- Email: [email protected]
Neutralization of Muscarinic Receptor Autoantibodies by Intravenous Immunoglobulin in Sjögren Syndrome
Neutralization of Muscarinic Receptor Autoantibodies by Intravenous Immunoglobulin in Sjögren Syndrome Anthony J. Smith, Michael W. Jackson, Fang Wang, Dana Cavill, Maureen Rischmueller, and Tom P. Gordon ABSTRACT: Autoantibodies that inhibit M3 muscarinic receptor (M3R)-mediated neurotransmission and cause bladder and bowel dysfunction have been reported in patients with Sjögren syndrome and belong to a family of functional autoantibodies that includes the thyroid-stimulating hormone receptor antibody present in Graves disease. We have recently reported that antiidiotypic antibodies present in pooled immunoglobulin (Ig) G or IgG from healthy individuals neutralize anti-M3R antibodymediated inhibition of smooth muscle contraction in vitro. Here we extend these studies to the clinic by examining whether therapeutic doses of intravenous immunoglobulin (IVIG) provided to patients with autoimmune diseases neutralize anti-M3R activity in vivo and improve bladder and bowel symptoms. Three patients with primary Sjögren syndrome, dermatomyositis, and celiac disease, respectively, all of whom had anti-M3R activity on a functional bladder contractile assay, were provided a single course of IVIG at a dose of 400 mg/kg per day for 5 days. Anti-M3R activity was neutralized at 4 weeks after ABBREVIATIONS ELISA enzyme-linked immunosorbent assay IVIG intravenous immunoglobulin M3R M3 muscarinic receptor OD optical density
INTRODUCTION Primary Sjögren syndrome (SS) is a systemic autoimmune disorder characterized by lymphocytic infiltration From the Department of Immunology, Allergy, and Arthritis, Flinders Medical Centre and Flinders University, Bedford Park, South Australia (A.J.S., M.W.J., D.C., T.P.G.); Beijing Hospital, Beijing, People’s Republic of China (F.W.); and Department of Rheumatology, The Queen Elizabeth Hospital, Woodville South, South Australia (M.R.). Address reprint requests to: Dr. Tom Gordon, Room 2E107, Department of Immunology, Allergy and Arthritis, Flinders Medical Centre and Flinders University, Bedford Park, South Australia, 5042, Australia; Tel: 61-882044097; Fax: 61-8-82044158; E-mail: [email protected]. Received October 19, 2004; accepted January 19, 2005. Human Immunology 66, 411– 416 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
IVIG infusion, whereas levels of specific autoantibodies (anti-La, anti–Jo-1, and anti–tissue transglutaminase) were unchanged. Bladder and bowel scores revealed variable improvement after IVIG. Neutralization of anti-M3R activity by IVIG in vivo, presumably as a result of antiidiotypic antibodies directed specifically against anti-M3R autoantibodies, provides a clinical correlate of our in vitro findings. This offers a rationale for IVIG as a treatment for parasympathetic dysfunction in patients with autoantibodies inhibiting postganglionic cholinergic neurotransmission. We suggest the presence of a network of naturally occurring antiidiotypic antibodies that regulate the expression of functional autoantibodies against neuronal receptors and ion channels. Human Immunology 66, 411– 416 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: autoimmunity; Sjögren syndrome; autonomic dysfunction; muscarinic receptor; intravenous immunoglobulin
RA SS TTG
rheumatoid arthritis Sjögren syndrome tissue transglutaminase
and exocrine failure of salivary and lacrimal glands, leading to keratoconjunctivitis sicca and xerostomia (sicca syndrome). Extraglandular complications are common and include interstitial nephritis with renal tubular acidosis, bronchiolitis and pulmonary fibrosis, palpable purpura, and sensory peripheral neuropathy [1]. Secondary SS occurs in the setting of other autoimmune diseases such as rheumatoid arthritis (RA), systemic lupus erythematosus, scleroderma, and dermatomyositis. Primary SS is the second most common systemic autoimmune disease in humans after RA, but therapeutic options 0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.01.020
412
remain limited because of our primitive understanding of its etiopathogenesis. Features of parasympathetic nervous system dysfunction, including gastrointestinal and lower urinary tract symptoms, orthostatic hypotension, anhidrosis, and Adie’s tonic pupil have attracted considerable attention as complications of SS [2, 3]. Part of this spectrum is the overactive bladder, a syndrome characterized by involuntary contractions of the detrusor muscle that has recently been reported to be a significant complication of primary SS [4]. The dominant immunological finding in primary SS is B-cell hyperactivity, which is characterized by polyclonal hypergammaglobulinemia, circulating autoantibodies against the classical ribonucleoproteins Ro (SS-A) and La (SS-B), and production of rheumatoid factor [1]. More recently, functional autoantibodies that inhibit M3 muscarinic receptor (M3R)-mediated cholinergic neurotransmission have been detected by radio ligand binding assays, flow cytometry of M3R-transfected cells, and smooth muscle contractile assays [5–7]. These autoantibodies have been reported in patients with both primary SS, secondary SS, and scleroderma and are likely to contribute to both bladder and gut dysfunction [7, 8]. The presence of a pathogenic autoantibody that mediates both the sicca symptoms and peripheral autonomic symptoms in patients with SS raises the possibility of new therapeutic approaches for the treatment of refractory bladder, gut, and sicca symptoms. The pathophysiological role of anti-M3R now appears more complex than originally thought, with a recent study demonstrating that in vivo exposure to these antibodies upregulates M3R expression, leading to a state of cholinergic hyperresponsiveness [9]. Treatment with muscarinic receptor agonists such as pilocarpine may potentially worsen the urinary, gut, and lung symptoms in SS patients in the presence of smooth muscle hyperactivity. Thus, immune strategies aimed at removing or neutralizing the antiM3R antibodies may be preferable to pharmacological approaches in the management of autonomic complications in SS patients. Neutralization of Anti-M3R Autoantibodies In Vitro by Antiidiotypic Antibodies in Intravenous Immunoglobulin Intravenous immunoglobulin (IVIG) has been used in several autoimmune rheumatic diseases including inflammatory myositis, systemic lupus erythematosus, and RA, with proven effectiveness in adult dermatomyositis alone [10]. Studies of IVIG in SS have been limited, with case reports of improvement of autonomic and exocrine glandular symptoms [11, 12] and a pilot study demonstrating prolonged improvement in the neurological complication of ataxic sensory neuronopathy [13]. We hypothesized that antiidiotypic antibodies present in
A.J. Smith et al.
IVIG may neutralize anti-M3R autoantibodies and account for the therapeutic effect of IVIG in patients with SS. By use of an in vitro functional assay of colon smooth muscle contraction stimulated by the cholinergic agonist carbachol, we demonstrated that equimolar concentrations of IVIG completely reversed the inhibitory effects of purified immunoglobulin (Ig) G from patients with primary and secondary SS on cholinergic neurotransmission [14]. The blocking activity of the IVIG was mediated by the F(ab=)2 fraction, and the functional antiM3R activity of patient IgG was removed by absorption with Sepharose-bound F(ab=)2 from IVIG. These findings confirm that the neutralization of the pathogenic anti-M3R autoantibodies is mediated by antiidiotypic antibodies in the IVIG, providing a potential mechanism for the improvement in autonomic symptoms in the case reports. In addition, equimolar amounts of IgG purified from several healthy adults also neutralized functional anti-M3R autoantibodies, raising the possibility that naturally occurring antiidiotypic antibodies may prevent the emergence of potentially pathogenic anti-M3R autoantibodies. We now expand these studies by revealing that administration of IVIG to different patients with circulating anti-M3R autoantibodies selectively neutralizes the activity of these autoantibodies in vivo.
MATERIALS AND METHODS Patients Sera, clinical data, and approval to receive IVIG were obtained, with informed consent, from each patient. Patient 1 was a 66-year-old woman with a history of seropositive primary SS for 36 years. Apart from keratoconjunctivitis sicca, symptoms of bladder irritability/ detrusor hyperactivity dominated the clinical presentation. Patient 2 was a 36-year-old woman with celiac disease for 8 years, whose serum contained IgA anti– tissue transglutaminase (anti-TTG) antibodies. Despite a gluten-free diet with normalization of small bowel mucosa, 10 to 12 loose bowel motions a day persisted. Urinary frequency was also a feature. Patient 3 was a 35-year-old man with anti–Jo-1–positive dermatomyositis diagnosed 1 year before. Diarrhea commenced 6 months after diagnosis, along with urinary frequency and urgency. Treatment with prednisolone, methotrexate, azathioprine, and mycophenolate mofetil failed to control the myositis. IVIG was then provided therapeutically. All patients received infusions of 400 mg/kg per day of IVIG (Intragam P, CSL, Melbourne, Australia) on five consecutive days. Blood was collected
Neutralization of Muscarinic Receptor Autoantibodies
before and 1 month after IVIG treatment for conventional and functional autoantibody analysis. Functional Assay on Bladder Smooth Muscle Male Balb/c mice were killed by cervical dislocation and the bladder was removed. The dome of the bladder was dissected and divided into two strips after removal of the mucosa. Strips were mounted in 5-ml jacketed organ baths containing Krebs solution at 37°C, gassed with 95% O2/5% CO2. Guanethidine (3 M; Sigma, St. Louis, MO) and hexamethonium (100 M, Sigma) were present throughout all experiments to block transmitter release from sympathetic neurons and nicotinic ganglionic transmission, respectively. At the beginning of each experiment, preparations were desensitized to capsaicin (10 M; Sigma) to inactivate sensory neurons. Concentration-response curves to carbachol (Sigma) were produced by the cumulative addition of 0.1, 0.3, 1, 3, 10, 30, and 100 M carbachol at 75-second intervals [7]. Contractile responses were measured with MLT0201 force transducers connected to a PowerLab/8SP data acquisition system (ADInstruments, Sydney, Australia) and analyzed by Chart software version 4.0 (ADInstruments) on an eMac computer. The IgG fraction from patient serum was purified on a protein A–Sepharose column [15] (Sigma) by using serum taken before and 4 weeks after IVIG treatment. IgG was added to organ baths to give a final concentration of 1 mg/ml. After a 60-minute incubation, the effect of patient IgG was tested on carbachol-evoked responses of smooth muscle. Responses to carbachol are expressed as a percentage of the maximal response to carbachol in that preparation. Results are expressed as mean ⫾ SEM. The effect of IgG and agonist on responses of bladder muscle to carbachol were analyzed by two-way analysis of variance. Conventional Autoantibody Assays Serum from each patient was collected before and 4 weeks after IVIG treatment and tested for anti-La (SS-B) (patient 1), anti-TTG (patient 2), and anti–Jo-1 (patient 3). Affinity-purified recombinant human La was diluted to 2 g/ml in 0.03 M carbonate, pH 9.6, and coated onto enzyme-linked immunosorbent assay (ELISA) plates (Nunc, Roskilde, Denmark) for 1 hour at 37°C. Wells were blocked with 200l of 1% bovine serum albumin for 1 hour at 37°C. A total of 100 l of 1/500 dilution of the patient’s sera was added to the wells for 1 hour at 37°C. The plates were washed four times with phosphate-buffered saline– 0.05% Tween 20, and bound IgG was detected by alkaline phosphatase– conjugated goat anti-human IgG. Results were expressed as OD405 units [16]. IgA anti-TTG titers were assayed by QUANTA Lite tTG
413
(INOVA Diagnostics, San Diego, CA), and IgG anti–Jo-1 titers were assayed with RELISA (Immunoconcepts, Sacramento, CA). Results for both were expressed in ELISA units. RESULTS Smooth Muscle Assays Purified IgG taken from the three patients with primary SS, celiac disease, and dermatomyositis before administration of IVIG produced a noncompetitive inhibition of carbachol-induced contraction of detrusor muscle strips compared with control response curves (p ⬍ 0.001 by two-way analysis of variance for each patient), consistent with the presence of a functional autoantibody directed against M3Rs on the postsynaptic smooth muscle membrane [7] (Figure 1a,c,e). We have reported a similar pattern of anti-M3R antibody activity in patients with primary SS, RA with secondary SS, and scleroderma [7, 8]. Four weeks after IVIG, the patient IgG fractions no longer inhibited carbachol-induced detrusor contractions mediated by M3Rs, indicating that the IVIG had neutralized the functional activity of the inhibitory antiM3R autoantibodies in vivo (Figure 1b,d,f ). Conventional Autoantibody Assays To exclude a nonspecific general inhibitory effect of the IVIG on circulating autoantibodies, we measured levels of conventional IgG marker autoantibodies in the three patients. Levels of specific anti-La (SS-B) (primary SS), anti-TTG (celiac disease), and anti–Jo-1 (dermatomyositis) were essentially unchanged before and after the administration of IVIG (Table 1). Therefore, the IVIGinduced reversal of the inhibitory anti-M3R activity in vivo appears specific and is most likely to be mediated by antiidiotypes present in IVIG, as demonstrated earlier by in vitro mixing and depletion experiments [14]. Preliminary clinical data suggested an improvement in symptoms at 1 month after IVIG. The patient with primary SS had an improved score on the urinary urgency question on the American Urological Association 7 [15] symptom index and a modest improvement in salivary flow, although she did not report an improvement in sicca symptoms per se. The patients with celiac disease and dermatomyositis exhibited improvement in diarrhea and urinary frequency scores, consistent with an improvement in gut and bladder function (Table 1). DISCUSSION Anti-M3R autoantibodies can be considered part of a family of functional autoantibodies that bind to a receptor or ion channel and modify its function by activating it (agonistic antibody) or inhibiting it (antagonistic antibody). In the case of anti-M3R, the monovalent anti-
414
A.J. Smith et al.
FIGURE 1 Neutralization of anti-M3R autoantibodies after in vivo administration of IVIG. Concentration-response curves to carbachol were produced in strips of bladder smooth muscle before (solid squares) and after (solid triangles) the addition of patient IgG (1 mg/ml). (a, c, e) Carbachol-evoked contractile responses were significantly inhibited by IgG from each patient before IVIG treatment, consistent with functional antiM3R antibodies (p ⬍ 0.001; two-way analysis of variance). (b, d, f ) In contrast, IgG taken 4 weeks after IVIG from the three patients did not significantly inhibit contractions, indicating in vivo neutralization of anti-M3R antibodies (p ⬎ 0.05; two-way analysis of variance). Results are expressed as the mean ⫾ SEM of 3 experiments for each patient.
body retains full activity, indicating that receptor crosslinking is not required. Prototypes of functional autoantibodies are those against the thyroid-stimulating hormone receptor in Graves disease and the skeletal muscle acetylcholine receptor in myasthenia gravis. As opposed to classical high-titer autoantibodies against intracellular targets such as Ro(SS-A) and La(SS-B), antiM3R and other functional autoantibodies are typically present at low serum concentrations, reactive with highly conformational epitopes under physiological conditions,
and are effective in passive transfer experiments. As such, anti-M3R autoantibodies are likely to be of fundamental pathogenic importance in primary and secondary SS. These autoantibodies appear to be more widespread than originally thought, being reported in cases of celiac disease and dermatomyositis in the present study. It is intriguing to speculate that anti-M3R may be responsible for some of the autonomic complications and gastrointestinal manifestations in these diseases [18, 19]. Larger studies are required to determine the true prevalence of anti-M3R autoantibodies in celiac disease and dermatomyositis. Given the clinical importance of anti-M3R autoantibodies in a spectrum of autoimmune diseases, further investigation of immunological approaches to reverse their pathogenetic effects on the bladder, gut, and other organs is warranted. In the present study, a single therapeutic dose of IVIG to patients has produced neutralization of IgG-mediated inhibition of cholinergic neurotransmission. The neutralizing activity of the IVIG, most likely by an antiidiotypic mechanism as previously demonstrated in vitro, suggests that anti-M3R autoantibodies act by a reversible functional process and do not induce irreversible cytotoxic destruction through Fc-mediated effector mechanisms. This is supported by passive transfer experiments, which reveal a compensatory upregulation of M3R numbers in the bladder but no evidence of inflammation or damage at 6 weeks after injection of patient IgG [9]. Our preliminary studies demonstrate that IVIG is a promising treatment for bladder, gut, and other autonomic complications in SS and other autoimmune diseases with anti-M3R autoantibodies, and demonstrate for the first time in patients that administration of IVIG can specifically neutralize a functional human autoantibody. Randomized controlled trials are now required to confirm the beneficial effect of IVIG on parasympathetic dysfunction in these different diseases. Future studies will be directed at isolating or cloning the specific neutralizing antiidiotypic antibody, which is
TABLE 1 Conventional autoantibodies and symptoms before and after IVIGa Disease
Autoantibody/Clinical features
Pre IVIG
Post IVIG
Primary Sjögren syndrome
anti-La n ⬍ 0.18 OD units USF (ml/15 min) Urinary urgency (AUA) anti-TTG n ⬍ 20 ELISA units Diarrhea (bowel motions per day) Urinary frequency (AUA) anti–Jo-1 n ⬍ 20 ELISA units Diarrhea (bowel motions per day) Urinary urgency (AUA)
1.7 3.5 5 41.3 10–12 ⫻ loose 4 117 8–10 ⫻ loose 3
2 5.5 2 52.7 1 ⫻ formed 2 118 1–2 ⫻ formed 1
Celiac disease Dermatomyositis
a Abbreviations: USF ⫽ unstimulated salivary flow in 15 minutes; AUA Score ⫽ American Urological Association score of symptom severity; TTG ⫽ tissue transglutaminase; IVIG ⫽ intravenous immunoglobulin.
Neutralization of Muscarinic Receptor Autoantibodies
415
3.
FIGURE 2 (a) Network of naturally occurring antiidiotypic antibodies regulates the expression of functional autoantibodies against neuronal receptors and ion channels. (b) Loss of coordinated expression of neutralizing antiidiotypic antibodies allows binding of pathogenic autoantibodies to M3Rs (Sjögren syndrome) and L-type voltage-gated calcium channel (VGCCs) (type 1 diabetes) with subsequent smooth muscle dysfunction.
4.
5.
likely to be a more specific and less costly agent than pooled human preparations of IVIG. In addition to the obvious therapeutic applications, the finding of natural antiidiotypic antibodies in healthy humans that neutralize anti-M3R autoantibodies provides clues to the emergence of these autoantibodies in disease. We suggest that the regulation of potentially pathogenic functional autoantibodies by natural antiidiotypic antibodies may be a more general phenomenon in health and autoimmune disease. For example, we have also identified antiidiotypes in IVIG that neutralize the functional activity of a novel calcium channel–stimulating autoantibody in type 1 diabetes [20, 21]. Mixing experiments with type 1 diabetic IgG and SS IgG have revealed that the antiidiotypic antibody that neutralizes the diabetic autoantibody is distinct from the species that neutralizes anti-M3R (M. Jackson, unpublished data). We speculate that there is a network of naturally occurring antiidiotypic antibodies circulating in normal individuals that prevents the emergence of potentially pathogenic functional autoantibodies against neuronal receptors and ion channels (Figure 2a). As a corollary of this, an alteration of the coordinated expression of an antiidiotypic antibody and its corresponding functional autoantibody allows binding of functional autoantibodies to their targets with pathophysiological consequences (Figure 2b). The mechanisms leading to such a disturbance in the human B-cell repertoire are presently unknown. The nonobese diabetic (NOD) mouse, which also develops functionally active anti-M3R autoantibodies that are not found in normal littermates, may be a useful animal model to address these questions. REFERENCES 1. Jonsson R, Haga H-G, Gordon T: Sjögren’s syndrome. In Koopman WJ (ed): Arthritis and Allied Conditions—A Textbook of Rheumatology, 14th ed. Philadelphia, Lippincott, Williams and Wilkins, 2001. 2. Hocevar A, Tomsic M, Praprotnik S, Hojnik M, Kveder T,
6.
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
Rozman B: Parasympathetic nervous system dysfunction in primary Sjögren’s syndrome. Ann Rheum Dis 62:702, 2003. Kovács L, Papós M, Takács R, Róka R, Csenke Z, Kovács A, Várkonyi T, Pajor L, Pávics L, Pokorny G: Autonomic nervous system dysfunction involving the gastrointestinal and the urinary tracts in primary Sjögren’s syndrome. Clin Exp Rheum 21:697, 2003. Walker J, Gordon TP, Lester S, Downie-Doyle S, McEvoy D, Pile K, Waterman SA, Rischmueller M: Increased severity of lower urinary tract symptoms and daytime somnolence in primary Sjögren’s syndrome. J Rheum 30:2406, 2003. Bacman S, Sterin-Borda L, Camusso J, Arana R, Hubscher O, Borda E: Circulating antibodies against rat parotid gland M3 muscarinic receptors in primary Sjögren’s syndrome. Clin Exp Immunol 104:454, 1996. Gao J, Cha S, Jonsson R, Opalko J, Peck AB: Detection of anti-type 3 muscarinic acetylcholine receptor autoantibodies in the sera of Sjögren’s syndrome patients by use of a transfected cell line assay. Arthritis Rheum 50:2615, 2004. Waterman SA, Gordon TP, Rischmueller M: Inhibitory effects of muscarinic receptor autoantibodies on parasympathetic neurotransmission in primary Sjögren’s syndrome. Arthritis Rheum 43:1647, 2000. Goldblatt F, Gordon TP, Waterman SA: Antibodymediated gastrointestinal dysmotility in scleroderma. Gastroenterology 123:1144, 2002. Wang F, Jackson M, Maughan V, Cavill D, Smith AJ, Waterman SA, Gordon TP: Passive transfer of Sjögren’s syndrome IgG produces the pathophysiology of an overactive bladder. Arthritis Rheum 50:3637, 2004. Pyne D, Ehrenstein M, Morris V: The therapeutic uses of intravenous immunoglobulins in autoimmune diseases. Rheumatology 41:367, 2002. Dupond J-L, Gil H, de Wazieres B: Five year efficacy of intravenous gammaglobulin to treat dysautonomia in Sjögren’s syndrome. Am J Med 106:125, 1999. Amital H, Rewald E, Levy Y, Bar-Dayan Y, Manthorpe R, Engervall P, Sherer Y, Langevitz P, Shoenfeld Y: Fibrosis regression induced by intravenous gammaglobulin treatment. Ann Rheum Dis 62:175, 2003. Takahashi Y, Takata T, Hoshino M, Sakurai M, Kanazawa I: Benefit of IVIG for long-standing ataxic sensory neuronopathy with primary Sjögren’s syndrome. Neurology 60:503, 2003. Cavill D, Waterman SA, Gordon TP: Antiidiotypic antibodies neutralize autoantibodies that inhibit cholinergic neurotransmission. Arthritis Rheum 48:3597, 2003. Livingston M: Immunoaffinity chromatography of proteins. Methods Enzymol 34:723, 1974. Beer RG, Rischmueller M, Coates PT, Purcell AW, Keech CL, McCluskey J, Gordon TP: Nonprecipitating
416
anti-La(SSB) autoantibodies in primary Sjögren’s syndrome. Clin Immunol Immunopath 79:314, 1996. 17. Barry MJ, Fowler FJ Jr, O’Leary MP, Bruskewitz RC, Holtgrewe HL, Mebust WK, Cockett AT: The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol 148:1549, 1992. 18. Hughes AJ, Ferguson I, Rankin E, Kane K: Polymyositis as a cause of total gut failure. Ann Rheum Dis 61:305, 2002. 19. Usai P, Usai Satta P, Lai M, Corda MG, Piras E, Calcara
A.J. Smith et al.
C, Boy MF, Morelli A, Balestrieri A, Bassotti G: Autonomic dysfunction and upper digestive functional disorders in untreated adult coeliac disease. Eur J Clin Invest 27:1009, 1997. 20. Jackson M, Waterman SA, Gordon TP: Functional characterization of a calcium channel activating autoantibody specific for type 1 diabetes. Diabetes 53(Suppl 2):A210, 2004. 21. Jackson M, Gordon TP, Waterman SA: Disruption of intestinal motility by a calcium channel-stimulating autoantibody in type 1 diabetes. Gastroenterology 126:819, 2004.
INTRODUCTION Primary Sjögren syndrome (SS) is a systemic autoimmune disorder characterized by lymphocytic infiltration From the Department of Immunology, Allergy, and Arthritis, Flinders Medical Centre and Flinders University, Bedford Park, South Australia (A.J.S., M.W.J., D.C., T.P.G.); Beijing Hospital, Beijing, People’s Republic of China (F.W.); and Department of Rheumatology, The Queen Elizabeth Hospital, Woodville South, South Australia (M.R.). Address reprint requests to: Dr. Tom Gordon, Room 2E107, Department of Immunology, Allergy and Arthritis, Flinders Medical Centre and Flinders University, Bedford Park, South Australia, 5042, Australia; Tel: 61-882044097; Fax: 61-8-82044158; E-mail: [email protected]. Received October 19, 2004; accepted January 19, 2005. Human Immunology 66, 411– 416 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
IVIG infusion, whereas levels of specific autoantibodies (anti-La, anti–Jo-1, and anti–tissue transglutaminase) were unchanged. Bladder and bowel scores revealed variable improvement after IVIG. Neutralization of anti-M3R activity by IVIG in vivo, presumably as a result of antiidiotypic antibodies directed specifically against anti-M3R autoantibodies, provides a clinical correlate of our in vitro findings. This offers a rationale for IVIG as a treatment for parasympathetic dysfunction in patients with autoantibodies inhibiting postganglionic cholinergic neurotransmission. We suggest the presence of a network of naturally occurring antiidiotypic antibodies that regulate the expression of functional autoantibodies against neuronal receptors and ion channels. Human Immunology 66, 411– 416 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: autoimmunity; Sjögren syndrome; autonomic dysfunction; muscarinic receptor; intravenous immunoglobulin
RA SS TTG
rheumatoid arthritis Sjögren syndrome tissue transglutaminase
and exocrine failure of salivary and lacrimal glands, leading to keratoconjunctivitis sicca and xerostomia (sicca syndrome). Extraglandular complications are common and include interstitial nephritis with renal tubular acidosis, bronchiolitis and pulmonary fibrosis, palpable purpura, and sensory peripheral neuropathy [1]. Secondary SS occurs in the setting of other autoimmune diseases such as rheumatoid arthritis (RA), systemic lupus erythematosus, scleroderma, and dermatomyositis. Primary SS is the second most common systemic autoimmune disease in humans after RA, but therapeutic options 0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.01.020
412
remain limited because of our primitive understanding of its etiopathogenesis. Features of parasympathetic nervous system dysfunction, including gastrointestinal and lower urinary tract symptoms, orthostatic hypotension, anhidrosis, and Adie’s tonic pupil have attracted considerable attention as complications of SS [2, 3]. Part of this spectrum is the overactive bladder, a syndrome characterized by involuntary contractions of the detrusor muscle that has recently been reported to be a significant complication of primary SS [4]. The dominant immunological finding in primary SS is B-cell hyperactivity, which is characterized by polyclonal hypergammaglobulinemia, circulating autoantibodies against the classical ribonucleoproteins Ro (SS-A) and La (SS-B), and production of rheumatoid factor [1]. More recently, functional autoantibodies that inhibit M3 muscarinic receptor (M3R)-mediated cholinergic neurotransmission have been detected by radio ligand binding assays, flow cytometry of M3R-transfected cells, and smooth muscle contractile assays [5–7]. These autoantibodies have been reported in patients with both primary SS, secondary SS, and scleroderma and are likely to contribute to both bladder and gut dysfunction [7, 8]. The presence of a pathogenic autoantibody that mediates both the sicca symptoms and peripheral autonomic symptoms in patients with SS raises the possibility of new therapeutic approaches for the treatment of refractory bladder, gut, and sicca symptoms. The pathophysiological role of anti-M3R now appears more complex than originally thought, with a recent study demonstrating that in vivo exposure to these antibodies upregulates M3R expression, leading to a state of cholinergic hyperresponsiveness [9]. Treatment with muscarinic receptor agonists such as pilocarpine may potentially worsen the urinary, gut, and lung symptoms in SS patients in the presence of smooth muscle hyperactivity. Thus, immune strategies aimed at removing or neutralizing the antiM3R antibodies may be preferable to pharmacological approaches in the management of autonomic complications in SS patients. Neutralization of Anti-M3R Autoantibodies In Vitro by Antiidiotypic Antibodies in Intravenous Immunoglobulin Intravenous immunoglobulin (IVIG) has been used in several autoimmune rheumatic diseases including inflammatory myositis, systemic lupus erythematosus, and RA, with proven effectiveness in adult dermatomyositis alone [10]. Studies of IVIG in SS have been limited, with case reports of improvement of autonomic and exocrine glandular symptoms [11, 12] and a pilot study demonstrating prolonged improvement in the neurological complication of ataxic sensory neuronopathy [13]. We hypothesized that antiidiotypic antibodies present in
A.J. Smith et al.
IVIG may neutralize anti-M3R autoantibodies and account for the therapeutic effect of IVIG in patients with SS. By use of an in vitro functional assay of colon smooth muscle contraction stimulated by the cholinergic agonist carbachol, we demonstrated that equimolar concentrations of IVIG completely reversed the inhibitory effects of purified immunoglobulin (Ig) G from patients with primary and secondary SS on cholinergic neurotransmission [14]. The blocking activity of the IVIG was mediated by the F(ab=)2 fraction, and the functional antiM3R activity of patient IgG was removed by absorption with Sepharose-bound F(ab=)2 from IVIG. These findings confirm that the neutralization of the pathogenic anti-M3R autoantibodies is mediated by antiidiotypic antibodies in the IVIG, providing a potential mechanism for the improvement in autonomic symptoms in the case reports. In addition, equimolar amounts of IgG purified from several healthy adults also neutralized functional anti-M3R autoantibodies, raising the possibility that naturally occurring antiidiotypic antibodies may prevent the emergence of potentially pathogenic anti-M3R autoantibodies. We now expand these studies by revealing that administration of IVIG to different patients with circulating anti-M3R autoantibodies selectively neutralizes the activity of these autoantibodies in vivo.
MATERIALS AND METHODS Patients Sera, clinical data, and approval to receive IVIG were obtained, with informed consent, from each patient. Patient 1 was a 66-year-old woman with a history of seropositive primary SS for 36 years. Apart from keratoconjunctivitis sicca, symptoms of bladder irritability/ detrusor hyperactivity dominated the clinical presentation. Patient 2 was a 36-year-old woman with celiac disease for 8 years, whose serum contained IgA anti– tissue transglutaminase (anti-TTG) antibodies. Despite a gluten-free diet with normalization of small bowel mucosa, 10 to 12 loose bowel motions a day persisted. Urinary frequency was also a feature. Patient 3 was a 35-year-old man with anti–Jo-1–positive dermatomyositis diagnosed 1 year before. Diarrhea commenced 6 months after diagnosis, along with urinary frequency and urgency. Treatment with prednisolone, methotrexate, azathioprine, and mycophenolate mofetil failed to control the myositis. IVIG was then provided therapeutically. All patients received infusions of 400 mg/kg per day of IVIG (Intragam P, CSL, Melbourne, Australia) on five consecutive days. Blood was collected
Neutralization of Muscarinic Receptor Autoantibodies
before and 1 month after IVIG treatment for conventional and functional autoantibody analysis. Functional Assay on Bladder Smooth Muscle Male Balb/c mice were killed by cervical dislocation and the bladder was removed. The dome of the bladder was dissected and divided into two strips after removal of the mucosa. Strips were mounted in 5-ml jacketed organ baths containing Krebs solution at 37°C, gassed with 95% O2/5% CO2. Guanethidine (3 M; Sigma, St. Louis, MO) and hexamethonium (100 M, Sigma) were present throughout all experiments to block transmitter release from sympathetic neurons and nicotinic ganglionic transmission, respectively. At the beginning of each experiment, preparations were desensitized to capsaicin (10 M; Sigma) to inactivate sensory neurons. Concentration-response curves to carbachol (Sigma) were produced by the cumulative addition of 0.1, 0.3, 1, 3, 10, 30, and 100 M carbachol at 75-second intervals [7]. Contractile responses were measured with MLT0201 force transducers connected to a PowerLab/8SP data acquisition system (ADInstruments, Sydney, Australia) and analyzed by Chart software version 4.0 (ADInstruments) on an eMac computer. The IgG fraction from patient serum was purified on a protein A–Sepharose column [15] (Sigma) by using serum taken before and 4 weeks after IVIG treatment. IgG was added to organ baths to give a final concentration of 1 mg/ml. After a 60-minute incubation, the effect of patient IgG was tested on carbachol-evoked responses of smooth muscle. Responses to carbachol are expressed as a percentage of the maximal response to carbachol in that preparation. Results are expressed as mean ⫾ SEM. The effect of IgG and agonist on responses of bladder muscle to carbachol were analyzed by two-way analysis of variance. Conventional Autoantibody Assays Serum from each patient was collected before and 4 weeks after IVIG treatment and tested for anti-La (SS-B) (patient 1), anti-TTG (patient 2), and anti–Jo-1 (patient 3). Affinity-purified recombinant human La was diluted to 2 g/ml in 0.03 M carbonate, pH 9.6, and coated onto enzyme-linked immunosorbent assay (ELISA) plates (Nunc, Roskilde, Denmark) for 1 hour at 37°C. Wells were blocked with 200l of 1% bovine serum albumin for 1 hour at 37°C. A total of 100 l of 1/500 dilution of the patient’s sera was added to the wells for 1 hour at 37°C. The plates were washed four times with phosphate-buffered saline– 0.05% Tween 20, and bound IgG was detected by alkaline phosphatase– conjugated goat anti-human IgG. Results were expressed as OD405 units [16]. IgA anti-TTG titers were assayed by QUANTA Lite tTG
413
(INOVA Diagnostics, San Diego, CA), and IgG anti–Jo-1 titers were assayed with RELISA (Immunoconcepts, Sacramento, CA). Results for both were expressed in ELISA units. RESULTS Smooth Muscle Assays Purified IgG taken from the three patients with primary SS, celiac disease, and dermatomyositis before administration of IVIG produced a noncompetitive inhibition of carbachol-induced contraction of detrusor muscle strips compared with control response curves (p ⬍ 0.001 by two-way analysis of variance for each patient), consistent with the presence of a functional autoantibody directed against M3Rs on the postsynaptic smooth muscle membrane [7] (Figure 1a,c,e). We have reported a similar pattern of anti-M3R antibody activity in patients with primary SS, RA with secondary SS, and scleroderma [7, 8]. Four weeks after IVIG, the patient IgG fractions no longer inhibited carbachol-induced detrusor contractions mediated by M3Rs, indicating that the IVIG had neutralized the functional activity of the inhibitory antiM3R autoantibodies in vivo (Figure 1b,d,f ). Conventional Autoantibody Assays To exclude a nonspecific general inhibitory effect of the IVIG on circulating autoantibodies, we measured levels of conventional IgG marker autoantibodies in the three patients. Levels of specific anti-La (SS-B) (primary SS), anti-TTG (celiac disease), and anti–Jo-1 (dermatomyositis) were essentially unchanged before and after the administration of IVIG (Table 1). Therefore, the IVIGinduced reversal of the inhibitory anti-M3R activity in vivo appears specific and is most likely to be mediated by antiidiotypes present in IVIG, as demonstrated earlier by in vitro mixing and depletion experiments [14]. Preliminary clinical data suggested an improvement in symptoms at 1 month after IVIG. The patient with primary SS had an improved score on the urinary urgency question on the American Urological Association 7 [15] symptom index and a modest improvement in salivary flow, although she did not report an improvement in sicca symptoms per se. The patients with celiac disease and dermatomyositis exhibited improvement in diarrhea and urinary frequency scores, consistent with an improvement in gut and bladder function (Table 1). DISCUSSION Anti-M3R autoantibodies can be considered part of a family of functional autoantibodies that bind to a receptor or ion channel and modify its function by activating it (agonistic antibody) or inhibiting it (antagonistic antibody). In the case of anti-M3R, the monovalent anti-
414
A.J. Smith et al.
FIGURE 1 Neutralization of anti-M3R autoantibodies after in vivo administration of IVIG. Concentration-response curves to carbachol were produced in strips of bladder smooth muscle before (solid squares) and after (solid triangles) the addition of patient IgG (1 mg/ml). (a, c, e) Carbachol-evoked contractile responses were significantly inhibited by IgG from each patient before IVIG treatment, consistent with functional antiM3R antibodies (p ⬍ 0.001; two-way analysis of variance). (b, d, f ) In contrast, IgG taken 4 weeks after IVIG from the three patients did not significantly inhibit contractions, indicating in vivo neutralization of anti-M3R antibodies (p ⬎ 0.05; two-way analysis of variance). Results are expressed as the mean ⫾ SEM of 3 experiments for each patient.
body retains full activity, indicating that receptor crosslinking is not required. Prototypes of functional autoantibodies are those against the thyroid-stimulating hormone receptor in Graves disease and the skeletal muscle acetylcholine receptor in myasthenia gravis. As opposed to classical high-titer autoantibodies against intracellular targets such as Ro(SS-A) and La(SS-B), antiM3R and other functional autoantibodies are typically present at low serum concentrations, reactive with highly conformational epitopes under physiological conditions,
and are effective in passive transfer experiments. As such, anti-M3R autoantibodies are likely to be of fundamental pathogenic importance in primary and secondary SS. These autoantibodies appear to be more widespread than originally thought, being reported in cases of celiac disease and dermatomyositis in the present study. It is intriguing to speculate that anti-M3R may be responsible for some of the autonomic complications and gastrointestinal manifestations in these diseases [18, 19]. Larger studies are required to determine the true prevalence of anti-M3R autoantibodies in celiac disease and dermatomyositis. Given the clinical importance of anti-M3R autoantibodies in a spectrum of autoimmune diseases, further investigation of immunological approaches to reverse their pathogenetic effects on the bladder, gut, and other organs is warranted. In the present study, a single therapeutic dose of IVIG to patients has produced neutralization of IgG-mediated inhibition of cholinergic neurotransmission. The neutralizing activity of the IVIG, most likely by an antiidiotypic mechanism as previously demonstrated in vitro, suggests that anti-M3R autoantibodies act by a reversible functional process and do not induce irreversible cytotoxic destruction through Fc-mediated effector mechanisms. This is supported by passive transfer experiments, which reveal a compensatory upregulation of M3R numbers in the bladder but no evidence of inflammation or damage at 6 weeks after injection of patient IgG [9]. Our preliminary studies demonstrate that IVIG is a promising treatment for bladder, gut, and other autonomic complications in SS and other autoimmune diseases with anti-M3R autoantibodies, and demonstrate for the first time in patients that administration of IVIG can specifically neutralize a functional human autoantibody. Randomized controlled trials are now required to confirm the beneficial effect of IVIG on parasympathetic dysfunction in these different diseases. Future studies will be directed at isolating or cloning the specific neutralizing antiidiotypic antibody, which is
TABLE 1 Conventional autoantibodies and symptoms before and after IVIGa Disease
Autoantibody/Clinical features
Pre IVIG
Post IVIG
Primary Sjögren syndrome
anti-La n ⬍ 0.18 OD units USF (ml/15 min) Urinary urgency (AUA) anti-TTG n ⬍ 20 ELISA units Diarrhea (bowel motions per day) Urinary frequency (AUA) anti–Jo-1 n ⬍ 20 ELISA units Diarrhea (bowel motions per day) Urinary urgency (AUA)
1.7 3.5 5 41.3 10–12 ⫻ loose 4 117 8–10 ⫻ loose 3
2 5.5 2 52.7 1 ⫻ formed 2 118 1–2 ⫻ formed 1
Celiac disease Dermatomyositis
a Abbreviations: USF ⫽ unstimulated salivary flow in 15 minutes; AUA Score ⫽ American Urological Association score of symptom severity; TTG ⫽ tissue transglutaminase; IVIG ⫽ intravenous immunoglobulin.
Neutralization of Muscarinic Receptor Autoantibodies
415
3.
FIGURE 2 (a) Network of naturally occurring antiidiotypic antibodies regulates the expression of functional autoantibodies against neuronal receptors and ion channels. (b) Loss of coordinated expression of neutralizing antiidiotypic antibodies allows binding of pathogenic autoantibodies to M3Rs (Sjögren syndrome) and L-type voltage-gated calcium channel (VGCCs) (type 1 diabetes) with subsequent smooth muscle dysfunction.
4.
5.
likely to be a more specific and less costly agent than pooled human preparations of IVIG. In addition to the obvious therapeutic applications, the finding of natural antiidiotypic antibodies in healthy humans that neutralize anti-M3R autoantibodies provides clues to the emergence of these autoantibodies in disease. We suggest that the regulation of potentially pathogenic functional autoantibodies by natural antiidiotypic antibodies may be a more general phenomenon in health and autoimmune disease. For example, we have also identified antiidiotypes in IVIG that neutralize the functional activity of a novel calcium channel–stimulating autoantibody in type 1 diabetes [20, 21]. Mixing experiments with type 1 diabetic IgG and SS IgG have revealed that the antiidiotypic antibody that neutralizes the diabetic autoantibody is distinct from the species that neutralizes anti-M3R (M. Jackson, unpublished data). We speculate that there is a network of naturally occurring antiidiotypic antibodies circulating in normal individuals that prevents the emergence of potentially pathogenic functional autoantibodies against neuronal receptors and ion channels (Figure 2a). As a corollary of this, an alteration of the coordinated expression of an antiidiotypic antibody and its corresponding functional autoantibody allows binding of functional autoantibodies to their targets with pathophysiological consequences (Figure 2b). The mechanisms leading to such a disturbance in the human B-cell repertoire are presently unknown. The nonobese diabetic (NOD) mouse, which also develops functionally active anti-M3R autoantibodies that are not found in normal littermates, may be a useful animal model to address these questions. REFERENCES 1. Jonsson R, Haga H-G, Gordon T: Sjögren’s syndrome. In Koopman WJ (ed): Arthritis and Allied Conditions—A Textbook of Rheumatology, 14th ed. Philadelphia, Lippincott, Williams and Wilkins, 2001. 2. Hocevar A, Tomsic M, Praprotnik S, Hojnik M, Kveder T,
6.
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
Rozman B: Parasympathetic nervous system dysfunction in primary Sjögren’s syndrome. Ann Rheum Dis 62:702, 2003. Kovács L, Papós M, Takács R, Róka R, Csenke Z, Kovács A, Várkonyi T, Pajor L, Pávics L, Pokorny G: Autonomic nervous system dysfunction involving the gastrointestinal and the urinary tracts in primary Sjögren’s syndrome. Clin Exp Rheum 21:697, 2003. Walker J, Gordon TP, Lester S, Downie-Doyle S, McEvoy D, Pile K, Waterman SA, Rischmueller M: Increased severity of lower urinary tract symptoms and daytime somnolence in primary Sjögren’s syndrome. J Rheum 30:2406, 2003. Bacman S, Sterin-Borda L, Camusso J, Arana R, Hubscher O, Borda E: Circulating antibodies against rat parotid gland M3 muscarinic receptors in primary Sjögren’s syndrome. Clin Exp Immunol 104:454, 1996. Gao J, Cha S, Jonsson R, Opalko J, Peck AB: Detection of anti-type 3 muscarinic acetylcholine receptor autoantibodies in the sera of Sjögren’s syndrome patients by use of a transfected cell line assay. Arthritis Rheum 50:2615, 2004. Waterman SA, Gordon TP, Rischmueller M: Inhibitory effects of muscarinic receptor autoantibodies on parasympathetic neurotransmission in primary Sjögren’s syndrome. Arthritis Rheum 43:1647, 2000. Goldblatt F, Gordon TP, Waterman SA: Antibodymediated gastrointestinal dysmotility in scleroderma. Gastroenterology 123:1144, 2002. Wang F, Jackson M, Maughan V, Cavill D, Smith AJ, Waterman SA, Gordon TP: Passive transfer of Sjögren’s syndrome IgG produces the pathophysiology of an overactive bladder. Arthritis Rheum 50:3637, 2004. Pyne D, Ehrenstein M, Morris V: The therapeutic uses of intravenous immunoglobulins in autoimmune diseases. Rheumatology 41:367, 2002. Dupond J-L, Gil H, de Wazieres B: Five year efficacy of intravenous gammaglobulin to treat dysautonomia in Sjögren’s syndrome. Am J Med 106:125, 1999. Amital H, Rewald E, Levy Y, Bar-Dayan Y, Manthorpe R, Engervall P, Sherer Y, Langevitz P, Shoenfeld Y: Fibrosis regression induced by intravenous gammaglobulin treatment. Ann Rheum Dis 62:175, 2003. Takahashi Y, Takata T, Hoshino M, Sakurai M, Kanazawa I: Benefit of IVIG for long-standing ataxic sensory neuronopathy with primary Sjögren’s syndrome. Neurology 60:503, 2003. Cavill D, Waterman SA, Gordon TP: Antiidiotypic antibodies neutralize autoantibodies that inhibit cholinergic neurotransmission. Arthritis Rheum 48:3597, 2003. Livingston M: Immunoaffinity chromatography of proteins. Methods Enzymol 34:723, 1974. Beer RG, Rischmueller M, Coates PT, Purcell AW, Keech CL, McCluskey J, Gordon TP: Nonprecipitating
416
anti-La(SSB) autoantibodies in primary Sjögren’s syndrome. Clin Immunol Immunopath 79:314, 1996. 17. Barry MJ, Fowler FJ Jr, O’Leary MP, Bruskewitz RC, Holtgrewe HL, Mebust WK, Cockett AT: The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol 148:1549, 1992. 18. Hughes AJ, Ferguson I, Rankin E, Kane K: Polymyositis as a cause of total gut failure. Ann Rheum Dis 61:305, 2002. 19. Usai P, Usai Satta P, Lai M, Corda MG, Piras E, Calcara
A.J. Smith et al.
C, Boy MF, Morelli A, Balestrieri A, Bassotti G: Autonomic dysfunction and upper digestive functional disorders in untreated adult coeliac disease. Eur J Clin Invest 27:1009, 1997. 20. Jackson M, Waterman SA, Gordon TP: Functional characterization of a calcium channel activating autoantibody specific for type 1 diabetes. Diabetes 53(Suppl 2):A210, 2004. 21. Jackson M, Gordon TP, Waterman SA: Disruption of intestinal motility by a calcium channel-stimulating autoantibody in type 1 diabetes. Gastroenterology 126:819, 2004.