Skip Navigation

Cardiovascular Research 2004 62(2):415-425; doi:10.1016/j.cardiores.2004.01.027
© 2004 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Camelliti, P.
Right arrow Articles by Green, C. R
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Camelliti, P.
Right arrow Articles by Green, C. R
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2004, European Society of Cardiology

Spatially and temporally distinct expression of fibroblast connexins after sheep ventricular infarction

Patrizia Camellitia, Gerard P Devlinb, Kenneth G Matthewsc, Peter Kohl*,a and Colin R Greend

aLaboratory of Physiology, University of Oxford, Parks Road, Oxford, OX1 3PT, UK
bDepartment of Cardiology, Waikato Hospital, Hamilton, New Zealand
cGrowth Physiology Department, AgResearch, Ruakura, Hamilton, New Zealand
dDepartment of Anatomy with Radiology, University of Auckland, New Zealand

* Corresponding author. Tel.: +44-1865-272114; fax: +44-1865-272554. Email address: peter.kohl{at}physiol.ox.ac.uk

Received 1 September 2003; revised 19 January 2004; accepted 20 January 2004


   Abstract
Top
 Abstract
1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
 References
 
Objectives: Myocardial infarction leads to extensive changes in the organization of cardiac myocytes and fibroblasts, and changes in gap junction protein expression. In the immediate period following ischemia, reperfusion causes hypercontraction, spreading the necrotic lesion. Further progressive infarction continues over several weeks. In reperfusion injury, the nonspecific gap junction channel uncoupler heptanol limits necrosis. We hypothesize that gap junction coupling and fibroblast invasion provide a substrate for progressive infarction via a gap junction mediated bystander effect. Methods: A sheep coronary occlusion infarct model was used with samples collected at 12, 24 and 48 h, and 6, 12 and 30 d (days) post-infarction. Immunohistochemical labelling of gap junction connexins Cx40, Cx43, and Cx45 was combined with cell-specific markers for fibroblasts (anti-vimentin) and myocytes (anti-myomesin). Double and triple immunolabelling and confocal microscopy were used to follow changes in cardiac myocyte morphology, fibroblast content and gap junction expression after myocardial infarction. Gap junction protein levels and fibroblast numbers were quantified. Results: Within 12 h of ischemia, myocyte viability is impaired within small islands in the ischemic region. These islands spread and fuse into larger infarct zones until 12 d post-infarction. Thereafter, surviving myocytes within the infarct and in the border-zone appear to become stabilized. Distant from the infarct, continuing myocyte disruption is regularly observed, even after 30 d. Cx43 becomes redistributed from intercalated discs to the lateral surface of structurally compromised myocytes within 12 d. Cx45 expressing fibroblasts infiltrate the damaged region within 24 h, becoming most numerous at 6–12 d post-infarction, with peak Cx45 levels at 6 d. Later, Cx43 expressing fibroblasts are observed, and the related Cx43 label increases over the 30 d observation period, even though fibroblast numbers decline after 12 d. Cx40 was only seen in vascular endothelium. Conclusions: Progressive infarction, identified by myocyte sarcomere disruption and subsequent cell loss, occurs in parallel with fibroblast invasion and gap junction remodeling. Two fibroblast phenotypes occur within infarcts, expressing either Cx43 or Cx45. Coupled fibroblasts may play a number of roles in tissue remodeling following myocardial infarction, including provision of a possible substrate for progressive infarction via a gap junction mediated bystander effect.

KEYWORDS Ischemia; Infarction; Gap junctions; In vivo; Bystander effect


   1. Introduction
Top
 Abstract
 1. Introduction
2. Materials and methods
 3. Results
 4. Discussion
 References
 
Myocardial infarction leads to complex changes in the three-dimensional organization of cardiac myocytes, fibroblasts, and gap junction proteins, which may contribute to reentrant arrhythmias [1,2]. Infarction causes a pronounced inflammatory response and lesion spread, mediated in part via gap junctions, as highlighted by the fact that the nonspecific gap junction channel uncoupler heptanol prevents progression of hypercontracture and limits necrosis and infarct size [3,4]. The role of non-myocytes in this process is not clear. We used a sheep coronary occlusion ischemic infarct model for immunohistochemical labelling and confocal laser scanning microscopy to study the time-course and extent of changes in cardiac myocyte morphology, fibroblast content, and gap junction remodeling. Using double and triple immunolabelling, we correlated expression of the three major cardiac connexins, Cx43, Cx45 and Cx40, with positive cell type identification of myocytes (myomesin label) and fibroblasts (vimentin label), during a period of 12 h–30 d post-infarction. Our results indicate that at least two fibroblast phenotypes invade the infarct region, expressing different connexin proteins. We suggest that gap junction remodeling and fibroblast invasion may contribute to progressive infarction via a gap junction mediated bystander effect.


   2. Materials and methods
Top
 Abstract
 1. Introduction
 2. Materials and methods
3. Results
 4. Discussion
 References
 
2.1. Animal model
Sheep infarct tissue was obtained as previously described [5]. Briefly, infarcts were induced in 13 sheep (2–4 years old) by occlusion of the left coronary artery via injection of 200 mg sterile gel foam (Upjohn, USA) suspended in 5 ml sterile saline, delivered through a perfusion catheter positioned in either the left anterior descending or circumflex artery. This regularly induced transmural myocardial infarcts. Only animals with left ventricular ejection fraction of ≤40% (determined echocardiographically) were included. Hearts were excised at 12, 24, 48 h, 6, 12 and 30 d post-infarction, and sectioned in blocks containing viable tissue distant from the infarct, infarct border tissue, and main infarct samples. Tissue blocks were embedded in Tissue-Tek (Miles, USA), frozen in liquid nitrogen, and stored at –80 °C. For reference, left ventricular samples from two normal sheep hearts were used.

The investigation conforms to the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes.

2.2. Antibodies
For Cx43 labelling, we used mouse monoclonal and rabbit polyclonal antibodies, raised against a short peptide sequence (amino acids 131–142) of the protein cytoplasmic loop [6,7]. For Cx40, a guinea pig polyclonal antibody, raised against an oligopeptide matching the cDNA sequence of the protein carboxyl-terminus (amino acids 256–270), was used [8,9]. For Cx45, we used a commercially available rabbit polyclonal antibody (AB1745; Chemicon, USA) that recognizes the carboxyl-terminal amino acids 354–367 [8,10]. The M-lines in sarcomeres of myocytes were labelled using mouse anti-myomesin antibodies (clone B4) [11,12]. Fibroblasts were labelled using a monoclonal mouse anti-vimentin antibody (clone V9, Sigma Aldrich, USA) [13,14], which labels intermediate filaments (this antibody also labels endothelial cells, lymphoid cells and melanocytes).

2.3. Immunolabelling
Cryosections (16 µm), cut parallel to the prevailing myocyte direction, were collected on SuperFrost® slides (Menzel-Glaser, Germany) and stored at –80 °C. Sections were thawed, dried and blocked for 1 h at room temperature (RT) in 10% goat serum/0.3% bovine serum albumin (BSA) in PBS containing 0.1% Triton X-100.

For single labelling, slides were incubated overnight (O/N) at 4 °C with primary antibodies (diluted in PBS/0.3% BSA) and with CY3- (Jackson Immuno Research Laboratories, USA) or Alexa488- (Molecular Probes, USA) conjugated secondary antibodies (2 h RT). A washing step in PBS was performed between antibody incubations.

For double labelling of Cx43 and Cx45, sections were incubated with the primary rabbit anti-Cx45 (1:100, O/N 4 °C) and the secondary goat anti-rabbit Alexa488 (1:200, 2 h RT) and subsequently with the primary mouse anti-Cx43 (1:100, O/N 4 °C) and the secondary goat anti-mouse CY3 (1:300, 2 h RT).

For double labelling of myocytes and fibroblasts, sections were incubated with a mix of mouse monoclonal anti-myomesin (1:100) and anti-vimentin antibodies (1:10,000) O/N at 4 °C and with the secondary goat anti-mouse CY3 (1:500, 2 h RT).

For triple labelling experiments, slides were incubated with anti-connexin antibodies (rabbit anti-Cx43 (1:100), rabbit anti-Cx45 (1:100) or guinea pig anti-Cx40 (1:200) O/N at 4 °C) and secondary goat anti-rabbit Alexa568 (1:500), goat anti-rabbit Alexa488 (1:200) or goat anti-guinea pig Alexa488 (1:500) for 2 h (RT), then with a mix of mouse anti-myomesin (1:100) and mouse anti-vimentin (1:10,000) antibodies O/N (4 °C) and secondary goat anti-mouse Alexa488 or CY3-conjugated antibodies (1:500) for 2 h (RT).

For triple labelling experiments of mouse anti-Cx43 and cell types, Molecular Probes' Zenon immunolabelling kits (Molecular Probes) were used, enabling application of multiple pre-conjugated mouse monoclonal antibodies in the same protocol. After blocking for 1 h (RT) in PBS containing 1% BSA/0.1% Triton X-100, sections were incubated with mouse anti-vimentin antibodies (1:10,000, O/N 4 °C) and secondary donkey anti-mouse CY5 (1:300, 2 h RT, Jackson Immuno Research Laboratories). Subsequently slides were incubated with a mix of the fluorochrome conjugated mouse anti-myomesin-Zenon Alexa568 (1:100) and mouse anti-Cx43-Zenon Alexa488 (1:100) O/N at 4°C.

Control experiments involved omission of primary antibodies, and incubation with inappropriate secondary antibodies to exclude cross-reactivity. No connexin-related fluorescence was detected in the absence of primary antibodies. Positive control experiments involved immunolabelling, using the same antibodies, of rabbit ventricular and sinoatrial node tissue sections.

2.4. Confocal microscopy
Immunolabelled sections, mounted in Citifluor mounting medium (Agar Scientific, UK), were examined on a TCS-SP2 confocal laser-scanning microscope (Leica Microsystems, Germany) using 488 nm excitation and 500–535 nm emission for Alexa488 labelling, 543 nm excitation and 555–630 nm emission for CY3 and Alexa568 labelling, and 633 nm excitation and 650–750 nm emission for CY5 labelling. Single optical slices or z-series were recorded. Where multiple labelling was used, images were combined to reveal localization of gap junctions with respect to myocytes and fibroblasts, or different connexin isoforms.

2.5. Quantification of connexin immunolabelling and fibroblast density
For connexin quantification, single optical slice images showing Cx43 or Cx45 immunolabelling in infarct areas devoid of myocytes were obtained from 6, 12 and 30 d sheep infarcts (four, three and one animals, respectively). Images were quantitatively analysed using ImageJ1.29 (http://rsb.info.nih.gov/ij/). Cx43 fluorescent spots were counted in eight regions from 6 d, nine regions from 12 d, and eight regions from 30 d infarcts. Cx45 fluorescent spots were counted in 30 regions from 6 d, 38 regions from 12 d and 9 regions from 30 d infarcts. Using a threshold value algorithm, only spots of sufficient intensity (assumed to represent gap junctions) were included. Total area occupied by connexin immunoreactivity in each single image was also calculated. Connexin density was expressed as (i) the cumulative number of gap junctions per tissue area, and (ii) the area occupied by connexin fluorescence relative to the total area scanned.

For fibroblast density quantification, fibroblast-labelled confocal images were obtained from infarct areas of 6, 12 and 30 d infarcts. Fibroblast density was calculated as the area occupied by vimentin fluorescence relative to the total image area in 21 regions from 6 d, 12 regions from 12 d and 13 regions from 30 d samples, and expressed as fibroblast-occupied percentage area.

Statistical comparison for connexin density or fibroblast density at the three time points were made using one-way analysis of variance (ANOVA); tests of significance between groups were performed using Dunn's multiple comparison tests in Prism software (GraphPad San Diego, USA). Data are presented as mean±S.E.M. All quantification was carried out by the same person (PC) to avoid inter-person variability.


   3. Results
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
4. Discussion
 References
 
3.1. Cell type identification and morphology
Cell spatial interrelation and morphology were investigated using anti-vimentin and anti-myomesin antibodies. In addition, myomesin labelling of M-lines allowed assessment of myocyte viability: a regular banding pattern is typical of normal myocytes (Fig. 1A), and irregular, disrupted patterns indicate infarcted/damaged cells (Fig. 1B).


Figure 1
View larger version (140K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Progressive infarction. Myocytes (striated cells, M) and fibroblasts (bright elongated cells, F) in normal ventricular myocardium (A), and at 24 h (B), 48 h (C), 6 d (D), 12 d (E) and 30 d (F) post-infarction (sheep heart). In normal myocardium, M-lines show a regular banding pattern (A). Within 24 h of infarction the myomesin pattern indicates sarcomere disruption (B). By 48 h, fibroblasts invade the infarct and areas of disrupted myocytes enlarge (C). By 6 d, the infarct is densely packed with fibroblasts (D) and only few islands of myocytes are left. In the border-zone, strands of disrupted myocytes extend from the myocardium into the fibroblast-rich infarct. By 12 d, fibroblast density is still high and myocyte islands have almost disappeared from the infarct. Border-zone myocytes show sarcomere disruption (E), indicating continued lesion spread. By 30 d, fibroblast density has decreased in the infarct, and collagen deposition is increased. Only very few islands of healthy myocytes are left within the infarct; border-zone myocytes appear to be viable cells with a regular myomesin pattern. Scale bar: 20 µm.

 
Sarcomere disruption occurs within 12 h of infarction in small patches only. By 24 h, these patches enlarge (Fig. 1B), indicating continuing cell damage. Fibroblast invasion is already present at this time point, although it is difficult to delineate the infarct area on the basis of fibroblasts alone. At 48 h, the islands of disrupted myocytes are further enlarged or merged, and fibroblast invasion continues (Fig. 1C).

At 6, 12 and 30 d post-infarction three distinct zones can be identified: the core infarct area, a border-zone containing a mix of disrupted and viable tissue (Fig. 1D–F), and viable tissue away from the infarct zone (Fig. 2D–F).


Figure 2
View larger version (143K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 (A–C) Fibroblast labelling in the infarct area at 6 d (A), 12 d (B) and 30 d (C) post-infarction. Fibroblasts become very dense within 6 d (A), remaining high at 12 d (B), but showing a significant decrease again by 30 d (C). (D–F) Myocyte and fibroblast co-labelling of viable tissue distant from the infarct area in 6 d (D), 12 d (E) and 30 d (F) infarcts shows an increase in fibroblast content between healthy myocytes with infarction time. Scale bar: 20 µm. (G) Quantification of fibroblast density in the infarct area of 6, 12 and 30 d infarcts. Fibroblast density at 30 d is significantly smaller than that observed at 6 and at 12 d (P<0.001, P<0.01, respectively).

 
By 6 d, the infarct is densely packed with fibroblasts. Remaining islands of myocytes largely show disrupted myomesin label. A distinct border-zone shows strands of disrupted myocytes with interstitial fibroblasts that extend from viable myocardium into the fibroblast-rich infarct (Fig. 1D). By 12 d post-infarction, the lesion contains very few remaining myocytes, and fibroblast density, while still high, appears to decline. Border-zone myocytes show sarcomere disruption (Fig. 1E) indicating continuing spread of the lesion. By 30 d post-infarction, the infarct zone contains very rare patches of viable myocytes that show no sign of sarcomere disruption. Fibroblast density decreases further, while simultaneously there is significant collagen deposition. Most surviving border-zone myocytes show the regular banding pattern of normal muscle cells (Fig. 1F).

Quantitative assessment of the infarct area at 6, 12 and 30 d post-infarct (Fig. 2A–C) shows that, after the initial fibroblast invasion, fibroblast density decreases (6 d: 18.1±0.9% of image area show vimentin label, n=21; 12 d: 14.5±0.8%, n=12; 30 d: 8.0±0.7%, n=13), with significant reductions from 6 to 30 d and 12 to 30 d (P<0.001, P<0.01, respectively; Fig. 2G).

Tissue distant from the infarct of 6, 12 and 30 d post-infarct animals shows a significant progressive increase in fibroblast numbers and interstitial collagen deposition (Fig. 2D–F), consistent with previous studies showing excess fibrous tissue in non-infarcted areas after myocardial infarction [15]. Surprisingly even after 12 and 30 d, occasional foci of disrupted myocytes were found in tissue distant from the infarct (not shown), indicating continued cell death.

3.2. Connexins and cell type labelling
3.2.1. Connexin43
Triple labelling experiments for Cx43, myomesin and vimentin allowed us to relate Cx43 expression patterns to underlying cell types. Cx43 becomes disorganized in myocytes with disrupted sarcomere striation as early as 24–48 h post-infarction. No longer exclusively in intercalated discs, it acquires a more dispersed pattern along the myocyte surface. By 6 d when the border-zone region is clearly identifiable, Cx43 is undergoing complex remodeling, thought to contribute to reentrant arrhythmias [16,17]. Fig. 3 illustrates Cx43 label in viable ventricular myocardium distant from the infarct at 6 d (Fig. 3A), and in the border-zone at 6 d (Fig. 3B), 12 d (Fig. 3C) and 30 d (Fig. 3D). Cx43 is organized in the typical intercalated disc pattern in viable tissue. In the border-zone the number of intercalated discs decreases with time, with few intercalated discs left at 6 d (Fig. 3B), and none at 12 or 30 d (Fig. 3C,D).


Figure 3
View larger version (199K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Cx43 expression in control tissue and infarct border-zone. Triple labelling for Cx43 (green), myomesin (striated red) and vimentin (red in A–C, blue in D). In the viable ventricle (A) Cx43 is restricted to intercalated discs between myocytes. In the infarct border-zone at 6 d post-infarct (B), some intercalated discs remain, but Cx43 is becoming increasingly dispersed along the longitudinal sides of myocytes (arrow). By 12 d post-infarction few intercalated discs remain in the border-zone (C), and at 30 d post-infarct Cx43 is mainly on the longitudinal membrane of the myocytes which border the infarct (D). Fibroblasts in the border-zone express Cx43 at 6 d (B), 12 d (C) and 30 d (D) infarcts (arrowheads). Scale bar: 20 µm.

 
Fibroblasts in the viable muscle region distant from the infarct (and fibroblasts of normal ventricular control tissue) show no Cx43 labelling (Fig. 3A). However, Cx43 is present in fibroblasts facing the border-zone at all time points analysed (6, 12 and 30 d, Fig. 3B–D) and in some cases these come into close apposition with border-zone myocytes.

In the infarct region itself, we found significant levels of Cx43, associated with islands of both sarcomere-disrupted and healthy myocytes. At 6 d post-infarction, numerous islands of disrupted myocytes were found in the infarct, with Cx43 still partially organized in intercalated discs (Fig. 4A). At 12 d, few islands of disrupted myocytes remain in the infarct region, with Cx43 in a punctate pattern around the myocyte surface (Fig. 4B). By 30 d, no islands of disrupted myocytes were found. Instead, we observed occasional groups of healthy myocytes in the infarct region. These were characterised by clear sarcomere patterns and showed Cx43 in intercalated discs (Fig. 4C). It appears, therefore, that the overwhelming majority of myocytes in the infarct succumb to continued Cx43 remodeling, involving disruption of intercalated discs, which is largely completed by 30 d post-infarction.


Figure 4
View larger version (166K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 Cx43 in the infarct area. (A–C) Cx43 is associated with islands of myocytes (M) in the infarct. At 6 d post-infarction, islands of disrupted myocytes express Cx43, still partially organized in intercalated discs, but also undergoing remodeling (A). By 12 d post-infarction, islands of disrupted myocytes have almost disappeared and Cx43 shows an unorganized punctate pattern associated with the few remaining islands (B). At 30 d post-infarction, no islands of disrupted myocytes are seen. Very occasional sites of healthy myocytes express Cx43 in intercalated discs (C). (D–F) In areas of the infarct devoid of myocytes at 6 d (D), 12 d (E) and 30 d (F), Cx43 has a punctate pattern and levels increase significantly with post-infarction time (compare D with F). (G–I) Triple labelling for Cx43 (green dots), myocytes and fibroblasts (red, F) at 6 d (G), 12 d (H) and 30 d (I) reveals that Cx43 is expressed by fibroblasts in the infarct region. Scale bar: 20 µm.

 
Surprisingly, in infarct areas devoid of myocytes, we still observed Cx43. Single labelling experiments for Cx43 show a well-defined punctate pattern of Cx43 (Fig. 4D–F), which is distributed apparently homogeneously throughout the entire infarct area. Both numbers and area of Cx43 spots increase with time after infarction (6 d: 2,940±211.2 spots/mm2, 2,542±224.4 µm2/mm2, n=8; 12 d: 6,377±838.5 spots/mm2, 4,738±484.7 µm2/mm2, n=9; 30 d: 11,670±2,157 spots/mm2, 11,820±1,854 µm2/mm2, n=8; Fig. 6A,B). Triple labelling experiments for Cx43, myomesin and vimentin, show that Cx43 in infarct zones devoid of myocytes is associated with fibroblasts (6, 12 and 30 d, Fig. 4G–I).


Figure 6
View larger version (27K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6 (A, B) Quantification of Cx43 (solid bars) and Cx45 (open bars) immunolabelling in the infarct area from single label experiments. Connexin density is expressed as number of gap junctions per tissue area (spots/mm2; A) and as area of gap junction fluorescence per tissue area (µm2/mm2; B). Cx43 density increases with infarction time and reaches the highest levels at 30 d, where it is significantly greater than at 6 and 12 d (P<0.001, P<0.05, respectively). In contrast, Cx45 density decreases progressively from 6 to 12 d and 30 d (P<0.001, P<0.05, respectively). Both connexins have similar levels at 6 d. At 12 d, Cx43 is significantly greater than Cx45 and this difference becomes more evident by 30 d (P<0.001). (C) Double labelling for Cx43 (red dots) and Cx45 (green dots) in the infarct area of a 12 d post-infarct heart. Cx43 and Cx45 are both present in the infarct area but generally do not co-localize (virtually no yellow spots). Scale bar: 20 µm.

 
3.2.2. Connexin45
Cx45 is absent from normal ventricular myocardium of control sheep, but is found associated with fibroblast aggregates in regions of infarcted tissue as soon as 24 to 48 h after infarction (Fig. 5A,B). It is also found at 6, 12 and 30 d, homogeneously distributed throughout the entire infarct, where it is associated with fibroblasts and apparently involved in fibroblast–fibroblast coupling (Fig. 5C,D for 6 d; E,F for 12 d; G,H for 30 d). These Cx45-expressing fibroblasts are often found in close apposition with myocytes bordering the infarct and could potentially be involved in myocyte-fibroblast coupling. Single immunolabelling for Cx45 shows initial high levels of this connexin (6 d: 3,085±204.5 spots/mm2, 1,805±104.9 µm2/mm2, n=30; Fig. 5C; Fig. 6A,B), followed by progressive reduction during post-infarction period (12 d: 1,372±31.2 spots/mm2, 683.8±27.1 µm2/mm2, n=38; 30 d: 396.2±55.6 spots/mm2, 175.9±22.6 µm2/mm2, n=9; Fig. 5E,G; Fig. 6A,B).


Figure 5
View larger version (75K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5 Cx45 at the border-zone of ischemic infarct in sheep heart. (A, B) Single Cx45 label (A) and triple labelling for Cx45 (green dots), myocytes and fibroblasts (B) at 48 h post-infarction. Cx45 is between fibroblasts (arrowheads) but is also present at the infarct border-zone. (C–H) Cx45 in the infarct area at 6 d (C, D), 12 d (E, F) and 30 d (G, H) post-infarct. Single labelling for Cx45 (green dots, C, E, G) shows a progressive reduction in Cx45 expression levels during the infarction period. Triple labelling for Cx45, myocytes and fibroblasts (D, F, H) shows that Cx45 is associated with fibroblasts in the infarct. Very little Cx45 remains by 30 d (G, H). Scale bar: 20 µm.

 
3.2.3. Connexin43 and connexin45
Single connexin immunolabelling results have demonstrated that Cx43 and Cx45 are both expressed in fibroblast-rich infarct areas at 6, 12 and 30 d, albeit with different expression dynamics (Cx45 decreasing and Cx43 increasing over time).

Double immunolabelling for Cx43 and Cx45 in infarct tissue devoid of myocytes in 12 d post-infarct hearts, where both connexins are well represented, shows virtually no co-localization of Cx43 and Cx45 (Fig. 6C). Thus there are two temporally and spatially distinct connexin isoforms expressed by fibroblasts within the infarct.

3.2.4. Connexin40
The only significant Cx40 label found in working ventricular myocardium from control sheep, or in viable myocardium distant from the infarct site, was associated with blood vessel endothelial cells (Fig. 7A). It is not detected in infarct border-zones or within the infarct up to 6 d post-infarction. By 12 d, an increase in Cx40 in blood vessel and capillary endothelial cells was observed in the infarcted tissue (Fig. 7B), probably owing to neo-angiogenesis as part of the repair process.


Figure 7
View larger version (107K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7 Cx40 expression in control tissue (A) and in the infarct area of a 12 d infarct (B). Triple labelling for Cx40 (green), myocytes and fibroblasts shows that Cx40 is associated with blood vessel and capillary endothelial cells. By 12 d post-infarction, vascularization of the infarct increases (B, arrows). Scale bar: 10 µm.

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
References
 
In this study, we correlate myocyte and fibroblast morphology changes with gap junction distribution and density within a time window from 12 h to more than 4 weeks after transmural myocardial infarction in a sheep coronary occlusion model. We show that disruption of the myocyte sarcomeric structure occurs as early as 12 h, initially affecting only small patches, which subsequently enlarge and merge. Contractile apparatus disruption coincides with fibroblast infiltration of the ischemic region that commences within the first 24 h and increases throughout the first week. Sarcomere disruption is ongoing at 12 d and although the infarct itself and the infarct border-zone appear to stabilize by 30 d post-infarction, foci of disrupted myocytes within the viable myocardium distant from the main infarct site are still seen. Although there was only one 30 d animal in this study, our results suggest that cell damage is still continuing at 4 weeks after the ischemic insult, consistent with previous studies of a sheep model of microinfarction reporting apoptotic myocytes in both infarct border-zone and remote viable muscle after 6 weeks [18].

However, not all the myocytes within the infarct area become disrupted, some survive and appear as islands of viable cells, with Cx43 in typical intercalated discs. In contrast to prior reports [16,19], we did not detect any myocyte strands that would have connected these islands with the surrounding healthy myocardium.

Myocyte sarcomere disruption also correlates with progressive Cx43 remodeling, which begins during the first 24 h after infarction and becomes more pronounced by 12 and 30 d post-infarction. Cx43 is normally organized in intercalated discs between myocytes. After infarction, it acquires a more dispersed pattern along the myocyte surface in infarcted tissue within 24–48 h. By 30 d, Cx43 appears very disorganized in infarct border-zone myocardium, with no identifiable intercalated disc-like structures, consistent with earlier reports [16,20,21].

In contrast to previous studies [16,22], we find connexins in the central infarct area. Possible explanations for this include discrepancies in infarct age and species differences in connexin isoforms expression. A significant proportion of these connexins is found among fibroblasts in the infarct area. Cx45 is expressed by early-infiltrating fibroblasts, which appear within 24 h between disrupted myocytes. Cx45 expressing fibroblasts were also found to invade the interstitial space in viable myocardium, where they may subsequently contribute to myocardial stiffening. Fibroblast-related Cx45 increases during the first week post-infarction, and then declines. The increase in fibroblast-related Cx43 levels starts later and continues over the 30 d study period, by which time scar tissue is being deposited. Thus cardiac fibroblasts in sheep heart infarcts express two different connexins with separate dynamics, consistent with the notion that infarct scars are living tissue, where fibroblasts are involved in many different activities [15,23,24].

More significantly, Cx43 and Cx45 do not generally co-localize in the infarct zone, suggesting that different fibroblast phenotypes are involved in post-infarct remodeling. Furthermore, these appear, in part, to be distinct from the fibroblast populations in rabbit sinoatrial node, which express Cx40 and Cx45, but not Cx43 [25]. We were not able to differentiate whether they are changing fibroblast cell populations, as opposed to a single fibroblast population, expressing initially Cx45, and later on Cx43 (although for the lack of co-localization this is less likely). Also, vimentin labelling will not separate fibroblasts from myofibroblasts, for example. Still, two distinct fibroblast populations could have quite different roles during myocardial remodeling including the regulation of proliferation, migration, extracellular matrix remodeling, and the production of specific cytokines and growth factors [26]. Direct cell–cell coupling will also contribute to scar tissue remodeling, and may imply an involvement of fibroblasts in nutrient and metabolite transfer. An efficient fibroblast metabolism seems important for efficient synthesis and deposition of new connective tissue components during infarct scarring, as has been hypothesized for the wound healing process [27]. Finally, homotypic and heterotypic channels may differently modulate intercellular communication between cells in the infarct and at the border-zone where these connexins are expressed, but despite recent studies showing that heterotypic channels may occur in cell lines transfected with DNA encoding for Cx43 or Cx45 [28–30], our co-localization studies provided no support for these in vivo.

Cx40 was first observed at 6 d in the infarct region, in parallel with angiogenesis (which has been reported to begin 4 d after infarction [31]).

In this study, we have not specifically identified fibroblast–myocyte coupling, which does occur in the rabbit sinoatrial node [25]. Fibroblast–myocyte coupling could have a significant effect on the electrophysiology in the infarct [32] and could interfere with the spread and regulation of excitation in areas such as the infarct border-zones that are known for their highly irregular electrical properties [33]. Heterogeneous coupling might also hold the key to integrating surviving groups of myocytes inside the infarct region with surrounding tissue, since—at least in vitro—cardiac fibroblasts may bridge gaps between myocytes of up to 300 µm [34]. Fibroblasts are also mechano-sensitive and, when subjected to transient stretch in cardiac scar models, induce local depolarization and trigger action potentials in connected myocytes [35].

An increase in Cx43 during early phases of repair could be explained by increasing fibroblast infiltration. This would parallel skin wounds where fibroblasts also penetrate the wound within the first 3 d and reach a maximum at around 5–7d post-wounding [36]. The high expression level of Cx43 at 30 d post-infarction does appear surprising in the light of the overall reduction in fibroblast density during maturation of the granular tissue. In contrast, the levels of Cx45 correlated with the early increase and subsequent reduction in fibroblast content (and healthy myocyte numbers in the infarct area and border-zone). This rapid infiltration of potentially highly coupled fibroblasts would provide a substrate for a bystander effect, which could contribute to the spread of remodeling-related signals from infarcted myocytes to surrounding tissue via gap junctions. Progressive infarction may therefore parallel the bystander effect reported in neural tissues, in which upregulated and highly coupled astrocyte populations contribute to spread of the neuronal damage [37,38]. This is also in keeping with the finding that Cx43-deficient mice develop smaller infarcts than wild-type mice following coronary occlusion [39].

Intercellular coupling has been reported to play a role in lesion spread that occurs as a result of reperfusion injury during the first few minutes of blood flow restoration after ischemia. In this case, the nonspecific gap junction channel uncoupler heptanol prevents hypercontracture and limits necrosis [3]. The model we have used here is not a reperfusion model; nonetheless, it would appear that the initial necrosis resulting from ischemia is increased with gap junction-mediated hypercontracture upon reperfusion, and our results suggest that a substrate exists for a gap junction-mediated bystander effect on progressive infarction. Reducing Cx43 and/or Cx45 expression after myocardial infarction, ideally targeted at cardiac fibroblasts, might limit cell-to-cell mediated propagation of damage and thus limit the extent of cardiac dysfunction.


   Acknowledgements
 
We are grateful to David Becker, University College London, Steven R Coppen and Nick Severs, Imperial College London, Robert G Gourdie, MUSC Charleston, and Hans M. Eppenberger, ETH Zürich for providing antibodies. This study was conducted at the Auckland University Biomedical Imaging Research Unit. Supported by the British Heart Foundation (#98035), the UK Medical Research Council (#49498) and the UK Biotechnology and Biological Sciences Research Council (#18561). PK is a Royal Society Research Fellow.


   Notes
 
Time for primary review 29 days


   References
Top
 Abstract
 1. Introduction
 2. Materials and methods
 3. Results
 4. Discussion
 References
 

  1. Peters N.S. New insights into myocardial arrhythmogenesis: distribution of gap-junctional coupling in normal, ischaemic and hypertrophied human hearts. Clin. Sci. (Lond.) (1996) 90:447–452.[Medline]
  2. Peters N.S, Wit A.L. Myocardial architecture and ventricular arrhythmogenesis. Circulation (1998) 97:1746–1754.[Free Full Text]
  3. Garcia-Dorado D, Rodriguez-Sinovas A, Ruiz-Meana M. Gap junction-mediated spread of cell injury and death during myocardial ischemia-reperfusion. Cardiovasc Res. (2004) 61:386–401.[Abstract/Free Full Text]
  4. Saltman A.E, Aksehirli T.O, Valiunas V, et al. Gap junction uncoupling protects the heart against ischemia. J. Thorac. Cardiovasc. Surg. (2002) 124:371–376.[Abstract/Free Full Text]
  5. Devlin G, Matthews K, McCracken G, et al. An ovine model of chronic stable heart failure. J. Card. Fail. (2000) 6:140–143.[CrossRef][Web of Science][Medline]
  6. Wright C.S, Becker D.L, Lin J.S, Warner A.E, Hardy K. Stage-specific and differential expression of gap junctions in the mouse ovary: connexin-specific roles in follicular regulation. Reproduction (2001) 121:77–88.[Abstract]
  7. Gourdie R.G, Harfst E, Severs N.J, Green C.R. Cardiac gap junctions in rat ventricle: localization using site-directed antibodies and laser scanning confocal microscopy. Cardioscience (1990) 1:75–82.[Web of Science][Medline]
  8. 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]
  9. Yeh H.I, Rothery S, Dupont E, Coppen S.R, Severs N.J. Individual gap junction plaques contain multiple connexins in arterial endothelium. Circ. Res. (1998) 83:1248–1263.[Abstract/Free Full Text]
  10. 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]
  11. Grove B.K, Kurer V, Lehner C, et al. A new 185,000-dalton skeletal muscle protein detected by monoclonal antibodies. J. Cell Biol. (1984) 98:518–524.[Abstract/Free Full Text]
  12. Agarkova I, Auerbach D, Ehler E, Perriard J.C. A novel marker for vertebrate embryonic heart, the EH-myomesin isoform. J. Biol. Chem. (2000) 275:10256–10264.[Abstract/Free Full Text]
  13. Osborn M, Debus E, Weber K. Monoclonal antibodies specific for vimentin. Eur. J. Cell Biol. (1984) 34:137–143.[Web of Science][Medline]
  14. Bohn W, Wiegers W, Beuttenmuller M, Traub P. Species-specific recognition patterns of monoclonal antibodies directed against vimentin. Exp. Cell Res. (1992) 201:1–7.[CrossRef][Web of Science][Medline]
  15. Sun Y, Weber K.T. Infarct scar: a dynamic tissue. Cardiovasc. Res. (2000) 46:250–256.[Abstract/Free Full Text]
  16. Smith J.H, Green C.R, Peters N.S, Rothery S, Severs N.J. Altered patterns of gap junction distribution in ischemic heart disease. An immunohistochemical study of human myocardium using laser scanning confocal microscopy. Am. J. Pathol. (1991) 139:801–821.[Abstract]
  17. Green C.R, Severs N.J. Distribution and role of gap junctions in normal myocardium and human ischaemic heart disease. Histochemistry (1993) 99:105–120.[CrossRef][Web of Science][Medline]
  18. Dispersyn G.D, Mesotten L, Meuris B, et al. Dissociation of cardiomyocyte apoptosis and dedifferentiation in infarct border zones. Eur. Heart J. (2002) 23:849–857.[Abstract/Free Full Text]
  19. Kostin S, Rieger M, Dammer S, et al. Gap junction remodeling and altered connexin43 expression in the failing human heart. Mol. Cell Biochem. (2003) 242:135–144.[CrossRef][Web of Science][Medline]
  20. Peters N.S, Green C.R, Poole-Wilson P.A, Severs N.J. Reduced content of connexin43 gap junctions in ventricular myocardium from hypertrophied and ischemic human hearts. Circulation (1993) 88:864–875.[Abstract/Free Full Text]
  21. Peters N.S, Coromilas J, Severs N.J, Wit A.L. Disturbed connexin43 gap junction distribution correlates with the location of reentrant circuits in the epicardial border zone of healing canine infarcts that cause ventricular tachycardia. Circulation (1997) 95:988–996.[Abstract/Free Full Text]
  22. Matsushita T, Takamatsu T. Ischaemia-induced temporal expression of connexin43 in rat heart. Virchows Arch. (1997) 431:453–458.[CrossRef][Web of Science][Medline]
  23. Sun Y, Kiani M.F, Postlethwaite A.E, Weber K.T. Infarct scar as living tissue. Basic Res. Cardiol. (2002) 97:343–347.[CrossRef][Web of Science][Medline]
  24. Gabbiani G, Hirschel B.J, Ryan G.B, Statkov P.R, Majno G. Granulation tissue as a contractile organ. A study of structure and function. J. Exp. Med. (1972) 135:719–734.[Abstract]
  25. Camelliti P, Green C.R, LeGrice I, Kohl P. Fibrolast network in rabbit sino-atrial node: structural and functional identification of homo- and heterogeneous cell coupling. Circ. Res. (2004 April 2) 94. (in press).
  26. Long C.S, Brown R.D. The cardiac fibroblast, another therapeutic target for mending the broken heart? J. Mol. Cell Cardiol. (2002) 34:1273–1278.[CrossRef][Web of Science][Medline]
  27. Moyer K.E, Davis A, Saggers G.C, Mackay D.R, Ehrlich H.P. Wound healing: the role of gap junctional communication in rat granulation tissue maturation. Exp. Mol. Pathol. (2002) 72:10–16.[CrossRef][Web of Science][Medline]
  28. Elenes S, Martinez A.D, Delmar M, Beyer E.C, Moreno A.P. Heterotypic docking of Cx43 and Cx45 connexons blocks fast voltage gating of Cx43. Biophys. J. (2001) 81:1406–1418.[Web of Science][Medline]
  29. Bukauskas F.F, Angele A.B, Verselis V.K, Bennett M.V. Coupling asymmetry of heterotypic connexin 45/connexin 43-EGFP gap junctions: properties of fast and slow gating mechanisms. Proc. Natl. Acad. Sci. U. S. A. (2002) 99:7113–7118.[Abstract/Free Full Text]
  30. Martinez A.D, Hayrapetyan V, Moreno A.P, Beyer E.C. Connexin43 and connexin45 form heteromeric gap junction channels in which individual components determine permeability and regulation. Circ. Res. (2002) 90:1100–1107.[Abstract/Free Full Text]
  31. Yang F, Liu Y.H, Yang X.P, et al. Myocardial infarction and cardiac remodelling in mice. Exp. Physiol. (2002) 87:547–555.[Abstract]
  32. Kohl P. Heterogeneous cell coupling in the heart: an electrophysiological role for fibroblasts. Circ. Res. (2003) 93:381–383.[Free Full Text]
  33. Kohl P, Noble D. Mechanosensitive connective tissue: potential influence on heart rhythm. Cardiovasc. Res. (1996) 32:62–68.[CrossRef][Web of Science][Medline]
  34. Gaudesius G, Miragoli M, Thomas S.P, Rohr S. Coupling of cardiac electrical activity over extended distances by fibroblasts of cardiac origin. Circ. Res. (2003) g3:421–428.
  35. Kohl P, Varghese A, Dekanski J, Noble D, Winslow R.L. Computational study of the impact of cardiac mechanosensitive fibroblasts on heart rhythm. Exp. Clin. Cardiol. (1996) 1:80–86.
  36. Martin P. Wound healing-aiming for perfect skin regeneration. Science (1997) 276:75–81.[Abstract/Free Full Text]
  37. Frantseva M.V, Kokarovtseva L, Naus C.G, et al. Specific gap junctions enhance the neuronal vulnerability to brain traumatic injury. J. Neurosci. (2002) 22:644–653.[Abstract/Free Full Text]
  38. Lin J.H, Weigel H, Cotrina M.L, et al. Gap-junction-mediated propagation and amplification of cell injury. Nat. Neurosci. (1998) 1:494–500.[CrossRef][Web of Science][Medline]
  39. Kanno S, Kovacs A, Yamada K.A, Saffitz J.E. Connexin43 as a determinant of myocardial infarct size following coronary occlusion in mice. J. Am. Coll. Cardiol. (2003) 41:681–686.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
N. Morita, A. A. Sovari, Y. Xie, M. C. Fishbein, W. J. Mandel, A. Garfinkel, S.-F. Lin, P.-S. Chen, L.-H. Xie, F. Chen, et al.
Increased susceptibility of aged hearts to ventricular fibrillation during oxidative stress
Am J Physiol Heart Circ Physiol, November 1, 2009; 297(5): H1594 - H1605.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. M. Maleckar, J. L. Greenstein, W. R. Giles, and N. A. Trayanova
K+ current changes account for the rate dependence of the action potential in the human atrial myocyte
Am J Physiol Heart Circ Physiol, October 1, 2009; 297(4): H1398 - H1410.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
Y. Xie, A. Garfinkel, J. N. Weiss, and Z. Qu
Cardiac alternans induced by fibroblast-myocyte coupling: mechanistic insights from computational models
Am J Physiol Heart Circ Physiol, August 1, 2009; 297(2): H775 - H784.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
V. Jacquemet and C. S. Henriquez
Loading effect of fibroblast-myocyte coupling on resting potential, impulse propagation, and repolarization: insights from a microstructure model
Am J Physiol Heart Circ Physiol, May 1, 2008; 294(5): H2040 - H2052.
[Abstract] [Full Text] [PDF]

Home page
 J. Physiol.Home page
L. Chilton, W. R. Giles, and G. L. Smith
Evidence of intercellular coupling between co-cultured adult rabbit ventricular myocytes and myofibroblasts
J. Physiol., August 15, 2007; 583(1): 225 - 236.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
K. Ryu, L. Li, C. M. Khrestian, N. Matsumoto, J. Sahadevan, M. L. Ruehr, D. R. Van Wagoner, I. R. Efimov, and A. L. Waldo
Effects of sterile pericarditis on connexins 40 and 43 in the atria: correlation with abnormal conduction and atrial arrhythmias
Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H1231 - H1241.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
T. A. Baudino, W. Carver, W. Giles, and T. K. Borg
Cardiac fibroblasts: friend or foe?
Am J Physiol Heart Circ Physiol, September 1, 2006; 291(3): H1015 - H1026.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
M. Miragoli, G. Gaudesius, and S. Rohr
Electrotonic Modulation of Cardiac Impulse Conduction by Myofibroblasts
Circ. Res., March 31, 2006; 98(6): 801 - 810.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
P. Kohl, P. Camelliti, F. L Burton, and G. L Smith
Fibroblast-Myocyte Interrelation in the Mammalian Heart: Experiments and Models
FASEB J, March 1, 2006; 20(5): A847 - A847.


Home page
 Cardiovasc ResHome page
R. B. Driesen, G. D. Dispersyn, F. K. Verheyen, S. M. van den Eijnde, L. Hofstra, F. Thone, P. Dijkstra, W. Debie, M. Borgers, and F. C.S. Ramaekers
Partial cell fusion: A newly recognized type of communication between dedifferentiating cardiomyocytes and fibroblasts
Cardiovasc Res, October 1, 2005; 68(1): 37 - 46.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
L. Chilton, S. Ohya, D. Freed, E. George, V. Drobic, Y. Shibukawa, K. A. MacCannell, Y. Imaizumi, R. B. Clark, I. M. C. Dixon, et al.
K+ currents regulate the resting membrane potential, proliferation, and contractile responses in ventricular fibroblasts and myofibroblasts
Am J Physiol Heart Circ Physiol, June 1, 2005; 288(6): H2931 - H2939.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
P. Camelliti, T. K. Borg, and P. Kohl
Structural and functional characterisation of cardiac fibroblasts
Cardiovasc Res, January 1, 2005; 65(1): 40 - 51.
[Abstract] [Full Text] [PDF]

Home page
 Cardiovasc ResHome page
S. Dhein and H. J Jongsma
Forming the network--gap junctions in the cardiovascular system
Cardiovasc Res, May 1, 2004; 62(2): 225 - 227.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Camelliti, P.
Right arrow Articles by Green, C. R
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Camelliti, P.
Right arrow Articles by Green, C. R
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?