| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Medicine, University of Cambridge Clinical School, Cambridge, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In human cord blood, almost all CD8+ T cells are CD45RAhigh. With increasing age, the proportion of cells expressing CD45RO increases. Upon stimulation with Ag or mitogen in vitro, CD45RAhighCD8+ T cells become activated and express CD45RO and down-regulate CD45RA (5). In human CD8+ T cells, it has been reported that the ability to respond to recall Ags mainly resides within the CD45RO+ T cell population (6). It has therefore been proposed that CD45RAhighCD8+ T cells constitute the naive T cell pool and that memory T cells express CD45RO, as has been proposed for CD4+ T cells (7). However, there is evidence to suggest that for CD8+ T cells this proposal may be an oversimplification. In adult PBMC, there is a subpopulation of CD45RAhighCD8+ T cells that express high levels of CD11a (LFA-1); the size of this subpopulation increases with age, raising the possibility that these CD45RAhigh cells may be primed cells (5). Indirect evidence from studies of the turnover of human T cell populations in vivo has suggested that CD45ROhigh T cells may revert to CD45RAhigh (8, 9), but these studies did not examine Ag-specific T cells. A recent study demonstrated that a substantial subset of CD8+CD45RA+ cells was stained with a EBV peptide-specific tetramer complex, suggesting that CD45RA is not a marker of naive T cells within the CD8+ population (10).
To address the extent to which human Ag-experienced CD8+ T cells revert from CD45ROhigh to CD45RAhigh in vivo, we studied the memory CD8+ CTL response against human CMV (HCMV).3 HCMV is a ubiquitous betaherpesvirus that infects between 60–90% of individuals depending on the population studied. Following primary HCMV infection, the virus persists lifelong in a latent state in cells of the myeloid lineage, under the control of the immune system, including CTL (11, 12, 13, 14, 15, 16).
During T cell development in the thymus, the TCR 
ß heterodimer is
generated by V(D)J gene recombination. Via the TCR, CD8+
CTL recognize viral peptides presented by MHC class I molecules on the
surface of infected cells. We have previously found high frequencies of
memory CD8+ CTL precursors (CTLp) specific for the HCMV
tegument protein pp65 in the peripheral blood of healthy virus carriers
(16). We have recently analyzed the clonal composition of the memory
CTL response to defined pp65 epitopes by generating multiple
independent peptide-specific CTL clones and sequencing the
hypervariable region of the TCR ß-chain. In each healthy virus
carrier, most of the CTL clones specific to a defined pp65 peptide use
the same TCR ß-chain at the level of nucleotide sequence, indicating
that the large population of circulating peptide-specific memory CTL
contains individual CTL clones that have greatly expanded in vivo (17).
These clonal expansions are sufficiently large to permit direct
molecular quantitation of the size of defined virus-specific CTL clones
in PBMC.
In this study, we separated highly purified populations of CD45RAhigh and CD45ROhigh CD8+ T cells from PBMC of long-term HCMV carriers, and analyzed the distribution of functional pp65 peptide-specific memory CTLp and the distribution of defined peptide-specific CTL clonotypes. We also studied two patients through acute primary HCMV infection into long-term memory, and investigated the distribution of the pp65-specific memory CTL population as a whole, and of an individual pp65-specific CTL clone, in relation to changing CD45RA and CD45RO expression in CD8+ T cells.
In long-term virus carriers, we found that both the CD45RAhigh and CD45ROhigh populations contained peptide-specific CTLp; the same peptide-specific CTL clonotype was present in both CD45RAhigh and CD45ROhigh populations. During primary HCMV infection, we observed fresh ex vivo pp65-specific cytotoxicity associated with a large increase in CD45RO+CD8+ T cells in PBMC. At subsequent time points, both the CD45RAhigh and CD45ROhigh populations contained peptide-specific CTLp, and a virus-specific CTL clone that was initially CD45ROhigh subsequently became distributed in both the CD45ROhigh and the CD45RAhigh populations.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Five healthy HCMV-seropositive laboratory donors and two
patients with primary HCMV were included in this study. All of the
laboratory donors were HCMV seropositive, as determined by an IgG ELISA
(Captia HCMV IgG immunoassay; Centocor, Malvern, PA). The patients with
primary HCMV were confirmed by HCMV-specific IgM and subsequent
seroconversion for HCMV-specific IgG (PHLS; Addenbrookes Hospital,
Cambridge, U.K.). The MHC class I tissue type of each donor (Table I
) was determined by serologic typing
(Lymphotype ABC-120; Biotest, Dreieich, Germany).
|
HCMV AD169 (ATCC VR-538) was grown in GMO5387 fibroblasts (Coriell Cell Repositories, Camden, NJ) infected at a multiplicity of infection (moi) of 0.01. Whole infected cultures were harvested 5 days after 100% cytopathic effect was evident, and were spun at 8000 rpm. Pellets were pooled, sonicated in an ice-cooled water sonication bath, aliquoted, and frozen at -70°C. Stocks were titered on 12-well plates of GMO5387s using 10-fold dilutions of virus; cytopathic effect was read after incubation for 10 days. Recombinant vaccinia viruses expressing the HCMV protein pp65 (vac pp65; gift of Dr. S. Riddel, Fred Hutchinson Cancer Research Centre, Seattle, WA), and a negative control expressing bacteriophage RNA polymerase T7 (vac T7) were grown in BHK cells infected at an moi of 0.1. After 48–72 h, the infected cells were harvested and subjected to three rounds of freeze-thaw, followed by sonication. The cell debris was removed by centrifugation, and supernatant-containing virus was aliquoted and stored at -70°C. A sample was subsequently titered on Vero cells; titers were normally between 1.5 and 10 x 107 plaque-forming units/ml.
Peptides based on the lower matrix protein pp65
Minimal peptides corresponding to proposed MHC-restricted epitopes of HCMV pp65 were synthesized and purified (>95%) by HPLC (Affiniti Research Products (Exeter, U.K.) (16, 17). All peptides were dissolved in RPMI 1640 at 200 µg/ml and frozen in small aliquots at -70°C.
mAbs and surface phenotyping
PBMC were prepared from fresh heparinized venous blood samples
by Ficoll-Hypaque (Lymphoprep; Nyegaard, Oslo, Norway) density-gradient
centrifugation. The surface phenotype of fresh PBMC and cells purified
by cell sorting was determined by flow cytometry. mAbs conjugated with
FITC, PE, or peridinin chlorophyll protein (PerCP) used were
anti-CD4 FITC (Leu3a), anti-CD8 FITC/PE/PerCP (Leu2a),
anti-CD16 FITC (Leu11a), anti-CD19 FITC (SJ25C1),
anti-CD45RA FITC (Leu18), anti-CD45RO PE (UCHL-1),
anti-CD56 PE (Leu19). The proportion of cells using specific TCR
Vß-chains was determined by three-color immunofluorescence using a
panel of TCR-specific FITC-conjugated mAbs (the panel included: Vß 1,
2, 3, 5.1, 5.2, 6.7, 7, 8.1, 11, 12, 13.6, 14, 16, 17, 19, 20, 21.3,
and 22 (Coulter, Palo Alto, CA); Vß 13.1/13.3 and V2 (Serotech))
with PE-conjugated anti-CD8 and PerCP-conjugated CD3. Samples were
analyzed on a FACSort flow cytometer (Becton Dickinson, Oxford, U.K.)
and using WinMDI software (Joseph Trotter, Scripps,
http://facs.scripps.edu).
Preparation of cell populations
Total CD8+, CD8+CD45ROhigh,
and CD8+CD45RAhigh cells were prepared from
PBMC using negative cell sorting to minimize any effects of ligation of
CD8, CD45RA, or CD45RO on the cell surface. It is not possible to
purify intermediate CD45RA+RO+ cells by
negative selection. PBMC were stained with FITC-conjugated
anti-CD4, anti-CD19, anti-CD16, and anti-CD56 mAbs (to
remove CD4+ T cells, B cells, and NK cells, respectively)
and sorted for nonstained cells using a FACSVantage cell sorter (Becton
Dickinson). CD8+CD45ROhigh cells were isolated
by staining another aliquot of PBMC with anti-CD4, anti-CD19,
anti-CD16, anti-CD56, and anti-CD45RA, and sorting for
nonstained cells. CD8+CD45RAhigh cells were
isolated by staining another aliquot of PBMC with anti-CD4,
anti-CD19, anti-CD16, anti-CD56, and anti-CD45RO, and
sorting for unstained cells. The purity of the CD45RAhigh
cells was >99% (CD45ROhigh cell contamination
0.1–0.65%); the purity of the CD45ROhigh cells was >99%
(CD45RA high cell contamination 0.1–0.27%) (Fig. 1, C and
F).
|
Replicate microcultures (n = 27) of purified CD8+ T cell populations were set up in 96-well round-bottom plates in which the number of responder T cells per well was progressively reduced over an appropriate range of dilutions in RPMI 1640 supplemented with 10% human AB serum (HCMV seronegative; Blood Transfusion Service, Addenbrookes Hospital, Cambridge, U.K.), 2 mM L-glutamine, 105 IU/L penicillin, and 100 mg/L streptomycin (referred to as RPMI-HuAB). As stimulator cells, autologous PBMC were pulsed for 1 h with HCMV (moi 0.01), irradiated (2400 rad), and added at 5 x 104 cells/well. Although HCMV does not appear to establish a productive infection of unstimulated PBMC, exogenous pp65 protein is able to enter the MHC class I processing pathway (15) for stimulation of pp65-specific CD8+ T cells. The medium was further supplemented with human rIL-2 to give a final concentration of 5 IU/ml. LDA cultures were incubated at 37°C in 5% CO2 and refed with RPMI-HuAB supplemented with 5 IU/ml IL-2 on days 5 and 10. On day 14, using split-well analysis, the cells in each individual well were resuspended and divided into aliquots that were assayed simultaneously for cytotoxicity against radiolabeled target cells in 4-h 51Cr release assays. Target cells comprised 4 x 103 cells/well autologous and MHC-mismatched lymphoblastoid B cell lines that were pulsed or unpulsed with pp65 peptide (40 µg/ml) for 1 h, or infected for 18 h with vac pp65 or control vac T7 (moi = 10). Target cells were washed three times and counted before use in chromium release assays. The LDA results were analyzed as described previously (16).
PCR amplification of TCR Vß regions
Total RNA was extracted from 1–10 x 105 cells using an RNA extraction kit (Qiagen, West Sussex, U.K.). First strand cDNA was reverse transcribed with an oligo(dT) primer and avian myeloblastosis virus reverse transcriptase using a reverse-transcription kit (Promega, Madison, WI), according to the manufacturer’s instructions. PCR was performed using a panel of 36 TCR Vß family-specific primers (17) together with the corresponding C region-specific primer (synthesized by Genosys Biotechnologies, Cambridge, U.K.). Each reaction was conducted in a total volume of 50 µl containing 1 mM of each dNTP (Boehringer Mannheim, Indianapolis, IN), 3.75 mM MgCl2, each primer at a final concentration of 1 mM, and 1 U of Taq polymerase (Promega), in buffer supplied by the manufacturer. A total of 2 µl of cDNA was used for PCR amplification in each case. The reaction was overlaid with mineral oil, and amplification was performed for 45 cycles of PCR. Conditions were 1-min denaturation at 94°C, 30 s of annealing at 60°C, and 30 s of extension at 72°C on a DNA thermal cycler (Perkin-Elmer Cetus 9600 Instruments, Norwalk, CT). A total of 45 µl of each PCR product was separated on a 1.3% agarose gel. Expression of Vß genes was considered positive where an approximately 300–400 nt rearranged band could be visualized with ethidium bromide staining.
Quantitative clonotypic analysis
We previously generated peptide-specific CTL clones in vitro and
sequenced the TCR ß-chain (17). To quantify a defined
peptide-specific CTL clone in PBMC by molecular assay, we designed a
complementary oligonucleotide 15–20 nt long based on the TCR ß-chain
hypervariable n-D-n region of that immunodominant CTL clone (referred
to as the clonotypic probe (Table I
)). We also designed a conserved TCR
C region probe to quantify total amplified TCR sequences. Vß
family-specific PCR was performed in duplicate on cDNA from total
CD8+, CD8+CD45RAhigh, and
CD8+CD45ROhigh cells from each donor. PCR
products were also produced from the defined peptide-specific CTL clone
(positive control) and from a mixed population of PBMC derived from
four HCMV-seronegative donors (negative control). The duplicate PCR
samples were separated on a 1.3% agarose gel, and the DNA was
transferred onto a filter (Bio-Rad, Richmond, CA) using standard
techniques in blotting buffer (0.6 M NaOH, 0.2 M NaCl) overnight.
Filters were washed in 2x SSC and baked at 80°C for 30 min. Filters
were prehybridized in 15 ml of 7% SDS, 0.25 M
Na2HPO4 for 1 h at 10–15°C below Tm of
the oligonucleotide probe. The oligonucleotide probes were end labeled
with [-32P]dATP using T4 polynucleotide kinase;
unincorporated label was separated from the probe using a G-25 Sephadex
spin column (Amersham, Arlington Heights, IL). Filters were first
hybridized with the clonotypic probe overnight at 10–15°C below the
Tm of the clonotypic probe in hybridization buffer (7% SDS, 0.25 M
Na2HPO4, 10% polyethelyne glycol 8000). After
hybridization, the filters were washed three times (5% SDS, 20 mM
Na2HPO4) for 15 min each. Binding of the probe
was imaged and quantitated using a phosphor imager (Beckman, Fullerton,
CA); the filters were also exposed to x-ray film. Bound clonotypic
probe was removed from the filters by soaking in 0.4 M NaOH for 30 min,
followed by one wash in 2x SSC. The filter was then rehybridized with
the TCR ß-chain C region probe (AGATCTCTGCTTCTGATG) following the
same protocol as for the clonotypic probe. In each experiment, it was
possible to correct for differences in radioactive labeling and binding
of the clonotypic and C region probes by measuring the radioactivity of
each probe bound to the positive control sample as a reference standard
(TCR ß-chain sequences amplified from the biologically derived CTL
clone contain the unique hypervariable n-D-n region and C region in
equal amounts). The proportion of clonotype sequence within the total
amplified TCR Vß sequence for a given population of CD8+
cells was calculated as: test sample cpm (probed with clonotypic
probe)/test sample cpm (probed with constant probe) x 100; positive
control cpm (probed with clonotypic probe)/positive control cpm (probed
with constant probe).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD8+ T cells purified from PBMC of HCMV carriers by
negative cell sorting were a mixture of CD45RAhigh,
CD45ROhigh, and CD45RA+RO+
intermediates (Fig. 1A).
Because CD8+CD45RAhigh and
CD8+CD45ROhigh T cells are mutually exclusive
populations, high purity CD8+CD45RAhigh cells
were obtained by sorting, using staining with anti-CD45RO and
collecting negative cells (Fig. 1C); high purity
CD8+CD45ROhigh cells were obtained using
staining with anti-CD45RA and collecting the negative cells (Fig. 1E). Each of the populations was stimulated in LDA with
autologous irradiated HCMV-infected PBMC. Ag-specific cytotoxicity
was assessed using target cells that expressed whole pp65
(vaccinia recombinant) and target cells that were pulsed with the
immunodominant pp65 peptide appropriate for donor HLA type (Table I).
In five long-term HCMV carriers, the CD8+ cell population
as a whole contained high frequencies of pp65-specific CTLp (Fig. 1B; Table II). Both the
CD8+CD45RAhigh and
CD8+CD45ROhigh subpopulations contained
pp65-specific CTLp, including CTLp specific to the same immunodominant
pp65 peptide (Fig. 1, D and F; Table II).
|
). In two healthy HCMV carriers (001 and 009), 10-fold
more of the pp65 peptide-specific CTLp in PBMC were
CD45RAhigh than were
CD45ROhighCD45RA-; in two other HCMV carriers
(005 and 002), equivalent numbers of the pp65 peptide-specific CTLp in
PBMC were CD45RAhigh and CD45ROhigh. Large clone sizes of peptide-specific CTL clonotypes are present in CD45RAhigh cells in long-term HCMV carriers
In each HCMV carrier, most of the independently derived pp65
peptide-specific CTL clones use the same TCR ß-chain at the level of
nucleotide sequence, indicating a high degree of clonal focusing (17).
To quantify an individual peptide-specific CTL clone, we designed an
oligonucleotide probe complementary to the hypervariable ß-chain
sequence of the CTL clone (Table I). In order first to determine the
relationship between the amount of clonotypic DNA and the amount of
bound radiolabeled probe in the blotting assay (see Materials and
Methods), cDNA from a biological pp65 peptide-specific CTL clone
was PCR amplified and added in increasing amounts to TCRs derived from
a pool of cDNA from PBMC of four HCMV-seronegative donors. Each of the
mixtures of TCR DNAs was probed first with the clonotypic probe (Fig. 2B) and, after stripping the
filter, with the C region probe that binds to all TCR ß-chain DNA
sequences (Fig. 2C). The results show a close correlation
between the actual amount of clonotypic sequence added into the test
samples and the experimentally derived values using clonotypic probing:
the relationship is linear over a wide range (Fig. 2D) that
includes the experimental values observed in subsequent experiments
(Table III). The technique has the
sensitivity to detect approximately 1% clonotype sequence against the
background of 99% different n-D-n sequences within the corresponding
Vß family.
|
|
, Table III). The absolute clone sizes of individual peptide-specific CTL
clones in the total CD8+ population were very large, up to
0.8% of CD8+ PBMC. In all donors tested, we found that the
CD45RAhigh population contributed 6- to 10-fold more than
the CD45ROhigh population to the total clone size in
CD8+ PBMC (Table III).
|
The large clone sizes of pp65-specific memory CTL in CD45RAhighCD8+ cells of long-term virus carriers suggest that following primary HCMV infection, some of the clonally expanded activated CD45ROhigh pp65-specific CTL may have reverted to less activated memory CTL that reexpress the CD45RAhigh phenotype. To address this possibility directly, we studied two patients during and after symptomatic primary HCMV infection.
Fresh PBMC were obtained from patient CMV104 at 3, 4, and 8 wk after
the onset of symptoms (fever, anorexia, headache associated with
atypical lymphocytosis, and abnormal liver function tests). PBMC were
stained with anti-CD45RA, anti-CD45RO, and anti-CD8, and
analyzed by flow cytometry (Fig. 4,
A, D, G, B, E,
and H); at each time point, PBL counts were determined in
the clinical haematology laboratory, from which the absolute number of
CD8+ cells and the CD8+CD45RAhigh
cells were calculated. Aliquots of unstimulated PBMC were depleted of
CD16+ NK cells and used as effector cells in direct ex vivo
cytotoxicity assays against HLA-matched target cells pulsed with the
appropriate pp65 peptides (Fig. 4, C, F, and
I). At 3 wk, there was a substantial CD8+
lymphocytosis of 4.9 x 109/L (normal range of
0.4–1 x 109/L), the large majority of which were
CD45ROhigh (4.7 x 109/L). At this time,
there was a very high level of ex vivo cytotoxicity against two
different HLA-B7-restricted pp65 peptides, p31 and p56. Similar results
were obtained 1 wk later. At 8 wk after the onset of symptoms, the
absolute CD8 count had decreased to 0.7 x 109 cells/L
(within normal range) and there was a substantial reduction in the
CD8+CD45ROhigh population. The absolute number
of CD8+CD45RAhigh cells remained relatively
stable at 3, 4, and 8 wk (0.2 x 109 cells/L, 0.5
x 109 cells/L, and 0.4 x 109 cells/L,
respectivly). Thirty-eight weeks after onset of symptoms,
CD8+CD45RAhigh cells were purified by
negative sorting from PBMC and stimulated with autologous irradiated
PBMC pulsed with peptide 56 in LDA. The frequency of peptide
56-specific CTLp in CD8+CD45RAhigh cells was
1240 per 106 CD8+ T cells (frequency against
control target <20 per 106
CD8+CD45RAhigh T cells).
|
Longitudinal analysis of pp65-specific clonotypes following primary HCMV infection shows clonal reversion from CD45ROhigh to CD45RAhigh T cells
Multiple independent pp65 peptide 56-specific CTL clones were
derived from the PBMC of patient CMV104 4 wk after onset of symptoms,
and the Vß TCR usage of these clones was determined by PCR analysis
with a panel of Vß TCR primers, as described (16). A large proportion
of the clones was found to utilize TCR Vß 6.4. A clonotypic probe
specific to the n-D-n region from one of these clones was used to
quantitate the size of this pp65 peptide-specific CTL clone within the
CD45RAhigh and CD45ROhigh CD8+ T
cell populations at 3, 4, 8, and 37 wk after onset of symptoms. The
results show that at 3 wk after onset of symptoms, there was very
little of the clonotype within the Vß 6.4+
CD45RAhigh cell population as compared with the Vß
6.4+ CD45ROhigh population (Fig. 5). Four weeks after onset of symptoms
there was still a pronounced difference in the amount of clone detected
in the CD45RAhigh as compared with the
CD45ROhigh populations (Fig. 5). At 8 wk after onset,
similar or higher levels of the clonotype were detected in the Vß
6.4+ CD45RAhigh as compared with the Vß
6.4+ CD45ROhigh population. At 37 wk after
onset of symptoms, the clonotype was distributed in both the
CD45RAhigh and CD45ROhigh T cell populations
(Fig. 5), as seen previously in the long-term virus carriers (Table III, Fig. 3). As a mAb to TCR Vß 6.4 is not commercially available,
we were unable to calculate the absolute clone size within the
CD45RAhigh and CD45ROhigh populations at each
of the time points sampled.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies of the human CD8+ CTL response against another persistent virus, EBV, reported substantially higher CTLp frequencies in the CD45ROhigh population (110–355 per 106 CD45ROhigh cells) compared with the CD45RAhigh population (3–27 per 106 CD45RAhigh cells) (6). In most healthy HCMV carriers, we also observed higher HCMV-specific CTLp frequencies in CD45ROhigh cells (200–4610 per 106 CD45ROhighCD8+ T cells) compared with the CD45RAhigh population (160–980 per 106 CD45RAhighCD8+ T cells). However, the absolute frequencies of pp65-specific CTLp that we observed in the CD45RAhigh population were 5–35-fold higher than previously observed for EBV-specific CTLp in the CD45RAhigh population. Because it is impossible to purify intermediate CD45RA+RO+ by negative selection, we were not able to determine directly the contribution of intermediate CD45RA+RO+ cells. Thus, the sum of the calculated frequencies of CTLp in the CD45ROhigh and CD45RAhigh fall short of the total CD8+ frequencies; this is probably due to contributions from the RO+RA+ subset that are not assayed in this study. The higher frequencies in our experiments may reflect improved cloning efficiency due to the addition of exogenous IL-2 and the longer period of LDA culture that we used before cytotoxicity assay (14 days, compared with 7–9 days in 6). Following stimulation, the initial rate of proliferation of CD45ROhigh cells is greater than that of CD45RAhigh cells (4). Therefore, with a longer period of culture in the presence of IL-2, CTLp within the CD45RAhigh population had sufficient time to grow into detectable CTL clones in vitro, leading to substantially higher estimates of CTLp frequency in the CD45RAhigh population. Because there are many more CD45RAhighCD8+ T cells than CD45ROhighCD8+ T cells in PBMC, in absolute terms up to 10-fold more of the HCMV-specific CTLp in PBMC are in the CD45RAhigh population.
The size of individual peptide-specific CTL clonotypes determined by clonotype probing was 10–100-fold larger than the corresponding peptide-specific CTLp frequency determined by LDA (which represents the aggregate functional response of all of the CTL clonotypes specific to the same peptide) in the same blood sample. Clonotype probing is sufficiently sensitive to detect individual clonotype sequences that constitute as little as 1% of sequences in a Vß family, and hence about 0.1% of CD8+ cells in PBMC, because unlike LDA clonotype detection, it does not involve biological cloning. Staining with fluorescent peptide-MHC class I tetramers (18) also provides greater sensitivity than LDA for detecting the aggregate of peptide-specific CTL clones that can bind to a given peptide-MHC (19). In this study, we used clonotype probing rather than tetramer staining not only because we wished to study memory CTL responses specific for different viral peptides restricted by three different MHC alleles, but also because the clonotype probing technique provides a higher level of resolution by allowing the study of reversion from CD45ROhigh to CD45RAhigh at the level of an individual CTL clone.
In response to acute virus infections, there are large expansions of virus-specific CD8+ T cells (10, 19, 20). Highly activated human CD45RO+CD8+ T cells show increased susceptibility to apoptosis, but the survival of these activated CD8+ cells in vitro can be prolonged by provision of help from nonlymphoid cells (21, 22). Human CD45ROhigh cells appear to turn over more rapidly than CD45RAhigh cells in vivo, with indirect evidence that CD45ROhigh T cells may revert to CD45RAhigh cells in vivo (8). Our finding of large clone sizes of the same pp65-specific CTL clonotype in both the CD45ROhigh and the CD45RAhigh CD8+ cells of long-term virus carriers suggested that after primary HCMV infection, some of the clonally expanded activated CD45ROhigh pp65-specific CTL reverted to (less activated) memory CTL that reexpress the CD45RAhigh phenotype. Our observations in primary HCMV infection provide direct evidence that cells of defined virus-specific CTL clonotypes revert from CD45ROhigh to CD45RAhigh in vivo.
The relative distribution of memory CD8+ T cells between the CD45ROhigh and CD45RAhigh populations may be influenced by the degree to which these cells are exposed to Ag in vivo. HCMV persists in cells of the myeloid lineage (11, 12) from which infectious virus can reactivate (23). Productive HCMV infection is rare in most body compartments in healthy subjects, and HCMV DNA cannot be detected in normal plasma by highly sensitive PCR (11, 12). At least for this persistent human virus infection, memory CD8+ T cells are a phenotypically heterogenous population of clonally expanded cells; some are CD45ROhigh cells that as a population are more activated and may turn over more rapidly (8), while others that are CD45RAhigh are less activated and divide infrequently. This is consistent with in vivo studies of murine models, in which phenotypic markers of activation are not identical to those in humans (24). Using adoptive transfer of CD8+ T cells from transgenic mice expressing an LCMV peptide-specific TCR, Zimmermann et al. (25) were able to visualize memory T cells by flow cytometry following LCMV infection in vivo. They observed heterogeneity both in cell surface expression of memory markers (most CD8+ memory T cells were CD11b- and CD45RBhigh, like naive mouse T cells) and in intermitotic interval (one-half of the transgenic memory T cells persisted for 5–7 wk without undergoing cell division). Following acute infection in intact mice, virus-specific memory CTL responses can be maintained at a stable level for many months (26, 27). These CD8+ T cell responses can nevertheless be modulated by subsequent infections with different viruses. In response to virus infection in mice, most of the activated CD44highCD8+ T cells undergo cell division; this preferential proliferation of CD44highCD8+ T cells can be mimicked by type I IFN (28). Once the immune system returns to homeostasis following infection with a different virus, the CTL memory pool to earlier viruses diminishes (27). The clonal composition differs between primary and memory T cell responses in some murine models, but not others (29, 30). Because the clonotype probing technique we describe is able to distinguish individual CTL clones, we are currently using this technique to determine whether the clonal composition of the virus-specific CTL response changes with time following primary infection.
The extent to which CD4+ T cells may undergo interconversion between CD45RAhigh and CD45ROhigh either in vitro or in vivo is much debated (31). Recent evidence that human memory CD4+ T cell responses are confined to the CD45ROhigh population (32) suggests that, at least in humans, the capacity of memory CD8+ T cells to reexpress CD45RAhigh in vivo may genuinely be different from that of CD4+ T cells.
We have shown that cells of defined virus-specific CD8+ CTL clonotypes revert from CD45ROhigh to CD45RAhigh in vivo. Because CD45RAhigh cells contain both naive cells and memory cells, we conclude that CD45RA is not a reliable marker of naivety in human CD8+ T cells and that CD45RO is a marker of cell activation that does not identify all CD8+ memory T cells.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Drs. Mark R. Wills, Andrew J. Carmichael, and Michael P. Weekes, Department of Medicine, University of Cambridge Clinical School, Hills Road, Cambridge, CB2 2QQ, U.K. E-mail address:
3 Abbreviations used in this paper: HCMV, human CMV; CTLp, CTL precursor; LDA, limiting dilution analysis; moi, multiplicity of infection; PerCP, peridinin chlorophyll protein.
Received for publication January 6, 1999. Accepted for publication April 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Crompton, N. Khan, R. Khanna, L. Nayak, and P. A. H. Moss CD4+ T cells specific for glycoprotein B from cytomegalovirus exhibit extreme conservation of T-cell receptor usage between different individuals Blood, February 15, 2008; 111(4): 2053 - 2061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. L. Elliott, A. Suhrbier, J. J. Miles, G. Lawrence, S. J. Pye, T. T. Le, A. Rosenstengel, T. Nguyen, A. Allworth, S. R. Burrows, et al. Phase I Trial of a CD8+ T-Cell Peptide Epitope-Based Vaccine for Infectious Mononucleosis J. Virol., February 1, 2008; 82(3): 1448 - 1457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. C. P. Waller, N. McKinney, R. Hicks, A. J. Carmichael, J. G. P. Sissons, and M. R. Wills Differential costimulation through CD137 (4 1BB) restores proliferation of human virus-specific "effector memory" (CD28 CD45RAHI) CD8+ T cells Blood, December 15, 2007; 110(13): 4360 - 4366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Yang, H. Peng, Z. Zhu, G. Li, Z. Huang, Z. Zhao, R. A. Koup, R. T. Bailer, and C. Wu Persistent memory CD4+ and CD8+ T-cell responses in recovered severe acute respiratory syndrome (SARS) patients to SARS coronavirus M antigen J. Gen. Virol., October 1, 2007; 88(10): 2740 - 2748. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. K. Day, A. J. Carmichael, I. J. M. ten Berge, E. C. P. Waller, J. G. P. Sissons, and M. R. Wills Rapid CD8+ T Cell Repertoire Focusing and Selection of High-Affinity Clones into Memory Following Primary Infection with a Persistent Human Virus: Human Cytomegalovirus J. Immunol., September 1, 2007; 179(5): 3203 - 3213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Precopio, M. R. Betts, J. Parrino, D. A. Price, E. Gostick, D. R. Ambrozak, T. E. Asher, D. C. Douek, A. Harari, G. Pantaleo, et al. Immunization with vaccinia virus induces polyfunctional and phenotypically distinctive CD8+ T cell responses J. Exp. Med., June 11, 2007; 204(6): 1405 - 1416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Northfield, C. P. Loo, J. D. Barbour, G. Spotts, F. M. Hecht, P. Klenerman, D. F. Nixon, and J. Michaelsson Human Immunodeficiency Virus Type 1 (HIV-1)-Specific CD8+ TEMRA Cells in Early Infection Are Linked to Control of HIV-1 Viremia and Predict the Subsequent Viral Load Set Point J. Virol., June 1, 2007; 81(11): 5759 - 5765. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. O. Kvale, Y. Floisand, F. Lund-Johansen, H. Rollag, L. Farkas, S. Ghanekar, P. Brandtzaeg, F. L. Jahnsen, and J. Olweus Plasmacytoid DCs regulate recall responses by rapid induction of IL-10 in memory T cells Blood, April 15, 2007; 109(8): 3369 - 3376. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Monteiro, C. Evaristo, A. Legrand, A. Nicoletti, and B. Rocha Cartography of gene expression in CD8 single cells: novel CCR7- subsets suggest differentiation independent of CD45RA expression Blood, April 1, 2007; 109(7): 2863 - 2870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kobayashi, T. Kondo, H. Takata, S. Yokota, and M. Takiguchi Functional and phenotypic analysis of human memory CD8+ T cells expressing CXCR3. J. Leukoc. Biol., August 1, 2006; 80(2): 320 - 329. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Munks, K. S. Cho, A. K. Pinto, S. Sierro, P. Klenerman, and A. B. Hill Four Distinct Patterns of Memory CD8 T Cell Responses to Chronic Murine Cytomegalovirus Infection J. Immunol., July 1, 2006; 177(1): 450 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. O. Kvale, J. Dalgaard, F. Lund-Johansen, H. Rollag, L. Farkas, K. Midtvedt, F. L. Jahnsen, J. E. Brinchmann, and J. Olweus CD11c+ dendritic cells and plasmacytoid DCs are activated by human cytomegalovirus and retain efficient T cell-stimulatory capability upon infection Blood, March 1, 2006; 107(5): 2022 - 2029. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Dunne, L. Belaramani, J. M. Fletcher, S. F. de Mattos, M. Lawrenz, M. V. D. Soares, M. H. A. Rustin, E. W.-F. Lam, M. Salmon, and A. N. Akbar Quiescence and functional reprogramming of Epstein-Barr virus (EBV)-specific CD8+ T cells during persistent infection Blood, July 15, 2005; 106(2): 558 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Almanzar, S. Schwaiger, B. Jenewein, M. Keller, D. Herndler-Brandstetter, R. Wurzner, D. Schonitzer, and B. Grubeck-Loebenstein L |
) and as a control vaccinia T7 (
).
