| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
24-Invariant Natural Killer T Cells in T-Cell-Reactive Leprosy Together with a Highly Biased T Cell Receptor V Repertoire
From the Institut Pasteur,*
Unité de Biologie
Moléculaire du Gène, INSERM U277, Département
d’Immunologie, Paris; and the Institut de Recherche sur la
Peau,
INSERM U312, l’Hôpital
St.-Louis, Paris, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
repertoire in the two diseases. In addition to V24+
NKT cells, all patients with T-cell-reactive leprosy showed a
very restricted T-cell-reactive V repertoire with a strong bias
toward the use of the V6 and V14 segments. V6 and
V14+ T cells were polyclonal in terms of CDR3 length and
J usage. In contrast, most sarcoidosis patients showed a
diverse usage of V chains associated with clonal or oligoclonal
expansions reminiscent of antigen-driven activation of conventional T
cells. Thus the origin and perpetuation of the two kinds of
granulomatous lesions appear to depend on altogether distinct T-cell
recruiting mechanisms.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Pathologically, granulomatous reactions highly similar to the
T-cell-reactive forms of leprosy are found in diseases of unknown
origin, such as sarcoidosis. Many cell types of the innate and adaptive
immune systems participate in the elaboration of cutaneous granuloma
formation in both diseases, with the typical feature of a central
accumulation of histiocytes surrounded and infiltrated by lymphocytes.
The lymphocytic rim is composed of ß, 
T lymphocytes and
occasionally NK cells. These cells invade the site of granulomatous
inflammation after a variety of stimuli, including antigenic ones, such
as mycobacterial peptides, lipoproteins, and
glycolipids,1-8
or chemoattraction by cytokines and
chemokines.9-11
In view of the pathological similarity of
the two kinds of diseases, a common etiology has been suggested,
inasmuch as sarcoidosis represents the inflammatory response to an as
yet unidentified microorganism, most probably of mycobacterial
origin.12-14
The causative association of mycobacteria
and sarcoidosis is still a focus of controversial
discussions.15,16
Over recent years the T cell response to mycobacteria has been extensively studied and found to display original features; the response to various antigens, and particularly to mycobacterial glycolipids, is well documented.17,18 In an animal model, NKT cells, a CD1d-restricted T-cell subpopulation with very unique features,19 have been found to be recruited by mycobacterial glycolipids.20 Moreover, in this model, NKT cells are needed for the granulomatous lesions to develop. These NKT cells, however, have to our knowledge not yet been identified in human mycobacteria-induced granulomas.
One way to test the "mycobacterial" origin of sarcoidosis is to analyze and compare the T-cell populations and the T-cell receptor usage in cutaneous mycobacteria-induced granulomas such as leprosy, and in the cutaneous granulomatous lesions of sarcoidosis, as performed for pulmonary patients in earlier studies.21-23 As cutaneous manifestations are currently found in a subgroup of sarcoidosis patients, the comparison of cutaneous sarcoidosis and leprosy lesions offers a convenient model for the study of the characteristics of granulomas in identical tissue compartments and without the difficulties of pulmonary specimen sampling or the bias of using bronchoalveolar lavage fluids.
Using a combination of immunohistochemistry, polymerase chain reaction (PCR)-based technology, and DNA sequencing, we have analyzed the T-cell population in human leprosy patients and compared them to cutaneous sarcoidosis. The present data lead to the conclusion that the two kinds of granulomas, although histologically similar, are composed of different T-cell populations and argue against a mycobacterial origin of sarcoidosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The diagnosis of leprosy and sarcoidosis was confirmed by clinical
and histological criteria. Leprosy patients were classified according
to the criteria of Ridley and Jopling into polar tuberculoid (TT),
borderline tuberculoid (BT), polar lepromatous (LL), borderline
lepromatous (BL), and midborderline (BB).24
A total of
five patients with tuberculoid leprosy (TT, BT), three patients with
lepromatous leprosy (LL, BL), two patients with the reversal form of
leprosy (RR), and six patients with cutaneous sarcoidosis were included
in the study. Furthermore, the skin biopsy of one healthy individual
undergoing plastic surgery served as the control for unaffected skin.
After informed consent was obtained, 4-mm lesional punch biopsies were
taken in local anesthesia and snap-frozen in liquid nitrogen, and
specimens were cut into two fragments for immunohistochemistry and
molecular biology analysis, respectively. A summary of the included
patients is shown in Table 1
.
|
Frozen sections were analyzed for the expression of CD3, CD4, CD8,
CD16, CD56, CD57, CD68, panTCRß, and panTCR (Dako, Trappes,
France); CD1a, CD1b, CD1c, CD83, and V24 (Immunotech, Marseille,
France); and CD1d (Biosource, Camarillo, Calif., USA), using
commercially available monoclonal antibodies, and a streptavidin-biotin
immunoperoxidase technique, using 3-amino-ethylcarbazole to produce
positive brownish staining (Beckman-Coulter, Paris, France). For
staining with anti-V24 we used a noncoupled primary antibody,
followed by a biotin-labeled secondary antibody to reduce nonspecific
background staining.
The percentage of positive cells was calculated by determining the number of stained cells among the total number of cells within the granulomas (eg, hematoxylin-stained nuclei).
As control samples, healthy skin from patients undergoing plastic breast surgery were used to test the staining pattern in unaffected skin.
RNA Extraction and cDNA Preparation
Specimens were disrupted in Trizol (Gibco) with a Polytron homogenizer. RNA extraction and cDNA preparation were carried out according to standard protocols, using AMV reverse transcriptase (RT) from Boehringer Mannheim (Mannheim, Germany).
TCR -Chain Analysis
The "Immunoscope" technique for determining complementary
region 3 (CDR3) length and distribution has been described
elsewhere.25-27
Briefly, standardized amounts of cDNA
(ie, the product of the reverse transcription of 10 µg of total RNA)
were PCR amplified, using each of the 29 V-specific probes and a
common C-specific probe. Each V-C PCR product was analyzed by
electrophoresis in an agarose gel. For CDR3 length diversity,
PCR-amplified products were submitted to five cycles of primer
extension, using an internal, fluorescent, C-specific probe. The
labeled material was loaded on a sequencing gel and analyzed with an
automatic sequencer (Applied Biosystem) equipped with a computer
program (Immunoscope; Applied Biosystems) that enables the
determination of the intensity of fluorescence of each band as well as
its actual size. The results are depicted as peaks, the surfaces of
which are proportional to the amount of material and the locations of
which are dictated by the length of the CDR3 region. The size
distribution of the V-C is Gaussian in the case of nonactivated
or polyclonally activated lymphocytes, whereas proliferating T cells
generated a non-Gaussian distribution with amplified peaks
corresponding to clones with a definite CDR3 length within a V-C
combination.
Primer and PCR Conditions and Detection of Invariant V Chains
To amplify the V repertoire of the invaded T cells we used a
panel of 29 V-specific primer together with a
C-specific primer as well as an internal C run-off primer, which
have previously been described.26
The run-off primer for
the J33 (5'-CCAGATTAACTGATAGTTGCTATC-3'), the J29
(5'-AAGAGGTGTGTTTCCTACGTC-3'), the J48
(5'-TAATTTCTCATTTCCAAAGTT-3'), and the J18
(5'-GCCTCCCCAGGGTTGAGCCTCTG-3') clonotypes were designed according to
the publications of Porcelli et al and Han et al.28,29
Primers for human CD3 and CD1d were as follows: CCAGGCTGATAGTTCGGTGA
(CD3–5') and TGTCTGAGAGCAGTGTTCCCAC (CD3–3') and AGCCTGTATGGGTGAAGTGG
(CD1d-5') and TAAAGCCCACAATGAGGAGG (CD1d-3'). For the primary PCR
reaction cDNA was amplified in 40 cycles (30' 94°—30' 60°—30'
72°) followed by five cycles (30' 94°—30' 60°—30' 72°) with
the fluorescent run-off primer. To exclude a higher affinity of
individual primers as a reason for a positive amplification product,
various primers were tested in a kinetics of 20/25/30/35 and 40 cycles
with a positive control of peripheral blood mononuclear cell
(PBMC) cDNA of a healthy individual in which all V chains had
been found to be amplified.
Direct DNA Sequencing
Direct sequencing of PCR products was carried out with a Sequenase kit (USB-Amersham, Cleveland, OH) according to the manufacturer‘s instructions, after separation of the PCR products by electrophoresis in a 2% agarose gel and electroelution of the proper DNA fragments.
Cloning and Sequencing of PCR Products
PCR products were ligated into a commercially available vector (Zero blunt; Invitrogen, Groningen, the Netherlands) and transformed into Escherichia coli (Invitrogen). Sequence analysis was carried out according to standard protocols (Perkin Elmer) and analyzed with ABI-Prism 373 software (Applied Biosystems).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Biopsies were frozen, sectioned, and immunostained. We first
determined the number of macrophages, dendritic cells, and NK
lymphocytes, using simple staining for CD68, CD1a, CD16, CD56, CD57,
and CD83 molecules, respectively. The staining pattern showed
comparable numbers of cells within the T-cell-reactive forms of leprosy
and sarcoidosis patients, whereas lepromatous leprosy patients showed
no or only weak staining for the analyzed cell types, with the
exception of CD68, which stained positive in all lepromatous leprosy
patients, confirming the presence of macrophages within the lesions
(data not shown). In addition, comparable numbers of T cells as
well as a similar ratio of CD4+ to
CD8+ cells were detected in TCR leprosy as well
as in sarcoidosis. Lepromatous leprosy patients showed only poor
infiltration by any of the investigated T-cell subpopulations. The
identified T cells dominated in both types of granuloma (sarcoidosis
and leprosy) in the areas surrounding the central eptheloid cell
accumulation. The immunochemical features of the samples under study
are in agreement with previous findings showing that the composition of
granuloma infiltrating immunocytes in cutaneous sarcoidosis is similar
to that of T-cell-reactive leprosy.4
In view of the involvement of CD1 molecules in the response to
mycobacterial antigens, the expression of CD1a–d molecules was
examined. Few or no CD1a+,
CD1b+, CD1c+, or
CD1d+ cells were found in lepromatous leprosy
patients, but remarkable levels of expression of CD1a, b, c were found
in all TCR leprosy patients, as reported in a previous
publication.30
To our surprise, even higher numbers of
CD1b+ cells and a remarkable expression of the
other CD1 molecules was found within the investigated sarcoidosis
patients (Figures 1 and 2)
. The positive staining cells in both
diseases accumulated within the dermal granulomas, with the exception
of CD1a, which, in addition, constantly stained positive for epidermal
cells, most probably because of the presence of epidermal dendritic
cells. This epidermal staining pattern was also seen in healthy control
biopsies. CD1b and CD1c stained positive for epidermal cells only
occasionally, without showing associations with one of the investigated
disorders. Some of the biopsies showed positive staining of CD1d, also
in the basal layer of the epidermis, suggesting an expression of this
molecule by human keratinocytes, as described earlier,31
but this pattern was seen for leprosy and sarcoidosis patients and was
also found in the uninvolved skin of healthy control subjects. The
positive staining pattern was paralleled by the detection of CD1d-RNA
by RT-PCR in all sarcoidosis and TCR leprosy patients (data not shown).
|
|
NKT Cells Are Present in TCR Leprosy but Not in Sarcoidosis Patients
Human NKT cells are T cells with NK cell surface markers, using an
invariant V24-J18 TCR chain, associated preferably with the
Vß11 chain and restricted by the MHCIb CD1d molecule. Their murine
V14-J281 counterparts have been shown to be necessary for the
development of the granulomatous lesions that follow the injection of
deproteinized mycobacterial cell walls.20
All samples were assayed for the presence of
V24+ transcripts, followed by the research of
the NKT-specific, V24-J18 transcripts.28
The
V24-C PCR products were studied by primer extension, using a
fluorescent C-specific probe. Patients T1, T5, R1, and R2 displayed
a Gaussian-like distribution of their CDR3 length; patients T2-T4
patients displayed a unique peak with a size compatible with that of
the invariant chain of the TCR chain of NKT cells (Figure 3a)
. On primer extension with the
V24-J18 clonotypic probe, all TT and RR samples showed the
presence of NKT cells. The resulting peak was superimposable on the
V24-C products in patients T2-T4, showing that NKT cells are
predominant among the V24+ T cells of these
patients.
|
24-C
amplification product, but we could not detect the clonotypic
V24-J18 transcript. Patients S2, S3, S4, and L1–L3 did not show
any V24-C amplification signal and were also found to be negative
in the V24-J18 runoff experiments.
We then stained tissue sections, using a V24-specific monoclonal
antibody raised against a V24-J18+ DN
T-cell clone.32,33
No positive cells were found in any of
the sarcoidosis or LL specimens, whereas two T-cell-reactive leprosy
patients also scored positive in immunohistochemistry. Figure 3b
shows
the staining pattern for V24+ in patient R2.
To confirm the presence of NKT cells seen in the immunoscope analysis,
we then cloned and sequenced the V24-C transcripts in all seven
T-cell-reactive leprosy patients. Whereas we could find in all patients
the canonical sequence V24-TGT GTG GTG AGC GAC
AGA-J18, patient T3 showed, in addition to this sequence, a
V24-TGT GTG GTG TCT GAC AGA-J18 rearrangement in which
the serine at the V-J junction was encoded by a triplet (TCT)
other than the germline transcribed triplet AGC, which is usually
associated with the NKT cell population.28
This finding
argues for a TCR-mediated selection process within the lesion of this
patient.
Thus classical V24-J18 NKT cells are present in T-cell-reactive
leprosy lesions but are absent from cutaneous sarcoidosis lesions.
TCR -Chain Usage in Leprosy Lesions
To further define the T cells detected by immunohistochemistry, we
examined their TCR -chain usage with a panel of 29 different primers
covering almost 95% of the described V families.34
The
analysis of the chain was preferred to that of the ß chain
because it allows simultaneously the characterization of the T-cell
repertoire together with the screening for possible invariant TCR
chains.
In addition to the V24 segment, which we have shown to be mostly
associated with NKT cells, all seven T-cell-reactive leprosy patients
showed the preferential usage of either V6 and/or V14 (Figure 4)
. On Immunoscope analysis of the
V6/14 -C PCR products with a C-specific primer, a
Gaussian-like or, in some instances, a comb-like distribution of the
TCR -chain CDR3 profile was observed. The absence of dominant peaks
is indicative of a polyclonal origin of the infiltrating T cells.
|
6- and V14-C PCR products in
various patients were cloned and sequenced to examine recurrency in
their CDR3 region sequence (including the presence of invariant
chains) or biases in the usage of J segments.
We did not find invariant sequences within the analyzed clones;
however, the DNA sequences showed a bias of the associated J,
because J segments were preferentially associated with V6 and
others were more often found to be associated with V14. They were
segments J 4, 5 with V6 and J 30, 31, 32, 33, 34, 48, and 49
with V14. The J 1, 2, 14, 18, 24, 25, 28, 35, 36, 38, 46, 50, 51,
59, 60, 61 segments were not detected among the 220 sequences we have
determined within the different leprosy patients (Figure 5A)
.
|
6 in patients T1, T2, and T5) could indicate a dominant
sequence within an otherwise polyclonal pattern. Moreover, one
rearrangement sequence (V6-J5) was found to be quite similar in
CDR 3 size and amino acid sequence in patients T1 and T2, which could
be suggestive of a common antigenic structure that selected the similar
clones in the two patients (Table 2)
.
|
ß-chain usage.35
However, the V7.2-J33, the
V4-J29, and the V19-J48 rearrangements have been described
to be preferentially associated with DN or
CD4-/CD8+ T
cells28,29,36
As immunohistochemical analysis of the DN
population was not found to be a suitable screening technique, we
decided to search for the described invariant TCR rearrangements by
PCR and Immunoscope techniques. Fluorescent primers coding for the
J33, J29, and J48 segments were designed and used for runoff
experiments of the V7-C, V4-C, and V19-C PCR
products. No signal was observed in the material recovered from the
T-cell-reactive leprosy lesions, showing the absence of V-invariant
T cells other than the V24-invariant NKT cell population.
Thus the local ß T-cell response in cutaneous leprosy consists
chiefly of NKT cells and oligoclonal/polyclonal V6/V14 T cells.
TCR -Chain Usage in Cutaneous Sarcoidosis Lesions
All biopsies of cutaneous sarcoidosis granulomas were studied in a
similar way. We have already shown the absence of NKT cells by RT-PCR.
On agarose gel electrophoresis of the V-C PCR products, and in
contrast with what we observed in leprosy, all patients with cutaneous
sarcoidosis showed the usage of a limited number of individual V
chains (except patient S1, who yielded usage of 28 of the screened 29
TCR chains) within their granulomatous lesions, without any
markable bias toward the usage of a particular V region. It is worth
noting that, as previously reported for pulmonary
specimens,37
we found in five of six patients an
amplification for V2 in cutaneous sarcoidosis lesions (Figure 4)
.
On immunoscope analysis of the distribution of the CDR3 length of the
different TCR chains in all patients, we found Gaussian-like
distributions of the V-C peaks, representing polyclonal
expansions of the infiltrating T lymphocytes, as well as distorted
profiles with dominant peaks, for V chains that were different in
each patient and indicated the presence of T-cell populations with a
conserved length of the CDR3 region of the TCR chain, correlating
with oligo- or monoclonal expansions. In this respect, the patterns
observed in cutaneous sarcoidosis are very reminiscent of those
observed in a number of human diseases involving antigen-driven T-cell
proliferation.27
We determined the sequence of the most dominating expansions in each
patient, either by direct sequencing (which proved successful in
patients S1 and S4) or following the cloning of the V-C PCR
product and determination of the expanded sequence (S2, S3, S5, and S6)
by sequencing several clones. In four patients (S1, S2, S3, and S4) a
single sequence of the CDR3 region typical for clonal expansions was
recurrently determined. The sequences obtained from the different
patients showed no similarities at the amino acid level, even
when the same V was studied (V9 for patients S1 and S2) (Table 3
).
|
20 in patient S5
and within the 10-aa expansion of V16 in patient S6. The comparison
of the CDR3 amino acid sequences, however, was reminiscent of the
recognition of the same antigen in a given patient: in patient S5 the
CDR3 sequences had a very similar sequence containing a CLVG-X-X-DTG
motif (16 of 19) or a motif in which D-T was replaced by another
hydrophilic amino acid (S or G); in addition, 10 of 19 of the 10-aa
CDR3-containing sequences were found to include the J5 chain.
Although the similarities were less striking in the CDR3 region
analysis of the V16-C amplification in patient S6, we still found
some common features shared by most amino acid sequences (Table 3)
.
To compare for possible similarities between the T cells invading
sarcoidosis and leprosy at a molecular level, we extended the sequence
analysis of the CDR3 region in the V6+ and
V14+ T-cell populations to the sarcoidosis
patients showing infiltration of the respective chains, who were S1,
S2, and S5 for V6 and S1, S5, and S6 for V14. This analysis
showed, in parallel to the T-cell-reactive leprosy patients, an
oligoclonal pattern with few clones on expansion in an otherwise
polyclonal composition of the V6+ or
V14+ T-cell populations (Table 2)
. When
comparing for the association with various J chains, we did not find
the same segments preferably used in cutaneous sarcoidosis as in
leprosy (eg, J33 was not used at all in sarcoidosis but was used in
several leprosy specimens); still a preferred pairing of the two
analyzed TCR chains could be observed (Figure 5B)
.
Using the Immunoscope technique and the V7.2-J33-, the
V4-J29-, and the V19-J48-specific clonotypic primers, we
found that patients S1 and S5 showed a clonotypic peak of 190 bp
(indicative of the V7.2-J33 rearrangement), and S1 showed the
196-bp peak indicative of the V19-J48 rearrangement. The
canonical V4-J29 rearrangement was not detected in any of the
patients. To rule out a bias on the level of J usage, we looked for
transcription of the J segments 18 and 33 in all identified V
chains. These experiments showed that the use of the segments J18,
J29, J33, and J48 was not restricted to the populations of
V4+, V7+,
V19+, or V24+ cells
as identified by runoff experiments (data not shown) with all
identified V chains in the sarcoidosis patients and sequence
analysis of the V6 and V14 populations (Table 2)
.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The samples we have collected possess the immunohistochemical features
already described4
and are thus characteristic of the
different forms of leprosy as well as of cutaneous sarcoidosis, and the
cell content of the sarcoidosis lesions mimics that of T-cell reactive
leprosy. However, the T-cell responses in these two families of
diseases, analyzed by a combination of RT-PCR, Immunoscope, and
systematic DNA sequencing carried out on material with similar T-cell
content, appear to be profoundly different. First, NKT cells are
present among the T cells infiltrating T-cell-reactive leprosy lesions
but are absent from sarcoidosis lesions. Second, the immune response in
leprosy is dominated by a polyclonal V6 and V14 usage, whereas
the immune response in cutaneous sarcoidosis is dominated by a mixture
of polyclonal infiltrates associated with monoclonal or oligoclonal
cell expansions, reminiscent of an antigen-driven proliferation of T
cells and profoundly different from the patterns observed in the
presence of mycobacteria. The oligoclonality we have observed in
cutaneous sarcoidosis patients correlates well with previous analyses
of pulmonary granulomas,22,38,39
although a restricted
repertoire has not been found in all studies.40
Oligoclonality cannot be due to the limited amount of material,
inasmuch as we have observed a large panel of V segments in patient
S1, who otherwise did not show higher percentages of invading T cells;
moreover, the Immunoscope technique has been shown to detect one cell
in 105
cells,41
a sensitivity
sufficient to identify even very minor lymphocytic populations.
However, biases in the cutaneous TCR repertoire may be associated with
the processes of homing to inflamed skin, which are highly complex and
require a variety of cellular interactions favoring activated and/or
antigen-specific T-cell populations.42
In this respect,
Klein and co-workers showed a restricted Vß usage with oligoclonal
expansions within the sites of Kveim reagent injections, a diagnostic
reaction appearing 4 weeks after intradermal injection of extracts of
sarcoidosis spleen or lymph nodes, which leads to granuloma formation
virtually identical to that of primary cutaneous
sarcoidosis.43
The expansions identified in our patients
were seen within different V-C expansions and revealed CDR3
shapes with only limited similarities between the clones present in
different individuals, thus suggesting an individual predisposition to
a yet unidentified antigenic stimulus. This idea is further supported
by the observations we made in patients S5 and S6, where the
peak-shaped sequences were not identical but contained a limited number
of similar sequences, which could well have been stimulated by a single
antigen. In conclusion, our results suggest that altogether different
mechanisms underlie the development of granulomas in T-cell-reactive
leprosy and reversal leprosy and in cutaneous sarcoidosis, and this may
be taken as meaning that (atypical) mycobacteria do not contribute to
the progression of cutaneous sarcoidosis. Moreover, were it to be
proved that the observed expansions are locally induced by some
antigen-driven process, the presence of oligoclonally expanded T cells
in the lesions leads to the conclusion that a systematic search for
sarcoidosis-associated antigens, whether of self or nonself origin,
could be a fruitful alternative approach to the etiology of the
disease.
The T-cell response in leprosy lesions is worth discussing. In the
investigated patients with leprosy we found the well-known differences
in APC and T-cell infiltration between the T-cell-reactive forms with a
high immunocellular content and the lymphocyte-poor form of lepromatous
leprosy. This expected finding represents one of the criteria for the
classification of leprosy patients.24
When analyzing the
V repertoire of the patients with TT and RR, we found a striking
bias of the chains used versus V6, V14, and
V24. Because major differences in the efficiency of the primers used
were ruled out,34
only three V segments are
predominantly used by the T cells infiltrating T-cell-reactive leprosy
patients. The strong bias versus V6 and V14 was
somehow surprising, although Strohal had shown in earlier studies that
T cells using these two V segments are preferentially found in
normal skin.44
The V6 and V14 T cells of all lesions
were polyclonally expanded in terms of CDR3 length, amino acid
sequence, and J usage. No invariant chain could be identified on the
basis of identical recurrent sequences and unique J usage. The
polyclonal expansion of cells using solely V6 and 14 is difficult to
account for. The bias in the Vß usage (Vß 6,12,14,19) previously
described in leprosy patients45
associated with the
currently reported selective usage of V6 and V14 (two
segments that are structurally related)46
may reveal a
particular response to a nominal mycobacterial antigen. However, one
would expect a highly skewed diversity within the expanded chains
that is due to an antigen-driven process or to preferential selection
of an invariant chain and thus a constant J usage, none of which
was observed in the present studies. The usage of V6 and V14
segments may also reveal the stimulation of T cells by some TCR
-specific superantigen secreted by proliferating mycobacteria as
described for TCR ß chains.47
As TCR -specific
superantigens have not yet been identified, this remains questionable,
but the contribution of the chain to superantigen recognition is
currently under debate, with some authors favoring an important role in
TCR-superantigen interaction.48,49
The analysis of the V24+ cells showed, in most
of our T-cell-reactive leprosy patients, a disturbed distribution of
the Immunoscope profile with an expanded peak corresponding to the CDR3
length of the known V24-J18 clonotype. The presence of this
clonotype, which is typical of human NKT cells, was further confirmed
by J18-specific runoff experiments and sequence analysis, which
showed an amplification product in all T-cell-reactive but not in
lepromatous leprosy patients. The presence of NKT cells in leprous
granulomas had not previously been described, and the presence of this
distinctive T-cell subset is clearly associated with T-cell-reactive
forms of leprosy. The finding that the invading NKT cell population at
least in one patient displayed differences at the nucleotide level in
their V24-J18 transcript with an identical sequence at the amino
acid level enforces the idea that NKT cells in the leprous granulomas
have undergone a selection process.
The V24-J18 TCR invariant NKT cells are restricted by the CD1d
molecule and have a homolog in the murine V14-J281 TCR invariant
NKT cells.50
These cells have been found to carry
markers of the NK lineage, such as CD161 or NK1.1 together with the
ßTCR. They are predominantly found in the DN or
CD4+CD8- pool. The
physiological role of the NKT cells in humans and mice is pleiotropic;
they have been implicated in the induction of a Th2 immune
response,51
the control of autoimmune
disease,52
and an antitumoral activity after activation
via interleukin-12 or specific ligands.53-55
In a murine
model of granuloma development by mycobacterial cell wall components,
these cells have recently been demonstrated to be almost exclusively
recruited by glycolipids belonging to the phosphatidylinositolmannoside
family and to be of crucial importance for granuloma
formation.20
However, an activation of NKT by
CD1d+ cells supplemented with mycobacterial cell
wall extracts has not yet been observed in vitro, and the
nature of the activation process of NKT cells remains elusive. It is
possible that the mycobacterial cell wall-induced NKT-cell-dependent
granuloma formation is not dependent on mCD1; an alternative activation
could be induced through lectin-like structures that are common to NK
and NKT cells.56
NKT cells were found in all patients with
T-cell-reactive leprosy but not in lepromatous leprosy patients who
lack a sufficient cellular immune response against M.
leprae. They were not found in sarcoidosis patients, although CD1
molecules were detectable at high levels in the lesions. In any event,
the formation of a granuloma in the absence of an obvious bacterial
load, as in cutaneous sarcoidosis, clearly does not require the local
recruitment of NKT cells. The identification of two further T-cell
rearrangements previously described with DN T cells (V7.2-J33 and
V19-J48) within the sarcoidosis lesions is difficult to explain.
As no functional role for these TCR-invariant cells has been
described, their presence might be due to a yet unknown specific
mechanism. However, as we found the V7.2-J33 rearrangement in
only two of six and the V19-J48 rearrangement in only one of six
patients, we would favor a nonspecific infiltration into the inflamed
skin.
In contrast, the recruitment of NKT cells in T-cell-reactive leprosy strengthens the hypothesis that NKT cells have an important role in the innate immune response toward substances produced by mycobacteria in the present case and by intracellular bacteria in general.57,58
|
|
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by grants from La Ligue Nationale centre le cencer l’Association pour la Recherche contre le Cencer, the European Community, and the College de France.
M.M. was supported by a scholarship from the Deutsche Forschungsgemeinschaft.
Accepted for publication May 19, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Peterfalvi, E. Gomori, T. Magyarlaki, J. Pal, M. Banati, A. Javorhazy, J. Szekeres-Bartho, L. Szereday, and Z. Illes Invariant V{alpha}7.2-J{alpha}33 TCR is expressed in human kidney and brain tumors indicating infiltration by mucosal-associated invariant T (MAIT) cells Int. Immunol., December 1, 2008; 20(12): 1517 - 1525. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Iannuzzi, B. A. Rybicki, and A. S. Teirstein Sarcoidosis N. Engl. J. Med., November 22, 2007; 357(21): 2153 - 2165. [Full Text] [PDF]
|
|
|
|
 |
|
 J. Grunewald and A. Eklund State of the Art. Role of CD4+ T Cells in Sarcoidosis Proceedings of the ATS, August 15, 2007; 4(5): 461 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Im, N. Tapinos, G.-T. Chae, P. A. Illarionov, G. S. Besra, G. H. DeVries, R. L. Modlin, P. A. Sieling, A. Rambukkana, and S. A. Porcelli Expression of CD1d Molecules by Human Schwann Cells and Potential Interactions with Immunoregulatory Invariant NK T Cells J. Immunol., October 15, 2006; 177(8): 5226 - 5235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kobayashi, Y. Kaneko, K.-i. Seino, Y. Yamada, S. Motohashi, J. Koike, K. Sugaya, T. Kuriyama, S. Asano, T. Tsuda, et al. Impaired IFN-{gamma} production of V{alpha}24 NKT cells in non-remitting sarcoidosis Int. Immunol., February 1, 2004; 16(2): 215 - 222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Illes, M. Shimamura, J. Newcombe, N. Oka, and T. Yamamura Accumulation of V{alpha}7.2-J{alpha}33 invariant T cells in human autoimmune inflammatory lesions in the nervous system Int. Immunol., February 1, 2004; 16(2): 223 - 230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Gansert, V. Kiebler, M. Engele, F. Wittke, M. Rollinghoff, A. M. Krensky, S. A. Porcelli, R. L. Modlin, and S. Stenger Human NKT Cells Express Granulysin and Exhibit Antimycobacterial Activity J. Immunol., March 15, 2003; 170(6): 3154 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ingen-Housz-Oro, A. Bussel, B. Flageul, L. Michel, L. Dubertret, P. Kourilsky, G. Gachelin, H. Bachelez, and P. Musette A prospective study on the evolution of the T-cell repertoire in patients with Sezary syndrome treated by extracorporeal photopheresis Blood, August 28, 2002; 100(6): 2168 - 2174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mempel, C. Ronet, F. Suarez, M. Gilleron, G. Puzo, L. Van Kaer, A. Lehuen, P. Kourilsky, and G. Gachelin Natural Killer T Cells Restricted by the Monomorphic MHC Class 1b CD1d1 Molecules Behave Like Inflammatory Cells J. Immunol., January 1, 2002; 168(1): 365 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ronet, M. Mempel, N. Thieblemont, A. Lehuen, P. Kourilsky, and G. Gachelin Role of the Complementarity-Determining Region 3 (CDR3) of the TCR-{{beta}} Chains Associated with the V{{alpha}}14 Semi-Invariant TCR {{alpha}}-Chain in the Selection of CD4+ NK T Cells J. Immunol., February 1, 2001; 166(3): 1755 - 1762. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), which is
suggestive of clonal populations. T-cell-reactive leprosy patients
(T, R), however, showed a
bias toward V
), which is
suggestive of poly- or oligoclonal expansions. The V