| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,,
* Center for Cell and Gene Therapy, Departments of
Molecular and Human Genetics and
Immunology, Baylor College of Medicine, Houston, TX 77030; and
Department of Bioengineering, Rice University, Houston, TX 77005
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
production, and MHC I
peptide-tetramer staining. By contrast, simultaneous exposure of the
immune system to this agonist/antagonist pair not only fails to
generate T cell antagonism in vivo, but also avoids original antigenic
sin. These observations suggest that simultaneous immunization with
vaccines containing mutant epitopes, even T cell antagonists, can
indeed generate a diverse array of T cell responses and that at least
some immune interference can be avoided by delivering mutant Ags to the
immune system simultaneously. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
A variety of mechanisms could provoke immune interference of T cell responses after vaccination with mutant Ags. T cell antagonism and original antigenic sin (OAS)3 are two such mechanisms, since both can be activated by exposure of the immune system to mutant epitopes or altered peptide ligands (APLs) (4). T cell antagonism occurs when an existing memory T cell is functionally inactivated by exposure to a point mutant of its cognate epitope presented on MHC I (reviewed in Ref. 5). A number of studies using the immunodominant OVA epitope SIINFEKL demonstrate that certain epitopes with single amino acid changes have the ability to antagonize T cell function by binding the TCR in the context of MHC I without inducing receptor activation (6). Given these observations, it is possible that immunization with a multiple mutant vaccine could generate T cell antagonism between mutant epitopes rather than multivalent CTL responses.
The dictum of OAS, discovered first in B cell biology, indicates that when an individual is exposed to an Ag that is similar to, but not identical to, an Ag to which he/she was previously exposed by either infection or immunization, the immune response to the second Ag will still be directed against the first Ag (7, 8). This phenomenon has subsequently been observed in T cells against lymphocytic choriomeningitis virus (LCMV) viral variants (4). In this case, exposure to first one and then a second LCMV strain generated CTLs specific to only the first strain. This observation suggests that vaccines delivering mutant epitopes, particularly at different times, have the potential to induce OAS to interfere with vaccine efficacy. Furthermore, these data suggest that a vaccine with limited epitopes may have the potential to prevent subsequent naive T cell responses against legitimate pathogens if their epitopes diverge from those of the original vaccine Ags.
In view of these observations, this study was performed to determine
whether immune interference by T cell antagonism, OAS, or other
mechanisms will limit the application of multiple mutant vaccines. To
test this, we used an epitope system representing the "worst case"
scenario for a vaccine in which an agonist epitope is combined with its
own antagonist epitope. To do this, C57BL/6
(H2-Kb) mice were genetically immunized using a
gene gun with minigenes encoding the
H2-Kb-restricted immunodominant epitope of OVA
(OVA257–264 = SIINFEKL, OVA hereafter) and its
point mutant antagonist (9) epitope SIINFEDL (OVANT
hereafter). After immunization, Ag-specific memory and effector CTLs
were analyzed by standard 4-h 51Cr release assay,
intracellular IFN- staining, and peptide-MHC-tetramer staining.
Results presented here demonstrate that both OVA and OVANT are mutual
in vitro antagonists and that sequential immunization with these
epitopes in either order generates a novel T cell immune response which
could be accommodated in the existing definition of OAS of CTLs.
Furthermore, data presented here demonstrate that simultaneous
immunization with this in vitro agonist/antagonist pair generates an in
vivo T cell repertoire consisting of both OVA- and OVANT (mono-)-
specific T cells without any sign of in vivo T cell antagonism or OAS
of T cells even if the epitopes are expressed in the cytoplasm of the
same cells. Since a number of persistent pathogens use antagonism and
OAS to perpetuate in the host, the results presented in this study may
help in the design and delivery of vaccines to effectively control
these immunoevasive and mutable pathogens.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Abs were purchased from BD PharMingen (San Diego, CA). Tissue culture reagents were purchased from Life Technologies (Gaithersburg, MD). T-Stim was purchased from Collaborative Biosciences (Bedford, MA). Endotoxin-free plasmid DNA purification columns were purchased from Qiagen (Chatsworth, CA). Peptides were synthesized by Research Genetics (Huntsville, AL). MHC I-SIINFEKL and MHC I-SIINFEDL tetramers were supplied by the National Institutes of Health Tetramer Core. All other reagents were purchased from Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Pittsburgh, PA), or VWR Scientific (Houston, TX).
Mice
Female C57BL/6 (H-2b) mice were purchased from Harlan Sprague Dawley (Indianapolis, IN). All mice were maintained in the American Association of Laboratory Animal Care-approved Center for Comparative Medicine at Baylor College of Medicine (Houston, TX).
Plasmids and genetic immunization
OVA was PCRed from a full-length OVA gene. OVANT was prepared by site-directed mutagenesis of a full-length OVA gene. Each construct was cloned in the pCMVi backbone. pCMVi was constructed from the pGL3 backbone from Promega (Madison, WI) to include the CMV promoter, a chimeric 5' intron, and the human growth hormone polyadenylation sequence. Each epitope was cloned into the plasmid with the addition of a Kozak methionine (ACC ATG) before the epitope and a TAA stop after the last amino acid of the epitope. Four- to 6-wk-old mice were genetically immunized by gene gun transfection of the epidermis using the Helios biolistic device (Bio-Rad, Hercules, CA) using 250 psi of helium. A total of 2.5 µg of plasmid DNA was delivered per shot of 1.6-µm gold particles, which were prepared as per the device instructions. In most cases, two shots were delivered into the backs of the ears on each mouse at each immunization except where noted in the text. In all cases, mice were vaccinated in a single round of immunization without further boosts.
CTL assay
Four weeks after immunization, splenocytes were harvested and stimulated in vitro for 6 days with 1 µg of relevant peptide in RPMI 1640 medium containing 10% v/v FBS, 10 mM HEPES, 50 µM 2-ME, 10% v/v T-Stim (Collaborative Biomedical Products) as a source of IL-2 and antibiotic/antimycotic. On day 7, CTL activity of the cultured splenocytes was assayed by a 4-h 51Cr release assay as detailed previously (10). Each experiment was done in triplicate and was repeated at least three times.
RMA target cells were labeled with 1 µM of the indicated peptide for 1 h at 37°C and excess peptide was removed by washing. Target cells were labeled with 100 µCi of 51Cr (as sodium chromate) for 1 h in a 1-ml volume. After washing, 1 x 104 labeled target cells were added to effector cells at the indicated E:T ratios. Percent specific lysis was calculated as: ((cpm of sample - cpm of spontaneous release)/cpm of maximum release - cpm of spontaneous release)) x 100. Spontaneous release was defined as the mean cpm released from six replicates of 1 x 104 labeled cells incubated in medium alone. Maximum release was defined as the mean cpm released from six replicates of 1 x 104 labeled cells incubated in medium containing 1% SDS.
Tetramer staining
Four weeks after immunization, splenocytes from nonimmunized or
immunized mice were harvested. Fresh lymphocytes or lymphocytes
stimulated for 6 days with 1 µM peptide were used for tetramer
staining. These stimulated cells were stained with 100 ng each of
anti-CD3-PerCP (PerCP-conjugated hamster anti-mouse CD3
(CD3 chain) mAb) and anti-CD8-FITC (FITC-conjugated rat
anti-mouse CD8a (Ly-2) mAb) or with isotype control of
anti-CD8-FITC conjugate and PE-conjugated MHC I-SIINFEKL and MHC
I-SIINFEDL tetramers for 30 min at 4°C in the dark in 100 µl of
FACS buffer (PBS containing 2% FBS and 0.02% of sodium azide). Cells
were then washed twice with FACS buffer and sample data were acquired
and analyzed on a BD Biosciences FACScan instrument and analyzed using
CellQuest software BD Biosciences, Mountain View, CA).
Intracellular IFN- staining
Fresh lymphocytes or lymphocytes stimulated for 6 days with
peptides or cells were stimulated in vitro at 1 x
106 cells/ml for 4–6 h with 10 µM (OVA or
OVANT) peptide in the presence of 5 µg/ml Golgi Plug (brefeldin A).
These stimulated cells were then stained with 100 ng each of
anti-CD3-PerCP (PerCP-conjugated hamster anti-mouse CD3
(CD3 chain) mAb) and anti-CD8-FITC (FITC-conjugated rat
anti-mouse CD8a (Ly-2) mAb) or with isotype control of
anti-CD8-FITC conjugate for 30 min at 4°C in the dark in 100 µl
of FACS buffer (PBS containing 2% FBS and 0.02% of sodium azide).
Cells were then washed twice with FACS buffer and intracellular IFN-
staining was performed using a Cytofix/Cytoperm Plus kit from BD
PharMingen according to the manufacturer’s instruction. Briefly, CD3-
and CD8-labeled cells were fixed and permeabilized for 20 min at 4°C
in 250 µl of Cytofix/Cytoperm solution and then were washed twice
with Perm/Wash solution. The permeabilized cells were then incubated
with 100 ng of PE-conjugated anti-IFN--labeled Ab for 30 min at
4°C. The cells were then washed twice with Perm/Wash solution and
were resuspended in FACS buffer for further analysis.
Sample data were acquired and analyzed on a BD Biosciences FACScan instrument and analyzed using CellQuest software.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
To determine whether T cell antagonism and OAS could
interfere with immune responses to vaccines, we tested a defined
epitope system in vivo by genetic immunization. The
H-2Kb-restricted immunodominant peptide epitope
of OVA (Ova257–264 = SIINFEKL) was
tested in combination with its point mutant SIINFEDL, which
is a well-defined T cell antagonist of OVA-specific CD8 T cells
(9) and carries a K 
D point mutation at position 7, a
primary TCR contact site. These epitopes were expressed as cytoplasmic
peptides from minigene plasmids expressing either
M-SIINFEKL or M-SIINFEDL. C57BL/6
H-2b mice were genetically immunized with each
plasmid once with a gene gun. Splenocytes harvested 4 wk later from
both OVA and OVANT plasmid-immunized mice demonstrated potent
epitope-specific CTL activity with little cross-reactivity (Fig. 1
). The observed CTL response induced by
minigenes occurred in the absence of delivery of any CD4 Th epitope.
This is consistent with several reports demonstrating CD8 T cell
responses in the absence of CD4 T cell help by the gene gun (Ref.
10 and the references therein). The antagonistic
properties of the OVANT peptide were confirmed by the pre-pulse CTL
lysis assay (9) and demonstrated that OVANT peptide does
inhibit OVA-specific CTL activity in vitro (Fig. 1A).
Moreover, OVA peptide reciprocally inhibited OVANT-specific CTLs in
vitro (Fig. 1D). Although inhibition by OVA was detectable,
this in vitro inhibition was substantially weaker than that mediated by
OVANT.
|
), CD8/IFN- production (Fig. 3, A–C), and MHC
I-tetramer-binding activity (Fig. 3, D and E)
against both the OVA and OVANT epitopes. The specificity of CTL
activity from the coimmunized mice depended entirely upon the peptide
used for 6-day in vitro expansion of CTL effectors (Fig. 1
), indicating
that the splenocytes contained CTL precursors against both
epitopes.
|
|
, C and D).
Partial cross-reactivity of the T cells for the alternate epitope was
expected given that a T cell antagonist must bind TCR to evoke its
inhibitory effects (9). Although there was some low level
cross-binding of TCR by the two epitopes on MHC I, this did not
translate into the production of functionally cross-reactive T cells,
as 1) splenocytes from mice immunized with only OVA and stimulated with
OVANT peptide failed to generate OVANT-specific CTLs (Fig. 1C) or CD8/IFN- production (Fig. 3) and 2) stimulation of
splenocytes from OVANT-immunized mice with OVA peptide failed to
produce OVA-specific CTLs (Fig. 1B) or CD8/IFN-
production (Fig. 3). Therefore, although both OVA and OVANT inhibit each other’s T cells to varying degrees in vitro, they fail to induce immune interference in vivo when they are presented to the immune system at the same time or even on the same cell. These results suggest, but do not formerly prove, that T cell antagonism or epitope competition may largely be an in vitro phenomenon where high levels of an inhibitor peptide can attenuate agonist-specific T cell responses. By contrast, too little inhibitor peptide may be produced by vaccines in vivo to effectively antagonize CD8 T cells (5). This model is consistent with observations of in vivo MHC II-restricted T cell antagonism in which antagonism can be avoided by decreasing the ratio of antagonist to agonist (12).
Sequential immunization with agonist/antagonist pair results in OAS of CTLs
In vivo immune interference was not observed after coimmunization.
In contrast, sequential exposure of the immune system to mutant
epitopes or APLs can reveal OAS in which pre-existing CD8 T cell
responses against one epitope reduces subsequent CD8 responses against
a point mutant of the same epitope (4). Given this
observation, we tested whether the OVA/OVANT system would generate OAS
when these APLs were delivered sequentially rather than simultaneously.
C57BL/6 mice were immunized by gene gun with either the OVA or OVANT
plasmid. One month later, mice were reimmunized with the same epitope
or with the counter epitope and their splenocytes were assayed 4 wk
later for CTL and tetramer binding. Splenocytes from mice immunized
twice with OVA plasmid produced OVA-specific CTLs,
CD8/IFN- production, and OVA tetramer binding, but
negligible OVANT-specific responses (Fig. 4 and data not shown). Mice immunized
with OVA plasmid and then boosted with OVANT plasmid showed high
OVA-specific CD8 responses, but no OVANT-specific CD8 responses.
Reciprocal immunizations demonstrated that OVANT plasmid priming
ablated subsequent OVA-specific CD8 responses (Fig. 4).
|
Although the exact mechanism of OAS for T cells remains unclear, at least three models have been proposed (13): 1) During OAS, a weakly cross-reactive second epitope reactivates the larger number of memory CTLs against the first epitope more effectively than it activates the small number of naive T cells specific for the second epitope. 2) OAS is due to killing or deactivation of cells presenting the second Ag by primary CTLs that cross-react against the second Ag. 3) OAS occurs because CTLs responding to the variant epitope experience "a form of T cell antagonism" where CTL clones could react suboptimally to the variant epitope and become partially activated, leading to anergy.
In OAS with the OVA/OVANT system, models 1 and 2 appear unlikely, since these mechanisms both require cross-recognition and cross-activation of the original T cells by the second Ag which does not appear to occur between OVA and OVANT. However, it is possible that more subtle interactions that are not represented by our in vitro assays may occur in vivo to invoke these models. Model 3 cannot be ruled out, but seems unlikely since T cell antagonism in vivo in the MHC II system required the antagonist to be present in higher amounts than the agonist (12). In contrast, after genetic immunization with a gene gun, 95% of expressed transgene product was lost within 5 days (14). Therefore, the amount of second Ag produced upon second immunization should be in marked excess over the residual potentially antagonistic epitope remaining from the first immunization. However, it is conceivable that the first Ag could be sequestered in high concentrations on dendritic cells in the draining lymph nodes that could overcome the Ag ratio requirement for in vivo antagonism.
Although our data cannot unequivocally exclude models 1 and 2, we
propose an alternative model to account for OAS for the OVA/OVANT
system. We hypothesize that the large population of pre-existing memory
T cells may cross-bind the APLs on MHC I dendritic cells in the
draining lymph nodes, thereby blocking access to the APCs by low
frequency naive T cells. Complete blockade of all MHC I molecules would
not be required for OAS. Rather, memory T cells would need only to
block sufficient numbers of MHC I to prevent the naive T cells from
triggering the 
8000 TCRs needed for T cell activation
(15). The "masking" model predicts that any
cross-reactive APLs will generate OAS. OAS involving full T cell
antagonists will ablate subsequent responses and not boost primary
responses. By contrast, OAS by partial T cell agonists/antagonists may
only partially attenuate subsequent responses and will likely partially
boost primary responses of T cells against the first mutant epitope
(4).
For mutable, immunoevasive pathogens like HIV-1, there might be a selective advantage for viruses bearing antagonistic APLs over other APLs, since antagonistic peptides cannot only escape pre-existing CTLs, but also inactivate pre-existing CTLs (at least in vitro) (16, 17). Our data favoring a masking model suggest that antagonist APLs may also activate OAS to suppress new CTLs against variant viruses. Although continuous large-scale production of divergent APLs by HIV-1 may eventually overcome OAS, even transient reductions or delays in the production of new CTLs may confer significant selective advantages to mutable viruses over the course of latent infection.
This work also has practical implications for the delivery of multimutant vaccines. Sequential immunization with APLs generates the extreme case of immune interference by activating OAS. By contrast, simultaneous delivery of the same problematic epitopes avoids both overt antagonism and OAS in vivo. These data, along with data demonstrating the ability to provoke multivalent T cell responses by simultaneous delivery of 64 Th epitopes (18) or a 32-plasmid genetic vaccine (19), suggest that the immune system can indeed respond effectively to multiple APLs, even if they include T cell antagonists. Simultaneous APL delivery contrasts with the sequential APL exposure produced by annual vaccines like influenza (7, 8) and during on-going infections by mutating pathogens like HIV-1 (16, 17), in which OAS and T cell antagonism are observed. Problematic natural immune interference mechanisms can be readily avoided by simultaneous delivery of multigene, multimutant vaccines.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Michael Barry, One Baylor Plaza, BCM505 Houston, TX 77030. E-mail address: mab{at}bcm.tmc.edu
3 Abbreviations used in this paper: OAS, original antigenic sin; APL, altered peptide ligand; LCMV, lymphocytic choriomeningitis virus.
Received for publication May 6, 2002. Accepted for publication October 3, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
by gene gun DNA vaccination with a minigene encoding a CTL epitope of Listeria monocytogenes. Vaccine 19:4297.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. Liu, B. A. Ewald, D. M. Lynch, A. Nanda, S. M. Sumida, and D. H. Barouch Modulation of DNA Vaccine-Elicited CD8+ T-Lymphocyte Epitope Immunodominance Hierarchies J. Virol., December 15, 2006; 80(24): 11991 - 11997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. A. Gillespie, G. Stewart-Jones, J. Rengasamy, T. Beattie, Job. J. Bwayo, F. A. Plummer, R. Kaul, A. J. McMichael, P. Easterbrook, T. Dong, et al. Strong TCR Conservation and Altered T Cell Cross-Reactivity Characterize a B*57-Restricted Immune Response in HIV-1 Infection J. Immunol., September 15, 2006; 177(6): 3893 - 3902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. S. Liu, J. Dyer, G. R. Leggatt, G. J. P. Fernando, J. Zhong, R. Thomas, and I. H. Frazer Overcoming Original Antigenic Sin to Generate New CD8 T Cell IFN-{gamma} Responses in an Antigen-Experienced Host. J. Immunol., September 1, 2006; 177(5): 2873 - 2879. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Dorrell, H. Yang, B. Ondondo, T. Dong, K. di Gleria, A. Suttill, C. Conlon, D. Brown, P. Williams, P. Bowness, et al. Expansion and Diversification of Virus-Specific T Cells following Immunization of Human Immunodeficiency Virus Type 1 (HIV-1)-Infected Individuals with a Recombinant Modified Vaccinia Virus Ankara/HIV-1 Gag Vaccine. J. Virol., May 1, 2006; 80(10): 4705 - 4716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Allen, X. G. Yu, E. T. Kalife, L. L. Reyor, M. Lichterfeld, M. John, M. Cheng, R. L. Allgaier, S. Mui, N. Frahm, et al. De Novo Generation of Escape Variant-Specific CD8+ T-Cell Responses following Cytotoxic T-Lymphocyte Escape in Chronic Human Immunodeficiency Virus Type 1 Infection J. Virol., October 15, 2005; 79(20): 12952 - 12960. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Sugimoto, D. E. Kaplan, F. Ikeda, J. Ding, J. Schwartz, F. A. Nunes, H. J. Alter, and K.-M. Chang Strain-Specific T-Cell Suppression and Protective Immunity in Patients with Chronic Hepatitis C Virus Infection J. Virol., June 1, 2005; 79(11): 6976 - 6983. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Howe, J. K. Craigo, C. J. Issel, and R. C. Montelaro Specificity of serum neutralizing antibodies induced by transient immune suppression of inapparent carrier ponies infected with a neutralization-resistant equine infectious anemia virus envelope strain J. Gen. Virol., January 1, 2005; 86(1): 139 - 149. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Splenocytes stimulated
with OVA peptide; 
, splenocytes stimulated with OVANT peptide.
D and E, Peptide-MHC I-tetramer staining
of OVA- and OVANT-specific CD8 cells in mice coimmunized with pOVA and
pOVANT. For tetramer staining, fresh (D) or 6-day
cultured splenocytes (E; 1 x
106/sample) were stained with anti-CD3-PerCP,
CD8-FITC-conjugated Abs, and PE-conjugated OVA-MHC I-tetramer or
OVANT-MHC I-tetramer (both from the National Institutes of Health
Tetramer Core) for 30 min at room temperature. On a FACScan, CD3/CD8
double-positive cells (5–10 x 104/sample) were
analyzed for the presence of tetramer-positive cells using CellQuest
software. All of the experiments were repeated at least three times.
Data shown are the mean ± SD from six mice from two different
immunizations. 