BACKGROUND OF THE INVENTION Type 1 diabetes is a chronic autoimmune disease of unknown aetiology. It is characterised by destruction of cells in the endocrine pancreas, and in as many as 50% of patients, by the development of autonomic neuropathy. Indeed, diabetes is the commonest cause of autonomic failure l, but the mechanisms causing autonomic dysfunction are poorly understood.

Autonomic neuropathy in diabetes is characterised by early and widespread changes to the function of small diameter unmyelinated nerve fibres in sympathetic, parasympathetic and enteric pathways. Its clinical manifestations include gastroparesis, nocturnal diarrhoea, bladder atony, postural hypotension, persistent tachycardia, gustatory sweating, pupil abnormalities and erectile dysfunction 1t2. Once clinical manifestations of autonomic neuropathy occur, the estimated 5-year mortality is approximately 50%, worse than for any other complication of diabetes 3. Recent studies suggest that autonomic neuropathy is likely to underlie sudden death in type 1 diabetes'-. Nevertheless, despite its clinical importance, few studies have addressed the mechanisms underlying autonomic neuropathy in diabetes; this may reflect the relative difficulty of studying small, unmyelinated autonomic fibres, as opposed to large diameter, myelinated fibres that are affected in diabetic sensorimotor neuropathy and have been studied extensively e. g. 6t7&num 8.

Although type 1 diabetes is an autoimmune disease, an autoimmune-mediated cause of autonomic neuropathy has received relatively limited attention. Lymphocytic infiltrates occur around unmyelinated nerves and autonomic ganglia in some diabetic patients 9, complement-fixing antibodies are detected to autonomic nerves in 10-30% of patients with type 1 diabetes'oil and autoantibodies are produced to the neuronal and islet cell antigen, glutamic acid decarboxylase (GAD) in 75-84% of patients with

recent onset type 1 diabetes 12. However, there is a poor correlation between the presence of anti-GAD autoantibodies and autonomic neuropathy 13, and complement- fixing autoantibodies can also be found in patients with type 2 diabetes l0 ll. GAD is an intracellular enzyme and anti-GAD autoantibodies would therefore not be expected to have a functional effect on intact, living cells. Similarly, other autoantibodies such as anti-IA-2 (ICA512) autoantibodies bind to intracellular targets and are unlikely to play a role in disease pathogenesis, although they are useful disease markers.

Identification of an autoantibody with a functional effect on intact cells leading to altered autonomic function would represent a major advance in understanding diabetic autonomic neuropathy and provide new approaches to the diagnosis and monitoring of disease, and to treatment.

The present inventors recently developed a new approach using physiological assays on intact tissues to identify novel autoantibodies that interfere with autonomic neurotransmission through a functional effect on neurotransmitter receptors or ion channels"'. These techniques have enabled the identification of functional autoantibodies in patients with the systemic autoimmune diseases, scleroderma and Sjogren's syndrome, that block muscarinic M3-receptors in organs innervated by parasympathetic and enteric neurons 14/15. The present inventors have now developed an even more sensitive assay to identify novel functional autoantibodies using an intact isolated colon undergoing spontaneous, neuronally mediated migrating motor complex (MMC) activity. This physiological approach has the advantage of allowing effects of autoantibodies on different activation and inactivation states of ion channels and receptors to be detected, which is not possible with more traditional approaches.

Such a physiological approach has not previously been used to investigate whether functional autoantibodies are produced in type 1 diabetes that alter autonomic neuroeffector transmission and contribute to the autonomic dysfunction.

SUMMARY OF THE INVENTION In a first aspect the present invention provides a method of treating type 1 diabetes neuropathy in a subject, the method comprising administering to the subject an effective amount of a composition comprising a ligand which binds to L-type voltage gated calcium channels or to a diabetic autoantibody, wherein the diabetic

autoantibody is directed against L-type voltage gated calcium channels, the binding of the ligand to the autoantibody or L-type voltage gated calcium channel ameliorating the affect of the autoantibody on the L-type voltage gated calcium channel.

In a second aspect the present invention provides a composition, the composition comprising a ligand which binds to L-type voltage gated calcium channels or to a diabetic autoantibody, wherein the diabetic autoantibody is directed against L-type voltage gated calcium channels, the binding of the ligand to the autoantibody or L- type voltage gated calcium channel ameliorating the affect of the autoantibody on the L-type voltage gated calcium channel.

In a third aspect the present invention provides a diagnostic assay, the assay comprising detecting the presence of a diabetic autoantibody, wherein the diabetic autoantibody is directed against L-type voltage gated calcium channels.

In a fourth aspect the present invention provides a method of treating diabetic neuropathy in a subject, the method comprising detecting the presence of the diabetic autoantibody in the subject and if present administering to the subject a composition according to the second aspect of the present invention.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 Immunoglobulin G from type 1 diabetics disrupts migrating motor complex activity recorded in mouse isolated colon. a, Normal pattern of migrating motor complex (MMC) activity recorded using 4 force transducers placed along the colon from proximal (top trace) to distal (bottom trace). The spontaneous contractile complex occurs every 4-6 minutes, separated by periods of relative quiescence. b, The pattern of MMC activity is disrupted by IgG (0.2 mg/ml) from a patient with type 1 diabetes. c, IgG from a healthy control has no effect on MMCs. d, Heterochronic contractions (contractions occurring out of phase with the MMCs) were evoked only by diabetic IgG and are indicated by asterisks. e, Diabetic IgG also evoked ectopic contractions (marked by asterisks) i. e. isolated contractions occurring in a region other than the proximal colon. f, Following addition of diabetic IgG, some MMCs aborted

before propagating the entire length of the colon (aborted MMC marked by asterisks).

Time bars represent 1000 s (a, c, e, f), 3000 s (b) and 1500 s (d).

Figure 2 The dihydropyridine class of drugs can mimic or reverse the effect of diabetic IgG on MMC activity. a, The dihydropyridine agonist, Bay K8644 (10 nM) mimics the effect of 0.2 mg/ml IgG from type 1 diabetics. Heterochronic contractions occur in the proximal colon and ectopic contractions in the distal colon (marked by asterisks) in the presence of Bay K8644 (right panel) but not in the untreated control colon (left panel). b, The effect of diabetic IgG (left panel) is reversed by the addition of the dihydropyridine antagonist, nicardipine (1 nM : right panel). Time bars represent 1500 s (a) and 1000 s (b).

Figure 3 IgG from type 1 diabetics competitively inhibits the effect of nicardipine on MMCs. a, Nicardipine causes a concentration-dependent inhibition of the amplitude of MMCs (solid curve, filled squares). The concentration-response curve is shifted to the right by the dihydropyridine agonist, Bay K8644 (2 nM: broken curve, open triangles). b, IgG from 3 patients with type 1 diabetes that disrupted MMCs caused a parallel shift to the right of the nicardipine concentration-response curve.

The average for the 3 patients is shown (broken curve, open triangles) compared with the response in the presence of disease control IgG (solid line, filled squares). c, IgG from 3 patients that did not disrupt MMC activity also had no effect on the nicardipine concentration-response curve. The effect of nicardipine in the presence of disease control IgG is shown by the solid curve (filled squares), and in the presence of autoantibody-negative diabetic IgG by the broken curve (open triangles). do The nicardipine concentration-response curve in the presence of IgG from a patient with scleroderma (broken line, open triangles) was not significantly different from that in the presence of nicardipine alone (solid line, filled squares).

Figure 4 Diabetic IgG interferes non-competitively with the phenylakylamine calcium channel blocker, verapamil. Verapamil causes a concentration-dependent inhibition of MMC amplitude (solid curve, filled squares); the maximum inhibition is decreased and IC50 increased in the presence of diabetic IgG (broken curve, open triangles), indicating a non-competitive interaction.

Figure 5. Passive transfer of immunoglobulin G from human type 1 diabetics to mice disrupts the MMC activity of isolated colon. A. Severe MMC disruption is displayed by a colon isolated from a mouse injected with 20 mg of IgG containing the anti-VGCC autoantibody. B. The pattern of colonic MMC activity is not affected in a mouse injected with 20 mg of IgG from a type 1 diabetes patient without the autoantibody.

Figure 6. Electrical field stimulation (EFS) of mouse vas deferens. Left trace: EFS at 10 Hz produces a characteristic ATP mediated twitch and noradrenaline mediated tonic contraction in untreated mouse vas deferens. Right trace: passive transfer of IgG (20 mg) from a patient with type 1 diabetes previously demonstrated to contain anti- VGCC autoantibodies severely disrupts the twitch and tonic contraction in ex-vivo vas deferens.

Figure 7. A. Filling cystometry tracing recorded in ex-vivo bladders from IgG- injected mice. IgG (20 mg) from a patient with type 1 diabetes (solid line) previously demonstrated to contain anti-VGCC autoantibodies reduced the compliance and increased phasic contractions of bladder wall smooth muscle following passive immunisation over 48 hours. In contrast, injection of mice with 20 mg of IgG from healthy controls (dashed line) did not result in compliance changes compared to untreated bladders (doted line). Infused volume is 150 1L B. Bladder wall compliance in mice injected with IgG from patients with type 1 diabetes with the anti-VGCC autoantibody (ff=3) was significantly reduced when compared to untreated bladders.

By comparison, the compliance of bladders from mice injected with IgG from healthy c.

DETAILED DESCRIPTION The present invention provides methods and compositions for inhibiting pathology associated with Type 1 diabetes.

The present inventors have found that neuropathy associated with type 1 diabetes is caused, at least in part, by auto-antibodies that stimulate voltage gated calcium channels. Further, the present inventors have found that the effects of this stimulation

of voltage gated calcium channels can be reduced by treatment with ligands that inhibit the binding of the autoantibody to the voltage dependent calcium channel.

Accordingly, in a first aspect the present invention provides a method of treating type 1 diabetes neuropathy in a subject, the method comprising administering to the subject an effective amount of a composition comprising a ligand which binds to L-type voltage gated calcium channels or to a diabetic autoantibody, wherein the diabetic autoantibody is directed against L-type voltage gated calcium channels, the binding of the ligand to the autoantibody or L-type voltage gated calcium channel ameliorating the affect of the autoantibody on the L-type voltage gated calcium channel.

In a preferred embodiment the ligand binds to the autoantibody such that the ability of the autoantibody to bind to L-type voltage gated calcium channel is inhibited. The ligand is preferably isolated and is preferably an antibody or portion thereof. Most preferably the ligand is an antiidiotypic antibody. The ligand is preferably a monoclonal antibody. More preferably, the ligand is a human antibody. The ligand is preferably of IgG class.

In another embodiment the ligand is a peptide derived from a L-type voltage gated calcium channel, wherein the peptide comprises the epitope recognised by the autoantibody.

In yet another embodiment the ligand binds to the L-type voltage gated calcium channel such that the ability of the autoantibody to bind to L-type voltage gated calcium channel is inhibited.

In a second aspect the present invention provides a composition, the composition comprising a ligand which binds to L-type voltage gated calcium channels or to a diabetic autoantibody, wherein the diabetic autoantibody is directed against L-type voltage gated calcium channels, the binding of the ligand to the autoantibody or L- type voltage gated calcium channel ameliorating the affect of the autoantibody on the L-type voltage gated calcium channel.

In a preferred embodiment the autoantibody is directed against the dihydropyridine (DHP) binding site of alc subunits of smooth muscle L-type calcium channels.

In a preferred embodiment of the invention the composition comprises a ligand that is an antibody or portion thereof. Preferably, the ligand is an antiidiotypic antibody.

The ligand is preferably a monoclonal antibody. More preferably, the ligand is a human antibody. The ligand is preferably of IgG class.

The composition of the present invention preferably comprises a mixture of antibodies. Most preferably, the composition comprises normal human IgG. More preferably, the composition comprises normal intravenous immunoglobulin (IVIg).

In a further embodiment of the present invention the composition comprises a ligand that is a peptide derived from a L-type voltage gated calcium channel, wherein the peptide comprises the epitope recognised by the autoantibody.

In yet another embodiment of the present invention the composition comprises a ligand that binds to the L-type voltage gated calcium channel such that the ability of the autoantibody to bind to L-type voltage gated calcium channel is inhibited.

The ability of the ligand to ameliorate the affect of the autoantibody on the L-type voltage gated calcium channel may be assessed in a number of ways. For example, the effect of the autoantibody on migrating motor complex activity in mouse isolated colon may be assessed in the presence or absence of ligand.

As is well understood antibodies refer to immunoglobulin molecules comprised of four polypeptide chains, two heavy (H) chains and two light (L) chains inter- connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable region (HCVR or VH) and a heavy chain constant region. The heavy chain constant region is comprised of three domains, CH1, CH2 and CH3. Each light chain is comprised of a light chain variable region (LCVR or VL) and a light chain constant region. The light chain constant region is comprised of one domain, CL. The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDR), interspersed with regions that are more conserved, termed framework regions (FR). Each VH and VL is composed of three CDRs and four FRs, arranged from amino-terminus to carboxy-terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4.

The term"antibody portion"as used herein refers to one or more fragments of an antibody that retain the ability to specifically bind to an antigen, ie the diabetic autoantibody. It has been shown that the antigen-binding function of an antibody can be performed by fragments of a full length antibody. Examples of binding fragments encompassed within the term"antigen-binding portion"of an antibody include (I) a Fab fragment, a monovalent fragment consisting of the VL, VH, CL and CH1 domains; (ii) a F (ab') 2 fragment, a bivalent fragment comprising two Fab fragments linked by a disulfide bridge at the hinge region; (iii) a Fd fragment consisting of the VH and CH1 domains ; (iv) a Fv fragment consisting of the VL and VH domains of a single arm of an antibody; (v) a dAb fragment (Ward et al, 1989, Nature 341: 544-546) which consists of a VH domain; and (vi) an isolated complementarity determining region (CDR).

Furthermore, although the two domains of the Fv fragment, VL and VH, are coded by separate genes, they can be joined, using recombinant methods, by a synthetic linker that enables them to be made as a single protein chain in which the VL and VH regions pair to form monovalent molecules (known as single chain Fv (scFv); (see eg Bird et al, 1988, Science 242: 423-426 and Huston et al., 1988 Proc. Natl. Acad. Sci. USA 85 : 5879-5883). Such single chain antibodies are also intended to be encompassed within the term"antigen-binding portion"of an antibody. Other forms of single chain antibodies, such as diabodies or triabodies are also encompassed. Diabodies are bivalent, bispecific antibodies in which VH and VL domains are expressed on a single polypeptide chain, but using a linker that is too short to allow for pairing between the two domains on the same chain, thereby forcing the domains to pair with complementary domains of another chain and creating two antigen binding sites (see e. g. , Holliger, P. , et al 1993 Proc. Natl. Acad. Sci. USA 90: 6444-6448; Poljak, R. j., et al.

1994 Structure 2: 1121-1123.

In a third aspect the present invention provides a diagnostic assay, the assay comprising detecting the presence of a diabetic autoantibody, wherein the diabetic autoantibody is directed against L-type voltage gated calcium channels.

The presence of the diabetic autoantibody can be indicative of diabetic neuropathy in the patient. The presence of the antibody may also play a role in autonomic neuropathy, sensorimotor neuropathy, cardiac autonomic neuropathy, gastrointestinal dysmotility and other neurologic symptoms including epilepsy. Accordingly, the

presence of the diabetic autoantibody in the diagnostic assay of the present invention can be used as an indicative of a disease selected from the group consisting of diabetic neuropathy, diabetic cystopathy, autonomic neuropathy, sensorimotor neuropathy, cardiac autonomic neuropathy, gastrointestinal dysmotility, epilepsy and sexual dysfunction in males. Most preferably, the presence of the diabetic autoantibody is an indicative of diabetic neuropathy in a patient.

In the diagnostic assay of the present invention the presence of the diabetic autoantibody can be detected using an assay selected from the group consisting of migrating motor complex activity assay, muscle contraction assay, ELISA-type assay with the antigen being the L-type voltage gated calcium channels, irnmunoassay, ligand binding inhibition assay, immunoprecipitation assay and enzyme linked assay.

Preferably, the assay is a colonic migrating motor complex activity assay, vas deferens muscle contraction assay or a bladder muscle contraction assay.

The DNA sequences encoding human calcium channel oct 2-, ß-and-subunits, including splice variants of primary transcripts, are disclosed in US 5, 846, 757, the disclosure of which is incorporated herein by cross-reference.

The presence of the diabetic autoantibody can be detected in a number of ways. For example, the presence of the autoantibody can be detected using the migrating motor complex activity assay described herein. Alternatively an assay system such as described in US 5,846, 757 may be used. It is presently preferred that the assay involves detecting the presence of the autoantibody using an ELISA-type assay with the antigen being the L-type voltage gated calcium channels, in particular the oq-subunit.

Another possibility is a ligand binding inhibition assay, in which the autoantibody is assayed for its ability to competitively inhibit the binding of radiolabelled dihydropyridine to cell membranes purified from a) a cell line expressing the L-type calcium channel, or b) a cell line transfected with the dihydropyridine binding domain (Huber I, Wappl E, Herzog A, Mitterdorfer J, Glossman H, Langer T, Striessing J.

Conserved Ca-antagonist-binding properties and putative folding structure of a high- affinity dihydropyridine-binding domain. Biochem J 347, 829-836. 2000). As the

diabetic autoantibody may bind only to specific conformational forms of the channel, manipulations such as depolarisation of channel-expressing cells may be required.

Another possibility is an irnmunoprecipitation assay, in which the diabetic autoantibody is used to immunoprecipitate radiolabelled L-type voltage-gated calcium channels from a cell line expressing the channel.

Another possibility is to create an antigen mimic of the dihydropyridine binding site by raising an antibody against a calcium channel antagonist of the dihydropyridine group as described (Matalka K, El-Thaher T, Saleem M, Arafat T, Jehanli A, Badwan A.

Enzyme linked assay for determination of amlodipine in plasma. J Clin lab Anal 15, 47-53,2001). This antibody can then be used as a surrogate antigen in an ELISA-type assay or similar immunoassay. Of course, based on our current research, nicardipine would be the most suitable channel antagonist to use as the immunogen, however any antagonist against the dihydropyridine binding site, including nitrendipine or nifedipine may also prove effective.

Another possibility is to raise an anti-idiotypic antibody against the purified diabetic autoantibody and use it as a surrogate antigen. The anti-idiotypic antibody will be the image of the epitope on the receptor (the L-type voltage-gated calcium channel) and can be used as the antigen in an ELISA-type assay or similar immunoassay.

Another possibility is to create a chemical synthetic analogue of the dihydropyridine binding site and use it as an antigen in an ELISA-type assay or similar immunoassay.

In a fourth aspect the present invention provides a method of treating diabetic neuropathy in a subject, the method comprising detecting the presence of the diabetic autoantibody in the subject and if present administering to the subject the composition of the second aspect of the present invention.

The present invention also provides tolerogenic compositions that contain a voltage gated calcium channel or fragment thereof and are useful in inducing tolerance in a patient with IDDM. In particular, the invention provides compositions including a voltage gated calcium channel or tolerogenic fragment thereof, combined with a tolerogizing molecule. Various molecules are known in the art to cause, promote or

enhance tolerance. See, e. g. , U. S. Pat. No. 5, 268, 454, and citations therein, which are incorporated herein by reference. As used herein, the term"tolerogizing molecule" means a molecule, compound or polymer that causes, promotes or enhances tolerogenic activity when combined with a voltage gated calcium channel, or fragment thereof. A tolerogizing molecule can be, for example, conjugated to an L-type voltage gated calcium channel. Such tolerogizing molecules include, e. g. , polyethylene glycol and are well known in the art (see, e. g. , U. S. Pat. No. 5,268, 454).

A voltage gated calcium channel or tolerogenic fragment of a voltage gated calcium channel can be produced or synthesized using methods well known in the art. Such methods include recombinant DNA methods and chemical synthesis methods for production of a peptide. Recombinant methods of producing a peptide through expression of a nucleic acid sequence encoding the peptide in a suitable host cell are well known in the art and are described, e. g. , US 5, 846, 757.

Throughout this specification the word"comprise", or variations such as"comprises" or"comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

All publications mentioned in this specification are herein incorporated by reference.

Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed anywhere before the priority date of each claim of this application.

In order that the nature of the present invention may be more clearly understood preferred forms thereof will now be described with reference to the following non- limiting examples.

Example 1 Materials and Methods Patients Blood samples were obtained with informed consent from patients with type 1 diabetes (n=16), type 2 diabetes (n=5), other endocrine diseases: Grave's disease (n=2) and Addison's disease (n=2); systemic rheumatic diseases: primary Sjogren's syndrome with autonomic dysfunction (n=7), scleroderma with gastrointestinal dysfunction (n=3), polymyositis (n=l) and rheumatoid arthritis with secondary Sjogren's syndrome (n=2); idiopathic dysautonomia (n=l), Crohn's disease (n=l), idiopathic abdominal pain (n=2), and from healthy controls (n=4). Clinical features of patients with type 1 diabetes appear in Table 2. The study was approved by the Clinical Ethics Committee of Flinders Medical Centre.

Spontaneous intestinal motility in vitro MMCs were studied using minor modifications of previously described methods Briefly, male Balb/c mice weighing 25-35 grams were killed by raised atmospheric COZ followed by cervical dislocation. The colon was excised and placed in Krebs' solution continuously gassed with 95% °2/5% CO2, pH 7.4 at 37°C. The composition of the Krebs'solution was (mM) : NaCl 118, Ka1 4. 7, KH2PO4 1. 2, MgSO4, 1.5, NaHCO3 25.0, D-Glucose 11.0 and CaCl2 2.5. A stainless steel rod was inserted through the lumen and attached to the base of a 100 ml organ bath. The mechanical activity of the circular muscle was recorded by four force-displacement transducers attached at approximately 1.2 cm intervals to the mesenteric border of the colon under an initial tension of 6 mN. Transducer output was fed into a Quad bridge amplifier (ADInstruments, Sydney, Australia) and contraction of the circular smooth muscle recorded and stored on a Power Macintosh G3 computer via Chart v4.2 software and PowerLab/8s data acquisition system (ADInstruments).

Experimental Protocol Following a 60 minute equilibration period, five control MMCs were recorded. Patient or control IgG, purified on a protein A-sepharose column according to standard methods Sigma, St Louis, Missouri; 51, was added to give a final concentration of 0.2 mg/ml. Preliminary studies indicated that this is the optimal concentration of IgG in this assay, and reflects the assay's greater sensitivity than the experiments on tissue strips, in which 0.6-1 mg/ml was optimal 14-After a 30 minute incubation, MMCs were recorded for a 30 minute period in the continued presence of IgG. Each IgG sample was tested on at least two occasions. In some experiments, Bay K8644 (1-10 nM; Sigma) was added instead of patient IgG. To test the ability of VGCC blockers to reverse the effects of diabetic IgG or Bay K8644, nicardipine (Sigma) or verapamil (Knoll, Lane Cove, New South Wales, Australia) were added at increasing concentrations at the end of the 30 minute recording period and 5 MMCs recorded following a further 10 minute incubation for each concentration. In some experiments, samples of patient IgG were pre-incubated with an equal concentration of intravenous immunoglobulin (IVIg ; Intragam P, Commonwealth Serum Laboratories, Melbourne, Australia) or normal healthy control IgG for 45 min at 37°C prior to addition to the organ bath. MMC activity in untreated preparations was stable for the 4-5 hour duration of the experiments.

Data Analysis and Statistics Colonic MMCs were defined as regularly occurring spontaneous contractions, which alternated with periods of relative quiescence, and were observed to occur at more than one site along the colon 17. Contractions occurring only in the proximal colon out of phase with MMCs were defined as heterochronic contractions. Ectopic contractions were defined as contractions occurring in a region other than the proximal colon and out of phase with MMCs. MMC contractions which started in the proximal colon but which did not propagate through to the mid-distal and distal regions of the colon were defined as aborted MMCs.

The maximal amplitude of MMCs was measured using the peak parameters-peak amplitude function of the DataPad on Chart v4.2 software (ADInstruments). The

maximal amplitude of at least 4 consecutive MMCs was measured prior to the addition of any drugs, 30 minutes after the addition of IgG, and after the addition of increasing concentrations of nicardipine or verapamil. The percentage inhibition of MMC amplitude by nicardipine or verapamil was calculated and the resulting data analysed by non-linear regression analysis using GraphPad Prism (version 3. Oa for Macintosh, GraphPad Software, San Diego, California). Sigmoidal concentration- response curves with unitary slope and baseline set at 0% were calculated. The concentration of nicardipine or verapamil required to cause 50% inhibition of MMC maximum contraction amplitude (ICso) was determined. Results are reported as the mean ICso and the upper and lower 95% confidence intervals (CI), and the R2 (dfl value for the goodness of curve fit, where df is the degrees of freedom.

Results Migrating motor complex activity Migrating motor complex activity was significantly disrupted in the presence of IgG from each of the 7 patients with type 1 diabetes and neuropathy, but not by IgG from any of the other patients or controls.

IgG from type 1 diabetic patients with neuropathy triggered the appearance of heterochronic contractions (IgG from 7/7 patients), ectopic contractions (6/7) and aborted MMCs (6/7).

Heterochronic contractions, ectopic contractions and aborted MMCs were not observed in any of the experiments testing IgG from patients with type 1 diabetes without neuropathy, type 2 diabetes, Addison's disease, Crohn's disease or healthy controls.

Pre-adsorption of IgG from type 1 diabetic patients with neuropathy with an equal concentration of IvIg completely prevented the disruptive effect of the patient IgG on MMCs (n=2). Heterochronic contractions, ectopic contractions and aborted MMCs were not observed. Similarly, preabsorption of diabetic IgG with an equal concentration of IgG from 2/2 healthy individuals prevented the disruption of MMCs.

This suggests that anti-idiotypic antibodies to the novel diabetic autoantibody are present in the serum of healthy individuals.

Unique IgG autoantibodies are present in the blood specifically of patients with type 1 diabetes and neuropathy. These autoantibodies alone can cause profound disruption of gastrointestinal function. The design of the experiments rules out contributions due to complement fixation, apoptosis, loss of nerve growth factor support, altered glucose concentrations or glycosylated antibodies.

MMC activity is an example of a physiological behaviour controlled by small, unmyelinated nerve fibres. The profound effect of the newly discovered diabetic autoantibodies on MMC activity may extend to any neuronal function involving small, unmyelinated (autonomic or sensory) fibres. The effect of the diabetic autoantibodies can be prevented by treatment with anti-idiotype antibodies present in normal healthy control serum The present inventors have described a new class of functional autoantibodies, probably of the IgG class, that directly modify autonomic neuroeffector function in the colon in the absence of complement fixation. The time course of these effects is consistent with a pharmacological-type action of the antibodies and suggests that antibody-evoked apoptosis or loss of growth factor support are not the mechanism of action. The experiments were performed in the presence of normal glucose concentrations, demonstrating that high glucose concentration is not a factor. The effects are specific for type 1 diabetes and were not observed using immunoglobulin from patients with type 2 diabetes; the effect is thus independent of blood glucose levels in the patient, suggesting that abnormally glycosylated antibodies are not the cause.

IgG from type 1 diabetics disrupts colonic migrating motor complex activity To determine whether IgG from patients with type 1 diabetes contains a functional autoantibody that alters autonomic neurotransmission, MMCs were recorded in mouse isolated colon before and after the addition of 0.2 mg/ml purified IgG (final concentration). Spontaneous, regular MMCs were recorded as contractions migrating from proximal to distal colon at intervals of 4-6 minutes, separated by periods of

quiescence, during which inhibitory motor activity occurs (Figure la)",". Tlus activity was profoundly disrupted by purified IgG from 8 of 15 patients with type 1 diabetes (Figure lb) but not by 0.2 mg/ml IgG from healthy controls (n=4; Figure 1c).

The typical pattern of MMCs was lost in the presence of diabetic IgG, and a number of new features became evident: heterochronic contractions (Figure la) occurred at the proximal recording site out of phase with the MMCs, during the normally quiescent phase; abnormal, isolated, non-propagating contractions occurred at sites other than the proximal lead (ectopic contractions: Figure le) and some MMCs aborted before propagating the entire length of the colon (Figure 1 complement-dependent mechanism for this effect can be ruled out; similarly, an effect of high glucose concentrations can be ruled out because the experiments were performed in Kreb's solution with a normal glucose concentration (11 mM).

To confirm that the disruptive activity of IgG from type 1 diabetics is disease specific, IgG from patients with other endocrine or autoimmune diseases, or with gastrointestinal or autonomic disorders was tested. Heterochronic contractions, ectopic contractions and aborted MMCs were not observed in any of the experiments testing IgG from patients with type 2 diabetes (n=5), Grave's disease (n=2), Addison's disease (n=2), primary Sjogren's syndrome with autonomic dysfunction (n=7), scleroderma with gastrointestinal dysfunction (n=3), rheumatoid arthritis with secondary Sjogren's syndrome (n=2), polymyositis (n=l), idiopathic dysautonomia (n=1), Crohn's disease (n=1) and idiopathic abdominal pain (n=2). The lack of effect of IgG from patients with type 2 diabetes rules out a non-specific effect of glycosylated antibodies. Thus some patients with type 1 diabetes produce a novel functional autoantibody that modifies autonomic neurotransmission in the colon.

A calcium channel agonist mimics, and an antagonist reverses the effect of the diabetic autoantibody To understand the mechanism of disruption of the MMC activity by the novel functional autoantibody, we attempted to mimic the autoantibody effect pharmacologically. Based on our experience using receptor and ion channel modulators in the MMC assay (unpublished data and 17), we predicted that the effect of the diabetic IgG should be mimicked by a drug that opens L-type voltage-gated

calcium channels (VGCCs) in the colonic smooth muscle. These channels are sensitive to the dihydropyridine (DHP) class of calcium channel activators and inhibitors 19 and are crucial for contraction. We therefore tested increasing concentrations of the DHP agonist, Bay K8644, on the MMC preparation and found that concentrations of 2-10 nM Bay K8644 produced effects similar to those of 0.2 mg/ml diabetic IgG (Figure 2a).

If an agonist at L-type VGCCs is able to mimic the effect of diabetic IgG, then a DHP blocker of L-type VGCCs should reverse the effect of the IgG, just as it reverses the effect of Bay K8644. To test this hypothesis, 1 nM nicardipine was added to the MMC preparations 30 minutes after the addition of patient IgG. Nicardipine reversed the disruptive effect of diabetic IgG from each of 3 patients tested (Figure 2b).

The diabetic autoantibody binds to the DHP binding site on L-type VGCCs We next investigated whether the novel functional autoantibody was acting by binding directly to the DHP binding site on L-type VGCCs, or to another site. We tested this by determining whether the diabetic autoantibody competitively or non- competitively altered the effect of the DHP antagonist, nicardipine, on MMCs. The DHP binding site is fully conserved between human and mouse (refs).

In normal mouse colon, nicardipine concentration-dependently inhibits the amplitude of MMCs with an IC50 of 1. 8 nM (Figure 3a, solid line). Pretreatment of the colon with 2 nM Bay K8644 (a concentration that mimicked the effects of diabetic IgG) caused a significant parallel shift of the nicardipine concentration-response curve to the right (Figure 3a, broken line), confirming that the agonist and antagonist interact competitively at the DHP binding site and that more nicardipine is required to inhibit MMC contractions in the presence of the agonist. The IC50 for nicardipine in the presence of 2 nM Bay K8644 was 2. 8 nM (Table 1). If diabetic IgG mimics Bay K8644 and acts at the DHP binding site to enhance the opening of L-type VGCCs, then in colon treated with the diabetic IgG there should also be a parallel shift to the right of the nicardipine concentration-response curve and an increase in the IC50. IgG from each of 3 patients whose IgG was previously shown to disrupt MMCs caused a parallel shift to the right of the nicardipine concentration-response curve, resulting in a significant increase in the IC50. The magnitude of this increase varied from patient to patient (Table 1) and is a measure of the activity of that patient's autoantibody.

Pooling the data for these patients showed an average increase in the IC50 to 3.2 nM (95% CI 2.7-3. 7 nM ; Figure 3b). By contrast IgG from each of 3 patients that did not disrupt MMC activity, did not have any significant effect on the nicardipine concentration-response curve or IC50 (Figure 3c ; Table 1). Thus IgG from some type 1 diabetics contains an autoantibody that competitively inhibits DHP binding to L-type VGCCs.

To confirm that the shift in the nicardipine concentration-response curve is a specific marker of the autoantibody activity in diabetic IgG, identical experiments were performed using IgG from disease controls. IgG from patients with type 2 diabetes, Sjogren's syndrome with autonomic dysfunction or scleroderma with gastrointestinal dysfunction (Figure 30 did not significantly increase the IC50 for nicardipine (Table 1).

Since the diabetic autoantibody appears to act specifically at the DHP binding site, we tested whether the autoantibody would cause a non-competitive inhibition of the action of an antagonist acting at a different site on the L-type VGCC. The phenylalkylamine class of drugs block L-type VGCCs through an action at a site distinct from the DHP binding site 19. In the presence of disease control IgG, the IC50 for the phenylalkylamine, verapamil, was 36 nM (95% CI 21-63 nM; Rgo) = 0.79) (Figure 4, solid line) and the maximum inhibition of MMC amplitude was 95 + 5%. Diabetic IgG non-competitively inhibited the effect of verapamil ; the maximum inhibitory effect was decreased to 85 4% and the IC50 increased to 50 nM (95% CI 31- 82 nM; R2 (25) = 0.89) (Figure 4, broken line). This confirms that the novel functional diabetic autoantibody acts at the DHP binding site and modifies the effect of verapamil only through an allosteric (i. e. non-competitive) interaction.

Passive transfer of the diabetic autoantibody causes intestinal dysmotility Having demonstrated the presence of a novel autoantibody in diabetic IgG that has functional effects in vitro, the question arises whether the autoantibody has access to the L-type calcium channel in vivo and is able to induce dysfunction. Passive transfer is crucial to confirm a specific pathogenic role of the autoantibody 20. We therefore undertook passive transfer experiments in which mice were injected intraperitoneally with 10 mg purified IgG on each of two consecutive days, sacrificed on the third day, and colons removed for measurement of MMC activity. Colons from mice injected with diabetic IgG that contains the novel diabetic autoantibody showed profound disruption of MMC activity; by contrast normal MMC activity was recorded in the colon from mice injected with either diabetic IgG that lacked the novel autoantibody or control IgG.

Table 1 Nicardipine inhibition of MMCs following acute exposure to diabetic or control IgG in vitro IgG/drug ICSO (nM) 95% CI (d for IC50 Untreated control 1.8 1.4-2. 2 0.94 (29) Healthy control IgG Pooled disease controls 1 1.5 1.2-1. 8 0. 88 (77) Bay K 8644 (2 nM) 2. 8* 2.3-3. 4 0.96 (14) Type 1 diabetes D1 4. 5# 3.3-6. 2 0.94 (20) D4 2. 5# 2. 1-2. 9 0. 98 (17) D17 1.5 1.1-1. 8 0.96 (20) D18 2. 7# 2. 1-3. 4 0.97 (17) D19 0.89 0.57-1. 4 0.86 (17) D20 1.3 0.94-1. 7 0.94 (20) D21 Type 2 diabetes D9 0. 98 0.71-1. 3 0.93 (17) D14 1.2 0.94-1. 6 0.96 (17) Primary Sjogren's syndrome S6 2.1 0. 89-5. 2 0.52 (8) S7 1.2 0.82-1. 9 0.80 (11) Scleroderma Scl2 1.8 1. 1-2. 8 0.90 (17) * Significantly different from IC50 in untreated control # Significantly different from IC50 in pooled disease control ^ Significantly different from IC50 in healthy control 1 pooled results of experiments using primary Sjogren's syndrome and scleroderma IgG Table 2 Clinical and serological features of patients with type 1 diabetes<BR> Px Age (yrs) / Novel diabetic IA2/GAD Neurological symptoms Nephropathy Gastrointestinal Other<BR> sex autoantibody<BR> D1 56/M + -/- Large fibre neuropathy + Gastroparesis* Postural hypotension,<BR> retinopathy,<BR> nephropathy, Pupillary<BR> abnormalities<BR> D2 48/F + +/+ Large fibre neuropathyl + Gastroparesis -<BR> D3 24/M - -/+ - - - -<BR> D4 45/M + -/-Large fibre neuropathy, - Gastroparesis -<BR> epilepsy<BR> D5 64/F + -/- Large fibre neuropathy - Constipation -<BR> D8 51/F + +/- Large fibre neuropathy - Gastroparesis, -<BR> Diarrhoea Table 2 Clinical and serological features of patients with type 1 diabetes<BR> Px Age (yrs)/Novel diabetic IA2/GAD Neurological symptoms Nephropathy Gastrointestinal Other<BR> sex autoantibody<BR> D10 61/M - -/- Large fibre neuropathy - - Retinopathy,<BR> Neuropathic joints<BR> D12 54/M + -/- Large fibre neuropathy + Gastroparesis Retinopathy,<BR> Postural hypotension<BR> D13 74/M +? ND/ND - - - -<BR> D15 55/M - -/- Large fibre neuropathy + - Postural hypotension<BR> D16 34/F - +/+ - - - Retinopathy<BR> D17 20/M - ND/ND - - - -<BR> D18 75/M + ND/ND Large fibre neuropathy + Gastroparesis Postural hypotension<BR> D19 47/F - ND/ND - - - - Table 2 Clinical and serological features of patients with type 1 diabetes<BR> Px Age (yrs)/Noval diabetic IA2/GAD Neurological symptoms Nephropathy Gastrointestinal Other<BR> sex autoantibody<BR> D20 60/M - ND/ND Large fibre neuropathy - - Retinopathy<BR> D21 63/M ?- ND/ND Large fibre neuropathy - - Retinopathy<BR> Peripheral neuropathy refers to a large fibre neuropathy with a loss of vibration sense. ND:not tested.<BR> <P>*defined on the basis of symptoms of abdominal bloating or fullness 21

Neutralization of functional diabetic autoantibodies with IVIg The characterisation of a novel functional autoantibody in diabetes suggests new approaches to the treatment of diabetic symptoms. As an alternative to interfering with the autoantibody with drugs acting directly or indirectly on the L-type VGCC, we tested whether intravenous immunoglobulin (IVIg) is able to neutralise the effect of the autoantibody in vitro. Pre-adsorption of IgG from type 1 diabetic patients with the autoantibody with an equal concentration of IVIg completely prevented the disruptive effect of the patient IgG on MMCs (n=2), probably through an idiotype- anti-idiotype interaction. Heterochronic contractions, ectopic contractions and aborted MMCs were not observed. IVIg alone at a total concentration of 0.4 or 0.6 mg/ml had no effect on MMCs. This provides a rationale for a trial of IVIg to ameliorate gastrointestinal and autonomic dysfunction in patients with type 1 diabetes and the novel autoantibody.

Similarly, preabsorption of diabetic IgG with an equal concentration of IgG from 2/2 healthy individuals prevented the disruption of MMCs. This suggests that anti- idiotypic antibodies to the novel diabetic autoantibody are present in the serum of healthy individuals.

Discussion Diabetic autonomic neuropathy is a common but poorly understood complication of type 1 diabetes. It is assumed to be multifactorial, with hyperglycaemia, microangiopathy, deficiency of growth factors and autoimmune destruction contributing. However, strong evidence for a role for these factors in autonomic dysfunction is lacking. Thus autonomic neuropathy can occur in the face of excellent glycaemic control levels of glycated haemoglobin do not always correlate with neuropathic symptoms reversal of chronic hyperglycaemia does not reverse autonomic dysfunction 29 and nerve growth factor therapy has not been successful in treating human diabetic neuropathy 30.

Functional autoantibodies that act at ion channels or receptors and disrupt autonomic or cardiovascular function have been described in autoimmune diseases including

Lambert-Eaton myasthenic syndrome 31 scleroderma 14 and Sjögren's syndrome l5.

Autoantibodies specifically to L-type VGCCs have been proposed as pathogenic in amyotrophic lateral sclerosis 32 and in Sjogren's syndrome 33 although these results have not been confirmed 16/34. Serum from patients with type 1 and 2 diabetes may contain a factor, possibly IgG, that causes complement-independent 6 or complement- dependent"apoptosis of cultured neuroblastoma cells after incubation for 24-96 hours; however the target was not identified. An IgM autoantibody in type 1 diabetes has also been described that does not have a direct effect on L-type VGCCs in pancreatic-cells, but activates them indirectly via an unidentified soluble mediator to cause apoptosis 36. The precise targets of these putative antibodies are unclear, and it is unknown whether they interfere with autonomic function.

We have identified a novel IgG autoantibody in patients with type 1 diabetes that acts as an agonist specifically at the DHP, and not the phenylalkylamine, binding site of L-type VGCCs. The autoantibody mimics the effect of Bay K8644 and is therefore likely to bind and stabilise preferentially the open channel state of the L-type VGCC The autoantibody can be detected using our new physiological assay on isolated whole colon, in which there is regular, spontaneous neuronal and muscle activity with ongoing activation of the L-type VGCCs. However, the requirement for ongoing activation of L-type VGCCs is not met in more conventional assays and we were therefore unable to detect the autoantibody immunohistochemically by binding of the autoantibody to sections of frozen human pancreas or to fixed mouse colon, by immunoblotting mouse colon lysates, by immunoprecipitation using lysates from cell lines expressing L-type VGCCs, or by fluorescence activated cell sorting analysis of human IgG binding to cell lines expressing L-type VGCCs (data not shown). This highlights the fundamental difference between autoantibodies with functional effects at ion channels and receptors, and marker autoantibodies to intracellular antigens.

Because the functional autoantibody is not detectable by routine methods, we had to develop an alternative, pharmacological approach utilising competitive versus non- competitive interaction studies of calcium channel activators and inhibitors with the autoantibody at previously well-defined 37~39 highly specific binding sites on the L- type VGCC.

L-type VGCCs are characterised by a pore-forming 1 subunit containing the DHP binding site that is absent in non-L-type VGCCs. Four main isoforms of L-type 1 subunit exist : α1C, α1D, α1F and α1S, 37,38; tissue-specific isoforms of the oclc subunit are present in smooth muscle, cardiac muscle and brain; the (Xig isoform is specific to skeletal muscle and the tothe retina, whereas the OC1D isoform is found predominantly in neuroendocrine cells and neurons 3'38. Our pharmacological data indicate that the novel functional diabetic autoantibody recognises specifically the DHP binding site on the oh subunit of L-type VGCCs in colonic smooth muscle, i. e. the smooth muscle-specific isoform of the Kic subunit \ The novel functional autoantibody caused significant disruption of gastrointestinal motilityboth in vitro, and following passive transfer in vivo, demonstrating that the autoantibody has access to the L-type calcium channel under physiological conditions. We therefore propose that the autoantibody is likely to cause similar disruption of cc subunit-dependent smooth muscle contraction in the gastrointestinal and genitourinary tracts and blood vessels in diabetic patients, contributing to the pathogenesis of symptoms that reflect intestinal, bladder and vascular dysfunction. Furthermore, the DHP binding site is completely conserved between different oclc isoforms 37 SO the novel diabetic autoantibody would be expected to recognise the cardiac and brain alc subunits. If the autoantibody has access to these subunits in vivo, it may thus contribute to the characteristic cardiac changes in diabetic patients that are associated with sudden death. Whilst diabetes is not often associated with changes in CNS function, an association has been noted with epilepsy, and anti-GAD autoantibodies occur in some patients with both disorders". Interestingly one of our patients with the novel functional autoantibody also had epilepsy (Table 2), although he is anti-GAD autoantibody negative. The functional autoantibody detected in this patient mimics the L-type VGCC agonist, Bay K8644. Since Bay K8644 induces epilepsy in mice and DHP antagonists such as nicardipine can prevent some forms of seizures in mice e. g. 46 we speculate that in the presence of a leaky blood-brain-barrier, the novel functional autoantibody could be responsible for the epilepsy in our patient with type 1 diabetes. Neutralisation of the autoantibody by treatment with IVIg may be effective in treating the epilepsy in this patient, just as IVIg has proven useful in a previous case of epilepsy associated with type 1 diabetes 42 and was able to neutralise the functional effect of the autoantibody on smooth muscle in vitro.

In conclusion, unique IgG autoantibodies are present in the blood specifically of patients with type 1 diabetes. These autoantibodies cause profound disruption of gastrointestinal function by stimulating L-type VGCCs at the DHP binding site, and their effects are neutralized by IVIg and reversed by L-type calcium channel blockers.

Since L-type VGCCs are present in smooth muscles throughout the body, and in peripheral and central neurons, these newly discovered diabetic autoantibodies may also contribute to symptoms affecting other peripheral organs, sensory fibres and the CNS.

EXAMPLE 2 Materials and methods Passive transfer of IgG to mice Passive transfer of IgG to mice (8-10 week old Balb/c males) was achieved by intraperitoneal injection (maximum injected volume of 0.5 ml) with 10 mg patient or control IgG on each of two consecutive days. Mice were sacrificed by cervical dislocation 48 hours following the initial IgG injection. Bladder or vas deferentia were excised immediately following animal sacrifice and placed in Krebs solution.

Colon Assay The analysis of functional alteration of colon activity was performed ex vivo as described in example 1, except that recordings were performed on excised tissues without the addition a patient or control IgG. The omission of patient or control IgG from the assays allowed analysis of the effects of the injected IgG on colonic MMC function.

Vas deferens Assay.

Untreated male mice (n = 4) or mice injected with IgG from patients with diabetes (n = 4) were euthanased and the vas deferentia removed and mounted in 5 ml organ baths containing Krebs solution. Preparations were fixed at one end and connected by

cotton to a force-displacement transducer (ADInstruments) at the other, under a resting tension of 10 mN. Hexamethonium (500 uM) and capsaicin (10 uM) were added to block nicotinic ganglionic transmission and desensitize sensory fibres respectively (1). Following a 30 minute equilibration period, platinum ring electrodes were used to stimulate the sympathetic nerve fibres with 5 trains of 90 pulses at 10 Hz (pulse duration 0.3 ms, voltage 60V) at 90 second intervals and the resulting contractions recorded using Chart software.

Functional Assay of Intact Bladder Whole bladders, with ureters and urethra, were excised and placed in an organ bath containing Krebs solution at 37°C, gassed with 95% °2/5% CO2. Both ureters were tied off, and the urethra cannulated using a 28-gauge needle (Terumo) connected to a pressure transducer and 1 ml syringe mounted on a syringe pump (Terumo). The preparation was allowed to equilibrate for 30 minutes, followed by infusion over 5 min of 150 ul warmed Krebs solution (37°C), after which the contents of the lumen were emptied by gravity and the preparation allowed to equilibrate for 10 min. This procedure was repeated 3 times. Luminal pressure change during infusion was recorded and stored on a Power Macintosh G4 computer via Chart v4.2 software and PowerLab/8s data acquisition system (ADInstruments). Bladder compliance was calculated using Chart v4.0 data pad using volume infused divided by change in pressure. Results from patient and control groups were compared using one-way ANOVA and Dunnett's test (GraphPad Prism). In untreated bladders, there was no change in bladder compliance following multiple infusions of Krebs solution over 390 minutes (data not shown), demonstrating the stability of the preparations over the experimental time course. A significant increase in compliance was recorded in separate preparations following addition of 10 pM of the muscarinic antagonist, atropine (P= 0.03 ; paired t-test), and 10 uM capsaicin (P= 0.03 ; paired t-test) indicating that cholinergic motor neurons and sensory neurons, respectively, were active in the preparations.

Results/discussion It has been proposed that passive transfer experiments are crucial to confirm a specific pathogenic role for postulated Abs20v52. To date, anti-acetylcholine receptor antibodies in myasthenia gravis remain one of the few Abs with a demonstrated role in the pathogenic mechanisms associated with human disease53. If the diabetic anti-VGCC Abs are responsible for the symptoms of autonomic dysfunction in type 1 diabetes, passive transfer to mice should reproduce in vivo the autonomic dysfunction observed in the isolated organs. Colons from mice injected with the diabetic anti-VGCC autoantibody showed profound disruption of MMC activity (/ ? =3 separate patient IgG samples), while injection of mice with IgG from healthy individuals did not alter MMC activity (Figure 5).

Males with type 1 diabetes suffer a variety of reproductive and sexual problems, including erectile dysfunction, infertility and retrograde ejaculation54-56. The hypothesis that anti-VGCC autoantibody disrupts the function of reproductive tissues was tested using a physiological assay of vas deferens activity, in which smooth muscle contraction is mediated by L-type VGCCsl6. Passive transfer to mice (n=3 separate patient IgG samples) of diabetic IgG containing the anti-VGCC autoantibody profoundly altered the normal ATP mediated rapid contraction phase, delaying contraction recover. In addition, contractions were noted at the end of the electrical stimulation period, suggesting alterations in the threshold for smooth muscle contraction (Figure 6). There was no significant effect of IgG from diabetic patients without the anti-VGCC Ab (=3 different IgG samples tested in duplicate), of healthy control IgG (n=3 different IgG samples tested in duplicate) or of disease control IgGl6 on vas deferens contraction. It is interesting to speculate that the diabetic anti-VGCC Ab contributes to sexual dysfunction in males with type 1 diabetes, as recent evidence suggests alterations of vas deferens smooth muscle contraction results in reduced fertility57.

Bladder dysfunction, including cystopathy, is a major complication of type 1 diabetes, with an estimated prevalence of 43-87% 54, 58. Diabetic cystopathy covers a broad range of abnormalities ranging from detrusor instability with irritative symptoms to atonic bladder. Cystometry analysis is the most useful investigation of bladder detrusor

instability59. In studies using a novel, whole-bladder filling cystometry assay, we assessed the effect of passive transfer of the anti-VGCC autoantibody on bladder compliance. A significant reduction in bladder wall compliance and increased detrusor phasic contractions were detected in bladders from mice injected with 20 mg of IgG containing anti-VGCC Abs (/ ? =3) compared to untreated bladders (n=5) (Figure 7). By comparison, the compliance of bladders from mice injected with IgG from healthy individuals (n=5) did not differ to that from untreated bladders.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

References 1. Vinik, A. I. , Park, T. S. , Stansberry, K. B. & Pittenger, G. L. Diabetic neuropathies.

Diabetologia 43,957-73 (2000).

2. Heart rate variability: standards of measurement, physiological interpretation and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. CirczzvEion 93,1043-65 (1996).

3. Ewing, D. J., Campbell, I. W. & Clarke, B. F. The natural history of diabetic autonomic neuropathy. QJMed 49, 95-108 (1980).

4. Weston, P. J. & Gill, G. V. Is undetected autonomic dysfunction responsible for sudden death in Type 1 diabetes mellitus? The'dead in bed'syndrome revisited.

Diabet Med 16, 626-31 (1999).

5. McNally, P. G. , Lawrence, I. G., Panerai, R. B. , Weston, P. J. & Thurston, H.

Sudden death in type 1 diabetes. Diabetes ObesMetab 1,151-8 (1999).

6. Srinivasan, S. , Stevens, M. J., Sheng, H. , Hall, K. E. & Wiley, J. W. Serum from patients with type 2 diabetes with neuropathy induces complement-independent, calcium-dependent apoptosis in cultured neuronal cells. J CZin Xnnest 102,1454-62 (1998).

7. Pittenger, G. L. , Malik, R. A. , Burcus, N. , Boulton, A. J. & Vinik, A. I. Specific fiber deficits in sensorimotor diabetic polyneuropathy correspond to cytotoxicity against neuroblastoma cells of sera from patients with diabetes. Diabetes Care 22, 1839-44 (1999).

8. Greene, D. A. , Sima, A. A. , Stevens, M. J., Feldman, E. L. & Lattimer, S. A.

Complications: neuropathy, pathogenetic considerations. Diabetes Care 15,1902-25 (1992).

9. Duchen, L. W. , Anjorin, A. , Watkins, P. J. & Mackay, J. D. Pathology of autonomic neuropathy in diabetes mellitus. Ann Intern Med 92, 301-3 (1980).

10. Sundkvist, G. , Lind, P. , Bergstrom, B. , Lilja, B. & Rabinowe, S. L. Autonomic nerve antibodies and autonomic nerve function in type 1 and type 2 diabetic patients. I InternMed 229, 505-10 (1991).

11. Zanone, M. M., Peakman, M. , Purewal, T., Watkins, P. J. & Vergani, D.

Autoantibodies to nervous tissue structures are associated with autonomic neuropathy in type 1 (insulin-dependent) diabetes mellitus. Diabetologia 36,564-9 (1993).

12. Bieg, S. & Lemmark, A. Diabetes. in The autoimmune diseases (eds. Rose, N. R.

& Mackay, I. R. ) 431-457 (Academic Press, San Diego, 1998).

13. Zanone, M. M. et al. High prevalence of autoantibodies to glutamic acid decarboxylase in long-standing IDDM is not a marker of symptomatic autonomic neuropathy. Diabetes 43, 1146-51 (1994).

14. Goldblatt, F. , Gordon, T. P. & Waterman, S. A. Antibody-mediated gastrointestinal dysmotility in scleroderma. GastroenteroGogy 123, 1144-1150 (2002).

15. Waterman, S. A., Gordon, T. P. & Rischmueller, R. Inhibitory effects of muscarinic receptor autoantibodies on parasympathetic neurotransmission in Sjogren's syndrome. Arthritis and Rheumatism 43,1647-54 (2000).

16. Ohlsson, M. , Gordon, T. P. & Waterman, S. A. Role of anti-calcium channel and anti-receptor autoantibodies in autonomic dysfunction in Sjogren's syndrome. J Neuroimmunol 127, 127-33 (2002).

17. Brierley, S. M. , Nichols, K. , Grasby, D. J. & Waterman, S. A. Neural mechanisms underlying migrating motor complex formation in mouse isolated colon. Br Pharrnacol132, 507-17 (2001).

18. Fida, R. , Lyster, D. J., Bywater, R. A. & Taylor, G. S. Colonic migrating motor complexes (CMMCs) in the isolated mouse colon. Neurogastroenterol Motil 9, 99-107 (1997).

19. Catterall, W. A. & Striessnig, J. Receptor sites for Ca2+ channel antagonists.

Trends Pharmacol Sci 13, 256-62 (1992).

20. Drachman, D. B. Autonomic"myasthenia" : the case for an autoimmune pathogenesis. J Clin Invest 111, 797-9 (2003).

21. Jones, K. L. et al. Predictors of delayed gastric emptying in diabetes. Diabetes Care 24,1264-9 (2001).

22. Karavanaki, K. , Davies, A. G. , Morgan, M. H. & Baum, J. D. Autonomic function in a cohort of children with diabetes., fPedialrEvdocrivolMetab 10,599-607 (1997).

23. Verrotti, A. , Chiarelli, F. & Morgese, G. Autonomic dysfunction in newly diagnosed insulin-dependent diabetes mellitus children. Pediatr Neurol 14, 49-52 (1996).

24. Merio, R. et al. Slow gastric emptying in type I diabetes: relation to autonomic and peripheral neuropathy, blood glucose, and glycemic control. Diabetes Care 20, 419-23 (1997).

25. Nazhel, B. , Yetkin, I., Irkec, C. & Kocer, B. Sympathetic skin response in diabetic neuropathy. Electromyogr Clin Neurophysiol 42, 181-5 (2002).

26. Donaghue, K. C. , Fung, A. T. , Fairchild, J. M. , Howard, N. J. & Silink, M.

Prospective assessment of autonomic and peripheral nerve function in adolescents with diabetes. DiabetMed l3, 65-71 (1996).

27. Gupta, R. C. , Chittora, M. D. & Jain, A. A study of autonomic neuropathy in diabetes mellitus in relation to its metabolic control. J Assoc Physicians India 43, 464-6 (1995).

28. Tuch, P. S. , Gill, G. V. & Huddle, K. R. Autonomic neuropathy in African diabetic patients. PostyradMedJ70, 188-91 (1994).

29. St Thomas's Diabetic Study Group. Failure of improved glycaemic control to reverse diabetic autonomic neuropathy. DiabetMed 3, 330-4 (1986).

30. Apfel, S. C. et al. Efficacy and safety of recombinant human nerve growth factor in patients with diabetic polyneuropathy: A randomized controlled trial. rhNGF Clinical Investigator Group./4 284,2215-21 (2000).

31. Waterman, S. A. , Lang, B. & Newsom-Davis, J. Effect of Lambert-Eaton myasthenic syndrome antibodies on autonomic neurons in the mouse. Ann Neurol 42, 147-56 (1997).

32. Kimura, F. et al. Amyotrophic lateral sclerosis patient antibodies label Ca2+ channel alpha 1 subunit. Ann Neurol35, 164-71 (1994).

33. Xiao, G. Q. , Hu, K. & Boutjdir, M. Direct inhibition of expressed cardiac 1- and t- type calcium channels by IgG from mothers whose children have congenital heart block. Circulation 103,1599-604 (2001).

34. Arsac, C. et al. Immunoassays fail to detect antibodies against neuronal calcium channels in amyotrophic lateral sclerosis serum. Ann Neurol 40, 695-700 (1996).

35. Pittenger, G. L. , Liu, D. & Vinik, A. I. The neuronal toxic factor in serum of type 1 diabetic patients is a complement-fixing autoantibody. DirbetMed l2, 380-6 (1995).

36. Juntti-Berggren, L. et al. Increased activity of L-type Ca2+ channels exposed to serum from patients with type I diabetes. Science 261, 86-90 (1993).

37. Hockerman, G. H. , Peterson, B. Z. , Johnson, B. D. & Catterall, W. A. Molecular determinants of drug binding and action on L-type calcium channels. Avvu Rep Pharnzacol Toxicol37, 361-96 (1997).

38. Striessnig, J. Pharmacology, structure and function of cardiac L-type Ca" channels. Cell Physiol Biochem 9, 242-69 (1999).

39. Catterall, W. A. Structure and regulation of voltage-gated Ca2+ channels. Afaau Rev Cell Dev Biol 16, 521-55 (2000).

40. Liu, X., Rusch, NJ., Striessnig, J. & Sama, S. K. Down-regulation of L-type calcium channels in inflamed circular smooth muscle cells of the canine colon.

Gastroenterology 120, 480-9 (2001).

41. Lyford, G. L. et al. aic (Cav1. 2) L-type calcium channel mediates mechanosensitive calcium regulation. Am jPhysiol283, C1001-8 (2002).

42. Olson, J. A. , Olson, D. M. , Sandborg, C. , Alexander, S. & Buckingham, B. Type 1 diabetes mellitus and epilepsia partialis continua in a 6-year-old boy with elevated anti-GAD65 antibodies. Pediatrics 109, E50 (2002).

43. Lenti, C. , Bognetti, E. , Bonfanti, R. , Bonifacio, E. & Meschi, F. Myoclonic encephalopathy and diabetes mellitus in a boy. Dev Med Child Neurol 41, 489-90 (1999).

44. Peltola, J. et al. Autoantibodies to glutamic acid decarboxylase in patients with therapy-resistant epilepsy. NeuroloS, 45. Solimena, M. et al. Autoantibodies to glutamic acid decarboxylase in a patient with stiff-man syndrome, epilepsy, and type I diabetes mellitus. NV 'ngl jMed 318, 1012-20 (1988).

46. Palmer, G. C. et al. Anticonvulsant properties of calcium channel blockers in mice: N-methyl-D-, L-aspartate- and Bay K 8644-induced convulsions are potently blocked by the dihydropyridines. Epilepgia 34,372-80 (1993).

47. Ristic, H. , Srinivasan, S. , Hall, K. E., Sima, A. A. & Wiley, J. W. Serum from diabetic BB/W rats enhances calcium currents in primary sensory neurons. J Neurophysiol 80, 1236-44 (1998).

48. Wang, D. W., Kiyosue, T. , Shigematsu, S. & Arita, M. Abnormalities of K+ and Ca2+ currents in ventricular myocytes from rats with chronic diabetes. Am f Phygiol 269, H1288-96 (1995).

49. Wang, R. , Wu, Y. , Tang, G. , Wu, L. & Hanna, S. T. Altered L-type Ca2+ channel currents in vascular smooth muscle cells from experimental diabetic rats. Am J Physiol 278, H714-22 (2000).

50. Beroukas, D. et al. Up-regulation of M3-muscarinic receptors in labial salivary gland acini in primary Sjogren's syndrome. Lab Invest82, 203-10 (2002).

51. Livingston, D. M. Immunoaffinity chromatography of proteins. Methods EnzyBnoS34, 723-31 (1974).

52. Rose, N. R. & Bona, C. (1993) Immunol Today 14,426-30.

53. Toyka, K. V. , Brachman, D. B., Pestronk, A. , et al. (1975) Science 190,397-9.

54. Vinik, A. I. , Maser, R. E. , Mitchell, B. D. , et al. (2003) Diabetes Care 26,1553-79.

55. Ziegler, D. (2001) Curr Diab Rep 1,216-27.

56. Ewing, D. J. & Clarke, B. F. (1986) Diabetes Care 9,648-65.

57. Mulryan, K., Gitterman, D. P. , Lewis, C. J. , et al. (2000) Nature 403, 86-9.

58. Nijhawan, S. , Mathur, A. , Singh, V. , et al. (1993) J Assoc Physicians India 41, 565-6.

59. McLellan, A. & Cardozo, L. (2001) Int Urogynecol J Pelvic Floor Dysfunct 12, 266-70.