Sensitized CD8+ T Cells Fail To Control Organism Burden but Accelerate
http://www.100md.com
《感染与免疫杂志》
Departments of Pediatrics Microbiology and Immunology, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642
ABSTRACT
While CD8+ cells have been shown to contribute to lung injury during Pneumocystis carinii pneumonia (PCP), there are conflicting reports concerning the ability of CD8+ cells to kill P. carinii. To address these two issues, we studied the effect of the presence of CD8+ cells in two mouse models of PCP. In the reconstituted SCID mouse model, depletion of CD8+ cells in addition to CD4+ cells after reconstitution did not result in increased numbers of P. carinii cysts compared to the numbers of cysts in mice with only CD4+ cells depleted. This result was observed regardless of whether the mice were reconstituted with nave or P. carinii-sensitized lymphocytes. In contrast, reconstitution with sensitized lymphocytes resulted in more rapid onset of lung injury that was dependent on the presence of CD8+ cells. The course of organism replication over a 6-week period was also examined in the CD4+-T-cell-depleted and CD4+- and CD8+-T-cell-depleted mouse model of PCP. Again, the organism burdens were identical at all times regardless of whether CD8+ cells were present. Thus, in the absence of CD4+ T cells, CD8+ T cells are a key contributor to the inflammatory lung injury associated with PCP. However, we were unable to demonstrate an in vivo effect of these cells on the course of P. carinii infection.
INTRODUCTION
Efficient control of Pneumocystis carinii infection requires normally functioning CD4+ T cells (9, 12). Although CD8+ T cells have been reported to be able to produce a modest decrease in the level of P. carinii infection in the absence of CD4+ T cells (2), the mechanism for this effect is not obvious since P. carinii is, as far as we know, an extracellular pathogen. Our observations have failed to demonstrate any killing of P. carinii by CD8+ T cells (16; unpublished observations). On the other hand, there is clear evidence that CD8+ T cells make a biologically significant contribution to the immune-mediated inflammatory insult to the lung during P. carinii pneumonia (PCP) (16, 17). Therefore, defining the contribution of CD8+ T cells to the control of infection compared to the exacerbation of lung injury is important in developing improved adjunctive therapies for PCP targeted at suppressing CD8+ T cells.
To more precisely define the role of CD8+ T cells in the control of P. carinii f. sp. muris, we carried out a series of experiments with two mouse models of PCP. Using the immune reconstituted SCID mouse model, we hypothesized that if CD8+ T cells can in fact kill P. carinii, then exposure (i.e., sensitization) of CD8+ T cells to P. carinii by immunization before they are infused into P. carinii-infected SCID mice should result in enhanced clearance of organisms. Using donor splenocytes from P. carinii-immunized mice also allowed us to further examine the effect of CD8+ T cells on immune-mediated inflammatory injury during PCP. With regard to the latter point, we hypothesized that if CD8+ T cells play a key role in initiating the inflammatory response during PCP, then CD8+ T cells from immunized donor mice should produce a more rapid immune response and onset of lung injury.
Because the SCID mouse model of PCP involves immune reconstitution and resolution of disease, we also wanted to examine the effect of CD8+ T cells in the CD4+-T-cell-depleted model of PCP. This is a model of disease progression rather than disease resolution; thus, it allowed us to monitor the progression of P. carinii infection over time in the presence and absence of CD8+ T cells. The results obtained with these two models of PCP provided additional details concerning the immunopathogenesis of PCP.
MATERIALS AND METHODS
Mouse models of PCP. Two models of PCP were used for the experiments described here: the SCID mouse model and the CD4+-T-cell-depleted mouse model. CB.17 scid/scid mice were obtained from Taconic Laboratories (Germantown, NY). The mice were maintained in microisolator cages and fed sterilized food and water. P. carinii f. sp. muris infection was established by either cohousing or intranasal inoculation. Animals were infected by cohousing by placing P. carinii-free SCID mice in cages with P. carinii-infected SCID mice for 5 weeks. For infection by intranasal inoculation, P. carinii cysts were isolated as described by Wang et al. (15), with the following modifications. Lungs were removed without perfusion, and the DNase step was not performed. Additionally, the length of incubation on antibody-coated plates was decreased from 1.5 h to 30 min, and organisms were ultimately resuspended in sterile saline instead of serum-free medium. SCID mice that were 4 to 5 weeks old were anesthetized with halothane gas and given 1.0 x 105 P. carinii organisms in 50 μl of sterile saline using a 100-μl pipette tip. The mice were observed until they recovered from anesthesia. Each mouse received gentamicin at a dose of 5 μg per g body weight given subcutaneously with a 27-gauge needle on the day of inoculation and again 24 h later to minimize any secondary infections. The infection was allowed to progress for at least 3 to 4 weeks, at which point the mice had a well-established infection. The mice were then immunologically reconstituted by intraperitoneal (i.p.) injection of 5 x 107 splenocytes (16, 17) from either P. carinii-immunized or nave congenic CB.17 mice. The mice then received anti-CD4 monoclonal antibody (MAb) (clone GK 1.5; ATCC), both anti-CD4 and anti-CD8 MAb (clone TIB210; ATCC), or no antibody, as previously described (9). This antibody treatment resulted in 98 to 99% reductions in the numbers of CD4+ or CD8+ T cells, as determined by analysis of cells in bronchoalveolar lavage fluid. At various times the respiratory rate and dynamic lung compliance (CDYN) were measured. The lungs were removed after measurement of CDYN and homogenized, and aliquots were used for real-time PCR and staining with ammoniacal silver nitrate to confirm and quantify the P. carinii burden.
CB.17 mice that were 4 to 5 weeks old obtained from Taconic Laboratories were used for the CD4+-T-cell-depleted model of PCP. The mice were maintained in microisolator cages and fed sterilized food and water. After 4 to 5 days to acclimatize, the mice were divided into two groups and received either anti-CD4 MAb or both anti-CD4 and anti-CD8 MAb. Two hundred fifty micrograms of each antibody was given by i.p. injection twice a week to maintain depletion of the desired lymphocytes throughout the experiment. Flow cytometry with splenocytes demonstrated that there was 99% or greater depletion of CD4+ cells and 97% or greater depletion of CD8+ T cells when this regimen was used. After 1 week of antibody injections mice were infected with P. carinii via intranasal inoculation with 1 x 105 cysts per mouse as described above. At 3, 4, and 6 weeks after infection the mice were sacrificed by using a lethal dose of sodium pentobarbital administered by i.p. injection. Lungs were removed, snap frozen in liquid nitrogen, and stored at –70°C until they were used.
Immunization of donor mice with P. carinii. CB.17 mice were maintained in microisolator cages with sterilized food and water. Mice received 250-μl i.p. injections of either whole P. carinii lung homogenate containing approximately 5 x 106 P. carinii cysts/ml or sterile saline (7). The injection procedure was repeated at 2 weeks and 6 weeks after the initial immunization. Serum was obtained for an enzyme-linked immunosorbent assay (ELISA) to confirm the presence or absence of P. carinii-specific antibody (8). One week after the third immunization the mice were sacrificed by using a lethal dose of sodium pentobarbital, and splenocytes were harvested from each group, pooled, and used to reconstitute the P. carinii-infected SCID mice. Nonimmunized mice served as donors of P. carinii-nave splenocytes. The reconstituted SCID mice also were assayed for a serum antibody response to P. carinii on the day of sacrifice.
Physiologic assessment of pulmonary compliance. Pulmonary compliance in live mice was measured using a method described previously, with modifications (3, 16). Mice were anesthetized with 0.13 mg of sodium pentobarbital/g of body weight by i.p. injection. Mice were then surgically cannulated through the trachea with an 18-gauge cannula advanced 3 mm into an anterior incision in the exposed trachea. The thorax was opened to equalize airway and transpulmonary pressure, and the mice were immediately placed on a Harvard rodent ventilator (Harvard Apparatus, South Natick, Mass.) at a respiratory rate of 150 strokes per min. Each animal was placed in a pressure plethysmograph and ventilated at 2.5 Hz with a tidal volume of 0.01 ml per g of body weight. Signals for airway pressure and volume were passed through an analogue-to-digital converter and used to calculate pulmonary compliance using the method of Amdur and Mead (1). Compliance was normalized for body weight.
Enumeration of P. carinii burden. The intensity of infection in mouse lungs was determined by real-time PCR for all experiments and by both PCR and P. carinii cyst counting in the CD4+-T-cell-depleted mouse model experiment. For real-time PCR crude lung homogenates were boiled for 15 min and centrifuged at 13,000 x g for 15 min, and the supernatants were either used immediately or stored at –70°C. A 2.5-μl portion of infected mouse lung genomic DNA, diluted 1:3 to minimize any PCR inhibition, was assayed by quantitative PCR using TaqMan primer-fluorogenic probe chemistry (Applied Biosystems, Foster City, CA). The primer-probe set used is specific for a 96-bp region of the P. carinii KEX1 gene (GenBank accession number AF093132) (4). The primer and probe were designed using the Primer Express software, version 2.0.0 (Applied Biosystems). The forward primer (5'-3') was GCACGCATTTATACTACGGATGTT (sequence positions 1192 to 1215) (4), and the reverse primer (3'-5') was GAGCTATAACGCCTGCTGCAA (sequence positions 1268 to 1288). The kexin probe was CAGCACTGTACATTCTGGATCTTCTGCTTCC (sequence positions 1230 to 1260). The real-time PCR mixtures (total volume, 25 μl) consisted of 2x TaqMan universal PCR master mixture (Applied Biosystems), 900 nM forward primer, 900 nM reverse primer, 150 nM Taqman probe, and 2.5 μl DNA template. To generate a standard curve for the assay, a section of the mouse P. carinii kexin gene was subcloned into the pRSET B plasmid (Invitrogen Corp., Carlsbad, CA). The copy number of the plasmid vector was calculated from the DNA concentration determined by A260 spectrophotometric measurement. Real-time PCR quantitation of the organism burden was performed by using the ABI Prism 7000 sequence detection system and its associated SDS software (version 1.0; Applied Biosystems) and extrapolating the amplification curve threshold cycle against the threshold cycles of a standard curve constructed by using serial 10-fold dilutions for predetermined copy numbers of the pRSETB:kexin vector. The thermocycler profile used with the ABI Prism 7000 system was 50°C for 2 min and 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The method used for specimen processing and determining the gene target sequence was evaluated in our laboratory and was found to give results that closely approximated the numbers of P. carinii cysts determined by conventional staining techniques. While the PCR results were checked by microscopy, the P. carinii counts given below are the counts obtained by PCR.
For microscopy P. carinii cysts were visualized and counted using an ammoniacal silver nitrate staining method. Ten-microliter portions of various dilutions of raw P. carinii-infected lung homogenate were spotted on glass microscope slides, air dried, and heat fixed. The slides were then sequentially submerged in 4% periodic acid, a preheated ammoniacal silver nitrate solution, 0.2% gold chloride, 2% sodium thiosulfate, and a fast green solution, with water washes between the different solutions. Finally, the slides were dehydrated with a graded ethanol series and then subjected to multiple changes of xylene. Coverslips were applied, and the cysts in adjacent fields were counted with a 100x oil immersion objective; the results were expressed as the number of P. carinii organisms per mouse lung homogenate.
Statistical analysis. Results were analyzed for statistical significance by analysis of varaince or Student's t test using the SigmaStat computer software.
RESULTS
CD8+ T cells do not enhance P. carinii clearance in the reconstituted SCID mouse model of PCP. P. carinii-infected SCID mice were immunologically reconstituted with nave splenocytes from congenic donor mice and then treated with monoclonal antibody to deplete CD4+ T cells or both CD4+ and CD8+ T cells. Mice were sacrificed at different times, and the P. carinii burden was measured by real-time PCR. This general procedure was repeated three times.
In the first experiment either mice were fully reconstituted or mice were reconstituted and then only CD4+ T cells were depleted. The results of these experiments are shown in Table 1. By day 21 after reconstitution there was a small increase in the number of P. carinii organisms in the mice in which CD4+ T cells were depleted compared to mice in which CD4+ T cells were not depleted (Table 1, experiment 1). The fact that P. carinii was not cleared by day 21 after reconstitution in the fully reconstituted mice differs from our previous results (16) and likely reflects the increased sensitivity of PCR compared to the sensitivity of microscopy and the slower onset of immune reconstitution when splenocytes are injected i.p., which were two differences from our previous study.
For the following two experiments we added a group in which CD4+ T cells and CD8+ T cells were both depleted with monoclonal antibodies. In this way, we could more clearly examine the contribution of the CD8+-T-cell population to the control of the organism by comparing the group with only CD4+ T cells depleted and the group with both CD4+ T cells and CD8+ T cells depleted. As shown in Table 1, at 7, 10 and 14 days after reconstitution, CD4+-T-cell-depleted mice and mice with CD4+ T cells and CD8+ T cells depleted had similar P. carinii burdens. There were no significant differences between the two groups at any time (Table 1), indicating that the additional depletion of CD8+ T cells did not allow increased replication of P. carinii.
Sensitization of CD8+ T cells prior to transfer into P. carinii-infected SCID mice has no effect on their capacity to control P. carinii. To determine whether sensitization of CD8+ T cells to P. carinii before they were infused into P. carinii-infected SCID mice enhanced the capacity of the CD8+ T cells to control the organism, we reconstituted infected SCID mice with splenocytes from mice immunized with P. carinii. To ensure that splenocytes were truly sensitized, serum was collected from immunized donor mice and assayed for antibody to P. carinii by ELISA. Nave donor mice had no detectable anti-P. carinii antibody. The optical densities in these mice were consistent with the results for negative controls, and the values were less than 0.10 at 15 and 30 min. Sensitized donor mice had significantly higher optical densities, with values of 0.6 at 15 min and 0.9 at 30 min. Recipient mice were also tested to determine their antibody response after they received either nave or sensitized splenocytes. The P. carinii-infected mice reconstituted with sensitized splenocytes exhibited an accelerated immune response with markedly higher ELISA values than the mice that received nave splenocytes (Fig. 1). Furthermore, this accelerated antibody response was dependent on the presence of CD4+ T cells (Fig. 1).
After we demonstrated that reconstitution of P. carinii-infected SCID mice with splenocytes from P. carinii-immunized donor mice resulted in an increased immune response, P. carinii-infected SCID mice were reconstituted with sensitized splenocytes and then CD4+ T cells or both CD4+ and CD8+ T cells were depleted. Consistent with the finding that there was an increased antibody response, there was a trend toward more rapid clearing of P. carinii in the mice reconstituted with P. carinii-sensitized splenocytes than in the mice reconstituted with nave cells by day 21 in experiment 1 (3.35 x 106 organisms/mouse versus 1.58 x 107 organisms/mouse; P = 0.31) (Tables 1 and 2) and by day 14 in experiment 2 (8.1 x 105 organisms/mouse versus 2.3 x 106 organisms/mouse; P = 0.12). In contrast to the results obtained with nave splenocytes, when P. carinii-infected SCID mice were reconstituted with sensitized splenocytes and CD4+ T cells were subsequently depleted, the expected blunted clearance of P. carinii was evident by day 14 after reconstitution (Table 2, experiment 2) (8.1 x 105 P. carinii organisms/mouse versus 2.1 x 106 P. carinii organisms/mouse; P = 0.04). However, as observed with the mice reconstituted with nave splenocytes, also depleting the CD8+ T cells in the mice had no effect on the number of P. carinii organisms in the mice compared to the number of organisms in mice in which only CD4+ cells were depleted (Table 2). In fact, at 14 days in experiment 2, the number of P. carinii organisms was actually higher in mice with CD8+ T cells than in mice in which CD8+ T cells were depleted (P = 0.04). This increase in the number P. carinii organisms could not be explained by differences in antibody levels, since both CD4+-T-cell-depleted mice and CD4+- and CD8+-T-cell-depleted mice had no detectable antibody to P. carinii (the ELISA optical density was 0.06 for both groups). The total number of inflammatory cells in bronchoalveolar lavage fluid was higher in CD4+-T-cell-depleted mice (3.4 x 105 ± 3.9 x 104 cells) than in CD4+- and CD8+ -T-cell-depleted mice (9.7 x 104 ± 1.7 x 104 cells; P = 0.05). The levels of tumor necrosis factor alpha and gamma interferon were also higher in the CD4+-T-cell-depleted mice than in the CD4+- and CD8+-T-cell-depleted mice (74 ± 48 pg/ml versus 39 ± 19 pg/ml and 60 ± 2 pg/ml versus 22 ± 22 pg/ml, respectively), but the differences were not statistically significant. Thus, despite documenting that there was an accelerated immune response with enhanced clearance of P. carinii in the mice fully reconstituted with P. carinii-sensitized splenocytes, after depletion of CD4+ T cells we could not document any difference in P. carinii replication between when CD8+ T cells were present and when CD8+ T cells were absent.
CD8+ T cells do not enhance the clearance of P. carinii in the CD4+-T-cell-depleted mouse model of PCP. Because in the experiments described above we used a reconstitution model, we wanted to examine the effect of CD8+ T cells over a longer time using a model of disease progression. Thus, as previously described (2, 16), we depleted CD4+ T cells or both CD4+ and CD8+ T cells in normal CB.17 mice and then infected the mice with P. carinii. Mice were sacrificed at 3, 4, and 6 weeks after infection with P. carinii to ensure that any contribution of CD8+ T cells was not missed during the course of infection. As shown in Table 3, the P. carinii counts, as determined by real-time PCR, in the CD4+-T-cell-depleted group were the same as or higher than the counts in the CD4+- and CD8+- T-cell-depleted mice at each time. The PCR results were confirmed by silver staining of P. carinii organisms, which also showed that there was no difference in the number of organisms at any of the times. Therefore, we were not able to demonstrate any significant contribution of CD8+ T cells to the control of the organism using the CD4+-T-cell-depleted model of PCP.
Effect of prior exposure to P. carinii on CD8+-T-cell-mediated lung injury. Using the SCID mouse model of PCP, we also looked at the effect of immunologic reconstitution with sensitized splenocytes on CDYN as a marker of impaired pulmonary physiology. Mice reconstituted with nave splenocytes were compared to mice reconstituted with sensitized splenocytes. To isolate the previously described effect of CD8+ T cells on lung injury during PCP, CD4+ cells were depleted from both groups. Despite similar organism burdens, mice with sensitized CD8+ T cells exhibited earlier onset of lung injury, by day 10 after reconstitution, as shown by a CDYN (1.47) that was significantly lower than the CDYN of the mice with P. carinii-nave CD8+ T cells (CDYN, 1.86; P = 0.004) (Fig. 2 and Table 4, experiment 1). However, over the next 11 days, the CDYN declined more gradually and the organism burden did not increase in mice reconstituted with sensitized splenocytes. In contrast, in the mice reconstituted with nave splenocytes, the CDYN fell more rapidly and the organism burden continued to rise, as would be expected in the absence of CD4+ T cells (Fig. 2 and Tables 1 and 2, experiment 1). We hypothesized that the slower fall in lung compliance was due to antibody-mediated control of the organism that may have limited the CD8+ T-cell inflammatory lung damage in the mice reconstituted with sensitized splenocytes (6).
When mice were reconstituted at a lower organism burden, the accelerated lung injury in the presence of sensitized CD8+ T cells was not evident (Table 4, experiment 2). However, this effect appeared to depend on the overall organism burden as a third experiment using mice with an organism burden of 107 organisms per mouse showed that there was a 26% drop in CDYN by day 10 after reconstitution in CD4+-T-cell-depleted mice reconstituted with sensitized splenocytes compared to mice reconstituted with nave splenocytes (Table 4, experiment 3) (P = 0.003).
Exacerbated lung injury in mice reconstituted with sensitized splenocytes requires CD8+ T cells. As noted above and shown in Fig. 3A, reconstituted and CD4+-T-cell-depleted P. carinii-infected SCID mice exhibited a significant (26%; P = 0.003) decrease in CDYN if they were reconstituted with P. carinii-sensitized splenocytes rather than nave splenocytes. However, the difference in early lung injury disappeared if the CD8+ cells in mice were also depleted (Fig. 3B), indicating that the presence of P. carinii-sensitized CD8+ T cells is a critical component of the accelerated onset of inflammatory lung injury in this model.
DISCUSSION
By reconstituting P. carinii-infected SCID mice with splenocytes that were presensitized to P. carinii, we were able to document an accelerated immune response to P. carinii in the recipient mice. However, in this model the presence of sensitized CD8+ T cells, in the absence of CD4+ T cells, had no effect on the organism burden but did result in accelerated onset of lung injury. These results confirm and extend our previous observations that CD8+ T cells play little or no role in the control of P. carinii but do contribute substantially to the inflammatory lung injury associated with PCP (16). Our inability to demonstrate that CD8+ T cells have an effect on the control of P. carinii, which is an extracellular pathogen, is consistent with the classical function of CD8+ T cells in controlling intracellular pathogens, such as viruses and some intracellular bacteria, rather than extracellular microorganisms.
We used two models of PCP to examine the effect of CD8+ T cells on P. carinii. Using the reconstituted SCID mouse model, we were able to show that, as expected, presensitized CD4+ T cells had a more pronounced effect on control of organism replication than nave CD4+ T cells had. However, despite demonstrating that reconstitution of infected mice with sensitized cells enhanced the known function of CD4+ T cells in protecting against P. carinii (9, 12, 13), we were unable to show any similar effect of CD8+ T cells. In fact, although the data were not statistically significant, with the exception of a single time (day 14) (Table 2, experiment 2), CD4+ T-cell-depleted mice with CD8+ T cells present, especially mice reconstituted with sensitized splenocytes, often contained higher numbers of P. carinii organisms, which is consistent with previous observations that the numbers of P. carinii organisms seem to be higher when there is more inflammation (5, 16). To confirm our observation with the SCID mouse model, we used the CD4+-T-cell-depleted model of PCP that enabled us to observe the effect of CD8+ T cells over a longer time, as PCP got progressively worse. We previously demonstrated that CD8+ T cells had no effect at a single time in this model (16). In this experiment we analyzed the P. carinii burden at 3, 4, and 6 weeks after inoculation of CD4+-T-cell-depleted mice. Using quantitative PCR and silver staining of organisms, we were not able to demonstrate any difference between the number of P. carinii organisms in CD4+-T-cell-depleted mice (CD8+ T cells present) and the number of P. carinii organisms in CD4+- and CD8+-T-cell-depleted mice at any time.
Our results differ from those of Beck et al., who showed that there was a modest reduction in the P. carinii burden in CD4+- T-cell-depleted mice compared to the burden in CD4+- and CD8+-T-cell-depleted mice (2). While the general experimental designs used by us and by Beck et al. were similar, Beck et al. performed a semiquantitative analysis based on silver staining of lung sections from mice at a single time. Thus, assaying cysts only in single lung sections may not have resulted in accurate representation of the overall organism burden in the entire lung. In contrast, we analyzed CD4+-T-cell-depleted mice at several times during progression of the disease. We also assayed lung homogenates using three methods of enumeration, Diff Quik staining, silver staining (data not shown), and real-time PCR, all of which produced consistent results. Finally, we corroborated our observations with the CD4+-T-cell-depleted mouse model using the reconstituted SCID mouse model of PCP.
Our results may also appear to conflict with several recent reports that gamma interferon stimulation of CD8+ T cells can protect CD4+-T-cell-depleted mice from development of PCP (10, 11, 14). However, these interesting studies of the effect of gamma interferon addressed a different issue about the biology of CD8+ T cells and the pathogenesis of PCP than our studies addressed. In the gamma interferon studies the workers used mice that were challenged with P. carinii 24 h after transfection of lung cells with a gamma interferon gene (10, 14) or 24 h before infusion of CD8+ T cells isolated from mice that had been transfected with the gamma interferon gene (11). In these studies, gamma interferon-stimulated CD8+ T cells protected against the development of PCP rather than controlled an established infection. In our experiments we examined the effect of CD8+ T cells both on the development of PCP (CD4+-T-cell-depleted model) and on eradication of an established infection (reconstituted SCID mouse model of PCP). Our interpretation of these two different experimental approaches is that our studies demonstrated that in the usual or "physiologic" response to infection with P. carinii, CD8+ T cells have little importance in the control of P. carinii; however, after pharmacologic stimulation with gamma interferon, CD8+ T cells can be made to have a biologic effect on the control of P. carinii that they normally do not express.
These studies also extended our previous experiments suggesting that CD8+ T cells are critical to the lung injury associated with PCP (16-18). For the current experiments, we hypothesized that if CD8+ T cells were indeed a critical component of the P. carinii-driven inflammatory response, then presensitized CD8+ T cells should accelerate the inflammatory response. This is in fact what we observed, since when P. carinii-infected SCID mice were reconstituted with P. carinii-sensitized splenocytes, there was a statistically significant more rapid onset of lung injury when subsequently CD4+ cells but not CD8+cells were depleted. This effect was lost after CD8+ T cells were also depleted from the reconstituted mice, confirming the role of these cells in exacerbating lung injury during PCP.
We previously demonstrated that P. carinii, in the absence of an adaptive immune response, did not cause much lung injury until relatively late in the course of the illness (16). However, the present study showed that there is a "threshold" effect of P. carinii on the induction of the CD8+-T-cell-mediated response since with lower numbers of P. carinii organisms (105 versus 107 organisms) the acceleration of lung injury by presensitized CD8+ T cells could not be demonstrated.
In summary, we concluded that CD8+ T cells play an insignificant role in control of P. carinii replication but are a key component of the inflammatory response and consequent lung injury induced by P. carinii in a susceptible host. Therefore, development of treatment strategies to blunt or eliminate the CD8+ T-cell-mediated inflammatory response to P. carinii, combined with antibiotic treatment of P. carinii, should result in improved lung function without an adverse impact on clearance of P. carinii.
ACKNOWLEDGMENTS
We acknowledge the expert technical assistance of Stephanie Campbell and Margaret Chovaniec.
This work was supported by grant P01 HL071659 from the National Institutes of Health.
FOOTNOTES
Corresponding author. Mailing address: University of Rochester School of Medicine & Dentistry, 601 Elmwood Avenue, Box 690, Rochester, NY 14642. Phone: (585) 275-5944. Fax: (585) 273-1104. E-mail: francis_gigliotti@urmc.rochester.edu.
Published ahead of print on 28 August 2006.
F.G. and E.L.C. contributed equally to the paper.
REFERENCES
1. Amdur, M. O., and J. Mead. 1958. Mechanics of respiration in unanesthetized guinea pigs. Am. J. Physiol. 192:364.
2. Beck, J. M., R. L. Newbury, B. E. Palmer, M. L. Warnock, P. K. Byrd, and H. B. Kaltreider. 1996. Role of CD8+ lymphocytes in host defense against Pneumocystis carinii in mice. J. Lab. Clin. Med. 128:477-487.
3. Bergmann, K. C., B. Lachmann, and K. Noack. 1984. Lung mechanics in orally immunized mice after aerolized exposure to influenza virus. Respiration 46:218-221.
4. Lee, L. H., F. Gigliotti, T. W. Wright, P. J. Simpson-Haidaris, G. A. Weinberg, and C. G. Haidaris. 2000. Molecular characterization of KEX1, a kexin-like protease in mouse Pneumocystis carinii. Gene 242:141-150.
5. Chen, W., E. A. Havell, F. Gigliotti, and A. G. Harmsen. 1993. Interleukin-6 production in a murine model of Pneumocystis carinii pneumonia: relation to resistance and inflammatory response. Infect. Immun. 61:97-102.
6. Empey, K. M., M. Hollifield, K. Schuer, F. Gigliotti, and B. Garvy. 2004. Passive immunization of neonatal mice against Pneumocystis carinii f. sp. muris enhances control of infection without stimulating inflammation. Infect. Immun. 72:6211-6220.
7. Gigliotti, F., and A. G. Harmsen. 1997. Pneumocystis carinii host origin defines the antibody specificity and protective response induced by immunization. J. Infect. Dis. 176:1322-1326.
8. Gigliotti, F., and T. McCool. 1996. Glycoprotein A is the immunodominant antigen of Pneumocystis carinii in mice following immunization. Parasitol. Res. 82:90-91.
9. Harmsen, A. G., and M. Stankiewicz. 1990. Requirement for CD4+ cells in resistance to Pneumocystis carinii pneumonia in mice. J. Exp. Med. 172:937-945.
10. Kolls, J. K., S. Habetz, M. K. Shean, C. Vazquez, J. A. Brown, D. Lei, P. Schwarzenberger, P. Ye, S. Nelson, W. R. Summer, and J. E. Shellito. 1999. IFN-gamma and CD8+ T cells restore host defenses against Pneumocystis carinii in mice depleted of CD4+ T cells. J. Immunol. 162:2890-2894.
11. McAllister, F., C. Steele, M. Zheng, E. Young, J. E. Shellito, L. Marrero, and J. K. Kolls. 2004. T cytotoxic-1 CD8+ T cells are effector cells against Pneumocystis in mice. J. Immunol. 172:1132-1138.
12. Roths, J. B., and C. L. Sidman. 1992. Both immunity and hyperresponsiveness to Pneumocystis carinii result from transfer of CD4+ but not CD8+ T cells into severe combined immunodeficiency mice. J. Clin. Investig. 90:673-678.
13. Shellito, J., V. V. Suzara, W. Blumenfeld, J. M. Beck, H. J. Steger, and T. H. Ermak. 1990. A new model of Pneumocystis carinii infection in mice selectively depleted of helper T lymphocytes. J. Clin. Investig. 85:1686-1693.
14. Steele, C., M. Zheng, E. Young, L. Marrero, J. E. Shellito, and J. K. Kolls. 2002. Increased host resistance against Pneumocystis carinii pneumonia in T-cell-deficient mice: protective role of gamma interferon and CD8+ T cells. Infect. Immun. 70:5208-5215.
15. Wang, J., F. Gigliotti, S. Maggirwar, C. Johnston, J. N. Finkelstein, and T. W. Wright. 2005. Pneumocystis activates the NF-B signaling pathway in alveolar epithelial cells. Infect. Immun. 73:2766-2777.
16. Wright, T. W., F. Gigliotti, J. N. Finkelstein, J. T. McBride, C. L. An, and A. G. Harmsen. 1999. Immune-mediated inflammation directly impairs pulmonary function, contributing to the pathogenesis of Pneumocystis carinii pneumonia. J. Clin. Investig. 104:1307-1317.
17. Wright, T. W., R. H. Notter, Z. Wang, A. G. Harmsen, and F. Gigliotti. 2001. Pulmonary inflammation disrupts surfactant function during Pneumocystis carinii pneumonia. Infect. Immun. 69:758-764.
18. Wright, T. W., G. S. Pryhuber, P. R. Chess, Z. Wang, R. H. Notter, and F. Gigliotti. 2004. TNF receptor signaling contributes to chemokine secretion, inflammation, and respiratory deficits during Pneumocystis pneumonia. J. Immunol. 172:2511-2521.(Francis Gigliotti, Elliott L. Crow, Sami)
ABSTRACT
While CD8+ cells have been shown to contribute to lung injury during Pneumocystis carinii pneumonia (PCP), there are conflicting reports concerning the ability of CD8+ cells to kill P. carinii. To address these two issues, we studied the effect of the presence of CD8+ cells in two mouse models of PCP. In the reconstituted SCID mouse model, depletion of CD8+ cells in addition to CD4+ cells after reconstitution did not result in increased numbers of P. carinii cysts compared to the numbers of cysts in mice with only CD4+ cells depleted. This result was observed regardless of whether the mice were reconstituted with nave or P. carinii-sensitized lymphocytes. In contrast, reconstitution with sensitized lymphocytes resulted in more rapid onset of lung injury that was dependent on the presence of CD8+ cells. The course of organism replication over a 6-week period was also examined in the CD4+-T-cell-depleted and CD4+- and CD8+-T-cell-depleted mouse model of PCP. Again, the organism burdens were identical at all times regardless of whether CD8+ cells were present. Thus, in the absence of CD4+ T cells, CD8+ T cells are a key contributor to the inflammatory lung injury associated with PCP. However, we were unable to demonstrate an in vivo effect of these cells on the course of P. carinii infection.
INTRODUCTION
Efficient control of Pneumocystis carinii infection requires normally functioning CD4+ T cells (9, 12). Although CD8+ T cells have been reported to be able to produce a modest decrease in the level of P. carinii infection in the absence of CD4+ T cells (2), the mechanism for this effect is not obvious since P. carinii is, as far as we know, an extracellular pathogen. Our observations have failed to demonstrate any killing of P. carinii by CD8+ T cells (16; unpublished observations). On the other hand, there is clear evidence that CD8+ T cells make a biologically significant contribution to the immune-mediated inflammatory insult to the lung during P. carinii pneumonia (PCP) (16, 17). Therefore, defining the contribution of CD8+ T cells to the control of infection compared to the exacerbation of lung injury is important in developing improved adjunctive therapies for PCP targeted at suppressing CD8+ T cells.
To more precisely define the role of CD8+ T cells in the control of P. carinii f. sp. muris, we carried out a series of experiments with two mouse models of PCP. Using the immune reconstituted SCID mouse model, we hypothesized that if CD8+ T cells can in fact kill P. carinii, then exposure (i.e., sensitization) of CD8+ T cells to P. carinii by immunization before they are infused into P. carinii-infected SCID mice should result in enhanced clearance of organisms. Using donor splenocytes from P. carinii-immunized mice also allowed us to further examine the effect of CD8+ T cells on immune-mediated inflammatory injury during PCP. With regard to the latter point, we hypothesized that if CD8+ T cells play a key role in initiating the inflammatory response during PCP, then CD8+ T cells from immunized donor mice should produce a more rapid immune response and onset of lung injury.
Because the SCID mouse model of PCP involves immune reconstitution and resolution of disease, we also wanted to examine the effect of CD8+ T cells in the CD4+-T-cell-depleted model of PCP. This is a model of disease progression rather than disease resolution; thus, it allowed us to monitor the progression of P. carinii infection over time in the presence and absence of CD8+ T cells. The results obtained with these two models of PCP provided additional details concerning the immunopathogenesis of PCP.
MATERIALS AND METHODS
Mouse models of PCP. Two models of PCP were used for the experiments described here: the SCID mouse model and the CD4+-T-cell-depleted mouse model. CB.17 scid/scid mice were obtained from Taconic Laboratories (Germantown, NY). The mice were maintained in microisolator cages and fed sterilized food and water. P. carinii f. sp. muris infection was established by either cohousing or intranasal inoculation. Animals were infected by cohousing by placing P. carinii-free SCID mice in cages with P. carinii-infected SCID mice for 5 weeks. For infection by intranasal inoculation, P. carinii cysts were isolated as described by Wang et al. (15), with the following modifications. Lungs were removed without perfusion, and the DNase step was not performed. Additionally, the length of incubation on antibody-coated plates was decreased from 1.5 h to 30 min, and organisms were ultimately resuspended in sterile saline instead of serum-free medium. SCID mice that were 4 to 5 weeks old were anesthetized with halothane gas and given 1.0 x 105 P. carinii organisms in 50 μl of sterile saline using a 100-μl pipette tip. The mice were observed until they recovered from anesthesia. Each mouse received gentamicin at a dose of 5 μg per g body weight given subcutaneously with a 27-gauge needle on the day of inoculation and again 24 h later to minimize any secondary infections. The infection was allowed to progress for at least 3 to 4 weeks, at which point the mice had a well-established infection. The mice were then immunologically reconstituted by intraperitoneal (i.p.) injection of 5 x 107 splenocytes (16, 17) from either P. carinii-immunized or nave congenic CB.17 mice. The mice then received anti-CD4 monoclonal antibody (MAb) (clone GK 1.5; ATCC), both anti-CD4 and anti-CD8 MAb (clone TIB210; ATCC), or no antibody, as previously described (9). This antibody treatment resulted in 98 to 99% reductions in the numbers of CD4+ or CD8+ T cells, as determined by analysis of cells in bronchoalveolar lavage fluid. At various times the respiratory rate and dynamic lung compliance (CDYN) were measured. The lungs were removed after measurement of CDYN and homogenized, and aliquots were used for real-time PCR and staining with ammoniacal silver nitrate to confirm and quantify the P. carinii burden.
CB.17 mice that were 4 to 5 weeks old obtained from Taconic Laboratories were used for the CD4+-T-cell-depleted model of PCP. The mice were maintained in microisolator cages and fed sterilized food and water. After 4 to 5 days to acclimatize, the mice were divided into two groups and received either anti-CD4 MAb or both anti-CD4 and anti-CD8 MAb. Two hundred fifty micrograms of each antibody was given by i.p. injection twice a week to maintain depletion of the desired lymphocytes throughout the experiment. Flow cytometry with splenocytes demonstrated that there was 99% or greater depletion of CD4+ cells and 97% or greater depletion of CD8+ T cells when this regimen was used. After 1 week of antibody injections mice were infected with P. carinii via intranasal inoculation with 1 x 105 cysts per mouse as described above. At 3, 4, and 6 weeks after infection the mice were sacrificed by using a lethal dose of sodium pentobarbital administered by i.p. injection. Lungs were removed, snap frozen in liquid nitrogen, and stored at –70°C until they were used.
Immunization of donor mice with P. carinii. CB.17 mice were maintained in microisolator cages with sterilized food and water. Mice received 250-μl i.p. injections of either whole P. carinii lung homogenate containing approximately 5 x 106 P. carinii cysts/ml or sterile saline (7). The injection procedure was repeated at 2 weeks and 6 weeks after the initial immunization. Serum was obtained for an enzyme-linked immunosorbent assay (ELISA) to confirm the presence or absence of P. carinii-specific antibody (8). One week after the third immunization the mice were sacrificed by using a lethal dose of sodium pentobarbital, and splenocytes were harvested from each group, pooled, and used to reconstitute the P. carinii-infected SCID mice. Nonimmunized mice served as donors of P. carinii-nave splenocytes. The reconstituted SCID mice also were assayed for a serum antibody response to P. carinii on the day of sacrifice.
Physiologic assessment of pulmonary compliance. Pulmonary compliance in live mice was measured using a method described previously, with modifications (3, 16). Mice were anesthetized with 0.13 mg of sodium pentobarbital/g of body weight by i.p. injection. Mice were then surgically cannulated through the trachea with an 18-gauge cannula advanced 3 mm into an anterior incision in the exposed trachea. The thorax was opened to equalize airway and transpulmonary pressure, and the mice were immediately placed on a Harvard rodent ventilator (Harvard Apparatus, South Natick, Mass.) at a respiratory rate of 150 strokes per min. Each animal was placed in a pressure plethysmograph and ventilated at 2.5 Hz with a tidal volume of 0.01 ml per g of body weight. Signals for airway pressure and volume were passed through an analogue-to-digital converter and used to calculate pulmonary compliance using the method of Amdur and Mead (1). Compliance was normalized for body weight.
Enumeration of P. carinii burden. The intensity of infection in mouse lungs was determined by real-time PCR for all experiments and by both PCR and P. carinii cyst counting in the CD4+-T-cell-depleted mouse model experiment. For real-time PCR crude lung homogenates were boiled for 15 min and centrifuged at 13,000 x g for 15 min, and the supernatants were either used immediately or stored at –70°C. A 2.5-μl portion of infected mouse lung genomic DNA, diluted 1:3 to minimize any PCR inhibition, was assayed by quantitative PCR using TaqMan primer-fluorogenic probe chemistry (Applied Biosystems, Foster City, CA). The primer-probe set used is specific for a 96-bp region of the P. carinii KEX1 gene (GenBank accession number AF093132) (4). The primer and probe were designed using the Primer Express software, version 2.0.0 (Applied Biosystems). The forward primer (5'-3') was GCACGCATTTATACTACGGATGTT (sequence positions 1192 to 1215) (4), and the reverse primer (3'-5') was GAGCTATAACGCCTGCTGCAA (sequence positions 1268 to 1288). The kexin probe was CAGCACTGTACATTCTGGATCTTCTGCTTCC (sequence positions 1230 to 1260). The real-time PCR mixtures (total volume, 25 μl) consisted of 2x TaqMan universal PCR master mixture (Applied Biosystems), 900 nM forward primer, 900 nM reverse primer, 150 nM Taqman probe, and 2.5 μl DNA template. To generate a standard curve for the assay, a section of the mouse P. carinii kexin gene was subcloned into the pRSET B plasmid (Invitrogen Corp., Carlsbad, CA). The copy number of the plasmid vector was calculated from the DNA concentration determined by A260 spectrophotometric measurement. Real-time PCR quantitation of the organism burden was performed by using the ABI Prism 7000 sequence detection system and its associated SDS software (version 1.0; Applied Biosystems) and extrapolating the amplification curve threshold cycle against the threshold cycles of a standard curve constructed by using serial 10-fold dilutions for predetermined copy numbers of the pRSETB:kexin vector. The thermocycler profile used with the ABI Prism 7000 system was 50°C for 2 min and 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The method used for specimen processing and determining the gene target sequence was evaluated in our laboratory and was found to give results that closely approximated the numbers of P. carinii cysts determined by conventional staining techniques. While the PCR results were checked by microscopy, the P. carinii counts given below are the counts obtained by PCR.
For microscopy P. carinii cysts were visualized and counted using an ammoniacal silver nitrate staining method. Ten-microliter portions of various dilutions of raw P. carinii-infected lung homogenate were spotted on glass microscope slides, air dried, and heat fixed. The slides were then sequentially submerged in 4% periodic acid, a preheated ammoniacal silver nitrate solution, 0.2% gold chloride, 2% sodium thiosulfate, and a fast green solution, with water washes between the different solutions. Finally, the slides were dehydrated with a graded ethanol series and then subjected to multiple changes of xylene. Coverslips were applied, and the cysts in adjacent fields were counted with a 100x oil immersion objective; the results were expressed as the number of P. carinii organisms per mouse lung homogenate.
Statistical analysis. Results were analyzed for statistical significance by analysis of varaince or Student's t test using the SigmaStat computer software.
RESULTS
CD8+ T cells do not enhance P. carinii clearance in the reconstituted SCID mouse model of PCP. P. carinii-infected SCID mice were immunologically reconstituted with nave splenocytes from congenic donor mice and then treated with monoclonal antibody to deplete CD4+ T cells or both CD4+ and CD8+ T cells. Mice were sacrificed at different times, and the P. carinii burden was measured by real-time PCR. This general procedure was repeated three times.
In the first experiment either mice were fully reconstituted or mice were reconstituted and then only CD4+ T cells were depleted. The results of these experiments are shown in Table 1. By day 21 after reconstitution there was a small increase in the number of P. carinii organisms in the mice in which CD4+ T cells were depleted compared to mice in which CD4+ T cells were not depleted (Table 1, experiment 1). The fact that P. carinii was not cleared by day 21 after reconstitution in the fully reconstituted mice differs from our previous results (16) and likely reflects the increased sensitivity of PCR compared to the sensitivity of microscopy and the slower onset of immune reconstitution when splenocytes are injected i.p., which were two differences from our previous study.
For the following two experiments we added a group in which CD4+ T cells and CD8+ T cells were both depleted with monoclonal antibodies. In this way, we could more clearly examine the contribution of the CD8+-T-cell population to the control of the organism by comparing the group with only CD4+ T cells depleted and the group with both CD4+ T cells and CD8+ T cells depleted. As shown in Table 1, at 7, 10 and 14 days after reconstitution, CD4+-T-cell-depleted mice and mice with CD4+ T cells and CD8+ T cells depleted had similar P. carinii burdens. There were no significant differences between the two groups at any time (Table 1), indicating that the additional depletion of CD8+ T cells did not allow increased replication of P. carinii.
Sensitization of CD8+ T cells prior to transfer into P. carinii-infected SCID mice has no effect on their capacity to control P. carinii. To determine whether sensitization of CD8+ T cells to P. carinii before they were infused into P. carinii-infected SCID mice enhanced the capacity of the CD8+ T cells to control the organism, we reconstituted infected SCID mice with splenocytes from mice immunized with P. carinii. To ensure that splenocytes were truly sensitized, serum was collected from immunized donor mice and assayed for antibody to P. carinii by ELISA. Nave donor mice had no detectable anti-P. carinii antibody. The optical densities in these mice were consistent with the results for negative controls, and the values were less than 0.10 at 15 and 30 min. Sensitized donor mice had significantly higher optical densities, with values of 0.6 at 15 min and 0.9 at 30 min. Recipient mice were also tested to determine their antibody response after they received either nave or sensitized splenocytes. The P. carinii-infected mice reconstituted with sensitized splenocytes exhibited an accelerated immune response with markedly higher ELISA values than the mice that received nave splenocytes (Fig. 1). Furthermore, this accelerated antibody response was dependent on the presence of CD4+ T cells (Fig. 1).
After we demonstrated that reconstitution of P. carinii-infected SCID mice with splenocytes from P. carinii-immunized donor mice resulted in an increased immune response, P. carinii-infected SCID mice were reconstituted with sensitized splenocytes and then CD4+ T cells or both CD4+ and CD8+ T cells were depleted. Consistent with the finding that there was an increased antibody response, there was a trend toward more rapid clearing of P. carinii in the mice reconstituted with P. carinii-sensitized splenocytes than in the mice reconstituted with nave cells by day 21 in experiment 1 (3.35 x 106 organisms/mouse versus 1.58 x 107 organisms/mouse; P = 0.31) (Tables 1 and 2) and by day 14 in experiment 2 (8.1 x 105 organisms/mouse versus 2.3 x 106 organisms/mouse; P = 0.12). In contrast to the results obtained with nave splenocytes, when P. carinii-infected SCID mice were reconstituted with sensitized splenocytes and CD4+ T cells were subsequently depleted, the expected blunted clearance of P. carinii was evident by day 14 after reconstitution (Table 2, experiment 2) (8.1 x 105 P. carinii organisms/mouse versus 2.1 x 106 P. carinii organisms/mouse; P = 0.04). However, as observed with the mice reconstituted with nave splenocytes, also depleting the CD8+ T cells in the mice had no effect on the number of P. carinii organisms in the mice compared to the number of organisms in mice in which only CD4+ cells were depleted (Table 2). In fact, at 14 days in experiment 2, the number of P. carinii organisms was actually higher in mice with CD8+ T cells than in mice in which CD8+ T cells were depleted (P = 0.04). This increase in the number P. carinii organisms could not be explained by differences in antibody levels, since both CD4+-T-cell-depleted mice and CD4+- and CD8+-T-cell-depleted mice had no detectable antibody to P. carinii (the ELISA optical density was 0.06 for both groups). The total number of inflammatory cells in bronchoalveolar lavage fluid was higher in CD4+-T-cell-depleted mice (3.4 x 105 ± 3.9 x 104 cells) than in CD4+- and CD8+ -T-cell-depleted mice (9.7 x 104 ± 1.7 x 104 cells; P = 0.05). The levels of tumor necrosis factor alpha and gamma interferon were also higher in the CD4+-T-cell-depleted mice than in the CD4+- and CD8+-T-cell-depleted mice (74 ± 48 pg/ml versus 39 ± 19 pg/ml and 60 ± 2 pg/ml versus 22 ± 22 pg/ml, respectively), but the differences were not statistically significant. Thus, despite documenting that there was an accelerated immune response with enhanced clearance of P. carinii in the mice fully reconstituted with P. carinii-sensitized splenocytes, after depletion of CD4+ T cells we could not document any difference in P. carinii replication between when CD8+ T cells were present and when CD8+ T cells were absent.
CD8+ T cells do not enhance the clearance of P. carinii in the CD4+-T-cell-depleted mouse model of PCP. Because in the experiments described above we used a reconstitution model, we wanted to examine the effect of CD8+ T cells over a longer time using a model of disease progression. Thus, as previously described (2, 16), we depleted CD4+ T cells or both CD4+ and CD8+ T cells in normal CB.17 mice and then infected the mice with P. carinii. Mice were sacrificed at 3, 4, and 6 weeks after infection with P. carinii to ensure that any contribution of CD8+ T cells was not missed during the course of infection. As shown in Table 3, the P. carinii counts, as determined by real-time PCR, in the CD4+-T-cell-depleted group were the same as or higher than the counts in the CD4+- and CD8+- T-cell-depleted mice at each time. The PCR results were confirmed by silver staining of P. carinii organisms, which also showed that there was no difference in the number of organisms at any of the times. Therefore, we were not able to demonstrate any significant contribution of CD8+ T cells to the control of the organism using the CD4+-T-cell-depleted model of PCP.
Effect of prior exposure to P. carinii on CD8+-T-cell-mediated lung injury. Using the SCID mouse model of PCP, we also looked at the effect of immunologic reconstitution with sensitized splenocytes on CDYN as a marker of impaired pulmonary physiology. Mice reconstituted with nave splenocytes were compared to mice reconstituted with sensitized splenocytes. To isolate the previously described effect of CD8+ T cells on lung injury during PCP, CD4+ cells were depleted from both groups. Despite similar organism burdens, mice with sensitized CD8+ T cells exhibited earlier onset of lung injury, by day 10 after reconstitution, as shown by a CDYN (1.47) that was significantly lower than the CDYN of the mice with P. carinii-nave CD8+ T cells (CDYN, 1.86; P = 0.004) (Fig. 2 and Table 4, experiment 1). However, over the next 11 days, the CDYN declined more gradually and the organism burden did not increase in mice reconstituted with sensitized splenocytes. In contrast, in the mice reconstituted with nave splenocytes, the CDYN fell more rapidly and the organism burden continued to rise, as would be expected in the absence of CD4+ T cells (Fig. 2 and Tables 1 and 2, experiment 1). We hypothesized that the slower fall in lung compliance was due to antibody-mediated control of the organism that may have limited the CD8+ T-cell inflammatory lung damage in the mice reconstituted with sensitized splenocytes (6).
When mice were reconstituted at a lower organism burden, the accelerated lung injury in the presence of sensitized CD8+ T cells was not evident (Table 4, experiment 2). However, this effect appeared to depend on the overall organism burden as a third experiment using mice with an organism burden of 107 organisms per mouse showed that there was a 26% drop in CDYN by day 10 after reconstitution in CD4+-T-cell-depleted mice reconstituted with sensitized splenocytes compared to mice reconstituted with nave splenocytes (Table 4, experiment 3) (P = 0.003).
Exacerbated lung injury in mice reconstituted with sensitized splenocytes requires CD8+ T cells. As noted above and shown in Fig. 3A, reconstituted and CD4+-T-cell-depleted P. carinii-infected SCID mice exhibited a significant (26%; P = 0.003) decrease in CDYN if they were reconstituted with P. carinii-sensitized splenocytes rather than nave splenocytes. However, the difference in early lung injury disappeared if the CD8+ cells in mice were also depleted (Fig. 3B), indicating that the presence of P. carinii-sensitized CD8+ T cells is a critical component of the accelerated onset of inflammatory lung injury in this model.
DISCUSSION
By reconstituting P. carinii-infected SCID mice with splenocytes that were presensitized to P. carinii, we were able to document an accelerated immune response to P. carinii in the recipient mice. However, in this model the presence of sensitized CD8+ T cells, in the absence of CD4+ T cells, had no effect on the organism burden but did result in accelerated onset of lung injury. These results confirm and extend our previous observations that CD8+ T cells play little or no role in the control of P. carinii but do contribute substantially to the inflammatory lung injury associated with PCP (16). Our inability to demonstrate that CD8+ T cells have an effect on the control of P. carinii, which is an extracellular pathogen, is consistent with the classical function of CD8+ T cells in controlling intracellular pathogens, such as viruses and some intracellular bacteria, rather than extracellular microorganisms.
We used two models of PCP to examine the effect of CD8+ T cells on P. carinii. Using the reconstituted SCID mouse model, we were able to show that, as expected, presensitized CD4+ T cells had a more pronounced effect on control of organism replication than nave CD4+ T cells had. However, despite demonstrating that reconstitution of infected mice with sensitized cells enhanced the known function of CD4+ T cells in protecting against P. carinii (9, 12, 13), we were unable to show any similar effect of CD8+ T cells. In fact, although the data were not statistically significant, with the exception of a single time (day 14) (Table 2, experiment 2), CD4+ T-cell-depleted mice with CD8+ T cells present, especially mice reconstituted with sensitized splenocytes, often contained higher numbers of P. carinii organisms, which is consistent with previous observations that the numbers of P. carinii organisms seem to be higher when there is more inflammation (5, 16). To confirm our observation with the SCID mouse model, we used the CD4+-T-cell-depleted model of PCP that enabled us to observe the effect of CD8+ T cells over a longer time, as PCP got progressively worse. We previously demonstrated that CD8+ T cells had no effect at a single time in this model (16). In this experiment we analyzed the P. carinii burden at 3, 4, and 6 weeks after inoculation of CD4+-T-cell-depleted mice. Using quantitative PCR and silver staining of organisms, we were not able to demonstrate any difference between the number of P. carinii organisms in CD4+-T-cell-depleted mice (CD8+ T cells present) and the number of P. carinii organisms in CD4+- and CD8+-T-cell-depleted mice at any time.
Our results differ from those of Beck et al., who showed that there was a modest reduction in the P. carinii burden in CD4+- T-cell-depleted mice compared to the burden in CD4+- and CD8+-T-cell-depleted mice (2). While the general experimental designs used by us and by Beck et al. were similar, Beck et al. performed a semiquantitative analysis based on silver staining of lung sections from mice at a single time. Thus, assaying cysts only in single lung sections may not have resulted in accurate representation of the overall organism burden in the entire lung. In contrast, we analyzed CD4+-T-cell-depleted mice at several times during progression of the disease. We also assayed lung homogenates using three methods of enumeration, Diff Quik staining, silver staining (data not shown), and real-time PCR, all of which produced consistent results. Finally, we corroborated our observations with the CD4+-T-cell-depleted mouse model using the reconstituted SCID mouse model of PCP.
Our results may also appear to conflict with several recent reports that gamma interferon stimulation of CD8+ T cells can protect CD4+-T-cell-depleted mice from development of PCP (10, 11, 14). However, these interesting studies of the effect of gamma interferon addressed a different issue about the biology of CD8+ T cells and the pathogenesis of PCP than our studies addressed. In the gamma interferon studies the workers used mice that were challenged with P. carinii 24 h after transfection of lung cells with a gamma interferon gene (10, 14) or 24 h before infusion of CD8+ T cells isolated from mice that had been transfected with the gamma interferon gene (11). In these studies, gamma interferon-stimulated CD8+ T cells protected against the development of PCP rather than controlled an established infection. In our experiments we examined the effect of CD8+ T cells both on the development of PCP (CD4+-T-cell-depleted model) and on eradication of an established infection (reconstituted SCID mouse model of PCP). Our interpretation of these two different experimental approaches is that our studies demonstrated that in the usual or "physiologic" response to infection with P. carinii, CD8+ T cells have little importance in the control of P. carinii; however, after pharmacologic stimulation with gamma interferon, CD8+ T cells can be made to have a biologic effect on the control of P. carinii that they normally do not express.
These studies also extended our previous experiments suggesting that CD8+ T cells are critical to the lung injury associated with PCP (16-18). For the current experiments, we hypothesized that if CD8+ T cells were indeed a critical component of the P. carinii-driven inflammatory response, then presensitized CD8+ T cells should accelerate the inflammatory response. This is in fact what we observed, since when P. carinii-infected SCID mice were reconstituted with P. carinii-sensitized splenocytes, there was a statistically significant more rapid onset of lung injury when subsequently CD4+ cells but not CD8+cells were depleted. This effect was lost after CD8+ T cells were also depleted from the reconstituted mice, confirming the role of these cells in exacerbating lung injury during PCP.
We previously demonstrated that P. carinii, in the absence of an adaptive immune response, did not cause much lung injury until relatively late in the course of the illness (16). However, the present study showed that there is a "threshold" effect of P. carinii on the induction of the CD8+-T-cell-mediated response since with lower numbers of P. carinii organisms (105 versus 107 organisms) the acceleration of lung injury by presensitized CD8+ T cells could not be demonstrated.
In summary, we concluded that CD8+ T cells play an insignificant role in control of P. carinii replication but are a key component of the inflammatory response and consequent lung injury induced by P. carinii in a susceptible host. Therefore, development of treatment strategies to blunt or eliminate the CD8+ T-cell-mediated inflammatory response to P. carinii, combined with antibiotic treatment of P. carinii, should result in improved lung function without an adverse impact on clearance of P. carinii.
ACKNOWLEDGMENTS
We acknowledge the expert technical assistance of Stephanie Campbell and Margaret Chovaniec.
This work was supported by grant P01 HL071659 from the National Institutes of Health.
FOOTNOTES
Corresponding author. Mailing address: University of Rochester School of Medicine & Dentistry, 601 Elmwood Avenue, Box 690, Rochester, NY 14642. Phone: (585) 275-5944. Fax: (585) 273-1104. E-mail: francis_gigliotti@urmc.rochester.edu.
Published ahead of print on 28 August 2006.
F.G. and E.L.C. contributed equally to the paper.
REFERENCES
1. Amdur, M. O., and J. Mead. 1958. Mechanics of respiration in unanesthetized guinea pigs. Am. J. Physiol. 192:364.
2. Beck, J. M., R. L. Newbury, B. E. Palmer, M. L. Warnock, P. K. Byrd, and H. B. Kaltreider. 1996. Role of CD8+ lymphocytes in host defense against Pneumocystis carinii in mice. J. Lab. Clin. Med. 128:477-487.
3. Bergmann, K. C., B. Lachmann, and K. Noack. 1984. Lung mechanics in orally immunized mice after aerolized exposure to influenza virus. Respiration 46:218-221.
4. Lee, L. H., F. Gigliotti, T. W. Wright, P. J. Simpson-Haidaris, G. A. Weinberg, and C. G. Haidaris. 2000. Molecular characterization of KEX1, a kexin-like protease in mouse Pneumocystis carinii. Gene 242:141-150.
5. Chen, W., E. A. Havell, F. Gigliotti, and A. G. Harmsen. 1993. Interleukin-6 production in a murine model of Pneumocystis carinii pneumonia: relation to resistance and inflammatory response. Infect. Immun. 61:97-102.
6. Empey, K. M., M. Hollifield, K. Schuer, F. Gigliotti, and B. Garvy. 2004. Passive immunization of neonatal mice against Pneumocystis carinii f. sp. muris enhances control of infection without stimulating inflammation. Infect. Immun. 72:6211-6220.
7. Gigliotti, F., and A. G. Harmsen. 1997. Pneumocystis carinii host origin defines the antibody specificity and protective response induced by immunization. J. Infect. Dis. 176:1322-1326.
8. Gigliotti, F., and T. McCool. 1996. Glycoprotein A is the immunodominant antigen of Pneumocystis carinii in mice following immunization. Parasitol. Res. 82:90-91.
9. Harmsen, A. G., and M. Stankiewicz. 1990. Requirement for CD4+ cells in resistance to Pneumocystis carinii pneumonia in mice. J. Exp. Med. 172:937-945.
10. Kolls, J. K., S. Habetz, M. K. Shean, C. Vazquez, J. A. Brown, D. Lei, P. Schwarzenberger, P. Ye, S. Nelson, W. R. Summer, and J. E. Shellito. 1999. IFN-gamma and CD8+ T cells restore host defenses against Pneumocystis carinii in mice depleted of CD4+ T cells. J. Immunol. 162:2890-2894.
11. McAllister, F., C. Steele, M. Zheng, E. Young, J. E. Shellito, L. Marrero, and J. K. Kolls. 2004. T cytotoxic-1 CD8+ T cells are effector cells against Pneumocystis in mice. J. Immunol. 172:1132-1138.
12. Roths, J. B., and C. L. Sidman. 1992. Both immunity and hyperresponsiveness to Pneumocystis carinii result from transfer of CD4+ but not CD8+ T cells into severe combined immunodeficiency mice. J. Clin. Investig. 90:673-678.
13. Shellito, J., V. V. Suzara, W. Blumenfeld, J. M. Beck, H. J. Steger, and T. H. Ermak. 1990. A new model of Pneumocystis carinii infection in mice selectively depleted of helper T lymphocytes. J. Clin. Investig. 85:1686-1693.
14. Steele, C., M. Zheng, E. Young, L. Marrero, J. E. Shellito, and J. K. Kolls. 2002. Increased host resistance against Pneumocystis carinii pneumonia in T-cell-deficient mice: protective role of gamma interferon and CD8+ T cells. Infect. Immun. 70:5208-5215.
15. Wang, J., F. Gigliotti, S. Maggirwar, C. Johnston, J. N. Finkelstein, and T. W. Wright. 2005. Pneumocystis activates the NF-B signaling pathway in alveolar epithelial cells. Infect. Immun. 73:2766-2777.
16. Wright, T. W., F. Gigliotti, J. N. Finkelstein, J. T. McBride, C. L. An, and A. G. Harmsen. 1999. Immune-mediated inflammation directly impairs pulmonary function, contributing to the pathogenesis of Pneumocystis carinii pneumonia. J. Clin. Investig. 104:1307-1317.
17. Wright, T. W., R. H. Notter, Z. Wang, A. G. Harmsen, and F. Gigliotti. 2001. Pulmonary inflammation disrupts surfactant function during Pneumocystis carinii pneumonia. Infect. Immun. 69:758-764.
18. Wright, T. W., G. S. Pryhuber, P. R. Chess, Z. Wang, R. H. Notter, and F. Gigliotti. 2004. TNF receptor signaling contributes to chemokine secretion, inflammation, and respiratory deficits during Pneumocystis pneumonia. J. Immunol. 172:2511-2521.(Francis Gigliotti, Elliott L. Crow, Sami)