8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

1/8

ARTHRITIS & RHEUMATISMVol. 43, No. 7, July 2000, pp 16471654 2000, American College of Rheumatology

INHIBITORY EFFECTS OF MUSCARINIC RECEPTOR

AUTOANTIBODIES ON PARASYMPATHETIC NEUROTRANSMISSIONIN SJOGRENS SYNDROME

SALLY A. WATERMAN, TOM P. GORDON, and MAUREEN RISCHMUELLER

Objective. Sjogrens syndrome (SS) is an auto-

immune disorder characterized by dry eyes and mouth

(sicca syndrome) and lymphocytic infiltration of the

lacrimal and salivary glands. Abnormalities of para-

sympathetic neurotransmission may contribute to the

glandular dysfunction. In this study, we used a func-tional assay to investigate autoantibody-mediated ef-

fects on parasympathetic neurotransmission and

smooth muscle contraction.

Methods. Serum and purified IgG were obtained

from patients with primary and secondary SS and from

control subjects. Contraction of isolated bladder strips

in response to stimulation of M3

-muscarinic receptors

by a muscarinic receptor agonist, carbachol, or by

endogenous acetylcholine released from postganglionic

parasympathetic nerves was measured before and after

the addition of patient serum or IgG.

Results.Sera from 5 of 9 patients with primary SS

and from 6 of 6 patients with secondary SS inhibitedcarbachol-evoked bladder contraction by 50%. Sera

from these patients also inhibited the action of neuro-

nally released acetylcholine at M3

-muscarinic receptors.

Sera from 7 of 8 healthy individuals, from patients with

rheumatoid arthritis without sicca symptoms, and from

patients with systemic lupus erythematosus had no

effect. The antimuscarinic receptor activity was local-

ized in the IgG fraction, since purified IgG from pa-

tients with SS also inhibited agonist- and nerve-evoked

contractions. In this preliminary study, the autoanti-

bodies seemed to be associated with the presence of

bladder symptoms and other autonomic features.

Conclusion. Autoantibodies that act as antago-

nists at M3

-muscarinic receptors on smooth muscle

occur in a subset of patients with primary and second-ary SS. Their presence in secondary SS was unexpected

and provides new evidence for a common pathogenetic

link between primary and secondary SS. These auto-

antibodies appear to contribute to sicca symptoms and

may explain associated features of autonomic dysfunc-

tion in some patients.

Primary Sjogrens syndrome (SS) is a systemicautoimmune disorder characterized by keratoconjuncti-

vitis sicca and xerostomia (sicca syndrome), and by thepresence of serum autoantibodies that target the Ro/Laribonucleoprotein. Secondary SS occurs in the setting of

other autoimmune diseases such as rheumatoid arthritis(RA), systemic lupus erythematosus (SLE), and sclero-derma. Although sicca symptoms occur in both primaryand secondary SS, the clinical and autoantibody profilesand immunogenetic characteristics differ, and they aretherefore usually considered separate entities (1).

Sicca symptoms in SS have commonly been at-tributed to infiltration and destruction of the lacrimaland salivary glands by T lymphocytes (2). The severity ofthe sicca symptoms, however, does not always correlate

with the degree of glandular destruction seen in labialbiopsy tissue (3), and alternative hypotheses are that

sicca symptoms result from suboptimal function of theacinar and ductal cells due to the action of cytokines, ordue to parasympathetic neural dysfunction (3). In thenonobese diabetic (NOD) mouse model of SS, salivarygland dysfunction precedes lymphocytic infiltration ofthe gland (4). However, gland function is normal in theabsence of antibody production in the Ignull NODmouse, suggesting that the presence of antibodies is

Supported by a grant from the National Health and MedicalResearch Council of Australia. Dr. Waterman is a National Health and

Medical Research Council R. D. Wright Fellow.Sally A. Waterman, PhD: University of Adelaide, Adelaide,

Australia; Tom P. Gordon, MD: Flinders University of South Austra-lia, Flinders, Australia; Maureen Rischmueller, FRACP: The QueenElizabeth Hospital, Adelaide, Australia.

Address reprint requests to Sally A. Waterman, PhD, Depart-ment of Physiology, University of Adelaide, Adelaide, South Australia,5005, Australia.

Submitted for publication November 23, 1999; accepted inrevised form February 18, 2000.

1647

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

2/8

necessary for gland dysfunction in this model (4). In-deed, autoantibodies to muscarinic receptors have beendescribed in the sera of NOD mice (5), raising thepossibility that autoantibodies with the same specificitymay be present in the sera of patients with SS.

In support of this notion, glandular dysfunctioncan be produced in Ignull NOD mice by passive transferof IgG from SS patients or NOD mice, but not with IgGfrom healthy human or murine controls. It has beenproposed that the putative antimuscarinic receptorautoantibodies mediating glandular dysfunction in thismodel interfere with transmission from postganglionicparasympathetic neurons innervating the lacrimal andsalivary glands. Although radioligand-binding studieshave demonstrated that some patients with primary SSproduce autoantibodies to muscarinic receptors (6,7),the functional effects of these autoantibodies remain un-

defined, and their wider clinical significance is uncertain.Features suggestive of dysfunction of the auto-nomic nervous system, including Adies tonic pupil(814), fluctuating blood pressure and orthostatic hypo-tension (10,15), anhidrosis (14), and constipation(11,16), have been reported to occur in SS. A prospec-tive study of 32 patients with primary SS revealedevidence of autonomic neuropathy in the majority ofpatients (16). Bladder symptoms have also been re-ported, including frequency of micturition, nocturia,urgency, and incomplete bladder emptying, frequently inassociation with interstitial cystitis (17).

Muscarinic receptors of the M3subtype mediate

the stimulatory effects of acetylcholine on the lacrimaland salivary glands, as well as the contractile effects ofacetylcholine on the bladder and intestinal smooth mus-cle and on nitric oxide production in blood vessels (forreview, see ref. 18). We tested the hypothesis thatpatients with SS produce autoantibodies that act asantagonists at this receptor, and that these antibodiesare capable of causing autonomic dysfunction. We pro-

vide evidence that such autoantibodies are present insera from patients with primary SS and in sera frompatients with secondary SS in conjunction with RA.

PATIENTS AND METHODSPatients and controls. Sera and clinical data were

obtained, with informed consent, from consecutive patientswith primary SS who were attending rheumatology clinics andwho fulfilled at least 4 of the 6 European consensus criteria(including the presence of anti-Ro [SSA] and anti-La [SSB]antibodies) (19), from patients with RA and secondary SS(xerostomia and xerophthalmia with a positive result on theSchirmers test), from patients with RA or SLE without sicca

symptoms, and from healthy controls. All patients with RA andsecondary SS had seropositive, erosive disease without hyper-gammaglobulinemia or antinuclear antibody positivity. Pa-tients were questioned with regard to symptoms of autonomicdysfunction, including orthostatic dizziness, anhidrosis, consti-pation, and bladder symptoms (frequency of micturition, noc-

turia, and urgency). Serum IgG and rheumatoid factor (RF)concentrations were measured by nephelometry (BeckmanArray, Brea, CA). Counterimmunoelectrophoresis was per-formed for the detection of extractable nuclear antigens usingrabbit thymus extract (Pel-Freez, Rogers, AR) and extracts ofK562 (a human myeloid cell line) (20). The IgG fraction waspurified from serum using the caprylic acid precipitationtechnique (21).

Functional assay on bladder smooth muscle. MaleSwiss mice were killed by CO2 inhalation and the urinarybladder was removed. Two strips of tissue were dissected perbladder from the bladder dome, and the mucosa were re-moved. Strips were mounted in 10-ml jacketed organ bathscontaining Krebs solution at 37oC, gassed with 95% O2/5%CO2. The composition of the Krebs solution was 119 mM

NaCl, 4.7 mMKCl, 1.2 mMKH2PO4, 25 mMNaHCO3, 1.5 mMMgSO4, 11.0 mMD-glucose, and 2.5 mMCaCl2. Guanethidine(3 M; Sigma, St. Louis, MO) and hexamethonium (100 M;Sigma) were present throughout all experiments to blocktransmitter release from sympathetic neurons and nicotinicganglionic transmission, respectively. At the beginning of eachexperiment, preparations were desensitized to capsaicin (10M; Sigma) to inactivate sensory neurons. Ring electrodesdelivered square wave pulses (90V amplitude, 0.3 msec pulseduration, trains of 20 pulses) at frequencies of 1, 2, 5, 10, 20,and 50 Hz. Concentration-response curves to carbachol (Sig-ma) were produced by the cumulative addition of 1, 3, 10, 30,and 100 M carbachol at 75-second intervals. Consecutiveconcentration-response curves performed at intervals of3090 minutes were superimposable. Our previous studies

(Waterman SA et al: unpublished observations) have demon-strated that both nerve- and carbachol-evoked contractions areblocked by low concentrations (0.1 M) of the M3-muscarinicreceptor antagonist, 1,1-dimethyl-4-diphenylacetoxypiperi-dinium iodide (4-DAMP; Tocris Cookson, Bristol, UK), butnot by high concentrations (10 M) of M1- and M2-receptorantagonists (pirenzepine [Sigma] and himbacine [Alexis Cor-poration, San Diego, CA], respectively). Contractile responseswere measured using Kent isometric transducers connected toa PowerLab/8s data acquisition system (ADInstruments, Syd-ney, Australia) and analyzed using Chart, version 3.6.1, soft-ware (ADInstruments) on a PowerMac computer.

Serum and the IgG fraction from patients and controlswere tested both for their acute effects on the bladder detrusormuscle and for their effects on bladder muscle contractionevoked by the muscarinic receptor agonist, carbachol, and bynerve stimulation. Serum (or IgG) was added to organ baths togive a final concentration of 2% (or 0.33 mg/ml). Theseconcentrations are within the range used previously in bindingstudies (for example, see refs. 7 and 22). The effect of serum onnerve-mediated responses in the bladder was tested 30, 60, and90 minutes after addition of the serum. The effect of serum oncarbachol-evoked responses was then tested. At the end of theexperiment, the muscarinic M3-receptor antagonist, 4-DAMP(0.1 M), was added and frequency-response and carbachol

1648 WATERMAN ET AL

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

3/8

concentrationresponse curves were repeated. In some exper-iments, the effect of serum was tested after the addition of 0.1M4-DAMP. In most experiments, preparations were desen-sitized to ,-methylene ATP (-MeATP; Sigma) to blockpurinergic responses to electrical stimulation and therebyisolate the muscarinic cholinergic component.

Responses to carbachol, -MeATP, and capsaicin areexpressed as a percentage of the maximal response to carba-chol in that preparation. Responses to electrical field stimula-tion (EFS) are expressed as a percentage of the maximalresponse to EFS in that preparation prior to the addition ofdrugs or serum. Results are expressed as the mean SEM.The effect of serum and drug treatments on responses ofbladder muscle to electrical stimulation or to carbachol were

analyzed using two-way analysis of variance (ANOVA). Re-sponses to -MeATP in the absence and presence of serumwere analyzed using unpaired t-tests.

RESULTS

Inhibition of muscarinic receptor agonist

evoked bladder contraction by patient sera.Sera from 5of 9 patients with primary SS (56%) and from 6 of 6

patients with secondary SS (100%) inhibited the con-tractile responses of mouse bladder smooth muscle tocarbachol (Figures 1a and b). At a final serum concen-tration of 2%, contractions were inhibited by 4050%.The remaining contraction was abolished by the M3-muscarinic receptor antagonist, 4-DAMP (0.1 M). Serafrom 3 patients with RA without sicca symptoms, 4patients with SLE, and 1 patient with high-titer RF,hypergammaglobulinemia, and antiU1 RNP autoanti-bodies did not inhibit contractile responses (Figure 1c).Serum from a patient with mild sicca symptoms who didnot fulfill the diagnostic criteria for SS also did not

inhibit responses. Sera from 7 of 8 healthy controlsubjects without sicca symptoms had no effect oncarbachol-evoked contractions (Figure 1d). The abilityto inhibit M3-mediated responses did not correlate withthe presence of anti-Ro/La antibodies because 1) only 5of 9 patients with primary SS (all of whom had anti-Ro/La) had antiM3 receptor activity, 2) no patients withsecondary SS (all with antiM3 receptor activity) had

Figure 1. Responses to carbachol in the presence of patients sera. Concentration-response curves to carbacholwere produced in strips of bladder muscle in the absence and presence of 2% serum from patients witha,primary

Sjogrens syndrome (SS) (n 13, using sera from 5 patients), b, secondary SS (n 14, using sera from 6

patients), and c, rheumatoid arthritis without sicca symptoms (n 5, using sera from 3 patients), and from d,healthy controls (n 12, using sera from 6 individuals). Significant inhibition of contractile responses was

produced by sera from patients with primary and secondary SS (P 0.0005 and P 0.0001, respectively, by

repeated-measures analysis of variance followed by Newman-Keuls test for comparison of responses before andafter addition of serum), but not by sera from patients with rheumatoid arthritis without sicca symptoms or

healthy controls (P 0.05 in each case). 4-DAMP 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide.

MUSCARINIC RECEPTOR AUTOANTIBODIES IN SS 1649

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

4/8

anti-Ro/La, and 3) 2 of 4 patients with SLE had anti-Rowithout inhibition of M3-mediated responses.

Specific inhibition of M3

-muscarinic receptors by

SS sera.To confirm that the inhibitory effect of patientsera was not due to a nonspecific effect on smooth

muscle, responses to the P2X purinoceptor agonist, -MeATP (10 M), were measured in the absence andpresence of patient sera. Contraction amplitudes werenot significantly altered by the sera from patients withprimary SS; response amplitudes before and after addi-tion of sera were 38.7 5.8% (mean SEM) and51.4 13.7%, respectively (n 20 and n 9, respec-tively;t 0.7,P 0.05). Splitting the data according tothe presence or absence of putative M3-receptor auto-antibodies also demonstrated no effect of the sera onresponses to -MeATP (effect of serum: F[1,25] 0.018,P 0.9 by two-way ANOVA; effect of category ofpatient: F[1,25] 0.096,P 0.8). Similarly, responses to-MeATP were not inhibited by normal healthy con-trol sera (27.3 4.5% [n 10] and 23.5 4.3% [n 9]in the absence and presence of sera, respectively) or bysera from patients with SLE (30.5 10.2% [n 3] and43.8 6.3% [n 6] in the absence and presence of sera,respectively). Sera from patients with secondary SSincreased responses to purinoceptor stimulation from29.7 4.0% to 52.7 1.7% (n 10 and n 2,respectively; t 26, P 0.04), indicating a possibleup-regulation of P2X purinoceptors in response to theinhibition of muscarinic receptors.

Inhibition of the effects of endogenous acetylcho-

line by SS sera. Although serum from SS patientsinhibited responses to an exogenous muscarinic receptoragonist, it was important to establish whether the puta-tive autoantibodies interfere with the action of endoge-nous acetylcholine released from postganglionic para-sympathetic neurons on the bladder detrusor muscle.Parasympathetic neurons were stimulated electrically ata range of frequencies, and contractile responses weremeasured in the absence and presence of patient serum.The effect of neuronally released acetylcholine at M3-muscarinic receptors was significantly inhibited by 3550% within 30 minutes of the addition of 2% serum frompatients with SS. Nerve-mediated contractions were

inhibited by serum from each of the patients with primarySS whose serum inhibited responses to carbachol. Serumfrom patients without sicca symptoms and from healthycontrols did not alter nerve-evoked contractions.

Acute agonist effects of SS sera at M3

-muscarinic

receptors.Sera from 3 of 9 patients with primary SS and2 of 6 patients with secondary SS also evoked an initialcontraction of the bladder detrusor muscle (Figure 2)

prior to the antagonistic effect described above. Theamplitude of these contractions declined with time,indicating receptor desensitization. To investigate

whether these acute responses were due to an action ofthe autoantibodies at M3-muscarinic receptors, serum

was added after M3receptors had been blocked using 0.1M4-DAMP. Figure 2 shows that the acute effect wasprevented by the antagonist, confirming that it is medi-

ated by stimulation of M3-muscarinic receptors.Antimuscarinic receptor activity in the IgGfraction. The effect of IgG purified from patient sera

was tested on bladder responses to carbachol stimula-tion. Responses were inhibited 60 80% by 3 mg/ml IgGpurified from the serum of patients with primary andsecondary SS, confirming that the antimuscarinic re-ceptor activity is in the IgG fraction. IgG from patients

with secondary SS also produced slight inhibition (1530%) when used at the lower concentration of 0.3mg/ml. Antibodies from these patients thus appear to bemore potent as antagonists at muscarinic receptors,consistent with the observation that greater inhibition of

responses was produced by sera from patients withsecondary SS compared with sera from patients withprimary SS (Figure 1b versus Figure 1a).

The IgG fraction from patients with SS was alsotested for its ability to inhibit the effect of endogenousacetylcholine released by nerve stimulation. Figure 3demonstrates that the IgG inhibited the whole bladdermuscle contraction (due to the combined effects of

Figure 2. Acute agonist effect of patient sera. The findings from 2patients with primary Sjogrens syndrome are shown, in which the sera

caused an acute contraction of the bladder smooth muscle. The same

sera caused inhibition of carbachol-induced contraction after30 minutes, as shown in Figure 1. These contractions were prevented

by prior treatment of the preparations with the M3-muscarinic receptor

antagonist, 1,1-dimethyl-4-diphenylacetoxypiperidinium iodide (0.1 M).

1650 WATERMAN ET AL

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

5/8

acetylcholine and ATP) by 50% within 30 minutes.Incubation of the tissue with IgG for a further 30minutes and 60 minutes did not produce further inhibi-

tion. Desensitization of the tissue to -MeATP, toblock the purinergic component of transmission, abol-ished the response. The M3-muscarinic receptor antag-onist, 4-DAMP, did not have any further effect. Thus,there was no cholinergic component of the contraction

remaining after treatment with patient IgG, indicatingthat the IgG had abolished the cholinergic, but not thepurinergic, component of the contraction. Inhibition ofthe total contraction by 50% by IgG is consistent withthe observation that 50% of the contraction in thistissue is mediated by acetylcholine (23).

Responses of bladder muscle to the purinoceptoragonist, -MeATP, were not altered by the IgG frac-tion from patients with primary or secondary SS(F[2,66] 1.3,P 0.29 by one-way ANOVA), confirmingthat the effect of the IgG is specific for muscarinicreceptors and not due to a nonspecific effect on smoothmuscle contractility. Contraction amplitudes were31.5 2.6% in untreated preparations (n 59), 36.3 9.1% in preparations treated with primary SS IgG (n 4), and 47.4 7.7% in preparations treated with sec-ondary SS IgG (n 4).

Clinical features suggestive of autonomic dys-

function in patients with muscarinic receptor autoanti-

bodies. Of the 11 patients whose sera were found tocontain antimuscarinic receptor autoantibodies, 4 werenoted to have autonomic features such as abdominalpain and bloating, constipation, orthostatic dizziness,postural hypotension, and hyperhidrosis (Table 1).

Figure 3. Effect of IgG on nerve-evoked bladder contractions. Para-

sympathetic fibers were electrically stimulated to produce acetylcho-

line release and muscle contraction. Addition of 3 mg/ml IgG frompatients with either primary or secondary Sjogrens syndrome (the

latter shown here) reduced these responses by 50% within 30

minutes (n 3). This effect did not alter with time, since responses 60minutes and 90 minutes after the addition of IgG were superimposable

on those observed 30 minutes after its addition. For clarity, only the

responses after 60 minutes (60) are shown. The remaining responseswere abolished by desensitization of the preparation to ,-methylene

ATP (,-MeATP), and subsequent addition of 0.1M1,1-dimethyl-

4-diphenylacetoxypiperidinium iodide (4-DAMP) did not have a sig-nificant effect. Thus, the IgG abolished the cholinergic component of

the response to nerve stimulation, and the response remaining in the

presence of IgG was purinergic. EFS electrical field stimulation.

Table 1. Characteristics of patients with sera containing muscarinic receptor autoantibodies*

Diagnosis,patient Age/sex Autonomic features

Acute

agonisteffect

Antagonistaction

Primary SS1 60/F Abdominal pain, bloating, constipation 2 62/F Bladder irritability, fluctuating BP, orthostatic

dizziness, postural hypotension, tonic pupil

3 30/F None 4 81/F Bladder irritability, suprapubic pain, increased

residual volume with recurrent cystitis, abdominalbloating, constipation, orthostatic dizziness

5 64/M Bladder irritability, interstitial cystitis Secondary SS

(with RA)6 72/M Bladder irritability, increased residual volume 7 62/F Bladder irritability 8 78/F Bladder irritability 9 58/F Bladder irritability, orthostatic dizziness, hyperhidrosis

10 86/M Bladder irritability 11 75/F None

* Bladder irritability denotes frequency, nocturia, and urgency. denotes serum with agonist orantagonist effect on smooth muscle contraction; denotes serum without agonist effect on smoothmuscle. SS Sjgrens syndrome; BP blood pressure; RA rheumatoid arthritis.

MUSCARINIC RECEPTOR AUTOANTIBODIES IN SS 1651

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

6/8

Symptoms of bladder irritability (frequency of micturi-tion, nocturia, urgency, and in some cases, suprapubicpain) were observed in 8 patients, 2 of whom hadbiopsy-proven interstitial cystitis. Two SS patients withmuscarinic receptor autoantibodies did not have a his-

tory of autonomic symptoms, and 1 healthy individualwith these autoantibodies did not have a history of eithersicca symptoms or autonomic symptoms. One of the 4patients with primary SS without muscarinic receptorautoantibodies reported symptoms of bladder irritabil-ity. However, apart from the presence of sicca symp-toms, there were no other distinguishing features evi-dent between the muscarinic autoantibodypositive andnegative patients.

DISCUSSION

We have demonstrated that subsets of patientswith primary or secondary SS produce functional auto-antibodies that act as antagonists at the M3-muscarinicreceptor and interfere with parasympathetic neurotrans-mission to smooth muscle. Serum and IgG from thesepatients were able to abolish bladder muscle contractionin response to nerve stimulation and to carbachol. Thepresence of these autoantibodies in all of the patients

with RA and secondary SS tested in this study wasunexpected and provides new evidence for a commonpathogenetic link between primary and secondary SS

with regard to their sicca and autonomic nervous systemdysfunction features.

The rat and human M3-muscarinic receptors arehighly homologous; the rat protein contains 589 aminoacids with a molecular mass of 64.79 kd (GenPeptaccession no. P08483), and the human protein contains590 amino acids with a molecular mass of 64.90 kd(GenPept accession no. P20309). There are 50 differ-ences in the amino acid sequences (including a glycinesubstitution at position 54 in the human M3-muscarinicreceptor), giving these proteins a 91.5% similarity, whichincreases to 95.4% after allowing for conservative aminoacid changes. In view of this strong interspecies homol-ogy, SS sera could be expected to have similar effects onhuman and rodent M3-muscarinic receptors, although this

requires confirmation on isolated human smooth muscle.We have also directly demonstrated, with the use

of our functional assay, the presence of muscarinicreceptor autoantibodies in NOD mice (results notshown), as predicted from earlier studies (4). The auto-antibodies against the M3-muscarinic receptor describedhere could be considered part of a family of autoanti-bodies against metabotropic receptors that were recently

recognized in Chagas disease (24), dilated cardiomyo-pathy (25), and preeclampsia (26). We propose thatabnormalities of secretory and autonomic function insome patients with primary and secondary SS can be

viewed as manifestations of anti-receptor autoimmunity.

Muscarinic receptors of the M3 subtype arepresent in many tissues, including the lacrimal andsalivary glands, where they mediate the secretomotoreffects of acetylcholine. It is therefore possible that in asubpopulation of SS patients, the sicca syndrome is anautonomic disorder produced by autoantibodies to M3-muscarinic receptors. M3-muscarinic receptors also me-diate the effects of acetylcholine on the bladder detrusorand intestinal smooth muscle (causing contraction), vas-cular smooth muscle (causing nitric oxide synthesis inthe endothelium and subsequent vasodilation; for re-

view, see ref. 18), and eccrine sweat glands (2729), andare present in iris sphincter muscle (30). In our prelim-inary study, the autoantibodies seemed to be associated

with bladder symptoms and other autonomic features(see Table 1).

We speculate that the autoantibodies to M3-muscarinic receptors present in some SS patients mayproduce bladder irritability, constipation, fluctuatingblood pressure, and dilated pupils, each of which hasbeen reported in patients with SS. Detailed controlledclinical studies of large numbers of patients with primarySS, secondary SS associated with other autoimmunediseases such as scleroderma and SLE, and sicca symp-toms from other causes such as fibromyalgia or anticho-

linergic medications, and age- and sex-matched controlswithout sicca symptoms will help to clarify the preva-lence of bladder irritability and other symptoms andsigns of autonomic dysfunction and their correlation

with antibodies directed against M3-muscarinic recep-tors. Such future studies, incorporating objective indicesof autonomic dysfunction, such as cardiac responses totilt-top tables, will aid in the determination of the clinicalutility, sensitivity, and specificity of assays for theseautoantibodies.

Autonomic dysfunction due to defective musca-rinic receptor function in a patient with chronic intesti-nal pseudoobstruction, Adies tonic pupils, atonic blad-

der, and anhidrosis which were unresponsive tomuscarinic receptor agonists and anticholinesterasedrugs was proposed by Bannister and Hoyes in 1981(31). An autoantibody-mediated antagonism of the mus-carinic receptors may have produced the autonomicdysfunction in this case as well as in other cases ofidiopathic autonomic dysfunction that have symptomssuggesting a cholinergic neuropathy. We propose that a

1652 WATERMAN ET AL

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

7/8

similar mechanism of autonomic bladder dysfunctionmay account for the symptoms in our patients. Bladderirritability is a feature of interstitial cystitis, a knownmanifestation of primary SS (17). Two patients in thepresent study who had antibodies directed against M3-

muscarinic receptors had biopsy-proven interstitial cys-titis of severe degree. It is tantalizing to speculate thatanti-receptor autoimmunity may be related to the devel-opment of localized lymphocytic infiltration and bladderinflammation. Although the number of subjects in ourstudy is too small to make any assertions, we believe thatfurther investigation is warranted in view of the debili-tating nature of symptoms experienced by these patients.

In summary, we have described the functionaleffects of antiM3-muscarinic receptor antibodies onparasympathetic neurotransmission in patients with SS,and speculate that the antagonistic effects of these

antibodies play a role in the development of siccasymptoms and other features of autonomic dysfunctionin these patients. Pending the outcome of larger clinicalstudies, identification of antiM3-muscarinic receptorantibodies in SS patients may provide useful clinicalinformation to aid both diagnosis and the elucidation ofunexplained symptoms. These findings suggest severalnew approaches to treatment of bladder and otherautonomic symptoms in these patients. Treatment withselective muscarinic receptor agonists might displaceautoantibodies from their receptors and partially reversethe inhibitory effect on the smooth muscle. Strategies toremove the autoantibodies by immunoadsorption may

also be feasible, as has been shown in patients withdilated cardiomyopathy who have anti1-adrenoceptorautoantibodies (32). Assuming that autoantibody pro-duction is driven by autoreactive T cells in these pa-tients, identification of the T cell determinants of theM3-muscarinic receptor may allow the induction oftolerance through the use of mucosal delivery of syn-thetic peptide determinants, as shown recently in exper-imental myasthenia gravis (33).

REFERENCES

1. Moutsopoulos HM, Webber BL, Vlagopoulos TP, Chused TM,

Decker JL. Differences in the clinical manifestations of siccasyndrome in the presence and absence of rheumatoid arthritis.

Am J Med 1979;66:7336.2. Fox RI. Sjogrens syndrome. Curr Opin Rheumatol 1995;7:40916.3. Fox RI, Tornwall J, Maruyama T, Stern M. Evolving concepts of

diagnosis, pathogenesis and therapy of Sjogrens syndrome. CurrOpin Rheumatol 1998;10:44656.

4. Robinson CP, Brayer J, Yamachika S, Esch TR, Peck AB, StewartCA, et al. Transfer of human serum IgG to nonobese diabeticIgnull mice reveals a role for autoantibodies in the loss of

secretory fu nction of exocrine tissues in Sjogrens syndrome. ProcNatl Acad Sci U S A 1998;95:753843.

5. Yamamoto H, Sims NE, Macauley SP, Nguyen K-HT, NakagawaY, Humphreys-Beher MG. Alterations in the secretory responseof non-obese diabetic (NOD) mice to muscarinic receptor stimu-lation. Clin Immunol Immunopathol 1996;78:24555.

6. Bacman S, Sterin-Borda L, Camusso JJ, Arana R, Hubscher O,Borda E. Circulating antibodies against rat parotid gland M3muscarinic receptors in primary Sjogrens syndrome. Clin ExpImmunol 1996;104:4549.

7. Bacman S, Perez Leiros C, Sterin-Borda L, Hubscher O, Arana R,Borda E. Autoantibodies against lacrimal gland M3 muscarinicacetylcholine receptors in patients with primary Sjogrens syn-drome. Invest Ophthalmol Vis Sci 1998;39:1516.

8. Bachmeyer C, Zuber M, Dupont S, Blanche P, Dhote R, Mas JL.Adie syndrome as the initial sign of primary Sjogrens syndrome.Am J Ophthalmol 1997;123:6912.

9. Font J, Valls J, Cervera R, Pou A, Ingelmo M, Graus F. Puresensory neuropathy in patients with primary Sjogrens syndrome:clinical, immunological, and electromyographic findings. AnnRheum Dis 1990;49:7758.

10. Griffin JW, Cornblath DR, Alexander E, Campbell J, Low PA,Bird S, et al. Ataxic sensory neuropathy and dorsal root

ganglionitis associated with Sjogrens syndrome. Ann Neurol1990;27:30415.

11. Katz JS, Houroupian D, Ross MA. Multisystem neuronal involve-ment and sicca complex: broadening the spectrum of complica-tions. Muscle Nerve 1999;22:4047.

12. Rossi P, Fossaluzza V, Tosato F. Coexistence of Sjogrens syn-drome associated systemic sclerosis with Adies syndrome [letter].J Rheumatol 1986;13:8235.

13. Wright RA, Grant IA, Low PA. Autonomic neuropathy associatedwith sicca complex. J Auton Nerv Syst 1999;75:706.

14. Gemignani F, Manganelli P, Pavesi G, Marbini A. Polyneuropathyin Sjogrens syndrome: a case of prevalently autonomic neuropathy

with tonic pupil and hypohidrosis. Funct Neurol 1988;3:33748.15. Andonopoulos AP, Ballas C. Autonomic cardiovascular neuropa-

thy in primary Sjogrens syndrome. Rheumatol Int 1995;15:1279.16. Andonopoulos AP, Christodoulou J, Ballas C, Bounas A, Alexo-

poulos D. Autonomic cardiovascular neuropathy in Sjogrens

syndrome: a controlled study. J Rheumatol 1998;25:23858.17. Van de Merwe J, Kamerling R, Arendsen E, Mulder D, Hooijkaas

H. Sjogrens syndrome in patients with interstitial cystitis. J Rheu-matol 1993;20:9626.

18. Buckley NJ, Caulfield M. Transmission: acetylcholine. In: Burn-stock G, Hoyle CHV, editors. Autonomic neuroeffector mecha-nisms. Chur, Switzerland: Harwood Academic Publishers; 1992. p.257322.

19. Vitali C, Bombardieri S, Moutsopoulos HM, Balestrieri G, Ben-civelli W, Bernstein RM, et al. Preliminary criteria for theclassification of Sjogrens syndrome: results of a prospectiveconcerted action supported by the European Community. Arthri-tis Rheum 1993;36:3407.

20. Nakamura RM, Peebles CL, Rubin RL, Molder DP, Tan EM.Autoantibodies to nuclear antigens (ANA). 2nd ed. Chicago:American Society of Clinical Pathology Press; 1985.

21. Steinbuch M, Audran R. The isolation of IgG from mammalian

sera with the aid of caprylic acid. Arch Biochem Biophys 1969;134:27984.

22. Bacman SR, Berra A, Sterin-Borda L, Borda ES. Human primarySjogrens syndrome autoantibodies as mediators of nitric oxiderelease coupled to lacrimal gland muscarinic acetylcholine recep-tors. Curr Eye Res 1998;17:113542.

23. Waterman SA. Multiple subtypes of voltage-gated calcium chan-nel mediate transmitter release from parasympathetic neurons inthe mouse bladder. J Neurosci 1996;16:415561.

24. Goin JC, Borda E, Perez Leiros C, Storino R, Sterin-Borda L.

MUSCARINIC RECEPTOR AUTOANTIBODIES IN SS 1653

8/10/2019 1529-0131(200007)43-7-1647--AID-ANR31-3.0

8/8

Identification of antibodies with muscarinic cholinergic activity inhuman Chagas disease: pathological implications. J Auton NervSyst 1994;47:4552.

25. Magnusson Y, Marullo S, Hoyer S, Waagstein F, Andersson B,Vahlne A, et al. Mapping of a functional autoimmune epitope onthe beta 1-adrenergic receptor in patients with idiopathic dilatedcardiomyopathy. J Clin Invest 1990;86:165863.

26. Wallukat G, Homuth V, Fischer T, Lindschau C, Horstkamp B,Jupner A, et al. Patients with preeclampsia develop agonisticautoantibodies against angiotensin AT1 receptor. J Clin Invest1999;103:94552.

27. Torres NE, Zollman PJ, Low PA. Characterization of muscarinicreceptor subtype of rat eccrine sweat gland by autoradiography.Brain Res 1991;550:12932.

28. Vilches JJ, Navarro X, Verdu E. Functional sudomotor responsesto cholinergic agonists and antagonists in the mouse. J Auton NervSyst 1995;55:10511.

29. Landis SC. Development of muscarinic receptors and regulation ofsecretory responsiveness in rodent sweat glands. Life Sci 1999;64:3815.

30. Ishizaka N, Noda M, Yokoyama S, Kawasaki K, Yamamoto M,Higashida H. Muscarinic acetylcholine receptor subtypes in thehuman iris. Brain Res 1998;787:3447.

31. Bannister R, Hoyes AD. Generalised smooth-muscle disease withdefective muscarinic-receptor function. Br Med J Clin Res Ed1981;282:10158.

32. Dorffel WV, Felix SB, Wallukat G, Brehme S, Bestvater K,Hofmann T, et al. Short-term hemodynamic effects of immuno-adsorption in dilated cardiomyopathy. Circulation 1997;95:19947.

33. Barchan D, Souroujon MC, Sin-Hyeog I, Antozzi C, Fuchs S.Ant ige n-s pecific mod ul ati on of experim ental mya sth eniagravis: nasal tolerization with recombinant fragments of the hu-man acetylcholine receptor. Proc Natl Acad Sci U S A1999;96:808691.

Clinical images: Orbital granuloma and subglottic tracheal stenosis in Wegeners granulomatosis

The patient was a 32-year-old woman with Wegeners granulomatosis of 6 years duration. She was being treated with methotrexatefor the maintenance of apparent disease remission. A protrusion of the bulbus of the right eye and a subglottic tracheal stenosisdeveloped. Magnetic resonance imaging of the orbits and paranasal sinuses (coronal non-enhanced T1-weighted spin-echo image,low signal intensity) demonstrated masses obliterating the right orbit, encasing the optic nerve, and infiltrating the intraorbitalmuscles, and bony destruction of the nasal cavity and maxillary sinuses with granuloma formation. After contrast administration, thecorresponding image(A)disclosed nonhomogeneous enhancement of intraorbital masses (arrows),suggestive of orbital granulomaformation and intense uptake of maxillary sinus granulomas (arrowheads). Spiral computed tomography (B) revealed a thickenedaryepiglottic fold seen on an axial scan and a voluminous mass obliterating the vocal cord (arrows) seen on coronal reformation.Orbital granuloma formation and subglottic tracheal stenosis are late organ manifestations of Wegeners granulomatosis, which arenot always responsive to systemic immunosuppressive treatment. If immunosuppressive treatment is not effective, percutaneoushigh-voltage radiotherapy, retrobulbar alcohol injections, or even enucleation may be necessary to treat the orbital granuloma, andintratracheal dilation and corticosteroid injection to treat the subglottic tracheal stenosis.

P. Lamprecht, MDE. Reinhold-Keller, MDW. L. Gross, MD

Medizinische Universitat Lubeck and Rheumaklinik Bad BramstedtBad Bramstedt, GermanyM. Reuter, MD

Klinikum der Christian-Albrechts-UniversitatKiel, Germany

1654 WATERMAN ET AL