HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Smith, P. A. Articles by Potter, T. A. Search for Related Content PubMed PubMed Citation Articles by Smith, P. A. Articles by Potter, T. A.

Alloreactive T Cells That Do Not Require TCR and CD8 Coengagement Are Present in Naive Mice and Contribute to Graft Rejection¹

Pamela A. Smith^*, and Terry A. Potter^2,*,

^* Division of Basic Immunology, Department of Medicine, National Jewish Medical and Research Center, Denver, CO 80206; and the Department of Immunology and the Cancer Center, University of Colorado Health Sciences Center, Denver, CO 80262

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Class I alloreactive CTL populations have been defined as either CD8 dependent or CD8 independent, based upon their ability to kill target cells in the presence of Ab to CD8. The CD8-dependent population uses CD8 in a coreceptor role with the TCR, and mutations in the class I molecule that destroy the CD8 binding site abrogate CTL killing, even if the target cell expresses other allelic forms of class I molecules with an intact binding site for CD8. The CD8-independent population apparently does not require CD8, as Ab to CD8 has no effect on the ability of these cells to kill appropriate target cells. We have isolated a third population of CTL that is inhibited by the addition of CD8 Ab yet can kill target cells that express the alloantigenic molecule incapable of binding CD8, provided that the target cells also express non antigenic class I molecules that contain an intact binding site for CD8. We refer to these cells as CD8 bystander-dependent CTL. Many (10 of 12) of these CTL were able to kill H-2K^b-expressing transfectants of T2 cells, consistent with the idea that they recognize a peptide-independent determinant that may be expressed at a high density on the cell surface. These CD8 bystander-dependent CTL are only readily detectable in vitro when spleen cells from mice primed in vivo with a skin graft are used.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
In general, T cells that recognize antigenic peptides in association with MHC class I molecules and perform cytolytic functions express the surface marker CD8. Similarly, T cells that recognize antigenic peptides in association with MHC class II molecules and perform functions such as the secretion of cytokines express an analogous marker, CD4. The term coreceptor was invoked for CD4 and CD8 to imply that for T cell activation, these molecules must bind to the same MHC molecule as the TCR. Under most physiologic conditions CD4 and CD8 can, and do, act as coreceptors. Perhaps some of the strongest evidence that the activation of most T cells requires CD8 to behave in a coreceptor role is provided by the finding that for recognition by most class I restricted CTL, there is an absolute dependence on the interaction of CD8 with the same MHC molecule as that engaged by the TCR (1, 2, 3, 4, 5). For example, target cells that express H-2K^b with a mutation in the 3 domain that abrogates CD8 engagement are not lysed by primary anti-H-2K^b CTL, even though these cells also express H-2^d class I molecules that can engage CD8 (3). In addition, these mutant class I molecules are unable to stimulate a primary alloresponse (6). This requirement for coreceptor binding is also apparent during thymic selection, as destruction of the CD8 binding site in the 3 domain grossly perturbs selection events to the mutated molecule even though class I molecules of other allelic forms that can bind CD8 are present (7, 8).

The term accessory molecule has been used to describe the activities of CD4 and CD8 when they are unable to bind the same MHC molecule as the TCR. There are conflicting data as to whether the binding of MHC molecules in an accessory manner contributes to T cell activation. While some studies demonstrated that the effects mediated by CD4 or CD8 as accessory molecules do not contribute to T cell activation (9, 10, 11, 12), several T cell hybridomas have been derived in which an accessory role for CD4 or CD8 enhances responsiveness (13, 14, 15, 16). In addition, other studies have shown that following Ab-mediated ligation of the TCR, the binding of CD8 to nonantigenic, or bystander, class I molecules leads to T cell activation, as measured by serine esterase release and hydrolysis or phosphatidylinositides (9, 10, 11). What is unclear in this system is whether the Ab-mediated ligation of the TCR has effects equivalent to those of the physiologic signaling through TCR only or, because of the high degree of TCR occupancy, the signaling events initiated are similar to those that occur upon TCR and CD8 coengagement. In a previous study we found that in CD8-dependent CTL, TCR and CD8 coengagement is required for phosphatidylinositide hydrolysis, whereas in CD8-independent CTL, engagement of the TCR is sufficient for killing, but not for phosphatidylinositol hydrolysis (12). Furthermore, in these CD8 independent CTL, which seemingly do not require CD8 to stabilize the binding of the TCR to the MHC molecule, CD8 and TCR engagement by distinct MHC class I molecules is capable of activating phosphatidylinositide hydrolysis (12). In this report we have further examined the contribution of CD8 to T cell activation under conditions where it is unable to function as a coreceptor.

Most of the previous studies on the coreceptor vs accessory molecule function of CD8 and CD4 have focused on their roles in isolated effector cells. There are, however, a number of investigations that demonstrate differences in the requirements for stimulation of naive or activated T cells. We had noted previously that the in vitro allo-response of naive T cells to cells expressing the 3 domain mutant molecules was very weak (6), implying a requirement for CD8 coreceptor function for the activation of naive T cells. We therefore have investigated the fate of a skin graft in which the only incompatibility is the expression on the donor graft of a mutant class I molecule incapable of binding CD8. We describe herein that such grafts are rapidly rejected. One of the components of this alloresponse is CD8⁺ class I reactive CTL, which, for cytolysis and secretion of IFN-, require CD8 engagement, although not necessarily coengagement with the same MHC class I molecule as the TCR. This population of alloreactive CTL, which we term CD8 bystander dependent, can be distinguished from the classically described CD8-dependent and CD8-independent CTL populations. These CTL represent a distinct subset of CD8-dependent CTL, as they require the binding of CD8; however, unlike most class I-restricted T cells, it is sufficient for CD8 to function as an accessory molecule and not necessarily as a coreceptor. Furthermore, this CTL population is not unique to stimulation with the mutant class I molecule and can also be isolated following stimulation with the wild-type molecule.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice

B10.D2, BALB/c, C57BL/6, and SJL mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and were bred to produce (BALB/c x B10.D2) F₁ (referred to herein as CD2F₁) and (B6 x SJL) F₁ progeny. 28.1 mice are transgenic for the H-2K^b gene on a B10.D2 background. 29.7 mice are transgenic for the mutant (non-CD8 binding) H-2K^b gene on a B10.D2 background (8). B6D8 mice are transgenic for the wild-type H-2D^d gene on a C57BL/6 background (13). 29.2 mice are transgenic for the mutant H-2D^d gene (non-CD8 binding) on a C57BL/6 background.

Target cell lines

M12.C3 (H-2^d) (14), the peptide-processing defective cell line, T2 (15, 16, 17, 18, 19), and the nonmutant parental cells from which T2 were derived, T1 cells (15), expressing the wild-type (H-2K^b.wt) or non-CD8 binding 3 domain mutant (H-2K^b.m) H-2K^b class I molecules, were generated by transfection. These cells are referred to herein as M12-, T2-, or T1-K^b wild-type (or K^b.wt) or as M12-, T2-, or T1-K^b mutant (or K^b.m). RMA/S (20, 21, 22) cells are a peptide-processing defective line of the H-2^b genotype.

Generation of T cell clones

All T cell clones described were isolated by limiting dilution from the spleens and lymph nodes of CD2F₁ mice that had been engrafted with trunk skin from a 29.7 strain transgenic mouse. In addition, several T cell lines were isolated by limiting dilution from the spleens and lymph nodes of CD2F₁ mice that had been engrafted with trunk skin from a 28.1 strain transgenic mouse. Lymph node and spleen cells were harvested on day 14 following engraftment and stimulated in vitro with irradiated 29.7 (clones) or 28.1 (lines) spleen cells. The clones were also restimulated weekly with irradiated 29.7 stimulator spleen cells. Responder and stimulator cells were cocultured in DMEM/FCS (10%) containing Con A-stimulated rat spleen supernatant (T cell medium). All clones characterized in this study were confirmed, by flow microfluorometric analysis, to be CD8⁺ /ß T cells (data not shown). The alloreactive H-2K^b CTL clone 4.1 was generated by limiting dilution cloning from a primary MLR of B10.D2 stimulated with C57BL/6 spleen cells.

Assay of cytotoxic reactivity

Cytotoxic activity was measured in a standard ⁵¹Cr release assay. Target cells were labeled with ⁵¹Cr (100 µCi/10⁶ cells) for 1.5 h at 37°C. Labeled target cells (5000/well) were added to each well of a V-bottom microtiter plate and incubated with effectors at the indicated E:T ratios at 37°C for 4 to 6 h. The percent specific release was calculated using the following formula [(E - S/(M - S)] x 100, where E is the average experimental release of duplicate samples, S is the average spontaneous release of triplicate samples, and M is the average maximum release of triplicate samples. Spontaneous release was measured in the absence of effector cells, and maximum release was measured in the presence of 1% Nonidet P-40 instead of effector cells.

Preparation of peptide extract

Cell suspensions prepared from the spleens of C57BL/6 mice were washed extensively and then homogenized in 0.1% trifluoroacetic acid. The suspension was adjusted to pH 2.0 with HCl, sonicated 20 times for 1 s each time, and then incubated at 4°C for 60 min. Cellular debris was removed by centrifugation at 2000 x g for 20 min, and the supernatant was passed through a Centricon 10 membrane, lyophilized, and resuspended in sterile distilled water.

Precursor frequency analysis

Spleen cells from CD2F₁ mice were cultured at titrated concentrations, each concentration repeated 24 times, in V-bottom 96-well plates. The CD2F₁ mice were either unprimed or primed with an H-2K^b-incompatible skin graft. The engrafted mice were sacrificed 2 wk following engraftment, a period sufficient in all cases for complete graft rejection. CD2F₁ cells were stimulated with 10⁶ irradiated spleen cells/well and, in addition, 10⁵ irradiated spleen cells from the responder. All cultures included 10% rat spleen cell Con A supernatant as a source of IL-2. Following 1 wk of stimulation, the cultures were assayed for cytolysis of ⁵¹Cr-labeled target cells during a 4-h incubation. The precursor frequency was calculated by determining the cell concentration at which 37% of the wells were negative, using a graph of the log percent negative wells on the ordinate and the responder cell concentration on the abscissa. Determination of the cutoff for positive wells was made by culturing 24 wells of responder cells at the highest concentration used in a particular assay with no stimulators, and then calculating the average CTL value readout plus 3 SDs.

Assay for IFN- secretion

The supernatant test samples used in this assay were prepared by incubating 10⁶ T cells (7 days after restimulation) with 10⁵ stimulators in a total of 200 µl of DMEM/FCS in a flat-bottom 96-well plate at 37°C overnight. The T cells and stimulators were washed thoroughly before coculture to remove any IFN- that may have been in the medium in which they were grown. The IFN- standards were prepared with recombinant mouse IFN- (Genzyme, Cambridge, MA) diluted in 1% BSA/PBS buffer. Starting with 200 U/well, the standard wells were prepared from serial doubling dilutions, which resulted in wells that ranged from 200 U of IFN-/well to 1.56 U of IFN-/well.

Supernatant IFN- levels were quantitated using an ELISA with two mAbs to mouse IFN-. Immulon-3 plates were coated with the capture Ab, XMG 1.2, at 100 ng/well, and after incubation with the samples, a biotin-conjugated Ab, R4.6A2, was added at 100 ng/well. The plates were developed colorimetrically with the ABTS Microwell Peroxidase Substrate System (Kirkegaard & Perry Laboratories, Gaithersburg, MD), and they were read on an ELISA reader at 415 nM. Units of IFN- were calculated from an equation derived from the linear portion of a curve generated by graphing the IFN- unit standards against the absorbance ELISA readings.

Skin grafting analysis

CD2F₁ mice were engrafted with trunk skin from either 28.1 (K^b.wt) or 29.7 (K^b.m) mice. The grafts were monitored daily, and rejection was determined by complete graft ablation. Depletion of T cells in vivo was accomplished by injection of Abs 2 wk following thymectomy. Mice were thymectomized at 3 to 4 wk of age and 2 wk later were injected i.p. three times (at 3-day intervals) with 200 µl of ascites fluid containing either CD4 (GK 1.5) or CD8 (YTS 169) Ab. The effectiveness of the Ab-mediated depletion was monitored with biotin-conjugated Abs to CD4 or CD8 that reacted with an epitope distinct from the depleting Ab (data not shown).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Skin grafts that express a disparate class I molecule that does not engage CD8 are rejected as rapidly as grafts that express wild-type class I

To examine the requirement for TCR and CD8 coengagement in skin graft rejection, CD2F₁ (H-2^d) mice were engrafted with H-2K^b.wt- or H-2K^b.m-incompatible skin. Skin grafts expressing either form of H-2K^b are rejected with similar kinetics (Table I). Rejection is T cell mediated, because depletion of both CD4⁺ and CD8⁺ in thymectomized recipients results in indefinite (120⁺ days) acceptance of the graft. The 28.1 transgenic mouse was initially derived in (C57BL/6 x BALB/c) F₂ fertilized eggs and, despite repeated backcrossing, could theoretically still retain some minor histocompatibility genes derived from BALB/c. By using CD2F₁ mice as recipients we have avoided the possibility that the rejection is directed at any BALB/c histocompatibility genes retained by the 28.1 mouse. One possibility for the vigorous response against the mutant H-2K^b molecule, which is not observed in vitro, is that the H-2K^b.m molecule is the source of a unique peptide not present in the H-2K^b.wt molecule. When presented by an H-2^d-encoded molecule, this theoretical peptide could constitute the alloreactive component upon which the rejection response is based. To address this possibility, several 28.1 (B10.D2 (H-2^d) H-2K^b.wt transgenic) mice were grafted with skin from 29.7 (B10.D2 H-2K^b.m) mice. The grafts showed no sign of rejection for 120+ days, indicating that there is no peptide derived from H-2K^b.m that is presented by H-2^d to provide an allodeterminant. To confirm that rapid skin graft rejection across a single MHC class I incompatibility in the absence of CD8/MHC coengagement is not unique to H-2K^b, we performed similar grafts in which the only disparity was the expression of H-2D^d mutant class I molecules incapable of binding CD8. These grafts were rejected as rapidly as those in which the disparity was a wild-type H-2D^d molecule (Table I). As was noted with the H-2K^b transgenic mice, using (B6 x SJL)F₁ avoided the potential problem of there being any remaining SJL minor histocompatibility genes in the B6.D^d.m transgenic line (which was derived in (C57BL/6 x SJL)F₂ eggs) contributing to rejection. We also grafted B6.D8 (H-2D^d.wt) transgenic mice with skin from 29.2 (H-2D^d.m) mice to test the possibility that the H-2D^d.m molecule might be the source of a unique epitope that could allow graft recognition and rejection. The grafts survived indefinitely, suggesting that the mutant H-2D^d molecule does not provide a unique peptide for presentation by H-2^b-encoded MHC molecules.

CD8 bystander-dependent T cells can kill cells that express other nonantigenic class I molecules in addition to the mutated H-2K^b, but they cannot kill cells that express only the H-2K^b class I molecule with the 3 domain mutation

H-2K^b alloreactive CTL clones were isolated from CD2F₁ mice that had been primed in vivo with an H-2K^b.m skin graft. Following rejection of the graft, CTL clones were isolated from the spleen and lymph nodes by in vitro restimulation with spleen cells from mice expressing the H-2K^b.m transgene. These CTL clones were assayed for their ability to kill M12.C3 (H-2^d) cells transfected with the H-2K^b.m or the H-2K^b.wt gene (Fig. 1). The M12.K^b.m cells are genotypically similar to the spleen cells used to elicit the CTL; thus, the killing of M12.K^b.m cells is consistent with the fact that these CTL clones were isolated by stimulation with cells that expressed the H-2K^b.m molecule. The ability to lyse M12.K^b.m cells suggests that in these CTL, CD8 does not have to engage the same MHC class I molecule as the TCR. We also examined whether these CTL would kill T2 cells expressing the wild-type or mutant H-2K^b molecule. T2 cells are unable to process Ag for loading into the MHC class I molecule; thus, they express very low levels of the human HLA molecules (23). Therefore, T2 cells transfected with murine class I MHC genes, such as H-2K^b.wt or H-2K^b.m, express these molecules either without peptide, or in association with a limited array of peptides (24). Furthermore, the human class I molecules that are expressed do not significantly engage murine CD8 (25, 26, 27); consequently, class I molecules expressed by the transfected gene are the only possible target for the binding of murine CD8. We have previously described several other CTL clones that were determined to be CD8 independent, as they were able to lyse T2.K^b.m cells and were unaffected by the addition of CD8 Ab (28). In contrast to the previously described CTL clones, which could kill both T2.K^b.wt and T2.K^b.m cells, 10 of the 12 clones represented in Figure 1 were able to kill T2.K^b.wt cells, but none were able to kill T2.K^b.m cells. These 12 CTL clones that were able to recognize M12.K^b.m, but not T2.K^b.m, cells are referred to herein as CD8 bystander-dependent T cells. Of these 12 clones, the 10 that were also able to recognize T2.K^b.wt cells are considered to be peptide independent, whereas the two clones that were unable to kill T2.K^b.wt cells were considered to be peptide dependent. The CTL that we had previously characterized as able to kill T2.K^b cells were shown to be peptide independent based on several criteria, including the ability to kill T2.K^b cells treated with acid to remove any MHC-bound peptide and the ability to respond to plate-bound H-2K^b produced by transfected Drosophila melanogaster cells (28). In the current study we did not perform such exhaustive analysis to establish the peptide-independent nature of the recognition by the 10 clones that killed T2.K^b.wt; thus, we cannot exclude the possibility that these CTL are, in fact, specific for one of the limited array of peptides expressed in association with H-2K^b on T2.K^b cells. Cells that express the 3 domain mutation are also killed by CD8-independent CTL (28); however, in contrast to CD8-independent CTL, CD8 bystander-dependent CTL are inhibited by Ab to CD8 (see below). This dependence on CD8 for killing of M12.K^b.m cells, in which the H-2K^b molecule is incapable of initial engagement of CD8, strongly suggests that the CD8 molecule is binding to the H-2^d-encoded class I molecules.

View larger version (14K): [in this window] [in a new window]   FIGURE 1. Cytolysis of M12.Kb.wt, M12.Kb.m, T2.Kb.wt, and T2.Kb.m target cells by 12 clones isolated from a CD2F1 mouse following rejection of a skin graft expressing H-2Kb.m. Lysis at a 10:1 E:T cell ratio for an individual clone is represented by a symbol. These T cell clones cannot kill target cells that express only the H-2Kb class I molecule with the 3 domain mutation (i.e., T2.Kb.m), but can kill cells that express other nonantigenic class I molecules in addition to the mutated H-2Kb (i.e., M12.Kb.m). Also shown is that two of these clones (no. 5 and 10) do not kill T2.Kb.wt cells.

View larger version (15K): [in this window] [in a new window]   FIGURE 2. Cytolysis of T2, T2.Kb.wt, and T2.Kb.m target cells by CD8 bystander-dependent CTL clones 5 (denoted by squares) and 10 (circles). These CTL assays were conducted in the absence (top) or the presence (bottom) of an extract from C57BL/6 spleen cells that contains MHC binding peptides. These clones are cytolytic for T2.Kb.wt cells only when pulsed with the extract from C57BL/6 spleen cells, suggesting that they are peptide dependent. These clones cannot, however, kill T2.Kb.m cells when pulsed with the C57BL/6 peptide extract, indicating that they require CD8 engagement.

View larger version (12K): [in this window] [in a new window]   FIGURE 3. Cytolysis by the CD8 bystander-dependent CTL clones (open squares indicate clone 15; closed circles indicate clone 5) and the CD8-dependent CTL clone 4.1 (closed triangles) of M12.C3 (H-2d), M12.Kb.wt, and M12.Kb.m target cells. The other two bystander-dependent clones, 10 and 28, were indistinguishable from clones 15 and 5 and are therefore not included in the figure. Failure to kill M12.C3 cells demonstrates specificity for H-2Kb.

View larger version (15K): [in this window] [in a new window]   FIGURE 4. Cytolysis of M12.Kb.wt (top panels), M12.Kb.m (middle panels), and T2.Kb.wt target cells by the two CD8 bystander-dependent CTL clones, 15 and 28, in the absence (closed circles) or the presence (empty circles) of the CD8 Ab, YTS.169. Addition of the Ab inhibited the killing of M12.Kb.wt, M12.Kb.m, and T2.Kb cells by both these clones, confirming that the CD8 bystander-dependent CTL phenotype requires engagement of CD8 for activation of lysis.

T cells that are CD8 bystander dependent for lysis are also CD8 bystander dependent for IFN- release

We assayed IFN- secretion by the CD8 bystander-dependent CTL clones upon incubation with M12.K^b and T1.K^b cells. T1.K^b cells were used instead of T2.K^b cells so that the IFN- responses of both the peptide-dependent and the peptide-independent clones could be determined. The clones secreted IFN- upon incubation with M12.K^b.m cells, but did not produce IFN- upon incubation with T1.K^b.m cells (Fig. 5). The difference in the response to these two cell lines suggests that the clones require CD8 engagement for IFN- secretion, and furthermore, that CD8 engagement of nonantigenic (i.e., a different allelic form that engaged by the TCR) MHC class I molecules is sufficient for IFN- release. For comparative analysis, the CTL clone, 4.1, that is CD8 coreceptor dependent for killing and for IFN- release is shown. This clone could be stimulated for IFN- release by M12.K^b.wt or T1.K^b.wt, but not by M12.K^b.m or T1.K^b.m cells.

View larger version (10K): [in this window] [in a new window]   FIGURE 5. IFN- secretion by two CD8 bystander-dependent CTL clones, 10 and 15, as well as the CD8 coreceptor-dependent clone 4.1 in response to M12.C3, M12.Kb, T1, and T1.Kb cells. All clones are stimulated by M12.Kb.wt and T1.Kb cells. Neither of the CD8 bystander-dependent clones secreted IFN- in response to incubation with T1.Kb.m cells, which did not express any murine class I molecules that engage CD8, indicating that these CTLs require CD8 engagement for IFN- secretion. They did secrete IFN- upon incubation with M12.Kb.m cells, indicating that CD8 engagement of nonantigenic MHC class I molecules (H-2d) is sufficient for IFN- release. The response is Ag specific because they do not secrete IFN- in response to M12.C3 cells, which express the H-2d haplotype only.

The precursor frequency of CTL reactive with M12.K^b.wt and M12.K^b.m cells was determined in naive mice and in mice primed in vivo with H-2K^b.wt or H-2K^b.m incompatible skin grafts (Table II). Primary in vitro stimulation with spleen cells that express H-2K^b.wt elicited coreceptor-dependent, H-2K^b-specific alloreactive T cells that killed M12.K^b.wt, but not M12.K^b.m, cells. In contrast, primary in vitro stimulation with the H-2K^b.m molecule produced very few H-2K^b alloreactive T cells that killed either M12.K^b.wt or M12.K^b.m cells, suggesting that the mutant H-2K^b molecule is a very poor primary allo-stimulus. Following in vivo priming with a skin graft expressing H-2K^b.m, the frequency of H-2K^b alloreactive T cells was increased to a level comparable to that obtained when the skin grafts expressed the wild-type form of H-2K^b (Table II). Thus, the mutated form of H-2K^b, which does not engage CD8, is a potent in vivo priming stimulus for eliciting H-2K^b-specific alloreactive CTL. The CTL generated following in vivo priming with either the wild-type or mutant H-2K^b molecule includes CTL that can kill M12.K^b.m cells, and thus do not require TCR/CD8 coengagement. The stimulation with H-2K^b.m is a stronger in vitro stimulus than H-2K^b.wt for CTL that can kill M12.K^b.m cells. This is apparent from the pCTL frequency of mice that were grafted with wild-type H-2K^b and restimulated with either H-2K^b.wt (1/216,423) or H-2K^b.m (1/6,761). Furthermore, in vitro stimulation of spleen cells from skin graft-primed mice with equal amounts of spleen cells expressing either H-2K^b.wt or H-2K^b.m elicited a high frequency of CTL (1/10,829), which can kill M12.K^b.m cells. This frequency is similar to the pCTL frequency obtained when only H-2K^b.m-expressing cells were used for the in vitro stimulation..

The precursor frequency of CTL that can kill M12.K^b.m cells is about 10-fold lower if the skin graft and the spleen cells used for the in vitro stimulation both express the H-2K^b.wt, rather than the H-2K^b.m, class I molecule. This difference may reflect a selection bias toward T cells that do not require CD8/TCR coreceptor engagement when the mutant H-2K^b molecule is used for stimulation. In addition it may also reflect a selection by the wild-type H-2K^b for coreceptor-dependent CTL. Therefore, we further addressed the question of whether the CD8 bystander-dependent phenotype exists as a normal component of the alloresponse to the wild-type H-2K^b molecule. We isolated CTL lines from a mouse that had been primed with an H-2K^b.wt-incompatible skin graft and restimulated in vitro with spleen cells that expressed H-2K^b.wt. The CTL lines selected were CD8 dependent, as defined by their ability to kill T2.K^b.wt, but not T2.K^b.m, cells. Three of these 11 CTL lines could kill M12.K^b.wt, but not M12.K^b.m, target cells; five lines killed M12.K^b.m cells less efficiently than M12.K^b.wt cells; and three lines (shown in Fig. 6) killed both M12.K^b.wt and M12.K^b.m cells equally well. These data confirm that CD8 bystander-dependent T cells are a component of the secondary alloresponse to a MHC class I disparity.

View larger version (19K): [in this window] [in a new window]   FIGURE 6. Cytolysis of H-2Kb-transfected M12.C3 (top) and T2 (bottom) target cells by three CTL lines isolated from a mouse that was primed in vivo with an H-2Kb.wt-incompatible skin graft. These lines were restimulated in vitro with spleen cells from an H-2Kb.wt transgenic mouse. They are CD8 bystander dependent because they can kill the M12.Kb.m cells, which express nonantigenic class I molecules that engage CD8; however, they are not CD8 independent because they cannot kill the T2.Kb.m cells, which only express class I molecules that do not engage CD8.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We had previously noted that the generation of CTL during a primary in vitro MLR response to stimulator cells expressing the 3 domain mutant class I molecules was either weak or nonexistent. It was therefore surprising that skin grafts in which the only disparity was the mutant class I molecule were rapidly rejected. Furthermore, in vitro restimulation of spleen cells from mice that had rejected such a graft was very effective in generating CTL that were characterized as CD8 bystander dependent. The precursor frequency data also confirmed the development of a strong alloresponse following in vivo priming and in vitro stimulation with the H-2K^b.m molecule. These CTL could also be isolated following priming with a skin graft that expressed H-2K^b.wt, although the pCTL frequency was lower than that obtained following H-2K^b.m stimulation in vivo and in vitro. Previous attempts to raise CD8-independent CTL have also involved in vivo priming (29) or stimulation in vitro in the presence of cross-linking Abs to CD8 (30). Alloreactive CTL that have been previously isolated in the absence of such approaches are dependent on CD8 coreceptor activity. The critical question is why do most class I-restricted T cells require CD8 coreceptor activity, whereas for other T cells CD8 can function as an accessory molecule? In addition, there are T cells that are totally CD8 independent. While many previous studies have assumed that CD8-independent T cells can recognize APCs because their TCR has a high affinity for the target MHC molecule, several studies have shown that, at least for some CD8-dependent T cells, the requirement for CD8 can be overcome by increasing the density of antigenic MHC-peptide complexes on the cell surface (12, 31, 32, 33, 34, 35). Thus, perhaps some CD8-independent T cells recognize a ligand present at high density on the target cell. Alloantigenic determinants that are not dependent on a unique peptide or do not require a peptide at all would be examples of determinants that could be expressed at higher levels than conventional peptide-unique epitopes. We have previously described a number of CD8-independent CTL that are capable of recognizing such determinants and, unlike most conventional CD8 coreceptor-dependent H-2K^b alloreactive CTL, are able to kill T2.K^b cells. Interestingly, most (10 of 12) of the CD8 bystander-dependent CTL were also able to kill T2.K^b. Whether the ability to recognize high density ligands increases the likelihood that a TCR binding to the mutant molecule is in the vicinity of a CD8 molecule that is bound to a nonantigenic class I molecule is an unknown, yet intriguing, possibility. In contrast, a low occupancy of the TCR would mandate that CD8 be bound by the same MHC molecule and thus function as a coreceptor. We are currently further analyzing the nature of the allo-determinants recognized by CD8-independent and CD8 bystander-dependent alloreactive CTL.

A puzzling issue arising from our observations is why the CD8 bystander cell requires in vivo priming, whereas the CD8 dependent T cell does not. Several studies have documented situations in which activation of naive T cells requires greater stimulation than activation of primed effector cells (36). One possibility is that specialized APC, such as Langerhans cells, at the site of the skin graft are ideal for activation of naive cells, particularly in the setting of an inflammatory response. Perhaps such conditions are more potent than primary in vitro stimulation with spleen cells for the activation of naive T cells and thus are able to activate T cells that express TCR that are able to recognize peptide-independent epitopes with low affinity.

The coreceptor role of CD8 is defined as the requirement for CD8 to engage the same MHC class I molecule as the TCR and has been demonstrated in many systems (3, 6, 7, 8, 37). The coreceptor role of CD8 provides two functions in T cell activation: 1) the binding of CD8 to class I activates the kinase p56^lck (38, 39), which, in association with TCR-mediated signal transduction events, augments signaling pathways; and 2) CD8 adds to the affinity of the TCR/MHC interaction. In this model, the CD8-mediated transduction events only occur when CD8 engages an MHC class I molecule in conjunction with the TCR. CD8-mediated signal transduction events are required for serine esterase release, hydrolysis of phosphatidylinositides, and an increase in intracellular Ca²⁺ (9, 10, 11) in a system in which CD8 was bound to immobilized, nonantigenic class I molecules subsequent to Ab-induced TCR ligation. Furthermore, we have found that coengagement of CD8 and the TCR is essential for the hydrolysis of phosphatidylinositides (PI)³ in CD8-dependent CTL (12). Moreover, in CD8-independent CTL, although CD8 engagement is not required for cytolysis, it is required for PI hydrolysis. Interestingly, in these CD8-independent CTL, TCR and CD8 coengagement was not required for PI hydrolysis providing that the target cell also expressed a nonantigenic class I molecule to which CD8 could bind (12). Thus, these CD8-independent CTL were CD8 bystander dependent for PI hydrolysis. Furthermore, even though the TCR expressed by these CD8-independent CTL provided sufficient affinity/avidity for killing, the signaling pathway leading to phosphatidylinositol bisphosphate hydrolysis was absolutely dependent on CD8 engagement. Therefore, T cell functions dependent upon this signaling pathway require CD8 activation regardless of the affinity of the TCR for its ligand.

Whether the primary role of CD8 is to increase the affinity of the TCR/MHC interaction is controversial. A recent study directly demonstrated that CD8 effectively increases the duration and extent of TCR engagement by stabilizing the TCR/MHC complex once it is formed (40). Other experiments using CD8 blocking Abs indicate that the relationship between the individual TCR and CD8 molecules is more complex than CD8 simply adding affinity to this interaction (41, 42). For example, it has been demonstrated that binding of the 2C CTL clone to H-2L^d in association with the allopeptide is not affected by Ab to CD8, even though the Ab is sufficient to completely inhibit cytolysis (42). Similarly, the binding of soluble MHC class II to the same TCR expressed on CD4⁺ and CD4^- T cells was indistinguishable (43). Therefore, despite the fact the CD8 has been shown to stabilize the TCR/MHC interaction, this scenario alone does not adequately account for much of the data describing CD8 activity. Furthermore, the observation that bystander CD8 engagement is sufficient for cytolysis or IFN- release indicates that the contribution of CD8 is not confined to increasing the affinity of the TCR for the MHC class I molecule. Apparently, CD8 bystander CTL do not require the increase in TCR affinity provided by CD8 coengagement of the same MHC class I molecule; whether this is due to a high affinity TCR or, alternatively, a high extent of TCR occupancy is unclear.

CD8 can function as an accessory molecule upon binding to nonantigenic MHC class I molecules distinct from those engaged by the TCR. There are several situations in which an accessory function of CD8 or CD4 molecules has been shown to enhance T cell reactivity (44, 45, 46, 47). Most of these studies, however, were performed in hybridomas, and it is unclear whether the T cell from which the hybridoma was derived required CD8 coreceptor activity. Furthermore, in one of the studies, the accessory molecule effect for CD4 was only apparent when variants of the hybridoma that expressed very low levels of TCR were analyzed (44). As an accessory molecule, CD8 could 1) increase the overall avidity of the T cell-target cell interaction, or 2) transduce signals through p56^lck that either act independently of or intersect downstream with TCR-mediated signal transduction. Currently, there is substantial evidence to indicate that following initial TCR/CD8 coengagement, binding of CD8 to other MHC class I molecules increases the overall avidity of the T cell-target cell interaction. This evidence includes studies that demonstrate that cross-linking the TCR with soluble mAb induces the binding of alloreactive CTL to immobilized class I molecules of any MHC allelic form (9, 10, 11, 48). This binding of T cells to the immobilized MHC class I is inhibited by Ab to CD8. According to a model proposed to explain these observations, an initial activation stimulus through the TCR converts CD8 to a high affinity state (9). In these studies, Ab to the TCR or CD3 was used as the activating stimulus. In a physiologic setting, however, activation of most CTL requires the coengagement of TCR and CD8 by the same MHC class I molecule (6). Recently, the two events, 1) coengagement of CD8 and the TCR, and 2) binding of CD8 to other MHC class I molecules, have been clearly separated. In this scheme (49) the Glu²²⁷lysine mutation in the MHC class I 3 domain inhibits the initial binding by CD8 and, hence, TCR and CD8 coengagement. This mutation has no effect on the CD8/MHC interaction subsequent to the activation through the TCR and CD8 coengagement event that converts CD8 to the high affinity state. It is unclear whether the high affinity CD8 overcomes the effect of the mutation by binding to the 222–229 loop or at the other regions of the class I molecule that contact CD8 (50, 51, 52, 53). Thus, in T cells activated with Abs to the TCR, CD8 can bind to nonantigenic class I molecules even if they contain the mutation at residue 227 (49). Therefore, CD8 can mediate an overall increase in T cell/target cell avidity through a generalized class I binding following the activation of CD8 when CD8 is coengaged with the TCR. This coengagement of the TCR and CD8, however, is dependent on the binding site in the 3 domain.

	Acknowledgments

 
We thank Ms. Heather Zwickey for a critical review of the manuscript.

	Footnotes

 
¹ This work was supported by Grant A128115 from the National Institutes of Health.

² Address correspondence and reprint requests to Dr. Terry Potter, National Jewish Medical and Research Center, 1400 Jackson St., Denver, CO 80206-2761.

³ Abbreviations used in this paper: PI, phosphatidylinositides.

Received for publication November 21, 1997. Accepted for publication February 5, 1998.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Potter, T. A., J. A. Bluestone, T. V. Rajan. 1987. A single amino acid substitution in the 3 domain of an H-2 class I molecule abrogates reactivity with CTL. J. Exp. Med. 166:956.[Abstract/Free Full Text]
2. Connolly, J. M., T. A. Potter, E.-M. Wormstall, T. H. Hansen. 1988. The Lyt-2 molecule recognizes residues in the class I 3 domain in allogeneic cytotoxic T cell responses. J. Exp. Med. 168:325.[Abstract/Free Full Text]
3. Potter, T. A., R. F. Rajan, II R. F. Dick, J. A. Bluestone. 1989. Substitution at residue 227 of H-2 class I molecules abrogates recognition by CD8-dependent, cytotoxic T lymphocytes. Nature 337:73.[Medline]
4. Salter, R. D., R. J. Benjamin, P. K. Weseley, S. E. Buxton, T. P. J. Garrett, C. Clayberger, A. M. Krensky, A. M. Normant, D. R. Littman, P. Parham. 1990. A binding site for the T-cell coreceptor CD8 on the 3 domain of HLA-A2. Nature 345:41.[Medline]
5. Sherman, L. A., S. V. Hesse, M. J. Irwin, D. La Face, P. Peterson. 1992. Selecting T cell receptors with high affinity for self-MHC by decreasing the contribution of CD8. Science 258:815.[Abstract/Free Full Text]
6. Connolly, J. M., T. H. Hansen, A. L. Ingold, T. A. Potter. 1990. Recognition by CD8 on cytotoxic T lymphocytes is ablated by several substitutions in the class I3 domain: CD8 and the T-cell receptor recognize the same class I molecule. Proc. Natl. Acad. Sci. USA 87:2137.[Abstract/Free Full Text]
7. Killeen, N., A. Moriarty, H.-S. Teh, D. R. Littman. 1992. Requirement for CD8-major histocompatibility complex class I interaction in positive and negative selection of developing T cells. J. Exp. Med. 176:89.[Abstract/Free Full Text]
8. Ingold, A. L., C. Landel, C. Knall, G. A. Evans, T. A. Potter. 1991. Co-engagement of CD8 with the T cell receptor is required for negative selection. Nature 352:721.[Medline]
9. O’Rourke, A. M., J. Rogers, M. F. Mescher. 1990. Activated CD8 binding to class I protein mediated by the T-cell receptor results in signalling. Nature 346:187.[Medline]
10. O’Rourke, A. M., M. F. Mescher. 1992. Cytotoxic T-lymphocyte activation involves a cascade of signalling and adhesion events. Nature 358:253.[Medline]
11. O’Rourke, A. M., M. F. Mescher. 1994. Signals for activation of CD8-dependent adhesion and costimulation in CTLs. J. Immunol. 152:4358.[Abstract]
12. Knall, C., P. A. Smith, T. A. Potter. 1995. CD8-dependent CTL require co-engagement of CD8 and the TCR for phosphatidylinositol hydrolysis, but CD8-independent CTL do not and can kill in the absence of phosphatidylinositol hydrolysis. Int. Immunol. 7:995.[Abstract/Free Full Text]
13. Bieberich, C., G. Scangos, K. Tanaka, G. Jay. 1986. Regulated expression of a murine class I gene in transgenic mice. Mol. Cell. Biol. 6:1339.[Abstract/Free Full Text]
14. Glimcher, L. H., D. J. McKean, E. Choi, J. G. Seidman. 1985. Complex regulation of class II gene expression: analysis with class II mutant cell lines. J. Immunol. 135:3542.[Abstract]
15. Henderson, R. A., M. Hanspeter, K. Sakaguchi, J. Shabanowitz, E. Appella, D. F. Hunt, V. H. Engelhard. 1992. HLA-A2.1-associated peptides from a mutant cell line: a second pathway of antigen presentation. Science 255:1264.[Abstract/Free Full Text]
16. Wei, M. L., P. Cresswell. 1992. HLA-A2 molecules in an antigen-processing mutant cell contain signal sequence-derived peptides. Nature 356:443.[Medline]
17. Hosken, N. A., M. J. Bevan. 1992. An endogenous antigenic peptide bypasses the class I antigen presentation defect in RMA-S. J. Exp. Med. 175:719.[Abstract/Free Full Text]
18. Cerundolo, V., J. Alexander, K. Anderson, C. Lamb, P. Cresswell, A. McMichael, F. Gotch, A. Townsend. 1990. Presentation of viral antigen controlled by a gene in the major histocompatibility complex. Nature 345:449.[Medline]
19. Salter, R. D., D. N. Howell, P. Cresswell. 1985. Genes regulating HLA class I antigen expression in T-B lymphoblast hybrids. Immunogenetics 21:235.[Medline]
20. Ljunggren, H.-G., N. J. Stam, C. Ohlen, J. Neefjes, J. P. Hoglund, M.-T. Heemels, J. Bastin, T. N. M. Schumacher, A. Townsend, K. Karre, H. L. Ploegh. 1990. Empty MHC class I molecules come out in the cold. Nature 346:476.[Medline]
21. Schumacher, T. N. M., M.-T. Heemels, J. J. Neefjes, W. M. Kast, C. J. M. Melief, H. L. Ploegh. 1990. Direct binding of peptide to empty MHC class I molecules on intact cells and in vitro. Cell 62:563.[Medline]
22. Townsend, A., C. Ohlen, J. Bastin, L. Ljunggren, L. Foster, K. Karre. 1989. Association of class I major histocompatibility heavy and light chains induced by viral peptides. Nature 340:443.[Medline]
23. DeMars, R., R. Rudersdorf, C. Chang, J. Petersen, J. Strandtmann, N. Korn, B. Sidwell, H. T. Orr. 1985. Mutations that impair a posttranscriptional step in expression of HLA-A and -B antigens. Proc. Natl. Acad. Sci. USA 82:8183.[Abstract/Free Full Text]
24. Hosken, N. A., M. J. Bevan. 1990. Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 248:367.[Abstract/Free Full Text]
25. Samberg, N. L., E. C. Scarlett, H. J. Stauss. 1989. The alpha 3 domain of major histocompatibility complex class I molecules plays a critical role in cytotoxic T lymphocyte stimulation. Eur. J. Immunol. 19:2349.[Medline]
26. Vitiello, A., D. Marchesini, J. Furze, L. A. Sherman, R. W. Chesnut. 1991. Analysis of the HLA-restricted Influenza-specific cytotoxic T lymphocyte response in transgenic mice carrying a chimeric human-mouse class I major histocompatibility complex. J. Exp. Med. 173:1007.[Abstract/Free Full Text]
27. Kalinke, U., B. Arnold, G. J. Hammerling. 1990. Strong xenogeneic HLA response in transgenic mice after introducing an 3 domain into HLA B27. Nature 348:642.[Medline]
28. Smith, P. A., A. Brunmark, M. R. Jackson, T. A. Potter. 1997. Peptide-independent recognition by alloreactive cytotoxic T lymphocytes (CTL). J. Exp. Med. 185:1023.[Abstract/Free Full Text]
29. MacDonald, R. H., A. L. Glasebrook, C. Bron, A. Kelso, J.-C. Cerottini. 1982. Clonal heterogeneity in the functional requirement for Lyt-2/3 molecules on cytolytic T lymphocytes (CTL): possible implications for the affinity of CTL antigen receptors. Immunol. Rev. 68:89.[Medline]
30. McCarthy, S. A., E. Kaldjian, A. Singer. 1989. Characterization of anti-CD8-resistant cytolytic T lymphocytes induced by multivalent cross-linking of CD8 on precursor cells. J. Immunol. 143:2112.[Abstract]
31. Cai, Z., J. Sprent. 1994. Resting and activated T cells display different requirements for CD8 molecules. J. Exp. Med. 179:2005.[Abstract/Free Full Text]
32. Alexander, M. A., C. A. Damico, K. M. Wieties, T. H. Hansen, J. M. Connolly. 1991. Correlation between CD8 dependency and determinant density using peptide-induced, Ld-restricted cytotoxic T lymphocytes. J. Exp. Med. 173:849.[Abstract/Free Full Text]
33. Shimonkevitz, R., B. Leuscher, J.-C. Cerottini, R. H. MacDonald. 1985. Clonal analysis of cytolytic T lymphocyte-mediated lysis of target cells with inducible antigen expression: correlation between antigen density and requirement for Lyt-2/3 function. J. Immunol. 135:892.[Abstract]
34. Goldstein, S. A., M. F. Mescher. 1987. Cytotoxic T cell activation by class I protein on cell-size artificial membranes: antigen density and Lyt-2/3 function. J. Immunol. 138:2034.[Abstract]
35. Al-Ramadi, B. K., M. T. Jelonek, L. F. Boyd, D. H. Margulies, A. L. M. Bothwell. 1995. Lack of strict correlation of functional sensitization with the apparent affinity of MHC/peptide complexes for the TCR. J. Immunol. 155:662.[Abstract]
36. Tanchot, C., F. A. Lemonnier, B. Perarnau, A. A. Freitas, B. Rocha. 1997. Differential requirements for survival and proliferation of CD8 naive or memory T cells. Science 276:2057.[Abstract/Free Full Text]
37. Salter, R. D., A. M. Norment, B. P. Chen, C. Clayberger, A. M. Krensky, D. R. Littman, P. Parham. 1989. Polymorphism in the 3 domain of HLA-A molecules affects binding to CD⁸. Nature 338:345.[Medline]
38. Barber, E. K., J. D. Dasgupta, S. F. Scholassman, J. M. Trevillyan, C. E. Rudd. 1989. The CD4 and CD8 antigens are coupled to a protein-tyrosine kinase (p56-Lck) that phosphorylates the CD3 complex. Proc. Natl. Acad. Sci. USA 86:3277.[Abstract/Free Full Text]
39. Luo, K., B. M. Sefton. 1990. Cross-linking of T-cell surface molecules CD4 and CD8 stimulates phosphorylation of the lck tyrosine protein kinase at the autophosphorylation site. Mol. Cell. Biol. 10:5305.[Abstract/Free Full Text]
40. Garcia, K. C., C. A. Scott, A. Brunmark, F. R. Carbone, P. A. Peterson, I. A. Wilson, L. Teyton. 1996. CD8 enhances formation of stable T-cell receptor/MHC class I molecule complexes. Nature 384:577.[Medline]
41. Luescher, I. F., E. Viver, A. Layer, J. Mahiou, F. Godeau, B. Malissen, P. Romero. 1995. CD8 modulation of T-cell antigen receptor-ligand interactions on living cytotoxic T lymphocytes. Nature 373:353.[Medline]
42. Sykulev, Y., A. Brunmark, M. Jackson, R. J. Cohen, P. A. Peterson, H. N. Eisen. 1994. Kinetics and affinity of reactions between an antigen-specific T cell receptor and peptide-MHC complexes. Immunity 1:15.[Medline]
43. Matsui, K., J. J. Boniface, P. A. Reay, H. Schild, B. Fazekas De St. Groth, M. M. Davis. 1991. Low affinity interaction of peptide-MHC complexes with T cell receptors. Science 254:1788.[Abstract/Free Full Text]
44. Gay, D., P. Maddon, R. Sekaly, M. A. Talle, M. Godfrey, E. Long, G. Goldstein, L. Chess, R. Axel, J. Kappler, P. Marrack. 1987. Functional interaction between human T-cell protein CD4 and the major histocompatibility complex HLA-DR antigen. Nature 328:626.[Medline]
45. Greenstein, J. L., J. Kappler, P. Marrack, S. J. Burakoff. 1984. The role of L3T4 in recognition of Ia by a cytotoxic, H-2Dd-specific T cell hybridoma. J. Exp. Med. 159:1213.[Abstract/Free Full Text]
46. Miceli, C. M., P. von Hoegen, J. R. Parnes. 1991. Adhesion versus coreceptor function of CD4 and CD8: role of the cytoplasmic tail in coreceptor activity. Proc. Natl. Acad. Sci. USA 88:2623.[Abstract/Free Full Text]
47. Lustgarten, J., T. Waks, Z. Eshhar. 1991. CD4 and CD8 accessory molecules function through interactions with major histocompatibility complex molecules which are not directly associated with the T cell receptor-antigen complex. Eur. J. Immunol. 21:2506.
48. O’Rourke, A. M., B. Ybarrondo, M. F. Mescher. 1993. CD8 and antigen-specific T cell adhesion cascades. Semin. Immunol. 5:263.[Medline]
49. Shen, L., T. A. Potter, K. P. Kane. 1996. Glu²²⁷ to Lys substitution in the acidic loop of major histocompatibility complex class I 3 domain distinguishes low avidity CD8 coreceptor and avidity-enhanced CD8 accessory functions. J. Exp. Med. 184:1671.[Abstract/Free Full Text]
50. Sun, R., S. E. Shepherd, S. S. Geier, C. E. Thompson, J. M. Sheil, S. G. Nathenson. 1995. Evidence that the antigen receptors of cytotoxic T lymphocytes interact with a common recognition pattern of the H-2K^b molecule. Immunity 3:573.[Medline]
51. Newberg, M. H., H. S. D., S. B. Haertel, D. R. Vining, E. Lacy, V. H. Engelhard. 1996. Importance of MHC class I 2 and 3 domains in the recognition of self and non-self MHC molecules. J. Immunol. 156:2473.[Abstract]
52. LaFace, D. M., M. Vestberg, Y. Y. R. Srivastava, J. DiSanto, N. Flomenberg, S. Brown, L. A. Sherman, P. A. Peterson. 1995. Human CD8 transgene regulation of HLA recognition by murine T cells. J. Exp. Med. 182:1315.[Abstract/Free Full Text]
53. Gao, G. F., J. Tormo, U. C. Gerth, J. T. Wyer, A. J. McMichael, D. I. Stuart, J. E. Bell, Y. E. Jones, B. K. Jakobsen. 1997. Crystal structure of the complex between human CD8 and HLA-A2. Nature 387:630.[Medline]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Smith, P. A. Articles by Potter, T. A. Search for Related Content PubMed PubMed Citation Articles by Smith, P. A. Articles by Potter, T. A.