Discussion of Background It has been expected that clinical rheumatoid arthritis can be diagnosed by assaying an antibody to a type II collagen. However, a conventional method of assaying the antibody to a type II collagen is not suitable for adequate diagnosis of clinical rheumatoid arthritis for use in practice due to the low sensitivity of the assay AND high non-specific reactions. In other words, it is impossible to perform precise diagnosis of the probability of the onset of autoimmune arthritis, autoimmune clinical rheumatoid arthritis, or the severity of autoimmune clinical rheumatoid arthritis by the conventional method.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an immunological assay method free from the conventional method's shortcomings, which makes it possible to perform the above-mentioned diagnosis with high assay sensitivity AND low, non-specific reaction, namely the diagnosis of the probability of the onset of autoimmune arthritis, the diagnosis of autoimmune clinical rheumatoid arthritis, and also the diagnosis of the severity of autoimmune clinical rheumatoid arthritis.

Another object of the present invention is to provide an agent for conducting the above- mentioned assay method.

BRIEF DESCRIPTION OF THE DRAWINGS A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein: Fig. 1 is a schematic illustration of examples of ELISA plates for assaying antibodies to chick (CII), bovine (BII), and porcine (PII) and human type II collagen (HII) in human sera.

Fig. 2 is a graph showing a typical standard curve obtained by the ELISA of the present invention.

Fig. 3A and Fig. 3B are graphs showing the assay results with respect to patients with RA, and negative control sera, obtained by a conventional ELISA using BSA-Tween as a blocking agent and an improved ELISA using heterologous animal serum as a blocking agent, which is used as a blocking agent in the present invention.

Fig. 4 is a graph showing standard curves for the assay of human type II collagen antibodies obtained by conventional ELISA using Peroxidase-conjugated rabbit anti-human IgG antibodies as a secondary antibody, further improved ELISA using an Avidin-Biotin system (A-B), and the ELISA of the present invention.

Fig. 5 shows graphs indicating a correlation between serum IgG autoantibody levels to type II collagen and arthritis markers in patients with RA (SJ: the number of swollen joints, ESR: erythrocyte sedimentation rate, CRP: concentration of C-reactive protein, RF: rheumatoid factor).

Fig. 6 shows graphs indicating a correlation between serum IgG autoantibody level to type II collagen in patients with RA and serum immunoglobulin levels thereof.

Fig. 7 shows graphs indicating the changes in the antibody levels in sera from patients with relapsing polychondritis who were treated with prednisolone.

Fig. 8 shows the results of tests analyzing the antibody reactivity to CB-peptide fragments of human type II collagen in individual patients with RA.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The inventor of the present invention has discovered a method of assaying type II collagen with high assay sensitivity in the course of this study of removing various shortcomings of the conventional method.

More specifically, the present invention provides a method of assaying cross-reactivity between an antibody to a heterologous type II collagen and an antibody to an autologous type II collagen, which can be performed by ELISA.

Furthermore, the present invention provides an ELISA in which such a type II collagen is assayed under conditions where the presence of bacteria is suppressed, based on the discovery that non-specific reactions are induced in the course of the assay of the type II collagen.

The ELISA which is performed under conditions where the presence of bacteria is suppressed makes it possible to assay with high assay sensitivity the antibody to the type II collagen, so that the diagnosis of the probability of the onset of autoimmune arthritis, the diagnosis of autoimmune clinical rheumatoid arthritis, and also the diagnosis of the severity of autoimmune clinical rheumatoid arthritis, which cannot be performed by the conventional method, can be performed by the method of the present invention.

In the present invention, (1) assaying cross-reactivity between an antibody to a heterologous type II collagen and an antibody to an autologous type II collagen, and (2) determining the ratio of the reactivity of an antibody to a heterologous type II collagen to the reactivity of an antibody to an autologous type II collagen for diagnosing early or autoimmune clinical rheumatoid arthritis, are carried out by an immunological method. Varieties of immunological methods can be employed in the present invention. However, ELISA is preferable for use in the present invention because non-specific reactions can be easily controlled.

As the type II collagen for use in the present invention, for instance, human, chick, bovine, and porcine type II collagens, can be employed. These collagens are easily available.

The reaction is carried out in a buffer, preferably in a phosphate buffer. It is preferable to add an antibacterial agent such as sodium azide to such a buffer to prevent the buffer from being contaminated with bacteria or to prevent the growth of bacteria in the buffer, in an amount sufficient for exhibiting a sufficient antibacterial effect, but which does not have adverse effects on immunological reactions and assays, usually in a range of about 0.005% to 0.5%.

In order to bind the type II collagen to a well, a mixed solution is incubated in the well at a temperature in a range of 0 to 35 C, preferably at about 4 C, for 2 to 24 hours.

Furthermore, the surface of the type II collagen bound well is then treated with animal serum. It is preferable that the animal serum used here be serum from the same animal as that derived from a second antibody used for the detection of human immunoglobulin.

In performing the assay, a sample such as serum is diluted with buffered animal serum containing an antibacterial agent such as azide and placed in the wells of an ELISA plate, and incubated at a temperature in a range of 0 to 35 C, preferably at about 4 C, for 2 to 24 hours.

After incubation, any unreacted antibody is removed by washing the well.

A conjugated antibody (the second antibody) is then added to the well and incubated.

After incubation, any unreacted conjugated antibody is removed by washing the well. This incubation is carried out at about room temperature for about 30 minutes to 12 hours. Examples of conjugated antibodies for use in the assay include enzyme-, biotin-, and hapten-conjugated antibodies. Examples of antibodies are animal derived antibodies such as anti-human IgG antibody, anti-human IgA antibody, and anti-human IgM antibody. Such antibodies may be antibody fragments such as Fab'and F (ab') 2. It is preferable that the antibodies be antibodies specifically bound to an Fc portion of human immunoglobulin.

When an enzyme is used as a labelling material in the conjugated antibody, the activity of the enzyme can be measured by adding a substrate to the enzyme. Examples of enzymes are peroxydase (POD), alkali phosphatase, and fÀ-galactosidase. Examples of substrates are varieties of coloring substrates, fluorescent substrates, and luminescent substrates. When biotin or hapten is used as the labeling material, the assay can be performed using enzyme-conjugated avidin, enzyme-conjugated streptoavidin, and enzyme-conjugated anti-hapten antibody. As the enzyme, the above-mentioned enzymes can be used, and as the substrate, the above-mentioned corresponding substrate can be used.

The result of the measured cross-reactivity between the antibody to the heterologous type II collagen and an antibody to the autologous type II collagen be used for diagnosing the probability of the onset of autoimmune arthritis in comparison with the type of a major histocompability complex (MHC).

Furthermore, the result of the measured ratio of the reactivity of the antibody to a heterologous type II collagen to the reactivity of the antibody to the autologous type II collagen can be used for diagnosing early or autoimmune clinical rheumatoid arthritis in comparison with the type of a major histocompability complex (MHC). The ratio of the reactivity can be calculated by dividing the antibody level for the autologous type II collagen by the antibody level for the heterologous type II collagen.

Other features of this invention will become apparent in the course of the following description of exemplary embodiments, which are given for illustration of the invention and are not intended to be limiting thereof.

Example 1 [Preparation of type II collagen coated ELISA plates] ELISA plates were coated overnight at 4 C with chick (CII) or human (HII) type II collagen (5 g/ml) dissolved in 0.15M phosphate buffer containing 0.05% of sodium azide, with the respective solution placed in the wells thereof.

The wells of the plates were washed 3 times with wash buffer. With the placement of 100 1 of full strength buffered NGS in the wells, the plates were incubated at room temperature for 1 hour and blocked. The wells were then washed 3 times with wash buffer, whereby ELISA plates (Azide+) were prepared.

The above-mentioned procedure was repeated except that the 0.15M phosphate buffer containing 0.05% of sodium azide employed in the above was replaced by a 0.15M phosphate buffer free of sodium azide, whereby ELISA plates (Azide-) were prepared.

ELISA plates for reference blank wells were prepared by repeating the above- mentioned procedure except that neither the chick (CII) nor human (HII) type II collagen (5 g/ml) was dissolved in the 0.15M phosphate buffer containing the 0.05% of sodium azide therein, and dissolved in the 0.15M phosphate buffer without containing the 0.05% of sodium azide therein.

Example 2 [Assay of anti-collagen antibodies in patients with rheumatoid arthritis (RA)] Anti-CII collagen antibodies or anti-HII collagen antibodies in seven samples of sera from patients with RA were assayed, using the ELISA plates (Azide+) and ELISA plates (Azide-) prepared in Example 1.

Each serum sample was diluted 100 fold and 100 1 of each diluted serum sample was placed in each well of the ELISA plates and incubated at 4 C overnight. Each well of the ELISA plates was washed with wash buffer.

Biotin-conjugated goat anti-human IgG (5 1, Fc specific) antibody was dissolved in 10 ml of 25% NGS, and added to each of the wells (100 I/well).

The ELISA plates were incubated at room temperature for 2 hours, and were then

washed 6 times with wash buffer.

2.5 1 of Strept-avidin-Peroxidase was dissolved in 10 ml of 1% ovalbumin (OVA)- 0.05% Tween solution, pH 7.5, and 100 fÊl ofthe solution was added to each of the wells.

The ELISA plates were incubated at room temperature for 1 hour and washed with wash buffer.

Immediately after washing the ELISA plates, one Urea-H202 plus buffer tablet and one o-phenylenediamine (OPD) tablet (Sigma, St. Louis, MO) were dissolved in 20 ml of glass- distilled water, and 100 1 of the solution was added to each of the wells. The ELISA plates were incubated at room temperature for 30 minutes so as to develop a color in each of the wells.

The incubation was stopped by adding 50 1 of 2. 5N sulfuric acid to each well.

OD values were determined at 490 nm (a 650 nm filter was used as a reference). The results are shown in TABLE 1.

TABLE 1 Samples Azide-Azide+ Buffer-CII-coated HII-coated Buffer-CII-coated HII-coated coated coated Blank 0. 053 0. 021 0. 051 0. 012 0. 014 0.012 JP77 0. 055 0. 113 0. 107 0. 042 0. 090 0.073 JP78 0. 068 0. 135 0. 089 0. 134 0. 086 0.061 JP79 0. 321 0. 305 0. 198 0. 061 0. 173 0.073 JP80 0. 452 0. 192 0. 242 0. 098 0. 104 0.080 JP81 0. 181 1. 131 0. 270 0. 063 0. 541 0.095 JP82 0. 140 0. 293 0. 248 0. 045 0. 144 0.087 JP83 0. 446 0. 207 0. 634 0. 100 0. 128 0.178 Example 3 [Assay of anti-type II collagen antibodies] Assays of anti-type II collagen antibodies were performed, using an Avidin-Biotin method. ELISA plates used in the assays are shown in Fig. 1.

The ELISA plates were coated overnight at 4 C with chick (CII), bovine (BII), porcine (PII) or human (HII) type II collagen (5 g/ml) dissolved in 0.15M phosphate buffer containing 0.05% of sodium azide. To determine the background level of samples, well for this purpose were not coated with collagen but were treated with buffer only containing sodium azide, with the placement of the respective solution in the wells of each ELISA plate.

The wells of the plates were washed 3 times with wash buffer. After the addition of 100 1 of full strength buffered NGS to the wells, the plates were incubated at room temperature for 1 hour and blocked. The wells were then washed 3 times with wash buffer.

Reference standards were prepared from the serum of a single monkey which was immunized with CII/CFA. The serum was diluted 1: 100,000 with buffered NGS and defined as 16 units/ml and kept at-20 C as a stock solution. This stock solution was diluted serially with buffered NGS to make 8,4,2,1,0.5 and 0.25 units/ml just before use.

All serum samples were diluted 1: 100 with buffered NGS containing 0.05% Azide, for example, by diluting 20 1 of each sample with 1.98 ml of the buffered NGS. This solution was kept as a stock solution for future assays. The samples were further diluted with buffered NGS 1: 200-1: 10,000 when necessary. Buffered NGS (100 1) was placed in all blank (B) wells.

The diluted samples (100 1) were placed in the above-mentioned type II collagen- coated wells and also in uncoated wells. The reference standards (0.25-16 units/ml, 100 1) were added to the standard strips as shown in Fig. 1.

The plates were incubated overnight at 4 C and washed with wash buffer. Further, Biotin-conjugated goat anti-human IgG (5 l, Fc specific) antibody was dissolved in 10 ml of 25% NGS, and placed in all wells (100 I/well).

The plates were incubated at room temperature for 2 hours and washed 6 times with wash buffer. Furthermore, 2.5 1 of Strept-avidin-Peroxidase was dissolved in 10 ml of 1% OVA-0.05% Tween 20 solution, pH 7.5, and added to all wells (100 I/well). The plates were incubated at room temperature for 1 hour and washed with wash buffer.

Immediately after washing the plates, one Urea-H202 plus buffer tablet and one OPD tablet were dissolved in 20 ml of distilled water, and 100 1 of the solution was added to each of the wells. The plates were incubated at room temperature for 30 minutes so as to develop a color in each of the wells.

The reaction was stopped by adding 50 1 of 2. 5N sulfuric acid to each well.

OD values were determined at 490 nm (a 650 nm filter was used as a reference). When the OD values were more than the OD value of the highest reference standard (16 units/ml), all the samples were re-assayed at a higher dilution (1: 200-1: 10,000).

(Calculation of Antibody Titers) Step 1

The duplicate OD values for test samples and reference standards were averaged.

Step 2 The average values of blank values (B) were subtracted from the averaged OD values in the same column obtained in the above Step 1.

Step 3 With respect to samples calculated in the above Step 2, the OD values of the samples tested in uncoated wells (background values, BG) were subtracted from their counterpart OD values of the collagen-coated wells (test values, T).

Step 4 The OD values of reference standards calculated in the above Step 1 were plotted as the ordinate against the units/ml of antibody standard as the abscissa. Fig. 2 shows the results of a representative experiment where the standard range was from 0.25 to 16 units/ml.

Example 4 [Assay of anti-type II collagen antibodies by an improved ELISA system using NGS as a blocking agent] In accordance with the same procedure as in Example 1, collagen uncoated, human type I collagen (HI) coated, and human type II collagen (HII) coated ELISA plates were prepared and were then blocked with NGS.

Using these ELISA plates and HRP-conjugated goat anti-human IgG antibody (Fc Specific), 30 anti-HII collagen antibody positive, but anti-HI antibody negative, sera from RA patients were diluted with buffered NGS containing Azide and assayed (so-called"Improved ELISA").

The results of the assays are shown in Figs. 3 (a) and 3 (b).

The above-mentioned procedure was repeated except that the NGS employed in the above was replaced by a mixed liquid of BSA and Tween 20 for blocking the ELISA plates and for diluting serum specimens.

The results are shown in Fig. 3 (a), which indicates that it was impossible to perform the assays, using the conventional ELISA system due to the high non-specific reactions of the collagen antibodies in the human sera.

Example 5 [Comparison of the assay sensitivity between the ELISA of the present invention and conventional ELISA] In accordance with the procedures in Examples 3 and 4, anti-human type II collagen antibodies in cynomolgus monkeys were assayed by the ELISA of the present invention and also by a conventional ELISA, whereby the standard curve of the ELISA of the present invention and the standard curve of a conventional ELISA as shown in Fig. 4 were obtained.

Furthermore, the above-mentioned procedure was repeated except that the HRP- conjugated goat, anti-human IgG antibody (Fc Specific) employed in Example 4 was replaced by Biotin-conjugated goat anti-human IgG antibody and Strept-avidin-Peroxidase, whereby the standard curve of a further improved ELISA (A-B) as shown in Fig. 4 was obtained.

The ELISA of the present invention exhibits an assay sensitivity as high as 100 times the assay sensitivity of the improved ELISA, and an assay sensitivity as high 10 times the assay sensitivity of the further improved ELISA (A-B).

Example 6 [Prevalence of anti-collagen antibodies in patients with RA] The anti-collagen antibodies in sera from patients with RA were assayed by the ELISA of the present invention and the prevalence of the anti-collagen antibodies determined by the ELISA of the present invention and that determined by the prior ELISA (A-B) were compared.

In the further improved ELISA (A-B), ELISA plates were coated overnight at 4 C with chick (CII), bovine (BII), porcine (PII) or human (HII) type II collagen (5 g/ml) dissolved in 0.15M phosphate buffer, with the placement of each of the respective solutions in the wells of the ELISA plates, and the coated ELISA plates were blocked, using a buffered NGS at room temperature for 1 hour.

The results of the assays are shown in TABLE 2.

TABLE 2 Further Improved A-B ELISA Psesent A-B ELISA (serum dilution: 1/50) (serum dilution: 1/100) Positive/Total % Positive/Total % IgA CII 86/259 33. 2 32/53 60. 3 BII 96/259 37. 1 29/53 54. 7 PII nt. 22/53 41. 6 HII 34/259 23. 6 24/53 453 IgG CII 78/259 30. 1 49/53 92. 5 BII 87/259 33. 6 48/53 90. 1 PII nt. 41/53 77. 4 13.144/5383.0HII34/259

In the ELISA ot the present invention, serum samples were diluted 1: 1 ut, whereas the conventional ELISA (A-B), serum samples were diluted 1: 50.

The prevalence of the autoantibodies to IgA and IgG anti-type II collagens were respectively determined to be only 23.6% and 13.1% of patients with RA by the further improved ELISA (A-B). In contrast to this, by the ELISA of the present invention, with respect to IgG antibodies (10 units/ml or more), the prevalence was determined to be as high as 83% of Japanese patients with RA and 90% of American patients with RA. Similarly, with respect to IgG antibodies (26 units/ml or more), the prevalence was determined to be as high as 77% of Japanese patients with RA and 63% of American patients with RA.

Similarly, the prevalences of IgA and IgG antibodies to heterologous type II collagen such as CII and BII determined by the ELISA of the present invention were 2-3 times higher than those determined by the further improved ELISA (A-B).

Especially, IgG antibodies to CII and BII were detected in 90% or more of patient sera by the ELISA of the present invention. This indicates that antibody production to dietary collagen is a very common biological phenomenon in humans. Antibodies to heterologous type II collagen cross-reacted with autologous human type II collagen to some degree in most patients with RA.

Example 7 [Correlation between HLA-DR types and autoantibody levels to type II collagen] To confirm the mechanism of autoantibody production in patients with RA with respect to MHC-type, the specificity of each of IgA and IgG anti-type II collagen antibodies was investigated, using sera samples from 53 Japanese and 60 American RA patients whose HLA- DR types were known, using the ELISA of the present invention.

The crossreaction ratio of the heterologous type II collagen antibodies with an autologous II collagen (HII) was determined by the steps of obtaining a unit value of each collagen antibody, and dividing the antibody level of the autologous II collagen (HII) by a highest antibody level of the heterologous type II collagen antibodies. TABLE 3 shows the results.

TABLE 3 Group HLA type HLA IgG Antibodies to % Crossreact to group (Units/ml (mean} SE)) HII CII BII PII HII GI 0405-1 1 982535 143+73 161100 9775 0. 210.15 2106#56431#19833#737#90.14#0.04G20405-2 G30405-32264609624635332285292490. 760.08 2607#241238#87153#6777#240.24#0.09G40101 G5 1001 2 251179 15998 14781 316227 1. 000.59 G6 0901 3 265127 23789 17469 19975 0. 620.09 3197#108106#57120#9630#9035#0.15G7Others

In view of the results shown in TABLE 3, it was confirmed that 1) IgA and IgG antibodies in patient sera dominantly react to either CII or BII, and partly crossreact with HII, 2) there is an inverse correlation between IgA and IgG antibody levels, and 3) the IgG antibody levels to HII differ significantly among patient groups with different DR-types, regardless of antibody specificity and levels to heterologous type II collagen. This indicates that DR-types play a regulatory role in the antibody production. For example, IgG anti-HII antibody levels in patients with DR-0405/0405 (G1: HLA +/+) and DR0405/1405, and those in patients with 0405/1501 (G2: HLA +/-) were respectively as low as 98+75 and 37+9 units/ml, since the cross-reactivity of anti-BII or CII antibodies to HII was less than 20% ("+"denotes alleles that are believed to be RA-susceptible,"-"denotes alleles that are believed not to be RA- susceptible). In contrast, patients with the DR-0405 type combined with a type other than 0101 or 1501 (G3: HLA +/-), IgG anti-HII antibody levels were 52924 units/ml (G1 and G2 vs. G3: P<0.05), since the cross-reactivity of anti-CII and BII antibodies to HII was significantly high (76%). Thus, the data indicates that the combination of Allele 1 and Allele 2 rather than a single allele has an effect on the immune-response to type II collagen and epitope specificity of resulting antibodies. The patient group with higher IgG anti-HII antibody titers has more severe arthritis as shown in TABLE 4.

TABLE 4 Group SJ3)RF4)IA5)IgM6)IgG7)A/GESR2) ratio8 Gl4. 52. 165185. 01. 748163062886201288137 0.250.0 25#74.2#1.5104#33345#62108#91240#540.28#0.0G21.4#0.4 68#128.8#2.0124#34317#33172#182110#1830.16#0.0G35.3#1.6 G4 3. 31. 7 4114 9. 01. 5 19190 34251 13738 1337191 0. 270.0 G5 2. 01. 0 467 4. 61. 2 10432 37848 9513 1757159 0. 220.0 40#66.0#2.0107#44380#26128#131470#1520.28#0.0G62.2#0.7 27#102.4#0.956#23279#22111#111496#1490.19#0.0G70.4#0.2

Similar results were observed in American RA patients when the prevalence of IgG anti-HII antibody positive sera (more than 50 units/ml) was compared between individual HLA- type groups. IgG antibody levels in patients with RA-susceptible types such as 0401,0404, 0405 or 10101 were not always high, while the levels in patients with non-RA types were not always low. High IgG anti-HII antibody levels were observed in patients with a particular combination of HLA-DR types, such as 0401/0701 and 0404/03 (HLA +/-), and 1301/0701 in HLA (-/-). The results are shown in TABLE 5.

TABLE 5 Allele 2 4034044057011112131415/16101otherTotal03401 Allele @ G1 01 2/2 2/4 3/3 1/1 0/1 0/1 1/2 0/1 9/15 G2 03-0/4-2/2-0/1 0/1----0/2-2/10 G3 0401-0/1 0/2 0/2 1/1 3/4 0/1--0/2 2/4--8/17 G4 0403-------0/1-----0/1 G5 0404--------0/1----0/1 G6 -----0/1-0/1----0/2- G7 13-----1/1-------1/1 G8 14----------0/1-0/1 0/2 G9 1001-------1/1----1/1 G10 nt.-----1/1 1/1-0/1---4/7 6/10 2/90/22/41/18/101/41/21/50/33/70/24/925/6Total2/2 0 The above results indicate that the HLA DR-type influences the epitope specificity of type II collagen antibodies. The reactivity of type II collagen antibodies to HII are regulated by epitope specificity to type II collagen. Furthermore, no differences were observed in species- specificity or antibody levels to heterologous CII or BII among individual groups; some patients had high levels of antibodies to CII, while others had high levels to BII even though their DR alleles were identical. This indicates that a combination of 2 DR alleles rather than a single allele is involved in the regulation of autoantibody production to HII. This is supported by the

fact that I-E genes in mice which are the counterpart to human DR genes, regulate immune- responses to type II collagen in mice.

Furthermore, humanized HLA-DRwl2 transgenic mice are resistant to CIA, since DRwl2, an RA-resistant type, down-regulates the immune-responses to type II collagen.

Furthermore, in order to confirm the association of anti-type II collagen autoantibodies in RA, IgG anti-HII antibody levels in individual patients were compared to clinical markers in these patients by regression analysis. As shown in Fig. 5 and 6, it was confirmed that a positive correlation exists between the anti-HII antibody levels and CRP, ESR and SJ (the number of swollen joints) values, and also between the anti-HII antibody levels and the serum IgM, IgG concentrations, whereas no positive correlation was observed between the anti-HII antibody levels and either serum rheumatoid factor (RF) nor serum IgA levels.

The above results indicate that the presence of autoantibodies to type II collagen correlates with clinical markers of RA in patients with certain HLA-types.

Since various factors other than the anti-type II collagen autoantibodies are involved in triggering, exacerbating or perpetuating arthritis, the combined information of DR-types and serum IgG anti-HII antibody levels may provide important information in order to predict the risk of the onset of RA, distinguish the type of RA, and to predict disease severity. Such information is useful for preventing the onset of arthritis in patients. The results are shown in TABLE 6.

TABLE 6 Antibodies to Collagen Arthritis Heterologous Autologous MHC Probability Gl Low (1) G2 + Low Predisposition G3 + Low G4 + + Low G5 + + Moderate G6 + Hi h Antibodies to Collagen Class of Arthritis Heterologous Autologous MHC Arthritis Severity G1---Non-Autoimmune N/A G2 - + Non-Autoimmune N/A Diagnosis G3 +-Non-Autoimmune N/A G4 + +-Autoimmune Mild G5 Autoirnmune Moderate G6 + + + Autoimmune Severe Example 8 [Therapeutic effect of prednisolone in patients with relapsing polychondritis]

In order to clarify the relationship between the disease activity and the antibody levels to type II collagen, a time course of change in the anti-type II collagen antibody level in the blood were measured in two patients with relapsing polychondritis (RP) undergoing treatment with prednisolone (5 mg/day), a steroid with anti-inflammatory and immunomodulating activity, using the ELISA of the present invention. The results are shown in Fig. 7.

Serum antibody levels to chick (CII) and human type II (HII) collagen in these patients were determined and in both the patients, a significant decrease in antibody levels was observed over the course of treatment. Notably, the decrease of antibody level in patient Ju4 was remarkable and was almost undetectable after 14 months. This patient was released from the hospital due to a complete remission. In contrast to this, Tj7, who had a remaining high level of anti-collagen antibodies died at 10 months. The data implicate the pathological role of anti-type II collagen antibodies in RP and their measurement can be used as a diagnostic marker for immunomodulating therapies.

Example 9 [Correlation of antibody crossreactivity with CIA-susceptibility in cynomolgus monkeys] According to the assay of each type II collagen antibody in serum, using the ELISA of the present invention, as observed in the patients with RA, an association between the percent crossreactivity of anti-heterologous collagen antibodies to autologous type II collagen and the development of arthritis was observed in outbred cynomolgus monkeys immunized with CII or HII. The results are shown in TABLE 7.

TABLE7 HIIMkIIHII/CIIMkII/CIICII/HIIMkII/CIICII/HIIMkII/HIIGroupCII 12.6#1.110.5#2.09.6#3.50.83#0.130.76#0.24//CII-Arthritic @mun. #2.24.4#3.81.1#0.40.45#0.470.11#0.05//Non-Arthritic10.9 HII-Arthritic 3. 4 0. 6 8. 0 3. 0 6. 0 3. 0 0. 44 0. 09 0. 77 0. 09 immun.Non-Arthritic 0. 9 i 0. 2 3. 6 i 3. 3 3. 0 + 3. 1//0. 17 i 0. 11 0. 78 + 0. 10 Although no difference was observed in the immune response to the collagen used for immunization between CIA-susceptible and resistant monkeys, autoantibody levels to monkey type II collagen (MkII) in arthritic monkeys were approximately 10 times higher than those in non-arthritic monkeys.

This difference depends upon the percent crossreactivity of anti-heterologous type II collagen antibodies to autologous type II collagen as similarly observed among the patients with RA of a different HLA DR-type.

It was also demonstrated that the epitope specificity of the antibodies has a significantly close relationship with the onset of autoimmune arthritis.

Example 10 [Autoantibody Epitopes on Type II Collagen in RA] As observed in the CIA models, the immune response and the epitope specificity of autoantibodies to type II collagen greatly depend on MHC types in the patients with RA.

As in the case of the patients with RA, remarkably high levels of autoantibody to type II collagen were frequently observed in sera from patients with relapsing polychondritis (RP).

In order to investigate the fine association between the MHC type and the epitope specificity of autoantibodies to type II collagen in the patients with RA and the patients with RP, the reactivity of anti-type II collagen antibodies against CB-peptides (cyanogen-bromide generated peptides) of human type II collagen was studied. These peptides are designated as CB12, CB11, CB8, CB10, and CB9.7. CB-peptides prepared from HII collagen were renatured into their original triple-helical conformation.

ELISA plates were coated with the renatured CB-peptides (10 g/ml) dissolved in phosphate buffer, pH 7.5 at 4 C overnight.

Antibodies purified from RA sera by affinity chromatography and extracted from RA cartilage were diluted with full-strength buffered NRS and were allowed to react with each CB-peptide at 4 C overnight. Antibodies bound to CB-peptides were detected by peroxidase- conjugated anti-human IgG antibodies diluted 1: 2,000 with 25% NRS. The results of this assay are shown in Fig. 8, which indicate that the antibodies in each RA serum recognized various antigenic determinants in different regions of HII.

In 19 sera tested, 7 samples dominantly reacted with CB11,5 samples with CB10,5 samples with CB8, and one sample reacted with CB12. One sample reacted with multiple CB-peptides although CB9.7 was dominant.

In contrast to this, when autoantibodies purified from sera from three patients with PR were investigated, 3 out of 3 sera reacted dominantly with CB9.7, so that the specificity of the autoantibodies of the patients with PR is obviously different from the specificty of the autoantibodies of the patients with RA. This result indicates that the epitope specificity of autoantibodies is linked to the MHC-types of individual patients.