Association with HLA DQ of early onset myasthenia gravis in Southeast Texas region of the United States.

Forty-four Caucasian American myasthenia gravis (MG) patients from Southeast Texas underwent high resolution HLA DQ analysis. For the majority of patients who were late onset or male, no significant associations with DQ were observed. However, associations with DQ increased in female patients and early onset patients. At the allele level, DQB1 *0503, *0604, *0502 and *0402 collectively contributed to a positive association of the DQ locus with early onset MG (EOMG), while individually failing to show significant association. At DQ level, the novel haplotype DQA1*0401:DQB1*0201 was the primary factor in the association of combined DQ loci with early onset. In addition, *0104:*0503, *0102:*0604, *0102:*0502 and *0303:*0402 collectively contributed to the positive association of the haplotype loci. DR3-DQ2.5cis, a well known risk factor for MG in Western Eurasia, was not found associated with disease in any group. For typical EOMG [early onset, no thymoma, anti-acetylcholine receptor (AChR) antibody (Ab) positive] no association with DQA1 locus was found, however DQB1*0604 demonstrated an 'uncorrected' positive association. A few DQ haplotype (DQA1:DQB1) were positively associated with typical EOMG; a positive individual association for *0401:*0201 was complimented by the contributions of *0102:*0604 and *0303:*0402 haplotypes. A small minority of patients that were atypical and EOMG had a strong genetic association with DQA1*0104:DQB1*0503, the group included an anti-MuSK Ab positive and an anti-AChR negative patient. This report finds common ground with European studies regarding MuSK association; however similarities in association for typical early onset disease resembled HLA risk factors in East Asia and Southern Europe.

On the initial trigger of myasthenia gravis and suppression of the disease by antibodies against the MHC peptide region involved in the presentation of a pathogenic T-cell epitope.

Myasthenia gravis (MG) is a disabling autoimmune disease caused by autoantibodies (auto-Abs) against the self-acetylcholine receptor (AChR). Although a great deal of information is known about the molecular and cellular parameters of the disease, its initial trigger, however, is not known. To study the possibility of the involvement of microbial antigens that mimic AChR in triggering MG, we have searched the microbial proteins in the data bank for regions that are similar in structure to the regions of human (h) AChR alpha chain recognized by auto-Abs in MG patients. Hundreds of candidate structures on a large number of bacterial and viral proteins were identified. To test the feasibility of the idea, we synthesized four microbial regions similar to each of the major autodeterminants of hAChR (alpha12-27, alpha111-126, alpha122-138, alpha182-198) and investigated their ability to bind auto-Abs in MG and normal sera controls. It was found that MG sera potentially recognized a significant number of these microbial regions. The results indicate that in some MG cases, immune responses to microbial antigens may cross-react with self-antigen (in this case hAChR) and could constitute initial triggers of the disease. Although anti-AChR Abs directly contribute to the degradation of AChR at the neuromuscular junctions, autoreactive T cells provide help to B cells that synthesize anti-AChR auto-Abs. To cause MG, T cells must recognize the pathogenic epitopes in the context of MHC class II molecules related to MG. The ability to regulate AChR presentation (hence AChR-reactive T-cell activation) could form the basis of an effective strategy for the control of autoimmunity in MG by selectively inhibiting the function of the Ir gene loci linked to disease susceptibility. An animal model of MG (experimental autoimmune MG, EAMG) can be induced in C57BL/6 (B6, H-2b) mice by immunization with Torpedo californica (t) AChR. A mutant mouse of B6, B6.C-H-2bm12 (bm12), which has three amino acid changes (at residues 67, 70, and 71) in the I-A beta(b) subunit, is resistant to EAMG development. Recently, we showed that region 62-76 of I-A beta(b), which contains the above residues, is involved in the binding to a pathogenic T-cell epitope within peptide t alpha146-162. We have prepared several monoclonal antibodies (mAbs) against peptide I-A beta(b)62-76, which are highly cross-reactive with I-A(b) molecules. These mAbs inhibited in vitro the proliferation of disease-related T cells of B6 specific to tAChR peptide t alpha146-162. Passive transfer of these mAbs suppressed the occurrence of clinical EAMG, which was accompanied by lower T-cell and Ab responses to tAChR. The results indicated that blocking disease-related MHC by targeting a disease-associated region on MHC molecules could be an effective, straightforward, and feasible strategy for immunointervention in MG.

Suppression of experimental myasthenia gravis by monoclonal antibodies against MHC peptide region involved in presentation of a pathogenic T-cell epitope.

We have prepared monoclonal antibodies (mAbs) against an antigen-binding region of I-A, region 62-76 of I-Abeta(b), which is involved in the T-cell participation in the pathogenesis of EAMG. The mAbs reacted with its parent molecules and inhibited the proliferation of disease-related T-cells. Passive transfer of these mAbs suppressed the occurrence of clinical EAMG, which was accompanied by decreased T-cell and Ab responses to tAChR. The results indicated that blocking the function of disease-related MHC by targeting a disease-associated region on MHC molecules could be an effective, straightforward and feasible strategy for immunointervention in MG.

Antigen mimicry in autoimmune disease. Can immune responses to microbial antigens that mimic acetylcholine receptor act as initial triggers of Myasthenia gravis?

Myasthenia gravis (MG) is an autoimmune disease caused by autoantibodies against self acetylcholine receptor (AChR). Although a great deal of information is known about the molecular and cellular parameters of the disease, its initial trigger is not known. In order to study the possibility of the involvement of microbial antigens that mimic AChR in triggering MG, we have searched the microbial proteins in the data bank for regions that are similar in structure to the regions of human (h) AChR alpha chain recognized by autoAbs in MG patients. Hundreds of candidate structures on a large number of bacterial and viral proteins were identified. To test the feasibility of the idea, we synthesized four microbial regions similar to each of the major autodeterminants of hAChR (alpha12-27, alpha111-126, alpha122-138, alpha182-200) and investigated their ability to bind autoAbs in MG and normal sera controls. It was found that MG sera recognized a significant number of these microbial regions. The results indicate that in some MG cases immune responses to microbial antigens may cross-react with self antigen (in this case hAChR) and could constitute initial triggers of the disease.

T cell responses in EAMG-susceptible and non-susceptible mouse strains after immunization with overlapping peptides encompassing the extracellular part of Torpedo californica acetylcholine receptor alpha chain. Implication to role in myasthenia gravis of autoimmune T-cell responses against receptor degradation products.

To study the role in myasthenia gravis (MG) of peptides resulting from acetylcholine receptor (AChR) degradation, we examined the ability of AChR peptides to induce T cell responses that are capable of cross-reacting with intact AChR. The studies were carried out in an experimental autoimmune MG (EAMG)-susceptible mouse strain [C57BL/6 (B6)] as well as in two non-susceptible strains [B6.C-H-2bm12 (bm12) and C3H/He]. A set of overlapping peptides encompassing the extracellular part (residues 1-210) of the alpha-chain of Torpedo californica (t) AChR were used, individually or in equimolar mixtures, as immunogens. In B6, immunization with peptides alpha45-60, alpha111-126, alpha146-162 and alpha182-198 gave T cells that responded in vitro to the correlate immunizing peptide. Only the T cells against the latter three peptides cross-reacted with tAChR. Peptide alpha146-162 exhibited the highest in vitro reaction with the immunizing peptide and cross-reaction with tAChR. T cells obtained by immunization of B6 with an equimolar mixture of the peptides responded in vitro to peptides alpha111-126, alpha146-162 and alpha182-198 and cross-reacted very strongly with tAChR. In bm12 and C3H/He, a number of peptides evoked, when used individually as immunogens, strong or moderate T cell responses that recognized in vitro the correlate immunizing peptide but cross-reacted poorly with tAChR. Immunization with the mixture of the peptides gave T cells that recognized several peptides in each strain butdid not cross-react with alpha146-162 or tAChR. The results indicate that the ability to recognize alpha146-162 or AChR by T cells against peptides resulting from receptor degradation can account for the susceptibility to, and aggravation of, MG in B6.

Preliminary three-dimensional model for nematode thick filament core.

Understanding the structure and the mechanism of assembly of thick filaments have been long-standing problems in the field of muscle biology. Cores which represent the backbones of thick filaments and consist of paramyosin and associated proteins were isolated from the nematode Caenorhabditis elegans. Electron microscopy of negatively stained and frozen hydrated cores was performed. The resulting images were analyzed by computing their Fourier transforms, three-dimensional reconstruction, and by modeling. A preliminary three-dimensional model is proposed in which the paramyosin constitutes an outer sheath of seven subfilaments about a set of inner 54-nm-long tubules which repeat every 72 nm. The subfilaments are not closely packed but require cross-linking by the internal tubules. Each subfilament consists of two strands of paramyosin molecules which are staggered by 72 nm with respect to one another. This stagger introduces a 22-nm gap between consecutive paramyosin molecules in each strand. An offset of the center of the inner tubules relative to the center of the gap of 6 nm was consistent with the images and their transforms. This model suggests that the nonhelical ends of paramyosin and the unpaired gap between adjacent paramyosin molecules contain sites for the interaction with the inner tubular proteins. The molecular interactions at this locus would appear to be critical in the assembly of thick filaments and their regulation.

Thick filament substructures in Caenorhabditis elegans: evidence for two populations of paramyosin.

The thick filaments of the nematode Caenorhabditis elegans contain two myosin heavy chain isoforms A and B and paramyosin, the products of the myo-3, unc-54, and unc-15 genes, respectively. Dissociation of paramyosin from native thick filaments at pH 6.36 shows a biphasic function with respect to NaCl concentration. Electron microscopy of the remaining structures shows 15-nm core structures that label with monoclonal anti-paramyosin antibody at 72.5-nm intervals. Purified core structures also show 72.5 nm repeats by negative staining. Structural analysis of native thick filaments and dissociated structures suggests that the more dissociable paramyosin is removed radially as well as processively from the filament ends. Minor proteins with masses of 20, 28, and 30 kD cosediment stoichiometrically with paramyosin in purified core structures.

Accumulation and assembly of myosin in hypertrophic cardiomyopathy with the 403 Arg to Gln beta-myosin heavy chain mutation.

The sarcomeric proteins and organization of cardiac myofibrils appeared intact in multiple unrelated patients with hypertrophic cardiomyopathy. In two subjects demonstrating the missense mutation at position 403 (Arg to Gln) in the beta-myosin heavy chain gene, total myosin and immunoreactive beta-myosin heavy chain levels were similar to those found in other patients with hypertrophic cardiomyopathy and various disease control subjects. No alteration in expression of the cardiac alpha-myosin heavy chain gene was observed. These results are consistent with the examined myosin heavy chain mutation, permitting proper accumulation and assembly of myosin while primarily impairing contractile function. The characteristic myocyte disarray would appear likely to be a secondary consequence of the mutations.

Assembly-dependent phosphorylation of myosin and paramyosin of native thick filaments in Caenorhabditis elegans.

Phosphorylation of the thick filament proteins myosin and paramyosin was studied in Caenorhabditis elegans. We have incubated partially purified, native thick filaments with [gamma 32P] ATP in the presence of 50-750 mM NaCl, pH 6.5-8.0. Myosin heavy chain and paramyosin were phosphorylatable only upon solubilization at 450 mM and higher NaCl concentrations. Under conditions preserving native structures, no phosphorylation of these proteins occurred. The phosphorylation required Mg2+ but was unaffected by cAMP, cGMP or Ca2+. The specific inhibitor of cAMP and cGMP kinase catalytic subunits, H8, inhibits the activity. Sedimentation experiments show that the kinase may associate with but is not an intrinsic component of thick filaments. In C. elegans, phosphorylation by the thick filament associated activity of myosin and paramyosin is dependent upon the state of their assembly.