Large- and Medium-sized Arteries Remaining in Transmural Scar Distal to Permanent Coronary Ligation Undergo Neointimal Hyperplasia and Inward Remodeling

Abstract

This study aimed to investigate the structural integrity and dynamic changes in chronically occluded residual arteries found in post–myocardial infarction (MI) scar. A transmural MI was induced in middle-aged, male Sprague-Dawley rats by left coronary artery ligation. The rats were euthanized 3 days and 1, 2, 4, 8, and 12 weeks after MI, and their hearts were processed into paraffin for histology, immunohistochemistry, and quantitative morphometry. It has been found that large- and medium-sized arteries were able to survive inside the transmural scars for 12 post-MI weeks. Furthermore, most residual arteries preserved their structural integrity for up to 2 weeks post-MI, but gradually all disused vessels had undergone neointimal hyperplasia and inward remodeling at later time periods. In addition, the replacement of vascular smooth muscle cells in the wall of residual arteries by extracellular matrix components led to a disruption of the vessel integrity and progressive obliteration of their lumen between 4 and 12 post-MI weeks. Taken together, this study demonstrate that residual arteries in post-infarcted region were capable of maintaining their structural integrity, including the patent lumen, during two post-MI weeks, suggesting that during this period they can be used as potential conduits for conceivable reflow of arterial blood within the scarred region of the heart

Introduction

Myocardial infarction (MI) remains one of the major causes of heart failure with reduced ejection fraction (HFrEF) worldwide.¹ The loss of cardiac muscle tissue due to ischemia and its subsequent replacement with fibrotic scar tissue lead to a marked reduction in left ventricular (LV) contractile performance.² A consequent systolic dysfunction activates a series of compensatory mechanisms that initiates adverse LV remodeling leading to HFrEF.³ Although, in the past several decades, a vast body of basic and clinical research has been devoted to refining the therapeutic approaches for patient with HFrEF,⁴ they are still unable to prevent progressive deterioration of the cardiac function primarily due to inability to restore the damaged LV wall with the functional myocardium.

However, in recent years, there have been numerous experimental and clinical attempts to cure MI-induced HFrEF through regeneration of contractile muscle tissue in ischemia-damaged areas of the heart by utilizing a broad range of cell-based therapeutic approaches.^5,6 Despite some encouraging results,⁷ in most cases, the regenerative strategies have failed to de novo produce the mature, contractile myocytes, in part, because of the inability to persistently retain the transplanted cells.⁸ According to the latest retrospective analyses of the cardiac regeneration field, the nature of the cellular material and the condition of the host environment have been identified as the most fundamental factors influencing the efficacy of cell-based therapies.^6,8–10 Surprisingly, whereas the challenges surrounding the cell therapy products have been well addressed in the modern literature,^5,8,11 the various components of the host environment, particularly the availability of a patent vascular network, necessary to supply the developing myocytes with oxygen and nutrients, have not received a thorough attention.^10,11 Nonetheless, the establishment in the healing cardiac wall, a well-organized coronary vascular system composed of both microvasculature and large conductive vessels, has been recognized as one of the strategic priorities for successful myocardial repair.^9–11

Currently, the presence of viable microvessels in ischemia-damaged areas of the post-MI heart has been well documented. They include both the preexisting capillaries surrounding the surviving cardiac myocytes^12–15 and the neovasculature of the developing granulation tissue.^16–18 Therefore, lately, there had been several encouraging experimental attempts to utilize these endogenous microvascular systems in the course of cardiac regenerative interventions to either sustain the residual cardiac myocytes¹⁵ or to nourish the in vitro engineered tissue grafts.¹⁹ However, the prospective origin of the large- and medium-sized distributing arteries, which are necessary to provide a patent route for the blood flow from the aorta to the microvascular network within the healing myocardium, remained obscure. Despite the initial promising report demonstrating that the large coronary vessels could be formed in situ from the injected progenitor cells,²⁰ the further progress in this direction remained vague. At the same time, there were a number of studies which revealed the existence of residual coronary arteries within scarred areas of the LV free wall following a permanent MI.^{14,17,18,21,22} Surprisingly, considering a plausible assumption that such vessels, if become conjoined with the intact segment of the original coronary vascular system, may regain their functionality as endogenous conduits for blood distribution in the infarcted areas, the studies in which such possibility has been experimentally explored are extremely rare.^14,21

Therefore, this study has been primarily undertaken to thoroughly investigate the structural integrity and dynamic changes in permanently occluded coronary arteries remaining inside the developing and mature transmural scars following a large MI. The findings obtained in this study have provided essential information regarding the time frame during which the disused arterial vessels remaining in post-MI scars appeared structurally suitable to serve as the patent conduits for conceivable reflow of the arterial blood to the infarcted/fibrotic region.

Material and Methods

Animals and Experimental Protocol

All animal procedures were approved by the Institutional Animal Care and Use Committee and performed in accordance with the Guide for the Care and Use of Laboratory Animals, 8th edition (National Research Council 2011, National Academic Press).

A transmural MI was induced in 12-month-old male (n=42) Sprague-Dawley rats (Charles River Laboratories, Inc.; Wilmington, MA) under ketamine (100 mg/kg intraperitoneally [i.p.]) and xylazine (10 mg/kg i.p.) anesthesia by permanent ligation of the left anterior descending coronary artery, as previously detailed.^23–25 Sham-operated rats (n=5) served as an age-matched control. Following surgery, the rats were housed under climate-controlled conditions at a 12-hr light/dark cycle and provided with standard rat chow and water ad libitum. The mortality rate among post-MI rats was ~28% with all death occurring within first 48 hr. The experimental rats were euthanized 3 days and 1, 2, 4, 8, and 12 weeks after MI (n=5 per experimental group), and their hearts were collected for further evaluation.

In two groups of rats (1 week and 2 weeks post-MI), starting on days 4 and 11 after surgery, respectively, the animals had received 5-bromo-2′-deoxyuridine (BrdU; Sigma, St Louise, MO) in phosphate-buffered saline (PBS) at a dose 12.5 mg/kg/day for 3 days i.p. via ALZET osmotic pumps (Durect; Cupertino, CA).

At the end of each time period, rats were anesthetized with 4% isoflurane in pure oxygen, and their hearts were arrested in diastole by an intracardiac injection of 2 ml of 100-mmol/l potassium chloride. The excised hearts were attached to a Langendorff apparatus and perfusion-fixed with 4% paraformaldehyde (PFA) in PBS for 20 min at constant pressure (100 mmHg). Then the hearts were immersed in a fresh portion of 4% PFA solution and stored at 4C for 48 hr, before being processed for tissue sampling. Each heart was washed in PBS and cut transversely into five parallel slices with a four-blade guillotine. The two midventricular slices were processed into paraffin blocks for routine histology, immunohistochemistry, and quantitative morphometry.

Histology and Immunohistochemistry

Transverse, 8.0-µm-thick serial sections were cut from the paraffin-embedded slices of the hearts onto microscope slides. From each ventricular slice, several serial sections were routinely stained with hematoxylin and eosin (H&E), Masson’s trichrome, picrosirius red, and Verhoeff’s elastic tissue stains.

Other serial sections were subjected to immunolabeling with various combinations of the primary (Table 1) and secondary (Table 2) antibodies. Briefly, sections were deparaffinized and antigen retrieval was performed by enzymatic digestion with 20-µg/ml proteinase K (cat. no. P4850; Sigma) in Dulbecco’s PBS containing calcium and magnesium ions (cat. no. 14040; Life Technologies, Carlsbad, CA) for 20 min at 37C. For immunofluorescence staining, incubation with primary antibodies was conducted for 2 hr followed by labeling with fluorophore-conjugated secondary antibodies for 45 min at 37C in a moist chamber. The sections were then coverslipped with ProLong Gold antifade mounting medium (cat. no. P36931; Molecular Probes, Inc., Eugene, OR) containing 4′, 6-diamidino-2-phenylindole (DAPI) to counterstain the nuclei. For immunohistochemistry, incubation with a mouse anti-α-smooth muscle actin antibody (Table 1) was conducted for 2 hr followed by labeling with a peroxidase-conjugated anti-mouse secondary antibody (cat. no. MP-7402; ImmPress Peroxidase Reagent kit; Vector Labs, Burlingame, CA) for 30 min at 37C in a moist chamber. The peroxidase-conjugated antibody was visualized using 3, 3′-diaminobenzidine (DAB) substrate (cat. no. SK-4105; ImmPACT DAB Peroxidase Substrate kit; Vector Labs) at room temperature. The nuclei were counterstained with hematoxylin 7211 (Richard-Allan Scientific; Kalamazoo, MI). Omission of primary antibodies served as negative controls.

Table 1.

Primary Antibodies Used for Immunostaining.

Antigen	Host	Clone	Isotype	Dilution	Cat. No.	Source
Alpha (α)-smooth muscle actin	Mouse	1A4	IgG2a	1:400	A2547	Sigma; St Louis, MO
Alpha (α)-smooth muscle actin, Cy3 conjugate	Mouse	1A4	IgG2a	1:600	C6198	Sigma; St Louis, MO
Desmin	Mouse	DE-U-10	IgG1	1:80	D1033	Sigma; St Louis, MO
Laminin	Rabbit	—	—	1:30	L9393	Sigma; St Louis, MO
Bromodeoxyuridine (BrdU)	Mouse	131-14871	IgG1	1:1000	MAB4072	EMD Millipore; Temecula, CA

Table 2.

Fluorophore-Conjugated Secondary Antibodies Used for Fluorescence Staining.

Description	Dilution	Catalog No.	Source
Alexa Fluor 488-conjugated goat anti-mouse IgG (H+L)	1:200	A-11001	Molecular Probes; Eugene, OR
Alexa Fluor 488-conjugated goat anti-rabbit IgG (H+L)	1:200	A-11008	Molecular Probes; Eugene, OR
Rhodamine Red-X-conjugated goat anti-mouse IgG (H+L)	1:200	115-295-146	Jackson ImmunoResearch Lab.; West Grove, PA
Cy3-conjugated goat anti-mouse IgG (FCγ subclass 2a)	1:400	115-165-206	Jackson ImmunoResearch Lab.; West Grove, PA

Microscopy and Quantitative Morphometry

From each heart, the entire profiles of midventricular cross-sections stained with Masson’s trichrome and H&E stains were captured onto a computer at high resolution (20× objective) using a MikroScan D2 slide scanner and Q-Scan software (MikroScan Technologies; Carlsbad, CA). The scanned digital images were used to accurately identify the profiles of the same coronary arteries within the serial sections of the same post-MI scar stained with different histological and immunohistological techniques.

The ventricular slices stained with H&E, Masson’s trichrome, picrosirius red, and Verhoeff’s elastic tissue stains and immunohistochemically with an antibody against α-smooth muscle actin were examined under a Leica DM IRE2 microscope (Leica Microsystems; Buffalo Grove, IL), and the high-resolution images (20× and 40× objectives) were captured onto a computer with a Leica DFC 450 digital camera (Leica Microsystems). The immunofluorescence-labeled sections were examined under a Leica DM 4000B fluorescence microscope (Leica Microsystems), and the high-resolution double and triple fluorescence-labeled images were captured onto a computer with a Leica DFC 360FX digital camera (Leica Microsystems) using Leica Application Suite software version 4.7.1. The final figures were digitally assembled from the captured images using Adobe Photoshop CC software (Adobe Systems; San Jose, CA).

Morphometric analysis were conducted on digital images of the large- and medium-sized coronary arteries (the vessels with an external diameter ranging from 220 to 40 µm) collected from the LV free wall of sham-operated rats and the LV scars of post-MI rats using Image-Pro Analyzer 7.0 software (Media Cybernetics, L.P.; Silver Spring, MD). Briefly, the inner (luminal) and the outer (media–adventitia border) outlines of each cross-sectioned arterial vessel stained with Masson’s trichrome were electronically traced, and lumen diameter, external (outer) diameter, and the total wall thickness (intima–media thickness) were measured. The wall-to-lumen ratio was calculated in each artery based on the measurements of the total wall thickness and lumen diameter. In average, about four to six cross-sectioned arterial profiles were analyzed in each LV section (five hearts per group).

Statistical Analysis

Data are expressed as the mean ± standard error of the mean. Statistical analysis was performed using Prism 6 software package (IBM Corp.; Armonk, NY). A one-way analysis of variance followed by the Tukey’s post hoc tests was performed for multigroup comparisons. A probability of p≤0.05, 0.01, 0.001, and 0.0001 was considered to indicate different levels of statistical significance.

Results

Considering that some arterial vessels remaining within the scar tissue in a close proximity to the non-infarcted myocardium could retain blood flow derived from the non-occluded branches of the left coronary artery, this study was focused exclusively on the residual arteries found at the center of the large transmural scars induced by the permanent ligation of the left anterior descending coronary artery.

Large- and Medium-sized Coronary Arteries Can Persistently Survive in Transmural Post-MI Scars

The examination of the hearts during 12 post-MI weeks revealed that the identifiable profiles of the large- and medium-sized residual coronary arteries with an external diameter ranging from 40.4 to 217.5 µm could be clearly seen in the transmural scars at all time points following permanent coronary artery occlusion (Fig. 1C, E, G, I, K, and M: an outline at the tip of an arrow). Most importantly, although the remaining coronary arteries became surrounded, at first, by the granulation tissue (Fig. 1D, F, and H: asterisks) and later by the mature fibrous tissue (Fig. 1J, L, and M: asterisks), they continued to maintain the key structural features, such as the patent vessel lumen, the tunica intima, and the tunica media, similar to the large- and the medium-sized arteries in the intact myocardium of the sham-operated rats (Fig. 1A and B).

Figure 1.

Residual coronary arteries in developing and mature transmural scars. The hematoxylin and eosin–stained micrographs demonstrate the profiles of surviving large- and medium-sized arteries in 3-day-old (C, D), 1-week-old (E, F), 2-week-old (G, H), 4-week-old (I, J), 8-week-old (K, L), and 12-week-old (M, N) post-MI scars. Note that the coronary artery shown in micrographs A and B serves as a representative example of a large arterial vessel in structurally intact LV myocardium of a sham-operated rat. In each low-power view (A, C, E, G, I, K, and M), an arrow points to a rectangular outline designating a location of the corresponding coronary arteries which are displayed in high-power micrographs B, D, F, H, J, L, and N, respectively. The arrowheads in micrographs C, E, G, I, K, and M mark the edges of the transmural scar. The asterisks in micrographs D, F, and H identify the vascularized granulation tissue, whereas in micrographs J, L, and M, they denote the fibrous tissue which surrounds the residual coronary arteries in the scars. Scale bars are 4 mm (C, G, I, and K), 3 mm (A, E, and M), 100 µm (B, D, F, and H), and 50 µm (J, L and N). Abbreviations: D, days; WK, week(s); MI, myocardial infarction.

Vascular Smooth Muscle (VSM) Cells in Chronically Occluded Residual Coronary Arteries Maintain Their Contractile Phenotype

To further confirm that the long-term surviving coronary arteries, which were initially recognized in transmural scars stained with histological dyes only (Fig. 1), continued to preserve the characteristics of the arterial vessels for the duration of 12 post-MI weeks, they were immunostained with the antibodies against muscle-specific proteins essential for the integrity of a contractile apparatus in VSM cells (Fig. 2). The histological and immunofluorescence staining of the same residual coronary arteries, which were identified on the adjacent serial sections, revealed that the VSM cells in the wall of chronically disused arteries had persistently expressed α-smooth muscle actin (Fig. 2B, J, and L) as well as desmin (Fig. 2D, F, and H) at all time points following permanent coronary occlusion.

Figure 2.

Expression of muscle-specific proteins in vascular smooth muscle (VSM) cells of the residual coronary arteries in transmural scars 3 days (A, B), 1 week (C, D), 2 weeks (E, F), 4 weeks (G, H), 8 weeks (I, J), and 12 weeks (K, L) after MI. Note, in each time point, the two micrographs demonstrate the same arterial vessel found on adjacent serial sections which were stained either with histological dyes or with immunofluorescence technique. A combination between these two staining methods has allowed to confirm that the vessels demonstrating the expression of muscle-specific protein, such as α-smooth muscle actin (α-SMA) and desmin, were the same which were initially identified on the sections stained with histological dyes only. The histological stains used to visualize the residual coronary arteries in the scars were Masson’s trichrome, in A and E, and H&E, in C, G, I, and K. The corresponding serial sections were immunofluorescence stained either with an antibody against α-SMA (red color in B, J, and L) or with a combination of the antibodies against desmin (red color) and laminin (green color) in D, F, and H. In the latter, the presence of laminin-positive labeling beneath the endothelial layer and around VSM cells confirms the maintenance of the basement membranes in the wall of the residual arteries. All immunofluorescence-labeled sections were counterstained with DAPI (blue color) to visualize the nuclei. Scale bars are 100 µm (A, B, C, D, E, F, F, H, J, and K) and 50 µm (C, I, and L). Abbreviations: D, days; WK, week(s); H&E, hematoxylin and eosin; α-SMA, α-smooth muscle actin; DAPI, 4′, 6-diamino-2-phenylindole; MI, myocardial infarction.

Permanent MI Causes Necrosis Predominantly in Small Coronary Arteries and Arterioles

In contrast to surviving large- and medium-sized arterial vessels, almost all small arteries and arterioles had undergone necrosis during first 2 weeks following a permanent MI (Fig. 3). The examination of the infarcted regions 3 days and 1 and 2 weeks following a large, transmural MI revealed that with exception of a few necrotic large- or medium-sized vessels, the entire network of small coronary arteries and arterioles had experienced the evident degenerative changes, including fibrinoid necrosis (Fig. 3B, C, F, G, J, and K: asterisks) and hyaline degeneration of the tunica media (Fig. 3D to F and H to I). Occasionally, the necrotic small-diameter arterial branches could be seen in alignment with the surviving portion of the larger residual coronary arteries (Fig. 3A). Most importantly, considering that the structurally intact large arteries could be often observed in close proximity to the profiles of the smaller necrotic vessels even 2 weeks after MI (Fig. 3J and K: arrows), it appeared that degeneration of small size vessels had occurred progressively throughout the earlier stages of post-MI scar formation, including both inflammatory reaction and development of the granulation tissue.

Figure 3.

Degenerative changes observed in coronary arteries remaining in transmural scars during two post-MI weeks. (A) An H&E-stained section of a 3-day-old post-MI scar demonstrating hyaline degeneration in the tunica media of a small-diameter arterial branch (arrowheads) that remains in alignment with a surviving portion of the large residual coronary artery. Note a massive accumulation of extravasated erythrocytes surrounding the degenerating arterial branch. The arrows denote the point of transition between necrotic and surviving segments in the artery. (B, C) and (D, E) Two representative necrotic coronary arteries identified in serial sections of a 1-week-old post-MI scar stained with H&E, in B and D, and with Masson’s trichrome, in C and E, respectively. Note that the branches of the same arterial vessel shown in B and C display the features of fibrinoid necrosis associated with leukocytic infiltration (arrows) and accumulation of fibrin-like material (asterisks) in the vessel wall, whereas the large artery depicted in D and E reveals the characteristic manifestations of hyaline degeneration associated with the homogeneously stained, acellular tunica media and a well-preserved endothelial lining (arrowheads). (F, G) and (H, I) Two representative necrotic coronary arteries recognized in serial sections of a 2-week-old post-MI scar stained with H&E, in F and H, and with Masson’s trichrome, in G and I, respectively. Note that the same arterial vessel shown in F and G displays the signs resembling fibrinoid necrosis (asterisks), whereas the artery illustrated in H and I shows the typical appearance of hyaline degeneration of the tunica media. The arrowheads in H and I point to the nuclei of surviving endothelial cells. (J, K) The same representative region from a 2-week-old post-MI scar stained with H&E (J) and Masson’s trichrome (K) demonstrating the presence of a small arterial vessel with the features of fibrinoid necrosis (asterisk) in close proximity to a structurally intact larger artery (arrows). All scale bars are 100 µm. Abbreviations: D, days; WK, week(s); H&E, hematoxylin and eosin; MI, myocardial infarction.

Coronary Arteries Remaining in the Same Region of the Scar Display a Different Degree of Structural Alterations

Although the residual arteries were exposed to relatively similar milieu inside the developing post-MI scar, the individual vessels from the same region often revealed the different scale of structural changes. In some instances, even one of the two closely adjacent vessels showed more advanced structural alterations compared with another (Fig. 4), suggesting some level of inner heterogeneity among the remaining arterial vessels in their response to the factors causing vessel wall alterations.

Figure 4.

Local variability in the degree of structural alterations among the residual arteries in 1-week-old (A–C) and 2-week old (D–F) post-MI scars. Note that each individual set of the micrographs (A–C or D–F) displays the profiles of the same representative arteries identified in adjacent serial sections stained with H&E (A, D), Masson’s trichrome (B, E), and picrosirius red (C, F). In each micrograph, the asterisk denotes an artery that demonstrates more advanced structural changes, such as the thickening of the arterial wall and accumulation of the extracellular matrix, including collagen fibers, in the tunica media. Scale bars are 100 µm. Abbreviations: WK, week(s); H&E, hematoxylin and eosin; MI, myocardial infarction.

Permanently Occluded Residual Coronary Arteries Undergo Progressive Neointimal Thickening, Intravascular Fibrosis, and Inward Remodeling

Despite the fact that most residual arteries distal to permanent coronary ligation were able to maintain their structural integrity for up to 2 weeks post-MI (Figs. 1D, F, 2A to F, 3J and K: arrows, and Fig. 4), some of the disused arterial vessels began to demonstrate visible structural alterations 1 week following the total cessation of anterograde blood flow (Fig. 5). The principal structural changes which were observed in the affected vessels included the neointimal hyperplasia that narrowed the vessel lumen (Fig. 5A to D and I to L) and accumulation of the extracellular matrix components, including collagen fibers, around activated or synthetic VSM cells in the neointima as well as between contractile VSM cells in the tunica media (Fig. 5C and K). In addition, the BrdU labeling experiments have confirmed that the neointimal hyperplasia seen in disused coronary arteries was in part caused by proliferation of both synthetic VSM cells and endothelial cells (Fig. 5E to H and M to P). The analogous, but more advanced, structural modifications were observed in the majority of, if not all, remaining vessels at later time points during 12 post-MI weeks (Fig. 6). However, the further expansion of neointima in these residual arteries had been accompanied by an increased accumulation of the elastic extracellular material around individual synthetic VSM cells and by the periodic breaks of the internal elastic lamina (Fig. 6G and K: arrows). It is important to emphasize that the quantitative morphometric assessment of the residual arteries found in the scars revealed that in comparison with the intact coronary arteries of the sham-operated rats, these vessels had undergone progressive inward remodeling associated with a reduction in external diameter and a significant increase in the vessel wall thickness-to-lumen ratio (Fig. 7).

Figure 5.

The pattern of structural modifications detected in coronary arteries surviving in 1-week-old (A–H) and 2-week-old (I–P) post-MI scars. Note that each individual set of the micrographs (A–D or I–L) demonstrates the same representative artery recognized on adjacent serial sections stained with H&E (A, I), Masson’s trichrome (B, J), and picrosirius red (C, K) and with an antibody against α-SMA that was visualized using 3, 3′-diaminobenzidine chromogen (brown color in D, L). In all light-microscopy images, the black, dotted line delineates the border between the tunica media and neointimal hyperplasia. A combination of different stains applied to serial profiles of the same arteries has confirmed that most of the cells forming the neointima in these vessels were activated or synthetic vascular smooth muscle (VSM) cells (D and L). A switch from a contractile to a synthetic phenotype in these cells is supported by the absence of a spindle-shaped morphology (A, B and I, J) and progressive accumulation of collagen fibers around the individual cells (C and K). In addition, two sets of fluorescence micrographs (E–H and M–P) obtained from 1- and 2-week-old post-MI scar, respectively, display the representative residual coronary arteries immunostained with the antibodies against BrdU (green color in G, H and O, P) and α-SMA (red color in E–H and M–P) in a combination with the nuclear counterstain by DAPI (blue color in E, F and M, N). Note that micrographs F, H, N, and P represent the high-power views of the areas outlined by a box in micrographs E, G, M, and O, respectively. The nature of the growing (BrdU-labeled) cells in the neointima was established on the basis of two principal criteria: (1) the proximity of a BrdU-positive nucleus to the vessel lumen and (2) the presence or absence of α-SMA immunostaining in the cytoplasm. Such methodological approach has allowed to confirm that the arrowheads in F and H denote BrdU-positive/α-SMA-positive VSM cells within the neointima, whereas the arrows in N and P point to BrdU-positive/α-SMA-negative cells in the endothelial lining. To facilitate the tracing of all BrdU-labeled cells in corresponding images of the same arterial wall, each arrow or arrowhead in micrographs E, G or M, O points to the matching nucleus. According to these observations, the neointimal growth in the residual coronary arteries involved proliferation of both synthetic VSM cells (arrowheads in E, G and M, O) and the cells of the endothelial lining (arrowheads in E, G and M, O). Scale bars are 50 µm (A–D, E, G, I–L, M, and O) and 25 µm (F, H, N, and P). Abbreviations: WK, week(s); H&E, hematoxylin and eosin; α-SMA, α-smooth muscle actin; DAPI, 4′, 6-diamino-2-phenylindole; BrdU, 5-bromo-2′-deoxyuridine; MI, myocardial infarction.

Figure 6.

The pattern of structural alterations observed in coronary arteries surviving in 4-week-old (A–D), 8-week-old (E–H), and 12-week-old (I–L) post-MI scars. Note that each individual set of the micrographs (A–D, E–H, or I–L) displays the same representative artery identified on adjacent serial sections stained with Masson’s trichrome (A, E), H&E (I), Verhoeff’s elastic tissue stain (C, G, and K), and picrosirius red (D, H, and L) and with an antibody against α-SMA that was visualized with 3, 3′-diaminobenzidine chromogen (brown color in B, F, and J). The white or black dotted outline seen on the arterial profiles in A, B, D, E, F, H, I, J, and L delineates the border between the tunica media and the neointima. In micrographs C, G, and K, the border between the tunica media and neointimal hyperplasia is demarcated by the internal elastic lamina visualized by Verhoeff’s elastic tissue stain as a black circumferential outline. Note that the discontinuous staining pattern of the internal elastic lamina seen in micrographs G and K (arrows) suggests some disruption of structural integrity in this essential component of the arterial wall. Furthermore, a combination of different stains applied to serial profiles of the same residual arteries has revealed that synthetic VSM cells of the neointima continued to deposit the amorphous and fibrous components of the extracellular matrix (A and E), including elastin (C, G, and K) and collagen fibers (D, H, and L), thereby thickening the arterial wall. However, it is important to highlight that despite substantial neointimal expansion, most surviving arteries were able to maintain a narrowed, but a patent, lumen. Scale bars are 50 µm. Abbreviations: WK, weeks; H&E, hematoxylin and eosin; α-SMA, α-smooth muscle actin; MI, myocardial infarction; VSM, vascular smooth muscle.

Figure 7.

Time course of changes in external (A) and lumen (B) diameters and total wall thickness (C) and the total wall thickness-to-lumen ratio (D) in the residual coronary arteries remaining in transmural scars during 12 post-MI weeks. Note, the pattern of progressive structural modifications observed in a majority of the residual arteries resembles that of inward arterial remodeling characterized by a marked increase in the wall-to-lumen ratio (D) caused by prominent thickening of the vessel wall in a combination with a reduction of external (outer) and lumen (inner) vessel diameters. Data are mean ± SEM. Abbreviations: MI, myocardial infarction; SEM, standard error of the mean. *p≤0.05, **p≤0.01, ***p≤0.001 vs sham-operated (SH) rats; ^§p≤0.05, ^§§p≤0.01, ^§§§p≤0.001 vs 3-day (3D) post-MI rats; ^†p≤0.05, ^††p≤0.01, ^†††p≤0.001 vs 1-week (1W) post-MI rats; ^‡‡‡p≤0.001 vs 2-week (2W) post-MI rats; ^#p≤0.05 vs 4-week (4W) post-MI rats.

Persistent Replacement of VSM Cells in Remaining Coronary Arteries by Extracellular Matrix Leads to a Disruption of the Vessel Wall Integrity and a Loss of Luminal Patency

Whereas all residual coronary arteries remaining in the mature scars demonstrated progressive structural modifications (Fig. 6), some of the vessels had experienced the more profound changes (Fig. 8). The detrimental character of such alteration was reflected in an apparent disruption of the internal elastic lamina (Fig. 8C, G: arrows) and a patchy disappearance of VSM cells in both the tunica media (Fig. 8B and F: arrowheads) and the neointima (Fig. 8A to D: double-headed arrows; E to H and I to L: asterisks) with a noticeable accumulation of extracellular matrix macromolecules, including elastin and collagen fibers, in the space between the remaining cells. Most importantly, a progressive loss of VSM cells which was accompanied by replacement intravascular fibrosis led to marked narrowing and, in some case, to total obliteration of the arterial lumen (Fig. 8E to H and K to L).

Figure 8.

Detrimental structural changes detected in some residual coronary arteries remaining in 4-week-old (A–D), 8-week-old (E–H), and 12-week-old (I, J and K, L) post-MI scars. Note that each individual set of the micrographs (A–D or E–H) demonstrates the same representative artery identified on adjacent serial sections stained with Masson’s trichrome (A, E), Verhoeff’s elastic tissue stain (C, G), and picrosirius red (D, H) and with an antibody against α-SMA that was visualized with 3, 3′-diaminobenzidine chromogen (brown color in B, F). The white or black dotted outline seen on the arterial profiles in E, F, and H defines the border between the tunica media and the neointima. In micrographs C and G, the border between the tunica media and neointimal hyperplasia is delineated by the internal elastic lamina visualized by Verhoeff’s elastic tissue stain as a black circumferential outline. A combination of different stains applied to serial profiles of the same residual arteries has revealed a patchy loss of VSM cells in both the tunica media (corresponding arrowheads in A, B and E, F) and the neointimal formation (corresponding double-headed arrows in A–D and an asterisk in E–H). Note that the areas which were formerly occupied by either contractile or synthetic VSM cells become filled with the extracellular material (A and E), including elastin (C and G) and collagen fibers (D and H). The analogous sporadic pattern of VSM cell disappearance has been also detected in the wall of some residual arteries from 12-week-old post-MI scars (I, J and K, L). In the latter micrographs, two representative arterial vessels were immunostained with the antibodies against desmin (red color in I–L) and laminin (green color in J and L) in a combination with the nuclear counterstain by DAPI (blue color in I–L). The laminin immunostaining was utilized to visualize the basement membranes of the VSM cells throughout the entire arterial wall. The patchy absence of desmin-positive immunostaining within the laminin-outlined arterial wall (asterisks in I–L) evidently confirms that these areas lack the VSM cells. Furthermore, it is important to emphasize that, in some residual vessels, the neointimal formation continued to accumulate the extracellular material around the remaining VSM cells (E), causing fibrosis of the arterial wall and marked narrowing of the vessel lumen. Moreover, the fragmented staining pattern of the internal elastic lamina in such vessels (arrows in C and G) suggests a complete loss of structural integrity in the arterial wall. Scale bars are 50 µm. Abbreviations: WK, weeks; α-SMA, α-smooth muscle actin; MI, myocardial infarction; VSM, vascular smooth muscle; DAPI, 4′, 6-diamino-2-phenylindole.

Discussion

The key findings in this study were as follows: (1) the large- and medium-sized coronary arteries distal to permanent coronary ligation were able to survive for 12 post-MI weeks in large transmural scar, whereas all original arterioles and small coronary arteries had undergone necrosis during the first 2 weeks following total arterial occlusion; (2) the VSM cells seen in the wall of chronically disused residual coronary arteries were able to persistently maintain the production of muscle-specific proteins; (3) the majority of coronary arteries surviving in transmural scars preserved their structural integrity, including the patent lumen, for up to two post-MI weeks; (4) starting at the second post-MI week, all residual coronary arteries had undergone progressive neointimal hyperplasia, intravascular fibrosis, and inward remodeling; and (5) in advanced cases, the residual coronary arteries had experienced the fragmentation of internal elastic lamina and fibroelastic replacement of VSM cells that eventually led to obliteration of the vessel lumen between 4th and 12th post-MI weeks.

Although in the past three decades there have been a number of studies that have reported the presence of residual coronary arteries in the LV free wall of the rat heart affected by a large, transmural MI caused by permanent ligation of the left coronary artery,^{12,14,17,21,22,26} there were only a few reports among these earlier investigations which provided some details on the condition of the remaining vessels.^14,21,22 Furthermore, considering that these previous reports were focused predominantly at a single time point after cessation of arterial blood flow, this study is the first to offer a comprehensive insight into the pattern of dynamic structural alterations that has been detected in coronary arteries remaining in transmural scars for the duration of 12 post-MI weeks.

Is Prolonged Survival of Residual Coronary Arteries in Post-MI Scars Primarily Supported by Direct Communications Between Their Luminal Space and LV Cavity?

According to this study, the profiles of original large- and medium-sized coronary arteries could be recognized inside the transmural post-infarcted region of the rat hearts between 3 days and 12 weeks after permanent ligation of the left anterior descending coronary artery. This finding further substantiates the observations reported previously by others,^21,22,26 who have demonstrated the existence of large diameter arterial vessels in mature post-MI scars caused by permanent coronary occlusion. Surprisingly, even though such results have been repeatedly documented during the last few decades, the mechanism that would elucidate long-term preservation of large coronary arteries inside the hostile environment of the ischemic/fibrotic region remained unclear. The only explanation that was provided in the past by other investigators has speculated on the idea that the residual coronary arteries of the original arterial bed might become reperfused by the vessels originated from the non-infarcted myocardium.²¹ However, although this concept is plausible for the experimental models that involve the animal species which are known to have a network of well-developed collateral vessels between the coronary arteries, such as dogs,¹³ it seems less feasible in a permanent MI model that employs rats, which lack the collateral vasculature in their coronary arterial system.^27,28 Moreover, because most of the large residual arteries seen in this study as well as in earlier investigations^12,14 were detected far from the border between the scar and non-infarcted myocardium, a potential model in which the ingrowing vessels derived from the non-occluded arteries would preserve the residual arteries right from the onset of coronary blood flow cessation is doubtful. However, a plausible explanation of prolonged survival of the residual coronary arteries in the rat transmural scars can be drawn from the fact that the large coronary arteries of the LV free wall have retained direct communications with the LV cavity via Thebesian vessels, specifically, via arterio-luminal vessels. Despite the fact that distribution and organization of these vessels have been predominantly explored in human hearts,²⁹ the existence of such communications was also confirmed in the hearts of other mammalian species, including rats.³⁰ In our recent study, the presence of patent communications between the Thebesian vessels and the remnants of the original venous system, which persisted inside the post-MI scars, has been considered the probable mechanism that allowed prolonged survival of the mature cardiac myocytes in the vicinity of the intramural venous sinusoids and large subepicardial veins.¹² Same idea can be generalized to the tentative mechanism that may explain the long-term maintenance of the residual coronary arteries in large, transmural post-MI scars. Taking into account the fact that predominantly large- and medium-sized diameter arteries were able to persistently survive in the ischemic/fibrotic region, while almost all small-diameter arterial branches and arterioles located in the same region had become necrotized during the two post-MI weeks, it is feasible to consider the existence of direct routes of communication between LV cavity and the luminal space of large surviving arterial vessels that would permit the constant delivery of oxygen- and nutrient-rich blood to the endothelial lining and the layers of VSM cells composing the thick arterial wall. To a certain extent, such hypothesis is in tune with an idea, which has been originally proposed in mouse and rat experimental models of the permanently ligated common carotid artery, considering that oscillations of arterial blood flow between the aorta and a patent proximal segment of a totally occluded artery might be the key factor that supported survival of the cells, and hence, preserved structural integrity of the vessel wall for an extended period of time.^31,32 However, a rat model of the doubly ligated segment of the common carotid artery had repeatedly demonstrated the presence of necrotic endothelium in the tunica intima and degenerated VSM cells in the tunica media inside the non-perfused, ischemic vascular segment during the first 24 hr.^31,33 Therefore, the presence of residual coronary arteries in large transmural scars during 12 post-MI weeks has strengthened the belief that these vessels should retain the patent communications with LV cavity which could permit intraluminal oscillations of blood flow coupled with cardiac cycles to continually maintain viability of endothelial and VSM cells in their walls.

Does the Pattern of Structural Alterations Detected in Residual Coronary Arteries of Post-MI Scars Resemble Those Encountered in Arterial Vessels With Reduced But Not Absent Luminal Flow?

As it has been mentioned above, the complete absence of blood flow through the lumen of a doubly ligated segment of the common carotid artery in rats was associated with ischemic necrosis of the endothelium as well as the majority of VSM cells, during the first 24 hr, and a complete obliteration of the vessel lumen between 2 and 4 weeks post-ligation.³³ However, in a rat as well as mouse model of the distally ligated common carotid artery in which a proximal vessel segment has remained in communication with the lumen of the aorta, and hence, continued to receive some nutrients and oxygen from the arterial blood, the vessel wall appeared intact, though it has demonstrated the development of the prominent neointimal hyperplasia between two and four post-ligation weeks.^32,34,35 In light of these observations, it is important to highlight that one of the primary findings in this study was the development of neointimal formations in the wall of the residual coronary arteries which led to progressive narrowing of the vessel lumens between 2 and 12 post-MI weeks. A similar pattern of structural modifications in residual arteries has been reported previously by Kalkman et al.²¹ and Whittaker et al.,²² who found the presence of aberrant arteries with a high wall-to-lumen ratio in the post-MI scars of rats subjected to permanent ligation of the left coronary artery. Unfortunately, both studies did not elaborate on the potential mechanism of such arterial modifications detected in a rat coronary artery ligation model. However, considering the published results from other experimental models showing that the reduction of blood flow, and hence, the decrease in sheer stress either in the common carotid artery of rats^36,37 or in coronary arteries of pigs,³⁸ had led to progressive neointimal hyperplasia and, in some cases, to inward arterial remodeling during the 4-week period, it can be assumed that the same hemodynamic trigger, that is, reduced, but not absent, intraluminal flow, might be responsible for the progressive development of neointima and inward remodeling in the residual arteries investigated in this study. In this regard, bearing in mind the previously established fact that the increased stagnation of the blood flow in the arterial lumen had led to augmented centripetal growth of the neointima in a common carotid artery ligation model,³⁵ the progressive nature of the residual artery modifications observed in this study between 4th and 12th post-MI weeks can to a large extent be viewed as a consequence of the continuing scar maturation and stiffening³⁹ that would gradually reduce the patency of the Thebesian vessels and, therefore, could attenuate the sheer stress generated by intraluminal oscillation of blood flow leading to marked narrowing, and in advanced cases, to a complete obliteration of the vessel lumen in the affected residual arteries. Nevertheless, considering that the residual arteries remaining within the ischemic/fibrotic areas during 12 post-MI weeks had been exposed to a wide variety of regulatory as well as modulatory humoral factors, which were concurrently produced, on one hand, by endothelial and synthetic VSM cells inside the vessels itself^40,41 and, on the other hand, by various cells of the granulation/scar tissue surrounding them,⁴² the precise mechanism that might be responsible for orchestrating the arterial wall alterations in such vessels, including neointimal growth, intravascular fibrosis, and inward remodeling, seems to be much more complex.

Can the Residual Coronary Arteries of the Transmural Scar Be Suitable for Blood Reflow?

Although the existence of residual coronary arteries within the large transmural post-MI scars of rats is a long-established fact,^14,17,21,22 the information regarding the functionality of these vessels is scarce. For instance, in the past three decades, the vasoactivity of such arteries has been assessed in experimental settings only once.²¹ According to that investigation, the residual arteries found in 3-week-old transmural post-MI scars were still responsive to vasoactive stimuli despite evident structural abnormalities, which to a large extent appeared to be comparable with those seen in this study. Considering this prior finding, it would be feasible to presume that the residual arteries seen in this study might also maintain their vasoactive sensitivity, particularly during the first 2 weeks after MI when most of the remaining vessels did not yet demonstrate noticeable structural alterations. In this regard, it is important to draw attention to a recent study done by Hayashi et al.⁴³ on a rat common carotid artery ligation model demonstrating that blood flow, which had been reduced in the distal carotid segment to ~40–50% of the original level for 2 weeks due to partial ligation of the proximal artery, had been restored back to the normal level during 4 weeks following removal of the vessel constriction. This observation provided an additional support to the concept that the residual coronary arteries residing in the transmural post-MI scars can be considered as plausible routes for reinstatement of blood supply to the scarred region during at least a 2-week period after total coronary occlusion. However, because almost all residual arteries examined in this study between 4th and 12th post-MI weeks had experienced progressive neointimal hyperplasia accompanied by accumulation of the fibroelastic extracellular matrix that is known to markedly stiffen the arterial wall,⁴⁴ it seems reasonable to assume that following a 2-week window, the suitability of the remaining arteries for potential blood reflow inside the transmural scar would gradually decline.

Study Limitations

Despite the strong evidence provided by the current investigation in support of long-term survival of patent coronary arteries inside the large transmural post-MI scars, this study encounters several limitations. First of all, it would be of a great value to demonstrate on a model of isolated beating heart the presence of retrograde inflow of tracer microspheres from the LV cavity into the luminal space of the residual arteries resided in the scar to corroborate the existence of direct communications between these two compartments. Second, considering that this study did not assessed the changes of blood perfusion through the scarred region in response to vasoactive agents, the functional responsiveness of chronically disused arteries to humoral stimuli remained undetermined. Third, it would be essential to conduct either angiographic or three-dimensional micro-computed tomography assessment of the residual arterial bed to confirm its continuity inside the large scarred region. Finally, there has been no attempt made in this study to reconnect the residual coronary arteries located distal to permanent coronary ligation with the proximal segment of the left coronary artery to validate the likelihood of reinstatement of anterograde blood flow through these vessels.

Taken together, the findings in this study demonstrate that large- and medium-sized coronary arteries, remaining in transmural scar distal to total coronary occlusion, could maintain their structural integrity, including the tunica media composed of contractile VSM cells and the patent lumen, up to two post-MI weeks. However, all chronically disused arterial vessels residing in the mature scars had eventually undergone progressive neointimal hyperplasia, intravascular fibrosis, and inward remodeling during the following weeks that ultimately led to severe disruption of the arterial wall integrity and obliteration of the vessel lumen. These observations support a concept that in middle-aged rats, the residual coronary arteries of the transmural scar could be explored, at least during the first two post-MI weeks, as the potential endogenous routes for reestablishing the arterial blood supply to facilitate any cell-based regenerative therapies inside a large fibrotic area of the myocardium formed in response to total proximal coronary occlusion. Moreover, considering that in humans, each phase of the wound healing process following an MI proceeds slower than that in rats,^45,46 the time frame during which the residual coronary arteries of the human scar can remain suitable to support the reestablished arterial blood flow may be much wider than that found in this study. In this regard, although the current investigation does encounter several limitations, it provides important information that can be exploited to optimize the future cardiac regenerative therapy strategies in humans.