Connexin gene transfer preserves conduction velocity and prevents atrial fibrillation

Abstract

Background: Several lines of evidence have suggested that maintenance of atrial fibrillation (AF) depends on reentrant mechanisms. Maintenance of reentry necessitates a sufficiently short refractory period and/or delayed conduction, and AF has been associated with both alterations. Fibrosis, cellular dysfunction, and gap junction protein alterations occur in AF and cause conduction delay. We performed this study to test the hypothesis that gap junction protein overexpression would improve conduction and prevent AF.

Methods and results: Thirty Yorkshire swine were randomized into 2 groups (sinus rhythm and AF), and each group into 3 subgroups: sham-operated control, gene therapy with adenovirus expressing connexin (Cx) 40, and gene therapy with adenovirus expressing Cx43 (n=5 per subgroup). All animals had epicardial gene painting; the AF group had burst atrial pacing. All animals underwent terminal study 7 days after gene transfer. Sinus rhythm animals had strong transgene expression but no atrial conduction changes. In AF animals, controls had reduced and lateralized Cx43 expression, and Cx43 gene transfer restored expression and cellular location to sinus rhythm control levels. In the AF group, both Cx40 and Cx43 gene transfer improved conduction and reduced AF relative to controls.

Conclusions: Connexin gene therapy preserved atrial conduction and prevented AF.

Figure 1. Telemetry recordings

A. Representative ECG tracings in the AF group at day 7. Sinus rhythm percentage (SR%) was calculated using the formula. B. Summary data showing SR% for all animals in the AF group over time since gene therapy. See text for statistical analysis.

Figure 2. Conduction assessment

A. P-wave duration measured during SR 7 days after gene transfer. B. Intra-atrial conduction time measured while pacing the sinus node and recording at the tip of the left atrial appendage 7 days after gene transfer. C. Representative isochronal maps of left atrial epicardial activation time from optical mapping. Pacing site is marked. D. Summary data showing longitudinal and transverse conduction velocities from optical mapping. ** p<0.01 AF- vs. SR-controls; † p<0.05, †† p<0.01 between AF subgroups.

Figure 3. Atrial Histology

Representative images of atrial tissues stained with Hematoxylin and eosin (A) and Masson’s trichrome (B). C. Comparison of histological changes based on a previously reported, standardized scale. *p<0.05, **p<0.01 AF- vs. SR-controls; p=NS within AF and SR groups. D. Quantification of fibrosis from Masson’s trichrome-stained sections. p=NS among all animals.

Figure 4. Transgene expression

A. Representative immunoblots of Cx40 and GAPDH from control and AdCx40 animals. B. Quantification of band intensity for Cx40, normalized to GAPDH expression. C. Representative immunoblots of Cx43 and GAPDH from control and AdCx43 animals. P1 and P2 indicate phosphorylated Cx43 bands, and NP indicates non-phosphorylated Cx43. D. Quantification of band intensity for Cx43, normalized to GAPDH expression. *p<0.05, **p<0.01 AF- vs. SR-controls; † p<0.05, †† p<0.01 for subgroup comparisons withinAF- or SR-groups.

Figure 5. Connexin localization

A. Illustration of procedure to quantify connexin localization. Connexin location was determined using immunohistochemistry and confocal imaging. Only longitudinal sections of myocytes were evaluated. An investigator blinded to study group identified intercalated disk (white dotted boxes) of all myocytes in a high power field. We used image quantification software to quantify signal intensity and then calculated the ratio of intercalated disk connexin to total. B. Representative immunohistochemistry of Cx40 (yellow) and Cx43 (green) in the SR and AF groups. C. Quantification of signal intensity showing intact localization of Cx40 and Cx43 to the intercalated disk after gene transfer. **p<0.01 AF- vs. SR-controls. †p<0.05 within SR subgroups, ††p<0.01 within AF subgroups.

Figure 6. Cross-regulation of non-targeted connexins

A. Representative immunoblots of Cx43 and GAPDH from control and AdCx40 animals. B. Quantification of band intensity for Cx43 normalized to GAPDH expression inAdCx40 and control animals. C. Representative immunoblots of Cx40 and GAPDH from control and AdCx43 animals. D. Quantification of band intensity for Cx40 normalized to GAPDH in AdCx43 and controls animals. **p<0.01 vs. SR-controls.