| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
T Cells Respond to Borrelia burgdorferi Lipoproteins and Lipidated Hexapeptides1
,
*
Divisions of Immunobiology and Rheumatology, Department of Medicine, University of Vermont College of Medicine, Burlington, VT 05405; Departments of
Internal Medicine and
Microbiology, University of Texas Southwestern Medical Center, Dallas, TX 75235;
§
Department of Medicine, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, New Brunswick, NJ 08903; and
¶
Lymphocyte Biology Section, Division of Rheumatology and Immunology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells that proliferates upon stimulation with the causative
spirochete, Borrelia burgdorferi. A panel of
Borrelia-reactive T cell clones was derived from
synovial fluid of two patients with Lyme arthritis. Each of six
clones from one patient used the V1 TCR segment but had otherwise
unique CDR3 sequences and diverse V segment usage. Stimulation of
the V1 clones was optimal in the presence of
Borrelia, dendritic cells, and exogenous IL-2, which was
reflected by proliferation, TCR down-modulation, as well as induction
of CD25 and Fas ligand expression. Stimulation by B.
burgdorferi-pulsed dendritic cells withstood chemical fixation
and was not restricted to class I or class II MHC, CD1a, CD1b, or CD1c.
In contrast, anti- antibody potently inhibited proliferation.
Extraction of B. burgdorferi lipoproteins with Triton
X-114 enriched for the stimulatory component. This was confirmed using
lipidated vs nonlipidated hexapeptides of Borrelia outer
surface proteins. These observations suggest that synovial V1 T
cells may mediate an innate immune response to common lipoprotein
products of spirochetes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production (6). Furthermore, administration of IL-4 can protect
susceptible strains, and neutralization of IFN- diminishes the
severity of infection (6, 7, 8).
T cells are found in increased numbers in inflamed synovial
fluid from both rheumatoid (9, 10, 11) and Lyme arthritis (12) as well as
at diverse idiopathic inflammatory sites (13, 14, 15, 16). The function of
T cells at these anatomically selected sites remains obscure.
Whereas the predominant cell in PBL expresses the V9V2
TCR, it is the V1 subset that accumulates in inflamed synovial fluid
(9, 10, 11) and in PBL of HIV-infected individuals (17, 18). The ligands
responsible for TCR activation of the V1 subset are largely unknown,
as are the Ag restriction elements, if any. Several reports suggest
that activated B cells can stimulate the expansion of V1 T cells
(19, 20, 21), but it is not clear whether direct TCR stimulation is
required for the activation of V1 T cells in these settings. A
variety of cell types is potentially available for processing and
presenting Ag within the inflamed synovium, including B cells,
macrophages, monocytes, and dendritic cells
(DC).3 At the site of
infection, an important APC appears to be the DC, which ingests
B. burgdorferi by coiling phagocytosis and subsequently can
present spirochetal antigens to conventional 
ß T cells (22).
Interestingly, degraded spirochetes can be found within the cytosol of
the DC as well as in association with MHC class II compartments (22).
One proposed function of the T cell population is as an arm of
the innate immune response, making use of the limited germline
TCR repertoire to respond to Ags unique to microbial pathogens. This
idea derives largely from the observation that V9V2 PBL uniformly
respond to mycobacterial prenyl pyrophosphate and similar
nonproteinaceous small molecules (23, 24). In addition, an effective
early protective immune response to Listeria monocytogenes
(25, 26, 27), Mycobacterium tuberculosis (28), and
Toxoplasma gondii (29) requires the subset, as
depletion leads to more severe infection.
While the role of synovial T cells in Lyme arthritis is unclear,
we have previously shown that the numbers of synovial T cells
expand dramatically when stimulated with Borrelia.
Furthermore, Lyme synovial fluid cells express Fas ligand (FasL)
and mediate apoptosis of the CD4+ subset of synovial ß
T cells when stimulated in vitro with B. burgdorferi (12).
In the present studies we further characterize clones from Lyme
arthritis synovial fluid that express the V1 TCR chain paired with
diverse V chains. Their optimal proliferative response requires DC,
B. burgdorferi, and exogenous IL-2 and occurs via the
TCR. Molecules other than traditional MHC restriction elements or group
I CD1 molecules mediate the V1 response to B.
burgdorferi. Finally, the stimulatory component of B.
burgdorferi is enriched in the membrane lipoprotein fraction and
is conferred by a minimal lipoprotein.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both Lyme arthritis patients were female and from areas endemic for Lyme disease. Each had a typical history for Lyme arthritis, and Abs to B. burgdorferi in both peripheral blood and synovial fluid were demonstrated by ELISA and confirmed by immunoblot. The patients were 12 and 16 yr old, with duration of symptoms for 12 and 18 mo, respectively, and were followed at the Lyme Disease Center Clinic at the University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School (New Brunswick, NJ).
Abs and flow cytometry
Abs included anti-human class I MHC (HLA-A, -B, and -C;
3F10, Ancell Corp., Bayport, MN), anti-human class II MHC (HLA-DP,
-DQ, and -DR; TDR31.1, Ancell), anti-CD1a (OKT6, American Type
Culture Collection, Manassas, VA), anti-CD1b (BCD1b5.1),
anti-CD1c (L161, BioSource International, Camarillo, CA),
anti-FasL-biotin (A11) (30), anti-CD25-phycoerythrin
(CD25-3g10, Caltag, Burlingame, CA), anti--FITC (5A6.E9,
Endogen, Woburn, MA), and anti-HLA-DR (YE2/36, Accurate Chemical &
Scientific Corp., Westbury, NY). Ascites from anti- clone
5A6.E9 (31) was a gift from Dr. Michael Brenner (Harvard Medical
School, Boston, MA). Flow cytometry was performed using a Coulter Elite
flow cytometer (Coulter, Hialeah, FL) and isotypic control Abs (Caltag)
to define negative staining.
Derivation of T cell clones and peripheral blood DC
T cell clones were derived as previously described (12)
by limiting dilution from primary cultures of Lyme arthritis synovial
fluid lymphocytes that had been stimulated for 10 days to expand the
subset using a sonicate of B. burgdorferi (10
µg/ml; strain N40) grown in BSK II medium (32). Clones were expanded
in 24-well plates by weekly stimulation of 106 cells with
3 x 106 irradiated (4000 rad) allogeneic PBMC in 2 ml
of serum-free AIM-V (with L-glutamine, streptomycin
(50 µg/ml), and gentamicin (10 µg/ml); Life Technologies,
Gaithersburg, MD), 40 U/ml recombinant human IL-2 (Cetus, Emeryville,
CA), and 10 µg/ml B. burgdorferi. Cultures were split and
fed with fresh medium as needed. The derivation and maintenance of the
B. burgdorferi-specific ß clone 114B have been
described previously (32).
DC were derived using a detailed protocol published previously (33). Briefly, PBMC were isolated by centrifugation over Histopaque 1077 (Sigma, St. Louis, MO), and 6–10 x 106 cells were incubated for 120 min at 37°C in 35-mm plastic culture wells. Nonadherent cells were then gently removed, and fresh medium was added containing IL-4 (500 U/ml; Genzyme, Cambridge, MA) and granulocyte-macrophage CSF (800 U/ml; Genetics Institute, Cambridge, MA). After 7 days at 37°C, the DC-enriched cultures were harvested and used either unsorted or following sterile sorting by flow cytometry into a large granular cell fraction that was highly enriched for DC. These preparations were used in proliferation assays or were frozen for subsequent experiments. The purity of the crude DC preparation ranged from 10–30% as judged by light scatter properties, whereas the purity of the sorted populations averaged 90%. The contaminating smaller cells consisted of approximately equal numbers of T and B cells based on CD3 and CD19 expression. The DC population was negative for CD14, CD3, and CD19 but positive for CD1a, CD1b, and CD1c and a variety of costimulatory and adhesion molecules, including CD11a, CD11c, CD40, CD54, CD58, CD80, and HLA-A, -B, -C, -DP, -DQ, and -DR as reported previously (33, and data not shown).
TCR sequencing
cDNA was prepared from RNA extracted from six Lyme synovial
clones using reverse transcriptase, and the CDR3 regions of the
-chains were amplified using previously published primer sequences
(12). The CDR3 region of the -chain was amplified using the
following primers: consensus V, 5'-AAATCTTCCAACTTGGAAGGGAGA-3';
consensus J1/2, 5'-TGTGACAACAAGTGTTGTTCCACT-3'; or consensus C,
5'-GTCGTTAGTCTTCATGGTGTTCCC-3'. Chain termination sequencing was
performed using a sequencing kit (Applied Biosystems, Foster City, CA),
a nested primer (5'-TTCACCAGACAAGCGACATT-3') in the case of the
-chain, or the above C primer for the -chain and was analyzed
by automated sequencing (Applied Biosystems model 373A).
Proliferation assays
T cell clones (2–3 x 104) that were at
least 14 days from previous Ag stimulation were incubated alone or with
3–10 x 104 allogeneic irradiated (8000 rad) crude DC
in the absence or the presence of B. burgdorferi sonicate at
3 or 10 µg/ml. All assays were performed in triplicate in
96-half-well plates in 200 µl of AIM-V with 5% FBS and 4 U/ml rIL-2
unless otherwise indicated. In experiments with PBMC as stimulators,
105 irradiated cells (4000 rad) were used under otherwise
identical conditions. Cells were harvested on day 3 after labeling with
1 µCi of [3H]TdR/well for the final 18 h of
culture. Values were corrected for basal proliferation in the presence
of medium alone, and error bars represent the SD of triplicate
determinations unless otherwise indicated. Assays with the
Borrelia-reactive ß T cell clone 114B were performed
similarly using 105 irradiated (4000 rad) autologous PBMC
feeders. The HLA haplotypes from donors were as follows: donor 1, A2,
A24, B44, B48, Cw3, Cw0704, DRB1*0101, DRB1*1301, DRB3*02, DQA1*0101,
DQA1*0103, DQB1*0501, DQB1*0603; donor 2, A2, A2501, B44, B39, Cw0501,
Cw1203, DRB1*01, DRB1*16, DQA1*0101, DQA1*0103, DQB1*0501, DQB1*0502;
and donor 3, A1, A2, B8, B1402, Cw8, DRB1*0301, DRB1*1302, DRB3*02,
DRB3*0301, DQA1*0501, DQA1*0102, DQB1*02, DQB1*0604. For Ab inhibition,
ascites or purified Ab was added at the indicated dilution before the
mixing of cells.
Chemical fixation of the DC preparation was performed by first incubating the cells in the presence or the absence of B. burgdorferi at 10 µg/ml overnight at 37°C, then washing with 5% FBS in RPMI (Life Technologies), and fixing by addition of either ice-cold 75 mM 1-ethyl-3-(3'-dimethyl-aminopropyl)-carbodiimide (EDCI; Sigma) in PBS for 60 min on ice or 0.5% paraformaldehyde (Ted Pella, Redding, CA) in PBS for 15 min at room temperature. Following fixation the cells were extensively washed with 5% FBS, AIM-V medium and used in proliferation assays as indicated.
Ags and fractionation of borrelial lipoproteins
Borrelia hermsii sonicate was a generous gift from
Dr. Allen Steere (New England Medical Center, Boston, MA).
Treponema pallidum was prepared as previously described
(34). Mycobacterium tuberculosis (strain H37RA) was
purchased from Difco (Detroit, MI). Candida albicans and
tetanus toxoid were purchased from the University Health Center
(Burlington, VT). LPS from Escherichia coli serotype 0111:B4
was obtained from Sigma. E. coli strain INVF'
(Invitrogen, Carlsbad, CA) was sonicated and spun 5 min at 13,000
x g, and the sonicate was passed through a 0.45-µm filter
(Millipore, Bedford, MA). Triton X-114 (Sigma) phase separation was
performed essentially as previously described (35). Briefly, B.
burgdorferi (109/ml) was solubilized in 2% (v/v)
Triton X-114 in PBS by end-over-end rotation overnight at 4°C.
Insoluble material was removed by centrifugation (20,000 x
g for 30 min), and the supernatant was heated to 37°C for
15 min. The detergent phase was separated by centrifugation
(13,000 x g for 15 min at 25°C). Proteins were
recovered by precipitation in 10 vol of -20°C acetone. Protein
concentrations were measured using the bicinchoninic acid assay
(Pierce, Rockford, IL) according to the manufacturer’s instructions.
Purification of native lipidated OspA from B. burgdorferi
was performed by mAb immunoaffinity chromatography as previously
described (34). Purity was assessed by SDS-PAGE. Nonacylated OspA was
prepared as a recombinant fusion protein with
glutathione-S-transferase in Escherichia coli
using the plasmid pGEX-2T (Pharmacia, Piscataway, NJ). Expression and
affinity column purification were performed according to standard
procedures of the manufacturer. The nonacylated OspA was cleaved and
isolated from the glutathione-S-transferase moiety by
thrombin cleavage of the immobilized fusion proteins bound to the
glutathione resin. To remove residual E. coli LPS
contaminants, protein solutions were subjected to Detoxi-Gel Endotoxin
Removing Gel columns (Pierce). Proteins were tested for the presence of
residual contaminating endotoxin by the Limulus
amoebocyte lysate gelatin assay (E-Toxate, Sigma).
Hexapeptides corresponding to the N-termini of OspA (Cys-Lys-Gln-Asn-Val-Ser) or OspC (Cys-Asn-Asn-Ser-Gly-Lys) were synthesized on an Applied Biosystems model 430A peptide synthesizer (Foster City, CA). Inasmuch as palmitate is the principal long chain fatty acid of native spirochetal lipoproteins (34), lipohexapeptide analogues corresponding to the acylated N-termini of the lipoproteins were synthesized as tripalmitoyl derivatives as previously described (34). Tripalmitoyl-S-glycerylcysteine (i.e., acylated cysteine lacking the pentapeptide) was also synthesized. All synthetic peptides and lipopeptides were assayed for contaminating endotoxin by the Limulus amoebocyte lysate assay after being dissolved or suspended in sterile endotoxin-free water.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
clones are diverse and of the V1
subset
Stimulation of Lyme arthritis synovial fluid lymphocytes with
B. burgdorferi leads to an expansion of the subset to
as much as 50% of the total cultured cells (12). To better
characterize the T cell response in Lyme arthritis, bulk
cultures of Borrelia-stimulated synovial fluid lymphocytes
were cloned by limiting dilution using autologous feeders and borrelial
sonicate. As expected from the predominant phenotype of the bulk
culture assessed by flow cytometry, most clones were
CD4-CD8- and expressed the V1 TCR (data
not shown). To determine whether the V1 clones were simply daughter
cells, the CDR3 segments of the - and -chains were PCR amplified
and sequenced from six clones from one patient. As shown for the
-chain in Fig. 1
A, no
duplicate sequences existed among the clones, ruling out expansion of a
single clone. There was no similarity in either CDR3 length or
junctional nucleotide sequence among the clones, and use was made of
both the J1 (four clones) and J2 (two clones) segments. The
TCR chain CDR3 sequences are shown in Fig. 1B and again are
notable for the lack of similarity at the nucleotide and amino acid
levels as well as the use of a several V regions, including V2,
V3, and V4. Despite the heterogeneity of the CD3 region of the
-chain and of V usage, all six synovial V1 clones proliferated
in response to B. burgdorferi (data not shown).
|
1 clones
Without preceding knowledge of a restriction element or Ag
processing requirements, we elected to use as accessory cells DC
prepared from adherent blood monocytes as previously described (33).
These DC express a large complement of adhesion and costimulatory
molecules as well as the MHC-like molecules, CD1a, CD1b, and CD1c (33)
(data not shown), which have been suggested to play a role in TCR
restriction of certain cells (36). For the clones shown
in Fig. 2A, the typical
stimulation in response to B. burgdorferi and DC is
severalfold over background proliferation of cells alone, and
the proliferative response to DC alone is intermediate between the two.
The Borrelia-induced proliferation was very reproducible in
both clones and was absolutely dependent on exogenous IL-2, which could
not be omitted from cultures without complete loss of viability (Fig. 2B). This is consistent with the lack of IL-2 production by
the clones in response to B. burgdorferi/DC
stimulation (data not shown). The necessary use of exogenous IL-2
probably accounted for the elevated backgrounds with DC alone,
resulting in a decreased stimulation index in the presence of
Borrelia. We have observed a similar decreased stimulation
index using a B. burgdorferi-specific ß T cell clone
when IL-2 was added to the cultures. In this case the increased
background proliferation with IL-2 caused the stimulation index to
decrease from 7.8 in the absence of exogenous IL-2 to 2.2 in its
presence (M. S. Vincent, unpublished observations). Thus, the
absolute requirement for IL-2 by the clones partially diminished
the magnitude of the stimulation index due to increased background
proliferation rather than poor responsiveness to Borrelia.
Other reports have also described the requirement for exogenous
IL-2 in the propagation of human V1 T cells (37, 38).
|
clones, we characterized the down-modulation of surface TCR
following stimulation with B. burgdorferi as well as the
induction of cell surface CD25 and FasL. Fig. 3 illustrates the expression of surface
TCR- and CD25 by a representative clone that was
stimulated with Borrelia and irradiated PBMC (APC) 2 days
previously. Compared with stimulation with PBMC alone, the addition of
B. burgdorferi considerably increased the proportion of
CD25+ cells. Furthermore, the CD25+
cells manifested down-modulated surface TCR- relative to the
+ cells that lacked CD25 expression. The addition of
cyclosporin A partially blocked the induction of CD25 and TCR
down-modulation, further suggesting that this change in phenotype with
Borrelia was due to a signaling pathway known to be
activated by TCR engagement. This finding was very consistent in a
second experiment.
|
A depicts a more
complete time course of CD25 expression by a representative
clone following stimulation with B. burgdorferi and
irradiated PBMC. CD25 expression peaked 3 days following stimulation,
consistent with previous reports of CD25 induction on human
clones (39). In contrast to the proliferative response, the induction
of surface CD25 was threefold higher than the levels in the presence of
PBMC alone, consistent with the view that Borrelia provided
a specific activation signal. A second marker of activated T cells is
FasL, which we have previously observed at high levels on cells
from Lyme arthritis synovial fluid stimulated with B.
burgdorferi (12). Fig. 4B shows that up-regulation of
FasL also peaked on day 3 in the presence of B. burgdorferi
and PBMC. Thus, the up-regulation of CD25 and FasL on V1 clones by
B. burgdorferi is consistent with TCR signaling to the
clones and can be used to corroborate findings obtained using
proliferation. The minimal background induction of these cell surface
molecules demonstrates that measurement of surface activation markers
provides a higher stimulation index than proliferation. As such, these
parameters may be more useful in investigating the fine specificity of
the borrelial components involved.
|
Because the DC preparations were not homogeneous, it was not clear
whether the contaminating T and B cells or the DC were stimulating
proliferation in the presence of B. burgdorferi. In fact, a
number of reports have suggested that activated B cells are a principal
costimulatory cell for the V1 subset of T cells (19, 20, 21).
When the DC were sorted flow cytometrically by light scatter properties
for large granular cells, essentially all the stimulatory activity
segregated with the DC population (Fig. 5). For this analysis, the absolute
number of sorted cells used as accessory cells was intentionally set
based on the proportion of a given cell subset that was in the unsorted
DC culture. It is clear that DC provided a potent stimulus for the
clones, which were active using as few as 8000 DC/well. The
non-DC population (including the B and T subsets), by contrast, induced
no Borrelia-specific response, even though they were used at
a 7.5-fold excess relative to DC (Fig. 5).
|
clones in the presence of B.
burgdorferi and DC could be related to a number of factors.
Stimulation may occur directly by traditional Ag presentation or
indirectly through Borrelia-induced expression of
costimulatory molecules or cytokines by DC, either of which would
require metabolically active DC. Alternatively, activation by DC
plus B. burgdorferi could involve a type of superantigen
response that would not require metabolically active accessory cells or
Ag processing (40). To distinguish between these possibilities chemical
fixation of DC was performed with either EDCI or paraformaldehyde to
preclude further metabolic activity, including cytokine production and
Ag processing (41). As shown in Fig. 6,
fixation of DC with either reagent after an initial overnight
incubation with B. burgdorferi did not diminish DC
stimulatory activity. In contrast, addition of B.
burgdorferi following fixation of accessory cells prevented
stimulation beyond background levels. These findings are consistent
with the view that the activation of the V1 clones by
Borrelia does not involve secreted molecules from DC or a
traditional superantigen effect of B. burgdorferi on
cells, but does require a metabolically active accessory cell.
|
clones lacks conventional MHC restriction
but is inhibited by anti- Ab
To determine whether the TCR was required for the activation of
V1 clones by Borrelia, blocking studies were performed
using a specific TCR Ab (31). Fig. 7 shows that a control
Borrelia-specific ß T cell clone, 114B (32),
proliferated in the presence of soluble pan- Ab, whereas V1
clone Bb1–3 was markedly inhibited in a dose-dependent manner across a
broad range of Ab concentrations.
|
clone in the presence of B. burgdorferi. If the
response of the clones were restricted by conventional
Ag-presenting molecules, one would expect disparate HLA haplotypes to
differ in their Borrelia-induced stimulatory capacity. As
shown in Fig. 8, PBMC from each of three
donors displayed the same ability to induce proliferation of the
clone Bb1-16. Clone Bb1-3 responded similarly to the same panel of
donors (data not shown). Although we cannot exclude a role for
classical MHC Ag presentation on the basis of these experiments, at a
minimum the data indicate a departure from the pattern of Ag
restriction typically observed with ß T cell clones. In fact, we
have previously shown that a panel of Borrelia-specific
ß T cell clones is indeed activated only by autologous PBMC (32).
As a separate measure of MHC restriction of the B.
burgdorferi response, Abs to class I and class II molecules were
used in blocking studies. As shown in Fig. 9, neither blocking Ab inhibited specific
proliferation of cells to Borrelia. The class I Ab
was nonspecifically toxic to the clones (Fig. 9A) at
a concentration of 1.0 µg/ml or greater as evidenced by its
inhibition of a class II-restricted CD4+ TCR-ß clone
(114B) that is specific for B. burgdorferi (Fig. 9B). The class II Ab produced no inhibition of the
clones (Fig. 9C) but did specifically inhibit the TCR-ß
clone 114B (Fig. 9D). It was still possible that Borrelia
presentation occurred through nonpolymorphic MHC-like molecules such as
those in the CD1 family. However, Fig. 10 shows no evidence of
inhibition by Abs to CD1a, CD1b, or CD1c. This was corroborated by lack
of specific lysis of target C1R cells transfected with CD1a, CD1b, or
CD1c (data not shown).
|
|
|
1 clones is specific for spirochetes and
mycobacteria, is enriched in the membrane lipoprotein fraction of
Borrelia, and is conferred by a minimal lipopeptide
Certain T cells are known to recognize microbial components
common to a variety of organisms. This was examined for the synovial
V1 clones using a panel of microbial preparations and purified Ags
(Fig. 11A). Interestingly,
two other spirochetes, B. hermsii and Treponema
pallidum, induced levels of proliferation nearly identical with
those seen with B. burgdorferi. In addition, M.
tuberculosis also stimulated the V1 clones. This is in contrast
to the organisms, Candida albicans and E. coli,
or the bacterial products tetanus toxoid and LPS, none of which
increased proliferation above background levels. These findings were
also substantiated using CD25 induction (Fig. 11B).
|
proliferative capacity was observed in the
lipoprotein extract compared with the unfractionated sonicate (Fig. 12A). This was further
substantiated using a series of lipidated vs nonlipidated variants of
Osp determinants from B. burgdorferi. Fig. 12B
shows the proliferative responses of two clones in response to
lipidated and nonlipidated versions of full-length OspA or N-terminal
hexapeptides of OspA or OspC. As is evident, both clones
responded only to the lipidated forms of these proteins/peptides. These
findings were extremely consistent in two other experiments.
Furthermore, if the hexapeptide was further truncated down to only the
single cysteine to which the tripalmitoyl residue was attached, the
response was also lost (Fig. 12C). Thus, the minimal
stimulatory unit of B. burgdorferi was a hexapeptide
tripalmitoyl residue.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 subset of cells was expanded
in fresh Lyme arthritis synovial fluid, and even further during in
vitro cultivation of synovial fluid lymphocytes with
Borrelia, suggested that synovial cells might respond
to borrelial components. V1 clones from Lyme patients proliferate in
vitro in the presence of B. burgdorferi, DC, and IL-2. This
response is not restricted by known classical or nonclassical HLA
molecules. Furthermore, the stimulatory component of
Borrelia is tripalmitoyl-hexapeptide whose lipid
modification is common to lipoproteins of other spirochetes. In
this regard synovial V1 cells may function in an innate immune
response.
Characterization of the V1 T cell response to Ag by proliferation is
complicated somewhat by its dependence on exogenous IL-2, which has
been noted for other cell populations (37, 38, 43). The absolute
requirement for exogenous IL-2 as well as the constitutive expression
of IL-2Rß found on the majority of resting T cells (44) and
the failure of development of subsets in IL-2Rß-deficient mice
(45) suggest that synovial V1 cells may require the presence of Th
cells to provide IL-2. Further support for the idea that there are
differences in trophic requirements for cells is found in mice
genetically deficient for IL-7R, which are devoid of only the
T cell subset (46). We have not observed augmented proliferation of
V1 clones in the presence of IL-7 (K. Roessner, unpublished
observations). Despite these differences, activation of T cells,
even in response to unconventional ligands, clearly does reflect
engagement of the TCR (47).
Other measures of activation, such as CD25 expression (39, 48, 49) and
cytolytic activity (36), have been used to characterize the specificity
of T cells. Up-regulation of the cell surface markers of
activation used here, CD25 and FasL, was in close agreement with the
proliferative responses observed and showed greater indexes of
stimulation than did proliferation.
The contribution of accessory cells in human activation has been
controversial, with data supporting a requirement for T cells (43, 50),
B cells (51), and mononuclear phagocytes (52). For our clones,
DC derived from PBMC clearly serve as efficient accessory cells as has
previously been noted for allostimulation of naive T cells (53).
The responsiveness of the clones following fixation of
Borrelia-pulsed DC and the potent inhibition by specific
anti-TCR Ab suggest that V1 cells recognize Ag via the TCR.
While consistent with the idea that the V1 clones respond to
B. burgdorferi via their TCR, we cannot rule out the
possibility that the observed inhibition by anti- Ab results
from a negative signal delivered to the T cell clone rather than
mere blocking. However, the down-modulation of surface TCR by the
responding (CD25+) cells further supports the view
that Borrelia activation of cells is via the TCR.
The current findings do not distinguish whether the synovial V1
cells respond directly to a component of B. burgdorferi or
whether activation is indirect through up-regulation of another
molecule on APCs. Either result would be interesting. We found no
evidence for restriction of the V1 response by either classical MHC
molecules or group I CD1 molecules. Despite a decade of intensive
study, the characterization of TCR ligands and restriction
elements has yet to yield a conserved molecular pattern of recognition
as is seen for the majority of ß T cells. The best described
members of the human population are V9V2 T cells that
recognize mycobacterial prenyl pyrophosphates, synthetic alkyl
pyrophosphates, and nucleoside phosphates (24). Although the precise
restriction element involved is unknown, activation is dependent on
cell-cell contact (54). In contrast, for the remainder of the human
repertoire, the TCR ligands are not well characterized. A number
of studies have noted expansions of the V1 and other subsets of
T cells in infectious (37) and autoimmune diseases (13, 14, 15, 16)
without formally identifying the relevant ligands or the requirement
for TCR signaling. Studies using cloned T cells have in some cases
identified restriction elements for TCR recognition, which
include MHC class I and class II molecules (55), and the nonpolymorphic
CD1c molecule (56), but the majority of clones, like those from
Lyme synovial fluid, are without classical MHC restriction. More
recently, the class I MHC-like molecules, MICA and MICB, have
been demonstrated to be recognition units for human intestinal V1
cells (57). We are currently examining whether our synovial V1 cells
are responsive to MICA and MICB. If similar to intestinal V1 cells,
then conceivably Borrelia may stimulate synovial V1 cells
indirectly by up-regulating surface MICA and MICB. The observed lack of
specificity for the hexapeptide moiety among the synovial V1 clones
would be consistent with this possibility.
The recently recognized ability of certain ß T cells to recognize
lipids (58) and glycolipids (59) in the context of CD1b has provided a
possible explanation for the difficulty in characterizing TCR
specificity and restriction by conventional means for studying ß T
cells. Like the majority of T cells, CD1-restricted ß cells
are often CD4-CD8-. In this regard, it is
interesting that the minimal component of Borrelia to which
the clones respond is a lipohexapeptide, although we did not find
evidence supporting restriction of the Borrelia response by
the group I CD1 molecules. Another suggestion, based on the similarity
of the CDR3 length distribution to that of Abs, is that
TCR may not require a restriction element, but, rather, may recognize
aggregated Ag, as in the case of a murine clone specific for a
herpesvirus glycoprotein, gI (60). This seems unlikely for our synovial
clones because their activation requires the presence of both
Borrelia and APCs and also because fixation before the
addition of Borrelia abrogates the response. Additionally,
we have not observed a response to B. burgdorferi sonicate
when coated onto plastic (M. S. Vincent, unpublished
observations). Our findings also argue against a superantigen effect
accounting for the Borrelia stimulatory activity.
T cells have also been noted in synovial fluid from rheumatoid
arthritis (9, 10, 11), but their function there is unknown. Given that
these synovial cells are also of the V1 subset, their
function in rheumatoid arthritis may parallel their role in Lyme
arthritis synovium. In Lyme arthritis synovial fluid, the expansion of
the subset correlates with apoptosis of CD4+
synovial lymphocytes when cultured in vitro with B.
burgdorferi (12). The expanded subset in synovial fluid
expresses high and sustained levels of FasL, which mediates the
apoptosis of Fas-bearing CD4+ cells. We have observed
pronounced FasL expression on these clones for up to 6 days
following Borrelia stimulation (M. S. Vincent,
unpublished observations). The current findings demonstrate that the
up-regulation of FasL on V1 clones is actually a consequence of
activation by B. burgdorferi. This places the spirochete at
the initiation of a circuit that would result in the elimination of
Fas-sensitive T cells at the inflammatory site via activation.
In this way the T cell infiltrate could play a regulatory role
in the immune response by suppressing the CD4+ T cell
expansion. Consistent with this model are the reports of two groups
that collagen-induced arthritis (61) and adjuvant-induced arthritis
(62) are exacerbated when T cells are depleted.
Our observations suggest a potential role for the V1 subset of
synovial T cells in response to conserved spirochetal lipoproteins that
contain tripalmitoyl residues that are also known to activate
macrophages and monocytes (34). Although B. burgdorferi is
devoid of LPS (63), the lipoprotein N-terminal modifications in the
cell membrane contain similar broad immunostimulatory properties (64).
The minimal structures necessary for the induction of TNF-, IL-1ß,
IL-6, and IL-12 production by mononuclear phagocytes are the same
hexapeptide derivatives of N-terminal palmitoylated cysteine that
stimulate the synovial V1 clones (34). The conservation of these
structures in other spirochetes suggests that V1 cells may
participate in an innate immune response. As such, this may have
parallels with the recognition of nonprotein components of
Mycobacteria by V9V2 cells (23, 24) or of
glycosylceramides by NK1.1+ T cells (65).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ralph C. Budd, Immunobiology Program, University of Vermont College of Medicine, Burlington, VT 05405-0068.
3 Abbreviations used in this paper: DC, dendritic cells; EDCI, 1-ethyl-3-(3'-dimethyl-aminopropyl)-carbodiimide; FasL, Fas ligand; OspA, outer surface protein A; MIC, MHC Class I chain-related genes.
Received for publication January 20, 1998. Accepted for publication July 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in modulation of immunity to Borrelia burgdorferi in mice. J. Immunol. 155:2020.[Abstract]
1 subset of rheumatoid synovial fluid and peripheral blood T lymphocytes. Eur. J. Immunol. 22:567.[Medline]
T lymphocytes in synovia from rheumatoid arthritis patients with active synovitis. J. Clin. Immunol. 12:130.[Medline]
+ cell subsets in the synovial membranes of patients with rheumatoid arthritis and juvenile rheumatoid arthritis. Scand. J. Immunol. 36:533.[Medline]
T cells in Lyme arthritis. J. Exp. Med. 184:2109. T cells in cerebrospinal fluid and peripheral blood of patients with multiple sclerosis: reactivity, cytotoxicity, and T cell receptor V gene rearrangements. J. Immunol. 154:4883.[Abstract]
T lymphocytes in Graves’ disease specifically recognize thyroid epithelial cells. J. Immunol. 156:804.[Abstract]
1+ T-cells in systemic sclerosis patients. Ann. NY Acad. Sci. 756:382.[Medline]
receptor. N. Engl. J. Med. 324:877.[Abstract]
9/V2 T cell deletion and anergy to nonpeptidic mycobacterial antigens in asymptomatic HIV-1 infected persons. J. Immunol. 157:449.[Abstract]
V1+ T cell expansion in the peripheral blood of HIV-infected persons. J. Immunol. 154:1418.[Abstract]
1+ T cells proliferates in response to Epstein-Barr virus-transformed B cell lines in vitro. Scand. J. Immunol. 38:335.[Medline]
1+ T cells. Eur. J. Immunol. 24:3199.[Medline]
1+ subset of human gamma delta T cells responds to ligands expressed by EBV-infected Burkitt lymphoma cells and transformed B lymphocytes. J. Immunol. 149:3984.[Abstract]
9V2 cells are stimulated in a cross-reactive fashion by a variety of phosphorylated metabolites. Eur. J. Immunol. 25:2052.[Medline]
T cells. Nature 375:155.[Medline]
ß and T cells in immunity against an intracellular bacterial pathogen. Nature 365:53.[Medline]
T cells in primary infection with Listeria monocytogenes in mice. J. Exp. Med. 175:49. T cells and their role in resistance to bacterial infection. J. Exp. Med. 178:971. T cells and /ß T cells in tuberculosis. Eur. J. Immunol. 25:2877.[Medline]
T cells play an important role in hsp65 expression and in acquiring protective immune responses against infection with Toxoplasma gondii. J. Immunol. 155:244.[Abstract]
subunit. Science 238:682. T cell V1+ subset. J. Infect. Dis. 174:1309.[Medline]
2 cells in adult peripheral blood, but of V1 cells in cord blood. J. Immunol. 157:1613.[Abstract]
T-cell stimulation by microfluorimetry. Immunol. Lett. 53:135.[Medline]
9-expressing T cells by Mycobacterium tuberculosis: requirement for Th1-type CD4 T cell help and inhibition by IL-10. J. Immunol. 152:4984.[Abstract]
T lymphocytes by interleukin-2. Eur. J. Immunol. 23:2092.[Medline]
is essential for the development of + T cells, but not natural killer cells. J. Exp. Med. 184:289.2V2 TCR-dependent recognition of non-peptide antigens and Daudi cells analyzed by TCR gene transfer. J. Immunol. 154:998.[Abstract]
T cells. J. Immunol. 151:1312.[Abstract]
T cells by a subset of memory helper T cells. Infect. Immun. 63:1211.[Abstract]
T cells and B cells regulating IgG production. Immunology 84:105.[Medline]
T cells activated by live Mycobacterium tuberculosis. Infect. Immun. 60:3480. T cell activation: role of accessory molecules in alloactivation. J. Clin. Invest. 95:296.
T cells. Immunity 3:495.[Medline]
/ cells. Annu. Rev. Immunol. 11:637.[Medline]
cells. Eur. J. Immunol. 20:703.[Medline]
T cells. Science 279:1737.ß+ T cells. Nature 372:691.[Medline]
cell. J. Immunol. 152:5392.[Abstract]
T cells in murine collagen-induced arthritis. J. Immunol. 151:6546.[Abstract]
T cells does not prevent or ameliorate, but rather aggravates, rat adjuvant arthritis. Arthritis Rheum. 39:204.[Medline]
14 NKT cells by glycosylceramides. Science 278:1626.This article has been cited by other articles:
|
|
 |
|
  C. Collins, C. Shi, J. Q. Russell, K. A. Fortner, and R. C. Budd Activation of {gamma}{delta} T Cells by Borrelia burgdorferi Is Indirect via a TLR- and Caspase-Dependent Pathway J. Immunol., August 15, 2008; 181(4): 2392 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Collins, J. Wolfe, K. Roessner, C. Shi, L. H. Sigal, and R. C. Budd Lyme Arthritis Synovial {gamma}{delta} T Cells Instruct Dendritic Cells via Fas Ligand J. Immunol., November 1, 2005; 175(9): 5656 - 5665. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Roessner, J. Wolfe, C. Shi, L. H. Sigal, S. Huber, and R. C. Budd High Expression of Fas Ligand by Synovial Fluid-Derived {gamma}{delta} T Cells in Lyme Arthritis J. Immunol., March 1, 2003; 170(5): 2702 - 2710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Huber, C. Shi, and R. C. Budd {gamma}{delta} T Cells Promote a Th1 Response during Coxsackievirus B3 Infection In Vivo: Role of Fas and Fas Ligand J. Virol., June 5, 2002; 76(13): 6487 - 6494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Steere Lyme Disease N. Engl. J. Med., July 12, 2001; 345(2): 115 - 125. [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Mbow, N. Zeidner, R. D. Gilmore Jr., M. Dolan, J. Piesman, and R. G. Titus Major Histocompatibility Complex Class II-Independent Generation of Neutralizing Antibodies against T-Cell-Dependent Borrelia burgdorferi Antigens Presented by Dendritic Cells: Regulation by NK and {gamma}{delta} T Cells Infect. Immun., April 1, 2001; 69(4): 2407 - 2415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Haake Spirochaetal lipoproteins and pathogenesis Microbiology, July 1, 2000; 146(7): 1491 - 1504. [Full Text]
|
|
|
|
 |
|
 L. Hviid, B. D. Akanmori, S. Loizon, J. A. L. Kurtzhals, C. H. Ricke, A. Lim, K. A. Koram, F. K. Nkrumah, O. Mercereau-Puijalon, and C. Behr High frequency of circulating {gamma}{delta} T cells with dominance of the V{delta}1 subset in a healthy population Int. Immunol., June 1, 2000; 12(6): 797 - 805. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Sellati, D. A. Bouis, M. J. Caimano, J. A. Feulner, C. Ayers, E. Lien, and J. D. Radolf Activation of Human Monocytic Cells by Borrelia burgdorferi and Treponema pallidum Is Facilitated by CD14 and Correlates with Surface Exposure of Spirochetal Lipoproteins J. Immunol., August 15, 1999; 163(4): 2049 - 2056. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
