Abstract
Dystrophin–dystroglycan complex (DGC) plays important roles for structural integrity and cell signaling, and its defects cause progressive muscular degeneration and intellectual disability. Dystrophin short product, Dp71, is abundantly expressed in multiple tissues other than muscle and is suspected of contributing to cognitive functions; however, its molecular characteristics and relation to dystroglycan (DG) remain unknown. Here, we report that DG physically interacts with Dp71 in cultured cells. Intriguingly, DG expression positively and DG knockdown negatively affected the steady-state expression, submembranous localization and subsequent phosphorylation of Dp71. Mechanistically, two EF-hand regions along with a ZZ motif of Dp71 mediate its association with the transmembrane proximal region, amino acid residues 788–806, of DG cytoplasmic domain. Most importantly, the pathogenic point mutations of Dp71, C272Y in the ZZ motif or L170del in the second EF-hand region, impaired its binding to DG, submembranous localization and phosphorylation of Dp71, indicating the relevance of DG-dependent Dp71 regulatory mechanism to pathophysiological conditions. Since Dp140, another dystrophin product, was also regulated by DG in the same manner as Dp71, our results uncovered a tight molecular relation between DG and dystrophin, which has broad implications for understanding the DGC-related cellular physiology and pathophysiology.
Introduction
Mutations in the dystrophin (Dp) gene are responsible for Duchenne or Becker muscular dystrophy (DMD/BMD) (1,2). The majority of Dp mutations result in prematurely truncated non-functional Dp (3–5); however, there are considerable cases in which one amino acid substitutions or deletions lead to clinical phenotypes (6–9). Among them, both C3340Y and L3238del mutations in Dp427, the full-length of Dp, result in intellectual disability (6,9). The DMD encodes not only Dp427 but also its multiple short products, Dp260, Dp140, Dp116, Dp71 and Dp40, due to the presence of alternative gene promoters and alternative splicing (10,11). From the distal part of DMD, Dp71 encoded by exons 63–79 is generated by gene promoter located in intron 62 and reported to be ubiquitously expressed in multiple tissues and cell lines (12–18). Two major Dp71 isoforms, Dp71d and Dp71f, are classified according to the presence or absence of exon 78 sequence (11,12). Although both brain and retina express various forms of Dp71 transcripts including Dp71d and Dp71f (12), Dp71f, which lacks exon 78, is the only product of Dp in HEK293 cells (19).
Dystroglycan (DG) is known as the causative gene of dystroglycanopathy and plays pivotal roles in the maintenance and function of muscle cells (20–24). DG is encoded by a single gene and processed into two mature proteins, α- and β-DG, which form a membrane-spanning protein. β-DG is a transmembrane protein and links to actin cytoskeleton as well as scaffold proteins such as syntrophins through directly interacting with Dp427 at the intracellular domain, whereas the extracellular domain non-covalently associates with α-DG that binds to extracellular matrix (24). Dp–DG complex (DGC) is expressed in various tissues and cells and plays a critical role in cell–extracellular matrix interactions (25–29). Dp427 is composed of the N-terminal actin-binding region, spectrin-like repeats, a cysteine-rich region containing a WW domain and EF-hand region followed by ZZ domain, and the C-terminus containing coiled-coil motifs (10,14). As the molecular basis, it was reported that WW, EF-hand and ZZ domains located at the distal region of Dp427 (residues 3026–3442) are important for the physical binding with DG (30–35). However, WW domain is not present in Dp71. Although the precise mechanisms of the molecular interactions of Dp short products with DG remain unclear, it is suggested that most Dp products would fulfill a function similar to Dp427 through interacting with DG in each Dp-expressing cells (11,14,18,36). In adult mouse hippocampal CA1 region, Dp products are known to present at astrocyte endfeet surrounding blood–brain barrier and inhibitory post-synapses on pyramidal neurons (25,37).
In this study, we characterized three distinct anti-dystrophin antibodies, Dp71d-specific, Dp71f-specific and pan-Dp71 antibodies, and found that HeLa cells express both Dp71d and Dp71f proteins, whereas HEK293T cells express only Dp71f. Immunoprecipitation (IP) after paraformaldehyde (PFA) crosslinking enabled us to detect physical molecular interaction between Dp71 and DG and clarified the critical regions responsible for its binding. Intriguingly, we found that steady-state expression of Dp71 was post-translationally upregulated upon DG expression, which led to accumulation of phosphorylated Dp71. Furthermore, DG is essential for proper submembranous expression and localization of wild-type Dp71. Notably, the pathogenic point mutations of Dp71, C272Y in the ZZ motif or L170del in the second EF-hand region, impaired its binding to DG, submembranous expression and DG-dependent phosphorylation of Dp71. Taken together, DG plays an essential role for the precise expression and submembranous localization of Dp71 through the physical molecular interaction and mediates the subsequent phosphorylation of Dp71.
Results
DG physically interacts with Dp71 proteins, Dp71d and Dp71f, in cultured epithelial cells
We checked the reactivities and specificities of anti-dystrophin antibodies derived from different sources, Abcam (ab15277), Proteintech (12715-1-AP) and Cosmo Bio (original antiserum), against main Dp71 isoforms, Dp71d and Dp71f, by western blotting and IP. Anti-dystrophin (Abcam, ab15277) and anti-dystrophin (Cosmo Bio, original antiserum) antibodies were generated by using 17 amino acid length of the peptides corresponding to Dp71d sequence and Dp71f sequence, respectively, whereas anti-dystrophin (Proteintech, 12715-1-AP) antibody was generated with a recombinant protein corresponding to the 350-amino acid length of the C-terminal region of Dp71f (Supplementary Material, Fig. S1A). Protein lysates from HEK293T cells overexpressing HA-tagged Dp71d spanning exons 63–79 of the dystrophin or Dp71f lacking the exons 71 and 78 were used for this study. As a result, anti-dystrophin antibody (Abcam, ab15277) detected ectopic Dp71d but not Dp71f (Supplementary Material, Fig. S1B), whereas anti-dystrophin antibody (Cosmo Bio, original antiserum) detected ectopic Dp71f, as well as endogenous Dp71f, but not Dp71d because ectopic expression level of Dp71d was higher than ectopic Dp71f when detected with anti-HA tag antibody. Anti-dystrophin antibody (Proteintech, 12715-1-AP) detected both ectopic Dp71d and Dp71f as well as endogenous Dp71f. Next, IPs with different anti-dystrophin antibodies (Abcam, Cosmo Bio, and Proteintech) followed by detection with anti-HA tag antibody was conducted, revealing that anti-dystrophin antibodies from Abcam (ab15277) and Cosmo Bio (original antiserum) specifically immunoprecipitated Dp71d and Dp71f, respectively, and anti-dystrophin antibody (Proteintech, 12715-1-AP) immunoprecipitated both Dp71d and Dp71f (Supplementary Material, Fig. S1C). Thus, anti-dystrophin (Abcam, ab15277) and anti-dystrophin (Cosmo Bio, original antiserum) antibodies are specific for Dp71d and Dp71f, respectively, and anti-dystrophin (Proteintech, 12715-1-AP) recognizes both Dp71d and Dp71f. Therefore, anti-dystrophin antibodies from Abcam, Cosmo Bio and Proteintech were hereinafter referred to as ‘anti-Dp71d’, ‘anti-Dp71f’ and ‘anti-pan Dp71’ antibodies, respectively.
To further examine the specificities of anti-dystrophin antibodies and to characterize the expression profile of Dp71 isoforms, Dp71d and Dp71f, in cultured cells, transient knockdown of dystrophin expression by siRNA targeting Dp71 was performed in HeLa and HEK293T cells (Fig. 1A). Dp71d detected in HeLa cells was prominently reduced according to the Dp71-siRNA treatment. However, doublet band detected in HEK293T cells was not quenched upon the Dp71-siRNA treatment, indicating that anti-Dp71d antibody strongly cross-reacts with unknown proteins around 65 kDa in HEK293T cells and Dp71d is not present or its expression level is under detection threshold in HEK293T cells. On the other hand, Dp71f was clearly downregulated by Dp71-siRNA in both HeLa and HEK293T cells, although one cross-reactive band was also detected around 80 kDa. Taken together, HeLa cells express both Dp71d and Dp71f; however, HEK293T cells express Dp71f but not Dp71d. Regarding the longer Dp products, Dp427, Dp260, Dp140 or Dp116 was not detected in HeLa and HEK293T cells (Fig. 1A).
DG physically interacts with Dp71 proteins, Dp71d and Dp71f, in cultured epithelial cells. (A) Western blots show HeLa cells express both Dp71d and Dp71f proteins; however, HEK293T cells preferentially express Dp71f. Specificities of the detected bands are validated by knockdown treatment with Dp71-siRNA. Ctl-siRNA is used as the negative control. Arrows indicate the specific bands. Arrowheads indicate cross-react of anti-Dp71d antibody or anti-Dp71f antibody. Actin is shown as the loading control. (B) IP using anti-Dp71d antibody (αDp71d) co-precipitates β-DG from HeLa cell extract. (C) IPs using anti-Dp71f antibody (αDp71f) co-precipitate β-DG from HeLa and HEK293T cell extracts. Arrows indicate the specific bands. Arrowheads indicate signals for IgG heavy chain used for IPs. (D) Co-IPs of FLAG-tagged β-DG with HA-tagged Dp71d or Dp71f from HEK293T cell extracts. pcDNA3 vector (Empty) is used as the negative control against Dp71 expressions. total, total lysate. (E) Western blot with anti-pan Dp71 antibody shows adult mouse hippocampus expresses Dp71, Dp140 and Dp427. (F) Confocal images of adult mouse hippocampal CA1 region with pan-Dp product(s), Dystrophin(s), detected by anti-pan Dp71 antibody and β-DG. Arrowheads, astrocyte endfeet; arrows, synaptic puncta; Or, stratum oriens; Pyr, stratum pyramidale; Rad, stratum radiatum; DAPI, nuclear staining. Scale bar, 10 μm.
Next, we examined the physical interaction potential of Dp71 isoforms with DG, a well-known binding partner of full-length dystrophin in muscle. In co-IP assay using native proteins, both anti-Dp71d and anti-Dp71f antibodies failed to co-precipitate β-DG, although another binding partner, syntrophin, was co-precipitated with Dp71d and Dp71f (Supplementary Material, Fig. S1D, left half). Alternatively, co-IPs using cross-linked proteins showed that both Dp71d and Dp71f physically interacted with β-DG as well as syntrophin (Supplementary Material, Fig. S1D, right half). This result suggests that physical interaction between Dp71 and β-DG is somewhat weak and cross-linking technique we used is a better way for the efficient biochemical detection of the molecular interaction. Thus, IPs using cross-linked proteins was applied to further examine the interaction of Dp71 with β-DG in the following studies and showed that Dp71d co-precipitated β-DG in HeLa cells (Fig. 1B). Similarly, Dp71f co-precipitated β-DG in both HeLa and HEK293T cells (Fig. 1C). Next, we carried out co-IPs using cross-linked proteins derived from HEK293T cells overexpressing ectopic FLAG-tagged DG in combination with HA-tagged Dp71d or HA-tagged Dp71f, revealing that β-DG was co-precipitated with Dp71d and Dp71f (Fig. 1D). These results indicate that both Dp71d and Dp71f physically interact with β-DG in the cultured epithelial cells.
In adult mouse hippocampus, three major Dp71 products, Dp71, Dp140 and Dp427, were detected by anti-pan Dp71 antibody (Fig. 1E). To demonstrate in vivo expressions of Dp product(s) and β-DG, we performed immunohistochemistry in adult mouse hippocampus, which revealed colocalizations of Dp product(s) and β-DG not only at astrocyte endfeet (arrowheads) but also at synaptic puncta (arrows) on the CA1 pyramidal neurons (Fig. 1F).
DG affects the steady-state expression and phosphorylation of Dp71
In transient expressions of HA-tagged Dp71d and FLAG-tagged DG in HEK293T cells, Dp71d was detected at a higher level in the double transfectant compared to the single transfectant without co-expression of DG (Fig. 2A and B). Moreover, Dp71d protein appeared to be retarded toward a higher molecular weight range on the western blot upon the co-expression with DG, suggesting that DG intracellularly affects Dp71 protein status, possibly protein stability and/or post-translational modifications. When we used HA-tagged Dp71f as well as Dp71d, we observed that DG affected Dp71f protein status in a similar way to Dp71d (Fig. 2C). This phenomena, upregulation and modification of Dp71, was DG-specific because ectopic expression of syntrophin α1 did not cause any changes in Dp71 protein status (Fig. 2D). DG-dependent modifications of Dp71 were detected in different cells, HeLa (Fig. 2E) and Neuro2a (Fig. 2F). Furthermore, Dp140, another dystrophin product, was also affected upon the co-expression with DG (Fig. 2G), suggesting that all the dystrophin isoforms that have binding potency for DG might be regulated by mutual molecular mechanism triggered by DG.
DG affects the steady-state expression and phosphorylation of Dp71. (A) Western blot of HA-tagged Dp71d (Dp71d-HA) expressed in HEK293T cells in the presence or absence of ectopic FLAG-tagged full-length DG (DG-FLAG). (B) The quantitative data obtained by densitometric analysis of the western blots shown in (A). Significantly increased levels of Dp71d-HA are detected in the presence (+DG) of ectopic DG when compared with that in the absence (−DG) of ectopic DG (*P < 0.01). Data were expressed as means ± standard deviation (SD; n = 3). (C) Western blot of HA-Dp71d or HA-Dp71f expressed in HEK293T cells in the presence or absence of ectopic DG-FLAG. (D) Western blot of Dp71d-HA expressed with ectopic DG-FLAG or green fluorescent protein (GFP)-tagged syntrophin-α1 (GFP-SNTA1). (E) Western blot of HA-Dp71d expressed in HeLa cells in the presence or absence of ectopic DG-FLAG. (F) Western blot of HA-Dp71d or HA-Dp71f expressed in Neuro2a cells in the presence or absence of ectopic DG-FLAG. (G) Western blot of HA-tagged Dp140 (Dp140-HA) expressed in HEK293T cells in the presence or absence of ectopic DG-FLAG. (H) Western blot of the Dp71d-HA after in vitro λ phosphatase treatment of anti-HA immunoprecipitates from HEK293T cells transiently expressing Dp71d-HA in the presence or absence of DG-FLAG. In the control experiment, the immunoprecipitates were incubated without λ phosphatase. (I) Western blot of the Dp71d-HA after in vitro λ phosphatase treatment of anti-FLAG immunoprecipitates from HEK293T cells transiently expressing Dp71d-HA in the presence or absence of DG-FLAG. pcDNA3 vector and p3XFLAG-CMV-14 vector are used as the negative controls against Dp71 and DG expressions, respectively. total, total lysate.
Next, to analyze the possibility that phosphorylation would be responsible for the Dp71 protein modification caused by co-expression with DG, we performed an in vitro phosphatase assay in which anti-HA-tagged Dp71d immunoprecipitates from the HEK293T cells co-expressing with or without DG were incubated in vitro with λ protein phosphatase (Fig. 2H). In this assay, immunoprecipitated samples but not total extracts were used for the protein phosphatase treatment in order to exclude contamination of cell-derived enzymes. As a result, protein phosphatase treatment caused a molecular shift of the Dp71 signal toward a lower molecular range, indicating that the Dp71 with a higher molecular weight caused by DG expression is a phosphorylated form. Additionally, anti-FLAG tagged DG IPs followed by in vitro λ protein phosphatase treatment showed that Dp71 protein bound to DG was a phosphorylated form at least in part (Fig. 2I). Taken together, DG physically interacts with Dp71 and affects the steady-state expression and post-translational modifications, especially phosphorylation, of Dp71 via unidentified mechanisms.
Identification of the critical region of DG for interacting with Dp71 and its protein modifications
To determine the critical region of DG for the phosphorylation of Dp71 and its physical interaction with Dp71, we carried out the co-expressions and co-IPs of several FLAG-tagged DG deletion mutants (Fig. 3A, Mut-1, −2, −3, −4 and −5) with the full-length of HA-tagged Dp71d. The full-length DG, Mut-2, Mut-3 and Mut-4 caused molecular shift of the Dp71 toward a higher molecular weight range, but the Mut-1 or Mut-5 did not affect the Dp71 status (Fig. 3B). The full-length DG and Mut-4 physically interacted with Dp71, but Mut-1 or Mut-5 did not bind to Dp71 (Fig. 3C). These results indicate that a span of 19 amino acid residues (788–806) locating near the membrane-proximal part within the DG cytoplasmic domain is essential for the physical interaction and modifications of Dp71 (Supplementary Material, Fig. S2A). To validate the necessity of the membrane-proximal region for interacting with Dp71, Mut-6 that lacks a span of 26 amino acid residues (781–806) was generated (Fig. 3A) and used for the co-IP assay. As a result, neither binding of the Mut-6 to Dp71 nor Dp71 modifications was not detected (Fig. 3D). All these results from the co-expression and co-IP assays using DG deletion mutants showed a tight relation of binding potency of DG with Dp71 and modifications of Dp71, suggesting that the Dp71 modifications would be caused as a consequence of the physical interaction between Dp71 and DG. As a molecular basis, the short stretch sequence of 19 residues at the membrane-proximal part of DG is responsible for interacting with Dp71 and necessary for the Dp71 modifications. This critical sequence is completely conserved throughout species (Supplementary Material, Fig. S2B), so that our consequence regarding the molecular mechanism for DG-dependent Dp71 modification will be shared among species.
Identification of the critical region of DG for interacting with Dp71 and its protein modifications. (A) A schematic illustration of mouse DG (top) and C-terminal FLAG-tagged DG deletion mutants used in this study (bottom). (B) Western blot of full-length Dp71d-HA co-expressed with full-length DG (Full) or DG-deletion mutants (Mut-1, Mut-2, Mut-3, Mut-4 or Mut-5). pcDNA3 vector (−) and p3XFLAG-CMV-14 vector (Empty) are used as the negative control against Dp71 and DG expression, respectively. (C) Co-IPs using full-length Dp71d-HA and DG deletion mutants (Full, Mut-1, Mut-4 or Mut-5). (D) Co-IPs using full-length Dp71d-HA and DG-deletion mutants (Full, Mut-1, or Mut-6). total, total lysate.
Molecular basis of DG-dependent Dp71 modifications
To determine the region of Dp71 that is critical for the interaction with DG, we constructed HA-tagged Dp71d deletion mutants (Fig. 4A) and examined their interaction with the full-length of FLAG-tagged DG. In this study, we used mouse sequences for both Dp71 and DG as tools to simulate Dp–DG interaction occurred in mammalian cell lines since the overall domain structures and the amino acid sequences of Dp71 are well conserved between human and mouse. We found that the deletion mutants 1-340 and 16-340 but not 1-238 or 288-617 interacted with DG (Fig. 4B), indicating that the N-terminal region consisting of two consecutive EF-hands and ZZ domain is sufficient for the physical binding with DG. However, the N-terminal sequence, 1–15 amino acid residues, in which a hypothetical actin-binding motif is suspected to be present, is not necessary for the binding with DG because the 16-340 mutant does not have the actin-binding domain. Since the responsible amino acid sequence 16-340 is completely conserved in all the human and mouse Dp products including Dp71d and Dp71f, common molecular basis would be applied to various Dp–DG interactions.
Molecular basis of DG-dependent Dp71 modifications. (A) A schematic illustration of mouse Dp71d (top) and HA-tagged Dp71d-deletion mutants used in this study (bottom). EF, EF-hand domain; ZZ, putative zinc finger domain; S, syntrophin-binding domain; CC, coiled-coil domain. (B) Co-IPs using full-length DG-FLAG and Dp71d-deletion mutants (1-340, 16-340, 1-238 or 288-617). (C) Western blot of Dp71d-deletion mutants (1-340, 16-340 or 1-238) expressed in the presence or absence of co-expression of ectopic DG-FLAG. (D) The quantitative data obtained by densitometric analysis of the western blots for Dp71d-deletion mutants (1-340, 16-340 or 1-238) shown in (C). Significantly increased levels of 1-340 and 16-340 are detected in the presence (+DG) of ectopic DG when compared with that in the absence (−DG) of ectopic DG (*P < 0.05; **P < 0.01; N.S., not significant). Data were expressed as means ± SD (n = 3). (E) Confocal images of HeLa cells co-transfected with full-length DG-FLAG and full-length Dp71d-HA or Dp71d mutants (16-340, 1-238) with anti-HA and anti-FLAG antibodies. Asterisk indicates a cell without ectopic DG. Blue, 4′,6-diamidino-2-phenylindole (DAPI) staining. (F) Confocal images of HeLa cells transfected with full-length Dp71d-HA or Dp71d mutants (1-340, 16-340, 1-238, 288-617) with anti-HA antibody. Red, pseudocolor DAPI staining. Scale bar, 20 μm.
Additionally, western blot showed that levels of both 1-340 and 16-340 were increased by co-expression with DG, whereas molecular shift toward a higher molecular weight region did not occur in 1-340 or 16-340 (Fig. 4C and D). These observations indicate that the N-terminal region of Dp71 is sufficient to form stabilized molecular complex with DG; however, DG-dependent Dp71 phosphorylation mediated by unidentified kinases requires not only the N-terminal region but also the C-terminal half, 341-617, of Dp71. It would be possible that potential phosphorylation sites of Dp71 and/or interaction site with kinases exist within the C-terminal half of Dp71. Furthermore, DG-dependent upregulation of Dp71 seems to be independent of phosphorylation of Dp71.
Next, we examined the subcellular localizations of HA-tagged Dp71d mutants as well as wild form in HeLa cells co-transfected with or without FLAG-tagged DG (Fig. 4E and F). The full-length of wild-type Dp71d (Dp71d full-HA) was restrictedly localized at subplasma membrane where FLAG-tagged DG was well colocalized with Dp71, although FLAG-tagged DG was also detected in the cytoplasmic regions as dot patterns possibly corresponding to the synthetic intermediates. This result demonstrates that Dp71 interacts with DG at subplasma membranous fraction and functions there. Similar localization pattern was seen for the mutant 16-340. In contrast, the mutants 1-238 and 288-617 were localized in both cytoplasm and nucleus. These results indicate that the full-length of Dp71 and its deletion mutants that have DG-binding potency show subplasma membranous localization but not nuclear localization, whereas the mutants that lack DG-binding potency show mislocalizations, diffuse cytoplasmic localization and nuclear localization. Notably, Dp71d full-HA, the mutants 1-340 and 16-340 demonstrated subplasma membranous patterns regardless of whether ectopic DG was present or not, suggesting that endogenous DG or the other interacting proteins would regulate submembranous localization of Dp71. Alternatively, it cannot be denied the possibility that Dp71 itself has submembranous targeting activity.
DG is necessary for the proper expression of Dp71 at subplasma membrane in the cultured epithelial cells. (A) Western blot of endogenous Dp71 proteins, pan-Dp71, Dp71d and Dp71f, in HeLa cells treated with DG-siRNA or negative control (Ctl)-siRNA for 48 h. Non-transfected cells are also analyzed as an experimental control (No siRNA). (B) Line scan analysis [regions examined are the western blot images for pan-Dp71 shown in (A)] showing the relative distributions of Dp71 protein as determined by their band intensities. Closed arrowheads indicate the higher molecular weight range in which Dp71 is detected. Open arrowhead indicates the peak position of Dp71 signal detected in the DG-knockdown sample. (C) The quantitative data obtained by densitometric analysis of the western blots for pan-Dp71 shown in (A). Significantly reduced levels of Dp71 protein are detected in the DG-knockdown sample compared to mock-treated or Ctl-siRNA (Ctl)-treated samples (*P < 0.01). Data were expressed as means ± SD (n = 5). (D) Quantitative PCR analysis of mRNA expression levels of Dp71 and DG in HeLa cells treated with Ctl-siRNA, DG-siRNA or without siRNA for 48 h. Data were expressed as means ± SD (n = 3; *P < 0.01). N.S., not significant. (E) Confocal images of pan-Dp71 and β-DG in HeLa cells treated with DG-siRNA or Ctl-siRNA for 48 h. Non-transfected cells are also analyzed as an experimental control (No siRNA). Z-stack of confocal microscope images was turned into a single 2D image by using a maximum intensity projection function of Zeiss ZEN2009 software (Z-stack, 4th column). Asterisks indicate siRNA-non-affected cells coexist with DG-knockdown cells. Blue, DAPI staining. (F) Confocal images of ectopic Dp71d-HA and endogenous β-DG in HeLa cells treated with DG-siRNA or Ctl-siRNA for 48 h. Non-siRNA-treated cells are also analyzed as an experimental control (No siRNA). Arrows and asterisks indicate Dp71d-HA-transfected cells with lower expression levels and higher expression levels, respectively. Line scan analysis (region examined indicated by the white dashed line) showing the relative distributions of Dp71d-HA as determined by their fluorescence intensities (4th column). Scale bar, 20 μm.
DG is necessary for the proper expression of Dp71 at subplasma membrane in the cultured epithelial cells
Since ectopic DG expression affects the Dp71 protein modifications, steady-state expression and phosphorylation status of Dp71, we carried out the knockdown of endogenous DG expression by DG-siRNA and examined endogenous Dp71 expressions by western blot and quantitative polymerase chain reaction (PCR). In the DG-knockdown HeLa cells, the levels of Dp71 proteins, pan-Dp71 (Dp71d and Dp71f), Dp71d and Dp71f, appeared to be slightly attenuated when compared with the controls, mock-treated or Ctl-siRNA-treated cells (Fig. 5A). The relative distributions of pan-Dp71 signals analyzed by line scan demonstrated that Dp71 in a higher molecular weight range was especially diminished in the DG-knockdown sample and the peak position in the broad band was relatively shifted toward the lower molecular weight range when compared with the controls (Fig. 5B). The protein levels of pan-Dp71 in the DG-knockdown samples were decreased up to 76 ± 11% (Fig. 5C). Quantitative PCR analysis showed that mRNA expressions of Dp71 were not changed among the siRNA treatments, whereas reduction of DG was clearly observed in the DG-siRNA-treated sample (Fig. 5D). These results indicate that endogenous DG affects the expression and modifications of Dp71 protein through a post-translational mechanism.
Next, we examined the subcellular localization of endogenous pan-Dp71 protein in HeLa cells treated with siRNAs, Ctl-siRNA or DG-siRNA, or without siRNA (Fig. 5E). In HeLa cells, Dp71 was restrictedly detected at subplasma membrane when analyzed in the controls, No siRNA-treated and Ctl-siRNA-treated cells. β-DG was also localized at plasma membrane although synthetic intermediate was detected as dots possibly localized in endoplasmic reticulum and Golgi apparatus. In DG-knockdown cells, the expression of Dp71 was clearly diminished, while siRNA-non-affected cells co-existed in the DG-siRNA-treated sample and still expressed both β-DG and Dp71. Z-stack images augmented the differences in expressions of Dp71 as well as β-DG between the knockdown cells and non-affected cells. The reduction of Dp71 submembranous expression in the DG-knockdown cells appeared to be prominent (Fig. 5E), although Dp71 in the DG-knockdown cells was detected at a slightly reduced level by western blotting when compared to the control samples (Fig. 5A and C). This discrepancy might be due to differences in the detection efficiency of different methods or reflect the possibility that Dp71 distributes diffusely within the DG-knockdown cells. These results suggest that Dp71 interacts and functions with DG at subplasma membrane fraction in cultured cells and DG is required for the proper submembranous expression of Dp71.
To validate this idea, we introduced siRNAs, Ctl-siRNA or DG-siRNA, into HeLa cells followed by ectopic expression of HA-tagged Dp71d or HA-tagged Dp71f and analyzed the subcellular localizations of HA-Dp71d or HA-Dp71f (Fig. 5F, Supplementary Material, Fig. S3A and B). In the control HeLa cells treated with or without Ctl-siRNA, ectopic Dp71d with lower expression levels (arrows) was detected as submembranous pattern, while Dp71d with higher levels (asterisks) was diffusely distributed in both cytoplasm and nucleus. In contrast, all the DG-knockdown cells demonstrated diffuse distributions of Dp71d regardless of expression levels of Dp71d. The relative distributions of Dp71d determined by their fluorescence intensities measured by line scan analysis in HeLa cells with lower expression levels of Dp71d showed mislocalizations, diffuse distribution throughout the cytoplasm and nucleus of Dp71d in the DG-knockdown cells. Similar results were obtained using HA-Dp71f (Supplementary Material, Fig. S3C). These results indicate that DG is necessary for the proper submembranous localization of Dp71 proteins, Dp71d and Dp71f, and insufficient expression of DG and/or an excess amount of Dp71 lead to the mislocalizations of Dp71. Our results obtained by ectopic expressions were consistent with the data describing the relations between endogenous Dp71 and DG.
In HEK293T cells, both pan-Dp71 and β-DG were detected as membranous pattern when analyzed in the Ctl-siRNA-treated cells (Supplementary Material, Fig. S3D). In the DG-knockdown HEK293T cells, not only β-DG but also pan-Dp71 were diminished, while siRNA-non-affected cells (asterisks) co-existed in the DG-siRNA-treated sample and still expressed both β-DG and pan-Dp71. Taken together, Dp71 interacts and functions with DG at membranous fraction in multiple cultured cell lines and DG is required for the proper submembranous expression of Dp71.
We also examined whether DG knockdown would affect syntrophin expression in HeLa cells and revealed that pan-syntrophin expressions were not drastically changed in the DG-siRNA-treated cells, although pan-Dp71 expressions were clearly diminished in most of the cells (Supplementary Material, Fig. S3E). In HeLa cells, pan-syntrophin was detected at subplasma membranes. This result suggests that DG knockdown preferentially affects Dp71 rather than an unspecified large number of components of DGC. However, we do not deny the possibility that the knockdown efficiency in this study was insufficient and knockout would be required for further evaluations of the relating proteins including syntrophins.
Next, we performed Dp71 knockdown in HeLa cells and examined the expression and subcellular localization of β-DG. Western blot showed prominent reductions of pan-Dp71, Dp71d and Dp71f in the Dp71-siRNA-treated HeLa cells. Expression levels of β-DG as well as pan-syntrophin were not changed in the Dp71-knockdown cells when compared to the controls (Supplementary Material, Fig. S4A). Immunocytochemical stains showed that expression and subcellular localization of β-DG appeared to be normal in the Dp71-knockdown cells, in which clear reductions of pan-Dp71 were detected (Supplementary Material, Fig. S4B). These results showed no obvious active involvement of Dp71 in the DG protein status as far as we examined. Steady-state expression and membranous localization of DG in cultured epithelial cells might be regulated in a Dp71-independent manner or a particular protein might compensate the Dp71 functions in the Dp71-knockdown condition.
Pathogenic mutation of Dp71 impairs precise expression, submembranous localization and phosphorylation of Dp71
To provide the molecular basis for understanding pathomechanisms of dystrophinopathy and dystroglycanopathy, we generated two distinct point mutants of Dp71, C272Y or L170del (Fig. 6A), which were previously identified in DMD patients demonstrating cognitive phenotypes (6,9). Neither C272Y nor L170del mutant showed DG-dependent Dp71 modifications, upregulation and phosphorylation (Fig. 6B and C). Additionally, physical binding with DG was not observed for the C272Y and L170del mutants (Fig. 6D). These results magnify the importance of EF-hand and ZZ domains of Dp71 and suggest that proper steric molecular structure composed of these domains might be essential for binding with DG, consequential Dp71 modifications and functions of DGC. The full-length of wild-type Dp71d (Dp71d full-HA) was restrictedly localized at subplasma membrane where FLAG-tagged DG was well colocalized with Dp71, whereas the mutants, C272Y and L170del, were localized in both cytoplasm and nucleus (Fig. 6E). These results indicate that Dp71 with pathogenic mutations, C272Y or L170del, impairs its precise expression, localization and phosphorylation, possibly due to a failure in molecular interactions with DG. Alternatively, it would be possible that DG deficiency causes mislocalization and hypophosphorylation of Dp, which leads to DGC dysfunctions.
Pathogenic mutation of Dp71 impairs precise expression, submembranous localization and phosphorylation of Dp71. (A) A schematic illustration of mouse Dp71d-HA with one amino acid substitution (C272Y) or one amino acid deletion (L170del) mutation used in this study. (B) Western blot of full-length Dp71d-HA (Wild), C272Y mutant or L170del mutant expressed in the presence or absence of co-expression of ectopic DG-FLAG. (C) The quantitative data obtained by densitometric analysis of the western blots for Dp71d wild form or Dp71d mutants (C272Y or L170del) shown in (B) (**P < 0.01). Data were expressed as means ± SD (n = 5). (D) Co-IPs using DG-FLAG and Dp71d-HA mutants (Wild, C272Y or L170del); total, total lysate. (E) Confocal images of HeLa cells co-transfected with full-length DG-FLAG and Dp71d wild form (Dp71d full-HA) or Dp71d mutants (C272Y or L170del) with anti-HA and anti-FLAG antibodies. Blue, DAPI staining; scale bar, 20 μm. (F) Western blot of the Dp71d-HA after in vitro λ phosphatase treatment of WGA-bound fractions from HEK293T cells transiently expressing Dp71d-HA in the presence or absence of DG-FLAG. p3XFLAG-CMV-14 vector is used as the negative control against DG expression. In the control experiment, the WGA-bound fractions are incubated without λ phosphatase. Anti-HA and anti-FLAG antibodies are used to detect Dp71d-HA and β-DG-FLAG, respectively. Anti-β-DG antibody detects both ectopic and endogenous β-DG. total, total lysate; WGA-capture, WGA-bound fractions. (G) Western blot of the Dp71d wild form-HA (Wild), Dp71d C272Y mutant form-HA (C272Y) or Dp71d L170del mutant form-HA (L170del) after in vitro λ phosphatase treatment of WGA-bound fractions from HEK293T cells transiently expressing each Dp71d form in the presence of DG-FLAG. In the control experiment, the WGA-bound fractions are incubated without λ phosphatase. (H) Western blots of Dp71 proteins, Dp71d and Dp71f, in WGA-bound fractions from HeLa cells treated with Ctl-siRNA, DG-siRNA or Dp71-siRNA for 48 h. Actin is shown as the loading control.
It remains unclear whether phosphorylated Dp71 would exist at subplasma membranous fraction in a DG-dependent manner or not. In order to address this question, we used wheat germ agglutinin (WGA) capture after PFA crosslinking, which enables purification of glycosylated mature forms of transmembrane proteins and its binding proteins. When we used ectopic expressions for both Dp71d and DG and examined the phosphorylation status of Dp71d in the subplasma membrane fraction captured by WGA-agarose, the levels of HA-tagged Dp71d in the WGA-bound fraction was prominently increased upon the co-expression of FLAG-tagged DG compared to the controls without ectopic DG (Fig. 6F). In vitro phosphatase treatment of the WGA-bound fractions caused a molecular shift of Dp71d toward a lower molecular weight range, indicating that Dp71 in the subplasma membrane fraction was highly phosphorylated. Additionally, in the absence of ectopic DG, slight Dp71d signals were detected in the WGA-bound fraction possibly due to the presence of endogenous DG. Taken together, proper expression of Dp71 in a subplasma membrane fraction and subsequent phosphorylation of Dp71 are dependent on the physical interaction with cell-surface DG. Thus, DG plays an important role in making unidentified kinases possibly phosphorylate Dp71 at submembranous fraction. Notably, neither C272Y nor L170del mutant of Dp71d showed comparable existence in the WGA-bound fractions even in the presence of ectopic DG when compared to the wild-form levels (Fig. 6G). In consistent with these results, when HeLa cells were treated with DG-siRNA, expression levels of DG in the total lysate as well as in the cell surface fraction captured by WGA-agarose were lower than those from control cells (Fig. 6H). Both the levels of Dp71d and Dp71f were also reduced in the WGA fraction from DG-knockdown cells, indicating that Dp71 anchors in a subplasma membrane fraction mainly through interacting with α-DG–β-DG complex. Dp71-siRNA was used as a positive control to show the downregulated expressions of Dp71 in the total lysate as well as the WGA fraction in which β-DG existed at a comparable level of the Ctl-siRNA-treated sample.
Discussion
Growing evidence suggests that DGC plays important roles for structural integrity and cell signaling in muscle (10,14). Although Dp71 is suspected of contributing to cognitive functions (11,18), molecular features of Dp71 and its relation to DG remain unclear.
Here, we characterized three distinct anti-dystrophin antibodies, Dp71d-specific, Dp71f-specific and pan-Dp71 antibodies, and found that both Dp71d and Dp71f isoforms were expressed in HeLa cells in which the other longer products, Dp116, Dp140, Dp260 or Dp427, were not detected since detection of other forms may be dependent on specific western blotting conditions or the use of other techniques. Similarly, Dp71f was the only Dp product we detected in HEK293T cells, which is consistent with a previous work reporting that Dp71 is the only Dp transcript in HEK293 cells and consists of two subisoforms, one lacking exon 78 and the other lacking exons 71 and 78 (19). Thus, these cultured epithelial cell lines are good models in which Dp71 protein expression and function are investigated.
By using these cell lines, we successfully found that both Dp71d and Dp71f physically interacted with DG and Dp71 was colocalized with β-DG at subplasma membrane. What is the biological significance of this molecular interaction? To answer the question, we performed knockdown experiments and revealed that DG played a role in proper expression and localization of Dp71 at subplasma membrane. Since submembranous localizations of both endogenous and exogenous Dp71 are tightly regulated by DG expression and DG insufficiency causes mislocalization of ectopic Dp71, experimental data to show expressions and localizations of exogenous Dp products as well as its interacting molecules should be carefully assessed whether it reflects endogenous properties or not. Similarly, specificities for endogenous Dp71 localizations also should be validated by knockdown or other techniques. Indeed, we herein showed that anti-dystrophin antibody (Abcam ab15277) reveals cross-reactivities by western blotting although the same antibody previously revealed nuclear staining (19). On the other hand, it is highly possible that Dp71 would be dysregulated in various tissues of dystroglycanopathy. Indeed, although DG null mice are embryonic lethal due to disruption of Reichert’s membrane (29), selective ablation of neuronal DG leads to specific loss of DGC in pyramidal cells (25). Since we demonstrated colocalizations of Dp product(s) and β-DG at astrocyte endfeet as well as synaptic puncta on the CA1 pyramidal neurons, astrocyte-specific DG-deletion would promote in vivo confirmation of our findings in the future.
Through the biochemical and immunocytological analyses using deletion mutants of Dp71 as well as DG mutants, we determined the domains and the minimum amino acid sequence required for the molecular interaction between Dp71 and DG (Supplementary Material, Fig. S5A). Within the cytoplasmic domain of DG, the short stretch of 19 amino acid residues (788–806) locating near the membrane-proximal part is essential for the physical interaction with Dp71. Additionally, C-terminal tail of β-DG is dispensable for binding to Dp71. At first glance, it looks like our result is not consistent with previous reports that proline-rich C-terminus of β-DG is important for binding with Dp (31,34,38). However, considering the fact that the C-terminus of β-DG interacts with the WW domain of Dp along with the EF-hand domains and Dp71 does not have an intact WW domain structure at the N-terminus, we can reveal an interpretable explanation that intermolecular interactions via proline-rich C-terminus of β-DG and the WW domain within Dp will possibly occur in the case of Dp long products except for Dp71. Furthermore, methods used in the previous studies were in vitro binding or competition assays with recombinant proteins as well as synthetic peptides (31,34,38). In contrast, we expressed deletion mutants in living cells and visualized its intermolecular interactions by IP after crosslinking. As the result, two EF-hand regions along with a ZZ motif within Dp71 are necessary and sufficient for the binding with DG. Our observation that the Dp71 deletion mutants possessing binding potency for DG (1-340 and 16-340 in Fig. 4) were localized at subplasma membrane, whereas Dp71 deletion mutants lacking the binding potency (1-238 and 288-617) showed mislocalization, indicates a positive correlation between the binding potency of Dp mutants and its membranous localization, which strengthens the idea that DG-dependent proper expression and subcellular localization of Dp71 are regulated by the physical intermolecular interaction and EF-hand regions as well as ZZ motif are responsible for the interface of Dp71.
We found that steady-state protein level of Dp71 is upregulated in response to its physical interaction with DG and subsequently phosphorylation of Dp71 is elicited (Figs 2–4). As far as we examined, EF-hand regions along with ZZ motif are necessary and sufficient for its DG-dependent upregulation and membranous localization; however, intact Dp71 sequence would be required for the subsequent phosphorylation of Dp (Fig. 4). It would be possible that increased level of Dp71 in the presence of DG reflects the stabilization of Dp71 at subplasma membrane fraction where unidentified kinase is involved in the subsequent phosphorylation of Dp71. The Dp71d isoform encoded by exons 63–79 contains 617 amino acids (GenBank, KY679207), including 55 serines, 29 threonines and 16 tyrosines (Supplementary Material, Fig. S5B), and the full sequence for Dp71 is well conserved between human and mouse. Within the C-terminal half, 341-617, of Dp71d, four tyrosines exist at 372, 386, 407 and 456. When we substituted all these four tyrosines with phenylalanines, this mutated Dp71 protein and the wild form showed a comparable degree of DG-dependent phosphorylation (data not shown), suggesting that serines and threonines but not tyrosines are responsible for the DG-dependent phosphorylation. However, to understand the significance of Dp71 phosphorylation, phosphorylation residues within Dp71 and its regulators, such as kinases and phosphatases, as well as effectors, should be clarified in the future. Additionally, since DGC is well known as a macromolecular complex, its predictable components, such as syntrophins, dystrobrevins, sarcoglycans and so on, would contribute to Dp71-related functions according to cell types. Although the DG knockdown in HeLa cells did not reveal a prominent effect on expression and submembranous localization of syntrophins, knockout strategies in vivo models rather than immortalized cell lines might unveil exact functional significance of DG-dependent Dp71 regulations within DGC macromolecular complex.
Previous studies reported that pathogenic single amino acid mutations, L3238del in Dp427 corresponding to L170del in Dp71 and C3340Y in Dp427 corresponding to C272Y in Dp71, caused decreased expression levels of mutated Dp, likely due to protein instability, in a patient-derived immortalized B-lymphocytes (6) or in an experimental mouse model (39). In this study, we tested molecular responses of Dp71 bearing C272Y or L170del mutations and showed that all the parameters, physical binding potency for DG, steady-state expression, submembranous localization and phosphorylation, are impaired in these Dp71 mutants, indicating the importance of these molecular features under physiological conditions and possible relevance of the molecular impairments to DMD pathogenesis. As a molecular mechanism under the DMD pathogenesis, it is possible that mutated Dp fails to interact with DG, which leads to mislocalization accompanied by instability and hypophosphorylation of Dp. As we demonstrated, cysteine at 272 and leucine at 170 on Dp71 are conserved between human and mouse and point mutations, C272Y and L170del, on mouse Dp71 simulate impairments in its molecular aspects, suggesting that generations of transgenic mouse lines with these point mutations would contribute to better understanding of relations between in vivo symptoms and molecular aspects.
Since Dp71 is the most abundant Dp product in the brain and is also expressed in peripheral tissues (12–18), Dp71 will play important roles in multiple biological processes in different cell types, possibly through DG-dependent Dp71 regulations we observed in this study. Although common molecular features, DG-dependent stabilization, submembranous localization and phosphorylation, can be applied to distinct Dp71 isoforms, Dp71d and Dp71f, different roles would be shared among various Dp71 alternative isoforms. Furthermore, Dp140, another Dp product, was shown to be regulated by DG in a similar manner to Dp71, suggesting that not only Dp71 but also the other Dp products might function through DG-dependent Dp modifications.
In this study, we characterized three distinct anti-dystrophin antibodies, Dp71d-terminus-specific, Dp71f-terminus-specific and pan-Dp antibodies, which can theoretically detect longer Dp products. We established an efficient method to detect physical interaction between Dp and DG by applying IP after PFA crosslinking. And then, we found that physical interaction of DG with Dp71 is essential for proper expression, membranous localization and subsequent phosphorylation of Dp71.
Materials and Methods
Antibodies
The antibodies used were as follows: anti-hemagglutinin (HA) (3724) from Cell Signaling (Massachusetts, USA), anti-Dystrophin (12715-1-AP) from Proteintech Japan (Tokyo, Japan), anti-Dystrophin (ab15277) from Abcam (Tokyo, Japan), anti-Actin (60008-1-Ig) from Proteintech Japan, anti-FLAG M2 (F1804) from Sigma (Tokyo, Japan), anti-β-Dysroglycan (43DAG1/8D5, NCL-β-DG) from Leica Biosystems (Newcastle, UK) and anti-syntrophin (1351, NB600-1294) from Novus Biologicals (Colorado, USA). Original anti-Dp71f rabbit serum was generated with synthesized peptide (CRAMESLVSVMTDEEGAE) as the antigen by outsourcing to Cosmo Bio (Tokyo, Japan). Pre-immune serum as the control was also provided by Cosmo Bio. Goat anti-rabbit IgG conjugated to HRP, goat anti-mouse IgG conjugated to HRP and goat anti-mouse IgM conjugated to HRP were from Southern Biotechnology Associates (Alabama, USA). Goat anti-rabbit IgG conjugated to Alexa Fluor 488 and goat anti-mouse IgG conjugated to Alexa Fluor 647 were from Life Technologies, Tokyo, Japan.
Constructs
The mammalian expression vector encoding C-terminal HA-tagged full-length mouse Dp71d (residues 1–617) (Accession no. KY679207) and vector encoding N-terminal HA-tagged mouse Dp71f (residues 1–622) (Accession No. KY679208) were described previously (40). To generate the N-terminal HA-tagged full-length mouse Dp71d protein, the cDNAs encoding mouse Dp71d (residues 1–617) and its 3′-untranslated region (UTR) were prepared from adult mouse hippocampus by 3′ rapid amplification of cDNA ends and cloned into the XhoI and NotI sites of the pCMV-HA vector (Clontech, Tokyo, Japan) as described previously (40). The cDNA encoding Dp71d deletion mutants (residues 1–340, 16–340 and 1–238) were cloned into pCMV-HA vector and mutant (residues 288–617) was cloned into pcDNA3 vector (Life Technologies) by using primer pairs listed in the Supplementary Material, Table S1. Site-directed mutants of Dp71d were generated by using the PrimeSTAR Mutagenesis Basal kit (Takara, Shiga, Japan). The cDNA encoding full-length human Dp140 (residues 1–1225) was prepared from human neural progenitor cells by reverse transcription (RT)-PCR and cloned into the NotI and XbaI sites of the pcDNA3.1+/C-HA vector (GenScript, Tokyo, Japan) by using primer pairs listed in the Supplementary Material, Table S1. The cDNA encoding full-length mouse DG (residues 1–893) and its deletion mutants (residues 1–777, 1–878, 1–819, 1–806, 1–787 and 1–780 fused with 807–878) were prepared from mouse hippocampus by RT-PCR and cloned into the EcoRI and BglII sites of the p3XFLAG-CMV-14 vector (Sigma) by using primer pairs listed in the Supplementary Material, Table S1. green fluorescent protein (GFP)-tagged syntrophin-α1 was previously generated (40).
Cell culture
HeLa, HEK293T and Neuro2a cells were maintained at 37°C with 5% CO2 in DMEM (Nacalai Tesque, Kyoto, Japan) containing 10% fetal bovine serum (Nichirei, Tokyo, Japan) and an antibiotic–antimycotic solution (Nacalai Tesque). The transfections of the plasmids into the HeLa and HEK293T cells were performed with EcoTransfect transfection reagent (OZ Biosciences, California, USA). HeLa and HEK293T cells were transfected with a small inhibitory RNA (siRNA) using Lipofectamine RNAiMAX Transfection Reagent (Thermo Fisher Scientific, Tokyo, Japan) according to the manufacturer’s protocol. Dicer-substrate short interfering RNA (DsiRNA) targeting the human dystroglycan gene (hs.Ri.DAG1.13.1, REF # 102581309) and negative control DsiRNA (REF # 101725097) were purchased from Integrated DNA Technologies (Tokyo, Japan). Predesigned stealth siRNA targeting the human Dp71 gene (siRNA id: HSS141851) was from Life technologies, Japan.
Immunoprecipitations
Native IPs were performed by using cell lysates with the lysis buffer A (25 mm Tris–HCl, pH 7.5, 150 mm NaCl, 1% Tx-100, 5 mm EDTA). The cell lysates were incubated with anti-Dystrophin antibody (Abcam, ab15277), anti-Dp71f serum (Cosmo Bio, original antiserum) or pre-immune serum, followed by capture with Protein G Sepharose 4 Fast Flow (GE Healthcare Life Sciences, UK). After four times washing with the lysis buffer A, immunoprecipitates were subjected to SDS-PAGE and western blotting as described previously (40). IPs after PFA crosslinking were used for the detection of molecular interaction between Dp71 and DG. Briefly, cultured cells were fixed with 4% PFA in phosphate buffer for 10 min at room temperature, washed with PBS and incubated with 125 mm glycine in PBS to quench PFA. After washing with PBS, cells were extracted with the lysis buffer B (25 mm Tris–HCl, pH 7.5, 150 mm NaCl, 1% NP40, 0.5% deoxycholate, 0.1% SDS, 5 mm EDTA) and sonicated with Vibra-Cell Ultrasonic Liquid Processors (SONICS, Connecticut, USA) for 30 s and inverted gently on a rotating wheel (ROTATOR RT-5, TAITEC, Saitama, Japan) for 30 min. Supernatants after centrifugation at 20 000g for 30 min were subjected to IPs using anti-Dystrophin antibody (Abcam, ab15277), anti-Dp71f serum (Cosmo Bio, original antiserum), pre-immune serum, Anti-HA Magnetic Beads (Thermo Fisher Scientific) or Anti-FLAG M2 Magnetic Beads (Merck, Germany). Immunoprecipitates were washed four times with the lysis buffer B, incubated in SDS sample buffer for 20 min at 96°C and subjected to SDS-PAGE and western blotting. Signals were developed by using Luminata Forte Western HRP substrate (Merck Ltd, Tokyo, Japan) and Image-Quant LAS500 (GE Healthcare Life Sciences). The data were evaluated by line scan analysis or densitometric analysis using ImageJ 1.51h software (NIH, Maryland, USA).
In vitro phosphatase treatment
In vitro phosphatase treatment was performed as described previously (40).
Animals
Animal experiment in this study was proved by the Institutional Review Board for Biomedical Research using animals at Kyoto Prefectural University of Medicine, and the animals were handled according to the institutional guidelines and regulations. The experiment was carried out on 8-week-old male ICR mice purchased from a breeder (SHIMIZU Laboratory Supplies Co. Ltd, Kyoto, Japan). The mouse brain was fresh frozen in Tissue-Tek O.C.T. Compound and subjected to preparation of coronal sections.
Immunocytochemistry
Immunocytochemical staining was performed as described previously with slight modification (41). To detect endogenous Dp product(s), anti-Dystrophin antibody (Proteintech Japan, 12715-1-AP) was used and the primary antibodies were visualized with the proper combination of secondary antibodies: goat anti-rabbit IgG conjugated to Alexa Fluor 546 and goat anti-mouse IgG conjugated to Alexa Fluor 488 or goat anti-rabbit IgG conjugated to Alexa Fluor 488 and goat anti-mouse IgG conjugated to Alexa Fluor 647. Fluorescence images were acquired by a confocal fluorescence microscope (LSM510 Ver. 4.0, Carl Zeiss, Wetzlar, Germany). Z-stack of confocal microscope images is turned into a single 2D image by using a maximum intensity projection function of Zeiss ZEN2009 software.
Quantitative RT-PCR
Total RNA from HeLa cells was isolated using the TRI Reagent (Molecular Research Center, Ohio, USA). From 1 μg of the total RNA, cDNA synthesis was performed with the ReverTra Ace qPCR RT Kit and quantitative PCR reactions were performed with the THUNDERBIRD SYBR QPCR MIX (both from TOYOBO, Osaka, Japan) according to the manufacturer’s protocols. Samples were normalized to Actin expression levels. Data were analyzed using the ∆∆CT method. Primers used were described in the Supplementary Material, Table S2.
WGA capture
Cultured cells were fixed with 4% PFA in phosphate buffer for 10 min, washed with PBS and incubated with 125 mm glycine in PBS to quench PFA as described above. After washing with PBS, cells were extracted with the lysis buffer C (25 mm Tris–HCl, pH 7.5, 150 mm NaCl, 1% Tx-100, 2 mm CaCl2) and sonicated for 30 s and inverted gently on a rotating wheel for 30 min as described above. Supernatants after centrifugation at 20 000g for 30 min were incubated with WGA-agarose (J-CHEMICAL, Tokyo, Japan) on a rotating wheel for 4 h at 4°C. The resultant resin was washed three times with the lysis buffer C, incubated in SDS sample buffer for 20 min at 96°C and subjected to SDS-PAGE and western blotting.
Statistical analysis
The data were shown as the mean ± standard deviation of the mean. Data comparisons between two groups were performed by two-tailed Student’s t-test. Differences among groups were assessed using one-way ANOVA with a Tukey post hoc analysis. P < 0.05 was considered to be statistically significant.
Acknowledgements
We would like to thank the past and present members of K.I.’s laboratory for fruitful discussions. We also thank Professor emeritus Shinji Fushiki for valuable comments and suggestions when reviewing an earlier draft of this paper.
Conflict of Interest statement:None declared.
Funding
MEXT KAKENHI (grant number 18K07883 to T.F.).
