Failure to Detect connexin43 Mutations in 38 Cases of Sporadic and Familial Heterotaxy
Background Heterotaxy results from failure to establish normal left/right asymmetry during embryonic development. Typical manifestations include complex heart defects and malpositioning of abdominal organs. Missense base substitutions clustered in a 150–base pair region of the gap-junction gene connexin43 (cx43) have been implicated in the pathogenesis of heterotaxy.
Methods and Results cx43 was studied in 38 cases of sporadic and familial heterotaxy. A 400–base pair region containing the previously reported mutation sites was amplified and directly sequenced in 19 patients. Nineteen additional patients were tested for restriction fragments predicted by two of the previously reported missense substitutions. No difference from normal control subjects was detected in any of the patients.
Conclusions Randomly selected cases of heterotaxy are unlikely to be the result of mutations in cx43.
All vertebrates establish internal left/right asymmetry during embryonic development. Failure to do so results in heterotaxy, a laterality defect characterized in humans by complex heart malformations, malposition of the abdominal viscera, and spleen abnormalities.1 Clinical and molecular studies of familial heterotaxy document variable expressivity, incomplete penetrance, and genetic heterogeneity.2 3 4 5 No genes definitively associated with human laterality defects have been isolated.
One approach to the identification of such genes may come from an understanding of molecular mechanisms establishing left/right asymmetry in other organisms. Among vertebrates, research with Xenopus and mice has characterized several genetic and nongenetic disruptions of normal left/right asymmetry.6 7 8 Recent experiments in chicks have uncovered asymmetrical expression along the left/right axis of at least four genes, one of which−sonic hedgehog−can randomize heart laterality if symmetrically distributed.9 10
Sonic hedgehog and its homologues serve a variety of functions in vertebrate embryogenesis. Other gene products also proposed to be essential for normal development are the connexins, a group of proteins that form intercellular gap junctions.11 Connexin43 (cx43) is one of the most thoroughly studied of the gap junction genes. It shows restricted spatial and temporal expression among and within several organ systems during murine development. In mouse embryos of gestation day 8.5, for example, cx43 expression cannot be detected in the heart but is quite prominent in the ventricle by day 10.5.12 Transgenic mice lacking cx43 die at birth due to right ventricular outflow obstruction.13
cx43 is also expressed in the human heart from at least fetal week 15 through adulthood.14 Recently, Britz-Cunningham et al15 speculated that mutations in cx43 may result in structural or functional heart defects in humans. They studied a heterogeneous group of 30 heart transplant patients by sequencing the 400–base pair (bp) portion of cx43 encoding the cytoplasmic tail of the protein, a region thought to contain many potential sites of posttranslational modification.16 One patient with familial atrial septal defect and six of six sporadic heterotaxy patients had cx43 mutations. Among the heterotaxy patients, five of six harbored an identical S364P missense substitution.
These results suggest that cx43 mutations may be a common cause of heterotaxy and that the S364P missense substitution occurs frequently among mutant cx43 chromosomes. Previously, we described briefly the absence of cx43 mutations in 18 cases of heterotaxy.17 Here, we report analysis of cx43 in 20 additional individuals with heterotaxy and summarize the results to date for all 38 patients studied.
Clinical information and material for analysis were provided by geneticists, cardiologists, and pathologists from institutions throughout North America and Europe. Informed consent was obtained from patients participating in this study, which was approved by the Institutional Review Board at Baylor College of Medicine. All of the patients carry the clinical diagnosis of heterotaxy based on the presence of a complex heart malformation and altered abdominal organ position (including abnormalities of spleen number or location) without features diagnostic of other identifiable syndromes or associations. Normal karyotypes were obtained from all individuals.
Polymerase Chain Reaction Amplification of cx43
DNA was extracted from whole blood or cell lines (lymphoblast or fibroblast) through the use of the Puregene DNA Isolation Kit (Gentra Systems, Inc) according to the manufacturer's protocol. Due to the presence of a processed cx43 pseudogene,18 a semi-nested polymerase chain reaction (PCR) was performed to amplify the expressed cx43 gene for subsequent analysis. Fig 1⇓ shows the relative location of the primers used in the study: cx43-1, 5′-GAAATACGTGAAACCGTTGG-3′ (derived from intronic sequence to avoid amplification of the pseudogene in the first round of amplification); cx43-2, 5′-CCTGGTGCACTTTCT-ACAGC-3′; cx43-3, 5′-AAAGAGCGACCCTTACCATGC-3′; and cx43-4, 5′-AAGCAAGTGAGCAAAACTGGGC-3′. First-round PCR was performed under the following conditions: 30 ng DNA, 1× Perkin-Elmer PCR buffer (200 mmol/L dNTPs, 1 mmol/L primers cx43-1 and cx43-2, 1.2 U Taq polymerase in 25 μL final volume) at 94°C×5 minutes for one cycle; 94°C×1 minute, 60°C×1 minute, and 72°C×1 minute for 30 cycles; and 72°C×7 minutes for one cycle. Second-round PCR was performed with 1 μL of 1:105 dilution of the first-round product amplified with primer cx43-2 combined with cx43-3 (if product was to be used for direct sequencing) or cx43-4 (if product was to be used for restriction enzyme digestion), with the final concentration of buffer, dNTPs, primers, and Taq polymerase identical to that of first-round PCR in a final volume of 50 μL. A touchdown protocol was used for the second-round PCR: 94°C×1 minute for one cycle; 94°C×1 minute, Z×1 minute, and 72°C×1 minute for one cycle, where Z ranged from 66°C to 51°C and decreased 1° each cycle; 94°C×1 minute, 50°C×1 minute, and 72°C×1 minute for 15 cycles; and 72°C×7 minutes for one cycle.
PCR products from amplification with primers cx43-3 and cx43-2 were purified after electrophoresis through a 2% agarose gel. Sequencing reactions were performed with a Sequenase Version 2.0 DNA Sequencing Kit (32P-dCTP) according to the manufacturer's instructions except for the addition of DMSO to a final concentration of 10% in the labeling reaction. Approximately 0.2 pmol of DNA and 10.0 pmol of primers were used in each reaction. Primers cx43-4 and cx43-7 (5′-TGCATGGGAGTTA-GAGATGG-3′) were used in separate sequencing reactions for each patient sample. Sequencing reactions were run on 6% denaturing polyacrylamide gels and autoradiographed at −70°C for 2 to 16 hours.
Five-microliter aliquots of PCR product from the second-round amplification with primers cx43-4 and cx43-2 were digested in separate reactions with 1 U Mnl I and Hae II in a total volume of 20 μL for 2 hours at 37°C. After digestion, the samples were electrophoresed through 2.5% low-melt agarose in Tris-acetate buffer (0.04 mol/L Tris-acetate and 0.001 mol/L EDTA) at 300 V/m (3 V/cm), stained with ethidium bromide, and photographed.
All sequences showed identity to the published cx43 cDNA sequence (HSCGJP, GenBank accession No. x52947) within the last 402 bp (nucleotides 904 to 1306) of the coding region. Fig 2a⇓ shows a portion of sequence from three heterotaxy patients spanning three of the six previously reported mutation sites, including S364P.
None of the 38 patients showed additional restriction sites for either Hae II (predicted by the T326A substitution) or Mnl I (predicted by the S364P mutation). Fig 2b⇑ shows a representative restriction-site analysis for 3 of the patients and 1 normal control subject. In all cases studied, the fragments of size and number predicted by the wild-type sequence were observed (see Fig 1b⇑).
In the present study, our aim was to determine the frequency with which regional or specific cx43 mutations occur among 38 heterotaxy patients. For 19 of the patients, the cx43 region coding for the protein cytoplasmic tail was amplified and directly sequenced. All of the nucleotide changes previously reported by Britz-Cunningham et al are contained within this portion of the gene. No base changes in the coding sequence were detected. Subsequently, these and the remaining 19 patients in the study population were assayed for gains in restriction enzyme sites, Hae II and Mnl I, predicted by two of the previously reported mutations, T326A and S364P, respectively. Only the restriction fragments predicted by the published cx43 sequence were detected. Other mutations in the cytoplasmic tail portion of cx43 cannot be excluded in those patients receiving only restriction enzyme site analysis.
Our results differ strikingly from those reported by Britz-Cunningham et al.15 We failed to detect mutations through direct sequencing in 19 of 19 heterotaxy cases studied compared with the previous report describing mutations in each of the 6 heterotaxy patients analyzed. In 19 additional heterotaxy cases, we failed to detect a specific mutation−S364P−that was reported in 5 of 6 patients from the previous study. Furthermore, Penman Splitt et al19 have found no evidence of cx43 mutations in 12 British patients with heart malformations and defects of laterality. Several explanations, alone or in combination, could account for these discordant findings, including ascertainment bias, types of tissue studied, and laboratory artifact leading to false-negative or false-positive results.
Ascertainment bias appears unlikely given the similarity of phenotypes in the two study populations. Our patient population comprises individuals of diverse ethnic background, including European, Latin American, and African American (the ethnic composition of the Britz-Cunningham group was not reported). The Table⇓ provides a summary of the phenotypic features of these patients. All carry the clinical diagnosis of heterotaxy, with heart malformations−usually complex and often fatal−combined with altered abdominal organ position. Fifteen have pulmonary outflow obstruction, a feature suggested by Britz-Cunningham et al15 to be correlated with the S364P mutation. At least 12 of the study patients are dead, as a result of the severity of their heart defects. Overall, the cases reported here encompass the ethnic diversity and range of phenotypes commonly seen among heterotaxic individuals. As a group, they do not appear to be substantially different from those analyzed by Britz-Cunningham et al.
We studied genomic DNA extracted from fresh peripheral blood, lymphoblasts, or fibroblasts, whereas Britz-Cunningham et al studied genomic DNA extracted from frozen heart tissue. Theoretically, somatic mutation of cx43 in the developing heart could occur, leading to mosaicism that would be undetectable through analysis of lymphocyte or fibroblast DNA. Although this possibility seems unlikely, it could be excluded easily through analysis of peripheral blood lymphocytes from individuals in whom apparent mutations have been identified in heart tissue.
Laboratory artifact leading to false-positive or false-negative results could account for the divergent findings reported here and elsewhere. Both possibilities could be addressed through a study of putative mutation-positive and mutation-negative cases in independent laboratories with the use of similar standard methods of mutation detection.
In summary, we failed to identify cx43 mutations in 38 sporadic and familial heterotaxy patients through direct sequencing in 19 cases and through restriction-site analysis in 19 other cases. We conclude that mutations in the terminal 400 bp of the cx43 coding region account for, at most, only a very small percentage of heterotaxy cases. An understanding of the underlying molecular defect for the majority of cases awaits further study.
Support from the Core Facilities of the Baylor Mental Retardation Research Center (grant P30 HD24064) is gratefully acknowledged. We thank Dr Glen Fishman for providing cx43 intronic sequence; Dr William Fletcher for unpublished data regarding previously identified cx43 base changes; Dr Jim Lupski, Dr Andrea Ballabio, and Dr David Nelson for critical reading of the manuscript; and Veronica Goytia for excellent technical assistance.
Guest editor for this article was Martina Bruekner, MD, Yale School of Medicine, New Haven, CT.
- Received December 27, 1995.
- Revision received April 8, 1996.
- Accepted April 24, 1996.
- Copyright © 1996 by American Heart Association
Brueckner M, McGrath J, D'Eustachio P, Horwich A. Establishment of left-right asymmetry in vertebrates: genetically distinct steps are involved. In: Bock GR, Marsh J, eds. Symposium on Biological Asymmetry and Handedness, Ciba Foundation Symposium 162. Chichester, UK: John Wiley; 1991:202-218.
Danos M, Yost H. Linkage of cardiac left-right asymmetry and dorsal-anterior development in Xenopus. Development. 1995;121:1467-1474.
Yokoyama T, Copeland NG, Jenkins NA, Montgomery CA, Elder FFB, Overbeek PA. Reversal of left-right asymmetry: a situs inversus mutation. Science. 1993;260:679-682.
Reaume A, de Sousa P, Kulkarni S, Langille B, Zhu D, Davies T, Juneja S, Kidder G, Rossant J. Cardiac malformation in neonatal mice lacking Connexin43. Science. 1995;267:1831-1834.
Fishman G, Eddy R, Shows T, Rosenthal L, Leinwand L. Molecular characterization and functional expression of the human cardiac gap junction channel. J Cell Biol. 1990;111:589-598.
Musil LS, Goodenough DA. Biochemical analysis of connexin43 intracellular transport, phosphorylation, and assembly into gap junctional plaques. J Cell Biol. 1991;115:1357-1374.
Penman Splitt M, Burn J, Goodship J. Connexin43 mutations in sporadic and familial defects of laterality. N Engl J Med. 1995;333:941-942.