We have shown that alloreactive CD8 T cell activation may proceed via CD4-dependent and CD4-independent pathways, and that CD8 T cell activation in Ag-primed animals is independent of CD154 costimulation. In this report, we further analyzed the activation and function of alloreactive CD8 CTL effectors in CD4 knockout (KO) skin/cardiac allograft recipients. FACS analysis showed that alloreactive CD8 T cells were activated at a significantly reduced level in CD4 KO mice. Importantly, these helpless CD8 T cells failed to develop CD154 blockade resistance following reactivation by the same alloantigen, indicative of defective memory formation. Only transient CD4 help was required, as short-term CD4 blockade at the time of first skin graft challenge only delayed alloreactive CD8 activation, without affecting the CD8 T cell memory response to a second skin graft. Moreover, postoperative CD4 blockade had no effect on alloreactive CD8 activation. Alloreactive CD8 cells generated in the absence of CD4 help exhibited decreased effector responses. Interestingly, intragraft induction of T cell-targeted chemokines early after transplant was also dependent on CD4 help, as the induction kinetics of CXCL9 and CCL5 in CD4 KO recipients was significantly delayed, coupled with similarly delayed infiltration by CD3/CD8 cells. Remarkably, helpless CD8 cells ultimately entering the graft still displayed significantly diminished T cell effector molecules (IFN-γ, granzyme B). Thus, CD4 help is critical for alloreactive CD8 activation, function, and memory formation.

By recognizing allogeneic MHC class I Ags and exerting cytolytic function, primarily via granzyme and perforin mechanisms, alloreactive CD8 T lymphocytes are major effectors in the allograft rejection cascade (1, 2). Despite their key role in transplant immunology, however, the mechanisms of alloreactive CD8 T cell activation and memory formation are far from being elucidated. Our knowledge of CD8 T cell function derives primarily from models of antiviral responses, which are distinct from alloimmune responses, particularly in the aspect of Ag recognition. Alloreactive T cells are capable to interact with allo-MHC Ags via a unique direct pathway, which leads to their extremely high precursor frequency in T cell repertoire and high MHC-binding affinity (3). In antiviral T cell responses, in contrast, Ags are processed and presented to T cells as peptides by the host’s own MHC molecules. Depending on the type of viral infection, the CD8 activation in vivo may proceed through both CD4 help-dependent and -independent pathways (4). Although details of defective memory CD8 response in the absence of CD4 remain to be defined, the CD8 memory development has been shown in most cases to require CD4 help (4, 5, 6).

We have used murine transplant models to study alloreactive CD8 T cell activation and its costimulation requirements. By using MHC fully mismatched skin grafts as alloantigen stimulant, we showed that alloreactive CD8 activation may proceed through CD4-dependent and -independent pathways, and that both pathways require CD154 costimulation (7, 8). Furthermore, alloantigen-primed CD8 T cells in skin-sensitized recipients become CD154 independent when reactivated by a second skin graft, indicating the functional disparity between naive and memory alloreactive CD8 T cells (9). The present work further dissects CD4-independent alloreactive CD8 activation, memory formation, and in vivo function. As cardiac allografts survive long-term in CD4 knockout (KO)3 recipients despite activated CD8 T cells in the periphery, the question as to why these activated alloreactive CD8 fail to reject the graft was also addressed. As the execution of graft rejection in vivo requires the presence of activated T cells at the graft site, we focused on the role of CD4 in coordinating peripheral T cell activation, intragraft chemokine expression, and local T cell infiltration.

Wild-type (WT) BALB/c (H-2d), C57BL/6 (H-2b), and CD4-deficient mice (CD4 KO; B6; intercrossed at least 10 generations) were used. All mice (age of 8–12 wk; 20–25 g) were obtained from The Jackson Laboratory, and were housed in the University of California (Los Angeles, CA) animal facilities under pathogen-free conditions. Orthotopic full-thickness skin grafts (∼0.5 cm in diameter) from BALB/c donors were sutured bilaterally onto the flanks of C57BL/6 recipients. Heterotopic vascularized BALB/c hearts were transplanted using standard microsurgery techniques (7, 8, 9). Graft survival was assessed daily by palpation of ventricular activity. The day of rejection was defined as the day of cessation of heartbeat, and was verified by autopsy and pathological examination.

Rat anti-mouse CD4 nondepleting Ab YTS177.9 (courtesy of Dr. H. Waldmann, Oxford University, Oxford, U.K.) was administrated either perioperationally: two doses at days −1 and +1 or four doses at −1, +1, +3, +5, or postoperationally: two doses at days +2 and +4 or +4 and +6, in skin-grafted WT recipients (0.5 mg/mouse/dose i.v.). A single dose of anti-CD154 Ab (MR1, hamster Ig), purchased from Bioexpress, was administered i.v. at day 0. Control recipients were treated with relevant doses of rat or hamster Ig.

RBC-free lymphocytes from PBL, splenocytes, or lymph node cells were prepared. One million cells were used for Ab staining in ice-cold PBSA (PBS with 1% BSA). Cells were first incubated with 10 μg of normal rat IgG to block Fc-binding sites. After washing, cells were stained with 0.5–1 μg of rat anti-mouse CD8a-FITC (Clone 53-6.7), CD62L-R-PE (clone MEL-14), and CD44-CyChrome (clone IM7; BD Pharmingen). After washing, three-color flow cytometry was performed on a FACScan cytometer (BD Biosciences). Cells in lymphocyte gate and stained positive for CD8a were analyzed for their CD62L and CD44 expression. CTLeff were identified as CD8+CD62LlowCD44high population (10).

Responder splenocytes from B6 recipients were labeled with CFSE (Molecular Probes) at 4 mM in PBS for 15 min at 37°C. The unconjugated CFSE was eliminated by washing the cells with FBS (20%) supplemented RPMI 1640. The labeled cells were resuspended in culture medium, and incubated with irradiated B6 (syngeneic), B/c (donor-type) stimulator cells (2 × 106/ml), or Con A (4 μg/ml; Sigma-Aldrich). At day 4, cells were harvested, and stained with anti-mouse CD8-R-PE, CD4-Cy5 (eBioscience). Topro 3 (1 nM; Molecular Probes) was added as viable dye. Four-color flow cytometry was performed on a FACSCalibur dual-laser cytometer (BD Biosciences). Cells in lymphocyte gate, Topro 3 negative (viable cells), stained positive for CD4/CD8 were analyzed for CFSE intensities. Cytokine production was measured by restimulating cells at day 4 with PMA (50 ng/ml) and ionomycin (250 ng/ml) for 6 h in the presence of GolgiStop (BD Biosciences). Cells were then stained first with CD8-allophycocyanin, followed by fixation and permeabilization. PE labels anti-IFN-γ and -IL-4 were used to stain for intracellular cytokines.

A total of 2.5 μg of RNA was reverse-transcribed into cDNA using SuperScriptTM III First-Strand Synthesis System (Invitrogen Life Technologies). Quantitative PCR was performed using DNA Engine with Chromo 4 Detector (MJ Research). In a final reaction volume of 25 μl, the following were added: 1× SuperMix (Platinum SYBR Green qPCR kit; Invitrogen Life Technologies), cDNA, and 0.5 mM of each primer. Amplification conditions were: 50°C (2 min), 95°C (5 min) followed by 50 cycles of 95°C (15 s), 60°C (30 s). Table I lists primers used to amplify a specific gene fragment.

Table I.

Primer sequences for quantitative RT-PCR

GeneSequences (5′–3′)a
HPRT S: tcaacgggggacataaaagt 
 AS: tgcattgttttaccagtgtcaa 
CXCL10 S: gctgccgtcattttctgc 
 AS: tctcactggcccgtcatc 
CXCL9 S: cttttcctcttgggcatcat 
 AS: gcatcgtgcattccttatca 
CCL5 S: cgcacctgcctcaccatatg 
 AS: tgacaaacacgactgcaagattg 
CD3 S: tgctcttggtgtatatctcattgc 
 AS: aacagagtctgcttgtctgaagc 
CD8 S: tccttgatcatcactctcatctg 
 AS: actagcggcctgggacat 
CXCR3 S: gcagcacgagacctgacc 
 AS: ggcatctagcacttgacgttc 
Granzyme B S: gctgctcactgtgaaggaagta 
 AS: ttaccatagggatgacttgctg 
Perforin S: tcttggtgggacttcagctt 
 AS: aaggcccaggaggaacag 
IFN-γ S: ggaggaactggcaaaaggat 
 AS: ttcaaagagtctgaggtagaaagagat 
GeneSequences (5′–3′)a
HPRT S: tcaacgggggacataaaagt 
 AS: tgcattgttttaccagtgtcaa 
CXCL10 S: gctgccgtcattttctgc 
 AS: tctcactggcccgtcatc 
CXCL9 S: cttttcctcttgggcatcat 
 AS: gcatcgtgcattccttatca 
CCL5 S: cgcacctgcctcaccatatg 
 AS: tgacaaacacgactgcaagattg 
CD3 S: tgctcttggtgtatatctcattgc 
 AS: aacagagtctgcttgtctgaagc 
CD8 S: tccttgatcatcactctcatctg 
 AS: actagcggcctgggacat 
CXCR3 S: gcagcacgagacctgacc 
 AS: ggcatctagcacttgacgttc 
Granzyme B S: gctgctcactgtgaaggaagta 
 AS: ttaccatagggatgacttgctg 
Perforin S: tcttggtgggacttcagctt 
 AS: aaggcccaggaggaacag 
IFN-γ S: ggaggaactggcaaaaggat 
 AS: ttcaaagagtctgaggtagaaagagat 
a

S, sense or 5′ primer; AS, antisense or 3′ primer.

The results are shown as mean ± SD. Statistical analyses were performed using Student’s t test with p < 0.05 considered as significant.

To determine the impact of CD4 help on alloreactive CD8 activation by allogeneic skin, we quantitated CD8 activation and its kinetics by measuring the frequency of CD8 CTLeff (CD8+CD44highCD62Llow) by serial PBLs sampling from CD4 KO B6 recipients of BALB/c skin grafts. BALB/c skin grafts were rejected in CD4 KO B6 recipients with slightly delayed kinetics as compared with WT (14 ± 3 vs 10 ± 2 days), consistent with a published report (11). Indeed, CD8 T cells in CD4 KO recipients were activated, but at much reduced levels, as compared with WT (Fig. 1,a). Although the kinetics of alloreactive CD8 activation were comparable in CD4 KO and WT hosts (both peaked at day 10 post-skin grafting), the frequency of activated CTLeff in CD4 KO mice was 18.5 ± 1.7% of total CD8, a ∼50% reduction of that in WT (44.3 ± 4%; p < 0.0001). We observed the same degree of CTLeff reduction in spleens and lymph nodes of CD4 KO vs WT recipients (data not shown). The reduction of CTLeff frequency also reflected the overall lower numbers of CTLeff in CD4 KO vs WT recipients, as the total number of spleen cells and the percentage of CD8 T cells were similar in WT and CD4 KO recipients (at day 10 post-skin graft: 50 ± 10 × 106/spleen; 15 ± 4% of CD8; n = 6/group, in both groups of mice). Furthermore, similar reduction of alloreactive CD8 activation was observed in WT recipients in which CD4 T cells were either depleted or their function was blocked by Ab treatment (data not shown, and see Fig. 3). The numbers of CTLeff decreased in time, and by day 40 constituted <10% of all CD8 T cells in both animal groups. Interestingly, unlike the kinetics itself, the distribution of CD8 subsets was markedly different between alloantigen-primed WT and CD4 KO recipients (Fig. 1 b). Thus, the CD44highCD62Lhigh subset, representing central memory cells, was present in WT recipients at a significantly higher frequency as compared with CD4 KO recipients (at day +40: 61.08 ± 0.96% vs 29 ± 2.6%, n = 6; p < 0.001). In fact, the latter was comparable with naive animals (30.83 ± 1.57% of the central memory subset; n = 6). These results indicate a potential failure to generate a memory repertoire in alloantigen-primed CD4 KO mice.

FIGURE 1.

Primary alloreactive CD8 activation in WT and CD4 KO recipients. BALB/c skin grafts were transplanted onto WT or CD4 KO C57BL/6 mice. Alloreactive CD8 activation and its kinetics following allogeneic skin grafts were measured serially by FACS analysis of the frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood, as described in Materials and Methods. a, The percentages of CD8 CTLeff in total CD8 measured at days 0, 10, 20, and 40 posttransplant in WT and CD4 KO recipients were plotted (n = 3–6/group). b, The representative density plots are expressed as the percent of central memory (CD44highCD62Lhigh) and effector/effector memory (CD44highCD62Llow) frequency in the total CD8 population.

FIGURE 1.

Primary alloreactive CD8 activation in WT and CD4 KO recipients. BALB/c skin grafts were transplanted onto WT or CD4 KO C57BL/6 mice. Alloreactive CD8 activation and its kinetics following allogeneic skin grafts were measured serially by FACS analysis of the frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood, as described in Materials and Methods. a, The percentages of CD8 CTLeff in total CD8 measured at days 0, 10, 20, and 40 posttransplant in WT and CD4 KO recipients were plotted (n = 3–6/group). b, The representative density plots are expressed as the percent of central memory (CD44highCD62Lhigh) and effector/effector memory (CD44highCD62Llow) frequency in the total CD8 population.

Close modal
FIGURE 3.

Transient CD4 blockade and primary alloreactive CD8 activation in WT recipients. BALB/c skin grafts were transplanted onto WT C57BL/6 mice that remained untreated or treated with CD4-blocking mAb (YTS177.9), as described in Materials and Methods. Alloreactive CD8 activation, as assessed by frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood was measured serially by FACS. The kinetics of CD8 CTLeff induction in the presence or absence of CD4 help was plotted. Mean and SD are shown; n = 3–6/group.

FIGURE 3.

Transient CD4 blockade and primary alloreactive CD8 activation in WT recipients. BALB/c skin grafts were transplanted onto WT C57BL/6 mice that remained untreated or treated with CD4-blocking mAb (YTS177.9), as described in Materials and Methods. Alloreactive CD8 activation, as assessed by frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood was measured serially by FACS. The kinetics of CD8 CTLeff induction in the presence or absence of CD4 help was plotted. Mean and SD are shown; n = 3–6/group.

Close modal

We have shown that second donor-type alloantigen challenge in primed WT recipients stimulates alloreactive CD8 activation in a CD154 costimulation-independent fashion (9). This may represent one of the key features in memory CD8 T cell activation. To functionally test whether such a memory CD8 activation occurs in primed CD4 KO mice, a second BALB/c skin was transplanted 40 days after the first skin. The CD8 activation, as measured by increased frequency of CTLeff in PBLs, was readily detected in primed recipients (Fig. 2,a). However, unlike in WT, it remained sensitive to CD154 blockade in CD4 KO recipients, as evidenced by the ability of MR1 mAb at the time of second skin to inhibit CTLeff (Fig. 2 b). Thus, alloantigen-primed CD8 T cells in CD4 KO skin-grafted mice were functionally comparable with naive CD8 T cells in their response to Ag rechallenge. This indicates that CD4 KO recipients fail to or develop defective memory alloreactive CD8 T cells.

FIGURE 2.

Reactivation of primed alloreactive CD8 T cells and CD154 blockade. Primed C57BL/6 recipients, both WT and CD4 KO, were rechallenged with second BALB/c skin graft with or without CD154 blockade 40 days after first skin graft. Alloreactive CD8 activation was measured by FACS and representative density plots are shown: (a) as a percent of central memory (CD44highCD62Lhigh) and effector/effector memory (CD44highCD62Llow) percentages in total CD8 T cells. Similar results were recorded in three separate experiments, n = 3–6/group. b, Statistical analysis of a.

FIGURE 2.

Reactivation of primed alloreactive CD8 T cells and CD154 blockade. Primed C57BL/6 recipients, both WT and CD4 KO, were rechallenged with second BALB/c skin graft with or without CD154 blockade 40 days after first skin graft. Alloreactive CD8 activation was measured by FACS and representative density plots are shown: (a) as a percent of central memory (CD44highCD62Lhigh) and effector/effector memory (CD44highCD62Llow) percentages in total CD8 T cells. Similar results were recorded in three separate experiments, n = 3–6/group. b, Statistical analysis of a.

Close modal

To determine the timing and duration of CD4 help required for alloreactive CD8 activation, WT mice were treated with CD4-blocking Ab at different time points post-skin graft challenge. Perioperative CD4 blockade (day −1, +1, or day −1, +1, +3, +5) effectively inhibited early CD8 activation, with CTLeff at day 10 in both treatment protocols significantly lower, as compared with untreated controls (15.3 and 14.7%, vs 49.9%; p < 0.005, Fig. 3), and similar to those found in CD4 KO recipients (Fig. 1,a). However, the inhibition was transient because once CD4-blocking Ab was removed from the circulation, the CD8 CTLeff increased and peaked by day +20 in the two treated groups (37.1 and 43.7%), at levels similar to untreated controls. The frequency of CTLeff in the two dose Ab-treated recipients declined to baseline by day 40, similar to untreated group. Interestingly, the four dose Ab regimen resulted in a decline but with a somewhat delayed kinetics, with the percent CTLeff remaining as high as 22.2% by day 40. Interestingly, postoperative CD4 blockade (two-dose regimen starting at day +2 or +4) had no effect whatsoever on alloreactive CD8 activation (Fig. 3). Thus, peritransplant CD4 blockade delayed, but did not inhibit, alloreactive CD8 activation.

Although at reduced levels, alloreactive CD8 T cells were activated in CD4 KO hosts by allogeneic skin grafts. We have also shown previously that CD8 T cells harvested from CD4 KO recipients were capable of lysing target cells in vitro (7). To determine why CD8 cells failed to reject MHC-fully mismatched (B/c) hearts in CD4-deficient hosts, we first confirmed the activation of alloreactive CD8 T cells by cardiac allografts in CD4 KO mice. Although at significantly lower levels than in WT, increased numbers of CTLeff were consistently detected in PBLs of CD4 KO cardiac allograft recipients (Fig. 4,a, CD4 KO = 12.45 ± 1.03% vs WT = 21.88 ± 1.73%, n = 4, p < 0.005). In MLR in vitro, CFSE-labeled spleen CD8 T cells from CD4 KO heart graft recipients proliferated at a lower degree against donor Ag (B/c) than those from WT counterparts (Fig. 4,b). Most importantly, CD8 T cells from CD4 KO hosts produced much lower levels of IFN-γ, as compared with WT (Fig. 4,c). Next, we analyzed the intragraft chemokine induction profile. Cardiac allografts were harvested from CD4 KO recipients at days 5 and 15; syngeneic grafts served as controls. In parallel, cardiac allografts and syngeneic grafts were harvested from WT hosts at day 5. As shown in Fig. 5, marked reduction of T cell-targeted chemokines, including CXCL9, 10, and CCL5, was noted in cardiac allografts of CD4 KO mice at day 5, as compared with WT. By that time, activated CD8 T cells in WT were readily accumulating in allografts to trigger rejection (as evidenced by increased expression of chemokines, CD3/CD8/CXCR3, and granzyme B/perforin/IFN-γ). The induction of T cell chemokines, particularly CXCL9 and CCL5, recovered in CD4 KO recipients by day 15, in parallel with the peak of T cell accumulation (as shown by CD3, CD8, and CXCR3 levels). However, graft expression of T cell functional molecules (granzyme B and IFN-γ) in CD4 KO mice remained low. Thus, delayed induction of intragraft T cell-targeted chemokines in CD4 KO recipients correlated with delayed infiltration by T cells unable to express cytotoxic molecules.

FIGURE 4.

Alloreactive CD8 T cell activation in cardiac allograft recipients. a, In vivo CD8 activation, as assessed by FACS analysis of the frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood at day 10 post cardiac allograft. b, Splenocytes, harvested from cardiac allograft recipients of WT (dark line) or CD4 KO (dotted line) (BALB/c → C57BL/6) at day 10, were labeled with CFSE and cultured with irradiated syngeneic (B6) or allogeneic (B/c) splenocytes or Con A for 4 days. Cells were stained with fluorochrome-labeled anti-CD8 Abs. Viable CD8+ cells were gated and representative CFSE histograms are shown (n = 3). c, Lymphocytes from the same MLRs were also stained for intracellular IFN-γ expression, as described in Materials and Methods.

FIGURE 4.

Alloreactive CD8 T cell activation in cardiac allograft recipients. a, In vivo CD8 activation, as assessed by FACS analysis of the frequency of CD8 CTLeff (CD8+CD44highCD62Llow) in peripheral blood at day 10 post cardiac allograft. b, Splenocytes, harvested from cardiac allograft recipients of WT (dark line) or CD4 KO (dotted line) (BALB/c → C57BL/6) at day 10, were labeled with CFSE and cultured with irradiated syngeneic (B6) or allogeneic (B/c) splenocytes or Con A for 4 days. Cells were stained with fluorochrome-labeled anti-CD8 Abs. Viable CD8+ cells were gated and representative CFSE histograms are shown (n = 3). c, Lymphocytes from the same MLRs were also stained for intracellular IFN-γ expression, as described in Materials and Methods.

Close modal
FIGURE 5.

Immune-related gene induction in WT and CD4 KO cardiac allograft recipients (BALB/c to C57BL/6). Allografts (▦) and spleens (▪) were harvested from WT (day 5) and CD4 KO (days 5 and 15) groups. Total RNA was isolated and subjected to quantitative RT-PCR analysis. Target gene expression profiles were determined by their ratios to the housekeeping HPRT gene. Gene expression levels in syngeneic C57BL/6 hearts, which were transplanted into WT or CD4 KO recipients and harvested at day 10, served as controls. The target gene expression ratios were plotted. Similar results were recorded in three separate experiments.

FIGURE 5.

Immune-related gene induction in WT and CD4 KO cardiac allograft recipients (BALB/c to C57BL/6). Allografts (▦) and spleens (▪) were harvested from WT (day 5) and CD4 KO (days 5 and 15) groups. Total RNA was isolated and subjected to quantitative RT-PCR analysis. Target gene expression profiles were determined by their ratios to the housekeeping HPRT gene. Gene expression levels in syngeneic C57BL/6 hearts, which were transplanted into WT or CD4 KO recipients and harvested at day 10, served as controls. The target gene expression ratios were plotted. Similar results were recorded in three separate experiments.

Close modal

In this study, we compared the activation and function of alloreactive CD8 T cells in the presence or absence of CD4 help in transplant recipients. In addition to quantitative reduction of activated CD8 cells in Ag-challenged CD4 KO mice, at least three previously unrecognized defects were revealed. First, CD4-helpless CD8 T cells, after initial alloantigen priming, failed to develop the CD154 costimulation blockade-resistant phenotype in their reactivation. Second, CD4-helpless CD8 T cells exhibited decreased priming and effector responses. Third, CD4 T cells were involved in the induction of intragraft T cell-targeted chemokines. The delay in chemokine gene induction observed in CD4 KO recipients correlated with delayed CD8 T cell infiltration into cardiac allografts. More importantly, the late infiltrating CD8 cells did not express cytotoxic molecules, which may explain why they failed to reject the graft. The CD4 help required for alloreactive CD8 was transient, as posttransplant CD4 blockade failed to ameliorate CD8 activation. Moreover, peritransplant CD4 blockade, although effective in reducing early CD8 activation, had no long-lasting impact as the CD8 activation fully recovered and CD154 costimulation blockade-resistant phenotype developed.

The impact of CD4 help on CD8 activation and function varies at different stages of CD8 T cell responses (4, 5, 12). For the primary response, depending on alloantigen type and the chosen endpoint (proliferation vs viral clearance), CD8 T cells can respond in both CD4-dependent and -independent manners. In general, for Ag stimulation capable of directly activating APCs via innate immune receptors (e.g., TLRs), such as viruses or bacteria, CD8 T cell activation (clonal expansion/cytokine production) remains Th independent. In contrast, for Ag stimulation that lacks the innate immune signaling, such as organ grafts, T cell help is necessary to activate quiescent APCs, and is thus required for CD8 activation. CD154 expression by alloantigen-stimulated CD4 T cells represents one such “help” mechanism promoting APC activation via CD40 (10, 13, 14, 15). However, CD8 T cells may respond to noninfectious Ags independent of CD4 help when Ag-specific precursors are present at high frequencies (16, 17), or the affinity of Ag epitopes to MHC class I was high (18). Indeed, activation of cross-primed OVA-specific TCR-transgenic CD8 T cells became CD4 independent when increasing numbers of CD8 T cells were infused into class II-deficient mice. Furthermore, this CD4-independent CD8 activation did not require CD154 signaling (16). In allograft models, CD4-independent alloreactive CD8 activation and organ rejection have been described (11, 19, 20, 21, 22, 23, 24, 25, 26), although the latter constituted an endpoint with little or no screening for actual alloreactive CD8 activation. Interestingly, CD4-independent CD8 activation was found to be CD154 costimulation blockade resistant (22, 23, 27). We have shown that alloreactive CD8 T cells become activated in a CD4-dependent and -independent fashion, both of which do require CD154 costimulation (7). Here, we further evaluated the impact of CD4 help on alloreactive CD8 activation using CD4-deficient mice or CD4-blocking Abs in WT mice. In both settings in the absence of CD4 help, we noted a ∼50% reduction in the numbers of activated CTLeff in the periphery.

As MHC-fully mismatched cardiac allografts survived long-term in CD4-deficient hosts despite the presence of activated CD8 T cells, we assessed their function. Although the kinetics of CD8 cell activation were comparable, CD8 cells in CD4KO cardiac allograft recipients exhibited decreased frequency of CTLeff, and decreased proliferation and cytokine production in response to rechallenge with donor alloantigen. In addition to these defects in priming, we assessed the intragraft trafficking of activated CD8 T cells, a critical step for executing allograft rejection in vivo. Compared with WT, cardiac allografts in CD4KO recipients exhibit delayed induction of T cell-targeted chemokines, including CCL5 (RANTES) and CXCL9 (Mig). In parallel, the absence of CD4 T cells delayed accumulation of CD8 cells within the graft and diminished the up-regulation of cytotoxic molecules throughout. Because the kinetics of alloreactive CD8 activation in CD4 KO and WT hosts was comparable, the reduced level of activation and delayed intragraft T cell trafficking are both likely to contribute to low intragraft cytotoxicity levels seen in CD4 KO mice. This in turn may lead to the failure of helpless CD8 T cells to reject the grafts. Moreover, tissue inflammation at this later phase might be diminished, which in turn reduces the activation-signaling threshold for intragraft T cells and prevents them from expressing cytotoxic molecules.

In addition to the primary CD8 response, we also addressed the impact of CD4 help on CD8 memory generation and function. In antiviral T cell responses, CD4 help is required for the generation of an optimal functional memory CD8 T cell pool (28, 29, 30). However, controversy exists as to the timing of CD4 help during CD8 activation, and the nature of helpless memory CD8 T cell defects (28, 31). Cases of functional CD8 memory generation and response independent of CD4 help have also been reported (25, 32). For example, CD8 activation by vesicular stomatitis virus in the absence of CD4 help, although quantitatively reduced, resulted in a comparable memory response. It was qualitatively similar to CD4 help in CD8 activation, as assessed by cell proliferation, cytokine production, and cytolytic activity (32). In allograft settings, alloreactive-transgenic CD8 T cells also differentiated into fully functional memory cells in the absence of CD4 T cells (25). Our current study explored a novel aspect of memory CD8 T cells, i.e., CD154 costimulation blockade-resistant activation. We and others (9, 33, 34, 35) have demonstrated that alloreactive CD8 T cells in the absence of CD4 help fail to develop this phenotype, as reactivation of helpless alloreactive CD8 memory T cells remains susceptible to the CD154 blockade suggesting that these cells are functionally naive.

In summary, our results identify several novel defects in the activation and memory response of helpless alloreactive CD8 T cells in vivo. These findings further highlight the critical role of CD4 help in the activation and function of CD8 T cells in transplant recipients.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institutes of Health Grants AI23847, AI42223, DK62357 (to J.W.K.-W.), the American Heart Association (to Y.Z.), and the Dumont Research Foundation.

3

Abbreviations used in this paper: KO, knockout; WT, wild type; CTLeff, CTL effector.

1
Rocha, P. N., T. J. Plumb, S. D. Crowley, T. M. Coffman.
2003
. Effector mechanisms in transplant rejection.
Immunol. Rev.
196
:
51
-64.
2
Le Moine, A., M. Goldman, D. Abramowicz.
2002
. Multiple pathways to allograft rejection.
Transplantation
73
:
1373
-1381.
3
Rogers, N. J., R. I. Lechler.
2001
. Allorecognition.
Am. J. Transplant.
1
:
97
-94.
4
Rocha, B., C. Tanchot.
2004
. Towards a cellular definition of CD8+ T-cell memory: the role of CD4+ T-cell help in CD8+ T-cell responses.
Curr. Opin. Immunol.
16
:
259
-263.
5
Bourgeois, C., C. Tanchot.
2003
. Mini-review CD4 T cells are required for CD8 T cell memory generation.
Eur. J. Immunol.
33
:
3225
-3231.
6
Northrop, J. K., H. Shen.
2004
. CD8+ T-cell memory: only the good ones last.
Curr. Opin. Immunol.
16
:
451
-455.
7
Zhai, Y., L. Meng, R. W. Busuttil, M. H. Sayegh, J. W. Kupiec-Weglinski.
2003
. Activation of alloreactive CD8+ T cells operates via CD4-dependent and CD4-independent mechanisms and is CD154 blockade sensitive.
J. Immunol.
170
:
3024
-3028.
8
Zhai, Y., X. D. Shen, F. Gao, A. J. Coito, B. A. Wasowska, A. Salama, I. Schmitt, R. W. Busuttil, M. H. Sayegh, J. W. Kupiec-Weglinski.
2002
. The CD154-CD40 T cell costimulation pathway is required for host sensitization of CD8+ T cells by skin grafts via direct antigen presentation.
J. Immunol.
169
:
1270
-1276.
9
Zhai, Y., L. Meng, F. Gao, R. W. Busuttil, J. W. Kupiec-Weglinski.
2002
. Allograft rejection by primed/memory CD8+ T cells is CD154 blockade resistant: therapeutic implications for sensitized transplant recipients.
J. Immunol.
169
:
4667
-4673.
10
Mobley, J. L., M. O. Dailey.
1992
. Regulation of adhesion molecule expression by CD8 T cells in vivo, I: differential regulation of gp90MEL-14 (LECAM-1), Pgp-1, and VLA-4α during the differentiation of cytotoxic T lymphocytes induced by allographs.
J. Immunol.
148
:
2348
-2356.
11
Youssef, A. R., C. Otley, P. W. Mathieson, R. M. Smith.
2004
. Role of CD4+ and CD8+ T cells in murine skin and heart allograft rejection across different antigenic disparities.
Transpl. Immunol.
13
:
297
-304.
12
Castellino, F., R. N. Germain.
2006
. Cooperation between CD4+ and CD8+ T cells: when, where, and how.
Annu. Rev. Immunol.
24
:
519
-540.
13
Grewal, I. S., H. G. Foellmer, K. D. Grewal, J. Xu, F. Hardardottir, J. L. Baron, C. A. Janeway, Jr, R. A. Flavell.
1996
. Requirement for CD40 ligand in costimulation induction, T cell activation, and experimental allergic encephalomyelitis.
Science
273
:
1864
-1867.
14
Fujii, S., K. Liu, C. Smith, A. J. Bonito, R. M. Steinman.
2004
. The linkage of innate to adaptive immunity via maturing dendritic cells in vivo requires CD40 ligation in addition to antigen presentation and CD80/86 costimulation.
J. Exp. Med.
199
:
1607
-1618.
15
Bennett, S. R., F. R. Carbone, F. Karamalis, R. A. Flavell, J. F. Miller, W. R. Heath.
1998
. Help for cytotoxic-T-cell responses is mediated by CD40 signalling.
Nature
393
:
478
-480.
16
Mintern, J. D., G. M. Davey, G. T. Belz, F. R. Carbone, W. R. Heath.
2002
. Cutting edge: precursor frequency affects the helper dependence of cytotoxic T cells.
J. Immunol.
168
:
977
-980.
17
Wang, B., C. C. Norbury, R. Greenwood, J. R. Bennink, J. W. Yewdell, J. A. Frelinger.
2001
. Multiple paths for activation of naive CD8+ T cells: CD4-independent help.
J. Immunol.
167
:
1283
-1289.
18
Franco, A., D. A. Tilly, I. Gramaglia, M. Croft, L. Cipolla, M. Meldal, H. M. Grey.
2000
. Epitope affinity for MHC class I determines helper requirement for CTL priming.
Nat. Immunol.
1
:
145
-150.
19
Vu, M. D., F. Amanullah, Y. Li, G. Demirci, M. H. Sayegh, X. C. Li.
2004
. Different costimulatory and growth factor requirements for CD4+ and CD8+ T cell-mediated rejection.
J. Immunol.
173
:
214
-221.
20
Zhan, Y., A. J. Corbett, J. L. Brady, R. M. Sutherland, A. M. Lew.
2000
. CD4 help-independent induction of cytotoxic CD8 cells to allogeneic P815 tumor cells is absolutely dependent on costimulation.
J. Immunol.
165
:
3612
-3619.
21
Manning, T. C., L. A. Rund, M. M. Gruber, F. Fallarino, T. F. Gajewski, D. M. Kranz.
1997
. Antigen recognition and allogeneic tumor rejection in CD8+ TCR transgenic/RAG−/− mice.
J. Immunol.
159
:
4665
-4675.
22
Guo, Z., L. Meng, O. Kim, J. Wang, J. Hart, G. He, M. L. Alegre, J. R. Thistlethwaite, Jr, T. C. Pearson, C. P. Larsen, K. A. Newell.
2001
. CD8 T cell-mediated rejection of intestinal allografts is resistant to inhibition of the CD40/CD154 costimulatory pathway.
Transplantation
71
:
1351
-1354.
23
Bishop, D. K., S. Chan Wood, E. J. Eichwald, C. G. Orosz.
2001
. Immunobiology of allograft rejection in the absence of IFN-γ: CD8+ effector cells develop independently of CD4+ cells and CD40-CD40 ligand interactions.
J. Immunol.
166
:
3248
-3255.
24
He, G., J. Hart, O. S. Kim, G. L. Szot, C. T. Siegel, J. R. Thistlethwaite, K. A. Newell.
1999
. The role of CD8 and CD4 T cells in intestinal allograft rejection: a comparison of monoclonal antibody-treated and knockout mice.
Transplantation
67
:
131
-137.
25
Jones, N. D., M. Carvalho-Gaspar, S. Luo, M. O. Brook, L. Martin, K. J. Wood.
2006
. Effector and memory CD8+ T cells can be generated in response to alloantigen independently of CD4+ T cell help.
J. Immunol.
176
:
2316
-2323.
26
Halamay, K. E., R. L. Kirkman, L. Sun, A. Yamada, R. C. Fragoso, K. Shimizu, R. N. Mitchell, D. B. McKay.
2002
. CD8 T cells are sufficient to mediate allorecognition and allograft rejection.
Cell. Immunol.
216
:
6
-14.
27
Jones, N. D., A. Van Maurik, M. Hara, B. M. Spriewald, O. Witzke, P. J. Morris, K. J. Wood.
2000
. CD40-CD40 ligand-independent activation of CD8+ T cells can trigger allograft rejection.
J. Immunol.
165
:
1111
-1118.
28
Janssen, E. M., E. E. Lemmens, T. Wolfe, U. Christen, M. G. von Herrath, S. P. Schoenberger.
2003
. CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes.
Nature
421
:
852
-856.
29
Shedlock, D. J., H. Shen.
2003
. Requirement for CD4 T cell help in generating functional CD8 T cell memory.
Science
300
:
337
-339.
30
Sun, J. C., M. J. Bevan.
2003
. Defective CD8 T cell memory following acute infection without CD4 T cell help.
Science
300
:
339
-342.
31
Sun, J. C., M. A. Williams, M. J. Bevan.
2004
. CD4+ T cells are required for the maintenance, not programming, of memory CD8+ T cells after acute infection.
Nat. Immunol.
5
:
927
-933.
32
Marzo, A. L., V. Vezys, K. D. Klonowski, S. J. Lee, G. Muralimohan, M. Moore, D. F. Tough, L. Lefrancois.
2004
. Fully functional memory CD8 T cells in the absence of CD4 T cells.
J. Immunol.
173
:
969
-975.
33
Valujskikh, A., A. Pantenburg, P. S. Heeger.
2002
. Primed allospecific T cells prevent the effects of costimulatory blockade on prolonged cardiac allograft survival in mice.
Am. J. Transplant.
2
:
501
34
Pantenburg, B., F. Heinzel, L. Das, P. S. Heeger, A. Valujskikh.
2002
. T cells primed by Leishmania major infection cross-react with alloantigens and alter the course of allograft rejection.
J. Immunol.
169
:
3686
-3693.
35
Vu, M. D., M. R. Clarkson, H. Yagita, L. A. Turka, M. H. Sayegh, X. C. Li.
2006
. Critical, but conditional, role of OX40 in memory T cell-mediated rejection.
J. Immunol.
176
:
1394
-1401.