| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,¶,||,¶
,¶
,¶
Departments of
*
Medicine,
Laboratory Medicine,
Immunology, and
Epidemiology, University of Washington, Seattle, WA 98195;
¶ Fred Hutchinson Cancer Research Center, Clinical Division, Seattle, WA 98109; and
||
Virginia Mason Research Center, Seattle, WA 98101
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Another clone
recognized an immediate early protein, ICP0. Lymphocytes specific for
these lesion-defined epitopes could be reactivated from the PBMC of
additional subjects. These data are consistent with an influence of HSV
immune evasion genes upon the selection of proteins recognized by CD8
CTL in lesions. Tegument proteins, identified for the first time as Ags
recognized by HSV-specific CD8 CTL, are rational candidate vaccine
compounds. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Clinical and experimental data indicate that HSV-2-specific CD8 CTL responses are a functionally important component of the acquired immune response to this prevalent infection. CD8 CTL localize to the site of recurrent HSV-2 genital lesions (10, 11), and the clearance of infectious virus from lesions correlates temporally with the infiltration of CD8 T cells and HSV-specific CTL (12). In HIV-HSV-2-coinfected subjects, the frequency of HSV-specific CD8 CTL in the PBMC correlated inversely with the severity of recurrent anogenital HSV-2 infection in a cross-sectional study (13). In mice, CD8 CTL are involved in clearance of infectious virus from the infected ganglia and prevention of reactivation from latency (5).
Two HSV-2 proteins interfere with CD8 T cell recognition. The infected
cell protein No. 47
(ICP47),3 encoded by
gene US12, is one of the five viral "immediate early"
proteins that are synthesized within 1–2 h of infection. ICP47 of
HSV-1 and HSV-2 directly inhibit human TAP; their relative inactivity
against murine TAP (14) somewhat complicates the
interpretation of murine pathogenesis studies. The virion host shutoff
(vhs) protein, encoded by gene UL41, rapidly
degrades host cell mRNA, contributing to decreased synthesis of new HLA
class I. As a functional consequence of these activities,
HSV-2-infected human fibroblasts and keratinocytes are poorly
recognized by HSV- or allospecific CD8 CTL clones (10).
Using knockout viruses, contributions of both genes can be discerned.
They can be overcome, and CTL lysis restored, if fibroblasts are
pretreated with IFN- (15). IFN- up-regulates
proteasome, TAP, and HLA genes involved in class I Ag processing and
presentation. In vivo, IFN- levels are very high in recurrent HSV-2
lesions (16).
HSV encodes 
85 proteins (1). Little is known concerning
the targets of HSV-specific CD8 CTL or how the CD8 repertoire is shaped
by immune evasion genes. Clones recognizing type-common epitopes in HSV
envelope glycoproteins B and D (15, 17) have been derived
from PBMC using secondary in vitro restimulation. Other clones were
active against targets infected in the presence of transcriptional
inhibitors (17), consistent with recognition of viral
protein(s) loaded into APC upon virion binding and entry. The molecular
identity of these targets was not determined. Recently, Mikloska et al.
(18) found a high prevalence of bulk CD8 CTL responses
specific for immediate early proteins ICP27 and ICP4, using
IFN--treated keratinocytes as stimulator and readout cells. The
kinetics of HSV-induced down-regulation of HLA class I may predict that
either viral proteins injected directly into the cytoplasm, as is the
case for CMV protein pp65, or viral immediate early proteins, might
"outpace" immune evasion. Alternatively, it is possible that
TAP-independent Ag processing might sidestep TAP inhibition.
We used genetic approaches to test these predictions. T cell clones were recovered from recurrent genital HSV-2 lesions, without secondary in vitro restimulation with Ag. We have tried to minimize bias that might be introduced by such restimulation, as recently reported for CMV pp65 (19), and to study cells that have physiologically localized to the site of infection. Expression cloning was used to assign antigenic specificity. This paper describes the first identification of three HSV-2 proteins as targets of the local HSV-2-specific CD8 T cell response. We further investigate the ability of these CTL to recognize skin-derived cells, their TAP dependence, and their reactivation from the PBMC of HLA-appropriate persons.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
EBV-transformed lymphocyte cell lines (EBV-LCL) were derived from PBMC and maintained in LCL medium (RPMI 1640, 25 mM HEPES, 2 mM L-glutamine, 1% penicillin-streptomycin, 2 x 10-5 M 2-ME, 1 mM pyruvate) as described (10). Autologous EBV-LCL 1874, 5101, and 5491 and EBV-LCL 5085 were initiated in house, and HLA B*45-bearing EBV-LCL HM0983-022A-3216 was provided by Dr. L. Musey. TAP-deficient cell lines 721.174 (20), T2 (21), and T2/B7.63 (T2 transfected with HLA B*0702) were maintained in LCL medium. Cell line T2/B7.63, made by Dr. Peter Cresswell and provided by Dr. Charles Lutz, was maintained in 600 µg/ml G418. Primate kidney epithelial COS-7 cells (22) were grown in MEM with 10% FCS, 2 mM L-glutamine, and 1% penicillin-streptomycin. Fibroblasts were grown from skin biopsies as described (10). Single donor neonatal foreskin keratinocytes (Cascade Biologics, Portland, OR) were HLA typed, and HLA A*0201 cells were expanded per the manufacturer’s recommendations with Epilife medium containing growth supplement (HKGS), penicillin-streptomycin-amphotericin, trypsin, and trypsin inhibitor supplied by the Cascade Biologics.
Donors for lesion and PBMC studies had their HSV-1 and HSV-2 serostatus determined by type-specific serology (23). All subjects gave informed consent. Lesion HSV-specific T cells were obtained from three subjects from HSV-2 culture-positive recurrent lesions as previously described (11, 12). Lesions were 4 to 5 days old at the time of specimen collection. For subjects 1874 and 5491, lesion lymphocytes were expanded in bulk with PHA and IL-2 in the presence of acyclovir (11, 12) and CD8+ cells were positively selected with CD8 immunomagnetic beads (Minimacs; Miltenyi Biotec, Auburn, CA) and cloned (11). Two input numbers (3 and 1 cell/well) were used, and clones selected for workup were from plates with <37% of wells positive for growth. For subject 5101, tissue was digested with collagenase IV-S (C1889; Sigma, St. Louis, MO) in PBS, and the resultant cell suspension was immediately cloned (24) in serial 10-fold dilutions. Clones selected for study were from plates with <37% of wells positive for growth. To expand clones after positive screening assays, 5 x 104 cells were mixed in 25 ml T cell medium (TCM) (11) with 2.5 x 107 irradiated (3300 rad) allogeneic PBMC, 5 x 106 irradiated allogeneic (8000 rad) EBV-LCL, and 30 ng/ml anti-CD3 mAb OKT3 (Ortho, Raritan, NJ) (25). At 24 h and approximately every 3 days, human recombinant IL-2 (50 U/ml; Chiron, Emeryville, CA) was added with fresh medium, and cells were split as needed. The nomenclature for clones lists subject number, year of specimen, and clone number.
CTL were restimulated from PBMC by incubating 4 x
106 PBMC (Ficoll-Hypaque density gradient
centrifugation) with 100 µM peptide (synthesized by F-moc chemistry)
in 100 µl TCM for 1 h (1% DMSO was present). Cells were
diluted to 106/ml and cultured in
1.88-cm2 wells in the presence of 20 ng/ml
recombinant human (rh) IL-7 (R&D systems). rhIL-2 (Chiron) was included
at 20 U/ml starting on day 3. Cultures were fed with 0.5 volume medium
containing the same cytokines every 2–3 days without further addition
of peptide and assayed on days 12–14. For ICP0 responses, a modified
protocol was used. PBMC (4 x 106) were
stimulated with 1 µM peptide in 1.88-cm2 wells
with rhIL-2 (10 U/ml) added on day 3. Cells were restimulated on day 7
with 2 x 106 autologous PBMC, 1 µM
peptide, and IL-2 and fed with 0.5 volume medium containing IL-2 every
2–3 days until assay on days 14–16. For some bulk PBMC-derived
cultures, CD8 cells were positively selected, cloned at 1 cell/well,
and expanded as described above.
HSV-1 strain E115 (26) and HSV-2 strains 333 (27) and HG52 (28) were raised and titered in Vero cells (29). The ICP4 deletion mutant of HSV-2, hr259, was grown on complementing E4 cells (30). Recombinant vaccinia expressing ICP0 of HSV-2 (31) (kindly provided by Dr. B. Rouse) and wild-type vaccinia NY were raised and titered in BSC-40 cells.
Expression cloning
The strategy of Boon et al. (32) was adapted to
genomic HSV-2 DNA. Cytoplasmic and supernatant virus purified from
HSV-2 strain HG52-infected Vero cells (33) were combined.
DNA was digested with Sau3AI, reextracted, and partially
filled in with DNA polymerase Klenow fragment
(3'-5-exo-) (New England Biolabs, Beverly, MA),
dTTP, and dCTP. Plasmids pcDNA3.1(+)/His A, B, and C (Invitrogen,
Carlsbad, CA) were digested with XhoI and partially filled
with Klenow fragment, dATP, and dGTP. After ligation of repurified
(organic extraction, ethanol precipitation) insert mixture (100 ng)
and individual vectors (1 µg), DNA was precipitated, washed, and
electroporated into Escherichia coli strain DH10B (Life
Technologies, Gaithersburg, MD). Each library contained several
thousand primary transformants. The majority of each library was
immediately amplified in bulk. Among 20 random clones, all contained
single HSV-2 Sau3AI fragments. Sequencing (Taq
Dye-Deoxy FS; Perkin-Elmer ABI, Foster City, CA) was performed per the
manufacturer’s instructions.
To make library DNA for transfection, 96-well plates (140504; Beckman,
Fullerton, CA) were inoculated either with libraries at 15
colonies/well or with selected individual clones. After overnight
incubation at 37°C at 300 rpm agitation, DNA was prepared with
96-well filters (Millipore, Bedford, MA) per the manufacturer. Some
bacteria in each well were saved. Average yield was 10 µg of DNA per
well. IFN- secretion as the primary readout of T cell activation.
COS-7 cells seeded at 9000 cells/well in 96-well flat-bottom plates
were transfected the next day (day 2) with 50 ng HLA heavy chain cDNA
and Fugene-6 (Boerhinger Mannheim-Roche, Indianapolis, IN), using the
manufacturer’s protocol. On day 3, cells were infected with HSV-2
strain 333 at an estimated MOI of 10. On day 4, 0.7–1.0 x
105 cloned CD8 T cells in 150 µl FCS-based medium were
added. Supernatants were saved on day five for IFN- ELISA
(below).
To screen libraries, COS-7 were cotransfected with 50 ng HLA cDNA and 100 ng library DNA (pool or single colony). Two days later, 1 x 105 cloned T cells/well (with rhIL-2 at 5 U/ml for some assays) were added, and supernatants were saved as above. DNA was prepared as above from 96 individual colonies derived from positive pools, and the process was repeated to identify individual active plasmids which were then sequenced to identify antigenic regions of HSV-2.
To make HLA B*4501 cDNA, total RNA (5 µg) from subject 1 (HLA A1, A*0201, B7, B*4501) EBV-LCL, purified using guanidinium-acid phenol (34), was incubated in 15 µl water with 0.5 µg oligo(dT)12–18 (Pharmacia, Piscataway, NJ) at 70°C for 10 min and chilled. The reaction mixture (25 µl) containing 1 µl Moloney murine leukemia virus reverse transcriptase, 2.5 mM MgCl2, 5 mM DTT, 2.5 µl 10x PCR buffer II (500 mM KCl, 100 mM Tris, pH 9.0), and 500 µM each dNTP was incubated at 42°C for 1 h and boiled for 5 min. For PCR, a 50-µl reaction containing 1x pfu buffer and 1 U pfu DNA polymerase (Invitrogen), 5 µl 2.5 mM each dNTP, 1 µl cDNA, and 50 pmol each primer (AAGGTACCATGCGGGTCACGGCACCCCGAA and GGTCTAGAAGTTCGACACTCTCTGTGTAGT; KpnI and XbaI sites underlined) was heated to 94°C for 2 min; cycled four times at 96°C for 30 s, 50°C for 30 s, and 72°C for 3 min; cycled 30x with 60°C annealing; and extended for 10 min at 72°C. Amplimer purified by organic extraction/alcohol precipitation was digested with appropriate enzymes and ligated into pcDNA3.0 (Invitrogen). The sequence was identical with GenBank 61710. HLA A*0201 and B*0702 cDNAs previously similarly cloned by RT-PCR, and shown by sequencing to be identical with wild-type sequences, were obtained from Dr. Stanley Riddell.
To study the cDNA species derived from the positive genomic clone containing portions of ICP0 (Results), COS-7 cells (100 mm2) were transfected with the ICP0 genomic clone, and total RNA was prepared after 48 h. The primer used for cDNA synthesis (TGCTCTAGAGACTCGATCCCTGCGCGTCGG; XbaI site underlined) was from the 3'-end of the HSV-2 DNA in the ICP0 genomic clone. Moloney murine leukemia virus reverse transcriptase (Life Technologies) was used per the manufacturer. To examine splicing, PCR used pfu cDNA polymerase, the above 3'-primer, and 5'-primer TAAGGTACCTGAACCCCGGCCCGGCACGAGC (KpnI site). To isolate exon 1 (28) of ICP0, PCR used the same 5'-primer and 3'-primer TGCTCTAGACCAGGCGTGCGGGGCGGCGGG (XbaI site). Reaction conditions were individually optimized. Product was digested with Acc65I and XbaI, gel purified, and ligated into similarly treated pCDNA 3.1-His-B, and in-frame insertion was confirmed by sequencing.
Full-length UL47 of HSV-2 was cloned by PCR into pCDNA3.1/His-C using 5'-primer CTAGGATCCCCTCCGGCCACCATGTCC and 3'-primer CGATCTAGACCTATGGGCGTGGCGGGC (BamHI and XbaI sites underlined). Full-length UL46 of HSV-2 was cloned by PCR into pcDNA3.1/His-C with 5'-primer CGAGGATCCGTCTCCGCCATGCAACGCCG and 3'-primer CGCTCTAGATTTTAATGGCTCTGGTGTCG (BamHI and XbaI sites underlined). Similarly, a construct expressing aa 1–590 of UL47 was made by PCR, using the above 5'-primer, an appropriate 3'-primer, and pCDNA3.1/His-C. Expression of aa 1–535 and 536–696 of UL47 was driven by constructs derived from full-length UL47 using a naturally occurring NotI site at aa 535. In-frame vector-HSV-2 fusion at the 5'-end of the HSV-2 DNA was confirmed by sequencing in each case.
We investigated some CD8 CTL clones in the COS-7 system using a panel
of cloned HSV-2 genes. Cells were cotransfected with HLA class I heavy
chain cDNA (50 ng/well) and HSV-2 constructs (25 ng/well) and assayed
for stimulation of IFN-. Our gene panel included UL46 and
UL47 (above) and full-length HSV-2 UL19, UL21,
UL50, and US8, each cloned into the pcDNA3.1/His
series, and UL49 cloned into pEGFP-C1 (Clontech, Palo Alto,
CA) as described and validated (35, 36).
Cytotoxicity assays
Cytotoxicity assays were performed as previously described
(10) using 4-h 51Cr release. Target
EBV-LCL were typically infected for 18 h with HSV or vaccinia
strains at multiplicity of infection (MOI) 10, and the usual E:T was
20:1, or loaded with peptide for 90 min at 37°C in 200-µl volumes.
In some assays, purified mAb W6/32 (37) was included at 10
µg/ml. To inhibit viral RNA expression, EBV-LCL were preincubated
with actinomycin D (Sigma) at 5 µg/ml for 30 min before infection.
Actinomycin D was maintained throughout the 90-min infection, wash, and
assay periods. For fibroblast targets, cells were plated in
9.6-cm2 plates and infected at 70–90%
confluence, usually the next day, with the exception of IFN-
experiments, in which some wells were treated with 500 U/ml IFN-
(Endogen, Woburn, MA) and all wells were held for 72 h until
infection. The IFN- wells were maintained with IFN- throughout
the infection. For keratinocyte targets, cells were plated in
9.6-cm2 plates at 2.5 x
103 cells/cm2, and IFN-
treatment was begun 4 days later when cells reached 60% confluence.
Adherent cells were chromated overnight, and were peptide loaded (1
µM) for 90 min, before trypsinization and washing. All targets were
used in CTL assays at 2 x 103/well, and
2.5% Ipegal (Sigma, St. Louis, MO) was used to measure total release.
Assays were performed in triplicate and spontaneous release was usually
<25%.
HLA-peptide tetramers
A tetrameric reagent containing HLA A*0201 heavy
chains, 
2-microglobulin, and peptide UL47
551–559 was synthesized by the Tetramer Facility of the National
Institutes of Allergy and Infectious Diseases (Bethesda, MD). Tetramers
were biotinylated and labeled with streptavidin-PE.
Flow cytometry
Lymphocytes were washed and incubated with
anti-CD4-FITC/anti-CD8-PE (Sigma),
anti-CD3-FITC/anti-CD16PE and CD56-PE, or anti-TCR
-FITC or anti-TCR 
-PE (Becton Dickinson, San Jose, CA),
or a mixture of control FITC- and PE-labeled mAb (Sigma) on ice for 30
min. For tetramer staining, cells were centrifuged, resuspended in 100
µl TCM, and incubated with 1 µl tetramer for 1 h at room
temperature. After this, 1 µg CD8-FITC (Caltag) was added, and cells
were incubated for 30 min on ice. To measure HLA transfection,
trypsinized COS-7 cells were stained with 1 µg FITC-labeled mAb B12
reactive with HLA B*4501 (One Lambda, Canoga Park, CA) or PE-labeled
mAb 1288 reactive with HLA B*0702 (Chemicon, Temecula, CA), or
supernatant of MA2.1 cells (38) reactive with HLA A*0201,
followed by FITC-labeled goat anti-mouse IgG (Sigma). Washed cells
fixed with 1% paraformaldehyde in PBS were analyzed with a FACScan
cytometer (Becton Dickinson) and WinMDI version 2.8 shareware
(http://facs.scripps.edu). Cells in the appropriate gates on forward vs
side scatter were analyzed. To measure the infection of wild-type and
mutant LCL with HSV-2, uninfected or infected (18 h, MOI 10) cells were
stained with mAb 18B3 specific for envelope glycoprotein D or
control mouse IgG1, followed by FITC-conjugated goat anti-mouse IgG
(Sigma) as previously described (10).
HLA typing
Subjects were typed serologically and by DNA methods at the Puget Sound Blood Center (Seattle, WA).
ELISA
IFN- was measured with reagents from Endogen. Costar No. 3369
plates (Corning, NY) were coated with 100 µl 0.25 µg/ml capture mAb
(M700A-E) diluted in 0.1 M sodium carbonate (pH 9.4) overnight at
4°C, and blocked with 1% BSA in 0.2 M NaCl, 3 mM KCl, 0.05 M Tris,
pH 9 (TBS) for 1 h. Subsequent incubations were each 100 µl,
preceded by three to five washes with PBS/0.2% Tween 20, and performed
with rotation at room temperature. Samples and standards diluted in TBS
with 0.1% BSA, 0.05% Tween 20, and 4 µg/ml Ig-Inhibiting Reagent
No. 6LD1068 (Bioreclamation, East Meadow, NY; sample buffer) were added
for 2 h. Biotinylated detection mAb (M701B) diluted to 100 ng/ml
in sample buffer was added for 1 h. Avidin D:HRP (A-2004) diluted
to 100 ng/ml in TBS with 1% BSA, 0.05% Tween 20 was added for 1
h. TMB substrate was added for 10 min, reactions were stopped with 1 M
phosphoric acid, and results were read on a plate reader at 450 and 650
nm. The lower limit of detection was 10 pg/ml.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD8 CTL clones specific for HSV-2 were obtained from herpetic
lesions, without secondary in vitro restimulation with Ag, as
previously described (12). Clones with HSV-specific
cytotoxic activity in screening assays, and CD8 but not CD4 expression,
were expanded for further study. For subjects 1874 and 5491, multiple T
cell clones with the same apparent pattern of HSV-2 type specificity
and HLA restriction were derived. Representative clones, based on HLA
restriction analysis and HSV-2 type specificity (39), were
chosen for detailed study (Table I
).
Clone 5101.1999.23 was obtained by collagenase digestion of a lesion
biopsy and was the single CD8 CTL clone obtained in screening 60
clones. Each clone was CD3+,
CD8+, TCR +,
CD4-, and CD16/56- and
recognized an HSV-2 type-specific epitope. For each clone, most HLA-A-
and B-mismatched target cells were not lysed, regardless of viral
infection. HLA-restricting alleles were preliminarily assigned as HLA
A*0201, B*0702, or B*4501 by HLA typing the source subjects and using
partially matched EBV-LCL as APC (Table I). A transfection/infection
assay (Fig. 1) confirmed the CTL results
(Table I) and also established the suitability of HLA-transfected COS-7
for expression cloning.
|
|
To evaluate whether the T cell clones recognized virion input
proteins, we checked their cytolytic activity against EBV-LCL infected
in the presence of actinomycin D. Clones 1874.1991.22, 5101.1999.23,
and 5491.2000.48 lysed these target cells (Table I
). Lysis by clone
1874.1997.51 was significantly inhibited by blockade of transcription.
Each of the clones was able to lyse target cells infected with the
mutant virus hr259, which lacks the ICP4 trans activator
protein, and is only able to newly express the other immediate early
proteins, ICP0, ICP27, ICP22, ICP47, and the small unit of
ribonucleotide reductase (30).
Recognition of tegument HSV-2 Ags by CD8 T cells
For expression cloning, HSV-2 genomic DNA fragments were
cotransfected together with HLA class I cDNA into COS-7 cells, and
IFN- secretion again used as the readout for T cell activation. The
genomic HSV-2 Sau3aI libraries, in each reading frame, were
screened to oversample the HSV-2 genome 6-fold. Each of the first
three CD8 T cell clones studied responded to cells transfected with
plasmid DNA prepared from individual bacterial colonies, which were
sequenced to preliminarily identify T cell Ags (Table I).
CD8 clone 5101.1999.23 recognized COS-7 cells cotransfected with HLA
A*0201 and a HSV-2 Sau3aI fragment from bp 102,943–102,876
(28) (Table I). The predicted fusion protein contains
HSV-2 UL47 aa 278–298. Reactivity with UL47 was confirmed by
cotransfection of A*0201 and full-length HSV-2 UL47 (Table II).
|
). This fragment was predicted to contain the
DNA encoding UL47 aa 300–696, intervening DNA, and then aa 1–71
of UL46. Analysis of the 5'-vector-insert junction in C:2:C10:B9
revealed out-of-frame translation of the initial UL47 DNA.
The insert is expected to contain the UL46 promoter
(40). The epitope, therefore, could be encoded by
UL46. C:2:C10:B9 also contains potential sites of internal
initiation of translation within UL47. We assayed
UL46 and UL47 separately in the COS-7
cotransfection assay (Table II). UL47 was active, whereas
UL46 was not. Truncation analysis localized the epitope to
aa 535–590 (Table II).
The specificity of CD8 clone 5491.2000.48 was determined with a panel
of partial- and full-length HSV-2 genes. The HSV-2 genes studied were
previously shown to be recognized by CD4 T cell clones (35, 36, 41). Only HSV-2 UL49, when cotransfected with HLA
B*0702, stimulated IFN- release by clone 5491.2000.48 (Fig. 2).
|
). UL47 535–590 (Table II) was
similarly analyzed. Peptide 551–559 was active at 1 nM (Fig. 3).
Potential HLA B*0702-binding peptides in UL49 of HSV-2 were
synthesized, and two (aa 47–55 and 14–22) were active at 1 µM (data
not shown). Titration (Fig. 3) showed that UL49 49–55 was highly
active, with an EC50 of <10 nM, whereas UL49 14–22 had
activity only at 1 µM (not shown). The antigenic peptides in UL47 and
UL49 contain significant amino acid sequence differences from the
corresponding predicted HSV-1 peptides (28, 43),
explaining type-specific recognition of HSV-2 (Table I).
|
For clone 1874.1997.51, positive reactions to plasmid pools were
present in each library. The active plasmids in each library contained
a genomic Sau3AI fragment from nucleotides 1858–3022
(28). Nucleotide 2007 listed as T in the published
sequence was read as C. In addition to 445 bp of 5'-untranslated
sequence, all of predicted exon 1, intron 1, and the first 234 bp of
predicted exon 2 of ICP0 were present, preliminarily identifying the Ag
as ICP0. Because alternative splicing of HSV-1 ICP0 has been documented
at both the RNA and protein levels (44, 45), we first
identified the Ag-encoding mRNA species in COS-7 cells to determine how
the ICP0 genomic clone was spliced in our system. COS-7 cells were
transfected with genomic clone C:1:H3:B8 (Table I), and cDNA was
synthesized from total cellular RNA followed by PCR designed to amplify
the spliced transcript. The size of the PCR product (300 bp) was
consistent with the splicing out of intron 1. Sequencing showed a
slight difference from the reported (28) splice point for
mature HSV-2 ICP0 mRNA. Three base pairs encoding aa Q26
(28) were missing. We have retained Q26 for peptide
numbering (below). To determine whether the antigenic peptide lay
within exon 1 or exon 2, PCR was repeated with specific primers. The
exon 1-partial exon 2 cDNA, but not exon 1 cDNA, was stimulatory for T
cell clone 1874.1997.51 (Table II), localizing the epitope to aa
26–105 in exon 2. Reactivity was confirmed in CTL assays using a
recombinant vaccinia virus expressing ICP0. At E:T 20:1, lysis of
vaccinia ICP0 (31)-infected target cells was 52.1%
compared with 2.3% for vaccinia wild type. Having determined the RNA
splicing pattern, we proceeded to find the peptide epitope.
Two reported HLA B45-restricted epitopes (46, 47, 48),
AEEAAGIGIL and GAETFYVDGA, share with the B44 supertype
(49) a preference for negatively charged and hydrophobic
amino acid side chains at the P2 and P9 anchor positions. ICP0 (HSV-2)
92–105, containing this motif, was active at 1 µM (not shown).
Truncation yielded ICP0 (HSV-2) 92–101, with an
EC50 in the 1 nM range (Fig. 3).
Recognition of skin-derived fibroblasts and keratinocytes by CD8 CTL clones
Within lesions, HSV-2 is mainly present in keratinocytes
(16). We investigated how MOI (amount of virus), time of
infection, and pretreatment with IFN- influenced lysis of dermal
fibroblasts and keratinocytes. For fibroblasts (Fig. 4), in the absence of IFN-
pretreatment, infection for 2 h led to detectable lysis, which
increased with increasing MOI. Lysis was undetectable (<5% specific
release at E:T of 20:1) after overnight infection with MOI 1, 5, or 25.
With IFN- pretreatment, lysis was generally increased, but 2-h
infection was still superior. HLA-mismatched target cells were not
lysed, even after peptide loading (data not shown).
|
). IFN- pretreatment generally increased
recognition, without leading to lysis of control cells. In contrast to
fibroblasts, 18-h infection was generally required. Weak cytolysis of
cells infected for 2 h was noted only for IFN--pretreated
targets. Chromium release again correlated directly with the amount of
infectious virus added, because no specific lysis was noted at MOI 1 or
5 (data not shown). PBMC responses to HSV-2 T cell epitopes
The A*0201-restricted responses to UL47 were studied in
six HLA A*0201-bearing, HSV-2 infected persons (Fig. 5). No response was seen in a
HSV-uninfected control subject. One HSV-2-infected subject, 9383, who
has infrequently recurring genital HSV-2, had strong cytolytic
responses, with effector populations killing both HSV-2-infected
A*0201-bearing EBV-LCL and peptide-loaded targets. Subject 1874, from
whom the 551–559-specific clone was derived, also had low but
detectable PBMC CTL responses to peptide 551–559. In contrast, subject
5101, from whom the 289–298-specific clone was derived, did not have a
detectable PBMC CTL response after peptide restimulation. Confirmation
of CTL activity was obtained by deriving CD8 clones from
peptide-stimulated PBMC from donor 9383. For both UL47 289–298 and
551–559, clones were obtained which lysed HLA A*0201-bearing EBV-LCL
loaded with the stimulating peptide or infected with HSV-2 (data not
shown).
|
).
The tetramer bound specifically to the index T cell clone 1874.1991.22.
Flow cytometric assays showed the highest level of
CD8+ and tetramer+ cells
for subject 9383. The index subject, 1874, also had
tetramer+ CD8 cells detected. However, other
subjects who did not have detectable CTL, such as 7282 and 5101,
appeared to have enriched numbers of tetramer-binding cells compared
with the HSV-seronegative control subject.
Three HSV-2-infected, HLA-compatible (B*4501) persons were available
for study of the epitope in ICP0. Two subjects had strong CTL responses
(Table III). Lysis of infected targets
was inhibited by anti-class I mAb, and not observed if the target
cells did not express B*4501.
|
For each of the three CD8 clones studied, lysis of TAP-deficient
cells after HSV-2 infection was greatly reduced in comparison to
wild-type EBV-LCL (Table IV). Greater
than 90% of each of the TAP-deficient cell lines, as well as control
wild-type LCL, were permissive for viral infection and protein
synthesis as evaluated by flow cytometry using mAb specific for
envelope glycoprotein gD. Peptide loading was able to sensitize the
TAP-deficient cells, confirming HLA class I expression.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The possible functional importance of HSV-specific CD8 CTL in humans has been addressed in several recent studies (4, 12, 13, 16, 54). Murine studies also illustrate protective roles for CD8 CTL (55, 56, 57). Because the HSV-2-encoded protein ICP47 is a relatively inefficient inhibitor of murine TAP (14), we chose the human system for our Ag discovery work.
Earlier work has shown that the majority of human PBMC- and lesion-derived HSV-2-reactive CD8 CTL clones are type specific for HSV-2 (12, 15, 17, 58). This is also consistent with a possible functional role for CD8 responses in host defense, because prior HSV-1 infection provides poor protection against subsequent HSV-2 infection (59), whereas both neutralizing Ab and CD4 responses have a strong type-common component (60, 61). We emphasized study of HSV-2 type-specific clones. Clones recovered directly from the site of infection, derived without secondary in vitro restimulation with Ag (62), were used to study physiological responses at the site of disease.
Little is known about the specificity of human HSV-2-specific CD8 CTL. The two published epitopes are type-common peptides within glycoproteins B and D (10, 15). At the nonclonal level, experiments using restimulation of PBMC, drug blocks, and vaccinia recombinants show that HSV-1 ICP4, ICP27, ICP0, all immediate early proteins, HSV-1 early protein ICP6, and possibly other true early proteins may be targets of human CTL (18, 63, 64). HSV-1 early protein thymidine kinase (tk) is recognized by CD8 clones from PBMC of subjects treated with tk-transfected autologous cells, but this is likely a primary immune response (65). A PBMC-derived CD8 T cell clone specific for a melanoma-associated protein (Melan A/MART-1) also reacted with a peptide from HSV-1 glycoprotein C (66).
It has been hypothesized that the selection of Ags
recognized by HSV-specific CD8 CTL is influenced by immune evasion
genes within HSV (7, 9). ICP47 blocks assembly of mature
HLA-2-microglobulin-peptide complexes by
inhibition of TAP. This effect occurs quickly during viral infection
(67). In addition, vhs destabilizes host mRNA
and reduces synthesis of new HLA class I. It is rational to predict
that processing of both virion input proteins and immediate early
proteins might outpace the down-regulation of HLA class I. A similar
model has been proposed for human CMV, which also potently
down-regulates HLA class I (9, 68, 69, 70).
In general, our results support an effect of immune evasion genes on the selection of CD8 Ags, given that we uncovered reactivity with two virion tegument proteins and one immediate early protein. The HSV open reading frame UL47 was detected twice. UL47 encodes tegument protein VP13/14, which is present in large amounts in virions (71). The physiological function(s) of UL47 are incompletely studied. The protein enhances trans activation by VP16 (72) but is dispensable for replication in culture (72, 73). Direct trafficking of input UL47 into the cytoplasm has been detected shortly after virion binding (42). UL49 encodes VP22, a tegument protein required for viral replication (1). UL49 protein is also abundant in virions and delivered into the cytoplasm by virus entry (42). Lysis of EBV-LCL by tegument-specific CD8 CTL was not inhibited by blockade of gene transcription or infection with a replication-incompetent virus, consistent with the processing and presentation of preformed virion input protein.
CD8 T cell specificity for the immediate early protein ICP0 (74) was also detected. ICP0 may be present in small amounts in HSV-1 virions (75). However, we found that lysis by the ICP0-specific CD8 clone 1874.1997.51 was substantially inhibited by infection in the presence of actinomycin D. Infection with the replication-incompetent virus hr259, which is able to direct the synthesis of ICP0, was able to sensitize target cells to lysis. These data are consistent with recognition of endogenously synthesized viral protein. As manipulation of virus or host is not possible with the natural host species for HSV-2, formal proof of an effect of immune evasion genes on the CD8 CTL repertoire will be difficult to obtain. If additional epitopes can be accrued with this or other approaches, they can be examined to determine whether a pattern consistent with antigenicity of virion input and immediate early proteins is present.
TAP-independent processing has been reported in other viral systems (76, 77, 78). Thus far, in our examination of three discrete epitopes in tegument proteins, we did not find evidence for TAP-independent Ag processing of HSV epitopes. The CD8 response seems to "evade the evasion," at least in the cases examined to date, while continuing to rely on TAP for Ag processing. The TAP dependence of responses to immunodominant Ags, if these can be identified as such, will also be of interest.
Most studies of clonal CD8 responses have used EBV-LCL as target cells.
These cells are relatively resistant to HSV-mediated class I
down-regulation (10). For dermal fibroblasts, we found
that a short time of infection (2 h) was adequate for target cell
sensitization for lysis by tegument protein-specific CTL. Because the
UL47 and UL49 tegument proteins are synthesized with "late"
kinetics (1), typically starting after 6 h or more of
viral infection, these data are also consistent with recognition of
preformed Ag in fibroblasts. Lysis was MOI dependent. Because HSV
preparations typically contain a large number of defective particles
(1), it is likely that tegument proteins were also being
delivered into fibroblasts by noninfectious particles. After 18 h
of infection, the fibroblasts were not lysed, regardless of MOI,
similar to previous results with CD8 CTL clones of unknown fine
specificity (10). IFN- pretreatment was able to
partially restore lysis of 18-h-infected cells. In contrast to
fibroblasts, recognition of keratinocytes after 18 h of infection
was superior to recognition after 2 h of infection. The reason for
the difference between fibroblasts and keratinocytes is unknown.
IFN- pretreatment was able to restore some lysis of 2-h-infected
cells, and further improved recognition of 18-h-infected cells. In
future studies, we hope to compare the recognition of IFN--treated
keratinocytes by both CD4 and CD8 CTL (79, 80, 81), given that
both are present in lesions.
Tegument proteins have not previously been described as targets of the HSV-specific CD8 T cell response. CD4 responses to HSV-1 UL47 have been detected in HSV-mediated acute retinal necrosis (82). CD4 responses to UL49 are commonly detected among lesion-infiltrating HSV-2-specific clones (41). Because responses to UL49 are also present in the cornea in herpes stromal keratitis in humans (35), a disease that may be driven by pathogenic Th1-like T cells (82), caution is clearly warranted in using this protein as a vaccine. Overall, UL49 is the only known HSV-2 protein recognized by both CD4 and CD8 T cell clones recovered from herpetic lesions. A unique intercellular transport pathway allows highly efficient uptake of soluble UL49 protein into a variety of epithelial cell types (83, 84, 85) which could also intersect Ag processing pathways.
A few technical aspects of the methods warrant brief comment. Our library approach worked each of the first three consecutive times it was applied to CD8 CTL clones, so we do not think that technical factors greatly biased the viral kinetic or structural classes of the CTL epitopes detected. A library created, as was ours, with a single restriction endonuclease will contain "holes." The initial positive ICP0 genomic clone started well upstream of the ATG start, extending to bp -445. As noted above, the ATG start of ICP0 was out of frame with the vector-derived peptide expressed by the positive genomic clone. The ICP0 promoter appears to be functioning without additional viral factors such as the major trans activator VP16 as previously reported for HSV-1 (86). Not all viral promoters will necessarily be active outside of the context of natural viral infection. This problem can be overcome by fragmenting the HSV-2 DNA with alternative methods before library creation.
We chose the sequenced strain (28) of HSV-2, HG52, for library creation. HSV-2 strains are relatively invariant due to the high fidelity of the HSV DNA polymerase (1). Possibly, our approaches may fail if strain-specific epitopes are recognized in vivo. The library is a relatively efficient method for epitope/Ag discovery once conditions are optimized. Neither "holes" in the library nor strain-specific epitopes have interfered as of yet.
In summary, reactivity of lesion-infiltrating, HSV-2 type-specific CD8 T cell clones with the tegument proteins encoded by genes UL47 and UL49 (VP13/14 and VP22, respectively), and ICP0, are described for the first time. The data are consistent with a modulatory effect of ICP47 and/or vhs on the CD8 response to HSV. TAP function, but not viral gene transcription, is required for recognition by UL47- and UL49-specific clones, consistent with processing of preformed virion input protein. Tegument-specific CD8 clones were able to recognize skin-derived fibroblasts and keratinocytes. Responses were also detectable in the PBMC of additional subjects. Further studies are required to define the prevalence and dominance of these virus-specific responses and the potential role of these Ags in immunologic approaches to reduce HSV-2 infection and disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. David M. Koelle, Virology Division, Harborview Medical Center, University of Washington, Mail Stop 359690, 325 9th Avenue, Seattle, WA 98104-2499.
3 Abbreviations used in this paper: ICP47, infected cell protein number 47 (other protein numbers are similar); US12, unique short region of the HSV genome, gene 12 (other gene numbers are similar); vhs, virion host shutoff; UL47, unique long region of the HSV genome, gene 47 (other gene numbers are similar); EBV-LCL, EBV-transformed lymphocyte cell line; MOI, multiplicity of infection, the number of PFUs of virus added per cell; TCM, T cell medium; rh, recombinant human; EC50, 50% effective concentration.
Received for publication November 3, 2000. Accepted for publication January 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
or when virion host shutoff functions are disabled. J. Immunol. 156:3901.[Abstract]
and IL-12 production. J. Immunol. 164:5167.0 accumulate in the cytoplasm of infected cells. Proc. Natl. Acad. Sci. USA 93:12535. bias. J. Virol. 69:5849.[Abstract]
. J. Infect. Dis. 173:7.[Medline]
TIF-mediated transcriptional induction: characterization of three viral deletion mutants. J. Virol. 65:829. interferon in recurrent herpes simplex infection. J. Infect. Dis. 153:979.[Medline]
This article has been cited by other articles:
|
|
 |
|
  W. J. Muller, L. Dong, A. Vilalta, B. Byrd, K. M. Wilhelm, C. L. McClurkan, M. Margalith, C. Liu, D. Kaslow, J. Sidney, et al. Herpes simplex virus type 2 tegument proteins contain subdominant T-cell epitopes detectable in BALB/c mice after DNA immunization and infection J. Gen. Virol., May 1, 2009; 90(5): 1153 - 1163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Jing, D. H. Davies, T. M. Chong, S. Chun, C. L. McClurkan, J. Huang, B. T. Story, D. M. Molina, S. Hirst, P. L. Felgner, et al. An Extremely Diverse CD4 Response to Vaccinia Virus in Humans Is Revealed by Proteome-Wide T-Cell Profiling J. Virol., July 15, 2008; 82(14): 7120 - 7134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Koelle, A. Magaret, C. L. McClurkan, M. L. Remington, T. Warren, F. Teofilovici, and A. Wald Phase I Dose-Escalation Study of a Monovalent Heat Shock Protein 70-Herpes Simplex Virus Type 2 (HSV-2) Peptide-Based Vaccine Designed To Prime or Boost CD8 T-Cell Responses in HSV-Naive and HSV-2-Infected Subjects Clin. Vaccine Immunol., May 1, 2008; 15(5): 773 - 782. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Chentoufi, X. Zhang, K. Lamberth, G. Dasgupta, I. Bettahi, A. Nguyen, M. Wu, X. Zhu, A. Mohebbi, S. Buus, et al. HLA-A*0201-Restricted CD8+ Cytotoxic T Lymphocyte Epitopes Identified from Herpes Simplex Virus Glycoprotein D J. Immunol., January 1, 2008; 180(1): 426 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Bettahi, A. B. Nesburn, S. Yoon, X. Zhang, A. Mohebbi, V. Sue, A. Vanderberg, S. L. Wechsler, and L. BenMohamed Protective Immunity against Ocular Herpes Infection and Disease Induced by Highly Immunogenic Self-Adjuvanting Glycoprotein D Lipopeptide Vaccines Invest. Ophthalmol. Vis. Sci., October 1, 2007; 48(10): 4643 - 4653. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Jing, T. M. Chong, B. Byrd, C. L. McClurkan, J. Huang, B. T. Story, K. M. Dunkley, L. Aldaz-Carroll, R. J. Eisenberg, G. H. Cohen, et al. Dominance and Diversity in the Primary Human CD4 T Cell Response to Replication-Competent Vaccinia Virus J. Immunol., May 15, 2007; 178(10): 6374 - 6386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Zhu, D. M. Koelle, J. Cao, J. Vazquez, M. L. Huang, F. Hladik, A. Wald, and L. Corey Virus-specific CD8+ T cells accumulate near sensory nerve endings in genital skin during subclinical HSV-2 reactivation J. Exp. Med., March 19, 2007; 204(3): 595 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Diaz and D. M. Koelle Human CD4+ CD25high Cells Suppress Proliferative Memory Lymphocyte Responses to Herpes Simplex Virus Type 2. J. Virol., August 1, 2006; 80(16): 8271 - 8273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Hosken, P. McGowan, A. Meier, D. M. Koelle, P. Sleath, F. Wagener, M. Elliott, K. Grabstein, C. Posavad, and L. Corey Diversity of the CD8+ T-Cell Response to Herpes Simplex Virus Type 2 Proteins among Persons with Genital Herpes. J. Virol., June 1, 2006; 80(11): 5509 - 5515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Jing, T. M. Chong, C. L. McClurkan, J. Huang, B. T. Story, and D. M. Koelle Diversity in the Acute CD8 T Cell Response to Vaccinia Virus in Humans J. Immunol., December 1, 2005; 175(11): 7550 - 7559. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Bosnjak, M. Miranda-Saksena, D. M. Koelle, R. A. Boadle, C. A. Jones, and A. L. Cunningham Herpes Simplex Virus Infection of Human Dendritic Cells Induces Apoptosis and Allows Cross-Presentation via Uninfected Dendritic Cells J. Immunol., February 15, 2005; 174(4): 2220 - 2227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. N. Milligan, C.-F. Chu, C. G. Young, and L. R. Stanberry Effect of Candidate Vaginally-Applied Microbicide Compounds on Recognition of Antigen by CD4+ and CD8+ T Lymphocytes Biol Reprod, November 1, 2004; 71(5): 1638 - 1645. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Koelle, Z. Liu, C. L. McClurkan, R. C. Cevallos, J. Vieira, N. A. Hosken, C. A. Meseda, D. C. Snow, A. Wald, and L. Corey Immunodominance among herpes simplex virus-specific CD8 T cells expressing a tissue-specific homing receptor PNAS, October 28, 2003; 100(22): 12899 - 12904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. A. Macdonald, A. W. Purcell, N. A. Mifsud, L. K. Ely, D. S. Williams, L. Chang, J. J. Gorman, C. S. Clements, L. Kjer-Nielsen, D. M. Koelle, et al. A Naturally Selected Dimorphism within the HLA-B44 Supertype Alters Class I Structure, Peptide Repertoire, and T Cell Recognition J. Exp. Med., September 2, 2003; 198(5): 679 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. BenMohamed, G. Bertrand, C. D. McNamara, H. Gras-Masse, J. Hammer, S. L. Wechsler, and A. B. Nesburn Identification of Novel Immunodominant CD4+ Th1-Type T-Cell Peptide Epitopes from Herpes Simplex Virus Glycoprotein D That Confer Protective Immunity J. Virol., September 1, 2003; 77(17): 9463 - 9473. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F.-X. Mbopi-Keou, L. Belec, J. Dalessio, J. Legoff, G. Gresenguet, P. Mayaud, D. W. G. Brown, and R. Ashley Morrow Cervicovaginal Neutralizing Antibodies to Herpes Simplex Virus (HSV) in Women Seropositive for HSV Types 1 and 2 Clin. Vaccine Immunol., May 1, 2003; 10(3): 388 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Koelle and L. Corey Recent Progress in Herpes Simplex Virus Immunobiology and Vaccine Research Clin. Microbiol. Rev., January 1, 2003; 16(1): 96 - 113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gierynska, U. Kumaraguru, S.-K. Eo, S. Lee, A. Krieg, and B. T. Rouse Induction of CD8 T-Cell-Specific Systemic and Mucosal Immunity against Herpes Simplex Virus with CpG-Peptide Complexes J. Virol., June 5, 2002; 76(13): 6568 - 6576. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Koelle, H. B. Chen, C. M. McClurkan, and E. W. Petersdorf Herpes simplex virus type 2-specific CD8 cytotoxic T lymphocyte cross-reactivity against prevalent HLA class I alleles Blood, May 15, 2002; 99(10): 3844 - 3847. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, clone 1874.1991.22 and UL47
289–298; 
, clone 1874.1997.51 and ICP0 92–101; 
, clone
5491.2000.48 and UL49 49–57. Lysis of mock-loaded targets was <5%
specific release for each clone.
