POPDC1(S201F) causes muscular dystrophy and arrhythmia by affecting protein trafficking

Abstract

The Popeye domain-containing 1 (POPDC1) gene encodes a plasma membrane-localized cAMP-binding protein that is abundantly expressed in striated muscle. In animal models, POPDC1 is an essential regulator of structure and function of cardiac and skeletal muscle; however, POPDC1 mutations have not been associated with human cardiac and muscular diseases. Here, we have described a homozygous missense variant (c.602C>T, p.S201F) in POPDC1, identified by whole-exome sequencing, in a family of 4 with cardiac arrhythmia and limb-girdle muscular dystrophy (LGMD). This allele was absent in known databases and segregated with the pathological phenotype in this family. We did not find the allele in a further screen of 104 patients with a similar phenotype, suggesting this mutation to be family specific. Compared with WT protein, POPDC1(S201F) displayed a 50% reduction in cAMP affinity, and in skeletal muscle from patients, both POPDC1(S201F) and WT POPDC2 displayed impaired membrane trafficking. Forced expression of POPDC1(S201F) in a murine cardiac muscle cell line (HL-1) increased hyperpolarization and upstroke velocity of the action potential. In zebrafish, expression of the homologous mutation (popdc1(S191F)) caused heart and skeletal muscle phenotypes that resembled those observed in patients. Our study therefore identifies POPDC1 as a disease gene causing a very rare autosomal recessive cardiac arrhythmia and LGMD, expanding the genetic causes of this heterogeneous group of inherited rare diseases.

Figure 7. Expression of Popdc1 and Popdc2 in the trunk skeletal muscle of the adult zebrafish.

(A and B) Transversal sections through the trunk skeletal muscles of WT (popdc1^+/+) and homozygous popdc1^S191F mutants (popdc1^S191F/S191F) were stained with antibodies for α-dystroglycan and (A) Popdc1 or (B) Popdc2. Scale bars: 10 μm. (C and D) Quantification of the plasma membrane staining of (C) Popdc1 and (D) Popdc2 in WT and homozygous Popdc1^S201F mutant animals, respectively. The regions of interest of 30 fibers were analyzed for each sample. The membrane localization of both Popdc1^S191F and Popdc2 is reduced in the popdc1^S191F/S191F mutants. In agreement with the reduction in plasma membrane localization, increased amounts of the Popdc1^S191F mutant protein are found in cytoplasmic vesicles (arrowheads in A). For statistical testing, mean values of the Popdc/dystroglycan ratios of WT (n = 3) and mutants (n = 3) were compared by paired 2-tailed Student’s t test (C and D). *P < 0.05.

Figure 6. The popdc1^S191F zebrafish mutant displays muscular dystrophy and cardiac arrhythmia.

(A–D) Confocal analysis of actin (phalloidin staining, red channel) and (A and C) DYS (green channel) or (B and D) vinculin (green channel) expression in trunk skeletal muscle of (A and B) WT (popdc1^+/+) and (C and D) homozygous mutants (popdc1^S191F/S191F). Lateral view of the tail musculature in larvae at 5 dpf. Homozygous mutants develop lesions in skeletal muscle. Muscle fibers detach from the MTJ and retract. Arrowheads indicate the presence of DYS and vinculin immunoreactivity at the end of the ruptured fibers, suggesting that the fibers that have lost connection to the MTJ are probably physically intact. The results shown are representative of 3 independent experiments. (E–H) TEM analysis of trunk skeletal muscle of (E and F) WT and (G and H) homozygous popdc1^S191F 5-dpf mutant zebrafish embryos. The MTJ in the homozygous mutant is characterized by a lack of electron-dense material. The results shown are representative of 3 independent experiments. (I–L) Analysis of heart function in WT and homozygous popdc1^S191F mutant embryos. (I) Mean heart rate (HR). (J) Mean stroke volume. (K) Percentage of embryos displaying a 2:1 AV block. (L) Percentage of embryos displaying a 2:1 AV block after Iso stimulation. Number of embryos in each group is indicated on the corresponding bar plot. Results are presented as mean ± SEM. Two groups of measurements in parts I–L were compared by paired 2-tailed Student’s t test. *P < 0.05. Scale bars: 10 μm (A–D); 1 μm (E and F); 2 μm (G and H). mf, myofibrils.

Figure 5. POPDC1^S201F hyperpolarizes the membrane potential of spontaneously beating HL-1 cells.

(A) Representative action potentials recorded for HL-1 cells or HL-1 cells transfected with POPDC1 or POPDC1^S201F. The boxes at the bottom represent an enlarged view to show the effects on maximal hyperpolarization and afterhyperpolarization. (B) Analyses of the action potential duration, analyzed at 50 % of repolarization (APD₅₀). (C) Analyses of maximal hyperpolarization (Max. HP), (D) afterhyperpolarization (AHP), and (E) activation threshold for HL-1 cells or HL-1 cells transfected with POPDC1 or POPDC1^S201F. (F) Enlarged view showing the upstroke phase of an action potential of HL-1 cells or HL-1 cells transfected with POPDC1 or POPDC1^S201F. Traces were aligned to the activation threshold. (G) Analyses of the upstroke velocity (Up. vel.). Results are presented as mean ± SEM. One-way ANOVA was used to compare multiple variables presented in B–E and G. **P < 0.01; ***P < 0.001.

Figure 4. The POPDC1^S201F mutant protein displays a reduced cAMP affinity.

(A) cAMP affinity precipitation of WT or POPDC1^S201F protein. Left panel: precipitated proteins, unbound fraction, and input samples as well as protein subjected to control precipitation using ethanolamine-agarose (agarose) were subjected to Western blot detection of POPDC1. Right panel: quantification of cAMP affinity, calculating the ratios between input and cAMP agarose–bound protein fractions. (B) Left: example of a FRET measurement of 293A cells transfected with YFP–TREK-1 together with POPDC1-CFP or POPDC1^S201F-CFP. Right: relative change in FRET signal as a measure of cAMP affinity. (C) Chemiluminescence assay in Xenopus oocytes analyzing the surface expression of extracellularly HA-tagged TREK-1 in the absence or presence of POPDC1 or POPDC1^S201F. Control, noninjected oocytes. Measurements after incubation with 0.5 mM theophylline are illustrated with white bars. (D) Representative 2-electrode voltage clamp measurements in oocytes injected with TREK-1 alone (black) or coinjected with Popdc1 (gray) or POPDC1^S201F (red). Voltage was ramped from –110 to +35 mV. (E) Relative current amplitudes at 0 mV. (F) 2-Electrode voltage clamp measurements in oocytes injected with TREK-1 or coinjected with POPDC1 or POPDC1^S201F bathed after cRNA injection in a storage solution containing 1 mM 8-Br-cAMP. (G) Relative current amplitudes of TREK-1 coinjected with POPDC1, measured in recording solution or directly after 10 minutes of perfusion with 2 mM 8-Br-cAMP. (A–C and E–G) Numbers of experiments are included in the bars. Results are presented as mean ± SEM. Two groups of measurements in A were compared by paired 2-tailed Student’s t test. One-way ANOVA was used for the data presented in B, C, and E–G. *P < 0.05; ***P < 0.001.

Figure 3. Membrane trafficking of POPDC2 is affected in patients with a homozygous POPDC1^S201F mutation.

Skeletal muscle biopsies of PTI-1 and PTIII-2 and 2 controls were immunostained for (A–D) POPDC2 (green signal) and SCGA (red signal). (E) Quantification of the relative intensities of the plasma membrane staining of POPDC2 and SGCA in 10 fibers each from 3 sections per biopsy. The signals of POPDC2 and SGCA and the ratio of both were plotted relative to the means of both controls, which were set at 1. The normalized intensities revealed a downregulation of the membrane localization of POPDC2 in the patient material. Scale bars: 10 μm. (F) Western blot analysis of POPDC2 expression in the muscle biopsy material of the 2 controls (CT1 and CT2) and PTI-1 and PTIII-2. Results are representatives of 3 independent experiments.

Figure 2. Membrane trafficking of POPDC1 is affected in muscle biopsies.

Skeletal muscle biopsies of PTI-1 and PTIII-1 and 2 controls were immunostained for (A–D) POPDC1 (green signal) and SCGA (red signal). (E) The immunofluorescence signals were quantified in 10 muscle fibers of 3 sections per biopsy. The signals of POPDC1 and SGCA and the ratio of both were plotted relative to the means of both controls, which were set at 1. (F–Q) Subcellular localization of POPDC1 in (F–H) CT1, (I–K) CT2, (L–N) PTI-1, and (O–Q) PTIII-2. Perinuclear localization of POPDC1 is only weakly present in control samples, whereas in patients’ muscle biopsies, a significant perinuclear accumulation of the mutant protein was seen. Scale bars: 10 μm (A–D, G, H, J, K, M, N, P, and Q); 20 μm (F, I, L, and O). (R and S) Western blot analysis of (R) POPDC1 expression in skeletal muscle biopsies and of (S) POPDC1 and POPDC2 in dermal fibroblasts of 2 controls (CT1 and CT2) and PTI-1 and PTIII-2. The main POPDC1 isoform in R is labeled by POPDC1. Additional isoforms are indicated by arrows, while bands, which are considered to be unspecific and remain present after peptide competition (Supplemental Figure 4, A and B), are indicated by an asterisk. Results are representative of 3 independent experiments.

Figure 1. Identification of the POPDC1^S201F mutation associated with AV block and LGMD.

(A) Pedigree of a family homozygous for the POPDC1^S201F mutation. The index patient (PTIII-2) and his brother (PTIII-1) suffered from AV block and the grandfather (PTI-1) from LGMD and AV block. (B) TEM analysis of skeletal muscle of PTI-1 revealed the presence of plasma membrane discontinuities (arrows) and subsarcolemmal and intermyofibrillar vacuoles (asterisks). C and D show higher magnifications of the boxed areas in B. Scale bars: 2 μm (B); 1 μm (C and D). Data depicted are representative of results derived from a single biopsy. (E) Holter ECG of the index patient (PTIII-2) displaying an episode of a paroxysmal AV block. (F) Electropherogram of the Sanger sequencing of genomic DNA isolated from family members. (G) Sequence alignment of part of the Popeye domain demonstrating sequence conservation. Many residues are identical (turquoise) or similar (green); however, S²⁰¹ (yellow) is ultraconserved. (H) 3D model of the Popeye domain. Only the phosphate-binding cassette (PBC) is shown. Two hydrogen bonds are formed by S²⁰¹, but these are lost when the F²⁰¹ (turquoise) mutant residue is present, possibly affecting the ligand-binding affinity of the Popeye domain.