Plekhh2, a novel podocyte protein downregulated in human focal segmental glomerulosclerosis, is involved in matrix adhesion and actin dynamics
2012; Elsevier BV; Volume: 82; Issue: 10 Linguagem: Inglês
10.1038/ki.2012.252
ISSN1523-1755
AutoresLjubica Perisic, Mark Lal, Jenny Hulkko, Kjell Hultenby, Björn Önfelt, Ying Sun, Fredrik Dunér, Jaakko Patrakka, Christer Betsholtz, Mathias Uhlén, Hjalmar Brismar, Karl Tryggvason, Annika Wernerson, Timo Pikkarainen,
Tópico(s)Genetic and Kidney Cyst Diseases
ResumoPleckstrin homology domain–containing, family H (with MyTH4 domain), member 2 (Plekhh2) is a 1491-residue intracellular protein highly enriched in renal glomerular podocytes for which no function has been ascribed. Analysis of renal biopsies from patients with focal segmental glomerulosclerosis revealed a significant reduction in total podocyte Plekhh2 expression compared to controls. Sequence analysis indicated a putative α-helical coiled-coil segment as the only recognizable domain within the N-terminal half of the polypeptide, while the C-terminal half contains two PH, a MyTH4, and a FERM domain. We identified a phosphatidylinositol-3-phosphate consensus-binding site in the PH1 domain required for Plekhh2 localization to peripheral regions of cell lamellipodia. The N-terminal half of Plekkh2 is not necessary for lamellipodial targeting but mediates self-association. Yeast two-hybrid screening showed that Plekhh2 directly interacts through its FERM domain with the focal adhesion protein Hic-5 and actin. Plekhh2 and Hic-5 coprecipitated and colocalized at the soles of podocyte foot processes in situ and Hic-5 partially relocated from focal adhesions to lamellipodia in Plekhh2-expressing podocytes. In addition, Plekhh2 stabilizes the cortical actin cytoskeleton by attenuating actin depolymerization. Our findings suggest a structural and functional role for Plekhh2 in the podocyte foot processes. Pleckstrin homology domain–containing, family H (with MyTH4 domain), member 2 (Plekhh2) is a 1491-residue intracellular protein highly enriched in renal glomerular podocytes for which no function has been ascribed. Analysis of renal biopsies from patients with focal segmental glomerulosclerosis revealed a significant reduction in total podocyte Plekhh2 expression compared to controls. Sequence analysis indicated a putative α-helical coiled-coil segment as the only recognizable domain within the N-terminal half of the polypeptide, while the C-terminal half contains two PH, a MyTH4, and a FERM domain. We identified a phosphatidylinositol-3-phosphate consensus-binding site in the PH1 domain required for Plekhh2 localization to peripheral regions of cell lamellipodia. The N-terminal half of Plekkh2 is not necessary for lamellipodial targeting but mediates self-association. Yeast two-hybrid screening showed that Plekhh2 directly interacts through its FERM domain with the focal adhesion protein Hic-5 and actin. Plekhh2 and Hic-5 coprecipitated and colocalized at the soles of podocyte foot processes in situ and Hic-5 partially relocated from focal adhesions to lamellipodia in Plekhh2-expressing podocytes. In addition, Plekhh2 stabilizes the cortical actin cytoskeleton by attenuating actin depolymerization. Our findings suggest a structural and functional role for Plekhh2 in the podocyte foot processes. The kidney glomerulus is a micro-organ comprising a molecular filtration barrier that prevents the loss of blood proteins into the primary filtrate. The capacity of the glomerular filtration barrier to facilitate this function is dependent on the coordinated function of its three constituent layers: the endothelium, the glomerular basement membrane (GBM), and the podocytes. Although each of the three layers contributes to the permselectivity of the glomerular filtration barrier, the podocyte forms the final barrier to filtration. As such, the podocyte is critically positioned to determine the ultimate composition of the primary urine that passes into Bowman's space.1.Patrakka J. Tryggvason K. New insights into the role of podocytes in proteinuria.Nat Rev Nephrol. 2009; 5: 463-468Crossref PubMed Scopus (139) Google Scholar The essential role of the podocyte in preventing the filtration of large blood proteins, such as albumin, across the glomerular filtration barrier is provided by genetic studies identifying podocyte genes that, when mutated, result in podocyte dysfunction and nephrotic syndrome.2.Tryggvason K. Patrakka J. Wartiovaara J. Hereditary proteinuria syndromes and mechanisms of proteinuria.N Engl J Med. 2006; 354: 1387-1401Crossref PubMed Scopus (444) Google Scholar Focal segmental glomerulosclerosis (FSGS) is one of the most common causes of nephrotic syndrome among adults and is a significant cause of chronic renal failure. It consists of several different categories based on histological features, and recent studies have identified a number of mutations in podocyte-expressed genes that cause podocyte dysfunction and result in some of the hereditary forms of FSGS. The cytoarchitecture of the glomerular podocyte is defined by three parts: the cell body, the major processes, and the interdigitating foot processes. The foot processes of individual podocytes enwrap the capillary and are interconnected with each other via specialized cell junctions, the slit diaphragms,3.Pavenstadt H. Kriz W. Kretzler M. Cell biology of the glomerular podocyte.Physiol Rev. 2003; 83: 253-307Crossref PubMed Scopus (1203) Google Scholar whereas their basal domains sit on the GBM. Formation of this highly articulated morphology is stringently dependent on the actomyosin cytoskeleton. The actin filament network is the predominant cytoskeletal component of podocyte foot processes and is formed by a unique assembly of linker and adaptor molecules.4.Takeda T. Podocyte cytoskeleton is connected to the integral membrane protein podocalyxin through Na+/H+-exchanger regulatory factor 2 and ezrin.Clin Exp Nephrol. 2003; 7: 260-269Crossref PubMed Scopus (56) Google Scholar The actin cytoskeleton determines proper podocyte foot process structure and provides mechanical support for morphology of glomerular podocytes. Moreover, its dynamic behavior is necessary during podocyte development, in response to injury, for counteraction of the forces that distend the capillary wall and generally for podocyte foot process flexibility. Notably, mutations/deletions in many podocyte genes that result in foot process effacement (FPE) and albuminuria ultimately impact cytoskeletal dynamics. Normally, the podocyte foot process cytoskeleton is organized as highly ordered bundles of parallel actin filaments. When effaced, however, the podocyte cytoskeleton is reorganized and transformed into a dense network.5.Oh J. Reiser J. Mundel P. Dynamic (re)organization of the podocyte actin cytoskeleton in the nephrotic syndrome.Pediatr Nephrol. 2004; 19: 130-137Crossref PubMed Scopus (68) Google Scholar Podocyte FPE is a common ultrastructural finding in proteinuric diseases where the shape of the foot process is widened and elongated; however, the mechanisms are not yet well understood. Cultured podocytes in vitro develop large lamellipodia, which are adhesion and migration structures that may functionally resemble podocyte processes in vivo. Lamellipodia are thin, protrusive actin sheets giving rise to ruffles of the plasma membrane at the leading edge of migrating cells or extending cellular processes.6.Attias O. Jiang R. Aoudjit L. et al.Rac1 contributes to actin organization in glomerular podocytes.Nephron Exp Nephrol. 2010; 114: e93-e106Crossref PubMed Scopus (32) Google Scholar We have identified over 300 glomerulus-upregulated genes using expressed sequence tag profiling and microarray analysis.7.Takemoto M. He L. Norlin J. et al.Large-scale identification of genes implicated in kidney glomerulus development and function.EMBO J. 2006; 25: 1160-1174Crossref PubMed Scopus (181) Google Scholar Plekhh2, a previously uncharacterized gene, was shown to be most strongly expressed by the testes and the podocyte cells of the kidney glomerulus.8.Patrakka J. Xiao Z. Nukui M. et al.Expression and subcellular distribution of novel glomerulus-associated proteins dendrin, ehd3, sh2d4a, plekhh2, and 2310066E14Rik.J Am Soc Nephrol. 2007; 18: 689-697Crossref PubMed Scopus (65) Google Scholar It is a protein of unique domain organization, having a coiled-coil segment at the N terminus followed by two adjoined PH domains, a MyTH domain and a FERM domain at the C terminus (Figure 1a). Here we show that Plekhh2 expression is altered in human FSGS, as well as in two mouse models of glomerular nephropathy. In vitro studies of Plekhh2 subcellular localization and interaction partners suggest that Plekhh2 has a role in adhesion structures linking the podocyte foot processes to the GBM. It may also contribute to the connection of actin filaments to these structures and affect the dynamics of the filaments. These findings therefore reveal Plekhh2 as a novel, important component of the podocyte foot processes. Immunofluorescence and immunoelectron microscopy (iEM) revealed that in the normal human glomerulus Plekhh2 localizes to the podocyte foot processes where it is mainly found centrally in the foot process cytoplasm (56%) or close to the plasma membrane (within a distance of 100nm, 28%). Some expression was also seen close to the GBM (12%) but rarely in the slit diaphragm (4%; Figure 1b, c and e). Analysis of renal biopsies from patients with perihilar and tip lesion FSGS, which are the most common FSGS subtypes in our patient population, indicated that there was a significant reduction in total podocyte Plekhh2 expression in perihilar FSGS compared with controls (0.7±0.2 vs. 1.8±0.5Au/μm2; Figure 1f; all clinical data presented in Supplementary Table S1 online). However, no correlation between the degree of proteinuria, FPE (measured as slits/μm GBM), and the amount of Plekhh2 (Au/μm2) was found. Low labeling density for Plekhh2 was seen in areas of FPE (Figure 1d) and a tendency toward decreased labeling was also noted in areas without FPE (data not shown). In tip lesion FSGS, we detected reduced levels of Plekhh2 in areas with FPE. As mentioned above, a tendency toward decreased labeling was seen in areas without FPE, but this was not statistically significant. Intriguingly, a tendency for redistribution of Plekhh2 from the plasma membrane to the central parts of the foot process was seen in both FSGS subtypes, although this was statistically significant only in tip lesion FSGS (Figure 1e). We have also examined renal biopsies from patients with minimal change nephrotic syndrome, membranous nephropathy, and diabetic nephropathy, but found no significant change in the overall amount of Plekhh2 in the podocytes, except for a reduction in areas with FPE in minimal change nephrotic syndrome (data not shown). Download .doc (.05 MB) Help with doc files Supplementary Table S1 To study the cellular function of Plekhh2, we turned our attention to investigations of its subcellular localization. When expressed heterologously as a Myc-tagged protein in human podocytes, we found that Plekhh2 localized predominantly to the peripheral regions of lamellipodia (Figure 2a). Peripheral lamellipodia staining of Plekhh2 was similarly observed in other cell types expressing the protein either transiently or stably (Supplementary Figure S1 online). Notably, this pattern of staining was confined to those regions of individual subconfluent cells that did not juxtapose neighboring cells, and no Plekhh2 staining of the plasma membrane was otherwise noted in confluent cells that abut against one another (Figure 2b). To exclude the possibility that artificially high expression levels of Plekhh2 accounted for the observed lamellipodial expression, we detergent-treated cells with 0.5% saponin before fixation and immunostaining. This permeabilization strategy results in the removal of the soluble pool of proteins while sparing those that are more stably anchored. Under these conditions, we found that Plekhh2 staining was partially retained, suggesting association with detergent-insoluble cytoskeletal/membrane structures (Supplementary Figure S2a and b online). In addition, subcellular analysis of Plekhh2-expressing podocytes indicated that Plekhh2 is present in the membrane, cytoskeletal, and cytosolic (but not nuclear) fractions (Supplementary Figure S2c online). Download .jpg (.04 MB) Help with files Supplementary Figure S1 Download .jpg (.03 MB) Help with files Supplementary Figure S2 We also transiently transfected human podocytes with a construct encoding green fluorescent protein (GFP)–Plekhh2 and followed the localization of the fusion protein in living cells during their attachment, spreading, and migration on fibronectin-coated surfaces. Plekhh2 was found in initial adhesion structures present at the basal aspect of podocytes very soon after cell plating (Supplementary Figure S3a/Supplementary Video a, Supplementary Figure S3/Supplementary Video b online). In spreading cells, GFP–Plekhh2 localized to lamellipodia (Supplementary Figure S3d and f/Supplementary Video d, Supplementary Video f online), whereas in migrating podocytes it was detected at the migration front (Supplementary Figure S3c and e/Supplementary Video c, Supplementary Video e online). Supplementary Figure S4 online shows that the construct encoding for GFP–Plekhh2 gives a band of expected size on western blot analysis (170kDa).eyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI2ODA3M2NiMzJkYTRiMDJlYTRlZjRkNjk2YWNhYjFkMSIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.XdhY5g4Lfym_OClTnGFhXXZY9b5-mEHC7AIVgbCdY3C8hhVXTjYr0bATTGYZw6N1TvJD7Zat8yxaYd46F5PYf6EvpOmRqDph-oN3sdve-yJVMVG1e4al2TT7v3L-OenjtuwG-fs7VPa3yiP0oBFcbcFQgHkh8BGB4dUgo1GGbBuS8bwr-kR2mAg-SGbF1BrSmxTDj-M6FncvLApFvua3EKOAmiXv4RyOeih5FdH_VcHdP1e_QPq2x0jdOSOEtixe3n0W0k-fRR2DAmh1ZstkOOWwq84plZ_JVBDdehjobpbrsdg6SLVlLSdC-F3Keb2m02GXkN_axzbNpu17hGPA4A Download .mp4 (1.36 MB) Help with .mp4 files Supplementary Video a Download .jpg (.02 MB) Help with files Supplementary Figure S3eyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI3MmYwZWMwNmMwYWVlNWYwNGE5ZjRkNjQ0OGQ1ZTU2MSIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.guNOGZ5Ue7bTiciG3zjc4yCdhYQcNRmMdWd2AIRLuhWpmIr5DGmkrEclbZ3FRhIf67WF__4QOMhk5TiHiz3vylW7D7ULzZRmaLYHtroUr27UWuWOVuTgtGfi15q9SgZrZiqgOXdtghZa022ghAqnr_f0vzKEHnG1kj0PyWNfjNL5lfD8_cgSguWoTFbgyseQDz9Dr91VHmu2zCNZB7Bm3z1He_G_KX-IJXH5JA6J9ggChwZnIIR2TP57Vib4_CyIzLRh-OUT8hX0ubkdwns9rDFRSfmL9g6nmykAAniGMzzV6KT5xG3AVvsbLF53tf3grdQQges-6KDvYMqBiEg3mA Download .mp4 (0.71 MB) Help with .mp4 files Supplementary Video beyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI4M2MyNmZjNGMzYTg1M2VmY2ExOThiYTk3MGNhYTZmNCIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.pD_cCezI9PJK-yRABNuSOIoevejcWvB91c5FQEkIc25ZLlN0J5kCsml42XEUTBgo1GUBkvriBK6WpgfLtS9tpy0Kl-vjMCKlBdGyABtXHDNj8Umm4nPZXRkRtzYxDAaHZHr7zxPRUTfZj2Ig1H27PVrtLdVj6raknY0q8eCenZPdtNjsuzroUX93pZsEpwT93o79J9IO4JPZn8jydCX_bDTfcg94Gi6b4kDby6BkqEKH-KnOFtjpJ0g7n6y_Wq-Fauy6FExHiyN9NnFQ_tmOyHjfSU-vG3S-ocytTHinKBs4X-Fj-UCbBx3nbCrlFuboO7f00FvQDhxa1ePRNB8MlQ Download .mp4 (1.29 MB) Help with .mp4 files Supplementary Video deyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI0YjUzYTM1ZTVkZDI4YjhiOGEyNmM1MjE5OTY2YjczNyIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.WhLeh0gtaxMT1MrGh_zn7yMnib9kd1h4bUjJNXHtkT9JJU4dDCt8prL0wT3zwJJZbqmDlJrcI1rUbmWbw-4_8cc7lvUYZCbb3fmw_KyiXz1LtYIH5X0KPiXo8017gXdBliQFoumnFfJuuYPMCLjRdiVE7SDnrAKnqFh08rHvt0c1baVHPXc2Qm3kPDqWg9fzvRrA6jOuVmCpitdZc9ULUiCIKG8L9tq6eLk-Y4-_hUcQAErBjS5Pyd1xG33_PCj2Zqt0a9dbQiTsWM6ac17-6HViyBxYRHwXkJysER5WqkPXdofEwZV2zXdczTYd2JFKb923334Z1Fz9FbKi9QYjgg Download .mp4 (1.05 MB) Help with .mp4 files Supplementary Video feyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiIwNDY4OTY1ZmE5ZWQ4YmQ4YTQzZDEzZGU1MjgwYjNmNSIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.bNWy0ttOlWa6xtRacFAjxoIgHWKEaKUsXQJWgHHPfLE-PqrSU6JYExzaoICVxL2yUBfjGRnsfxJmMrVG-PCKqPGiFPs7kN7vDw0qhgl4oe3NATIBLXKlx6AaYd0sFhAv5NMY5W4yNXOZKkC2gziUv_A2w_DxwfpzR4CiBprA2rZPTKT8MQmnQwQEu4z-Uo8qtz-qpkglHuZXl5DZNlMG7FxON4QXbQ3En11Pyog4mffbx3e_khDktneLP7g_O2GVC7WN9STPzoNxmfNOeWoX-2pgqa3fYZLi4cdNPf1QbAJVK5i9RhHt5aVhFIL0s85YtwaSTZSOa-wCwtiIaNr8hQ Download .mp4 (0.34 MB) Help with .mp4 files Supplementary Video ceyJraWQiOiI4ZjUxYWNhY2IzYjhiNjNlNzFlYmIzYWFmYTU5NmZmYyIsImFsZyI6IlJTMjU2In0.eyJzdWIiOiI3ODNjNmVjNzc0NDdiMDBlNmY4ZmQ0YWU1YTI1YTIzMCIsImtpZCI6IjhmNTFhY2FjYjNiOGI2M2U3MWViYjNhYWZhNTk2ZmZjIiwiZXhwIjoxNjc4MjM3NTAxfQ.qCowCOHFPRaKVCo-nFhi_znyu0cebR1zHja_Bf9wzsaZIMoVT9K6zk4C4Lupi798IlikwPqnN0aMUFNrByx7-4-p6_2TueoNuSZ7fA6AXY6a6GZnYldq8swoDNNzB_mss6fshFPfKZMXOrdwiqB4KaK-Tcrg_M0uslL5jIAjBd-xHnNEV2CwrqElOlO0wapBpIJCBkyF2VTDl38zm93_2H1yT79tMk3uKjXdd2JMX4Jzpl-Ay0C1LoqCoFl98dh71G0jhZ09Rz3bjGjopECjZEWgn9QhSjK2-8GHZ-JPWXmtbdLKa1QVLiM0xwGMTjbU835LQvw7_rp-SOL2Wk4d-A Download .mp4 (0.89 MB) Help with .mp4 files Supplementary Video e Download .jpg (.03 MB) Help with files Supplementary Figure S4 To additionally assess the specificity of the observed Plekhh2 localization, we turned our attention to the various domains of Plekhh2. On the basis of comparative database analysis, Plekhh2 is predicted to contain multiple putative interactive domains including PH and FERM domains that have previously been described as key determinants of protein localization.9.Lemmon M.A. Membrane recognition by phospholipid-binding domains.Nat Rev Mol Cell Biol. 2008; 9: 99-111Crossref PubMed Scopus (1121) Google Scholar,10.Fehon R.G. McClatchey A.I. Bretscher A. Organizing the cell cortex: the role of ERM proteins.Nat Rev Mol Cell Biol. 2010; 11: 276-287Crossref PubMed Scopus (742) Google Scholar In the following set of experiments, we examined whether deletion of one or more of the individual domains in the C-terminal half of Plekhh2 affected its subcellular localization (Figure 3). Expression of a truncation mutant lacking the entire C-terminal half (two PH domains, MyTH domain, FERM domain) prevented Plekhh2 lamellipodial expression, and resulted in its nuclear accumulation. Conversely, expression of a Plekhh2 mutant composed solely of its C-terminal half was targeted to the lamellipodia. These studies demonstrate the importance of the MyTH-FERM cassette and/or PH domains in targeting Plekhh2 to lamellipodial structures. Both FERM and PH domains, when singly expressed, demonstrated a propensity to accumulate at lamellipodia. Deletion of either PH or FERM domains of full-length Plekhh2 was not sufficient to completely abolish Plekhh2 localization to lamellipodial structures. This finding suggests that the PH and FERM domains cooperate to ensure the proper localization of Plekhh2. An analysis of the amino acid sequence of Plekhh2 indicated the presence of a PtdIns(3,4,5)P3 (PIP3) consensus binding site in the PH1 domain (Figure 4a). To test the significance of this finding, we mutated the selected conserved residues (K711A and R722C, asterisk) and studied the localization of the protein in podocytes. Although the K711A mutation had no effect on Plekhh2 localization, the double K711A/R722C mutation markedly affected the 'linear' lamellipodial localization of Plekhh2, particularly in the case of the mutant lacking the FERM domain. The protein seemed to be transported toward the cell periphery, but the staining was very diffuse (Figure 4b). We also treated Plekhh2-expressing cells with wortmannin or LY294002, potent phosphatidylinositol 3-kinase inhibitors. Both of these treatments markedly reduced Plekhh2 localization to the plasma membrane of lamellipodia (Figure 4c, Supplementary Figure S5 online), with only about 25% of the Plekhh2 signal remaining at this cellular location (Figure 4d). In contrast, Plekhh2 variants composed either of only the PH domains (PHPH) or lacking the FERM domain (ΔFERM) were completely relocalized to the cytosol upon wortmannin or LY294002 treatment (Supplementary Figure S5b and c online). As predicted, localization of the ΔPHPH variant was not altered by either of these two treatments. These findings strongly suggest that interaction of the PH1 domain with PIP3 contributes to Plekhh2 localization to the peripheral regions of lamellipodia, and confirm that the FERM domain at the C terminus of the polypeptide also serves to target Plekhh2 to these sites. Download .jpg (.03 MB) Help with files Supplementary Figure S5 The only recognizable domain within the N-terminal half of the 1491-residue-long Plekhh2 polypeptide is a putative α-helical coiled-coil domain located between amino acids 19 and 177 (Supplementary Figure S6a online). According to the secondary structure prediction by the PSIPRED program,11.McGuffin L.J. Bryson K. Jones D.T. The PSIPRED protein structure prediction server.Bioinformatics. 2000; 16: 404-405Crossref PubMed Scopus (2725) Google Scholar this segment folds into a continuous coiled-coil, except for one short interruption (Supplementary Figure S6b online). Coiled-coil domains are common oligomerization domains, the most predominant oligomeric forms being dimers and trimers. Analysis of the segment with the SCORER 2.0 program,12.Armstrong C.T. Vincent T.L. Green P.J. et al.SCORER 2.0: an algorithm for distinguishing parallel dimeric and trimeric coiled-coil sequences.Bioinformatics. 2011; 27: 1908-1914Crossref PubMed Scopus (38) Google Scholar which distinguishes parallel dimeric and trimeric coiled-coils, predicts that the coiled-coil domain of Plekhh2 assembles into a dimer. To provide experimental evidence for these predictions, we tested by coimmunoprecipitation whether Plekhh2 self-associates. To this end, HEK293 cells were cotransfected with full-length Plekhh2 tagged with either Myc or hemagglutinin (HA) epitopes, followed by immunoprecipitation with anti-HA antibodies and western blotting with those against the Myc-tag. As shown in Figure 5a, these two full-length versions of Plekhh2 associated with each other. As there is some evidence from structural studies indicating that the FERM domain is capable of forming dimeric structures,13.Kitano K. Yusa F. Hakoshima T. Structure of dimerized radixin FERM domain suggests a novel masking motif in C-terminal residues 295-304.Acta Crystallogr Sect F Struct Biol Cryst Commun. 2006; 62: 340-345Crossref PubMed Scopus (13) Google Scholar we tested the involvement of this domain in Plekhh2 self-association. However, no interaction was seen when HA-tagged full-length Plekhh2 was coexpressed with the Myc-tagged FERM domain (Supplementary Figure S7 online), indicating that association does not occur through the FERM domain. Download .jpg (.03 MB) Help with files Supplementary Figure S6 Download .jpg (.02 MB) Help with files Supplementary Figure S7 We also used fluorescent resonance energy transfer (FRET) to confirm that Plekhh2 self-associates. In this experiment, we coexpressed in human podocytes Plekhh2 proteins tagged at their N termini with either GFP or Myc and measured the change in GFP donor fluorescence following bleaching of the red (Myc) acceptor fluorophore (Figure 5b). Under these conditions, we detected a significant increase in the GFP signal in lamellipodia of the transfected cells (Figure 5c). If the coiled-coil domain located in the N terminus of Plekhh2 is indeed biologically functional, we reasoned that full-length Plekhh2 may be able to alter the nuclear expression pattern of the Plekhh2 mutant that lacks the entire C-terminal half, but that still possesses its coiled-coil domain (the form ΔPHPHMyTHFERM). Immunofluorescent examination of cells coexpressing these two proteins revealed that the truncated Plekhh2 partially relocalized from the nucleus to the plasma membrane of lamellipodia in the presence of the full-length protein (Figure 5d). Such a partial rescue of ΔPHPHMyTHFERM localization to lamellipodia was not similarly observed in cells overexpressing Ezrin. Altogether, these studies provide evidence that the N-terminal half of Plekhh2, containing the coiled-coil region, is sufficient for the self-association of Plekhh2, which likely assembles into a dimer. To predict the cellular function of Plekhh2, we used an unbiased yeast two-hybrid approach for discovering protein interactions, and screened a mouse kidney glomerular complementary DNA (cDNA) library with full-length Plekhh2 and its deletion variants. This screen identified Hic-5, a paxillin-related focal adhesion protein (TGFB1i1), and β-actin (ActB) as potential interacting partners of Plekhh2 (Figure 6). Hic-5 was isolated in the screens with full-length Plekhh2, the PHPHMyTHFERM form, and the FERM domain as baits, whereas ActB was identified in the screens with the latter two baits. These findings, which were confirmed in a yeast-mating assay (Figure 6a, Supplementary Figure S8 online online), indicate that it is the FERM domain of Plekhh2 that mediates interaction of Plekhh2 with these two proteins. The interaction of Plekhh2 with Hic-5 was verified in a coimmunoprecipitation assay using HEK293 cells coexpressing HA-tagged Plekhh2 and Myc-tagged Hic-5 (Figure 6b, Supplementary Figure S7 online). Download .jpg (.03 MB) Help with files Supplementary Figure S8 To assess the subcellular relationship between these two proteins, we next examined their spatial distribution in cultured human podocytes. In these cells, endogenous Hic-5 localizes mainly to focal adhesions (Figure 6c, Supplementary Figure S9a online). Following transfection of Plekhh2, endogenous Hic-5 was found to partially relocalize to lamellipodial extensions, a response that was observed as a linear staining pattern located at the periphery of lamellipodia, near the plasma membrane (Figure 6c, Supplementary Figure S9b online). This was not due to overall changes in Hic-5 expression (Supplementary Figure S9c online). Consistent with the yeast two-hybrid results, such a relocalization was also seen upon transfection of only the FERM domain of Plekhh2, but not the ΔFERM form. Download .jpg (.03 MB) Help with files Supplementary Figure S9 To examine the location of these two proteins in situ, we analyzed human kidney samples by iEM (Figure 7a and b). Hic-5 was occasionally found in the podocyte foot processes, localized close to the GBM. Coimmunolabeling of Plekhh2 and Hic-5 with 10- and 5-nm gold particles, respectively, indicated that there was significant colocalization of the two proteins and that they are less than 20nm apart from each other, providing strong evidence that Plekhh2 and Hic-5 interact in vivo as well (Figure 7c and d). Glomerular expression of both proteins was also verified by reverse transcription-polymerase chain reaction (PCR; Supplementary Figure S9d online). Examination of peripheral Plekhh2 by confocal microscopy indicated a significant overlap with the cortical actin cytoskeleton both in the XY- and Z-scan (Figure 8a). Together with the finding that ActB was isolated in the yeast two-hybrid screen with the C-terminal half of Plekhh2, as well as with its FERM domain, and the fact that Plekhh2 is found in the cytoskeletal fraction of podocyte cell lysates, we raised
Referência(s)