Author + information
- Received January 25, 1996
- Revision received October 3, 1996
- Accepted October 9, 1996
- Published online February 1, 1997.
- ↵*Dr. Cornel Badorff, Düsseldorfer Strasse 171, 40545 Düsseldorf, Germany.
Objectives. The purpose of this study was to investigate the expression of perforin and T-cell intracellular antigen-1, two crucial components of lymphocyte-mediated cytotoxicity, in endomyocardial biopsies from patients with idiopathic dilated cardiomyopathy.
Background. Previous reports have demonstrated the presence of myocardial interstitial fibrosis and increased infiltrating lymphocytes in patients with dilated cardiomyopathy. However, the pathogenic significance of these lymphocytic infiltrates remains unclear.
Methods. Endomyocardial biopsies from 134 patients with idiopathic dilated cardiomyopathy were histologically and immunohistologically analyzed. Monoclonal antibodies against diverse T-lymphocyte antigens, perforin and T-cell intracellular antigen-1 were used with the highly sensitive avidin–biotin complex technique. Positive cells were counted in at least 10 high power fields.
Results. Perforin and T-cell intracellular antigen-1 were immunohistologically detected in all biopsies. Immunoreactivity was restricted to the cytoplasm and was granular in nature, indicating specific staining of cytoplasmic granules. Correlations were established between the expression of perforin and T-cell intracellular antigen-1 and the abundance of foci of various T-lymphocyte subpopulations and, most importantly, the degree of interstitial fibrosis on routine histologic examination (p = 0.015).
Conclusions. Cytotoxic activity is clearly present in endomyocardial biopsies from patients with idiopathic dilated cardiomyopathy. Local activation—that is, focal accumulation of T lymphocytes—seems to be important for the generation of lymphocyte-mediated cytotoxicity. The interstitial fibrosis commonly seen in dilated cardiomyopathy may be caused by cytotoxic T-lymphocyte damage to the myocardium.
(J Am Coll Cardiol 1997;29:429–34)
By definition, idiopathic dilated cardiomyopathy is a heart muscle disease of unknown etiology. Viral myocarditis has emerged as being intimately involved in the pathogenesis of dilated cardiomyopathy (). Several lines of evidence suggest a myocardial enteroviral infection as one possible initiating event in a significant proportion of patients with dilated cardiomyopathy. The viral infection may trigger an autoreactive immunologic process, resulting from either molecular mimicry or expression of cardiac neoantigens (). In addition to the involvement of the humoral immune system, characterized by the generation of autoantibodies directed against various cardiac antigens, such as the adenine nucleotide translocator (), the cellular immune system seems to play a key role in the pathogenesis of dilated cardiomyopathy ().
Endomyocardial biopsy is often performed as an adjunct in establishing the diagnosis of dilated cardiomyopathy () to rule out any specific heart muscle disease (). The incidence of any effectively treatable specific heart muscle disease may be low (), but certain unspecific findings such as myocardial replacement or interstitial fibrosis can be demonstrated in the majority of patients (). The pathogenesis of this increased fibrous tissue deposition within the myocardium remains obscure.
Although the histologic diagnosis of acute or borderline myocarditis, according to the Dallas criteria (), can only be made in a minority of patients with dilated cardiomyopathy ([9, 10]), taking advantage of the more sensitive and specific immunohistologic approach has demonstrated the presence of increased myocardial mononuclear cell infiltrates in at least one-third of the patients with dilated cardiomyopathy (). Furthermore, increased expression of major histocompatibility complex antigens () and adhesion molecules () suggests an ongoing (auto-) immunologic reaction in at least some patients with idiopathic dilated cardiomyopathy. Although several mechanisms of both humoral and cellular immune-mediated myocyte injury are conceivable (), the pathogenic significance of myocardial infiltrating lymphocytes in idiopathic dilated cardiomyopathy is still poorly understood. Furthermore, no data are available to show a direct myocytotoxic effect of these infiltrating lymphocytes.
Recently, the molecular mechanisms of lymphocyte target cell killing were elucidated. On target cell recognition, cytotoxic T lymphocytes exocytose preformed cytoplasmic granules that contain perforin and T-cell intracellular antigen-1 (TIA-1) (). Perforin monomers, after release and cleavage by chymases, are ∼68-kDa proteins that bind to target cell membranes and subsequently aggregate to form polymeric pore structures. These pores perturb membrane permeability and result in osmotic lysis by sodium and water influx (). The physiologic importance of perforin-mediated target cell killing was documented in vitro by the administration of antisense oligonucleotides () and in vivo by the generation of perforin-deficient mice (). Both resulted in greatly impaired lymphocyte-mediated cytotoxicity. In contrast, expression of perforin, detected with monoclonal antibodies, correlated with the cytolytic activity of lymphocytes (). Apart from the direct lytic effects of perforin, the perforin pores act as a conduit for other mediators such as TIA-1. T-cell intracellular antigen-1 is a 15-kDa polyadenylate binding protein that is known to induce deoxyribonucleic acid (DNA) degradation on entry into the target cells and triggers FAS-mediated apoptosis of the target cell ([19, 20]).
In the present study we immunohistologically demonstrated the expression of perforin and TIA-1 in endomyocardial biopsies taken from patients with dilated cardiomyopathy and, in some cases, the direct release of these mediators onto the surface of cardiomyocytes. Furthermore, correlations with other immune markers of endomyocardial biopsies were established. The hypothesis to be tested was that mononuclear cellular infiltrates within the myocardium have a cytotoxic effect on TIA-1 and may cause the interstitial fibrosis commonly seen in dilated cardiomyopathy.
1.1 Study group.
This study included 134 patients (53 women and 81 men; mean [±SD] age 48.3 ± 12.7 years) with idiopathic dilated cardiomyopathy. All patients underwent echocardiography as well as right and left heart catheterization. Duration of symptoms was longer than 6 months (range 6 months to 10 years) without any clinical history of previous viral infection. All patients had a left ventricular diastolic diameter ≥6 cm, with diffuse hypokinesia and an ejection fraction ≤40% (). They did not have any evidence of hypertensive, valvular, systemic, ischemic or congenital heart disease ().
Multiple endomyocardial biopsies from the right side of the ventricular septum were performed by way of the right femoral vein using a CORDIS bioptome (CORDIS, Haan, Germany) under fluoroscopic guidance.
Cardiac control tissue was taken at autopsy from patients with no cardiac pathology who had died from noncardiac causes (6 from trauma; 6 from suicide; 2 other deaths). Most samples were taken from the right ventricular septum.
1.2 Sample preparation.
For conventional histologic examination, two biopsies were fixed in 10% formalin. For immunohistologic study, the remaining biopsy specimens were covered with TissueTec embedding medium (SLEE, Mainz, Germany) and snap-frozen in methylbutane in liquid nitrogen at −70°C . For immunohistologic study, biopsy specimens were serially cut into 5-μm thick sections, and six to nine serial sections were analyzed for each antibody. Biopsies were examined by two observers in blinded fashion.
1.3 Light microscopy.
Specimens were stained with hematoxylin-eosin according to standard protocols. Active or borderline myocarditis was a histologic diagnosis according to the Dallas criteria (). Active myocarditis required myocyte necrosis with an adjacent inflammatory infiltrate, whereas borderline myocarditis was diagnosed when infiltrating mononuclear cells were seen in the absence of myocyte injury, with or without interstitial or reparative fibrosis. The degree of fibrosis was semiquantitated from 0 = no; 1 = mild; 2 = moderate; to 3 = severe.
After fixation with acetone for 10 min, cryosections were incubated with commercially available monoclonal antibodies against CD3 (pan T cells), CD4 (helper/inducer T cells), CD8 (suppressor/cytotoxic T cells), UCHL-1 (activated memory T cells) and major histocompatibility complex (MHC) class I and II (DR subtype) antigens, as described previously ().
For both antibodies, the sensitive avidin–biotin complex (ABC) procedure () was performed using the Vectastain elite ABC peroxidase kit (Vector). All incubations were performed at room temperature. Briefly, after blockade of endogenous peroxidase with hydrogen superoxide (0.3% in phosphate-buffered saline [PBS]) for 20 min and blockade of endogenous avidin and biotin binding sites (avidin/biotin blocking kit, Vector) for 15 min, slides were incubated with the appropriately diluted primary antibody (antiperforin 1:10, anti-TIA-1 1:400 in PBS/5% fetal calf serum) for 1 h. Then, biotinylated antimouse immunoglobulin G was added for 1 h, and the ABC reagent was subsequently allowed to react for 30 min. Staining was performed using 3-ethyl-9-aminocarbazol as a chromogen against hematoxylin counterstain. Sections were finally covered with Kaiser’s glycerol gelatine.
Cells positive for CD3, CD4, CD8, UCHL-1, perforin and TIA-1 were counted in at least 10 high power fields (HPF) under 400× magnification (1 HPF = 0.28 mm2). Major histocompatibility complex antigen expression was semiquantitated, as previously described (). The immunohistologic diagnosis of increased lymphocytic infiltrates was made for a mean lymphocyte cell count ≥2.0 cells/HPF (∼7 cells/mm2) accompanied by activated lymphocytes or increased MHC antigen expression (). A lymphocytic focus within the myocardium was defined as clustering of at least three positively stained cells. The abundance of foci within a biopsy specimen was graded as follows: 0 = no focus in the entire biopsy specimen; + = focus in at least one high power field; ++ = foci in at least half of the analyzed high power fields; +++ = foci in virtually all analyzed high power fields.
In dilated cardiomyopathy, positive staining for perforin (≥2.0 positive cells/HPF) and TIA-1 (≥1.5 positive cells/HPF) was defined as the mean value ± SD. Positive staining for cell-mediated cytotoxicity was defined as positive staining for both perforin and TIA-1.
1.5 Statistical analysis.
Values are expressed as mean ± SD. Statistical analyses were performed using the SPSS software package. The Student ttest was used to compare quantitative data, and the chi-square test was used to compare qualitative data. Pearson’s correlation coefficient was used for quantitative data and Spearman’s correlation coefficient for semiquantitative data. A probability value <0.05 was considered statistically significant.
In control hearts, no gross cardiac pathology and low lymphocyte counts were observed (≤1.5 CD3+cells/HPF). Perforin and TIA-1 immunoreactivity was detected in the majority of specimens (85%), but their expression was low (0.83 ± 0.52 cells/HPF and 0.58 ± 0.40 cells/HPF, respectively).
In routine hematoxylin-eosin–stained sections from patients with dilated cardiomyopathy, 6 (4.5%) of the 134 biopsy specimens had an increased number of interstitial inflammatory cells without myocytolysis, yielding the diagnosis of borderline myocarditis according to the Dallas criteria. Acute myocarditis was not seen in any of the examined biopsies.
In the patient group, immunohistologic cell marker analysis of frozen sections, however, revealed a prevalence of 43% (57 of 134 biopsies) of pathologically increased lymphocytic infiltrates. The mean cell counts for CD3+cells were 3.3 ± 2.3 cells/HPF in the positive samples and 1.3 ± 0.5 cells/HPF in the negative samples.
Perforin and TIA-1 were immunohistologically detected in all dilated cardiomyopathy biopsy specimens. The staining pattern for both perforin and TIA-1 showed an immunoreactivity that was restricted to the cytoplasm and appeared granular as well as focal in nature (Fig. 1). Usually, single cytotoxic cells were detected in the myocardium (Fig. 1). Occasionally, perforin- and TIA-1-positive cells were clustered together in foci as a sign of local cytotoxic effector-cell activation (Fig. 1). In some cases, staining of adjacent interstitium and myolemma suggested a direct release of perforin and TIA-1 onto the surface of cardiomyocytes (Fig. 1). There was a highly significant correlation between the expression of perforin and TIA-1 (p < 0.0001), indicating the specificity of both monoclonal antibodies used. The mean cell count for perforin and TIA-1 was 1.31 cells/HPF and 0.94 cells/HPF, respectively. The corresponding standard deviations were 0.67 cells/HPF and 0.51 cells/HPF, respectively.
On average, cell counts for perforin (p < 0.01) and TIA-1 (p < 0.02) per HPF were significantly higher in the tissue specimens taken from patients with dilated cardiomyopathy in comparison with the control group.
The expression of perforin and TIA-1 did not differ significantly between immunohistologically positive and negative dilated cardiomyopathic biopsies (Table 1). Immunoreactivity of perforin and TIA-1 in the group with cardiomyopathy, either expressed as cell count per HPF or positively labeled biopsy, correlated with the abundance of foci of CD3+(pan T) cells, CD8+(cytotoxic/suppressor T) cells and UCHL-1+(activated memory T) cells (Table 1). Correlations with the mere cell count for CD3, CD8 and UCHL-1 were of little or no significance. CD4 (helper/inducer T cell) immunoreactivity correlated less consistently with the various variables of perforin and TIA-1 expression.
Positive staining for cell-mediated cytotoxicity (perforin ≥2.0 cells/HPF and TIA-1 ≥1.5 cells/HPF, defined as mean value ± SD) in dilated cardiomyopathy correlated with the expression of vascular and interstitial MHC class I antigens (p = 0.0253, p = 0.0081), whereas no relation to MHC class II antigen expression was observed. Furthermore, positive staining for cell-mediated cytotoxicity furthermore correlated with the degree of interstitial fibrosis on routine histologic examination (p = 0.015).
The findings of the present study demonstrate that increased cytotoxic activity—immunohistologically detected as perforin and TIA-1 expression—is present in endomyocardial biopsies from patients with dilated cardiomyopathy. There was a linear relation between local activation, reflected by focal accumulation of suppressor/cytotoxic and activated memory T lymphocytes within the myocardium and the degree of cell-mediated cytotoxicity variables. Positive staining for perforin and TIA-1 correlated with the vascular and interstitial major histocompatibility complex class I antigen expression and, most importantly, with the degree of interstitial fibrosis on routine histologic study.
3.1 Immunologic factors in the pathogenesis of dilated cardiomyopathy.
The pathogenesis of dilated cardiomyopathy, which is still largely unknown, has been linked to viral myocarditis (). The hypothesis that dilated cardiomyopathy may be a sequela of viral myocarditis is based on the presence of an enteroviral genome within the myocardium of a significant proportion of patients with both myocarditis and dilated cardiomyopathy (). Furthermore, although the incidence of acute myocarditis, according to the DALLAS criteria (), in patients with dilated cardiomyopathy varies greatly between the different studies (), immunohistologic studies have shown that pathologically increased lymphocytic infiltrations can be detected in at least one-third of the patients with dilated cardiomyopathy (). These findings and the demonstration of increased expression of MHC antigens () and adhesion molecules (), both characteristic of lymphocyte activation, suggest a chronic inflammatory process in dilated cardiomyopathic hearts (). However, the pathogenic significance of these myocardial lymphocytic infiltrates remains to be established, as there are currently no data on the mechanism of T-lymphocyte–mediated myocyte injury in dilated cardiomyopathy.
3.2 Mechanisms of myocyte injury by lymphocytes.
Principally, there are several possible mechanisms of immunomediated myocyte injury, as reported elsewhere (). According to the most widely supported granule–exocytosis model, cytotoxic effector cells degranulate on target cell recognition and release cytolytic mediators such as perforin and TIA-1 (). Perforin acts by forming large, nonselective ion channels in the target cell membrane, which results in osmotic cell lysis (). In isolated guinea pig ventricular myocytes, perforin induces a shortening of the action potential duration and a decline in both the rest potential and action potential amplitude, leading to myocyte contracture and destruction (). T-cell intracellular antigen-1 is known to induce DNA degradation and thus triggers FAS-mediated apoptosis of the target cell ([19, 20]).
3.3 Perforin and myocarditis.
In a murine model of Coxsackie virus B3–induced myocarditis, perforin was found in the cytoplasmic granules of heart-infiltrating cells expressing perforin messenger ribonucleic acid (). In addition, the release of perforin directly onto the surface of cardiac myocytes and circular lesions characteristic of perforin pores in the membrane of cardiomyocytes could be demonstrated by using immunoelectron microscopy (). Young et al. () extended these experimental findings to human myocarditis, where they were able to show perforin expression in endomyocardial biopsies from seven patients with acute myocarditis. Taken together, these results suggest an important role for perforin-mediated myocardial damage in myocarditis. Until now, no data were available regarding the expression of cytolytic mediators and the importance of cell-mediated cytotoxicity in dilated cardiomyopathy.
3.4 Perforin and TIA-1 in dilated cardiomyopathy.
To our knowledge, our findings are the first to show increased expression of perforin and TIA-1 in endomyocardial biopsies from patients with idiopathic dilated cardiomyopathy. Our results confirm and expand the findings by Seko et al. (), demonstrating perforin expression in the hearts of four patients with dilated cardiomyopathy. However, their study neither included a control group nor quantitated the perforin expression.
Cardiac control tissue taken from patients who had died from noncardiac causes displayed significantly lower perforin (p < 0.01) and TIA-1 expression (p < 0.02). Using specific anti-perforin and anti-TIA-1 monoclonal antibodies ([19, 22]) and the sensitive ABC technique (), we were able to demonstrate perforin and TIA-1 immunoreactivity restricted to the cytoplasm, often appearing focal and granular in nature. This characteristic staining pattern is in good agreement with previous antibody-based studies ([18, 28, 29]) and implies specific staining of cytoplasmic granules. Furthermore, there was a highly significant correlation between the expression of perforin and TIA-1 (p < 0.0001), indicating the specificity of both antibodies used and the co-expression of these cytolytic mediators. In some cases, we observed staining of the adjacent interstitium and myolemma, indicating a direct release of perforin and TIA-1 onto the surface of cardiomyocytes. This phenomenon, which is also known from experimental and human myocarditis ([26, 27]), suggests an important role for cell-mediated cytotoxicity in dilated cardiomyopathic hearts.
3.5 Correlations of perforin and TIA-1 with further immune markers in dilated cardiomyopathic hearts.
Myocyte injury and subsequent necrosis leads to interstitial replacement fibrosis, a frequent but nonspecific finding in the myocardium of patients with dilated cardiomyopathy (). In our study, positive staining for cell-mediated cytotoxicity (defined as positive staining for perforin and TIA-1, based on mean ± SD) correlated with the degree of interstitial fibrosis on routine histologic study (p = 0.015). Although this observation is not necessarily indicative of a causal relation, the interstitial fibrosis commonly seen in dilated cardiomyopathy may be caused by cytotoxic T-cell damage to the myocardium. The fact that no correlation was found between interstitial fibrosis and perforin or TIA-1 expression alone may be explained by the lower stringency of positive staining for just one antigen and underlines the importance of investigating multiple antigens in immunologic myocardial cell phenotyping (). In addition, other factors, such as fibroblast-activating cytokines like interferon-gamma, may be important, as they modulate extracellular matrix composition ().
Apart from structural myocardial damage due to the “lethal hit” delivery, cytotoxic T-cell–derived perforin (and TIA-1) is able to cause reversible target cell injury without lysis. This is associated with a transient decrease in contractility (). This “sublethal hit”—invisible on light microscopy—may explain the often variable clinical course of dilated cardiomyopathy ().
In the past, there was considerable controversy as to whether a biopsy specimen with lymphocytes can be labeled as myocarditis in the absence of any visible myocyte injury (). Indeed, in our study, the immunohistologic diagnosis of increased lymphocytic infiltrates did not correlate with any variables of cell-mediated cytotoxicity. This finding implies that the mere presence of immune cells within the myocardium does not automatically imply a cytotoxic effect on the cardiomyocytes. However, other mechanisms such as cytokine generation may be relevant.
Rather, we found significant correlations between the expression of perforin and TIA-1 and the abundance of foci of CD3+(pan T) cells, CD8+(cytotoxic/suppressor T) cells and UCHL-1+(activated memory T) cells. Correlations with the mere cell count for CD3, CD8 and UCHL-1 were mostly of little or no significance. These results suggest that local activation—that is, focal accumulation—of effector cells seems to be important for the generation of cell-mediated cytotoxicity.
Positivity for cell-mediated cytotoxicity correlated with the presence of increased major histocompatibility complex class I antigen expression on the vascular endothelium and interstitium. This may be of pathogenic relevance because target-cell-increased MHC class I antigens facilitate the interaction between cytotoxic effector cells and their target cells ().
3.6 Study limitations.
The conclusions that can be drawn from the present study are somewhat limited, because the study provides observational data but no definite proof that expression of perforin and TIA-1 has a functional impact on cardiac myocytes. Nevertheless, we believe that this first description of cell-mediated cytotoxicity in idiopathic dilated cardiomyopathic hearts greatly expands the currently poor knowledge regarding the pathophysiology of this medically important disease.
We elected to use autopsy specimens from noncardiac deaths as a control group rather than biopsy specimens from patients with atypical chest pain or serious arrhythmias. The rationale was to exclude biopsy specimens from patients with a patchy myocardial process such as lymphocytic myocarditis, which was missed owing to sampling errors (). Therefore, the role of cell-mediated cytotoxicity in other cardiac diseases remains to be established.
This study is the first to demonstrate increased expression of perforin and TIA-1 in idiopathic dilated cardiomyopathic hearts. Correlations with other biopsy markers of immune activation and the degree of interstitial fibrosis, as well as the direct release of these cytolytic mediators onto cardiomyocyte surfaces, suggest a pathogenic role for cell-mediated cytotoxicity in idiopathic dilated cardiomyopathy. Dilated cardiomyopathy and myocarditis appear to share common mechanisms of immune-mediated myocyte injury, which further substantiates the link between these two clinically different entities.
- avidin–biotin complex
- high power field
- major histocompatibility complex
- phosphate-buffered saline
- T-cell intracellular antigen-1
- Received January 25, 1996.
- Revision received October 3, 1996.
- Accepted October 9, 1996.
- The American College of Cardiology
- Martino TA,
- Liu P,
- Sole MJ
- Schultheiss HP
- Limas CL,
- Goldenberg IF,
- Limas C
- Mason JW,
- O’Connell JB
- Kasper EK,
- Agema WRP,
- Hutchins GM,
- Deckers JW,
- Hare JM,
- Baugham KL
- Kühl U,
- Noutsias M,
- Seeberg B,
- Schultheiss H-P
- Barry WH
- Nakata M,
- Kawasaki A,
- Azuma M,
- et al.
- Tian Q,
- Taupin J,
- Elledge S,
- Robertson M,
- Anderson P
- ↵Report of the WHO/ISFC Task Force on the definition and classification of cardiomyopathies. Br Heart J 1980;44:672–3.
- Felzen B,
- Berke G,
- Rosen D,
- et al.
- Seko Y,
- Shinkai Y,
- Kawasaki A,
- et al.
- Ojcius DM,
- Zheng LM,
- Sphicas EC,
- Zychlinsky A,
- Young JDE
- Holzinger C,
- Schöllhammer A,
- Imhof M,
- et al.
- Keren A,
- Gottlieb S,
- Tzivoni D,
- et al.