CD8+ T cells play an essential role in immunity to Chlamydiapneumoniae (Cpn). However, the target Ags recognized by Cpn-specific CD8+ T cells have not been identified, and the mechanisms by which this T cell subset contributes to protection remain unknown. In this work we demonstrate that Cpn infection primes a pathogen-specific CD8+ T cell response in mice. Eighteen H-2b binding peptides representing sequences from 12 Cpn Ags sensitized target cells for MHC class I-restricted lysis by CD8+ CTL generated from the spleens and lungs of infected mice. Peptide-specific IFN-γ-secreting CD8+ T cells were present in local and systemic compartments after primary infection, and these cells expanded after pathogen re-exposure. CD8+ T cell lines to the 18 Cpn epitope-bearing peptides were cytotoxic, displayed a memory phenotype, and secreted IFN-γ and TNF-α, but not IL-4. These CTL lines lysed Cpn-infected macrophages, and the lytic activity was inhibited by brefeldin A, indicating endogenous processing of CTL Ags. Finally, Cpn peptide-specific CD8+ CTL suppressed chlamydial growth in vitro by direct lysis of infected cells and by secretion of IFN-γ and other soluble factors. These studies provide information on the mechanisms by which CD8+ CTL protect against Cpn, furnish the tools to investigate their possible role in immunopathology, and lay the foundation for future work to develop vaccines against acute and chronic Cpn infections.

hlamydiapneumoniae (Cpn)3 is an obligate intracellular bacterial pathogen that is estimated to cause at least one infection during the lifetime of nearly every human being (1). Although most infections are mild or subclinical, Cpn is a common cause of community-acquired pneumonia, bronchitis, pharyngitis, and sinusitis (2). Like Chlamydia trachomatis and Chlamydia psittaci, the two other chlamydial human pathogens, Cpn can persist in the host and cause chronic infection (3), which is associated with many inflammatory conditions, including asthma, chronic obstructive pulmonary disease, and multiple sclerosis (4, 5, 6). Of greatest significance is the compelling association of Cpn infection with atherosclerosis and cardiovascular events (7, 8). Although antibiotics can treat acute Cpn infection, cells can remain persistently infected despite treatment (9). Therefore, a logical approach to reduce respiratory and systemic morbidity from Cpn is to develop an effective vaccine to prevent or ameliorate acute and chronic infection from this pathogen. However, developing vaccines against Chlamydia has been hindered by the limited knowledge of pathogen Ags and immune mechanisms that lead to protective or adverse immune responses.

CD8+ T cells play a critical role in protection against most intracellular pathogens, including Chlamydia. Pathogen-derived Ags from organisms that replicate in the host cell cytosol, such as Listeria monocytogenes and Trypanosoma cruzi, readily induce a CD8+ T cell response, as microbial proteins are directly accessible to the MHC class I Ag-processing machinery. In contrast, for Chlamydia, which resides in a membrane-bound vacuole termed an inclusion, and for other intravacuolar pathogens, such as Mycobacterium tuberculosis, Ags need to traffic into the cytosol for CD8+ CTL induction. Nevertheless, CD8+ CTL responses are induced to mycobacterial Ags (10), and CD8+ T cells primed during C. trachomatis infection lyse chlamydia-infected cells (11, 12). Moreover, depletion and adoptive transfer of CD8+ T cells have, respectively, abrogated and conferred protection to C. psittaci- and C. trachomatis-challenged mice (12, 13, 14). Despite the clear role of CD8+ T cells in resistance to chlamydial pathogens, only two C. trachomatis CD8+ CTL target Ags have been identified to date (15, 16).

Information on immunity to Cpn is sparse, but studies using a mouse model that faithfully mimics important aspects of human Cpn infection (17, 18, 19) indicate that CD8+ T cells and IFN-γ are critical for protection (20, 21, 22). In the absence of CD8+ T cells, Cpn-infected mice have increased bacterial burdens and disease severity (20, 21), and in animals lacking IFN-γ signaling, bacterial loads are higher, and clearance of organisms is greatly hampered (22). Nevertheless, it is uncertain whether CD8+ T cells recognize Cpn-infected cells and whether this T cell subset contributes to protection through cytokine production or a lytic mechanism, as cytokine-producing CD8+ CTL have not been documented during Cpn infection. Furthermore, the Cpn-derived Ags contributing to MHC class I-restricted CD8+ T cell responses remain unidentified.

We report in this work that Cpn-infected mice generate pathogen-specific CD8+ CTL with a type 1 cytokine secretion pattern and that these effector cells recognize multiple MHC class I-restricted epitopes from Cpn Ags endogenously processed by productively infected macrophages. We also show that Cpn peptide-specific CD8+ CTL and their soluble factors significantly inhibit chlamydial growth in vitro.

These studies used Mycoplasma-free stocks of the Cpn Kajaani 6 (K6) (obtained from Dr. M. Puolakkainen, University of Helsinki, Helsinki, Finland), AR39 (University of Washington, Seattle, WA), and CWL029 (obtained from Dr. C. M. Black, Centers for Disease Control, Atlanta, GA) strains. For propagation to high titers, each bacterial strain diluted in Chlamydia medium was centrifuged (500 × g, 1 h, 35°C) onto monolayers of HL cells (23) grown in 12-well plates (BD Biosciences, Franklin Lakes, NJ). Plates were incubated for 1 h at 37°C in 6% CO2 before replacing the inocula with cycloheximide-containing medium that was then used to incubate infected cultures for 72 h. Infected monolayers were harvested with glass beads and sonicated on ice for 20 s. After removing cell debris by low speed centrifugation (200 × g, 10 min, 4°C), bacteria were pelleted (33,100 × g, 35 min, 4°C), resuspended in cold sucrose-phosphate-glutamate solution, and aliquoted for titration and storage at −70°C. Similarly processed uninfected HL cell monolayers were used to prepare control material.

Six- to 10-wk-old female C57BL/6J (B6) mice (H-2b) (The Jackson Laboratory, Bar Harbor, ME) were used in all experiments. Mice were kept in microisolator cages and housed in a pathogen-free environment. B6 mice were infected by intranasal (i.n.) inoculation with 106 inclusion-forming units (IFU) of Cpn K6 in 40 μl PBS under methoxyflurane anesthesia. In most experiments animals were reinfected i.n. with the same infectious dose 35–100 days after the initial infection. Control mice were inoculated with material prepared from uninfected cells. To induce T. cruzi trypomastigote surface Ag (TSA)-1-specific CTL, B6 mice were infected with this parasite as previously described (24). The institutional animal care and use committee approved all procedures involving animals.

RMA-S (H-2b; TAP2, T cell lymphoma; provided by Dr. H.-G. Ljundggren, Karolinska Institute, Stockholm, Sweden) (25), mAM (H-2b; murine alveolar macrophage cell line; Z. Chroneos, unpublished observations), HL (University of Washington) (23), and HEp-2 (ATCC CCL 23, American Type Culture Collection, Manassas, VA) were maintained in complete RPMI 1640 medium containing 10% heat-inactivated FBS (HyClone, Logan, UT), 20 mM HEPES, 2 mM l-glutamine, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, and 20 μg/ml gentamicin (all from Invitrogen-Life Technologies, Gaithersburg, MD). T2Kb and T2Db (T2 cells transfected with the Kb and Db genes; supplied by Dr. P. Cresswell, Yale University, New Haven, CT) (26) were maintained in complete IMDM with 0.4 mg/ml Geneticin (Invitrogen). T cell medium (TCM) was prepared by supplementing complete RPMI 1640 with 50 μM 2-ME (Invitrogen). TCM-RCAS was made by addition of 5% supernatant from Con A-stimulated rat splenocytes (T-STIM without Con A; Collaborative Biomedical Products, Bedford, MA). Chlamydia medium consisted of complete DMEM/F-12 medium (Invitrogen) with an additional 0.5 mg/ml glucose and 0.26 mg/ml sodium bicarbonate. Cycloheximide (1 μg/ml; Sigma-Aldrich, St. Louis, MO) was included when indicated.

Lungs collected at various time points between 2 and 165 days after primary and secondary Cpn or mock infections (three to five mice per time point) were perfused with 10 ml PBS via the right ventricle and inflated by intratracheal instillation with Excell fixative (American MasterTech Scientific, Lodi, CA). After postfixation, tissues were embedded in paraffin, and sections (6 μm) were stained with H&E for histological analysis. To detect Cpn, deparaffinized tissue sections were treated for 15 min with 3% H2O2, blocked for 20 min with 5% BSA, and then incubated overnight at 4°C with a 1/2000 dilution of the Cpn major outer membrane protein (MOMP)-specific mAb RR-402 (University of Washington) (27, 28). A 1/1000 dilution of biotinylated goat anti-mouse IgG (ICN, Costa Mesa, CA) was then applied to the sections for 30 min, followed by 15 min with a 1/10 dilution of streptavidin-HRP (Innovex Biosciences, Richmond, CA). Color development and counterstaining was achieved using Turbo AEC (Innovex) and Contrast Blue (KPL, Gaithersburg, MD), respectively. After each staining step, sections were rinsed with wash solution (KPL) and signal enhancing buffer (Innovex).

mAbs used for cell surface staining were FITC and PE anti-CD8α (53-6.7), FITC anti-CD4 (H129.19), PE anti-CD44 (IM7), PE anti-TCRαβ (H57-597), FITC anti-CD3ε (145-2C11), FITC- and PE isotype-matched control mAbs (all from BD PharMingen, San Diego, CA), purified anti-Db (28-14-8S; ATCC HB176; ATCC), and purified anti-Kb (Y3; ATCC HB176; ATCC). Cells (5 × 105–1 × 106) were washed with cold FACS buffer (1% BSA/0.05% NaN3 in PBS) and stained for 45 min at 4°C in 100 μl buffer with saturating concentrations of Abs. For purified mAbs, cells were then stained for 30 min on ice with a 1/50 dilution of FITC-F(ab′)2 goat anti-mouse IgG (Southern Biotechnology, Birmingham, AL). Cells were washed twice after each staining step and then analyzed on an EPICS C flow cytometer (Beckman Coulter, Hialeah, FL).

Single-cell suspensions of perfused lungs from Cpn- and mock-infected mice were prepared by homogenizing the organs in 100-μm pore size mesh cell strainers (BD Biosciences). After lysing RBC, cells were washed and resuspended at 107/ml in TCM. Isolated lung mononuclear cells (LMNC) were plated (106/100 μl/well) into TCM-washed, 40-h cultures of Cpn K6-infected and uninfected mAM monolayers prepared in 96-well, flat-bottom plates (Costar, Cambridge, MA). The mAM (2 × 104/well) were infected at 4 IFU/cell, then incubated in medium without cycloheximide. Parallel Cpn-infected mAM monolayers, fixed for 10 min in methanol and stained with an FITC-conjugated Chlamydia genus-specific mAb (Pathfinder Chlamydia Culture Confirmation System; Bio-Rad, Hercules, CA), indicated that ∼60–70% of mAM had Cpn inclusions. After 1.5 h of coculture, 100 μl TCM with 2 μl/ml GolgiPlug (brefeldin A (BFA); BD PharMingen) was added to each well. LMNC were harvested 3.5 h later, washed once in FACS buffer, incubated for 15 min on ice with a 1/100 dilution of anti-CD16/CD32 (2.4G2) mAb (BD PharMingen), and then surface-stained with FITC anti-CD8α. Cells were washed, fixed, and permeabilized (Cytofix/Cytoperm kit, BD PharMingen), then incubated for 30 min on ice with a 1/100 dilution of PE-conjugated anti-IFN-γ (XMG1.2) before analysis by flow cytometry. An isotype-matched mAb (rat IgG1) was used to control for the specificity of intracellular cytokine staining.

H-2b motif-bearing Cpn and control peptides (Table I) were synthesized by F-moc-based solid phase chemistry using an ABI 430A peptide synthesizer (Applied Biosystems, Foster City, CA) and were purified by reverse phase HPLC. To confirm their purity (95%) and identity, peptides were analyzed by mass spectrometry. The control peptides used were Kb-restricted T. cruzi TSA-1515 epitope VDYNFTIV (TcTSA) (24), Kb-restricted OVA257 epitope SIINFEKL (OVA) (29), and Db-restricted influenza A NP366 epitope ASNENMETM (FLUnp) (30). Lyophilized peptides were dissolved in DMSO (Sigma) at 20 mg/ml and stored at −70°C. Before use, peptides were further diluted with RPMI 1640. No cell toxicity was associated with any peptide.

Table I.

Chlamydia pneumoniae synthetic peptides: protein sources and H-2b binding affinities

IDPeptide SequenceaProtein(s)bcCpn Genome AnnotationsdH-2b MotifeRatio MFIfBinding (IC50 nM)g
NTVVFDAL 60 kDa Omp (Omp2, OmcB)bh CPn0557, CP0195 Kb 1.69 155,000 
QESCYGRL   Kb 2.78 344 
ISVSNPGDL   Db 2.54 220 
VLSFNLGDM   Db 1.49 62,857 
AEDTNVSLI   Db 0.95 440,000 
SKLQYKII FKBP-type PPIASE (Mip) CPn0661, CP0086 Kb 1.13 155,000 
SSEGNNEPIL   Db/Db 1.01 293,333 
QLPPNSLLI   Db 1.07 293,333 
DDEEYVIL 10 kDa chaperonin (GroES)bh CPn0135, CP0637 Kb 0.99 124,000 
10 ANEGYDAL 60 kDa chaperonin (GroEL)bh CPn0134, CP0638 Kb 0.97 11,923 
11 TAGANPMDL   Db 1.47 22,000 
12 ISANNDSEI   Db 1.80 880 
13 STEINQPFITM DnaK (heat shock protein-70)bh CPn0503, CP0251 Db/Kb 2.01/2.07 124,000/13,209 
14 VLSTNGDTL   Db 3.30 427 
15 FLLFFEFLLV 76 kDabh CPn0728, CP0018 Kb/Kb 1.85 20,667 
16 LMSGFRQM   Kb 1.43 6,889 
17 YASDNQAIL   Db 2.89 480 
18 GFKSNFNKI   Kb/Db 1.2/0.95 155,000/440,000 
19 LVYNYPGV 43 kDa homologs 1–4:4 CPn0929, CP0937 Kb 3.20 23 
20 LIYNYPGV CPn0928, CP0938 Kb 2.99 4.1 
21 LLVFNYPGI CPn0927, CP0939 Db/Kb 1.69/1.48 5,167/62,857 
22 LIFNYPGV CPn0562, CP0188 Kb 2.81 40 
23 HPYLYRLL CPn0562, CP0188 Kb 2.10 194 
24 HPTLFKVL CPn0927, CP0939 Kb 2.87 620 
25 SIILFLPL CPn0927, CP0939 Kb 3.61 1.0 
26 KICQNFILL CPn0562, CP0188 Db/Kb 2.80/2.95 2,480/2,200 
27 ISNGNSDCL CPn0562, CP0188 Db 3.62 314 
28 YSQGNSGLM CPn0927, CP0939 Db 2.35 889 
29 TGKLNLENL CPn0928, CP0938 Db 1.79 17,600 
30 QAPTNRWML CPn0929, CP0937 Db 3.07 5.6 
31 SLLGNATAL MOMP (OMP1, OmpA)bh CPn0695, CP0051 Db 2.98 160 
32 SHYAFSPMFEVL Omp5bh (Pmp10) CPn0449, CP0303 Kb/Kb 2.66 689 
33 ISFAFCQL   Kb 2.91 1.3 
34 QPQNYLRL Omp4b.h (Pmp11) CPn0451, CP0302 Kb 1.54 51,667 
35 HDQLFSLL   Kb 1.89 2,296 
36 GTYHFTKL Omp85 homolog (YaeT) CPn0300, CP0458 Kb 2.79 18 
37 FQLCNSYDL OmpB (PorB) CPn0854, CP1015 Db 4.29 13 
38 NHPVFSPL IncA homolog CPn0585, CP0163 Kb 2.61 56 
39 LQQRYSRL   Kb 2.91 0.6 
TcTSA VDYNFTIV T. cruzi TSA1  Kb 3.37 36 
OVA SIINFEKL OVA  Kb 3.38 11 
FLUnp ASNENMETM Influenza A nucleoprotein  Db 3.47 6.2 
IDPeptide SequenceaProtein(s)bcCpn Genome AnnotationsdH-2b MotifeRatio MFIfBinding (IC50 nM)g
NTVVFDAL 60 kDa Omp (Omp2, OmcB)bh CPn0557, CP0195 Kb 1.69 155,000 
QESCYGRL   Kb 2.78 344 
ISVSNPGDL   Db 2.54 220 
VLSFNLGDM   Db 1.49 62,857 
AEDTNVSLI   Db 0.95 440,000 
SKLQYKII FKBP-type PPIASE (Mip) CPn0661, CP0086 Kb 1.13 155,000 
SSEGNNEPIL   Db/Db 1.01 293,333 
QLPPNSLLI   Db 1.07 293,333 
DDEEYVIL 10 kDa chaperonin (GroES)bh CPn0135, CP0637 Kb 0.99 124,000 
10 ANEGYDAL 60 kDa chaperonin (GroEL)bh CPn0134, CP0638 Kb 0.97 11,923 
11 TAGANPMDL   Db 1.47 22,000 
12 ISANNDSEI   Db 1.80 880 
13 STEINQPFITM DnaK (heat shock protein-70)bh CPn0503, CP0251 Db/Kb 2.01/2.07 124,000/13,209 
14 VLSTNGDTL   Db 3.30 427 
15 FLLFFEFLLV 76 kDabh CPn0728, CP0018 Kb/Kb 1.85 20,667 
16 LMSGFRQM   Kb 1.43 6,889 
17 YASDNQAIL   Db 2.89 480 
18 GFKSNFNKI   Kb/Db 1.2/0.95 155,000/440,000 
19 LVYNYPGV 43 kDa homologs 1–4:4 CPn0929, CP0937 Kb 3.20 23 
20 LIYNYPGV CPn0928, CP0938 Kb 2.99 4.1 
21 LLVFNYPGI CPn0927, CP0939 Db/Kb 1.69/1.48 5,167/62,857 
22 LIFNYPGV CPn0562, CP0188 Kb 2.81 40 
23 HPYLYRLL CPn0562, CP0188 Kb 2.10 194 
24 HPTLFKVL CPn0927, CP0939 Kb 2.87 620 
25 SIILFLPL CPn0927, CP0939 Kb 3.61 1.0 
26 KICQNFILL CPn0562, CP0188 Db/Kb 2.80/2.95 2,480/2,200 
27 ISNGNSDCL CPn0562, CP0188 Db 3.62 314 
28 YSQGNSGLM CPn0927, CP0939 Db 2.35 889 
29 TGKLNLENL CPn0928, CP0938 Db 1.79 17,600 
30 QAPTNRWML CPn0929, CP0937 Db 3.07 5.6 
31 SLLGNATAL MOMP (OMP1, OmpA)bh CPn0695, CP0051 Db 2.98 160 
32 SHYAFSPMFEVL Omp5bh (Pmp10) CPn0449, CP0303 Kb/Kb 2.66 689 
33 ISFAFCQL   Kb 2.91 1.3 
34 QPQNYLRL Omp4b.h (Pmp11) CPn0451, CP0302 Kb 1.54 51,667 
35 HDQLFSLL   Kb 1.89 2,296 
36 GTYHFTKL Omp85 homolog (YaeT) CPn0300, CP0458 Kb 2.79 18 
37 FQLCNSYDL OmpB (PorB) CPn0854, CP1015 Db 4.29 13 
38 NHPVFSPL IncA homolog CPn0585, CP0163 Kb 2.61 56 
39 LQQRYSRL   Kb 2.91 0.6 
TcTSA VDYNFTIV T. cruzi TSA1  Kb 3.37 36 
OVA SIINFEKL OVA  Kb 3.38 11 
FLUnp ASNENMETM Influenza A nucleoprotein  Db 3.47 6.2 
a

Sequence in single-letter amino acid code.

b

Cpn proteins from which H-2b motif-bearing sequences were selected for peptide synthesis; synonyms shown in parentheses.

c

Originally reported in Cpn genome sequencing Refs. 33–35.

d

Gene name designations as annotated from the sequenced genomes of Cpn strains CWL029 (prefix CPn; Chlamydia Genome Project; reported in Ref. 33; http://chlamydia-www.berkeley.edu;4231) and AR39 (prefix CP; reported in Ref. 34; http://www.tigr.org/tigr-scripts/CMR2/GenomePage3.spl?database=bcp). Cpn strain J138 gene names are commonly those of CWL029 (http://w3.grt.kyusbu-u.ac.jp/J138/; http://www.tigr.org/tigr-scripts/CMR2/GenomePage3.spl?database=ntcp02).

e

H-2b binding motif present in synthetic peptides; H-2Kb binding motif is FY at position 5, and LMIV at position 8; H-2Db binding motif is N at position 5, and MIL at position 9 (36).

f

Binding affinity as assessed by ability of peptides (10 μM) to up-regulate and stabilize Kb or Db molecules on RMA-S cells. Ratio MFI is MFI in the presence of test peptide over the fluorescence intensity in the absence of peptide (high, ≥2.5; intermediate, 2.0–2.4; low, 1.5–1.9; negative, ≤1.4).

g

Binding affinity as measured by quantitative binding assay. IC50 is the nanomolar concentration of peptide capable of inhibiting by 50% the binding of 125I-labeled index peptide to purified soluble H-2b molecules (high, ≤50 nM; intermediate, 50–500 nM; low, 500–10,000 nM; negative, >10,000).

h

Originally reported in Refs. 39–44.

Peptide binding to Kb or Db was measured by the stabilization of class I molecules on the surface of RMA-S cells (31) and by a quantitative molecular binding assay that measures the inhibition of binding of a radiolabeled probe peptide to soluble Kb or Db molecules (32). For the MHC class I stabilization assay, RMA-S cells (106/ml) were cultured at 26°C in 6% CO2 for 24 h, followed by 1 h in the presence of peptide (0.1–50 μM). Cells were then transferred to 37°C for 2 h and washed with FACS buffer, and up-regulated cell surface expression of Kb and Db was detected by FACS analysis after staining for immunofluorescence. Results were expressed as the mean fluorescence intensity (MFI) ratio: MFI of peptide-treated cells/MFI of untreated cells. For the quantitative binding assay, test peptides (1 nM to 100 μM) were coincubated with radiolabeled probe peptides (1–10 nM; SGPSNTYPEI for Db; RGYVFQGL for Kb), purified soluble Db or Kb H chain (5–500 nM), and β2-microglobulin (1 μM; Scripps Laboratories, San Diego, CA) for 48 h at room temperature in the presence of protease inhibitors. The percentage of MHC-bound radioactivity was determined by gel filtration, and the concentration required to inhibit 50% (IC50) of the binding of radiolabeled peptide was calculated.

To generate Cpn peptide-specific CTL, mice were killed 2 wk to 6 mo after the first or second Cpn infection. Immune spleen cells (SC) were washed, resuspended in TCM, and seeded in 24-well plates (Costar) at 5 × 106 cells/well. Individual peptides (2 μM) were included in each 2-ml culture. After 2 days of incubation at 37°C, 6% CO2, cultures were made to 5% RCAS and incubated for 4 additional days. Peptide-stimulated effectors were also generated from LMNC removed 1 wk to 45 days after primary or secondary infection.

To prepare RMA-S targets, cells preincubated for 24 h at 26°C in 6% CO2, were seeded into 24-well plates (106/2 ml/well) and incubated overnight in the presence of peptide (0.1 μM) and 100 μCi Na251CrO4 (51Cr; Amersham, Arlington Heights, IL). Cells were shifted to 37°C for 2 h before processing for CTL assays. T2Kb and T2Db targets were prepared by overnight incubation at 37°C with 51Cr and peptide. To prepare Cpn-infected targets, 24-h mAM monolayers growing in 12-well plates (3 × 105/well) were centrifuged with 0.4 ml/well Chlamydia medium containing live or heat-killed Cpn (4 IFU/cell). Heat-killed Cpn was prepared by incubating bacteria at 60°C for 1 h. Control mAM targets were inoculated with HL cell-derived material. After 1 h of incubation at 37°C, the inocula were removed, and 3 ml medium with 100 μCi 51Cr was added to each well. The plates were incubated at 37°C for 20 h, at which time BFA (10 μg/ml) was added to a subset of wells containing Cpn-infected mAM. Two hours later, monolayers were washed with RPMI 1640 and treated with Cell Dissociation Buffer (Invitrogen) to prepare single-cell suspensions. Chlamydia-specific immunofluorescent staining of infected mAM incubated for 24 h more indicated that ∼60–70% of the cells were infected. Similar rates of infection were achieved for the K6, AR39, and CWL029 Cpn strains. BFA treatment did not affect Cpn growth, as determined by subculture of 5-h BFA-treated infected mAM monolayers.

Cytotoxic activity of effector cells on target cells was assessed by 51Cr release assays, as previously described (24). Briefly, peptide-sensitized, Cpn-infected, and control 51Cr-labeled target cells (5 × 103/well) were incubated for 5-h with effector cells at various E:T cell ratios in 96-well, round-bottom plates (Corning, Corning, NY). BFA (10 μg/ml) was included in those wells containing Cpn-infected BFA-treated target cells. Effectors depleted of CD8+ and CD4+ T cells were only tested at the highest E:T cell ratio. Depletions were conducted with magnetic beads coated with anti-CD8 or anti-CD4 mAbs (Miltenyi Biotec, Auburn, CA) or by incubating cells on ice for 30 min with anti-CD8 (3.155; ATCC TIB 211, ATCC) or anti-CD4 (GK1.5; ATCC TIB 207; ATCC), followed by 30 min at 37°C with rabbit complement. Supernatants were harvested (Skatron SCS System; Molecular Devices, Sunnyvale, CA), and released 51Cr was counted on a gamma counter. The percent specific lysis was calculated from the mean of triplicates as 100 × (experimental release − spontaneous release)/(maximum release − spontaneous release). Spontaneous release did not exceed 26% of the maximum release. SEs were <6% of the mean. A CTL response was considered positive when, at the highest E:T ratio, a difference of ≥10% lysis was obtained by subtracting the percentage lysis for control peptide from the percentage lysis for the test peptide.

Peptide-specific CD8+ T cell lines were generated as previously described (24). In brief, 4 × 107 immune SC from Cpn-infected mice were incubated with CTL peptides (2 μM) for 6 days at 37°C in 6% CO2 in 10 ml TCM using T25 flasks (Corning). RCAS (5%) was added on day 2 of culture. In each of two to four subsequent 6-day cycles of restimulation, 4 × 106 viable effector cells were cultured with 4 × 107 peptide-pulsed irradiated (3000 rad) syngeneic SC in 15 ml TCM-RCAS. A similar protocol was used to generate CD8+ T cell lines from immune LMNC.

Culture supernatants of peptide-stimulated SC were harvested 48 h into the first and second cycles of stimulation, and levels of IFN-γ, IL-4, and TNF-α were determined by sandwich ELISA. Capture and detection Ab pairs were R4-6A2/XMG1.2 (BD PharMingen) for IFN-γ, BVD4-1D11/BVD6-24G2 (Caltag, Burlingame, CA) for IL-4, and Ag affinity-purified goat polyclonal Ab/MP6-XT3 (R&D Systems and BD PharMingen) for TNF-α. The lower detection limits for IFN-γ, IL-4, and TNF-α were 40, 8, and 40 pg/ml, respectively.

Nitrocellulose-backed 96-well plates (MultiScreen MAHA S4510; Millipore, Bedford, MA) were coated overnight at 4°C with 75 μl PBS containing 10 μg/ml anti-IFN-γ mAb R4-6A2. After washing with PBS, the wells were blocked for 2 h at 37°C with 100 μl TCM. Two-fold serial dilutions of freshly isolated LMNC or SC starting at 107 cells/ml were added in TCM-RCAS in triplicate wells (100 μl/well) containing peptide-pulsed (1 μM) irradiated (16 krad) RMA-S cells (105/100 μl/well). Irradiated unpulsed RMA-S cells were used as a control for Ag-independent IFN-γ secretion. As positive controls, LMNC and SC containing unpulsed RMA-S cells were stimulated with PMA (50 ng/ml) and ionomycin (500 ng/ml; Sigma). After incubation at 37°C in 5% CO2 for 24 h, plates were washed three times with PBS, followed by three times with PBS/0.05% Tween 20 (PBS/T). Wells then received 75 μl of a solution of 3 μg/ml biotinylated anti-IFN-γ mAb XMG1.2 in PBS/T/0.5% FBS. After a 16-h incubation at 4°C, plates were washed six times with PBS/T, and 100 μl alkaline phosphatase-ExtrAvidin (Sigma; 1/800 dilution) was added to each well. Following 1 h at 26°C wells were washed four times with PBS/T and twice with PBS. Spots were developed with 75 μl/well 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium substrate (Sigma). After 20 min the plates were rinsed with water and dried, and spots in each well were counted under a stereomicroscope.

To evaluate the inhibition of Cpn growth by CD8+ T cell-derived soluble factors, 24-h HEp-2 cell monolayers prepared in 48-well plates (5 × 104/well) were treated with supernatants from CTL lines for 18 h before Cpn infection. Supernatants preincubated with anti-IFN-γ XMG1.2 or IgG1 control mAbs (20 μg/ml; BD PharMingen) were also used for HEp-2 cell pretreatment. After Cpn infection (3 IFU/cell), plates were incubated for 1 h at 37°C, the inocula were removed, and monolayers were incubated for 24 h more with new samples of the same supernatants. Following 48 h of incubation in Chlamydia medium, cultures were stained with the Pathfinder Chlamydia genus-specific mAb, and the number of inclusions in each well was counted in 10 ×400 high powered fields (HPF). Supernatants were tested in triplicate. The numbers of inclusions in Cpn-infected HEp-2 cells treated with supernatants from cultures lacking CD8+ T cells, peptide, or both were also assessed. All results were compared with those obtained using supernatants from immunomagnetically purified naive CD8+ T cells stimulated with Cpn CTL peptides.

To test the chlamydial growth-inhibiting activity of Cpn peptide-specific CTL, mAM monolayers prepared 1 day earlier in 48-well plates (4 × 104/well) were infected with Cpn (4 IFU/cell) and incubated for 20 h before adding graded numbers of CD8+ T cells (2.5 × 104–4 × 105/ml/well). Cultures with or without effectors were washed 4 h later, and 1 ml Chlamydia medium with cycloheximide was added to each well. After 48 h at 37°C, mAM monolayers were scraped, plates were frozen and thawed once, and serial dilutions of cleared pooled material from triplicate wells were used to infect fresh HEp-2 cell monolayers. Total numbers of mAM recovered from parallel cultures were similar for all effector cell types and densities tested, ranging between 1–1.4 × 105/triplicate. Inoculated HEp-2 cells were incubated for 72 h in cycloheximide-containing medium and processed for immunofluorescence, and chlamydial inclusions were counted in 10 HPF. Control effector cells were T. cruzi TSA-1515-specific CD8+ CTL and purified naive CD8+ T cells.

Mouse models of Cpn infection indicate that CD8+ T cells and IFN-γ play important roles in protective immunity (20, 21, 22). However, the induction of Cpn-specific CD8+ T cells during infection has not been reported. Therefore, we first determined whether the pulmonary inflammatory response of B6 mice during primary and secondary Cpn infections included pathogen-specific CD8+ T cells capable of IFN-γ production. Lung sections from mock-infected animals showed no evidence of pulmonary inflammation up to 100 days after each of two inoculations (Fig. 1,A). In contrast, a mild interstitial mononuclear cell infiltration with a perivascular and peribronchiolar lymphoid cuffing was present in the lungs from Cpn-infected mice from days 5–60 after infection (Fig. 1, C and D). During reinfection, a moderate to marked pneumonia and lymphoid infiltrate was detected 2 days postinoculation, peaked on days 8–12 (Fig. 1,B), and gradually declined, but remained present for up to 2 mo. Pulmonary bacterial loads commonly peaked 2–4 days before maximal inflammatory responses with mean IFUs per lung of 3 × 105 ± 9 × 104 after primary infection and 2.9 × 104 ± 8 × 103 after reinfection. Staining with a Cpn MOMP-specific mAb localized chlamydial Ag within epithelial and mononuclear cells throughout the interstitium, especially in the perivascular and peribronchiolar inflammatory foci (Fig. 1,D). No staining was detected using an IgG3 control mAb (Fig. 1 C). The MOMP-specific mAb did not stain lung tissue from mock-infected mice (data not shown).

FIGURE 1.

Detection of intracellular Cpn Ag and pathogen-specific IFN-γ-producing CD8+ T cells in the lungs of infected mice. Lungs from Cpn- and mock-infected B6 mice were obtained 8–12 days after i.n. inoculation with 40 μl PBS containing 106Cpn IFUs or material obtained from mock-infected HL cells, respectively. A and B, Histological analysis of representative 6-μm H&E-stained sections from the lungs of mock-infected (×200; A) and Cpn-infected (×100; B) mice 12 days postinoculation. C and D, Pulmonary sections from infected mice prepared 8 days postinoculation and immunostained with isotype-matched (IgG3) control mAb (×400; C) or Cpn-specific mAb RR-402 (×600; D; arrowheads point at some cells with positive intracellular staining for Cpn Ag). EH, Pooled LMNC isolated 12 days postinoculation from mock-infected (n = 8; E and F) and Cpn-infected (n = 4; G and H) mice were stimulated in vitro for 5 h with mock-infected (E and G) and Cpn-infected (F and H) mAM. GolgiPlug was added for the last 3.5 h of culture. After staining for CD8, LMNC were fixed, permeabilized, stained for intracellular IFN-γ, and analyzed by flow cytometry. Results are representative of three experiments.

FIGURE 1.

Detection of intracellular Cpn Ag and pathogen-specific IFN-γ-producing CD8+ T cells in the lungs of infected mice. Lungs from Cpn- and mock-infected B6 mice were obtained 8–12 days after i.n. inoculation with 40 μl PBS containing 106Cpn IFUs or material obtained from mock-infected HL cells, respectively. A and B, Histological analysis of representative 6-μm H&E-stained sections from the lungs of mock-infected (×200; A) and Cpn-infected (×100; B) mice 12 days postinoculation. C and D, Pulmonary sections from infected mice prepared 8 days postinoculation and immunostained with isotype-matched (IgG3) control mAb (×400; C) or Cpn-specific mAb RR-402 (×600; D; arrowheads point at some cells with positive intracellular staining for Cpn Ag). EH, Pooled LMNC isolated 12 days postinoculation from mock-infected (n = 8; E and F) and Cpn-infected (n = 4; G and H) mice were stimulated in vitro for 5 h with mock-infected (E and G) and Cpn-infected (F and H) mAM. GolgiPlug was added for the last 3.5 h of culture. After staining for CD8, LMNC were fixed, permeabilized, stained for intracellular IFN-γ, and analyzed by flow cytometry. Results are representative of three experiments.

Close modal

Analysis of LMNC isolated 12 days postinfection revealed that the percentage of CD8+ T cells in lungs from infected mice was almost twice that in lungs from mock-infected animals (Fig. 1, G and H vs E and F). When LMNC were cocultured for 5 h with mock-infected mAM, the percentage of CD8+ T cells that were IFN-γ+ was 6.5% in Cpn-infected animals ((1.1/17) × 100), but only 1.1% in mock-inoculated mice ((0.1/8.9) × 100; Fig. 1, G vs E). This level of IFN-γ production in pulmonary CD8+ T cells from infected mice may reflect specific activation in vivo. After coculture with Cpn-infected mAM, the percentage of CD8+ T cells expressing IFN-γ was 27.1% in cells from infected mice ((5/18.4) × 100), and only 2.2% in cells from mock-infected animals ((0.2/9.3) × 100; Fig. 1, H vs F). Coculture with Cpn-infected mAM elicited ∼4-fold increase in the percentage of CD8+ IFN-γ+ T cells (Fig. 1, H vs G). These data establish that Cpn-specific CD8+ T cells are present in the lungs from infected B6 mice and suggest that they may participate in the immune response against this pathogen by producing IFN-γ.

Based on the foregoing results, we asked: Which Cpn Ags reach the cytosol of infected cells and are recognized as MHC class I-peptide complexes by CD8+ T cells? From the available Cpn genome sequence databases (33, 34, 35) we selected 35 proteins that included outer and inclusion membrane proteins, chaperones, and selected proteins unique to Chlamydia or Cpn. In the sequences of these proteins we identified 461 segments of 8–9 aa conforming to the murine H-2Kb and Db class I binding motifs (36). Of these, 39 sequences from 16 proteins were selected for peptide synthesis (Table I). When the ability to bind to Kb and Db was examined by the RMA-S stabilization assay (31), 22 peptides were classified as high or intermediate binders (MFI ≥ 2.0). Using a quantitative molecular binding assay (32), which provides precise measurements of binding affinity that better correlate with CD8+ T cell immunogenicity (37), 17 peptides bound with high or intermediate affinity (IC50 ≤ 500) to purified Kb or Db molecules. Four additional peptides with affinities of 500 nM but ≤1000 nM were considered good binders, as CD8+ T cell immunogenicity has also been reported in this affinity range (38). Overall, 20 peptides were identified as high to intermediate binders by both binding assays (Table I).

All synthetic peptides were next assayed for their ability to target H-2b-bearing cells for lysis by SC from Cpn-infected mice obtained after culture with each individual peptide. Of the 39 peptides, 18 generated effector cells that specifically lysed RMA-S cells (H-2b; MHC class II) pulsed with the respective peptide, but not RMA-S cells pulsed with control OVA (Kb) and FLUnp (Db) CTL peptides (Fig. 2). The 18 peptides represent sequences in 12 Cpn Ags: five outer membrane proteins (omp), a 76-kDa protein, a family of four hypothetical 43-kDa proteins, heat shock protein-70, and an inclusion membrane protein (33, 34, 35, 39, 40, 41, 42, 43, 44) (Table I and Fig. 2). At the highest E:T cell ratio, net peptide-specific lysis for effectors stimulated with the 18 CTL peptides ranged from 10–27.9% after primary infection and from 10.4–34.2% after reinfection. As illustrated for a subset of peptides, CTL activity was detectable after primary infection (Fig. 2,A). For most of the 18 positive peptides, the in vitro recall CTL responses were slightly enhanced in reinfected mice, and absent in mock-infected animals (Fig. 2, B and C). Thus, most peptides capable of binding to H-2b molecules with high to intermediate affinity elicit functional CTL, and the polyclonal multi-Ag-specific CTL response is induced as a result of infection.

FIGURE 2.

Cpn infection primes a CTL response to multiple chlamydial Ags. SC from Cpn-infected (106 IFU) and mock-infected B6 mice were stimulated in vitro with H-2b motif-bearing peptides (2 μM) from pathogen-derived Ags. After 6 days effector cells were tested for recognition of RMA-S cells pulsed with homologous peptide, negative control Kb OVA or Db FLUnp peptides (0.1 μM), or no peptide in a CTL assay at the indicated E:T ratios. A, Lytic activity of immune SC from 40-day Cpn-infected mice (20 of 39 peptides are shown). B, Lytic activity of immune SC from 65-day Cpn-infected mice tested 1 mo after reinfection (106 IFU; 18 positive peptides and negative peptides 4 and 16 are shown). C, Lytic activity of SC from 80-day mock-infected mice tested 1 mo after the second inoculation (18 positive peptides and negative peptides 11 and 35 are shown).

FIGURE 2.

Cpn infection primes a CTL response to multiple chlamydial Ags. SC from Cpn-infected (106 IFU) and mock-infected B6 mice were stimulated in vitro with H-2b motif-bearing peptides (2 μM) from pathogen-derived Ags. After 6 days effector cells were tested for recognition of RMA-S cells pulsed with homologous peptide, negative control Kb OVA or Db FLUnp peptides (0.1 μM), or no peptide in a CTL assay at the indicated E:T ratios. A, Lytic activity of immune SC from 40-day Cpn-infected mice (20 of 39 peptides are shown). B, Lytic activity of immune SC from 65-day Cpn-infected mice tested 1 mo after reinfection (106 IFU; 18 positive peptides and negative peptides 4 and 16 are shown). C, Lytic activity of SC from 80-day mock-infected mice tested 1 mo after the second inoculation (18 positive peptides and negative peptides 11 and 35 are shown).

Close modal

To further characterize the lytic response to the positive Cpn CTL peptides, cell depletion and MHC class I restriction analyses were conducted. As illustrated for a subset of positive peptides, effector cells generated from SC of Cpn-reinfected mice lysed RMA-S cells pulsed with Cpn peptide but not with irrelevant peptides, and target cell lysis was eliminated by depleting effectors of CD8+, but not of CD4+ T cells (Fig. 3). T2 target cells pulsed with Kb-binding CTL peptides (e.g., peptides 19, 20, and 39) were lysed by the respective peptide-specific effectors only if the targets expressed Kb but not Db molecules. In contrast, effectors generated with Db-binding CTL peptides (e.g., peptides 27, 30, 31, and 37) lysed T2Db but not T2Kb cells sensitized with the homologous peptide (Fig. 3). These results demonstrate that SC from Cpn-infected B6 mice recognize pathogen-derived peptides in an Ag-specific, Kb- or Db-restricted, and CD8+ T cell-dependent manner.

FIGURE 3.

Recognition of Cpn peptides by splenic CTL is specific, Kb- or Db-restricted, and CD8+ T cell-dependent. Immune SC from 3-mo infected B6 mice were obtained 2 mo after reinfection (106 IFU) and cultured for 6 days with Cpn CTL peptides (2 μM). The lytic activity of undepleted effector cells was tested against RMA-S cells pulsed with the homologous CTL peptide or with negative control Kb OVA or Db FLUnp peptides (0.1 μM). T2Kb and T2Db target cells were used to define the H-2b class I restricting molecule. After depletion of CD4+ and CD8+ T cells, the cytotoxicity of depleted effectors was measured at the highest E:T ratio (50:1) against RMA-S sensitized with homologous peptide. Results for seven of 18 Cpn CTL peptides are shown.

FIGURE 3.

Recognition of Cpn peptides by splenic CTL is specific, Kb- or Db-restricted, and CD8+ T cell-dependent. Immune SC from 3-mo infected B6 mice were obtained 2 mo after reinfection (106 IFU) and cultured for 6 days with Cpn CTL peptides (2 μM). The lytic activity of undepleted effector cells was tested against RMA-S cells pulsed with the homologous CTL peptide or with negative control Kb OVA or Db FLUnp peptides (0.1 μM). T2Kb and T2Db target cells were used to define the H-2b class I restricting molecule. After depletion of CD4+ and CD8+ T cells, the cytotoxicity of depleted effectors was measured at the highest E:T ratio (50:1) against RMA-S sensitized with homologous peptide. Results for seven of 18 Cpn CTL peptides are shown.

Close modal

To further characterize Cpn peptide-specific T cells, SC from Cpn-infected mice were expanded by three weekly cycles of stimulation with each peptide and cell lines were evaluated for cytokine secretion profiles and surface phenotypes (Table II). During the first cycle, supernatants contained IFN-γ but not IL-4. After the second cycle, IFN-γ levels were 4- to 9-fold higher, but IL-4 levels remained undetectable. All the CTL peptides elicited significant TNF-α production. IFN-γ and TNF-α levels varied for each CTL peptide, but were significantly higher than the levels produced by stimulation with four Cpn peptides without CTL activity (Table II). All T cell lines were >90% CD3+CD8+TCRαβ+CD44high (Table II) and highly lytic to Cpn peptide-pulsed targets (data not shown). These results indicate that the 18 Cpn peptides stimulate memory CD8+ T cells in the spleens of infected mice displaying cytolytic function and a Tc1 cytokine secretion pattern.

Table II.

Chlamydia pneumoniae peptide-stimulated effector cells: cytokine and cell surface phenotype profiles

IDPeptideIFN-γ (ng/ml)aIL-4 (pg/ml)aTNF-α (pg/ml)aTCRαβ+CD8+CD44+b
Stimulation cycleStimulation cycleStimulation cycleStimulation cycle
121223 (%)
Cpn CTL peptides        
QESCYGRL 2.5 18.5 <8 <8 243.8 93.7 
ISVSNPGDL 1.9 10.8 <8 <8 198.4 93.1 
13 STEINQPFITM 12.5 61.2 <8 <8 597.1 98.9 
17 YASDNQAIL 3.0 15.4 <8 <8 206.3 96.8 
19 LVYNYPGV 5.1 24.6 <8 <8 281.0 97.0 
20 LIYNYPGV 4.3 24.8 <8 <8 275.2 97.7 
22 LIFNYPGV 2.8 11.5 <8 <8 211.9 90.0 
23 HPYLYRLL 3.0 13.9 <8 <8 187.7 90.2 
24 HPTLFKVL 2.8 18.1 <8 <8 266.5 93.3 
25 SIILFLPL 6.1 39.9 <8 <8 494.1 95.9 
27 ISNGNSDCL 3.2 26.1 <8 <8 289.9 94.1 
30 QAPTNRWML 5.2 30.3 <8 <8 328.6 95.8 
31 SLLGNATAL 1.9 13.3 <8 <8 185.3 90.7 
32 SHYAFSPMFEVL 2.9 19.9 <8 <8 292.0 94.0 
33 ISFAFCQL 3.2 30.1 <8 <8 338.7 92.8 
36 GTYHFTKL 4.9 22.6 <8 <8 303.8 97.6 
37 FQLCNSYDL 5.6 49.5 <8 <8 386.7 98.9 
39 LQQRYSRL 12.8 57.9 <8 <8 504.6 99.2 
Negative Cpn peptides        
11 TAGANPMDL 0.4 1.1 <8 <8 67.2 15.1 
12 ISANNDSEI 0.05 0.6 <8 <8 55.4 10.6 
16 LMSGFRQM <0.04 0.6 <8 <8 <40.0 11.7 
21 LLVFNYPGI 0.9 1.6 <8 <8 62.5 14.9 
        
TcTSA VDYNFTIV 5.3 38.4 <8 <8 354.3 96.2 
IDPeptideIFN-γ (ng/ml)aIL-4 (pg/ml)aTNF-α (pg/ml)aTCRαβ+CD8+CD44+b
Stimulation cycleStimulation cycleStimulation cycleStimulation cycle
121223 (%)
Cpn CTL peptides        
QESCYGRL 2.5 18.5 <8 <8 243.8 93.7 
ISVSNPGDL 1.9 10.8 <8 <8 198.4 93.1 
13 STEINQPFITM 12.5 61.2 <8 <8 597.1 98.9 
17 YASDNQAIL 3.0 15.4 <8 <8 206.3 96.8 
19 LVYNYPGV 5.1 24.6 <8 <8 281.0 97.0 
20 LIYNYPGV 4.3 24.8 <8 <8 275.2 97.7 
22 LIFNYPGV 2.8 11.5 <8 <8 211.9 90.0 
23 HPYLYRLL 3.0 13.9 <8 <8 187.7 90.2 
24 HPTLFKVL 2.8 18.1 <8 <8 266.5 93.3 
25 SIILFLPL 6.1 39.9 <8 <8 494.1 95.9 
27 ISNGNSDCL 3.2 26.1 <8 <8 289.9 94.1 
30 QAPTNRWML 5.2 30.3 <8 <8 328.6 95.8 
31 SLLGNATAL 1.9 13.3 <8 <8 185.3 90.7 
32 SHYAFSPMFEVL 2.9 19.9 <8 <8 292.0 94.0 
33 ISFAFCQL 3.2 30.1 <8 <8 338.7 92.8 
36 GTYHFTKL 4.9 22.6 <8 <8 303.8 97.6 
37 FQLCNSYDL 5.6 49.5 <8 <8 386.7 98.9 
39 LQQRYSRL 12.8 57.9 <8 <8 504.6 99.2 
Negative Cpn peptides        
11 TAGANPMDL 0.4 1.1 <8 <8 67.2 15.1 
12 ISANNDSEI 0.05 0.6 <8 <8 55.4 10.6 
16 LMSGFRQM <0.04 0.6 <8 <8 <40.0 11.7 
21 LLVFNYPGI 0.9 1.6 <8 <8 62.5 14.9 
        
TcTSA VDYNFTIV 5.3 38.4 <8 <8 354.3 96.2 
a

Levels of IFN-γ, IL-4, and TNF-α were determined by ELISA as described in Materials and Methods. Culture supernatants were harvested 48 h poststimulation with peptide (2 μM) and kept at −20°C until assayed. Negative and positive control supernatants were obtained from Cpn- and T. cruzi-infected mice using SC stimulated with Cpn peptides with no CTL activity and TcTSA, respectively. Supernatants from wells containing Cpn CTL peptide-stimulated naive SC contained minimal to undetectable levels of cytokines.

b

Effector cells purified at the end of the third cycle of stimulation with Cpn CTL peptides were analyzed by flow cytometry for surface expression of CD3, TCRαβ, CD8, CD4, and CD44 after staining with FITC- and PE-conjugated mAbs. Negative and positive control effector cell lines were SC from Cpn- and T. cruzi-infected mice stimulated for three cycles with Cpn peptides negative for CTL activity and TcTSA, respectively. Similar cytokine levels and cell surface phenotypes were obtained in two to four additional experiments.

To ascertain whether the 18 Cpn peptides represent CTL epitopes that are presented on the surface of infected cells, we used Cpn K6-infected mAM as targets for peptide-stimulated effectors. As shown for seven Cpn CTL peptides, effector cells generated after two cycles of stimulation with individual peptides lysed Cpn-infected but not mock-infected or BFA-treated infected mAM (Fig. 4). Short term CTL lines did not lyse mAM inoculated with heat-killed Cpn (Fig. 4). Similar results were obtained when mAM targets were treated with the Cpn strains AR39 and CWL029 (data not shown). Thus, all 12 Cpn CTL Ags are processed by the endogenous pathway, and the 18 epitope-bearing peptides are presented to CTL only in productively infected cells.

FIGURE 4.

Infected macrophages are targeted for lysis by Cpn peptide-specific CD8+ CTL. Immune SC from Cpn-infected B6 mice inoculated 6 and 4 mo (106 IFU each dose) before sacrifice were stimulated for two 6-day cycles with CTL peptides (2 μM). At the end of the second cycle, effectors (79–86% CD8+) were tested for their ability to kill mAM inoculated 22 h earlier with live Cpn (□), HL cell-derived material (⋄), or heat-killed bacteria (▴). Lytic activity of effector cells was also tested against BFA-treated Cpn-infected mAM (○). BFA (10 μg/ml) was added for the last 2 h of infection and during the CTL assay. Results for seven of 18 Cpn peptide-specific, short term CTL lines are shown.

FIGURE 4.

Infected macrophages are targeted for lysis by Cpn peptide-specific CD8+ CTL. Immune SC from Cpn-infected B6 mice inoculated 6 and 4 mo (106 IFU each dose) before sacrifice were stimulated for two 6-day cycles with CTL peptides (2 μM). At the end of the second cycle, effectors (79–86% CD8+) were tested for their ability to kill mAM inoculated 22 h earlier with live Cpn (□), HL cell-derived material (⋄), or heat-killed bacteria (▴). Lytic activity of effector cells was also tested against BFA-treated Cpn-infected mAM (○). BFA (10 μg/ml) was added for the last 2 h of infection and during the CTL assay. Results for seven of 18 Cpn peptide-specific, short term CTL lines are shown.

Close modal

To determine whether Cpn peptide-specific CD8+ CTL were present in the lungs of infected mice, we purified LMNC 12 days after reinfection and stimulated them with Cpn CTL peptides. Six days later, the lytic activity of undepleted and CD8+ T cell-depleted effectors was tested against peptide-pulsed target cells. For five representative peptides, each from a different Cpn CTL target Ag, resulting effectors displayed CD8+ T cell-dependent CTL activity against RMA-S cells sensitized with homologous peptide, but not against the same cells pulsed with control CTL peptides or with no peptide (Fig. 5,A). As shown for a Db- and a Kb-restricted Cpn CTL peptide, T cell effectors generated after four cycles of stimulation were highly enriched for peptide-specific, H-2b-restricted lytic activity and a CD8+ CD44high memory phenotype. Moreover, these LMNC-derived CTL lines lysed mAM inoculated with live AR39 Cpn bacteria, but not with heat-killed organisms unless they were also sensitized with peptide (Fig. 5 B). These results demonstrate that Cpn-specific CD8+ CTL are present in the lungs of infected animals and further support the natural presentation of identified epitopes by Cpn-infected cells.

FIGURE 5.

Presence of Cpn peptide-specific CD8+ CTL in the lungs of infected mice. Immune LMNC from 2-mo infected B6 mice were obtained 12 days after reinfection (106 IFU) and cultured for four 6-day cycles with Cpn CTL peptides (2 μM). A. At the end of the first cycle, the lytic activity of undepleted effector cells was tested against RMA-S (H-2b) cells pulsed with the homologous CTL peptide or with negative control Kb OVA or Db FLUnp peptides (0.1 μM) in a CTL assay at the indicated E:T ratios. The cytotoxicity of CD8+ T cell-depleted effectors and complement-treated controls was measured at the highest E:T ratio (28:1) against RMA-S sensitized with homologous peptide. B, After four cycles of stimulation, effector cells were analyzed for surface expression of CD8 and CD44 by flow cytometry and for their ability to lyse the indicated target cells inoculated 22 h earlier with live or heat-killed Cpn AR39 bacteria or were sensitized for 16 h with CTL or control peptide (0.1 μM). Results for five (A) and two (B) of 18 Cpn peptide-specific CTL lines are shown.

FIGURE 5.

Presence of Cpn peptide-specific CD8+ CTL in the lungs of infected mice. Immune LMNC from 2-mo infected B6 mice were obtained 12 days after reinfection (106 IFU) and cultured for four 6-day cycles with Cpn CTL peptides (2 μM). A. At the end of the first cycle, the lytic activity of undepleted effector cells was tested against RMA-S (H-2b) cells pulsed with the homologous CTL peptide or with negative control Kb OVA or Db FLUnp peptides (0.1 μM) in a CTL assay at the indicated E:T ratios. The cytotoxicity of CD8+ T cell-depleted effectors and complement-treated controls was measured at the highest E:T ratio (28:1) against RMA-S sensitized with homologous peptide. B, After four cycles of stimulation, effector cells were analyzed for surface expression of CD8 and CD44 by flow cytometry and for their ability to lyse the indicated target cells inoculated 22 h earlier with live or heat-killed Cpn AR39 bacteria or were sensitized for 16 h with CTL or control peptide (0.1 μM). Results for five (A) and two (B) of 18 Cpn peptide-specific CTL lines are shown.

Close modal

We initially observed that LMNC from Cpn-infected B6 mice contained CD8+ T cells capable of rapidly producing IFN-γ in response to Cpn-infected cells (Fig. 1, EH). Once Cpn CTL determinants became defined, we asked whether epitope-specific CD8+ T cells were detectable in the lungs and spleens of infected mice without stimulation, and if so, we determined precursor frequencies of these cells during primary and secondary Cpn infections. Immune LMNC and SC from B6 mice were obtained 11 days after infection or reinfection and were stimulated for 24 h with peptide-pulsed and unpulsed RMA-S cells, followed by enumeration of IFN-γ secreting CD8+ T cells by ELISPOT. We measured the CD8+ T cell precursor frequencies for peptides 13, 37, and 39, which were associated with the highest IFN-γ production (Table II). The mean number of LMNC isolated at 11 days postinfection (5.3 × 106 ± 8 × 105) or reinfection (6.7 × 106 ± 1.4 × 106) was 6- to 7-fold higher than the yields obtained for mock-infected animals (9 × 105 ± 3 × 105). Based on these data and the frequencies of IFN-γ spot-forming cells (SFC) obtained by ELISPOT, the lungs of infected animals contained an estimated 165, 145, and 250 peptide 13-, 37-, and 39-specific IFN-γ-producing CD8+ T cells, respectively. For each peptide the number of IFN-γ SFCs was 2- to 2.5-fold higher after reinfection (Fig. 6,A). Parallel studies with SC showed that the precursor frequencies of peptide 13-, 37-, and 39-specific CD8+ T cells were 2- to 5-fold higher after reinfection compared with primary infection (69–172 vs 20–39 IFN-γ SFCs/106 cells; Fig. 6 B). IFN-γ secretion was specific, as spots were rarely observed when immune cells were incubated with unpulsed RMA-S cells. Furthermore, peptide-specific IFN-γ SFCs were not detected using cells from mock-infected animals. Thus, Cpn CTL peptide-specific, IFN-γ-secreting CD8+ T cells are present in local and systemic compartments after primary chlamydial infection, and these cells expand after reinfection.

FIGURE 6.

Enumeration of pulmonary and splenic Cpn CTL peptide-specific IFN-γ-producing CD8+ T cells by ex vivo ELISPOT. Cpn-infected (n = 5) and mock-infected (n = 25) B6 mice were sacrificed 11 days after each of two i.n. inoculations with 106Cpn IFUs or with material obtained from mock-infected HL cells, respectively. A and B, Irradiated RMA-S cells pulsed (1 μM), or not, with the indicated CTL peptides were cocultured for 24 h with pooled isolated LMNC (A) and SC (B) under the conditions described in Materials and Methods, and the number of specific IFN-γ SFC (CD8+ T cells) was quantified. Values represent the mean ± SEM of triplicate determinations from a representative of three experiments.

FIGURE 6.

Enumeration of pulmonary and splenic Cpn CTL peptide-specific IFN-γ-producing CD8+ T cells by ex vivo ELISPOT. Cpn-infected (n = 5) and mock-infected (n = 25) B6 mice were sacrificed 11 days after each of two i.n. inoculations with 106Cpn IFUs or with material obtained from mock-infected HL cells, respectively. A and B, Irradiated RMA-S cells pulsed (1 μM), or not, with the indicated CTL peptides were cocultured for 24 h with pooled isolated LMNC (A) and SC (B) under the conditions described in Materials and Methods, and the number of specific IFN-γ SFC (CD8+ T cells) was quantified. Values represent the mean ± SEM of triplicate determinations from a representative of three experiments.

Close modal

IFN-γ contributes to protection against Cpn infection (21, 22) and inhibits chlamydial growth in vitro (45). Our results suggested that Cpn-specific CD8+ T cells could participate in protective immunity through secretion of IFN-γ and/or killing of infected cells. To evaluate the ability of CD8+ T cell-derived soluble factors to inhibit bacterial growth, we counted the number of Cpn inclusions in infected HEp-2 cells that had been treated with supernatants collected from CD8+ T cell cultures (Fig. 7 A). As shown for peptides 13, 20, 37, and 39, undiluted culture supernatants from peptide-specific CD8+ T cells inhibited inclusion development by 95–100% compared with undiluted culture supernatant from Cpn peptide-stimulated naive CD8+ T cells. Serial dilutions of the supernatants resulted in 70–92% (1:5) and 8–51% (1:25) inhibition, indicating a dose-dependent effect on chlamydial growth. Addition of anti-IFN-γ mAb to the undiluted supernatants from peptide-specific CD8+ T cells reduced inhibition by 18–65%. No reversal of inhibition was observed using control mAb (data not shown). Supernatants from cultures containing peptide alone or peptide-stimulated naive CD8+ T cells failed to inhibit chlamydial multiplication. Minimal inhibition was detected using supernatant from unstimulated peptide-specific CD8+ T cells, perhaps due to accumulation of inhibitory factors during the 48 h before collection of the supernatants for testing. Thus, Cpn-specific CD8+ CTL suppress chlamydial growth by production of IFN-γ and other unidentified soluble factors.

FIGURE 7.

Inhibition of chlamydial multiplication by soluble factors and cells from Cpn peptide-specific CD8+ CTL lines. A, Culture supernatants from Cpn-immune LMNC-derived CD8+ T cell lines were obtained 48 h into the third cycle of stimulation with the indicated CTL peptides (2 μM) and irradiated syngeneic SC. Serial dilutions and undiluted supernatants preincubated, or not, with anti-IFN-γ neutralizing or isotype-matched mAbs (20 μg/ml) were used to pretreat triplicate HEp-2 cell cultures for 18 h before Cpn infection. Infected monolayers were treated for 24 h with new samples of the same supernatants and cultured for an additional 48 h in Chlamydia medium. Cultures were fixed and stained with a Chlamydia genus-specific mAb, and the number of developing chlamydial inclusions in 10 HPF/well was counted. Supernatants collected from cultures lacking CD8+ T cells, peptide, or both were also tested. All results were compared with those obtained with the respective dilution of culture supernatant from immunomagnetically purified naive CD8+ T cells stimulated with a mixture of the indicated CTL peptides. B, The indicated numbers of CD8+ T cells obtained at the end of the third cycle of stimulation with Cpn CTL peptide were cocultured for 4 h with Cpn-infected mAM monolayers. Washed cultures were incubated for an additional 48 h and passed into fresh HEp-2 cell monolayers as described in Materials and Methods. After 72 h, the number of inclusions in 10 HPF/well was counted. T. cruzi TSA-1515 (TcTSA)-specific CD8+ CTL and purified naive CD8+ T cells were used as control effector cells. Values represent the mean ± SEM. Data are representative of three experiments.

FIGURE 7.

Inhibition of chlamydial multiplication by soluble factors and cells from Cpn peptide-specific CD8+ CTL lines. A, Culture supernatants from Cpn-immune LMNC-derived CD8+ T cell lines were obtained 48 h into the third cycle of stimulation with the indicated CTL peptides (2 μM) and irradiated syngeneic SC. Serial dilutions and undiluted supernatants preincubated, or not, with anti-IFN-γ neutralizing or isotype-matched mAbs (20 μg/ml) were used to pretreat triplicate HEp-2 cell cultures for 18 h before Cpn infection. Infected monolayers were treated for 24 h with new samples of the same supernatants and cultured for an additional 48 h in Chlamydia medium. Cultures were fixed and stained with a Chlamydia genus-specific mAb, and the number of developing chlamydial inclusions in 10 HPF/well was counted. Supernatants collected from cultures lacking CD8+ T cells, peptide, or both were also tested. All results were compared with those obtained with the respective dilution of culture supernatant from immunomagnetically purified naive CD8+ T cells stimulated with a mixture of the indicated CTL peptides. B, The indicated numbers of CD8+ T cells obtained at the end of the third cycle of stimulation with Cpn CTL peptide were cocultured for 4 h with Cpn-infected mAM monolayers. Washed cultures were incubated for an additional 48 h and passed into fresh HEp-2 cell monolayers as described in Materials and Methods. After 72 h, the number of inclusions in 10 HPF/well was counted. T. cruzi TSA-1515 (TcTSA)-specific CD8+ CTL and purified naive CD8+ T cells were used as control effector cells. Values represent the mean ± SEM. Data are representative of three experiments.

Close modal

To examine whether Cpn peptide-specific CD8+ T cells directly inhibit chlamydial growth, we evaluated inclusion development in HEp-2 cell monolayers inoculated with lysates of chlamydia-infected mAM that had been briefly cocultured with CD8+ T cells (Fig. 7 B). At high CD8+ T cell densities, Cpn peptide 13-, 20-, 37-, and 39-specific CTL inhibited 94–100% the number of Cpn inclusions that developed in the presence of naive CD8+ T cells, and the extent of inhibition decreased with decreasing numbers of Cpn peptide-specific CD8+ T cells. Significant inhibition was observed for all Cpn peptide-specific CD8+ CTL lines compared with studies in the absence of CD8+ T cells or in the presence of irrelevant T. cruzi TSA-specific CD8+ T cells. These data indicate that Cpn-specific CD8+ CTL decrease the output of viable organisms from infected cells.

Like all chlamydiae, Cpn has an intracellular biphasic developmental cycle, alternating between the infectious, nonreplicating, and relatively metabolically inert elementary body and the noninfectious, replicating, and metabolically active reticulate body. Hundreds of infectious progeny are released from each infected cell. Like C. trachomatis and C. psittaci, Cpn can persist for long periods in infected cells without dividing before resuming a productive cycle (46, 47). Thus, the induction of an early potent immune response that eliminates Chlamydia-infected cells or renders developing bacteria noninfectious would reduce the pathogen’s ability to grow and spread in the infected host. Herein, we provide the first evidence that Cpn-infected mice mount pathogen-specific CD8+ T cell responses that exhibit potent anti-chlamydial activity through lysis of infected cells and secretion of soluble factors, including IFN-γ. We also report the identification of 18 H-2b-restricted Tc1 epitope-bearing sequences in the first 12 bona fide Cpn target Ags of the CD8+ T cell response induced in infected mice.

Development of vaccines that induce CD8+ T cell responses against Cpn requires knowledge of the mechanisms by which these cells mediate resistance to infection. Equally important for the design of anti-Cpn vaccines is to determine the response kinetics and functional attributes of CD8+ T cells found in the lungs and systemic compartments of Cpn-infected hosts and whether these cells associate exclusively or collectively with protection, persistence, or immunopathology. However, understanding Cpn-specific CD8+ T cell responses and designing vaccines to maximize protection and minimize tissue damage are contingent upon the ability to assess such responses and on the identification of the Cpn Ags or epitopes that are recognized by CD8+ T cells. As a first step to characterize the anti-Cpn CD8+ T cell response, we structured a strategy based on three components: 1) the mouse model of Cpn infection, which is an excellent system to study the immune mechanisms thought to control this pathogen in humans; 2) the Cpn genome sequence databases, from which putative target Ags of CD8+ T cells were selected; and 3) the Cpn-competent mAM cell line, which served to stimulate and detect the activity of Cpn-specific CD8+ T cells. We selected the macrophage because Cpn disseminates from the respiratory tract by infecting this cell (48), and a similar mechanism may permit this pathogen to spread systemically in humans.

We first asked whether chlamydial infection primes Cpn-specific CD8+ T cells in the lungs of infected B6 mice. To select a time point when we could detect T cells, we conducted a kinetic analysis of histopathologic changes and bacterial loads in the lungs of Cpn-infected mice. From this analysis we determined that mononuclear cells are the main inflammatory cell type during infection, peak inflammatory responses are stronger and emerge faster after reinfection than after primary infection, and fewer pulmonary bacteria are recovered from Cpn-reinfected mice than from infected animals. These results indicated that Cpn infection induces immunoprotective responses. Because Cpn MOMP+ foamy mononuclear cells were seen adjacent to lymphoid cells composing the inflammatory foci, we postulate that these consist of macrophages priming pathogen-specific CD8+ T cells. Indeed, LMNC from 12-day infected mice contained Cpn-specific CD8+ T cells, and nearly one-fourth of the infiltrating CD8+ T cells produced IFN-γ upon stimulation with Cpn-infected mAM. Furthermore, CD8+ T cells play a key role in early and late resistance to Cpn infection (20, 21). Early in the infection, CD8+ T cells are thought to contribute to protection by modifying the CD4+ T cell cytokine pattern from a Th2 to a protective Th1 phenotype (20). We speculate that Cpn-specific IFN-γ-producing CD8+ T cells contribute to the type 1 cytokine milieu required to control early Cpn growth in the lungs of infected B6 mice.

We next conducted a genome-wide search for sequences encoding potential target Ags of murine anti-Cpn CD8+ T cells, selecting Cpn proteins that we believed were most likely to reach the cytosol of infected cells for processing and MHC class I presentation. These included membrane proteins specific to Chlamydia or with relatives in other organisms, chaperones, and Ags unique to Cpn, or with orthologs in other chlamydial species. Within the primary sequences of 16 Cpn Ags, 39 of 146 Kb- or Db motif-fitting segments were synthesized as peptides and tested for the capacity to sensitize target cells for lysis by peptide-stimulated SC and LMNC from Cpn-infected mice. Remarkably, a specific, H-2b-restricted, and CD8+ T cell-dependent CTL response was detected for 18 peptides in 12 Cpn Ags. Of the 18 CD8+ CTL epitope-bearing peptides, seven represent sequences within five outer membrane protein complex Ags, namely, Omp2, MOMP, OmpB, Omp85 homolog, and Omp5 (Pmp10) (33, 34, 35, 39, 43, 44). In the group of four chaperones, only a peptide from DnaK (41) elicited recall CD8+ CTL responses. Interestingly, DnaK is associated with the outer membranes of C. trachomatis (49). Finally, in the group of Cpn Ags with or without chlamydial orthologs, eight CTL peptides represent sequences within a family of four 43-kDa protein homologs, one within a C. trachomatis-like 76-kDa protein and another within an inclusion membrane protein similar to C. psittaci IncA (33, 34, 35, 42). The 43- and 76-kDa proteins may also be surface exposed, as Cpn Ags of these molecular masses are recognized by human and rabbit immune sera, the 43-kDa primary sequences revealed potential transmembrane domains, and a specific anti-76-kDa antisera neutralizes Cpn infectivity in vitro (42, 50, 51).

Aside from macrophages, Cpn can infect and multiply in nonprofessional APC, including airway epithelial, endothelial, and smooth muscle cells (52, 53). However, macrophages and dendritic cells are the APC most likely responsible for priming a Cpn-specific CD8+ CTL response. Because in addition to the endogenous MHC class I pathway, professional APC can process and present exogenous Ag to CD8+ T cells (54), detected Cpn-specific CTL could have been primed by processed Ags from phagocytosed bacteria and not from chlamydial Ags accessing the cytosol of infected APC. Arguing against this possibility was the finding that all 18 peptide-specific CD8+ CTL effectors killed mAM only when inoculated with live Cpn bacteria. Moreover, the fact that BFA inhibited this lytic activity strongly suggested that the 12 Cpn Ags access the endogenous MHC class I processing machinery. The cytotoxic mechanism that prevailed in the killing of Cpn-infected mAM was probably not Fas/Fas ligand dependent, as BFA inhibits this pathway in CTL effectors (55). BFA did not affect Cpn growth, a finding consistent with that described in a study in which the endogenous Ag processing pathway was shown to target C. trachomatis-infected fibroblasts for lysis by pathogen-specific CD8+ CTL (11). In support of our findings, C. trachomatis MOMP and the inclusion membrane protein Cap1 are the targets of infection-primed CD8+ CTL that kill C. trachomatis-infected nonprofessional APC (15, 16). Thus, although our results do not rule out the contribution of alternative pathways to the processing of Cpn Ags by professional APC, they imply that CD8+ CTL primed by endogenously processed Ags are the most likely to recognize all Cpn-infected cells.

Despite their intravacuolar location, chlamydiae interact with multiple host cell processes to ensure that the inclusion is a safe niche for their survival and replication. These interactions are needed to acquire nutrients, avoid fusion with lysosomes, obtain membrane components from Golgi-derived exocytic vesicles, and modify host cell functions (56, 57). The chlamydial products that control these processes are thought to be proteins translocated through or inserted into the inclusion membrane via a type III secretion apparatus (58). However, of the 12 identified Cpn CTL Ags, none is known to be secreted into the host cell cytosol, and only one, the Ag similar to C. psittaci IncA, is likely to belong to the Inc family of inclusion membrane proteins. Identified Inc proteins have a unique 50- to 60-aa hydrophobic region and domains that localize to the cytoplasmic face of the inclusion (59, 60, 61). These exposed domains may be cleaved by proteases in the cytosol or on the membranes of interacting vesicles, and after proteasomal processing, Inc-derived CTL epitopes, such as peptide 39, may be generated. How do the Cpn CTL target Ags located or associated with the envelopes of developing chlamydiae become accessible to the endogenous MHC class I presentation pathway? As no report has localized Cpn envelope Ags in the cytosol of infected cells, we speculate that these and other envelope Ags may reach the cytosol in a denatured or preprocessed form. Unfolded envelope Ags or fragments thereof may arise during the extensive membrane remodeling that occurs during chlamydial replication and differentiation, from the “ghost-like” membranous material present alongside organisms within a typical inclusion (47), or from a small fraction of developing chlamydiae undergoing autolysis.

There is little variation in the sequences of various genes from multiple Cpn isolates, and nearly identical sequences were reported for the genomes of Cpn strains CWL029 and AR39 from the United States and J138 from Japan (33, 34, 35). Our CTL data are in line with these findings, as all 18 peptide-specific CD8+ T cell effectors were generated from mice infected with the Finnish K6 Cpn strain, and these CTL lysed mAM infected with Cpn strains K6, AR39, or CWL029. These results are encouraging, as the CTL Ags or epitopes included in vaccines against Cpn will require this level of sequence conservation among strains from different geographic locations. Moreover, because 80% of the predicted coding sequences for Cpn have an ortholog in C. trachomatis, and, on average, orthologs from the two species share 62% aa identity (33), some CTL epitopes may be conserved between orthologous chlamydial Ags. Indeed, the sequence of Cpn Omp2-derived peptide 3 is identical with its ortholog in C. trachomatis. In contrast, the six Cpn CTL peptides derived from the 76-kDa protein, Omp85 homolog, DnaK, OmpB, and MOMP differed by one to eight residues from the respective sequences in the orthologous C. trachomatis Ags. Nevertheless, these Ags could still be the targets of CTL responses during human or experimental C. trachomatis infections, as recently shown for C. trachomatis MOMP (16).

Lung-derived Cpn peptide-specific CD8+ T cells lysed Cpn-infected cells, expressed a CD44high memory phenotype and produced Tc1 cytokines. IFN-γ is crucial for the control of chlamydial infections (21, 22, 62), and IFN-γ-producing CD44high CD8+ CTL mediate protection in murine models of M. tuberculosis and T. cruzi infection (24, 63). Similarly, we found that peptide-specific CD8+ T cells can inhibit Cpn inclusion development by producing soluble factors that limit Cpn growth and by direct effects of CD8+ T cells themselves on the infectious titers produced by Cpn-infected macrophages. The capacity of supernatants to suppress Cpn growth correlated with the levels of IFN-γ. However, neutralization of IFN-γ only partially restored inclusion development, suggesting that other factors inhibit Cpn multiplication. Cpn peptide-specific CD8+ T cells produce TNF-α, which may contribute to the suppressive effect, as it synergizes with IFN-γ in inhibiting Cpn growth (45). Cpn multiplication may also be influenced by other CD8+ T cell-derived factors, including RANTES, macrophage inflammatory proteins-1α and -1β, and IL-16, which suppress HIV replication (64, 65). The anti-chlamydial activity of purified Cpn peptide-specific CD8+ T cells was probably due to direct lysis of Cpn-infected mAM. Because most organisms released by lysed mAM were most likely in the form of noninfectious reticulate bodies, Cpn viability may have been affected by the release of molecules with antimicrobial activity, similar to the human CD8+ T cell-derived granulysin (66). This possibility is being evaluated using Cpn-infected cells, where organisms have differentiated into infectious elementary bodies. Altogether these results suggest that Cpn-specific CD8+ T cells may also inhibit Cpn growth in vivo through the combined actions of both effector functions. We are conducting adoptive transfer experiments to determine whether CD8+ T cells to each of the identified Cpn CTL epitopes can reduce pulmonary chlamydial loads in Cpn-challenged mice and to define the mechanisms by which CTL protect against infection.

Although CD8+ CTL might protect against Cpn, they may also be partially responsible for inducing persistent infection and immunopathology. IFN-γ may have a dual role in controlling the outcome of chlamydial infections in vivo, as exposure of infected cells to high IFN-γ concentrations irreversibly inhibits chlamydial replication, while lower concentrations induce the formation of persistent forms (45). Thus, production of IFN-γ by Cpn-specific CD8+ T cells may be critical to achieve the local threshold concentration required to inhibit Cpn growth and prevent persistence. Persistent Cpn infection in humans and rodents (3, 19) may be due to a suboptimal CD8+ T cell response during the course of an active Cpn infection. Nevertheless, we believe that an efficacious anti-Cpn CD8+ CTL response can be induced through vaccination. Vaccines against Cpn will most likely include multiple determinants from various chlamydial Ags that induce CD8+ CTL capable of eliminating productively and possibly persistently infected cells without causing serious tissue inflammation. In the current studies we provide direction for the development of such vaccines, as we have demonstrated the generation of Cpn-specific MHC class I-restricted CD8+ CTL with anti-chlamydial growth activity in infected mice and have identified 12 CTL target molecules. Work is in progress to further delineate the dynamics of the murine CD8+ T cell response to the Cpn CTL epitope-bearing sequences described in this study and to determine whether Cpn-infected humans also generate CTL responses against the Ags identified using the murine model. Such information is important to validate the use of the mouse model to provide insight into the aspects of the CTL response that contribute to protection and those that mediate immunopathology, and to identify and test CTL target Ags that may ultimately be used to develop vaccines against Cpn.

1

This work was supported by the National Institutes of Health (RO1 HL70641-01), the American Heart Association Texas Affiliate, the Center for Pulmonary and Infectious Disease Control, and the Cain Foundation for Infectious Disease Research. B.W. is a recipient of the Lyndon Baines Johnson Research Award from the American Heart Association.

3

Abbreviations used in this paper: Cpn, Chlamydia pneumoniae; BFA, brefeldin A; HPF, high powered field; IFU, inclusion-forming units; i.n., intranasal; LMNC, lung mononuclear cell; mAM, murine alveolar macrophage; MFI, mean fluorescence intensity; MOMP, major outer membrane protein; RCAS, rat Con A supernatant; SC, spleen cell; SFC, spot-forming cell; TCM, T cell medium; TSA, trypomastigote surface Ag.

1
Grayston, J. T..
2000
. Background and current knowledge of Chlamydiapneumoniae and atherosclerosis.
J. Infect. Dis.
181
: (Suppl. 3):
S402
2
Kuo, C. C., L. A. Jackson, L. A. Campbell, J. T. Grayston.
1995
. Chlamydia pneumoniae (TWAR).
Clin. Microbiol. Rev.
8
:
451
3
Hammerschlag, M. R., K. Chirgwin, P. M. Roblin, M. Gelling, W. Dumornay, L. Mandel, P. Smith, J. Schachter.
1992
. Persistent infection with Chlamydiapneumoniae following acute respiratory illness.
Clin. Infect. Dis.
14
:
178
4
Hahn, D. L., R. W. Dodge, R. Golubjatnikov.
1991
. Association of Chlamydiapneumoniae (strain TWAR) infection with wheezing, asthmatic bronchitis, and adult-onset asthma.
JAMA
266
:
225
5
Von Hertzen, L., H. Alakarppa, R. Koskinen, K. Liippo, H. M. Surcel, M. Leinonen, P. Saikku.
1997
. Chlamydia pneumoniae infection in patients with chronic obstructive pulmonary disease.
Epidemiol. Infect.
118
:
155
6
Sriram, S., C. W. Stratton, S. Yao, A. Tharp, L. Ding, J. D. Bannan, W. M. Mitchell.
1999
. Chlamydia pneumoniae infection of the central nervous system in multiple sclerosis.
Ann. Neurol.
46
:
6
7
Saikku, P., M. Leinonen, K. Mattila, M. R. Ekman, M. S. Nieminen, P. H. Makela, J. K. Huttunen, V. Valtonen.
1988
. Serological evidence of an association of a novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction.
Lancet
2
:
983
8
Elkind, M. S., I. F. Lin, J. T. Grayston, R. L. Sacco.
2000
. Chlamydia pneumoniae and the risk of first ischemic stroke: The Northern Manhattan Stroke Study.
Stroke
31
:
1521
9
Gieffers, J., H. Fullgraf, J. Jahn, M. Klinger, K. Dalhoff, H. A. Katus, W. Solbach, M. Maass.
2001
. Chlamydia pneumoniae infection in circulating human monocytes is refractory to antibiotic treatment.
Circulation
103
:
351
10
Lalvani, A., R. Brookes, R. J. Wilkinson, A. S. Malin, A. A. Pathan, P. Andersen, H. Dockrell, G. Pasvol, A. V. Hill.
1998
. Human cytolytic and interferon γ-secreting CD8+ T lymphocytes specific for Mycobacterium tuberculosis.
Proc. Natl. Acad. Sci. USA
95
:
270
11
Beatty, P. R., R. S. Stephens.
1994
. CD8+ T lymphocyte-mediated lysis of Chlamydia-infected L cells using an endogenous antigen pathway.
J. Immunol.
153
:
4588
12
Starnbach, M. N., M. J. Bevan, M. F. Lampe.
1994
. Protective cytotoxic T lymphocytes are induced during murine infection with Chlamydia trachomatis.
J. Immunol.
153
:
5183
13
Buzoni-Gatel, D., L. Guilloteau, F. Bernard, S. Bernard, T. Chardes, A. Rocca.
1992
. Protection against Chlamydiapsittaci in mice conferred by Lyt-2+ T cells.
Immunology
77
:
284
14
Igietseme, J. U., D. M. Magee, D. M. Williams, R. G. Rank.
1994
. Role for CD8+ T cells in antichlamydial immunity defined by Chlamydia-specific T-lymphocyte clones.
Infect. Immun.
62
:
5195
15
Fling, S. P., R. A. Sutherland, L. N. Steele, B. Hess, S. E. D’Orazio, J. Maisonneuve, M. F. Lampe, P. Probst, M. N. Starnbach.
2001
. CD8+ T cells recognize an inclusion membrane-associated protein from the vacuolar pathogen Chlamydia trachomatis.
Proc. Natl. Acad. Sci. USA
98
:
1160
16
Kim, S. K., M. Angevine, K. Demick, L. Ortiz, R. Rudersdorf, D. Watkins, R. DeMars.
1999
. Induction of HLA class I-restricted CD8+ CTLs specific for the major outer membrane protein of Chlamydia trachomatis in human genital tract infections.
J. Immunol.
162
:
6855
17
Yang, Z. P., C. C. Kuo, J. T. Grayston.
1993
. A mouse model of Chlamydiapneumoniae strain TWAR pneumonitis.
Infect. Immun.
61
:
2037
18
Penttila, J. M., M. Anttila, M. Puolakkainen, A. Laurila, K. Varkila, M. Sarvas, P. H. Makela, N. Rautonen.
1998
. Local immune responses to Chlamydiapneumoniae in the lungs of BALB/c mice during primary infection and reinfection.
Infect. Immun.
66
:
5113
19
Malinverni, R., C. C. Kuo, L. A. Campbell, J. T. Grayston.
1995
. Reactivation of Chlamydiapneumoniae lung infection in mice by cortisone.
J. Infect. Dis.
172
:
593
20
Penttila, J. M., M. Anttila, K. Varkila, M. Puolakkainen, M. Sarvas, P. H. Makela, N. Rautonen.
1999
. Depletion of CD8+ cells abolishes memory in acquired immunity against Chlamydiapneumoniae in BALB/c mice.
Immunology
97
:
490
21
Rottenberg, M. E., A. C. Gigliotti Rothfuchs, D. Gigliotti, C. Svanholm, L. Bandholtz, H. Wigzell.
1999
. Role of innate and adaptive immunity in the outcome of primary infection with Chlamydiapneumoniae, as analyzed in genetically modified mice.
J. Immunol.
162
:
2829
22
Rottenberg, M. E., A. Gigliotti Rothfuchs, D. Gigliotti, M. Ceausu, C. Une, V. Levitsky, H. Wigzell.
2000
. Regulation and role of IFN-γ in the innate resistance to infection with Chlamydiapneumoniae.
J. Immunol.
164
:
4812
23
Kuo, C. C., J. T. Grayston.
1990
. A sensitive cell line, HL cells, for isolation and propagation of Chlamydiapneumoniae strain TWAR.
J. Infect. Dis.
162
:
755
24
Wizel, B., M. Nunes, R. Tarleton.
1997
. Identification of Trypanosoma cruzi trans-sialidase family members as targets of protective CD8+ TC1 responses.
J. Immunol.
159
:
6120
25
Ljunggren, H. G., K. Karre.
1985
. Host resistance directed selectively against H-2-deficient lymphoma variants: analysis of the mechanism.
J. Exp. Med.
162
:
1745
26
Alexander, J., J. A. Payne, R. Murray, J. A. Frelinger, P. Cresswell.
1989
. Differential transport requirements of HLA and H-2 class I glycoproteins.
Immunogenetics
29
:
380
27
Puolakkainen, M., J. Parker, C. C. Kuo, J. T. Grayston, L. A. Campbell.
1995
. Further characterization of Chlamydiapneumoniae specific monoclonal antibodies.
Microbiol. Immunol.
39
:
551
28
Wolf, K., E. Fischer, D. Mead, G. Zhong, R. Peeling, B. Whitmire, H. D. Caldwell.
2001
. Chlamydia pneumoniae major outer membrane protein is a surface-exposed antigen that elicits antibodies primarily directed against conformation-dependent determinants.
Infect. Immun.
69
:
3082
29
Rotzschke, O., K. Falk, S. Stevanovic, G. Jung, P. Walden, H. G. Rammensee.
1991
. Exact prediction of a natural T cell epitope.
Eur. J. Immunol.
21
:
2891
30
Townsend, A. R., J. Rothbard, F. M. Gotch, G. Bahadur, D. Wraith, A. J. McMichael.
1986
. The epitopes of influenza nucleoprotein recognized by cytotoxic T lymphocytes can be defined with short synthetic peptides.
Cell
44
:
959
31
Ljunggren, H. G., N. J. Stam, C. Ohlen, J. J. Neefjes, P. Hoglund, M. T. Heemels, J. Bastin, T. N. Schumacher, A. Townsend, K. Karre.
1990
. Empty MHC class I molecules come out in the cold.
Nature
346
:
476
32
Vitiello, A., L. Yuan, R. W. Chesnut, J. Sidney, S. Southwood, P. Farness, M. R. Jackson, P. A. Peterson, A. Sette.
1996
. Immunodominance analysis of CTL responses to influenza PR8 virus reveals two new dominant and subdominant Kb-restricted epitopes.
J. Immunol.
157
:
5555
33
Kalman, S., W. Mitchell, R. Marathe, C. Lammel, J. Fan, R. W. Hyman, L. Olinger, J. Grimwood, R. W. Davis, R. S. Stephens.
1999
. Comparative genomes of Chlamydiapneumoniae and C. trachomatis.
Nat. Genet.
21
:
385
34
Read, T. D., R. C. Brunham, C. Shen, S. R. Gill, J. F. Heidelberg, O. White, E. K. Hickey, J. Peterson, T. Utterback, K. Berry, et al
2000
. Genome sequences of Chlamydia trachomatis MoPn and Chlamydiapneumoniae AR39.
Nucleic Acids Res.
28
:
1397
35
Shirai, M., H. Hirakawa, M. Kimoto, M. Tabuchi, F. Kishi, K. Ouchi, T. Shiba, K. Ishii, M. Hattori, S. Kuhara, et al
2000
. Comparison of whole genome sequences of Chlamydiapneumoniae J138 from Japan and CWL029 from USA.
Nucleic Acids Res.
28
:
2311
36
Falk, K., O. Rotzschke, S. Stevanovic, G. Jung, H. G. Rammensee.
1991
. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules.
Nature
351
:
290
37
Sette, A., A. Vitiello, B. Reherman, P. Fowler, R. Nayersina, W. M. Kast, C. J. Melief, C. Oseroff, L. Yuan, J. Ruppert.
1994
. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes.
J. Immunol.
153
:
5586
38
Wentworth, P. A., A. Vitiello, J. Sidney, E. Keogh, R. W. Chesnut, H. Grey, A. Sette.
1996
. Differences and similarities in the A2.1-restricted cytotoxic T cell repertoire in humans and human leukocyte antigen-transgenic mice.
Eur. J. Immunol.
26
:
97
39
Watson, M. W., S. al-Mahdawi, P. R. Lamden, I. N. Clarke.
1990
. The nucleotide sequence of the 60 kDa cysteine rich outer membrane protein of Chlamydiapneumoniae strain IOL-207.
Nucleic Acids Res.
18
:
5299
40
Kikuta, L. C., M. Puolakkainen, C. C. Kuo, L. A. Campbell.
1991
. Isolation and sequence analysis of the Chlamydiapneumoniae GroE operon.
Infect. Immun.
59
:
4665
41
Kornak, J. M., C. C. Kuo, L. A. Campbell.
1991
. Sequence analysis of the gene encoding the Chlamydiapneumoniae DnaK protein homolog.
Infect. Immun.
59
:
721
42
Perez Melgosa, M., C. C. Kuo, L. A. Campbell.
1994
. Isolation and characterization of a gene encoding a Chlamydia pneumoniae 76-kDa protein containing a species-specific epitope.
Infect. Immun.
62
:
880
43
Perez Melgosa, M., C. C. Kuo, L. A. Campbell.
1991
. Sequence analysis of the major outer membrane protein gene of Chlamydiapneumoniae.
Infect. Immun.
59
:
2195
44
Knudsen, K., A. S. Madsen, P. Mygind, G. Christiansen, S. Birkelund.
1999
. Identification of two novel genes encoding 97- to 99-kDa outer membrane proteins of Chlamydiapneumoniae.
Infect. Immun.
67
:
375
45
Summersgill, J. T., N. N. Sahney, C. A. Gaydos, T. C. Quinn, J. A. Ramirez.
1995
. Inhibition of Chlamydiapneumoniae growth in HEp-2 cells pretreated with γ interferon and tumor necrosis factor α.
Infect. Immun.
63
:
2801
46
Beatty, W. L., R. P. Morrison, G. I. Byrne.
1994
. Persistent chlamydiae: from cell culture to a paradigm for chlamydial pathogenesis.
Microbiol. Rev.
58
:
686
47
Kutlin, A., C. Flegg, D. Stenzel, T. Reznik, P. M. Roblin, S. Mathews, P. Timms, M. R. Hammerschlag.
2001
. Ultrastructural study of Chlamydiapneumoniae in a continuous-infection model.
J. Clin. Microbiol.
39
:
3721
48
Moazed, T. C., C. C. Kuo, J. T. Grayston, L. A. Campbell.
1998
. Evidence of systemic dissemination of Chlamydiapneumoniae via macrophages in the mouse.
J. Infect. Dis.
177
:
1322
49
Raulston, J. E., C. H. Davis, D. H. Schmiel, M. W. Morgan, P. B. Wyrick.
1993
. Molecular characterization and outer membrane association of a Chlamydia trachomatis protein related to the hsp70 family of proteins.
J. Biol. Chem.
268
:
23139
50
Iijima, Y., N. Miyashita, T. Kishimoto, Y. Kanamoto, R. Soejima, A. Matsumoto.
1994
. Characterization of Chlamydiapneumoniae species-specific proteins immunodominant in humans.
J. Clin. Microbiol.
32
:
583
51
Puolakkainen, M., C. C. Kuo, A. Shor, S. P. Wang, J. T. Grayston, L. A. Campbell.
1993
. Serological response to Chlamydiapneumoniae in adults with coronary arterial fatty streaks and fibrolipid plaques.
J. Clin. Microbiol.
31
:
2212
52
Gaydos, C. A., J. T. Summersgill, N. N. Sahney, J. A. Ramirez, T. C. Quinn.
1996
. Replication of Chlamydiapneumoniae in vitro in human macrophages, endothelial cells, and aortic artery smooth muscle cells.
Infect. Immun.
64
:
1614
53
Jahn, H. U., M. Krull, F. N. Wuppermann, A. C. Klucken, S. Rosseau, J. Seybold, J. H. Hegemann, C. A. Jantos, N. Suttorp.
2000
. Infection and activation of airway epithelial cells by Chlamydiapneumoniae.
J. Infect. Dis.
182
:
1678
54
Jondal, M., R. Schirmbeck, J. Reimann.
1996
. MHC class I-restricted CTL responses to exogenous antigens.
Immunity
5
:
295
55
Kataoka, T., N. Shinohara, H. Takayama, K. Takaku, S. Kondo, S. Yonehara, K. Nagai.
1996
. Concanamycin A, a powerful tool for characterization and estimation of contribution of perforin- and Fas-based lytic pathways in cell-mediated cytotoxicity.
J. Immunol.
156
:
3678
56
Al-Younes, H. M., T. Rudel, T. F. Meyer.
1999
. Characterization and intracellular trafficking pattern of vacuoles containing Chlamydiapneumoniae in human epithelial cells.
Cell. Microbiol.
1
:
237
57
Wolf, K., T. Hackstadt.
2001
. Sphingomyelin trafficking in Chlamydiapneumoniae-infected cells.
Cell. Microbiol.
3
:
145
58
Hsia, R. C., Y. Pannekoek, E. Ingerowski, P. M. Bavoil.
1997
. Type III secretion genes identify a putative virulence locus of Chlamydia.
Mol. Microbiol.
25
:
351
59
Rockey, D. D., D. Grosenbach, D. E. Hruby, M. G. Peacock, R. A. Heinzen, T. Hackstadt.
1997
.
Chlamydia psittaci IncA is phosphorylated by the host cell and is exposed on the cytoplasmic face of the developing inclusion. Mol. Microbiol.
24
:
217
60
Scidmore-Carlson, M. A., E. I. Shaw, C. A. Dooley, E. R. Fischer, T. Hackstadt.
1999
. Identification and characterization of a Chlamydia trachomatis early operon encoding four novel inclusion membrane proteins.
Mol. Microbiol.
33
:
753
61
Bannantine, J. P., R. S. Griffiths, W. Viratyosin, W. J. Brown, D. D. Rockey.
2000
. A secondary structure motif predictive of protein localization to the chlamydial inclusion membrane.
Cell. Microbiol.
2
:
35
62
Lampe, M. F., C. B. Wilson, M. J. Bevan, M. N. Starnbach.
1998
. γ interferon production by cytotoxic T lymphocytes is required for resolution of Chlamydia trachomatis infection.
Infect. Immun.
66
:
5457
63
Bonato, V. L., V. M. Lima, R. E. Tascon, D. B. Lowrie, C. L. Silva.
1998
. Identification and characterization of protective T cells in hsp65 DNA-vaccinated and Mycobacterium tuberculosis-infected mice.
Infect. Immun.
66
:
169
64
Cocchi, F., A. L. DeVico, A. Garzino-Demo, S. K. Arya, R. C. Gallo, P. Lusso.
1995
. Identification of RANTES, MIP-1α, and MIP-1β as the major HIV-suppressive factors produced by CD8+ T cells.
Science
270
:
1811
65
Baier, M., A. Werner, N. Bannert, K. Metzner, R. Kurth.
1995
. HIV suppression by interleukin-16.
Nature
378
:
563
66
Stenger, S., D. A. Hanson, R. Teitelbaum, P. Dewan, K. R. Niazi, C. J. Froelich, T. Ganz, S. Thoma-Uszynski, A. Melian, C. Bogdan, et al
1998
. An antimicrobial activity of cytolytic T cells mediated by granulysin.
Science
282
:
121