International Immunology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (32) Request Permissions Google Scholar Articles by Labalette, M. Articles by Dessaint, J.-P. Search for Related Content PubMed PubMed Citation Articles by Labalette, M. Articles by Dessaint, J.-P. Social Bookmarking          What's this?

International Immunology, Vol. 11, No. 8, 1327-1336, August 1999
© 1999 Japanese Society for Immunology

Peripheral human CD8⁺CD28⁺T lymphocytes give rise to CD28^–progeny, but IL-4 prevents loss of CD28 expression

Myriam Labalette, Emmanuelle Leteurtre, Caroline Thumerelle, Claudine Grutzmacher, Béatrice Tourvieille and Jean-Paul Dessaint

Service d'Immunologie, EA 2686, Centre Hospitalier et Universitaire de Lille, 59045 Lille, France

Correspondence to: M. Labalette

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
At birth, virtually all peripheral CD8⁺ T cells express the CD28 co-stimulatory molecule, but healthy human adults accumulate CD28^–CD8⁺ T cells that often express the CD57 marker. While these CD28^– subpopulations are known to exert effector-type functions, the generation, maintenance and regulation of CD28^– (CD57⁺ or CD57^–) subpopulations remain unresolved. Here, we compared the differentiation of CD8⁺CD28^brightCD57^– T cells purified from healthy adults or neonates and propagated in IL-2, alone or with IL-4. With IL-2 alone, CD8⁺CD28^brightCD57^– T cell cultures yielded a prevailing CD28^– subpopulation. The few persisting CD28^dim and the major CD28^– cells were characterized by similar telomere shortening at the plateau phase of cell growth. Cultures from adults donors generated four final CD8⁺ phenotypes: a major CD28^–CD57⁺, and three minor CD28^–CD57^–, CD28^dimCD57^– and CD28^dimCD57^dim. These four end-stage CD8⁺ subpopulations displayed a fairly similar representation of TCR V_ß genes. In cultures initiated with umbilical cord blood, virtually all the original CD8⁺CD28^bright T cells lost expression of CD28, but none acquired CD57 with IL-2 alone. IL-4 impacted on the differentiation pathways of the CD8⁺CD28^brightCD57^– T cells: the addition of IL-4 led both the neonatal and the adult lymphocytes to keep their expression of CD28. Thus, CD8⁺CD28^brightCD57^– T cells can give rise to four end-stage subpopulations, the balance of which is controlled by both the cytokine environment, IL-4 in particular, and the proportions of naive and memory CD8⁺CD28⁺ T cells.

Keywords: antigens, CD8⁺ T lymphocytes, CD28, CD57, cell aging, cell culture, cell differentiation, down-regulation, human, IL-4, T lymphocyte subsets, telomere

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
CD28 ligation is considered as the major co-stimulatory signal for activation of naive T cells (1,2). Once established, however, the proliferative response appears to be less dependent on CD28 co-stimulation, since exogenous growth factors such as IL-2 can restore mitotic progression. Given the important role of CD28 in the immune response, a perplexing observation is the presence in humans of CD8⁺ T cells lacking membrane expression of CD28. In fact, heterogeneity of the human CD8⁺ population progressively increases with age. Human CD8⁺ T lymphocytes are largely homogenous at birth and essentially all express the CD28 receptor, but healthy adults accumulate CD28^–CD8⁺ T cells, most of which express the CD57 glycoprotein. Moreover, CD28^– and CD57⁺CD8⁺ T cells are increased in numbers and maintained for many years by viral carrier status, viral latency or other chronic immuno-stimulative conditions (3–8).

In adults, peripheral lymphocytes contain both naive and memory T cells. Inasmuch as >80% memory-type human CD8⁺ T cells are reported to be CD28⁺ (9), the unusual proliferative and functional profiles of the CD8⁺CD28^– subpopulation raise a number of concerns in interpretation. First, these cells demonstrate poor proliferative responses to mitogenic stimulations (8–11). While lack of co-stimulatory signal by deficient CD28 expression may account for this hyporesponsiveness, CD8⁺CD28^– T cells have shortened telomeres and may have reached a state of replicative senescence (12,13). Second, CD28^– and/or CD57⁺CD8⁺ T cells are efficient producers of a variety of cytokines (14,15), they express perforin and exert potent cytotoxic activity (7, 8,10). Accordingly, CD8⁺CD57⁺ T lymphocytes can respond to human cytomegalovirus (CMV) in vitro and exert virus-specific cytotoxicity (5). Likewise, freshly purified blood CD8⁺CD28^– lymphocytes from HIV-infected individuals exert HIV-specific cytotoxicity (7,8). However, the full extent of effector cell removal by apoptosis (16) after an immune response has subsided and T cell homeostasis has been re-established is still unknown.

Contrary to their recognized functional properties, basic questions regarding the generation, maintenance and regulation of CD8⁺CD28^– and CD8⁺CD57⁺ subpopulations remain unresolved. It is not proven whether CD28^– and CD57⁺ cells belong to a separate lineage of peripheral CD8⁺ T cells or whether loss of CD28 expression and acquisition of CD57 result from a normal pattern of differentiation of human CD8⁺CD28⁺CD57^– T cells. On the one hand, CD57⁺ T cells have been shown to appear in allogeneic secondary mixed lymphocyte cultures initiated with CD57^– cells (14), but CD28 expression was not investigated in this study. On the other hand, there are conflicting reports on the long-term stability of CD28 expression. The level of CD28 expression has been reported to transiently increase or decrease in short-term studies (17,18), but CD28 expression is reportedly stable on human CD8⁺CD28⁺ T cell clones (10). To investigate whether activation of human naive (i.e. neonatal) or memory (in adults) CD8⁺CD28⁺ T lymphocytes generates different types of descendants and their lineage relationships, we have developed long-term cultures of purified CD8⁺CD28⁺ T cells, in which we tested the differentiating role of IL-4 by monitoring the expression of CD28 and CD57 antigens throughout activation and cell expansion.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Antibodies and immunofluorescent flow cytometry
The following mouse mAb were used: anti-CD4 (T4, FITC-conjugated IgG1; Coulter, Hialeah, FL); anti-CD8 [CD8, phycoerythrin (PE)–cyanin 5.1-conjugated IgG1; Immunotech, Marseilles, France]; anti-CD28 (Leu-28, PE-conjugated IgG1) and anti-CD57 (Leu-7, FITC-conjugated IgM) were from Becton Dickinson (Erembodegem, Belgium). Irrelevant mouse IgG1 and IgM directly conjugated with the appropriate fluorochrome were used as negative controls. Staining procedures were carried out as described previously (4,19). Usually, 10,000 viable cells were analyzed using an Epics Elite (Coulter) FACS, with particular attention paid to preservation of blast cells after gating on forward and side scatter.

Cell isolation and cell cultures
Peripheral blood mononuclear cells (PBMC) were prepared from 15 healthy adult volunteers (aged 20–50 years) and from five human umbilical cord bloods by centrifugation over Ficoll-Paque. All cell cultures were conducted in RPMI 1640 supplemented with 10% FCS, 2 mM L-glutamine, 2 mM sodium pyruvate and antibiotics, at a starting density of 1x10⁶/ml as 1 ml aliquots in flat-bottom 24-well plates. PBMC were stimulated by soluble anti-CD3 mAb (IOT3, 2.5 ng/ml; Immunotech) added only once at the time of seeding (day 0). From day 3 on, 20 U/ml of recombinant (r) human IL-2 (rIL-2; Boeringher Mannheim, Mannheim, Germany) was added and replaced every 2–3 days. CD8⁺CD28^bright T cells [CD28 mean fluorescence intensity (MFI) >10] were purified by positive selection from 5 day stimulated PBMC cultures to achieve purity >98%. The sorted cells were re-cultured immediately with anti-CD3 (2.5 ng/ml) and 40 U/ml of rIL-2, and then propagated with rIL-2 (20 U/ml) added under the same conditions as for unfractionated PBMC. In some experiments, cultures were propagated in parallel with rIL-2 alone or with rIL-2 (20U/ml) + rIL-4 (5 U/ml; Boeringher Mannheim).

TCR analysis
Semi-quantitative RT-PCR assays were carried out according to published protocols (20). Briefly, total cellular RNA was isolated from lymphocytes with RNAzol B (Bioprobes Systems, Montreuil, France), and cDNA was synthesized using oligo-dT primers and murine MLV reverse transcriptase (Gibco/BRL, Cergy-Pontoise, France) in the presence of RNase inhibitor (Pharmacia, Biotech, Les Ulis, France). V_ß-specific cDNA was then amplified with one of a series of 5' sense V_ß-specific primers (21) and a 3' C_ß antisense primer, using 5 U of rTth DNA polymerase (Perkin Elmer). As an internal control, C region primers were included in the reaction tube. The amplified products were size-separated by electrophoresis on a 2% agarose gel and transferred to a nylon membrane. Blots were pre-hybridized and hybridized with ³²P-labeled 5'-C and 5'-C_ß probes, and the autoradiograms were analyzed by densitometry.

Analysis of expression of CD28 mRNA
RNA was isolated from a constant number (5x10⁵) of cells for each purified subpopulation and cDNA synthesized under the same conditions as used for V_ß repertoire analysis. CD28, actin primers and PCR conditions were those used by Lake et al. (17), with a number of cycles within the exponential phase of the PCR reactions. CD28 and actin were co-amplified in triplicate. The PCR products were size-separated by electrophoresis on a 2% agarose gel. Ratios of the densitometry of CD28 (539 bp splice variant) to actin were then compared. Doubling dilutions of initial cDNA were tested to check the linearity of the PCR reactions.

Measurement of telomeric length
Mean terminal restriction fragment length was determined as described in (13,22). Briefly, a constant number (5x10⁵) of cells was harvested for each purified subpopulation. DNA was extracted using a genomic DNA purification kit (Promega, Charbonnières, France), double-digested with HinfI and RsaI (Boehringer), and electrophoresed on a 0.5% agarose gel. Hybridization was performed with a ³²P-labeled telomeric repeat-specific probe, and the gel was analyzed by densitometry and normalized to the intensity of the less DNA-loaded line.

Cytokine production assays
Cells sorted from the cultures were re-cultured at 1x10⁵ cells/well in flat-bottom 96-well plates in parallel with rIL-2 or in the presence of phorbol myristate acetate (PMA; 3 ng/ml) and ionomycin (250 ng/ml). After 48 and 96 h of secondary culture, the cell-free supernatants were collected from each well, and assayed using specific ELISAs for IFN- (Immunotech), IL-4 and IL-13 (Biotest, Buc, France).

	Results

Top Abstract Introduction Methods Results Discussion References

 
Generation of distinct subpopulations in IL-2-propagated cultures of purified CD8⁺CD28^bright T cells from adult donors
Stimulation of PBMC with mitogenic anti-CD3 mAb and exogenous rIL-2 preferentially expanded the CD8⁺ over the CD4⁺ T cells after 5–7 days, to reach a plateau at ~80% of the total T cells. Over this period, the contribution of the CD28 subset to the expanding CD8⁺ population declined progressively, concomitant with the progressive emergence of CD28^–CD57^– and CD28^–CD57⁺CD8⁺ T cells (Fig. 1A). To investigate whether the decline in CD8⁺CD28⁺ cell numbers stems from the differentiation of primary CD8⁺CD28^bright T cells, whereby CD28 expression is lost with or without acquisition of the CD57 molecule, we purified CD8⁺CD28^bright T cells at day 5 from the primary PBMC cultures. The culture initiated with CD8⁺CD28^brightCD57^– cells (purity >98%) remained exponentially growing for 2–3 weeks. Then, the proliferative capacity decreased until week 5, without an increase in cell death. At this stage, lymphocytes were medium sized. The expression of CD28 and CD57 evolved the same as in whole PBMC cultures, with a gradual fall over time in the proportion of CD28⁺ cells and the progressive appearance of CD57⁺ cells (Fig. 1B). The residual fluorescence from the sorting disappeared after 2–3 days, ruling out loss of CD28 due to the positive selection. There was a declining surface density of the CD28 molecule over time and all the CD28⁺ cells that persisted after spontaneous cessation of proliferation kept a dim expression of CD28 (Fig. 2A): the MFI of the original CD28^bright population was 17.6 ± 3.5 versus 3.9 ± 1.5 for the final CD28^dim cells. The expression level of CD57 increased with culture age (Fig. 2B). In fact, at the time when CD57 expression just arose in the culture, most lymphocytes with this marker were also CD28⁺ (Figs 1C and 3A). The percentage of triple-labeled cells remained stable at ~15% thereafter (Figs 1C and 3D), but these triple-positive cells kept a low expression level of CD57 throughout (MFI < 30; Fig. 3B). Conversely, the steadily expanding CD8⁺CD28^–CD57⁺ cells displayed an increasingly high density of CD57 (MFI > 70; Fig. 3C).

View larger version (44K): [in this window] [in a new window]   Fig. 1. Progressive replacement of CD8+CD28bright T cells by CD8+CD28– cells in long-term cultures. Data from a representative experiment (from 15 healthy adults donors tested). (A) Unfractionated PBMC cultures. PBMC were stimulated with soluble anti-CD3; after 3 days, the cultures were supplemented with rIL-2 (20 U/ml) and fresh medium containing 20 U/ml rIL-2 was replaced every 2–3 days thereafter. (B) Cultures established with purified CD8+CD28bright T cells. Lymphocytes harvested from 5 days stimulated PBMC cultures were FACS-sorted for double-positive cells. The sorted cells (>98% CD8+CD28brightCD57–) were re-cultured immediately and periodically fed rIL-2 (20 U/ml). Cells were harvested at the indicated time-points (presented relative to the initiation of the primary PBMC culture), washed, and analyzed by direct immunofluorescence and triple-color flow cytometry. Gate windows were set to include small to large sized lymphocytes. (C) Three-color flow cytometric analysis of the evolution of CD28 and CD57 expression by CD8+ T cells (the acquisition gate) in the culture of purified CD8+CD28bright T cells presented in (B). Numbers indicate the proportions of cells in each quadrant, determined at days 5 (FACS sorting and initiation of re-culture), 15 and 39. The fourth subpopulation of CD8+CD28+CD57+ cells shown in the dot-plots is not presented in (B) to avoid line superposition.

 

View larger version (25K): [in this window] [in a new window]   Fig. 2. Inverse changes in the densities of cell surface expression of CD28 and CD57 in long-term cultures of purified CD8+CD28brightCD57– lymphocytes. Representative flow cytometric profiles of cultures (as presented in Fig. 1B) analyzed over sequential time-points are shown as overlays and the vertical line delineates the specific staining to exclude ~99% of the isotypic control. (A) Expression level of the CD28 molecule and (B) of the CD57 molecule. Because no significant staining was obtained with the anti-CD57 mAb over week 1 of culture, the corresponding profile is not presented.

 

View larger version (23K): [in this window] [in a new window]   Fig. 3. Comparison of CD8+CD28+CD57+ and CD8+CD57+CD28– T cells generated in a culture initiated with adult CD8+CD28bright T cells (representative of 15 tested) and analyzed by triple-color flow cytometry for the expression of CD28 versus CD57. (A) Dot-plot shows the expression of CD57 by CD28+ (quadrant 2) and CD28– (quadrant 4) CD8+ lymphocytes (the acquisition gate) at day 20. Numbers indicate the percentage of cells in the quadrant. CD8+CD28+ cells (B) and CD8+CD28– cells (C) were gated to draw the FITC histograms to compare their intensity of CD57 expression. The MFI are indicated next to the peaks. (D) Evolution in culture of the percentages of CD8+CD28+CD57+ and CD8+CD28–CD57+ T lymphocytes.

 
Telomeric lengths of subpopulations generated in IL-2-propagated cultures of purified CD8⁺CD28^bright T cells from adult donors
CD8⁺CD28^bright blastic cells could differ in their proliferative capacities. The CD28^dim or the CD28^– cells that persist could then be either the terminally differentiated progeny of CD8⁺CD28^bright T cells or a subset that fails or stops prematurely to proliferate. Thus, we sought to compare the telomeric length of cell subpopulations sorted at different time points of the cultures. As expected, the average telomeric length of the CD8⁺CD28⁺ lymphocytes diminished between day 5 and 19 (–0.6 to –1.2 kb), that could represent six to 12 rounds of cell division (22), consistent with the mean number of population doublings (12 ± 2) of the cultures. Interestingly, telomere shortening was found to be similar in the CD8⁺CD28^–CD57⁺ and the CD8⁺CD28^dimCD57^– subpopulations sorted after 3 weeks of culture (Fig. 4).

View larger version (38K): [in this window] [in a new window]   Fig. 4. Comparison of telomeric length shortening of the subpopulations derived from adult CD8+CD28bright T lymphocytes after different times of culture. The resultant CD8+CD28dimCD57– and CD8+ CD28–CD57+ cells were sorted by FACS at the indicated time. A representative Southern hybridization blot of terminal restriction fragments is presented. The lane corresponding to the CD8+CD28brightCD57– lymphocytes on day 5 shows the telomeric length of these cells just after sorting from the primary PBMC culture. The results are typical of four similar experiments.

 
TCR diversity of subpopulations generated in IL-2-propagated cultures of purified CD8⁺CD28^bright T cells from adult donors
Differences in the expression pattern of surface markers could be linked to the accumulation of cells already primed in vivo and with a skewed repertoire. The cells were therefore harvested after 3 weeks of culture and FACS-sorted into the four CD8⁺ subpopulations, according to the presence or the absence of CD28 and CD57. The data obtained from four donors did not reveal marked differences in the representation of the TCR V_ß transcripts between the four output subpopulations, and whenever a few V_ß transcripts were over-represented, it was among several final subpopulations (Fig. 5).

View larger version (32K): [in this window] [in a new window]   Fig. 5. Comparison of TCR Vß expression between the four CD8+ subpopulations generated in long-term cultures of adult CD8+CD28bright cells. After 3 weeks, the resultant cells were sorted by FACS according to the presence or the absence of CD28 and CD57 molecules. The individual subsets were analyzed by a semi-quantitative RT-PCR assay, run on the same day. The histograms show the relative levels of Vß segments in the four CD8+ subpopulations from a representative experiment (out of four).

 
Pattern of cytokine production by subpopulations generated in IL-2-propagated cultures of purified CD8⁺CD28^brightT cells from adult donors
CD8⁺CD28^dimCD57^– and CD8⁺CD28^–CD57⁺ cells generated after 2 weeks of culture were FACS-sorted and re-cultured to compare their production of IFN-, IL-4 and IL-13. Only IL-13 was detected in the supernatants of CD8⁺CD28^dim cells re-cultured with rIL-2 alone. However, when the sorted cells were re-cultured in the presence of PMA and ionomycin for 48–96 h, the two subpopulations displayed a similar pattern of cytokine release (Table 1).

View this table: [in this window] [in a new window]   Table 1. Production of cytokines by the CD8+CD28–CD57+ and the CD8+CD28dimCD57– subpopulations generated in cultures of purified CD8+CD28bright T cells

 
Prevailing generation of CD8⁺CD28^–CD57^– cells in IL-2-propagated cultures of purified CD8⁺CD28^brightT cells from umbilical cord blood
In long-term cultures of CD8⁺CD28^brightCD57^– T cells from umbilical cord blood, the expression of CD28 declined steadily. Few (<5–25%) CD8⁺CD28^dim T cells persisted and virtually no cells acquired the CD57 molecule: the prevailing final subpopulation was lacking both CD28 and CD57 expression (Fig. 6A and B).

View larger version (22K): [in this window] [in a new window]   Fig. 6. Comparison of long-term cultures of CD8+CD28bright cells purified from human umbilical cord blood and from healthy adults, propagated either in IL-2 or in IL-2 + IL-4. Purified CD8+CD28bright T cells were cultured in parallel and periodically supplemented with either rIL-2 alone (20 U/ml) (A and B) or rIL-2 (20 U/ml) + recombinant IL-4 (5 U/ml) (C and D). Cell cultures were harvested at the indicated time-points, and analyzed for their percentages of CD8+CD28+ (A and C) and CD8+CD57+ T cells (B and D). The results are typical of five similar experiments.

 
Effect of IL-4 on the differentiation of purified CD8⁺CD28^bright T cells
Human T cells respond to IL-4, but not to IL-13 (23). To determine whether the differentiating cytokine IL-4 could impact on the expression pattern of surface markers, CD8⁺CD28^brightCD57^– T cells were propagated comparatively in rIL-2 alone and in rIL-2 + rIL-4. No differences in the mean number of population doublings and cell death were observed between these two conditions of culture. Exogenous rIL-4 maintained a significantly higher proportion of CD28^dim cells in both neonatal and adult lymphocyte cultures (Fig. 6C). Loss of cell surface expression of CD28 in the cultures incubated with rIL-2 alone was linked to a decrease in CD28 message over week 3 of culture, while level of CD28 message was maintained in lymphocytes cultured in rIL-2 + rIL-4 (Table 2). These results were confirmed in one experiment using real-time quantitative PCR (TaqMan PCR; data not shown). Delayed addition of rIL-4 (from day 10 or 14 onwards) to the cultures did not change the declining pattern of CD28 expression (data not shown). In cultures from adult donors, addition of rIL-4 also resulted in earlier acquisition of the CD57 molecule and more CD8⁺CD28^dimCD57^dim cells, but the final proportion of CD57⁺ lymphocytes was the same as with rIL-2 alone. In cultures from umbilical cord blood, rIL-4 could induce the expression of the CD57 molecule on a small proportion (5–20%) of the final cells (Fig. 6D).

View this table: [in this window] [in a new window]   Table 2. Exogenous recombinant human IL-4 maintains CD28 message and cell-surface expression in long-term cultures of purified CD8+CD28bright T cells

 

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
The experiments presented here indicate that the original CD8⁺CD28^brightCD57^– T cells give rise to CD28^– progeny, but that IL-4 prompts CD8⁺ T cells to maintain CD28 expression in long-term cultures. The purified CD8⁺CD28^bright T cells could sustain their expansion and their differentiation similar to that in unfractionated cultures, presumably because a physiological level of co-stimulation had already been provided by accessory cells through CD28 engagement during the few days before cell sorting. This finding indicates that, once activated, CD8⁺CD28^bright T cells no longer require to be intermixed with CD4⁺ T cells and/or adherent cells to regulate their phenotypic profile. Provision of IL-2 was required because, normally, most CD8⁺ T cells do not produce enough IL-2 to support their own expansion.

All cultures initiated with pure CD8⁺CD28^brightCD57^– T cells and maintained in recombinant human IL-2 alone were characterized by a declining expression of CD28, yielding many CD28^– and some CD28^dim cells. We cannot formally exclude the possibility that CD28^– T cells derive from the expansion of CD28^– contaminants, but it seems unlikely because CD8⁺CD28^bright cells were sorted at purity >98% and, given the poor proliferative capacity of CD28^– cells (8,10,11), such a minute input could not overgrow the bulk of CD28⁺ cells. In agreement with our data, Posnett et al. (24) have recently used BrdU labeling to show that CD8⁺CD28^– cells derive from CD8⁺CD28⁺ precursors in vitro. Down-regulation of CD28 transcripts and cell-surface expression following in vitro activation have been investigated only in short-term cultures (1,17,18). The CD28^dim cells might correspond to a transitional population that can progress further to CD28^–. Nevertheless, some cells were found to maintain CD28 at a low density until the end of the cultures and others have reported that clones derived from CD8⁺CD28⁺ T cells continue to express CD28 at reduced densities (10). Persistence versus loss of CD28 expression might reflect differences in mitotic progression, but telomere shortening was found to be similar in the CD28^– and the CD28^dim subpopulations sorted after 3 weeks of culture. This indicates that either cells have not prematurely stopped proliferating and end-stage CD8⁺CD28^dim T cells would then correspond to a minor, but stable, terminally differentiated phenotype.

In cultures initiated with purified CD8⁺CD28^brightCD57^– T cells from healthy adults, a sizable proportion of cells exhibited a progressive shift towards a CD57⁺ phenotype, but not all CD28^– lymphocytes acquired the CD57 marker. In vitro conversion of CD57^– to CD57⁺ lymphocytes by a secondary allogeneic mixed lymphocyte reaction has been reported previously, but expression of CD28 was not addressed (14). The expression of CD28 and CD57 markers is largely reciprocal on blood lymphocytes. Accordingly, most CD57⁺ cells lacked CD28 expression at the plateau phase of growth, but their accumulation was preceded by the emergence of a conspicuous proportion of CD8⁺ cells expressing both CD28 and CD57. That their surface density of CD57 was always lower than on CD28^– cells might designate these cells as a transitional population, but some CD8⁺CD28^dimCD57^dim cells persisted throughout, suggesting that triple-positive cells can emerge through a distinct, albeit minor, differentiation process.

Several observations suggest that the phenotypic evolutions we detected in long-term cultures from healthy adult donors also occur in vivo. First, CD8⁺ T cells with high, moderate and low CD28 density can be identified in normal adult blood (25). Second, progressive emergence of CD28^– and/or CD57⁺CD8⁺ T cells has been observed in healthy human adults, particularly but not exclusively in elderly subjects, and considerably more during many clinical situations associated with chronic immune responses (3–8,11). In human CMV infection, for instance, CD8⁺CD28^brightCD57^– T cells that expand initially are progressively replaced by CD8⁺CD28^– and, finally, CD8⁺CD57⁺ T cells (4), whereas expansion of CD8⁺CD28⁺CD57⁺ T lymphocytes can be detected early in secondary CMV infection (19). Likewise, CD8⁺ lymphocytosis in HIV-seropositive individuals appears to be sustained by cells lacking the CD28 marker (11,12,26). Third, shortened telomeres have been reported in blood CD8⁺CD28^– cells (12,13), and, interestingly, it was estimated that this subset had a replicative history in vivo analogous to the telomere length shortening and population-doubling levels achieved in our cell cultures.

Oligoclonality is common in blood CD8⁺ T lymphocytes of the CD57⁺ or CD28^– subsets in healthy adults or in pathological conditions (5, 27). Our data show that the representation of TCR V_ß genes is fairly similar in the four subpopulations differentiated from CD8⁺CD28^brightCD57^– T cells. While this result might be expected after polyclonal stimulation by anti-CD3 mAb and IL-2, it rules out any developmental link between a difference in the expression pattern of surface markers and skewing of the expressed TCR. Although the cultures retained much of the input TCR diversity, in some cultures, such as the one presented in Fig. 5, a few V_ß transcripts were over-represented among several final subpopulations. Purified CD8⁺CD28^bright lymphocytes from adult donors can indeed contain primed T cells (9,28) that would expand faster and accumulate upon in vitro stimulation. The oligoclonalities observed in vivo would therefore reflect the persistent accumulation of CD28^– and/or CD57⁺CD8⁺ T cells after antigen-driven expansion and differentiation. Consistent with this view, oligoclonal CD8⁺CD57⁺ T cells have been shown to respond to CMV (5), and identical expanded clones within both CD28⁺ and CD28^–CD8⁺ T cell subsets have been detected in HIV-infected individuals (29).

The ability to produce a variety of cytokines is considered to be a typical feature of primed T cells. Only IL-13 was detected in the supernatants of CD8⁺CD28^dim cells recultured in IL-2, consistent with studies showing that this cytokine is mainly produced by CD8⁺ T cells (23). The failure to detect other cytokines in the supernatants may indicate that they are immediately used as autocrine factors. Indeed, after stimulation by PMA and ionomycin, we found that both CD8⁺CD28^dim CD57^– and CD8⁺CD28^–CD57⁺ cells released IFN-, IL-4 and IL-13. While in agreement with several reports (14,15), these results also suggest that differentiation towards either end-stage phenotype imparts no distinction in cytokine production profiles.

The range of CD8⁺ T cells that go on to loose CD28 expression in culture and, among these, to acquire CD57 expression was broad between the adult donors, but in no donor was this trend not apparent. This is in line with the wide inter-individual differences in the proportions of CD28^– and CD57⁺ CD8⁺ T cells in adult blood. Because such cells only accumulate in adults, we investigated whether the phenotypic evolution was different in umbilical cord blood lymphocytes. Contrary to cultures from adult donors, more of the original neonatal CD8⁺CD28^bright cells differentiated to CD8⁺CD28^– and none acquired CD57 expression. Cord blood lymphocytes are considered as a most enriched source of naive T cells. Therefore, conversion to CD28^–CD57^– might be the prevailing differentiation pathway of naive CD8⁺CD28⁺ T cells upon priming. Expression of CD28 is shared between circulating naive and memory CD8⁺ T cells in adult blood (9). Comparison of long-term cultures initiated with neonatal (naive) and adult (naive and memory) CD8⁺CD28⁺ T lymphocytes suggests that re-stimulation of memory cells can push down additional differentiation pathways, including conversion to CD28^–CD57⁺ instead of CD28^–CD57^– as with naive CD8⁺ T cells. Because CD45RA is not a reliable marker of naive cells, since effector-type primed T cells are found within the CD8⁺CD28⁺CD45RA⁺ subpopulations of adult blood lymphocytes (9,28,30), this hypothesis could not be readily tested in adults.

In vivo as well as in vitro, the different outcomes could also be determined by differences within the endogenous cytokine environment, as cord blood CD8⁺ T lymphocytes cannot readily produce IL-4 upon stimulation (31). Human IL-4 initiated a significant shift in the proportion of cells that maintain their membrane expression of CD28 throughout the culture and CD28 message was steadily expressed, while with IL-2 alone CD28 message fell sharply over week 3 of culture. The CD28 promoter contains a signal transducer and activator of transcription (STAT)-6 site, which might be involved in CD28 regulation by IL-4 (32). However, the effect of IL-4 would depend on whether the cell is receiving concomitant activation signals, because a recent report shows that quiescent CD8⁺ T cells down-regulate CD28 when treated by IL-4 (25). Besides, delayed addition of IL-4 to our cultures did not re-induce CD28 on cells that have already lost the marker, indicating that IL-4 can prevent, but cannot reverse, the switch in CD28 expression. In line with a previous report (14), we found that IL-4 also initiated earlier acquisition of the CD57 molecule, but its effect seems to be transient on lymphocytes from adult donors, since the final proportion of CD57⁺ cells was the same as with IL-2 alone. Altogether, it can be surmised from these data that differences in the capacity, and timing, of IL-4 production (and possibly of other cytokines to be tested) could impact on the prevailing phenotype generated from the original CD8⁺CD28^bright T cells. Accordingly, our preliminary results indicate that adding a neutralizing anti-human IL-4 mAb could reduce the proportion of CD28^dim cells that persists in cell cultures from adult donors propagated in IL-2 alone.

In conclusion, our cell culture experiments indicate that the balance between end-stage CD8⁺ subpopulations is controlled by the cytokine environment, IL-4 in particular. Considering that ex vivo generation of these subpopulations precedes spontaneous cessation of proliferation but that end-stage cells exhibit readily detectable effector functions, it may be postulated that most differentiated CD8⁺ T cells would reduce or terminate expression of CD28 co-stimulatory receptors in order to limit their expansion and avoid clonal exhaustion. Conversely, induction of CD57 might change lymphocyte adhesive interactions (33). According to this direct lineage relationship, expansions of CD28^– and CD57⁺CD8⁺ T cells are not a distinctive feature of some special antigenic stimulations, but instead would merely reflect repeated stimulations of the human immune system by any persisting antigen. The prevailing differentiated CD8⁺ subpopulation that accumulates in vivo would then be determined by both the extent of memory cell mobilization and the prevailing cytokine(s) produced during immune responses.

	Acknowledgments

 
This work was supported by grants from the Centre Hospitalier et Universitaire de Lille and the Conseil Régional du Nord-Pas de Calais. The technical assistance of Ms Patricia Dussart is gratefully acknowledged.

	Abbreviations

 

CMV	cytomegalovirus
MFI	mean fluorescence intensity
PBMC	peripheral blood mononuclear cells
PE	phycoerythrin
PMA	phorbol myristate acetate
r	recombinant

	Notes

 
Transmitting editor: M. Feldmann

Received 6 January 1999, accepted 6 May 1999.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Linsley, P. S. and Ledbetter, J. A. 1993. The role of the CD28 receptor during T cell responses to antigen. Annu. Rev. Immunol. 11:191.[Web of Science][Medline]
2. Sperling, A. I., Auger, J. A., Ehst, B. D., Rulifson, I. C., Thompson, C. B. and Bluestone, J. A. 1996. CD28/B7 interactions deliver a unique signal to naive T cells that regulates cell survival but not early proliferation. J. Immunol. 157:3909.[Abstract]
3. Effros, R. B., Boucher, N., Porter, V., Zhu, X., Spaulding, C., Walford, R. L., Kronenberg, M., Cohen, D. and Schachter, F. 1994. Decline in CD28⁺ T cells in centenarians and in long-term T cell cultures: a possible cause for both in vivo and in vitro immunosenescence. Exp. Gerontol. 29:601.[Web of Science][Medline]
4. Labalette, M., Salez, F., Pruvot, F. R., Noel, C. and Dessaint, J. P. 1994. CD8 lymphocytosis in primary cytomegalovirus (CMV) infection of allograft recipients: expansion of an uncommon CD8⁺CD57^– subset and its progressive replacement by CD8⁺CD57⁺ T cells. Clin. Exp. Immunol. 95:465.[Web of Science][Medline]
5. Wang, E. C. Y., Moss, P. A. H., Frodsham, P., Lehner, P. J., Bell, J. I. and Borysiewicz, L. K. 1995. CD8^highCD57⁺ T lymphocytes in normal, healthy individuals are oligoclonal and respond to human cytomegalovirus. J. Immunol. 155:5046.[Abstract]
6. Fagnoni, F. F., Vescovini, R., Mazzola, M., Bologna, G., Nigro, E., Lavagetto, G., Franceschi, C., Passeri, M. and Sansoni, P. 1996. Expansion of CD8⁺ CD28^– T cells in healthy ageing people, including centenarians. Immunology 88:501.[Web of Science][Medline]
7. Fiorentino, S., Dalod, M., Olive, D., Guillet, J. G. and Gomard, E. 1996. Predominant involvement of CD8⁺CD28^– lymphocytes in human immunodeficiency virus-specific cytotoxic activity. J. Virol. 70:2022.[Abstract]
8. Mollet, L., Sadat-Sowti, B., Duntze, J., Leblond, V., Bergeron, F., Calvez, V., Katlama, C., Debre, P. and Autran, B. 1997. CD8^hi+CD57⁺ T lymphocytes are enriched in antigen-specific T cells capable of down-modulating cytotoxic activity. Int. Immunol. 10:311.[Abstract/Free Full Text]
9. Hamann, D., Baars, P. A., Rep, M. H. G., Hooibrink, B., Kerkhof-Garde, S. R., Klein, M. R. and Van Lier, A. W. 1997. Phenotypic and functional separation of memory and effector human CD8⁺ T cells. J. Exp. Med. 186:1407.[Abstract/Free Full Text]
10. Azuma, M., Phillips, J. H. and Lanier, L. L. 1993. CD28^– T lymphocytes. Antigenic and functional properties. J. Immunol. 150:1147.[Abstract]
11. Borthwick, N. J., Bofill, M., Gombert, W. M., Akbar, A. N., Medina, E., Sagawa, K., Lipman, M. C., Johnson, M. A. and Janossy, G. 1994. Lymphocyte activation in HIV-1 infection. II. Functional defects of CD28^– T cells. AIDS 8:431.[Web of Science][Medline]
12. Effros, R. B., Allsopp, R., Chiu, C. P., Hausner, M. A., Hirji, K., Wang, L., Harley, C. B., Villeponteau, B., West, M. D. and Giorgi, J. V. 1996. Shortened telomeres in the expanded CD28^–CD8⁺ cell subset in HIV disease implicate replicative senescence in HIV pathogenesis. AIDS 10:F17.[Web of Science][Medline]
13. Monteiro, J., Batliwalla, F., Ostrer, H. and Gregersen, P. K. 1996. Shortened telomeres in clonally expanded CD28^–CD8⁺ T cells imply a replicative history that is distinct from their CD28⁺CD8⁺ counterparts. J. Immunol. 156:3587.[Abstract]
14. Dupuis d'Angeac, A., Monier, S., Pilling, D., Travaglio-Encinoza, A., Rème, T. and Salmon, M. 1994. CD57⁺ T lymphocytes are derived from CD57^– precursors differentiation occurring in late immune responses. Eur. J. Immunol. 24:1503.[Web of Science][Medline]
15. Hebib, C., Leroy, E., Rouleau, M., Fornairon, S., Metivier, D., Hirsch, F., Kroemer, G., Legendre, C., Senik, A. and Charpentier, B. 1998. Pattern of cytokine expression in circulation CD57⁺ T cells from long-term renal allograft recipients. Transplant. Immunol. 6:39.[Web of Science][Medline]
16. Borthwick, N. J., Bofill, M., Hassan, I., Panayiotidis, P., Janossy, G., Salmon, M. and Akbar, A. N. 1996. Factors that influence activated CD8⁺ T cells apoptosis in patients with acute herpesviruses infections: loss of costimulatory molecules CD28, CD5 and CD6 but relative maintenance of Bax and Bcl-X expression. Immunology 88:508.[Web of Science][Medline]
17. Lake, R. A., O'Hehir, R. E., Verhoef, A. and Lamb, J. R. 1993. CD28 mRNA rapidly decays when activated T cells are functionally anergized with specific peptide. Int. Immunol. 5:461.[Abstract/Free Full Text]
18. Linsley, P., Bradshaw, J., Urnes, M., Grosmaire, L. and Ledbetter, J. 1993. CD28 engagement by B7/BB-1 induces transient down-regulation of CD28 synthesis and prolonged unresponsiveness to CD28 signaling. J. Immunol. 150:3161.[Abstract]
19. Hazzan, M., Labalette, M., Noel, C., Lelievre, G. and Dessaint, J. P. 1997. Recall response to cytomegalovirus in allograft recipients: mobilization of CD57⁺, CD28⁺ cells before expansion of CD57⁺, CD28^– cells within the CD8⁺ T lymphocyte compartment. Transplantation 63:693.[Web of Science][Medline]
20. Pannetier, C., Cochet, M., Darche, S., Casrouge, A., Zoller, M. and Kourilsky, P. 1993. The sizes of CDR3 hypervariable regions of the murine T cell receptor ß chains vary as a function of the recombined germ-line segments. Proc. Natl Acad. Sci. USA 90:4319.[Abstract/Free Full Text]
21. Genevee, C., Diu, A., Nierat, J., Caignard, A., Dietrich, P. Y., Ferradini, L., Roman-Roman, S. F., Triebel, F. and Hercend, T. 1992. An experimentally validated panel of subfamily-specific oligonucleotide primers (V1-w29/V_ß1-w24) for the study of human T cell receptor variable V gene segment usage by polymerase chain reaction. Eur. J. Immunol. 22:1261.[Web of Science][Medline]
22. Vaziri, H., Schächter, F., Uchida, I., Wei, L., Zhu, X., Effros, R., Cohen, D. and Harley, C. B. 1993. Loss of telomeric DNA during aging of normal and trisomy 21 human lymphocytes. Am. J. Hum. Genet. 52:661.[Web of Science][Medline]
23. Minty, A., Asselin, S., Bensussan, A., Shire, D., Vita, N., Vyakarnam, A., Wijdenes, J., Ferrara, P. and Caput, D. 1997. The related cytokines interleukin-13 and interleukin-4 are distinguished by differential production and differential effects on T lymphocytes. Eur. Cytokine Netw. 8: 203.[Web of Science][Medline]
24. Posnett, D. N., Edinger, J. W., Manavalan, J. S., Irwin, C. and Marodon, G. 1999. Differentiation of human CD8 T cells: implications for in vivo persistence of CD8⁺CD28^– cytotoxic effector clones. Int. Immunol. 11: 229.[Abstract/Free Full Text]
25. Lloyd, T. E., Yang, L., Tang, D. N., Bennett, T., Schober, W. and Lewis, D. E. 1997. Regulation of CD28 costimulation in human CD8⁺ T cells. J. Immunol. 158:1551.[Abstract]
26. Caruso, A., Licenziati, S., Canaris, A. D., Cantalamessa, A., Fiorentini, S., Ausenda, S., Ricotta, D., Dima, F., F. Malacarne, F., Balsari, A. and Turano, A. 1998. Contribution of CD4⁺, CD8⁺ CD28⁺ and CD8⁺ CD28^– T cells to CD3⁺ lymphocyte homeostasis during the natural course of HIV-1 infection. J. Clin. Invest. 101:137.[Web of Science][Medline]
27. Gorochov, G., Debre, P., Leblond, V., Sadat-Sowti, B., Sigaux, F. and Autran, B. 1994. Oligoclonal expansion of CD8⁺ CD57⁺ T cells with restricted T cell receptor beta chain variability after bone marrow transplantation. Blood 83:587.[Abstract/Free Full Text]
28. Wack, A., Cossarizza, A., Heltai, S., Barbieri, D., D'Addato, S., Fransceschi, C., Dellabona, P. and Giulia Casorati, G. 1998. Age-related modifications of the human ß T cell repertoire due to different clonal expansions in the CD4⁺ and CD8⁺ subsets. Int. Immunol. 10:1281.[Abstract/Free Full Text]
29. Mugnaini, E. N., Spurkland, A., Egeland, T., Sannes, M. and Brinchmann, J. E. 1998. Demonstration of identical expanded clones within both CD8⁺CD28⁺ and CD8⁺CD28^– T cell subsets in HIV type 1-infected individuals. Eur. J. Immunol. 28:1738.[Web of Science][Medline]
30. Höflich, C., Döcke, W.-C., Busch, A., Kern, F. and Volk, H.-D. 1998. CD45RA^bright/CD11a^bright CD8⁺ T cells: effector T cells. Int. Immunol. 10:1837.[Abstract/Free Full Text]
31. Byun, D. G., Demeure, C., Yang, L., Shu, U., Ishihara, H., Vezzio, N., Gately, M. and Delespesse, G. 1994. In vitro maturation of neonatal human CD8 T lymphocytes into IL-4 and IL-5 producing cells. J. Immunol. 153:4862.[Abstract]
32. Lee, K. P., Taylor, C., Bronislawa, P., Turka, L. A., June, C. H. and Thompson, C. B. 1990. The genomic organization of the CD28 gene. Implications for the regulation of CD28 mRNA expression and heterogeneity. J. Immunol. 145:344.[Abstract]
33. Needham, L. K. and Schnaar, R. L. 1993. The HNK-1 reactive sulfoglucuronyl glycolipids are ligands for L-selectin and P-selectin but not E-selectin. Proc. Natl Acad. Sci. USA 90:1359.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				R. J. Simpson, G. D. Florida-James, C. Cosgrove, G. P. Whyte, S. Macrae, H. Pircher, and K. Guy High-intensity exercise elicits the mobilization of senescent T lymphocytes into the peripheral blood compartment in human subjects J Appl Physiol, July 1, 2007; 103(1): 396 - 401. [Abstract] [Full Text] [PDF]

				W. K. Chiu, M. Fann, and N.-p. Weng Generation and Growth of CD28nullCD8+ Memory T Cells Mediated by IL-15 and Its Induced Cytokines J. Immunol., December 1, 2006; 177(11): 7802 - 7810. [Abstract] [Full Text] [PDF]

				N. L. Alves, F. A. Arosa, and R. A. W. van Lier IL-21 Sustains CD28 Expression on IL-15-Activated Human Naive CD8+ T Cells J. Immunol., July 15, 2005; 175(2): 755 - 762. [Abstract] [Full Text] [PDF]

				M. P. Weekes, M. R. Wills, J. G. P. Sissons, and A. J. Carmichael Long-Term Stable Expanded Human CD4+ T Cell Clones Specific for Human Cytomegalovirus Are Distributed in Both CD45RAhigh and CD45ROhigh Populations J. Immunol., November 1, 2004; 173(9): 5843 - 5851. [Abstract] [Full Text] [PDF]

				J. Bukczynski, T. Wen, K. Ellefsen, J. Gauldie, and T. H. Watts Costimulatory ligand 4-1BBL (CD137L) as an efficient adjuvant for human antiviral cytotoxic T cell responses PNAS, February 3, 2004; 101(5): 1291 - 1296. [Abstract] [Full Text] [PDF]

				H. Uda, T. Mima, N. Yamaguchi, Y. Katada, M. Fukuda, N. Fujii, K. Nakamura, and O. Saiki Expansion of a CD28-Intermediate Subset among CD8 T Cells in Patients with Infectious Mononucleosis J. Virol., June 5, 2002; 76(13): 6602 - 6608. [Abstract] [Full Text] [PDF]

				M. R. Wills, G. Okecha, M. P. Weekes, M. K. Gandhi, P. J. G. Sissons, and A. J. Carmichael Identification of Naive or Antigen-Experienced Human CD8+ T Cells by Expression of Costimulation and Chemokine Receptors: Analysis of the Human Cytomegalovirus-Specific CD8+ T Cell Response J. Immunol., June 1, 2002; 168(11): 5455 - 5464. [Abstract] [Full Text] [PDF]

				C. J. Pitcher, S. I. Hagen, J. M. Walker, R. Lum, B. L. Mitchell, V. C. Maino, M. K. Axthelm, and L. J. Picker Development and Homeostasis of T Cell Memory in Rhesus Macaque J. Immunol., January 1, 2002; 168(1): 29 - 43. [Abstract] [Full Text] [PDF]

				S. Fiorentini, S. Licenziati, G. Alessandri, F. Castelli, S. Caligaris, M. Bonafede, M. Grassi, E. Garrafa, A. Balsari, A. Turano, et al. CD11b Expression Identifies CD8+CD28+ T Lymphocytes with Phenotype and Function of Both Naive/Memory and Effector Cells J. Immunol., January 15, 2001; 166(2): 900 - 907. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (32) Request Permissions Google Scholar Articles by Labalette, M. Articles by Dessaint, J.-P. Search for Related Content PubMed PubMed Citation Articles by Labalette, M. Articles by Dessaint, J.-P. Social Bookmarking          What's this?

Online ISSN 1460-2377 - Print ISSN 0953-8178

Copyright © 2009 Japanese Society for Immunology

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics