close

Вход

Забыли?

вход по аккаунту

?

Neutrophil chemotactic factors in synovial fluids of patients with lyme disease.

код для вставкиСкачать
770
BRIEF REPORT
NEUTROPHIL CHEMOTACTIC FACTORS IN SYNOVIAL FLUIDS OF
PATIENTS WITH LYME DISEASE
KOSTIS GEORGILIS, RICHARD NORING, ALLEN C. STEERE. and MARK S. KLEMPNER
We examined synovial fluid samples from 14
patients with Lyme arthritis for the presence of neutrophi1 chemotactic factors. Thirteen of the synovial fluids
stimulated chemotaxis of normal human neutrophils.
The chemotactic activity was heat-sensitive and was not
inhibited by antibody to C5a or antibody to interleukin-8, or by a competitive inhibitor of the chemotactic
peptide f-Met-Leu-Phe. A culture supernatant of Borrelia burgdorferi also contained neutrophil chemoattractants. Chromatography demonstrated that the chemoattractants in the synovial fluid samples were different
from those in the B burgdorferi culture supernatant.
One of the major chemotactic factors in Lyme disease
synovial fluid had a calculated molecular weight of
13,900. We conclude that a novel, host-derived
chemoattractant is present in the synovial fluid of patients with Lyme disease.
Lyme disease, or Lyme borreliosis, is a tickborne spirochetal infection that affects populations in
From the Division of Geographic Medicine and Infectious
Diseases and the Division of Rheumatology/lmmunology, Department of Medicine, Tufts University School of Medicine, New
England Medical Center Hospitals, Boston, Massachusetts.
Dr. Georgilis’ work was supported by a Whitaker Health
Sciences Fund award. Dr. Steere’s work was supported by NIH
grant AM-20358. Dr. Klempner’s work was supported by NIH grant
AI-16732.
Kostis Georgilis, MD, PhD: Division of Geographic Medicine and Infectious Diseases; Richard Noring, BS: Division of
Geographic Medicine and Infectious Diseases; Allen C. Steere, MD:
Division of Rheumatology/Immunology; Mark S . Klempner, MD:
Division of Geographic Medicine and Infectious Diseases.
Address reprint requests to Kostis Georgilis, MD, PhD,
Division of Geographic Medicine and Infectious Diseases, Tufts
University School of Medicine, New England Medical Center
Hospitals, 750 Washington Street, Box 67, Boston, MA 021 11.
Submitted for publication March 26, 1990; accepted in
revised form December 31, 1990.
Arthritis and Rheumatism, Vol. 34, No. 6 (June 1991)
many parts of the world (1). The most common sites of
involvement are the skin, joints, nervous system,
heart, and lymphatic system (1,2). The causative spirochete, Borrelia burgdorferi, probably spreads hematogenously to joints early in the illness, and migratory arthralgias may be noted during this stage. After
several months, some patients may experience brief
attacks of oligoarticular arthritis, which may become
chronic, sometimes with erosion of cartilage and bone
(1,2). The synovial lesion and immune reactants in
these patients are similar to those of the other chronic
inflammatory arthritides (3-5). As in rheumatoid arthritis, joint fluid white cell counts in Lyme arthritis
range from 500 cells/mm3 to 100,000 cells/mm3 and
consist primarily of polymorphonuclear neutrophils
(PMN) (2,3). Although B burgdorferi has been cultured from joint fluid and identified in synovial lesions
in a few cases (1,5), the scarcity of organisms in the
synovial lesions, the wide spectrum of joint involvement, and the slow response, if any, of chronic arthritis to antibiotic therapy suggest that host factors play a
major role in the pathogenesis of Lyme arthritis (5).
A paradox of the chronic inflammatory arthritides is that the synovial lesion contains a lymphoplasmacellular infiltrate, while the joint fluid consists primarily of PMN (3,4). This cell may participate in the
destruction of cartilage and bone by the release of acid
and neutral proteinases into the joint (6). A major
chemotactic factor for PMN is the activated fifth
component of complement, C5a, formed in serum by
its interaction with immune complexes (7). Since immune complexes and C5a can be detected in synovial
fluid of patients with chronic inflammatory arthritides,
it is thought that C5a is a major chemoattractant for
PMN in articular effusions in these diseases (7). How-
BRIEF REPORTS
ever, platelet-activating factor (PAF), leukotrienes,
such as leukotriene B, (LTB,), and certain cytokines
have also been shown to be chemoattractants for PMN
in the joint fluids of patients with rheumatoid arthritis
(6-9). In bacterial diseases, the principal chemoattractants for neutrophils are the N-formylated methionyl
peptides, such as f-Met-Leu-Phe (FMLP), which are
released by the dividing bacteria as products of protein
synthesis (7).
Since the cause of Lyme arthritis is known, this
disease presents a special opportunity to determine the
chemotactic stimuli for PMN in this form of inflammatory arthritis. We report here that synovial fluids from
patients with Lyme arthritis have marked chemotactic
activity for neutrophils, but at least one of the principal chemoattractants is a small molecular weight protein, which is not C5a, LTB,, PAF, interleukin-8
(IL-8), or a formylated methionyl peptide.
PATIENTS AND METHODS
Patients. For this study we selected synovial
fluids from 14 patients with Lyme arthritis who were
representative of the range of joint inflammation observed in this disease. All of the patients had swollen
and painful knees; they had elevated antibody responses to B burgdovferi, as determined by enzymelinked immunosorbent assay ( I ) , and their joint fluid
white cell counts ranged from 2,000 cells/mm3 to
110,000 cells/mm3. All samples were centrifuged in a
microfuge for 5 minutes, and the cell-free supernatants
were stored at -70°C until used.
Chemotaxis of PMN. The isolation of PMN and
the chemotaxis assay were performed as previously
described (10). Heparinized peripheral venous blood
(20 units of heparin per ml of blood) was obtained from
healthy adult volunteers. Leukocytes containing 9598% PMN were isolated by Hypaque-Ficoll density
gradient centrifugation and dextran sedimentation, followed by 2 cycles of hypotonic lysis of the residual
erythrocytes. The cells were suspended in Hanks'
balanced salt solution (HBSS), pH 7.4. The bottom
wells of blind-well chambers were filled with 250 pl of
chemoattractant in HBSS, and the top wells were filled
with 600 p1 of HBSS containing 2% bovine serum
albumin (Sigma, St. Louis, MO) and 2.5 x lo6 PMN/ml.
Positive controls were the chemotactic peptide
FMLP (Sigma), the arachidonate metabolite L'TB,
(Sigma), zymosan-activated serum (ZAS) prepared
according to a standard protocol and used as the
77 1
source of C5a (1 I ) , and IL-8 provided by Dr. C. A.
Dinarello (Tufts-New England Medical Center, Boston, MA). To inhibit IL-8 activity (where indicated),
we used antiserum to IL-8, which was also provided
by Dr. C. A. Dinarello, and boc-Phe-Leu-Phe-Leu-Phe
(BPLPLP; Sigma) was used as a competitive inhibitor
of FMLP.
The 2 wells of the chamber were separated by a
cellulose nitrate filter with 3-pm pores (Sartorius SM
1 1302; Vangard International, Neptune, NJ). The
chambers were incubated at 37°C in humidified air
with 5% CO, for 1 hour. The filters were removed and
stained, and chemotaxis was determined by microscopy as the mean distance (in pm) that neutrophils
migrated in each of 5 randomly selected fields for each
filter. Each condition was run in triplicate, and the
mean migration was calculated. Results are expressed
as chemotactic index, i.e., the percentage of increase
of stimulated migration over random migration, given
by the formula [(stimulated migration/random migration - I)] x 100.
In preliminary experiments, it was shown that
synovial fluids at dilutions of 10% and 20% in HBSS
showed chemotactic activity comparable to that of
FMLP (1OP8M).Synovial fluids at dilutions of 0.1%
and 1% were not active.
Immunoadsorption. Protein A-Sepharose CL4B beads (Sigma) were coupled to an anti-CS (C5a)
goat antibody provided by Dr. J. A. Gelfand (TuftsNew England Medical Center). Immunoadsorption of
synovial fluids, buffer (HBSS), FMLP, or ZAS was
carried out in Eppendorf microtubes at room temperature for 1 hour. The tubes were then centrifuged in a
microfuge for 3 minutes, and the supernatants were
carefully aspirated and assayed for chemotactic activity as described above.
Culture supernatant of Borrelia burgdorferi.
Strain G39/40 of B burgdorfevi was cultured at 32°C in
modified Barbour-Stoenner-Kelly medium (12). To
obtain cell-free supernatant, the spirochete cultures
were centrifuged at 12,OOOg at 15°C for 20 minutes.
The supernatant was then centrifuged for another 20
minutes at 20,OOOg and 15°C. After the second centrifugation, no sediment was present, and all but I ml of
the culture medium at the bottom of each tube was
aspirated and stored at -70°C until used.
Chromatography. Synovial fluid (pooled from
patients F, G , and L) and B burgdorfevi culture supernatants were chromatographed in a 30 x 0.7-cm column (Bio-Rad, Richmond, CA) packed with Sephacryl
S-200 (Pharmacia, Piscataway, NJ). For each fraction-
BRIEF REPORTS
772
120
100
Y
4
80
-
60
..-u
0
c
i
C
40
20
.
0
FMLPA
B
C
D
E
F
G
H
I
J
0
.
K
.
L
.
M
.
N
,
Lyme disease synovial fluid
Figure 1. Chemotactic activity for polymorphonuclear neutrophils
of cell-free synovial fluids (20%) from 14 patients with Lyme
disease, as compared with that of FMLP (lO-'M).
ation, 30 mg of protein was used, and 25 fractions of 1
ml were collected using HBSS as the elution buffer.
The fractions were assayed for chemotactic activity as
described above and for protein concentration. For
determination of the molecular weight of a major
chemoattractant in Lyme disease synovial fluid, a 100
x 1-cm column was used, and 71 fractions of 1 ml were
collected and assayed for chemotactic activity. Fractions 5&55, containing the peak chemotactic activity,
were pooled, reapplied to the column, and 68 fractions
of 1 ml were collected and assayed for chemotactic
activity. The column was calibrated using gel filtration
molecular weight standards (Bio-Rad).
Chloroform-methanol extraction. To separate
proteins from lipids in Lyme disease synovial fluids, 2
volumes of ch1oroform:methanol 2: 1 (volume/volume)
and 1 volume of cell-free synovial fluid (pooled Lyme
synovial fluid from patients I, J, and K) at 20% in
HBSS were added to a glass tube and mixed manually
at 26°C for 5 minutes. The tube was then centrifuged at
500g at 26°C for 10 minutes. Separated aqueous and
organic phases were collected, and the organic phase
was dried under N, and resuspended in HBSS by
water-bath sonication.
RESULTS
Chemotactic activity of synovial fluids. Of the
synovial fluids from 14 patients with Lyme arthritis,
screened at 20% dilution (Figure I), all but 1 showed
chemotactic activity for PMN. One sample was as
active and 10 were more active than the chemotactic
peptide FMLP (lO-'M). The chemotactic activity did
not correlate with the absolute neutrophil count in the
synovial fluids (r = 0.04, P not significant). In addition,
none of these fluids activated the neutrophil respiratory burst, as assessed by superoxide production (data
not shown).
Chemotactic factors. To determine whether the
chemotactic activity in Lyme disease synovial fluids
could be attributed to C5a, 3 of the active Lyme
disease synovial fluids (J, K, and L), at a 10% dilution,
were immunoadsorbed with antibody to C5a. As a
positive control, ZAS (0.5%) and, as a negative control, the chemotactic peptide FMLP (lO-'M) were also
immunoadsorbed with the same antibody. As depicted
in Figure 2A? immunoadso~tionwith anti-C5a abolished the chemotactic activity of ZAS, which contains
large quantities of C5a ( I I ) , but did not appreciably
alter the chemotactic activity of FMLP or of the Lyme
disease synovial fluids.
TO determine whether heat-stable substances
were principal chemotactic factors in the Lyme disease synovial fluids, we examined the stability of the
chemotactic activity in the synovial fluids after heating. Lyme disease synovial fluid (10%) was compared
with the heat-stable chemoattractants LTB4 and IL-8.
As illustrated in Figure 2B, heating had no effect on
the bioactivity of LTB, ( 10-6A4) or IL-8 (10 ng/ml), but
abolished the chemotactic activity of the Lyme disease
synovial fluid samples.
To confirm that the chemotactic activity of
Lyme disease synovial fluids could not be ascribed to
the presence of 1L-8, we examined the effect of rabbit
antiserum against IL-8 on the chemotactic activity of
these fluids. Figure 2C shows that while antiserum
(150) inhibited the chemotactic activity of IL-8 (10
ng/ml), it had no effect on the chemotactic activity of
FMLP (lO-'M) or of Lyme disease synovial fluid
samples (10%). This suggests that IL-8 is not the major
chemoattractant in these arthritic fluids.
The tripeptide FMLP stimulates neutrophil
chemotaxis by binding to a specific receptor on the cell
surface (7). If the chemotactic activity in the Lyme
disease synovial fluids results from activation of the
neutrophil receptor for FMLP, then a competitive
inhibitor of the chemotactic peptide should block the
activity of the synovial fluids. Figure 2D shows that
BPLPLP (lO-,M), a competitive antagonist of FMLP,
blocked the migration induced by FMLP (IO-'M) but
had no effect on the chemotactic activity of Lyme
BRIEF REPORTS
773
H
x
-
0
[A1
control
immunoadsorption
LT
T
L
01
D
S
e3
control
heat treatment
-
0 40
c
0
m
40
w
E 20
2
20
0
0
K
J
FMLP
L
LTB 4
IL-8
Lyme synovial fluid
1
0
0
control
IL-8 antiserum
control
BPLPLP
[Dl
X
a,
I
D
.r 60
.-0
CI
0
40
0
5
c
0 20
0
FMLP
IL-8
I
J
FMLP
Lyme synovial fluid
Figure 2. Chemotactic factors in Lyme disease synovial fluid. A, Effect of immunoadsorption with antibody to C5a on the
chemotactic activity for polymorphonuclear neutrophils (PMN) of cell-free synovial fluids (10%) from 3 patients with Lyme
disease (J, K , and L), as compared with that of FMLP (10 -8M)and zymosan-activated serum (ZAS) (0.5%). Values are the mean
and SEM of 2-4 experiments. B, Effect of heat treatment on the chemotactic activity for PMN of cell-free pooled synovial fluid
(10%) from 3 patients with Lyme disease (F, G, and L), as compared with that of the known heat-stable chemoattractants
leukotriene B, (LTB,) (10-6M) and interleukin-8 (IL-8) (10 ng/ml). Values are the mean and SEM of 3 experiments. C, Effect of
IL-8 antiserum (150) on the chemotactic activity for PMN of cell-free synovial fluid (10%) from 2 patients with Lyme disease (I
and J), as compared with that of IL-8 (10 ng/ml) and FMLP (10-8M). Values are the mean and SEM of 2-3 experiments. D, Effect
of boc-Phe-Leu-Phe-Leu-Phe (BPLPLP) ( lO-,M), a competitive antagonist of FMLP, on the chemotactic activity for PMN of
cell-free pooled synovial fluid (10%) from 3 patients with Lyme disease (F, G , and L), as compared with that of FMLP (lO-'M).
Values are the mean and SEM of 2 experiments.
disease synovial fluid (10%). This suggests that the
chemoattractants in these synovial fluids do not bind
to the FMLP receptor on human neutrophils.
Chromatography. To further clarify whether the
chemotactic factors in the Lyme disease synovial
fluids were host-derived o r microorganism-derived,
we compared the chemotactic activity of fractions of
synovial fluid to that of fractions of a spirochete-free
culture supernatant of B burgdorferi. The results,
shown in Figure 3, demonstrated that the synovial
fluid chemotactic activity did not completely coincide
with that in the culture supernatant. The chromatographic pattern suggested that the synovial fluid contained some of the chemoattractants present in the
culture supernatant as well as other chemoattractants.
Furthermore, the chemotactic activity in the synovial
fluid was associated mostly with protein (Figure 3A),
whereas the activity in the culture supernatant was
contained in small molecular weight protein-deficient
fractions (Figure 3B). After extraction of the lipids
from Lyme disease synovial fluids by chloroformmethanol partitioning, more than 96% of the chemo-
BRIEF REPORTS
774
-
too>
5
-40
- -- - ~- -m - tr .m - m- r h" m" "m
o
-
~
m
t
I
fraction
Figure 3. Chemotactic (chemot.) activity of chromatographed cellfree pooled synovial fluid from 3 patients with Lyme disease (F, G,
and L) (A) and of chromatographed Borvelia burgdorferi culture
supernatant (B). For both fractionations, 30 mg of protein was used.
The corresponding protein concentrations (conc.) of the fractions
are also presented. The chemotactic activity is compared with that
of buffer (Hanks' balanced salt solution [HBSS]) and FMLP (lO-'M).
tactic activity was recovered in the aqueous phase,
and no activity was present in the organic phase (data
not shown). Therefore, the major chemoattractant in
Lyme disease synovial fluid is water-soluble. By sieving chromatography, the molecular weight of a major
chemoattractant in Lyme disease synovial fluids was
found to be approximately 13,900. This size corresponds to none of the previously described chemotactic factors for neutrophils.
DISCUSSION
In this report, we present evidence that cell-free
synovial fluids from patients with Lyme arthritis contain chemotactic activity for neutrophils. The Lyme
disease spirochetes have been shown to induce superoxide production by human PMN (13). Our cell-free
Lyme disease synovial fluid samples failed to stimulate
the neutrophil respiratory burst, which suggests that
the chemotactic factors in these fluids are not stimuli
m
that activate all PMN functions. Antigens extracted
from Borrelia burgdorferi have been previously tested
for their chemotactic effect on PMN, but this was
found to be very low compared with that of FMLP
(14). The chemotactic activity could not be inhibited
by boc-Phe-Leu-Phe-Leu-Phe, which competes for the
chemotactic peptide f-Met-Leu-Phe receptor on neutrophils (Figure 2D). This suggests that the chemotactic factors are not N-formylated methionyl peptides,
which are released by bacteria as products of protein
synthesis (7). In addition, chromatographic analysis
(Figure 3) demonstrated that the chemotactic activity
in the synovial fluid overlaps, but does not coincide,
with that in the B burgdorferi culture supernatant,
suggesting that the synovial fluid contained additional
chemoattractants.
The major chemotactic factor for PMN that is
formed in serum by its interaction with immune complexes is the activated fifth component of complement,
CSa, and this chemoattractant accounts for part of the
chemotactic activity in articular effusions of patients
with rheumatoid arthritis (7). Unlike the ZAS control,
the chemotactic activity of the Lyme disease synovial
fluids was not imrnunoadsorbed by an antibody to CSa
(Figure 2A), even though we used the lowest dilution
of Lyme disease synovial fluids that was chemotactic
(10%). Therefore, the chemotactic activity in these
fluids cannot be ascribed entirely to CSa.
Leukotrienes and platelet-activating factor
have been shown to be responsible for chemotactic
activity for PMN in rheumatic disease joint fluids (7,s).
However, it is unlikely that these factors have a major
chemotactic role in Lyme arthritis, since the chemoattractants in Lyme disease synovial fluids are heatlabile (Figure 2B), in contrast to lipid factors, such as
LTB, and PAF, which are heat-stable. Chromatographic analysis showed that the host-derived chemoattractants in Lyme arthritis joint fluids are most likely
proteins (Figure 3A). Furthermore, no chemotactic
activity was present in the organic phase after lipid
extraction from Lyme disease synovial fluids, whereas
more than 96% of the chemotactic activity could be
recovered in the aqueous phase. This indicates that the
major chemotactic factor in these fluids is watersoluble and provides additional evidence for its protein
nature.
A possible candidate for the major chemoattractant in Lyme disease joint fluids would be the
cytokine interleukin-8, also termed neutrophil-activating peptide, which is chemotactic for PMN and has
been detected in tissues of patients with inflammatory
BRIEF REPORTS
diseases (15). However, IL-8 is heat-stable (Figure
2B), and specific antiserum against IL-8 did not neutralize the chemotactic activity of the synovial fluids,
but did eliminate the chernotactic activity of IL-8
(Figure 2C).
We conclude that a novel host-derived chemoattractant is present in the synovial fluids of patients
with Lyme disease. Chromatographic analysis showed
that the peak chemotactic activity in the fluids corresponds to a molecular weight of approximately 13,900.
To our knowledge, none of the known chemotactic
factors for neutrophils has a similar molecular weight.
More precise definition of this major chemotactic
factor in Lyrne disease synovial fluids will require
additional studies.
Acknowledgments. We gratefully acknowledge the
helpful contributions of Bari Sue Brodsky and Bruce Reinhardt. We also thank Dr. C. A. Dinarello for providing IL-8
and antiserum to IL-8 and Dr. J. A. Gelfand for providing
antibody to C5a.
REFERENCES
1. Steere AC: Lyme disease. N Engl J Med 321:586-596,
1989
2. Steere AC, Malawista S E , Hardin JA, Ruddy S, Askenase PW, Andiman WA: Erythema chronicum migrans
and Lyme arthritis: the enlarging clinical spectrum. Ann
Intern Med 86:685-698, 1977
3. Steere AC, Gibofsky A, Patarroyo ME, Winchester RJ,
Hardin J A , Malawista SE: Chronic Lyme arthritis:
clinical and immunogenic differentiation from rheumatoid arthritis. Ann Intern Med 90:896-901, 1979
4. Steere AC, Duray PH, Butcher EC: Spirochetal antigens
and lymphoid cell surface markers in Lyme synovitis:
comparison with rheumatoid synovium and tonsillar
lyumphoid tissue. Arthritis Rheum 3 1 :487-495, 1988
775
5 . Sleere AC: The pathogenesis of Lyme disease: implications for rheumatic disease. Ann N Y Acad Sci 539:8792, 1988
6. Harris E D Jr: Pathogenesis of rheumatoid arthritis: a
disorder associated with dysfunctional immunoregulation, Inflammation: Basic Principles and Clinical Correlates. Edited by JI Gallin, IM Goldstein, R Snyderman.
New York, Raven Press, 1988
7. Snyderman R: Mechanisms of inflammation and leukocyte chemotaxis in the rheumatic diseases. Med Clin
North Am 70:217-235, 1986
8. Klickstein LB, Shapleigh C , Goetzl EJ: Lipoxygenation
of arachidonic acid as a source of polymorphonuclear
leukocyte chemotactic factors in synovial fluid and
tissue in rheumatoid arthritis and spondyloarthritis. J
Clin Invest 66: 1166-1 170, 1980
9. Ziff M: Role of cytokines in rheumatoid synovitis. Scand
J Rheumatol [Suppl] 76:211-215. 1988
10. Georgilis K. Schaefer C. Dinarello CA, Klempner MS:
Human recombinant interleukin Ip has no effect on
intracellular calcium o r on functional responses of human neutrophils. J Immunol 138:3403-3407, 1987
I I . Hammerschmidt D E , Bowers TK, Lammi-Keefe CJ,
Jacob HS, Craddock PR: Granulocyte aggregometry: a
sensitive technique for the detection of C5a and complement activation. Blood 55:898-902, 1980
12. Barbour AG: Isolation and cultivation of Lyme disease
spirochetes. Yale J Biol Med 57:521-525, 1984
13. Szczepanski A. Fleit HB: Interaction between Borrelia
burgdorferi and polymorphonuclear leukocytes: phagocytosis and the induction of the respiratory burst. Ann
NY Acad Sci 539:425428, 1988
14. Benach JL. Coleman J L , Garcia-Monco JC, Deponte
PC: Biological activity of Borrelia burgdorferi antigens.
Ann NY Acad Sci 539: 115-125, 1988
15. Baggiolini M , Walz A, Kunkel SL: Neutrophil-actvating
peptide- liinterleukin-8. a novel cytokine that activates
neutrophils. J Clin Invest 84: 1045-1049, 1989
Документ
Категория
Без категории
Просмотров
2
Размер файла
577 Кб
Теги
factors, neutrophils, patients, lyme, chemotactic, disease, synovial, fluid
1/--страниц
Пожаловаться на содержимое документа