Autoimmune Response in Chronic Ongoing Myocarditis Demonstrated by Heterotopic Cardiac Transplantation in Mice
Background Autoimmune mechanisms have been implicated in the pathogenesis of chronic ongoing myocarditis. To investigate this relation, we used an A/J mouse model inoculated with coxsackievirus B3 and determined whether myocarditis would be transferred to normal hearts that were heterotopically transplanted.
Methods and Results Inbred 3-week-old A/J mice were inoculated intraperitoneally with coxsackievirus B3 (Nancy strain; 2×104 plaque-forming units) and housed for >60 days. The presence of the viral genome in the myocardium was determined by the polymerase chain reaction with primers specific for the 5′ end of the coxsackievirus B3 genome performed at 40, 50, or 60 days after inoculation. Normal A/J mouse hearts were transplanted into the same strain of mice without myocarditis (group A) and into mice with chronic ongoing myocarditis (group B). The hearts were evaluated histologically 2 weeks after transplantation. Conventional histological examination of infiltrated T cells and macrophages was performed, and the expression of intercellular adhesion molecule-1, major histocompatibility complex (MHC) class I antigen, and MHC class II antigen was evaluated by immunoenzymatic staining. The concentrations of interleukin-1α (IL-1α) and tumor necrosis factor (TNF-α) in the grafts were measured with an ELISA. The viral RNA genomes were not detected in the mice with chronic ongoing myocarditis, but their transplanted hearts did show myocarditis. In the hearts with induced myocarditis, infiltrated mononuclear cells consisted of CD4+ T cells, CD8+ T cells (CD4+ cell number >CD8+ cell number), and macrophages. Intercellular adhesion molecule-1, MHC class I antigen, and MHC class II antigen were expressed in the vascular endothelial cells and myocardial cells in and around the infiltrated lesions. The concentrations of IL-1α and TNF-α in group B were significantly higher than those in group A (group A versus group B: IL-1α, 125±35 versus 180±34 pg/mL; TNF-α, 45±15 versus 96±40 pg/mL; P<.05).
Conclusions Results suggest that an autoimmune response may play a key role in the progression of chronic ongoing myocarditis.
Myocarditis is characterized histologically by the infiltration of inflammatory cells in the myocardium that is accompanied by myocardial cell damage in a pattern differing from that of ischemic heart disease; most cases of acute myocarditis in patients living in developed countries are thought to be the result of viral infection.1 Although an causal link between viral myocarditis and DCM has long been recognized, the nature of this relation is controversial. The progression from viral myocarditis to DCM is thought to involve autoimmunity and the presence of persistent viral infection.2 When viral genome was detected in the myocardium of heart transplant recipients with DCM, the persistent viral infection theory gains prominence. However, recent data demonstrated that the number of transplant cases with such viral infection is less than expected.3
We previously demonstrated the abnormal immunological responses in patients with DCM.4 We also developed a murine model of chronic ongoing viral myocarditis in A/J mice that were inoculated with coxsackievirus B3. This model demonstrated that during the chronic ongoing myocarditis, CD4+ T cells were found to be the main infiltrates in this model. The ICAM-1 was expressed on the endothelial cells of vessels in and around the infiltrated lesions; in contrast, MHC class I and II antigens were expressed on infected myocardial cells. More interestingly, the coxsackievirus B3 genome was not detected in the myocardia of animals with chronic ongoing myocarditis.5
In this study, to investigate an autoimmune link to the pathogenesis of chronic ongoing myocarditis, we heterotopically transplanted normal A/J mouse hearts into A/J mice with chronic ongoing myocarditis and analyzed immunological factors such as IL-1α and TNF-α and the expression of ICAM-1 and MHC classes I and II in transplanted hearts.
Sixty inbred 3-week-old male A/J (H-2a) mice, certified to be virus free, were purchased from Japan SCL (Shizuoka, Japan). Room temperature was maintained at 22±2°C, and the humidity at 45±5%. Animals were housed five per cage and given water and rodent chow ad libitum.
Virus Preparation and Infection of Mice
The coxsackievirus B3 (Nancy strain) used in this study was kindly provided by Akira Matsumori, MD (Kyoto University, Kyoto, Japan). The virus was stored at −80°C until use. Sixteen mice were administered an intraperitoneal injection of 2×104 plaque-forming units of the coxsackievirus in 0.2 mL of saline. The infected mice were isolated, five per cage, in a special unit for >8 weeks.
The presence of viral genome was investigated by the PCR with primers specific for the 5′ end of the coxsackie genome (CVB1 446-640). This was done at 40 (n=2), 50 (n=1), or 60 (n=1) days after the inoculation, as previously reported.6 7 Also, three age-matched noninfected control mice were killed, and at 2 weeks after inoculation, one infected mouse was prepared for positive control. In brief, nucleic acid extracted from the heart was purified by the acid guanidinium thiocyanate/phenol/chloroform method. α-Tubulin oligonucleotide primers were used as a positive control to confirm the extraction of nucleic acid. cDNA was then synthesized using the Moloney murine leukemia RT (200 U/μL, BRL) and was amplified by PCR. DNA was amplified in a DNA thermal cycler (Perkin Elmer Cetus), as described by Chapman et al,7 as follows: heart denaturation for 1 minute at 94°C, 10 cycles (1 cycle consisted of 94°C for 1 second, 50°C for 1 second, and 72°C for 1 second), and then 40 cycles (one cycle consisted of 94 for 1 second and 50 for 1 second). A total of 10 μL of a PCR mixture was electrophoresed at 80 V into 1.5% agarose gels in 0.2× TAE (1× TAE consisted of 40 mmol/L Tris-acetate and 1 mmol/L EDTA, pH 8.0). Gels were stained in ethidium bromide (1 μg/mL) to detect DNA by illumination with ultraviolet light and photographed.
Heterotopic Cardiac Transplantation
At 8 to 12 weeks after the inoculation, 12 mice with viral myocarditis served as the recipients, and 12 age-matched normal A/J mice served as the donors. Heterotopic cardiac transplantation was performed by a modification of the method of Corry et al.8 Donors and recipients were anesthetized by the administration of 5% pentobarbital (0.015 mL/g). Donor hearts were perfused with chilled heparinized saline via the inferior vena cava and were harvested after the ligation of the vena cava and the pulmonary veins. Harvested hearts were preserved in chilled saline. The aorta and pulmonary artery of each donor heart were anastomosed to the abdominal aorta and inferior vena cava of the recipient animal by microsurgical techniques. The total operative time was ≈60 minutes, and the duration of ischemia was ≈30 minutes. The surgical success rate was ≈90%. Technical failures observed within the first 72 hours were excluded from evaluation. Viability of the cardiac grafts was assessed by palpation of the transplanted heart. Normal A/J mouse hearts were transplanted into the same strain of nonmyocarditis mice (group A), as well as into mice with chronic ongoing myocarditis (group B). At 2 weeks after surgery, the grafts were excised and compared immunologically with the hearts grafted into normal A/J mice. No immunosuppressive agents were administered.
A total of 10 mice with heart transplants (group A, n=5; group B, n=5) were killed 14 days after the operation. The transplanted hearts were cut into halves along the midline. One half was fixed in 10% buffered formalin and embedded in paraffin according to standard procedures. These specimens were serially cut into 4-μm-thick sections, stained with hematoxylin and eosin and Azan solutions, and examined with light microscopy. The other half was quickly frozen and used for immunoenzymatic staining. Frozen specimens were embedded with O.C.T. compound (Milesc) and sectioned into 4-μm slices with the use of a cryostat. Immunoenzymatic staining was performed with the Dako LSAB Kit (Dako) according to the manufacturer's instructions. Briefly, the sections were preincubated with 0.3% hydrogen peroxide and normal goat serum to block nonspecific reactions. Antibodies against mouse CD4 (1:10 dilution, Caltag Laboratories), CD8 (1:10 dilution, Caltag Laboratories), macrophage (1:10 dilution, Caltag Laboratories), ICAM-1 (1:100 dilution, Seratec), MHC class I antigen (1:25 dilution, Cedarlane Lab), or MHC class II antigen (1:10 dilution, Seratec) were added and incubated for 30 minutes at room temperature. Sections were then incubated with biotinylated rabbit anti-rat immunoglobulins for 20 minutes, followed by incubation with horseradish peroxidase–labeled streptavidin solution for 10 minutes. The slides were rinsed in cold Tris-buffered saline after each step of incubation. Peroxidase activity was visualized with substrate-chromogen solution. Finally, slides were counterstained with hematoxylin solution. Negative controls were carried out with nonimmune serum instead of primary antibody.
For quantification of lymphocyte subsets, the sections were examined by two independent examiners in a blinded fashion. The extent of CD4+ and CD8+ T-cell and macrophage infiltration in each group was assessed by determining cell counts per high-power field. The number of lymphocytes per high-power field in each section that were stained by each monoclonal antibody was recorded, and the number of stained lymphocytes was then calculated. This process was repeated for >20 fields.
The intensity of other immunostainings was graded semiquantitatively on a four-point scale from 0 to 3. Grade 0 indicated the absence of any staining; grade 1, positive immunostaining associated with <10% of the cells; grade 2, positivity of 10% to 30% of the cells; and grade 3, positivity of >30% of the cells. Grading was performed independently by two of the investigators who had no knowledge of the data. Grades assigned were agreed to within one grade, with any differences resolved by consensus.
The concentration of IL-1α and TNF-α in the grafts was determined with ELISA kits for mouse IL-1α and TNF-α (Genzyme). For cytokine measurements, 10 harvested whole grafts (group A, n=5; group B, n=5) were immediately immersed in liquid nitrogen. A volume of 500 μL of PBS was added to a graft, and this mixture was homogenized with an ultrasonic homogenizer. A 96-well microtiter plate precoated with monoclonal antibody was first blocked with protein-based buffer and then used to capture the cytokine present in standards and test samples. All standards and samples were measured in duplicate. After washing to remove unbound material, a tagged polyclonal antibody that bounds the captured cytokine was added in microtiter plate with 96 wells. In the IL-1α assay, after washing to remove unbound material, a peroxidase-conjugated avidin that bound the biotin-tagged immune complexes captured on the plate was added to each well. In the TNF-α assay, after washing of the plate to remove unbound material, a peroxidase-conjugated polyclonal anti-mouse TNF-α (horseradish peroxidase–conjugate) that bound captured mouse TNF-α was added. These plates were rewashed to remove unbound material. The change in color catalyzed by peroxidase was subsequently stopped by acidification, and the absorbance was measured at 450 nm. A standard curve was constructed to determine the concentration of each cytokine in the samples. The minimum detectable concentration level was 15 pg/mL in each assays. The extraction rate was 20.7% and 29.0% for IL-1α and TNF-α in the whole heart.
Data are expressed as mean±SD. The normal distribution of each value was confirmed before were performed the statistical computations and significance testing. Student's t test was used to compare differences between the two groups. A level of P<.05 was acceptable as statistically significant.
Fig 1⇓ shows the PCR products obtained from both groups of mice. A 195–base-pair band that corresponded to the size of the coxsackievirus B3 genome PCR product was detected only in the positive controls. On the other hand, none of the samples from the noninfected control or from mice inoculated with CVB3 for 40-, 50-, or 60-day PCR products corresponding to the 195 base pairs were detectable with RT-PCR.
Histological Findings and Cytokine Concentrations
Photomicrographs are shown of representative cardiografts in a normal mouse (Fig 2A⇓) and in a mouse with myocarditis (Fig 2B⇓). The myocardium of the transplanted heart from group B appeared to be edematous and swollen compared with that of group A. There were focal infiltrations of mononuclear cells in the myocardia of the transplanted heart from group B (Fig 2C⇓). In addition, no pericarditis resulting from the operation was discernible in the transplanted hearts obtained from both group A and B.
The infiltrated mononuclear cells were composed of CD4+ T cells, CD8+ T cells, and macrophages (Fig 3A through 3C⇓). The cell counts per high-power field of both CD4+ T-cell and CD8+ T-cell infiltration into the grafts of group B were significantly increased compared with those of group A (Table⇓). In addition, the increase in the cell number of CD4+ T cells infiltrated into the grafts obtained from group B was twice that of CD8+ T cells in the same grafts.
Fig 4⇓ demonstrates that ICAM-1 and MHC class I and II antigens were constitutively expressed on the myocardial cells and vascular cells in and around the lesions in the heart in both group A and B. ICAM-1 was expressed mainly in the endothelium of the small vessels (Fig 4F⇓). On the other hand, MHC classes I and II were expressed not only on the vascular endothelial cells but also on the myocardial cells (Fig 4D through 4F⇓). The expression of ICAM-1 and MHC classes I and II was graded semiquantitatively on a four-point scale from 0 to 3. The semiquantitative assessment of the intensity of these immunostainings demonstrated the significant difference of the intensity between both groups (group A versus group B: MHC class I, 0.2±0.4 versus 1.6±0.5; MHC class II, 0.4±0.5 versus 2.4±0.5; ICAM-1, 0.6±0.5 versus 2.8±0.4; P<.01, respectively). There were few cellular infiltrates and little expression of MHC class I and II stains in group A in comparison with group B (Fig 4A and 4B⇓⇓).
The concentrations of both IL-1α and TNF-α in the myocardium of group B were significantly higher than those of group A (P<.05, respectively; Fig 5⇓).
Heart transplantation has become an established method of treating patients with end-stage heart disease in the past decade. To date, >13 000 cardiac transplants have reportedly been performed at 230 transplantation centers. Approximately 50% of heart transplant recipients have DCM as his or her primary myocardial disease.9 On the other hand, substantiation of a definite link between myocarditis and the development of DCM was initially based on clinical studies.10 Although the pathogenesis of DCM is uncertain, it has been suggested that an episode of subclinical viral myocarditis can initiate an autoimmune reaction that culminates in the development of DCM.11
In this study, we demonstrated that (1) the viral RNA genomes were not detected in the mice with chronic ongoing myocarditis, but the transplanted hearts obtained from chronic ongoing myocarditic mice did show myocarditis; (2) CD4+ T cells were mainly found in inflammatory lesions, with ICAM-1, MHC class I antigen, and MHC class II antigen being expressed in the myocardium of transplanted hearts in the chronic ongoing myocarditic mice; and (3) the concentration of both IL-1α and TNF-α significantly increased in the transplanted hearts into chronic ongoing myocarditic mice.
Witebsky et al12 proposed rationales for the autoimmune basis of clinical disease. They were consciously modeled on Koch's postulates and required that an autoimmune response be recognized, as either an autoantibody or a cell-mediated immunity; that the corresponding antigen be identified; and that an analogous autoimmune response be induced in an experimental animal. In addition, the immunized animal must also develop a similar disease. Recently, Rose and Bona13 proposed novel criteria for autoimmune disease based on new information gained from the use of molecular biological and hybridoma techniques; these include direct evidence of the transfer of a pathogenetic antibody or pathogenetic T cells, indirect evidence based on the reproduction of the autoimmune disease in experimental animals, and circumstantial evidence from clinical clues. Using an animal model, we first demonstrated that myocarditis was transferred to a normal heart that was transplanted into a mouse with chronic ongoing myocarditis. This finding suggested the possibility that circulating pathogenetic T cells in the A/J mouse with chronic ongoing myocarditis responded to the myocardium of the transplanted normal heart.
Our results showed that the infiltrated mononuclear cells consisted of CD4+ T cells, CD8+ T cells, and macrophages in Fig 3⇑. The cell number of CD4+ T cells infiltrated into the transplanted hearts obtained from chronic ongoing myocarditic mice was 20-fold higher; on the other hand, that of CD8+ T cells was 10-fold higher than that of noninfected mice in this experiment (Table⇑). These CD4+ T cells are expected to react primarily with adhesion molecules and MHC class II antigens, and they may induce graft injury through secretion of cytokines and recruitment of nonspecific macrophages and monocytes, which can also cause graft injury through a delayed-type hypersensitivity reaction. On the other hand, CD8+ T cells, which interact primarily with monocytes, expressing MHC class I antigens, can induce cell injury directly through a calcium-dependent or -independent mechanism.14 Importantly, a particular subset of CD4+ T cells, the T helper 1 cells, has been shown to contribute to the development of organ-specific autoimmune disease.15 Results of experimental T cell–mediated organ-specific autoimmune diseases, such as experimental autoimmune encephalitis16 and insulin-dependent diabetes mellitus,17 strongly support an immunological role in disease progression. Nevertheless, there remain some questions about the source of mononuclear cell infiltration from the host cell: whether it is triggered by the transplanted heart, which is immunologically slightly different from the host; whether it is activated by cross-reacting autoantibodies; or whether this is a reaction resulting from cross-reacting antigens or molecular mimicry. We previously confirmed that 12-week-old male A/J (H-2a) mice that were certified to be virus free accepted skin grafts from the same strain of mice (data not shown). We therefore speculated that the mononuclear cell infiltration was not the result of only cross-reacting antigens and molecular mimicry but also an autoimmune response that occurred in the recipient animal.
With regard to rejection of the cardiac graft, there are two possible pathways proposed: the direct and the indirect pathway. First, the direct pathway involves peptides in the groove of allogeneic MHC molecules that are expressed on donor antigen-presenting cells and reflects the recognition of these peptide differences within the α-helices in contact with T-cell receptors. Second, the indirect pathway involves the T-cell receptor recognition of donor MHC peptides in the groove of self–MHC class II molecules presented by recipient antigen-presenting cells.18 In our experimental model, we used the same inbred strain of mice, so it is unlikely that the myocarditis that occurred in the infected mice was due to the rejection. It is possible that the induced myocarditis in the donor hearts was caused by a circulating autoreactive T cells to the myocardium of the recipient.
The concentrations of IL-1α and TNF-α were elevated in the hearts that were transplanted into the infected mice compared with the control animals shown in Fig 5⇑. Huber et al19 showed that coxsackievirus B3–induced myocarditis may depend on the release of specific cytokines, IL-1 or IL-2, during infection. These authors also demonstrated that the activation of Th1 cells may be important in disease pathogenesis in Balb/c mice with the myocarditic H3 virus variant and the nonmyocarditic H310A1 virus variant. These cytokines are also known to regulate the expression of ICAM-1 on the vascular endothelial cells.20 21 Furthermore, in this experiments, we showed that in the grafts transplanted into the infected animals, ICAM-1, MHC class I antigen, and MHC class II antigen were expressed not only in the vascular endothelial cells but also in myocardial cells in Fig 4⇑. These findings implicate an involvement of the immune system in the induction of myocarditis in these animals.
The relation between chronic myocarditis and DCM is controversial. If these two diseases are related, our model may suggest the possibility of the post-transplantation immune response to the myocarditis-related DCM patient who underwent cardiac transplantation. It is likely that many patients with myocarditis have undergone cardiac transplantation and have received long-term immunosuppressive agents that would have concealed the post-transplantation immune response in myocarditis-related patients.
There are several reports indicating that candidates for autoantigens are β-adrenergic receptors,22 laminin,23 mitochondrial adenine nucleated translocator,24 branched-chain α-keto acid dehydrogenase protein,25 and cardiac myosin.26 We also proposed that α-gal antigen and three kinds of myocardial cell membrane proteins are possible antigens for autoreactive T cells to the myocardium.4
Finally, the adoptive transfer experiments show that the virus can surreptitiously persist in immunocytes at low levels and then reinfect the heart in the new host.27 We could not exclude this possibility through the use of the PCR method in these experiments.
In conclusion, in the present study, we demonstrated that the viral RNA genomes were not detected in the mice with chronic ongoing myocarditis, but their transplanted hearts did show myocarditis, suggesting that an autoimmune response may play a key role in the progression of chronic ongoing myocarditis. Further experiments are required to verify the autoimmune mechanism of chronic ongoing myocarditis and the development of subsequent DCM and the autoantigen that is responsible for the transmission of myocarditis in cardiac transplant models.
Selected Abbreviations and Acronyms
|ICAM-1||=||intercellular adhesion molecule-1|
|MHC||=||major histocompatibility complex|
|PCR||=||polymerase chain reaction|
|TNF||=||tumor necrosis factor|
We acknowledge the excellent technical advice by Dr Kimikazu Hamano (First Department of Surgery, Yamaguchi University School of Medicine) for cardiac transplantation and by Kazuto Yamaguchi and Tomoaki Murata (Institute of Laboratory Animals, Yamaguchi University School of Medicine). We also would like to thank Ms. Ishihara for her excellent technical assistance.
- Received June 14, 1996.
- Revision received July 31, 1996.
- Accepted August 7, 1996.
- Copyright © 1996 by American Heart Association
Sole MJ, Liu P. Viral myocarditis: a paradigm for understanding the pathogenesis and treatment of dilated cardiomyopathy. J Am Coll Cardiol. 1993;22(suppl 4A):99A-105A.
Keeling PJ, Jeffery S, Caforio ALP, Taylor R, Bottazzo GF, Davies MJ, Mckenna WJ. Similar prevalence of enteroviral genome within the myocardium from patients with idiopathic dilated cardiomyopathy and controls by the polymerase chain reaction. Br Heart J. 1992;68:554-559.
Nakamura H, Yamamura T, Kohro J, Fukuta S, Matsuzaki M, Matsumori M. The pathogenesis of perpetuating myocarditis: A/J mouse inoculated with Coxsackievirus B3. J Mol Cell Cardiol. 1994;26(suppl):117. Abstract.
Chapman NM, Tracy S, Gauntt CJ, Fortmuller U. Molecular detection and identification of enteroviruses using amplification and nucleic acid hybridization. J Clin Microbiol. 1990;28:843-850.
Woodruff JF. Viral myocarditis: a review. Am J Pathol. 1980;101:427-479.
Makino S, Kunimoto K, Muraoka Y, Mizushima Y, Katagiri K, Tochino Y. Breeding of a non-obese, diabetic strain of mice. Exp Anim. 1980;29:1-13.
Huber SA, Polgar J, Schultheiss P, Schwimmbeck P. Augmentation of pathogenesis of coxsackievirus B3 infections in mice by exogenous administration of interleukin-1 and interleukin-2. J Virol. 1994;68:195-206.
Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA. Induction by IL-I and interferon-gamma: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J Immunol. 1986;137:245-254.
Pober JS, Gimbrone MJ, Lapierre LA, Mendrick DL, Fiers W, Rothlein R, Springer TA. Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol. 1986;137:1893-1896.
Maish B, Wedeking U, Kochsiek K. Quantitative assessment of antilaminin antibodies in myocarditis and perimyocarditis. Eur Heart J. 1987;8(suppl J):233-236.
Schulze K, Becker BF, Schultheiss HP. Antibodies to the ADP/ATP carrier, an autoantigen in myocarditis and dilated cardiomyopathy, penetrate into myocardial cells and disturb energy metabolism in vivo. Circ Res. 1989;64:179-192.
Herskowitz A, Ahmed AA, Neumann DA, Beschorner WE, Rose NR, Soule LM, Burek CL, Sell KW, Baughman KL. Induction of major histocompatibility complex antigens within the myocardium of patients with active myocarditis: a nonhistologic marker of myocarditis. J Am Coll Cardiol. 1990;15:624-632.
Inamoto T, Hanada H, Miyanishi T, Yajima E, Nakayama S, Maita T, Kodama M, Izumi T, Shibata A, Abo T. Localization of porcine cardiac myosin epitopes that induce experimental autoimmune myocarditis. Circ Res. 1995;76:726-733.
Schwimmbeck PL, Badorff C, Schultheiss HP, Strauer BE. Transfer of human myocarditis into severe combined immunodeficiency mice. Circ Res. 1994;75:156-164.