close

Вход

Забыли?

вход по аккаунту

?

Serum antibody levels against t mycoplasmas in two north american indian populations predisposed to spondylitis.

код для вставкиСкачать
1328
SERUM ANTIBODY LEVELS AGAINST
T MYCOPLASMAS IN TWO NORTH
AMERICAN INDIAN POPULATIONS
PREDISPOSED TO SPONDYLITIS
DENYS K . FORD and ELIZABETH HENDERSON
Serum antibody levels against T mycoplasmas
(Ureaplasma urealyticum) were determined by the metabolic inhibition method in several populations. A higher
prevalence o f antibody was found in Haida Indians and
Bella Coola Indians than in blood donors, patients with
rheumatoid arthritis, and patients attending a VD clinic.
Antibody levels did not correlate with the presence of
spondylitis or the histocompatibility antigen HLA-B27,
although both these Indian populations have a high prevalence of spondylitis and HLA-B27.
The metabolic inhibition technique (M IT) of
Purcell et a1 (1 ) was the first practical method for detecting antigenic differences between strains of T mycoplasmas. Initial studies showed multiple antigenic serotypes (2,3), but no satisfactory evidence was found of
significant serum antibody elevations in response to T
mycoplasma “infection” (4). Subsequent reports have
defined eight antigenically distinct subtypes of T mycoplasmas ( 5 ) , one being Shepard’s type organism (6) and
From the Division of Kheumatology. Department of Medicine, University of British Columbia, Vancouver, British Columbia,
Canada.
Supported by the Canadian Arthritis and Rheumatism
Society and by Federal Health Grant 610-1055-28.
Denys K . Ford, M.D.: Professor of Medicine: Elizabeth Henderson. B.Sc.: Research Assistant.
Address reprint requests to Denys K . Ford. M.D.. Division
of Rheumatology. Department of Medicine. 700 West 10th Avenue,
Vancouver, British Columbia, Canada V5Z I M9.
Submitted for publication December 17. 1975; accepted May
20. 1976.
Arthritis and Rheumatism, Vol. 19, No. 6 (November-December 1976)
the other seven being serotypes initially characterized in
the present authors’ laboratory and previously reported
(3). Recently the test of Purcell was modified to prevent
the inactivation of complement by NH: ions produced
from the hydrolysis of urea (73). In this laboratory, this
modification was evaluated but was not found to alter
significantly the sensitivity of the test: the original test of
Purcell and the modification yielded equivalent data.
Because the Purcell method was simpler to perform, it
was chosen for the following study.
This investigation arose because a pilot survey of
stored frozen sera revealed antibody in a sample from a
Haida Indian woman with spondylitis. Immediate follow-up of this observation showed that her son, also a
spondylitic, had a significant serum antibody titer, but 5
other members of the family without spondylitis had no
demonstrable antibody. This family had been the source
of a special study in 1963 (9) when polyarthritis was
observed as a sequel to a flu-like illness. Serum samples
from over 400 Haida Indians were available for further
investigation, having been collected in 1962 during an
epidemiologic study of the Haidas in the Queen Charlotte Islands of British Columbia. At that time it was
found that the prevalence of rheumatoid arthritis among
the Haidas was not distinctive, but that the prevalence
of spondylitis was unexpectedly high (10). A subsequent
epidemiologic study of Bella Coola Indians also showed
an increased prevalence of spondylitis ( I 1).
In 1973 a pilot study of a random sample of these
stored sera indicated that the Haida Indian samples had
1329
a higher prevalence of T mycoplasma antibody than did
samples from Red Cross donors. Consequently, when
visits were made t o Bella Coola in 1974 a n d the Queen
Charlotte Islands in 1975 for H L A histocompatibility
typing of the Bella Coola and Haida Indians respectively
(12), serum samples were taken for further T mycoplasma antibody studies a n d the following investigation
was subsequently completed.
MATERIALS AND METHODS
Serum Samples. Haida Indian Serum samples were
collected at the time of HLA typing in the spring of 1975. Bella
Coola Indian serum samples were gathered at the time of HLA
typing in May 1974. The Bella Coola Indians live about 200
miles north of Vancouver, whereas the Haidas live about 400
miles north of Vancouver. Both serum collections were designed to yield a random sample of each Indian population, as
described in the report of the HLA-B27 study (12).
Serum samples were also taken from Red Cross donors attending Blood Donor Clinics in Vancouver and from
patients attending the VD Control Clinic with a variety of
diagnoses. Patients with rheumatoid arthritis, ankylosing
spondylitis, and Reiter's syndrome were under the care of the
rheumatologists of the Division of Rheumatology of the University of British Columbia Medical School at the Arthritis
Centre in Vancouver. Sera from patients with Reiter's syndrome and also from the VD Clinic patients were collected
when the patients were not on antimycoplasmal antibiotics.
All the sera were stored at -20°C before testing.
Purification of IgG Antibody. The IgG fraction of
serum was prepared by the DEAE-Sephadex method of Webb
(13). The purity of the fractions was confirmed by immunoelectrophoresis, which demonstrated single immunoprecipitin
IgG lines in each of the fractions. Quantitative radial immunodiffusion determined the IgG concentrations in the fractions,
and the concentration of IgG was adjusted to approximate
that in the original serum.
HLA Histocompatibility Testing. Tests for H L A typing
were performed by the Vancouver General Hospital tissuetyping laboratory. The Terasaki microlymphocytotoxicity
procedure was employed, with antisera provided by Dr. Terasaki.
T Mycoplasma Serotypes and Media. All of the eight
serotypes of Black ( 5 ) were employed in the pilot phases of the
study. Seven of these were originally isolated in this laboratory, and the remaining strain was the type organism isolated
by Shepard et a1 (6). I n the final testing, serotype V was
selected because this organism had been the main organism
studied in this laboratory during the past 14 years. The culture
medium consisted of Difco PPLO medium, supplemented with
10% unheated horse serum, I%Oxoid yeast. urea at a concentration o f 0.05% for growth and I % for MIT, 0.002% phenol
red, and 1000 units/ml of penicillin.
Metabolic Inhibition Test. Antibody levels were measured by the Microtiter method of Purcell, with 50 color change
units (CCU) of organisms per well. Each well finally contained
0.05 ml of serum serially diluted twofold in 1% urea mycoplasma medium, 0.05 ml of T mycoplasma culture containing
50 CCU in the same medium, and 0.05 ml of I:20 guinea pig
complement (Microbiological Associates), also in the same
medium. The serial serum dilutions and pipetting were performed in the Cooke Minipipetter and Minidiluter. All tests
were done in duplicate.
The disposable Linbro plates were sealed with Microtiter Plate Sealers and incubated overnight at 32°C and at
37°C on the following day, until the control wells showed a pH
rise to 7.5 by comparison with standards. At t h a t time the
titers were read as the highest dilution of serum added to the
test wells that showed an inhibition of pH rise.
A modified Lin and Kass procedure (7,8) was developed for comparison with the Purcell method. The T mycoplasmas, in a urea-free medium containing dialyzed horse
serum, were exposed to human serum and 1:20 guinea pig
complement for 1 hour at 37°C before the addition of ureacontaining medium. In this laboratory the 10 CCU inoculum
suggested by Lin and Kass was found to give erratic results,
and therefore a 50 CCU inoculum was used.
RESULTS
Preliminary studies were required to define the
M I T before proceeding to the main investigation. As
noted previously, an initial comparison was made between a modified Lin and Kass method a n d that of
Purcell, but no significant difference in sensitivity was
found between the two techniques in this laboratory.
Table I shows the comparison when 3 human sera and 3
hyperimmune rabbit antisera were tested against 3 serotypes. This table demonstrates the low levels of inhibition given by human sera in contrast to hyperimmune
rabbit antisera; it also shows the greater serotype specificity of t h e rabbit antisera in contrast to a broader
cross-reactivity of the inhibition by human serum.
In order t o establish that the metabolic inhibition
test with human serum was a measure of antibody and
not a manifestation of nonspecific inhibition, the IgG
fraction of serum was prepared by the Webb procedure
(13) from the sera of 9 subjects, selected because they
had high, moderate, or negligible inhibitory titers. The
comparison of titers between the original serum and the
IgG fraction, at a concentration equivalent t o that in the
serum, is shown in Table 2. N o loss of inhibition resulted from the IgG fractionation procedure.
Table 2 also confirms the observation noted in
Table 1 that the inhibitory factor in human serum was
broadly reactibe against the T mycoplasma serotypes.
Because of this broad reactivity of human antibody as
contrasted t o rabbit antibody, a n d because the pilot
study had shown no more significant antibody levels
against any other serotype, it was decided t o test all sera
against a single serotype, V.
In the MIT tests on serum samples from the
FORD A N D HENDERSON
I330
Table 1. Comparison of Purcell's Standard MIT Method with a Modification
of
the Lin and Kass Method
Inhibitory Titers (Reciprocal) Against Specific Serotypes
Purcell Method
Modified Lin and Kass Method
Human
T-Ill
T-V
T-VII
T-111
2
4
8
16
16
32
4
32
8
4
8
64
32
64
64
4
16
8
10.240
80
160
40
5,120
. 160
20
20
2,560
10,240
20
80
20
10,240
80
20
20
5,120
Sera
LL
HL
DT
T-V
T-VII
Antisera
Rabbit
111
V
VII
population groups shown in Table 3, about 20 serum
samples were tested at one time and sera from several
population groups were included in each testing. The
sera were unselected except for two factors. First, individuals in the control groups were selected to achieve a
male predominance similar to the spondylitic and Reiter's groups. Second, the individuals i n the con,trol
groups were selected to achieve an average age similar to
the Haida, spondylitic, and Reiter's groups. Columns 3
and 4 of Table 3 demonstrate the similarity of the
groups in regard to sex and age. Red Cross donors in
Vancouver were chosen as controls to compare to the
Indian population, patients attending the Vancouver
VD Control Clinic were selected for comparison with
the patients having Reiter's syndrome, and the rheumatoid arthritics were chosen for comparison with the nonHaida spondylitics. In view of the resulting low titers,
Table 3 shows the percentage of sera that gave inhibition
at each of the twofold serum dilutions up to 1:16. The
data in columns 6-9 of Table 3 show uniformity of the
comparisons between the population categories, whichever inhibitory titer is considered, with the single exception that the number of 1:16 titers in the rheumatoid
arthritic groups is unexpectedly high.
Table 2. Comparison of MIT Titers Between Serum and Derived IgG Fractions Againsr Seven Serotypes
Inhibitory Titers (Reciprocal) Against Specific Serotypes
Patient
Diagnosis
RA
RA
RS
RA
RA
RA
RS
HS
HS
Sample
Tested
IgG (mg%)
T-I
T-I1
T-111
T-IV
I360
I080
1 I30
750
1360
I I70
1240
900
I890
I360
1 I70
I090
I550
1961
2320
4520
I700
1250
0
0
0
0
0
0
0
0
4
I
8
2
32
16
32
32
ND
ND
0
0
0
0
0
0
0
0
0
0
2
0
0
0
16
8
8
2
16
4
16
32
16
32
8
8
1
1
0
0
2
0
0
4
8
8
I
16
16
4
8
32
32
0
8
0
8
2
4
8
16
16
ND
ND
T-V
0
0
1
0
4
4
0
0
8
4
32
32
16
16
32
128
64
64
T-VI
0
0
0
0
0
0
0
0
2
0
8
4
4
4
8
4
16
64
RA: rheumatoid arthritis: RS: Reiter's syndrome: HS: Haida spondylitic; N D : not done.
T-VII
0
0
0
0
I
1
0
0
8
2
8
4
32
64
16
32
16
16
1331
SERUM ANTIBODY LEVELS
Table 3. Serum Antibody Titers to T Mycoplasma Serotype V
~
Population
Nonspondylitic
Haidas
Spondylitic Haidas
Bella Coola Indians
Reiter’s syndrome
patients
Non-Haida
spondylitics
Rheumatoid
arthritics
Red Cross donors
VD ClinicControls
<Age 3 I
>Age31
Total
Total
No.
Percent
Male
Average
Age
Percent
with
0 Titer
I08
88
42
41.6
58.4
46.4
30.5
10.2
22
106
40
86
58.5
48
86.4
66.0
75.0
59.1
48.0
45.0
41.0
35.0
25.0
13.6
11.6
36
18.0
34.0
25.0
43
34.0
66.0
37.5
28.1
15.6
53
69.4
30.0
20.4
18.0
10.0
37
65.8
34.2
21.2
13.8
4.6
25
39
31
59.0
59.3
59.0
41.2
40.7
41.0
29.5
30.5
29.9
14.2
15.5
14.6
2.6
5. I
3.7
44
100
32
87.5
50
I00
108
68.5
78
59
137
I00
I00
I00
Table 4 presents the statistical analysis of the
Percent Positiveat Titer
I :2
1:4
1:8
1.16
15.0
whom HLA-B27 status was determined, and again there
was no definite correlation.
2 1.4 serum dilution data from which column 7 of Table
3 was derived. The Haida and Bella Coola Indians are
significantly difyerent from the Red Cross donors, the
VD Control Clinic patients, and the group with rheumatoid arthritis. The group with Reiter’s syndrome is significantly different from the Red Cross donors and the
rheumatoid arthritics, but not from the VD Clinic controls. The non-Haida spondylitics are only different
from the Red Cross donors at a significance level of P <
0.05.
Because Haida Indians have eight to ten times
the prevalence of ankylosing spondylitis as compared to
populations studied elsewhere (12), and also because
they have eight to ten times the prevalence of the histocompatibility antigen HLA-B27 (12), it was of interest
to determine if there was a relationship between T
mycoplasma antibody titers and the presence of HLAB27. Table 5 shows that n o correlation existed. It also
includes data on 23 patients with Reiter’s syndrome in
DISCUSSION
An unexpected finding of T mycoplasma antibody in a Haida mother and her son initiated this
study, which was performed to discover if etiologic clues
to the cause of spondylitis would emerge. The data
derived from the investigation indicate that the Haida
and Bella Coola Indian populations have a high prevalence of low levels of antibody against T mycoplasmas,
when compared to Vancouver Red Cross donors,
patients attending the Vancouver VD Control Clinic, or
patients with rheumatoid arthritis. There is, unfortunately, no suggestion that this observation is related to
the high prevalence of spondylitis in these Indian populations. Moreover, the findings show that antibody levels were not correlated with HLA-B27 status. Consequently, these observations d o not now have clinical
Table 4. Statistical Analysis of Data in Column 7 of Table 3 (Titer 2 1:4/
Bella Coola
Indians
RedCrossdonors
VDcliniccontrols
R heumatoid
arthritics
Nonspondylitic
Haidas
Spondylitic
Haidas
Reiter’s
Syndrome Pts
Non-Haida
Spondylitics
X2
p
x2
p
x2
p
x2
P
x2
P
17.0
8.5
11.3
<0.01
15.0
6.8
10.1
<0.01
<0.01
<0.01
10.4
5.2
8.6
<0.01
8.1
3.2
6.5
<0.01
3.5
0.74
3.0
0.05
0.4
0.08
<0.01
<O.Ol
0.02
<0.01
0.08
0.01
1332
F O R D A N D HENDERSON
Table 5. HLA-27 Status of Individuals Compared
Serorype V Antibody Levels
lo
T Mycoplasnia
HLA-27 Status
106 Haidas
23 Reiter’s Cases
Antibody Titer
Pos
Neg
Pos
Neg
2 1.4
23
29
21
29
10
I
4
< 1:4
8
or immunologic relevance, and it is impossible to hypothesize profitably on the basis of the observations.
I n this laboratory a modified Lin and Kass
method was not found superior to the Purcell technique
for measuring T mycoplasma antibody, and thus
emphasis on complement lysis of organisms does not
seem important in the practical determination of antibody levels in human serum.
Previous studies employing hyperimmune rabbit
serum have differentiated distinct serotypes of T
mycoplasmas, and there is currently no explanation
of the broader cross-reactivity of human serum or pure
IgG fractions of human serum.
The significance of T mycoplasma antibody in
patients with Reiter’s syndrome is not clear at this time.
The difficulties in adequate follow-up of patients with
nongonococcal urethritis have prevented a satisfactory
study of antibody responses to urethritis associated with
T mycoplasmas, but pilot studies have indicated that
antibody responses are unpredictable and often not demonstrable. Currently, T mycoplasma serology does not
add significant support to the possibility of an etiologic
role for these organisms in Reiter’s syndrome, although
the gats provide no evidence against such an hypothesis.
ACKNOWLEDGMENTS
The authors are grateful to Ms. Judy Smith and Ms.
Beverley Wort for expert technical assistance in the initial
phases of the study. They qlso acknowledge the help of Dr. J.
P. Gofton in collecting the Haida and Bella Coola sera, and of
the Canadian Red Cross and Vancouver VD Control Clinic in
securing the control sera.
REFERENCES
I . Purcell RH, Taylor-Robinson D, Wong D, et al: Color
test for measurement of antibody to T-strain mycoplasmas. J Bacteriol 92:6-12, 1966
2. Purcell R H , Wong D, Channock RM, et al: Significance
of antibody to mycoplasma as measured by metabolicinhibition techniques. Ann NY Acad Sci 143:664-675.
I967
3. Ford DK: Relationships between mycoplasma and the
etiology of non-gonococcal urethritis and Reiter‘s syndrome. A n n NY Acad Sci 143501-504, 1967
4. Ford DK: Serological studies on T-strain mycoplasmas.
Arthritis Rheum 9503-504, 1966
5. Black FT: Modifications of the growth inhibition test and
its application to human T-mycoplasmas. Appl Microbiol
251528-533, 1973
6. Shepard MC, Lunceford CD, Ford DK, et al: Ureaplasma
urealyticum gen., sp.nov.: proposed nomenclature for the
human T (T-strain) mycoplasmas. Int J Syst Bacteriol
24: 160-17 I , 1974
7. Lin JS, Kass EM: Immune inactivation of T-strain mycoplasmas. J Infect Dis 122:93-95, 1970
8. Lin JS, Kendrick MI, Kass EH: Serologic typing of h u man genital T-mycoplasmas by a complement-dependent
mycoplasmacidal test. J Infect Dis 126:658-663, 1972
9. Price G E , Ford D K , Gofton JP, et al: An outbreak of
“infectious” polyarthritis in a Haida Indian family. Arthritis Rheum 6:633-638, 1963
10. Robinson HS, Gofton JP, Price GE: A study of rheumatic
disease in a Canadian Indian population. A n n Rheum Dis
221232-236, 1963
1 I . Gofton JP, Bennet PH, Smythe HA, et al: Sacroiliitis and
ankylosing spondylitis i n North American Indians. Ann
Rheum Dis 31:474-481, 1972
12. Gofton JP, Chalrners A, Price GE, et al: HLA-B27 and
ankylosing spondylitis in B.C. Indians. J Rheumatol
2~314-318, 1975
13. Webb AJ: A 30-minute preparative method for isolation
of IgG from human serum. Vox Sang 23:279-290, 1972
Документ
Категория
Без категории
Просмотров
4
Размер файла
379 Кб
Теги
two, population, north, level, predisposed, serum, indian, mycoplasma, american, antibody, spondylitis
1/--страниц
Пожаловаться на содержимое документа