The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ramage, J. M.
Right arrow Articles by Hill Gaston, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramage, J. M.
Right arrow Articles by Hill Gaston, J. S.
The Journal of Immunology, 1999, 162: 704-710.
Copyright © 1999 by The American Association of Immunologists

T Cell Responses to Heat-Shock Protein 60: Differential Responses by CD4+ T Cell Subsets According to Their Expression of CD45 Isotypes1

Judith M. Ramage, Joyce L. Young, Jane C. Goodall and J. S. Hill Gaston2

Department of Medicine, University of Cambridge, Addenbrooke’s Hospital, Cambridge, United Kingdom


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
We demonstrate that human T lymphocytes proliferate in vitro to highly purified human heat-shock protein 60 (Hu.hsp60). The response to this self Ag was confined to the CD45RA+RO- T cell subset, with minimal responses by adult CD45RA-RO+ T cells. Experiments using keyhole limpet hemocyanin as a prototypic novel Ag, or tetanus toxoid as a recall Ag, were consistent with the notion that CD45RA+RO- and CD45RA-RO+ T cell subsets can be designated as naive and memory cells, respectively; thus, responses to Hu.hsp60 were confined to the putative naive subset. In contrast, both CD45RA+RO- and CD45RA-RO+ T cell populations proliferated to bacterial hsp60 from Mycobacterium leprae, Escherichia coli, or Chlamydia trachomatis. However, only CD45RA-RO+ (memory) T cells responded to a mycobacterial hsp60-derived peptide previously defined as a major bacteria-specific epitope. Experiments with cord blood T cells, which are CD45RA+RO- and can be considered truly naive, showed that the peptide could elicit responses from naive T cells in vitro; cord blood cells also responded to Hu.hsp60. Since bacterial hsp60 Ags contain both conserved and nonconserved epitopes, we speculate that in vivo challenge with bacterial hsp60 will activate T cells capable of seeing either type of epitope, but only those that see nonconserved epitopes maintain the CD45RA-RO+ memory phenotype. However, T cells recognizing conserved epitopes, while not apparently being recruited to the memory pool, may nevertheless play a role in immunoregulation, particularly in the context of inflammation, when expression of Hu.hsp60 is increased.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The peripheral pool of T lymphocytes consists of two broad categories, those that have not yet encountered their cognate Ag, i.e., naive or virgin cells, and those that have (1, 2, 3, 4). The latter may be either effector cells, if they have recently encountered Ag and are still in an activated state as a result of antigenic stimulation, or memory cells if they have returned to a quiescent state following Ag encounter, but are primed for subsequent responses to the same Ag. There is evidence that each of these categories of T cell differs in their expression of surface Ags, particularly the isoforms of the CD45 (5, 6). These isoforms are generated by alternative splicing of the CD45 gene to include 0–3 of exons 4, 5, and 6 (termed A, B, and C) (7). High m.w. forms are identified using mAbs specific for different exons, and these vary in different species: Abs to A, B, and C are available for humans, for B in the mouse, and for C in the rat. An Ab identifying the isoform with none of the ABC exons (CD45RO) is also available in humans (8). Responses to recall Ags have been easily demonstrated in human T cells expressing CD45RO and lacking CD45RA (9, 10, 11, 12). Their response to stimulation in vitro is faster than cells expressing high m.w. forms, e.g., CD45RA+RO- cells, which generally respond to nonrecall Ags. In addition, after in vitro stimulation, CD45RA+RO- cells decrease expression of CD45RA and gain expression of CD45RO (5). All of these properties are consistent with the idea that effector/memory cells are CD45RA-RO+ and naive cells CD45RA+RO-.

In this study, we have addressed the question of which T cell subset is capable of mounting responses to three categories of Ag: 1) classical recall Ags, i.e., those against which an individual has been specifically immunized; 2) novel Ags, unlikely to have been previously encountered in vivo; and 3) self Ags. In line with previous work, we show that responses to classical recall Ags are indeed mounted by CD45RA-RO+ T cells, and those to novel Ags by CD45RA+RO- T cells. For a self Ag, the human 60-kDa heat-shock protein (Hu.hsp60)3 responses were confined to the CD45RA+RO- subset. We also examined responses to mycobacterial hsp60 (M.hsp60), an Ag that can be considered a recall Ag in populations such as those tested in this study that have been immunized with Mycobacterium bovis bacille Calmette-Guérin (BCG). However, it shares epitopes with Hu.hsp60, since there is considerable sequence homology between prokaryotic and eukaryotic members of the hsp60 family (13, 14). We find that both T cell subsets are capable of responding to M.hsp60, but that the response to a mycobacteria-specific epitope is confined to the CD45RA-RO+ (memory) subset. Therefore, the peripheral T cell repertoire of healthy individuals contains cells capable of recognizing epitopes in Hu.hsp60 in vitro, although these cells are either not activated in vivo, or following activation, they do not remain within the CD45RA-RO+ subset in peripheral blood.

There is extensive literature on the role of hsp60-specific T cells in the pathogenesis of autoimmune diseases, particularly arthritis, with evidence for both pathogenic and protective T cells responding to hsp60 (15, 16, 17). In adjuvant arthritis, the pathogenic T cells have been found to recognize bacteria-specific epitopes (18), whereas protective cells are directed against epitopes that are conserved (19, 20). This study provides in vitro evidence that T cells specific for epitopes conserved in Hu.hsp60 have the potential to be activated in normal humans following challenge with bacterial hsp60. Unlike bacteria-specific T cells, they are not recruited to the memory pool, a feature that would be consistent with the immunoregulatory role of these cells inferred from experiments in arthritis models.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Antigens

The following Ags were used, at previously determined optimal concentrations: tetanus toxoid (TT) (Statens Seruminstitut, Copenhagen, Denmark) at a 1/200 final dilution; PPD (Statens Seruminstitut) at 10 µg/ml; KLH (Calbiochem, Nottingham, U.K.) at 50 µg/ml; GroEL (Sigma, Poole, U.K.) at 10 µg/ml; and Hu.hsp60 and M.hsp60 (prepared as described below) at 20 and 10 µg/ml, respectively. The peptide M1–15 (1–15 amino acids of M.hsp60) was synthesized using FMOC chemistry (a gift from Dr. P. Life, Glaxo/Wellcome, Stevenage, U.K.) and used at 10 µg/ml.

Purification of Hu.hsp60

It was vital that no contaminating Escherichia coli proteins were present in the recombinant preparations. For this reason, two purification steps were performed for Hu.hsp60. To allow purification using nickel affinity columns, the protein was expressed using the histidine-tag system in E. coli (M15)-transformed cells expressing the vector pQE60 and the repressor (pREP4) plasmid (Qiagen, Chatsworth, CA). Lysates of bacteria expressing Hu.hsp60 were resuspended in buffer (pH 7.8, 50 mM sodium phosphate, 300 mM NaCl) and sonicated on ice (three times, 2 min, using 100 W, vibracell sonicator (Jencons, Bedfordshire, U.K.)) before centrifugation at 10,000 x g for 5 min. This procedure was repeated up to six times. Protein was then solubilized in 6 M urea. The protein sample was resolved by electrophoresis using a 3-mm 10% SDS polyacrylamide gel (Bio-Rad Protean II x1 cell; Bio-Rad Hemel Hempstead, Hertfordshire, U.K.) The position of the 60-kDa band was determined and the protein was extracted by electroelution (Bio-Rad model 422 Electro-Eluter; Bio-Rad) from the appropriate strip of gel. Hu.hsp60 was further purified by elution from a nickel-containing HiTrap affinity column (Pharmacia, Herts, U.K.) using 300 mM imidazole (Sigma).

Recombinant M.hsp60 was also prepared from E. coli disrupted by sonication on ice (three times, 1 min) and centrifugation, but in this instance, the pellet was resuspended each time in CIT-E buffer (20 mM Tris-sodium citrate, 10 mM Di-sodium EDTA). The protein was solubilized in urea (6 M urea, 50 mM sodium phosphate, pH 7, 20 mM EDTA). The sample was then subjected to centrifugation at 8000 x g for 10 min, and the supernatant was collected. Addition of ammonium sulfate to a final concentration of 0.9 M (just below the precipitable point of M.hsp60) removed some unwanted proteins. This protocol was adapted from a procedure described by Lamb et al. (21). The remaining proteins were resolved by electrophoresis and electroeluted as described for Hu.hsp60.

Samples were concentrated using a vivaspin 15-ml concentrator with a 30-kDa molecular mass cutoff (Vivascience, Lincoln, U.K.) and then filter sterilized before using in cell culture. Western blot analysis demonstrated no detectable contamination of either Hu.hsp60 or M.hsp60 by GroEL (the E. coli hsp60).

Preparation of mononuclear cells

Blood packs were obtained from blood transfusion centers (Birmingham and Cambridge) and from volunteers within the laboratory; cord blood was obtained from the Rosie Maternity Hospital (Cambridge, U.K.). PBMC and cord blood mononuclear cells were obtained following centrifugation on Ficoll-Paque (Pharmacia Biotechnology, Milton Keynes, U.K.) at 220 x g for 30 min. To remove immature RBC from cord blood mononuclear cells, anti-glycophorin A (PharMingen, San Diego, CA) was added, followed by two rounds of immunomagnetic depletion with sheep anti-mouse IgG-coated Dynabeads (Dynal A.S., Oslo, Norway).

Separation of CD4+CD45RA+RO- and CD4+CD45RA-RO+ T cells and APCs

APCs were obtained by adhering PBMC in 175-cm3 flasks for 1 h at 37°C. CD45RA+RO- and CD45RA-RO+ T cells were negatively selected from the nonadherent cells using a mixture of Abs followed by immunomagnetic depletion with sheep anti-mouse IgG-coated Dynabeads (Dynal). The following mAbs were used: SN130 (anti-CD45RA (5)) UCHL1 (anti-CD45RO+ (6, 8)), BU12 (anti-CD19 (22)), OKM1 (anti-CD11b (23)), UCHT4 (anti-CD8 CD19 (22)), anti-CD16 (Dako A/S, Glostrup, Denmark), and anti-glycophorin A (PharMingen). Successive rounds of Abs and Dynabeads were incubated with the cells. Two rounds of treatment with Ab and four with beads were required for optimal purity of cells. Purity of the selected T cell subsets was determined by flow-cytometric analysis; >98% purity of subsets was obtained with less than 1% contamination with double-positive (CD45RA+RO+) T cells (results not shown).

Culture of cells

Cells were cultured in RPMI (Sigma) with 5% human male AB+ or A+ serum and 5 mM HEPES (Sigma), as follows: 1 x 106 mononuclear cells/ml or 1 x 106 separated T cells/ml were cocultured with 1 x 105 adherent cells/ml, in 2-ml vol in a 24-well plate (Nunc, Life Technologies, Paisley, U.K.), or in triplicate in 200-µl microtiter tissue culture plates (Falcon; Becton Dickinson, Lincoln Park, CA) with the appropriate concentration of Ag. Two-milliliter cultures were sampled at different time points, between days 4 and 10, by transferring three 100-µl aliquots from each well into 96-well microtiter plates and pulsing with 0.5 µCi [3H]thymidine/well for 6 h. Samples were harvested onto printed filter mats (Wallac, Milton Keynes, U.K.) by means of a cell harvester (Skatron, Oslo, Norway), and [3H]thymidine incorporation was measured using a beta plate counter (LKB Wallac, Turku, Finland).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
PBMC proliferative responses to Hu.hsp60

To determine whether T cells from healthy individuals could respond to Hu.hsp60, PBMC were cultured for up to 10 days with this Ag, and proliferative responses were assessed. Fig. 1Go represents one of four such experiments and shows that in vitro responses to this self Ag could be detected. The proliferative response to Hu.hsp60 peaked at day 7, later than the response to the recall Ag TT, but with kinetics similar to the response to the nonrecall Ag, KLH.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 1. Proliferative responses to Hu.hsp60 by PBMC. PBMC were cultured at 1 x 106/ml in 2-ml wells either alone or with TT, KLH, or Hu.hsp60. Results are expressed as the mean [3H]thymidine uptake (cpm + SEM) of triplicate aliquots. This represents one of four experiments.

 
CD45RA+ T cells are the main responders to self Ags

T cell subsets expressing the isoforms of CD45 associated with memory or naive subsets were rigorously purified from healthy adults and stimulated with KLH (Fig. 2GoA), TT (Fig. 2GoB), or Hu.hsp60 (Fig. 2GoC). The CD45RA+RO- T cells responded to KLH with a response that had not peaked by day 9, whereas the response by CD45RA-RO+ T cells to the same Ag was very low. For TT, a reciprocal pattern was evident; the CD45RA-RO+ T cells showed a response that peaked on day 6, whereas there was no proliferation by the CD45RA+RO- T cells (maximum stimulation index, 4.1 (1456 cpm), as compared with 315 (96,361 cpm) for CD45RA-RO+ cells), even though the same cells were capable of responding to KLH. These results are similar to those that we have published previously (10), and are included in this work to show that the purified subsets used in this study to test responses to hsp60 had identical properties. Responses to Hu.hsp60 were similar to those observed with KLH; the largest proliferative responses were demonstrated in the CD45RA+RO- population, with only minimal responses by the CD45RA-RO+ T cells (Fig. 2GoC). A similar pattern of response was also observed to another purified human Ag, phospholipid-binding protein (data not shown).



View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 2. CD45RO+ and CD45RA+ T cell responses to Hu.hsp60 compared with KLH and to TT. Purified T cells were cultured at 1 x 106/ml with 1 x 105 adherent cells/ml in 2-ml wells either alone, or with KLH (A), TT (B), or Hu.hsp60 (C) for up to 9 days. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) of triplicate aliquots. This represents one of six experiments (A), five experiments (B), and three experiments (C), all giving similar results. This figure illustrates the results from the same individual in the same experiment.

 
As an additional source of purified CD45RA+RO- T cells, we obtained cord blood mononuclear cells (98% CD45RA+, data not shown). Fig. 3Go shows that these cells were also able to proliferate in response to Hu.hsp60, with similar kinetics to the Ag PPD. The CD45RA+ T cells recognize this recall Ag, since cord blood T cells are naive, having not yet encountered PPD or other recall Ags in vivo. Together these experiments are in keeping with the idea that Hu.hsp60-specific T cells are present in the repertoire of the putative naive subset, and absent from the repertoire of the putative memory subset.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 3. Proliferative responses to Hu.hsp60 by cord blood mononuclear cells. Cord blood mononuclear cells were cultured at 1 x 106/ml in 2-ml wells either alone, or with PPD or Hu.hsp60. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) of triplicate aliquots. This represents one of two experiments, both of which gave similar results.

 
Both CD45RA-RO+ and CD45RA+RO- T cells respond to bacterial hsp60

As heat-shock proteins have highly conserved amino acid sequences, there are regions within bacterial hsp60 that are homologous to Hu.hsp60. When proliferative responses to M.hsp60 were investigated, both the CD45RA-RO+ and CD45RA+RO- T cells responded (Fig. 4Go), unlike the findings with either KLH or TT. In the experiment shown in Fig. 4Go, the response by CD45RA-RO+ T cell peaked on day 5, while the response by CD45RA+RO- T cells peaked on day 8. In additional experiments, the precise days on which responses peaked varied: days 4 to 6 for CD45RA-RO+ T cells, and days 6 to 9 for CD45RA+RO- T cells, but in all cases, in which cells from the same donor were assessed in parallel, the response by CD45RA-RO+ T cells always peaked before that of CD45RA+RO- T cells. Thus, in contrast to their response to TT, healthy individuals appeared to have both memory and naive T cells capable of responding to M.hsp60 in vitro, as judged by both the responding T cell subsets and the kinetics of their responses. Similar results were found using GroEL, the E. coli hsp60 (Fig. 5Go) and Chlamydia trachomatis hsp60 (data not shown).



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 4. CD45RO+ and CD45RA+ T cell proliferative responses to M.hsp60. Purified T cells were cultured at 1 x 106/ml with 2 x 105 adherent cells/ml in 2-ml wells either alone or with M.hsp60 for up to 9 days. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) of triplicate. This represents one of seven experiments, all giving similar results.

 


View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 5. CD45RA-RO+ and CD45RA+RO- T cell proliferative responses to GroEL. Purified T cells were cultured at 1 x 106/ml with 1 x 105 adherent cells/ml in 2 ml either alone or with GroEL for up to 9 days. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) of triplicate aliquots. This represents one of three experiments, all giving similar results.

 
Only CD45RA-RO+ cells proliferate to a previously defined epitope in M.hsp60

We and others have shown previously that a peptide consisting of amino acids 1–15 M.hsp60 (M1–15) contains an epitope that is recognized by M.hsp60-specific T cells in the context of HLA-DR3 (24, 25, 26, 27). This peptide is able to bind to HLA-DR3 (shown by using a biotinylated peptide and detecting binding by flow cytometry) and is recognized by M.hsp60-specific T cell clones (data not shown). DR3+ individuals commonly mount strong responses to this epitope, and to date, no other DR3-restricted has been described; the sequence of the epitope is not conserved in Hu.hsp60. We examined responses to this peptide mounted by purified CD45RA+RO- and CD45RA-RO+ T cells from a DR3+ individual. Fig. 6Go shows proliferation by the CD45RA-RO+ T cells at day 5 both to intact M.hsp60 and to M1–15. In contrast, on day 8, the CD45RA+RO- T cells proliferated strongly to M.hsp60, but only minimal responses to M1–15 were observed.



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 6. CD45RO+ and CD45RA+ T cell proliferative responses to M.hsp60 and M1–15 by a HLA-DR3+ individual. Purified T cells were cultured at 1 x 106/ml with 1 x 105 adherent cells/ml in triplicate in 96U wells with and without Ag. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) on day 5 for CD45RA+RO- T cells and on day 8 for CD45RA+RO- T cells. This represents one of two experiments, both giving similar results.

 
To rule out the possibility that a lack of a response to M1–15 by CD45RA+RO- T cells was due to inadequate culture conditions for detecting proliferative responses to peptides by this subset, we again used cord blood as a source of CD45RA+RO- T cells. Fig. 7GoA shows the results using cord blood from a HLA-DR3+ neonate. Clear proliferative responses to the peptide M1–15 were observed which peaked on day 7 when the same culture conditions (i.e., cell number and Ag concentration) were used as in those testing adult CD45RA+RO- cells. The kinetics of this response was similar to that observed when adult CD45RA+RO- T cells were cultured with the novel Ag KLH, and as previously shown, these cells, unlike adult CD45RA+RO- cells, also responded to TT (10 . Fig. 7GoB demonstrates that cord blood cells can respond to various bacterial hsp60 Ags and other recall Ags with similar kinetics to the nonrecall Ag KLH.



View larger version (24K):
[in this window]
[in a new window]
  		FIGURE 7. Proliferative responses of cord blood mononuclear cells. A, Cord blood mononuclear cells from a HLA-DR3+ individual respond to peptide M1–15. Mononuclear cells obtained from cord blood were cultured at a concentration of 2 x 106 cells in 2-ml wells with Ag for up to 9 days. Results are expressed as the mean [3H]thymidine uptake (cpm ± SEM) of triplicate aliquots. This represents one of six experiments, all giving similar results. B, Cord blood mononuclear cells respond to bacterial hsp60, recall, Ags, and KLH with similar kinetics.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This work demonstrates that healthy individuals have the potential to mount T cell responses to human hsp60 in vitro, showing that the T cell repertoire post thymic selection includes potentially autoreactive T cells capable of recognizing self hsp60. Studies in transgenic mice have also indicated that a population of T cells able to recognize self hsp60 can be released to the periphery, even when hsp60 is overexpressed in the thymus (28). Nevertheless, the same experiments suggest that hsp60 can participate in negative selection in the thymus, since overexpression abolished recognition of an otherwise dominant epitope by peripheral T cells. Although there are previous reports of human T cell responses to Hu.hsp60, in most cases the studies were in patients with inflammatory conditions such as oligoarticular juvenile chronic arthritis, Behcet’s disease, recurrent oral ulcers, or salpingitis (29, 30, 31, 32, 33). Furthermore, in some of these studies, peptides were used, introducing the possibility of observing responses to cryptic epitopes, i.e., those that would not be generated by normal Ag processing in vivo.

The results of the present study also show that the subset of CD4+ cells mainly responsible for the response to Hu.hsp60 is CD45RA+RO-. Similar results were described for responses to other self Ags, including phospholipid-binding protein in this study, and, in a previous study, the rhesus Ag expressed by human RBC (34). We have also previously shown that highly purified CD45RA+RO- T cells respond well in vitro to novel foreign Ags, such as KLH, which the individual is unlikely to have previously encountered, but poorly to recall Ags such as TT (10). Conversely, CD45RA-RO+ T cells are the main responders to TT and show minimal responses to KLH. These results are consistent with the idea that expression of CD45RO is a marker of memory T cells and CD45RA a marker of naive T cells. This concept has been challenged, mainly as a result of work in the rat, in which it has been possible to demonstrate that T cells can revert from expressing the low m.w. isoform of CD45 (identified as CD45RC-) to expressing high m.w. isoforms (CD45RC+) (35, 36). Evidence for reversion in humans is controversial. Within CD4+ cells, only the CD45RO+ population showed the elevated expression of LFA-1, which is associated with primed cells (37). Contrariwise, studies of patients following radiotherapy indicate that the CD45RO+ population is relatively short-lived (38), a surprising property for the subset that is believed to contain T cell memory that can last many years. In addition, a recent paper claims that the CD45RA+ subset in humans also contains memory T cells, but that they are refractory to normal stimulation with Ag, requiring additional stimulation through CD28 (39). Nevertheless, under the conditions of our assays, and following our rigorous purification of CD45RO+RA- and CD45RA+RO- subsets, there was a reproducible difference in the response and kinetics of the two subsets, which is consistent with the notion that they represent memory and naive cells, respectively. If memory cells specific for TT resided in the CD45RA+RO- cells that we purified, they were either not stimulated by the conditions of our assay or present in very small numbers, since we did not record any responses to TT under conditions in which significant responses to other Ags (KLH, M.hsp60, Hu.hsp60) were readily demonstrated. It should be noted, however, that our studies have been confined to highly purified subsets, and we have not examined responses by the double-positive CD45RA+RO+ T cells.

If T cells responsive to Hu.hsp60 are found within the putative naive subset, does this imply that activation does not normally occur in vivo? Following activation in vitro, CD45RA+RO- cells switch CD45 isoform expression to CD45RA-RO+, but it is clear that in vivo T cells recognizing Hu.hsp60 have not been permanently recruited to the CD45RA-RO+ T cells subset; otherwise, we would have been able to detect responses in this subset. (It is formally possible that they are recruited to the CD45RA-RO+ subset, but are undetectable, having become unresponsive to Ag, i.e., anergic). Therefore, either Hu.hsp60-specific T cells fail to be recruited to the CD45RA-RO+ subset following activation, or are never activated in vivo, remaining in an immunologically ignorant state, no different from any other naive T cell population. Activation in vivo without recruitment to the CD45RA-RO+ subset could also be a property of a regulatory subset, as has been postulated for Hu.hsp60-specific T cells (see below), and our experiments could not distinguish such cells from those that are truly naive.

To investigate these possibilities further, we studied responses to M.hsp60, a bacterial hsp60, representing a hybrid self/foreign Ag, containing both bacteria-specific epitopes, and epitopes conserved within Hu.hsp60. The subjects were previously immunized with M. bovis BCG 10–20 yr previously. For M.hsp60, responses were clearly demonstrated by both CD45RA-RO+ and CD45RA+RO- T cells, in sharp contrast to the result obtained with the recall Ag TT. Similar results were obtained for two other bacterial hsp60 Ags, GroEL from E. coli and chlamydial hsp60. All individuals are exposed to E. coli as a gut commensal, and ~50% of adults have been infected with Chlamydia pneumoniae or Chlamydia trachomatis (mainly the former), organisms with highly conserved hsp60 Ags in which we have recently demonstrated an identical epitope (40).

Having demonstrated responses to bacterial hsp60 by both T cell subsets, we were able to make use of a previously mapped immunodominant epitope in M.hsp60 (24, 25) to show that these subsets were responding to different epitopes. Responses to the bacteria-specific epitope were readily demonstrated in the CD45RA-RO+ subset, while they were absent from the CD45RA+RO- subset. Thus, in vitro processing of M.hsp60 allows presentation of epitopes that can be recognized by both subsets; although it is possible to postulate that processing will occur differently in vivo, it is difficult to envision a mechanism whereby bacteria-specific epitopes are presented, but those resembling self are not, simply because of differential processing. The inference must be that in vivo challenge with M.hsp60 should result in the activation of T cells recognizing both bacteria-specific and self-specific epitopes.

To gain further insight into T cell recognition of hsp60, we used cord blood both as a source of CD45RA+RO- cells, and of cells that are unlikely to have previously encountered Ag, i.e., a truly naive population (41). It lacks CD45RA-RO+ cells, and a recent study demonstrated that the murine placenta was an efficient barrier against transfer of hepatitis B virus e Ag (42). With respect to Hu.hsp60 and novel Ags such as KLH, the responses of cord blood T cells and the adult CD45RA+RO- cells were similar, consistent with both being naive populations. The principal difference was that adult naive cells failed to recognize the recall Ag TT, or the bacteria-specific epitope in M.hsp60, whereas cord blood cells responded to both (Fig. 7Go). The apparent lack of responses by the adult naive subset to recall Ags and epitopes requires explanation. It might be supposed that a truly naive subset would be able to respond to any Ag, irrespective of its classification as recall or novel. We have explained previously our results by suggesting that immunization with TT in vivo results in the conversion of all of the CD45RA+RO- T cells capable of recognizing TT into CD45RO+-expressing cells, thereby effectively purging the CD45RA+RO- T cell population of this specificity, and that persistent Ag allows purging to continue. The same argument can apply to M.hsp60; following immunization with BCG, mycobacterial Ags persist in vivo for long periods of time: M. bovis BCG establishes a low level persistent infection, and strong T cell responses to mycobacterial Ags are maintained indefinitely. This would allow continuous purging of the naive repertoire. These concepts receive support from work in the rat, demonstrating that reexpression of CD45RC by activated T cells only occurs in the absence of Ag; otherwise, the CD45RC- phenotype is retained (43).

In this context, our observation of CD45RA+RO- T cells that recognize the recall Ag M.hsp60 in the adult population is surprising. We suggest that the adult CD45RA+RO- T cell response to M.hsp60 is likely to be directed against epitopes other than those selected as targets by the memory T cell population, and that these are most probably epitopes that are also present in Hu.hsp60. Cloning studies are currently underway to address this point directly. If CD45RA+RO- T cells are activated by recognition of conserved epitopes in M.hsp60 in vivo, why are they not recruited to the CD45RA-RO+ subset, as appears to occur in vitro? One possibility is that, following their initial activation by bacterial hsp60 presented by dendritic cells, later exposure to continuous and abundant expression of Hu.hsp60 by nonprofessional APCs could result in the induction of anergy, or even their deletion from the CD45RA-RO+ T cell subset. It might also be argued that whereas responses to conserved epitopes in Hu.hsp60 would be absent from normal individuals, they might be evident in patients with autoimmune conditions such as rheumatoid arthritis. However, against this hypothesis, we have noted lower responses to M.hsp60 in rheumatoid arthritis patients as compared with controls (J. M. Ramage and J. S. H. Gaston, manuscript submitted), and in rheumatoid arthritis synovial fluid T cells that are dominated by CD45RA-RO+ cells, responses to Hu.hsp60 were not evident (M. Lillicrap and H. Gaston, unpublished observations).

Is there any physiologic significance in the presence of a T cell subset that is able to recognize self hsp60? It is known that expression of Hu.hsp60 is up-regulated at sites of inflammation (44), and that these conditions also enhance Ag processing and presentation. In addition, self hsp60-reactive T cells have been demonstrated in both humans and rodents, although in most cases they have been observed in the context of immune responses to bacterial hsp60 (45, 46, 47). In the rat adjuvant arthritis model, this population of autoreactive self hsp60-specific T cells has been shown to be important in preventing the onset of arthritis after challenge with adjuvant (20). However, studies in rodents have not determined whether the self hsp60-specific cells are generated from cells expressing the naive phenotype. Likewise, it has not been determined whether, after either activation in vivo or following passive transfer, these effectors are recruited to the memory pool. Therefore, one possible role for human hsp60-specific cells might be the control of the inflammatory response to bacterial Ags through production of cytokines such as TGF-ß, IL-4, and IL-10. If, as suggested by results in the rat adjuvant arthritis model, challenge with bacterial hsp60 results in concomitant activation of regulatory self hsp60-specific T cells, this could serve as an important general mechanism in the control of immune responses to bacterial pathogens. Defective function in such a mechanism could have clinical consequences, leading to the chronicity of inflammation that characterizes diseases such as rheumatoid arthritis and juvenile chronic arthritis. This idea receives some support from the recent correlation between favorable outcome and responses to Hu.hsp60 in juvenile chronic arthritis (48) and from our observations of relatively defective T cell responses to M.hsp60 in early rheumatoid arthritis. To assess the significance of these observations, further characterization of the T cell response to Hu.hsp60 in humans is required.

A separate body of work in the rat has shown that cells with the naive phenotype (CD45RC+) have the potential to induce autoimmune disease, including diabetes and thyroiditis, when transferred into athymic recipients (49). These effects are prevented by cotransfer of CD45RC- cells (50, 51). Likewise, inflammatory bowel disease is observed in SCID mice when CD45RBhigh cells are transferred and the disease is ameliorated by CD45RBlow cells (52, 53). Hence, it may be that the self-reactive properties of cells within the naive subset, which may be regulated by appropriate responses to foreign Ags, may be capable of inducing autoimmune disease in the absence of reciprocal regulation by cells of the memory subset. Results of our continuing studies will address whether the Hu.hsp60-reactive T cells found within the naive subset in normal adults may function to induce autoimmunity or, as outlined above, regulate inflammation.


   Footnotes
 
1 This work was supported by the Medical Research Council and the Arthritis Research Campaign. Back

2 Address correspondence and reprint requests to Dr. J. S. Hill Gaston, Department of Medicine, University of Cambridge, Level 5, Addenbrooke’s Hospital, Cambridge, CB2 2QQ, U.K. E-mail address: Back

3 Abbreviations used in this paper: Hu.hsp60, human heat-shock protein 60; BCG, bacille Calmette-Guérin; GroEL, E. coli hsp60; KLH, keyhole limpet hemocyanin; M.hsp60, mycobacterial heat-shock protein 60; PPD, purified protein derivative; TT, tetanus toxoid. Back

Received for publication June 5, 1998. Accepted for publication September 28, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Sprent, J.. 1994. T-memory and B-memory cells. Cell 76:315.[Medline]
  2. Sprent, J.. 1997. Immunological memory. Curr. Opin. Immunol. 9:371.[Medline]
  3. Ahmed, R., D. Gray. 1996. Immunological memory and protective immunity: understanding their relation. Science 272:54.[Abstract]
  4. Gray, D.. 1993. Immunological memory. Annu. Rev. Immunol. 11:49.[Medline]
  5. Akbar, A. N., L. Terry, A. Timms, P. C. L. Beverley, G. Jannossy. 1988. Loss of CD45R and gain of UCHL1 reactivity is a feature of primed T cells. J. Immunol. 140:2171.[Abstract]
  6. Smith, S. H., M. H. Brown, D. Rowe, R. E. Callard, P. C. Beverley. 1986. Functional subsets of human helper-inducer cells defined by a new monoclonal antibody, UCHL1. Immunology 58:63.[Medline]
  7. Streuli, M., L. Hall, Y. Saga, S. Schlossman, H. Saito. 1987. Differential usage of three exons generates at least five different mRNAs encoding human leukocyte common antigens. J. Exp. Med. 166:1548.[Abstract/Free Full Text]
  8. Terry, L. A., M. H. Brown, P. C. Beverley. 1988. The monoclonal antibody, UCHL1, recognizes a 180,000 MW component of the human leukocyte-common antigen, CD45. Immunology 64:331.[Medline]
  9. Young, J. L., A. Daser, P. C. L. Beverley. 1995. In vitro proliferative responses of human peripheral blood mononuclear cells to non-recall antigens. J. Immunol. Methods 182:177.[Medline]
  10. Young, J. L., J. M. Ramage, J. S. H. Gaston, P. C. L. Beverley. 1997. In vitro responses of human CD45R0brightRA- and CD45R0-RAbright T cell subsets and their relationship to memory and naive T cells. Eur. J. Immunol. 27:2383.[Medline]
  11. Plebanski, M., M. Saunders, S. S. Burtles, S. Crowe, D. C. Hoooper. 1992. Primary and secondary human in vitro T-cell responses to soluble antigens are mediated by subsets bearing different CD45 isoforms. Immunology 75:86.[Medline]
  12. Plebanski, M., S. S. Burtles. 1994. In vitro primary responses of human T cells to soluble protein antigens. J. Immunol. Methods 170:15.[Medline]
  13. Young, D., A. Mehlert, V. Bal, P. Mende-Samperio, J. Ivanyi, J. Lamb. 1988. Stress proteins and the immune response to mycobacteria antigens as virulence factors?. Antonie Van Leeuwenhoek 54:431.[Medline]
  14. Kaufmann, S., B. Schoel, E. J. Van, T. Koga, A. Wand-Wurttenberger, M. Munk, U. Steinhoff. 1991. Heat shock protein 60: implications for pathogenesis of and protection against bacterial infections. Immunol. Rev. 121:67.[Medline]
  15. Cohen, I.. 1991. Autoimmunity to chaperonins in the pathogenesis of arthritis and diabetes. Annu. Rev. Immunol. 9:567.[Medline]
  16. Gaston, J.. 1991. Heat shock proteins and autoimmunity. Semin. Immunol. 3:35.[Medline]
  17. Gaston, J. S. H., J. H. Pearce. 1994. Immune responses to heat shock proteins in reactive arthritis. W. van Eden, ed. Stress Proteins in Medicine Marcel Dekker,
  18. van Eden, W., J. Thole, R. van der Zee, A. Noordzij, J. van Embden, E. Hensen, I. Cohen. 1988. Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature 331:171.[Medline]
  19. Anderton, S. M., R. van der Zee, A. Noordzij, W. van Eden. 1994. Differential mycobacterial 65-kDa heat shock protein T cell epitope recognition after adjuvant arthritis-inducing or protective immunization protocols. J. Immunol. 152:3656.[Abstract]
  20. Anderton, S. M., R. van der Zee, B. Prakken, A. Noordzij, W. van Eden. 1995. Activation of T cells recognizing self 60-kD heat shock protein can protect against experimental arthritis. J. Exp. Med. 181:943.[Abstract/Free Full Text]
  21. Lamb, F., A. Kingston, I. Estrada, M. Colston. 1988. Heterologous expression of the 65 kilodalton antigen of Mycobacterium leprae and murine T cell responses to the gene product. Infect. Immun. 56:1237.[Abstract/Free Full Text]
  22. Dorken, B., G. Moldenhauer, P. Moller, A. Pezzutto, and A. Schwartz-Albiez. 1989. Leukocyte Typing IV. W. Knapp, B. Dorken, W. R. Gilks, E. P. Rieber, R. E. Schmidt, H. Stein, and A. E. G. Kr.von dem Borne, eds. Oxford University Press, Oxford, p. 15.
  23. Breard, J., E. L. Reinherz, P. C. Kung, G. Goldstein, S. F. Schlossman. 1980. A monoclonal antibody reactive with human peripheral blood monocytes. J. Immunol. 124:1943.[Abstract]
  24. Gaston, J. S. H., P. F. Life, P. J. Jenner, M. J. Colston, P. A. Bacon. 1990. Recognition of a mycobacteria specific epitope in the 65 kD heat shock protein by synovial fluid derived T cell clones. J. Exp. Med. 171:831.[Abstract/Free Full Text]
  25. van Schooten, W., D. Elferink, J. van Embden, D. Anderson, R. de Vries. 1989. DR3 restricted T cells from different HLA DR3 positive individuals recognize the same peptide (amino acids 2–12) of the mycobacterial 65 kDa heat shock protein. Eur. J. Immunol. 19:2075.[Medline]
  26. Geluk, A., K. E. van Meijgaarden, S. Southwood, C. Oseroff, J. W. Drijfhout, R. R. P. De Vries, T. H. M. Ottenhoff, A. Sette. 1994. HLA-DR3 molecules can bind peptides carrying two alternative specific submotifs. J. Immunol. 152:5742.[Abstract]
  27. Geluk, A., X. T. Fu, K. E. Van Meijgaarden, Y. Y. R. Jansen, R. R. P. de Vries, R. W. Karr, T. H. M. Ottenhoff. 1994. T cell receptor and peptide-contacting residues in the HLA-DR17(3) ß1 chain. Eur. J. Immunol. 24:3241.[Medline]
  28. Birk, O., D. Douek, D. Elias, K. Takacs, H. Dewchand, S. Gur, M. Walker, R. van der Zee, I. Cohen, D. Altmann. 1996. A role of hsp60 in autoimmune diabetes: analysis in a transgenic model. Proc. Natl. Acad. Sci. USA 93:1032.[Abstract/Free Full Text]
  29. De GraeffMeeder, E., R. van der Zee, G. Rijkers, H. Schuurman, W. Kuis, J. Bijlsma, B. Zegers, W. van Eden. 1991. Recognition of human 60 kD heat shock protein by mononuclear cells from patients with juvenile chronic arthritis. Lancet 337:1368.[Medline]
  30. De GraeffMeeder, E. R., W. van Eden, G. T. Rijkers, B. J. Prakken, W. Kuis, M. M. Voorhorstogink, R. van der Zee, H. J. Schuurman, P. J. M. Helders, B. J. M. Zegers. 1995. Juvenile chronic arthritis: T cell reactivity to human HSP60 in patients with a favorable course of arthritis. J. Clin. Invest. 95:934.
  31. Witkin, S. S., J. Jeremias, M. Toth, W. J. Ledger. 1994. Proliferative response to conserved epitopes of the Chlamydia trachomatis and human 60-kilodalton heat-shock proteins by lymphocytes from women with salpingitis. Am. J. Obstet. Gynecol. 171:455.[Medline]
  32. Hasan, A., F. Fortune, A. Wilson, K. Warr, T. Shinnick, Y. Mizushima, R. van der Zee, M. Stanford, J. Sanderson, T. Lehner. 1996. Role of {gamma}{delta} T cells in pathogenesis and diagnosis of Behcet’s disease. Lancet 347:789.[Medline]
  33. Stanford, M. R., E. Kasp, R. Whiston, A. Hasan, S. Todryk, T. Shinnick, Y. Mizushima, D. C. Dumonde, R. van der Zee, T. Lehner. 1993. Heat shock protein peptides reactive in patients with Behcet’s disease are uveitogenic in Lewis rats. Clin. Exp. Immunol. 97:226.
  34. Barker, R. N., C. J. Elson. 1994. Multiple self epitopes on the Rheus polypeptides stimulate immunologically ignorant human T cells in vitro. Eur. J. Immunol. 24:1578.[Medline]
  35. Bell, E., S. Sparshott. 1990. Interconversion of CD45R subsets of CD4 T cells in vivo. Nature 348:163.[Medline]
  36. Sparshott, S. M., E. B. Bell, S. R. Sarawar. 1991. CD45R CD4 T cell subset-reconstituted nude rats: subset-dependent survival of recipients and bi-directional isoform switching. Eur. J. Immunol. 21:993.[Medline]
  37. Okumura, M., Y. Fujii, K. Inada, K. Nakahara, H. Matsudfa. 1993. Both CD45RA+ and CD45RA- subpopulations of CD8+ T cells contain cells with high levels of lymphocyte function-associated antigen-1 expression, a phenotype of primed T cells. J. Immunol. 150:429.[Abstract]
  38. Michie, C. A., A. McLean, C. Alcock, P. C. L. Beverley. 1992. Life span of human lymphocyte subsets defined by CD45 isoforms. Nature 360:264.[Medline]
  39. Pilling, D., A. N. Akbar, P. A. Bacon, M. Salmon. 1996. CD4+ CD45RA+ T cells from adults respond to recall antigens after CD28 ligation. Int. Immunol. 8:1737.[Abstract/Free Full Text]
  40. Deane, K., R. Jecock, J. Pearce, J. Gaston. 1997. Identification and characterization of a DR4-restricted T cell epitope within chlamydia hsp60. Clin. Exp. Immunol. 109:439.[Medline]
  41. Hannet, I., F. Erkeller-Yuksel, P. Lydyard, V. Deneys, M. Debruyere. 1992. Developmental and maturational changes in human blood lymphocyte subpopulations. Immunol. Today 13:215.[Medline]
  42. Reifenberg, K., T. Deutschle, J. Wild, R. Hanano, I. Gastrock-Balitsche, R. Schirmbeck, H.-J. Schlicht. 1998. The hepatitis B virus e antigen cannot pass the murine placenta efficiently and does not induce CTL immune tolerance in H-2b mice in utero. Virology 243:45.[Medline]
  43. Bunce, C., E. B. Bell. 1997. CD45RC isoforms define two types of CD4 memory T cells, one of which depends on persisting antigen. J. Exp. Med. 185:767.[Abstract/Free Full Text]
  44. De Graeff-Meeder, E., M. Voorhorst, W. van Eden, H. Schuurman, J. Huber, D. L. Barkley, R. Maini, W. Kuis, G. Rijkers, B. Zegers. 1990. Antibodies to the mycobacterial 65 kd heat shock protein are reactive with synovial tissue of adjuvant arthritic rats and patients with rheumatoid arthritis and osteoarthritis. Am. J. Pathol. 137:1013.[Abstract]
  45. Hermann, E., A. W. Lohse, R. van der Zee, W. van Eden, W. J. Mayet, P. Probst, T. Poralla, K. H. Meyer zum Buschenfelde, B. Fleischer. 1991. Synovial fluid derived Yersinia reactive T cells responding to human 65 kDa heat shock protein and heat stressed antigen presenting cells. Eur. J. Immunol. 21:2139.[Medline]
  46. Quayle, A. J., W. K. Black, S. G. Li, J. Kjeldson-Kragh, F. Oftung, M. Sioud, O. Forre, D. Capra, J. B. Natvig. 1992. Peptide recognition, T cell receptor usage and HLA restriction of human heat-shock protein 60 and mycobacterial 65-kDa hsp-reactive T cell clones from rheumatoid synovial fluid. Eur. J. Immunol. 22:1315.[Medline]
  47. Anderton, S. M., R. van der Zee, J. A. Goodacre. 1993. Inflammation activates self hsp60-specific T-cells. Eur. J. Immunol. 23:33.[Medline]
  48. Prakken, A., W. van Eden, G. Rijkers, W. Kuis, E. Toebes, E. de Graeff-Meeder, R. van der Zee, B. Zegers. 1996. Autoreactivity to human heat shock protein 60 predicts disease remission in oligoarticular juvenile rheumatoid arthritis. Arthritis Rheum. 39:1826.[Medline]
  49. Fowell, D., D. Mason. 1993. Evidence that the T cell repertoire of normal rats contains cells with the potential to cause diabetes: characterization of the CD4+ T cell subset that inhibits this autoimmune potential. J. Exp. Med. 177:627.[Abstract/Free Full Text]
  50. Saoudi, A., S. Simmonds, I. Huitinga, D. Mason. 1995. Prevention of experimental allergic encephalomyelitis in rats by targeting autoantigen to cells: evidence that the protective mechanism depends on changes in the cytokine response and migratory properties of the autoantigen-specific T cells. J. Exp. Med. 182:335.[Abstract/Free Full Text]
  51. Saoudi, A., B. Seddon, V. Heath, D. Fowell, D. Mason. 1996. The physiological role of regulatory T cells in the prevention of autoimmunity: the function of the thymus in the generation of the regulatory T cell subset. Immunol. Rev. 149:195.[Medline]
  52. Powrie, F., R. Correaoliveira, S. Mauze, R. L. Coffman. 1994. Regulatory interactions between CD45RB(high) and CD45RB(low) CD4(+) T cells are important for the balance between protective and pathogenic cell-mediated immunity. J. Exp. Med. 179:589.[Abstract/Free Full Text]
  53. Powrie, F., J. Carlino, M. W. Leach, S. Mauze, R. L. Coffman. 1996. A critical role for transforming growth factor-ß but not interleukin 4 in the suppression of T helper type 1-mediated colitis by CD45RB(low) CD4(+) T cells. J. Exp. Med. 183:2669.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 StrokeHome page
M. Knoflach, S. Kiechl, D. Penz, A. Zangerle, C. Schmidauer, A. Rossmann, M. Shingh, R. Spallek, A. Griesmacher, D. Bernhard, et al.
Cardiovascular Risk Factors and Atherosclerosis in Young Women: Atherosclerosis Risk Factors in Female Youngsters (ARFY Study)
Stroke, April 1, 2009; 40(4): 1063 - 1069.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
J. E. Fildes, S. M. Shaw, N. Yonan, and S. G. Williams
The Immune System and Chronic Heart Failure: Is the Heart in Control?
J. Am. Coll. Cardiol., March 24, 2009; 53(12): 1013 - 1020.
[Abstract] [Full Text] [PDF]

Home page
 Int ImmunolHome page
M. S. Lillicrap, R. C. Duggleby, J. C. Goodall, and J. S. H. Gaston
T cell recognition of a highly conserved epitope in heat shock protein 60: self-tolerance maintained by TCR distinguishing between asparagine and aspartic acid
Int. Immunol., March 1, 2004; 16(3): 405 - 414.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
L. M. Biasucci, G. Liuzzo, A. Ciervo, A. Petrucca, M. Piro, D. J. Angiolillo, F. Crea, A. Cassone, and A. Maseri
Antibody Response to Chlamydial Heat Shock Protein 60 Is Strongly Associated With Acute Coronary Syndromes
Circulation, June 24, 2003; 107(24): 3015 - 3017.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
S. A. Johnson, S. J. Rozzo, and J. C. Cambier
Aging-Dependent Exclusion of Antigen-Inexperienced Cells from the Peripheral B Cell Repertoire
J. Immunol., May 15, 2002; 168(10): 5014 - 5023.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
A. G. Pockley
Heat Shock Proteins, Inflammation, and Cardiovascular Disease
Circulation, February 26, 2002; 105(8): 1012 - 1017.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ramage, J. M.
Right arrow Articles by Hill Gaston, J. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ramage, J. M.
Right arrow Articles by Hill Gaston, J. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS