- Email: [email protected]
Myocardial inflammatory infiltrate in human chronic chagasic cardiomyopathy: Immunohistochemical findings
ELSEVIER
SYMPOSIUM
Myocardial Inflammatory Infiltrate in Human Chronic Chagasic Cardiomyopathy: Immunohistochemical Findings Jo& Milei, MD,* Graciela Fernhdez Alonso, MD,+ Silvia Vanzulli, MD,+ Rub& Storino, MD,*’ Luigi Matturri, MD,” and Lino Rossi, MD” From *Cardiopsis,
the tNational Academy of Medicine, and the $INCALP Foundation, Buenos Aires, Argentina; and the $Department of Pathology, University of Milan, Milan, Italy
Chagas’ disease is the most common form of chronic myocarditis in the world. It is characterized by a progressive chronic myocarditis that leads to cardiomegaly, arrhythmias, cardiac failure, and thromboembolic phenomena. This communication reports studies on the immunohistochemistry of chronic infiltrates in 30 endomyocardial biopsies and in contracting and specialized myocardium of autopsies of four patients suffering from Chagas’ cardiomyopathy. Expression of the following antigens was studied: common leucocyte antigen (CLA-CDQSR), L-26(CD20), CD68, kappa and lambda light chains and T-UCLH-1 (CD45RO), and MB-l. Streptavidin-peroxidase and streptavidin-alkaline phosphatase with biotinylated anti-mouse IgG were used as detection systems. Double immunostaining for the simultaneous demonstration of T lymphocytes (CD45RO) and macrophages was performed using both immunoenzymatic techniques consecutively. Expression of CD31 was detected for the demonstration of endothelial cells. In endomyocardial biopsies, tissue forms of trypanosomes were not found. The percentage of fibrous tissue was 24.1% ? 12.8% (range 8.2%49%). Eosinophils were scarce (l/high-power field), but associated with necrotic areas of the myocardium. Mast cells were scarce or absent. They were always situated in fibrotic areas. The most remarkable finding was the presence of infiltrates consisting of macrophages and CLA-positive mononuclear cells. Twenty-six and one-half percent of them were T lymphocytes, and 10.5% were B lymphocytes. Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial lesions. Thirty percent of the infiltrate was composed of macrophages (positive CD68 cells). The remaining infiltrate was composed of mononuclear cells resembling macrophages and CLA-negative mononuclear cells. Contacts between CD68-positive cells and T lymphocytes were frequently found. CD31 antibodies clearly pointed out normal endothelial cells, in either normal or damaged vessels. No isolated cells positive for these antibodies were found within the mononuclear infiltrate. In autopsied hearts, myocardial lesions consisted of a chronic inflammatory process with fibrotic scars and extensive mononuclear infiltrates. No amastigote nests were found. A statistically significant difference @ < 0.05) was obtained when the percentage of fibrosis was compared in the specialized and contracting myocardiums (5 1.6% -t 18% vs. 43.4 % -C 8%). Eosinophils were scarce in infiltrates, reaching 5%, and they were associated with necrotic myocardium. Mast cells also were scarce or absent in specialized and in contracting myocardium. Almost all the lymphocytic population was T lymphocytes. Such infiltrates were more prominent in the working myocardium (39%) and in the specialized cells of the left branch of the His bundle than in the atrioventricular node and in the right Hisian branch (34.4%). The 31% of mononuclear cells were CD68 positive, thus corresponding to macrophages. Contacts among T lymphocytes and macrophages were frequently observed. Although much that is concerned with Chagas’ cardiomyopathy is controversial, these may be the major ingredients for its pathogenesis: the parasite or a part of it, lymphocytes and macrophages, and fibrosis. Then a multifactorial or “combined theory” may be suggested to explain the sequence of events that lead to the chronic stage of the disease. Cardiovasc Path01 1996;5:209-219 Manuscript received December 15, 1995; accepted January 10, 1996. Address for reprints: Dr. Jose Milei, CARDIOPSIS, Tucuman 2163, 4B, 1050 Buenos Aires, Argentina; fax: 00541-951-0366. Cardiovascular Pathology Vol. 5, No. 4, July/August 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
Presented at an international held in Milan, Italy, June 26-28,
symposium 1995.
on Chagas’ heart disease
1996:209-219 1054-8807/96/$15.00 PI1 SlO54-8807(96)00006-3
210
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Pathol Vol. 5, No. 4 July/August 1996~209-219
INFKTRATION
Chagas’ disease (American trypanosomiasis) is the most common form of chronic myocarditis in the world (1,2). It is estimated that at least 20 million individuals are infected with Trypan~~rna crxzi and that 90 million live in endemic areaswhere the risk of this infection exists (3). The progressive chronic cardiac form appearsin 10% to 20% of previously asymptomatic patients, 15 to 20 years after the infection (4). It is found in adults 30 to 50 years of age and is manifestedby atrioventricular (AV) block and/or right bundle branch block plus left anterior hemiblock and ventricular ectopic beats (4). It is characterized by a progressive chronic myocarditis that leads to cardiomegaly, arrhythmias, cardiac failure, and thromboembolic phenomena(5). Despite the importance of the disease,little is known of its pathogenesis(6). At the chronic stage, there is persistence of scarceparasitesin the myocardium and low-level parasitemias.Mononuclear infiltrates and extensive areasof fibrosis with apical ventricular aneurysmsare the usual lesions (7). These findings support the hypothesis that the chronic chagasic cardiomyopathy (CCC) is mediated, at least in part, by immune phenomena. Recently, identification of 7: cruzi parasites in hearts with CCC was facilitated with the use of a polyclonal antiZ cruzi antibody serum.There was a significant correlation between the presenceof the parasite antigensand moderate or severeinflammatory infiltrates (8). This observation was confirmed when an amplification of a 7: cruzi DNA sequence from inflammatory lesions was shown in human CCC (9). These studiesdemonstratedthat r cruzi or a portion of its genomeis presentin the inflammatory portion of infiltrates and supports a direct role for the parasite in the perpetuation of the inflammation (8). However, the chronic chagasicinfiltration hasrarely been characterized by means of immunohistochemicalmethods (7,10,11). This communication reports studies on the immunohistochemistry of chronic infiltrates in endomyocardial biopsies and in contracting and specialized myocardium of autopsies of patients suffering from CCC.
Material and Methods The study was conducted on 30 endomyocardial biopsies from an equal number of patients (7) and on four heartsbelonging to autopsiedpersonswho suffered from CCC (10). There were malesand females. Ages ranged from 20 to 69 years (mean, 45.1 years). Both the clinical and the pathologic features establishedthe diagnosisof Chagas’disease. The clinical and pathologic characteristics of each case were already reported (7,lO). Chagas’ serology was studied for four standardreactions against T cruzi (immunofluorescence, complement fixation, passivehemagglutination, and latex test). Patients were divided into 3 groups, according to the Chagas’ DiseaseCouncil of the Argentine Society of Cardiology (12): Group I: asymptomatic patients, normal thorax X ray and ECG; Group II: conduction disturbances
and/or arrhythmia; and Group III: cardiomegaly with heart failure and.abnormal ECG.
Procedure Endomyocardial biopsies were performed under fluoroscopic and bidimensional echocardiographic guidance by percutaneous puncture of the right internal jugular vein; Four to five specimenswere immediately placed in 10% buffered formalin. Autopsies were performed within 4 hours of death, and tissue sampleswere also immediately fixed in 10% buffered formalin (pH 7.0).
Histological
Methods
The biopsy samplesmeasured2 to 3 mm in the greatest dimension. The specimenswere routinely processed for light microscopy by embeddingin paraffin. Sections (4 pm in thickness) were serially cut and mounted so that each slide contained three to four sections. These serial slides were processed consecutively for immunohistochemical methodsand stainedwith hematoxylin and eosin, BarbeitoLopez trichrome, the PAS method, and the silver-methenamine technique. For the morphometric assessmentof fibrosis, the Mallory trichrome stain was used, and each tissue section was examined under a grid. The Alcian blue technique (pH 2.5) was used to detect and quantify mast cells in 100 high-power fields (HPF; X 400). Giemsa stain was also used to quantify eosinophils. All the slides were examined by two observers who had no knowledge of the clinical findings. In the autopsiedcases,histologic examination of the conduction systemon serial sectionswas carried out following the technique usually employed by one of the authors (LR) (13). Four to eight sectionsof each case,from the right and left ventricular walls were also examined. The hematoxylin and eosin, Alcian blue (pH 2.5), Giemsa, and silver-methenamine methods were routinely followed for all sections; Bielschowsky-silver impregnation was usedfor the nervous system. Each section stained with the Mallory trichrome was examined under a grid to establishthe percentageof fibrosisin the specializedand ordinary myocardium. As far as the intrinsic cardiac nervous system is concerned, observationswere carried out on the stageand number of neuronal bodies inside individual ganglia of the AV groove plexus, the ganglia of the sinoatrial (SA) nerve plexus, the intercaval funnel adventitia, and the posterior wall of the left atria.
Immunohistochemical
Methods
For cellular immunophenotyping of the mononuclear infiltrate, nnmunoenzymatic techniqueswith monoclonal primary antibodieswere performed. Serial sectionsof endomyocardial biopsies from 14 se-
Cardiovasc July/August
Path01 Vol. 5, NO. 4 1996:X9-219
CHRONIC
lected cases and random sections from the anterior, medial, and posterior areas of the bundle of His and its branches, as well as several sections from the right and left free ventricular walls from the autopsied cases were processed. Expression of the following antigens was studied: common leukocyte antigen (CLA-CD45R), L-26(CD20), CD68, kappa and lambda light chains (Dakopatts, Santa Barbara, CA), and T-UCLH-1 (CD45RO) and MB-l (Biogenex). Streptavidin-peroxidase and streptavidin-alkaline phosphatase with biotynilated anti-mouse IgG were used as detection systems and reactions were developed with 3.3’ diaminobenzidine (DAB) and with fast red. Double immunostaining for the simultaneous demonstration of T lymphocytes (CD45RO) and macrophages was performed using both immunenzymatic techniques consecutively. Peroxidase and DAB was used for T lymphocytes and alkaline phosphatase and fast red for macrophages (14). Expression of CD31 and Ulex Europaeus I was detected to demonstrate endothelial cells in order to study both the relationship between infiltrates and vessels and endothelial integrity. Positive cells for the different antigens studied were quantified in both contractile myocardium and specialized myocardium. The number of negative and positive mononuclear cells for each marker was counted on three HPF/slide. These numbers were correlated with the extent of interstitial fibrosis and/or myocyte damage on each sample. According to previous studies (15-17), we considered the normal mean number of lymphocytes to be less than 5/ HPF. Then, S/HPF was considered as the lower limit of myocarditis if it was associated with myocyte injury (16). Specifically for grading purposes and independently of previous grading using hematoxylin and eosin-stained preparations, any patient having a mean of 5 CLA-positive cells/HPF would be classified as having 1+ infiltration of the heart. A mean of 6 to 10 mononuclear cells/HPF was graded as 2-t) and a mean greater than 10 lymphocytes/HPF was graded as 3+ (17).
Results Endomyocardial
Biopsies
Clinical data and histological features of these patients were previously reported (7). Patients belonged to groups II or III (12). In brief, the 30 biopsy specimens showed mild to moderate hypertrophy of the myocytes and variable degrees of interstitial fibrosis (mild in 15 patients, moderate in 11, and severe in 4). Lesions consisted of a chronic inflammatory process with fibrotic areas and mononuclear cellular infiltrates. The lesions showed a microfocal disposition. Coagulative necrosis and/or myocytolysis were frequently observed. A marked subendocardial sclerosis, presumably due to an increased end diastolic pressure, was also observed.
CHAGASIC
MILE1 ET AL. INFILTRATION
211
Figure 1. A granulating mast cell surrounded by dense connective tissue. Alcian blue, pH 2.5, X400.
Tissue forms of trypanosomes were not found. Diffuse or focal thickening of the basement membranes was observed after staining with the silver-methenamine technique. Mononuclear cell infiltrates were found in 25 biopsies. The endocardium ranged from 4.8 to 43.6 pm in thickness, with a mean (2 SD) of 24 + 12.6 pm, and the diameter of the myocytes ranged from 11.4 to 40 km, with a mean of 20 & 7.33 pm. None of these data could be correlated with the ejection fraction in each patient. Fibrosis. The percentage of fibrous tissue was 24.1% + 12.8% (range 8.2%-49%). Only 1 of the 30 biopsies contained less than 10% fibrous tissue, and 20 of the 30 biopsies contained more than 15% fibrous tissue. Eosinophils and mast cells. Eosinophils were scarce (l/I-IPF), but associated with necrotic areas of the myocardium. Mast cells were scarce or absent. They were always situated in fibrotic areas (Figure 1). Lymphocytes, T and B lymphocytes. The most remarkable finding was the presence of infiltrates consisting of macrophages and CLA-positive mononuclear cells (true lymphocytes; Figure 2). Although a high mean of apparently mononuclear cells may be found in a microscopic HPF, only 28.5% of them may result in true lymphocytes (Table 1). Immunostaining for K and X chains and L-26 antigen or MB1 showed that only the 10.5% of the CLA-positive lymphocytes belonged to the B subset while the normal (2.5/ l)K/X ratio was maintained. Conversely, almost all the lymphocytic population was due to T lymphocytes (26.5% of the total number of mononuclear cells; Table 1; Figure 2). Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial lesions (Figure 3). From the 14 samples previously diagnosed as CCC, 12 showed 5 or more CLA-positive mononuclear cells/I-IPF. From these results, 12 of the 14 cases resulted in “active chronic chagasic myocarditis.” The remaining two samples were classified as being “chronic chagasic cardiomyopathies.”
212
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Path01 Vol. 5, No. 4 July/August 1996~209-219
EWILTRATION
mal or damagedvessels.No isolated cells positive for these antibodies were found within the mononuclear infiltrate (Figure 6).
Autopsied Hearts Grossly, the four heartspresentedsevere dilatation, flattening of the papillary muscles,and a marked subendocardial sclerosisrelated to an increasedend diastolic pressure. A remarkable thinning of the left ventricular wall was characteristic at the apical areas.Intracavitary tbrombi were also found. Microscopically, myocardial lesions consisted of a chronic inflammatory processwith fibrotic scarsand extenFigure 2. Clusterof mononuclearcells. Doubleimmunostaining sive mononuclearinfiltrates (Figure 7). No amastigotenests for the simultaneous demonstration of T lymphocytes(CD45RO) were found. Extensive myocytolysis and spotty contraction and macrophages (CD68). PeroxidaseandDAB (T lymphocytes, band necrosiswere observed. Cell hypertrophy in the apparbrown)andalcalinephosphatase and fast red (macrophages, negaently preserved myocytes was revealed by hypertrophic bitive in this section), X 1,000. zarre nuclei. Overall results in the 12 caseswith active chronic chaAs previously described(4-lo), dilated lymphatic changasic myocarditis showed that T and B lymphocytes reprenels widespread in the ventricular septum and in the AV sented27.3% and 10.8% of the total mononuclearinfiltrates. node, His bundle, and in the root of the right and left bunMacrophages. The 30% of the infiltrate was composed dles branch were observed. Around the apical aneurysm of of macrophages(positive CD68 cells; Figure 4). The rethe left ventricle, such dilated lymphatics were distributed maining infiltrate was composedof mononuclear cells resubepicardially. sembling macrophages and CLA-negative mononuclear The serial sectioning in the conducting system showed cells (Table 1, Figures 2 and 3). Of note, contacts between mild to moderatediffuse fibrosis of the AV node and of the CD68-positive cells and T lymphocytes were frequently “penetrating” and “branching” portions of the His bundle, found (Figure 5). complete destruction of the proximal segmentsof the right Endothelial cells. CD3 1 and Ulex europaeusantibodies and left bundlesbranch by varying degreesof replacement clearly pointed out normal endothelial cells, either in norby densecollagen tissue(Figure 8).
Table 1. Monoclonal
Antibody Marker Quantitative Analysis of Interstitial Infiltrates in 14 Patients with Chronic Chagasic
Cardiomyopathy
Patient No. 1
Total Lymphocytes/ Total Mononuclear Cells” (X)
Total T Cells/ Mononuclear Cell?
36170 44/80 19/40
(>5)
(>5) (>5) (b-5) (>5)
29165 29154 19147
(~5)
(<5) (>5)
6141 33/66
12
(>5) (>5) (>5)
57/129 46192 28/89
13 14
(>5) (<5)
20/55 3/20
2 3 4 5 6 7 8 9
10 11
Total The numbers
(>5) p-5)
(X) (4)
--
386/969
(>5)
10/37 23/63 26162 32162
p-5) (<5) (>5)
15/45 2132 36/100
(>5) (<5)
68/1&I 6126
(>5) (>5)
18/64 17/51
(<5)
l/12
(X)
17/121
26/58 20/56
(28.5%)
of the cells are given. Mast cells and eosinophils
were always
‘Represents
the sum of the number
of specific
cells counted
in HPF.
(<5)
42170 20139 25168
(4) (4)
5/43 2/69 9/28
(<5) (<5) ((5) (<5)
7149 8/34 4/24 6/31
(<5) (<5)
2/38 o/54
(<5) (<5) (<5)
5176 3/81 o/79
(<5) (<5)
7155 5/49
18/46 19162 4134 36176 611125 47170 27182
18154 4124 369/867
Total B Cells/ Mononuclear Cellsb
22l64
26153
300/832(26.5%)
“(>5) or (<5) lymphocytes, T or B lymphocytes/HPF.
Total CD68 Positive Macrophage Cells/Mononuclear Cell?
and were therefore
(29.8%)
not tabulated.
83/710(10.5%)
Cardiovasc July/August
Path01 Vol. 5, No. 4 1996:20!-219
CHRONIC
CHAGASIC
MILEIETAL. INFILTRATION
213
Figure 3. Doubleinnnunostaining for the simultaneous demonstration of T lymphocytes(CD45RO) andmacrophages (CD68).(A) T lymphocytes(brown)in closecontactwith a macrophage (pinkcytoplasm).X 1000.(B) Two closemononuclear cellsin theneighborhood of a damagedmyocyte.The T lymphocyte (brown) has a close contact with the sarcolemma. The macrophage, negative for CD68 antibody (large nucleus and pink cytoplasm) is also very close to a round cell, presumably a B lymphocyte. X 1,000.
The remaining specialized fibers presentedatrophy and mild fatty infiltration and were surroundedin most casesby infiltrates consisting mainly of lymphocytes and macrophages(seelater) (Figures 9 and 10). Venular and lymphatic vascular dilatations were frequently observedin thoseareas. No selective involvement of the right half of the His bundle was observed, as previously reported (10). The subendocardial Purkinje fibers were usually damagedby chronic inflammation and fibrosis, aswas describedfor contracting myocytes. Diffuse or focal thickening of basal membranes of contracting and specialized myocytes were frequently observed asrevealed by the silver-methenaminemethod. In the neurovegetative ganglia, no actual neuronal loss has been found; only early increasein satellite cells and/or mild mononuclearinfiltration was rarely observed.
Spotty round cell epiperineuritis was detected in one case in the subepicardiumof the wall of a left aneurysm. Morphometric data are summarizedin Table 2. Fibrosis. Statistical significance @ < .05) was obtained when the percentage of fibrosis was compared in the specialized and contracting myocardium (51.6% Ifr 18% vs. 43.4% ? 8%). It suggestedthat the fibrotic lesionsof CCC were randomly distributed through the conducting system and the ordinary myocardium. A similar degree of fibrosis was observed in the sectionsbelonging to the free wall of the left ventricle (41.3% & 6%). Eosinophils and mast cells. Eosinophils were relatively scarcein infiltrates, reaching 5% and frequently associated
Figure 4. Extensive mononuclearinfiltrates. Roundcells are in
Figure 5. Area of myocardiuminfiltrated by smallclustersof ei-
contact with damaged myocytes of forming extensive clusters.Approximately more than half of them show positivity for CD68 anti-
ther positively stained T lymphocytes (brown, CD45RO antibody), or macrophages (red, CD68 antibody). Note the cell-to-cell interactions. Immunoperoxidase, X 1,000.
gen(macrophages). X 1,000.
214
MILE1 ET AL. CHRONIC CHAGASIC
Cmliovasc Path01 Vol. 5, No. 4 July/August 1996209-219
lNFIIXRATION
Figure 6. Imtnunostaining to show endothelial cells. (A) Capillaries and small vessels are clearly showed by the expression of CD31. Vessels are rnidly or moderately distorted because of the surrounding fibrosis. X 100. (B) A small, normal vessel surrounded by dense fibrous tissue. CD31 antigen, inununoperoxidase, X250.
with necrotic areas of myocardium. Mast cells were scarce or absent as well in specialized and in contracting myocardium. Lymphocytes, T and B lymphocytes. Almost all the lymphocytic population was due to T lymphocytes (see Table 2). Such infiltrates were more prominent in the working myocardium and in the specialized cells of the left branch of the His bundle than in the AV node and in the right Hisian branch (Figures 9 and 10). Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial fibers. Macrophages. Thirty-one percent of mononuclear cells were CD68 positive, corresponding to macrophages. Contacts between T lymphocytes and macrophages were frequently observed.
Discussion Findings In this study, the immunohistochemical features of chronic chagasic infltrates revealed by immunoenzymatic techniques was defined. Fibrosis. Lesions due to a healed myocarditis would explain the extensive and widespread distribution of fibrosis in this stage of CCC. Eosinophils. Although histologic studies of CCC have assessed inflammatory infiltrates (6,18-23), quantitative studies linking the amount of eosinophils with the myocardial lesions have been lacking. Only Kierszenbaum (23,24) has shown that the quantity of eosinophils correlates with the extent of myocardial damage. However, these inflam-
Figure 7. Severe myocarditis. Extensive mononuclear infiltrates, myocyte loss, and subendocardial fibrosis. Hematoxylin and eosin stain, X25.
Cardiovasc Pathol Vol. 5, No. 4 July/August 1996x209-219
CHRONIC CHAGASIC
MILE1 ETAL. INFILI’RA”ON
215
Figure 9. Serial section as in Figure 8. A detail of the left bundle of His is shown. Immunostaining for T lymphocytes. Positive cells express CD45RO antigen (brown); specialized myocardial cells have almost disappeared.Extensive mononuclear infiltrate, the majority of them being T lymphocytes. X250.
Figure 8. Bifurcating His bundle showing severe fibrosis at the left branch (between arrows). The right branch (asterisk) is intramyocardial and surrounded by connective tissue. Mallory trichome, X 25.
matory cells were always > UHPF of the infiltrate. In these reports (23,24) the percentage of eosinophils ranged from 5.1 to 9.2 in the cases of the most severe myocarditis, figures not much different from those found in our autopsied cases (10). Anyhow, it may be considered that eosinophilderived neuro-toxins and eosinophil peroxidases may contribute to tissue damage in CCC (24). Mast cells. Mast cells were reported to be more numerous in eosinophilic myocarditis in Africa (25). However, we did not find any mast cells or degranulating close to damaged myocytes. In a recent paper from our laboratory, these cells were neither observed close to neurons in intrinsic cardiac ganglia nor in the neighborhood of the conducting system and the contracting myocardium (10). These results suggest that mast cells do not play a role at least in the chronic state of the disease. Lymphocytes. Our finding that 26.5% to 39% of T lymphocytes associated with areas of myocardial damage is important because as it suggests a cytotoxic effect and/or helper effect (26-28). The higher percentages were observed in the autopsied
cases (34.4% and 39% in the specialized and working myocardium, respectively). The severity of the lesions found in the conducting system and the terminal stage of the chagasic myocardiopathy may be related to those figures, compared to the less severe cases submitted to endomyocardial biopsies (26.5%). Laguens (29) demonstrated that the infiltrates in murine chronic chagasic infection consisted of 50% of T-helper lymphocytes and 50% of macrophages. Recently, Higuchi et al. (11) characterized the infiltrating cells in human chronic myocarditis in endomyocardial biopsies and compared them with the myocardial rejection process. T cells formed 96.3% of the infiltrate, predominantly CDS+ (cytotoxickuppressor) T cells. Mean numbers of
Figure 10. Serial section as in Figure 8. Left bundle of His. Extensive fibrosis and mononuclear infiltrates. B lymphocytes are de-
picted by the immunoperoxidasetechnique. Positive cells express L-26 antigen (brown). Venula and dilated lymphatic channels are also shown. X 100.
216
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc
Path01 Vol. 5, No. 4
July/August 1996:209-219
LNF’KTRATION
Table2. Percentages of MononuclearCells,Eosinophils,Mast Cells,Lymphocytes,T andB Lymphocytes,Macrophages, and Percentage of Fibrosisin SpecializedandWorking Myocardiumsin Four Autopsiesof PatientsSufferingfrom ChronicChagasic Myocardiopathy Lymphocytes/T Lymphocytes/ Mononuclear Cells
Mononuclear Cells
CD68 Positive Macrophagesl Mononuclear Cells
Eosinophils” or Eosinophils/ Mononuclear Cells
Mast Cells”
B Lymphocytes/ Mononuclear Cells
Fibrosis
(%)
Specialized myocardium x 5 SD
51.6 i
18
1,421/2,108 (40.2%)
1,319/2,5113 (34.4%)
1,204/2,614 (31.5%)
0.48
0.01
21612,435 (10.18%)
Working myocardium x 2 SD
43.4 2 8b
1,562/2,004 (43.8%)
I,3 12/2,036 (39%)
1,109/2,416 (31.4%)
136/2,344 (5.5%)
0.14
142/2,041 (6.5%)
41.3 t 6’ “Results were expressed as the average number of cells per 100 high-power b Interventricular septum. ’ Free wall of the left ventricle distant from aneurysms.
CD8+ T cells were similar in both groups, and CD4+ T-cell counts, CD4+/CD8+ ratios, and CD4+ were significantly lower in the chagasic group compared to the myocardial rejection group. Based on experimental studies, the authors suggestedthat the parasiteor its antigenspresent in the chronic infiltrates may explain the relative paucity of helper T cells. In fact, 7: cruzi decreaseshelper T-cell surface markers (30) and inhibits their function induced by active suppressormacrophages,which is recovered by the presenceof interleukin-2 (3 1). Accordingly, Reis et al. (32) demonstratedthat the infiltrates in chronic chagasic cardiomyopathy were composedmainly of CD8+ T cells; killer cells and B lymphocytes were scarce,as well as somemacrophagesexpressingthe tumor necrosisfactor. Theseresults are rather different from thoseobtained in this paper and are discussedin the next paragraphs. Macrophages. Since the pioneering work of Metchnikoff, it becameevident that macrophagesplay an essential role as crucial mediators in the development of immune responseand inflammation t.33). On the other hand, our results also stressthe fact that about 50% to 60% of the infiltrating cells are mononuclear cells, namely, monocytes and macrophages.Although one must examine the possibility that macrophagesmay be recruited at the areasof necrosisto clear tissuedebris, the possibility that they might play a specific role as “facilitating macrophages”or “immunoregulatory monocytes” alsoshould be considered(7). Also, although the cellular immune responseis T-lymphocyte dependent (34), macrophagesappear as the most important effector cells in chronic chagasic infiltrates (35) where they form the bulk of the cellular response(17). In our study a certain percentage of morphologically macrophagecells were negative for the CD68 antigen both for peroxidase and alkaline phosphataseimmunostaining.
fields.
Although this may have been due to a technical problem, the antibody Kpl reacts against a specific epitope different from other antigens that are recognized by the different monoclonal antibodies. According to macrophageheterogeneity established by immunocytochemistry (36), only a wide range of macrophage-specificmonoclonal antibodies would recognize all the mononuclearphagocyte system. One may theorize that if there is a wide expression of membranereceptors of macrophages,these cells could also exhibit wide functional aspects. The various functions of macrophagesare exerted either in a cell-to-cell contact setupwith the target cells or by various biologically and pharmacologically active factors released by these cells. Among the most salient factors releasedby macrophagesare various monokinesand products of arachidonicacid (prostaglandinsand leukotrienes)(37). One of the consequencesof macrophage-mediatedcytotoxicity may be permeability alteration of the plasmamembranesof nearby target cells, possibly becauseof structural integrity changeof the membraneresulting from peroxidation of cell membranelipids (35). Macrophages are known to secrete large amounts of oxygen-free radicals. The secretion of their products such astumor necrosisfactor or interleukin-1 may enhancethe generationof oxygen-derived radicals(38). Macrophages present highly immunogenic antigens to the T-Lymphocyte system (39). Individual coexistence of, and contacts between, T lymphocytes and CD68-positive macrophageswere clearly demonstratedin this paper (see Figure 5). A clustering phenomenonalso was observed in other areas(seeFigure 2). Thesephenomenaare believed to be important for cell-to-cell interaction and subsequent T-cell activation and proliferation (40). Differences in r cruzi strainsand or cardiomyopathy severity may account for the different counts found in our paper from thoseof Higuchi (11) and Reis (32).
Cardiovasc July/August
Pathol Vol. 5, No. 4 1996:209-219
Figure 11. Combined theory. Factors involved in the production of myocardial damage in chronic Chagas’ cardiomyopathy. The major factors are in thick boxes. Reproduced from (4) with permission.
CHRONIC
Delayed
A Combined Theory for Chronic Chagas’ Cardiomyopathy It hasbeen saidthat the pathogenesisof chronic chagasic cardiac lesions remains an enigma (8). Different mechanismshave been proposed:autonomic involvement (22,41), lymphatics obstructed by fibrosis (42), microvascular disease(43,44), and autoimmune mechanisms(28,29). However no theory hasbeen definitively proved. Accordingly, Tafuri suggestedthat pathogenic mechanismsin chronic chagasiccardiomyopathy may be multifactorial (45). A “combined theory” also has been recently issued(4) basedon three ingredients: the parasite, macrophagesand lymphocytes, and the fibrosis. The parasite. It was generally accepted that 7: cruzi is scarcein the chronic phaseand that the inflammatory infiltrate was unrelated to the parasites(6). However, recent papers have shown that the parasites are present more frequently than it was thought (8). In fact, using dissimilar methods, different authors demonstratedeither the persistence of 7: cruzi or parasite antigensin mice (46), the parasite DNA sequenceamplified by the polymerasechain reaction (9), or T. cruzi antigens (8) from inflammatory lesions in human chagasiccardiomyopathy. The three studies undoubtedly showed the presenceof the parasiteor a part of it (B-46) or a portion of its genome (9) related to myocardial lesionsand chronic inflammatory infiltrates, suggestinga direct role for the parasitein the perpetuation of myocardial inflammation (8). In other words, the antigen stimulation would persistthroughout the chronic stage,even though the parasitesare not morphologically detectable by light microscopy (47). A delayed type of hypersensitivity reaction might also play an important role in this pathology (47). Teixeira et al. (48) have postulated the insertion of T. cvuzi DNA sequencesinto the host cell genome.Regardless,whatever the form of lesional parasite DNA, the chronic chagasiccardiomyopathy may be a consequenceof the continued persistence of ?: cruzi or its parts at the site of the lesion (9).
type
CHAGASIC
MILE1 ETAL. INFILTRATION
217
hypersensitivy
Macrophages and lymphocytes. As shown, mononuclear cells persistin the chronic stage,contributing to the inflammation through its products of secretion or through its own cytotoxicity (suppressorT cells) and cytolytic action (macrophages)(4). Apparently only autoimmunity may explain that all the processis restricted to the myocardium (49). Autoimmunity may be explained by the presenceof cross-reactivity antigens between myocardium and T cruzi or becausethe polyclonal activation might liberate antiautoclonesduring the acutephasethat may persista long time in the host (50). The fibrosis. According to Tafuri (45), the fibrosis causesthe progressivelossof the contracting capacity of the heart. Patients with chronic chagasic myocardiopathy exhibit heart failure in the late stageof the disease(1,4). Ultrastructural studieshave describeddegenerationof organelles, hypertrophy of myocytes, and increased fibrosis (51-52). Schaper(53) in chronic dilated cardiomyopathy showedimpairment of the myocardial ultrastructure and changesof the cytoskeleton. They concluded that the increaseof fibrosis, the degeneration of hypertrophied myocytes, and the alterations of the cytoskeleton (increaseof desmin, tubulin, and vinculin) are the morphologic correlatesof reducedmyocardial function in chronic heart disease.A similar pattern may be suspectedin chronic chagasicmyocardiopathy. Although much that is concerned with Chagas’ cardiomyopathy is controversial and much remains to be discovered, three factors may be the major ingredients for its pathogenesis:(1) the parasite: (2) lymphocytes and macrophages;and (3) the fibrosis. A multifactorial theory may then be suggestedto explain the sequenceof events that lead to the chronic stageof the disease(4; Figure 11). The authors wish to express their thanks to Miss Carina Elizabeth Kripelz for her skillfull editorial assistance. This investigation received partial financial support from the interuniversity cooperation between the Department of Pathology of the University of Milan, Italy, and the University of Buenos Aires, Argentina.
218
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Path01 Vol. 5, No. 4 July/August 1996:209-219
INFILTRATION
References
21. Romatia C. Enfermedad 1963;97--145.
1. Milei J, Storino R, Sanchez JA, Ferrans VJ. Does Chagas’ disease exist as an undiagnosed form of cardiomyopathy in the United States? Am Heart J 1992;123:1732-1735. 2. Editorial:
New Light on Chagas’
disease. Lancet
1965;i:1150.
3. UNDPiWorld Bank, WHO. Special Program for Research and Training in Tropical Disease. IV Programme Reports 1.7, Dec. 31, 1982, Geneva. 4. Storino Doyma,
R, Milei J. Enfermedad 1994,150.
de Chagas.
Buenos
Aires:
Mosby-
de Chagas.
Buenos
23. Molina HA, Kierszenbaum F. A study of human myocardial Chagas’ ‘disease: Distribution and frequency of inflammatory Int J Parasitol 1987;17:1297-1305.
6. Andrade Z. A patologia da doenca de Chagas Belge Med Trop 1985;65(suppl 1):15-30.
27. Acosta AM. 731256-258.
26. Kierszenbaum 72:201-211.
F. Autoimmunity Critical
comment
in Chagas’ (letter
28. Santos-Buch CA, Acosta AM. Pathology zard I, ed. Immunology and Pathogenesis Raton, FL; CRC Press, 1985;145-183.
8. Higuchi ML, Brito T, Martins Reis M, et al. Correlation between Trypanosoma cnrzi parasitism and myocardial inflammatory infiltrate in human chronic chagasic myocarditis: light microscopy and immunohistochemical findings. Cardiovasc Path01 1993;2:101-106.
29. Laguens RP. Immunologic mechanisms mental Chagas’ disease. Rev Fed Argent
10. Milei J, Storino RA, Beigelman R, Fem&ndez Alonso G, Mattuni L, Rossi L. Histopathology of specialized and ordinary myocardium and nerves in chronic Chagas’ disease, with a morphometric study of inflammation and fibrosis. Cardiologia 1991;36: 107-l 15. 11. Higuchi ML, Gutierrez PS, Aiello VD, Palomino S, Bocchi E, Pileggi F. Human chagasic cardiopathy and myocardial rejection: a similar pattern of myocarditis. Virchows Arch A Path01 Anat Histopathol 1993;423:157-160. 12. Consejo Argentino de Enfermedad de Chagas. “Dr. Salvador Editorial. Clasificacion de la enfermedad de Chagas cr6nica. Cardiol 1988;56:108-109. 13. Rossi L. Histopathology Ambrosiana, 1978;60.
of cardiac arrhythmias.
Milan:
Mazza.” Rev Arg
Casa Editrice
14. Guesdon JL, Temynck T, Anameas S. The use of avidin-biotin action in immunoenzymatic techniques. J Histochem Cytochem 27:1131-1134.
inter1979;
15. Edwards WD, Holmes DR, Reeder GS. Diagnosis of active lymphocytic myocarditis by endomyocardial biopsy. Quantitative criteria for light microscopy. Mayo Clin Proc 1982;57:419425. 16. Steenbergen C, Kolbeck PC, Wolfe JA, Anthony RY, Sanfilippo FP, Jennings RB. Detection of lymphocytes in endomyocardium using immunohistochemical techniques. Relevance to evaluation of endomyocardial biopsies in suspected cases of lymphocytic myocarditis. J Appl Cardiol 1986;1:63-73.
of
tissue in cell type.
mast cells? Hum Path01 disease. J Parasitol
to the editor).
7. Milei J, Storino R, Femandez Alonso Cl, Beigelman R, Vanzulli S, Ferrans VJ. Endomyocardial biopsies in chronic chagasic cardiomyopathy. Cardiology 1992;80:424-437.
9. Jones EM, Colley DG, Tostes S, Reis Lopez E, Vnencak-Jones CL, McCurley TL. Amplification of a Trypanosoma cruzi DNA sequence from inflammatory lesions in human chagasic cardiomyopathy. Am J Trop Med Hyg 1993;48:348-357.
Libreros,
24. Molina HA, Kierszenbaum F. Immunohistochemical detection of deposits of eosinophil-derived neurotoxin and eosinophil peroxidase in the myocardium of patients with Chagas’ disease. Immunology 1988; 64:725”3 1. 25. Esterisen RD. What is the role of myocardial 1985;16:536-538.
Ann Sot
L6pez
22. Koberle !F. Chagas’ disease and Chagas’ syndromes. The pathology American trypanosomiasis. Adv Parasitol 1968;6:63-116.
5. Sanchez IA, Milei J, Yu ZX, Storino R, Wenthold R, Ferrans VJ. Immunohistochemical localization of laminin in the hearts of patients with chronic chagasic cardiomyopathy: relationship to thickening of basement membranes. Am Heart J 1993;126:1392-1401. no homen.
Aires:
1986;
J Parasitol
1987;
of Chagas’ disease. of Trypanosomiasis.
In: TiBoca
of cardiac damage in experiCardiol 1988;17:166168.
30. Sztein hl, Washington RC, Kierszenbaum F. Trypanosomu cruzi inhibits the expression of CD3, CD4, CD8, and IL-2R by mitogen-activated helper and cytotoxic human lymphocytes. J Immunol 1990; 144:3558-3562. 31. Reed SG, Inverso JA, Roters SB. Heterologous antibody responses in mice with chronic Trypanosoma cruzi infection: depressed T-helper function restored with supematants containing interleukin-2. J Immuno1 1984;133:1558-1563. 32. Reis DD, Jones EM, Tostes S Jr, et al. Characterization of inflammatory infiltrates in chronic chagasic myocardial lesions: presence of tumor necrosis factor-alpha + cells and dominance of Granzyme A+, CD8+ lymphocytes. Am J Trop Med Hyg 1993;48:637-642. 33. Adams DO, Hamilton TA. Phagocyte cell: cytotoxic activities of macrophages. In: Gallin JI, Goldstein IM, Snyderman R, eds. Inflammation: Basic Principles and Clinical Correlates. New York Raven Press, 1988;471-492. 34. Hontebcyrie-Joskowicz M, Said G, Milon G, Marchal G, Eisen H. L3T4+ T cells able to mediated parasite-specific delayed-type hypersensitivity play a role in the pathology of experimental Chagas’ disease. Em J Immunol 1987;17:1027-1033. 35. Rossi MA. Myocardial role for macrophages.
damage in Trypanosoma cruzi myocarditis: Can J Cardiol 1990;6:293-298.
36. Dijkstra CD, Damoiseaux JGMC. Macrophage heterogeneity lished by immunocytochemistry. In: Progress in Histochemistry Cytochemistry, vol. 27/2. New York: Gustav Fisher, 1993.
A
estaband
37. Ben-Efrain S. Interactions between macrophage cytokines and ecosanoids in expression of antitnmor activity. Mediators of Inflammation 1992;1:295-308. 38. Ward PA, Johnson KJ, Warren JS, Kunkel RG. Immune complexes, oxygen radicals, and lung injury. In: Halliwell B, ed. Oxygen Radicals and Tissue Injury. Bethesda, MD: Federation of American Societies of Experimental Biology, 1988;107-114.
17. Milei J, Bortman G, Femandez Alonso G, Grancelli H, Beigelman R. Immunohistochemical staining of lymphocytes for the reliable diagnosis of myocarditis in endomyocardial biopsies. Cardiology 1990; 77~77-85.
39. Spencer SC, Fabre JW: Characterization of the tissue macrophage and the interstitial dendritic cell as distinct leukocytes normally resident in the connective tissue of rat heart. J Exp Med 1990;171:1841-1851.
18. Andrade ZA, Andrade SG. Chagas’ disease (American trypanosomiasis). In: Martial-Rojas RA, ed. Pathology of Protozoal and Hehnintic Diseases. Baltimore: Williams &Wilkins, 1971; chap 2.
40. Zhang J, Yu ZX, Shinsuke F, Yamaguchi ML, Ferrans V. Interstitial dendritic cells of the rat heart. Quantitative and ultrastructural changes in experimental myocardial infarction. Circulation 1993;87:909-920.
19. Andrade ZA, Andrade SG, Oliveira GB, Alonso DR. Histopathology of the conducting tissue of the heart in Chagas’ myocarditis. Am Heart J 1978;95:316-323.
41. Oliveira JSM. A natural model of intrinsic heart nervous system denervation: Chagas’ Cardiopathy. Am Heart J 1985:110:1092-1098.
20. Mazza S, J&g E, Canal Feijoo EF. Primer case cronico mortal de forma cardfaca de enfermedad de Chagas demostrado en Santiago de1 Estero. MEPRA 1938;38:1-17.
42. Rossi L, Storino R, Milei J, Matturri L. Deplecidn neuronal en la enfermedad de Chagas: todo deberfa reverse. Rev Arg Cardiol 1994, 62:239-246. 43. Morris SA, Tanowitz HB, Wittner M, Bilezikian JP. Patbophysiologi-
Cardmvasc July/August
Pathol Vol. 5, No. 4 1996:209-219
cal insights into the cardiomyopathy 1990;82:1900-1909.
CHRONIC
of Chagas’
44. Rossi MA, Mengel JO. Patogenese da miocardite papel de factores autoimunes e microvasculares. Sao Paulo, 1992;34:593-596.
disease.
Immunol
219
49. Andrade ZA. 126 (abstract).
chagasica cronica: o Rev Inst Med Trop
50. Minoprio PM, Eisen H, Fomi L, D’lmperio-Lima MR, Joskowicz MS, Coutinho A. Polyclonal lymphocyte responses to murine Ttypanosonn cruzi infection. Quantitation of both T- and B-cell responses. Stand J Immunol 1986;24:661-666.
46. Younees-Chennouti AB, Hontebeyre-Joskowicz M, Tricottet V, Eisen H, Reynes M. Said G. Persistence of Ttypanosoma cruzi antigens in the inflammatory lesions of chronically infected mice. Trans Roy Sot Trop Med Hyg 1988;82:77.
of Chagas’
MILE1 ET AL. INFILTRATION
disease. Circulation
45. Tafuri WL. Patogenese. VIII Reunao Anual de Pesquisa Aplicada em Doenca de Chagas. Rev Sot Bras Med Trop 1992;25(suppl III): 19 (abstract).
Pathogenesis
CHAGASIC
1991; 142:
51. Ferrans VJ, Milei J, Tomita Y, Storino RA. Basement membrane thickening in cardiac myocytes and capillaries in chronic Chagas’ disease. Am J Cardiol 1988;6 1: 1 137-I 140.
47. Andrade Z. Novos aspectos sobre a patogenia da doenca de Chagas. Rev Sot Bras Med Trop 1992;25(suppl III):8 (abstract).
52. Carrasco Guerra HA, Palacios-Pm E, Dagert de Scorza C, Molina C, Inglessis G, Mendoza RV. Clinical, histochemical, and ultrastructural correlation in septal endomyocardial biopsies from chronic chagasic patients: detection of early myocardial damage. Am Heart J 1987; 113:716-724.
48. Teixeira ARL, Lacava Z, Santana JM, Luna H. Insercao de DNA de Ttypanosoma cruzi no genoma de celula hospedeira de mamifero por meio de infeccao. Rev Sot Bras Med Trop 1991;24:55-61.
53. Schaper J, Froede R, Hein ST, et al. Impairment of the myocardial ultrastructure and changes of the cytoskeleton in dilated cardiomyopathy. Circulation. 199 I ;83:504.
SYMPOSIUM
Myocardial Inflammatory Infiltrate in Human Chronic Chagasic Cardiomyopathy: Immunohistochemical Findings Jo& Milei, MD,* Graciela Fernhdez Alonso, MD,+ Silvia Vanzulli, MD,+ Rub& Storino, MD,*’ Luigi Matturri, MD,” and Lino Rossi, MD” From *Cardiopsis,
the tNational Academy of Medicine, and the $INCALP Foundation, Buenos Aires, Argentina; and the $Department of Pathology, University of Milan, Milan, Italy
Chagas’ disease is the most common form of chronic myocarditis in the world. It is characterized by a progressive chronic myocarditis that leads to cardiomegaly, arrhythmias, cardiac failure, and thromboembolic phenomena. This communication reports studies on the immunohistochemistry of chronic infiltrates in 30 endomyocardial biopsies and in contracting and specialized myocardium of autopsies of four patients suffering from Chagas’ cardiomyopathy. Expression of the following antigens was studied: common leucocyte antigen (CLA-CDQSR), L-26(CD20), CD68, kappa and lambda light chains and T-UCLH-1 (CD45RO), and MB-l. Streptavidin-peroxidase and streptavidin-alkaline phosphatase with biotinylated anti-mouse IgG were used as detection systems. Double immunostaining for the simultaneous demonstration of T lymphocytes (CD45RO) and macrophages was performed using both immunoenzymatic techniques consecutively. Expression of CD31 was detected for the demonstration of endothelial cells. In endomyocardial biopsies, tissue forms of trypanosomes were not found. The percentage of fibrous tissue was 24.1% ? 12.8% (range 8.2%49%). Eosinophils were scarce (l/high-power field), but associated with necrotic areas of the myocardium. Mast cells were scarce or absent. They were always situated in fibrotic areas. The most remarkable finding was the presence of infiltrates consisting of macrophages and CLA-positive mononuclear cells. Twenty-six and one-half percent of them were T lymphocytes, and 10.5% were B lymphocytes. Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial lesions. Thirty percent of the infiltrate was composed of macrophages (positive CD68 cells). The remaining infiltrate was composed of mononuclear cells resembling macrophages and CLA-negative mononuclear cells. Contacts between CD68-positive cells and T lymphocytes were frequently found. CD31 antibodies clearly pointed out normal endothelial cells, in either normal or damaged vessels. No isolated cells positive for these antibodies were found within the mononuclear infiltrate. In autopsied hearts, myocardial lesions consisted of a chronic inflammatory process with fibrotic scars and extensive mononuclear infiltrates. No amastigote nests were found. A statistically significant difference @ < 0.05) was obtained when the percentage of fibrosis was compared in the specialized and contracting myocardiums (5 1.6% -t 18% vs. 43.4 % -C 8%). Eosinophils were scarce in infiltrates, reaching 5%, and they were associated with necrotic myocardium. Mast cells also were scarce or absent in specialized and in contracting myocardium. Almost all the lymphocytic population was T lymphocytes. Such infiltrates were more prominent in the working myocardium (39%) and in the specialized cells of the left branch of the His bundle than in the atrioventricular node and in the right Hisian branch (34.4%). The 31% of mononuclear cells were CD68 positive, thus corresponding to macrophages. Contacts among T lymphocytes and macrophages were frequently observed. Although much that is concerned with Chagas’ cardiomyopathy is controversial, these may be the major ingredients for its pathogenesis: the parasite or a part of it, lymphocytes and macrophages, and fibrosis. Then a multifactorial or “combined theory” may be suggested to explain the sequence of events that lead to the chronic stage of the disease. Cardiovasc Path01 1996;5:209-219 Manuscript received December 15, 1995; accepted January 10, 1996. Address for reprints: Dr. Jose Milei, CARDIOPSIS, Tucuman 2163, 4B, 1050 Buenos Aires, Argentina; fax: 00541-951-0366. Cardiovascular Pathology Vol. 5, No. 4, July/August 0 1996 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
Presented at an international held in Milan, Italy, June 26-28,
symposium 1995.
on Chagas’ heart disease
1996:209-219 1054-8807/96/$15.00 PI1 SlO54-8807(96)00006-3
210
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Pathol Vol. 5, No. 4 July/August 1996~209-219
INFKTRATION
Chagas’ disease (American trypanosomiasis) is the most common form of chronic myocarditis in the world (1,2). It is estimated that at least 20 million individuals are infected with Trypan~~rna crxzi and that 90 million live in endemic areaswhere the risk of this infection exists (3). The progressive chronic cardiac form appearsin 10% to 20% of previously asymptomatic patients, 15 to 20 years after the infection (4). It is found in adults 30 to 50 years of age and is manifestedby atrioventricular (AV) block and/or right bundle branch block plus left anterior hemiblock and ventricular ectopic beats (4). It is characterized by a progressive chronic myocarditis that leads to cardiomegaly, arrhythmias, cardiac failure, and thromboembolic phenomena(5). Despite the importance of the disease,little is known of its pathogenesis(6). At the chronic stage, there is persistence of scarceparasitesin the myocardium and low-level parasitemias.Mononuclear infiltrates and extensive areasof fibrosis with apical ventricular aneurysmsare the usual lesions (7). These findings support the hypothesis that the chronic chagasic cardiomyopathy (CCC) is mediated, at least in part, by immune phenomena. Recently, identification of 7: cruzi parasites in hearts with CCC was facilitated with the use of a polyclonal antiZ cruzi antibody serum.There was a significant correlation between the presenceof the parasite antigensand moderate or severeinflammatory infiltrates (8). This observation was confirmed when an amplification of a 7: cruzi DNA sequence from inflammatory lesions was shown in human CCC (9). These studiesdemonstratedthat r cruzi or a portion of its genomeis presentin the inflammatory portion of infiltrates and supports a direct role for the parasite in the perpetuation of the inflammation (8). However, the chronic chagasicinfiltration hasrarely been characterized by means of immunohistochemicalmethods (7,10,11). This communication reports studies on the immunohistochemistry of chronic infiltrates in endomyocardial biopsies and in contracting and specialized myocardium of autopsies of patients suffering from CCC.
Material and Methods The study was conducted on 30 endomyocardial biopsies from an equal number of patients (7) and on four heartsbelonging to autopsiedpersonswho suffered from CCC (10). There were malesand females. Ages ranged from 20 to 69 years (mean, 45.1 years). Both the clinical and the pathologic features establishedthe diagnosisof Chagas’disease. The clinical and pathologic characteristics of each case were already reported (7,lO). Chagas’ serology was studied for four standardreactions against T cruzi (immunofluorescence, complement fixation, passivehemagglutination, and latex test). Patients were divided into 3 groups, according to the Chagas’ DiseaseCouncil of the Argentine Society of Cardiology (12): Group I: asymptomatic patients, normal thorax X ray and ECG; Group II: conduction disturbances
and/or arrhythmia; and Group III: cardiomegaly with heart failure and.abnormal ECG.
Procedure Endomyocardial biopsies were performed under fluoroscopic and bidimensional echocardiographic guidance by percutaneous puncture of the right internal jugular vein; Four to five specimenswere immediately placed in 10% buffered formalin. Autopsies were performed within 4 hours of death, and tissue sampleswere also immediately fixed in 10% buffered formalin (pH 7.0).
Histological
Methods
The biopsy samplesmeasured2 to 3 mm in the greatest dimension. The specimenswere routinely processed for light microscopy by embeddingin paraffin. Sections (4 pm in thickness) were serially cut and mounted so that each slide contained three to four sections. These serial slides were processed consecutively for immunohistochemical methodsand stainedwith hematoxylin and eosin, BarbeitoLopez trichrome, the PAS method, and the silver-methenamine technique. For the morphometric assessmentof fibrosis, the Mallory trichrome stain was used, and each tissue section was examined under a grid. The Alcian blue technique (pH 2.5) was used to detect and quantify mast cells in 100 high-power fields (HPF; X 400). Giemsa stain was also used to quantify eosinophils. All the slides were examined by two observers who had no knowledge of the clinical findings. In the autopsiedcases,histologic examination of the conduction systemon serial sectionswas carried out following the technique usually employed by one of the authors (LR) (13). Four to eight sectionsof each case,from the right and left ventricular walls were also examined. The hematoxylin and eosin, Alcian blue (pH 2.5), Giemsa, and silver-methenamine methods were routinely followed for all sections; Bielschowsky-silver impregnation was usedfor the nervous system. Each section stained with the Mallory trichrome was examined under a grid to establishthe percentageof fibrosisin the specializedand ordinary myocardium. As far as the intrinsic cardiac nervous system is concerned, observationswere carried out on the stageand number of neuronal bodies inside individual ganglia of the AV groove plexus, the ganglia of the sinoatrial (SA) nerve plexus, the intercaval funnel adventitia, and the posterior wall of the left atria.
Immunohistochemical
Methods
For cellular immunophenotyping of the mononuclear infiltrate, nnmunoenzymatic techniqueswith monoclonal primary antibodieswere performed. Serial sectionsof endomyocardial biopsies from 14 se-
Cardiovasc July/August
Path01 Vol. 5, NO. 4 1996:X9-219
CHRONIC
lected cases and random sections from the anterior, medial, and posterior areas of the bundle of His and its branches, as well as several sections from the right and left free ventricular walls from the autopsied cases were processed. Expression of the following antigens was studied: common leukocyte antigen (CLA-CD45R), L-26(CD20), CD68, kappa and lambda light chains (Dakopatts, Santa Barbara, CA), and T-UCLH-1 (CD45RO) and MB-l (Biogenex). Streptavidin-peroxidase and streptavidin-alkaline phosphatase with biotynilated anti-mouse IgG were used as detection systems and reactions were developed with 3.3’ diaminobenzidine (DAB) and with fast red. Double immunostaining for the simultaneous demonstration of T lymphocytes (CD45RO) and macrophages was performed using both immunenzymatic techniques consecutively. Peroxidase and DAB was used for T lymphocytes and alkaline phosphatase and fast red for macrophages (14). Expression of CD31 and Ulex Europaeus I was detected to demonstrate endothelial cells in order to study both the relationship between infiltrates and vessels and endothelial integrity. Positive cells for the different antigens studied were quantified in both contractile myocardium and specialized myocardium. The number of negative and positive mononuclear cells for each marker was counted on three HPF/slide. These numbers were correlated with the extent of interstitial fibrosis and/or myocyte damage on each sample. According to previous studies (15-17), we considered the normal mean number of lymphocytes to be less than 5/ HPF. Then, S/HPF was considered as the lower limit of myocarditis if it was associated with myocyte injury (16). Specifically for grading purposes and independently of previous grading using hematoxylin and eosin-stained preparations, any patient having a mean of 5 CLA-positive cells/HPF would be classified as having 1+ infiltration of the heart. A mean of 6 to 10 mononuclear cells/HPF was graded as 2-t) and a mean greater than 10 lymphocytes/HPF was graded as 3+ (17).
Results Endomyocardial
Biopsies
Clinical data and histological features of these patients were previously reported (7). Patients belonged to groups II or III (12). In brief, the 30 biopsy specimens showed mild to moderate hypertrophy of the myocytes and variable degrees of interstitial fibrosis (mild in 15 patients, moderate in 11, and severe in 4). Lesions consisted of a chronic inflammatory process with fibrotic areas and mononuclear cellular infiltrates. The lesions showed a microfocal disposition. Coagulative necrosis and/or myocytolysis were frequently observed. A marked subendocardial sclerosis, presumably due to an increased end diastolic pressure, was also observed.
CHAGASIC
MILE1 ET AL. INFILTRATION
211
Figure 1. A granulating mast cell surrounded by dense connective tissue. Alcian blue, pH 2.5, X400.
Tissue forms of trypanosomes were not found. Diffuse or focal thickening of the basement membranes was observed after staining with the silver-methenamine technique. Mononuclear cell infiltrates were found in 25 biopsies. The endocardium ranged from 4.8 to 43.6 pm in thickness, with a mean (2 SD) of 24 + 12.6 pm, and the diameter of the myocytes ranged from 11.4 to 40 km, with a mean of 20 & 7.33 pm. None of these data could be correlated with the ejection fraction in each patient. Fibrosis. The percentage of fibrous tissue was 24.1% + 12.8% (range 8.2%-49%). Only 1 of the 30 biopsies contained less than 10% fibrous tissue, and 20 of the 30 biopsies contained more than 15% fibrous tissue. Eosinophils and mast cells. Eosinophils were scarce (l/I-IPF), but associated with necrotic areas of the myocardium. Mast cells were scarce or absent. They were always situated in fibrotic areas (Figure 1). Lymphocytes, T and B lymphocytes. The most remarkable finding was the presence of infiltrates consisting of macrophages and CLA-positive mononuclear cells (true lymphocytes; Figure 2). Although a high mean of apparently mononuclear cells may be found in a microscopic HPF, only 28.5% of them may result in true lymphocytes (Table 1). Immunostaining for K and X chains and L-26 antigen or MB1 showed that only the 10.5% of the CLA-positive lymphocytes belonged to the B subset while the normal (2.5/ l)K/X ratio was maintained. Conversely, almost all the lymphocytic population was due to T lymphocytes (26.5% of the total number of mononuclear cells; Table 1; Figure 2). Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial lesions (Figure 3). From the 14 samples previously diagnosed as CCC, 12 showed 5 or more CLA-positive mononuclear cells/I-IPF. From these results, 12 of the 14 cases resulted in “active chronic chagasic myocarditis.” The remaining two samples were classified as being “chronic chagasic cardiomyopathies.”
212
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Path01 Vol. 5, No. 4 July/August 1996~209-219
EWILTRATION
mal or damagedvessels.No isolated cells positive for these antibodies were found within the mononuclear infiltrate (Figure 6).
Autopsied Hearts Grossly, the four heartspresentedsevere dilatation, flattening of the papillary muscles,and a marked subendocardial sclerosisrelated to an increasedend diastolic pressure. A remarkable thinning of the left ventricular wall was characteristic at the apical areas.Intracavitary tbrombi were also found. Microscopically, myocardial lesions consisted of a chronic inflammatory processwith fibrotic scarsand extenFigure 2. Clusterof mononuclearcells. Doubleimmunostaining sive mononuclearinfiltrates (Figure 7). No amastigotenests for the simultaneous demonstration of T lymphocytes(CD45RO) were found. Extensive myocytolysis and spotty contraction and macrophages (CD68). PeroxidaseandDAB (T lymphocytes, band necrosiswere observed. Cell hypertrophy in the apparbrown)andalcalinephosphatase and fast red (macrophages, negaently preserved myocytes was revealed by hypertrophic bitive in this section), X 1,000. zarre nuclei. Overall results in the 12 caseswith active chronic chaAs previously described(4-lo), dilated lymphatic changasic myocarditis showed that T and B lymphocytes reprenels widespread in the ventricular septum and in the AV sented27.3% and 10.8% of the total mononuclearinfiltrates. node, His bundle, and in the root of the right and left bunMacrophages. The 30% of the infiltrate was composed dles branch were observed. Around the apical aneurysm of of macrophages(positive CD68 cells; Figure 4). The rethe left ventricle, such dilated lymphatics were distributed maining infiltrate was composedof mononuclear cells resubepicardially. sembling macrophages and CLA-negative mononuclear The serial sectioning in the conducting system showed cells (Table 1, Figures 2 and 3). Of note, contacts between mild to moderatediffuse fibrosis of the AV node and of the CD68-positive cells and T lymphocytes were frequently “penetrating” and “branching” portions of the His bundle, found (Figure 5). complete destruction of the proximal segmentsof the right Endothelial cells. CD3 1 and Ulex europaeusantibodies and left bundlesbranch by varying degreesof replacement clearly pointed out normal endothelial cells, either in norby densecollagen tissue(Figure 8).
Table 1. Monoclonal
Antibody Marker Quantitative Analysis of Interstitial Infiltrates in 14 Patients with Chronic Chagasic
Cardiomyopathy
Patient No. 1
Total Lymphocytes/ Total Mononuclear Cells” (X)
Total T Cells/ Mononuclear Cell?
36170 44/80 19/40
(>5)
(>5) (>5) (b-5) (>5)
29165 29154 19147
(~5)
(<5) (>5)
6141 33/66
12
(>5) (>5) (>5)
57/129 46192 28/89
13 14
(>5) (<5)
20/55 3/20
2 3 4 5 6 7 8 9
10 11
Total The numbers
(>5) p-5)
(X) (4)
--
386/969
(>5)
10/37 23/63 26162 32162
p-5) (<5) (>5)
15/45 2132 36/100
(>5) (<5)
68/1&I 6126
(>5) (>5)
18/64 17/51
(<5)
l/12
(X)
17/121
26/58 20/56
(28.5%)
of the cells are given. Mast cells and eosinophils
were always
‘Represents
the sum of the number
of specific
cells counted
in HPF.
(<5)
42170 20139 25168
(4) (4)
5/43 2/69 9/28
(<5) (<5) ((5) (<5)
7149 8/34 4/24 6/31
(<5) (<5)
2/38 o/54
(<5) (<5) (<5)
5176 3/81 o/79
(<5) (<5)
7155 5/49
18/46 19162 4134 36176 611125 47170 27182
18154 4124 369/867
Total B Cells/ Mononuclear Cellsb
22l64
26153
300/832(26.5%)
“(>5) or (<5) lymphocytes, T or B lymphocytes/HPF.
Total CD68 Positive Macrophage Cells/Mononuclear Cell?
and were therefore
(29.8%)
not tabulated.
83/710(10.5%)
Cardiovasc July/August
Path01 Vol. 5, No. 4 1996:20!-219
CHRONIC
CHAGASIC
MILEIETAL. INFILTRATION
213
Figure 3. Doubleinnnunostaining for the simultaneous demonstration of T lymphocytes(CD45RO) andmacrophages (CD68).(A) T lymphocytes(brown)in closecontactwith a macrophage (pinkcytoplasm).X 1000.(B) Two closemononuclear cellsin theneighborhood of a damagedmyocyte.The T lymphocyte (brown) has a close contact with the sarcolemma. The macrophage, negative for CD68 antibody (large nucleus and pink cytoplasm) is also very close to a round cell, presumably a B lymphocyte. X 1,000.
The remaining specialized fibers presentedatrophy and mild fatty infiltration and were surroundedin most casesby infiltrates consisting mainly of lymphocytes and macrophages(seelater) (Figures 9 and 10). Venular and lymphatic vascular dilatations were frequently observedin thoseareas. No selective involvement of the right half of the His bundle was observed, as previously reported (10). The subendocardial Purkinje fibers were usually damagedby chronic inflammation and fibrosis, aswas describedfor contracting myocytes. Diffuse or focal thickening of basal membranes of contracting and specialized myocytes were frequently observed asrevealed by the silver-methenaminemethod. In the neurovegetative ganglia, no actual neuronal loss has been found; only early increasein satellite cells and/or mild mononuclearinfiltration was rarely observed.
Spotty round cell epiperineuritis was detected in one case in the subepicardiumof the wall of a left aneurysm. Morphometric data are summarizedin Table 2. Fibrosis. Statistical significance @ < .05) was obtained when the percentage of fibrosis was compared in the specialized and contracting myocardium (51.6% Ifr 18% vs. 43.4% ? 8%). It suggestedthat the fibrotic lesionsof CCC were randomly distributed through the conducting system and the ordinary myocardium. A similar degree of fibrosis was observed in the sectionsbelonging to the free wall of the left ventricle (41.3% & 6%). Eosinophils and mast cells. Eosinophils were relatively scarcein infiltrates, reaching 5% and frequently associated
Figure 4. Extensive mononuclearinfiltrates. Roundcells are in
Figure 5. Area of myocardiuminfiltrated by smallclustersof ei-
contact with damaged myocytes of forming extensive clusters.Approximately more than half of them show positivity for CD68 anti-
ther positively stained T lymphocytes (brown, CD45RO antibody), or macrophages (red, CD68 antibody). Note the cell-to-cell interactions. Immunoperoxidase, X 1,000.
gen(macrophages). X 1,000.
214
MILE1 ET AL. CHRONIC CHAGASIC
Cmliovasc Path01 Vol. 5, No. 4 July/August 1996209-219
lNFIIXRATION
Figure 6. Imtnunostaining to show endothelial cells. (A) Capillaries and small vessels are clearly showed by the expression of CD31. Vessels are rnidly or moderately distorted because of the surrounding fibrosis. X 100. (B) A small, normal vessel surrounded by dense fibrous tissue. CD31 antigen, inununoperoxidase, X250.
with necrotic areas of myocardium. Mast cells were scarce or absent as well in specialized and in contracting myocardium. Lymphocytes, T and B lymphocytes. Almost all the lymphocytic population was due to T lymphocytes (see Table 2). Such infiltrates were more prominent in the working myocardium and in the specialized cells of the left branch of the His bundle than in the AV node and in the right Hisian branch (Figures 9 and 10). Lymphocytic infiltration was particularly associated with necrotic and degenerative myocardial fibers. Macrophages. Thirty-one percent of mononuclear cells were CD68 positive, corresponding to macrophages. Contacts between T lymphocytes and macrophages were frequently observed.
Discussion Findings In this study, the immunohistochemical features of chronic chagasic infltrates revealed by immunoenzymatic techniques was defined. Fibrosis. Lesions due to a healed myocarditis would explain the extensive and widespread distribution of fibrosis in this stage of CCC. Eosinophils. Although histologic studies of CCC have assessed inflammatory infiltrates (6,18-23), quantitative studies linking the amount of eosinophils with the myocardial lesions have been lacking. Only Kierszenbaum (23,24) has shown that the quantity of eosinophils correlates with the extent of myocardial damage. However, these inflam-
Figure 7. Severe myocarditis. Extensive mononuclear infiltrates, myocyte loss, and subendocardial fibrosis. Hematoxylin and eosin stain, X25.
Cardiovasc Pathol Vol. 5, No. 4 July/August 1996x209-219
CHRONIC CHAGASIC
MILE1 ETAL. INFILI’RA”ON
215
Figure 9. Serial section as in Figure 8. A detail of the left bundle of His is shown. Immunostaining for T lymphocytes. Positive cells express CD45RO antigen (brown); specialized myocardial cells have almost disappeared.Extensive mononuclear infiltrate, the majority of them being T lymphocytes. X250.
Figure 8. Bifurcating His bundle showing severe fibrosis at the left branch (between arrows). The right branch (asterisk) is intramyocardial and surrounded by connective tissue. Mallory trichome, X 25.
matory cells were always > UHPF of the infiltrate. In these reports (23,24) the percentage of eosinophils ranged from 5.1 to 9.2 in the cases of the most severe myocarditis, figures not much different from those found in our autopsied cases (10). Anyhow, it may be considered that eosinophilderived neuro-toxins and eosinophil peroxidases may contribute to tissue damage in CCC (24). Mast cells. Mast cells were reported to be more numerous in eosinophilic myocarditis in Africa (25). However, we did not find any mast cells or degranulating close to damaged myocytes. In a recent paper from our laboratory, these cells were neither observed close to neurons in intrinsic cardiac ganglia nor in the neighborhood of the conducting system and the contracting myocardium (10). These results suggest that mast cells do not play a role at least in the chronic state of the disease. Lymphocytes. Our finding that 26.5% to 39% of T lymphocytes associated with areas of myocardial damage is important because as it suggests a cytotoxic effect and/or helper effect (26-28). The higher percentages were observed in the autopsied
cases (34.4% and 39% in the specialized and working myocardium, respectively). The severity of the lesions found in the conducting system and the terminal stage of the chagasic myocardiopathy may be related to those figures, compared to the less severe cases submitted to endomyocardial biopsies (26.5%). Laguens (29) demonstrated that the infiltrates in murine chronic chagasic infection consisted of 50% of T-helper lymphocytes and 50% of macrophages. Recently, Higuchi et al. (11) characterized the infiltrating cells in human chronic myocarditis in endomyocardial biopsies and compared them with the myocardial rejection process. T cells formed 96.3% of the infiltrate, predominantly CDS+ (cytotoxickuppressor) T cells. Mean numbers of
Figure 10. Serial section as in Figure 8. Left bundle of His. Extensive fibrosis and mononuclear infiltrates. B lymphocytes are de-
picted by the immunoperoxidasetechnique. Positive cells express L-26 antigen (brown). Venula and dilated lymphatic channels are also shown. X 100.
216
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc
Path01 Vol. 5, No. 4
July/August 1996:209-219
LNF’KTRATION
Table2. Percentages of MononuclearCells,Eosinophils,Mast Cells,Lymphocytes,T andB Lymphocytes,Macrophages, and Percentage of Fibrosisin SpecializedandWorking Myocardiumsin Four Autopsiesof PatientsSufferingfrom ChronicChagasic Myocardiopathy Lymphocytes/T Lymphocytes/ Mononuclear Cells
Mononuclear Cells
CD68 Positive Macrophagesl Mononuclear Cells
Eosinophils” or Eosinophils/ Mononuclear Cells
Mast Cells”
B Lymphocytes/ Mononuclear Cells
Fibrosis
(%)
Specialized myocardium x 5 SD
51.6 i
18
1,421/2,108 (40.2%)
1,319/2,5113 (34.4%)
1,204/2,614 (31.5%)
0.48
0.01
21612,435 (10.18%)
Working myocardium x 2 SD
43.4 2 8b
1,562/2,004 (43.8%)
I,3 12/2,036 (39%)
1,109/2,416 (31.4%)
136/2,344 (5.5%)
0.14
142/2,041 (6.5%)
41.3 t 6’ “Results were expressed as the average number of cells per 100 high-power b Interventricular septum. ’ Free wall of the left ventricle distant from aneurysms.
CD8+ T cells were similar in both groups, and CD4+ T-cell counts, CD4+/CD8+ ratios, and CD4+ were significantly lower in the chagasic group compared to the myocardial rejection group. Based on experimental studies, the authors suggestedthat the parasiteor its antigenspresent in the chronic infiltrates may explain the relative paucity of helper T cells. In fact, 7: cruzi decreaseshelper T-cell surface markers (30) and inhibits their function induced by active suppressormacrophages,which is recovered by the presenceof interleukin-2 (3 1). Accordingly, Reis et al. (32) demonstratedthat the infiltrates in chronic chagasic cardiomyopathy were composedmainly of CD8+ T cells; killer cells and B lymphocytes were scarce,as well as somemacrophagesexpressingthe tumor necrosisfactor. Theseresults are rather different from thoseobtained in this paper and are discussedin the next paragraphs. Macrophages. Since the pioneering work of Metchnikoff, it becameevident that macrophagesplay an essential role as crucial mediators in the development of immune responseand inflammation t.33). On the other hand, our results also stressthe fact that about 50% to 60% of the infiltrating cells are mononuclear cells, namely, monocytes and macrophages.Although one must examine the possibility that macrophagesmay be recruited at the areasof necrosisto clear tissuedebris, the possibility that they might play a specific role as “facilitating macrophages”or “immunoregulatory monocytes” alsoshould be considered(7). Also, although the cellular immune responseis T-lymphocyte dependent (34), macrophagesappear as the most important effector cells in chronic chagasic infiltrates (35) where they form the bulk of the cellular response(17). In our study a certain percentage of morphologically macrophagecells were negative for the CD68 antigen both for peroxidase and alkaline phosphataseimmunostaining.
fields.
Although this may have been due to a technical problem, the antibody Kpl reacts against a specific epitope different from other antigens that are recognized by the different monoclonal antibodies. According to macrophageheterogeneity established by immunocytochemistry (36), only a wide range of macrophage-specificmonoclonal antibodies would recognize all the mononuclearphagocyte system. One may theorize that if there is a wide expression of membranereceptors of macrophages,these cells could also exhibit wide functional aspects. The various functions of macrophagesare exerted either in a cell-to-cell contact setupwith the target cells or by various biologically and pharmacologically active factors released by these cells. Among the most salient factors releasedby macrophagesare various monokinesand products of arachidonicacid (prostaglandinsand leukotrienes)(37). One of the consequencesof macrophage-mediatedcytotoxicity may be permeability alteration of the plasmamembranesof nearby target cells, possibly becauseof structural integrity changeof the membraneresulting from peroxidation of cell membranelipids (35). Macrophages are known to secrete large amounts of oxygen-free radicals. The secretion of their products such astumor necrosisfactor or interleukin-1 may enhancethe generationof oxygen-derived radicals(38). Macrophages present highly immunogenic antigens to the T-Lymphocyte system (39). Individual coexistence of, and contacts between, T lymphocytes and CD68-positive macrophageswere clearly demonstratedin this paper (see Figure 5). A clustering phenomenonalso was observed in other areas(seeFigure 2). Thesephenomenaare believed to be important for cell-to-cell interaction and subsequent T-cell activation and proliferation (40). Differences in r cruzi strainsand or cardiomyopathy severity may account for the different counts found in our paper from thoseof Higuchi (11) and Reis (32).
Cardiovasc July/August
Pathol Vol. 5, No. 4 1996:209-219
Figure 11. Combined theory. Factors involved in the production of myocardial damage in chronic Chagas’ cardiomyopathy. The major factors are in thick boxes. Reproduced from (4) with permission.
CHRONIC
Delayed
A Combined Theory for Chronic Chagas’ Cardiomyopathy It hasbeen saidthat the pathogenesisof chronic chagasic cardiac lesions remains an enigma (8). Different mechanismshave been proposed:autonomic involvement (22,41), lymphatics obstructed by fibrosis (42), microvascular disease(43,44), and autoimmune mechanisms(28,29). However no theory hasbeen definitively proved. Accordingly, Tafuri suggestedthat pathogenic mechanismsin chronic chagasiccardiomyopathy may be multifactorial (45). A “combined theory” also has been recently issued(4) basedon three ingredients: the parasite, macrophagesand lymphocytes, and the fibrosis. The parasite. It was generally accepted that 7: cruzi is scarcein the chronic phaseand that the inflammatory infiltrate was unrelated to the parasites(6). However, recent papers have shown that the parasites are present more frequently than it was thought (8). In fact, using dissimilar methods, different authors demonstratedeither the persistence of 7: cruzi or parasite antigensin mice (46), the parasite DNA sequenceamplified by the polymerasechain reaction (9), or T. cruzi antigens (8) from inflammatory lesions in human chagasiccardiomyopathy. The three studies undoubtedly showed the presenceof the parasiteor a part of it (B-46) or a portion of its genome (9) related to myocardial lesionsand chronic inflammatory infiltrates, suggestinga direct role for the parasitein the perpetuation of myocardial inflammation (8). In other words, the antigen stimulation would persistthroughout the chronic stage,even though the parasitesare not morphologically detectable by light microscopy (47). A delayed type of hypersensitivity reaction might also play an important role in this pathology (47). Teixeira et al. (48) have postulated the insertion of T. cvuzi DNA sequencesinto the host cell genome.Regardless,whatever the form of lesional parasite DNA, the chronic chagasiccardiomyopathy may be a consequenceof the continued persistence of ?: cruzi or its parts at the site of the lesion (9).
type
CHAGASIC
MILE1 ETAL. INFILTRATION
217
hypersensitivy
Macrophages and lymphocytes. As shown, mononuclear cells persistin the chronic stage,contributing to the inflammation through its products of secretion or through its own cytotoxicity (suppressorT cells) and cytolytic action (macrophages)(4). Apparently only autoimmunity may explain that all the processis restricted to the myocardium (49). Autoimmunity may be explained by the presenceof cross-reactivity antigens between myocardium and T cruzi or becausethe polyclonal activation might liberate antiautoclonesduring the acutephasethat may persista long time in the host (50). The fibrosis. According to Tafuri (45), the fibrosis causesthe progressivelossof the contracting capacity of the heart. Patients with chronic chagasic myocardiopathy exhibit heart failure in the late stageof the disease(1,4). Ultrastructural studieshave describeddegenerationof organelles, hypertrophy of myocytes, and increased fibrosis (51-52). Schaper(53) in chronic dilated cardiomyopathy showedimpairment of the myocardial ultrastructure and changesof the cytoskeleton. They concluded that the increaseof fibrosis, the degeneration of hypertrophied myocytes, and the alterations of the cytoskeleton (increaseof desmin, tubulin, and vinculin) are the morphologic correlatesof reducedmyocardial function in chronic heart disease.A similar pattern may be suspectedin chronic chagasicmyocardiopathy. Although much that is concerned with Chagas’ cardiomyopathy is controversial and much remains to be discovered, three factors may be the major ingredients for its pathogenesis:(1) the parasite: (2) lymphocytes and macrophages;and (3) the fibrosis. A multifactorial theory may then be suggestedto explain the sequenceof events that lead to the chronic stageof the disease(4; Figure 11). The authors wish to express their thanks to Miss Carina Elizabeth Kripelz for her skillfull editorial assistance. This investigation received partial financial support from the interuniversity cooperation between the Department of Pathology of the University of Milan, Italy, and the University of Buenos Aires, Argentina.
218
MILE1 ET AL. CHRONIC CHAGASIC
Cardiovasc Path01 Vol. 5, No. 4 July/August 1996:209-219
INFILTRATION
References
21. Romatia C. Enfermedad 1963;97--145.
1. Milei J, Storino R, Sanchez JA, Ferrans VJ. Does Chagas’ disease exist as an undiagnosed form of cardiomyopathy in the United States? Am Heart J 1992;123:1732-1735. 2. Editorial:
New Light on Chagas’
disease. Lancet
1965;i:1150.
3. UNDPiWorld Bank, WHO. Special Program for Research and Training in Tropical Disease. IV Programme Reports 1.7, Dec. 31, 1982, Geneva. 4. Storino Doyma,
R, Milei J. Enfermedad 1994,150.
de Chagas.
Buenos
Aires:
Mosby-
de Chagas.
Buenos
23. Molina HA, Kierszenbaum F. A study of human myocardial Chagas’ ‘disease: Distribution and frequency of inflammatory Int J Parasitol 1987;17:1297-1305.
6. Andrade Z. A patologia da doenca de Chagas Belge Med Trop 1985;65(suppl 1):15-30.
27. Acosta AM. 731256-258.
26. Kierszenbaum 72:201-211.
F. Autoimmunity Critical
comment
in Chagas’ (letter
28. Santos-Buch CA, Acosta AM. Pathology zard I, ed. Immunology and Pathogenesis Raton, FL; CRC Press, 1985;145-183.
8. Higuchi ML, Brito T, Martins Reis M, et al. Correlation between Trypanosoma cnrzi parasitism and myocardial inflammatory infiltrate in human chronic chagasic myocarditis: light microscopy and immunohistochemical findings. Cardiovasc Path01 1993;2:101-106.
29. Laguens RP. Immunologic mechanisms mental Chagas’ disease. Rev Fed Argent
10. Milei J, Storino RA, Beigelman R, Fem&ndez Alonso G, Mattuni L, Rossi L. Histopathology of specialized and ordinary myocardium and nerves in chronic Chagas’ disease, with a morphometric study of inflammation and fibrosis. Cardiologia 1991;36: 107-l 15. 11. Higuchi ML, Gutierrez PS, Aiello VD, Palomino S, Bocchi E, Pileggi F. Human chagasic cardiopathy and myocardial rejection: a similar pattern of myocarditis. Virchows Arch A Path01 Anat Histopathol 1993;423:157-160. 12. Consejo Argentino de Enfermedad de Chagas. “Dr. Salvador Editorial. Clasificacion de la enfermedad de Chagas cr6nica. Cardiol 1988;56:108-109. 13. Rossi L. Histopathology Ambrosiana, 1978;60.
of cardiac arrhythmias.
Milan:
Mazza.” Rev Arg
Casa Editrice
14. Guesdon JL, Temynck T, Anameas S. The use of avidin-biotin action in immunoenzymatic techniques. J Histochem Cytochem 27:1131-1134.
inter1979;
15. Edwards WD, Holmes DR, Reeder GS. Diagnosis of active lymphocytic myocarditis by endomyocardial biopsy. Quantitative criteria for light microscopy. Mayo Clin Proc 1982;57:419425. 16. Steenbergen C, Kolbeck PC, Wolfe JA, Anthony RY, Sanfilippo FP, Jennings RB. Detection of lymphocytes in endomyocardium using immunohistochemical techniques. Relevance to evaluation of endomyocardial biopsies in suspected cases of lymphocytic myocarditis. J Appl Cardiol 1986;1:63-73.
of
tissue in cell type.
mast cells? Hum Path01 disease. J Parasitol
to the editor).
7. Milei J, Storino R, Femandez Alonso Cl, Beigelman R, Vanzulli S, Ferrans VJ. Endomyocardial biopsies in chronic chagasic cardiomyopathy. Cardiology 1992;80:424-437.
9. Jones EM, Colley DG, Tostes S, Reis Lopez E, Vnencak-Jones CL, McCurley TL. Amplification of a Trypanosoma cruzi DNA sequence from inflammatory lesions in human chagasic cardiomyopathy. Am J Trop Med Hyg 1993;48:348-357.
Libreros,
24. Molina HA, Kierszenbaum F. Immunohistochemical detection of deposits of eosinophil-derived neurotoxin and eosinophil peroxidase in the myocardium of patients with Chagas’ disease. Immunology 1988; 64:725”3 1. 25. Esterisen RD. What is the role of myocardial 1985;16:536-538.
Ann Sot
L6pez
22. Koberle !F. Chagas’ disease and Chagas’ syndromes. The pathology American trypanosomiasis. Adv Parasitol 1968;6:63-116.
5. Sanchez IA, Milei J, Yu ZX, Storino R, Wenthold R, Ferrans VJ. Immunohistochemical localization of laminin in the hearts of patients with chronic chagasic cardiomyopathy: relationship to thickening of basement membranes. Am Heart J 1993;126:1392-1401. no homen.
Aires:
1986;
J Parasitol
1987;
of Chagas’ disease. of Trypanosomiasis.
In: TiBoca
of cardiac damage in experiCardiol 1988;17:166168.
30. Sztein hl, Washington RC, Kierszenbaum F. Trypanosomu cruzi inhibits the expression of CD3, CD4, CD8, and IL-2R by mitogen-activated helper and cytotoxic human lymphocytes. J Immunol 1990; 144:3558-3562. 31. Reed SG, Inverso JA, Roters SB. Heterologous antibody responses in mice with chronic Trypanosoma cruzi infection: depressed T-helper function restored with supematants containing interleukin-2. J Immuno1 1984;133:1558-1563. 32. Reis DD, Jones EM, Tostes S Jr, et al. Characterization of inflammatory infiltrates in chronic chagasic myocardial lesions: presence of tumor necrosis factor-alpha + cells and dominance of Granzyme A+, CD8+ lymphocytes. Am J Trop Med Hyg 1993;48:637-642. 33. Adams DO, Hamilton TA. Phagocyte cell: cytotoxic activities of macrophages. In: Gallin JI, Goldstein IM, Snyderman R, eds. Inflammation: Basic Principles and Clinical Correlates. New York Raven Press, 1988;471-492. 34. Hontebcyrie-Joskowicz M, Said G, Milon G, Marchal G, Eisen H. L3T4+ T cells able to mediated parasite-specific delayed-type hypersensitivity play a role in the pathology of experimental Chagas’ disease. Em J Immunol 1987;17:1027-1033. 35. Rossi MA. Myocardial role for macrophages.
damage in Trypanosoma cruzi myocarditis: Can J Cardiol 1990;6:293-298.
36. Dijkstra CD, Damoiseaux JGMC. Macrophage heterogeneity lished by immunocytochemistry. In: Progress in Histochemistry Cytochemistry, vol. 27/2. New York: Gustav Fisher, 1993.
A
estaband
37. Ben-Efrain S. Interactions between macrophage cytokines and ecosanoids in expression of antitnmor activity. Mediators of Inflammation 1992;1:295-308. 38. Ward PA, Johnson KJ, Warren JS, Kunkel RG. Immune complexes, oxygen radicals, and lung injury. In: Halliwell B, ed. Oxygen Radicals and Tissue Injury. Bethesda, MD: Federation of American Societies of Experimental Biology, 1988;107-114.
17. Milei J, Bortman G, Femandez Alonso G, Grancelli H, Beigelman R. Immunohistochemical staining of lymphocytes for the reliable diagnosis of myocarditis in endomyocardial biopsies. Cardiology 1990; 77~77-85.
39. Spencer SC, Fabre JW: Characterization of the tissue macrophage and the interstitial dendritic cell as distinct leukocytes normally resident in the connective tissue of rat heart. J Exp Med 1990;171:1841-1851.
18. Andrade ZA, Andrade SG. Chagas’ disease (American trypanosomiasis). In: Martial-Rojas RA, ed. Pathology of Protozoal and Hehnintic Diseases. Baltimore: Williams &Wilkins, 1971; chap 2.
40. Zhang J, Yu ZX, Shinsuke F, Yamaguchi ML, Ferrans V. Interstitial dendritic cells of the rat heart. Quantitative and ultrastructural changes in experimental myocardial infarction. Circulation 1993;87:909-920.
19. Andrade ZA, Andrade SG, Oliveira GB, Alonso DR. Histopathology of the conducting tissue of the heart in Chagas’ myocarditis. Am Heart J 1978;95:316-323.
41. Oliveira JSM. A natural model of intrinsic heart nervous system denervation: Chagas’ Cardiopathy. Am Heart J 1985:110:1092-1098.
20. Mazza S, J&g E, Canal Feijoo EF. Primer case cronico mortal de forma cardfaca de enfermedad de Chagas demostrado en Santiago de1 Estero. MEPRA 1938;38:1-17.
42. Rossi L, Storino R, Milei J, Matturri L. Deplecidn neuronal en la enfermedad de Chagas: todo deberfa reverse. Rev Arg Cardiol 1994, 62:239-246. 43. Morris SA, Tanowitz HB, Wittner M, Bilezikian JP. Patbophysiologi-
Cardmvasc July/August
Pathol Vol. 5, No. 4 1996:209-219
cal insights into the cardiomyopathy 1990;82:1900-1909.
CHRONIC
of Chagas’
44. Rossi MA, Mengel JO. Patogenese da miocardite papel de factores autoimunes e microvasculares. Sao Paulo, 1992;34:593-596.
disease.
Immunol
219
49. Andrade ZA. 126 (abstract).
chagasica cronica: o Rev Inst Med Trop
50. Minoprio PM, Eisen H, Fomi L, D’lmperio-Lima MR, Joskowicz MS, Coutinho A. Polyclonal lymphocyte responses to murine Ttypanosonn cruzi infection. Quantitation of both T- and B-cell responses. Stand J Immunol 1986;24:661-666.
46. Younees-Chennouti AB, Hontebeyre-Joskowicz M, Tricottet V, Eisen H, Reynes M. Said G. Persistence of Ttypanosoma cruzi antigens in the inflammatory lesions of chronically infected mice. Trans Roy Sot Trop Med Hyg 1988;82:77.
of Chagas’
MILE1 ET AL. INFILTRATION
disease. Circulation
45. Tafuri WL. Patogenese. VIII Reunao Anual de Pesquisa Aplicada em Doenca de Chagas. Rev Sot Bras Med Trop 1992;25(suppl III): 19 (abstract).
Pathogenesis
CHAGASIC
1991; 142:
51. Ferrans VJ, Milei J, Tomita Y, Storino RA. Basement membrane thickening in cardiac myocytes and capillaries in chronic Chagas’ disease. Am J Cardiol 1988;6 1: 1 137-I 140.
47. Andrade Z. Novos aspectos sobre a patogenia da doenca de Chagas. Rev Sot Bras Med Trop 1992;25(suppl III):8 (abstract).
52. Carrasco Guerra HA, Palacios-Pm E, Dagert de Scorza C, Molina C, Inglessis G, Mendoza RV. Clinical, histochemical, and ultrastructural correlation in septal endomyocardial biopsies from chronic chagasic patients: detection of early myocardial damage. Am Heart J 1987; 113:716-724.
48. Teixeira ARL, Lacava Z, Santana JM, Luna H. Insercao de DNA de Ttypanosoma cruzi no genoma de celula hospedeira de mamifero por meio de infeccao. Rev Sot Bras Med Trop 1991;24:55-61.
53. Schaper J, Froede R, Hein ST, et al. Impairment of the myocardial ultrastructure and changes of the cytoskeleton in dilated cardiomyopathy. Circulation. 199 I ;83:504.