**5. Conclusion and expected insights from the ETiCS study**

In the last two decades, much knowledge has accumulated with respect to the possible pathophysiologic and clinical implications of heart-directed aabs (Jahns et al., 2006; Fu, M., 2008).

Homologies between cardiomyocyte surface molecules (in particular membrane receptors) and viral or bacterial proteins have been proposed as a mechanism for the elaboration of endogenous cardiac autoantibodies by antigen mimicry (Elies et al., 1996; Hoebeke et al., 1996). Chagas' heart disease, a slowly evolving inflammatory cardiomyopathy, is one of the best investigated examples to highlight this mechanism. The disease is induced by the protozoon *Trypanosoma cruzi*. About 30% of the Chagas' patients develop antibodies that cross-react between the ribosomal P2beta-protein of *T. cruzi* and some specific amino acids present in the second extracellular loop of the human beta1-adrenoceptor (Elies et al., 1996).

Acute Myocarditis – A Trigger of

failure (Jahns et al., 2004).

Cardiac Autoimmunity? Expected Insights from the Etiology, Titre-Course… 395

(beta1-ECII). This might explain the successful use of ECII-mimicking peptides for the induction of specific anti-beta1-ECII antibodies in different animal-models (Magnusson et al., 1996; Jahns et al., 2004, 2006). Finally, by isogenic transfer of anti-beta1-ECII antibodies in a human-analogous rat model the (autoimmune-)attack against the hearts of healthy recipients by activating anti-beta1-antibodies has been identified as a possible cause of heart

In the last decade several research groups have independently shown that anti-beta1-ECIIantibodies preferentially recognize native membrane beta1-receptors in various immunological assays (cell-ELISA, immunoprecipitation, immunofluorescence), indicating that such antibodies or autoantibodies are conformational (Jahns et al., 1999, 2006; Fu 2008). In addition, functional tests revealed that such anti-beta1-ECII may also act as allosteric regulators of beta1-adrenoceptor activity through modulation of cellular cAMP-production and/or cAMP-dependent protein kinase (PKA) activity (Wallukat et al., 1991; Limas et al., 1992; Jahns et al., 1999; Nikolaev et al., 2007). The use of different screening-techniques renders direct comparisons of the available data difficult, however. Hence, so far reported prevalences should be always interpreted in the context of the detection method utilised. Nevertheless, taken all these reports together, there is wide consent that a substantial fraction of patients with DCM and ICM, but only very few healthy subjects have ciculating functionally active adrenoceptor-aabs (Wallukat et al., 1991; Magnusson et al., 1994; Jahns et al., 1999). An association of such aabs with impaired cardiac function (Jahns et al., 1999), a higher incidence of ventricular arrhythmias (Iwata et al., 2001; Störk et al., 2006), and a higher incidence of sudden cardiac death (Iwata et al., 2001) have been demonstrated. In addition, a previous clinical follow-up study also implies an increased risk for (cardiovascular) mortality in adrenoceptor aab-positive DCM patients (Störk et al., 2006). To further analyse the time-course and sequentially engaged immunologic processes in autoimmune-mediated heart disease, the ETiCS study will prospectively follow the evolution of cardiac morphology and function after a first inflammatory or ischemic myocyte damage. In total 400 patients will benefit from a structured follow-up and best standard available medical care. The central hypothesis of ETiCS is that both inflammatory and ischemic myocardial injury may trigger a sequence of immunologic reactions which result in the formation of functionally active cardiac receptor-aabs (Jahns et al., 2006). Thus, the pre-specified primary endpoint is the titre of receptor-aabs at diagnosis compared to 3, 6, and 12 months after the index event. Since these aabs are thought to confer additional risk, pre-specified secondary endpoints of the ETiCS study comprise occurrence of lifethreatening arrhythmias, changes in cardiac diameters and function, changes in clinical

status, time to cardiovascular death, and time to all-cause death.

Endomyocardial biopsies will allow for a correlation of cardiac aab-titres with the severity of myocardial inflammation (Aretz et al., 1987; Elliott et al., 2008; Kindermann et al., 2008) , with the kind of T cells engaged (Th1 vs. Th2-response, Treg, CTLs (Noutias et al. 2011)), and with the presence and/or activity of infective agents detected (that is, the type and load of pre-specified viral/microbial pathogens, as determined by PCR (Kühl et al., 2005)). Once available, the results of the ETiCS study will significantly contribute to a number of important diagnostic, pathophysiologic and prognostic issues in autoantibody-mediated heart disease (Fu, M., 2008). We expect insights on the role of inflammatory (and ischemic) cardiac damage in triggering autoimmune processes (including the involved immunologic paths), and on the relevance of heart-directed autoimmune reactions for the initiation or progression of heart failure (Jahns et al., 2006). In particular, we might learn and better

Because the large majority of functionally active beta1-aabs detected in DCM patients seem to be directed against the same receptor loop, it was speculated that these antibodies might also originate from molecular mimicry between the beta1-adrenoceptor and hitherto nonidentified viral pathogen (Hoebeke, 1996). Another example is Chlamydia–associated heart disease, which in BALBc mice appears to be induced by antigen mimicry between Chlamydia antigens and the alpha-myosin heavy chain molecule, resulting in activation of autoreactive T- and B-cells (Bachmaier et al., 1999).

However, irrespective of (potentially occurring) immunologic cross-reactions, to date no prospective clinical study has ever addressed the key question, whether structural damage to the heart muscle (e.g., necrosis or apoptosis) is a mandatory pre-requisite for the formation of heart-directed aabs.

In general, the immune system will not attack cardiac self-proteins. On a susceptible genetic background, however, this self-inhibition of immune effector cells after cardiac injury may be hampered. It is presently unclear, whether this autoreactivity depends on the kind and extent of injury, the kind and amount (or "dose") of self-antigens presented, or the kind and quality of the subsequently engaged immunologic paths. Regarding the latter aspect, a recent analysis of the expression of functional T cell markers in EMBs from patients with acute myocarditis and patients with chronic (post-inflammatory) DCM found an increased expression of CD3d, CD3z, and TRBC in both disease entities, whereas Th1 (more than) Th2 marker-genes as well as regulatory and cytotoxic T cells were differentially up-regulated in DCM EMBs only (Noutsias et al., 2011). Interestingly, in human DCM biopsies any clues for a Th17 response were lacking, which is in sharp contrast to findings in murine models of (experimental) autoimmune myocarditis (Valaperti et al., 2008). Nevertheless, from these recent molecular data it seems clear now that in human DCM a Th2 response also participates in the myocardial T cell infiltrates and serves as a pre-requisite for the stimulation (that is, maturation) of autoreactive B cells that, e.g., produce cardiac autoantibodies.

In acute myocarditis diffuse and/or focal inflammation causes structural damage to the heart. If myocardial inflammation persists, in a majority of cases cardiac function does not recover and finally may result in severe DCM. Acute ischemia also causes structural damage to the heart and to date represents the most common aetiology of heart failure. Thus, a structured follow up of patients with either disease is both pathophysiologically and clinically relevant.

The ETiCS study will follow such patients 3, 6, and 12 months after their index events, because the formation of autoreactive immunoglobulin G (IgG) is supposed to take place within the first 6 weeks (up to three months) after the index event.

A variety of autoreactive IgG have been identified, but only few of them have been investigated more in detail. Because myocyte surface receptors are easily accessible to circulating autoantibodies, the cardiac beta1-adrenergic receptor (which is the predominant adrenocepter subtype in the heart) and the M2-muscarinic acetylcholine receptor represent key targets for autoreactive antibodies that might affect heart function to some extent.

To generate an autoimmune response, membrane receptors must be degraded to small oligopeptides able to form a complex with a MHC class II or HLA molecule of the host (Hoebeke, 1996). In case of the human beta1-adrenoceptor, a computer-based search for potential immunogenic amino-acid streches within this (seven) transmembrane spanning protein revealed, that the only portion of the molecule containing B- and T-cell epitopes (and accessible to antibodies) was in fact the predicted second extracellular receptor loop

Because the large majority of functionally active beta1-aabs detected in DCM patients seem to be directed against the same receptor loop, it was speculated that these antibodies might also originate from molecular mimicry between the beta1-adrenoceptor and hitherto nonidentified viral pathogen (Hoebeke, 1996). Another example is Chlamydia–associated heart disease, which in BALBc mice appears to be induced by antigen mimicry between Chlamydia antigens and the alpha-myosin heavy chain molecule, resulting in activation of

However, irrespective of (potentially occurring) immunologic cross-reactions, to date no prospective clinical study has ever addressed the key question, whether structural damage to the heart muscle (e.g., necrosis or apoptosis) is a mandatory pre-requisite for the

In general, the immune system will not attack cardiac self-proteins. On a susceptible genetic background, however, this self-inhibition of immune effector cells after cardiac injury may be hampered. It is presently unclear, whether this autoreactivity depends on the kind and extent of injury, the kind and amount (or "dose") of self-antigens presented, or the kind and quality of the subsequently engaged immunologic paths. Regarding the latter aspect, a recent analysis of the expression of functional T cell markers in EMBs from patients with acute myocarditis and patients with chronic (post-inflammatory) DCM found an increased expression of CD3d, CD3z, and TRBC in both disease entities, whereas Th1 (more than) Th2 marker-genes as well as regulatory and cytotoxic T cells were differentially up-regulated in DCM EMBs only (Noutsias et al., 2011). Interestingly, in human DCM biopsies any clues for a Th17 response were lacking, which is in sharp contrast to findings in murine models of (experimental) autoimmune myocarditis (Valaperti et al., 2008). Nevertheless, from these recent molecular data it seems clear now that in human DCM a Th2 response also participates in the myocardial T cell infiltrates and serves as a pre-requisite for the stimulation (that is, maturation) of autoreactive B cells that, e.g., produce cardiac

In acute myocarditis diffuse and/or focal inflammation causes structural damage to the heart. If myocardial inflammation persists, in a majority of cases cardiac function does not recover and finally may result in severe DCM. Acute ischemia also causes structural damage to the heart and to date represents the most common aetiology of heart failure. Thus, a structured follow up of patients with either disease is both pathophysiologically and

The ETiCS study will follow such patients 3, 6, and 12 months after their index events, because the formation of autoreactive immunoglobulin G (IgG) is supposed to take place

A variety of autoreactive IgG have been identified, but only few of them have been investigated more in detail. Because myocyte surface receptors are easily accessible to circulating autoantibodies, the cardiac beta1-adrenergic receptor (which is the predominant adrenocepter subtype in the heart) and the M2-muscarinic acetylcholine receptor represent key targets for autoreactive antibodies that might affect heart function to some extent. To generate an autoimmune response, membrane receptors must be degraded to small oligopeptides able to form a complex with a MHC class II or HLA molecule of the host (Hoebeke, 1996). In case of the human beta1-adrenoceptor, a computer-based search for potential immunogenic amino-acid streches within this (seven) transmembrane spanning protein revealed, that the only portion of the molecule containing B- and T-cell epitopes (and accessible to antibodies) was in fact the predicted second extracellular receptor loop

within the first 6 weeks (up to three months) after the index event.

autoreactive T- and B-cells (Bachmaier et al., 1999).

formation of heart-directed aabs.

autoantibodies.

clinically relevant.

(beta1-ECII). This might explain the successful use of ECII-mimicking peptides for the induction of specific anti-beta1-ECII antibodies in different animal-models (Magnusson et al., 1996; Jahns et al., 2004, 2006). Finally, by isogenic transfer of anti-beta1-ECII antibodies in a human-analogous rat model the (autoimmune-)attack against the hearts of healthy recipients by activating anti-beta1-antibodies has been identified as a possible cause of heart failure (Jahns et al., 2004).

In the last decade several research groups have independently shown that anti-beta1-ECIIantibodies preferentially recognize native membrane beta1-receptors in various immunological assays (cell-ELISA, immunoprecipitation, immunofluorescence), indicating that such antibodies or autoantibodies are conformational (Jahns et al., 1999, 2006; Fu 2008). In addition, functional tests revealed that such anti-beta1-ECII may also act as allosteric regulators of beta1-adrenoceptor activity through modulation of cellular cAMP-production and/or cAMP-dependent protein kinase (PKA) activity (Wallukat et al., 1991; Limas et al., 1992; Jahns et al., 1999; Nikolaev et al., 2007). The use of different screening-techniques renders direct comparisons of the available data difficult, however. Hence, so far reported prevalences should be always interpreted in the context of the detection method utilised. Nevertheless, taken all these reports together, there is wide consent that a substantial fraction of patients with DCM and ICM, but only very few healthy subjects have ciculating functionally active adrenoceptor-aabs (Wallukat et al., 1991; Magnusson et al., 1994; Jahns et al., 1999). An association of such aabs with impaired cardiac function (Jahns et al., 1999), a higher incidence of ventricular arrhythmias (Iwata et al., 2001; Störk et al., 2006), and a higher incidence of sudden cardiac death (Iwata et al., 2001) have been demonstrated. In addition, a previous clinical follow-up study also implies an increased risk for (cardiovascular) mortality in adrenoceptor aab-positive DCM patients (Störk et al., 2006).

To further analyse the time-course and sequentially engaged immunologic processes in autoimmune-mediated heart disease, the ETiCS study will prospectively follow the evolution of cardiac morphology and function after a first inflammatory or ischemic myocyte damage. In total 400 patients will benefit from a structured follow-up and best standard available medical care. The central hypothesis of ETiCS is that both inflammatory and ischemic myocardial injury may trigger a sequence of immunologic reactions which result in the formation of functionally active cardiac receptor-aabs (Jahns et al., 2006). Thus, the pre-specified primary endpoint is the titre of receptor-aabs at diagnosis compared to 3, 6, and 12 months after the index event. Since these aabs are thought to confer additional risk, pre-specified secondary endpoints of the ETiCS study comprise occurrence of lifethreatening arrhythmias, changes in cardiac diameters and function, changes in clinical status, time to cardiovascular death, and time to all-cause death.

Endomyocardial biopsies will allow for a correlation of cardiac aab-titres with the severity of myocardial inflammation (Aretz et al., 1987; Elliott et al., 2008; Kindermann et al., 2008) , with the kind of T cells engaged (Th1 vs. Th2-response, Treg, CTLs (Noutias et al. 2011)), and with the presence and/or activity of infective agents detected (that is, the type and load of pre-specified viral/microbial pathogens, as determined by PCR (Kühl et al., 2005)).

Once available, the results of the ETiCS study will significantly contribute to a number of important diagnostic, pathophysiologic and prognostic issues in autoantibody-mediated heart disease (Fu, M., 2008). We expect insights on the role of inflammatory (and ischemic) cardiac damage in triggering autoimmune processes (including the involved immunologic paths), and on the relevance of heart-directed autoimmune reactions for the initiation or progression of heart failure (Jahns et al., 2006). In particular, we might learn and better

Acute Myocarditis – A Trigger of

**7. Acknowledgements** 

**8. References** 

283: 1335-1339.

115,1: 76-83.

*Heart J.* 16, Suppl.O: 68-70.

Cardiac Autoimmunity? Expected Insights from the Etiology, Titre-Course… 397

further understand and therapeutically modulate the immune system with respect to

The investigator-initiated ETiCS study receives public funding from the German Ministry for Education and Research (Bundesministerium für Bildung und Forschung, BMBF Project number 01ES0816). The ETiCS study has been acknowledged by the German Competence Network Heart Failure (CNHF, subproject 6b), and has equally been associated to the BMBF-funded Comprehensive Heart Failure Centre, University Hospital of Würzburg (CHFC, associated project C4). All authors had full access to the data and have read and

American Heart Association, Dallas (2009). Heart disease and stroke statistics - 2009 update.

Aretz, H.T., Billingham, M.E., Edwards, W.D., Factor, S., Fallon, J.T., Fenoglio, J.J., Jr., Olsen,

Baba, A., Yoshikawa, T., Fukuda, Y., Sugiyama, T., Shimada, M., Akaishi, M., Tsuchimoto,

Bachmaier, K., Neu, N., de la Maza, L.M., Pal, S., Hessel, A., Penninger, J.M. (1999).

Boivin, V., Jahns, R., Gambaryan, S., Ness, W. & Lohse, M. (2001). Immunofluorescent imaging of beta1- and 2-adrenergic receptors in rat kidney*. Kidney Int.* 59: 515-531. Borda, E., Pascual, J., Cossio, P., Vega, M., Araña, R. & Sterin-Borda, L. (1984). A circulating

Caforio, A.L., Goldman, J.H., Baig, M., Keeling, P.J., Bottazzo, G.F. & McKenna, W.J. (1995).

Caforio, A.L. & Iliceto, S. (2008). Genetically determined myocarditis: clinical presentation and immunological characteristics. *Curr. Opin. Cardiol.* 23,3: 219-226. Caforio, A.L., Mahon, N.G., Baig, M., Tona, F., Murphy, R.T., Elliott, P.M.& McKenna, W.J.

Caforio, A.L., Mahon, N.J., Tona, F. & McKenna, W.J. (2002). Circulating cardiac

Christ, T., Schindelhauer, S., Wettwer, E., Wallukat, G. & Ravens, U. (2006). Interaction

A report from the American Heart Association Statistics Committee and Stroke

E. & Schoen, F.J. (1987). Myocarditis. A histopathologic definition and

K., Ogawa, S. & Fu M (2004). Autoantibodies against M2-muscarinic acetylcholine receptors: new upstream targets in atrial fibrillation in patients with dilated

Chlamydia infections and heart disease linked through antigenic mimicry. *Science* 

IgG in Chagas disease which binds to -adrenoceptor of myocardium and

Organ-specific cardiac autoantibodies in dilated cardiomyopathy - an update. *Eur.* 

(2007). Prospective familial assessment in dilated cardiomyopathy: cardiac autoantibodies predict disease development in asymptomatic relatives. *Circulation*

autoantibodies in dilated cardiomyopathy and myocarditis: pathogenetic and

between autoantibodies against the beta1-adrenoceptor and isoprenaline in

immune-mediated (post-inflammatory) human cardiac disorders.

approved the final article and have declared no conflicts of interest.

Statistics Subcommittee. *Circulation* 119: 480-486.

classification. *Am. J. Cardiovasc. Pathol.* 1,1: 3-14.

cardiomyopathy. *Eur. Heart J.* 25: 1108-1115.

modulates its activity. *Clin. Exp. Immunol.* 57: 679-685.

clinical significance. *Eur. J. Heart Fail.* 4: 411-417.

understand whether –and if so, then to which extent– the specific target of a heart-directed autoantibody, but also its respective titre and biological activity and –last not least– its respective kinetics (that is, antibody-persistence or -clearance over time) relate to the complex process of cardiac wounding and healing. Different cardiac aabs might have distinct propensities to induce a certain cardiac phenotype, and the ETiCS study will allow for a differentiation of such features in a prospective manner. Thereby, additional prognostic markers for patients with an unfavourable course of autoimmune heart failure might be recognised and, as a consequence, conventional treatment modalities could be optimized earlier and/or novel -more specific- treatment strategies could be developed.
