Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Respiratory virus infection evokes potent immune responses in the lung, involving various arms of the innate and adaptive immune system. CD8+ T cells are particularly important in clearance of virus, but may also contribute to lung injury in the process (1), although the relative contribution of the virus infection itself and the T-cell antiviral effector activities in this context are unclear. The effector mechanisms involved in viral clearance and associated injury include both cytolytic and noncytolytic effector functions (5, 6). CD8+ T cells accumulate in the lung parenchyma in a variety of inflammatory disease states, but the nature of their specific contribution to lung injury is also unclear (4). One hypothesis concerning the pathogenesis of idiopathic interstitial pneumonia suggests that the immune response to a viral infection may become quantitatively or qualitatively inappropriate and may lead to acute and/or chronic lung injury. We used an in vivo model of interstitial pneumonia to examine the specific effects of CD8+ cytolytic T cells and their various effector activities on lung injury. We showed that activated antiviral T cells induce significant pulmonary inflammation and injury after adoptive transfer into transgenic animals expressing a viral antigen on alveolar epithelial cells, and in the absence of virus infection (5, 6). The injury results in considerable respiratory dysfunction and eventual death, in a time frame that is dependent upon cell dose. Alveolar epithelial cells are uniquely sensitive to the cytotoxic effect of transmembrane tumor necrosis factor (TNF)- expressed by CD8+ T cells, and are also activated in vitro and in vivo by CD8+ T-cell recognition to express inflammatory chemokines (5). We demonstrated in vivo that CD8+ T cell-mediated lung injury occurs in the absence of perforin and Fas, but neutralizing antibody to TNF- completely abrogates lung injury that occurs in absence of both mediators (7). In vitro, alveolar epithelial-derived cells appear sensitive to the cytotoxic effects of perforin and TNF- expressed by CD8+ T cells but are completely insensitive to induction of apoptosis by Fas ligand, despite expression of functional Fas on the alveolar cells (7). These cells are also significantly less susceptible to cytolysis induced by soluble TNF- than by TNF- expressed by T lymphocytes (7, 8).

CD8+ T cells express predominantly a transmembrane form of TNF- (8), which may initiate injury via direct cytotoxic effects on alveolar epithelial cells. This may contribute to the observed respiratory dysfunction that evolves in transgenic recipients. However, the inflammatory infiltration that ensues 3 to 4 d after adoptive transfer into hemagglutinin (HA)-transgenic mice consists largely of neutrophils, host lymphocytes, and (predominantly) activated macrophages; it is the presence of large numbers of these cells that correlates most strongly with the profound respiratory impairment observed after T-cell transfer (6). We have also demonstrated that alveolar epithelial cells are activated in vivo as a result of specific CD8+ T-cell antigen recognition to express the inflammatory chemokine monocyte chemotactic protein (MCP)-1 (5). This induction appears to be mediated primarily by transmembrane TNF- expressed on the surface of the antigen-specific CD8+ T cells, and in vivo neutralization of MCP-1 significantly abrogates the inflammatory infiltration that ensues after adoptive transfer of the T cells. In this study we sought to examine the relationship between alveolar epithelial cell cytotoxicity and activation that occurs as a direct result of CD8+ T-cell recognition of antigen presented by the epithelial cells. We present evidence that alveolar epithelial cells show considerably greater sensitivity to the cytotoxic effects of transmembrane TNF- than to perforin expressed by CD8+ T cells. Further, we show that alveolar epithelial cells may undergo activation or cytolysis as a result of antigen-specific CD8+ T-cell recognition, but that these outcomes are inversely related and vary with effector/target (E:T) ratio.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Effector T-Cell Populations

To obtain CD8+ T-cell populations with a defined HA specificity we immunized wild-type BALB/c mice with a recombinant vaccinia virus (vv) expressing an A/Japan/57 HA deletion mutant that lacks the transmembrane and cytoplasmic domain [vv(HAanchor-)]. The construct retains the HA204-212 and HA210-219 epitopes recognized in association with the H-2K^d major histocompatibility complex (MHC) class I molecule by H-2K^d-restricted CD8+ cytotoxic T lymphocytes (CTLs). HA-immune splenocytes from these animals, which contained CD8+ memory T cells specific for these two HA epitopes, were restimulated in vitro with splenocytes infected with the A/GV/17 (H2N2) influenza strain. This virus has a single nucleotide difference in the HA gene (compared with the A/Japan/57 strain) leading to an amino acid substitution in the 204-212 epitope (N-to-K change at residue 207 [6]). Consequently, this virus selectively activates and expands CD8+ memory CTL precursors from the vv(HAanchor-)-primed mice which are directed to the HA 210-219 epitope common to A/GV/17 and A/Japan/57 (6). These populations were maintained by weekly in vitro stimulation in fresh Iscove's complete media supplemented with 10 units/ml of interleukin (IL)-2. The clones were restimulated in vitro with irradiated syngeneic splenocytes that were infected with A/Japan/57 influenza. CD8+ T-cell clones used in these experiments were generated by limiting dilution as previously described (11). The clones were restimulated in vitro with irradiated syngeneic splenocytes that were infected with A/Japan/57 influenza, and placed into fresh Iscove's complete media supplemented with 10 units/ml of IL-2.

T-Cell Cytotoxicity Assay

CD8+ T-cell clones used in these experiments were generated either from wild-type or perforin-deficient mice (12). On Day 5 after in vitro stimulation, T cells were tested for cytolytic activity using a ⁵¹Cr-release assay against target cells infected with A/Japan/57 (multiplicity of infection [MOI} of approximately 100), or against target cells loaded with 10⁹ M synthetic peptide representing either the 204-212 (LYQNVQTYV) or the 210-219 (TYVSVGTSTL) epitope of the A/Japan/57 HA (13, 22). The target cells used in these assays were MLE-K^d, which are MLE-15 cells (13) stably transfected with the class I MHC molecule H-2K^d (6) or primary type II pneumocytes (see the following section). Cytotoxicity assays were carried out in 96-well plates with 0.2 ml/well for 6 h, after which 0.1 ml was harvested and counted on a -counter (Isomedic; ICN Biomedicals, Inc., Costa Mesa, CA). In some experiments, anti-TNF- antisera (IP400; Genzyme, Boston, MA) was added to wells at a final dilution of 1:100. Percent specific ⁵¹Cr release was calculated according to the formula: (test CPM  spontaneous release CPM)/(total CPM  spontaneous CPM) × 100 where CPM is counts per min. Spontaneous release from targets incubated with media alone was always less than 10%.

Specific lysis values represent the mean percent specific ⁵¹Cr release from four replicate wells. For analysis of MCP-1 expression, parallel plates were set up with unlabeled target cells, and supernatants were assayed for MCP-1 production using a sandwich enzyme-linked immunosorbent assay (ELISA) (Pharmingen, San Diego, CA) in accordance with the manufacturer's instructions.

Primary Type II Pneumocyte Preparation

Primary alveolar type II cells were prepared using a modification of a previously published method (14). Briefly, BALB/c mice were anesthetized and exsanguinated by severing the inferior vena cava and left renal artery. The tracheae were exposed and cannulated, and lungs were perfused with 10 to 20 ml sterile saline via the pulmonary artery until visually free of blood. Dispase (Collaborative Research, Inc., Bedford, MA) was instilled into the lungs via the tracheal catheter, followed by 1% low-melt agarose warmed to 45°C. The lungs were immediately covered with ice and incubated for 2 min to gel the agarose. Lungs were then dissected out, put in a culture tube containing an additional 1 ml of Dispase, and incubated for 45 min at room temperature. Lungs were then transferred to a culture dish containing DNAse I (Sigma, St. Louis, MO) and the tissue was gently teased away from the airways. The cell suspension was successively filtered and then pelleted. Crude cell suspensions were added to culture dishes coated with anti-CD45 and anti-CD32 antibodies (Pharmingen) and incubated for 1 to 2 h. Plates were removed from the incubator and gently "panned" to free settled type II cells, which were resuspended in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS). Purity of the type II cell preparations used for these studies was 92 ± 1% by morphologic criteria (14) and viability determined by trypan blue exclusion after overnight incubation was > 60%. (Purity for more than 100 separate preparations was consistently > 90%.)

Nuclear Factor-B Reporter Assays

MLE-K^d cells were transiently transfected with a nuclear factor (NF)-B-secreted alkaline phosphatase (SEAP) reporter construct (Clontech, La Jolla, CA), or control plasmids using Superfect (Qiagen, Valencia, CA). After transfection with 2 µg of DNA (in 0.5 ml) for 3 h, cells were then incubated for 48 h in DMEM with 5% FBS. Cells were harvested, plated in 96-well plates, and allowed to adhere overnight. T cells and peptide were then added at varying E:T ratios, or soluble TNF- was added at a saturating concentration (10,000 U/ml [7]), and plates were incubated at 37°C for 6 h, after which time supernatants were sampled for analysis. Chemiluminescent substrate (Clontech) was added according to the manufacturer's instructions, and samples were assayed on a luminometer (Turner Designs, Sunnyvale, CA).

Statistical Analysis

Significant differences were determined by the Mann-Whitney test. All error bars represent standard deviation.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Alveolar Epithelial-Derived Cells Efficiently Process and Present Class I MHC-Restricted Overlapping Viral Epitopes

Antiviral CD8+ T lymphocytes represent an important arm of the adaptive immune response to respiratory virus infection (11, 15). Influenza is a virus that primarily infects respiratory epithelial cells, and we have shown that cytolytic effector T cells can efficiently recognize a subdominant K^d-restricted epitope of the A/Japan/57 influenza HA expressed on virus-infected alveolar epithelial- derived cells leading to alveolar cell cytotoxicity (6, 7). This epitope, HA210-219, overlaps with a dominant K^d- restricted epitope, HA204-212, both of which are processed and presented efficiently by nonrespiratory cell lines in vitro (18). To confirm that alveolar epithelial-derived cells were also capable of efficient processing and presentation of both the dominant and subdominant epitopes, MLE-K^d cells were infected with A/Japan/57 influenza and then used as targets for recognition by epitope-specific CD8+ cytolytic T-cell clones. As shown in Figure 1, clones 40-2 (specific for HA210-219) and 14-1 (specific for HA204-212) recognize and lyse influenza-infected MLE-K^d cells with approximately equivalent efficiency. There was no overlap in the recognition pattern of 40-2 and 14-1, as shown in Figure 2, in which MLE-K^d cells were exogenously loaded with synthetic peptide. This also indicates that the pattern of cytolysis shown in Figure 1 represents a valid readout of epitope processing by alveolar epithelial-derived cells. We also tested peptide-dependent cytoxicity of MLE-K^d cells by perforin-deficient CD8+ T-cell clones PKOGV17 and PKOAA57, specific for the HA210-219 and HA204-212 epitopes, respectively. As shown in Figure 2, both of these T cells induce comparable cytotoxicity of MLE-K^d cells, loaded with appropriate peptide. Interestingly, the level of cytolysis of MLE-K^d cells induced by perforin-deficient CD8+ T-cell recognition was only slightly lower than that induced by wild-type T-cell recognition, suggesting that perforin-mediated cytotoxicity may play a relatively minor role in lysis of this particular target cell.

View larger version (12K): [in this window] [in a new window]   Figure 1.   Alveolar epithelial-derived cells process and present overlapping viral antigenic epitopes for MHC class I-restricted CD8+ T-cell recognition, and these target cells are efficiently lysed as a result of T-cell recognition. CD8+ T-cell cytotoxicity of influenza-infected alveolar epithelial-derived target cells. 51Cr-release assays were performed using two wild-type CD8+ T-cell clones, 14-1 (specific for HA204-212; circles) and 40-2 (specific for HA210-219; squares). MLE-Kd cells were infected with A/Japan/57 influenza (MOI of approximately 100) and incubated at several E:T ratios for 6 h before harvest of supernatants. Spontaneous release was derived from infected cells incubated for 6 h in the absence of T cells. The data shown are representative of four separate experiments. There were no significant differences between the two T-cell clones at any E:T ratio.

View larger version (33K): [in this window] [in a new window]   Figure 2.   CD8+ T-cell cytotoxicity of synthetic peptide-treated alveolar epithelial-derived target cells. 51Cr-release assays were performed using two CD8+ T-cell clones specific for HA210-219, wild-type 40-2 (A) and perforin-deficient PKO-GV (B); and two CD8+ T-cell clones specific for HA204-212, wild-type 14-1 (C) and perforin-deficient PKO-AA (D). MLE-Kd cells were treated with the appropriate peptide at 109 M and incubated with T cells at an E:T ratio of 10:1 (black bars) or 20:1 (gray bars) for 6 h before harvest of supernatants. The data shown are representative of three separate experiments. There was no significant overlap of the peptide-specific recognition of antigen presented on alveolar epithelial-derived cells, and the level of specific lysis was only slightly decreased in the absence of perforin. In A and B, there was no difference between the level of lysis with the 204-212 peptide and lysis with no peptide at either E:T ratio. There were significant differences between lysis with the 210-219 peptide compared with lysis with the 204-210 peptide at both E:T ratios (P < 0.01), as well as between the two E:T ratios on the 210-219 peptide (P < 0.01). In C and D, there was no difference between the level of lysis with the 210-219 peptide and lysis with no peptide at either E:T ratio. There were significant differences between lysis with the 210-219 peptide compared with lysis with the 204-210 peptide at both E:T ratios (P < 0.01), as well as between the two E:T ratios on the 204-212 peptide (P < 0.01). There were significant differences in specific lysis at both E:T ratios between wild-type (A) and perforin-deficient (B) T cells on 210-219 peptide-treated targets (P < 0.05), and between wild-type (C) and perforin-deficient (D) T cells on 204-210 peptide-treated targets (P < 0.05).

Cytotoxicity of Alveolar Epithelial Cells by Wild-Type CD8+ T Cells Is Mediated Primarily by TNF-

Previously we have shown that perforin-independent CD8+ T cell-mediated cytolysis of alveolar epithelial cells is independent of Fas/Fas ligand interaction and entirely dependent upon TNF- expressed by the T cell (7). We also showed that acute T cell-mediated lung injury in vivo is critically dependent upon TNF-, and is independent of Fas expression, though there may be a role for Fas in chronic injury (19). We have also previously shown that anti-TNF- completely inhibits cytotoxicity of MLE-K^d cells by the perforin-deficient T-cell clone PKOGV17 (7). As shown in Figure 3, this effect is not unique to this particular T-cell clone, nor is it unique to recognition of the subdominant HA210-219 epitope. Anti-TNF- inhibited all cytotoxicity of MLE-K^d cells by PKOAA57 (specific for HA204-212). It is interesting to note that this antibody also inhibited nearly all cytotoxicity of alveolar cells triggered by the wild-type clone 14-1 (also specific for HA204-212), though the level of lysis was still above background (not shown). These data further demonstrate that MLE-K^d cells are unusually sensitive to one particular T-cell effector activity, i.e., TNF-, and that this phenotype is not unique to one particular T-cell clone or unique to recognition of the subdominant K^d-restricted epitope of A/Japan/ 57 influenza HA. Further, MLE-K^d cells appear to have a limited susceptibility to perforin/granzyme-mediated cytolysis, inasmuch as the bulk of cytotoxicity of these cells appears to be mediated by TNF-, in contrast to lysis of nonrespiratory target cells, which is primarily mediated by the perforin pathway (3, 12, 20, 21). Previous studies in this system (22) and others (23) have shown that the sensitivity to perforin-mediated cytolysis may be directly related to the avidity of the T-cell receptor (TCR) for the peptide/MHC on the target cell. Further, these studies suggested that expression of low-avidity ligands on target cells may preferentially trigger perforin-independent cytolysis by the T cell. In such circumstances, cytotoxicity has been shown to be FasL-dependent, at least with respect to nonrespiratory target cells (22, 23). Because peptide/MHC complex avidity is directly related to peptide concentration (24), we characterized the contribution of perforin-dependent lysis to T cell- mediated cytotoxicity of MLE-K^d cells at varying peptide concentrations. As shown in Figure 4, cytolysis of MLE-K^d cells by the wild-type CD8+ T-cell clone 40-2 reached a plateau level at a peptide concentration of 10¹⁰ M, and most of the cytolysis was inhibited by anti-TNF-. At a 10-fold lower peptide concentration (10¹¹ M), TNF-dependent cytolysis was reduced but still significant (compared with backgound; P < 0.01), whereas perforin-dependent cytolysis (i.e., residual lysis in the presence of anti-TNF-) was not different from background levels. At 10¹² M peptide, cytolysis became undetectable (equal to background). These data suggest that MLE-K^d cells are extremely sensitive to T-cell cytotoxicity mediated by TNF-, and minimally sensitive to perforin/granzyme-mediated cytolysis. Thus, although the differential threshold for perforin-dependent and -independent cytotoxicity was also in evidence with these target cells, the effector activity associated with higher threshold activation events was mediated by TNF-. To confirm that this phenotype is not unique to the MLE-K^d cell line, we used primary type II cell isolates as targets of T-cell cytotoxicity, to assess the relative contribution of the two effector mechanisms. As shown in Figure 5, HA+ cells were lysed by both wild-type and perforin-deficient T-cell clones with comparable efficiency. This suggests that, similar to MLE-K^d cells, primary type II cells are also insensitive to perforin-mediated cytotoxicity.

View larger version (16K): [in this window] [in a new window]   Figure 3.   CD8+ T-cell cytotoxicity of synthetic peptide-treated alveolar epithelial-derived target cells in the presence (shaded bars) or absence (black bars) of neutralizing antibody to TNF-. 51Cr-release assays were performed using two CD8+ T-cell clones specific for HA204-212 (wild-type 14-1 and perforin-deficient PKO-AA). MLE-Kd cells were treated with 109 M peptide and incubated with T cells at an E:T ratio of 20:1 in the presence of anti-TNF- for 6 h before harvest of supernatants. The data shown are representative of three separate experiments. Anti- TNF- inhibited most of the cytotoxicity of alveolar epithelial- derived cells mediated by wild-type CD8+ T cells, and all of the cytotoxicity mediated by perforin-deficient CD8+ T cells. There were significant differences between lysis induced by PKO-AA and 14-1 in the absence (black bars; P < 0.05) and presence (gray bars; P < 0.01) of anti-TNF, as well as for each clone individually in the presence or absence of anti-TNF (P < 0.01 for both). PKO-AA plus anti-TNF produced lysis that was not different from background (not shown), but lysis triggered by 14-1 plus anti-TNF was above background (not shown; P < 0.01).

View larger version (24K): [in this window] [in a new window]   Figure 4.   Cytotoxicity of alveolar epithelial-derived target cells in the presence (gray bars) or absence (black bars) of neutralizing antibody to TNF-, at varying concentrations of antigenic peptide. 51Cr-release assays were performed using CD8+ T cell clone 40-2 (specific for HA210-219) at an E:T of 20:1. MLE-Kd cells were treated with the appropriate concentration of peptide and incubated with T cells at an E:T ratio of 20:1 for 6 h before harvest of supernatants. The data shown are representative of three separate experiments. The bulk of T cell-mediated cytotoxicity of alveolar epithelial-derived target cells was inhibited by anti-TNF- at all concentrations of peptide tested. At 1011 M peptide, specific lysis in the presence of antibody (gray bars) was not different from background, indicating that cytolysis of alveolar cells was exclusively mediated by TNF- at this level of T-cell activation. Lysis at 1011 M peptide was significantly different from lysis at 1010 M peptide in the absence (black bars; P < 0.01) or presence (gray bars; P < 0.01) of anti-TNF. Lysis in the presence of anti-TNF was significantly lower than in the absence of anti-TNF at all peptide concentrations greater than 1012 M (P < 0.01).

View larger version (15K): [in this window] [in a new window]   Figure 5.   Cytotoxicity of primary type II pneumocytes by two CD8+ T-cell clones specific for HA210-219, wild-type 40-2 and perforin-deficient PKO-GV. 51Cr-release assays were performed using type II cells isolated from surfactant protein C-HA transgenic mice (black bars) or littermate controls (gray bars) at an E:T ratio of 20:1 for 4 h before harvest of supernatants. Both T cells induced significant cytolysis of the HA+ target cells compared with the HA targets (P < 0.05). However, there was no significant difference between the level of cytotoxicity of type II pneumocytes induced by wild-type compared with perforin-deficient CD8+ T-cell recognition. The data shown are representative of two separate experiments.

CD8+ T-Cell Recognition of Alveolar Epithelial Cells Results in Epithelial Cell Activation

We previously demonstrated that recognition of type II pneumocytes by CTL may result not only in target cell apoptosis but also in alveolar cell activation, both in vitro and in vivo (5). This activation leads to alveolar MCP-1 expression, which contributes to the inflammatory infiltrates that ensue after CD8+ T-cell transfer in vivo (6). The relative contributions of alveolar cell cytotoxicity and activation to T cell-mediated lung injury in vivo, however, remains unclear. A critical determinant of cytotoxicity in vitro is the E:T ratio; however, the relationship between E:T and type II cell activation is unclear. To begin to analyze the relationship between E:T and alveolar cell activation, MLE-K^d cells were transfected with an NF-B reporter construct before incubation with varying numbers of CD8+ T cells. As shown in Figure 6, the degree of NF-B activation was inversely related to E:T, with the highest levels seen at 1:1 (P < 0.05 compared with 5:1). Further, at 6 h the level of NF-B activation at E:T of 1:1 was similar to the level of activation by soluble TNF- at a saturating concentration (7). This is consistent with the hypothesis that activation and cytolysis are competing processes, and that at the later time point, cytotoxicity limits the availability of viable cells capable of being activated. To further explore the relationship between cytotoxicity and activation after 6 h of exposure to T cells, MLE-K^d cells were used as targets of ⁵¹Cr-release assays in parallel with ELISA for MCP-1. We have previously demonstrated that MLE-K^d cells express MCP-1 upon activation by transmembrane TNF- expressed by CD8+ T cells (and also to a much lesser extent by soluble TNF-) and that CD8+ T cells themselves express no MCP-1 (5). As shown in Figure 7, the inverse relationship between cytolysis and activation of these alveolar epithelial-derived target cells was evident at the extremes of E:T, with very little cytotoxicity, and maximal MCP-1 expression observed at an E:T of 1:1. Conversely, MCP-1 expression became nearly undetectable at an E:T of 20:1, and cytotoxicity reached a plateau level of over 60%.

View larger version (16K): [in this window] [in a new window]   Figure 6.   CD8+ T-cell induction of NF-B reporter activity in alveolar epithelial-derived cells. MLE-Kd cells were transiently transfected with an NF-B-SEAP reporter before incubation with CD8+ T-cell clone 40-2, at varying E:T ratios, in the presence of 109 M antigenic peptide. The supernatants were harvested after 6 h, and SEAP activity was compared with that induced by soluble TNF- alone. T cell- triggered SEAP activity was highest at the lowest E:T of 1:1 (P < 0.05, compared with E:T of 5:1). The level of SEAP activity triggered by CD8+ T-cell recognition at an E:T of 1:1 was not significantly different from the level of activity induced by a saturating concentration of soluble TNF-. The levels of induction at E:T ratios of 2.5:1 and 5:1 were significantly different from the level induced by soluble TNF- (P < 0.01). The data shown are representative of two separate experiments.

View larger version (15K): [in this window] [in a new window]   Figure 7.   MCP-1 protein production by alveolar cells is triggered by CD8+ T-cell recognition, and varies inversely with cytotoxicity at varying E:T ratios. 51Cr-release assays were performed using the CD8+ T-cell clone 40-2 (squares), in parallel with ELISA performed on the supernatants (circles). MLE-Kd cells were treated with 109 M synthetic peptide and incubated with T cells at varying E:T ratios for 6 h before harvest. At 20:1 there was very little MCP-1 produced, whereas the cytolysis was very high. At 1:1 the cytolysis was low and the MCP-1 levels produced by the target cells were quite high. The levels of cytolysis varied directly with increasing E:T ratio (P < 0.01) until the effect reached a plateau at 10:1. The levels of MCP-1 induction decreased with increasing E:T ratio (P < 0.05), reaching background levels at 10:1. The data shown are representative of two separate experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Cytolytic CD8+ T cells represent an important arm of the adaptive immune response to virus infection, though T cell- mediated virus clearance may be associated with significant tissue injury (1). The immune response to respiratory virus infection leads to a complex inflammatory cascade in the lung, which may result in virus clearance, lung injury, or both. Virus infection of the respiratory epithelium triggers production of inflammatory mediators by the infected epithelial cells, such as IL-8 (25) and type 1 interferons (26). This leads to recruitment of cells and mediators of the innate immune system, and subsequently to those of the adaptive immune response. Epithelial cells which continue to present viral antigens during this phase become targets of antigen-specific cytolytic T-cell recognition. Specific CD8+ T-cell responses may play an even greater role in virus clearance (and tissue injury) in immune individuals, inasmuch as the memory CD8+ T-cell response emerges earlier and more vigorously than does the CD8+ T-cell response in primary virus infection (27, 28). Though specific CD8+ T-cell recognition of a virus-infected target cell may result in apoptotic death of the infected cell, the complexity of factors which influence susceptibility to cytolysis are increasingly being appreciated (2, 29). In addition, ample evidence exists to indicate that numerous soluble factors may trigger the expression of a variety of inflammatory mediators by respiratory epithelial cells. However, the specific contribution of lung epithelial cells to the inflammatory milieu which evolves in the context of an immune response to a respiratory virus infection has been difficult to assess because of the complicating effects of virus infection on epithelial cell activities. In this study we have shown that CD8+ T-cell recognition of alveolar epithelial cells leads to cytotoxicity in a manner much more dependent upon TNF- than upon perforin, even when both effector activities are available. We observed small but significant differences in lysis of alveolar cells triggered by perforin-deficient versus wild-type CD8+ T cells, indicating a residual role for perforin in alveolar cell cytotoxicity. These differences are unlikely to represent differences in the relative avidity of the different T-cell receptors for MHC/peptide complexes because the peptide dose used in these experiments (10⁹ M) was intentionally high enough to be well within the plateau range of cytotoxicity (100-fold higher than the break point in Figure 4).

Further, we have shown that specific CD8+ T-cell recognition of alveolar epithelial cells, in the absence of virus infection, may induce alveolar cell expression of a variety of inflammatory mediators in vitro and in vivo. The T cell may also produce soluble inflammatory mediators upon specific antigen recognition, and it is possible that both sources of chemokines contribute to the milieu that leads to parenchymal infiltration of host inflammatory cells, and the resultant respiratory embarrassment. Alveolar epithelial target cells may be important participants in the lung injury resulting from CD8+ T-cell recognition, inasmuch as significant macrophage accumulation occurs approximately 72 to 96 h after transfer and the transferred CD8+ T cells become undetectable in the lungs after approximately 48 h (Enelow and associates, unpublished observation). Further, CD8+ T-cell transfer appears to induce a similar degree of respiratory dysfunction in severe combined immunodeficiency mice expressing the HA transgene, which suggests against a critical role of the host lymphocyte component of the parenchymal infiltration (Enelow and coworkers, unpublished observation).

Chemokines are a large family of chemoattractant cytokines that appear to play a major role in the orchestration and amplification of acute and chronic inflammatory processes, particularly in the lung (32). The induction of chemokine expression by alveolar epithelial cells appears to be mediated primarily by TNF- expressed (and/ or secreted) by the CD8+ T cell, a mediator not commonly associated with T-cell effector activity (39). We have previously demonstrated that MCP-1 expression is preferentially triggered by transmembrane TNF-, whereas macrophage inflammatory protein-2 expression appears to be triggered preferentially by soluble TNF- (5). Distinct effector functions for the two different forms of TNF- have been described in several systems and may relate to the differential engagement of the two TNF receptors (9, 42). We speculate that activation of alveolar cells by CD8+ T-cell recognition as read out by MCP-1 production and as read out by NF-B activation are similar phenomena, at least to the extent that they are both inversely related to cytotoxicity. There are many possible triggers for activation of NF-B, TNF- being the prototype, but more work will be necessary to establish a direct link between NF-B activation and MCP-1 transcription.

The results presented in this study have been corroborated with several distinct CD8+ T-cell clones, indicating that the phenomena described are not unique to one T-cell clone. The activation of alveolar epithelium which occurs as a result of specific CD8+ T-cell recognition may reflect either a population effect or an effect of epithelial cells as individual responders to TCR engagement (or both). There are two potential mechanisms that might account for MCP-1 production by alveolar epithelial cells upon specific T-cell recognition, which are not mutually exclusive. The first involves the direct cytolysis of those target cells whose MHC/peptide complexes engage TCR, by either perforin or membrane-bound TNF- (though the contribution of the former appears quite limited) with associated antigen-dependent T-cell production of soluble TNF-. The soluble TNF- produced might therefore lead to the induction of bystander MCP-1 expression by other target cells, which stochastically did not engage the TCR of a lymphocyte and did not undergo apoptosis. The second possibility involves the direct ligation of TCR and TNF receptor(s) on an individual target cell, the result of which may be activation of the engaged target, and induction of chemokine expression in the target cell. A critical corollary to this hypothesis is that there may be interactions between cytolytic T cells and target cells that may be below threshold for induction of target-cell death, but above threshold for induction of transcriptional activation. The fact that induction of chemokine expression could occur in a target cell expressing MHC/peptide complexes that have been engaged by a TCR of a cytolytic CD8+ T cell suggests that the perforin/granzyme system may sometimes be quantitatively or qualitatively insufficient to induce cytolysis (at least with respect to alveolar epithelial-derived cells, which are rather insensitive to perforin-mediated cytolysis), or that other factors may influence target-cell susceptibility to cytotoxicity. Indeed, we have in vivo evidence that perforin plays a minimal role in the lung injury-mediated adoptive transfer of wild-type CD8+ T cells, the primary mediator being TNF- (Enelow and coworkers, unpublished observation). Further, the threshold for induction of chemokine expression by transmembrane TNF- may be lower than the threshold for induction of apoptosis. The identification of cellular regulators of protection against perforin/granzyme-mediated cytotoxicity (29) and TNF-mediated apoptosis (43) suggest potential mechanisms by which regulation of in vivo susceptibility of target cells to T cell-mediated cytolysis might occur.

It is likely that multiple mechanisms account for the activation of epithelial cells as a population in response to specific CD8+ T-cell recognition, the net result of which is the production of chemokines by the alveolar cells, which may in turn amplify inflammatory responses in the lung. In our adoptive transfer model, this may contribute to the recruitment of host inflammatory cells, particularly macrophages, into the lung parenchyma, the infiltration of which strongly correlates with the physiologic dysfunction associated with the interstitial pneumonitis, i.e., restrictive mechanics and diminished diffusing capacity (6). The data in this study suggest that alveolar epithelial cells may be active participants in the inflammation and injury associated with CD8+ T-cell recognition of alveolar antigens in the lung, regardless of the influence of virus infection. Further, apoptotic target-cell death is not a necessary outcome of CD8+ cytolytic T-cell antigen recognition, which may instead (or in addition) lead to inflammatory activation of the cell being recognized.