| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 and V2 Cells in Human Peripheral Blood Mononuclear Cells: TCR Selection Is Not Altered by Stimulation with Gram-Negative Bacteria
*
Pathology Research Laboratory, Massachusetts General Hospital, Charlestown, MA 02129;
Chair of Semeiotica Medica, Catholic University, Rome, Italy;
Vincent Memorial Obstetrical Service, Massachusetts General Hospital, Boston, MA 02114
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells, there
are several observations which suggest that the expressed
repertoire in the periphery of normal individuals is often quite
restricted. To assess selective expansions among T cells from
both adult and newborn blood samples, PBMC from 12 normal adults and
cord blood from 15 normal newborns were analyzed for TCRDV1 and TCRDV2
junctional diversity by CDR3 size spectratyping and single-strand
conformational polymorphism. Although TCRBV usage showed extensive
heterogeneity in adults and newborns, both populations often showed
CDR3 region restriction for TCRDV1 and TCRDV2. Analysis of the CDR3
spectratype patterns of newborn twins suggested that clonal selection
for TCRDV is independent of genetic background. The possible role of
Gram-negative bacteria in driving selective responsiveness of T
cells in PBMCs from adults was examined by in vitro stimulation with
Escherichia coli and Pseudomonas aeruginosa.
Donors whose TCRDV repertoire was highly clonal in the unstimulated
blood cells showed the same predominant clones among the
bacteria-stimulated cultures. In individuals whose T cells were
less restricted, in vitro stimulation did not select for clonality;
rather, the TCRDV repertoires were similar before and after bacterial
stimulation. Together, these data indicate that T cells are
often clonally restricted in adults as well as in newborns and suggest
that the prominent stimulatory activity of Gram-negative bacteria does
not by itself account for the restriction or diversity of the T
cell repertoire. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells (1, 2, 3, 4, 5). The responding T
cells in the peripheral blood of normal adults are almost exclusively
of the V2+ subset, while the responding cells from
newborn umbilical cord blood are predominantly V1+. It
has been proposed that the shift from V1 predominance in the blood
of newborns to V2-expressing cells in the blood of adults may
reflect a selective response to environmental stimuli, including
bacteria (5, 6, 7, 8, 9, 10, 11). Moreover, the restriction of TCRDJ junctional gene
usage associated with TCRDV2 in PBMC and intraepithelial lymphocytes
from normal adults (12, 13, 14, 15, 16, 17, 18) has been attributed to environmental
antigenic stimulation. Additionally, specific CDR3 region usage has
been demonstrated in TCR-mediated T cell responses to various
micro-organisms, although the results are conflicting, with data
arguing both for clonal selection (19, 20) and against it (3, 21, 22).
This situation is summarized in a recent review by De Libero, in which
the question of whether the diversity of the TCR CDR3 contributes to
the fine specificity of V9/V2+ T cells is addressed.
This review notes that ligand recognition by these T cells is still
debated, concluding that these cells are broadly cross-reactive to
phosphorylated metabolites (23).
To assess the role of Gram-negative bacteria in the selective expansion
of T cells (1, 24), it was first necessary to verify the status
of TCR clonality in normal unstimulated T cells. Some of the
previous reports have used TCR cloning in bacteria, followed by DNA
sequencing. As a limited number of clones could be evaluated, there is
a possible bias in the interpretation of results. Therefore, we have
used technologies that would minimize artifacts, while providing an
assessment of the CDR3 diversity among TCRDV1- and TCRDV2-expressing
cells.
Single-strand conformational polymorphism (SSCP)2 and heteroduplex conformational polymorphism have been used to assess TCR diversity based on sequence differences in the junctional region (25, 26, 27, 28). CDR3 size spectratyping can also be used to demonstrate size heterogeneity of the CDR3 within a V gene family (25, 29). SSCP is particularly powerful for detection of unique clonal expansions, while spectratyping is more applicable for evaluating the diversity of the overall repertoire, as it provides information on the frequency of transcripts of a given size.
In this study, we combined these two technologies to assess TCR
chain diversity in peripheral blood T cells from healthy adults and
newborns, and to investigate the TCRDV diversity in relationship to the
T cell responsiveness to Gram-negative bacteria. By examining
the selective expression of TCR among T cells in cord blood, we
were also able to assess possible clonal restriction before exposure to
external environmental Ags. In assessing both TCRDV1 and TCRDV2
transcripts, we have addressed selective TCR usage by both subsets of
T cells. Finally, by analyzing TCR selection from monozygotic
twins and fraternal twins from both newborn and adults we were able to
demonstrate that clonal selection among T cells is influenced by
factors in addition to genetic background, since the twins selected
different TCR clones among the restricted TCRDV populations
detected.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peripheral blood samples were obtained from 12 healthy adults
(age range, 22–50 yr). Umbilical cord blood samples were collected
from 15 newborns (nine normal deliveries of individual newborns and
three pairs of twins) at Massachusetts General Hospital (Boston, MA).
PBMC were isolated by Ficoll-Hypaque separation, and 5 to 10 x
106 cells were used for RNA extraction.
Immunophenotypes of fresh PBMC were also examined by direct
fluorescence analysis on a FACScan flow cytometer (FACScan II, Becton
Dickinson, Mountain View, CA) using mouse mAbs against CD3, CD4, and
CD8 (Becton Dickinson) and V1 and V2 (T Cell Diagnostics, Woburn,
MA). PBMC from adult subjects and cord blood samples in our study
showed <5% V1+ and V2+ cells before in
vitro culture.
Collection of T cell-enriched lymphocytes after
Gram-negative bacteria stimulation from adult PBMC and cloning by
limiting dilution
Escherichia coli 018K- (gift from Dr. G. Siber, Dana-Farber Cancer Institute, Boston, MA) and Pseudomonas aeruginosa type 6 (gift from Dr. A. Cross, Walter Reed Army Institute of Research, Washington, DC) were used for stimulation of PBMC in this study. The soluble bacteria preparations were obtained as previously described (1, 24).
PBMC from five adults (1 x 106/well) were incubated with E. coli and P. aeruginosa preparations at a dilution of 1/250. Human rIL-2 (100 IU/ml) was added on day 6 of culture to expand the reactive T cells for an additional 8 days.
After 2 wk of culture, the cells were analyzed for surface phenotype
with mouse mAbs against CD3, CD4, CD8, V1, and V2. Cells
(5–10 x 106) from these T cell-enriched
cultures were collected for RNA extraction and TCR analyses.
To evaluate the T cell repertoire on the basis of single cell
progeny, the E. coli-responsive cultured lymphocytes were
cloned by limiting dilution as previously described (30, 31). Although
T cells respond initially to E. coli, they do not
produce IL-2 in response to restimulation with the same bacterial
preparations and will be outstripped by 
ß T cells (1). Therefore,
limiting dilution was performed in the continued presence of 100 U of
IL-2 and in response to PHA, which drives the proliferation of
virtually all T cells. After 14 days in culture, the bulk culture
containing approximately 60% T cells, and 40% CD4+
ß T cells, was diluted to 0.5 cells/well in two 96-well plates
using irradiated autologous PBL and PHA at 1 µg/ml. Every 3 days, the
cultures were replenished with IL-2. As the wells became confluent (in
14–20 days), cells from positive wells were transferred to 24-well
plates for additional culture. To obtain sufficient cells for assays, a
second PHA and irradiated PBL restimulation was conducted in the
24-well plates. The progeny of these limiting dilution wells were
phenotyped by staining with anti-CD4- and anti-V2-specific
Abs. Of 42 wells from which growth was accomplished (from 192 starting
wells), 13 wells stained for V2, 27 stained for CD4, and 2 of the
wells showed a mixture of phenotypes indicative of wells that contained
more than one cell. The V2+ cells were further analyzed
for TCR usage, requiring RNA extraction, PCR amplification, and SSCP
analysis as described below.
RNA extraction and cDNA synthesis
Total RNA was extracted with TRIzol (Life Technologies, Gaithersburg, MD) according to the manufacturer’s specifications. Briefly, 5 to 10 x 106 PBMC were resuspended in 1 ml of TRIzol, with lysis facilitated by repetitive pipetting. Then, 0.2 ml of chloroform was added to the cell-TRIzol slurry. RNA was extracted from the upper (aqueous) phase of the mixture after microfuge centrifugation for 15 min at 12,000 x g. The aqueous phase was then transferred to a new tube, and 0.5 ml of isopropyl alcohol was added for 10 min at room temperature followed by a 10-min centrifugation (12,000 x g) at 4°C. The RNA pellet was washed once with 75% ethanol and then dissolved in 50 µl of RNase-free H2O. One microgram of total RNA was converted to first-strand cDNA using a (dT)12–18 primer together with avian myeloblastosis virus reverse transcriptase according to the manufacturer’s specifications (Promega, Madison, WI; incubation for 0.5–1 h at 42°C). Standard precautions were taken to avoid RNase contamination.
PCR and analysis of TCR by CDR3 size spectratyping and SSCP
PCR. TCR transcripts of TCRDV1, TCRDV2, and TCRBV2 were amplified using the following specific primers: TCRDV1, 5'-AGAGCTACATGCCCACATGCT; TCRDV2, 5'-AGAGGCAGAGCTG CCCCT; TCRDC, 5'-CTTATATCCTTGGGGTA GAA; TCRBV2, 5'-GCTCCAAGGCCACATACGAGCAAGGCGTCG; and TCRBC, 5'-CTCTTGACCATGGCCATC. The products generated by PCR were about 500 bp for TCRDV1 and TCRDV2, and 900 bp for TCRBV2. Thermocycling conditions were as follows: 5 min at 95°C (initial denaturation), followed by 35 cycles of 0.5 min at 95°C, 0.5 min at 55°C, and 1 min at 72°C. A final 5-min extension at 72°C concluded this first round of PCR.
To increase the sensitivity and specificity of PCR for SSCP and CDR3
size spectratyping analyses, a nested PCR amplification was used to
generate a 200- to 300-bp product. One microliter from a 1/100 dilution
of the first PCR reaction was reamplified respectively using internal
primers. The same C primers were used as described above: TCRDV1',
5'-CAAGAAAGCAGCGAAATCCGT; TCRDV2', 5'-CTATGGCCCTGGTT TCAAAGAC;
TCRBV2', 5'-CTGACCTTGTCCACTCTGACAG; and TCRBC',
5'-GGCCTTTTGGGTGTGGGAGA TCT.
CDR3 size spectratyping. The nested PCR products were labeled with 32P by incorporating 10 to 20 µCi/reaction of [32P]dCTP in the amplifications. Five microliters of the final PCR reaction volume (50 µl) was added to an equivalent volume of formamide/dye loading buffer, heated at 95°C for 2 to 5 min, then applied on a prewarmed (50°C) 8% polyacrylamide sequencing gel (National Diagnostics, Atlanta, GA) for approximately 4 to 5 h. The gels were run until the 200 to 300 products reached the lower half of the gels to assure adequate separation. The gels were dried, and autoradiography was performed at room temperature for 1 to 3 days with an enhancing screen. Each specific amplification of TCRDV1, TCRDV2, and TCRBV2 resulted in a ladder of bands separated by 3-bp spacing.
Densitometry and spectratyping graphs. To assess the intensity and distribution of the amplified TCR bands, the films were scanned using a computer-linked densitometer running ImageQuant (version 3.0) software (Molecular Dynamics, Sunnyvale, CA). The data were plotted on a Macintosh computer running DeltaGraph Pro (version 3.0); DeltaPoint, Monterey, CA). A curve was generated representing the intensity of each band in the ladder, corresponding to the distribution of amplified bands representing different size classes of CDR3 regions.
SSCP.
As described previously (32), the unlabeled nested PCR products of
TCRDV1 and TCRDV2 were purified on a S-300 column; 10 µl of purified
DNA was diluted in 2 µl of 0.5 M EDTA and 88 µl of formamide. Ten
to fifteen microliters of this mixture was heated to 95°C, followed
by ice-cooling immediately before loading on a 10% glycerol/8%
nondenaturing polyacrylamide (0.4-mm thick) sequencing format gel,
running at 15 to 20 watts for 15 to 17 h at room temperature in
1x TBE buffer. Thereafter, DNA was transferred to nylon membranes
(Magnagraph, MSI, Westborough, MA) by electroblotting at 4°C in 0.5x
TBE at 20 V for 3 h. The membrane was then UV cross-linked and
hybridized overnight at 50°C with the following TCRDC (constant
region) probe:
GAACAAATGTCGCTTGTCTGGTGAAGGAATTCTACCCCAAGGATATAAGA,
which was labeled with [-32P]ATP by T4 kinase
(Promega). The membranes were washed three times at 50°C (15 min/wash
step) at low stringency (0.1% SDS/1x SSC), medium stringency (1%
SDS/1x SSC), and high stringency (1% SDS/0.1x SSC). Films were
exposed to Kodak X-OMAT film at -80°C for 1 to 2 days with an
enhancing screen.
In some experiments, directly labeled DNA was generated by
incorporating [-32P]dCTP in the PCR reaction. For SSCP
gel analysis, 2 µl of the PCR product was combined with 98 µl of
formamide. Ten to fifteen microliters of this solution was then
electrophoresed directly on the gel. Following electrophoresis, the gel
was vacuum-dried to paper and autoradiographed, precluding the need for
hybridization.
Sequencing. For sequencing of the CDR3 region of the TCR genes, the PCR products were purified with a Wizard Prep PCR clean up kit (Promega), using the same internal TCRDV2-specific primer and TCRDC primers. Sequencing reactions were performed by the Sanger dideoxy method, using Sequenase according to the manufacturer’s specification (Sequenase version 2 kit, U.S. Biochemical Corp., Cleveland, OH), except that PCR products were denatured and quick-chilled in a dry ice/ethanol bath.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As shown in Figure 1
, two levels of
information can be obtained from CDR3 size spectratyping: band number
and band intensity. The number of distinguishable bands (or peaks in
the graphs) reflects the number of in-frame CDR3 sizes (size classes)
present. The intensity of each band is related to the frequency of
CDR3s of the same length. The spectratyping graphs obtained from
densitometry demonstrate the intensity distribution of the different
size classes. The TCRDV1 family usually has 12 to 14 distinguishable
bands (or peaks), while TCRDV2 usually shows eight to 10 bands. TCRBV2
transcripts can usually be discerned in six to eight size classes in
the peripheral polyclonal populations. A typical polyclonal population
usually displays a normally distributed pattern of TCR size frequency,
with modal bands being the most intense; larger and smaller size
classes contribute progressively less to the repertoire.
|
As shown in Figure 1
, combining the two techniques, three patterns of
TCR diversity could be discerned based on differences in the junctional
sequences. Pattern A (diverse, or polyclonal) shows a normal
distribution of band intensity in the spectratyping analyses and a
smear pattern in SSCP, suggesting a highly diverse population. Pattern
B (restricted) shows contractions and gaps between bands indicative of
a skewed intensity distribution pattern in spectratyping and
distinguishable, but often faint, bands in the SSCP, reflecting a less
diverse population. Pattern C (highly restricted) shows fewer than five
bands identified by spectratyping and an SSCP pattern of distinct
bands, representing a highly restricted population.
Diversity for TCRDV1 and TCRDV2 junctional usage in peripheral blood from adults
CDR3 size spectratype patterns in TCRDV1, TCRDV2, and TCRBV2 from
12 normal adults are shown graphically in Figure 2. The polyclonal nature of the TCRBV2 in
all but one of the samples is indicated by the normally distributed
size and intensity pattern expected of a diverse ß T cell
population. The only exception was donor 10, who showed a restricted
population with fewer than five peaks discernible.
|
T cells in the blood of normal
individuals. As shown in Figure 2, a highly restricted pattern was
observed in donors 1 through 5 in TCRDV1 and in donors 1 through 6 in
TCRDV2 peripheral populations, indicating oligoclonality in these
donors. Donors 6 through 10 in V1 and donors 7 through 10 in V2
showed a restricted pattern, suggesting populations that contain
relatively few clones but are more diverse than those showing the
highly restricted pattern. A typical polyclonal T cell
population in PBMC from normal donors was only observed in donors 11
and 12 with respect to both V1 and V2. Thus, combining the SSCP
and spectratype analyses, we were able to demonstrate restricted TCRDV1
and TCRDV2 junctional usage in PBMC from most normal adult donors.
These results for T cells contrast with the diversity of the
ß cells in the same donors.
The clonal nature of the highly restricted TCR repertoire was
examined further in donor 6, in whom a single dominant
TCRDV2-expressing clone was suggested by both spectratyping and SSCP
(Fig. 1). When the TCRDV2 product was amplified by PCR and subjected to
direct DNA sequence analysis, an unambiguous sequence could be
determined, indicating that this clone was predominant in the starting
population of T cells in the PBMC (Fig. 3). As shown in the analysis of single
cell progeny, this same TCR sequence could be identified in a high
proportion of the E. coli-responsive T cell clones
(see below).
|
2+ T cells from PBMC culture
stimulated with E. coli and P. aeruginosa
To determine whether the restricted or polyclonal patterns of
T cells in the blood of normal adults were related to responses
to environmental stimuli, we used the stimulation of T cells by
Gram-negative bacteria to assess the repertoire of T cells
responding to these stimuli. Adult donors 4, 6, 7, 8, and 10 were used
to evaluate their TCRDV usage after stimulation with E. coli
and P. aeruginosa.
Expansions of V1+ and V2+ T cells from
these donors are shown in Table I. The
percentage of V2 cells increased in all samples stimulated with
E. coli and P. aeruginosa, followed by expansion
in IL-2 for 2 wk. Stimulation with E. coli was generally
more effective at producing V2+ expansion (33–95%),
although P. aeruginosa was also effective in some
individuals, resulting in 15 to 84% V2+ cells as
assessed by flow cytometric analysis. In contrast, only a few adults
showed expansions of V1+ cells.
|
). In donor 4 (Fig. 4A), in whom the V2+ cells accounted
for 95 and 84% of the CD3+ cells, respectively, after
E. coli and P. aeruginosa stimulations, the
demonstration of an identical TCRDV2 SSCP pattern after stimulation
suggests the expansion of the clones present in peripheral blood. The
identical SSCP patterns obtained at different time points during the
same E. coli stimulation culture (lanes
2–4) show the consistency of the V2+
population during culture.
|
B). Their TCRDV2 SSCP patterns
also showed stability over time (data not shown). These data suggest
that the clonal populations of V2+ cells in PBL of these
donors are responsive to bacterial stimulation.
In donors 7, 8, and 10, who showed restricted patterns (but more
diverse than those of donors 4 and 6) with respect to TCRDV2 junctional
repertoire, their peripheral populations were also expanded by
E. coli and P. aeruginosa. PBMCs drawn
at up to three different times were stimulated by the same bacterial
preparations to compare their V2+ populations. As shown
in Figure 5, after E.
coli and P. aeruginosa stimulations,
predominant clones did not result among the cells responding to the two
bacterial preparations. Samples 8 and 10 were cultured three times at,
respectively, 3-wk and 1-yr intervals, and sample 7 was cultured twice
at a 6-mo interval. Consistently, no predominant clones were detected
from the stimulations with either bacterial preparation, and their
peripheral starting TCRDV2 analyses demonstrated virtually identical
CDR3 spectratype patterns (Fig. 5). These data suggest that the
population of T cells responsive to Gram-negative bacterial
stimulation is reflective of the starting population in the blood,
rather than a selective, clonally restricted response toward
Gram-negative bacteria.
|
Cloning from the E. coli-stimulated population showed the same predominant clone as that present in PBL
After the expansion of PBMC from adult donor 6 with E.
coli, the cells were cloned by limiting dilution. Thirteen
V2+ clones were obtained and grown from the culture, and
they were analyzed by SSCP to evaluate the dominant clonal response. As
shown in Figure 6A, 10 of 13
clones showed the same TCRDV2 SSCP pattern, which was identical with
that of the predominant clone observed in the PBL. These
receptors in these V2+ clones are paired with V9, and
their TCRGV9 SSCP analysis also showed the same dominance in the same
10 clones (Fig. 6B). However, because there may be
nonfunctional TCRGV9 transcripts in ß T cells, the same dominance
is not clearly present with respect to TCRGV9 in the PBL. Instead,
unlike peripheral TCRDV2 transcripts, which demonstrate one dominant
TCR, TCRGV9 usage from the same sample displays a much more complex
SSCP pattern (Fig. 6B, lane P). Taken
together, the above results again demonstrate the clonal predominance
among the V2+ cells in PBL from adult donor 6 and the
responsiveness of this dominant clone to E. coli in
vitro.
|
To clarify whether a restriction in the TCR repertoire is also
seen in newborns or represents a selection for a few clones during
extrauterine life, we collected cord blood samples from nine normal
deliveries. As shown in Figure 7, the
TCRBV2 spectratypes showed a polyclonal feature similar to that
observed in the adult donors. However, in both TCRDV1 and TCRDV2
analyses, a highly restricted pattern was observed in seven of nine
cord blood samples, again indicating that oligoclonality of
V1+ and V2+ cells is also present in
newborns. As shown in Figure 8, in an
additional three sets of twins, none of the six infants showed
polyclonal spectratype patterns, indicating that of the 15 newborns
analyzed, 13 showed restricted TCRDV usage.
|
|
To determine whether the selective clonal usage of TCR is
influenced predominantly by the genetic background of the individual or
if it reflects additional regulatory control, we evaluated the TCR
repertoire in twins. TCRDV1 and TCRDV2 spectratypes of six cord blood
samples from three pairs of twins are shown in Figure 8. Samples A1/A2
and B1/B2 were collected from two pairs of identical twins, while C1
and C2 were from one pair of nonidentical twins. As shown in Figure 8, the spectratypes for both TCRDV1 and TCRDV2 junctional usage showed
clear differences between twins regardless of their genetic identity.
An additional pair of adult identical twins also showed different
clonal predominance (donors 1 and 2 in Fig. 2).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells in normal PBMC, the data from
several reports are somewhat contradictory and have dealt almost
exclusively with the V2/V9 subset and not with
V1+ T cells. In particular, the T cells
responding to mycobacteria have been reported to be polyclonal with
respect to CDR3 (21), although unselected PBMC have been shown to be
clonally restricted in some adults (12, 13, 14, 15). It was our hope to clarify
the status of the T cell repertoire in adults and to determine
whether the selective stimulation of T cells by Gram-negative
bacteria might be responsible for the clonal restriction noted in some
individuals. Also, as the selection for V2/V9+ T
cells has been ascribed to environmental stimuli, we wished to
determine whether the V1 repertoire was similarly influenced by the
bacterial stimuli or whether the clonal restriction was confined to the
V2/V9+ subset. Furthermore, we wanted to determine
whether the clonal pattern seen in some adults represented a selection
from a polyclonal T cell repertoire at birth or, rather, if the
restricted pattern was already present at birth. As the adult
repertoire would be expected to be influenced by the exposure to a wide
variety of environmental stimuli, the sheltered environment of the
fetus coupled with immunologic immaturity would be expected to provide
less opportunity for in vivo selection in the newborn. Finally, by
evaluating the TCRDV repertoire in twins, we demonstrated that
selective TCRDV usage is influenced by more than the genetic background
of the blood donor.
The results favor the hypothesis that the T cells can be
clonally restricted in many normal individuals both at birth and in
adult life, although there are significant differences among different
individuals, some of whom show polyclonal T cell populations.
Both the V1+ and V2+ subsets show similar
restriction patterns both at birth and in adult life.
The question of clonality of T cells in newborns has been
addressed in one other study (18) that reported polyclonal junctional
usage in TCRDV1-DJ1 from four cord blood samples as judged by their
polyacrylamide electrophoresis patterns after PCR. Thirteen of the 15
newborns we studied showed a restricted pattern in spectratyping and
SSCP analyses, suggesting oligoclonality in TCRDV1 and TCRDV2
lymphocytic populations before extrauterine exposure. As two of the 15
cases we studied were polyclonal, the apparent discrepancy with the
results of Beldjord’s study (18) may be due to the limited number of
cases they examined. By studying the TCRDV1 and TCRDV2 repertoires of a
slightly larger population using two complementary techniques, our data
would indicate that selection for a limited repertoire is indeed
a common finding among newborns. Differences between the TCR
repertoire in newborn twins suggests that the genetic background alone
does not determine the selection for clones bearing the same
TCR, but, rather, like the unique attributes of fingerprints, the TCR
repertoire shows phenotypic variations between individuals with genetic
identity.
Although some previous studies have also noted clonal selection for
T cells in adults (12, 13, 14, 15), other groups have failed to confirm
this finding (20, 33, 34, 35). Our own data, derived from the combination
of SSCP and spectratyping analyses of whole blood lymphocyte
populations, indicate that most normal donors do show some restriction
in their TCRDV repertoire, although some individuals were quite diverse
in both TCRDV and TCRBV repertoires. Among the donors whose TCRDV
repertoire included some different clones, but were not truly
polyclonal (donors 7–10), both of the molecular approaches we used
showed that there was limited diversity. In such individuals, analyses
of T cell clones would require isolating large number of cellular
clones to demonstrate the selection we have found on the population
level. However, nothing in any of these studies clarifies the range of
TCR usage observed among normal subjects whose TCRDV repertoires varied
from highly clonal to polyclonal. Although the subjects studied were of
diverse racial and geographical backgrounds, the total number of
subjects is still too limited to draw conclusions that would allow us
to distinguish populations of individuals on the basis of TCR
diversity.
Although V2+ cells are the predominant subset in
PBMC in adults, while V1+ cells predominate in tissue or
cord blood (7), the range of restriction of the TCRDV repertoires was
similar in lymphocyte populations in both adults and newborns. This
suggests that restriction of TCRDVJ usage exists in PBMC regardless of
their relative numbers and V region usage, indicating that restriction
is a general property of cells. This concept is also supported
by others, who found a general restriction of TCRDV gene usage in
V1+, V2+, and V3+ T cells
from PBMC and intestinal tissues (15).
Importantly, the restricted repertoire seen in several of the
adults could be confirmed to be stable over a period of up to 12 mo,
reaffirming the findings of previous studies (13, 14, 15). Analysis of
clones of bacteria-stimulated PBL of one of the donors demonstrated
that 10 of 13 clones generated in response to E. coli had
the identical TCRDV2 sequences noted to be predominant among the TCRDV2
transcripts in the fresh PBL. These data indicate that the T
cells that respond to E. coli are representative of their
counterparts in the starting population.
Since the response to environmental stimuli, such as bacterial Ags,
could drive clonal selection in T cells, we investigated the
restriction of TCR repertoire in T cells derived from in vitro
cultures stimulated with E. coli or P.
aeruginosa. Our data show that when the unstimulated population
was highly restricted, a similar pattern could be detected in the
bacteria-stimulated cells. Similarly, in those individuals whose
T cells were polyclonal, the bacteria-stimulated cells remained
polyclonal. Although it has been reported that the response to
mycobacterial Ags results in polyclonal V2+ cells (21),
we found that the response to more ubiquitous bacteria did not skew the
expansion of the TCRDV repertoire. The T cell responsiveness of
mycobacteria-specific clones to Listeria monocytogenes and
E. coli demonstrated by Libero et al. (10) is reminiscent of
our results indicating reactivity of highly clonal T cells to
both E. coli and P. aeruginosa.
As the response to Gram-negative bacteria does not appear to influence
the T cell repertoire, one might postulate that the specificity
of the TCR is not directly involved in the response to either
E. coli or P. aeruginosa. These two Gram-negative
organisms are not cross-reactive at the level of Ag recognition by
CD4+ ß T cells (24, 36), although they could share
antigenic structures recognizable by T cells. However, the
vigorous response by both clonal and polyclonal T cells suggests
that either all of the clones have TCRs that recognize Gram-negative
bacterial Ags or that T cells are stimulated by structures other
than the CDR3 of the TCR. Cord blood T cells, which are
predominantly V1+, respond to the same bacteria as the
predominantly V2+ adult T cells. This further suggests
that a stimulus that is neither conventional Ag nor superantigen (1) is
capable of activating T cells via interaction with structures
other than the CDR3 region of the TCR.
Although our own previous study excluded LPS as the primary stimulus
for this polyclonal activator of T cells in humans (1), Leclercq
et al. (3) demonstrated that LPS influenced the response of T
cells to Gram-negative bacteria in mice. We do not know whether LPS
plays an adjuvant role in the human T cell response to the
proposed polyclonal response element. However, this polyclonal response
of T cells to Gram-negative bacteria may parallel the response
of B lymphocytes to bacterial products (such as LPS from Gram-negative
organisms and Staphylococcus aureus-derived protein A). Such
polyclonal activators do not interact with the Ag-combining sites of
the B cell’s Ig receptors, although protein A may activate by
cross-linking these Ag receptors by binding to Fc regions on the
surface Igs.
Recently, it has been found that V2/V9+ T cell
stimulation by monoethylphosphate involves the TCR complex, in that
transfection of cells with such TCR confers responsiveness to such
nonprotein Ags (37). It is possible that parallel to the B cell
response to protein A, determinants other than the presumed
Ag-combining sites encoded within the hypervariable CDR3 elements of
the TCR, including C and C structures, could interact with
cross-reactive bacterial components, leading to the stimulation of both
the V1+ and V2+ subsets we have noted
with Gram-negative bacteria. Such a polyclonal response element on
T cells would interact with structures on Gram-negative bacteria,
resulting in up-regulation of IL-2R. The addition of IL-2 then drives
further T cell proliferation following this bacterial
stimulation (1).
In summary, PBMC from both normal adults and newborns often show
restriction in the TCRDV1 and TCRDV2 repertoire. Furthermore, the
primary response of T cells to Gram-negative bacteria may be
independent of CDR3 region specificity. The restriction of TCRDV in
newborns argues against environmental antigenic selection for the
T cell repertoire in peripheral blood. As even identical twins
show different patterns of restriction, selection for CDR3 sequences
among T cells seems to be independent of the genetic background.
The possibility that T cells are responsive as a cell lineage to
some form of polyclonal activator shared among bacteria is
hypothesized, but further studies are needed to elucidate the
mechanisms involved.
|
Footnotes |
|---|
2 Abbreviations used in this paper: SSCP, single-strand conformation polymorphism; TBE, TRIS-borate-EDTA.
Received for publication May 19, 1997. Accepted for publication November 21, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 cells in adult peripheral blood but of V1 cells in cord bloods. J. Immunol. 157:1613.[Abstract]
+ and TCR - CD2+CD3+ double negative lymphocytes by bacterial stimulation. Int. Immunol. 2:775.3 cells by Gram-negative bacteria. J. Immunol. 154:5313.[Abstract]
/ T cells and their role in resistance to bacterial infection. J. Exp. Med. 178:971.-9-2 T lymphocytes during development. Scand. J. Immunol. 40:521.[Medline]
/ repertoire. J. Exp. Med. 171:1597. T cells at birth. J. Immunol. 153:3979.[Abstract]
T cell reactivity towards bacterial superantigens. Semin. Immunol. 5:41.[Medline]
T cells to staphylococcal enterotoxins: a restricted reaction pattern mediated by two distinct recognition pathways. Scand. J. Immunol. 38:89.[Medline]
/ T cells. J. Exp. Med. 173:1311.2-J1 junctional regions of human peripheral T lymphocytes. J. Immunol. 152:2860.[Abstract]
subset. Eur. J. Immunol. 23:804.[Medline]
1 T cell receptor repertoire in human small intestine and colon. J. Exp. Med. 180:183.1+ and V2+ cells increase with age and limit the repertoire of human gamma delta T cells. Eur. J. Immunol. 24:1914.[Medline]
T cell receptor repertoire in human colon and peripheral blood is oligoclonal irrespective of V region usage. J. Clin Invest. 96:1108.
rearrangements expressed in systemic lupus erythematosus (SLE) patients. Clin. Exp. Immunol. 97:430.[Medline]
T cell clones from intestinal biopsies of patients with celiac disease. Eur. J. Immunol. 23:499.[Medline]
1+ cells with restricted T cell receptor gene junctional repertoire in the peripheral blood of healthy donors. J. Exp. Med. 178:121./ T cell receptors. J. Exp. Med. 173:503./ T cells in response to a human pathogen. J. Exp. Med. 174:683. T cells reactive to Mycobacterium tuberculosis are encoded by specific V genes but diverse V-J junctions. J. Immunol. 147:3353.[Abstract]
/ß and / T cells from listeria-infected mice: synergistic effects of interleukin 1 and 7 on / T cells. J. Exp. Med. 178:985. T cells. Immunol Today. 18:22.[Medline]
1+ T cells in systemic sclerosis patients. J. Immunol. 153:881.[Abstract]
repertoire is skewed in cerebrospinal fluid of multiple sclerosis patients: molecular and functional analyses of antigen-reactive clones. Eur. J. Immunol. 25:355.[Medline]
9/V2 clones from human female reproductive tissues. J. Immunol. 155:2453.[Abstract]
2-V2 TCR-dependent recognition of non-peptide antigens and Daudi cells analyzed by TCR gene transfer. J. Immunol. 154:998.[Abstract]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
