British Journal of Rheumatology 1998;37:357–361
ANTI-JO-1 ANTIBODIES IN POLYMYOSITIS OR DERMATOMYOSITIS:
EVALUATION BY ELISA USING RECOMBINANT FUSION PROTEIN
JO-1 AS ANTIGEN
M. NISHIKAI, K. OHYA,* M. KOSAKA,* K. AKIYA and T. TOJO
Department of Internal Medicine and Clinical Research Institute, Second Tokyo National Hospital, Tokyo and *Medical and
Biological Laboratories, Inc., Ina, Japan
SUMMARY
We evaluated an enzyme-linked immunosorbent assay (ELISA) for detecting anti-Jo-1 antibodies in patients with polymyositis
(PM ) or dermatomyositis (DM ) by use of the recombinant fusion protein Jo-1. Sera from 64 patients with PM or DM, from
80 patients with other connective tissue diseases, and from 64 healthy subjects matched for age, sex and race, were studied by
the ELISA and by the double immunodiffusion (DID) method. Eight patients with myositis (six PM, one DM and one DM
with malignancy) with positive anti-Jo-1 by DID also showed positive results by the ELISA method, whereas five patients with
positive anti-Jo-1 by this ELISA showed negative results on DID. One of the five had non-specific results. The incidence of
positive results for anti-Jo-1 with the ELISA (18.8%) was greater than that for DID (12.5%), but the difference was not
statistically significant. All patients with positive results for anti-Jo-1 by DID were also positive by the ELISA. The ELISA
system with the recombinant Jo-1 antigen was useful in the detection of anti-Jo-1 antibodies in patients with PM/DM.
K : Jo-1 antibodies, Polymyositis, Dermatomyositis, Recombinant protein antigen, ELISA, Double immunodiffusion,
Jo-1 antibody syndrome, Anti-synthetase syndrome.
V autoantibodies to nuclear and cytoplasmic
antigens are characteristic of patients with polymyositis
(PM ) and dermatomyositis (DM ). Among the antigens to which these autoantibodies are targeted are
aminoacyl-tRNA synthetases [histidyl-tRNA synthetase (Jo-1), threonyl-tRNA synthetase (PL-7), alanyltRNA synthetase (PL-12), isoleucyl-tRNA synthetase
(OJ ) and glycyl-tRNA synthetase ( EJ )], signal recognition particle (SRP), Mi-2, PM-Scl and Ku [1]. The
anti-Jo-1 antibody [2] is most common, and is clinically
the most important, as a marker antibody for PM,
especially in those patients with interstitial lung
disease [3].
Anti-Jo-1 antibodies had been estimated variously
by double immunodiffusion (DID) [2, 3], counterimmunoelectrophoresis [4, 5], Western blotting [6 ],
enzyme-linked immunosorbent assay ( ELISA) using
the column-purified Jo-1 antigen [6, 7], or the enzyme
inhibitory activity method [6 ].
Recently, an ELISA with recombinant fusion proteins has been substituted for DID in the detection of
various autoantibodies, such as antibodies to Sm [8],
nRNP [9], Ro(SSA) [10], La(SSB) [11], topoisomerase-1 [12] and centromere (CEMP-B) [13]. AntiMi-2 antibody [14, 15], a DM-specific autoantibody
[14], is also useful in an ELISA system using recombinant antigen [16 ].
There are no previous published reports of an ELISA
system with recombinant fusion Jo-1 antigen for the
detection of anti-Jo-1 antibodies; this is the first such
report.
PATIENTS AND METHODS
Patients
Serum samples of 64 Japanese patients with idiopathic inflammatory muscle diseases were studied.
There were 19 males and 45 females, aged 15–85 (mean
50.9) yr. These cases consisted of 21 PM, 19 DM, 11
PM/DM with another connective tissue disease, three
childhood DM and 10 DM with malignancies. Serum
samples were also obtained from 20 patients with
systemic lupus erythematosus (SLE ), 20 patients with
systemic sclerosis, 20 patients with rheumatoid arthritis
(RA) and 20 patients with primary Sjögren’s syndrome. Diagnoses of PM and DM were based on the
criteria of Bohan and Peter [17]. All 64 patients were
diagnosed as ‘definite’ or ‘probable’ PM/DM. Fifteen
additional ‘possible’ PM/DM cases were also tested.
The diagnoses of SLE, systemic sclerosis and RA were
based on ARA criteria [18–20]. Sjögren’s syndrome
was diagnosed by the criteria of the Ministry of Health
and Welfare of Japan [21]. Interstitial lung disease was
diagnosed on the basis of persistent bibasilar infiltrates
on chest roentgenogram together with persistent crepitations on auscultation.
In addition, 64 serum samples of healthy subjects
matched for age, sex and race to the 64 PM/DM
patients were evaluated. These subjects were employees
of our hospital and students of our nurses’ school.
Double immunodiffusion
A kit to detect anti-Jo-1 antibodies in serum ( ENA-4
Test; Medical and Biological Laboratories, Inc.,
Nagoya, Japan) was used for the DID procedure.
Crude pig spleen extracts were the source of the
Jo-1 antigen.
Submitted 7 May 1997; revised version accepted 25 July 1997.
Correspondence to: M. Nishikai, Department of Internal
Medicine, Second Tokyo National Hospital, 2-5-1 Higashigaoka,
Meguro-ku, Tokyo 152, Japan.
Recombinant fusion Jo-1 antigen
The Jo-1 full-length complementary deoxyribonucleic acid (cDNA) was isolated from a cDNA library
© 1998 British Society for Rheumatology
357
358
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 37 NO. 4
(Clonetech Laboratories, Inc., Palo Alto, CA, USA)
obtained from KB cells (human pharyngeal epidermoid
carcinoma cell lines). Two cDNA fragments of 1530
base pairs were generated by polymerase chain reaction
with primer pairs. The primers were designed as
5∞-CAACCACCGCAGGTCGAGA-3∞ and 5∞-GTAGTGCCAGTCCCACTTCCT-3∞. This step was followed by cloning into the plasmid pGEM-T Vector
System (Promega Co., Madison, WI, USA). The Jo-1
cDNA was subcloned into a pGEX-3 vector
(Pharmacia Biotech, Uppsala, Sweden) for transformation of Escherichia coli (strain NM522). Expression of
the Jo-1 protein was induced by 1 m isopropyl-b-thiogalactopyranoside. The E. coli cells were harvested
by centrifugation. The E. coli pellet was resuspended
in 10 m Tris–HCl (pH 8.0), 20% sucrose and 1 m
EDTA (pH 7.6), and was dissolved in 50 m Tris–HCl,
100 m NaCl, 2 m phenylmethylsulphonyl fluoride
(PMSF ), 2 mg/ml pepstatin A and 2 mg/ml leupepsin
(pH 7.5). After disruption of E. coli and its DNA by
a sonicator ( VD201; Tomy, Tokyo, Japan) under the
conditions of ‘output’ at 5 for 30 min was repeated
four times, this solution was layered onto 40% (w/v)
sucrose, 10 m Tris–HCl (pH 7.5), 100 m NaCl,
1 m EDTAΩ2Na and 1 m PMSF. The solution was
centrifuged at 13 000 g at 4°C for 60 min. The pellet
containing inclusion bodies in which the recombinant
Jo-1 protein glutathione-S-transferase (GST )–Jo-1 resides was solubilized by pipetting and sonication in 6 
urea, 10 m Tris–HCl (pH 7.5) and 1 m PMSF.
The protein concentration was measured using the
Bio-Rad Protein Assay Kit (Bio-Rad Laboratories,
Hercules, CA, USA). The purity of the GST–Jo-1 was
examined on 12.5% polyacrylamide gels by sodium
dodecyl sulphate–polyacrylamide gel electrophoresis
(SDS–PAGE ). The Jo-1 antigenic activity of the
GST–Jo-1 was examined by the Western blotting
method, as described previously [22].
ELISA
For the ELISA, 10 mg protein/ml of the GST–Jo-1
antigen dissolved in 10 m Tris–HCl (pH 7.6), 6 
urea and 0.01% b-octyl-glucoside was prepared.
Polystyrene microtitre plates with a flat bottom (Nunc
Maxisoap; Intermed, Co., Roskilde, Denmark) were
coated with 100 ml of the GST–Jo-1 antigen solution
overnight at 4°C. Blocking of the sensitized wells was
carried out with phosphate-buffered saline (PBS), 1%
bovine serum albumin (BSA) (Intergen Co., New
York, USA) and 5% sucrose for 2 h at room temperature (20–25°C ). The plates were dried. Then
100 ml/well of the sera to be tested were diluted 1:200
in PBS (pH 7.4), and 1% BSA was added. This was
incubated for 1 h at room temperature. After four
washings with PBS (pH 7.4) and 0.1% Tween 20,
100 ml/well of horseradish peroxidase-conjugated
mouse anti-human c-chain (Medical and Biological
Laboratories, Co., Nagoya, Japan) diluted 1:5000 were
added and incubated for 1 h at room temperature.
After four washings with the same buffer, 100 ml/well
of 3,3∞,5,5∞-tetramethylbenzidine as substrate were
added and incubated for 30 min at room temperature.
The reaction was stopped by 0.1  phosphate.
Plates were read by a microplate reader MPR-A4i
( Tohsoh, Co., Tokyo, Japan) at 450 nm. Each sample
was estimated in duplicate.
For controls, antibodies to recombinant Sm antigen
(Immunovision, Co., Springdale, AR, USA) and to
GST were estimated using the same ELISA system.
Inhibition test
Inhibition of ELISA activity of the anti-Jo-1-positive
sera was carried out simultaneously by the recombinant
GST–Jo-1 antigen and native column-purified Sm antigen (Immunovision, Springdale, AR, USA) as control.
If the binding of these sera was not inhibited significantly [>50% of optical density (OD)] by the recombinant GST–Jo-1 antigen, it was considered a sign of nonspecific binding or of binding to a contaminant.
Statistical methods
x2 tests with correction were used for 2×2 tables.
RESULTS
Purity and antigenic specificity of the recombinant
GST–Jo-1 antigen
SDS–PAGE analysis of the recombinant GST–Jo-1
fusion protein showed a 76 kDa main band, which is
compatible with the molecular weight (MW ) of a
complex of two proteins: GST (MW=25 kDa) and
Jo-1 protein (MW~50 kDa) [5, 23]. The main band
had ~90% homogeneity, as determined by Coomassie
blue staining on SDS–PAGE (Fig. 1). The main band
and the thinner band located just beneath it, which
was possibly a degradation product of the main band,
had Jo-1 antigenic activity, as confirmed by the immunoblotting procedure (data not shown).
F. 1.—SDS–PAGE of the recombinant Jo-1 antigen. Lane 1,
marker proteins for the molecular weights. Lanes 2a and b, E. coli
lysates (control ) with different protein concentrations. Lanes 3a and
b, purified recombinant fusion protein (GST–Jo-1) with different
protein concentrations.
NISHIKAI ET AL.: JO-1 ANTIBODY DETECTION IN MYOSITIS
Anti-Jo-1 ELISA
Sera from the 64 healthy control subjects produced
a mean ± 3 .. OD for Jo-1 binding of 1.29 ± 3.70
450
absorbance units (AU ). The assay result was considered positive if the OD value at a serum dilution of
450
1:100 was 5.0 AU.
The incidence of anti-Jo-1 antibodies in the ELISA
and DID, and the actual AU values in the ELISA, are
shown in Table I and Fig. 2, respectively. In the ELISA
system, 13 patients showed positive results (5 AU ).
The positive sera of 12 of the 13 patients were specifically absorbed by the GST–Jo-1 antigen. However, the
anti-GST ELISA of all 13 anti-Jo-1-positive sera
yielded negative results. Representative results of the
absorption tests are shown in Fig. 3.
In one PM patient, the positive anti-Jo-1 activity
was absorbed not only by the GST–Jo-1 antigen, but
also by the Sm antigen used as the control. This
patient’s serum contained two precipitin lines in DID,
but their antigenic specificities differed from the Jo-1
and the Sm antibody system. In the other 12 patients
positive for anti-GST–Jo-1, four had negative DID
results: one had DM alone, one had PM–SLE overlap,
one had DM with malignancy and one had Sjögren’s
syndrome with polyarthritis and elevated serum creatine kinase levels but no proximal muscle weakness or
interstitial lung disease. The incidence of the anti-Jo-1
by the ELISA system (18.8%) vs DID (12.5%) did not
differ significantly ( x2 = 0.535, P = 0.4646). Seven
(58.3%) of the 12 anti-Jo-1-positive patients had interstitial lung disease. On the other hand, the anti-Jo-1
antibodies were lacking in three (37%) of eight PM
patients with interstitial lung disease. All patients who
were positive for anti-Jo-1 by DID also had positive
results in the ELISA system. No patients tested positive
for anti-Jo-1 in DID but not in the ELISA system.
Among the 15 patients with a diagnosis of ‘possible’
PM/DM, one tested positive for anti-Jo-1 in the
ELISA, but not in DID. This patient exhibited
359
F. 2.—Anti-Jo-1 antibodies by ELISA using the recombinant Jo-1
antigen in polymyositis (PM ), dermatomyositis (DM ), overlap of
PM/DM and other connective tissue diseases (OL), childhood DM
(cDM ), systemic lupus erythematosus (SLE ), systemic sclerosis
(SSc), Sjögren’s syndrome (SS ), rheumatoid arthritis (RA) and
healthy controls. *Patient with a false-positive result. Dashed line,
mean value + 3 .. in healthy controls.
interstitial lung disease, arthritis and Raynaud’s
phenomenon.
DISCUSSION
An ELISA that uses recombinant antigens has
replaced DID in tertiary care hospitals in Japan and
Table I
Incidence of antibodies to the recombinant GST–Jo-1
Diagnosis
No. of patients
No. positive
in ELISA (%)
No. positive
in DID* (%)
Idiopathic inflammatory myopathies
Polymyositis (PM )
Dermatomyositis (DM )
Overlap with other
connective tissue diseases
Childhood DM
PM/DM with malignancy
21
19
11
7 (33.3)
2 (10.5)
1 (9.0)
6 (28.6)
1 (5.3)
0 (0.0)
3
10
0 (0.0)
2 (20.0)
0 (0.0)
1 (10.0)
Subtotal
64
12 (18.8)
8 (12.5)
Other connective tissue diseases
Systemic lupus erythematosus
Systemic sclerosis
Sjögren’s syndrome
Rheumatoid arthritis
20
20
20
20
0
0
1
0
Healthy controls
64
0 (0.0)
*Double immunodiffusion.
(0.0)
(0.0)
(5.0)
(0.0)
0
0
0
0
(0.0)
(0.0)
(0.0)
(0.0)
0 (0.0)
360
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 37 NO. 4
F. 3.—Inhibition tests in two representative serum samples positive
for anti-Jo-1. Solid and dashed lines indicate the anti-Jo-1 antibody
activity absorbed by the GST–Jo-1 and column-purified native Sm
antigens, respectively.
synthetase syndrome [27]. ‘Possible’ PM/DM could be
an appropriate diagnosis for this distinct subset of
idiopathic inflammatory myopathies.
Recent advances in serology suggest that a disease
entity should be established by its serological characteristics as well as by its clinical and pathological characteristics. Revised criteria for SLE [18], RA [20] and
Sjögren’s syndrome [28] reflect this concept. For more
than 20 yr, however, physicians have used the diagnostic criteria for PM/DM [17] which lack serological
factors as elements of diagnosis. The classification
criteria for PM/DM as recently reported in Japan [29]
have introduced the Jo-1 antibody as a diagnostic
criterion for PM/DM.
In conclusion, an ELISA system that employed the
recombinant Jo-1 antigen proved to be a useful replacement for DID in the detection of anti-Jo-1 antibodies
in patients with PM/DM.
other countries for the detection of various autoantibodies, such as those to Sm, nRNP, Ro(SSA),
La(SSB) and topoisomerase-1. However, an ELISA
system for the anti-Jo-1 antibody has not been available until now. This report is the first to demonstrate
the feasibility of recombinant Jo-1 antigen for the antiJo-1 ELISA.
Previous studies [1–3] have shown that the Jo-1
antibody is highly specific for myositis, being much
more common in PM than in DM, especially in those
patients with interstitial lung disease [3]; it is rare in
childhood DM [24]. The present report confirmed
these findings.
Compared with DID, ELISA is generally superior,
with a greater sensitivity. Since Jo-1 antibodies are
found only in a small portion of the general population
with myositis [1–3], and since serum anti-Jo-1 antibody
levels can fluctuate [3, 25], detection of low titres of
Jo-1 antibodies may be useful. Such cases may be false
negative by DID and positive by ELISA, as reported
by Targoff and Reichlin [6 ] in one patient. We found
five additional patients with the ELISA using the
recombinant antigen. The weakly positive activities for
anti-Jo-1 in ELISA were genuine in most cases, as
evidenced by the absorption tests. One serum sample,
however, demonstrated a non-specific or false-positive
result. The presence of a non-specific reaction in the
ELISA system is a problem that needs to be resolved,
whatever the specificity of the antigen. In addition,
since our GST–Jo-1 antigen contains ~10% contaminants (components of E. coli ), the absorption tests used
in the present study were themselves incomplete. In
theory, a completely pure GST–Jo-1 antigen should be
used in the absorption tests. However, impurity of the
antigen does not seem to present a practical problem,
since anti-Jo-1 has thus far been found solely in
patients affected by myositis.
Anti-Jo-1 antibody was detected in the serum of a
patient with possible PM and interstitial lung disease,
arthritis and Raynaud’s phenomenon. This patient had
the so-called Jo-1 antibody syndrome [26 ] or anti-
1. Targoff IN, Reichlin M. Humoral immunity in polymyositis and dermatomyositis. Mt Sinai J Med 1988;55:
487–3.
2. Nishikai M, Reichlin M. Heterogeneity of precipitating
antibodies in polymyositis and dermatomyositis:
Characterization of the Jo-1 antibody system. Arthritis
Rheum 1980;23:881–8.
3. Yoshida S, Akizuki M, Mimori T, Yamagata H, Inada S,
Homma M. The precipitating antibody to an acidic
nuclear protein antigen, the Jo-1, in connective tissue
diseases: A marker for a subset of polymyositis with
interstitial pulmonary fibrosis. Arthritis Rheum 1983;
26:604–11.
4. Arnett FC, Hirsch TJ, Bias WB, Nishikai M, Reichlin M.
The Jo-1 antibody system in myositis: Relationships to
clinical features and HLA. J Rheumatol 1981;8:925–30.
5. Walker EJ, Tymms KE, Webb J, Jaffrey PD. Improved
detection of anti-Jo-1 antibody, a marker for myositis,
using purified histidyl-tRNA synthetase. J Immunol
Methods 1987;96:149–56.
6. Targoff IN, Reichlin M. Measurement of antibody to
Jo-1 by ELISA and comparison to enzyme inhibitory
activity. J Immunol 1987;138:2874–82.
7. Biswas T, Miller FW, Takagaki Y, Plotz PH. An enzymelinked immunosorbent assay for the detection and quantitation of anti-Jo-1 antibody in human sera. J Immunol
Methods 1987;98:243–8.
8. Hines JJ, Danho W, Elkon KB. Detection and quantification of human anti-Sm antibodies using synthetic
peptide and recombinant SmB antigens. Arthritis Rheum
1991;34:572–9.
9. Netter HJ, Guldner HH, Szostecki C, Lakomek H-J,
Will H. A recombinant autoantigen derived from the
human ( U1) small nuclear RNP-specific 68-kd protein:
Expression in Escherichia coli and serodiagnostic application. Arthritis Rheum 1988;31:616–22.
10. St. Clair EW, Burch JA Jr, Saitta M. Specificity of
autoantibodies for recombinant 60-kd and 52-kd Ro
autoantigens. Arthritis Rheum 1994;37:1373–9.
11. Whittingham S, Naselli G, McNeilage LJ, Coppel RL,
Sturgess AD. Serological diagnosis of primary Sjögren’s
syndrome by means of human recombinant La (SS-B)
as nuclear antigen. Lancet 1987;2:1–3.
12. D’Apra P, White-Cooper H, Cleveland DW, Rothfield
NF, Earnshaw WC. Use of molecular cloning methods
R
NISHIKAI ET AL.: JO-1 ANTIBODY DETECTION IN MYOSITIS
13.
14.
15.
16.
17.
18.
19.
20.
21.
to map the distribution of epitopes on topoisomerase-1
(Scl-70) recognized by sera of scleroderma patients.
Arthritis Rheum 1990;33:1501–11.
Rothfield NF, Whitaker D, Bordwell B, Weiner E,
Senecal J-L, Earnshaw W. Detection of anticentromere
antibodies using cloned autoantigen CENP-B. Arthritis
Rheum 1987;30:1416–9.
Targoff IN, Reichlin M. The association between Mi-2
antibodies and dermatomyositis. Arthritis Rheum
1985;28:796–803.
Nishikai M, Reichlin M. Purification and characterization of a nuclear non-histone basic protein (Mi-1) which
reacts with anti-immunoglobulin sera and the sera of
patients with dermatomyositis. Mol Immunol
1980;17:1129–41.
Seelig HP, Moosbrugger I, Ehrfeld H, Fink T, Renz M,
Genth E. The major dermatomyositis-specific Mi-2
autoantigen is a presumed helicase involved in transcriptional activation. Arthritis Rheum 1995;38:1389–99.
Bohan A, Peter JB. Polymyositis and dermatomyositis.
N Engl J Med 1975;292:344–7.
Tan EM, Cohen AS, Fries JF et al. The 1982 revised
criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7.
Masi AT, Rodnan GP, Medsger TA Jr et al. Preliminary
criteria for the classification of systemic sclerosis (scleroderma). Arthritis Rheum 1980;23:581–90.
Arnett FC, Edworthy SM, Bloch DA et al. The American
Rheumatism Association 1987 revised criteria for the
classification of rheumatoid arthritis. Arthritis Rheum
1988;31:315–24.
Homma M, Tojo T, Akizuki M, Yamagata H. Criteria
22.
23.
24.
25.
26.
27.
28.
29.
361
for Sjögren’s syndrome in Japan. Scand J Rheumatol
1986;61(suppl.):26–7.
Nishikai M, Okano Y, Yamashita H, Watanabe M.
Characterization of centromere (kinetochore) antigen
reactive with sera of patients with a scleroderma variant
(CREST syndrome). Ann Rheum Dis 1984;43:813–24.
Nishikai M, Aoyagi S, Kanazawa N, Sato A. Indirect
immunofluorescent staining pattern of myositisautoantibody, the anti-Jo-1. Jpn J Rheumatol
1990;2:245–50.
Chmiel JF, Wessel HU, Targoff IR, Pachman LM.
Pulmonary fibrosis and myositis in a child with anti-Jo-1
antibody. J Rheumatol 1995;22:762–4.
Bernstein RM, Morgan SH, Chapman J, Bunn CC,
Mathews MA, Turner-Warwick M, Hughes GRV. AntiJo-1 antibody: a marker for myositis with interstitial
lung disease. Br Med J 1984;289:151–2.
Ray DW, Bridger J, Hawnaur J, Waldek S, Bernstein
RM, Dornan TL. Transverse myelitis as the presentation
of Jo-1 antibody syndrome (myositis and fibrosing alveolitis) in long-standing ulcerative colitis. Br J Rheumatol
1993;32:1105–8.
Marguerie C, Bunn CC, Beynon HLC et al. Polymyositis,
pulmonary fibrosis and autoantibodies to aminoacyltRNA synthetase enzymes. Q J Med 1990;282:1019–38.
Vitali C, Bombardieri S, Moutsopoulos HM et al.
Assessment of the European classification criteria for
Sjögren’s syndrome in a series of clinically defined cases:
results of a prospective multicentre study. Ann Rheum
Dis 1996;55:116–21.
Tanimoto K, Nakano K, Kano S et al. Classification
criteria for polymyositis and dermatomyositis.
J Rheumatol 1995;22:668–74.