вход по аккаунту


Presence of autoantibodies to the glycolytic enzyme ╨Ю┬▒-enolase in sera from patients with early rheumatoid arthritis.

код для вставкиСкачать
Vol. 46, No. 5, May 2002, pp 1196–1201
DOI 10.1002/art.10252
© 2002, American College of Rheumatology
Presence of Autoantibodies to the Glycolytic Enzyme
␣-Enolase in Sera From Patients With
Early Rheumatoid Arthritis
Vincent Saulot,1 Olivier Vittecoq,2 Roland Charlionet,1 Patrice Fardellone,3 Catherine Lange,4
Laure Marvin,4 Nadine Machour,1 Xavier Le Loët,2 Danièle Gilbert,1 and François Tron1
Objective. To identify a new autoantigen/
autoantibody population in rheumatoid arthritis (RA)
Methods. Following a population-based recruitment effort, 255 patients with very early arthritis (median disease duration 4 months) were studied using
different clinical, biologic, and radiologic assessments.
After a followup period of 1 year, patients were classified
as having RA (n ⴝ 145), non-RA rheumatic diseases
(n ⴝ 70), and undifferentiated arthritis (n ⴝ 40).
Patients’ sera were analyzed by one-dimensional (1D)
and 2D Western blotting. The recognized 50-kd protein
was analyzed by matrix-assisted laser desorption
ionization–time-of-flight (MALDI-TOF) mass spectrometry (MS). RA serum reactivities were evaluated
against the recombinant protein synthesized by an in
vitro coupled transcription–translation system.
Results. On 1D Western blots, 36 of the 145 RA
sera bound to a 50-kd polypeptide. On 2D Western
blots, anti–50-kdⴙ RA sera recognized a triplet of
isoelectric point 6.5–7.0 and a molecular mass of 50 kd.
The 3 spots of the triplet were analyzed by MALDI-TOF
MS and were shown to correspond to human ␣-enolase.
A goat anti-enolase antiserum, which recognized a band
comigrating with the 50-kd antigen on 1D Western blots,
gave a labeling pattern on 2D Western blots similar to
that observed with anti–50-kdⴙ RA sera. Among the 36
RA sera that identified ␣-enolase in protein maps, only
8 recognized the recombinant (unmodified) ␣-enolase.
The specificity of anti–␣-enolase antibodies for RA was
97.1%. Half of the anti–␣-enolase–positive RA patients
were negative for both rheumatoid factor and antifilaggrin antibodies. The presence of anti–␣-enolase antibodies was the greatest predictive factor of radiologic
progression in the first 66 RA patients included.
Conclusion. Autoantibodies to ␣-enolase, an enzyme of the glycolytic pathway, are present in the sera of
patients with very early RA and have potential diagnostic and prognostic value for RA.
Rheumatoid arthritis (RA) has long been
thought to be an autoimmune disease because various
autoantigens have been shown to be the targets of
synovial T cells and autoantibodies present in patients’
synovial fluid and/or sera (1). Although the autoimmune
nature of RA remains controversial, these autoantibodies have received much attention because their characterization, namely, identification of their target antigen,
may shed some light on the nature and origin of the
immune process and, in fact, their presence might
constitute a marker of the disease (2).
Several autoantibody populations have been detected during the course of RA (1). For example,
rheumatoid factor (RF) has long been considered to be
the most sensitive and best predictive diagnostic marker
of RA but has relatively poor specificity. Recently,
Supported by INSERM, the Fondation de la Recherche
Médicale, and the Programme Hospitalier de Recherche Clinique
VERA-O (Very Early Rheumatoid Arthritis Outcome). Mr. Saulot is
the recipient of a fellowship from the Association de Recherche sur la
Vincent Saulot, MS, Roland Charlionet, PhD, Nadine Machour, PhD, Danièle Gilbert, PhD, François Tron, MD, PhD: Institut
National de la Santé et de la Recherche Médicale, Rouen, France;
Olivier Vittecoq, MD, PhD, Xavier Le Loët, MD: Institut National de
la Santé et de la Recherche Médicale, and Centre Hospitalier Universitaire de Rouen, Rouen, France; 3Patrice Fardellone, MD, PhD:
Centre Hospitalier Universitaire d’Amiens, Amiens, France; 4Catherine Lange, PhD, Laure Marvin, PhD: Centre de Spectrométrie de
Masse, Mont Saint Aignan, France.
Address correspondence and reprint requests to Danièle
Gilbert, PhD, INSERM Unité 519, IFRMP-23, Faculté Mixte de
Médecine et Pharmacie, 22, Boulevard Gambetta, 76183 Rouen,
France. E-mail:
Submitted for publication May 31, 2001; accepted in revised
form January 11, 2002.
antifilaggrin antibodies (AFA) were shown to be more
specific to RA (3). Furthermore, the antigenic determinants recognized by AFA and present in RA sera were
shown to be generated posttranslationally at various
sites on profilaggrin by arginine-residue deimination,
changing them into citrulline residues (4,5). These observations prompted us to search for new autoantibody
populations, including those directed against posttranslationally modified proteins, in the sera of RA patients
using Western blotting of a 2-dimensional (2D) protein
map and mass spectrometry.
Patients and sera. Two hundred fifty-five patients with
very early arthritis were recruited prospectively by rheumatologists and by general practitioners in private practice in 2
French regions: upper Normandy and the metropolitan area of
Amiens. For inclusion criteria, patients were required to have
swelling in at least 2 joints which had persisted for ⱖ4 weeks
and evolved for ⬍6 months (median disease duration 4
months), and to have taken no local or systemic corticosteroids
or disease-modifying antirheumatic drugs before inclusion.
Exclusion criteria were inflammatory back pain, pregnancy, or
breastfeeding. Patients were followed up for 1 year after the
appearance of the first symptoms, at which time diagnoses
were established by a committee of 5 expert rheumatologists.
Three subgroups of patients were thus defined: 1) 145 patients
with RA (subgroup 1), 2) 70 patients with non-RA rheumatic
diseases (subgroup 2), and 3) 40 patients with undifferentiated
arthritis (subgroup 3) (Table 1).
The following markers were evaluated as previously
described (3): RF measured by agglutination tests (latex
fixation test and Rose-Waaler test) and by enzyme-linked
immunosorbent assay (ELISA) (IgM, IgA, and IgG isotypes),
AFA detected by indirect immunofluorescence, and anticalpastatin antibodies detected by ELISA. Radiographs of the
hands and feet were carried out at inclusion and at the end of
followup. They were read chronologically (in order of patient
inclusion) by 2 experienced rheumatologists (OV and PF),
using the van der Heijde–modified Sharp method of scoring
for radiographic damage (6), for the first 66 RA patients
included in subgroup 1. The criterion for severity of the disease
was defined by progression of radiologic damage during the
followup period.
Sera were obtained at the time of inclusion from all
patients enrolled in the cohort, as well as from 26 patients with
systemic lupus erythematosus (SLE), 13 with systemic sclerosis
(SSc), and 14 with primary biliary cirrhosis.
Antibodies. Polyclonal goat IgG raised against a
carboxy-terminal–region peptide common to ␣, ␤, and ␥
isoforms of mouse, rat, and human enolase (200 ␮g/ml; Santa
Cruz Biotechnology, Santa Cruz, CA) was used at a dilution of
Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and Western blotting. Sample preparation. Fresh human placenta was minced and was resuspended
(1 gm/ml) in lysis buffer (50 mM Tris HCl, pH 8.0, 100 mM
Table 1. Frequencies of autoantibodies recognizing ␣-enolase in the
cohort of 255 patients with very early arthritides*
Non-RA rheumatic diseases
Psoriatic arthritis
Crystal-induced arthritis
Paramalignant arthritis
Parvovirus B19 arthritis
Connective tissue disease
Primary Sjögren’s syndrome
Systemic lupus erythematosus
Mixed connective tissue disease
Wegener’s granulomatosis
Behçet’s disease
Eosinophilic fasciitis
Polymyalgia rheumatica
Undifferentiated arthritis
No. of
* RA ⫽ rheumatoid arthritis.
NaCl, 0.02% NaN3, 1% Triton X-100, 1 mM ethylenediaminetetraacetic acid, 1% protease inhibitor cocktail; Sigma,
St. Louis, MO). The suspension was sonicated for 5 minutes,
boiled for 5 minutes, and centrifuged for 20 minutes at 15,000g.
IgG were eliminated by incubating the supernatant with a
mixture of Sepharose beads coupled to protein A (Sigma) for
1 hour at 4°C. After decantation, aliquots of supernatant were
stored at ⫺80°C.
Electrophoresis and Western blotting. Proteins were separated in SDS-PAGE according to the method of Laemmli,
and electrotransferred onto a nitrocellulose membrane. The
filters were then incubated with an RA serum (1:300) or goat
anti-enolase antibodies (1:100) in Tris buffered saline–0.05%
Tween 20 (TBST)–5% dry milk for 2 hours. After washing, the
filters were incubated for 1 hour with 1:15,000-diluted
peroxidase-conjugated goat anti-human IgG (Sigma) or
1:2,000-diluted peroxidase-conjugated donkey anti-goat IgG
(Santa Cruz Biotechnology) in 0.05% TBST–milk. The filters
were washed and revealed by a chemiluminescence reaction
(Supersignal; Pierce, Rockford, IL).
2D-PAGE and immunoblotting. MC-F12 cells (mammary epithelial cell line CRL 10782; American Type Culture
Collection, Rockville, MD) were scraped from plastic dishes,
centrifuged, immediately resuspended in acetone (⫺18°C),
10% trichloroacetic acid, 0.12% dithiothreitol (DTT), and kept
at ⫺18°C overnight. After centrifugation (35,000g for 30
minutes at 4°C), the supernatant was then discarded and the
pellet resuspended in acetone, 0.2% DTT. After 1 hour at
Figure 1. Analysis of rheumatoid arthritis (RA) sera by immunoblotting on an MC-F12 protein map. a, MC-F12
extract separated by 2-dimensional polyacrylamide gel electrophoresis and stained with Coomassie blue (MC-F12
protein map). The following antisera were incubated with the filters: b, serum from an RA patient that bound to
the 50-kd polypeptide; c, anti–50-kd antibodies affinity-purified from the same serum used in b, by elution from
the nitrocellulose strip containing the 50-kd band; d, serum from a patient with systemic lupus erythematosus that
bound to the 50-kd polypeptide. Molecular mass is indicated on the right and the isoelectric point (pI) is indicated
at the top of each figure. The 3 isoforms of ␣-enolase, ␣1, ␣2, and ␣3, are indicated by the arrows.
⫺18°C, the sample was spun again at 35,000g for 30 minutes.
The supernatant was discarded and the pellet was dried under
vacuum. The pellet was then resuspended in a solution of 8M
urea, 2% CHAPS, 1% (weight/volume) dithioerythritol, 7%
spermine (Sigma), 1% protease-inhibitor cocktail (Sigma), and
incubated for 1 hour at 4°C. After centrifugation, the supernatant was collected and stored at ⫺80°C.
The lysate was subjected to 2D-PAGE according to the
system previously described (7) using nonlinear, pH 3–10
gradients for the first dimension and 10% SDS-PAGE for the
second dimension. Finally, gels were either stained with Coomassie blue to obtain a protein map or transferred onto
polyvinylidene fluoride (PVDF) membranes. The immunoreactive spots blotted onto the PVDF membranes were detected
as described above. Autoradiographic films were superimposed on Coomassie blue–stained protein maps to localize the
recognized protein spots.
Trypsin digestion and mass spectrometry. The protein
spots were excised from polyacrylamide gels digested by trypsin, and peptides were analyzed with the matrix-assisted laser
desorption ionization–time-of-flight (MALDI-TOF) Reflector
spectrometer as previously described (7). The peptide masses
were matched within the largest window possible for isoelectric
point (pI) and molecular mass, and with species specificity
(Homo sapiens).
Antibody elution. Proteins extracted from human placenta were separated by SDS-PAGE and transferred onto a
nitrocellulose membrane as described above. After incubation
with the RA serum and visualization of the polypeptide of
interest, the band was cut out of the remaining membrane and
incubated for 20 minutes in 1 ml of elution buffer (0.2M
glycine, pH 2.8) and the supernatant was neutralized with 0.1
volume of 1M Tris. The eluted antibodies were used for
immunoblotting on 2D protein maps.
In vitro antigen expression. The complete complementary DNA (cDNA) encoding human ␣-enolase was isolated
from a cDNA expression library derived from synoviocytes
obtained from an RA patient (Stratagene, La Jolla, CA) and
immunoscreened with goat anti-enolase antibodies. This
cDNA was subcloned in frame in the pSPUTK in vitro
translation vector (Stratagene) using the Apa I and Bam HI
restriction sites. The translation product was synthesized as a
Figure 2. Western blot analysis of rheumatoid arthritis (RA) sera
using the biotinylated in vitro–synthesized recombinant ␣-enolase
as the antigen. Filters were incubated with the following reagents:
lane 1, peroxidase-conjugated streptavidin; lane 2, goat anti-enolase
antibodies; lane 3, secondary anti-goat IgG antibodies conjugated
to peroxidase (negative control); lanes 4 and 5, RA sera that bound
to ␣-enolase in the 2-dimensional (2D) protein map (only the sera
in lane 4 bound to the recombinant product); lane 6, systemic lupus
erythematosus serum that bound to ␣-enolase in the 2D protein
biotinylated polypeptide, purified by SoftLink Soft Release
Avidin Resin (Promega, Madison, WI), migrated in SDSPAGE, and subjected to Western blotting experiments as
described above.
1D Western blot analysis of RA sera. Two hundred fifty-five sera obtained from patients with early
arthritis were screened by Western blot on human
placenta extract. Among the sera obtained from the 145
patients diagnosed 1 year later as having RA, 36 (24.8%)
recognized a 50-kd band. Only 2 of the 70 sera obtained
from patients with early non-RA inflammatory arthritides reacted with the 50-kd protein (Table 1). Thus, the
anti–50-kd reactivity was mainly observed in early RA
sera with a specificity of 97.1% compared with sera from
patients with non-RA rheumatic diseases (subgroup 2).
Moreover, 17 of the 36 positive sera (47.2%) reacted
with the 50-kd band only, i.e., they were AFA and RF
negative. These results suggest that the 50-kd antigen–
anti–50-kd antibody system might contribute to the early
diagnosis of RA, and prompted us to identify the 50-kd
2D Western blot analysis of RA sera. To characterize the 50-kd polypeptide recognized by the 36 RA
sera, we first verified its presence in extracts of the
mammary epithelial cell line, MC-F12, currently used
for 2D-PAGE in our laboratory. All 36 anti–50-kd⫹ RA
sera were then analyzed by immunoblotting on MC-F12
proteins separated by 2D-PAGE (Figure 1a). All sera
recognized the same protein triplet, which consisted of
components with a molecular mass of ⬃50 kd and pI
between 6.5 and 7.0 (Figure 1b). Among RA and
non-RA sera that did not react with the 50-kd band in
1D Western blot from the cohort of patients with very
early arthritides, 30 were randomly chosen and analyzed
by immunoblotting on the MC-F12 protein map obtained by 2D-PAGE. None of them bound to the
identified triplet.
Immunoblotting with eluted antibodies. To demonstrate the identity between the 50-kd polypeptide
recognized by RA sera in 1D placenta-extract immunoblots and the triplet recognized in the MC-F12 protein
map, the antibodies reacting with the former were eluted
and allowed to react with the MC-F12 protein map.
Figure 1c shows that eluted antibodies bound exclusively
to the same triplet.
Mass spectrometry analysis of the immunoreactive MC-F12 triplet. To characterize the triplet recognized by RA sera, the 3 protein spots (␣1, ␣2, and ␣3)
were analyzed by MALDI-TOF mass spectrometry. The
comparison of the mass spectra with those contained in
other databases (MS-FIT and SWISS-PROT) allowed us
to identify, with high probability (7 ⫻ 104), the 3
isoforms of human ␣-enolase. The database-matched
peptides of the spectrum of the most basic and major
protein spot (␣3) covered 80% of the protein sequence
of ␣-enolase. Comparison of the 3 very similar spectra
given by ␣1, ␣2, and ␣3 revealed that the 3 isoforms of
␣-enolase had slightly different pI, mainly attributed to
differences in their phosphorylation (data not shown).
Reactivity of RA sera with recombinant human
␣-enolase. Recombinant human ␣-enolase was produced by an in vitro transcription–translation system to
test the reactivities of sera with the recombinant
polypeptide. First, correct expression of the recombinant
human ␣-enolase was verified by Western blot, in
which the translated polypeptide was visualized with
peroxidase-conjugated streptavidin at ⬃50 kd, in accordance with the expected size of the recombinant product. Second, goat anti-enolase antibodies incubated in
Western blot recognized the in vitro–synthesized
polypeptide, as expected. Then, the 36 RA sera that
bound to ␣-enolase in the 2D protein map were analyzed
by Western blot using the recombinant enolase as the
antigen. Only 8 of them recognized the recombinant
protein, while the other 28 RA sera were negative
(Figure 2).
Anti–␣-enolase antibodies in other systemic diseases. Sera from 26 patients with SLE, 13 with SSc, and
14 with primary biliary cirrhosis were screened by 1D
Western blot on human placenta extract. Five of the 26
SLE sera (19%) and 2 of the 13 SSc sera (15%) reacted
with the 50-kd protein. All of these sera tested on the
MC-F12 protein map bound to the different isoforms
of ␣-enolase (Figure 1d). All of these sera also reacted
with the recombinant protein obtained by in vitro
transcription–translation (Figure 2).
Prognostic value of anti–␣-enolase antibodies for
RA. Sixty-six RA patients (subgroup 1) who were enrolled in the cohort and who underwent evaluation of
radiologic damage were analyzed to determine the prognostic value of anti–␣-enolase antibodies. The independent variable was dichotomized into the presence (n ⫽
26) or absence (n ⫽ 40) of radiologic progression.
Dependent variables were all of the autoantibodies
tested (see Patients and Methods). Using a univariate
analysis (Fisher’s 2-tailed exact test), only positive results on the latex fixation test (P ⫽ 0.01) or presence of
anti–␣-enolase antibodies at inclusion (P ⫽ 0.01) was
predictive of the progression of radiologic damage.
Using a multivariate stepwise logistic regression analysis,
positivity for anti–␣-enolase antibodies was the strongest
predictor of radiologic progression (P ⫽ 0.004).
This study showed that autoantibodies directed
against human ␣-enolase are present at an early stage in
the sera of patients who are subsequently diagnosed as
having RA. Indeed, almost 25% of RA sera recognized
a 50-kd polypeptide on Western blot using human
placenta extract and a protein triplet in MC-F12 extracts
separated by 2D-PAGE. This protein triplet, identified
as human ␣-enolase by mass spectrometry, corresponded to the 50-kd band recognized by RA sera,
because 1) anti–50-kd antibodies eluted from the 50-kd
band blotted onto a nitrocellulose strip gave the same
immunoreactive pattern, and 2) goat anti–␣-enolase
antibodies that reacted with the 50-kd band comigrating
with the polypeptide bound by RA sera also recognized
the triplet, whose 3 protein spots were shown to correspond to 3 isoforms of human ␣-enolase (data not
shown). Thus, human ␣-enolase is a new putative target
autoantigen in RA.
Autoantibodies directed against ␣-enolase have
previously been described in various pathologic situations, including autoimmune diseases such as SLE (8)
and discoid lupus erythematosus (9), SSc, primary bili-
ary cirrhosis, and autoimmune hepatitis (10), chronic
inflammatory diseases such as primary sclerosing
cholangitis, inflammatory bowel diseases, primary membranous nephropathy, and cancer-associated retinopathy
(10). Thus, at first glance, the anti–␣-enolase response
does not seem to be restricted to RA, but rather, seems
to occur in several chronic disorders, which limits its
potential significance and diagnostic value in RA.
However, properties of the anti–␣-enolase antibodies detected in RA patients’ sera might encourage a
reconsideration of their role in the pathogenesis and
diagnosis of RA. First, Western blot analysis of the 255
sera obtained from patients with early arthritis showed
that the 50-kd response was observed almost exclusively
in sera obtained from patients subsequently shown to
have RA. Of note, they were not detected in the more
common non-RA rheumatic diseases, i.e., spondylarthropathies and primary Sjögren’s syndrome. Moreover,
half of the anti–␣-enolase–positive RA patients were
negative for both RF and AFA. Second, the presence of
anti–␣-enolase antibodies was correlated with the severity of the articular destruction and, in addition, our
preliminary data suggest that anti–␣-enolase antibodies
might constitute a better predictive marker of radiologic
progression than does RF. Third, the lack of reactivity
against human recombinant ␣-enolase in most RA patients’ sera, compared with the reactivity in SLE patients’ sera, suggests that RA sera predominantly bind to
posttranslationally modified or appropriately folded
epitopes of the protein.
The cytoplasmic and ubiquitous glycolytic enzyme ␣-enolase catalyzes the formation of phosphoenolpyruvate from 2-phosphoglycerate, the second of the two
high-energy intermediates that generate ATP in glycolysis (10). Of note, other glycolytic enzymes, such as
glucose-6-phosphate isomerase (GPI) (11) and aldolase
(12), were recently identified as target antigens of RA
sera, and GPI is the target of the arthritogenic IgG
antibodies produced in the K/BxN mouse model, which
shares several features with RA (13). Thus, antibodies
directed against glycolytic enzymes might very well be
implicated in the pathophysiology of RA, at least in
terms of dysregulation of glucose metabolism in hyperplastic synoviocytes in RA, as has been previously described (14).
1. Blä␤ S, Engel JM, Burmester GR. The immunologic homunculus
in rheumatoid arthritis. Arthritis Rheum 1999;42:2499–506.
2. Desprès N, Boire G, Lopez-Longo FJ, Ménard HA. The Sa
system: a novel antigen–antibody system specific for rheumatoid
arthritis. J Rheumatol 1994;21:1027–33.
Vittecoq O, Jouen-Beades F, Krzanowska K, Bichon-Tauvel I,
Ménard JF, Daragon A, et al. Rheumatoid factors, anti-filaggrin
antibodies, and low in vitro interleukin 2 and interferon-␥ production are useful immunological markers for early diagnosis of
community cases of rheumatoid arthritis: preliminary study. Joint
Bone Spine 2001;68:144–53.
Girbal-Neuhauser E, Durieux JJ, Arnaud M, Dalbon P, Sebbag M,
Vincent C, et al. The epitopes targeted by the rheumatoid
arthritis-associated antifilaggrin autoantibodies are posttranslationally generated on various sites of (pro)filaggrin by deimination
of arginine residues. J Immunol 1999;162:585–94.
Schellekens GA, de Jong BAW, van den Hoogen FHJ, van de
Putte LBA, van Venrooij WJ. Citrulline is an essential constituent
of antigenic determinants recognized by rheumatoid arthritisspecific autoantibodies. J Clin Invest 1998;101:273–81.
van der Heijde DM, van Riel PL, Nurver-Zwart IH, Gribnau FW,
van de Putte LB. Effects of hydroxychloroquine and sulphasalazine on progression of joint damage in rheumatoid arthritis.
Lancet 1989;1:1036–8.
Thébault S, Gilbert D, Machour N, Marvin, L, Lange C, Tron F,
et al. Two-dimensional electrophoresis and mass spectrometry
identification of proteins bound by a murine monoclonal anticardiolipin antibody: a powerful technique to characterize the
cross-reactivity of a single autoantibody. Electrophoresis 2000;21:
Pratesi F, Moscato S, Sabbatini A, Chimenti D, Bombardieri S,
Migliorini P. Autoantibodies specific for ␣-enolase in systemic
autoimmune disorders. J Rheumatol 2000;27:109–15.
Gitlits VM, Sentry JW, Matthew MLSM, Smith AI, Toth BH.
Autoantibodies to evolutionarily conserved epitopes of enolase in
a patient with discoid lupus erythematosus. Immunology 1997;92:
Pancholi V. Multifunctional ␣-enolase: its role in diseases. Cell
Mol Life Sci 2001;58:902–20.
Schaller M, Burton DR, Ditzel HJ. Autoantibodies to GPI in
rheumatoid arthritis: linkage between an animal model and human
disease. Nature Immunol 2001;2:746–53.
Ukaji F, Kitajima I, Kubo T, Shimizu C, Nakajima T, Maruyama I.
Serum samples of patients with rheumatoid arthritis contain a
specific autoantibody to “denatured” aldolase A in the osteoblastlike cell line, MG-63. Ann Rheum Dis 1999;58:169–74.
Matsumoto I, Staub A, Benoist C, Mathis D. Arthritis provoked by
linked T and B cell recognition of a glycolytic enzyme. Science
Henderson B, Bitensky L, Chayen J. Glycolytic activity in human
synovial lining cells in rheumatoid arthritis. Ann Rheum Dis
Без категории
Размер файла
180 Кб
presence, sera, patients, autoantibodies, enzymes, enolase, glycolytic, arthritis, early, rheumatoid
Пожаловаться на содержимое документа