© 2002 by European Society of Cardiology
Copyright © 2002, European Society of Cardiology
Re: The sinoatrial node, connexin distribution patterns and specific immunodetection of connexin45
National Heart and Lung Institute, Faculty of Medicine, Imperial College, London, UK
n.severs{at}ic.ac.uk
* Corresponding author. Cardiac Medicine, National Heart and Lung Institute, Imperial College of Science, Technology and Medicine, Royal Brompton Hospital, Sydney Street, London SW3 6NP, UK. Tel.: +44-20-7351-8140; fax: +44-20-7351-8476
Received 13 November 2001; accepted 30 November 2001
To the editor:
In a recent paper in Cardiovascular Research, Verheijck et al. [1] examine the important topic of the electrophysiological properties of the sinoatrial (SA) node in relation to the distribution of different connexin types. It is certainly pleasing to see that these authors' results and conclusions on connexin distribution in the mouse SA node in many respects parallel our earlier published findings in the rabbit [2]. Unfortunately, however, these parallels are largely concealed to the reader by the authors' confused representation of our work. Muddling up citations (as, for example, in their statement referring to our report of connective tissue between the atrium and nodal tissue; Ref. [2], not Ref. [3] as cited) would alone be too trivial a matter to merit comment. However, the authors go on to state that Coppen et al. [2] reported the rabbit SA node to be connexin43 (Cx43) positive when, in fact—just as in the study of Verheijck et al. [1] in the mouse—we found the central SA node to be Cx43 negative. The confusion is then compounded by a series of comments that appear to cast doubt on the reliability and specificity of the Cx45 antibody used in our work demonstrating the SA node to be connexin45 positive [2]. Such comments undermine the research effort we have devoted to re-appraising and advancing investigation of the spatial and temporal expression patterns of Cx45 in cardiovascular and other tissues [2–11]. From this perspective, we feel it imperative that the record be set straight.
The authors state that "Coppen et al. [3] showed that the Cx45 antibody they used cross-reacts with the same four-amino acid sequence of Cx43. Therefore, dual labelling which shows co-localisation of Cx43 and Cx45 immunostaining should be interpreted with great care. It cannot be ruled out that immunostaining with the anti-Cx45 antibody in these areas is due to cross-reaction with Cx43". In fact, the Cx45 antibody used by Coppen and colleagues to investigate Cx45 distribution manifestly does not react with Cx43; it was the anti-Cx45 antibody supplied by Chemicon that showed such cross-reactivity as does the anti-Cx45 antibody used by Verheijck et al. [1]. As no dual labelling is presented by Verheijck et al. [1], their comment on dual labelling appears to the reader to refer to the work of Coppen et al. cited (on Purkinje fibres; [3]) or, given the content of their paper, that on the SA node [2], where data from dual labelling for Cx45 and Cx43 formed a key element of the findings. As the Cx45 antibody used by Coppen et al. to carry out these dual labelling experiments is of proven specificity [3], any implication that the dual label data of Coppen et al. may be flawed is without foundation.
Let us clarify the background and evidence. The anti-Cx45 antibody [designated Q14E(GP42)] used by Coppen and colleagues to investigate connexin distribution in the SA node [2], and applied in a series of other studies on Cx45 [3–11], was developed and characterised by our laboratory [3]. This antibody was raised against a peptide corresponding to residues 354–367 of human Cx45, and when first applied 5 years ago, showed quite different labelling patterns from those previously reported for Cx45 in the heart, in particular revealing preferential distribution of Cx45 label in the ventricular conduction system but negligible or low levels of Cx45 label in the working ventricle [3,9]. By contrast, earlier studies using a Cx45 antibody raised to a different peptide sequence (corresponding to residues 285–298 of dog Cx45 which are identical in mouse and human Cx45) had been reported to give widespread labelling throughout ventricular myocardium [12–14] in common with the labelling patterns given by a Cx45 antibody marketed by Chemicon (catalogue No. AB1742) which had been raised against the same sequence as that used to produce the antibody used in the studies pre-dating the work of Coppen and colleagues. The peptide antigen to which this earlier antibody and its commercial counterpart was raised has a sequence of four amino acids (PPGY) which is also present in the Cx43 molecule, thereby raising the possibility of cross-reactivity of these antibodies with Cx43. Coppen et al. [3] demonstrated that the widespread labelling of working ventricle given by the Chemicon anti-Cx45 antibody is inhibited by concurrent incubation with a six-amino acid peptide corresponding to part of the Cx43 molecule in which the four-amino acid segment (SPPGYK) common to the Chemicon antigen is included, thereby confirming cross-reactivity of the Chemicon anti-Cx45 antibody with Cx43. Again, it must be emphasised that this non-specific Cx45 antibody has never been used by Coppen et al. to investigate the distribution of Cx45 in the SA node or any other tissues. Chemicon eventually admitted the cross-reactivity problem of their original Cx45 antibody when several years later they introduced new Cx45 antibodies which they had raised against the same peptide antigen as used by Coppen to generate the anti-Cx45 antibody Q14E(GP42) instead of their earlier antigen (Chemicon Communications Update, volume 9, number 4).
Verheijck et al. [1] go on to state that "the anti-Cx45 antibody in our study is from another source as the one used by Coppen et al." but "it is raised against the same peptide sequence". These statements need to be clarified. The antibody used by Verheijck et al. [1] was supplied by Alpha-Diagnostic International but was raised against the same sequence as the discredited Chemicon anti-Cx45 antibody, not the sequence used by Coppen et al. [3] to generate their Cx45 specific antibody, Q14E(GP42). Given the data presented by Coppen et al. [3] and the admission of cross-reactivity by Chemicon themselves, cross-reactivity of the Alpha-Diagnostic anti-Cx45 antibodies would be predicted as a distinct possibility on theoretical grounds alone and, as we report elsewhere, is indeed confirmed in practice [11]. Not surprisingly, the immunostaining patterns illustrated by Verheijck et al. [1] are consistent with cross-reaction of the Cx45 antibody they used; comparison of their Fig. 2A and B shows intensive labelling by the Cx45 antibody of the Cx43-rich regions above and below the SA node, but this receives no explicit comment in their paper.
On a more general note, publications reporting data using incompletely characterised anti-connexin antibodies are, unfortunately, on the increase rather than on the decline. Misinterpretation has become a very real pitfall for the unwary and inexperienced practitioner of connexin immunocytochemistry [11]. Commercial suppliers appear only too willing to sell supposedly isoform-specific connexin antibodies which, when subjected to even the most basic checks in the laboratory, are found to cross-react with other connexins. Users can be tempted to take any results obtained at face value, without critical evaluation of suppliers' unsubstantiated claims. The solution, as we have emphasised elsewhere [11], is characterisation of all connexin antibodies using a comprehensive range of techniques with valid positive and negative controls (e.g., transfected cells) and including immunogold electron microscopy. This is quite distinct from controls in which the primary antibody is omitted, pre-immune serum is substituted and competitive peptide inhibition applied—as conducted by Verheijck et al. [1]—which help check specificity of labelling, but do not confirm specificity of the antibody for its target protein. Successful peptide inhibition shows that the antibody binds to the peptide to which it was raised—it does not demonstrate specificity of detection for the connexin under investigation. As highlighted above, another connexin, or even an unrelated protein, may present an epitope that a sub-population of antibodies within an antiserum recognises. The act of purchase does not exonerate the user from demanding the same standard of evidence for specificity as would be required for an antibody made in their own laboratory. If this evidence is not forthcoming from the supplier, then, if the purchaser wishes to use the reagent, they must themselves undertake the necessary characterisation work.
 |
References |
|---|
 Top References |
|---|
- Verheijck E.E., Van Kempen M.J.A., Veereschild M., et al. Electrophysiological features of the mouse sinoatrial node in relation to connexin distribution. Cardiovasc Res (2001) 52:40–50.
[Abstract/Free Full Text] - Coppen S.R., Kodama I., Boyett M.R., et al. Connexin45, a major connexin of the rabbit sinoatrial node, is co-expressed with connexin43 in a restricted zone at the nodal-crista terminalis border. J Histochem Cytochem (1999) 47:907–918.
[Abstract/Free Full Text] - Coppen S.R., Dupont E., Rothery S., Severs N.J. Connexin45 expression is preferentially associated with the ventricular conduction system in mouse and rat heart. Circ Res (1998) 82:232–243.
[Abstract/Free Full Text] - Coppen S.R., Severs N.J., Gourdie R.G. Connexin45 (
6) expression delineates an extended conduction system in the embryonic and mature rodent heart. Dev Genet (1999) 24:82–90.[CrossRef][Web of Science][Medline] - Coppen S.R., Gourdie R.G., Severs N.J. Connexin45 is the first connexin to be expressed in the central conduction system of the mouse heart. Exp Clin Cardiol (2001) 6:17–23.
- Honjo H, Boyett MR, Coppen SR, et al. Heterogeneous expression of connexins in rabbit sinoatrial node cells: correlation between connexin isoform and cell size. Cardiovasc Res, in press.
- Kilarski W.M., Dupont E., Coppen S.R., et al. Identification of two further gap-junctional proteins, connexin40 and connexin45, in human myometrial smooth muscle cells at term. Eur J Cell Biol (1998) 75:1–8.[Web of Science][Medline]
- Ko Y.-S., Coppen S.R., Dupont E., Rothery S., Severs N.J. Regional differentiation of desmin, connexin43 and connexin45 expression patterns in rat aortic smooth muscle. Arterioscler Thromb Vasc Biol (2001) 21:355–364.
[Abstract/Free Full Text] - Vozzi C., Dupont E., Coppen S.R., Yeh H.-I., Severs N.J. Chamber-related differences in connexin expression in the human heart. J Mol Cell Cardiol (1999) 31:991–1003.[CrossRef][Web of Science][Medline]
- Dupont E., Ko Y.S., Rothery S., et al. The gap-junctional protein, connexin40, is elevated in patients susceptible to post-operative atrial fibrillation. Circulation (2001) 103:842–849.
[Abstract/Free Full Text] - Severs N.J., Rothery S., Dupont E., et al. Immunocytochemical analysis of connexin expression in the healthy and diseased cardiovascular system. Microsc Res Tech (2001) 52:301–322.[CrossRef][Web of Science][Medline]
- Kanter H.L., Saffitz J.E., Beyer E.C. Cardiac myocytes express multiple gap junction proteins. Circ Res (1992) 70:438–444.
[Abstract/Free Full Text] - Davis L.M., Rodefeld M.E., Green K., Beyer E.C., Saffitz J.E. Gap junction protein phenotypes of the human heart and conduction system. J Cardiovasc Electrophysiol (1995) 6:813–822.[Web of Science][Medline]
- Verheule S., van Kempen M.J., te Welscher P.H., Kwak B.R., Jongsma H.J. Characterization of gap junction channels in adult rabbit atrial and ventricular myocardium. Circ Res (1997) 80:673–681.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  N. J. Severs, A. F. Bruce, E. Dupont, and S. Rothery Remodelling of gap junctions and connexin expression in diseased myocardium Cardiovasc Res, October 1, 2008; 80(1): 9 - 19. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||