B. Lang

Antibodies to motor nerve terminals: an electrophysiological study of a human myasthenic syndrome transferred to mouse.

Immunoglobulin G(IgG) prepared from the plasma of patients with a presynaptic disorder of neuromuscular transmission (Lambert‐Eaton myasthenic syndrome, l.e.m.s.), or from normal pooled control human plasma, was injected into mice (10 mg daily) for up to 99 days. Micro‐electrodes were used to record end‐plate potentials from the diaphragm muscle bathed in normal Krebs solution containing tubocurarine (1.0‐4.6 microM). At 0.5 Hz nerve stimulation frequency, the quantal content was significantly reduced (P less than 0.01‐P less than 0.001) in mice treated with six l.e.m.s. patients’ IgG each compared with paired controls. The pooled quantal content was 55 +/‐ 3 (n = 110 end‐plates) for all test animals and 131 +/‐ 9 (n = 47) for all controls (P less than 0.001). During short trains at 20 or 40 Hz nerve stimulation, control muscles showed marked depression, while test muscles showed either facilitation or less marked depression. Quantal content throughout these trains remained lower than in controls. The results indicate that IgG antibody from l.e.m.s. patients can induce a similar physiologic disorder in injected mice, and they support the view that this antibody interferes with evoked release of transmitter in l.e.m.s. by binding to nerve terminal determinants.

Antibody heterogeneity and specificity in myasthenia gravis.

Anti-AChR is heterogeneous within individuals and between individuals. Anti-AChR idiotypes are not shared to any large extent. Ten monoclonal antibodies raised against human AChR: (a) bind to five partially overlapping regions; (b) are not idiotypically identical even within a region; (c) do not all bind to the main immunogenic region; (d) four distinguish between normal and denervated human AChR; (e) can be used to define the antigenic determinants in MG. Antigenic specificities vary in different clinical groups. Antigenic specificities can change during the course of the disease, but some remain relatively constant. Thymus cultures make antibodies with the same specificity as those present in the serum of the individual. All monoclonal antibodies bind to myoid cells of normal and MG thymus. We find no convincing evidence of naturally occurring antiidiotype antibodies in MG sera.

Selective action of myasthenic syndrome antibodies on calcium channels in a rodent neuroblastoma x glioma cell line.

1. The effect of Lambert‐Eaton myasthenic syndrome (LEMS) immunoglobulin G (IgG) on Ca2+ channels in undifferentiated mouse neuroblastoma x rat glioma hybrid cells (NG 108 15) was studied using the whole‐cell patch clamp technique. 2. Sustained inward Ca2+ channel currents were evoked by depolarizing pulses from holding potentials of ‐80 and ‐40 mV, and were blocked by 5 microM‐nitrendipine (L‐type currents). Transient inward Ca2+ channel currents were activated from a holding potential of ‐80 mV by small depolarizing steps (T‐type currents). Noradrenaline (10 microM) was without effect on transient currents. 3. LEMS IgG selectively reduced sustained (L‐type) Ca2+ channel current amplitudes evoked from either holding potential used. In the presence of nitrendipine (5 microM), there was no significant effect of LEMS IgG on the remaining transient (T‐type) Ca2+ channel current amplitudes. 4. Studies of the potential for maximal inward current indicated that voltage sensitivities of both L‐ and T‐type Ca2+ channel current amplitudes were unaffected by LEMS IgG, whether recorded in the presence or absence of nitrendipine. LEMS IgG had no significant effect on the time‐to‐peak or decay of Ca2+ channel currents. 5. It is concluded that LEMS IgG acts selectively to cause functional loss of L‐type, but not T‐type, Ca2+ channels in NG 108 15 cells. Any effect of LEMS IgG on N‐type channels (not present in these undifferentiated cells) was not studied here. LEMS IgG also acts at motor nerve terminal Ca2+ channels leading to muscle weakness. Thus antigenic similarities must exist between L‐type channels in NG 108 15 cells and Ca2+ channels at motor nerve terminals.

Passive Transfer of Lambert-Eaton Myasthenic Syndrome in Mice: Decreased Rates of Resting and Evoked Release of Acetylcholine from Skeletal Muscle

Abstract: Mice were injected for 1–2 months daily with 10 mg immunoglobulin G (IgG) from four patients with Lambert‐Eaton myasthenic syndrome (LEMS); control mice were injected with pooled human IgG from normal donors. Gastrocnemius muscles were homogenised for the assay of acetylcholine (ACh), choline acetyltrans‐ferase (ChAT), and cholinesterase (ChE). The ACh, ChAT, and ChE contents of gastrocnemius muscles from “LEMS mice” were about the same as the control values, which were 180 pmol, 40 nmol * h−1 (37°C), and 15 μmol * h−1 (37°C), respectively. Hemidiaphragms were treated with an irreversible ChE inhibitor (Soman) and incubated at 20°C for estimation of ACh release. Resting ACh release from experimental muscles was reduced by about 25% (P2 < 0.05) and the release evoked by 3 s−1 nervous stimulation by 50% (P2 < 0.05). On the other hand, 50 mM KCl‐induced transmitter release was not abnormal in LEMS mice. The findings indicate that IgG antibody from patients with LEMS may bind to nerve terminal determinants that are involved in quantal and nonquantal ACh release.

Lambert-Eaton syndrome antibodies: reaction with membranes from a small cell lung cancer xenograft

Lambert-Eaton myasthenic syndrome (LEMS) is a paraneoplastic autoimmune disorder caused by an IgG-mediated reduction in number of presynaptic voltage-gated calcium channels (VGCC) at the neuromuscular junction. In at least 50% of cases, the stimulus for antibody production may be VGCC on small cell lung cancer (SCLC). In this study membranes isolated from a human small cell lung cancer xenograft (Mar), that bound [3H]PN200-110, a VGCC antagonist, were subjected to Western blotting using plasma from 12 LEMS patients and eight controls. Although one band recognised by 3/12 LEMS IgGs might be associated with the VGCC, a number of other proteins were recognised both by LEMS plasma, and by plasma from patients with other disorders. The results illustrate the difficulties found using Western blotting with autoimmune plasma to identify specific polypeptides in a crude antigen preparation.

Site of action of Lambert-Eaton myasthenic syndrome antibodies at mouse nerve terminals

The Lambert-Eaton myasthenic syndrome (LEMS) is a disorder characterised by a marked reduction in the nerve evoked release of transmitter at the skeletal neuromuscular junction (Lamber & Elmqvist, 1971). The autoimmune nature of this disorder has been recently shown (Lang et al., 1981, 1983). The IgG fraction of patients’ plasma was injected into mice and this transferred the disorder of transmitter release to mice. In the experiments reported here, the evidence for this passive transfer is summarized, and preliminary results concerning the site of action of these antibodies are presented.

Expression of voltage-gated calcium channels in tumour cell lines of neuroectodermal or other origin

Although most normal neural ce l l s lack MHC products, greater expression can be seen in pathological or experimental s i tuat ions (reviewed in Lampson, Trends in NeuroSci, May, 1987). Whether these molecules serve an immunological ro le , or a non-immunological ro le in ce l l d i f f e r e n t i a t i o n or homeostasis is not known. We have focused on class I molecules. We confirm that class I proteins wi th the appropriate mRNA, structure ( I and 2-D gels) , and polymorphic s p e c i f i c i t y can be produced by human neural ce l l l ines. Yet, in v ivo , class I expression is not detected on neurons or g l i a l ce l l s in the adu l t , in any developing neural t issue in the mouse embryo, nor in the regenerating o l fac to ry epi thel ium. Nor is class I detected on neural ce l l s normally exposed to blood-borne elements (ba r r i e r f ree areas of normal brain, tumor metastases) or the external environment (o l fac to ry nerve endings), or fo l lowing trauma (stab wound). Thus, the accumulated evidence argues against a ro le fo r class I molecules in normal growth, d i f f e r e n t i a t i o n , maintenance or repair of neural ce l l s , and against class I modulation as a non-speci f ic response to in ju ry . These studies provide a background for in te rpre t ing the class I modulation that is seen in spec i f ic c l i n i c a l s i tua t ions.

ANTI‐ACETYLCHOLINE RECEPTOR ANTIBODY HETEROGENEITY IN DIFFERENT FORMS OF MYASTHENIA GRAVIS

In a study of 68 patients with myasthenia gravis (MG) we distinguished three main groups of patients by their age of onset and thymic pathology, which were associated with different sex incidence, HLA antigens and anti-acetylcholine receptor (anti-AChR) levels (Compston et al. 1980, Brain 103: 579). In addition those patients with symptoms restricted to extraocular muscles had low anti-AChR titers but reacted proportionately better with ocular muscle AChR. We therefore analyzed anti-AChR characteristics in these four groups, and compared them with those from patients with penicillamine-associated MG. The 54 patients used included 12 with thymoma (HLA-B7, n = 5; -B8, n = l ) , 11 with onset before 40 years (-B7, n = 0 ; -B8, n = lo), 12 with onset after 40 years (-B7, n = 9; -B8, n = 2), seven with pure ocular disease of more than two years’ duration, and 11 with penicillamine-associated MG whose anti-AChR fell fairly rapidly when the drug was withdrawn. The earliest available sample was tested. The duration of the disease at that time was variable (<1-40 years) hut not significantly different between the groups except for the penicillamine-associated patients in whom it was less than 2 m. Approximately half the patients were on prednisone. Anti-AChR antibodies were measured using denervated human leg muscle AChR and tested for: percentage of kappa and lambda light chains, IgG subclass and anti-AChR “affinity” (i.e., the reciprocal of the relative increase in titer when 4x the AChR concentration was used in the assay). Sera were tested for inhibition of a-hungarotoxin binding to denervated AChR and reactivity with AChR from normal human leg and extraocular muscle, normal and denervated rat muscle, and mouse cell line BC3Hl. Our results confirm the polyclonality and heterogeneity of anti-AChR antibody and the variation in anti-AChR characteristics between different patients, but we have not demonstrated any significant differences between the characteristics of anti-AChR in the three main groups of patients. This suggests that different factors precipitating loss of tolerance to AChR rather than the nature of the autoimmune response itself may he related to the different age of onset, sex distribution and HLA associations in these groups. The results from the 35 patients with acquired generalized MG were therefore pooled and compared with those from the ocular and pencillamineassociated groups. The main results are given in the table. Anti-AChR in penicillamine associated MG showed some distinctive features, notably a significantly lower affinity and lower proportion of kappa light chains. However, penicillamine-associated anti-AChR is presumably of recent origin and it will be important to test antibodies from patients of equally recent onset in order to confirm that anti-AChR characteristics are different (TABLE 1).