Molecular insights into RBR E3 ligase ubiquitin transfer mechanisms

RING‐in‐between‐RING (RBR) ubiquitin (Ub) ligases are a distinct class of E3s, defined by a RING1 domain that binds E2 Ub‐conjugating enzyme and a RING2 domain that contains an active site cysteine similar to HECT‐type E3s. Proposed to function as RING/HECT hybrids, details regarding the Ub transfer mechanism used by RBRs have yet to be defined. When paired with RING‐type E3s, E2s perform the final step of Ub ligation to a substrate. In contrast, when paired with RBR E3s, E2s must transfer Ub onto the E3 to generate a E3~Ub intermediate. We show that RBRs utilize two strategies to ensure transfer of Ub from the E2 onto the E3 active site. First, RING1 domains of HHARI and RNF144 promote open E2~Ubs. Second, we identify a Ub‐binding site on HHARI RING2 important for its recruitment to RING1‐bound E2~Ub. Mutations that ablate Ub binding to HHARI RING2 also decrease RBR ligase activity, consistent with RING2 recruitment being a critical step for the RBR Ub transfer mechanism. Finally, we demonstrate that the mechanism defined here is utilized by a variety of RBRs.     

Maternal Smad3 deficiency compromises decidualization in mice

Transforming growth factor (TGF)‐β and activin, members of TGF‐β superfamily, are abundantly expressed in the endometrium and regulate decidualization of endometrial stroma. Smad2 and Smad3 are receptor‐regulated Smads (R‐Smads) that transduce extracellular TGF‐β/activin/Nodal signaling. In situ hybridization results showed that Smad3 was highly expressed in the decidual zone during the peri‐implantation period in mice. By using artificial decidualization, we found that Smad3 null mice showed partially compromised decidualization. We therefore hypothesized that Smad2 might compensate for the function of Smad3 during the process of decidualization. Smad2 was also highly expressed in the decidual zone and phosphorylated Smad2 was much more abundantly increased in the deciduoma of Smad3 null mice than for wild‐type (WT) mice. We further employed an in vitro uterine stromal cell decidualization model, and found that decidual prolactin‐related protein (dPRP) and cyclin D3, which are well‐known markers for decidual cells, were significantly down‐regulated in Smad3 null decidual cells, and were much more significantly reduced when the expression of Smad2 was simultaneously silenced by its siRNA (P < 0.05). However, the expression levels of dPRP and cyclin D3 remained the same when Smad2 was silenced in WT decidual cells. Collectively, these findings provide evidence for an important role of Smad3 in decidualization and suggest that Smad2 and Smad3 may have redundant roles in decidualization. J. Cell. Biochem. 113: 3266–3275, 2012. © 2012 Wiley Periodicals, Inc.     

Structure of the human Parkin ligase domain in an autoinhibited state

Mutations in the protein Parkin are associated with Parkinson's disease (PD), the second most common neurodegenerative disease in men. Parkin is an E3 ubiquitin (Ub) ligase of the structurally uncharacterized RING‐in‐between‐RING(IBR)‐RING (RBR) family, which, in an HECT‐like fashion, forms a catalytic thioester intermediate with Ub. We here report the crystal structure of human Parkin spanning the Unique Parkin domain (UPD, also annotated as RING0) and RBR domains, revealing a tightly packed structure with unanticipated domain interfaces. The UPD adopts a novel elongated Zn‐binding fold, while RING2 resembles an IBR domain. Two key interactions keep Parkin in an autoinhibited conformation. A linker that connects the IBR with the RING2 over a 50‐Å distance blocks the conserved E2～Ub binding site of RING1. RING2 forms a hydrophobic interface with the UPD, burying the catalytic Cys431, which is part of a conserved catalytic triad. Opening of intra‐domain interfaces activates Parkin, and enables Ub‐based suicide probes to modify Cys431. The structure further reveals a putative phospho‐peptide docking site in the UPD, and explains many PD‐causing mutations.     

TGF‐β Stimulates Expression of Chondroitin Polymerizing Factor in Nucleus Pulposus Cells Through the Smad3, RhoA/ROCK1, and MAPK Signaling Pathways

The enzyme chondroitin polymerizing factor (ChPF) is primarily involved in extension of the chondroitin sulfate backbone required for the synthesis of sulfated glycosaminoglycan (sGAG). Transforming growth factor beta (TGF‐β) upregulates sGAG synthesis in nucleus pulposus cells; however, the mechanisms mediating this induction are incompletely understood. Our study demonstrated that ChPF expression was negatively correlated with the grade of degenerative intervertebral disc disease. Treatment of nucleus pulposus cells with TGF‐β induced ChPF expression and enhanced Smad2/3, RhoA/ROCK activation, and the JNK, p38, and ERK1/2 MAPK signaling pathways. Selective inhibitors of Smad2/3, RhoA or ROCK1/2, and knockdown of Smad3 and ROCK1 attenuated ChPF expression and sGAG synthesis induced by TGF‐β. In addition, we showed that RhoA/ROCK1 signaling upregulated ChPF via activation of the JNK pathway but not the p38 and ERK1/2 signaling pathways. Moreover, inhibitors of JNK, p38 and ERK1/2 activity also blocked ChPF expression and sGAG synthesis induced by TGF‐β in a Smad3‐independent manner. Collectively, our data suggest that TGF‐β stimulated the expression of ChPF and sGAG synthesis in nucleus pulposus cells through Smad3, RhoA/ROCK1 and the three MAPK signaling pathways. J. Cell. Biochem. 119: 566–579, 2018. © 2017 Wiley Periodicals, Inc.     

Structure of LNX1:Ubc13 ~ Ubiquitin Complex Reveals the Role of Additional Motifs for the E3 Ligase Activity of LNX1

LNX1 (ligand of numb protein-X1) is a RING and PDZ domain-containing E3 ubiquitin ligase that ubiquitinates human c-Src kinase. Here, we report the identification and structure of the ubiquitination domain of LNX1, the identification of Ubc13/Ube2V2 as a functional E2 in vitro, and the structural and functional studies of the Ubc13~Ub intermediate in complex with the ubiquitination domain of LNX1. The RING domain of LNX1 is embedded between two zinc-finger motifs (Zn-RING-Zn), both of which are crucial for its ubiquitination activity. In the heterodimeric complex, the ubiquitin of one monomer shares more buried surface area with LNX1 of the other monomer and these interactions are unique and essential for catalysis. This study reveals how the LNX1 RING domain is structurally and mechanistically dependent on other motifs for its E3 ligase activity, and describes how dimeric LNX1 recruits ubiquitin-loaded Ubc13 for Ub transfer via E3 ligase-mediated catalysis.     

Role of Smad3, acting independently of transforming growth factor‐β, in the early induction of Wnt‐β‐catenin signaling by parathyroid hormone in mouse osteoblastic cells

Parathyroid hormone (PTH) exerts an anabolic action on bone but the mechanisms are incompletely understood. We showed previously that PTH interacts with the canonical Wnt‐β‐catenin signaling pathway via the transforming growth factor (TGF)‐β signaling molecule, Smad3, to modulate osteoblast differentiation and apoptosis. Here, we examined which actions of Smad3 are TGF‐β‐independent in stimulating the osteoblast phenotype and PTH‐induced Wnt‐β‐catenin signaling. For this, the TGF‐β receptor type 1 [activin receptor‐like kinase (ALK5)] inhibitor (SB431542), and a Smad3 mutant in which the site normally phosphorylated by ALK5 is mutated from SSVS to AAVA, was used. PTH induced total β‐catenin and reduced phosphorylated β‐catenin levels at 1, 6, and 24 h in mouse osteoblastic MC3T3‐E1 cells. Transient transfection of Smad3AAVA inhibited the PTH induction of total β‐catenin and reduction of phosphorylated β‐catenin levels at 6 and 24 h, but not at 1 h, indicating that the early effects occur independently of TGF‐β receptor signaling. On the other hand, MC3T3‐E1 cell clones in which Smad3AAVA was stably expressed demonstrated elevated β‐catenin levels, although alkaline phosphatase (ALP) activity and mineralization were unaltered. In contrast, MC3T3‐E1 cell clones in which wild‐type Smad3 was stably expressed exhibited increased ALP activity and mineralization that were decreased by the ALK5 inhibitor, SB431542, although the β‐catenin levels induced in these cells were not modulated. In conclusion, the present study indicates that PTH induces osteoblast β‐catenin levels via Smad3 independently of, and dependently on, TGF‐β in the early and later induction phases, respectively. J. Cell. Biochem. 108: 285–294, 2009. © 2009 Wiley‐Liss, Inc.     

Enhancement of BRCA1 E3 ubiquitin ligase activity through direct interaction with the BARD1 protein

The breast and ovarian cancer-specific tumor suppressor RING finger protein BRCA1 has been identified as an E3 ubiquitin (Ub) ligase through in vitro studies, which demonstrated that its RING finger domain can autoubiquitylate and monoubiquitylate histone H2A when supplied with Ub, E1, and UBC4 (E2). Here we report that the E3 ligase activity of the N-terminal 110 amino acid residues of BRCA1, which encodes a stable domain containing the RING finger, as well as that of the full-length BRCA1, was significantly enhanced by the BARD1 protein (residues 8-142), whose RING finger domain itself lacked Ub ligase activity in vitro. The results of mutagenesis studies indicate that the enhancement of BRCA1 E3 ligase activity by BARD1 depends on direct interaction between the two proteins. Using K48A and K63A Ub mutants, we found that BARD1 stimulated the formation of both Lys(48)- and Lys(63)-linked poly-Ub chains. However, the enhancement of BRCA1 autoubiquitylation by BARD1 mostly resulted in poly-Ub chains linked through Lys(63), which could potentially activate biological pathways other than BRCA1 degradation. We also found that co-expression of BRCA1 and BARD1 in living cells increased the abundance and stability of both proteins and that this depended on their ability to heterodimerize.     

Ubiquitin‐Activated Interaction Traps (UBAITs) identify E3 ligase binding partners

We describe a new class of reagents for identifying substrates, adaptors, and regulators of HECT and RING E3s. UBAITs (Ub iquitin‐A ctivated I nteraction T raps) are E3‐ubiquitin fusion proteins and, in an E1‐ and E2‐dependent manner, the C‐terminal ubiquitin moiety forms an amide linkage to proteins that interact with the E3, enabling covalent co‐purification of the E3 with partner proteins. We designed UBAITs for both HECT (Rsp5, Itch) and RING (Psh1, RNF126, RNF168) E3s. For HECT E3s, trapping of interacting proteins occurred in vitro either through an E3 thioester‐linked lariat intermediate or through an E2 thioester intermediate, and both WT and active‐site mutant UBAITs trapped known interacting proteins in yeast and human cells. Yeast Psh1 and human RNF126 and RNF168 UBAITs also trapped known interacting proteins when expressed in cells. Human RNF168 is a key mediator of ubiquitin signaling that promotes DNA double‐strand break repair. Using the RNF168 UBAIT, we identify H2AZ—a histone protein involved in DNA repair—as a new target of this E3 ligase. These results demonstrate that UBAITs represent powerful tools for profiling a wide range of ubiquitin ligases.     

RING‐between‐RINGs—keeping the safety on loaded guns

EMBO J (2012) 31 19, 3833–3844 doi:10.1038/emboj.2012.217; published online September072012EMBO Rep (2012) 13 9, 840–846 doi:10.1038/embor.2012.105; published online September072012The ‘RING-between-RING''-type E3 ubiquitin ligase HOIP acts via a novel RING/HECT-hybrid ubiquitin transfer mechanism and catalyses the formation of linear ubiquitin chains by non-covalently binding the acceptor ubiquitin. But in the absence of a binding partner, HOIP is auto-inhibited. This explains why assembly of either HOIP/HOIL-1L or HOIP/SHARPIN is required to catalyse linear chain formation.Post-translational modification of a protein with Ubiquitin (Ub) requires the activity of three enzymes: a Ub activating enzyme (E1), a Ub conjugating enzyme (E2), and a Ub ligase (E3). Final Ub transfer is performed by an E3 enzyme, which mediates the ligation of Ub from an E2∼Ub conjugate (‘∼'' denotes a thioester) onto a substrate. E3s are commonly divided into two mechanistic classes: RING/U-box E3s and HECT E3s. RING/U-box E3s facilitate the transfer of Ub from the E2∼Ub directly onto a substrate amino group. In contrast, HECTs transfer Ub from the E2∼Ub to the substrate via a HECT∼Ub intermediate. This mechanistic difference leads to an important distinction regarding what determines the type of Ub product (i.e., the specific Ub-chain linkage) formed: in ubiquitination pathways involving RING-type E3 ligases, the E2 determines the product formed, whereas for HECT-catalysed pathways, the E3 governs product formation (Christensen et al, 2007; Kim and Huibregtse, 2009).RING-between-RING (RBR) E3s comprise a class of E3s that appear to have special properties. Although RBR E3s have been considered as a subfamily of RING E3s, the RBR E3 HHARI (Human Homologue of ARIadne) was recently shown to form a HECT-like E3∼Ub intermediate (Wenzel et al, 2011). Two other members of the RBR family, HOIL-1 and HOIP, form the Linear Ub Chain Assembly Complex (LUBAC), the only E3 ligase known to catalyse the synthesis of linear Ub chains (Kirisako et al, 2006). Linear Ub chains are produced by head-to-tail conjugation of Ub molecules through their N- and C-termini and have been shown to activate the canonical NF-κB pathway (Tokunaga et al, 2009).Two studies by the Rittinger and Sixma groups now reveal important insights regarding the formation of linear Ub chains by the dimeric RBR E3 complex HOIP/HOIL-1L (Smit et al, 2012; Stieglitz et al, 2012). Results from these studies highlight three emerging themes among RBR ligases: a RING/HECT-hybrid Ub transfer mechanism; auto-inhibition of RBR E3 activity, and a role for E3:Ub interactions.The RBR E3 ligase domain consists of two distinct RING domains, called RING1 and RING2, connected by an IBR (In-Between-Ring) domain. Despite its name, RING2 is not a canonical RING domain as it contains an active site Cysteine (Cys), which has recently been shown to form a thioester E3∼Ub intermediate, as directly detected for the RBR E3 HHARI. Although the Ub-loaded species could not be detected for the RBR E3 parkin, mutation of the analogous cysteine residue abrogated parkin''s ligase activity implying that it works via the same mechanism. On the basis of these observations, Wenzel et al (2011) proposed that the RBR E3s are a family of RING/HECT hybrids that use RING1 to bind an E2 (RING-like) and RING2 to present the active site Cys (HECT-like) as shown schematically in Figure 1. Both Smit et al (2012) and Stieglitz et al (2012) observed a HOIP∼Ub thioester, confirming that HOIP also acts via a RING/HECT-hybrid mechanism. Furthermore, Smit et al (2012) used a clever strategy to uncouple the first transfer event (E2∼Ub to E3) from the final transfer event (E3∼Ub to substrate Ub) to verify that the E3∼Ub intermediate is a prerequisite for Ub transfer onto a substrate and not just a serendipitous side product. The results extend the number of RBR E3s for which a thioester intermediate has been observed and support the notion that RBR E3s are indeed RING/HECT hybrids.Open in a separate windowFigure 1Three common themes are emerging among RBR ligases: a RING/HECT-hybrid Ub transfer mechanism; auto-inhibition of RBR E3 activity, and a role for E3:Ub interactions. RBR E3s are characterized by their RBR domain that consists of two distinct RING domains, RING1 that binds the E2, and RING2 that harbours the active site Cys. Two new studies on the RBR E3 HOIP show that (a) domain(s) in HOIP''s N-terminal region inhibits its ligase activity and (b) a domain C-terminal to HOIP''s RBR binds and orients an acceptor Ub to direct linear Ub-chain formation (‘Linear Ub chain Determining Domain'' or LDD). (A)Three ways in which auto-inhibition might occur are illustrated: (1) inhibition of E2∼Ub binding by RING1, (2) obstruction of the active site cysteine on RING2, and/or (3) occlusion of acceptor Ub binding on the LDD. (B) A possible flow of events that occur once auto-inhibition released is shown. Details of each step and how specifically auto-inhibition is released are still unknown.Previous studies have established that HOIP Ub ligase activity and subsequent activation of NF-κB require either the RBR-containing protein, HOIL-1L, or SHARPIN, an adaptor protein associated with LUBAC (Ikeda et al, 2011; Tokunaga et al, 2011). The two current studies now show that although full-length HOIP exhibits very low activity on its own, removal of the N-terminal ∼700 residues results in robust ligase activity. Thus, HOIP appears to be auto-inhibited in the absence of a binding partner. Further analysis revealed that HOIP''s UBA (Ub-Associated) domain is partly responsible for auto-inhibition, although additional N-terminal domains appear to have auto-inhibitory effects as well. SHARPIN, which contains a UBL (Ub-Like) domain, can relieve auto-inhibition of HOIP. Similarly, the addition of the HOIL-1L UBL domain, previously shown to interact with the HOIP UBA domain (Yagi et al, 2012), relieves inhibition. Interestingly, the addition of full-length HOIL-1L results in even greater ubiquitination activity.Stieglitz et al (2012) show that the RBR E3 HOIL-1L has very low E3 activity on its own. Intriguingly, they found that mutation of the HOIL-1L RING2 active site Cys (C460A) reduced activity of the HOIP/HOIL-1L complex back to levels comparable to HOIP activity in presence of HOIL UBL alone. This suggests a more active, catalytic role for HOIL-1L in linear Ub-chain formation than previously appreciated. The details regarding this role must await further studies, but involvement of an active site Cys residue on a second RING2 domain suggests a possible reciprocal transfer mechanism. Perhaps linear chains can be pre-built via such a mechanism and passed en bloc to substrate, similarly to mechanisms used by some HECT-type bacterial E3 ligases (Levin et al, 2010).Parkin, another RBR E3, also exhibits auto-inhibition (Chaugule et al, 2011), but the auto-inhibitory mechanism and the release thereof differ from HOIP. Unlike parkin''s N-terminal UBL, which is thought to interact within the RBR domain at RING2, HOIP''s UBA does not bind detectably in trans to any region in the RBR domain (Stieglitz et al, 2012). Furthermore, addition of its UBA in trans does not inhibit the activity of HOIP RBR E3 as was seen with parkin and its UBL domain. The auto-inhibition of parkin is likely released by substrate binding, because addition of either the UIM of Eps15 or the SH3 domain of endophilin-A, both known to bind the parkin UBL, can restore the activity of parkin (Chaugule et al, 2011). In addition, phosphorylation of Ser65 within the UBL of parkin by PINK-1 activates parkin, presumably by releasing the UBL from RING2 (Kondapalli et al, 2012). In contrast, HOIP overcomes its auto-inhibition through binding either HOIL-1L or SHARPIN. There is no additive effect when both binding partners are present, consistent with the notion that both proteins act via their UBL domains, although this remains to be demonstrated for SHARPIN. The activity of either SHARPIN/HOIP or HOIL-1L/HOIP can activate NF-κB (Ikeda et al, 2011; Tokunaga et al, 2011), but how the protein complexes differ in their cellular roles remains to be further analysed.The finding that HOIP and parkin exhibit auto-inhibition raises the question whether there is something special about the RBR E3s that require auto-inhibition. In this regard, we note that RBR E3s bind the E2 UbcH7 with significantly tighter affinity than canonical RING E3s bind their E2s (Dove and Klevit, unpublished). In the absence of a substrate, RING1 loaded with UbcH7∼Ub would lead to non-productive transfer of Ub from UbcH7∼Ub to the active site of RING2. Occlusion of the active site by auto-inhibition may therefore act as a safety check until its activity is required for transfer of Ub to a substrate. As yet, there is no evidence to indicate whether substrate binding will release HOIP auto-inhibition, as it does for parkin, but this remains a possibility.The revelation that removal of all domains N-terminal to the HOIP RING1 domain yields a highly active ligase allowed both groups to explore questions pertaining to how linear chains are built. Remarkably, constructs comprised of only the RBR domain through the C-terminus of HOIP are sufficient to specify linear Ub chains. (The two groups use HOIP constructs that differ by only two N-terminal residues (697/699–1072) but Stieglitz et al call their construct RBR whereas Smit et al call it RBR-LDD.) (Smit et al, 2012; Stieglitz et al, 2012). Smit et al (2012) demonstrate that the region immediately C-terminal to RING2 is required for linear chain building activity and name the region the ‘LDD'' (Linear Ub chain Determining Domain). Their results indicate that the LDD binds and orients the acceptor Ub to promote transfer of the donor Ub from the RING2 active site to the N-terminus of the acceptor Ub (Figure 1). Parkin has also been suggested to bind free Ub. Details about whether parkin binds acceptor or donor Ub and whether Ub binding determines Ub-chain specificity are still unknown.There is precedence for acceptor Ub binding by HECT E3s and this interaction is essential for chain formation by NEDD4 and its yeast orthologue Rsp5 (Kim et al, 2011; Maspero et al, 2011). In another example, the inactive E2 variant MMS2 binds an acceptor Ub and orients the Ub-Lys63 into the active site of Ubc13 thereby guaranteeing K63-linked chain formation by the E2 (Eddins et al, 2006). Besides proper orientation of the acceptor Ub, chemical differences between α- and ɛ-amino groups likely contribute to linear Ub-chain specificity. For example, E2s known to be active with RING-type E3s can transfer Ub onto the amino acid lysine, but not the other amino acids containing α-amino groups indicating specificity towards the ɛ-amino of lysine (Wenzel et al, 2011).Catalysed by the unexpected discovery that HHARI is a HECT/RING hybrid E3, details about how the RBR class of E3s function are beginning to emerge. We now know, either directly or indirectly, that at least 4 RBR E3s of the 13 identified in humans (HHARI, HOIL, HOIP, and parkin) require a trans-thiolation event using an active site cysteine within RING2. Conservation of this cysteine among all RBR E3s strongly suggests that the RING/HECT-hybrid mechanism is conserved and therefore defines the class. The hybrid mechanism also offers an explanation for the heretofore puzzling observation that, despite being categorized as a RING E3, HOIP determines the type of Ub chain formed. The ability to bind an acceptor Ub close to the RING2 active site likely contributes to how the RBR E3s dictate the type of product they produce. Finally, both HOIP and parkin are auto-inhibited. It remains to be seen whether HOIP''s auto-inhibitory domains work via inhibition of E2∼Ub binding by RING1, obstruction of the active site cysteine on RING2, and/or occlusion of acceptor Ub binding on the LDD (Figure 1). Regardless of the mechanistic details, the ability to modulate their activity may be a common trait of the RBR E3s. Given recent rapid progress, our understanding of this special class of E3s will continue to grow apace.     

High yield expression and NMR characterization of Arkadia E3 ubiquitin ligase RING-H2 finger domain

E3 ubiquitin ligases play a key role in the recognition of target proteins and the degradation by 26S proteasomes. Arkadia is the first example of an E3 ubiquitin ligase that positively regulates TGF-β family signaling. It has been shown to induce ubiquitin-dependent degradation of negative regulators of TGF-β signaling through its C-terminal RING domain. Structural analysis of Arkadia RING domain is needed to elucidate its enzymatic properties. For such studies efficient production of pure and correctly folded Arkadia protein is required. Here we report the recombinant expression in Escherichia coli and purification of the C-terminal RING domain of Arkadia. NMR analysis of the soluble construct reveals a stable folded protein suitable for high resolution structural studies.     

Quantitative phosphoproteomics of transforming growth factor-β signaling in colon cancer cells

The transforming growth factor‐β (TGF‐β) signaling pathway progresses through a series of protein phosphorylation regulated steps. Smad4 is a key mediator of the classical TGF‐β signaling pathway; however, reports suggest that TGF‐β can activate other cellular pathways independent of Smad4. By investigating the TGF‐β‐regulated phosphoproteome, we aimed to uncover new functions controlled by TGF‐β. We applied titanium dioxide to enrich phosphopeptides from stable isotope labeling with amino acids in cell culture (SILAC)‐labeled SW480 cells stably expressing Smad4 and profiled them by mass spectrometry. TGF‐β stimulation for 30 min resulted in the induction of 17 phosphopeptides and the repression of 8 from a total of 149 unique phosphopeptides. Proteins previously not known to be phosphorylated by TGF‐β including programmed cell death protein 4, nuclear ubiquitous casein and cyclin‐dependent kinases substrate, hepatoma‐derived growth factor and cell division kinases amongst others were induced following TGF‐β stimulation, while the phosphorylation of TRAF2 and NCK‐interacting protein kinase are examples of proteins whose phosphorylation status was repressed. This phosphoproteomic screen has identified new TGF‐β‐modulated phosphorylation responses in colon carcinoma cells.     

RIPK4 suppresses the TGF‐β1 signaling pathway in HaCaT cells

Receptor‐interacting serine/threonine kinase 4 (RIPK4) and transforming growth factor‐β 1 (TGF‐β1) play critical roles in the development and maintenance of the epidermis. A negative correlation between the expression patterns of RIPK4 and TGF‐β signaling during epidermal homeostasis‐related events and suppression of RIPK4 expression by TGF‐β1 in keratinocyte cell lines suggest the presence of a negative regulatory loop between the two factors. So far, RIPK4 has been shown to regulate nuclear factor‐κB (NF‐κB), protein kinase C (PKC), wingless‐type MMTV integration site family (Wnt), and (mitogen‐activated protein kinase) MAPK signaling pathways. In this study, we examined the effect of RIPK4 on the canonical Smad‐mediated TGF‐β1 signaling pathway by using the immortalized human keratinocyte HaCaT cell line. According to our results, RIPK4 inhibits intracellular Smad‐mediated TGF‐β1 signaling events through suppression of TGF‐β1‐induced Smad2/3 phosphorylation, which is reflected in the upcoming intracellular events including Smad2/3‐Smad4 interaction, nuclear localization, and TGF‐β1‐induced gene expression. Moreover, the kinase activity of RIPK4 is required for this process. The in vitro wound‐scratch assay demonstrated that RIPK4 suppressed TGF‐β1‐mediated wound healing through blocking TGF‐β1‐induced cell migration. In conclusion, our results showed the antagonistic effect of RIPK4 on TGF‐β1 signaling in keratinocytes for the first time and have the potential to contribute to the understanding and treatment of skin diseases associated with aberrant TGF‐β1 signaling.     

MeCP2‐421‐mediated RPE epithelial‐mesenchymal transition and its relevance to the pathogenesis of proliferative vitreoretinopathy

Proliferative vitreoretinopathy (PVR) is a blinding eye disease. Epithelial‐mesenchymal transition (EMT) of RPE cells plays an important role in the pathogenesis of PVR. In the current study, we sought to investigate the role of the methyl‐CpG‐binding protein 2 (MeCP2), especially P‐MeCP2‐421 in the pathogenesis of PVR. The expressions of P‐MeCP2‐421, P‐MeCP2‐80, PPAR‐γ and the double labelling of P‐MeCP2‐421 with α‐SMA, cytokeratin, TGF‐β and PPAR‐γ in human PVR membranes were analysed by immunohistochemistry. The effect of knocking down MeCP2 using siRNA on the expressions of α‐SMA, phospho‐Smad2/3, collagen I, fibronectin and PPAR‐γ; the expression of α‐SMA stimulated by recombinant MeCP2 in ARPE‐19; and the effect of TGF‐β and 5‐AZA treatment on PPAR‐γ expression were analysed by Western blot. Chromatin immunoprecipitation was used to determine the binding of MeCP2 to TGF‐β. Our results showed that P‐MeCP2‐421 was highly expressed in PVR membranes and was double labelled with α‐SMA, cytokeratin and TGF‐β, knocking down MeCP2 inhibited the activation of Smad2/3 and the expression of collagen I and fibronectin induced by TGF‐β. TGF‐β inhibited the expression of PPAR‐γ, silence of MeCP2 by siRNA or using MeCP2 inhibitor (5‐AZA) increased the expression of PPAR‐γ. α‐SMA was up‐regulated by the treatment of recombinant MeCP2. Importantly, we found that MeCP2 bound to TGF‐β as demonstrated by Chip assay. The results suggest that MeCP2 especially P‐MeCP2‐421 may play a significant role in the pathogenesis of PVR and targeting MeCP2 may be a potential therapeutic approach for the treatment of PVR.