1. Introduction
Congenital pseudomyotonia in cattle (PMT) is a rare skeletal muscle disorder with a recessive mechanism of inheritance. It is clinically characterized by stiffness and progressive impairment of muscle relaxation induced by exercise [
1]. Cattle PMT was first described in the
Chianina breed [
1], and at almost the same time, in the
Belgian Blue breed (termed congenital muscular dystonia1 or CMD1) [
2]. Subsequently, it was found in the
Romagnola cattle breed [
3] and as a single case in a
Dutch Improved Red and White (DRW) calf [
4]. It has been well established that the muscle relaxation impairment underlying cattle PMT is due to defective content in the muscles of Sarco(endo)plasmic Reticulum Ca
2+ ATPase isoform 1 (SERCA1), caused by missense variants in the
ATP2A1 gene [
5]. In vertebrates, three genes (
ATP2A1-3) encode three SERCA1 isoforms, differentially expressed in a tissue-dependent fashion and according to the stage of development [
6]. SERCA1 is the main protein component of the non-junctional sarcoplasmic reticulum (SR) membranes of adult fast-twitch (type 2) skeletal muscle fibers. At the end of the contraction cycle, SERCA1 transports two Ca
2+ ions from cytosol into the SR lumen at the expense of the hydrolysis of one ATP molecule, thus initiating muscle relaxation [
7].
In
Chianina, as well as
Dutch Improved Red White and
Belgian Blue, affected animals, DNA sequencing detected single causative homozygous missense mutations in SERCA1. In
Chianina, the causative mutation replaces an Arg at position 164 by a His (R164H). In
Dutch Improved Red White, as well as in
Belgian Blue, the variant leads to an Arg559Cys (R559C) substitution in a highly conserved domain of the protein. In the
Romagnola breed, two missense mutations in exon 8 (occurring in the same allele) were described, leading to Gly211Val and Gly286Val (G211V and G286V) substitutions. Most
Romagnola PMT-affected calves are in fact compound heterozygous, carrying the two point mutations in one allele, and the R164H mutation already found in the
Chianina breed, in the other allele. Only one PMT
Romagnola subject was found homozygous for the G211V/G286V variant [
3], found exclusively in this breed.
Irrespective of the type of genetic variant affecting the
ATP2A1 gene, a striking selective reduction of SERCA1 mutant protein expression has been observed in SR membranes isolated from PMT-affected bovine muscles from all different breeds [
3,
4,
8]. In
Chianina, this was because R164H SERCA1 amino acid substitution leads to a protein defective in proper folding, as we recently clarified [
9]. The R164H mutant, although still functionally active, is recognized as corrupted and prematurely degraded by the ubiquitin–proteasome system (UPS). Given the decrease in SERCA1 pump at the SR membrane, the re-uptake of calcium and the lowering of its cytosolic concentration slowed down, triggering muscle contracture. This study is focused on the G211V/G286V mutations exclusively found in the
Romagnola breed. We tested whether one or both amino acid substitutions would play a major role in causing the reduced expression level of the SERCA1 protein and whether, in the same way as the R164H mutant, the UPS is involved in the reduced expression at the SR membrane of this SERCA1 mutant, causing PMT in
Romagnola cattle.
3. Discussion
Bovine PMT is a rare muscular disorder described for the first time in
Chianina cattle [
1]. PMT pathophysiology has been fully clarified in this breed [
9]. We provided unequivocal evidence that the SERCA1 R164H mutation generates a protein most likely to be affected in its proper folding. Although still functionally active, the protein was ubiquitinated and removed by the UPS. The relevance of these studies, carried out on cattle PMT, remains in its similarity to human Brody disease, a rare autosomal recessive myopathy, characterized by muscle cramps and stiffness occurring after muscular activity or even mild exercise (such as climbing the stairs) [
15]. As in the case of bovine PMT, Brody disease is due to SERCA1 deficiency [
16], resulting from defects of the
ATP2A1 gene [
17], many of which are missense mutations [
18]. At present, a mouse model is not available for Brody disease. For these reasons, bovine PMT represents the only and widely accepted mammalian model for studying Brody disease.
Brody myopathy is also an orphan disease, since a specific therapy, to date, does not exist. Indeed, Brody patients are usually treated with generic muscle relaxant drugs that prevent calcium release from the SR [
19]. However, these molecules are unsuitable for long-term treatments and the therapy is often suspended due to side effects [
18]. The fact that the functional characteristics of the enzyme are preserved represents the fundamental prerequisite for any possible innovative pharmacological therapy based on the recovery of normal protein levels of SERCA1.
Bovine PMT has been more recently described in
Romagnola cattle. In a previous work, we reported that, in this breed, PMT missense mutations in the
ATP2A1 gene lead to G211V/G286V substitutions in the SERCA1 protein, in addition to the R164H substitution already described in
Chianina PMT (pathogenic SERCA1 variants described thus far in bovine species are summarized
Table S1). Compound heterozygosis is frequently reported in PMT-affected
Romagnola, except for one pathological case that was found to be homozygous for the novel complex variant G211V/G286V. Heterozygosity is most likely due to the accidental introgression of
Chianina PMT-affected animals in the ancestry of the
Romagnola PMT founder sire. Independently from differences in genotype, skeletal muscles of
Romagnola PMT cattle are characterized by a selective low expression level of the SERCA1 protein in SR membranes [
3], as found in
Chianina PMT.
Here, we analyzed the relative impact of the G211V and G286V mutations, exclusively found in the Romagnola breed. We used an experimental approach involving a heterologous cellular system transfected with the bovine SERCA1 expression constructs carrying G211/286V mutations together or independently. Our results clearly show that, of the two variants, G211V is the pathogenic one. In fact, the reduction in SERCA1 content, as well as in Ca2+ pumping and Ca2+ ATPase activity of SR membranes, is virtually identical when G211V is expressed alone or in combination with G286V. Vice versa, the G286V mutant does not show any difference with the WT in regard to those parameters.
In our previous work, we reported that the reduction in SERCA1 pump Ca
2+ ATPase activity in SR membranes from PMT
Romagnola muscles consistently correlated with the decreased protein expression in both heterozygous and homozygous subjects [
3,
12]. The results described here using both transfected cells and transverse cryosections from pathological skeletal muscles demonstrate that the G211V mutation does not interfere with proper localization of mutated SERCA1 pump. Using SR membrane fraction from the unique homozygous bovine patient (it must be remembered that PMT is a rare disease), we have clearly shown that the Ca
2+ dependence of the Ca
2+ ATPase activity was maintained, the pKCa50 value being almost identical to that of the WT SERCA1 protein (
Figure 3B,C). Moreover, the Ca
2+ ATPase activity in microsomes isolated from cells transfected with G211V SERCA1 and treated with the proteasome inhibitor was found to increase in correlation with the accumulation of this mutant protein within the ER membranes. Overall, these data indicate that the G211V mutation, although responsible for the reduced expression of the mutant SERCA1 protein, does not affect its catalytic function. Since in muscle specimens the mutation does not alter the transcription levels of the
ATP2A1 gene, the reduction in content of the SERCA1 protein both in cells and muscle [
3,
12] must reflect a posttranslational event. Our conclusion is that, as for the R164H SERCA1 mutant causing PMT in
Chianina cattle [
9], G211V/G286V mutations found in
Romagnola presumably originate in a folding-defective SERCA1 protein that is recognized and diverted to degradation by the UPS, while still being catalytically functional. Furthermore, in this event, the main role is played by the G211V mutation, the other variant, G286V, being almost ineffective.
Gly211 residue is located in the actuator (A) domain of SERCA1. The bovine SERCA1 molecular model, very similar to that of the rabbit enzyme, as expected by the high amino acid sequence identity [
20], revealed that Gly211 is surrounded by Val159 Pro160 residues, in addition to Asn39 (
Figure 6). This is in close agreement with that found by Miyauchi and colleagues [
21], who used the atomic structure of rabbit SERCA1. In
Romagnola PMT, the single neutral hydrogen atom side chain of Gly211 is substituted with the bulky aliphatic hydrophobic side chain of Val. It is conceivable that this group, which tends not to form hydrogen or ionic bonds with other groups in near proximity, interferes with side-chain interactions of surrounding residues, moving them away and ultimately leading to destabilization of the structure. It is possible to speculate that this substitution might be the cause of folding alteration. This is not surprising, since substitution of Gly211 residue in SERCA2b, the major isoform expressed in keratinocytes, has been depicted as one of the several mutations causing Darier disease, a skin disorder characterized by keratosis [
22]. The SERCA2b “house-keeping” variant and SERCA1 adult isoform selectively expressed in fast-twitch type II skeletal muscle fibers, although differentially expressed, share a high degree of homology [
6]. On the other hand, Gly286 residue is located in the second intraluminal loop of SERCA1, connecting M3 and M4 transmembrane alpha-helix domains [
23]. It is plausible that the substitution of the hydrogen atom of Gly with the long-branched side chain of Val was not relevant in perturbing the structure, due to the short length of this loop (almost 10 residues, as calculated by atomic structure) and to its distance from all other functional domains of SERCA1.
Recently, we proposed a pharmacological approach based on the use of CFTR correctors, specifically developed for rescuing type II mutations of the Cystic Fibrosis Transmembrane Regulator (CFTR) (e.g., F508del-CFTR), causing cystic fibrosis [
24]. It has been evidenced in vitro, as well as in vivo, that using these small molecules is possible to avoid the degradation of several alpha-sarcoglycan mutants (causing limb-girdle muscular dystrophy type R3, LGMDR3) that were consequently rescued at the plasma membrane level [
25,
26]. This new therapeutic strategy has also been proposed for Brody disease (European Patent EU 2925317). Cystic fibrosis, LGMDR3 and Brody myopathy, although different in symptoms and outcome, share the same pathogenetic mechanism: the “loss of function” due to the early disappearance of a protein [
9,
27,
28]. Acting to avoid the premature disposal of mutated SERCA1, this therapy could become a pharmacological approach addressing a specific population of Brody patients with documented reduced expression, but not activity, of the SERCA1 pump. Preliminary results obtained in vitro using cell models transiently expressing the R164H SERCA1 mutant are indeed quite encouraging.
As mentioned above, neither specific therapy nor mouse models exist for Brody disease. Interestingly, a SERCA1 knockout mouse was generated [
29]. Newborn SERCA1-null mice displayed limb contractures as in Brody human patients [
18,
29]. However, SERCA1-null mice displayed gasping respiration and severely impaired diaphragm function due to the high percentage of fast-twitch fibers in this muscle. In fact, a wide difference in fiber composition of the diaphragm muscle has been described between mice and large mammals, including humans [
29,
30,
31]. Fast-twitch fibers account for more than 90% of the diaphragm muscle in mice, compared to about 60% in humans. The SERCA1 protein, which is mutated in Brody disease, is exclusively expressed in skeletal muscle fast-twitch fibers. Consequently, newborn mice died by respiratory failure within two hours and the histological examination of the diaphragm revealed a severe hyper contracture injury of the muscle. This result may explain in part why no SERCA1 knockout in mouse models is available to date.
In conclusion, our data argue that reduced quantities of SERCA1 in Romagnola PMT-affected cattle were a consequence of the G211V mutation, the G286V mutation being practically ineffective. Moreover, results presented in this paper, together with data already collected for the R164H mutant, show that both SERCA1 point mutations found in Chianina and Romagnola cattle affect the expression of the protein but not its catalytic activity. Since null mice or natural mutants for Brody disease are not available, we hope that bovine PMT, although alternative and unconventional, might become a valuable tool for validating any pharmacological approaches in vivo, an essential step on the way to developing novel and specific treatment for the human Brody myopathy.
4. Materials and Methods
4.1. SERCA1 Construct and Site-Directed Mutagenesis
The full-length adult bovine SERCA1 cDNA was synthetically generated by Eurofins Genomics (Ebersberg, Germany), based on the published database sequence of Bos taurus (cow) SERCA1 mRNA (GenBank cDNA clone MGC:140007). HindIII and NotI restriction sites were added upstream of the ATG and downstream of the STOP codon, respectively, to allow the fragment cloning in the pcDNA3.1 expression vector. The G211V, G286V and G211V/G286V SERCA1 substitutions were generated by using the QuikChange site-directed mutagenesis kit (Stratagene, San Diego, CA, USA), according to the manufacturer’s specifications. The mutagenic primers (Eurofins Genomics, Ebersberg, Germany), annealing in the region where the mutation was introduced, were designed with a mismatch centered to the nucleotide to be changed. The construct was verified by sequencing.
4.2. Cell Culture, Transfection and Treatment with Proteasome Inhibitor
HEK293 or HeLa cells (ATCC, Manassas, VA, USA) were counted, seeded and grown in DMEM high glucose medium supplemented with 10% FBS. Cells were transfected with wild-type (WT) and G211V, G286V, G211/286V SERCA1 mutant cDNAs, using TransIT-293 (Mirus Bio) or jetOPTIMUS
® DNA (Polyplus Transfection, New York, NY, USA) transfection reagents, according to the manufacturer’s instructions. Sixteen hours after transfection, MG132 (Sigma-Aldrich, St. Louis, MO, USA), an inhibitor of the UPS, was added (10 µM final concentration dissolved in DMSO) and cells were incubated for 8 h [
8]. After treatment, cells were washed twice with ice-cold phosphate-buffered saline (PBS) and lysed in a buffer containing Tris-HCl 50 mM pH 7.5, NaCl 150 mM and NP40 1% (
v/
v), supplemented with protease inhibitors (Sigma-Aldrich, St. Louis, MO, USA). Protein concentrations were determined by the Bicinchoninic Protein Assay Kit (Quantum Protein Assay Kit, EuroClone Pero, MI, Italy).
4.3. Gel Electrophoresis and Immunoblotting
Proteins were resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a nitrocellulose membrane. The blots were exposed for an hour of 3% Bovine serum albumin in Tris-buffered saline, Tween 20 (TBS-T) (TBS + 0.05% Tween-20) and then probed with mouse-monoclonal antibodies against SERCA1 (dilution 1:5000, ThermoFisher Scientific, Waltham, MA, USA) or β-actin (dilution 1:30,000, Sigma Aldrich, St. Louis, MO, USA). The membranes were then incubated with secondary antibody (dilution 1:30,000, Sigma-Aldrich, St. Louis, MO, USA) alkaline phosphatase or horseradish peroxidase conjugated and developed with BCIP/NBT solution or ECL chemiluminescent substrate, respectively. The blots were imaged with iBright 1500 (ThermoFischer Scientific, Waltham, MA, USA). Quantification of protein bands was performed by densitometric analysis on Western blots from at least four independent experiments.
4.4. Immunofluorescence Analysis
HEK293 and HeLa cells were grown on 13 mm glass coverslips. After transfection and treatment with proteasome inhibitor, cells were gently washed with PBS pH 7.4 and fixed with 4% paraformaldehyde for 15 min at room temperature. Cells were washed with PBS and then permeabilized in 0.5% Triton X-100 in PBS. As primary antibodies, mouse monoclonal against SERCA1 (dilution 1:500, ThermoFisher Scientific, Waltham, MA, USA) and rabbit polyclonal against calreticulin (dilution 1:200, Enzo LifeScience, New York, NY, USA) were used in 1% BSA/PBS solution. After the incubation period, cells were gently washed with PBS and incubated with the Alexa Fluor 568 red or Alexa Fluor 488 green secondary antibody (dilution 2 μg/mL, ThermoFisher Scientific, Waltham, MA, USA). Nuclear morphology was characterized by staining with Hoechst 33342 (dilution 1 μg/mL, ThermoFisher Scientific, Waltham, MA, USA). Glass coverslips were closed with mowiol (Sigma-Aldrich, St. Louis, MO, USA).
Cryostat sections (10 μm) from muscle biopsies were incubated with the mouse-monoclonal primary antibody against SERCA1, followed by the Alexa Fluor 568 red secondary antibody.
Confocal microscopy was performed using a TCS-SP5 II confocal laser scanning microscope (Wetzlar, Germany).
4.5. Aequorin Ca2+ Measurements
HEK293 cells co-transfected with the cytosolic Ca
2+ probe aequorin (cytAEQ) with empty and with WT, G211V, G211/286V, G286G SERCA1-expressing vectors in a 1:1 ratio, were pre-incubated for 1–3 h with 5 μM coelenterazine. Measurements and calibration of aequorin signal were performed as previously described [
13].
4.6. Preparative Procedures and Biochemical Assay
All animal work was conducted according to the national and international guidelines for animal welfare. Semimembranosus muscle biopsy from G211V/G286V PMT-affected
Romagnola cattle was collected from the Veterinary Clinic of the University of Bologna under local anesthesia during diagnostic procedures that would have been carried out anyway, as described in our previous work [
3]. Control semimembranosus muscle samples were collected from healthy
Romagnola animals of the same gender and age, euthanized at the slaughterhouse. Biopsies were homogenized in 10 mM HEPES, pH 7.4, 20 mM KCl supplemented with protease inhibitors (Sigma-Aldrich, St. Louis, MO, USA), as described in [
8]. The myofibrils were sedimented by centrifugation at 650×
g for 10 min at 4 °C. The crude SR fraction was obtained from the previous supernatant by ultracentrifugation at 120,000×
g for 90 min at 4 °C. The final supernatant, representing the soluble sarcoplasm, was saved. Membrane fractions were resuspended in 0.3 M sucrose and 5 mM imidazole, pH 7.4, containing protease inhibitors and stored at −80 °C. The isolation of microsomal fraction from HEK293 cells was carried out according to Maruyama and MacLennan [
11]. Briefly, at the end of incubation time, cells were washed with PBS and collected in 10 mM Tris-HCl pH 7.5 and 0.5 mM MgCl
2 and homogenized with 30 strokes in a glass homogenizer. The homogenate was diluted with an equal volume of a solution of 10 mM Tris-HCl, pH 7.5, 0.5 M sucrose and 300 mM KCl. The suspension was centrifuged at 10,000×
g for 30 min at 4 °C to pellet nuclei and mitochondria. The pellet was discarded, and the supernatant was further centrifuged at 100,000×
g for 150 min at 4 °C to obtain the microsomal fraction. The final pellet was re-suspended in 0.25 M sucrose and 10 mM MOPS containing protease inhibitors. Protein concentrations were determined by the Bicinchoninic Protein Assay Kit.
ATPase activity of the SR microsomal fractions (20 μg/mL) was measured by spectrophotometric determination of NADH oxidation coupled to an ATP regenerating system, as previously described [
8]. The assay was performed at 37 °C in the presence of 2 μg/mL A23187 Ca
2+ ionophore at pCa5. For investigating the Ca
2+ dependence of ATPase activity, the concentration of free Ca
2+ was varied from pCa9 to pCa3 using EGTA-buffered solutions [
8].
NADH coupled with the ATPase assay protocol adapted for use on a 96-well microplate reader was used to analyze the ATPase activity of the microsomal fraction from HEK293 cells (5 μg/well), as previously described [
32]. The absorbance change at 340 nm was monitored using the EnVision (PerkinElmer, Waltham, MA, USA) plate reader. The experiments were performed at 37 °C in a final volume of 200 μL at pCa5 [
21] on the same buffer described above. Technical duplicates were performed for each experiment.
4.7. Immunoprecipitation
HEK293 cells were washed with PBS and subsequently lysed in RIPA solubilization buffer (50 mM TRIS-HCl pH 7.4, 150 mM NaCl, 1% NP40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate) supplemented with 2 mM N-ethyl-maleimide (NEM) (Sigma) and protease inhibitors. 450 μg of total-cell lysate (input) was subjected to immunoprecipitation. Samples were incubated overnight, tumbling at 4 °C with the polyclonal antibodies specific to SERCA1 (Cell Signaling, Danvers, MA, USA). The following day, protein G-magnetic beads (Millipore, Darmstadt, Germany) were added and incubated for 1 h at 4 °C. Beads were recovered, extensively washed with RIPA buffer containing N-ethyl-maleimide (NEM) and aspirated to dryness. SERCA1 was eluted by heating beads at 70 °C in sample-loading buffer with added beta-mercaptoethanol for subsequent detection by Western blotting. The blots were blocked with Tris-buffered saline, Tween 20 (TBS-T) with 5% milk powder and probed with mouse anti-ubiquitin (Merck Millipore, Burlington, MA, USA) and then incubated with secondary HRP-conjugated antibodies (Vector Laboratories, Newark, CA USA). Bands were detected as described above. Images were acquired by the Alliance mini HD9 Imaging System.
4.8. RNA Preparation and Quantitative Real-Time PCR
Total RNA extraction from semimembranosus muscle and cDNA synthesis were performed as previously described [
4,
8]. Real-time PCR was performed by the SYBR Green method with an Applied Biosystems 7500 Fast Real Time PCR System. Reaction conditions were as follows: 2 min at 50 °C, 10 s at 95 °C, and 40 cycles of 15 s at 95 °C, 1 min at 60 °C. Oligonucleotide primers used were: SERCA1 5′-GCACTCCAAGACCACAGAAGA-3′ (sense), 5′-GAGAAGGATCCGCACCAG-3′ (antisense); glyceraldehydes-3-phosphate dehydrogenase 5′-GGTCACCAGGGCTGCTTTTA-3′ (sense) and 5′-GAAGATGGTGATGGCCTTTCC-3′ (antisense).
4.9. Statistical Analysis
Data were expressed as means ± SD. Statistical differences among groups were determined by one-way ANOVA test, followed by Dunnett’s test for simultaneous multiple comparisons with control. When only two groups were considered, statistical analysis was performed by the unpaired two-tailed Student’s t-test. A level of confidence of p ≤ 0.05 was used for statistical significance.