IDDM is an autoimmune disease. It is caused by the autoimmune system-mediated destruction of pancreatic islet beta-cells. This leads to an imbalance in the levels of the hormones insulin and glucagon which disrupts glucose metabolism, leading to the well-known symptoms of diabetes, such as high glucose levels in urine. As IDDM is an autoimmune disease, autoantigens can often be detected in the blood of individuals suffering from IDDM.

Some characterisation of the autoantigens involved in IDDM has already been achieved. In particular, two autoantigens are known to be closely associated with the development of IDDM; these are known as the 37 kDa and 40kDa autoantigens, because of their respective molecular weights. It has recently been suggested that the 40kDa autoantigen is a protein known as IA-2 (Payton et al (1995) J. Clin. Invest. 96, 1506-1511) .

Human IA-2 (hIA-2) was originally cloned in a truncated form, and given the name ICA-512 (Islet Cell Antigen 512) (Rabin et al (1994) J. Immunol. 152, 3183) . Subsequently, the complete DNA and amino acid sequences were determined and the protein was given the name IA-2 (Lan et al (1994) DNA and Cell Biology vol 13, No 5, pp 505-514) .

A mouse homologue (mIA-2) is also known (Lu et al (1994) Biophys. Biochem. Res. Commun. vol 204, No 2) . Its cDNA shows 92% identity to that of hIA-2. Two possible rat homologues have also been identified. The first of these is known as PTPLP (Kambayashi et al (1995) Biochem. J. 306, 331-335) and has 79.3% identity to hIA-2 at the amino acid level. The second has a sequence identified from a series of overlapping clones (Passini et al (1995), Proc. Natl. Acad. Sci. USA 92, 9412-9416) . This DNA shows 93% homology

to the corresponding mIA-2 sequence and 86% homology to the corresponding hIA-2 sequence.

As stated above, it has been suggested that hIA-2 is the 40kDA autoantigen associated with IDDM. Other putative IDDM-associated autoantigens inc.ude glutamic acid decarboxylase (GAD) and the islet cell autoantigens (ICA) known as ICA-12, ICA-13, ICA-208, ICA- 302, ICA-313, ICA-123, ICA-525 and ICA-505 (EP-A-0, 383, 129; EP-A-0, 569,800) . Some of these IDDM-associated antoantigens have been proposed as diagnostic reagents for screening patients for IDDM, notably GAD, ICA-512 and ICA-12 (EP-A-0 569, 800) . Whilst searching for protein tyrosine phosphatases, the present inventors unexpectedly cloned and sequenced a cDNA fragment, and then a complete cDNA, encoding part of a new probable IDDM-associated autoantigen. This has been give the name IAR (Islet Cell Antigen-Related) and may also be referred to as IAR-PTP (Islet Cell Related Protein Tyrosine Phosphatase) . Analysis of this partial sequence shows that, like the known autoantigens mentioned above, it has an extracellulatr region and an intracellular region. The partial IAR sequence encodes around 460 amino acids from the extracellular domain and the whole of the intracellular domain. It is likely that the intracellular domain is an antigenic part of the protein.

The IAR sequence identified by the inventors shows around 70% nucleotide and amino acid sequence homology to the IA-2 sequence in the intracellular region. This provides further evidence that IAR is an IDDM-associated autoantigen. In the extracellular region, the nucleotide homology to IA-2 is about 35% and the amino acid sequence homology is about 20%.

Although previous work has suggested that hIA-2, PTLTP and mIA-2 do not have protein tyrosine phosphatase (PTP) activity, the present inventors have found that hIA-2 in fact has low but detectable PTP activity. Like hIA-2, IAR

has PTP activity, and is also referred to herein as IAR-PTP (Islet Cell Antigen-Related Protein Tyrosine Phosphatase) . Further, IAR has a similar tissue expression pattern to IA-2, being highly expressed in the brain and pancreas. Thus, IAR has three characteristics of an IDDM-associated autoantigen: (i) a characteristic structure, in terms of organisation into intra- and extra- cellular domains and the sequence of its intracellular domain; (ii) PTP activity; and (iii) a characteristic tissue distribution. IAR is also reactive with IDDM patient sera. A study with a group of IDDM patients shows that recombinant IAR is precipitated by autoantibodies present in more than 50% of the serum samples while IAR does not react with any sera from control patients. Nonetheless, it is clear that IAR is distinct from previously isolated autoantigens; in particular, its sequence identity to hIA-2, mIA-2 and PTPLP is too low for it to be any of these proteins, especially as it has only 20% homology to hIA-2 at the amino acid level in the extracellular region. Further, IAR and IA-2 can distinguish different populations of IDDM autoantibodies.

The inventors have also carried out a study to compare the IDDM sensitivity and predictive value of IAR antibodies (IAR Ab) and IA-2 antibodies (IA-2 Ab) . Antibodies to IAR and IA-2, measured by radiobinding assay, were correlated in recent-onset IDDM patients. Some individuals had IAR Ab in the absence of IA-2 Ab, and others had IA-2 Ab in the absence of IAR Ab. Therefore, IAR and IA-2 can distinguish different populations of IDDM autoantibodies since they identify overlapping but non-identical sets of IDDM patients. The diagnostic sensitivity for IDDM was similar for IAR Ab and IA-2 Ab. Accordingly, IAR is clearly a useful predictive tool for the diagnosis of IDDM, either as an alternative IA-2 or in combination with IA-2. Moreover, in a group of first degree relatives of IDDM sufferers, the presence of IAR Ab were signi icantly associated with progression to IDDM while IA-2 Ab were not.

This is significant because it establishes that IAR Ab are likely to be better predictors of IDDM than IA-2 Ab in first degree relatives of IDDM sufferers, who are a high risk group for contracting IDDM. Therefore, the IAR of the invention is likely to be effective in predicting the onset of IDDM in high-risk groups such as first-degree relatives. Further, the inventors have found that the cytoplasmic (intracellular) domain of IAR is itself a useful predictor of IDDM. It is likely to be as useful as, or more useful than, complete IAR.

A brain Expressed Sequence Tag (EST) is available which has a sequence corresponding to part of the IAR sequence and highly homologous to the human IA-2 sequence. This is EST03250 (Adams et al; Nature Genetics Vol. 4, July 1993) . The length of the EST is 313 nucleotides. It was identified during the partial sequencing of over 3400 ESTs from human brain cDNA, in an attempt to demonstrate the diversity of RNA transcripts present in brain cells. IAR of course differs in terms of sequence from this EST because it has a different, longer, sequence. Also, it is unknown whether or not the EST encodes a functional autoantigen.

IAR sequences of the invention can be used in the diagnosis of IDDM, for example to screen for or predict the onset, presence or development of IDDM. For example, individuals who have, or are predisposed to develop, IDDM can be expected to have autoantibodies to the autoantigenic IAR in their blood. Thus, blood, serum or other samples can be taken and contacted with the IAR of the invention. The anti-IAR autoantibodies will recognise the exogenous IAR and this recognition can be detected, for example by detecting the IAR/anti-IAR complex thus formed.

Accordingly, the present invention provides:

A nucleic acid encoding a polypeptide that has the properties of an Insulin-dependent diabetes mellitus (IDDM) -associated autoantigen, which nucleic acid comprises:

(a) the coding sequence of SEQ. ID. No. 1, 10 or 12 and/or the sequence complementary thereto;

(b) a sequence which hybridises to a sequence defined in (a) ; (c) a sequence that is degenerate as a result of the genetic code to a nucleic acid sequence defined in (a) or (b) ; (d) a sequence having at least 80% homology to a sequence defined in (a) , (b) or (c) ; a polypeptide encoded by such a nucleic acid; a nucleic acid vector including such a nucleic acid; ,- a cell harbouring such a vector; a method of diagnosing IDDM which comprises:

(a) contacting a sample from an individual with a polypeptide as defined above under conditions that permit the recognition of the polypeptide by autoantibodies that recognise IAR; and

(b) determining whether or not said recognition occurs. a method of producing a polypeptide as defined above which comprises culturing a cell as defined above under conditions which permit the expression of the polypeptide; and recovering the polypeptide; use of a polypeptide as defined above in diagnosing IDDM; a diagnostic test kit for diagnosing IDDM which comprises a polypeptide as defined above, and means for determining whether or not the polypeptide is recognised by autoantibodies that recognise IAR.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. Northern analysis of r75 tissue expression.

Human multiple tissue Northern blots are from Clontech and the source of the mRNA in each lane is indicated at the bottom of the figure. The r75 probe was labelled with ³² P- dCTP (Amersham) using the High Prime DNA Labeling Kit (Boehringer Mannheim) . Hybridizations were carried out at

65°C overnight in 0.5M sodium phosphate (pH 6.8) , 7% SDS, 15% forma ide, ImM EDTA. After washing 3 times with 50mM sodium phosphate (pH 6.8) , 1% SDS for 40 min at 65°C, the blots were exposed to film at -70°C. Blots were stripped of probe by incubation in freshly boiled 0.1% SDS (3 x 2 min) before reprobing. Top panels, r75; bottom panels, β - actin. The positions of molecular size markers (in kb) are shown to the left of each blot.

Figure 2. Schematic depiction of IAR clones and predicted protein product.

A. Overlapping cDNA clones C3 and 3-7. Nucleotide numbering is as in SEQ. ID. No.l.

B. Predicted IAR protein product aligned with the IA-2 protein. The predicted partial IAR polypeptide (top) translated from and aligned with the clones shown in A is depicted in alignment with the IA-2 protein (Lan et al . , 1994) . The % amino acid similarity between the intracellular regions of IAR and IA-2 is shown, as is that: between the partial extracellular region of IAR and the corresponding portion of IA-2. TM, transmembrane region.

C. Relationships of cDNA clones. The relative positions of clones Bll, C3 , 3-7 and 3-10 are shown. The coding portion of these clones is represented by the thick line on the top scale. Both strands of clones Bll, C3, and 3-7 were sequenced in their entirety. The angled 5' piece of C3 is not identical to the corresponding region of Bll and likely represents a cDNA library artifact, as does the boxed region (54 bp) of C3 which is a sequence found in the opposite orientation in the corresponding region of Bll.

Figure 3. Northern analysis of IA-2 tissue expression.

Human multiple tissue Northern blots (different membranes from those used for Fig.l) are from Clontech and the source of the mRNA in each lane is indicated at the bottom of the figure. The IA-2 probe (nucleotides 1590-1906, coding for a portion of the extracellular region, the transmembrane

region, and a small portion of the intracellular region; and non-homologous to IAR sequence) was labelled and hybridization carried out as described in the legend to Fig. 1. The position of the IA-2 transcript is indicated by the arrow at the right of the figure. The β-actin signal (position indicated by the parentheses to the right of each panel) results from a previous hybrization with a _/ 3-actin probe, and could not be completely stripped from the membranes before reprobing with IA-2. The positions of molecular size markers (in kb) are shown to the left of each blot.

Figure 4. Northern analysis of IAR tissue expression.

Human multiple tissue Northern blots are from Clontech and the source of the mRNA in each lane is indicated at the bottom of the figure. Membranes were stripped of IA-2 probe (Fig.2) by incubation in freshly boiled 0.1% SDS (3 x 2 min) . The IAR probe (nucleotides 790-1197 of SEQ. ID. No.l, coding for a portion of the predicted extracellular region and non-homologous to IA-2 sequence) was labelled and hybridization carried out as described in the legend to Fig. 1. The positions of IAR transcripts are indicated by the arrows at the right of the figure. The -actin signal (position indicated by the parentheses to the right of each panel) results from a previous hybrization with a -actin probe, and could not be completely stripped from the membranes before reprobing with IAR. The positions of molecular size markers (in kb) are shown to the left of each blot.

Figure 5. Comparison of IAR and IA-2 transcripts in human brain and pancreas. The indicated lanes from the Northern analyses of IA-2 (Fig. 3) and IAR expression (Fig.4) are shown together here for the purpose of direct comparison. The same membrane was hybridized with the IA-2 probe, then stripped and rehybridized with the IAR probe as described in the legends to Figs. 3 and 4.

Figure 6. Purification and phosphatase activity of the intracellular regions of IAR and IA-2.

A. Purified IA-2 (open arrow) and IAR polypeptides (closed arrows) obtained after expression, affinity purification and protease cleavage of GST-fusion proteins. Lane 1, IA-2 obtained from expression of pGEX-3C-IA-2 and cleavage with 3C protease; lane 2, IA-2 obtained from expression of pGEX- KG-IA-2 and cleavage with thrombin; lane 3, IAR obtained from expression of pGEX-3C-IAR and cleavage with 3C protease; lane 4, IAR obtained from expression of pGEX-KG- IAR and cleavage with thrombin. The positions of molecular size markers (kD) are shown to the right of the panel.

B. Phosphatase activity of IAR. The IAR was produced as n lane 3 of panel A. IAR was added to a reaction mixture containing 50mM sodium acetate (pH 4.5), 0.5 mg/ml BSA, 0.5mM dithiothreitol and 5mM para-nitrophenyl phosphate (PNPP) and incubated at 30°C. At the times indicated, aliquots of the reaction (each containing 16μg of purified IAR as measured by Bradford analysis) were removed and mixed with 13% KHP0 ₄ to terminate the reaction. The absorbance of the stopped reaction was monitored at 405 nm and is a measure of the amount of para-nitrophenol product formed by dephosphorylation of PNPP. All reactions were performed in duplicate and the absorbances averaged. Non- enzymatic hydrolysis of PNPP was accounted for by performing appropriate control reactions without added IAR, and this absorbance was subtracted from that attained in the presence of enzyme.

C. Phosphatase activity of IA-2. The IA-2 was produced as in lane 1 of panel A, and phosphatase assays carried out as described above for IAR except that reactions contained 2τ_M PNPP. Each reaction aliquot removed at the indicated times contained 20μg of purified IA-2 as measured by Bradford analysis.

Figure 7. Catalytic activity of IAR.

A. SDS-PAGE of IAR polypeptide (amino acids 646-1015)

obtained after expression, affinity purification, and proteolytic cleavage of GST-IAR fusion protein. Lane 1, molecular size markers (sizes in kDa shown at left) ; lane 2, IAR after cleavage with 3C protease. The two uppermost bands in lane 2 are bacterial proteins which are often present after purification of various GST-fusion proteins, including IAR(C945S) .

B. pH profile of IAR activity towards PNPP. All reactions were carried out for 30 min and contained 22.3μg/ml IAR and 2mM PNPP. The buffers used were 50mM sodium acetate (pH 4.0 - 5.0) , Mes (pH 5.5 - 6.5) or Tris (pH 7.0 - 8.5) .

C. Time-dependent dephosphorylation of PNPP by IAR. Reactions containing 44.4μg/ml IAR (■) or IAR(C945S) (o) and 5mM pNPP were carried out as described in Materials and Methods and stopped at the times indicated. IAR activity was also measured in the presence of ImM Na3V0 ₄ (Δ) . Control reactions lacking IAR were simultaneously performed and the resulting absorbance at 405nm subtracted from that obtained from reactions carried out in the presence of IAR. The data shown are the average of duplicate experiments.

Prior to use in the assays described above in B and C, the purified IAR was quantitated by densitometric scanning of Coomassie Blue-stained IAR protein band on SDS-PAGE (indicated by arrow in A) alongside known amounts of protein standards.

Figure 8. Reactivity of IAR with IDDM sera.

A. Immunoprecipitation of IAR with sera from control and recent-onset IDDM subjects. In vi tro translated ³⁵ S- methionine labelled IAR was incubated with sera (prebound to Protein A-Sepharose) from control (subjects 5 and 6 in Table 1) (lanes 1 and 2) and IDDM subjects (subjects 12, 3, and 18 in Table 1) (lanes 3-5) . Immunoprecipitates were analyzed by SDS-PAGE and autoradiography. The 41 and 32 kDa bands (arrowed) are IAR and an IAR proteolytic product, respectively.

B. Graphical depiction of control and IDDM subject sera

reactivity with IAR. The exact values of the data presented here and further subject information are given in Table 1. Experiments were carried out as in A and the immunoprecipitated IAR was resolved by SDS-PAGE and quantitated (in arbitrary density units) using a phosphorimager.

Figure 9. Relationship IAR Ab and IA-2 Ab in recent -onset IDDM patients Dotted lines represent thresholds of positivity. r=0.70,p<0.0001.

Figure 10. Effect of IAR Ab (A) and IA-2 Ab (B) on IDDM- free survival in first-degree relatives with ICA > 20JDFU or IAA > lOOnU/ml

Dotted line: antibody-negative subjects, solid line: antibody positive subjects. A:p<0.0001,B: Not Significant.

The invention provides nucleic acids encoding the IDDM-associated autoantigen known as IAR or IAR-PTP and other related polypeptides having the properties of IDDM- associated antigens. Nucleic acids of the invention include the partial IAR-PTP cDNA having the sequence shown in SEQ ID No. 1, the complete cDNA sequence (SEQ ID. No 10) and the cDNA sequence of the cytoplasmic region of the IAR

(SEQ ID. No. 12) .

The nucleic acids of SEQ ID Nos. 1 and 10 are preferred nucleic acids of the invention. In another preferred embodiment, the nucleic acid of the invention includes a sequence encoding the cytoplasmic domain of I ^, also known as the intracellular domain (SEQ ID No. 12) . The DNA sequence encoding the cytoplasmic domain is given in SEQ ID No. 12, along with the encoded ammo acid sequence of the cytoplasmic domain (SEQ ID NO. 13) . The nucleotide sequence of SEQ ID NO. 12 corresponds to the nucleotides which encode ammo acids 646 to 1015 of SEQ 3D NO.11. In this embodiment, the sequence of the invention

may consist of the sequence encoding the cytoplasmic domain or it may include other IAR sequences or non-IAR sequences at one or both ends.

However, it will be appreciated that other nucleic acids are included within the scope of the invention.

Nucleic acids of the invention may have substantially the sequence of SEQ ID No. 1, 10 or 12.

Nucleic acids of the invention may consist essentially of the sequence of SEQ ID No. 1, 10 or 12. The nucleic acid sequences of the present invention are preferably DNA, for example, cDNA or genomic DNA, though they may be RNA. It will be appreciated by those of skill in the art that, in RNA sequences according to the invention, the T residues shown in SEQ ID No. 1, 10 or 12 will be replaced by U. Further, the invention provides both single-stranded and double-stranded nucleic acids. The nucleic acids of the present invention are not limited to the nucleic acids of SEQ. ID. No. 1, 10 and 12. The nucleic acids of the invention include nucleic acids having sequences that are closely related to these sequences; these nucleic acids encode polypeptides of the invention.

Such nucleic acids of the invention may, for example, be prepared by altering that of SEQ ID No. 1, 10 or 12 by any conventional method, or isolated from any organism or made synthetically. Such alterations, isolations or syntheses may be performed by any suitable method, for example by the methods of Sambrook et al : (Molecular Cloning: A Laboratory Manual; 1989) . For example, nucleic acids of the invention may include substitutions, deletions, insertions, or extensions that distinguish them from the nucleic acid of SEQ. ID. No. 1, 10 or 12 as long as these do not destroy the ability of the nucleic acid to encode a polypeptide that functions as an IDDM-associated autoantigen.

A substitution, deletion or insertion may suitably involve one or more nucleotide positions. Thus, nucleic

acid sequences of the invention may differ from those of SEQ ID No. 1, 10 or 12 at any number of nucleotide positions, for example as a result of deletion, substitution, insertion or extension, as long as they encode a polypeptide of the invention. Sequences of the invention may, for example, differ from that of SEQ ID No. 1, 10 or 12 at 1, 2, 3, 4, 5, 6 to 10, 10 to 20, 20 to 30, 30 to 40, 40 to 50 or 50 to 100 nucleotide positions.

For example, the sequences of the invention include sequences that are capable of selective hybridisation to either strand of the sequence of SEQ. ID. No. 1, 10 or 12 and that encode a polypeptide of the invention. Such sequences capable of selectively hybridizing to the DNA of SEQ. ID. No. 1, 10 or 12 will generally be up to 80%, preferably at least 80%, more preferably at least 90%, at least 95% or 99% homologous to the DNA of SEQ. ID. No. 1, 10 or 12. Such homology will preferably apply over a region of at least 20, preferably at least 50, for instance 100, 500 or 1000 or more contiguous nucleotides. Such hybridisation may be carried out under any suitable conditions known in the art (see Sambrook et al (1989) : Molecular Cloning: A Laboratory Manual) . For example, if high stringency is required, suitable conditions include 0.2 x SSC at 60°C. If lower stringency is required, suitable conditions include 2 x SSC at 60°C. Also included within the scope of the invention are sequences that differ from those defined above and because of the degeneracy of the genetic code, encode the same polypeptide of the invention. For example, the invention provides degenerate variants of the sequence of SEQ ID No

1, 10 or 12 that also encode the polypeptide of SEQ ID No.

2, 11 or 13.

The nucleic sequences of the invention may be of any length as long as they encode a polypeptide of the invention. For instance, a nucleic acid sequence according to the invention will typically comprise one or more sequences including one or more of the antigenic epitopes

of IAR-PTP. A nucleic acid sequence according to the invention may be a contiguous fragment of the sequence of SEQ. ID. No. 1, 10 or 12 or a sequence that is related to it in any of the ways described above. Alternatively, nucleic acids of the invention may comprise DNA sequences that are not contiguous in the sequence of SEQ. ID. No. 1, 10 or 12. These sequences may be fragments of the sequence of SEQ. ID. No. 1, 10 or 12 or nucleic acid sequences that are related to such fragments in any of the ways described above. Nucleic acid sequences of the invention will preferably comprise at least 50 bases or base pairs, for example 50 to 100, 100 to 500, 500 to 1000, or 1000 to 2000 bases or base pairs.

In particular, it will be appreciated that SEQ. ID. No. 1 is a partial cDNA sequence of IAR-PTP. The invention also includes sequences extended at either the 3' end or the 5' end, or both ends, by further nucleic acid sequence. 5' extensions are preferred. This sequence may be of any nature and of any suitable length. For example, an extension may comprise up to 10, up to 20, up to 50, up to 100, up to 200, up to 500, up to 1000, up to 1500, up to 2000, up to 2500 or up to 3000 or more bases or base pairs. For example, an extension may comprise the sequence that is contiguous with the sequence of SEQ. ID. No. 1 in the complete IAR-PTP cDNA (SEQ ID. No. 10) . Alternatively, it may be any other sequence. Thus, preferred nucleic acids of the invention include complete IAR-PTP cDNAs and fragments thereof which include SEQ. ID. No. 1, particularly human cDNAs and fragments. Also, the invention provides nucleic acids encoding fusion proteins comprising polypeptides of the invention.

A particularly preferred nucleic acid of the invention has the sequence of SEQ ID No. 10, the complete IAR cDNA sequence. Another particularly preferred nucleic acid of the invention has the sequence of the cytoplasmic domain of IAR, SEQ ID NO. 12.

Similarly, the extension or extensions could have any

other sequence. Also, the sequence of SEQ. ID. No. 1, 10 or 12 may be interrupted in one or more places by non- coding sequence. In particular, the non-coding sequence may be introns that occur in the IAR-PTP genomic DNA. Further preferred nucleic acids of the invention are thus IAR-PTP genomic DNAs, and fragments thereof, that include the sequence of SEQ ID No. 1, 10 or 12, particularly human genomic DNAs and fragments thereof.

Three preferred polypeptides of the inventicn have the sequences shown in SEQ. ID. No. 2, (partial amino acid sequence of IAR) , SEQ. ID. No. 11 (complete amino acid sequence of IAR) and 13 (amino acid sequence of cytoplasmic domain of IAR) .

SEQ ID No. 13 corresponds to amino acids 646 to 1015 of SEQ ID No. 11. However, the polypeptides of the invention are not limited to the polypeptides of SEQ. ID. No. 2, 11 and 13. Rather, the polypeptides of the invention also include polypeptides with sequences closely related to those of SEQ. ID. No. 2, 11 and 13 that have suitable properties. Thus, the polypeptides of the invention typically have substantially the sequence of SEQ ID No. 2, 11 or 13.

Polypeptides of the invention may have a sequence which consists essentially of the sequence of SEQ ID No. 2 , 11 or 13.

Polypeptides of the invention may have a sequence which is substantially the sequence of SEQ. ID. No. 2, 11 or 13.

The polypeptides of the invention are encoded by the nucleic acids of the invention, as described herein. The polypeptides of the invention, IAR-PTP and closely related polypeptides, are IDDM-associated autoantigens, or have the properties of IDDM-related autoantigens.

Thus, the polypeptides of the invention are typically capable of provoking an autoimmune response in individuals who have, or are predisposed to have, or are likely to develop IDDM. Thus, the polypeptides of the invention may

provoke the production of IDDM-associated autoantibodies in such individuals. Also, or alternatively, they may provoke other elements of the autoimmune response.

Further, the polypeptides of the invention are typically recognised by autoantibodies that already present in the sera of such individuals, and are capable of recognising the natural IAR-PTP which contributes to the autoimmune response in these individuals.

Optionally, the polypeptides of the invention may have PTP activity.

Polypeptides of the invention having sequences related to the sequence may be prepared by altering the polypeptide of SEQ. ID. No. 2, 11 or 13 by any conventional method, or isolated from any organism or made synthetically. Such alterations, isolations or syntheses may be performed by any conventional method, for example by the methods of Sambrook et al (Molecular cloning: A Laboratory Manual; 1989) . In particular, polypeptides related to those of SEQ. ID. No. 2, 11 and 13 may be prepared by modifying DNA sequences as shown in SEQ. ID. No. 1, 10 or 12 and expressing them recombinantly.

Polypeptides of the invention may include substitutions, deletions, insertions, or extensions that distinguish them from the polypeptide of SEQ. ID. No. 2, 11 or 13 as long as these do not destroy the ability of the polypeptide to function as an IDDM-associated autoantigen.

A substitution, deletion or insertion may suitably involve one or more amino acids, typically from one to five, one to ten or one to twenty amino acids. For example, a substitution, deletion or insertion may involve one, two, three, four, five, eight, ten, fifteen, or twenty amino acids. Polypeptides of the invention are substantially homologous to that of SEQ. ID. No. 2, 11 or 13. Typically, a polypeptide of the invention has up to 80%, preferably at least 80%, at least 90%, at least 95%, at least 98% or at least 99% sequence identity to the polypeptide of SEQ. ID. No. 2, 11 or 13.

In general, the physicochemical nature of the polypeptide of SEQ. ID. No. 2, 11 or 13 should be preserved in a sequence of the invention. Such sequences will generally be similar in charge, hydrophobicity and size to that of SEQ. ID. No. 2, 11 or 13. Thus, substitution will typically be conservative. Examples of conservative substitutions that do not greatly affect the physicochemical nature of amino acid sequences are those in which an amino acid from one of the following groups is substituted by a different amino acid from the same group: H, R and K I, L, V and M A, G, S and T D, E, P and N. As far as extensions are concerned, a sequence of one or more amino acids may be provided at either or both of the C- and N- termini of the sequence of SEQ. ID. No. 2, 11 or 13 or a sequence related to it in any of the ways defined herein. An extension may comprise up to 5, up to 10, up to 20, up to 50, up to 100, up to 200, up to 300, up to 400, up to 500 or more amino acids. For example, an extension may comprise one, two, three, four, five, up to ten, up to 20, up to 30, up to 50, up to 100, up to 200 or more amino acids. In particular, extensions may include amino acid sequences which are found in the native complete IAR-PTP protein. Thus, preferred polypeptides of the invention include complete IAR-PTP proteins and polypeptide fragments thereof that include the amino acid sequence of SEQ. ID. No. 2, particularly human proteins and fragments thereof. Also, the invention provides fusion proteins including polypeptides of the invention.

A particularly preferred polypeptide of the invention has the sequence of SEQ. ID. No. 11, the sequence encoded by the complete IAR cDNA (SEQ. ID. No. 10) . A further particularly preferred polypeptide has the sequence of SEQ ID NO. 13, which is the sequence of the cytoplasmic domain

of IAR .

A polypeptide of the invention may be subjected to one or more chemical modifications, such as glycosylation, sulphation, COOH-amidation or acylation. A polypeptide of the invention may comprise multiple copies of the sequence of SEQ. ID. No. 2, 11 or 13 or a sequence related to it in any of the ways defined herein.

A polypeptide of the invention may be of any length as long as it has the ability to function as an IDDM- associated autoantigen. For instance, polypeptides of the invention preferably comprise one or more epitopes of IAR- PTP and therefore retain the ability to provoke an autoimmune response and/or be recognised by preexisting IDDM autoantibodies. Such polypeptides of the invention could be considerably smaller than the polypeptide of SEQ. ID. No. 2.

Polypeptides according to the invention may include one or more fragments of the sequence of SEQ. ID. No. 2, 11 or 13 or fragments that are related to SEQ. ID. No. 2 in any of the ways described above. Such fragments typically comprise at least 10, for example 10 to 20 or 20 to 50, amino acids.

Alternatively, polypeptides of the invention may comprise amino acid sequences that are not contiguous in SEQ. ID. No. 2, 11 or 13. These amino acid sequences may be identical to parts of the native amino acid sequence or related to such parts in any of the ways described herein. For example, polypeptides of the invention may comprise two or more epitopes of IAR-PTP, optionally separated by IAR sequence or non-IAR sequence.

Polypeptides according to the invention preferably comprise at least 10 amino acids, for example 10 to 20, 20 to 50, 50 to 100, 100 to 200, or 200 to 500 or 500 to 1000 amino acids. Polypeptides and nucleic acids according to the invention may be substantially isolated, isolated, purified or substantially purified. For example, they may be

present in preparations, for example solutions, vvhich have undergone one or more purification steps. Optionally, they may be completely pure. They may be present in preparations, for example solutions, which, apart from the nucleic acid and/or polypeptide of the invention consist essentially of solvents and/or biologically inert carriers. Polypeptides of the invention in substantially purified form will generally comprise the polypeptide in a preparation in which more than 90%, eg. up to 95£, up to 98% or up to 99% of the peptide material in the preparation is that of a polypeptide or polypeptides according to the invention.

The nucleic acids and polypeptides of the invention were originally derived from the human genome. However, nucleic acid sequences and/or polypeptides of the invention may also be obtained from other eukaryotic genomes, especially other mammalian genomes. They may be obtained either by conventional cloning techniques or by probing genomic or cDNA libraries with nucleic acid sequences according to the invention. This can be done by any conventional method, such as the methods of Sambrook et al (Molecular Cloning: A Laboratory Manual; 1989) . In particular, the complete cDNA and/or genomic DNA sequences of human IAR-PTP can be determined in this way, and the complete amino acid sequence of IAR-PTP can be determined. A nucleic acid sequence according to the invention may be included within a vector, suitably a replicable vector or an expression vector, for instance a replicable expression vector. Such an expression vector typically comprises an origin of replication so that the vector can be replicated in a host cell such as a bacterial host cell or a yeast host cell. A suitable vector will also typically comprise the following elements, usually in a 5' to 3 ' arrangement: a promoter for the directing expression of the nucleic acid sequence of the invention and optionally a regulator of the promoter, a translational start codon, a nucleic acid

sequence according to the invention.

The vector may also contain one or more selectable marker genes, for example an ampicillin resistance gene for the identification of bacterial transformants or a marker gene that allows selection of yeast or other eukaryotic transformants. Optionally, the vector may also comprise an enhancer for the promoter. The vector may also comprise a polyadenylation signal operably linked 3' to the nucleic acid of the invention. The vector may also comprise a transcriptional terminator 3' to the sequence encoding the polypeptide of the invention.

The vector may also comprise one or more introns or other coding sequences 3' to the nucleic acid sequence of the invention. The intron or introns may be from the human genome (the organism from which the sequences of the invention were originally derived) or the host organism which is to be transformed with the vector or from another eukaryotic organism.

In an expression vector, the nucleic acid sequence of the invention is operably linked to a promoter capable of expressing the sequence. "Operably linked" refers to a juxtaposition wherein the promoter and the nucleic acid sequence encoding the polypeptide of the invention are in a positional relationship permitting the coding sequence to be expressed under the control of the promoter. Thus, there may be elements such as non-coding sequence 3' or 5' to the coding sequence. These elements may be native either to the human genome or, for example, to the organism from which the promoter sequence is derived. Alternatively, the said element or elements may be native to neither the organism from which the promoter sequence is derived nor to the human genome. Such sequences can be included in the vector if they enhance or do not impair the correct control of the coding sequence by the promoter.

An expression vector may be of any type. The vector may be in linear or circular form. For example, it may be

a plasmid or viral vector. Those of skill in the art will be able to prepare suitable vectors comprising nucleic acid sequences encoding polypeptides of the invention starting with widely available vectors which will be modified by genetic engineering techniques such as those described by Sambrook et al (Molecular Cloning: A Laboratory Manual; 1989) .

In an expression vector, any promoter capable of directing expression of a sequence of the invention may be operably linked to the nucleic acid sequence of the invention. For example, suitable promoters include yeast: promoters, mammalian promoters, viral promoters and bacterial promoters. Such promoters may be constitutive cr regulable. Some suitable promoters include the cytomegalovirus (CMV) early promoter, the SV40 promoter, the mouse mammary tumour virus promoter, the human elongation factor 1 α-P promoter (EF-lα-P) , the SRα promoter and the mouse metallothienein gene I (mMTl) promoter. Typically, nucleic acid sequences according to the invention will be inserted into such vectors in a sense orientation. However, nucleic acid sequences according to the invention may also be inserted into the vectors described above in an antisense orientation in order to provide for the production antisense RNA. Antisense RNA may also be produced by synthetic means. Such antisense RNA may be used in a method of controlling the levels of the polypeptide of SEQ. ID. No. 2 or 11 or 13 or a protein encoded by a related nucleic acid sequence in a cell. Such a protein is, for example, a complete IAR-PTP protein. Thus, antisense sequences of the invention can form the basis of methods of gene therapy aimed at suppressing the autoantigenicity of IAR-PTP.

Vectors according to the invention may be used in vitro, for example for the production of RNA hybridisabl€: to cDNA, of the invention. Such vectors may be used to transfect or transform a host cell. Depending on the type

of vector, they may be used as cloning vectors to amplify DNA sequences according to the invention or to express this DNA in a host cell.

A further embodiment of the invention provides host cells harbouring one or more vectors and/or nucleic acids of the invention. Typically, such cells are transformed or transfected with the vectors for the replication and/or expression of nucleic acid sequences according to the invention, including the DNA of SEQ. ID. No. 1, 10 and 12. The cells will be chosen to be compatible with the vector and may for example be bacterial cells, mammalian, insect or yeast cells.

The cells may be transformed or transfected by any suitable method, such as the methods disclosed by Sambrook et al (Molecular Cloning: A Laboratory Manual; 1989) . For example, vectors comprising nucleic acid sequences according to the invention may be packaged into infectious viral particles, such as retroviral particles. The constructs may also be introduced by electroporation, calcium phosphate precipitation, biolistic methods or by contacting naked nucleic acid vectors with the cells in solution.

In the said nucleic acid vectors with which the host cells are transformed or transfected, the nucleic acid may be DNA or RNA, preferably DNA, it may be single-stranded or double-stranded.

The vectors with which the host cells are transformed or transfected may be of any suitable type. For example, the vectors may be able to effect integration of nucleic acid sequences of the invention into the host cell genome or they may remain free in the host cell. Typically, the vectors will be expression vectors, such as a retroviral vector or a DNA expression vector as defined herein. For example, the vector used for transformation construct may be a plasmid vector.

The transformed or transfected cells of the invention can be used in a process of production of polypeptides of

the invention. Such processes will typically comprise transforming or transfecting host cells with vectors comprising nucleic acid sequences according to the invention and expressing the nucleic acid sequence in these cells.

In this case, the nucleic acid sequence will be operably linked to a promoter capable of directing its expression in the host cell. Desirably, such a promoter will be a "strong" promoter capable of achieving high levels of expression in the host cell. It may be desirable to overexpress the polypeptide according to the invention in the host cell.

Suitable host cells for this purpose include bacterial cells, for example E. coli cells; yeast cells; mammalian cells, for example CHO cells; and insect cells, preferably in combination with a baculovirus expression system. Eukaryotic cells, such as mammalian, insect and yeast cells are preferred in situations where glycosylated polypeptide products are desired. The thus produced polypeptide of the invention may be recovered by any suitable method known in the art.. Optionally, the thus recovered polypeptide may be purified or partly purified by any suitable method, for example a method according to Sambrook et al (Molecular Cloning: A Laboratory Manual) .

The nucleic acid sequences of the invention may be used to prepare probes and primers. These will be useful in the isolation of nucleic acid sequences having sequences similar to that of SEQ. ID. No. 1, 10 or 12, which may encode polypeptides of the invention. Such probes and primers may be of any suitable length, desirably from 10 to 100, for example from 10 to 20, 20 to 50, 50 to 100 bases in length. Typically, the probes and primers will correspond exactly to part of the sequence of SEQ. ID. No. 1, 10 or 12 over their entire length, although they may vary from this sequence in any of the ways described above. Such probes and/or primers may be used, in particular to

isolate complete IAR-PTP cDNA and/or genomic DNA sequences from the human genome; or from the genomes of other organisms .

The present invention also provides antibodies, preferably monoclonal antibodies, to the polypeptides of the invention. Such antibodies may be useful in methods of diagnosing IDDM, as described herein.

Antibodies of the invention include fragments of whole antibodies which retain their binding activity for a target antigen. Such fragments include Fv, Fab' and Fab' ₂ fragments, as well as single chain antibodies.

The antibodies may be produced by any suitable method known in the art, such as the methods of Sambrook et al (Molecular Cloning: A Laboratory Manual; 1989) . For example, they may be prepared by conventional hybridoma techniques or, in the case of modified antibodies or fragments, by recombinant DNA technology, for example by the expression in a suitable host vector of a DNA construct encoding the modified antibody or fragment operably linked to a promoter, as described above. Suitable host cells include bacterial (for example E. coli) , yeast, insect and mammalian cells.

Therefore, the invention also provides nucleic acid sequences encoding antibodies of the invention, vectors comprising such sequences, cells harbouring such vectors or sequences, and methods for expressing antibodies of the invention by recombinant techniques.

Polyclonal antibodies may also be raised, using conventional techniques in the art, using the polypeptides of the invention. For example, polyclonal antibodies may be raised by inoculating a host animal, such as a rat, mouse or rabbit, with a polypeptide of the invention, and recovering the immune serum.

Antibodies according to the invention may be useful in methods of the invention, as described herein, for the diagnosis of IDDM.

The invention further provides methods of diagnosing

IDDM. Thus, for example, polypeptides of the invention can be used in methods of diagnosing any aspect of IDDM. For example, they may be used in diagnosing the onset of IDDM, the presence of IDDM, or predisposition to IDDM, or in monitoring the development or progress of IDDM, εtnd in methods of predicting the possible future occurrence of IDDM. For example, individuals who have, or are predisposed to develop, IDDM can be expected to have autoantibodies to the autoantigenic IAR-PTP. In particular, The IAR polypeptides of the invention are useful in methods carried out on individuals having a high likelihood of developing IDDM, for example, relations, especially close relations, more especially first: degree relations (parents and siblings), of known IDDM sufferers. For example, the IAR polypeptides of the invention are particularly useful in determining the likelihood that a close relation of an IDDM sufferer will contract IDDM. Close relations include, for example, parents, grandparents, siblings, children, grandchildren, cousins, uncles, aunts, nieces, grand aunts and uncles and grand nieces and nephews.

Thus, samples of biological fluid suspected of containing such autoantibodies can be taken from individuals and contacted with polypeptides of the invention. The anti-IAR autoantibodies, if present, will recognise the polypeptides of the invention and this recognition can be detected, for example by detecting the IAR/anti-IAR complex thus formed.

Thus, the methods of the invention typically comprise a step of determining whether or not recognition of the polypeptide of the invention by autoantibodies occurs in the future. Similarly, if no such recognition occurs, the test suggests that the individual does not have IDDM or is unlikely to develop it in the future. Recognition of polypeptides of the invention by antibodies present in such a sample correlates with IDDM or a potential for developing IDDM.

The methods of the invention are thus typically methods of assaying for the presence or absence of autoantibodies that recognise IAR. They may also quantitate the amount or concentration of such autoantibodies in a sample.

Thus, diagnostic methods of the invention typically comprise:

(a) contacting a sample from an individual with a polypeptide according to the invention under conditions that permit the recognition of the polypeptide by autoantibodies that recognise IAR; and

(b) determining whether or not said recognition occurs. Any suitable sample from the individual may be used.

Typically the sample will comprise a biological fluid, preferably blood, plasma or serum. The sample may be of any size.

The sample is contacted with a polypeptide of the invention and the presence or absence of immunoreactivity between them is determined. In some embodiments of the invention, a second IDDM- related autoantigen may be used. Thus, the invention provides methods, as described above, which further comprise, in addition to steps (a) and (b) , contacting the sample with at least one further IDDM-related autoantigen, preferably an islet cell antigen (ICA) . This may be done at the same time, or at a different time to the contacting with the polypeptide of the invention.

In such methods, it is determined whether or not autoantigens recognise the further autoantigen or autoantigens. The same sample or a different sample from the same individual may be contacted with the polypeptide of the invention and the further IDDM-associated autoantigen. Such a further autoantigen may be, for example, hIA-2, GAD, ICA-512, ICA-12, ICA-13, ICA-208, ICA- 302, ICA-313, ICA-525 or ICA-505; or a polypeptide that functions as one of these further autoantigens.

Preferred further autoantigens include: IA-2,

preferably hIA-2; ICA-512; and GAD; and antigenic fragments thereof. Preferred combinations of autoantigens therefore! include: a polypeptide of the invention and one, two or all three of IA-2, GAD and ICA-512. Further IDDM-associated autoantigens may be used, as appropriate, in any of the techniques described below.

Any conventional technique may be used to determine whether or not polypeptides of the invention are recognised by autoantibodies. This also applies to determining whether or not any further autoantigens which may be used are recognised by autoantibodies.

For example, suitable immunoassay formats employ a combination of solid phase or immobilized reagents and labelled reagents whereby the association of the label with the solid phase is a function of the presence or absence of recognition of the polypeptide of the invention cr a further autoantigen. In general, such a solid phase reagent comprises a binding substance such as a polypeptide of the invention, an anti-antibody (e.g., anti-IgG) , or other immunobinder or other binding agent according to the' assay protocol involved, bound or attached, covalently or non-covalently, to a solid phase matrix or in an otherwise' immobilized form. Suitable solid phase matrices are conventional in the art and include such matrices as microtiter plate wells, test tubes and other test containers or vessels, test strips, beads, and particles such as plastic microparticles and latexes. Where a solid phase reagent comprises more than one type of polypeptide of the present invention and/or a polypeptide of the invention and one or more further autoantigens, it will be recognised that each reagent can be physically separated or isolable from the others or two or more can be mixed or otherwise associated in an undifferentiated manner with the solid phase. For example, reagents can be immobilised in separate wells of a microtiter plate or can occupy the sane wells; or, where the solid phase is in the form of particles, each individual particle can have attached only

a single reagent or can have attached two or more reagents. Similarly, useful labelled reagents comprise a binding substance such as a polypeptide of the invention, a further autoantigen, an anti-antibody (e.g., anti-IgG), or other immunobinder or other binding agent according to the assay protocol involved, which is chemically coupled with a detectable chemical moiety. Useful labels are conventional in the art and include fluorescers, chemiluminescers, radioisotopes, and enzymes. Enzyme labels are particularly useful and are generally selected from alkaline phosphatase, peroxidase, and -galactosidase. Enzyme labels are readily detectable by addition of a corresponding chromogenic substrate and detecting the resulting colour or fluorescent response. One particular immunoassay format that can be applied to the present method employs an immobilized form of the polypeptides of the invention, and optionally an immobilized form of a further autoantigen, as an immunoreagent. A test sample is incubated with the solid phase reagent and preferentially washed to remove unbound material. A labelled antibody reagent is then added. Such antibody reagents can be specific for a particular class of immunoglobulin, e.g., IgG, IgM, IgA, etc., or can be a mixture of conjugates so that all immunoglobulin types are detectable. Islet cell antibodies of interest are generally of the IgG isotype and thus anti-IgG would normally be employed. The solid phase is washed to remove unbound labelled antibody reagent and the label activity remaining on the solid phase is measured qualitatively or quantitatively.

A variation of this protocol uses a ligand-modified form of the polypeptide of the invention, and optionally a ligand-modified form of a further autoantigen with immobilization to the solid phase being accomplished by using a solid phase bearing a binding partner to the ligand. For example, biotin or a hapten (e.g., fluroscein) can be used as the ligand and can be rendered immobilized

by contact with a solid phase form of avidin or anti-hapten antibody, respectively. The addition of the solid phase binding partner can occur at any convenient time in the assay, such as prior to contact of sample with the ligand- antigen(s) or thereafter.

Another immunoassay format that can be applied to the present method employs an immobilized form of an antibody reagent. Antibody specific for the desired ICA immunoglobulin type to be detected (e.g., IgG) , cr a mixture of antibodies against different IgG isotypes, is immobilized on a solid phase and incubated with the test sample. Resulting islet cell antibody that has become bound to the solid phase antibody reagent can then be detected in any suitable manner. For instance, one can add labelled forms of one or more types of polypeptide of the invention and/or of other IDDM autoantigens either as individually labelled reagents or as a combined labelled reagent. Alternatively, one can add a soluble form of the polypeptide of the invention (and optionally a further autoantigen) and then, together in the form of a complex, or later as a separate reagent addition, labellecL antibody that can bind with it. Further, one can add a soluble form of the polypeptide of the invention (and optionally a further autoantigen) which has been modified with a ligand and then adding a labelled form of a binding partner to the ligand. The previously described variation of using a ligand-modified form of the solid phase reagent, in this case, antibody, with immobilization to the solid phase being accomplished by using a solid phase bearing a binding partner to the ligand, can also be used.

A competitive immunoassay format is also useful. Immobilized polypeptide of the invention is employed along with a labelled form of islet cell antibodies, for example antibodies of the invention as described herein. Labelled antibody and ICA from the test sample are allowed to compete for binding to the polypeptide of the invention either simultaneously or sequentially, e.g, by exposing the

solid phase polypeptide of the invention first to the sample and thereafter to the labelled antibody reagent. The polypeptide of the invention can be immobilized directly to the solid phase or through a ligand-binding partner bridge as described above, with the immobilization step being performed at any convenient time in the assay, including as the last step.

Latex or particle agglutination methods are also suitable. Particles are coated or covalently coupled with the polypeptides of the invention, and optionally with one or more further IDDM-associated autoantigens. The particles are then incubated with the test sample and resulting agglutination of the particles due to the formation of ICA-antibody linkages between particles is detected. Detection can be accomplished, for example, by visual observation (a slide agglutination format) or quantitated by measuring turbidity change with a spectrophotometer or nephelometer. A variation is based on inhibition of particle agglutination. Each particle reagent comprises one or more monoclonal antibodies corresponding specifically with one or more particular polypeptide of the invention and optionally with one or more further IDDM-associated autoantigens, respectively. In addition, an agglutinator reagent is prepared comprising multiple antigens, e.g., a water soluble polymer backbone to which are attached multiples of one or more types of polypeptides of the invention, and optionally one or more further autoantigens. When mixed with sample, the absence of autoantibodies in the sample results in agglutination of the particles by formation of bridges formed by the agglutinator reagent. When autoantibodies are present, they bind to the agglutinator and thereby prevents or inhibits agglutination.

In methods according to the invention, any suitable quantity of polypeptides of the invention and/or of the further autoantigens may be used. For example, for each test according to the invention, a quantity of from

1 x 10 ^"12 g to 1 x IO ^"6 g of a polypeptide of the invention and optionally of one or more further autoantigens as defined herein may be used. Preferably, quantities of from 1 x 10 ^"10 g to 1 x 10 ^"7 g may be used, with quantities of from 10 ^"9 g to 10 ^"7 g being more preferred. For example, quantities of from 10 to 50 ng, for example from 20 ng to 30ng may be used (1 x 10 ^~8 g to 5 x 10 ^~8 g) . Such quantities of polypeptides of the invention and optionally of further autoantigens as defined herein, may, for example, be applied to a given microtiter well in a microtiter plate or to any other solid phase matrix.

Diagnostic methods of the invention are typically performed in vi tro or ex vivo although, where appropriate, in vivo methods are contemplated. The methods of the invention have a number cf clinical applications related to IDDM. For example, the methods of the invention are useful in screening of patients from the general population, particularly juveniles, wherein a positive test result (i.e., a finding that a patient's blood contains antibodies that bind to a polypeptide of the invention) indicates that such a patient is at risk of developing IDDM, that is, that such a patient may have a pre-IDDM condition. Furthermore, the present method is applicable to the screening of first degree relatives (i.e., siblings and children) of individuals alre:ady diagnosed as having IDDM. Patients with positive test results from the methods of the invention could be examined in more detail, for example by means of metabolic and genetic testing, to determine whether glucose intolerance has already occurred, or whether additional risk factors are present.

Other applications of the present method include the testing of a patient already diagnosed with diabetes for the purpose of determining if IDDM is involved, for instance in the case of a Type II diabetic who may also experience the onset of IDDM.

Furthermore, the methods of the invention are useful

in monitoring of diabetic therapy. For example, using the methods of the invention, it is possible to monitor the effectiveness of any type of treatment of IDDM, for example gene therapy, immunosuppressant therapy or immune therapy. it is also possible to monitor the effects of drug therapy, the impact of diet on IDDM, and the effect of other factors on IDDM.

The invention further provides medical uses of the polypeptides of the invention. Accordingly, the invention provides polypeptides of the invention for use in methods of treatment of the human or animal body, especially the treatment of IDDM; and for use in methods of diagnosis of IDDM. In methods of treatment of IDDM, polypeptides of the invention may, for instance, be used to saturate an individual's immune response, blocking the individual's autoimmune response to natural IAR-PTP. The invention also provides polypeptides of the invention for use in methods of treatment of the human or animal body, particularly for IDDM, in combination with one or more further autoantigens as described herein.

Further, the invention provides the use of polypeptides of the invention in the manufacture of medicaments for the treatment of the human or animal body, especially the treatment of IDDM; and for use in methods of diagnosis of IDDM, as described herein. Such medicaments may comprise a further IDDM-associated autoantigens as described herein, or may be administered in combination with such a further autoantigen.

Similarly, the invention provides pharmaceutical compositions comprising a polypeptide of the invention, and optionally a further autoantigen as defined herein; and a pharmaceutically acceptable carrier.

The invention further provides diagnostic test kits comprising polypeptides and/or antibodies of the invention, optionally together with one or more further IDDM- associated autoantigens, as described herein.

Any suitable quantity of these components may be

present in the test kits. Other components may also be included in the kits, for example buffers and/or other solutions; labels; tubes, plates, membranes, membrane sticks or other items of equipment, for example for immobilising reagents; and/or anti-human antibodies for monitoring immunoreactivity.

Test kits of the invention typically further comprise means for determining whether or not the polypeptides of the invention are recognised by autoantibodies. Test kits comprising one or more further autoantigens typically include means for determining whether or not the further autoantigen or autoantigens are recognised by autoantibodies.

Test kits according to the invention may, for example, enable the use of methods according to the invention. Thus, they may comprise, as means for determining whether or no: the polypeptides of the invention, and optionally one or more further IDDM-associated autoantigens as described herein, are recognised by autoantibodies, any reaigents suitable for putting into practice the techniques described above.

In test kits of the invention, the components may be in any suitable form; for example they may be packaged in any suitable container. The components may be intended or separate, for example sequential, use; or for concurrent use.

In test kits of the invention, the components may be in any suitable form, for example they may be in sterile solutions, buffered solutions or sterile and buffered solutions. They may be labelled or unlabelled. The components may be in substantially purified, purified, isolated or substantially isolated form.

The following Examples illustrate the invention.

EXAMPLES

1. Initial identification of IAR as a PTPase

Primers corresponding to conserved amino acid

sequences (DYINA (SEQ. ID. No. 14) and VHCSAGV (SEQ. ID No. 15) of the catalytic domain of PTPases were designed (with additional EcoRI sites added) and used to amplify a human colon carcinoma cell (SW480) cDNA library (Clontech, HL3014b) using PCR. The primers used had the following sequences: 5' -GGGAATTCNGAYTAYATHAAYGC-3 ' (SEQ ID No. 3) and 5' -GGGAATTCACNCCNGCRCTRCARTGNAC-3' (SEQ ID No. 4) . The PCR was carried out as follows: 94°C x 5 mm >(94°C x 40 sec >50°C x 30 sec >72°C x 90 sec) x 30 >72°C x 5 min- -- _> 4°C. The major amplified fragment of ~460bp was subcloned into the pGEM-T vector (Promega) and 94 clones were sequenced.

One clone contained an amplified fragment which did not have high homology to any known PTPase. This fragment was selected for further study and was termed r75. r75 was identical over a 312bp portion of its sequence to a human brain EST sequence (# 03250) [Adams et al . , 1993, Nature Genet. 4, 256-267] which was noted in this paper as having 80.6% similarity to ICA512) . To examine the size and tissue expression of RNA transcripts from which r75 originated, human multiple Northern blots (Clontech) were probed with ³² P-labelled r75 (Fig. 1) . Two major transcripts of about 3.7kb and 5.5kb were observed. The highest levels of expression were detected in brain and pancreas, followed by prostate and testis, with lower expression detectable in a few other tissues, including colon.

2. Cloning of IAR Based on the expression pattern of transcripts recognized by r75, a random primed human pancreas cDNA library in λgtll (Clontech, HL1163b) was screened using r75 as probe. The r75 was labelled with ³² P-dCTP (Amersham) using the High Prime Labeling Kit (Boehringer Mannheim) and denatured by heating before hybridization. Hybridization was carried out overnight at 65°C in 0.5M sodium phosphate (pH 6.8) , 7% SDS, 15% formamide, ImM EDTA. The membrane

was washed 3 times at 65°C with 50mM sodium phosphate (pH 6.8) , 0.1% SDS, and exposed to film at -70°C. Positive clones were reselected through another 3 rounds of subscreening with r75. Continuously positive clones were subcloned into the EcoRI site of Bluescript SK(+)

(Stratagene) and partially sequenced for comparison with r75. Some clones were found to be fragments of IA-2 cDNA (Lan et al . , 1994) . One of these clones, D2, contained nucleotides 1589-3166 of IA-2. Several other clones were distinct from IA-2. Of these, the largest clone (C3) containing an open reading frame of about 2.5kb was selected for sequencing. The C3 clone was digested by Pstl or Apal into shorter fragments. Each fragment was subcloned into Bluescript SK(+) and sequenced across both strands using either primers to the vector or synthesized primers hybridizing to the insert in conjunction with the Sequenase Version 2.0 Sequencing Kit (United States Biochemical Corporation) .

The C3 clone contained a coding sequence missing initiation and termination signals, and was significantly smaller than than the 3.7 and 5.5kb RNA transcrip>ts detected by Northern blotting with r75 (Fig.l) , suggesting that it was incomplete at both the 5' and 3' ends. The 3' end of the clone was obtained by 3' RACE using the Marathon cDNA Amplification Kit (Clontech) . The first-strand cDNA was reverse transcribed from human brain polyA ^* mRNA (Clontech) using the cDNA synthesis primer provided with the kit. The gene-specific primer used for 3 '-RACE was 5'- CCTGCCTCCTCAGGCGGAGCAAGA-3' (SEQ ID No. 9) , corresponding to nucleotides 1389-1412 of SEQ ID No.l. PCR was carried out at 94°C x 3 min >(94°C x 40 sec >68°C x 4 min) x 30-

-- _> 4°C using the Expand Long PCR Kit (Boehringer Mannheim) . The amplified fragments were cloned into the pGEM-T vector (Promega) and partially sequenced for comparison with C3. One clone, 3-7, with a region of identical overlap with C3, contained a stop codon and polyadenylation signal and was sequenced along both strands by Sanger' s dideoxy-mediated

chain-termination method. Another clone, 3-10, was about 2kb larger than 3-7. The 3-10 clone was identical to 3-7 from the 5' end up to the polyadenylation signal in 3-7, but was followed by approximately 2kb of additional 3' untranslated sequence.

The combined sequences of C3 and 3-7 comprise a partial cDNA encoding a novel putative receptor-like PTPase which we term IAR (for islet cell antigen-related) (Fig.2) . The predicted protein fragment (SEQ. ID. No.2) translated by the only possible open reading frame comprises an incomplete N-terminal extracellular region, a 21 amino acid transmembrane-spanning region, and a 379 amino acid intracellular region.

To obtain the 5' end of the cDNA, the pancreas library was rescreened with a 300 bp probe close to the 5' end of C3. No IA-2 clones were found. Of the positive clones, clone Bll appeared to have the longest 5' extension to C3 and was selected for further sequencing along both strands . The 3' end of Bll was identical to C3 sequence over a length of about 2 kb, but the 5' ends of Bll and C3 were different (Fig. 2C) . Bll had a unique 5' sequence of 412 bp, followed by a 54 bp sequence that is found in the opposite orientation in C3. The sequence of C3 found 5' to this inverted region was shorter and different from that of Bll. 5' RACE was used to determine which clone contained the correct 5' sequence and to try to obtain additional 5' sequence to the open reading frames of Bll and C3. Template cDNA was prepared from human brain polyA ^* mRNA (using the Marathon cDNA amplification Kit (Clontech) using random primers or a primer (P5, with the sequence 5'-

CGTGTGGGCCACATAGGTCAGGATGCTCTCGGAGAA-3' (SEQ. ID. No. 20) ) corresponding to shared B11/C3 sequence located about 170 bp 3 ' to the invert. The 5' RACE used a primer to the ligated adaptor sequence at the ends of the cDNA and an IAR specific reverse primer (Pll with the sequence

5' -TGGCGAGCACGTCTGAGGCTG-3' (SEQ. ID. No. 21)) to a region of shared B11/C3 sequence located about 90 bp 3' to the

invert. The amplified fragments were cloned into pGEM-T (Promega) and sequenced along both strands. Cloning and sequencing of the 5' RACE PCR products from both types of template cDNA showed that all had Bll sequence. Similarly, PCR amplification of the pancreas cDNA library using a primer to λgtll and either P5 or Pll primers gave DNA fragments which all corresponded to the unique BL1 sequence. Thus the Bll sequence is correct and the 5' end of the C3 clone is likely a library artifact. No PCR products were found with a 5' extension of the Bll sequence, and no 5'-extended clones were identified by further screening of human brain and pancreas cDNA libraries.

Although no in-frame stop codons were found 5' to the first ATG codon in the Bll sequence, we propose that this is the initiation codon for the following reasons: 1) a purine (G) in the -3 position conforms with the Kozak rules of initiation (Kozak (1989)), and 2) this methionine is followed by a 20 residue hydrophobic sequence that has the features of a signal sequence (Von Heijne 1986)) . To test for signal sequence function, cDNA encoding full-length IAR protein was transcribed and translated in vi tro in the presence of microsomal membranes. Subsequent treatment with Proteinase K reduced the size of the translated product to a fragment corresponding to the predicted size of the extracellular region of IAR (data not shown) . This suggests that the IAR protein is inserted into the membranes such that the extracellular region of IAR is in the Proteinase K-inaccessible interior of the membrane vesicles.

The complete cDNA (SEQ. ID. No. 10) is predicted to encode a receptor-like PTP of 1015 amino acids (SEQ. ID. No. 11) . This includes a 614 amino acid extracellular region, a 25 residue hydrophobic transmembrane segment, and a 376 amino acid intracellular region containing a single PTP catalytic domain (SEQ. ID. No. 11) . This protein has: about 43% overall sequence identity to the putative PTP IA-

2 (Lan et al (1994) ; Rabin et al (1994)) , a protein identified as an islet cell auto-antigen in IDDM, and another PTP-like molecule called PTPLP ( (Kambayashi et al (1995)) . We have thus named the novel PTP as islet cell antigen-related PTP or IAR PTP. The IAR extracellular region does not contain FN-III or Ig-like repeats, but has the adhesion recognition peptide sequence RDGS (amino acids 372-375) . There is one potential site for N-linked glycosylation. The catalytic domain of IAR has conserved sequences typical of other PTPases, although the active site is unusual in having an aspartate residue at position 947 (IVHCSDGAGRTG (SEQ. ID. No. 16)) in place of a conserved alanine residue.

In view of the homology of IAR and IA-2, we selected non-homologous fragments of the C3 and D2 clones to use as specific probes on Northern blots in order to determine whether distinct transcripts were identified. The IAR probe comprised nucleotides 790-1197 of SEQ. ID. No.l (corresponding to a part of the predicted extracellular region) , and the IA-2 probe comprised nucleotides 1665-1906 of IA-2 cDNA (Lan et al . , 1994), corresponding to a part of the extracellular region, the transmembrane region, and a small portion of the intracellular region. These probes were labelled with ³² P as described in the legends to Figs. 3 and 4. A set of human multiple tissue Northern blots (Clontech) was probed first with the IA-2 probe (Fig.3) , and later stripped and reprobed with the IAR probe (Fig. 4) . The two probes recognize distinct transcripts in terms of size and tissue specificity. The IA-2 probe detects a single transcript of about 3.9kb (Lan et al . , 1994) , whereas the IAR probe detects two transcripts of about 3.7 and 5.5kb. It is possible that the 3.7kb transcript has a 3' untranslated sequence corresponding to that of the 3-7 clone, while the larger 5.5kb transcript represents an alternative 3' untranslated sequence corresponding to that in the longer 3-10 clone. The highest expression of IA-2 is detected in brain, followed by spinal cord and pancreas,

with low levels expressed in small intestine and adrenal gland. The highest expression of IAR is in pancreas and brain, followed by trachea, prostate, stomach and spinal cord, with low levels detectable in small intestine and adrenal gland. Figure 5 shows an alignment of the IAR and IA-2 transcripts detected in brain and pancreas. It is obvious that the IA-2 transcript is larger than the smallest 3.7kb IAR transcript, and furthermore the 3.7kb IAR transcript is much more highly expressed in pancreas than brain while the 3.9kb IA-2 transcript is more highly expressed in brain than pancreas. Thus IAR and IA-2 appear to be encoded by distinct genes .

3. Expression and analysis of the intracellular region of IAR

The intracellular region of IAR was amplified by PCR using the 3-7 clone as template. The primers used had the sequences 5' -GGGCTCGAGTCTAGACAGGCTGAAGGAGAAGCTCTC-3 ' (SEQ ID No. 5) (corresponding to nucleotides 1517-1538 of SEQ. ID. No.l, with added Xbal and Xhol sites) and 5'- GGGGAATTCCATGGTTATAATAGAAGACACACA-3' (SEQ ID No. 6) (corresponding to nucleotides 2704-2723 of SEQ. ID. No.l, with added EcoRI and Ncol sites) . PCR was carried out at 94°C x 5 min >(94°C x 40 sec >60°C x 30 sec >72°C x 2 min) x 10 >72°C x 5 min >4°C using the Expand Long PCR

Kit (Boehringer Mannheim) . The amplified fragment was cloned into the Xbal and Ncol sites of pGEX-KG (Guan and Dixon, 1991) and sequencing of the entire insert confirmed that no mutations had resulted from PCR. Later, the insert was cut out of this plasmid with Xhol and EcoRI and clon«;d into these sites in pGEX-3C (Walker et al . , 1994) . Mutant IAR (C945S) was generated by PCR using overlapping forward and reverse primers corresponding to the desired mutation and surrounding sequence encoding IIVHSSDGA (SEQ. ID. No. 19) ;

5' -ATAATTGTTCATTCCAGTGACGGTGCA-3' (SEQ. ID. No. 17) and 5' -TGCACCGTCACTGGAATGAACAATTAT-3' (SEQ. ID. No. 18) . Each

primer was used in combination with one of the primers (of SEQ. ID. No. 5 or 6) (used to amplify the intracellular region of IAR) to generate two overlapping DNA fragments corresponding to nucleotides 1993-2904 and 2878-3197 of IAR. These fragments were mixed and used as template for a PCR reaction with the primers of SEQ. ID. No. 5 and 6. The above plasmids were transformed into E. coli (DH5αF' ) . The recombinant bacterial host was cultured and GST-IAR expression induced with 0.15mM IPTG. The GST-IAR was purified from bacterial lysates by standard methods (Smith and Johnson, 1988) . The IAR was cleaved from GST using thrombin or 3C protease as appropriate (Smith and Johnson, 1988; Walker et al . , 1994), and the purified IAR quantitated by SDS-PAGE followed by densitometric scanning alongside known amounts of standards as described (Wang and Pallen, 1991) . Cleavage with thrombin of GST-IAR produced from the pGEX-KG plasmid resulted in a major protein band on SDS-PAGE which migrated with an apparent size of about 36kD (Fig. 6A, lane 4) . This was smaller than the predicted size of the expressed IAR insert, suggesting that internal cleavage of the IAR polypeptide by thrombin might have occurred. This led to recloning of the insert into the pGEX-3C vector as described above. Cleavage with 3C protease of GST-IAR produced from this plasmid resulted in a major protein band on SDS-PAGE which migrated with the expected size of about 40kD (Fig.6A, lane 3) . The purified 40kD form of the IAR intracellular region was analyzed for phosphatase activity. Low but detectable phosphatase activity towards the substrate para-nitrophenyl phosphate was measured (Fig. 6B) . Since the related molecule IA-2 reportedly has no detectable phosphatase activity (Rabin et al . , 1994; Lu et al. , 1994) , we examined this in side-by-side experiments with IAR.

4. Expression and analysis of the intracellular region of IA-2

The intracellular region of IA-2 was amplified by PCR

using the D2 clone as template. The primers used had the sequences 5' -GGGCTCGAGTCTAGAGCGGCAGCAAGACAAGGAG-3 ' (SEQ IE No. 7) (corresponding to nucleotides 1885-1903 of IA-2 (Lan et al . , 1994) , with added Xbal and Xhol sites) and 5' -GGGGAATTCGAGCTCGCATGCCCAAGAGGTGGC-3' (SEQ ID No. 8)

(corresponding to nucleotides 3104-3121 of IA-2, with added EcoRI and Sad sites) . PCR was carried out as described above for the IAR intracellular region. The amplified fragment was cloned into the Xbal and Sad sites of pGEX-KG (Guan and Dixon, 1991) . Later, the insert was cut out of this plasmid with Xhol and EcoRI and cloned into these sites in pGEX-3C (Walker et al. , 1994) .

Recombinant bacterial GST-IA-2 protein was produced from both the above plasmids, cleaved, purified and quantitated as described above for IAR. However, unlike IAR, the type of protease used for cleavage from GST did not affect the size of the IA-2 produced, with the purified IA-2 migrating on SDS-PAGE in both cases as a major band with an apparent size of 44kD (Fig.6A, lanes 1 and 2) . Thus the IAR and IA-2 proteins differ in that IAR appears to possess an internal thrombin cleavage site which is absent in IA-2.

5. Catalytic activity of IAR The purified 44kD form of IA-2 (produced by cleavage with 3C protease) possessed phosphatase activity towards para-nitrophenyl phosphate which was roughly comparable to that of IAR (Fig. 6C) .

The intracellular region of IAR (amino acids 646-1015) was expressed as a GST-fusion protein and purified following 3C protease cleavage from GST. The IAR migrated on SDS-PAGE as a major protein band of about 41 kDa (Fig. 7A, lane 2) , in accord with the predicted size of 41.7 kDa. The purified IAR possessed low but detectable phosphatase activity towards pNPP. The IAR is active over εi narrow pH range, with optimal activity at pH 4.5, and is essentialLy inactive at pH 5.5 (Fig.7B) . IAR catalyzes the time-

dependent dephosphorylation of pNPP (Fig. 7C) with a specific activity of 21 nmol/min/mg. Like many PTPs, IAR activity is sensitive to inhibition by vanadate, and ImM sodium orthovanadate completely abolished activity (Fig. 7C) . Site-directed mutagenesis of the essential cysteine residue (to a serine) in the active site resulted in the expression of an IAR intracellular region polypeptide which did not catalyze pNPP hydrolysis (Fig. 7C) , indicating that the phosphatase activity measured with the wild type polypeptide was unlikely to be due to contaminants in the protein preparation. IAR dephosphorylated phosphotyrosyl casein but not phosphoseryl casein.

6. Reactivity of IAR with IDDM Patient Sera

Patients and Sera. Sera were obtained from healthy blood bank controls (n=10) and from recent-onset IDDM patients (n=20, age 4.0-15.7 years) prior to commencement of insulin therapy. ICA were measured by indirect immunofluorescence on frozen sections of human pancreas

(Bonifacio et al (1990) ) , and GAD antibodies were measured by enzymatic immunoprecipitation (Schmidli et al (1994)) .

Tn Vitro Translation and Immunoprecipitation Tests with Patient Sera. cDNAs encoding amino acids 646-1015 of IAR (representing the intracellular region) or full-length IA-2 protein (Lan et al (1994)) were subcloned into the pGEM-3 vector, under the control of the SP-6 promoter. Radiolabelled IAR and IA-2 proteins were synthesized by in vi tro translation and transcription in the presence of ^3S S- methionine, using a reticulocyte lysate system (Promega TNT Kit) . For immunoprecipitations, lμl of lysate containing labelled IAR or IA-2 in lOOμl TBST-BSA (50mM Tris-HCl, 150mM NaCl, 1% Tween-20, 0.1% BSA) was incubated overnight at 4 °C with 2μl of serum which had been prebound to lOμl (packed bead volume) Protein A-Sepharose (Pharmacia) .

Beads were washed once with TBST-BSA, once with high salt wash buffer (0.5% Triton X-100, 200mM NaCl, 50mM NaH,PO„) ,

twice in PBS plus 0.5% Triton X-100 and once in 0.5% Triton X-100, 4.17mM Tris, 192mM glycine, pH 7.4. Precipitated proteins were resolved by 10% SDS-PAGE. Gels were dried and exposed to a phosphorimager, and bands were quantified by densitometry after subtraction of local background.

Sera from 10 healthy blood bank control and 20 recent-onset IDDM subjects (ICA, GAD antibodies, age and sex cf subjects are shown in Table 1) were incubated with in vi tro translated ^3S S-methionine labelled IAR (amino acids 646- 1015) . Immunoprecipitates were analyzed by SDS-PAGE, where the translated IAR protein appeared as a band of 41 kDa with an associated proteolytic breakdown product of 32 kDa. Representative immunoprecipitations of IDDM and control sera are shown in Fig. 8A. Quantitative analysis of a panel of sera showed that band densities were greater than the mean +2SD of the 10 control sera (1818 density units) in 11 of the 20 (55%) recent-onset IDDM sera (Table 1) . There was no significant correlation, in this small number of subjects, between IAR and ICA, GAD antibodies, or age. The IAR reactivity of the control and IDDM subjects is shown graphically in Fig. 8B. Sera from 9 of the: 11 IAR- reactive patients also immunoprecipitated IA-2, some reacting more strongly with IAR than IA-2 and the: inverse seen with others (Table 1) . Of interest is the finding that 3 IDDM subjects had antibodies which immunoprecipitated IAR or IA-2, but not both (Table 1, subjects 4, 13, 20) . Thus IAR and IA-2 can distinguish different autoantibody populations in a subset of IDDM patients.

Table 1. Reactivity of sera with ICA, GAD, IAR and IA-2

8. Comparison of the predictive value of IAR and IA-2 antibodies for IDDM (see Figures 9 and 10)

Sera from recent-onset IDDM subjects (n=53) were

obtained piror to commencement of insulin therapy. The healthy control group (n=144) consisted of laboratory volunteers and healthy schoolchildren. First-degree relatives of subjects with IDDM (n=3315) were enrolled in the Melbourne Prediabetes Family Study and had ICA and I A measured at entry. These subjects form an "at-risk" group in that the risk of developing IDDM in ICA positive relatives has been previously shown to be 35% in five years and 60-70% in 10 years (reviewed in Bingley et al (1993) . As an alternative to population screening, we have studied this well-defined at risk-group of subjects who have been prospectively followed, to compare the predictive power of IAR Ab and IA-2 Ab.

Measurement of IAR Ab and IA-2 Ab in these sera has allowed us direct comparison of the abilities of these assays to predict IDDM, using identical assay formats and standardised reference ranges. In total, 33 of these subjects have developed IDDM. IAR and IA-2 antibodies were measured in 115 subjects with ICA > _. 20 and/or IAA _> 100, who have been followed for 3.52 years (mean, range 0.30- 7.28 years) . Of this group, 18 progressed to IDEM during follow-up over 2.28 years (range 0.3-4.52 years) .

IAR and IA-2 antibody radiobinding assays

Coupled in vi tro transcription and translation was used to synthesise ³⁵ S-labelled protein, which was then immunoprecipitated with sera using a modification of the method of Bonifacio et al (1995) . Complementary DNA (cDNA) encoding full length human IAR was cloned into the pBluescript vector under control of the T3 promoter; cDNA encoding amino acids 646-1015 of IAR, representing the intracellular region was inserted into the pGEM3 vector under control of the SP6 promoter and cDNA encoding full- length IA-2 was inserted into the pSP64 poly A vector under control of the SP6 promoter. Recombinant protein was synthesised using the Promega TNT reticulocyte lysate kit

(Promega, Madison, Wis., USA) , in the presence of ^b S methionine, according to the manufacturers' instructions, using the appropriate RNA polymerase. Two μg DNA template was used in all in vi tro translation and transcription reactions, which proceeded for two hours at 30°C.

The transcription/translation product (50μl) was diluted into 12ml TBST-BSA buffer (50mM Tris-HCl pH 7.2, 150mM NaCl, 1% Tween 20, ImM L-methionine, 0.1% BSA), and 50μl of this dilute lysate was added to 5μl test serum and incubated overnight at 4°C in a 96-well plate (well volune 2ml, Beckman, Palo Alto, CA, USA) . Fifty μl of protein A Sepharose suspension (50mg/ml dry weight) in TBST-BSA was added and the plate was shaken on a plate shaker for lh at 4°C. Beads were washed x3 with 1ml TBST-BSA in the plate. Immunoprecipitates were transferred to a 96-well Unifilter plate (Packard, meriden, CT, USA) , using a multichannel pipette, in lOOμl TBST-BSA. Scintillation fluid (30μl Microscint 40, Packard) was added to the wells and radioactivity was counted in a TopCount Microplate Scintillation Counter (Packard) . Results were expressed as arbitrary units, calculated using the formula:

Result= ( (sample - negative control) / (positive control - negative control) ) .

The negative control consisted of pooled normal serum and the positive control was strongly IAR and IA-2 positive plasmapheresis fluid from a patient with stiff-man syndrome and diabetes. The inter assay coefficient of variation of the IAR antibody assay was 5.8% for a sample with a mean value of 57.3 units, and 12.5% for the IA-2 Ab assay, using a sample with a mean value of 96.8 units.

ICA and IAA assays

ICA were measured by indirect immunofluorescence on frozen sections of blood group O human pancreas (Bonifacio

et al (1990)) and assigned a value in JDF units by dilution and comparison with standard sera. This assay has been standardised in all Immunology of Diabetes and Proficiency Workshops. Insulin autoantibodies were measured by a competitive fluid phase radiobinding assay (Vardi et al (1989) ) .

Intravenous glucose tolerance test (IVGTT)

The Islet Cell Antibody Registry Users Study (ICARUS) recommended protocol was used for the IVGTT, as previously described (McNair et al (1995) ) . Serum insulin was measured using a commercial RIA kit (Pharmacia, Uppsala, Sweden) .

Statistical analysis

Associations between parameters were assessed using regression analysis. To standardise the specificty of antibody assays for comparison of sensitivity, the threshold for positivity of each assay was adjusted to derive a specificity of 97-98%. This level of specificity was chosen because it detects a sufficiently small number of false positives to be useful in population screening. The effect of IAR and IA-2 antibody positivity or IDDM-free survival was analysed using the log-rank test.

Results

Before comparing IAR Ab with IA-2 Ab, the partial (intracellular domain) and full-length IAR clones were compared for recognition by IDDM sera, to determine whether additional reactivity was present on the extracellular domain of the full-length IAR molecule that was not present in the intracellular (cytoplasmic) domain. There was a strong relationship between levels of antibodies to full- length IAR, and antibodies to the intracellular domain

(r=94.9%, p<0.0001) . Use of a standardised specificity of 97.9% resulted in a sensitivity of 58.1% for antibodies to the cytoplasmic domain and 48.8% for antibodies to the full-length molecule. Because of the strong relationship between antibody levels and the similar assay sensitivities, the intracellular fragment was used as the antigen in the remaining experiments.

IAR Ab and IA-2 Ab were measured in 53 recent-onset IDDM and 144 healthy control subjects. The reference range for each assay was adjusted to give a specififcity of 97.9% for each assay, which was 0.5 units for IAR Ab and 3.0 units for IA-2 Ab. The corresponding sensitivities were 56.6% for IAR Ab and 62.3% for IA-2 Ab (Table 2) . Sensitivities and specificities of combinations of IAR Ab and IA-2 Ab, and IAR Ab or IA-2 Ab, using the reference ranges described above, are also shown in Table 2. The presence of either antibody resulted in a sensitivity of 73.6% and a specificity of 96.5%, and the presence of both antibodies gave a specificity of 100% and a sensitivity of 45.3%. Although levels of IAR Ab and IA-2 Ab were correlated in recent-onset subjects (r=0.7, p<0.0001, Figure 9) and there was a degree of overlap, some patients were positive for one antibody only. Thus, six (11.3%) were positive for IAR Ab and negative for IA-2 Ab, and nine (17%) were positive for IA-2 Ab and negative for IAR Ab.

In some cases there were marked differences between levels of IAR Ab and IA-2 Ab, for example there were six sera with IA-2 Ab levels greater than 50 units and IAR Ab less than 5 units, and one with IAR Ab of 25 units and IA-2 Ab of 2.9 units.

We next examined the ability of IAR Ab and IA-2 to discriminate between IDDM progression and non-progression, in our prospective study of first-degree relatives with ICA > 20JDFU or IAA >_ lOOnU/ml. IAR and IA-2 antibody positivity and combinations of these antibodies are summarised in Table 2. Of the 18 subjects who progressed to IDDM, IAR Ab were detectable in nine (50.0%) and IA-2 Ab

in 10 (55.6%) 2.6± 1.8 _. (mean, range 0.3-6.3) years prior to diagnosis, using the threshold for positivity defined above. Of the 97 nonprogressors, only 21 (21.6%) had IAR. antibodies, whereas 31 (31.9%) had IA-2 antibodies. The difference in IDDM-free survival between IAR Ab positive and negative subjects was statistically significεint (p<0.001 log-rank test), whereas IA-2 Ab positivity did not significantly influence IDDM-free survival. The combination of either IAR or IA-2 antibodies did not significantly affect IDDM-free survival, but the presence of both antibodies had a significant effect (p<0.0005) . Although IA-2 Ab positivity did not affect IDDM-free survival when these standardised thresholds of positivity were used, the use of higher thresholds for IA-2 Ab positivity resulted in significant differences. For example, when nine units was used as the threshold, IA-2 Ab positivity was significantly associated with progression to IDDM (p<0.0005, log-rank test) .

In addition to predicting the development of IDDM, IAR Ab were associated with markers of high risk of IDDM, in relatives with ICA or IAA. Insulin autoantibodies (Ziegler et al (1989)) and impaired intravenous glucose tolerance (Vardi et al (1991)) have both been previously shown to be associated with progression to IDDM in ICA positive first- degree relatives. IAA were significantly correlated with IAR antibodies (r=0.25, p=0.005) but not with IA-2 antibodies (r=0.11, N.S.) . Both IAR and IA-2 antibodies were negatively correlated with the sum of the 1 and 3 minute insulin levels during an intravenous glucose tolerance test (IAR: r=0.27, p<0.01; IA-2: r=0.25, p<0.05) . In addition to these relationships, both IAR and IA-2 antibodies were strongly correlated with ICA levels (IAR: r=0.31, p<0.001, IA-2 :r=0.38) .

There were some subjects in whom there were marked differences in the level of each antibody. Furthermore, in our group of high-risk ICA or IAA positive first degree relatives, the presence of IAR Ab was associated with

progression to IDDM whereas IA 2 Ab was not Neither combinations of IAR Ab or IA 2 Ab, nor IAR Ab and IA-2 Ab resulted m better prediction of IDDM than IAR Ab alone m the prospective study; thus, combined measurement of PTP antibodies does not appear to be preferable to measurement of IAR Ab alone IDDM screening programmes. In addition to discriminating subjects who progressed to IDDM, IAR Ab were correlated with other markers of progression to IDDM, namely IAA and impaired first-phase insulin release. These results provide evidence that IAR Ab are markers of β cell destruction.

Levels of antibodies against the cytoplasmic domain of IAR were closely correlated with those against the full- length clone, and sensitivities of both assays were similar, suggesting that antibodies from IDDM patients recognise the cytoplasmic domain of IAR.

In conclusion, we have demonstrated that IAR Ab have similar diagnostic sensitivity for recent-onset IDDM to IA- 2 Ab, but are better predictors of IDDM in our high-risk subjects As preclinical intervention therapies m IDDM are critically reliant on effective screening tests to define high-risk subjects, IAR Ab may well have an important role IDDM screening, either alone or m combination with other humoral markers of IDDM

TABLE 2

IAR Ab IA-2 Ab IAR Ab or IAR Ab or & IA-2 Ab. IA-2 Ab

Positivity threshold (units) <0.5 <3.0

Recen -onset IDDM Sensitivity (%) 56.6 62.3 73.6 45.3

Specificity (%) 97.9 97.9 96.5 100.0

ICA positive first-degree rela tives % Progressors positive 55.6 % Nonprogressors positive 21.6 pdogrank test) <0.001

Table 2 Comparison of diagnostic sensitivity and predictive power for IDDM, of IAR Ab, IA-2 Ab and combinations of IAR Ab and/or IA-2 Ab. Thresholds for positivity described in text

References:

Adams, M.D., Kerlavage, A.R., Fields, C. and Venter, J.C.

(1993) . Nature Genet. 4, 256-267. Guan, K. and Dixon, J.E. (1991) . Analyt . Biochem. 192,

262-267.

Lan, M.S., Lu, J., Goto, Y. and Notkins, A.L. (1994) . DNA

Cell Biol. 13, 505-514.

Lu, J., Notkins, A.L. and Lan, M.S. (1994) . Biochem. Biophys. Res. Commun. 204, 930-936.

Rabin, D.U., Pleasic, S.M., Shapiro, J.A., Yoo-Warren, H.,

Oles, J., Hicks, J.M., Goldstein, D.E. and Rae, P.M.M.

(1994) . J. Immunol.152, 3183-3188.

Smith, D.B. and Johnson, K.S. (1988) . Gene 67, 31-40.

Walker, P.A., Leong, L. E.-C, Ng, P.W.P., Tan, S.H.,

Waller, S., Murphy, D. and Porter, A.G. (1994) . Bio/Technology 12, 601-605.

Wang, Y. and Pallen, CJ. (1991) . EMBO J. 10, 3231-3237.

Kozak, M. (1989) J. Cell Biol. 108, 229-241. von Heijne, G. (1986) Nuc. Acids Res. 14, 4683-4690.

Kambayashi, Y., Takahashi, K. , Bardhan, S., & Inagami, T. (1995) Biochem. J. 306, 331-335.

Bonifacio, E., Bingley, P.J., Shattock, M. , Dean, B.M.,

Dunger, D., Gale, E.A.M., & Bottazzo, G.F. (1990) Lancet

335, 147-149.

Schmidli, R.S., DeAizpurua, H.J. , Harrison, L.C. & Colman, P.G. (1994) J. Autoimmun. 7, 55-66.

Bingley P.J., Bonifacio E., Gale E.A.M. (1993) Diabetes

42 :213-20.

Ziegler A.G., Ziegler R., Vardi P., Jackson RA, Soeldner

J.S., Eisenbarth G.S. (1989) . Diabetes 38:1320-5. Bonifacio E., Genovese S., Braghi S., Bazzigaluppi E. ,

La pasona V., Bingley P.J. et al , (1995) . Diabetologia

38 :816-822.

Vardi P., Dib S.A., Tuttleman M., Connelly J.E. , Grinbergs

M, Radizabeh A., Riley W.J., Maclaren N.K., Eisenbarth G.S., Soeldner J.S., (1987) . Diabetes 36:1286-91.

McNair P.D., Colman P,G, Alford FP, Harrison LC (1995) .

Diabetes Care 18:1168-73.

Vardi P., Crisa L., Jackson R.A. (1991) Diabetologia 34:93-

102.

SEQUENCE INFORMATION

SEQ. ID. Nos. 1 and 2: partial cDNA sequence of IAR-PTP (SEQ ID

NO. 1) and encoded amino acid sequence (SEQ ID NO. 2) .

AGT CGG GAG GGC GGT GCT GCC CTG GCC AAC GCC CTC CGA CGC CAC CTG

Ser Arg Glu Gly Gly Ala Ala Leu Ala Asn Ala Leu Arg Arg His Leu 5 10 15

CCC TTC CTG GAG GCC CTG TCC CAG GCC CCA GCC TCA GAC GTG CTC GCC

Pro Phe Leu Glu Ala Leu Ser Gin Ala Pro Ala Ser Asp Val Leu Ala 20 25 30

AGG ACC CAT ACG GCG CAG GAC AGA CCC CCC GCT GAG GGT GAT GAC CGC

Arg Th His Thr Ala Gin Asp Arg Pro Pro Ala Glu Gly Asp Asp Arg 3b 40 45

TTC TCC GAG AGC ATC CTG ACC TAT GTG GCC CAC ACG TC']' GCG CTG ACC

Phe Ser Glu Ser lie Leu Thr Tyr Val Ala His Thr Ser Ala Leu Thr 50 55 60

TAC CCT CCC GGG TCC CGG ACC CAG CTC CGC GAG GAC CTC CTG CCG CGG

Tyr Pro Pro Gly Ser Arg Thr Gin Leu Arg Glu Asp Leu Leu Pro Arg 65 70 75 80

ACC CTC GGC CAG CTC CAG CCA GAT GAG CTC AGC CCT AAG GTG GAC AGT

Thr Leu Gly Gin Leu Gin Pro Asp Glu Leu Ser Pro Lys Val Asp Ser 85 90 95

GGT GTG GAC AGA CAC CAT CTG ATG GCG GCC CTC AGT GCC TAT GCT GCC

Gly Val Asp Arg His His Leu Met Ala Ala Leu Ser Ala Tyr Ala Ala 100 105 110

CAG AGG CCC CCA GCT CCC CCC GGG GAG GGC AGC CTG GAG CCA CAG TAC

Gin Arg Pro Pro Ala Pro Pro Gly Glu Gly Ser Leu Glu Pro Gin Tyr 115 120 125

CTT CTG CGT GCA CCC TCA AGA ATG CCC AGG CCT TTG CTG GCA CCA GCC

Leu Leu Arg Ala Pro Ser Arg Met Pro Arg Pro Leu Leu Ala Pro Ala 130 135 140

GCC CCC CAG AAG TGG CCT TCA CCT CTG GGA GAT TCC GAA GAC CCC TCC

Ala Pro Gin Lys Trp Pro Ser Pro Leu Gly Asp Ser Glu Asp Pro Ser 145 150 155 160

AGC ACA GGC GAT GGA GCA CGG ATT CAT ACC CTC CTG AAG GAC CTG CAG

Ser Thr Gly Asp Gly Ala Arg lie His Thr Leu Leu Lys Asp Leu Gin 165 ^{^} 170 175

AGG CAG CCG GCT GAG GTG AGG GGC CTG AGT GGC CTG GAG CTG GAC GGC

Arg Gin Pro Ala Glu Val Arg Gly Leu Ser Gly Leu Glu Leu Asp Gly 180 185 190

ATG GCT GAG CTG ATG GCT GGC CTG ATG CAA GGC GTG GAC CAT GGA GTA

Met Ala Glu Leu Met Ala Gly Leu Met Gin Gly Val Asp His Gly Val

195 200 205

GCT CGA GGC AGC CCT GGG AGA GCG GCC CTG GGA GAG TCT GGA GAA CAG

Ala Arg Gly Ser Pro Gly Arg Ala Ala Leu Gly Glu Ser Gly Glu Gin

210 ^" 215 220

GCG GAT GGC CCC AAG GCC ACC CTC CGT GGA GAC AGC TTT CCA GAT GAC

Ala Asp Gly Pro Lys Ala Thr Leu Arg Gly Asp Ser Phe Pro Asp Asp 225 ^' 230 235 240

GGA GTG CAG GAC GAC GAT GAT AGA CTT TAC CAA GAG GTC CAT CGT CTG

Gly Val Gin Asp Asp Asp Asp Arg Leu Tyr Gin Glu Val His Arg Leu 245 250 255

AGT GCC ACA CTC GGG GGC CTC CTG CAG GAC CAC GGG TCT CGA CTC TTA

Ser Ala Thr Leu Gly Gly Leu Leu Gin Asp His Gly Ser Arg Leu Leu 260 265 270

CCT GGA GCC CTC CCC TTT GCA AGG CCC CTC GAC ATG GAG AGG AAG AAG

Pro Gly Ala Leu Pro Phe Ala Arg Pro Leu Asp Met Glu Arg Lys Lys 275 280 285

TCC GAG CAC CCT GAG TCT TCC CTG TCT TCA GAA GAG GAG ACT GCC GGA

Ser Glu His Pro Glu Ser Ser Leu Ser Ser Glu Glu Glu Thr Ala Gly 290 295 300

GTG GAG AAC GTC AAG AGC CAG ACG TAT TCC AAA GAT CTG CTG GGG CAG

Val Glu Asn Val Lys Ser Gin Thr Tyr Ser Lys Asp Leu Leu Gly Gin 305 310 315 320

CAG CCG CAT TCG GAG CCC GGG GCC GCT GCG TTT GGG GAG CTC CAA AAC

Gin Pro His Ser Glu Pro Gly Ala Ala Ala Phe Gly Glu Leu Gin Asn

3 25 3 3 0 3 35

CAG ATG CCT GGG CCC TCG /AC GAG GAG CAG AGC CTT CCA GCG GGT GCT

Gin Met Pre. Gly PΪ et y Glu Glu Gin Sei Leu Pio Ale Gly l.. 340 3 5 350

CAG GAG GCC CTC AGC GAG GGC CTG CAA TTG GAG GTC CAG CCT TCC GAG

Gin Glu Ala Leu Ser Asp Gly Leu Gin Leu Glu Val Gin Pro Ser Glu 355 360 365

GAA GAG GCG CGG GGC TAC ATC GTG ACA GAC AGA GAC CCC CTG CGC CCC

Glu Glu Ala Arg Gly Tyr lie Val Thr Asp Arg Asp Pro Leu Arg Pro 370 375 380

GAG GAA GGA AGG CGG CTG GTG GAG GAC GTC GCC CGC CTC CTG CAG GTG

Glu Glu Gly Arq Arg Leu Val Glu Asp Val Ala Arg Leu Leu Gin Val 385 390 395 400

CCC AGC AGG GCG TTC GC'l GAC GTG GAG GTT CTC GGA CCA GCA GTG ACC

Pro Ser Sci Ala Pht Ala Asp Val Glu Val Leu Gly Pro Ala Val Thr 405 10 15

TTC AAA GTG AGC GCC AAT GTC CAA AAC GTG ACC ACT GAG GAT GTG GAG

Phe Lys Val Ser Ala Asn Val Gin Asn Val Thr Thr Glu Asp Val Glu 420 425 430

AAG GCC ACA GTT GAC AAC AAA GAC AAA CTG GAG GAA ACC TCT GGA CTG

Lys Ala Thr Val Asp Asn Lys Asp Lys Leu Glu Glu Thr Ser Gly _^ eu 435 440 445

AAA ATT CTT CAA ACC GGA GTC GGG TCG AAA AGC AAA CTC AAG TTC CTG

Lys lie Leu Gin Thr Gly Val Gly Ser Lys Ser Lys Leu Lys Phe Leu 450 455 460

CCT CCT CAG GCG GAG CAA GAA GAC TCC ACC AAG TTC ATC GCG CTC ACC

Pro Pro Gin Ala Glu Gin Glu Asp Ser Thr Lys Phe lie Ala Leu Thr 465 470 475 480

CTG GTC TCC CTC GCC TGC ATC CTG GGC GTC CTC CTG GCC TCT GGC CTC

Leu Val Ser Leu Ala Cys lie Leu Gly Val Leu Leu Ala Ser Gly Leu 485 490 495

ATC TAC TGC CTC CGC CAT AGC TCT CAG CAC AGG CTG AAG GAG AAG CTC lie Tyr Cys Leu Arg His Ser Ser Gin His Arg Leu Lys Glu Lys Leu 500 505 510

TCG GGA CTA GGG GGC GAC CCA GGT GCA GAT GCC ACT GCC GCC TAC CAG

Ser Gly Leu Gly Gly Asp Pro Gly Ala Asp Ala Thr Ala Ala Tyr Gin

515 520 525

GAG CTG TGC CGC CAG CGT ATG GCC ACG CGG CCA CCA GAC CGA CCT GAG

Glu Leu Cys Arg Gin Arg Met Ala Thr Arg Pro Pro Asp Arg Pro Glu 530 535 540

GGC CCG CAC ACG TCA CGC ATC AGC AGC GTC TCA TCC CAG TTC AGC GAC

Gly Pro His Thr Ser Arg lie Ser Ser Val Ser Ser Gin Phe Ser Asp 545 550 555 560

GGG CCG ATC CCC AGC CCC TCC GCA CGC AGC AGC GCC TCA TCC TGG TCC

Gly Pro lie Pro Ser Pro Ser Ala Arg Ser Ser Ala Ser Ser Trp Ser 565 570 575

GAG GAG CCT GTG CAG TCC AAC ATG GAC ATC TCC ACC GGC CAC ATG ATC

Glu Glu Pro Val Gin Ser Asn Met Asp lie Ser Thr Gly His Met lie 580 585 590

CTG TCC TAC ATG GAG GAC CAC CTG AAG AAC AAG AAC CGG CTG GAG AAG

Leu Ser Tyr Met Glu Asp His Leu Lys Asn Lys Asn Arg Leu Glu Lys 595 600 605

GAG TGG GAA GCG CTG TGC GCC TAC CAG GCG GAG CCC AAC AGC TCG TTC

Glu Trp Glu Ala Leu Cys Ala Tyr Gin Ala Glu Pro Asn Ser Ser Phe 610 615 620

GTG GCC CAG AGG GAG GAG AAC GTG CCC AAG AAC CGC TCC CTG GCC GTG

Val Ala Gin Arg Glu Glu Asn Val Pro Lys Asn Arg Ser Leu Ala Val 625 630 635 640

CTG ACC TAT GAC CAC TCC CGG GTC CTG CTG AAG GCG GAG AAC AGC CAC

Leu Thr Tyr Asp His Ser Arg Val Leu Leu Lys Ala Glu Asn Ser His 645 650 655

AGC CAC TCA GAC TAC ATC AAC GCT AGC CCC ATC ATG GAT CAC GAC CCG

Ser His Ser Asp Tyr lie Asn Ala Ser Pro lie Met Asp His Asp Pro 660 665 670

AGG AAC CCC GCG TAC ATC GCC ACC CAG GGA CCG CTG CCC GCC ACC GTG

Arg Asn Pro Ala Tyr lie Ala Thr Gin Gly Pro Leu Pro Ala Thr Val 675 680 685

GCT GAC TTT TGG CAG ATG GTG TGG GAG AGC GGC TGC GTG GTG ATC GTC

Ala Asp Phe Trp Gin Met Val Trp Glu Ser Gly Cys Val Val He Val 690 695 700

ATG CTG ACA CCC CTC GCG GAG AAC GGC GTC CGG CAG TGC TAC CAC TAC

Met Leu Thr Pro Leu Ala Glu Asn Gly Val Arg Gin Cys Tyr His Tyr 705 710 715 ^' 720

TGG CCG GAT GAA GGC TCC AAT CTC TAC CAC ATC TAT GAG GTG AAC CTG

Trp Pro Asp Glu Gly Ser Asn Leu Tyr His He Tyr Glu Val Asn Leu 725 730 735

GTC TCC GAG CAC ATC TGG TGT GAG GAC TTC CTG GTG AGG AGC TTC TAT

Val Ser Glu His He Trp Cys Glu Asp Phe Leu Val Arg Ser Phe Tyr 740 745 750

CTG AAG AAC CTG CAG ACC AAC GAG ACG CGC ACC GTG ACG CAG TTC CAC

Leu Lys Asn Leu Gin Thr Asn Glu Thr Arg Thr Val Thr Gin Phe His 755 760 765

TTC CTG AGT TGG TAT GAC CGA GGA GTC CCT TCC TCC TCA AGG TCC CTC

Phe Leu Ser Trp Tyr Asp Arg Gly Val Pro Ser Ser Ser Arg Ser Leu 770 775 780

CTG GAC TTC CGC AGA AAA GTA AAC AAG TGC TAC AGG GGC CGT TCT TGT

Leu Asp Phe Arg Arg Lys Val Asn Lys Cys Tyr Arg Gly Arg Ser Cys 785 790 795 800

CCA ATA ATT GTT CAT TGC AGT GAC GGT GCA GGC CGG AGC GGC ACC TAC

Pro He He Val His Cys Ser Asp Gly Ala Gly Arg Ser Gly Thr Tyr 805 810 815

GTC CTG ATC GAC ATG GTT CTC AAC AAG ATG GCC AAA GGT GCT AAA GAG

Val Leu He Asp Met Val Leu Asn Lys Met Ala Lys Gly A a Lys Glu 820 825 830

ATT GAT ATC GCA GCG ACC CTG GAG CAC TTG AGG GAC CAG AGA CCC GGC

He Asp He Ala Ala Thr Leu Glu His Leu Arg Asp Gin Arg Pro Gly 835 840 845

ATG GTC CAG ACG AAG GAG CAG TTT GAG TTC GCG CTG ACA GCC GTG GCT

Met Val Gin Thr Lys Glu Gin Phe Glu Phe Ala Leu Thr Ala Val Ala 850 855 860

GAG GAG GTG AAC GCC ATC CTC AAG GCC CTT CCC CAG TGAGCCGCAG

Glu Glu Val 11.; u Lys Ala Leu Pro Gin 865 870 875

CCTCAGGGGC ( 5 CΛGGGGΛC CCCCTACCCC ACGGATCTTG TCAGGAΛTC .5 TGATCTGACT

TTAATTGTGT GTCTTCTΛTT ATAACTGCAT AGTAATAGGG CCCTTAGCTC TCCCGTAGTC

AGCGCAGTTT AGCAGTTAAA AGTGTATTTT TGTTTAATCA AACAATAATA AAGAGAGGTT

TGTCGGAAAA ATCCAAAAAA AA

SEQ. ID. Nos. 10 and 11 Complete nucleotide (SEQ ID No. 10) and amino acid (SEQ ID NO. 11) sequence of IAR.

In SEQ ID Ke 11, the signal peptide is underlined ana the transmemtuantr region ιr double-underlined. An extracellular RGDS sequence and .- potential s:Lf of N- linked glycosylation are m bold. Tlτ ^" active site or the phosphatase is found at ammo acids 942-954.

GCCTCCCGCC GCCTCCCGCG CGGCCATGGA CTGAGCGCCG CCGGCCAGGC CGCGGGG

ATG GGG CCC CCG CTC CCG CTG CTG CTG CTG CTA CTG CTG CTG CTG CCG 105

Met Gly Pro Pro Leu Pro Leu Leu Leu Leu Leu Leu Leu Leu Leu Pro 1 10 15

CCA CGC GTC CTG CCT GCC GCC CCT TCG TCC GTC CCC CGC GGC CGG CAG

Pro Arc Val Leu Pro Ala Ala Pro Ser Ser Val Pro Arg Gly Arg Gin 20 25 30

CTC CCG GGG CGT CTG GGC TGC CTG CTC GAG GAG GGC CTC TGC GGA GCG

Leu Pro Gly Arg Leu Gly Cys Leu Leu Glu Glu Gly Leu Cys Gly Ala 35 40 45

TCC GAG GCC TGT GTG AAC GAT GGA GTG TTT GGA AGG TGC CAG AAG GTT

Ser Glu Ala Cys Val Asn Asp Gly Val Phe Gly Arg Cys Gin Lys Val 50 55 60

CCG GCA ATG GAC TTT TAC CGC TAC GAG GTG TCG CCC GTG GCC CTG CAG

Pro Ala Met Asp Phe Tyr Arg Tyr Glu Val Ser Pro Val Ala Leu Gin 65 70 75 80

CGC CTG CGC GTG GCG TTG CAG AAG CTT TCC GGC ACA GGT TTC ACG TGG

Arg Leu Am Val Ala Leu Gin Lys Leu Ser Gly Thr Gly Phe Thr Trp 85 90 95

CAG GAT GAC TAT ACT CAG TAT GTG ATG GAC CAG GAA CTT GCA GAC CTC

Gin Asp Asp Tyr Thr Gin Tyr Val Met Asp Gin Glu Leu Ala Asp Leu 100 105 110

CCG AAA ACC TAC CTG AGG CGT CCT GAA GCA TCC AGC CCA GCC AGG CCC

Pro Lys Thr Tyr Leu Arg Arg Pro Glu Ala Ser Ser Pro Ala Arg Pro 115 120 125

TCA AAA CAC AGC GTT GGC AGC GAG AGG AGG TAC AGT CGG GAG GGC GGT

Ser Lys His Ser Val Gly Ser Glu Arg Arg Tyr Ser Arg Glu Gly Gly 130 135 140

GCT GCC CTG GCC AAC GCC CTC CGA CGC CAC CTG CCC TTC CTG GAG GCC

Ala Ala Leu Ala Asn Ala Leu Arg Arg His Leu Pro Phe Leu Glu Ala 145 150 155 160

CTG TCC CAG GCC CCA GCC TCA GAC GTG CTC GCC AGG ACC CAT ACG GCG

Leu Ser Gin Ala Pro Ala Ser Asp Val Leu Ala Arg Thr His Thr Ala 165 170 175

CAG GAC AGA CCC CCC GCT GAG GGT GAT GAC CGC TTC TCC GAG AGC ATC

Gin Asp Arg Pro Pro Ala Glu Gly Asp Asp Arg Phe Ser Glu Ser He 180 185 190

CTG ACC TAT GTG GCC CAC ACG TCT GCG CTG ACC TAC CCT CCC GGG TCC

Leu Thr Tyr Val Ala His Thr Ser Ala Leu Thr Tyr Pro Pro Gly Ser 195 200 205

CGG ACC CAG CTC CGC GAG GAC CTC CTG CCG CGG ACC CTC GGC CAG CTC

Arg Thr Gin Leu Arg Glu Asp Leu Leu Pro Arg Thr Leu Gly Gin Leu 210 215 220

CAG CCA GAT GAG CTC AGC CCT AAG GTG GAC AGT GGT GTG GAC AGA CAC

Gin Pro Asp Glu Leu Ser Pro Lys Val Asp Ser Gly Val Asp Arg His 225 ^" 230 235 240

CAT CTG ATG GCG GCC CTC AGT GCC TAT GCT GCC CAG AGG CCC CCA GCT

His Leu Met Ala Ala Leu Ser Ala Tyr Ala Ala Gin Arg Pro Pro Ala 245 250 255

CCC CCC GGG GAG GGC AGC CTG GAG CCA CAG TAC CTT CTG CGT GCA CCC

Pro Pro Gly Glu Gly Ser Leu Glu Pro Gin Tyr Leu Leu Arg Ala Pro 260 265 270

TCA AGA ATG CCC AGG CCT TTG CTG GCA CCA GCC GCC CCC CAG AAG TGG

Ser Arg Met Pro Arg Pro Leu Leu Ala Pro Ala Ala Pro Gin Lys Trp 275 280 285

CCT TCA CCT CTG GGA GAT TCC GAA GAC CCC TCC AGC ACA GGC GAT GGA

Pro Ser Pro Leu Gly Asp Ser Glu Asp Pro Ser Ser Thr Gly Asp Gly 290 295 300

GCA CGG ATT CAT ACC CTC CTG AAG GAC CTG CAG AGG CAG CCG GCT GAG

Ala Arg He His Thr Leu Leu Lys Asp Leu Gin Arg Gin Pro Ala Glu 305 310 315 320

GTG AGG GGC CTG AGT GGC CTG GAG CTG GAC GGC ATG GCT GAG CTG ATG

Val Arg Gly Leu Ser Gly Leu Glu Leu Asp Gly Met Ala Glu Leu Met 325 330 335

GCT GGC CTG ATG CAA GGC GTG GAC CAT GGA GTA GCT CGA GGC AGC CCT

Ala Gly Leu Met Gin Gly Val Asp His Gly Val Ala Arg Gly Ser Pro

340 345 350

GGG AGA GCG GCC CTG GGA GAG TCT GGA GAA CAG GCG GAT GGC CCC AAG

Gly Arg Ala Ala Leu Gly Glu Ser Gly Glu Gin Ala Asp Gly Pro Lys 355 360 365

GCC ACC CTC CGT GGA GAC AGC TTT CCA GAT GAC GGA GTG CAG GAC GAC

Ala Thr Leu Arg Gly Asp Ser Phe Pro Asp Asp Gly Val Gin Asp Asp 370 375 380

GAT GAT AGA CTT TAC CAA GAG GTC CAT CGT CTG AGT GCC ACA CTC GGG

Asp Asp Arg Leu Tyr Gin Glu Val His Arg Leu Ser Ala Thr Leu Gly 385 390 395 400

GGC CTC CTG CAG GAC CAC GGG TCT CGA CTC TTA CCT GGA GCC CTC CCC

Gly Leu Leu Gin Asp His Gly Ser Arg Leu Leu Pro Gly Ala Leu Pro

405 410 415

TTT GCA AGG CCC CTC GAC ATG GAG AGG AAG AAG TCC GAG CAC CCT GAG

Phe Ala Arg Pro Leu Asp Met Glu Arg Lys Lys Ser Glu His Pro Glu 420 425 430

TCT TCC CTG TCT TCA GAA GAG GAG ACT GCC GGA GTG GAG AAC GTC AAG

b ( >

Ser Ser Leu Ser Ser Glu Glu Glu Thr Ala Gly Val Glu Asn Val Lys 435 440 445

AGC CAG ACG TAT TCC AAA GAT CTG CTG GGG CAG CAG CCG CAT TCG GAG

Ser Gin Thr Tyr Ser Lys Asp Leu Leu Gly Gin Gin Pro His Ser Glu 450 455 460

CCC GGG GCC GCT GCG TTT GGG GAG CTC CAA AAC CAG ATG CCT GGG CCC

Pro Gly Ala Ala Ala Phe Gly Glu Leu Gin Asn Gin Met Pro Gly Pro 465 470 475 480

TCG AAG GAG GAG CAG AGC CTT CCA GCG GGT GCT CAG GAG GCC CTC AGC

Ser Lys Glu Glu Gin Ser Leu Pro Ala Gly Ala Gin Glu Ala Leu Ser 485 490 495

GAC GGC CTG CAA TTG GAG GTC CAG CCT TCC GAG GAA GAG GCG CGG GGC

Asp Gly Leu Gin Leu Glu Val Gin Pro Ser Glu Glu Glu ALa Arg Gly 500 505 510

TAC ATC GTG ACA GAC AGA GAC CCC CTG CGC CCC GAG GAA GGA AGG CGG

Tyr He Val Thr Asp Arg Asp Pro Leu Arg Pro Glu Glu Gly Arg Arg 515 520 525

CTG GTG GAG GAC GTC GCC CGC CTC CTG CAG GTG CCC AGC AGC GCG TTC

Leu Val Glu Asp Val Ala Arg Leu Leu Gin Val Pro Ser Ser Ala Phe 530 535 540

GCT GAC GTG GAG GTT CTC GGA CCA GCA GTG ACC TTC AAA GTG AGC GCC

Ala Asp Val Glu Val Leu Gly Pro Ala Val Thr Phe Lys Val Ser Ala 545 550 555 560

AAT GTC CAA AAC GTG ACC ACT GAG GAT GTG GAG AAG GCC ACA GTT GAC

Asn Val Gin Asn Val Thr Thr Glu Asp Val Glu Lys Ala Thr Val Asp 565 570 575

AAC AAA GAC AAA CTG GAG GAA ACC TCT GGA CTG AAA ATT CTT CAA ACC

Asn Lys Asp Lys Leu Glu Glu Thr Ser Gly Leu Lys He Leu Gin Thr 580 585 590

GGA GTC GGG TCG AAA AGC AAA CTC AAG TTC CTG CCT CCT CAG GCG GAG

Gly Val Gly Ser Lys Ser Lys Leu Lys Phe Leu Pro Pro Gin Ala Glu 595 600 605

CAA GAA GAC TCC ACC AAG TTC ATC GCG CTC ACC CTG GTC TCC CTC GCC

- e - -

Gin Glu Asp Ser Thr Lys Phe He Ala Leu Thr Leu Val Sei Leu_Al 610 615 620

TGC ATC CTG GGC GTC CTC CTG GCC TCT GGC CTC ATC TAC TGC CTC CGC

Cys Ile _^ Lejj Gly Val Leu Leu Ala 5g Gly Leu_ ^" JI Tyr__Cyχ_Leu Arg 625 630 ^" 635 640

CAT AGC TCT CAG CAC AGG CTG AAG GAG AAG CTC TCG GGA CTA GGG GGC

His Ser Ser Gin His Arg Leu Lys Glu Lys Leu Ser Gly Leu Gly Gly 645 650 655

GAC CCA GGT GCA GAT GCC ACT GCC GCC TAC CAG GAG CTG TGC CGC CAG

Asp Pro Gly Ala Asp Ala Thr Ala Ala Tyr Gin Glu Leu Cys Arg Gin 660 665 670

CGT ATG GCC ACG CGG CCA CCA GAC CGA CCT GAG GGC CCG CAC ACG TCA

Arg Met Ala Thr Arg Pro Pro Asp Arg Pro Glu Gly Pro His Thr Ser 675 ^" 680 685

CGC ATC AGC AGC GTC TCA TCC CAG TTC AGC GAC GGG CCG ATC CCC AGC

Arg He Ser Ser Val Ser Ser Gin Phe Ser Asp Gly Pro He Pro Ser 690 695 700

CCC TCC GCA CGC AGC AGC GCC TCA TCC TGG TCC GAG GAG CCT GTG CAG

Pro Ser Al Arg Ser Ser Ala Ser Ser Trp Ser Glu Glu Pro Val Gin 705 710 715 720

TCC AAC ATG GAC ATC TCC ACC GGC CAC ATG ATC CTG TCC TAC ATG GAG

Ser Asn Met Asp He Ser Thr Gly His Met He Leu Ser Tyr Met Glu 725 730 735

GAC CAC CTG AAG AAC AAG AAC CGG CTG GAG AAG GAG TGG GAA GCG CTG

Asp His Leu Lys Asn Lys Asn Arg Leu Glu Lys Glu Trp Glu Ala Leu 740 745 750

TGC GCC TAC CAG GCG GAG CCC AAC AGC TCG TTC GTG GCC CAG AGG GAG

Cys Ala Tyr Gin Ala Glu Pro Asn Ser Ser Phe Val Ala Gin Arg Glu 755 760 765

GAG AAC GTG CCC AAG AAC CGC TCC CTG GCC GTG CTG ACC TAT GAC CAC

Glu Asn Val Pro Lys Asn Arg Ser Leu Ala Val Leu Thr Tyr Asp His 770 775 780

TCC CGG GTC CTG CTG AAG GCG GAG AAC AGC CAC AGC CAC TCA GAC TAC

Ser Arg Val Leu Leu Lys Ala Glu Asn Ser His Ser His Ser Asp Tyr

- G2 -

785 790 795 800

ATC AAC GCT AGC CCC ATC ATG GAT CAC GAC CCG AGG MX CCG GCG TAC

He Asn Ala Ser Pro He Met Asp His Asp Pro Arg Asn Pre Ai a Tyr 805 810 ^~ 81 :

ATC GCC ACC CAG GGA CCG CTG CCC GCC ACC GTG GCT GAC TTT TGG CAG

He Ala Thr Gin Gly Pro Leu Pro Ala Thr Val Ala Asp Phe Trp Glr. 820 825 830

ATG GTG TGG GAG AGC GGC TGC GTG GTG ATC GTC ATG CTG ACA CCC CTC

Met Val Trp Glu Ser Gly Cys Val Val He Val Met Leu Thr Pro Leu 835 840 845

GCG GAG AAC GGC GTC CGG CAG TGC TAC CAC TAC TGG CCG GAT GAA GGC

Ala Glu Asn Gly Val Arg Gin Cys Tyr His Tyr Trp Pro Asp Glu Gly 850 855 860

TCC AAT CTC TAC CAC ATC TAT GAG GTG AAC CTG GTC TCC GAG CAC ATC

Ser Asn Leu Tyr His He Tyr Glu Val Asn Leu Val Ser Glu His He 865 870 875 880

TGG TGT GAG GAC TTC CTG GTG AGG AGC TTC TAT CTG AAG AAC CTG CAG

Trp Cys Glu Asp Phe Leu Val Arg Ser Phe Tyr Leu Lys Asn Leu Gin 885 890 895

ACC AAC GAG ACG CGC ACC GTG ACG CAG TTC CAC TTC CTG AGT TGG TAT

Thr Asn Glu Thr Arg Thr Val Thr Gin Phe His Phe Leu Ser Trp Tyr 900 905 910

GAC CGA GGA GTC CCT TCC TCC TCA AGG TCC CTC CTG GAC TTC CGC AGA

Asp Arg Gly Val Pro Ser Ser Ser Arg Ser Leu Leu Asp Phe Arg Arg 915 920 925

AAA GTA AAC AAG TGC TAC AGG GGC CGT TCT TGT CCA ATA ATT GTT CAT

Lys Val Asn Lys Cys Tyr Arg Gly Arg Ser Cys Pro He He Val His 930 935 940

TGC AGT GAC GGT GCA GGC CGG AGC GGC ACC TAC GTC CTG ATC GAC ATG

Cys Ser Asp Gly Ala Gly Arg Ser Gly Thr Tyr Val Leu He Asp Met 945 950 955 960

GTT CTC AAC AAG ATG GCC AAA GGT GCT AAA GAG ATT GAT ATC GCA GCC

Val Leu Asn Lys Met Ala Lys Gly Ala Lys Glu He Asp He Ala Ala 965 970 975

ACC CTG GAG CAC TTG AGG GAC CAG AGA CCC GGC ATG GTC CAG ACG AAG

Thr Leu Glu His Leu Arg Asp Gin Arg Pro Gly Met Val Gin Thr Lys

980 985 990

GAG CAG TTT GAG TTC GCG CTG ACA GCC GTG GCT GAG GAG GTG AAC GCC

Glu Gin Phe Glu Phe Ala Leu Thr Ala Val Ala Glu Glu Val Asn Ala

995 1000 1005

ATC CTC AAG GCC CTT CCC CAG T GAGCGGCAGC CTCAGGGGCC TCAGGGGAGC

He Leu Lys Ala Leu Pro Gin

1010 1015

CCCTACCCCA CGGATGTTGT CAGGAATCAT GATCTGACTT TAATTGTGTG TCTTCTATTA

TAACTGCATA GTAATAGGGC CCTTAGCTCT CCCGTAGTCA GCGCAGTTTA GCAGTTAAAA

GTGTATTTTT GTTTAATCAA ACAATAATAA AGAGAGGTTT GTCGGAAAAA TCCAAAA

SEQ ID Nos 12 and 13: Nucleotide (SEQ ID No. 12) and amino acid (SEQ ID NO. 13) sequences of the cytoplasmic (intracellular) domain of IAR. SEQ ID NO. 12 corresponds to the nucleotides of SEQ ID No. 10 which encode amino acids 646 to 1015 of SEQ ID No. 11. SEQ ID No. 13 corresponds to amino acids 646 to 1015 of SEQ ID No. 11.

AGG CTG AAG GAG AAG CTC TCG GGA CTA GGG GGC

Arg Leu Lys Glu Lys Leu Ser Gly Leu Gly Gly

650 655

GAC CCA GGT GCA GAT GCC ACT GCC GCC TAC CAG GAG CTG TGC CGC CAG

Asp Pro Gly Ala Asp Ala Thr Ala Ala Tyr Gin Glu Leu Cys Arg Gin 660 665 670

CGT ATG GCC ACG CGG CCA CCA GAC CGA CCT GAG GGC CCG CAC ACG TCA

Arg Met Ala Thr Arg Pro Pro Asp Arg Pro Glu Gly Pro His Thr Ser 675 680 685

CGC ATC AGC AGC GTC TCA TCC CAG TTC AGC GAC GGG CCG ATC CCC AGC

Arg He Ser Ser Val Ser Ser Gin Phe Ser Asp Gly Pro He Pro Ser

690 695 700

CCC TCC GCA CGC AGC AGC GCC TCA TCC TGG TCC GAG GAG CCT GTG CAG

Pro Sei Ada _^ rg Ser Ser Ala Ser Ser Trp Ser Glu Glu Pro Val Gin

705 710 715 720

TCC AAC ATG GAC ATC TCC ACC GGC CAC ATG ATC CTG TCC TAC ATG GAG

Ser Asn Met Asp He Ser Thr Gly His Met He Leu Ser Tyr Met Glu

725 730 735

GAC CAC CTG AAG AAC AAG AAC CGG CTG GAG AAG GAG TGG GAA GCG CTG

Asp His Leu Lys Asn Lys Asn Arg Leu Glu Lys Glu Trp Glu Ala Leu

740 745 750

TGC GCC TAC CAG GCG GAG CCC AAC AGC TCG TTC GTG GCC CAG AGG GAG

Cys Ala Tyr Gin Ala Glu Pro Asn Ser Ser Phe Val Ala Gin Arg Glu

755 760 765

GAG AAC GTG CCC AAG AAC CGC TCC CTG GCC GTG CTG ACC TAT GAC CAC

Glu Asn Val Pro Lys Asn Arg Ser Leu Ala Val Leu Thr Tyr Asp His

770 775 780

TCC CGG GTC CTG CTG AAG GCG GAG AAC AGC CAC AGC CAC TCA GAC TAC

Ser Arg Val Leu Leu Lys Ala Glu Asn Ser His Ser His Ser Asp Tyr

785 790 795 800

ATC AAC GCT AGC CCC ATC ATG GAT CAC GAC CCG AGG AAC CCC GCG TAC

He Asn Ala Ser Pro He Met Asp His Asp Pro Arg Asn Pro Ala Tyr

805 810 815

ATC GCC ACC CAG GGA CCG CTG CCC GCC ACC GTG GCT GAC TTT TGG CAG

He Ala Thr Gin Gly Pro Leu Pro Ala Thr Val Ala Asp Phe Trp Gin

820 825 830

ATG GTG TGG GAG AGC GGC TGC GTG GTG ATC GTC ATG CTG ACA CCC CTC

Met Val Trp Glu Ser Gly Cys Val Val He Val Met Leu Thr Pro Leu

835 840 845

GCG GAG AAC GGC GTC CGG CAG TGC TAC CAC TAC TGG CCG GAT GAA GGC

Ala Glu Asn Gly Val Arg Gin Cys Tyr His Tyr Trp Pro Asp Glu Glv

850 855 860

TCC AAT CTC TAC CAC ATC TAT GAG GTG AAC CTG GTC TCC GAG CAC ATC

Ser Asn Leu Tyj His He Tyr Glu Val Asn Leu Val Ser Glu His He 865 870 875 880

TGG TGT GAG GAC TTC CTG GTG AGG AGC TTC TAT CTG AAG AAC CTG CAG

Trr- CYS GIU Asp Phe Leu Val Arg Sei Phe Tyr Leu Lys Asn Leu Gin

885 890 ^' 895

ACC AAC GAG ACG CGC ACC GTG ACG CAG TTC CAC TTC CTG AGT TGG TAT

Thr Asn Glu Thr Arg Thr Val Thr Gin Phe His Phe Leu Ser Trp Tyr 900 905 910

GAC CGA GGA GTC CCT TCC TCC TCA AGG TCC CTC CTG GAC TTC CGC AGA

Asp Arg Gly Val Pro Ser Ser Ser Arg Ser Leu Leu Asp Phe Arg Arg 915 920 925

AAA GTA AAC AAG TGC TAC AGG GGC CGT TCT TGT CCA ATA ATT GTT CAT

Lys Val Asn Lys Cyc Tyr Arg Gly Arg Ser Cys Pro He He Val His 930 935 940

TGC AGT GAC GGT GCA GGC CGG AGC GGC ACC TAC GTC CTG ATC GAC ATG

Cys Ser Asp Gly Ala Gly Arg Ser Gly Thr Tyr Val Leu He Asp Met 945 950 955 960

GTT CTC AAC AAG ATG GCC AAA GGT GCT AAA GAG ATT GAT ATC GCA GCG

Val Leu Asn Lys Met Ala Lys Gly Ala Lys Glu He Asp He Ala Ala

965 970 975

ACC CTG GAG CAC TTG AGG GAC CAG AGA CCC GGC ATG GTC CAG ACG AAG

Thr Leu Glu His Leu Arg Asp Gin Arg Pro Gly Met Val Gin Thr Lys 980 985 990

GAG CAG TTT GAG TTC GCG CTG ACA GCC GTG GCT GAG GAG GTG AAC GCC

Glu Gin Phe Glu Phe Ala Leu Thr Ala Val Ala Glu Glu Val Asn Ala 995 1000 1005

ATC CTC AAG GCC CTT CCC CAG

He Leu Lys Ala Leu Pro Gin 1010 1015