close

Вход

Забыли?

вход по аккаунту

?

Studies on the infectious etiology of human rheumatoid arthritis.

код для вставкиСкачать
ARTHRITIS
435
&
RHEUMATISM
OFFICIAL JOURNAL O F T H E AMERICAN RHEUMATISM ASSOCIATION
SECTION OF T H E A R T H R I T I S FOUNDATION
STUDIES ON THE INFECTIOUS
ETIOLOGY OF HUMAN
RHEUMATOID ARTHRITIS
11. SEARCH FOR HUMORAL AND CELL-BOUND
ANTIBODIES AGAINST MYCOPLASMAL ANTIGENS
B. C. COLE, M. B. TAYLOR, and J. R. WARD
A total of 29 rheumatoid patients and 19 nonrheumatoid patients were tested for evidence of present or past infection b y M pneumoniae, M hominis,
M fermentam, M arthritidis, M pulmonis, and M hyorhinis. T h e techniques of lymphocyte transformation,
metabolic-inhibiting antibody test, and mycoplasmacidal antibody test indicated no significant difference
in the response of rheumatoid as opposed to nonrheumatoid patients.
Although a wide variety of immunologic processes are apparent within the joints of patients exhibiting rheumatoid arthritis (RA), the nature of the
initiating agent or event appears elusive. An infectious etiology for human rheumatoid arthritis remains
a n attractive hypothesis despite numerous conflicting
and inconclusive results on the isolation of an agent.
From the Division of Arthritis, Department of Medicine,
University of Utah College of Medicine, Salt Lake City, Utah
84132.
Supported by the National Institutes of Arthritis and
Metabolic Diseases grant No. AM02255 and by a grant from the
Kroc Foundation.
Barry C. Cole, Ph.D.: Division of Arthritis, Department
of Medicine, University of Utah College of Medicine; Mark B.
Taylor, M.D.: Division of Arthritis, Department of Medicine,
University of Ulah College of Medicine; John R. Ward, M.D.:
Division of Arthritis, Department of Medicine, University of
Utah College of Medicine.
Address reprint requests to Dr. Cole.
Submitted for publication December 13, 1974; accepted
January 3, 1975.
Arthritis and Rheumatism, Vol. IS, No. 5 (September-october lW5)
I n 1939, Sabin suggested that a mycoplasma
may be the cause of human RA after he showed that
a chronic arthritis closely resembling the human disease could be induced in mice by these organisms (1).
Attempts to implicate mycoplasmas in the etiology of
human RA have not been conclusive (2,3). Williams’
(4) observations on the isolation of Mycoplasma fermentans from synovial material using a preliminary
sucrose gradient fractionation or other techniques
have not been confirmed (2,3,5,6). These studies have
also failed to find a n association between the presence
of mycoplasmas or humoral antimycoplasma antibodies and human rheumatoid arthritis (3,5,6). Reports that the migration of leukocytes from human
rheumatoid arthritis patients is inhibited in the presence of M fermentans (7) have not been confirmed
(8,9). Recent studies by Brostoff et a1 (10) suggest that
leukocytes taken from RA patients exhibit inhibition
of migration when exposed to aggregated IgG. These
observations could explain the inhibition of migration observed with M fermentam because the latter
organism has been reported to bind gammaglobulin
(7). However additional studies indicated that M
fermentons-mediated inhibition of leukocyte migration correlated better with disease activity than did
gammaglobulin-mediated inhibition of leukocyte migration (1 l).
Additional evidence for the infectious etiology
of human RA was recently presented by Gottlieb et al
(12), who demonstrated that prior to onset of arthritis
COLE ET AL
436
Table 1. Mean Blastogenic Responses of Rheumatoid ( R A ) and Control (CON)
Lymphocytes to Phytohemagglutinin ( P H A ) and Mycoplasma Antigens
Mean Blastogenic Indices
PHA
RA (29)*
CON (19)*
P value$
Total (48)*
187.8 f 156.3t
94.9 f 63.1
< 0.01s
141.4
M pneumoniae
2.4 f 2.5
1.5 k 1.2
> 0.1
2.0
M hominis M fermentans
1.6 f 2.2
1.2 t 1.3
> 0.5
1.4
2.1 3z 2.9
1.6 3z 1.3
> 0.5
1.9
M arthritidis
M pulmonis
M hyorhinis
1.1 f 2.0
0.8 f 1.0
> 0.5
1.o
2.0 f 2.1
1.8 f 2.6
> 0.5
1.9
1.2 +- 2.7
1.3 -C 1.5
> 0.5
1.2
*Number of patients tested.
?Mean value 3z standard deviation.
$P values calculated using the Welch t test.
§See text.
RA patients h a d experienced a greater exposure to
household pets t h a n h a d control patients. Failure to
cultivate mycoplasmas does n o t necessarily imply their
absence. T h u s arthritis will persist in swine long after
the inducing agent, M hyorhinis, is no longer cultivable (13). I t was recently shown that M hyorhinis,
which usually grows readily in artificial media, can
become dependent on the presence of animal cells
after prolonged passage in cell culture (14). T h u s a
similar adaptation of a n agent to growth in h u m a n
synovium may negate attempts to detect it.
The present study was undertaken t o determine
whether two new immunologic procedures could be
utilized to detect evidence of present or past mycoplasma infection in patients exhibiting rheumatoid
arthritis. Firstly, the technique of lymphocyte transformation using whole blood culture (15) was employed t o determine the presence or absence of a cellmediated i m m u n e response to a variety of mycoplasma
antigens. Secondly, because some animal hosts do n o t
produce a complete antibody response t o mycoplasma
(16,17), a m o r e sensitive cidal test (17) was utilized
to detect the presence of humoral antibodies against
mycoplasmas.
MATERIALS AND METHODS
Mycoplasma Cultivation and Antigen Preparation.
The following mycoplasmas were used throughout this
study: M p n e u n o n i a e M710-001-084, M hominis M711-002084, M fermentans M713-002-085, and M hyorhinis M718001-084 were obtained from the National Institutes of
Health, courtesy of Dr. M. F. Barile. M pulrnonis J B was
obtained from Dr. J. G. Tully. M arthritidis strain 20-P
was isolated from human rheumatoid synovium (18) and
was kindly supplied by Dr. E. Jansson.
Mycoplasmas were grown in mycoplasma broth
(Difco Laboratories) supplemented to final concentrations
of 15-20% vol/vol horse serum, 5% vol/vol yeast extract,
and 1000 U per milliliter of penicillin G (19,ZO). T h e
cells were harvested by centrifugation at 27,000 x g, washed
three times in phosphate buffered saline, and finally re-
suspended in sterile distilled water. T h e suspensions were
sonified at 20,000 kc per second for 5 minutes and the protein content was determined by the method of Lowry (21).
Stock antigen suspensions containing 25 pg of protein per
milliliter were prepared and cultured to ensure absence
of viable mycoplasms or bacterial contamination.
Lymphocyte Transformation. T h e basic technique
used was that described by Pauiy et a1 (15). Human blood
collected with 10 U per milliliter of preservative-free
heparin was diluted 1:20 with RPMI 1640 medium containing 100 U per milliliter of penicillin and streptomycin.
T h e suspension was dispensed in 3-ml aliquots into 16 x
125 mm disposable polystyrene tubes (Falcon Plastics). Each
mycoplasma antigen (0.4 ml) was added to triplicate tubes.
Triplicate control (negative) tubes containing 0.4 ml of a
dilution of mycoplasma broth in KPMI medium were also
set up. Positive controls for each patient were prepared by
adding 10 pg of purified phytohemagglutinin (PHA, Burroughs Wellcome MR 68) in 0.4 ml RPMI medium to
triplicate tubes. T h e lymphocyte cultures were incubated
with loosened caps in a CO, incubator at 37°C. After 5
days the cultures were pulsed for 24 hours with 1 pCi SH
thymidine per tube. T h e cultures were harvested according
to Pauly et a1 (15), using NCS as solubilizer and spectraHuor as the scintillation fluid (Amersham Searle). The
samples were counted for 10 minutes on a Nuclear Chicago
Scintillation Counter and were adjusted to 100% efficiency.
T h e results were expressed as the index of the disintegrations per minute (DPM) obtained with the antigen-containing culture as compared with the DPM observed in
the broth-containing control cultures. Statistical analysis
of the results was performed using the Welch t test because
the variances between the groups could not be assumed to
be equal.
Metabolic-Inhibiting (MI) and hIycoplasmacida1
Antibody Tests. Serum samples prepared from clotted
blood, taken at the same time as the blood used for the
translormation studies, were tested for cidal antibodies
against all of the six mycoplasmas as has been previously
described (16). M I antibody tests employing the microtiter
system were also set up using either glucose (22) or arginine
(23) as substrate.
Patients. T h e 29 patients with active rheumatoid
arthritis studied were receiving ibuprofen, indomethacin,
prednisolone, aspirin, meclofenamic acid, or combinations
of these at the time of sampling.
437
INFECTIOUS ETIOLOGY OF HUMAN RA
Table 2. Percentages of Rheumatoid ( R A ) and Control ( C O N )
Patients Responding to Various Mycoplasmal Antigens
Antigens
Patient
Group
<1
M pneumoniae
RA(29)*
CON (19)*
Total (48)*
M hominis M fermentans M arthritidis M pulmonis M hyorhinis
27.6
31.6
29.2
55.2
57.9
56.3
41.4
42.1
41.7
79.3
73.7
77.1
24.1
36.8
29.2
41.4
57.9
47.9
> 2 RA(29)
CON (19)
Total (48)
27.6
21.1
25.0
17.2
10.5
14.6
17.2
26.3
20.8
6.9
10.5
R.3
27.6
21.1
25.0
17.2
15.8
16.7
> 4 RA
CON
Total
17.2
5.3
12.5
6.8
5.3
6.3
10.3
5.3
8.3
6.8
5.3
6.3
13.8
15.8
14.6
6.8
10.5
8.3
*Number of patients tested.
The control group consisted of 18 patients with
osteoarthritis and 1 patient with gout. The drug treatment
at time of sampling consisted of ibuprofen, indomethacin,
aspirin, or allopurinol.
RESULTS
T h e blastogenic responses of RA and control
lymphocytes to PHA and mycoplasmal antigens are
summarized in Tables 1 and 2. T h e lymphocytes from
all patients responded well to PHA. RA patients exhibited blastogenic indices ranging from 23.5 to 719
(mean: 187.8) as opposed to control patients who had
indices ranging from 7.3 to 251.5 (mean: 94.9). This
difference was statistically significant. Because of the
variability in "H-thymidine uptake in the control cultures, the results were calculated by comparing DPM
values (rather than individual indices) observed in
lymphocytes from control and arthritic patients exposed to PHA. N o significant difference was observed.
There was no obvious correlation between drug treatment and either inhibition or enhancement of the
PHA response, although the numbers of patients receiving any one drug were small.
Lymphocytes exposed to mycoplasmal antigens
exhibited varying degrees of blastogenic transformation; however there was no significant difference between RA and control patients. M pneumoniae antigen induced the greatest transformation of lymphocytes, ie a mean blastogenic index of 2.4 for RA patients and 1.5 for non-RA patients (Table l). Individual values ranged from 0.5 to 9.8 for RA patients
and from 0.6 to 5.2 for control patients. When RA
and control patient groups were combined, it was
shown that 29.27, of patients exhibited indices of
less than 1 and 25.0% had indices greater than 2
(Table 2).
T h e least degree of transformation was obtained with M urthritidis antigen, which exhibited a
mean index of 1.1 with RA patients and 0.8 with
control patients. Examination of individual patients
in both groups revealed that the uptake of 3H-thymidine by lymphocytes from 77.1% of patients was
less in the presence of M arthritidis antigen than in
the lymphocytes without M arthritidis antigen (Table
2). Only 8.3% of patients exhibited indices greater
than 2 in the presence of M arthritidis antigen. T h e
inhibitory activity of M arthritidis antigen has been
reported by other investigators (24).
No significant differences in blastogenic indices
were apparent between RA and non-RA patients
when their lymphocytes were exposed to M fermentans, M pzrlmonis, or M hyorhinis antigens.
T h e lymphocytes from some individuals responded to more than one antigen. T h u s 4 RA and
4 control patients exhibited indices greater than 3.0
to three or more antigens. There was no association
of these responses with drug therapy. One of the 4
KA patients receiving meclofenamic acid responded
to all six antigens; the blastogenic indices ranged
from 9.8 to 14.0. N o immediate explanation of this
response is apparent because mycoplasma medium
alone was not stimulatory.
Most patients were also examined for the presence of circulating M I and mycoplasmacidal antibodies against all six mycoplasma species tested. T h e
results are summarized in Tables 3 and 4. There was
no association between presence or absence of antibodies against any antigen and type of disease present.
Both mycoplasmacidal and hlI antibodies were detected more frequently against M pneumoniae and M
hyorhinis than against the other mycoplasmas. With a
few exceptions antibody titers were low and of doubt-
COLE E T AL
438
Table 3. Metabolic-Inhibiting Antibodies Present in Serum
of Rheumatoid ( R A ) and Control (CON) Patients
No. of Patients Exhibiting Following
Antigen
M pneumoniae
M hominis
M fermentans
M arthritidis
M pulmonis
M hyorhinis
Patient
Group
RA (28)*
CON (18)*
RA (28)
CON (18)
RA(28)
CON (18)
RA(28)
CON (18)
RA (28)
CON (18)
RA (28)
CON (18)
Titers Against Indicated Antigens
<1
2
1:2
1:4
1:s
23
13
28
17
28
16
28
18
25
17
16
10
*Total number of patients.
ful significance. T h e mycoplasmacidal test appeared
to be more sensitive i n detecting antibody than was
the metabolic inhibition test. Thus, for all antigens,
a total of 60 tests exhibited titers of 1:lO or greater
with the mycoplasmacidal reaction whereas only 35
tests exhibited MI antibody titers of 1:2 or greater.
Antibodies against M urthl-itidis were detected only
when the mycoplasmacidal antibody test was used.
A significant correlation between the presence
of mycoplasmacidal and MI antibodies occurred only
with sera exhibiting MI antibody titers of 1:4 or
greater. These results thus raise the question of the
reliability of lower antibody titers. Similarly only
those few patients who exhibited both MI and cidal
antibody titers greater than 1:2 and 1: 10 respectively
had a consistently elevated blastogenic index to the
homologous antigen.
DISCUSSION
T h e whole blood procedure (15) for lymphocyte stimulation, although possibly less sensitive than
other techniques employing concentrated lymphocyte
suspensions (25), appears to be highly suitable for
testing the immunologic responsiveness of individual
patients toward a battery of antigens.
Mycoplasma pneumoniae was utilized in this
study as a positive control because lymphocyte transformation has been shown in patients previously infected with this mycopktsma (25-2’7). I n this study
lymphocytes taken from patients exhibiting both cidal
and MI antibodies transformed significantly when exposed to M pneumoniae antigen.
T h e majority of both rheumatoid and non-
rheumatoid patients responded well to PHA, exhibiting mean combined indices of 141.4. Although the
PHA response of RA patients was somewhat higher
than that of control patients, statistical analysis of
the results indicated no significance due to variability
in background levels of 3H-thymidine incorporation.
Opelz, Terasaki, and Hirata (28) recently reported
that aspirin, when added to lymphocytes in physiologic dose ranges, inhibited the mitogenic response
to PHA. Other workers have shown that prednisolone
also inhibits lymphocyte mitosis (29). Although the
patient groups were small, we failed to demonstrate
any inhibitory effect of patient drug treatment on
lymphocyte reactivity to PHA in vitro. However negative results may be due to the dilution factor inherent
in the whole blood procedure (15) used for lymphocyte transformation.
I n this study the correlation observed between
M I ant1 cidal antibodies against the mycoplasmas was
not good at tlie lower titers, suggesting either that the
two tests were measuring antibodies or that the low
titers detected were not due to antibody activity. Although the inhibitory activity of some sera could be
due to drug therapy, this cause is unlikely because
there was no positive correlation between any one
drug treatment and enhancement of the titers.
N o evidence was found for an association of
Kh patients with the presence of humoral or cellmediated antibody against any of tlie mycoplasmas
tested. In contrast to the observations of Williams (4),
antibodies against i\.I feymcntuns were not detected
more frequently in tlie sera taken from rheumatoid
patients. Negative results have been reported by other
439
INFECTIOUS ETIOLOGY OF HUMAN RA
Table 4. Mycoplasmacidal Antibodies Present in Serum of
Rheumatoid (RAJ and Control (CON) Patients
Antigen
M pneumoniae
M hominis
M fermentans
M arthritidis
M pulmonis
M hyorhinis
Patient
Group
RA(28)*
CON (18)*
RA(28)
CON (18)
RA (28)
CON (18)
RA(28)
CON (18)
RA(28)
CON (18)
RA(28)
CON (18)
No. of Patients Exhibiting Following
Titers Against Indicated Antigens
< 10
1: 10
1:40
17
9
28
18
27
16
21
14
26
16
15
9
7
7
3
2
1
2
5
4
2
2
11
8
-
2
2
1
1:60
1
-
-
*Total number of patients.
investigators (3,5,6,30,31). Similarly the present results are consistent with others (8,9) in that no evidence of a cell-mediated response specific for RA patients was found.
Bartholomew, using cell culture techniques
(32), reported the isolation of M hyorhinis from patients with rheumatoid arthritis and systemic lupus
erythematosus and found complement-fixing antibodies against this organism in 90% of patients. We
have failed to find a relationship between rheumatoid arthritis and either a humoral or cell-mediated
antibody response toward M hyorhinis. Other investigators have likewise failed to find a n association of
humoral antibody to M hyorhinis and the presence
of arthritis. Our studies and those of Stewart et a1
(3) revealed a high incidence of humoral antibody
against M hyorhinis in both RA and control patients.
Despite the observations of Jansson and coworkers (18,33) we and others (30,31) have failed to
associate M arthritidis with human rheumatoid arthritis. I n the lymphocyte transformation test the presence of M arthritidis antigen resulted in less uptake
of 3HH-thymidine
as compared with control cultures
containing no antigen. Although M arthritidis and
other arginine-utilizing species of mycoplasma are
known to interfere with lymphocyte mitosis (24),
transformation can occur in lymphocytes taken from
sensitized hosts. T h u s we have shown that lymphocytes taken from mice chronically infected with M
arthritidis transform significantly when exposed to
specific antigen (34).
T h e remaining mycoplasmas, M pulmonis and
M hominis, were likewise found not to be associated
with the RA group of patients. Humoral antibody
tests in other laboratories have resulted in similar
findings (30,31).
What are the prospects of a mycoplasmal etiology for human rheumatoid arthritis? I n view of the
numerous cultural and serologic investigations that
report negative data, the species considered in this
article do not rank as likely etiologic candidates. It
is not impossible, however, that some other mycoplasma is responsible. T h e finding of Gottlieb et a1
(12) that RA patients as opposed to control patients
had experienced greater exposure to household pets
should stimulate work to detect the possible presence
of feline and canine species of mycoplasma in RA.
T h e fact that none of these species has been isolated
from RA patients could be due to the ability of some
mycoplasmas to enter into a nonisolable phase in
association with host cells (14). Unfortunately present
mycoplasma media remain unsuitable for maintaining the growth of many fresh isolates (35). New approaches such as that described by Person et a1 (36)
for the detection of nonhost DNA are clearly much
in need. Meanwhile experimental mycoplasmal models
of chronic arthritis, namely M arthritidis and M pulmonis induced disease of mice, and M hyorhinis disease of swine, have been receiving increasing attention. It is hoped that continued research into these
models of disease will provide clues into both the
mechanisms of pathogenesis of RA and the development of new methodology applicable to human
studies.
ACKNOWLEDGMENTS
The authors thank L. Golightly-Rowland, B. Santistevan, and J. Cadette for technical assistance.
COLE E T AL
440
REFERENCES
1. Sabin AB: Experimental proliferative arthritis in mice
produced by filterable pleuropneumonia-like microorganisms. Science 89:228-229, 1939
2. Cole BC, Ward JR, Smith CB: Studies on the infectious etiology of rheumatoid arthritis. Arthritis Rheum
16:191-198, 1973
3. Stewart SM, Duthie JJR, Mackay JMK, et al: Mycoplasmas and rheumatoid arthritis. Ann Rheum Dis 33:
349-352, 1974
4. Williams MH: Recovery of mycoplasma from rheumatoid synovial fluid. Edited by JJR Duthie, WRM Alexander. Third Pfizer International Symposium on Rheumatic Diseases. Edinburgh, Edinburgh University Press,
1968, pp 171-181
5. Mirdh PA, Nilsson FJ, Bjelle A: Mycoplasmas and
bacteria in synovial fluid from patients with arthritis.
Ann Rheum Dis 32:319-325, 1973
6. Windsor GD, Nicholls A, Maini RN, et al: Search for
mycoplasma in synovial fluids from patients with rheumatoid arthritis. Ann Rheum Dis 33:70-74, 1974
7. Williams MH, Brostoff J, Roitt JM: Possible role of
Mycoplasma fermentans in pathogenesis of rheumatoid
arthritis. Lancet 2277-280, 1970
8. Maini RN, Lemcke RM, Windsor GD, et a]: Mycoplasmas and rheumatoid arthritis: studies of cellular hypersensitivity in vitro and attempts at isolation from
synovial fluids, XI11 International Congress on Rheumatology. Excerpta Medica International Congress
Series No. 299. Amsterdam, Elsevier/Excerpta Medica,
1973, p 11
9. Taylor-Robinson D: Discussion, Pathogenic Mycoplasmas. Ciba Foundation Symposium. Amsterdam, London, New York, Associated Scientific Publishers, 1972,
p 257
10. Brostoff J, Howell A, Roitt IM: Leucocyte migration
inhibition with aggregated gammaglobulin in patients
with rheumatoid arthritis. Clin Exp Immunol 15:l-7,
1973
11. Brostoff J, Freedman A, Roitt IM: Leucocyte migration inhibition of Mycoplasma fermentans in patients
with rheumatoid arthritis. Int Arch Allergy Appl
Immunol45:69&696, 1973
12. Gottlieb NL, Ditchek N, Poiley J, et al: Pets and rheumatoid arthritis. An epidemiologic survey. Arthritis
Rheum 17:22%234, 1974
13. Ennis RS, Johnson JS, Decker JL: Persistent Mycoplasma hyorhinis (MH) antigen in chronic mycoplasma
arthritis of swine. Arthritis Rheum 15:108, 1972 (abstr)
14. Hopps HE, Meyer BC, Barile MF, et al: Problems concerning “non-cultivable” mycoplasma contaminants in
tissue cultures. Ann NY Acad Sci 225:265-276, 1973
15. Pauly JL, Sokal JE: A simplified technique for in vitro
studies of lymphocyte reactivity. Proc Soc Exp Biol
Med 140:40-44, 1972
16. Cole BC, Cahill JF, Wiley BB, et al: Immunological
responses of the rat to Mycoplasma arthritidis. J Bacteriol 98:930-937, 1969
17. Cole BC, Ward JR: Detection and characterization of
defective mycoplasmacidal antibody produced by rodents
against Mycoplasrna arthritidis. Infect Immunol 8: 199207, 1973
18. Jansson E, Wager 0: Mycoplasma in collagen diseases
and blood dyscrasia. Ann NY Acad Sci 143:535-543,
1967
19. Chanock RM, Hayflick L, Barile MF: Growth on an
artificial medium of an agent associated with atypical
pneumonia and its identification as a PPLO. Proc Natl
Acad Sci 48:41-49, 1962
20. Hayflick L: Tissue cultures and mycoplasmas. Tex Rep
Biol Med 23:285-303, 1965 (suppl 1)
21. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein
measurement with the folin phenol reagent. J Biol
Chem 193:265-275, 1951
22. Taylor-Robinson D, Purcell RM, Wong DC, et al: A
colour test for the measurement of antibody to certain
niycoplasma species based upon the inhibition of acid
production. J Hyg (Camb) 64:91-104, 1966
23. Purcell RH, Taylor-Robinson D, Wong DC, et al: A
color test for the measurement of antibody to the nonacid forming human mycoplasma species. Am J Epidemiol 84:51-56, 1966
24. Morton HE, Copperman R, Lam GT: Some properties of the inhibitor(s) of lymphocyte mitosis derived
from mycoplasma. J Bacteriol 95:2418-2419, 1968
25. Biberfield G, Biberfeld P, Sterner G: Cell-mediated
immune response following Mycoplasma pneumoniae
infection i n man. I. Lymphocyte stimulation. Clin Exp
Immunol 17:29-41, 1974
26. Levinthal BG, Smith CB, Carbone PP, et al: Lymphocyte transformation in response to M pneumoniae,
Proceedings of the Third Annual Leucocyte Culture
Conference. Edited by WO Rieke. New York, Appleton-Century-Crofts, 1974, p 519
27. Fernald GW: In vitro response to human lymphocytes
to Mycoplasma pneumoniae. Infect Immunol 5:552558, 1972
28. Opelz G, Terasaki PI, Hirata AA: Suppression of lymphocyte transformation by aspirin. Lancet 2:478480.
1973
29. Nowell PC: Inhibition of human leukocyte mitosis by
prednisalone in vitro. Cancer Res 21:1518-1521, 1961
30. Chandler RW, Robinson H, Masi AT: Serological investigations for evidence of an infectious etiology of
rheumatoid arthritis. Ann Rheum Dis 30:274-278, 1971
31. Marcolongo R, Carcassi A, Bianco G, et al: Isolation
of mycoplasma and antibody studies from patients with
I
Документ
Категория
Без категории
Просмотров
0
Размер файла
516 Кб
Теги
etiology, arthritis, infectious, human, studies, rheumatoid
1/--страниц
Пожаловаться на содержимое документа