Abstract
Background: Central core disease (CCD) and nemaline rod myopathy are generally considered two genetically and histologically distinct disorders. CCD is defined by the presence of well-demarcated round cores within most myofibers. Nemaline rod myopathy is distinguished by the presence of characteristic nemaline bodies within myofibers. The simultaneous occurrence of both cores and rods in the same muscle biopsy has been described, but no gene mutations have been reported yet for this condition.
Objective: To describe a family containing 16 affected individuals in six generations with an autosomal dominant congenital myopathy that shows clinical and histologic features of both CCD and nemaline myopathy, and to determine the genetic etiology and protein composition of the cores/rods in this family.
Methods and Results: The results of linkage analyses excluded involvement of the two autosomal dominant nemaline myopathy loci on chromosome 1, but were consistent with a localization of the disease gene at the CCD locus on chromosome 19q13.1 (ryanodine receptor). SSCP analysis and DNA sequencing identified a novel Thr4637Ala mutation in the transmembrane region of the ryanodine receptor protein. Immunofluorescence studies of patient muscle biopsies showed the central cores to stain for ryanodine receptor.
Conclusions: These data suggest that the occurrence of nemaline bodies can be a secondary feature of CCD, and that genetic studies on previously reported core/rod families should be targeted to the ryanodine receptor locus. The results of the immunofluorescence studies suggest that the cores contain excess abnormal ryanodine receptor protein.
Central core disease (CCD) is an autosomal dominant congenital myopathy distinguished by the presence of well-demarcated round regions, or cores, within most myofibers. The cores may be singular or multiple, central or peripheral, and extend through the length of the myofiber. They are found almost exclusively in type I fibers, are best seen in oxidative stains such as nicotinamide adenine dehydrogenase (NADH) as a central area of clearance, lacking in oxidative activity because of an absence of mitochondria. The clinical features of CCD typically present in early infancy and include hypotonia and proximal muscle weakness. Skeletal abnormalities are common and may include congenital dislocation of the hips, thoracic deformities, flat feet, pes cavus, clubfoot, kyphoscoliosis, or flexion deformities of the fingers. Patients can usually perform all movements against resistance and gravity, but running and climbing stairs remain difficult. The attainment of motor milestones is delayed, but the overall progression of the disease is slow, and most patients remain fully active throughout their lives. The only gene pathology identified to date in CCD are dominant missense mutations of the ryanodine receptor (RYR1) in five families.1-4⇓⇓⇓ Seventeen other RYR1 gene mutations have been associated with malignant hyperthermia.2,5⇓
Nemaline rod myopathy is defined by the occurrence of rod-shaped structures largely composed of α-actinin and actin. These threadlike structures assemble in an irregular distribution as clusters that are best visualized with toluidine blue or Gomori trichrome stains. The clinical spectrum of nemaline myopathy ranges from a severe fatal neonatal form to adult onset forms. The typical form of nemaline myopathy is a mild, nonprogressive, or slowly progressive myopathy with hypotonia and feeding difficulties in early life, small muscles, slender extremities, and proximal muscle weakness. Weakness is also found in the distal limb, neck flexor, and trunk muscles, as well as in facial and masticatory muscles. Dysmorphic features include narrow, high-arched palate, micrognathia or marked prognathism, chest deformities, contractures of the fingers, and pes cavus or talipes equinovarus. Nemaline myopathy may be caused by sporadic mutations, or may be inherited in a dominant or recessive fashion. The proportion of inherited versus sporadic cases remains uncharacterized.6 Mutations in either the slow α-tropomyosin7 or α-actin gene8 have been identified in both dominant and recessive nemaline myopathy. Mutations in nebulin are likely to account for a substantial portion of recessively inherited cases.9
The simultaneous occurrence of cores and rods in the same patient has been reported in several instances.10-14⇓⇓⇓⇓ Although one rod/core family was previously reported to be compatible with linkage to the CCD locus on chromosome 19,15 no gene mutations have yet been reported in a patient showing both rods and cores. We report a novel mutation in the RYR1 gene in an extended family whose muscle shows the coexistence of cores and rods. Clinical features, histology, and immunohistochemical studies on patient muscle biopsies which suggest that the abnormal protein may itself contribute to core formation are described.
Patients and methods.
Patients. Sixteen affected individuals were identified in a six-generation pedigree from Mississippi. Eight affected persons were clinically examined (four men and four women). The onset was uniform in early infancy, but delays in motor milestones (sitting and walking) were identified in only two. A description of children by older family members suggested a degree of calf prominence which disappeared later in life. None had neonatal difficulties. The disease was reported to become apparent at about 1 year of age, with difficulty standing or rising from a seated position. Affected children never became good runners or jumpers. Walking up an incline was easier when done backwards or sideways.
Age at neurologic examination varied from 17 to 73 years. Motor activities were impaired in every affected person; none participated in athletics and all had problems with running, jumping, and rising from low chairs. The disease course appeared to show little or no progression, and only one person needed a cane at age 70 years. A total of 10 affected individuals had undergone a total of 22 surgical procedures under general anesthesia. Patient records were too limited to determine the exact type of anesthetics (volatile or depolarizing agents) used during these surgeries, but there was no reported history of malignant hyperthermia or other anesthesia-related complications during any of 22 surgical procedures. The clinical data are summarized in the table.
Summary of clinical examination
Histopathology.
Light microscopy. Muscle biopsies from the deltoids of two affected individuals (father and son) were flash frozen in liquid nitrogen–cooled isopentane and stained with hematoxylin and eosin (H-E) and modified Gomori trichrome using standard methods.16 NADH and adenosine triphosphatase (ATPase) reactions were also performed.
Electron microscopy.
Two muscle biopsy specimens (father and son) were fixed in 2.5% glutaraldehyde in 0.1M cacodylate buffer pH 7.3, postfixed in 1% osmium tetroxide for 1 hour at 4 °C, dehydrated in graded series of ethanol concentrations and embedded using Epon epoxy resin at 60 °C. Ultrathin sections (500 Å) were cut using a Reichert ultramicrotome and mounted on 400-mesh copper grids. Thick sections (1 μm) were stained with toluidine blue to identify alterations in the muscle fiber. Sections were also stained with uranyl acetate and lead citrate. The grids were examined in a Jeol 100 CX ultramicroscope (Peabody, MA) at 80 kV.
Linkage analyses.
Genomic DNA was extracted from leukocytes. Eleven family members (eight affected, three unaffected) were genotyped for the STR markers linked to the CCD locus and the two nemaline myopathy loci (NEM1, ACTA1). The Généthon map reported the following sex-averaged distances between markers: NEM1: D1S442–4 cM–D1S305–18 cM–D1S194; ACTA1: D1S479–<1 cM–D1S439–3 cM–D1S251; CCD: D19S225–5 cM–D19S220–1 cM–D19S881–6 cM–D19S908–7 cM–D19S90217 (PCR conditions used to analyze these markers are described at http://carbon.wi.mit.edu:8000/ftp/pub/human_STS_releases/pcr_conditions.html32). P-labeled PCR products were electrophoresed on 6% denaturing polyacrylamide gels, dried on filter paper, and exposed to X-ray film (DuPont, Wilmington, DE) using standard methods. Alleles were hand scored and genotypes were generated for available family members. Multipoint linkage analyses were performed using Genehunter-Plus as previously described.18
RNA extraction and reverse transcription.
Total RNA was isolated from a frozen muscle biopsy from one affected family member and reverse transcribed with oligo dT (Boehringer Mannheim, Mannheim, Germany) essentially as described.19
PCR amplification of the RYR1 gene.
PCR primers flanking the first 62 exons of the RYR1 gene20 were designed based on the Genbank sequences U48449 –U48481, synthesized (Life Technologies, Rockville, MD) and used to amplify 50 to 100 ng of genomic DNA in the presence of α-32P dATP radioisotope. PCR primers were obtained (gift from Professor T. McCarthy) to amplify cDNA corresponding to the remainder of the coding sequence (Genbank accession number J05200). In each case, 10 unrelated control patients were amplified. All PCR reactions were performed on an MJ Research PTC-100 (Watertown, MA) using standard conditions.
Single-strand conformational polymorphism (SSCP) analysis and DNA sequencing.
Denatured PCR products were subjected to mutational analysis using two different types of SSCP gels: 1) 0.5 × TBE/5% (49:1) acrylamide, and 2) MDE gel (FMC Bioproducts, Rockland, ME). Aberrant conformers (bands showing an electrophoretic shift in mobility by SSCP) were excised from dried gels, eluted, and reamplified as previously described.21 Sequencing of the double-stranded PCR product was performed using CEQ Dye Terminator Cycle Sequencing Kits (Beckman; Fullerton, CA) according to the manufacturer’s instructions. Purified sequencing reaction products were analyzed on a Beckman CEQ 2000 capillary automated sequencer. Aberrant conformers were sequenced in both directions. Sequences were then aligned with those of the wild-type genes (Sequencher; Gene Codes Corp., Ann Arbor, MI). Primers for PCR amplification and cycle sequencing are available upon request.
Restriction enzyme analysis of missense mutations.
The presence of the A13996G base substitution in the RYR1 gene introduces a HhaI restriction enzyme site. To confirm the presence of the heterozygous A13996G base change, a 346-bp fragment containing exon 95 was amplified from all available family members and digested with HhaI (New England Biolabs, Beverly, MA). In affected family members, digestion of the 346-bp PCR fragment with HhaI was predicted to yield an uncut 346-bp fragment, and cut fragments of 224 and 122 bp. PCR products of 346 bp from unaffected family members were predicted to be resilient to restriction enzyme digests. This technique was used to study family members and 100 additional unaffected control individuals (200 chromosomes).
Immunohistochemical studies.
The following primary antibodies were used: ryanodine receptor (34C, Developmental Studies Hybridoma Bank, University of Iowa, Iowa City),22 tropomyosin (ICN Biomedicals, Costa Mesa, CA), fast-twitch myosin (F59),23 α-actinin (Accurate Chemical and Scientific, Westbury, NY), and desmin (Ventana Medical Systems, Tucson, AZ). Unfixed skeletal muscle from the deltoids of one affected family member, one patient with Duchenne muscular dystrophy, and one control patient with a mild myopathy but with normal findings for all muscular dystrophy proteins were flash frozen in liquid nitrogen–cooled isopentane and sectioned on a cryostat (4 um). Cryosections were thawed directly on Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA). Tissues were blocked in phosphate-buffered saline containing 10% horse serum. Primary antibodies were diluted in blocking reagent and were applied for 2 hours at room temperature. For indirect immunofluorescence, secondary Cy3-conjugated goat antimouse or donkey antirabbit (Jackson Immunochemicals, West Grove, PA) were used according to the manufacturer’s instructions. For indirect immunohistochemistry, biotinylated secondary antibody was applied, followed by avidin/streptavidin-enzyme conjugate, and detection using a chromogenic enzyme substrate according to the manufacturer’s instructions (Ventana Medical Systems). All slides were analyzed using a Nikon (FXA) (Tokyo, Japan) microscope. All three biopsies were subjected to immunohistochemical analyses with ryanodine receptor, fast-twitch myosin, and tropomyosin antibodies. To control for specificity of ryanodine receptor immunostaining, immunohistochemistry using the ryanodine receptor antibody (34C) was performed on muscle biopsies from two patients with target fibers due to denervation.
Results.
Histopathology. Light microscopy. H-E studies on muscle biopsies from two affected family members (father and son) revealed myopathic features including variability in fiber size diameter, central nuclei, and mild fatty infiltration (data not shown). Gomori trichrome staining on both biopsies revealed nemaline bodies located mainly in the subsarcolemmal zones (figure 1, a through d). In some fibers, the rod bodies appeared to circle the cores faintly visible in trichrome sections. Rods were observed in 5 to10% of the myofibers in the father’s biopsy and in 20 to 25% of the myofibers in the son’s biopsy. ATPase reactions at alkaline pH showed striking type I predominance with over 95% of the fibers classified as type I. The NADH oxidative enzyme reaction revealed the presence of sharply demarcated cores in >85% of myofibers in both biopsies ( see figure 1, e through g). In one biopsy, the cores were predominantly singular and centrally located ( see figure 1, e and f). In the other biopsy, multiple cores were often seen in the same fiber, and were more peripherally located ( see figure 1g). In a few myofibers of both biopsies, rods and cores coexisted ( see figure 1, a through b, d).
Figure 1. Histologic visualization of rods and cores in the same biopsy. Gomori trichrome stains (a through d) and NADH oxidative reactions (e through h) on muscle sections from two different family members (son: a, b, e, f; father: c, d, g, h). (a through d) Note the presence of both rods (→) and cores (→) within several myofibers seen by Gomori trichrome stains. (e through h) Secions reacted to NADH-dehydrogenase show a marked reduction of oxidative enzyme activity in the cores, which contrasts sharply to the normal enzyme activity in the rest of the fiber. Magnification: a, b, d, f, h = ×200 before reduction; c, e, g = ×100 before reduction.
Electron microscopy.
Analysis of ultrastructure confirmed the presence of cores, rods, and myofibrillar disruption in subsarcolemmal areas. In the core region the structure of the muscle fiber appeared disorganized, the number of mitochondria were reduced, and the Z disks had a zigzag appearance (data not shown). Numerous membrane areas of the sarcoplasmic reticulum were seen (data not shown). Rod body clusters were found in subsarcolemma, paranuclear, and central locations. Rod bodies were often surrounded by a scattered sarcoplasmic reticulum (figure 2). The rods were largely separate from the cores. These features are consistent with the ultrastructure of cores typically seen in CCD and the rod bodies seen in nemaline myopathy families.24
Figure 2. Electron micrographs of a rod cluster located in a subsarcolemmal area. Sarcoplasmic membranes were also seen near the rods (a) 20,750 × (b)43,500.
Linkage analyses.
We conducted linkage analyses to determine if the family was linked to either of the two nemaline myopathy (NEM1) loci on chromosome 1 (TPM3–1q2125 or ACTA1–1q42.18) or the CCD RYR1 locus on chromosome 19q13.1.26,27⇓ The results of multipoint linkage analyses excluded linkage to the two nemaline rod loci on chromosome 1 (Zmax < −3). Multipoint logarithm of odds (lod) scores were consistent with a localization of the disease gene at or near the ryanodine receptor gene (RYR1) on 19q13.1 (Zmax = 2.5 at θ = 0.0).
Mutation screening of RYR1 gene.
To determine whether mutations in RYR1 cause CCD in this family, SSCP analysis was used to screen the complete coding sequence of the RYR1 gene. The first 62 exons were screened from genomic DNA while the remainder of the coding sequence (40 exons) was screened from muscle biopsy RNA (complementary DNA [cDNA]). We identified an aberrant conformer in exon 95 of the RYR1 gene in the affected family member using RT-PCR of muscle biopsy cDNA. Sequencing of the aberrant conformer revealed a heterozygous A–G point mutation, at cDNA nucleotide 13996 (A13996G), resulting in a threonine-to-alanine substitution at amino acid 4637 (Thr4637Ala) (figure 3a). Heterozygosity for the A13996G mutation was confirmed by re-amplification of exon 95 from genomic DNA and subsequent restriction enzyme digestion with HhaI (see figure 3b). The A13996G base change cosegregated with the disease in all affected family members (see figure 3b), and was not found in 100 unrelated control individuals (200 chromosomes) (data not shown), indicating that the Thr4637Ala change is not a benign polymorphism. Alignment of the human RYR1 protein sequence to those in mice, pigs, rabbits, and chickens show evolutionary conservation of the threonine amino acid residue (see figure 3c). The Thr4637Ala change found in the RYR1 gene is most likely the mutation causing central core/rod disease in this family.
Figure 3. Ryanodine receptor mutation identified in the family with congenital myopathy. (a) Identification of G13996A mutation (Thr4637Ala) by direct sequencing of an aberrant conformer on single-strand conformational polymorphism analysis of ryanodine receptor complementary DNA. (b) Cosegregation of the A13996G mutation with the disease shown by re-amplification of exon 95 from genomic DNA and digestion with HhaI. (c) Alignment of the human RYR1 protein sequence to those in mice, pigs, rabbits, and chickens shows evolutionary conservation of the threonine amino acid residue.
Immunohistochemistry.
To investigate the role of the RYR1 protein in the cores and whether the rod bodies in the central core/rod myopathy muscle showed features similar to rod bodies in nemaline myopathy, we performed immunofluorescence studies using antibodies to the ryanodine receptor, tropomyosin, fast-twitch myosin, α-actinin, and desmin. Biopsy controls were from an age-matched individual showing a mild myopathy but with normal findings for all muscular dystrophy proteins, and a patient with Duchenne muscular dystrophy. To control for specificity, ryanodine receptor antibodies were also tested on a muscle biopsy with target fibers due to denervation.
The expression pattern of tropomyosin was similar between the RYR1 patient and the patient with Duchenne muscular dystrophy and normal controls (figure 4). Antibodies to fast-twitch myosin (F59) yielded a characteristic checkerboard staining pattern in the patient with Duchenne muscular dystrophy and normal control subjects, representing a mixed population of type I and II myofibers. This was in contrast to the Thr4637Ala patient biopsy (see figure 4), which showed reduced immunostaining of type II fibers. The predominance of type I fibers seen by immunofluorescence was consistent with our ATPase enzyme histochemistry, and is typically reported for CCD biopsies.24
Figure 4. Central cores in a RYR1 (Thr4637Ala) patient contain excess ryanodine receptor protein. Cores strongly immunoreacted to ryanodine receptor antibodies (34C) in the RYR1 patient (top row) but not in the patient with Duchenne muscular dystrophy (middle row) and normal control subjects (bottom row). Note the absence of type II fibers using antibodies to fast-twitch myosin (F59), indicating a predominance of type I fibers in our RYR1 patient compared with the patient with Duchenne muscular dystrophy and normal control subjects. The expression pattern of tropomyosin (tpm) is similar between the RYR1 patient and the patient with Duchenne muscular dystrophy and normal control subjects. All muscle biopsies are from the deltoid.
Antibodies directed against the ryanodine receptor reacted to the cores in the Thr4637Ala patient indicating that the cores may contain an excess of ryanodine receptor protein (see figure 4). This pattern of staining was absent in the patient with Duchenne muscular dystrophy and normal control subjects. To control for specificity, we tested our ryanodine receptor antibody on muscle biopsies with target fibers that, like CCD biopsies, show central regions devoid of NADH stain (figure 5, top). The ryanodine receptor antibody (34C) did not immunostain the target fibers of two unrelated individuals (see figure 5, bottom). These results illustrate the immunospecificity of the ryanodine receptor antibody for cores in patients with CCD.
Figure 5. Ryanodine receptor antibodies do not immunostain target fibers. Top: NADH oxidative reaction of a biopsy with target fibers. Bottom: Immunohistochemical results of a target fiber biopsy using the ryanodine receptor antibody (34C). In contrast to the central cores seen in our RYR1 patients, target fibers with central regions devoid of NADH stain (top) do not immunoreact to ryanodine receptor antibodies (compare figure 5, bottom to figure 6, top). Similar results were seen using a different biopsy with target fibers from an unrelated individual (not shown).
Anti–α-actinin antibodies immunostained the rod bodies in our CCD family, indicating that they are similar in composition to the rods found in patients with nemaline myopathy (data not shown).28-30⇓⇓ Also consistent with the findings of previous studies,31 the cores in our CCD biopsies immuoreacted to desmin antibodies. In particular, desmin appeared concentrated at the periphery of the cores (figure 6, bottom). This was in contrast to the ryanodine receptor immunostaining, which showed a more uniform pattern of staining throughout the core (see figure 6, top) .
Figure 6. Immunoreactivity of cores to ryanodine receptor (top) and desmin (bottom) antibodies. Top: Excess ryanodine receptor protein appears on the periphery of the cores (arrows), as well as throughout the core (arrowheads). Bottom: Desmin protein appears concentrated on the periphery of central cores.
Discussion.
CCD and nemaline myopathy are generally considered to be two genetically and histologically distinct disorders. Some cases of CCD are caused by mutations in the ryanodine receptor and are histologically defined by the presence of distinctive cores within myofibers.1-4⇓⇓⇓ Some cases of autosomal dominant nemaline myopathy are caused by mutations in slow α-tropomyosin (TPM3),7 and others by a mutation of α-actin (ACTA1).8 Previously described cases showing both cores and rods have either been familial11,12⇓ or isolated,13,14⇓ but no mutation studies have been conducted. Here we report a large family with a dominantly inherited congenital myopathy whose myofibers show the simultaneous existence of central cores and nemaline rods on light and electron microscopy. Through genetic linkage studies, we excluded the involvement of the two nemaline rod myopathy loci on chromosome one (tropomyosin [TPM3] and α-actin [ACTA1]). We instead found suggestive linkage data to the CCD, ryanodine receptor RYR1, locus on chromosome 19 (lod = 2.5, θ = 0.0). We subsequently identified a novel Thr4637Ala mutation in the RYR1 gene.
The ryanodine receptor is a 564 kDa protein consisting of a large amino-terminal cytoplasmic domain and a smaller carboxyl-terminal transmembrane domain. The C-terminal domain of the receptor contains four highly hydrophobic regions that are believed to represent transmembrane α-helices which form the functional calcium release channel.32,33⇓ According to one topology model,32 the Thr4637Ala mutation we identified occurs within the C-terminal/transmembrane domain of the RYR1 protein, within a luminal loop flanked by transmembrane domains M2 and M3. A separate but similar topology model34 predicts the Thr4637Ala mutation in a luminal end of the proposed transmembrane sequence M6. Of the 22 previously reported RYR1 mutations,5 only one was reported to occur in the C terminal transmembrane/luminal region. This was an I4898T mutation found in a Mexican family with a particularly severe form of CCD.2 Studies in which this mutation was coexpressed with a normal RYR1 transcript resulted in reduced maximal levels of calcium release by 67%. Analysis of cotransfected cells showed a significantly increased resting cytoplasmic calcium level and a significantly reduced luminal calcium level. The authors suggested that the I4898T mutation causes a “leaky channel” in which the luminal stores of calcium leak into the cytoplasm. Based on the results of these expression studies, it is likely that the transmembrane/luminal Thr4637Ala mutation alters intracellular calcium levels in a similar fashion.
CCD is closely associated with susceptibility to malignant hyperthermia, and all patients with CCD are considered susceptible to malignant hyperthermia unless in vitro contracture tests (IVCT) are normal.35 IVCT were not performed in our family with the Thr4631Ala mutation, but 10 affected individuals underwent a total of 22 surgical procedures under general anesthesia without any episodes of malignant hyperthermia. This is similar to the Mexican kindred of CCD with a I4898T C-terminal mutation, in which IVCT results in two individuals were consistent with malignant hyperthermia susceptibility, but several affected family members were exposed to halothane anesthetics without complications.2 These results suggest that carboxy terminal mutations in the RYR1 gene may not confer sensitivity to malignant hyperthermia.
To date, the composition of the cores in CCD has remained elusive. In a previous attempt to characterize core composition, immunocytochemistry was conducted using a battery of antibodies to cytoskeletal proteins including dystrophin, spectrin, vinculin, desmin, vimentin, and myosin heavy chain.31 Antibodies to desmin immunostained the cores of several patient biopsies, thereby implicating desmin as playing a possible role in the pathogenesis of cores, though it is unknown whether this role is primary or secondary. The results of our immunofluorescence studies confirm the immunoreactivity of cores to desmin in patients with both cores and rods. Our immunofluorescence studies using antibodies to the ryanodine receptor suggest that the cores also contain excess ryanodine receptor protein. We suspect that RYR1 mutations affect processing and transport of the ryanodine receptor protein, causing it to accumulate within the cell. Our results suggest that the occurrence of nemalinelike rods can be a secondary feature of CCD, and that genetic studies on other core/rod families and individuals might be targeted to the RYR1 locus.
Acknowledgments
Supported by a graduate student fellowship from the Western Pennsylvania American Heart Association (P.C.S.). Supported by Australian National Health and Medical Research Council grant # 970104 (N.G.L., M.R.D.). E.P.H. is an Established Investigator of the American Heart Association, and holds the A. James Clark Chair in Molecular Medicine.
Acknowledgments
The authors thank Tommie McCarthy for supplying RYR1 oligonucleotides and Mike R. Erdos for help with constructing figures. The ryanodine receptor antibody developed by Judith Airey and John Sutko was obtained from the Developmental Studies Hybridoma Bank, developed under the auspices of the NICHD, and maintained by The University of Iowa, Department of Biological Sciences, Iowa City.
- Received March 16, 2000.
- Accepted August 24, 2000.
References
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.