close

Вход

Забыли?

вход по аккаунту

?

Persistent arthritis in Borrelia burgdorferiinfected HLADR4positive CD28-negative mice postantibiotic treatment.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 58, No. 12, December 2008, pp 3892–3901
DOI 10.1002/art.24028
© 2008, American College of Rheumatology
Persistent Arthritis in Borrelia burgdorferi–Infected
HLA–DR4–Positive CD28-Negative Mice
Post–Antibiotic Treatment
Bettina Panagiota Iliopoulou,1 Joseph Alroy,2 and Brigitte T. Huber1
Objective. The immunologic events that lead to
persistent joint inflammation in certain patients with
Lyme arthritis post–antibiotic treatment have been elusive so far. The prevalence of this condition is highest in
individuals with rheumatoid arthritis–associated
HLA–DR alleles. This study was undertaken to generate
a murine model with persistent arthritis post–antibiotic
treatment.
Methods. We have previously shown that CD28ⴚ/ⴚ
mice develop intermittent monarticular Lyme arthritis
that is responsive to antibiotics. Since there seems to be
a link in humans between persistent arthritic manifestations post–antibiotic treatment and the HLA–DR4
allele, we generated DR4ⴙ/ⴙCD28ⴚ/ⴚMHCIIⴚ/ⴚ mice, infected them with Borrelia burgdorferi, and subsequently
treated them with antibiotics.
Results. Thirty-eight percent of the B burgdorferi–
infected DR4ⴙ/ⴙCD28ⴚ/ⴚMHCIIⴚ/ⴚ mice, but none of
the B burgdorferi–infected CD28ⴚ/ⴚMHCIIⴚ/ⴚ mice, remained arthritic post–antibiotic treatment. A significant fraction (36%) of these mice, but none of the mice
in which arthritis resolved, had serum antibodies to
outer surface protein A of B burgdorferi. After abrogation of active B burgdorferi infection, the inflammatory
reaction in mice with persistent joint inflammation
was restricted to the joints, since their draining lymph
nodes were no longer enlarged. Increased CD20 and
interferon-␥ messenger RNA expression in the inflamed
joints of these mice suggested a possible role of B cells
and inflammatory cytokines in the pathogenesis of
persistent arthritis post–antibiotic treatment.
Conclusion. The establishment of this murine
model allows, for the first time, the elucidation of the
immunologic events that lead to persistent Lyme arthritis post–antibiotic therapy in genetically susceptible
individuals.
Lyme disease, caused by the tick-borne spirochete Borrelia burgdorferi, is the most common vectorborne illness in the US. After inoculation into the skin,
B burgdorferi quickly disperses in the mammalian host by
binding to components of the extracellular matrix (1).
Three clinical stages of Lyme disease have been described in humans. Early infection consists of localized
erythema migrans, followed within days or weeks by
disseminated infection that affects the nervous system,
heart, or joints, and subsequently by late or persistent
infection (2). While the spirochetes can be eliminated
from patients with Lyme disease by antibiotic treatment, chronic arthritis may persist, mainly in patients
with rheumatoid arthritis (RA)–associated HLA–DR
alleles, such as HLA–DRB1*0401 (DR4) and HLA–
DRB1*0101 (3,4).
Two basic hypotheses have been proposed to
explain this phenomenon: persistent infection and
infection-induced autoimmunity. The latter hypothesis
is supported by the fact that manifestations of arthritis
continue despite the absence of B burgdorferi DNA,
documented by polymerase chain reaction (PCR) analysis of the synovial fluid (5–7). Interestingly, in 70% of
the patients who continued to experience arthritis after
antibiotic treatment, an antibody response to outer
surface protein A (OspA) of B burgdorferi that seemed
to parallel the severity and duration of arthritis was
Supported by the NIH (grant R01-AR-45386). Tufts Medical
Center’s Gastroenterology Research on Absorptive and Secretory
Processes (GRASP) Center is funded by National Institute of Diabetes
and Digestive and Kidney Diseases, NIH (grant DK-34924).
1
Bettina Panagiota Iliopoulou, PhD, Brigitte T. Huber, PhD:
Tufts University School of Medicine, Boston, Massachusetts; 2Joseph
Alroy, DVM: Tufts University School of Medicine, and Tufts Medical
Center, Boston, Massachusetts.
Address correspondence and reprint requests to Brigitte T.
Huber, PhD, Department of Pathology, Tufts University School of
Medicine, Jaharis 512, 150 Harrison Avenue, Boston, MA 02111.
E-mail: brigitte.huber@tufts.edu.
Submitted for publication May 5, 2008; accepted in revised
form August 1, 2008.
3892
HLA–DR4 AND PERSISTENT ARTHRITIS AFTER B BURGDORFERI INFECTION
mounted during periods of maximal arthritis (8,9). In
addition, an anti-OspA Th1 response has been documented in the synovial fluid of these patients (10–12).
Murine systems have been developed to analyze
the immunologic events that occur upon B burgdorferi
infection in humans. It is well established that manifestations of arthritis depend on the administered dose of
B burgdorferi and the age and genetic background of the
mice (13–15). It has also been suggested that T cells,
more specifically the CD4⫹ Th1 subset, as well as the
proinflammatory cytokine interferon-␥ (IFN␥), are responsible for exacerbation of arthritis upon B burgdorferi
infection (16–20). Murine Lyme arthritis peaks within
the first 2 weeks after infection and then resolves spontaneously; very few individual mice continue to exhibit
chronic arthritis (21,22). These manifestations are reminiscent of the acute phase of Lyme arthritis in humans
(15). Detailed characterization of the chronic phase of
Lyme arthritis has been elusive thus far.
Through the study of the development of Lyme
arthritis in different inbred mouse strains, it has become
apparent that arthritis severity depends on a fine balance between proinflammatory factors and immunoregulatory mechanisms. We hypothesized that by interfering with this balance, such as in the CD28⫺/⫺ mouse,
we could increase the incidence of chronic Lyme arthritis. We have recently demonstrated that CD28⫺/⫺ mice,
but not wild-type C57BL/6J (B6) mice, develop chronic
Lyme arthritis upon B burgdorferi infection. The persistent episodes of arthritis observed in these mice
lasted for ⬎6 months, but were sensitive to antibiotic
treatment (23).
Since a prerequisite for the development of persistent arthritis post–antibiotic treatment in humans is
the presence of HLA–DR4 or related alleles, we introduced the CD28⫺/⫺ genotype onto the DR4⫹/⫹MHCII⫺/⫺
background. In the present study we showed that a
significant fraction of B burgdorferi–infected DR4⫹/
⫺/⫺
⫹CD28
MHCII⫺/⫺ mice continue to manifest arthritis
after antibiotic therapy. One-third of these mice, but
none of the mice in which arthritis had resolved, maintained an OspA antibody titer after antibiotic treatment.
Furthermore, we showed that CD20 and IFN␥ expression are increased in the joints of DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mice, which suggests that B cells and inflammatory cytokines may be involved in the perpetuation of
inflammation. This animal system will allow, for the first
time, the direct examination of the inflammatory events
that lead to persistent Lyme arthritis after antibiotic
treatment.
3893
MATERIALS AND METHODS
Mice. CD28⫺/⫺ mice (B6.129S2-Cd28tm1Mak/J) were
bred at the Tufts University Division of Laboratory Animal
Medicine from breeding pairs (stock no. 002666) obtained
from The Jackson Laboratory (Bar Harbor, ME) and confirmed to have been backcrossed for at least 10 generations to
the B6 background. DR4⫹/⫹MHCII⫺/⫺ transgenic mice were a
gift from T. Forsthuber (Case Western Reserve University,
Cleveland, OH) and were bred in our facility. These mice were
generated with HLA–DRA–IE␣ and HLA–DRB1*0401–IE␤
chimeric genes and were also on a B6 background (24).
DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice were generated by breeding
the CD28⫺/⫺ genotype onto DR4⫹/⫹MHCII⫺/⫺ mice. The F1
DR4⫹/⫺CD28⫹/⫺MHCII⫹/⫺ mice were backcrossed to the
CD28⫺/⫺ mice, and the DR4⫹/⫺CD28⫺/⫺MHCII⫹/⫺ mice were
selected by fluorescence-activated cell sorting (FACS) and
then intercrossed to generate DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺
mice. All offspring were screened by FACS analysis for surface
expression of DR4, CD28, and I-Ab molecules (all antibodies
were purchased from PharMingen, San Diego, CA). All animal
experiments were approved by the Institutional Animal Care
and Use Committee at Tufts Medical Center.
B burgdorferi. Low-passage (passage-2) infectious B burgdorferi N40 clone D10E9A1-E (a kind gift from Dr. Jenifer
Coburn, Medical College of Wisconsin, Milwaukee) (25,26)
was used for all infections. B burgdorferi were cultured in
complete BSK medium (Sigma, St. Louis, MO) at 34°C until
mid-log phase (5 ⫻ 107 B burgdorferi/ml) and were counted by
darkfield microscopy.
Development of a murine model of persistent Lyme
arthritis after antibiotic treatment. Sex-matched DR4⫹/⫹
CD28 ⫺/⫺ MHCII ⫺/⫺ , CD28 ⫺/⫺ , and DR4 ⫹/⫹ MHCII ⫺/⫺
mice (4–5 weeks old) were infected intradermally in the skin of
the femoral area of both hind limbs with a total dose of 2 ⫻ 104
B burgdorferi per mouse (1 ⫻ 104 B burgdorferi in 50 ␮l per hind
limb). This protocol was used for all of the infections. No
difference in arthritis development was observed between male
and female mice; therefore, both male and female mice were
included in the experiments. For each experiment, however, great
care was taken that all of the mice in the experimental group
(DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice) were sex- and age-matched
with those of the control groups (CD28⫺/⫺ mice and DR4⫹/
⫺/⫺
⫹MHCII
mice). Arthritis was assessed, based on edema
formation, 2–3 times per week by an observer (BPI) who was
blinded with regard to experimental group. The anteroposterior
tibiotarsal joint thickness was measured using gauge calipers
(Mahr Federal, Providence, RI). Upon establishment of chronic
arthritis (within ⬃3 months after B burgdorferi infection), ceftriaxone (50 mg/kg/dose once a day for 5 days) was administered
intraperitoneally. This treatment has been reported to be 100%
effective in eradicating B burgdorferi (27,28). Mice were monitored for the presence of arthritis over a period of 2–5 months
after antibiotic therapy.
Since we previously established that when mice on the
CD28⫺/⫺ background are not treated with antibiotics they
continue to experience intermittent episodes of monarticular
arthritis for at least 5 months (which is the longest they have
been tested) (23), we did not include a group of mice that were
not treated with antibiotics in this set of data. In addition, when
DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice were infected with B burg-
3894
dorferi, but not treated with antibiotics, they developed chronic
joint inflammation similar to that in CD28⫺/⫺ mice. At the end
of the experiment, mice were killed by CO2 asphyxiation.
Hearts and ankles were harvested and processed for histologic
analysis. We also observed the development of anti-OspA
antibody throughout the course of the experiment. Mice were
bled weekly, and serum was subjected to enzyme-linked immunosorbent assay (ELISA) for the presence of anti-OspA
antibodies. In order to assess the effectiveness of the antibiotic
therapy, DNA was extracted from ear punch tissue before
treatment with antibiotics and at the end of the experiment.
Experiments were repeated 4 times.
Histopathologic analysis. Ankles were harvested and
decalcified overnight in Decalcifier I solution (SurgiPath,
Richmond, IL). The hearts were cut in half through bisection
across the atria and ventricles, and kept in formalin. Both
ankles and hearts were processed for histologic analysis as
previously described (23).
Anti-OspA ELISA. Flat-bottomed Immulon 2HB
plates (Fisher Scientific, Pittsburgh, PA) were coated overnight with 5 ␮g/ml of recombinant OspA or OspA fragments in
binding buffer (0.1M Na2HPO4 [pH 9]). ELISA was performed
as previously described (23). The cutoff used to calculate the
end point titer was set as the lowest serum concentration, in
which the optical density unit signal was twice that of the
background wells (all reagents included, expect the serum) for
each isotype.
In vitro restimulation and interleukin-17 (IL-17)
ELISA. Popliteal as well as inguinal lymph node cells were
stimulated with anti-CD3 (145.2C11) at a 1:150 dilution for 48
hours in vitro, and IL-17 production was assayed by ELISA.
IL-17 ELISA was performed as previously described (23), with
the following modifications. Plates were coated overnight with
3 ␮g/ml of capture anti-mouse IL-17 antibody (R&D Systems,
Minneapolis, MN) in 1⫻ phosphate buffered saline (PBS).
Coated plates were blocked with 2% bovine serum albumin
and 5% sucrose in 1⫻ PBS at room temperature for 1 hour.
Recombinant mouse IL-17 (standard curve) and the supernatant from the in vitro restimulation were added in duplicate to
the ELISA plates and incubated for 45 minutes at 37°C. Plates
were washed and incubated with biotinylated anti-mouse IL-17
(R&D Systems) for 1 hour at 37°C, followed by another wash
and incubation with neutrAvidin–alkaline phosphatase for 30
minutes at room temperature. Plates were then developed as
previously described (23).
Determination of B burgdorferi burden. DNA was
extracted from ear punch tissue and the B burgdorferi burden was determined by real-time quantitative PCR, as previously described (23,29).
Real-time reverse transcriptase–PCR (RT-PCR) analysis of messenger RNA (mRNA) for IL-17, IFN␥, and CD20.
Mouse ankles were harvested and immediately frozen in liquid
nitrogen. Frozen tissue was pulverized using a mortar and
pestle precooled in liquid nitrogen. RNA from pulverized
ankles and popliteal as well as inguinal lymph node cells was
extracted using the RNeasy Mini kit, according to the recommendations of the manufacturer (Qiagen, Valencia, CA).
Complementary DNA (cDNA) synthesis and removal of
genomic DNA were performed using the QuantiTect reverse
transcription kit (Qiagen). The cDNA was then diluted 1:20
and was used as template in a 20-␮l reaction mixture contain-
ILIOPOULOU ET AL
ing primers and probes specific for IL-17 mRNA, IFN␥
mRNA, CD20 mRNA, and 18S. Primers and FAM-labeled
probes specific for IL-17 mRNA (Mm00439619-m1), IFN␥
mRNA (Mm00801778-m1), and CD20 mRNA (Mm00545909m1), as well as VIC-labeled probes specific for murine 18S,
were purchased from Applied Biosystems (Foster City, CA)
and used as loading control. The PCR was carried out using
iTaq Supermix with ROX (Bio-Rad, Hercules, CA) under the
same cycling parameters as described above. The amount of
template DNA was first normalized by the signal of the 18S
housekeeping gene.
Statistical analysis. Statistical analysis was performed
using GraphPad Prism software (GraphPad Software, San
Diego, CA). All data were tested for Gaussian distribution,
using the Shapiro-Wilk normality test. Quantitative differences
were assessed by Student’s 2-tailed t-test for comparisons of 2
groups, and by analysis of variance for comparisons of ⬎2
groups, for normally distributed data. For skewed data, the
Mann-Whitney test was used for comparisons of 2 groups, and
the Kruskal-Wallis test was used for comparisons of ⬎2
groups. Statistical differences in the proportion of mice with
persistent joint inflammation post–antibiotic treatment as well
as in the proportion of mice with anti-OspA antibodies were
also determined by Fisher’s 2-tailed exact probability test.
Correlation between 2 variables was assessed by calculating
Pearson’s correlation coefficient for normally distributed data,
or Spearman’s correlation coefficient for skewed data. P values
less than 0.05 (2-tailed) were considered significant.
RESULTS
Persistent Lyme arthritis in DR4ⴙ/ⴙCD28ⴚ/ⴚ
MHCIIⴚ/ⴚ mice after antibiotic treatment. DR4⫹/⫹
CD28⫺/⫺MHCII⫺/⫺ mice were generated (Figure 1) and
subsequently infected with B burgdorferi, while DR4⫹/⫹
MHCII⫺/⫺ and CD28⫺/⫺ mice were used as controls.
Upon establishment of chronic arthritis in these mice,
intraperitoneal ceftriaxone, a treatment that has been
reported to be 100% effective in eradicating B burgdorferi (27,28), was administered, and mice were monitored
for the presence of arthritis for 2–5 months after antibiotic therapy. While joint inflammation was completely
eradicated after antibiotic treatment in the CD28⫺/⫺
mice, a significant fraction of the DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mice (38%) remained arthritic (Figures 2A
and B). The effectiveness of antibiotic therapy in all
groups was shown by the fact that virtually none of the
mice in the control groups (0 of 25 CD28⫺/⫺ mice and
1 of 31 DR4⫹/⫺MHCII⫺/⫺ mice) had joint inflammation
after treatment with antibiotics, as opposed to the DR4⫹/⫺
CD28⫺/⫺MHCII⫺/⫺ group.
If persistent spirochetes that would elicit an
inflammatory reaction in the joints of these mice were
present, the percentage of edema formation after antibiotic treatment should have been the same in all
HLA–DR4 AND PERSISTENT ARTHRITIS AFTER B BURGDORFERI INFECTION
Figure 1. Phenotype of the DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice. HLA–
DR4 and I-Ab surface expression on B220⫹-gated DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mouse lymphocytes was determined by fluorescenceactivated cell sorting. For the CD28 staining, the B220⫹ lymphocytes
were gated out.
groups. The elimination of B burgdorferi in these mice
systemically was also confirmed by real-time quantitative
PCR (Figure 2A), indicating that chronic arthritis may
occur in DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice in the absence of B burgdorferi. However, the possibility of persistent infection due to a small number of spirochetes
that survived in the microenvironment of the joint could
not be excluded. In accordance with observations made
in human patients with Lyme disease who had persistent
arthritis after antibiotic therapy, these mice continued
to manifest mostly persistent and in some cases intermittent episodes of arthritis, which was monarticular
in some mice, after antibiotic therapy (Figure 3A).
3895
More specifically, 8 of 11 mice had prolonged and persistent joint inflammation (ankle width 2.9–3.1 mm)
that did not resolve with time. In addition, in the 3
remaining mice, we never observed a remission phase
during which inflammation was completely gone. Instead,
we observed a fluctuation of ankle width that ranged
from 2.7 to 3.1 mm. These results imply that the
presence of HLA–DR4, or a related RA-associated DR
allele, is necessary for the development of persistent
Lyme arthritis after antibiotic treatment in a murine
model.
Joint inflammation over the course of arthritic
disease in mice is routinely measured with calipers,
which allows monitoring of edema formation. To examine whether cellular infiltration occurred in the inflammatory process that was established in these arthritic
mice after antibiotic treatment, we performed histologic
analyses of their ankles at the end of the experiment, 4
months after antibiotic therapy. Of 11 mice with persistent joint inflammation post–antibiotic treatment, 6
(55%) showed cellular infiltration, mostly of neutrophils, as opposed to the synovial infiltrate in human
patients, which contains mainly lymphocytes and macrophages and very few neutrophils (2) (Figure 3B). This
observation is consistent with the data obtained using
calipers, since not all of the mice had continuous joint
inflammation; instead, some exhibited recurring ankle
swelling over time that could be attributed only to
Figure 2. Persistence of Lyme arthritis in DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice post–antibiotic treatment. DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺, CD28⫺/⫺, and
DR4⫹/⫹MHCII⫺/⫺ mice were infected with Borrelia burgdorferi (Bb), and arthritis was assessed by measuring the ankles using calipers. On day 55
after infection, mice were treated with antibiotics. Mice were monitored for arthritis for 2 months. A, Arthritis incidence and B burgdorferi burden
in the 3 mouse strains. A significant fraction of DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice (11 of 29) remained arthritic post–antibiotic treatment, compared
with CD28⫺/⫺ mice (0 of 25) and DR4⫹/⫹MHCII⫺/⫺ mice (1 of 31). All mice were negative for the B burgdorferi recA gene post–antibiotic treatment.
The median and 25th to 75th percentile B burgdorferi burden before antibiotic treatment for each mouse strain is shown. Data are the pooled results
of 4 independent experiments. Differences in B burgdorferi burden were not statistically significant (P ⫽ 0.6 by Kruskal-Wallis test). ⴱ ⫽ P ⫽ 0.004;
ⴱⴱ ⫽ P ⫽ 0.0004, by Fisher’s 2-tailed exact probability test. B, Mean ⫾ SEM ankle width of 3 representative mice per group over 120 days. ⴱ ⫽ P ⬍
0.001 versus mock-infected mice, by Student’s 2-tailed t-test.
3896
ILIOPOULOU ET AL
Figure 3. Development of persistent or intermittent arthritis, which was monarticular in some cases, in antibiotic-treated DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mice. A, Width of the left (L) and right (R) ankles of 3 representative mice per group before and after antibiotic treatment. Red, green,
and blue lines represent individual mice. Shaded areas represent mock-infected mice. B, Representative hematoxylin and eosin–stained joint
sections from a CD28⫺/⫺ mouse (a and b) and a DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mouse (c and d) post–antibiotic treatment. b and d are
higher-magnification views of the boxed areas in a and c, respectively. Mice were killed 4 months after antibiotic therapy. Heavy infiltration, mostly
of neutrophils, occurred in the joints of DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice.
edema formation. Besides cellular infiltration, we did
not detect any significant changes in the synovial tissue
of these mice.
Another manifestation of Lyme disease in B burgdorferi–infected mice is carditis, an inflammatory response characterized by cellular infiltration of macrophages (30). To exclude the possibility that the edema in
the joints of these mice was due to obstruction of lower
limb circulation caused by carditis, we performed histologic analysis of their hearts. None of the mice examined
showed cellular infiltration in the heart, suggesting that
edema formation was due to continued inflammation of
the joints, even after elimination of B burgdorferi with
antibiotics (results not shown).
Incidence of anti-OspA antibodies in DR4ⴙ/ⴙ
ⴚ/ⴚ
CD28 MHCIIⴚ/ⴚ mice with persistent joint inflammation after antibiotic treatment. As mentioned above,
in 70% of patients with chronic Lyme arthritis, an
anti-OspA immune response is mounted that seems to
correlate with periods of maximal arthritis (8,9). Therefore, we monitored the OspA serum antibody titers in
our murine model before and after treatment. AntiOspA antibodies were observed in the majority
of B burgdorferi-infected DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺
mice before antibiotic therapy. Following antibiotic treatment, anti-OspA was retained in a significant proportion
of the mice with persistent joint inflammation (36%), as
opposed to none of the mice that no longer had inflammation in the joints (Figure 4). In addition, none of the
Figure 4. Incidence of anti–outer surface protein A (anti-OspA)
antibodies (Ab) in DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent
joint inflammation post–antibiotic treatment. Anti-OspA antibody
levels in DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice were determined by
enzyme-linked immunosorbent assay before and after antibiotic treatment. Squares represent individual mice. Horizontal lines represent
the median. Numbers in parentheses are the absolute numbers of
DR4⫹/⫹ CD28⫺/⫺MHCII⫺/⫺ mice with anti-OspA antibodies. ⴱⴱ ⫽
P ⫽ 0.003 by Mann-Whitney test; ⴱ ⫽P ⫽ 0.014 by Fisher’s 2-tailed
exact probability test.
HLA–DR4 AND PERSISTENT ARTHRITIS AFTER B BURGDORFERI INFECTION
3897
Figure 5. Absence of active immune response in the draining lymph nodes of mice with persistent joint inflammation post–antibiotic treatment.
A, Mean and SEM absolute number of popliteal lymph node cells in mock-infected mice, CD28⫺/⫺ mice actively infected with Borrelia burgdorferi,
and antibiotic-treated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent joint inflammation. Data are the pooled results of 3 independent
experiments. B, Mean and SEM expression of interleukin-17 (IL-17) mRNA in draining popliteal and in inguinal lymph node cells in B burgdorferi–
infected CD28⫺/⫺ mice (left) and in antibiotic-treated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice (right), as assessed by real-time reverse transcriptase–
polymerase chain reaction. Data were normalized to the 18S housekeeping gene and are representative of 3 independent experiments. C, Mean and
SEM expression of IL-17 cytokine in draining popliteal and in inguinal lymph node cells in B burgdorferi–infected CD28⫺/⫺ mice (left) and in
antibiotic-treated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice (right), upon anti-CD3 stimulation for 48 hours, as assessed by IL-17 enzyme-linked
immunosorbent assay. Data are the pooled results of 3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001, by Student’s 2-tailed t-test.
CD28⫺/⫺ or DR4⫹/⫹MHCII⫺/⫺ mice had an antibody
response to OspA post–antibiotic treatment (data not
shown). No significant differences were observed between mice that had anti-OspA antibodies and those
that did not.
Since mice with persistent joint inflammation were
bled weekly, they were further analyzed post–antibiotic
therapy to determine whether there was a correlation between arthritis development and anti-OspA antibody titers.
Consistent with the observations made in patients with
chronic Lyme arthritis, there was a significant correlation
over time between anti-OspA antibodies and development
of arthritis post–antibiotic treatment in these mice (r ⫽
0.53, P ⬍ 0.0001) (results not shown).
In addition, we determined the isotype of the
OspA-specific antibodies. We observed IgM and IgG3,
as well as Th1-dependent IgG2c (results not shown).
Based on the notion that IgG3 class switching can occur
in the presence of IFN␥ (31), we speculated that the
inflammatory milieu was biased toward a Th1 response.
The complete absence of Th2-dependent IgG1 antibodies in all 4 mice with anti-OspA antibodies and
persistent joint inflammation may also be indicative of a
Th1/Th2 imbalance.
Absence of active immune response in the draining lymph nodes of mice with persistent joint inflammation after antibiotic treatment. To investigate the
cellular mechanisms that contribute to persistent inflammation post–antibiotic treatment, we assessed the dynamics and the cytokine profile of the draining popliteal,
as well as the inguinal, lymph node cells in B burgdorferi–
infected, antibiotic-treated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺
mice. B burgdorferi–infected CD28⫺/⫺ mice were used
as a positive control. The size and absolute number of
popliteal lymph node cells was much larger in the
B burgdorferi–infected CD28⫺/⫺ mice with active arthritis compared with the B burgdorferi–infected CD28⫺/⫺
mice that no longer had inflammation in the joints. In
contrast, the absolute number of popliteal lymph node
cells in the antibiotic-treated DR4 ⫹/⫹ CD28 ⫺/
⫺/⫺
⫺MHCII
mice with persistent arthritis was identical
to that in the DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice in which
arthritis was resolved (Figure 5A). A similar pattern was
observed in the inguinal lymph nodes (data not shown),
3898
ILIOPOULOU ET AL
this observation in an animal model in vivo. To assess
CD20 and IFN␥ mRNA expression in antibiotic-treated
DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent joint
inflammation, we isolated the arthritic and nonarthritic
joints from each mouse and performed RT-PCR. There
was a significant increase in both CD20 and IFN␥
mRNA expression in the inflamed joints, compared with
the uninflamed ones, in each DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺
mouse examined (Figure 6). This observation is indicative of a possible role of B cells, as well as inflammatory
cytokines, in the manifestation of persistent inflammation after antibiotic therapy. No difference in IL-17
mRNA expression was observed in joints from DR4⫹/
⫺/⫺
⫹CD28
MHCII⫺/⫺ mice with persistent arthritis and
those from mice in which arthritis was resolved (data not
shown).
DISCUSSION
Figure 6. Mean and SEM CD20 (top) and interferon-␥ (IFN␥)
(bottom) mRNA expression in arthritic joints from 3 antibiotic-treated
DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent joint inflammation.
CD20 and IFN␥ mRNA expression was assessed by real-time reverse
transcriptase–polymerase chain reaction. Data were normalized to the
18S housekeeping gene. P values were determined by Student’s
2-tailed t-test.
providing independent confirmation that arthritis persists in these mice in the absence of systemic B burgdorferi infection.
The presence of IL-17–producing T cells has been
associated with development of inflammation in a wide
range of infectious, as well as autoimmune, diseases
(32–34). In this model of arthritis, IL-17 mRNA expression was observed only in the lymph nodes of the
B burgdorferi–infected CD28⫺/⫺ mice (Figure 5B), even
though an IL-17 memory T cell response was detectable
in the draining popliteal lymph nodes of antibiotictreated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent joint inflammation (Figure 5C).
Increased CD20 and IFN␥ mRNA expression in
the arthritic joints of DR4ⴙ/ⴙCD28ⴚ/ⴚMHCIIⴚ/ⴚ mice
with persistent joint inflammation. The presence of B
cells, as well as of the cytokine IFN␥, has been associated with persistent manifestations of arthritis in humans (10,11,35,36), providing a mechanism for the perpetuation of inflammation in the absence of B burgdorferi.
However, it has not been possible thus far to confirm
The immunologic mechanisms that trigger persistent joint inflammation after elimination of B burgdorferi
infection have been elusive thus far, due to the absence
of an animal model of this inflammatory disease (37).
Here we describe the development of a murine system in
which we observed self-perpetuating arthritis upon systemic eradication of B burgdorferi, as assessed by PCR
analysis. Clinical data have shown that persistent arthritis is manifested in a small percentage of antibiotictreated patients with Lyme arthritis who had experienced intermittent inflammation in the joints before
treatment. These patients had a preponderance of RAassociated HLA–DR alleles, such as HLA–DR4 (1,3,4).
The intriguing phenomenon of persistent arthritis in patients with Lyme disease who were treated
with antibiotics has been extensively studied. Two main
hypotheses have been formulated to explain its pathogenesis, namely, persistent B burgdorferi infection after
antibiotic treatment and infection-induced autoimmunity
(1). The fact that spirochetal DNA is no longer detectable in the inflamed joints after antibiotic treatment
seems to rule out the first hypothesis (5–7). However,
the strong HLA–DR restriction observed in patients
with persistent arthritis post–antibiotic treatment favors
the autoimmunity-based hypothesis. The importance of
understanding the mechanism(s) that trigger this chronic
arthritic disease is 2-fold. It contributes to the general
understanding of chronic inflammatory arthritis, and it
may provide a model for other infection-induced autoimmune diseases.
There seems to be a well-defined linkage between
certain HLA–DR haplotypes and various autoimmune
HLA–DR4 AND PERSISTENT ARTHRITIS AFTER B BURGDORFERI INFECTION
diseases. HLA alleles are responsible for the presentation of autoantigens, as well as the positive and negative
selection of autoreactive T cells in the thymus (38,39).
Patients with Lyme arthritis that is refractory to treatment have a high frequency of HLA–DR4–related alleles compared with patients with treatment-responsive
disease (1,3,4). Interestingly, the same HLA–DR alleles
have also been associated with susceptibility to RA
(40,41). Our murine studies indicate that the presence of
HLA–DR4 or a related allele accentuates persistent
Lyme arthritis post–antibiotic treatment, since a significant fraction of antibiotic-treated DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mice, but none of the CD28⫺/⫺ mice, showed
persistent joint inflammation after antibiotic treatment.
Interestingly, it has been shown that there is an increased frequency of CD4⫹CD28⫺/⫺ T cells in patients
with RA, which correlates with the severity of the
disease (42,43). This observation validates the use of
DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice in developing a murine model of persistent Lyme arthritis post–antibiotic
treatment.
The humoral response against B burgdorferi antigens in antibiotic-treated patients with Lyme arthritis
with persistent joint inflammation is of particular interest. Clinical studies have shown that antibodies against
OspA mark the initiation of prolonged episodes of
arthritis, as opposed to other B burgdorferi proteins that
are associated with mild arthritis of short duration (3,9).
While the majority of B burgdorferi–infected DR4⫹/⫹
CD28⫺/⫺MHCII⫺/⫺ mice with arthritic symptoms developed anti-OspA antibodies prior to antibiotic treatment,
the anti-OspA humoral response was detectable in
some, but not all, antibiotic-treated mice with persistent
joint inflammation. These results provide evidence that
in our murine model of persistent joint inflammation
post–antibiotic treatment, anti-OspA antibodies play an
important, but not necessarily exclusive, role in the pathogenesis of the disease. However, there was a significant correlation between anti-OspA antibody titers and
development of arthritis in the mice that developed an
anti-OspA humoral immune response after antibiotic
treatment. This is consistent with clinical data that indicate
that the levels of anti-OspA antibodies parallel the
severity and duration of arthritis (3,9).
B burgdorferi has been shown to induce IL-17
production in murine T cells in vitro, as well as in
synovial fluid T cells isolated from patients with Lyme
arthritis (44,45). In addition, B burgdorferi–induced arthritis was ameliorated by administration of anti–IL-17
(46). However, it has not yet been demonstrated
whether Th17 cells participate in the pathogenesis of
3899
persistent joint inflammation post–antibiotic therapy. In
the present study we did not detect any IL-17 mRNA
expression in the inflamed mouse joints, although the
B burgdorferi–infected mice showed increased IL-17
mRNA production in the inguinal and popliteal lymph
nodes before antibiotic treatment, consistent with previous reports. It is possible that the number of infiltrated
Th17 cells is small, rendering the IL-17 mRNA negligible in the context of an entire joint. Alternatively, IL-17
may not play an active role in the pathogenesis of
persistent inflammation upon antibiotic treatment. A
memory IL-17 response, however, was detectable in the
lymph nodes of the DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice in
which disease was refractory to treatment, indicating
that Th17 cells were induced during the B burgdorferi
infection phase in these mice, prior to antibiotic treatment.
Increased CD20 mRNA expression was found
in the joints of antibiotic-treated DR4⫹/⫹CD28⫺/⫺
MHCII⫺/⫺ mice with persistent joint inflammation, suggesting that inflammation is locally perpetuated in the
microenvironment of the joint. Taken together, these
data and the presence of anti-OspA antibodies in the
peripheral immune system of these mice are indicative
of a possible role of B cells in the murine model,
consistent with findings in previous clinical studies (3,9).
In addition to the presence of a humoral response in the
inflammatory milieu of the joints, a critical role of T
cells, in particular of Th1 cells, has been documented in
patients with Lyme disease who have persistent arthritis
after antibiotic therapy.
IFN␥ production has been associated with increased severity of Lyme arthritis in mice (20), and
IFN␥-producing cells were identified in the synovial
tissue of patients with Lyme disease who had persistent
arthritis post–antibiotic treatment (10,11,47,48). In this
regard, it is of special interest that we observed increased
IFN␥ mRNA expression in the joints of antibiotictreated DR4⫹/⫹CD28⫺/⫺MHCII⫺/⫺ mice with persistent joint inflammation, suggesting that IFN␥ plays a
pathogenetic role in the murine model. Interestingly, the
presence of IFN␥ in the inflamed joint may explain the
absence of IL-17 transcription, since IL-17 production is
negatively regulated by IFN␥ in vitro (49,50).
In conclusion, we have described a mouse model
with persistent joint inflammation post–antibiotic treatment. Although the possibility of residual infection due
to the presence of a small number of spirochetes in the
joints cannot be excluded, it is apparent from this model
that the expression of the HLA–DR4 allele accentuates
chronic inflammation in B burgdorferi–infected CD28⫺/⫺
mice after antibiotic therapy. The persistent joint inflam-
3900
ILIOPOULOU ET AL
mation observed in these mice may be due to an
autoimmune response that is linked to the presence of
the HLA–DR4 allele. Therefore, the murine model
developed in the present study will allow, for the first
time, precise definition of the mechanism involved in the
development of persistent Lyme arthritis post–antibiotic
treatment. Unraveling its manifestation represents a
unique opportunity, because it provides insights into
other chronic inflammatory arthritides with a possible
autoimmune etiology, such as RA, in which the infecting
agent is unknown.
ACKNOWLEDGMENTS
We are grateful to Dr. Jenifer Coburn for the gift of
B burgdorferi and to Dr. Jeffrey Bluestone for guiding us to
the CD28⫺/⫺ system. We thank members of the Huber laboratory for critical reading of the manuscript and Lin Miao and
Francesca Chang for technical help.
AUTHOR CONTRIBUTIONS
Dr. Huber had full access to all of the data in the study and
takes responsibility for the integrity of the data and the accuracy of the
data analysis.
Study design. Iliopoulou, Huber.
Acquisition of data. Iliopoulou.
Analysis and interpretation of data. Iliopoulou, Huber.
Manuscript preparation. Iliopoulou, Huber.
Statistical analysis. Iliopoulou.
Scoring of hematoxylin and eosin–stained sections. Alroy.
REFERENCES
1. Steere AC, Glickstein L. Elucidation of Lyme arthritis. Nat Rev
Immunol 2004;4:143–52.
2. Steere AC. Lyme disease. N Engl J Med 2001;345:115–25.
3. Kalish RA, Leong JM, Steere AC. Early and late antibody
responses to full-length and truncated constructs of outer surface
protein A of Borrelia burgdorferi in Lyme disease. Infect Immun
1995;63:2228–35.
4. Steere AC, Klitz W, Drouin EE, Falk BA, Kwok WW, Nepom GT,
et al. Antibiotic-refractory Lyme arthritis is associated with HLADR molecules that bind a Borrelia burgdorferi peptide. J Exp Med
2006;203:961–71.
5. Nocton JJ, Dressler F, Rutledge BJ, Rys PN, Persing DH, Steere
AC. Detection of Borrelia burgdorferi DNA by polymerase chain
reaction in synovial fluid from patients with Lyme arthritis. N Engl
J Med 1994;330:229–34.
6. Carlson D, Hernandez J, Bloom BJ, Coburn J, Aversa JM, Steere
AC. Lack of Borrelia burgdorferi DNA in synovial samples from
patients with antibiotic treatment–resistant Lyme arthritis. Arthritis Rheum 1999;42:2705–9.
7. Steere AC, Angelis SM. Therapy for Lyme arthritis: strategies for
the treatment of antibiotic-refractory arthritis [review]. Arthritis
Rheum 2006;54:3079–86.
8. Kalish RA, Leong JM, Steere AC. Association of treatmentresistant chronic Lyme arthritis with HLA-DR4 and antibody
reactivity to OspA and OspB of Borrelia burgdorferi. Infect
Immun 1993;61:2774–9.
9. Akin E, McHugh GL, Flavell RA, Fikrig E, Steere AC. The
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
immunoglobulin (IgG) antibody response to OspA and OspB
correlates with severe and prolonged Lyme arthritis and the IgG
response to P35 correlates with mild and brief arthritis. Infect
Immun 1999;67:173–81.
Gross DM, Steere AC, Huber BT. T helper 1 response is dominant
and localized to the synovial fluid in patients with Lyme arthritis.
J Immunol 1998;160:1022–8.
Shin JJ, Glickstein LJ, Steere AC. High levels of inflammatory
chemokines and cytokines in joint fluid and synovial tissue
throughout the course of antibiotic-refractory Lyme arthritis.
Arthritis Rheum 2007;56:1325–35.
Yssel H, Shanafelt MC, Soderberg C, Schneider PV, Anzola J,
Peltz G. Borrelia burgdorferi activates a T helper type 1-like T cell
subset in Lyme arthritis. J Exp Med 1991;174:593–601.
Barthold SW, Beck DS, Hansen GM, Terwilliger GA, Moody KD.
Lyme borreliosis in selected strains and ages of laboratory mice.
J Infect Dis 1990;162:133–8.
Ma Y, Seiler KP, Eichwald EJ, Weis JH, Teuscher C, Weis JJ.
Distinct characteristics of resistance to Borrelia burgdorferiinduced arthritis in C57BL/6N mice. Infect Immun 1998;66:161–8.
Wooten RM, Weis JJ. Host-pathogen interactions promoting
inflammatory Lyme arthritis: use of mouse models for dissection
of disease processes. Curr Opin Microbiol 2001;4:274–9.
McKisic MD, Redmond WL, Barthold SW. Cutting edge: T cellmediated pathology in murine Lyme borreliosis. J Immunol 2000;
164:6096–9.
Keane-Myers A, Nickell SP. Role of IL-4 and IFN-␥ in modulation
of immunity to Borrelia burgdorferi in mice. J Immunol 1995;155:
2020–8.
Kang I, Barthold SW, Persing DH, Bockenstedt LK. T-helper-cell
cytokines in the early evolution of murine Lyme arthritis. Infect
Immun 1997;65:3107–11.
Matyniak JE, Reiner SL. T helper phenotype and genetic susceptibility in experimental Lyme disease. J Exp Med 1995;181:1251–4.
Crandall H, Dunn DM, Ma Y, Wooten RM, Zachary JF, Weis JH,
et al. Gene expression profiling reveals unique pathways associated with differential severity of Lyme arthritis. J Immunol
2006;177:7930–42.
Barthold SW, de Souza MS, Janotka JL, Smith AL, Persing DH.
Chronic Lyme borreliosis in the laboratory mouse. Am J Pathol
1993;143:959–71.
Barthold SW, Persing DH, Armstrong AL, Peeples RA. Kinetics
of Borrelia burgdorferi dissemination and evolution of disease
after intradermal inoculation of mice. Am J Pathol 1991;139:
263–73.
Iliopoulou BP, Alroy J, Huber BT. CD28 deficiency exacerbates
joint inflammation upon Borrelia burgdorferi infection, resulting
in the development of chronic Lyme arthritis. J Immunol 2007;
179:8076–82.
Ito K, Bian HJ, Molina M, Han J, Magram J, Saar E, et al.
HLA-DR4-IE chimeric class II transgenic, murine class II-deficient mice are susceptible to experimental allergic encephalomyelitis. J Exp Med 1996;183:2635–44.
Coburn J, Barthold SW, Leong JM. Diverse Lyme disease spirochetes bind integrin ␣IIb␤3 on human platelets. Infect Immun
1994;62:5559–67.
Coburn J, Leong JM, Erban JK. Integrin ␣IIb␤3 mediates binding
of the Lyme disease agent Borrelia burgdorferi to human platelets.
Proc Natl Acad Sci U S A 1993;90:7059–63.
Moody KD, Adams RL, Barthold SW. Effectiveness of antimicrobial treatment against Borrelia burgdorferi infection in mice.
Antimicrob Agents Chemother 1994;38:1567–72.
Pavia C, Inchiosa MA Jr, Wormser GP. Efficacy of short-course
ceftriaxone therapy for Borrelia burgdorferi infection in C3H
mice. Antimicrob Agents Chemother 2002;46:132–4.
Morrison TB, Ma Y, Weis JH, Weis JJ. Rapid and sensitive
quantification of Borrelia burgdorferi-infected mouse tissues by
HLA–DR4 AND PERSISTENT ARTHRITIS AFTER B BURGDORFERI INFECTION
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
continuous fluorescent monitoring of PCR. J Clin Microbiol
1999;37:987–92.
Ruderman EM, Kerr JS, Telford SR III, Spielman A, Glimcher
LH, Gravallese EM. Early murine Lyme carditis has a macrophage
predominance and is independent of major histocompatibility
complex class II-CD4⫹ T cell interactions. J Infect Dis 1995;171:
362–70.
Stavnezer J. Immunoglobulin class switching. Curr Opin Immunol
1996;8:199–205.
Lubberts E, Koenders MI, Oppers-Walgreen B, van den Bersselaar L, Coenen-de Roo CJ, Joosten LA, et al. Treatment with a
neutralizing anti-murine interleukin-17 antibody after the onset of
collagen-induced arthritis reduces joint inflammation, cartilage
destruction, and bone erosion. Arthritis Rheum 2004;50:650–9.
Stockinger B, Veldhoen M. Differentiation and function of Th17
T cells. Curr Opin Immunol 2007;19:281–6.
Veldhoen M, Hocking RJ, Flavell RA, Stockinger B. Signals
mediated by transforming growth factor-␤ initiate autoimmune
encephalomyelitis, but chronic inflammation is needed to sustain
disease. Nat Immunol 2006;7:1151–6.
Ghosh S, Steere AC, Stollar BD, Huber BT. In situ diversification
of the antibody repertoire in chronic Lyme arthritis synovium.
J Immunol 2005;174:2860–9.
Ghosh S, Seward R, Costello CE, Stollar BD, Huber BT. Autoantibodies from synovial lesions in chronic, antibiotic treatmentresistant Lyme arthritis bind cytokeratin-10. J Immunol 2006;177:
2486–94.
Benoist C, Mathis D. Autoimmunity provoked by infection: how
good is the case for T cell epitope mimicry? Nat Immunol 2001;2:
797–801.
Marrack P, Kappler J, Kotzin BL. Autoimmune disease: why and
where it occurs. Nat Med 2001;7:899–905.
Ridgway WM, Fathman CG. MHC structure and autoimmune
T cell repertoire development. Curr Opin Immunol 1999;11:
638–42.
Buckner JH, Nepom GT. Genetics of rheumatoid arthritis: is there
a scientific explanation for the human leukocyte antigen association? Curr Opin Rheumatol 2002;14:254–9.
3901
41. Pascual M, Mataran L, Jones G, Shing D, van der Slik AR,
Giphart MJ, et al. HLA haplotypes and susceptibility to rheumatoid arthritis: more than class II genes. Scand J Rheumatol
2002;31:275–8.
42. Pawlik A, Ostanek L, Brzosko I, Brzosko M, Masiuk M, Machalinski B, et al. The expansion of CD4⫹CD28⫺ T cells in patients
with rheumatoid arthritis. Arthritis Res Ther 2003;5:R210–3.
43. Schmidt D, Martens PB, Weyand CM, Goronzy JJ. The repertoire
of CD4⫹ CD28⫺ T cells in rheumatoid arthritis. Mol Med
1996;2:608–18.
44. Knauer J, Siegemund S, Muller U, Al-Robaiy S, Kastelein RA,
Alber G, et al. Borrelia burgdorferi potently activates bone
marrow-derived conventional dendritic cells for production of
IL-23 required for IL-17 release by T cells. FEMS Immunol Med
Microbiol 2007;49:353–63.
45. Infante-Duarte C, Horton HF, Byrne MC, Kamradt T. Microbial
lipopeptides induce the production of IL-17 in Th cells. J Immunol
2000;165:6107–15.
46. Burchill MA, Nardelli DT, England DM, DeCoster DJ, Christopherson JA, Callister SM, et al. Inhibition of interleukin-17
prevents the development of arthritis in vaccinated mice challenged with Borrelia burgdorferi. Infect Immun 2003;71:3437–42.
47. Gross DM, Forsthuber T, Tary-Lehmann M, Etling C, Ito K, Nagy
ZA, et al. Identification of LFA-1 as a candidate autoantigen in
treatment-resistant Lyme arthritis. Science 1998;281:703–6.
48. Yin Z, Braun J, Neure L, Wu P, Eggens U, Krause A, et al. T cell
cytokine pattern in the joints of patients with Lyme arthritis and its
regulation by cytokines and anticytokines. Arthritis Rheum 1997;
40:69–79.
49. Harrington LE, Hatton RD, Mangan PR, Turner H, Murphy TL,
Murphy KM, et al. Interleukin 17-producing CD4⫹ effector
T cells develop via a lineage distinct from the T helper type 1 and
2 lineages. Nat Immunol 2005;6:1123–32.
50. Park H, Li Z, Yang XO, Chang SH, Nurieva R, Wang YH, et al.
A distinct lineage of CD4 T cells regulates tissue inflammation by
producing interleukin 17. Nat Immunol 2005;6:1133–41.
Документ
Категория
Без категории
Просмотров
0
Размер файла
346 Кб
Теги
persistence, treatment, mice, cd28, arthritis, hladr4positive, negativa, borrelia, burgdorferiinfected, postantibiotic
1/--страниц
Пожаловаться на содержимое документа