proximity, a process known as X pairing (; ; ). Xist transcription is inhibited in trans by pluripotency transcription factors including Rex1 () and in cis by the lnc RNA Tsix, which is transcribed antisense to Xist (; ).
The Rlim gene (also known as Rnf12) encodes a RING finger ubiquitin ligase (E3) (). During mouse development Rlim mRNA is widely expressed, while RLIM protein expression is more restricted in cell types and tissues (). In cells, RLIM protein shuttles between the nucleus and cytoplasm in a phosphorylation-dependent manner () but in most cell types, RLIM protein is detected in the nucleus, where many of its substrate proteins reside that include transcription factors and transcriptional co-regulators. Indeed, RLIM is involved in regulating the dynamics of DNA-bound multiprotein complexes in promoters/enhancers
(; ; ). RLIM can self-ubiquitinate and mutations of the RING finger results in gain-of function activities and stabilization of the mutated protein ().
Important in vivo functions of Rlim have been discovered in female mice. In mammary glands of pregnant and lactating females, RLIM serves as a survival factor specifically for milk-producing alveolar cells (, ). Moreover, Rlim/Rnf12 has been identified as a major activator of XCI in female ESCs () and required for iXCI in female mice (). In an ESC model RLIM interacts with Rex1/Zfp42, a transcriptional repressor of Xist, leading to proteasomal degradation and initiation of XCI (). However, Rlim was found dispensable for rXCI in epiblast tissues and in other ESC model systems (, ), and thus over the last years there was much confusion on the roles and importance of Rlim during rXCI. Recent work has identified Rex1 as the critical Rlim target during iXCI and directly compared XCI in various ESC model systems (; ). Results illuminate major roles of Rlim in conjunction with Rex1 during XCI in nuclei of cells, thereby partially clarifying the existing controversy.
RLIM as a RING Finger E3 Ubiquitin Ligase
RLIM/Rnf12 was first identified as an antigen recognized by autologous antibodies of renal cancer patients () and as a cofactor negatively affecting the transcriptional and developmental activity of LIM homeodomain transcription factors (). The gene Rlim maps to the X chromosome and is conserved from humans to chick (). RLIM protein in mice encompasses 600 amino acids (Figure 1A) and contains several conserved domains, including nuclear localization and export sequences (NLS and NES, respectively), a centrally located basic domain (BD) and a C-terminal RING-H2 zinc finger domain. Indeed, both the NLS and the NES are functional and rapid nucleocytoplasmic shuttling of the RLIM protein in a phosphorylation-dependent manner has been demonstrated (), even though in most cell types RLIM protein is detected predominantly in the nuclear compartment. The C-terminal RING H2 zinc finger motif identifies RLIM as an E3 ubiquitin protein ligase. RING finger E3 ligases are part of the ubiquitin proteasome system (UPS) involving an E1 activating enzyme, E2 ubiquitin-conjugating enzymes and E3 ubiquitin ligases which, by recognizing the substrate proteins, confer the specificity of the ubiquitination reaction (). Ubiquitinated proteins are often targeted for degradation by the 26S proteasome but this depends on mono-, multi- or poly-ubiquitination chains and the chain linkage type (; ; ; ). Unlike the cullin subclass of RING finger E3s (), RLIM belongs to a subclass of RING finger E3s in which substrate recognition and ubiquitination ligase functions are mediated by the same protein (), and many substrate proteins of RLIM are recognized and bound by the Basic Domain (Figure 1A). As RLIM locates mostly to the nucleus, it is not surprising that substrates comprise mainly nuclear proteins, including LIM-only (LMO) and CLIM/Ldb cofactors (; ), the histone de-acetylase HDAC2 (), the telomeric protein TRF1 (), the oncoprotein Stathmin (), estrogen receptor alpha (ERalpha) (), the TFIIIB subunit BRF1 () and SMAD7 involved in TGFbeta signaling (). Due to its E3 ligase activity, RLIM regulates cellular levels of transcriptional co-regulators and is involved in the dynamics of multiprotein complexes on DNA in promoters/enhancers (; ; ). Importantly, RLIM also targets Rex1 for proteasomal degradation (), a pluripotency factor expressed in early embryos () that acts as a transcriptional repressor of Xist. RLIM is able to self-ubiquitinate (), a hallmark of RING finger E3s, which likely contributes to its reported short half-life (). Moreover, RLIM targets another RING E3 ligase Mdm2 for degradation () and is the target of the RING E3 ligase SIAH1/2 (). Indeed, as part of regulatory networks, both self-ubiquitination of RING finger E3s and ubiquitination of RING E3s by heterologous E3s frequently occurs in vivo (). These activities likely contribute to the observed differences in the widespread developmental Rlim mRNA expression, and the more restricted RLIM protein expression ().
FIGURE 1
The activity of E3 ligases toward a specific substrate protein is influenced by many factors including the correct cellular location and enzyme stability (). Moreover, E3 activity is regulated via post-translational modifications both at the level of substrate or E3 enzyme (, ). Thus, mutations in many different domains within an E3 enzyme may ultimately influence levels of target proteins. Because instability domains in substrate proteins are generally recognized by multiple E3 ligases, mutations in the RING finger often lead to dominant-negative or gain-of-function effects, inhibiting the actions of the full lengths enzyme and/or other E3s, depending on the cellular context (). Indeed, forced expression of RING finger mutated RLIM in cells inhibits the functions of endogenous RLIM and likely other E3 ligases, leading to dramatically increased cellular levels of target proteins CLIM and HDAC2 (; ), while levels of these proteins in RlimKO cells only increase marginally ().
Not surprisingly, mutations in genes encoding E3 ligases are often associated with human diseases mostly due to mis-regulation of substrate protein levels (), and mutations in Rlim have been associated with intellectual disability and autism (; ). However, while evidence suggests altered neuronal differentiation in mouse male ESCs lacking Rlim/Rnf12 (), CNS development in mice lacking Rlim appears normal and thus, it is unclear at this stage whether human disability phenotypes are caused by the lack of functional RLIM protein, gain-of-function activity acquired by the mutated RLIM or a combination of both. Male mice systemically lacking Rlim are born at Mendelian ratios, appear healthy and grow up to be fertile (, ; ) and even though minor functions cannot be excluded, these results indicate no general, major developmental function of Rlim in male mice. However, during female embryogenesis, major roles of Rlim have been identified in conjunction with Rex1 during X dosage compensation.
Rlim/Rnf12, Rex1 and rXCI
In an attempt to validate a stochastic model of rXCI that postulates the existence of an X-linked activator of XCI (), Rlim/Rnf12 was identified as an activator of XCI (). Using the established F121 female ESC line, a KO ESC line was generated (; ), hereafter referred to as Rnf12KO. Indeed, upon differentiation, Rnf12KO ESCs failed to activate Xist transcription and initiate XCI, and in corresponding heterozygous ESCs rXCI was skewed toward inactivation of the KO allele (; ). Moreover, data from an Rnf12KO derived ESC line indicated that X-pairing was not needed for XCI (). Based on these results a model was proposed in which the dose of RLIM expressed from two alleles was required to trigger rXCI (). However, ESCs lacking Rlim, which were isolated from an independent mouse model containing a conditional Rlim KO allele (referred to as RlimKO) developed Xist clouds and initiated XCI upon differentiation in vitro and in vivo, and in tetraploid complementation experiments these ESCs were able to undergo rXCI and form post-gastrulation embryos (, ). rXCI in the absence of Rlim in mice was supported by targeting the cKO of Rlim specifically in epiblasts via Sox2-Cre. Moreover, RLIM protein levels in WT embryos are downregulated specifically in differentiating epiblast cells at the timepoint of X-reactivation, shortly before the onset of rXCI (). However, in contrast to differentiating epiblast cells, RLIM downregulation does not occur in ESCs differentiated in vitro (). Thus, while Rlim is dispensable for rXCI in vivo and in female ESCs freshly isolated from embryos, it was required in the female F121 ESCs.
The identification of the pluripotency factor Rex1/Zfp42 as RLIM target () has proven a major step in XCI regulation. Indeed, RLIM strongly interacts with and polyubiquitinates Rex1 in female ESCs leading to targeted proteasomal degradation. Rex1 acts as a potent repressor of Xist transcription both by repressing Xist directly in trans and in cis by activating the lncTsix (). Thus, the proteasomal targeting of Rex1 has provided a mechanism for Rlim’s positive action on XCI.
Two recent papers have provided much evidence not only that the dosage of nuclear Rex1 is critical in regulating the XCI process but have also shed light on the ESC controversy. One study generated various CRISPR/Cas9 mediated ESCs carrying an Rlim deletion in a Pgk12.1 () female ESC background (RlimKOp) (). These cells were able to undergo XCI at much higher rates of efficiency when directly compared with Rnf12KO ESCs. It was discovered that the Rnf12KO did not represent a full knockout but that a 45kD truncated RLIM protein consisting of 333 amino acids (referred to as RLIM333) is expressed from the “KO” Rlim locus (). As RLIM333 contains the NLS and a partial BD but lacks the NES and the RING finger (see Figure 1A), this protein can no longer shuttle but is trapped in the nucleus and consistent with the fact that Rex1 interacts with the BD, forced expression of RLIM333 in ESCs (RlimKO or WT) leads to nuclear accumulation of Rex1, indicating gain-of-function activity. Moreover, while Rnf12KO ESCs (Rlim333f) express Rex1 in the nucleus, cell derivatives with a CRISPR/Cas9-mediated deletion of RLIM333 (Rlim0f, in F121 background) are XCI competent at least to some degree and displayed mostly cytoplasmic Rex1 (Supplementary Figure 1). Together, these results indicate a direct connection between nuclear accumulation of Rex1 and the presence of RLIM333 and show that lack of Rlim in ESCs does not necessarily lead to nuclear accumulation of Rex1 (). However, in a newly generated ESC line (in F121) that lacks the entire Rlim ORF (referred to as Rnf12CR–/CR–) and is unable to undergo XCI, XCI activity was restored upon additional deletion of Rex1 (). In contrast to ESCs examined in the first study, however, Rex1 accumulated in nuclei of Rnf12CR–/CR– cells, presumably due to lack of E3 ligase activity (). These results are important as they demonstrate that in Rnf12KO it is Rex1 activity that prevents these cells from undergoing XCI. Thus, in various female ESC lines nuclear Rex1 accumulation is caused by the presence of RLIM333 or the lack of Rlim, while the latter does not affect Rex1 in all ESC lines.
The observed differences in Rex1 distribution in various ESCs models suggests the existence of additional pathways regulating both levels and subcellular localization of Rex1 that might include mechanisms affecting RLIM’s E3 ligase activity and/or differences in cellular competence factor repertoire. In this context it is important to mention that various ESC lines WT for Rlim display very different endogenous levels of Rex1, with Pgk12.1 ESCs expressing low but F121 and the male E14 () lines much higher levels of Rex1, and at least endogenous Rex1 levels and cellular localization in Pgk12.1 are Rlim-independent (). Moreover, ESC culture conditions may play an important role as physiological O2 levels improved XCI efficiency in RlimKO ESC lines (). A powerful Rlim-independent cellular pathway regulating Rex1 is induced upon ESC differentiation, when Rex1 protein rapidly drops to undetectable levels even in the absence of Rlim (; ). Thus, one common feature among various ESC lines lacking Rlim is that nuclear exclusion of Rex1 strongly correlates with their ability to undergo XCI in vitro.
Roles of Rlim/Rnf12 During iXCI
In contrast to rXCI, there is little controversy about the major role Rlim plays during iXCI. Indeed, female mice receiving a maternally transmitted Rlim KO allele (KOm) die () and not a single female receiving a KOm allele has been born in over 10 years of breeding. In contrast, females receiving a paternally transmitted KO allele (KOp) are born at Mendelian ratios and appear to undergo normal XCI. This parent-of-origin effect is a consequence of the fact that iXCI silences exclusively the Xp and the Rlim gene is X-linked. KOm females die around implantation due to defective development of trophoblast tissues caused by iXCI failure ().
Concerning the precise function during iXCI, a crucial role of Rlim in the maintenance of iXCI at blastocyst stages appears clear. This role was genetically established by inducing the Rlim cKO in embryos after ZGA via a paternally transmitted Rosa26-Cre transgene as well as by data obtained from RNA-seq experiments following expression profiles of X-linked genes and Xist transcription in single embryos across pre-implantation stages comparing RlimKO with WT embryos (Figures 2A,B; ). Another mouse model that targets the Rlim gene has recently been generated based on Rnf12KO ESCs. Like RlimKO mice, females receiving a maternal Rnf12KO allele do not undergo iXCI leading to defective trophoblast development (). Strikingly, iXCI is rescued in female mice containing a Rnf12/Rex1 double KO, establishing a genetic link between these two genes during iXCI. This result is exciting as it identifies Rex1 as the critical target of Rlim during iXCI in vivo. Moreover, Rex1 protein accumulates in nuclei of Rnf12KO blastocysts (). Indeed, in WT animals, Rex1 mRNA levels increase dramatically from 4 cell staged embryos to early blastocyst stages (Figure 2C; ), and Rex1 protein is detected in the cytoplasm of cells throughout mouse pre-implantation development, while it is undetectable in nuclei of cells in WT embryos up to morula stages (). These data emphasize the critical role that Rlim plays in the nuclear exclusion of Rex1 during iXCI.
FIGURE 2
As for the role of Rlim in initiating Xist expression and iXCI, the results are less clear-cut. At early pre-implantation stages single embryo RNA-seq data indicate that the process of X dosage compensation initiates at least to some degree in females lacking Rlim but is then interrupted at early blastocyst stages (Figures 2A–C; ), when Rex1 levels increase. Formation of Xist clouds in a significant number of cells up to E2.5 was confirmed by RNA-FISH in RlimKO females. Importantly, the RlimKO embryos were generated using mothers systemically lacking Rlim including in their germline (). Because the profiles of X-linked gene expression in RlimKO embryos mirror those observed in WT females at early pre-implantation stages (Figure 2A), this result argues against influences of maternally transmitted Rex1 which, like Rlim is expressed in oocytes (), on initiation of iXCI in next generation embryos. However, when compared to WT, the number of cells lacking a Xist cloud is diminished in RlimKO females at all timepoints investigated (; ). Therefore, while iXCI initiation is not absolutely dependent on Rlim at least in the C57Bl/6 genetic background, its presence likely promotes this process.
It is important to point out here that in the Rnf12KO system, embryos are expected to express the RLIM333 protein in all cells at pre-implantation stages, but this has not been examined. Because effects of RLIM333 expression and Rlim deletion on Rex1 are partially overlapping, additional repressive contributions of the truncated Rlim protein on the iXCI process cannot be excluded, and iXCI in Rnf12KO animals appears more severely inhibited when compared to RlimKO animals, especially during iXCI initiation (). Thus, while the Rex1KO in Rnf12KO mice removes the effects on XCI induced by both lack of Rlim and the presence of RLIM333, this model cannot distinguish what caused the nuclear accumulation of Rex1 protein.
Discussion
The discovery of Rex1 as the crucial RLIM target during iXCI represents a major finding, assigning critical roles for the interplay of both genes in controlling the process of X dosage compensation in mice, in particular for iXCI and possibly XCR in epiblast cells. A general theme that emerges is that the major role of Rlim during XCI is to regulate nuclear Rex1 levels in order to enable Xist transcription. This is supported by genetic data in mice, Rex1 mRNA and protein expression during pre-implantation development and some data from ESC model systems. Incorporating these data in a previously formulated model () (see also Figure 2D), X dosage compensation in mice might be carried out in a scenario similar to that outlined below: Initiation of Xist transcription at the end of 2-/beginning of 4 cell-stage likely occurs under low levels of embryonic Rex1 (Figure 2C) and at least in part independent of Rlim. As Rex1 mRNA levels increase during pre-implantation development, the function/requirement of Rlim becomes more important to keep Rex1 protein excluded from nuclei of cells to ensure the continuation/maintenance of Xist transcription and the iXCI process. Thus, when Rlim is downregulated specifically in epiblasts of late blastocysts, it is likely that high Rex1 mRNA levels translate into rapid nuclear accumulation of Rex1 protein. Because Rex1 represses Xist but activates Tsix transcription, this spike in nuclear Rex1 likely facilitates the XCR process and/or contributes to an introduction/establishment of Xist regulation in cis via Tsix during rXCI in vivo. However, to ensure rXCI, Rex1 needs to be eliminated from nuclei of differentiating epiblast cells in a Rlim-independent manner, to allow for initiation of Xist transcription. The mechanisms for this are unclear but it is tempting to speculate the involvement of a pathway that is activated upon epiblast differentiation, partially mimicked in ESCs by LIF removal/embryoid body formation. Xist activation and rXCI in differentiating epiblast cells would then occur in the absence of both RLIM and Rex1. Conceptually, such a model shifts the critical dose dependency of the XCI process from Rlim to nuclear Rex1, as during iXCI Rlim is transcribed mostly from a single allele.
Concerning initiation of iXCI, it is important to point out that both RLIM and Rex1 proteins are detectable in the female germline (; , ). However, while initiation of iXCI in embryos occurs at least to some degree in the C57Bl/6 genetic background, this likely depends on maternal Rex1 levels, which might vary in individual females or in females of other genetic backgrounds. Thus, in early female embryos RLIM may serve as a safeguard of iXCI, protecting against maternal Rex1. Indeed, the ability of RLIM to target DNA-bound proteins () is predicted to efficiently counteract repressive functions of both maternally transmitted and embryonic Rex1 on Xist regulatory sequences.
Concerning XCI in various female ESC models, the ability of female ESCs to undergo XCI correlates with low nuclear levels of Rex1. However, nuclear exclusion and overall cellular levels of Rex1 is controlled by RLIM in some ESC lines, but not in others, indicating the existence of alternative pathway(s) that may be sensitive to culture and/or differentiation conditions and possibly also involving multiple mechanisms on RLIM regulation such as E3 ligase activity and differences in the epigenetic make-up selected via clonal line selection. Moreover, various ESC lines might represent different developmental states with possibly leaky suppression of Rlim-independent pathway(s) controlling Rex1, thereby explaining different baseline levels and localization of Rex1 protein in cells. Thus, while RLIM appears the only factor controlling nuclear exclusion of Rex1 at early pre-implantation stages, in ESCs Rlim represents only one of multiple pathways. These data suggest that the dynamics of both Rlim and Rex1 regulation during mouse development is not always paralleled in ESCs in culture, and more work is needed to elucidate the cellular regulation of Rex1 protein.
Concerning RLIM333, the presence of this truncated protein in Rnf12KO-derived systems is based on a cloning artifact. RLIM333 causes nuclear accumulation of Rex1, an effect overlapping with that of the RlimKO and thus, this gain-of-function activity will be eliminated by an additional Rex1KO. However, because RLIM333 lacks specific protein domains, its activity is no longer associated with pathways regulating the full-length protein in terms of expression, stability and E3 ligase activity. In Rnf12KO-derived model systems including mice, there is likely enhanced effects on nuclear Rex1 due to the lack of RLIM and the presence of RLIM333, potentially influencing various aspects of the iXCI process in a temporary, qualitative and quantitative manner, and the effects of RLIM333 on XCI cannot be distinguished from those induced by the lack of Rlim. Outside the Rex1-Rlim axis, RLIM has other E3 ligase targets most of which are also targeted by other E3s. While the KO has little/no effect on such targets, the presence of RLIM333 might, and thus other phenotypes are likely. This is why the reported lethality in male Rnf12KO () is concerning as in RlimKO models including germline RlimKO models (Cre-recombinase negative) no lethality has been observed in males lacking RLIM over more than 10 years. Therefore, data obtained by the Rnf12KO model need to be interpreted not only from the KO perspective, but also from a gain-of-function perspective. Because the terms “KO” or “−/−” are incorrect and cause confusion, derivatives of the Rnf12KO ESCs should be renamed to account for the presence of RLIM333.
In summary, recent results have assigned a crucial importance to the Rlim-Rex1 axis in controlling XCI both in vivo and in vitro. Going forward, it will be important to elucidate the functional evolutionary conservation of this module in other species, the dynamics between both factors specifically in epiblast cells during implantation including effects on Xist/Tsix, and to identify factors that mediate rXCI in vivo.
Statements
Data availability statement
All datasets generated for this study are included in the article/Supplementary Material.
Ethics statement
The animal study was reviewed and approved by UMMS Institute of Animal Care and Usage Committee.
Author contributions
Both authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.
Funding
This work was supported by NIH grant R01GM128168.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcell.2019.00258/full#supplementary-material
References
1
AuguiS.FilionG. J.HuartS.NoraE.GuggiariM.MarescaM.et al (2007). Sensing X chromosome pairs before X inactivation via a novel X-pairing region of the Xic.Science3181632–1636. 10.1126/science.1149420
2
BachI.OstendorffH. P. (2003). Orchestrating nuclear functions: ubiquitin sets the rhythm.Trends Biochem. Sci.28189–195. 10.1016/s0968-0004(03)00055-0
3
BachI.Rodriguez-EstebanC.CarriereC.BhushanA.KronesA.RoseD. W.et al (1999). RLIM inhibits functional activity of LIM homeodomain transcription factors via recruitment of the histone deacetylase complex.Nat. Genet.22394–399. 10.1038/11970
4
BacherC. P.GuggiariM.BrorsB.AuguiS.ClercP.AvnerP.et al (2006). Transient colocalization of X-inactivation centres accompanies the initiation of X inactivation.Nat. Cell Biol.8293–299. 10.1038/ncb1365
5
BarakatT. S.GunhanlarN.PardoC. G.AchameE. M.GhazviniM.BoersR.et al (2011). RNF12 activates Xist and is essential for X chromosome inactivation.PLoS Genet.7:e1002001. 10.1371/journal.pgen.1002001
6
BarakatT. S.LoosF.van StaverenS.MyronovaE.GhazviniM.GrootegoedJ. A.et al (2014). The trans-activator RNF12 and cis-acting elements effectuate X chromosome inactivation independent of x-pairing.Mol. Cell.53965–978. 10.1016/j.molcel.2014.02.006
7
BeckerT.BossenzM.TursunB.SchluterA.PetersM. A.BeckerC. G.et al (2003). Comparing protein stabilities during zebrafish embryogenesis.Methods Cell Sci.2585–89. 10.1023/b:mics.0000006895.03556.9b
8
BrownC. J.HendrichB. D.RupertJ. L.LafreniereR. G.XingY.LawrenceJ.et al (1992). The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus.Cell71527–542. 10.1016/0092-8674(92)90520-m
9
BustosF.Segarra-FasA.ChauguleV. K.BrandenburgL.BraniganE.TothR.et al (2018). RNF12 X-Linked intellectual disability mutations disrupt E3 ligase activity and neural differentiation.Cell Rep.231599–1611. 10.1016/j.celrep.2018.04.022
10
ChenX.ShenJ.LiX.WangX.LongM.LinF.et al (2014). Rlim, an E3 ubiquitin ligase, influences the stability of stathmin protein in human osteosarcoma cells.Cell. Signal.261532–1538. 10.1016/j.cellsig.2014.03.018
11
CiechanoverA.SchwartzA. L. (2004). The ubiquitin system: pathogenesis of human diseases and drug targeting.Biochim. Biophys. Acta16953–17. 10.1016/j.bbamcr.2004.09.018
12
ClimentM.Alonso-MartinS.Perez-PalaciosR.GuallarD.BenitoA. A.LarragaA.et al (2013). Functional analysis of rex1 during preimplantation development.Stem Cells Dev.22459–472. 10.1089/scd.2012.0211
13
GaoK.WangC.JinX.XiaoJ.ZhangE.YangX.et al (2016). RNF12 promotes p53-dependent cell growth suppression and apoptosis by targeting MDM2 for destruction.Cancer Lett.375133–141. 10.1016/j.canlet.2016.02.013
14
GontanC.AchameE. M.DemmersJ.BarakatT. S.RentmeesterE.vanI. W.et al (2012). RNF12 initiates X-chromosome inactivation by targeting REX1 for degradation.Nature485386–390. 10.1038/nature11070
15
GontanC.Mira-BontenbalH.MagarakiA.DupontC.BarakatT. S.RentmeesterE.et al (2018). REX1 is the critical target of RNF12 in imprinted X chromosome inactivation in mice.Nat. Commun.9:4752. 10.1038/s41467-018-07060-w
16
GüngörC.Taniguchi-IshigakiN.MaH.DrungA.TursunB.OstendorffH. P.et al (2007). Proteasomal selection of multiprotein complexes recruited by LIM homeodomain transcription factors.Proc. Natl. Acad. Sci. U.S.A.10415000–15005. 10.1073/pnas.0703738104
17
HerY. R.ChungI. K. (2009). Ubiquitin ligase RLIM modulates telomere length homeostasis through a proteolysis of TRF1.J. Biol. Chem.2848557–8566. 10.1074/jbc.M806702200
18
HickeL. (2001). Protein regulation by monoubiquitin.Nat. Rev. Mol. Cell Biol.2195–201. 10.1038/35056583
19
HirataniI.YamamotoN.MochizukiT.OhmoriS. Y.TairaM. (2003). Selective degradation of excess Ldb1 by Rnf12/RLIM confers proper Ldb1 expression levels and Xlim-1/Ldb1 stoichiometry in Xenopus organizer functions.Development1304161–4175. 10.1242/dev.00621
20
HooperM.HardyK.HandysideA.HunterS.MonkM. (1987). HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells.Nature326292–295. 10.1038/326292a0
21
HuH.HaasS. A.ChellyJ.Van EschH.RaynaudM.de BrouwerA. P.et al (2016). X-exome sequencing of 405 unresolved families identifies seven novel intellectual disability genes.Mol. Psychiatry.21133–48. 10.1038/mp.2014
22
IkedaF.DikicI. (2008). Atypical ubiquitin chains: new molecular signals. protein modifications: beyond the usual suspects’ review series.EMBO Rep.9536–542. 10.1038/embor.2008.93
23
JiaoB.MaH.ShokhirevM. N.DrungA.YangQ.ShinJ.et al (2012). Paternal RLIM/Rnf12 is a survival factor for milk-producing alveolar cells.Cell149630–641. 10.1016/j.cell.2012.02.056
24
JiaoB.Taniguchi-IshigakiN.GungorC.PetersM. A.ChenY. W.RiethdorfS.et al (2013). Functional activity of RLIM/Rnf12 is regulated by phosphorylation-dependent nucleocytoplasmic shuttling.Mol. Biol. Cell243085–3096. 10.1091/mbc.E13-05-0239
25
JohnsenS. A.GungorC.PrenzelT.RiethdorfS.RiethdorfL.Taniguchi-IshigakiN.et al (2009). Regulation of estrogen-dependent transcription by the LIM cofactors CLIM and RLIM in breast cancer.Cancer Res.69128–136. 10.1158/0008-5472.CAN-08-1630
26
JonkersI.BarakatT. S.AchameE. M.MonkhorstK.KenterA.RentmeesterE.et al (2009). RNF12 is an X-Encoded dose-dependent activator of X chromosome inactivation.Cell139999–1011. 10.1016/j.cell.2009.10.034
27
KramerO. H.MullerS.BuchwaldM.ReichardtS.HeinzelT. (2008). Mechanism for ubiquitylation of the leukemia fusion proteins AML1-ETO and PML-RARalpha.FASEB J.221369–1379. 10.1096/fj.06-8050com
28
KramerO. H.ZhuP.OstendorffH. P.GolebiewskiM.TiefenbachJ.PetersM. A.et al (2003). The histone deacetylase inhibitor valproic acid selectively induces proteasomal degradation of HDAC2.EMBO J.223411–3420. 10.1093/emboj/cdg315
29
LeeJ. T. (2000). Disruption of imprinted X inactivation by parent-of-origin effects at tsix.Cell10317–27. 10.1016/s0092-8674(00)00101-x
30
LeeJ. T.DavidowL. S.WarshawskyD. (1999). Tsix, a gene antisense to Xist at the X-inactivation centre.Nat. Genet.21400–404. 10.1038/7734
31
MetzgerM. B.HristovaV. A.WeissmanA. M. (2012). HECT and RING finger families of E3 ubiquitin ligases at a glance.J. Cell Sci.125531–537. 10.1242/jcs.091777
32
MetzgerM. B.PrunedaJ. N.KlevitR. E.WeissmanA. M. (2014). RING-type E3 ligases: master manipulators of E2 ubiquitin-conjugating enzymes and ubiquitination.Biochim. Biophys. Acta184347–60. 10.1016/j.bbamcr.2013.05.026
33
MonkhorstK.JonkersI.RentmeesterE.GrosveldF.GribnauJ. (2008). X inactivation counting and choice is a stochastic process: evidence for involvement of an X-linked activator.Cell132410–421. 10.1016/j.cell.2007.12.036
34
NavarroP.OldfieldA.LegoupiJ.FestucciaN.DuboisA.AttiaM.et al (2010). Molecular coupling of tsix regulation and pluripotency.Nature468457–460. 10.1038/nature09496
35
NorrisD. P.PatelD.KayG. F.PennyG. D.BrockdorffN.SheardownS. A.et al (1994). Evidence that random and imprinted Xist expression is controlled by preemptive methylation.Cell7741–51. 10.1016/0092-8674(94)90233-x
36
OstendorffH. P.BossenzM.MinchevaA.CopelandN. G.GilbertD. J.JenkinsN. A.et al (2000). Functional characterization of the gene encoding RLIM, the corepressor of LIM homeodomain factors.Genomics69120–130. 10.1006/geno.2000.6311
37
OstendorffH. P.PeiranoR. I.PetersM. A.SchluterA.BossenzM.ScheffnerM.et al (2002). Ubiquitination-dependent cofactor exchange on LIM homeodomain transcription factors.Nature41699–103. 10.1038/416099a
38
OstendorffH. P.TursunB.CornilsK.SchluterA.DrungA.GungorC.et al (2006). Dynamic expression of LIM cofactors in the developing mouse neural tube.Dev. Dyn.235786–791. 10.1002/dvdy.20669
39
PayerB. (2016). Developmental regulation of X-chromosome inactivation.Semin. Cell Dev. Biol.5688–99. 10.1016/j.semcdb.2016.04.014
40
PennyG. D.KayG. F.SheardownS. A.RastanS.BrockdorffN. (1996). Requirement for Xist in X chromosome inactivation.Nature379131–137. 10.1038/379131a0
41
PetroskiM. D.DeshaiesR. J. (2005). Function and regulation of cullin-RING ubiquitin ligases.Nat. Rev. Mol. Cell Biol.69–20. 10.1038/nrm1547
42
PickartC. M. (2001). Mechanisms underlying ubiquitination.Annu. Rev. Biochem.70503–533. 10.1146/annurev.biochem.70.1.503
43
ScanlanM. J.GordanJ. D.WilliamsonB.StockertE.BanderN. H.JongeneelV.et al (1999). Antigens recognized by autologous antibody in patients with renal-cell carcinoma.Int. J. Cancer83456–464. 10.1002/(sici)1097-0215(19991112)83:4<456::aid-ijc4>3.0.co;2-5
44
ShinJ.BossenzM.ChungY.MaH.ByronM.Taniguchi-IshigakiN.et al (2010). Maternal Rnf12/RLIM is required for imprinted X-chromosome inactivation in mice.Nature467977–981. 10.1038/nature09457
45
ShinJ.WallingfordM. C.GallantJ.MarchoC.JiaoB.ByronM.et al (2014). RLIM is dispensable for X-chromosome inactivation in the mouse embryonic epiblast.Nature51186–89. 10.1038/nature13286
46
TonneE.HoldhusR.StansbergC.Stray-PedersenA.PetersenK.BrunnerH. G.et al (2015). Syndromic X-linked intellectual disability segregating with a missense variant in RLIM.Eur. J. Hum. Genet.231652–1656. 10.1038/ejhg.2015.30
47
WangF.McCannellK. N.BoskovicA.ZhuX.ShinJ.YuJ.et al (2017). Rlim-dependent and -independent pathways for X chromosome inactivation in female ESCs.Cell Rep.213691–3699. 10.1016/j.celrep.2017.12.004
48
WangF.ShinJ.SheaJ. M.YuJ.BoskovicA.ByronM.et al (2016). Regulation of X-linked gene expression during early mouse development by rlim.eLife5:e19127. 10.7554/eLife.19127
49
WangF.ZhaoK.YuS.XuA.HanW.MeiY. (2019). RNF12 catalyzes BRF1 ubiquitination and regulates RNA polymerase III-dependent transcription.J. Biol. Chem.294130–141. 10.1074/jbc.RA118.004524
50
WeissmanA. M.ShabekN.CiechanoverA. (2011). The predator becomes the prey: regulating the ubiquitin system by ubiquitylation and degradation.Nat. Rev. Mol. Cell Biol.12605–620. 10.1038/nrm3173
51
WutzA.JaenischR. (2000). A shift from reversible to irreversible X inactivation is triggered during ES cell differentiation.Mol. Cell.5695–705. 10.1016/s1097-2765(00)80248-8
52
XuN.TsaiC. L.LeeJ. T. (2006). Transient homologous chromosome pairing marks the onset of X inactivation.Science3111149–1152. 10.1126/science.1122984
53
ZhangL.HuangH.ZhouF.SchimmelJ.PardoC. G.ZhangT.et al (2012). RNF12 controls embryonic stem cell fate and morphogenesis in zebrafish embryos by targeting Smad7 for degradation.Mol. Cell46650–661. 10.1016/j.molcel.2012.04.003
Summary
Keywords
mouse genetics, XCI, RLIM, Rnf12, Rex1, Xist regulation
Citation
Wang F and Bach I (2019) Rlim/Rnf12, Rex1, and X Chromosome Inactivation. Front. Cell Dev. Biol. 7:258. doi: 10.3389/fcell.2019.00258
Received
22 August 2019
Accepted
16 October 2019
Published
31 October 2019
Volume
7 - 2019
Edited by
Celine Morey, UMR 7216 Epigénétique et Destin Cellulaire, France
Reviewed by
Atsushi Fukuda, Tokai University, Japan; Sha Sun, University of California, Irvine, United States
Updates
Copyright
© 2019 Wang and Bach.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Ingolf Bach, ingolf.bach@umassmed.edu
This article was submitted to Developmental Epigenetics, a section of the journal Frontiers in Cell and Developmental Biology
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.