Antibody-mediated gastrointestinal dysmotility in scleroderma

GASTROENTEROLOGY 2002;123:1144 –1150

Antibody-Mediated Gastrointestinal Dysmotility in Scleroderma FIONA GOLDBLATT, TOM P. GORDON, and SALLY A. WATERMAN Department of Immunology, Allergy and Arthritis, Flinders Medical Centre, Bedford Park, South Australia

Background & Aims: Defects in enteric excitatory neurotransmission have been proposed to underlie the gastrointestinal dysmotility associated with scleroderma (systemic sclerosis). This study investigated whether patients with scleroderma produce antibodies that inhibit M3-muscarinic or neurokinin receptor–mediated intestinal contractions, either directly or via an effect on L-type voltage-gated calcium channels (VGCCs). Methods: Responses of mouse colon longitudinal muscle to stimulation by the muscarinic agonist carbachol (1–300 ␮mol/L) and neurokinin-1 and -2 receptor agonists were measured in the absence and presence of serum (2%) or immunoglobulin G (IgG) (0.3–1.0 mg/mL) from patients with scleroderma, those with other autoimmune disorders, and healthy controls. The role of L-type VGCCs in carbachol- and tachykinin-evoked contractions was assessed using nicardipine. Results: M3-muscarinic receptor–mediated contractions were inhibited by Ig fractions from 7 of 9 patients with scleroderma (limited and diffuse forms), 4 of 4 patients with primary Sjo ¨gren’s syndrome, and 3 of 3 patients with secondary Sjo ¨gren’s syndrome. Ig fractions from healthy controls did not inhibit the M3-muscarinic receptor–mediated contractions. Inhibition by Ig was concentration-dependent; a maximum inhibition of approximately 40% occurred at 0.6 mg/mL IgG. Both M3-muscarinic and neurokinin receptor–mediated contractions were L-type VGCC dependent. Patient sera had no effect on responses to neurokinin receptor stimulation, demonstrating the lack of antibodies inhibiting L-type VGCCs. Conclusions: Functional antibodies specifically inhibiting M3-muscarinic receptor–mediated enteric cholinergic neurotransmission may provide a pathogenic mechanism for the gastrointestinal dysfunction seen in patients with scleroderma.

astrointestinal dysmotility occurs frequently in patients with the systemic autoimmune disease scleroderma (systemic sclerosis), but the underlying pathogenesis remains uncertain. Esophageal motility is affected in approximately 90% of patients; functional problems of the stomach, small intestine, colon, and anorectum may also occur and cause significant morbidity.1 Manometric studies suggest the initial pathologic event is neuropathic, with histologic changes of smooth

G

muscle atrophy and fibrosis occurring later in the disease.2–5 The neuropathic disturbance possibly affects cholinergic neurotransmission, although effects on other excitatory neurotransmitters such as tachykinins have not been studied. It has been proposed that the gastrointestinal dysmotility in scleroderma may be autoantibody mediated. Myenteric neuronal antibodies, distinct from anti-Hu antibodies, have been reported by indirect immunofluorescence in a subset of patients with gastrointestinal dysmotility and scleroderma.6 However, the target autoantigen remains undefined and the mechanism by which the autoantibodies functionally alter gastrointestinal motility unresolved. Contractile activity of the gastrointestinal tract is controlled predominantly by intrinsic neurons in the myenteric plexus. Acetylcholine, acting predominantly via the M3-muscarinic receptor, and tachykinins, acting at neurokinin (NK) receptors, are the principal excitatory neurotransmitters regulating motility.7 Both cholinergic and tachykininergic responses depend on calcium entry into the smooth muscle via L-type voltage-gated calcium channels (VGCCs).8 Antibodies that block M3-muscarinic receptors, NK receptors, or L-type VGCCs would therefore be expected to inhibit excitatory enteric neurotransmission and cause dysmotility. Whether such antibodies are present in the serum of patients with scleroderma and contribute to their symptoms remains to be investigated. There is a precedent in rheumatologic diseases for antibody-mediated disruption of autonomic neurotransmission. In primary Sjo¨gren’s syndrome, antibodies to the G-protein– coupled M3-muscarinic receptor have been detected by both radioligand-binding studies9,10 and functional assays on bladder smooth muscle.11 Muscarinic receptors of the M3 subtype are widespread, including in lacrimal and salivary glands and bladder detrusor and intestinal smooth muscle, and mediate the effects of acetylcholine in these tissues. Antibodies acting Abbreviations used in this paper: NK, neurokinin; RA, rheumatoid arthritis; RF, rheumatoid factor; VGCC, voltage-gated calcium channel. © 2002 by the American Gastroenterological Association 0016-5085/02/$35.00 doi:10.1053/gast.2002.36057

October 2002

as antagonists at M3-muscarinic receptors may contribute to sicca symptoms and explain the autonomic dysfunction in some patients with primary and secondary Sjo¨gren’s syndrome.11 Interestingly, antibodies to the M2-muscarinic receptor have also been described in patients with achalasia associated with Chagas’ disease.12 Using a functional colon assay, we tested the hypothesis that patients with scleroderma produce antibodies that inhibit M3-muscarinic- or NK-receptor–mediated neurotransmission of intestinal smooth muscle, either directly or via an effect on L-type VGCCs, thereby contributing to the gastrointestinal features seen in this disease.

Materials and Methods Patients and Controls Sera and clinical data were obtained with informed consent from eight healthy individuals and eleven patients with scleroderma who were attending rheumatology clinics (8 with limited disease and 3 with diffuse disease, as defined by the criteria proposed by the American College of Rheumatology).13 Nine of the scleroderma patients had oral and ocular sicca with a positive Schirmer’s test. Sera were also obtained from 8 patients with at least 4 of the 6 European consensus criteria for the diagnosis of primary Sjo¨gren’s syndrome, including presence of anti-Ro or anti-La autoantibodies,14 and from 9 patients with rheumatoid arthritis (RA) who fulfilled the criteria of the American Rheumatism Association.15 Of the patients with RA, 6 had secondary Sjo¨gren’s syndrome (xerostomia and xerophthalmia with a positive Schirmer’s test). Patients were questioned regarding symptoms of gastrointestinal dysfunction, including dysphagia, heartburn, constipation, diarrhea, postprandial fullness, and fecal incontinence. Serum immunoglobulin G (IgG) and rheumatoid factor (RF) were measured by nephelometry (ICS II; Beckman, Brea, CA). Antinuclear antibodies were detected by indirect immunofluorescence on Hep-2000 cells (Immunoconcepts, Sacramento, CA), and extractable nuclear antigens were measured by counterimmunoelectrophoresis with rabbit thymus extract (Pel-Freez, Rogers, AR) and extracts of K562 (a human myeloid cell line).16 The Ig fraction was purified from serum using the caprylic acid precipitation technique.17

Functional Assay on Colon Smooth Muscle Male Balb/c mice (age, 8 –10 weeks; average weight, 25.6 g) were killed by CO2 inhalation followed by cervical dislocation. The colon was quickly removed and placed in a Petri dish containing Krebs solution. The composition of the Krebs solution was 119 mmol/L NaCl, 4.7 mmol/L KCl, 1.2 mmol/L KH2PO4, 25 mmol/L NaHCO3, 1.5 mmol/L MgSO4, 11.0 mmol/L D-glucose, and 2.5 mmol/L CaCl2. Fecal material was removed by gently flushing the bowel with a Krebs solution–filled syringe. Attached mesentery and connective tissue were removed from the colon; then the colon was

ANTIBODY–MEDIATED DYSMOTILITY

1145

dissected into 8 pieces approximately 0.5– 0.8 mm long. These were mounted longitudinally in 10-mL jacketed organ baths containing Krebs solution at 37°C and gassed with 95% O2/5% CO2. Preparations were desensitized to capsaicin (10 ␮mol/L; Sigma, St. Louis, MO) at the beginning of each experiment to inactivate sensory neurons. Hexamethonium was added to block nicotinic ganglionic neurotransmission (100 ␮mol/L; Sigma) and guanethidine, to block neurotransmitter release from sympathetic neurons (0.3 ␮mol/L; Sigma). The prostaglandin inhibitor indomethacin (10 ␮mol/L; ICN Biomedicals, Aurora, OH) and the nitric oxide synthase inhibitor NG-nitro-L-arginine (L-NOArg) (100 ␮mol/L; Novabiochem, Beeston, England) were present throughout all experiments. Contractile responses were measured using Kent isometric transducers connected to a PowerLab/8s data acquisition system (AD Instruments, Sydney, Australia). Tissues were left to equilibrate for 60 minutes before the experiment was started. Three protocols were used. To assess the effect of serum on M3-muscarinic receptor–mediated neurotransmission, concentration–response curves to carbachol were produced by the cumulative addition of 1, 3, 10, 30, 100, and 300 ␮mol/L of carbachol at 60-second intervals. Carbachol concentration– response curves were repeated after a 30-minute incubation with whole serum (final concentration of 2%) or IgG fractions (at concentrations of 0.3, 0.6, and 1.0 mg/mL) from patients or controls. To determine the effect of time on contraction amplitude, 1 preparation in each experiment was tested without the addition of serum. There were no time-dependent changes in colon contractions. In agreement with earlier studies, carbachol-evoked colon contractions were mediated predominantly by M3-muscarinic receptors, with contractions inhibited by 1.0 ␮mol/L of the M3-receptor–specific antagonist 1,1-dimethyl-4diphenylacetoxypiperidinium iodide (4-DAMP; Tocris, Bristol, UK) (data not shown). In a second series of experiments, the effect of serum on NK receptor–mediated responses was tested. First, 3 ␮mol/L of either the NK-1 receptor agonist [Sar9, Met (O2)11]-Substance P (RBI/Sigma, Natick, MA) or NK-2 receptor agonist [␤-Ala8]-Neurokinin A (4-10) (RBI/Sigma) were added to the organ baths. Extensive washing to prevent desensitization was performed once the peak contraction was reached. Thirty minutes after the addition of 2% whole serum, responses to [Sar9, Met (O2)11]-Substance P and [␤-Ala8]-Neurokinin A (4-10) were again measured. To control for nonspecific effects of increased protein concentration, 1 preparation per experiment was tested with 2% bovine serum albumin (BSA) (Sigma). Responses decreased by 25% in response to BSA alone, and NK-1 and NK-2 responses were therefore corrected by this factor. To confirm the role of L-type VGCCs in contraction of colon smooth muscle, carbachol- and tachykinin-evoked contractions were measured before and after the addition of the L-type VGCC antagonist nicardipine (1 ␮mol/L; Sigma). Balb/c mice were used in experimental protocols approved by the Flinders University Committee for Animal Welfare.

1146

GOLDBLATT ET AL.

GASTROENTEROLOGY Vol. 123, No. 4

Human ethics approval was obtained from the Clinical Investigations (Ethics) Committee at Flinders Medical Centre.

Data Analysis Maximum contraction amplitudes were measured using Chart version 3.6.1 software (ADInstruments, Sydney, Australia) on a PowerMac computer (Apple Computers, Sydney, Australia). Responses to carbachol and tachykinins after the addition of whole serum, immunoglobulin fractions, or nicardipine were expressed as a percentage of the maximal response to the agonist in that preparation. Statistical analysis comparing the concentration–response curves in the absence and presence of serum/Ig was performed using 2-way analysis of variance (GraphPad Prism; GraphPad Software, San Diego, CA); paired Student t tests were used to compare amplitudes of tachykinin-evoked contractions in the absence and presence of serum. P values ⬍0.05 were considered significant.

Results Inhibition of M3-Muscarinic Receptor Agonist–Evoked Colon Smooth Muscle Contraction by Patient Serum In initial experiments, sera from 11 of 11 patients with scleroderma, 4 of 8 patients with primary Sjo¨gren’s syndrome, 5 of 6 patients with RA and secondary Sjo¨gren’s syndrome, 1 of 3 patients with RA alone, and 3 of 8 normal controls significantly inhibited carbacholevoked colon longitudinal smooth muscle contraction (n ⫽ 4 experiments per serum; P ⬍ 0.0001– 0.03). Contractions were inhibited by approximately 50%. Anti–M3-Muscarinic Receptor Activity is in the Immunoglobulin Fraction To determine whether inhibition of M3-muscarinic receptor–mediated contractions was due to a specific effect of antimuscarinic receptor antibodies, and not to nonspecific serum effects, we tested purified Ig on the functional assay. Ig fractions were prepared from all sera that had significantly inhibited carbachol-evoked colon contractions (except for 2 scleroderma and 2 RA and secondary Sjo¨gren’s syndrome patients, from whom additional sera were unavailable for Ig isolation). Ig was added to give final IgG concentrations of 0.3, 0.6, or 1.0 mg/mL. Specific inhibitory effects of Ig were observed in 7 of the 9 patients with scleroderma and in all of the patients with primary Sjo¨gren’s syndrome or RA and secondary Sjo¨gren’s syndrome who were positive in the initial assay (Figure 1). Contractile responses were inhibited in a concentration-dependent manner, with a maximum inhibition of approximately 40% occurring at 0.6 mg/mL total IgG (Figure 2A, B, and D). The

Figure 1. Raw traces showing contractile responses to cumulative addition of 1 –300 ␮mol/L carbachol in untreated tissue (upper trace) and after the addition of 1.0 mg/mL IgG from a patient with scleroderma (lower trace).

specific autoantibody would represent only a small percentage of this total IgG. In this preliminary study there was no relationship between inhibition of carbachol-induced contractions and hypergammaglobulinemia, RF levels, or the presence of anti-Ro/La, anticentromere and antinucleolar autoantibodies. Antimuscarinic receptor activity was not detected in Ig from the patient with RA alone or from healthy controls at IgG concentrations up to 1 mg/mL (Figure 2C ). From these data, we conclude that only patients with scleroderma or Sjo¨gren’s syndrome have specific antibodies blocking M3-muscarinic receptor–mediated enteric neurotransmission. We also conclude that the inhibitory effects seen in normal sera are secondary to nonspecific serum effects. Sera From Patients With Scleroderma Do Not Block Neurokinin Receptors To investigate whether patient antibodies also inhibit neurotransmission mediated by the other principal excitatory transmitters in the colon, effects on tachykinin-evoked colon contraction were assessed using patient sera, which inhibited the muscarinic receptor– mediated contractions, and normal control sera, which had not affected carbachol-induced contractions. Neither the NK-1 agonist [Sar9, Met (O2)11]-Substance P– nor the NK-2 agonist [␤-Ala8]-Neurokinin A (4-10)– evoked contractions were affected by the addition of 2% serum. Contractions evoked by 3 ␮mol/L [Sar9, Met (O2)11]-Substance P in the presence of serum were 108% ⫾10.0% of preserum values (n ⫽ 7; P ⬎ 0.05), and those evoked by 3 ␮mol/L [␤-Ala8]-Neurokinin A (4-10) in the presence of serum were 94% ⫾11.0% of preserum values (n ⫽ 7; P ⬎ 0.05). This suggests that the M3muscarinic receptor antibodies in patient serum do not cause nonspecific inhibition of smooth muscle contraction and, furthermore, that the patient serum did not contain functional antibodies to NK-1 or NK-2 receptors.

October 2002

ANTIBODY–MEDIATED DYSMOTILITY

1147

Figure 2. Responses to carbachol in the presence of Ig fraction from patients or healthy controls. Ig from a patient with scleroderma significantly inhibits carbachol-evoked responses at concentrations of (A) 0.3 mg/mL (P ⫽ 0.0002) and (B) 1 mg/mL (P ⬍ 0.0001). (C) IgG from a healthy control does not significantly alter carbachol-evoked responses (P ⬎ 0.05). (D) The maximum percentage inhibition of contractions evoked by 30 ␮mol/L carbachol with increasing concentrations of IgG from antibody-positive patients (n ⫽ 10) and healthy controls (n ⫽ 4) occurred at a concentration of 0.6 mg/mL.

Inhibition of Carbachol-Evoked Colon Contraction Is not Due to Anti–L-Type Voltage-Gated Calcium Channel Autoantibodies Carbachol responses were partially inhibited by 1 ␮mol/L of the L-type VGCC blocker nicardipine (n ⫽ 6, data not shown, P ⬍ 0.05), demonstrating that muscarinic responses in the mouse colon are coupled to calcium influx into smooth muscle cells via this channel. In addition, both the NK-1 agonist [Sar9, Met (O2)11]-Substance P– and NK-2 agonist [␤-Ala8]-Neurokinin A (4-10)–induced contractions were inhibited by 50%– 65% by 1 ␮mol/L nicardipine, confirming a role for L-type VGCCs in tachykinin-evoked colon contraction (n ⫽ 6, data not shown, P ⬍ 0.05). Antibodies to L-type VGCCs would thus be expected to inhibit both M3-muscarinic- and NKreceptor–mediated responses. Because only the carbachol responses, and not the tachykinin-evoked contractions, were inhibited by patient serum, we

conclude that functional autoantibodies to L-type VGCCs are not present in the serum of patients with scleroderma and Sjo¨gren’s syndrome and are not responsible for the immunoglobulin-induced inhibition of colon contraction detected in our study. Gastrointestinal Dysfunction in Patients With Antibodies Inhibiting M3-Muscarinic Receptor–Mediated Contraction The functional antibody was present in patients with both limited and diffuse forms of scleroderma. Not unexpectedly, all except 1 of the scleroderma patients screened in the present study had gastroesophageal reflux disease. All patients with scleroderma whose serum contained functional anti–M3-muscarinic receptor antibodies described additional gastrointestinal symptoms, including dysphagia, nausea, abdominal bloating, constipation, diarrhea, and fecal incontinence (Table 1). Interestingly, the 2 patients with scleroderma who were

1148

GOLDBLATT ET AL.

GASTROENTEROLOGY Vol. 123, No. 4

Table 1. Clinical and Serologic Characteristics of Patients With Scleroderma With and Without Functional Antibodies Inhibiting M3-Muscarinic Receptor-Mediated Enteric Neurotransmission Diagnosis

Functional anti–M3-muscarinic receptor antibody status

ANA/ENA

52/F

Limited scleroderma

⫹

Centromere/⫺

73/F

Limited scleroderma

⫹

Centromere/⫺

69/F

Limited scleroderma

⫹

⫺/⫺

74/F

Limited scleroderma

⫹

⫹/UPL

79/F

Diffuse scleroderma

⫹

Centromere/⫺

66/F

Diffuse scleroderma

⫹

⫹/Scl-70, Ro

60/F

Diffuse scleroderma

⫹

⫹/⫺

72/F 75/F

Limited scleroderma Limited scleroderma

⫺ ⫺

Centromere/⫺ Centromere/⫹

Age (yr)/sex

Gastrointestinal features GORD Nausea Abdominal bloating Constipation Sicca GORD Dysphagia Sicca GORD Dysphagia Barrett’s esophagus Esophageal stricture Diarrhea Fecal incontinence Rectal prolapse Sicca GORD Constipation Sicca GORD Dysphagia Esophageal dilatation Constipation Diarrhea Sicca GORD Dysphagia GORD Dysphagia Sicca Mild GORD No GI symptoms Sicca

ANA, antinuclear antibody; ENA, extractable nuclear antigen; UPL, unidentified protein line; GORD, gastroesophageal reflux disease.

antibody negative in our functional assay had only mild or no gastrointestinal symptoms. Gastrointestinal symptoms, including gastroesophageal reflux disease, dysphagia, and chronic constipation, were described by 3 of the 4 patients with primary Sjo¨gren’s syndrome and by 2 of the 3 patients with RA and secondary Sjo¨gren’s syndrome who were antibodypositive in our functional assay (data not shown). Dry eyes and/or dry mouth (sicca) were reported by all but 1 patient positive for the functional anti–M3-muscarinic receptor antibody. All healthy controls and other patients negative for the anti–M3-muscarinic receptor antibody had no gastrointestinal features.

Discussion We have demonstrated for the first time an antibody that specifically inhibits M3-muscarinic receptor mediated enteric neurotransmission in patients with scleroderma. We also detected the presence of

these antibodies in patients with primary Sjo¨gren’s syndrome and those with RA and secondary Sjo¨gren’s syndrome, but not RA alone, confirming our earlier work in a similar functional assay using bladder smooth muscle.11 Inhibition of colon contraction was specific for the cholinergic pathway, because serum did not affect tachykinin-evoked contractions and smooth muscle L-type VGCC– dependent contractions. We postulate that the antibody-mediated interference of M3-muscarinic receptor neurotransmission and subsequent inhibition of gastrointestinal tract smooth muscle contraction contributes to symptoms associated with esophageal dysmotility and lower gastrointestinal tract dysfunction experienced by scleroderma patients. Moreover, inhibition of cholinergic neurotransmission may contribute to the sicca symptoms (secondary Sjo¨gren’s syndrome) variably associated with scleroderma. In our functional assay in which the smooth muscle is stimulated nonphysi-

October 2002

ologically with receptor agonists, patient serum and Ig blocked cholinergic contractions by as much as 50%. Because the assay does not involve physiologic stimulation of the tissues, the degree of inhibition in vitro cannot be directly correlated with the degree of inhibition of intestinal motility that may occur in vivo. Previously described antimyenteric neuronal antibodies in scleroderma gave granular staining not only of the neuronal cell bodies, but also of the adjacent smooth muscle cells.6 The latter may be consistent with the staining pattern of autoantibody binding to M3-muscarinic receptors. Passive immunization of mice with purified IgG from a patient with scleroderma and high-titer antimyenteric neuronal antibodies caused disruption of intestinal myoelectrical activity, but no overt neuronal damage was detected.18 Because purified antimyenteric neuronal antibody was not used in this study, the observed gut dysmotility may be related to the effects of an autoantibody blocking M3-muscarinic receptors on the smooth muscle as reported herein. The effect of patient IgG in the present study is likely to be a specific effect. The IgG-mediated inhibition was specific for cholinergic contractions; tachykinin-mediated contractions were unaffected. Furthermore, IgG from healthy controls had no effect at up to 1 mg/mL, and serum from 2 of 9 scleroderma patients had no significant effect. Levels of total patient IgG in present study (0.6 mg/mL) are comparable to or lower than those used in other studies of autoantibodies in functional assays, which range from 0.8 to 10 mg/mL total Ig.19 –21 This argues strongly against a nonspecific inhibitory effect of IgG. Although we have demonstrated specific inhibition of cholinergic neurotransmission, the precise epitope(s) of the functional antibodies that inhibit M3-muscarinic receptor–mediated neurotransmission have yet to be determined. To date, the nonavailability of purified native receptor has prevented us from using affinity-purified antibody. It has not been possible to interpret absorption studies with crude smooth muscle lysates due to nonspecific inhibition of muscle activity by lysate. The most likely target is the M3-muscarinic receptor itself, a Gprotein– coupled receptor. Antibodies to G-protein– coupled receptors have been described in several other autoimmune conditions, including Graves’ disease (thyroid-stimulating hormone receptor),22 dilated cardiomyopathy (␤1-adrenoceptor and M2-muscarinic receptor),23,24 congenital heart block (M1-muscarinic receptor),25 Chagas’ disease (␤1 and ␤2 adrenoceptors and M2-muscarinic receptor),26 preeclampsia (angiotensin-1 receptor),27 and vitiligo (melanin-concentrating

ANTIBODY–MEDIATED DYSMOTILITY

1149

hormone receptor-1).28 Studies in our laboratory have not identified binding of patient IgG directly to M3-muscarinic receptors transfected and expressed on the surface of mammalian cells, or to recombinant M3-muscarinic receptors produced in baculovirus (data not shown). Additionally, patient sera do not react with extracellular loop peptides derived from the M3-muscarinic receptor.29 The functional antibody may react against a conformational epitope of the M3-muscarinic receptor that is not mimicked by recombinant receptors or by linear peptides. Alternatively, the antibodies may be reacting with either dimerized receptors or proteins and/or cofactors associated with the M3-muscarinic receptor. Identification of the precise epitope is important to enable the development of future therapeutic strategies, such as use of selective muscarinic agonists or removal of antibodies by immunoadsorption. In conclusion, this study describes for the first time functional antibodies interfering specifically with enteric cholinergic neurotransmission in patients with scleroderma. This is a potential cause for the early neuropathic gastrointestinal dysfunction seen in scleroderma.30 Longitudinal studies correlating the functional assay with manometry, gastrointestinal imaging, and clinical features will clarify the relationship between these antibodies and gastrointestinal disease in scleroderma.

References 1. Rose S, Young MA, Reynolds JC. Gastrointestinal manifestations of scleroderma. Gastroenterol Clin North Am 1998;27:563–594. 2. Rees WD, Leigh RJ, Christofides ND, Bloom SR, Turnberg LA. Interdigestive motor activity in patients with systemic sclerosis. Gastroenterology 1982;83:575–580. 3. Cohen S, Fisher R, Lipshutz W, Turner R, Myers A, Schumacher R. The pathogenesis of esophageal dysfunction in scleroderma and Raynaud’s disease. J Clin Invest 1972;51:2663–2668. 4. Treacy W, Baggenstoss A, Slocumb C, Code C. Scleroderma of the esophagus: a correlation of histologic and physiologic findings. Ann Intern Med 1963;59:351–356. 5. Battle WM, Snape WJ Jr, Wright S, Sullivan MA, Cohen S, Meyers A, Tuthill R. Abnormal colonic motility in progressive systemic sclerosis. Ann Intern Med 1981;94:749 –752. 6. Howe S, Eaker EY, Sallustio JE, Peebles C, Tan EM, Williams RC Jr. Antimyenteric neuronal antibodies in scleroderma. J Clin Invest 1994;94:761–770. 7. Furness JB. Types of neurons in the enteric nervous system. J Auton Nerv Syst 2000;81:87–96. 8. Kuriyama H, Kitamura K, Itoh T, Inoue R. Physiological features of visceral smooth muscle cells, with special reference to receptors and ion channels. Physiol Rev 1998;78:811–920. 9. Bacman S, Perez Leiros C, Sterin-Borda L, Hubscher O, Arana R, Borda E. Autoantibodies against lacrimal gland M3 muscarinic acetylcholine receptors in patients with primary Sjo¨gren’s syndrome. Invest Ophthalmol Vis Sci 1998;39:151–156. 10. Bacman S, Sterin-Borda L, Camusso JJ, Arana R, Hubscher O, Borda E. Circulating antibodies against rat parotid gland M3 muscarinic receptors in primary Sjo¨gren’s syndrome. Clin Exp Immunol 1996;104:454 – 459. 11. Waterman SA, Gordon TP, Rischmueller M. Inhibitory effects of

1150

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

GOLDBLATT ET AL.

muscarinic receptor autoantibodies on parasympathetic neurotransmission in Sjo¨gren’s syndrome. Arthritis Rheum 2000;43: 1647–1654. Goin JC, Sterin-Borda L, Bilder CR, Varrica LM, Iantorno G, Rios MC, Borda E. Functional implications of circulating muscarinic cholinergic receptor autoantibodies in chagasic patients with achalasia. Gastroenterology 1999;117:798 – 805. American Rheumatism Association Diagnostic and Therapeutic Criteria Committee. Preliminary criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980; 23:581–590. Vitali C, Bombardieri S, Moutsopoulos HM, Coll J, Gerli R, Hatron PY, Kater L, Konttinen YT, Manthorpe R, Meyer O, Mosca M, Ostuni P, Pellerito RA, Pennec Y, Porter SR, Richards A, Sauvezie B, Schiodt M, Sciuto M, Shoenfeld Y, Skopouli FN, Smolen JS, Soromenho F, Tishler M, Wattiaux MJ, and the European Study Group on Diagnostic Criteria for Sjo¨gren’s Syndrome. Assessment of the European classification criteria for Sjo¨gren’s syndrome in a series of clinically defined cases: results of a prospective multicentre study. The European Study Group on Diagnostic Criteria for Sjo¨gren’s Syndrome. Ann Rheum Dis 1996;55:116 –121. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS, Medsger TA Jr, Mitchell DM, Neustadt DH, Pinals RS, Schaller JG, Sharp JJ, Wilder RL, Hunder GG. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–324. Nakamura R, Peebles C, Rubin R, Molder D, Tan E. Autoantibodies to nuclear antigens (ANA). Chicago: American Society for Clinical Pathology Press, 1985. Steinbuch M, Audran R. The isolation of IgG from mammalian sera with the aid of caprylic acid. Arch Biochem Biophys 1969; 134:279 –284. Eaker EY, Kuldau JG, Verne GN, Ross SO, Sallustio JE. Myenteric neuronal antibodies in scleroderma: passive transfer evokes alterations in intestinal myoelectric activity in a rat model. J Lab Clin Med 1999;133:551–556. Houzen H, Hattori Y, Kanno M, Kikuchi S, Tashiro K, Motomura M, Nakao Y, Nakamura T. Functional evaluation of inhibition of autonomic transmitter release by autoantibody from LambertEaton myasthenic syndrome. Ann Neurol 1998;43:677– 680. Pinto A, Gillard S, Moss F, Whyte K, Brust P, Williams M, Stauderman K, Harpold M, Lang B, Newsom-Davis J, Bleakman D, Lodge D, Boot J. Human autoantibodies specific for the alpha-1A calcium channel subunit reduce both P-type and Q-type calcium currents in cerebellar neurons. Proc Natl Acad Sci U S A 1998; 95:8328 – 8333. Boutjdir M, Chen L, Zhang ZH, Tseng CE, DiDonato F, Rashbaum

GASTROENTEROLOGY Vol. 123, No. 4

22.

23.

24.

25.

26.

27.

28.

29.

30.

W, Morris A, el-Sherif N, Buyon JP. Arrhythmogenicity of IgG and anti-52-kD SSA/Ro affinity-purified antibodies from mothers of children with congenital heart block. Circ Res 1997;80:354 – 362. Atger M, Misrahi M, Young J, Jolivet A, Orgiazzi J, Schaison G, Milgrom E. Autoantibodies interacting with purified native thyrotropin receptor. Eur J Biochem 1999;265:1022–1031. Wallukat G, Wollenberger A, Morwinski R, Pitschner HF. Anti-beta 1-adrenoceptor autoantibodies with chronotropic activity from the serum of patients with dilated cardiomyopathy: mapping of epitopes in the first and second extracellular loops. J Mol Cell Cardiol 1995;27:397– 406. Wallukat G, Fu HM, Matsui S, Hjalmarson A, Fu ML. Autoantibodies against M2 muscarinic receptors in patients with cardiomyopathy display non-desensitized agonist-like effects. Life Sci 1999;64:465– 469. Borda E, Leiros CP, Bacman S, Berra A, Sterin-Borda L. Sjo¨gren autoantibodies modify neonatal cardiac function via M1 muscarinic acetylcholine receptor activation. Int J Cardiol 1999;70: 23–32. Sterin-Borda L, Borda E. Role of neurotransmitter autoantibodies in the pathogenesis of chagasic peripheral dysautonomia. Ann N Y Acad Sci 2000;917:273–280. Wallukat G, Homuth V, Fischer T, Lindschau C, Horstkamp B, Jupner A, Baur E, Nissen E, Vetter K, Neichel D, Dudenhausen JW, Haller H, Luft FC. Patients with preeclampsia develop agonistic autoantibodies against the angiotensin AT1 receptor. J Clin Invest 1999;103:945–952. Kemp EH, Waterman EA, Hawes BE, O’Neill K, Gottumukkala RV, Gawkrodger DJ, Weetman AP, Watson PF. The melanin-concentrating hormone receptor 1, a novel target of autoantibody responses in vitiligo. J Clin Invest 2002;109:923–930. Cavill D, Waterman SA, Gordon TP. Failure to detect antibodies to extracellular loop peptides of the muscarinic M3 receptor in primary Sjo¨gren’s syndrome. J Rheumatol 2002;29:1342–1344. Sjogren RW. Gastrointestinal motility disorders in scleroderma. Arthritis Rheum 1994;37:1265–1282.

Received January 9, 2002. Accepted July 5, 2002. Address requests for reprints to: Sally Waterman, Ph.D., Department of Immunology, Allergy and Arthritis, Flinders Medical Centre, Bedford Park, South Australia. e-mail: Sally.Waterman@flinders.edu.au; fax: (61) 8-8204 4158. Supported by a National Health and Medical Research Council of Australia (NH&MRC) medical postgraduate scholarship (to F.G.) and a NH&MRC project grant (to T.P.G and S.A.W). S.A.W is a NH&MRC R.D. Wright Fellow.