T-Cell Receptor Sequences Identify Combined Coxsackievirus- Streptococci Infections as Triggers for Autoimmune Myocarditis and Coxsackievirus- Clostridia Infections for Type 1 Diabetes

Abstract

Recent research suggests that T-cell receptor (TCR) sequences expanded during human immunodeficiency virus and SARS-CoV-2 infections unexpectedly mimic these viruses. The hypothesis tested here is that TCR sequences expanded in patients with type 1 diabetes mellitus (T1DM) and autoimmune myocarditis (AM) mimic the infectious triggers of these diseases. Indeed, TCR sequences mimicking coxsackieviruses, which are implicated as triggers of both diseases, are statistically significantly increased in both T1DM and AM patients. However, TCRs mimicking Clostridia antigens are significantly expanded in T1DM, whereas TCRs mimicking Streptococcal antigens are expanded in AM. Notably, Clostridia antigens mimic T1DM autoantigens, such as insulin and glutamic acid decarboxylase, whereas Streptococcal antigens mimic cardiac autoantigens, such as myosin and laminins. Thus, T1DM may be triggered by combined infections of coxsackieviruses with Clostridia bacteria, while AM may be triggered by coxsackieviruses with Streptococci. These TCR results are consistent with both epidemiological and clinical data and recent experimental studies of cross-reactivities of coxsackievirus, Clostridial, and Streptococcal antibodies with T1DM and AM antigens. These data provide the basis for developing novel animal models of AM and T1DM and may provide a generalizable method for revealing the etiologies of other autoimmune diseases. Theories to explain these results are explored.

Keywords: Clostridia; Streptococci; T-cell receptors; anti-idiotype; antigen complementarity; autoimmune myocarditis; autoimmunity; coxsackievirus; diabetes; insulin; laminin; mimicry; myosin; synergism.

Figure 1

Percentage of T-cell receptor (TCR) sequences from type 1 diabetes mellitus (T1DM) patients that mimic 37 common human viruses compared with the percentage mimicked by TCRs from healthy individuals. Asterisks (*) indicate that the difference is statistically significant after Bonferroni correction (see Appendix A, Table A1). HHV = human herpes virus; HTLV = human T-lymphotropic virus; RSV = respiratory syncytial virus.

Figure 2

Percentage of T-cell receptor (TCR) sequences from autoimmune myocarditis (AM) patients that mimic 37 common human viruses compared with the percentage mimicked by TCRs from healthy individuals. Asterisks (*) indicate that the difference is statistically significant after Bonferroni correction (see Appendix A, Table A1). HHV = human herpes virus; HTLV = human T-lymphotropic virus; RSV = respiratory syncytial virus.

Figure 3

Percentage of T-cell receptor (TCR) sequences from type 1 diabetes mellitus (T1DM) patients that mimic 25 bacteriophages compared with the percent mimicked by TCRs from healthy individuals. Asterisks (*) indicate that the difference is statistically significant after Bonferroni correction (see Appendix A, Table A2).

Figure 4

Percentage of T-cell receptor (TCR) sequences from type 1 diabetes mellitus (T1DM) patients that mimic 42 common human bacteria compared with the percentage mimicked by TCRs from healthy individuals. Asterisks (*) indicate that the difference is statistically significant after Bonferroni correction (see Appendix A, Table A3).

Figure 5

Percentage of T-cell receptor (TCR) sequences from autoimmune myocarditis patients that mimic 42 common human bacteria compared with the percentage mimicked by TCRs from healthy individuals. Asterisks (*) indicate that the difference is statistically significant after Bonferroni correction (see Appendix A, Table A3).

Figure 6

Significant similarities between the TCR DIA A6 sequence and the antigens listed in Table 1 above. Underlined numbers are the UniProtKB identifiers.

Figure 7

Significant similarities between the TCR DIA 4 sequence and the antigens listed in Table 1 above. Underlined numbers are the UniProtKB identifiers.

Figure 8

Similarities between Clostridium perfringens (C.perf.) proteins (UniProtKB identifiers are underlined) and the human insulin A chain (INS A), using LALIGN.

Figure 9

Similarities between Clostridium perfringens (C.perf.) proteins (UniProtKB identifiers are underlined) and the human insulin B chain (INS B), using LALIGN.

Figure 10

Similarities between coxsackievirus B3 (Cox B3) and the human insulin receptor (INSREC). UniProtKB identifiers are underlined.

Figure 11

Significant similarities between the TCR Heart 2.4 sequence and the antigens listed in Table 2 above. Underlined numbers are the UniProtKB identifiers.

Figure 12

Significant similarities between the TCR Heart 7.8 sequence and the antigens listed in Table 2 above. Underlined numbers are the UniProtKB identifiers.

Figure 13

Sequence similarities between coxsackievirus B3 and human dynactin revealed by LALIGN. UniProtKB identifiers are underlined.

Figure 14

Summary of mimicry and complementarity relationships predicted for T-cell receptors (TCRs) by the molecular mimicry theory.

Figure 15

Summary of mimicry and complementarity relationships predicted for T-cell receptors (TCRs) by the anti-idiotype theory.

Figure 16

Summary of mimicry and complementarity relationships predicted for T-cell receptors (TCRs) by the complementary antigen theory.

Figure 17

Summary of mimicry and complementarity relationships predicted for T-cell receptors (TCRs) by the antigen template theory in the case of complementary antigens. Note that the antigen template theory does not resolve any of the problems applying the molecular mimicry or anti-idiotype theories to the TCR mimicry data presented in the Results here.