CCDC78: Unveiling the Function of a Novel Gene Associated with Hereditary Myopathy

Abstract

CCDC78 was identified as a novel candidate gene for autosomal dominant centronuclear myopathy-4 (CNM4) approximately ten years ago. However, to date, only one family has been described, and the function of CCDC78 remains unclear. Here, we analyze for the first time a family harboring a CCDC78 nonsense mutation to better understand the role of CCDC78 in muscle.

Methods: We conducted a comprehensive histopathological analysis on muscle biopsies, including immunofluorescent assays to detect multiple sarcoplasmic proteins. We examined CCDC78 transcripts and protein using WB in CCDC78-mutated muscle tissue; these analyses were also performed on muscle, lymphocytes, and fibroblasts from healthy subjects. Subsequently, we conducted RT-qPCR and transcriptome profiling through RNA-seq to evaluate changes in gene expression associated with CCDC78 dysfunction in muscle. Lastly, coimmunoprecipitation (Co-Ip) assays and mass spectrometry (LC-MS/MS) studies were carried out on extracted muscle proteins from both healthy and mutated subjects.

Results: The histopathological features in muscle showed novel histological hallmarks, which included areas of dilated and swollen sarcoplasmic reticulum (SR). We provided evidence of nonsense-mediated mRNA decay (NMD), identified the presence of novel CCDC78 transcripts in muscle and lymphocytes, and identified 1035 muscular differentially expressed genes, including several involved in the SR. Through the Co-Ip assays and LC-MS/MS studies, we demonstrated that CCDC78 interacts with two key SR proteins: SERCA1 and CASQ1. We also observed interactions with MYH1, ACTN2, and ACTA1.

Conclusions: Our findings provide insight, for the first time, into the interactors and possible role of CCDC78 in skeletal muscle, locating the protein in the SR. Furthermore, our data expand on the phenotype previously associated with CCDC78 mutations, indicating potential histopathological hallmarks of the disease in human muscle. Based on our data, we can consider CCDC78 as the causative gene for CNM4.

Keywords: CCDC78; centronuclear myopathy-4; myopathy; sarcoplasmic reticulum.

Figure 1

Pedigree of the CCDC78-mutated patients (a), photographs of the proband showing slight hypertrophy of the calves bilaterally (b–d), muscle biopsy with E-H staining (20X) indicating nuclear centralizations (e,f), and TEM (scale bar = 200 nm) revealing peculiar dilated terminal SR (red arrows), whirls of redundant membranes (green arrow), and areas of dilated and swollen SR with numerous abnormal accumulations of membranous material (blue arrows) (g–j). Immunofluorescent analysis on muscle tissue (20X): by comparing control (k–m) and CCDC78-mutated muscles (n–p) after CCDC78 (k,n) and RyR1 staining (l,o), we showed CCDC78 aggregates, overlapping with RyR1.

Figure 2

PCR amplification of CCDC78 exons 10–14 showed the presence of four transcripts in healthy controls (a). Agarose gel electrophoresis (2% agarose) of PCR-amplified products using specific PCR primer sets. Lanes 1–3 display the sequenced PCR products in control muscle; lanes 4, 5, and 6 show the transcript pattern of exon regions 10–14 in PBLs, muscle, and fibroblasts from controls, respectively. Tissues transcripts in muscle, fibroblast, and blood from controls are shown (v = transcript sequenced, X = transcript not sequenced (b). RT-qPCR in muscle showed a significant increase in our patient (c) calculated by setting the ratio of CCDC78/reference genes expression in the control group to 1. Relative expression levels were calculated relative to HPRT1, SERPINC1, and ZNF80 mRNA levels. The bars show the mean ± SD (n = 2, ** p < 0.01). RT-qPCR showed the relative expression levels of CCDC78 in lymphocytes, fibroblasts, and muscle of control samples (d). CCDC78 expression in muscle tissue was set to 1 (SD ± 0.09), and the relative expressions in fibroblasts and lymphocytes were, respectively, 2.3 (SD ± 1.30045) and 25.62 (SD ± 9.39). Relative expression levels were calculated in relation to HPRT1 and ZNF80 mRNA levels (n = 2). To assess NMD due to the p.W402* mutation, as suggested by an apparently lower level of transcript NM_001031737.3 from muscle sample (e), we analyzed the CCDC78 expression in lymphocytes of the CCDC78-mutated patient and control subjects in basal conditions and after cycloheximide treatment. The patient showed a significant increase in transcript values compared to the control (f). Relative expression levels were calculated relative to HPRT1, SERPINC1, and ZNF80 mRNA levels and set to 1. (n = 3, p < 0.01). Western blot (WB) analysis showing the expression levels of CCDC78 isoforms in muscle tissue of patient and three controls (g). The bar graph (h) shows the isoform expression fold change in CCDC78, calculated by setting the ratio of CCDC78 protein/GAPDH protein band intensities in the control group to 100. The bars show the mean ± SD. (n ≥ 5, * p < 0.01).

Figure 3

Curves obtained through pairwise comparison of the proteins in the test set in an “each against the rest” fashion (a,b): there are two cutoffs indicated on the plots: a “strict” one at 2.2 DAS score, and a “loose” one at 1.7. The hit at 2.2 is informative in terms of the number of matching segments, while the hit at 1.7 gives the actual location of the transmembrane segment (TS). The segments reported in the feature table (FT) records of the SwissProt database are marked at 1.0 DAS score (“FT lines”). In (a), we report a staggered superposition of the potential TS predictions for both the WT 48 kDa protein (NM_001031737.3) (black lines) and the relative mutated protein (p.Trp402* (NM_001031737.3)) (orange lines): WT TS starts at 209 and stops at 216 (length ~8, cutoff ~1.7); in the mutated protein, TS starts at 210 and stops at 216 (length ~7, cutoff ~1.7). The plots are very similar; however, the TS is shorter in the mutated protein. In (b), we report a staggered superposition of the potential TS predictions for both the WT 52kDa protein (NM_001378030.1) (black lines) and the relative mutated protein (p.Glu404Lys (NM_001378030.1)) (blue lines): both WT and mutated TS start at 209 and stops at 216 (length ~8, cutoff ~1.7). The plots are identical and perfectly overlapping. Three-dimensional modelling: both the WT isoforms and the relative mutated protein are shown (A2IDD5_CCDC78_HUMAN, 438aa, WT (c), A2IDD5_CCDC78_HUMAN, 401aa, p.Trp402* (d), H3BLT8_CCDC78_HUMAN, 470aa, WT. (e), H3BLT8_CCDC78_HUMAN, 470aa, p.Glu404Lys (f)). For both the variants, we observed a different spatial orientation of the first (Glu56-Asp105) and the fourth (Asp381-Ser401) alpha-helices with respect to the second one (Asn156-Asp259).

Figure 4

Volcano plots showing gene expression differences between the CCDC78 mutated patient and controls (a). The plot shows the global transcriptional change across the groups compared. All the genes are plotted, and each data point represents a gene. The log2 fold change in each gene is represented on the x-axis, and the log10 of its adjusted p-value is on the y-axis. Genes with an adjusted p-value less than 0.05 and a log2 fold change greater than 1 are indicated by red dots and represent upregulated genes. Genes with an adjusted p-value less than 0.05 and a log2 fold change less than −1 are indicated by blue dots and represent downregulated genes. Hierarchical clustering analysis of the CCDC78 mutated patients and controls with heatmap density color representation of differentially expressed genes (DEGs) (b). This analysis was performed to visualize the expression profile of the top 30 genes sorted by their adjusted p-values. This analysis is useful to identify co-regulated genes across the conditions. In (c,d), a series of downregulated and upregulated genes associated with muscular function and SR are reported. Normalized counts for each gene in the controls and the CCDC78 mutated patient are reported. Gene ontology (GO) analysis (e): top GO terms of genes associated with differentially expressed transcripts identified in muscles from the CCDC78-mutated patient as compared to controls. p-values were adjusted by false discovery rate (FDR) multiple testing correction.

Figure 5

Co-immunoprecipitation (Co-Ip) of CCDC78-interacting proteins (a): muscle lysates were subjected to Co-Ip using anti-CCDC78 antibody and analyzed by SDS-PAGE followed by blue Coomassie staining. Two bands with molecular weights of around 50 and 110 kDa were detected (band 2, in red, and 4, in green). Lane 1, protein molecular weight ladder; lane 2, muscle lysate of healthy control. Identification of CCDC78-interacting proteins by nLC-nESI-HRMS/MS (b). Mass spectrometry (MS) analysis of band 2 (c): fragmentation spectra of 781.42053 m/z identifying AT2A1_HUMAN. The corresponding putative amino acid sequences were taken from MASCOT search. The VGEATETALTTLVEK producing an Ions Score of 52 (expect: 5.9 × 10⁻⁵). The m/z values of detected positive ion fragments are in red. The ‘b’ and ‘y’ ions are singly charged fragments (molecule + 1 H +) produced by fragmentation from the N- and C-terminus, respectively; ‘b ++’ and ‘y ++’ ions are the corresponding doubly charged fragments (molecule + 2 H +). The y* ions are y ions with loss of water. MS analysis of band 4 (d): fragmentation spectra of 789.38879 m/z identifying CASQ1_HUMAN and the corresponding putative amino acid sequences taken from MASCOT search. The ELQAFENIEDEIK producing an Ions Score of 47 (expect: 3.9 × 10⁻⁵). Western blot analysis (e). Proteins immunoprecipitated using CCDC78 antibody were immunoblotted on membranes using anti-SERCA1 and anti-CASQ1 antibodies. SERCA1 and CASQ1 were detected in wild-type pulldowns (WT1, WT2) but not in CCDC78-mutated patient pulldown (PT). ATP2A1 and CASQ1 were also detected in the total cell lysate (+Cnt) but not in the IgG control (−Cnt.). Membranes were immunoblotted with anti-GAPDH (negative Co-IP control) and anti-CCDC78 (positive Co-IP control). Colocalization between SERCA1 and CCDC78 in control muscle tissue (f–h). In the muscle seriated sections, we found a colocalization of CCDC78, SERCA1, and NADPH diaphorase (i–k).

Figure 6

SERCA1 and CCDC78 aggregates (asterisks) in CCDC78-mutated muscle (d–f) compared to control (a–c). RYR1-mutated muscle (g–i): costaining of SERCA1 and CCDC78 with cores (arrow heads).

Figure 7

Co-Ip of CCDC78-interacting proteins (a). Five bands were further detected: ~250 kDa (band 1, in red), ~100 kDa (band 3, in green), ~45 kDa (band 5, in blue), and two bands in the 25–37 kDa range (band 6, in orange, band 7, in purple). Lane 1, protein molecular weight ladder; lane 2, muscle lysate of healthy patient. Identification of CCDC78-interacting proteins by nLC-nESI-HRMS/MS (b). Mass spectrometry (MS) analysis of band 1 (c): fragmentation spectra of 859.91791 m/z identifying MYH1_HUMAN. The LQNEVEDLMIDVER sequence produced an Ions Score of 81 (expect: 6.7 × 10⁻⁸). Mass spectrometric analysis of band 3 (d): fragmentation spectra of 908.42621 m/z identifying ACTN2_HUMAN. The ISSSNPYSTVTMDELR sequence produced an Ions Score of 82 (expect: 2.1 × 10⁻⁸). Mass spectrometric analysis of band 5 (e): fragmentation spectra of 1276.58142 m/z identifying ACTS_HUMAN. The LCYVALDFENEMATAASSSSLEK sequence produced an Ions Score of 77 (expect: 5.6 × 10⁻⁸). Western blot analysis (f,g). Proteins immunoprecipitated using CCDC78 antibody were immunoblotted using anti-ACTN2, anti-MYH1, anti-ACTA1, anti-TPM1, and anti-TPM2 antibodies. ACTN2, MYH1, and ACTA1 were detected in wild-type pulldowns (WT1, WT2). ACTN2, MYH1, and ACTA1 were also detected in the total cell lysate (CTR+) but not in the IgG control (CTR−). Membranes immunoblotted with anti-TPM1 and anti-TPM2 detected protein only in the total cell lysate (CTR+) but not in the WT muscles and IgG control (CTR−). Membranes were immunoblotted with anti-GAPDH (negative Co-IP control) and anti-CCDC78 (positive Co-IP control). MYH1 staining in control (h–j) and CCDC78-mutated (k–m) muscles (20×): perinuclear small MYH1 aggregates were present in mutated muscle. HeLa cells showing a costaining of CCDC78 and gamma-tubulin (n).

Figure 8

Schematic representation of cytosolic shell (CS, in green) and channel and activation core (CAC, in yellow) of RyR1 (a): black star signs represent the analyzed RYR1 mutations. CCDC78 expression analysis by WB in RYR1-mutated patients (n = 4) compared to controls (n = 3) (b): significant reduction (**, p < 0.01) in 37 kDa, 48 kDa, and 52 kDa isoforms compared to controls. The two RYR1-mutated patients harboring a mutation in the CS showed a significant reduction in 37, 48, and 52 kDa isoforms compared to controls (p < 0.01); the patients carrying a RYR1 mutation in CAC showed a significant reduction only for 37 and 52 kDa isoforms. Western blot analysis (c): Proteins immunoprecipitated using CCDC78 antibody were immunoblotted using anti-RyR1 and anti-CCDC78 antibodies. CCDC78 was detected in wild-type pulldowns (WT1, WT2), in the total cell lysate (positive control) but not in the IgG control (negative control). Membranes immunoblotted with anti-RyR1 detected the protein only in the total cell lysate. Morphometric analysis of muscle fiber cross-sectional area (d): in CCDC78-mutated muscle, we observed a significant increase in %positive RyR1 and Trisk95 nuclei compared to controls. RyR1, DAPI, and Trisk95 staining in the CCDC78-mutated muscle (f) and control (e): in mutated muscle, RyR1 aggregates co-stained with DAPI both in peripheral and central regions of the fibers; we also observed Trisk95 aggregates that colocalized with RyR1.