Specific Immunotherapy by Genetically
Engineered APCs: The ''Guided Missile''
Strategy
This information is current as
of July 28, 2017.
Bo Wu, Jian-Ming Wu, Alexei Miagkov, Robert N. Adams,
Hyam I. Levitsky and Daniel B. Drachman
J Immunol 2001; 166:4773-4779; ;
doi: 10.4049/jimmunol.166.7.4773
http://www.jimmunol.org/content/166/7/4773
Subscription
Permissions
Email Alerts
This article cites 36 articles, 19 of which you can access for free at:
http://www.jimmunol.org/content/166/7/4773.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2001 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
References
Specific Immunotherapy by Genetically Engineered APCs:
The “Guided Missile” Strategy1
Bo Wu,* Jian-Ming Wu,* Alexei Miagkov,* Robert N. Adams,* Hyam I. Levitsky,†
and Daniel B. Drachman2*
O
ne of the major goals of immunotherapy is to eliminate
specific immune responses, without otherwise affecting
the immune system (1). We have developed a powerful
new strategy using genetically engineered APCs as “guided missiles” to target and eliminate Ag-specific T cells. Because virtually
all naturally occurring immune responses are not only highly heterogeneous but also unique to the individual (2–5), it is necessary
to devise a method capable of targeting the entire spectrum of each
individual’s unique Ag-specific T cell repertoire. The present strategy is based on the natural ability of APCs from a given individual
to process and present the Ag so as to target that individual’s entire
repertoire of Ag-specific T cells, no matter how heterogeneous it
may be. For T cell targeting in this study, we have engineered
APCs with a gene construct coding for the model Ag influenza
hemagglutinin (HA)3 fused to signals that direct the APCs to process and present the Ag in association with MHC class II (i.e., the
lysosome-associated membrane protein, LAMP-1) (6 – 8). As a
“warhead,” we have engineered the APCs to express Fas ligand
(FasL). When FasL interacts with Fas, which is highly expressed
*Neuromuscular Research Laboratory, Department of Neurology, and †Department of
Oncology, Johns Hopkins School of Medicine, Baltimore, MD 21287
Received for publication November 8, 2000. Accepted for publication January
26, 2001.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported in part by National Institutes of Health Grants NS07368
and 1R01NS37205 and by the Muscular Dystrophy Association. Generous support
was provided by the C. W. Parke Family Foundation, the Ann and Donald Brown
Family Foundation, and the Eleanor Denmead Ingram Foundation.
2
Address correspondence and reprint requests to Dr. Daniel B. Drachman, Department of Neurology, Johns Hopkins School of Medicine, Meyer 5-119, 600 N. Wolfe
Street, Baltimore, MD 21287-7519. E-mail address: [email protected]
3
Abbreviations used in this paper: HA, hemagglutinin (influenza); BrdU, bromodeoxyuridine; FADD, Fas-associated death domain; FasL, Fas ligand; LAMP-1, lysosome-associated membrane protein-1; PUVA, psoralen plus UV light A; TK, thymidine kinase; TrFADD, truncated FADD; VV, vaccinia virus; VVV, vaccinia virus
vector; wt, wild type.
Copyright © 2001 by The American Association of Immunologists
on the surface membranes of activated T cells, it induces apoptosis
and death of the T cells (9, 10). We have protected the APCs from
self-destruction by the FasL by inserting a third gene, expressing
a truncated form of the Fas-associated death domain (TrFADD),
which acts in a dominant negative fashion to prevent Fas-FasLinduced apoptosis of the APCs (11, 12). To transfer the three gene
constructs individually or simultaneously into APCs, we have prepared a series of recombinant vaccinia virus vectors (VVV) (13–
15). Attenuation of the VVV (by treatment with psoralen and UV
light (PUVA)) prevents replication of the virus but permits efficient infection of APCs and expression of the VVV-transferred
gene products (16).
In the present study, we have explored the ability of APCs with
VVV-transferred genes to target and eliminate Ag-specific T cells,
using a murine transgenic T cell model specific for influenza HA
(17) in vitro. Our findings indicate that the APCs express all three
gene products and effectively and specifically induce apoptosis of
HA-specific T cells, while sparing T cells of other specificities.
Materials and Methods
Mice
BALB/c transgenic mice expressing an ␣␤TCR specific for the HA epitope
(18) 110 –120 were bred and maintained in the animal care facilities at the
Johns Hopkins University. BALB/c transgenic mice expressing an ␣␤T
cell receptor specific for OVA (DO11.10 strain (19)) were a gift of Dr. K.
Murphy (Washington University, St. Louis, MO). Wild-type (wt) BALB/c
mice (8 –12 wk old) were purchased from The Jackson Laboratory (Bar
Harbor, ME) or the National Cancer Institute (Frederick, MD). All animal
experiments were performed in accordance with protocols approved by the
Animal Care and Use Committee of the Johns Hopkins University School
of Medicine.
Cell lines
A20, human TK⫺, and CV-1 cell lines were purchased from American
Type Culture Collection (Manassas, VA). The MC57G mouse fibroblast
cell line was provided by Dr. D. Pardoll (Johns Hopkins University).
0022-1767/01/$02.00
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
We tested the hypothesis that APCs genetically engineered to present an Ag and to express Fas ligand (FasL) simultaneously can
target and eliminate Ag-specific T cells. Transgenic T cells specific for influenza hemagglutinin (HA) were used as targets. We
prepared recombinant vaccinia virus vectors (VVV) to transfer the gene constructs individually or simultaneously into APCs. We
prevented unwanted viral replication by attenuating the VVVs with psoralen-UV light treatment. For presentation of the HA Ag,
APCs were transduced with cDNA for HA flanked by sequences of the lysosome-associated membrane protein that direct efficient
processing and presentation of the Ag by APCs. As a “warhead” for the APCs, we transduced them with the gene for FasL, which
induces apoptosis of Fas-expressing activated T cells. To protect the transduced APCs from self-destruction by FasL, we transferred cDNA for a truncated form of Fas-associated death domain, which inhibits Fas-mediated cell death. Our results show that
the engineered APCs effectively expressed the genes of interest. APCs transduced with VVV carrying all three gene constructs
specifically killed HA-transgenic T cells in culture. Coculture with T cells specific for an unrelated Ag (OVA) had no significant
effect. Our in vitro findings show that APCs can be genetically engineered to target and kill Ag-specific T cells and represent a
promising novel strategy for the specific treatment of autoimmune diseases. The Journal of Immunology, 2001, 166: 4773– 4779.
4774
SPECIFIC IMMUNOTHERAPY BY GENE TRANSFER TO APC
Insertion of genes into VV and selection
The genes of interest were first ligated into appropriate transfer plasmids
and then transfected into the vaccinia virus (VV) by homologous recombination in predetermined loci within the viral genome (14). To insert more
than one gene, different transfer plasmids are used sequentially, inserted
into different loci, and using different methods of selection at each successive step of recombination (13, 15). In these studies, we have inserted up
to three genes in a single VV (Fig. 1) and have demonstrated that all of the
genes are efficiently expressed by infection of APCs. The first gene (HALAMP), which induces processing and presentation of HA, was inserted in
a transfer plasmid that recombines in the J2R region of the virus and disrupts the vaccinia thymidine kinase (TK) gene. Recombined VVV was
selected with medium containing bromodeoxyuridine (BrdU), which is lethal to cells that express TK. VV with a cDNA construct for HA fused to
LAMP-1 was a generous gift of D. Pardoll. The second gene (FasL) was
inserted in a transfer plasmid that reinduces expression of TK (pTK7.5b),
and recombines in the HindIII-F region of vaccinia. Selection of the recombined virus, which now expresses TK, was conducted using medium
containing methotrexate, which is lethal to cells that do not express TK.
The third gene (TrFADD), which protects Fas-expressing APCs from FasL-mediated death, was inserted in a plasmid that recombines in the I4L
region of vaccinia and induces expression of ␤-glucuronidase (15). ␤-Glucuronidase converts the chromogenic substrate 5-bromo-4-chloro-3-indolyl-␤-glucuronide to a blue color that is used for selection of plaques.
For insertion of genes into VV, CV-1 cells were seeded at 2–5 ⫻ 105/well
in six-well plates in MEM with 10% FBS (13, 15, 21). One day later, the
cells were infected with the WR strain of VV at 0.1–1 PFU/cell in 1 ml
MEM with 2.5% FBS at 37°C with gentle shaking every 20 –30 min for
1–2 h. Transfection of the CV-1 cells with the pSCmcs2 plasmid was
carried out using the Mammalian Transfection Kit (Stratagene, La Jolla,
CA) according to the manufacturer’s recommendation. The transfected
cells were harvested, frozen, and thawed three times, and sonicated to
release the VVV, and four 10-fold dilutions were made in MEM-2.5. Selection of recombinant VVV was conducted as follows. Confluent TK⫺
cells in six-well plates were infected with 1 ml of each dilution of the
transfectant by incubation at 37°C for 2 h with gentle rocking. After the
medium was discarded, the cells were gently overlaid with 3 ml warmed
(45°C) 1% LMP agarose in 1⫻ plaque medium (Life Technologies, Gaithersburg, MD) with 5% FBS (HyClone, Logan, UT), 25 ␮g/ml BrdU (Sigma, St. Louis, MO). The agarose layer was allowed to solidify at room
temperature, and the cultures were incubated at 37°C for 2 days. For blue
color selection, the cultures were overlaid with 2 ml low melting point
(LMP) agarose in 1⫻ plaque medium with 10 mg/ml neutral red and 1/150
volume of 5% 5-bromo-4-chloro-3-indolyl-␤-D-galactoside (X-Gal) and incubated at 37°C for an additional 1–2 days. Discrete blue plaques were
picked, freeze-thawed three times, and used for reselection as above. Three
to five rounds of selection were conducted until pure, single plaques were
obtained. Individual plaques were checked by PCR and RT-PCR for each
recombinant gene before being amplified. Viral stocks were produced in
TK⫺ cells in T-175 flasks, titered, aliquoted, and stored at ⫺70°C.
For production of recombinant VVV containing a second gene, we used
the TK⫹ transfer plasmid, pTK7.5b (Fig. 1). The procedures for transfection of the CV-1 cells, and TK⫹ selection were similar to those described
above, with the one critical exception that the selection medium contained
MTAGG (methotrexate (3 ␮M), thymidine (15 ␮M), adenosine (50 ␮M),
guanosine (50 ␮M), and glycine (10 ␮M); the methotrexate component
selects for TK⫹ cells), instead of BrdU. As above, three to five rounds of
selection were conducted to obtain pure two-gene VVV.
For production of recombinant VVV containing the third gene, we used
the a ␤-glucuronidase-expressing “GUS” transfer plasmid (pIV113) construct containing the TrFADD gene, and a selection method that depends
on development of color by the expressed marker ␤-glucuronidase. The
procedures for transfection of the CV-1 cells were as described above. The
first agarose overlay did not contain any selection reagent. In the second
layer, 5-bromo-4-chloro-3-indolyl-␤-glucuronide (200 ␮g/ml; Clontech,
Palo Alto, CA) was used for the development of blue color. Because this
selection method does not involve negative selection of nonrecombined
VVV, five to six rounds of selection were carried out until pure single
plaques were obtained.
Construction of transfer plasmids
We designed primers for FasL (based on published sequences (20)), amplified FasL cDNA from a human lymphocyte cDNA library by PCR, and
verified it by sequencing. The FasL cDNA was cloned into the
pZeoSV2(⫹) plasmid (Invitrogen, San Diego, CA), amplified in Escherichia coli, selected in Zeocin-containing medium. The FasL fragment was
released from the plasmid by the restriction enzymes BamHI and EcoRV
and recloned into the transfer plasmids pSC11mcs2 and pTK7.5b for homologous recombination in vaccinia virus (Fig. 1). TrFADD fragment was
a gift from Dr. V. Dixit, Genentech, South San Francisco, CA in a
pcDNA3.1 vector. The TrFADD fragment was amplified by PCR using a
pair of primers that were designed to anchor on the pcDNA3.1 plasmid and
to have ApaI restriction sites at both ends. The primer sequences were as
follows: sense primer, AATACGACTGGGCCCAGGGAGACCCAAGC
TTGG; antisense primer, TATAGAATAGGGCCCTCTAG. The PCR
product was digested with ApaI restriction enzyme and ligated into the
pIV113 transfer plasmid (Fig. 1).
In vitro T cell proliferative responses
Different numbers of lymph node cells or splenocytes were seeded in 96well plates in complete medium (RPMI with 10% FBS, 5.5 ⫻ 10⫺5 M
2-ME, 1 mM HEPES buffer, 100 U/ml penicillin, 100 ␮g/ml streptomycin,
and 0.25 ␮g/ml Fungizone) and incubated at 37°C with 5% CO2 for 2–5
days. Ag, or APCs transduced with various VVV, were added to the wells.
Cultures were pulsed with 1 ␮Ci/well [3H]TdR for 8 –16 h and harvested
onto glass fiber filters. Dried glass fiber filters were counted, and the results
were expressed as ⌬cpm ⫾ SEM.
Flow cytometry
FIGURE 1. Transfer plasmids and sites of recombination within VV.
The pSC11mcs2 plasmid was used for insertion of the first gene: either the
HA-LAMP gene (as shown) or the FasL gene. It inserts in the J2R region
of the VV, disrupting the VV TK gene. The pTK7.5b plasmid was used for
insertion of the second gene in a VVV that already has the first gene
inserted via the pSC11mcs2 plasmid. Recombination of this plasmid with
the VVV inserts a new TK gene as well as the gene of interest in the
HindIII-F region of VV. The pIV113 plasmid recombines in the I4L region
of the VV and inserts the gene of interest as well as the marker ␤-glucuronidase (GUS), which is used for selection. HSVtk, Herpes simplex virus
thymidine kinase. See text for details.
Ab to human FasL and clonotypic Ab to the DO11.10 OVA-specific T cells
(KJ1-26) were purchased from Caltag (Burlingame, CA). All other Abs
used for flow cytometry were purchased from Pharmingen (San Diego,
CA). Cells (2–10 ⫻ 105) were dispensed in 96-well plates in 1⫻ HBSS
supplemented with 1% FBS and 0.5% sodium azide. To each well were
added 10 –20 ␮l of appropriately diluted Ab solutions, and the mixture was
incubated on ice for 30 min. After three washes, 10 –20 ␮l of properly
diluted secondary Ab were added and incubated for 30 min. After fixing
with Cytofix (Pharmingen), cells were analyzed immediately or within 1
wk using a FACScan/FACSorter (BD Biosciences, San Jose, CA).
DNA fragmentation induced by FasL expressed by
recombinant VV
To express FasL, MC57G mouse fibroblast cells were infected with recombinant VVV containing the gene for FasL, or control VV, at 20 multiples of infection for 5 h in six-well plates. A20 mouse B lymphoma cells,
which express Fas abundantly, were used as target cells after labeling overnight with [3H]TdR (5 ␮Ci/106 cells/ml in 100-mm petri dishes at 37°C).
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
Insertion of genes into VVV
The Journal of Immunology
Labeled A20 cells (104 per 1.5-ml tube) were cocultured with different
numbers of transduced MC57G cells, as indicated in Fig. 3 at 37°C. Negative controls contained labeled A20 cells alone; positive controls contained labeled A20 cells with added anti-Fas Ab. After 24 h coculture, all
tubes were centrifuged at 2500 rpm for 5 min. The supernatants (⬃200 ␮l)
were collected in scintillation vials with 2.5 ml scintillation fluid. The
pellets were further extracted by mixing vigorously with 200 ␮l 1⫻ TE
buffer with 1% Triton and centrifuging at 14,000 rpm. Radioactivity in
each supernatant (S), Triton-extracted supernatant (T), both containing
fragmented DNA, and pellet (P), containing unfragmented DNA, was
counted. The result is expressed as the percentage of total DNA that was
fragmented: % of DNA fragmented ⫽ [(fragmented DNA (S ⫹ T)/(total
DNA (S ⫹ T ⫹ P))] ⫻ 100%.
PUVA attenuation of recombinant VVs
Results
Construction of recombinant VVs with FasL and TrFADD
As described above, we prepared, selected, and amplified recombinant VVV containing the three gene constructs, in five single or
multiple combinations, as follows: HA-LAMP-1; FasL; HALAMP-1 ⫹ FasL; FasL ⫹ TrFADD; HA-LAMP-1 ⫹ FasL ⫹
TrFADD (“three-gene VVV”).
Each transfer plasmid was tested by PCR and sequencing for the
accuracy of the construct. Each recombinant VVV was tested by
PCR for the presence of the appropriate gene construct(s) and by
RT-PCR for transcription of mRNA for the recombined genes. In
all cases, the selected VVV contained and expressed the gene constructs of interest (data not shown).
FIGURE 2. Stimulation of TCR-transgenic HA-specific T cells by
APCs transduced with the HA-LAMP-1 VVV. T cells from lymph nodes
of HA-transgenic mice were enriched by depletion of MHC class II cells by
paramagnetic beads. APCs were transduced by infection with attenuated
HA-LAMP-1 VVV. Cultures were prepared in 96-well plates with 5 ⫻ 104
T cells and 5 ⫻ 105 APCs per well. Unstimulated cultures were used for
background counts. At daily intervals from 2 to 7 days, triplicate sets of
cultures were pulsed with [3H]TdR overnight, and incorporated radioactivity was expressed as ⌬cpm ⫾ SEM. T cells responded vigorously to
stimulation with VVV-transduced APCs.
duced MC57G cells overnight (see details in Materials and Methods). As shown in Fig. 3b, the FasL gene products expressed by the
VVV-infected cells induced apoptosis of the A20 cells. The percentage of DNA that was fragmented varied from 40 to 80%, with
different recombinant viruses, contrasted with only 15–25% in
A20 cells incubated with control uninfected MC57G cells or
MC57G cells infected with wt VV. The results indicate that these
different recombinant viruses express strongly functional FasL
gene products.
APCs expressing VVV-transferred “three-gene” products kill
HA-specific T cells
Lymph node cells from HA-specific TCR-transgenic mice and
from OVA-specific TCR transgenic mice were collected and cultured for 48 h in RPMI 10 with the corresponding Ag, either HA
APCs expressing VVV-transferred HA-LAMP stimulate HAspecific T cells
Lymph node cells from HA-specific TCR transgenic mice were
cocultured with congenic BALB/c mouse splenocytes that had
been infected overnight with the HA-LAMP VVV. Cultures were
harvested at different time points with the addition of [3H]TdR for
the last 16 –18 h (Fig. 2). The results showed pronounced stimulation of the HA-specific T cells. This experiment demonstrated
that the product of the VVV-transferred HA-LAMP gene was processed and presented by APCs and stimulated HA-specific T cells
to proliferate.
FasL gene product expressed by recombinant viruses kills
Fas⫹ cells
Expression of FasL was tested by infection of the MC57G cell line
with each FasL-containing VVV overnight at a multiple of infection of 20:1, and staining with PE-labeled specific Ab to FasL.
Flow cytometry demonstrated that up to 83% of the MC57G cells
expressed FasL (Fig. 3a). To test the function of the FasL gene
product expressed by the recombinant viruses, we used Fas-positive A20 B lymphoma cells as targets. A20 cells pulsed with
[3H]TdR were incubated with different numbers of VVV-trans-
FIGURE 3. Expression of FasL by VVV-transduced cells. a, Flow cytometry showing expression of FasL by MC57G fibroblast cells transduced
by infection with VVV, stained with PE-labeled Ab to FasL. Dashed line,
negative control. b, Functional assay, showing DNA fragmentation of A20
cells incubated overnight with MC57G cells transduced with VVV carrying the FasL gene. Marked apoptosis occurred at all E:T ratios.
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
The following procedure was used to determine the appropriate parameters
of PUVA treatment to attenuate each batch of VVV. Viral suspensions
were freshly made in the range of 108–109 PFU/ml in 1⫻ HBSS with 0.1%
BSA, and 1 ml was transferred to each 35-mm tissue culture dish. Psoralen
(Trioxsalen; Calbiochem, La Jolla, CA) was added to a final concentration
of 1–10 ␮g/ml and incubated at 20°C for 10 min. UV irradiation was
conducted in a Stratalinker 1800 UV irradiation unit (Stratagene) for 1–10
min. PUVA-treated virus suspensions were used immediately, stored at
4°C for days, or stored in aliquots at ⫺80°C for months. Confluent monolayers of CV-1 cells were infected with 103–106 PFU virus/well in MEM2.5 and incubated overnight at 37°C with 5% CO2. Plaques were counted
after staining with 0.5% crystal violet. CV-1 cells infected with PUVAtreated VVV in another six-well plate were used for RT-PCR. The appropriate PUVA attenuation parameter is defined as the minimal dose of psoralen and shortest UV irradiation time required to eliminate plaque
formation (replication), but with preserved production of mRNA as shown
by positive RT-PCR results.
4775
4776
peptide 10 ␮g/ml or OVA 40 ␮g/ml. In 96-well plates, 2 ⫻ 104
stimulated HA or OVA LNCs were then cocultured at a ratio of
1:10 with BALB/c splenocytes that had been infected overnight
with the three-gene VVV, or with wt VV. Cells were cocultured
for 5 days and pulsed for the last 18 h with [3H]TdR. [3H]TdR
incorporated by HA cells cultured with the three-gene transduced
APCs was reduced by ⬎40%, as compared with control HA cells
that had been cultured with wild-type VV-infected APCs (Fig. 4).
OVA-specific T cells showed only minimal inhibition after coculture with the three-gene-transduced APCs. This experiment suggests that APCs with the three-gene VVV kill only the HA-specific
T cells.
Ag targeting enhances FasL killing of HA-specific T cells
We compared the ability of APCs transduced with either the threegene VVV or the VVV expressing FasL and TrFADD (but without
the targeting construct HA-LAMP) to kill HA-specific T cells (Fig.
5). Target T cells from HA-transgenic mice were spleen cells that
were first stimulated for 48 h with HA. APCs from BALB/c mice
that had been infected overnight with either the three-gene VVV or
the FasL ⫹ TrFADD VVV or control APCs infected with wt VV
were added to the cultures and coincubated overnight. The cultures
were harvested, and apoptosis was measured by standard ELISA
(Boehringer Mannheim Cell Death Detection ELISA; Boehringer
Mannheim, Indianapolis, IN). The background ODs were measured in supernatants from separately cultured APCs transduced
with each of the VVVs. The background OD for APCs transduced
with each specific VVV was subtracted from the results of the
cocultures with the corresponding APCs, to give the ⌬OD. The
enrichment factor was calculated by dividing the ⌬OD by the OD
of HA-transgenic T cells that had been cultured alone (i.e., without
APCs or Ag). The enrichment factor therefore represents apoptosis
of T cells due to coculture with the particular VVV-transduced
APCs. The results showed that HA-transgenic T cell cultures that
were coincubated with three-gene VVV-transduced APCs exhibited marked apoptosis (enrichment factor ⫽ 2.43). By contrast,
coincubation of HA-specific T cells with APCs that expressed
FasL and TrFADD, but did not present HA, did not induce apoptosis (enrichment factor ⬍1, which is not significant).
FIGURE 5. Ag targeting of VVV-transduced APCs enhances FasL effect. Spleen cells from HA-specific TCR-transgenic mice were used as
target T cells. They were prestimulated for 48 h with HA peptide (10
␮g/ml) and then cocultured overnight with BALB/c APCs infected with
various attenuated VVV, as indicated (5 ⫻ 104 target cells and 5 ⫻ 105
APCs, in triplicate microcentrifuge tubes). The supernatant was collected,
and fragmentation of DNA indicative of apoptosis was determined by an
ELISA method that detects histone-bound mono- and oligonucleotides (see
text). The background ODs were measured in supernatants from separately
cultured APCs transduced with each of the VVVs and subtracted from the
results of the cocultures with the corresponding APCs to give the ⌬OD.
The “Enrichment Factor” was calculated by dividing the ⌬OD by the OD
of HA-transgenic T cells that had been cultured alone (i.e., without APCs
or Ag). The enrichment factor therefore represents apoptosis of T cells due
to coculture with the particular VVV-transduced APCs. The enrichment factor
is considered positive if it is significantly greater than 1. Coculture with threegene-transduced APCs induced a marked increase of DNA fragmentation;
coculture with APCs transduced with FasL and TrFADD did not.
Time course of stimulation and killing of naive HA-specific T
cells by APCs transduced with three-gene VVV
In preliminary experiments, we found that naive transgenic HAspecific T cells had to be stimulated for ⬃2 days to render them
vulnerable to killing either by APCs transduced with the threegene VVV or by Ab to Fas. This is consistent with the requirement
for activation of T cells to induce up-regulation of Fas and vulnerability to FasL (9, 10, 22). To determine whether our transduced APCs could first stimulate, and then kill, these naive HAspecific T cells, we coincubated the T cells from transgenic HA
mice with APCs that had been transduced with the three-gene
VVV. Control APCs were infected with HA-LAMP VVV or wt
VV. The cocultures with wt VV-infected APCs were either stimulated with 5 ␮g HA (as shown) or unstimulated and used for
determination of backgrounds. Cocultures were grown for 8 days,
and duplicate sets of cultures were pulsed with [3H]TdR for 18 h
on each day from day 2 through day 8. The results indicated that
all three groups were initially stimulated, reaching a peak on day
4 (Fig. 6). However, the T cells cocultured with three-gene APCs
showed a rapid reduction of [3H]TdR incorporation after the peak
and zero incorporation after day 5, suggestive of death of the cells.
[3H]TdR incorporation declined slowly in the HA-stimulated T
cells but remained active through the remainder of the 8-day
experiment.
Discussion
Although current treatments of autoimmune diseases using general
immunosuppressive agents are often effective, they have important
drawbacks. The immune system as a whole is suppressed, thereby
increasing the risks of infection and neoplasia, and the agents may
have numerous other adverse side effects (23, 24). Ideally, treatment should eliminate the specific pathogenic autoimmune response, without otherwise suppressing the immune system. It
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
FIGURE 4. Ag specificity of effect by VVV-transduced APCs. Lymph
node cells (LNC) from HA-specific TCR-transgenic mice and from OVAspecific TCR-transgenic mice were prestimulated in separate bulk cultures
with HA peptide (10 ␮g/ml) or OVA (40 ␮g/ml), respectively, for 48 h.
HA-specific or OVA-specific T cells were then cocultured with BALB/c
splenocytes that had been infected either with attenuated three-gene VVV
or with control wt VV for 5 days (2 ⫻ 104 T cells; 2 ⫻ 105 APCs, in
triplicate cultures), and pulsed for the last 18 h with [3H]TdR. Results are
expressed as cpm ⫾ SEM. Note marked inhibition of HA-specific T cells,
with negligible effect on the OVA-specific T cells.
SPECIFIC IMMUNOTHERAPY BY GENE TRANSFER TO APC
The Journal of Immunology
4777
each component of this strategy and have shown that it works
independently.
Ag presentation
should have minimal or no other adverse side effects and should be
long-lasting or permanent. The present study was designed to investigate and develop a novel strategy for specific immunotherapy,
involving genetic engineering of APCs to target and eliminate Agspecific T cells. Although conceptually simple, this ideal poses a
number of difficult challenges. 1) All the relevant Ag-specific lymphocytes must be targeted. Because virtually all naturally occurring immune responses are not only highly heterogeneous but also
unique to the individual (2–5), it is necessary to devise a method
capable of targeting the entire spectrum of each individual’s specific T lymphocytes. The approach used in the present study utilizes the individual’s own APCs, genetically engineered to process
and present the Ag of interest. These APCs can naturally target the
entire repertoire of the same individual’s Ag-specific T cells, no
matter how heterogeneous it may be. 2) The targeted T cells must
be destroyed. For this purpose, we have provided the APCs with
FasL as a warhead, which induces apoptosis of Fas-expressing T
cells. 3) The APCs must be protected from self-destruction by the
potentially dangerous FasL. We have used a gene that encodes a
truncated mutant of FADD to protect against Fas-mediated apoptosis. 4) It is essential that all three gene constructs, i.e., for targeting the Ag, for expressing FasL, and for protecting the APCs,
be inserted into and expressed simultaneously by the engineered
APCs. Expression of one of these genes without the others could
be counterproductive; Ag presentation alone could stimulate,
rather than kill, the Ag-specific T cells; the lethal warhead alone,
without the targeting system, might damage cells other than the
Ag-specific T cells; FasL without the protective TrFADD could
destroy the APCs. To assure simultaneous insertion of these gene
constructs into APCs, we have developed a VVV that carries all
three genes. 5) The vector must insert the genes only into the
APCs; if other cells were to become infected, expression of the Ag
or FasL could produce harmful results. To prevent such uncontrolled spread, we attenuate the VVV so that it cannot replicate and
spread, and we infect the APCs directly with VVV.
Our results clearly demonstrate that APCs expressing all 3 gene
constructs induce apoptosis and death of HA-specific T cells, while
sparing T cells with other specificities. Moreover, we have tested
Fas-mediated cell death
The Fas-FasL system is of fundamental importance in the regulation of T lymphocytes. Fas (CD95) is present at the surface membranes of T cells, and is upregulated when they are activated (9, 10,
22). When FasL molecules interact with Fas, they cross-link Fas,
initiating a series of steps that activate caspases and result in apoptosis of the Fas-bearing cells (33–35). In the present studies, we
transferred the gene for FasL by means of recombinant VVV. Our
results show that the FasL gene product is highly expressed by
transduced cells, as demonstrated both by flow cytometry and
functionally by killing stimulated HA-specific T cells as well as
Fas-bearing A20 cells. Because of its lethal effect on activated T
cells, FasL can therefore be used as a potent agent to eliminate
Fas-expressing lymphocytes. Consistent with our findings, a previous study has shown that infection of APCs with adenovirus
engineered to carry the gene for FasL induced T cell tolerance to
the adenovirus, by producing apoptosis of T cells specific for the
adenovirus, and infection of macrophages with FasL-expressing
adenovirus induced tolerance to cell surface Ags of the macrophage (36, 37).
Truncated FADD
Studies of the role of FADD have led to a robust strategy for
preventing Fas-mediated cell death. FADD is associated with the
cytoplasmic portion of Fas and normally participates as an intermediary in the Fas-mediated cell death pathway. However, a
FADD deletion mutant lacking aa 1–79 acts as a dominant negative, which inhibits Fas-mediated cell death (11, 12). We have
obtained cDNA for this TrFADD mutant. Expression of the gene
for TrFADD in the highly vulnerable A20 cells confers protection
against cell death mediated by the Fas pathway (J.-M. Wu unpublished results). In the present study, we used VVV with the
TrFADD gene in association with the gene for FasL, to avoid the
risk of self-destruction of the APCs.
VV as a vector for gene transfer
VV has been used for ⬎15 years as a vector for expression of
genes in mammalian cells (14) including APCs (29). For our purposes, vaccinia has several important characteristics, including:
1) the ability to transfer multiple genes simultaneously; 2) high
level production of the proteins encoded by the transferred genes;
3) the ability to be attenuated (by treatment with psoralen and UV
light (16, 38)), so that it does not replicate (and therefore does not
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
FIGURE 6. Time course of stimulation and killing of naive HA-specific
T cells by VVV-transduced APCs. Lymph node cells from HA-specific
TCR-transgenic mice were cocultured with BALB/c APCs transduced with
HA-LAMP1 VVV or three-gene VVV or infected with wt VV (with added
exogenous HA peptide). Each well contained 5 ⫻ 104 LNCs and 5 ⫻ 105
APCs. Triplicate sets of wells were pulsed with [3H]TdR for 18 h every day
from day 2 through day 8, and incorporated radioactivity was expressed as
⌬cpm ⫾ SEM. (Background counts from wells with LNC and wt VVinfected APCs without Ag were subtracted to give ⌬cpm.) Stimulation
peaked on day 4 in all Ag-stimulated cultures. T cells cocultured with
three-gene VVV showed a rapid reduction after the peak, and zero incorporation after day 5, suggestive of death of the cells.
“Professional” APCs normally process and present exogenously
derived Ag (25, 26). However, endogenously synthesized proteins
can also be processed and presented by APCs, provided that they
are efficiently directed to the class II processing pathway. Recent
studies have identified the protein signal LAMP-1, which can direct a variety of endogenously synthesized Ags to the class II pathways efficiently, resulting in greatly enhanced Ag presentation and
T cell immune responses (6, 7, 27–29). This requires the LAMP-1
signal sequence at the 5⬘-terminus of the Ag to ensure translocation into the endoplasmic reticulum, as well as the transmembrane/
cytoplasmic domain of LAMP-1 at its 3⬘-terminus (8, 30 –32). Our
results show that APCs transduced with recombinant VVV expressing the HA-LAMP-1 gene construct, alone or in combination
with other genes, produced vigorous stimulation (and therefore
targeting) of HA-specific T cells in vitro.
4778
Acknowledgments
We are grateful for helpful discussions with Dr. Drew Pardoll.
References
1. Hohlfeld, R., and H. Wekerle. 1998. Novel therapeutic approaches to autoimmunity. In The Autoimmune Diseases. N. Rose and I. Mackay, eds. Academic Press,
San Diego, p. 833.
2. Wraith, D. C., H. O. McDevitt, L. Steinman, and H. Acha-Orbea. 1989. T cell
recognition as the target for immune intervention in autoimmune disease. Cell
57:709.
3. Germain, R., and D. Margulies. 1993. The biochemistry and cell biology of
antigen processing and presentation. Annu. Rev. Immunol. 11:403.
4. Lehmann, P. V., E. E. Sercarz, T. Forsthuber, C. M. Dayan, and G. Gammon.
1993. Determinant spreading and the dynamics of the autoimmune T-cell repertoire. Immunol. Today 14:203.
5. Abbas, A. K., A. H. Lichtman, and J. S. Pober. 1997. Cellular and Molecular
Immunology. W. B. Saunders Co., Philadelphia.
6. Chen, J. W., Y. Cha, K. U. Yuksel, R. W. Gracy, and J. T. August. 1988. Isolation
and sequencing of a cDNA clone encoding lysosomal membrane glycoprotein
mouse LAMP-1: sequence similarity to proteins bearing onco-differentiation antigens. J. Biol. Chem. 263:8754.
7. Guarnieri, F. G., L. M. Arterburn, M. B. Penno, Y. Cha, and J. T. August. 1993.
The motif Tyr-X-X-hydrophobic residue mediates lysosomal membrane targeting
of lysosome-associated membrane protein 1. J. Biol. Chem. 268:1941.
8. Wu, J. M., B. Wu, F. Guarnieri, J. T. August, and D. B. Drachman. 2000. Targeting antigen-specific T cells by genetically engineered antigen presenting cells:
a strategy for specific immunotherapy of autoimmune disease. J. Neuroimmunol.
106:145.
9. Drappa, J., N. Brot, and K. B. Elkon. 1993. The Fas protein is expressed at high
levels on CD4⫹CD8⫹ thymocytes and activated mature lymphocytes in normal
mice but not in the lupus-prone strain, MRL lpr/lpr. Proc. Natl. Acad. Sci. USA
90:10340.
10. Boise, L. H., and C. B. Thompson. 1996. Hierarchical control of lymphocyte
survival. Science 274:67.
11. Chinnaiyan, A. M., K. O’Rourke, M. Tewari, and V. M. Dixit. 1995. FADD, a
novel death domain-containing protein, interacts with the death domain of Fas
and initiates apoptosis. Cell 81:505.
12. Hsu, H., H. B. Shu, M. G. Pan, and D. V. Goeddel. 1996. TRADD-TRAF2 and
TRADD-FADD interactions define two distinct TNF receptor 1 signal transduction pathways. Cell 84:299.
13. Coupar, B., M. Andrew, and D. Boyle. 1988. A general method for the construction of recombinant vaccinia viruses expressing multiple foreign genes. Gene
68:1.
14. Mackett, M., G. Smith, and B. Moss. 1990. Vaccinia virus: a selectable eukaryotic cloning and expression vector. Proc. Natl. Acad. Sci. USA 79:7415.
15. Howley, P. M., D. Spehner, and R. Drillien. 1996. A vaccinia virus transfer
vector using a GUS reporter gene inserted into the 14L locus. Gene 172:233.
16. Tsung, K., J. H. Yim, W. Marti, R. M. Buller, and J. A. Norton. 1996. Gene
expression and cytopathic effect of vaccinia virus inactivated by psoralen and
long-wave UV light. J. Virol. 70:165.
17. Lu, Z., L. Yuan, X. Zhou, E. Sotomayor, H. I. Levitsky, and D. M. Pardoll. 2000.
CD40-independent pathways of T cell help for priming of CD8(⫹) cytotoxic T
lymphocytes. J. Exp. Med. 191:541.
18. Staveley-O’Carroll, K., E. Sotomayor, J. Montgomery, I. Borrello, L. Hwang,
S. Fein, D. Pardoll, and H. Levitsky. 1998. Induction of antigen-specific T cell
anergy: an early event in the course of tumor progression. Proc. Natl. Acad. Sci.
USA 95:1178.
19. Murphy, K. M., A. B. Heimberger, and D. Y. Loh. 1990. Induction by antigen of
intrathymic apoptosis of CD4⫹CD8⫹TCRlo thymocytes in vivo. Science 250:
1720.
20. Takahashi, T., M. Tanaka, J. Inazawa, T. Abe, T. Suda, and S. Nagata. 1994.
Human Fas ligand: gene structure, chromosomal location and species specificity.
Int. Immunol. 6:1567.
21. Goebel, S. J., G. P. Johnson, M. E. Perkus, S. W. Davis, J. P. Winslow, and
E. Paoletti. 1990. The complete DNA sequence of vaccinia virus. Virology 179:
247.
22. Alderson, M. R., R. J. Armitage, E. Maraskovsky, T. W. Tough, E. Roux,
K. Schooley, F. Ramsdell, and D. H. Lynch. 1993. Fas transduces activation
signals in normal human T lymphocytes. J. Exp. Med. 178:2231.
23. Hohlfeld, R., and K. V. Toyka. 1993. Therapies. In Myasthenia Gravis.
M. DeBaets and H. G. J. H. Oosterhuis, eds. CRC Press, Boca Raton, FL.
24. Drachman, D. B. 1998. Immunosuppressive agents. In Current Neurologic
Drugs. L. Rowland, ed. Current Medicine, Philadelphia.
25. Grakoui, A., S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen,
and M. L. Dustin. 1999. The immunological synapse: a molecular machine controlling T cell activation. Science 285:221.
26. Underhill, D. M., M. Bassetti, A. Rudensky, and A. Aderem. 1999. Dynamic
interactions of macrophages with T cells during antigen presentation. J. Exp.
Med. 190:1909.
27. Mellman, I., P. Pierre, and S. Amigorena. 1995. Lonely MHC molecules seeking
immunogenic peptides for meaningful relationships. Curr. Opin. Cell Biol. 7:564.
28. Rowell, J. F., A. L. Ruff, F. G. Guarnieri, K. Staveley-O’Carroll, X. Lin, J. Tang,
J. T. August, and R. F. Siliciano. 1995. Lysosome-associated membrane protein1-mediated targeting of the HIV-1 envelope protein to an endosomal/lysosomal
compartment enhances its presentation to MHC class II-restricted T cells. J. Immunol. 155:1818.
29. Wu, T. C., F. G. Guarnieri, K. F. Staveley-O’Carroll, R. P. Viscidi, H. I. Levitsky,
L. Hedrick, K. R. Cho, J. T. August, and D. M. Pardoll. 1995. Engineering an
intracellular pathway for major histocompatibility complex class II presentation
of antigens. Proc. Natl. Acad. Sci. USA 92:11671.
30. Lin, K. Y., F. G. Guarnieri, K. F. Staveley-O’Carroll, H. I. Levitsky,
J. T. August,D. M. Pardoll, and T. C. Wu. 1996. Treatment of established tumors
with a novel vaccine that enhances major histocompatibility class II presentation
of tumor antigen. Cancer Res. 56:21.
31. Ruff, A. L., F. G. Guarnieri, K. Staveley-O’Carroll, R. F. Siliciano, and
J. T. August. 1997. The enhanced immune response to the HIV gp160/LAMP
chimeric gene product targeted to the lysosome membrane protein trafficking
pathway. J. Biol. Chem. 272:8671.
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017
escape to infect other cells) but does direct expression of the gene
products by the infected cells.
Our results clearly demonstrate the validity of the principles
embodied in the “guided missile” strategy. We have shown that
APCs genetically engineered to process and present epitopes of a
model Ag and to express FasL can target and induce apoptosis of
Ag-specific T cells. Our results demonstrated the specificity of the
lethal effect of Ag-directed APCs. Thus, OVA-specific T cells
were not affected significantly by three-gene APCs that presented
HA-specific epitopes, even though the T cells were appropriately
stimulated and in close physical contact with the APCs. Intimacy
of contact between T cells expressing Fas and APCs expressing
FasL appears to be necessary for induction of apoptosis. APCs
expressing the two gene products, FasL and TrFADD, but not
presenting HA did not induce apoptosis of HA-specific T cells.
This is consistent with reports that FasL-expressing cells must bind
their Fas-expressing victims to induce apoptosis (39, 40). Furthermore, for T cells to undergo Fas-mediated apoptosis, they must be
activated and must express Fas strongly. Our findings showed that
naive HA-transgenic T cells required Ag stimulation in order to be
susceptible to apoptosis induced by the FasL-expressing APCs.
The time course experiment showed that the three-gene APCs
were able to kill naive HA-specific T cells in culture but that they
first stimulated the T cells before inducing apoptosis.
The lethal (or inhibitory) effect of the three-gene-transduced
APCs could not be attributed to Ag-induced cell death, but
rather to the effect of the FasL expressed by transduction of the
APCs. Parallel experiments in which the HA-specific cells were
confronted with APCs that were transduced with the stimulatory HA-LAMP-1 VVV and the three-gene VVV showed that
stimulation per se did not induce apoptosis of the targeted cells,
whereas Ag targeting and expression of FasL effectively induced apoptosis.
Certain features of these studies require comment. First, as a
convenient source of abundant and highly reproducible Ag-specific T cells, we used a well-defined model murine system, with
transgenic T cells that express the ␣␤TCR for HA. Although the
Ag-specific T cells used here were actually monoclonal, we
have also had equally striking success with a heterogeneous
Ag-specific T cell population (41). Second, for the purposes of
the present experiments, we have used the VV for gene transfer.
The advantages of vaccinia for this purpose have been stated,
but other methods of gene transfer may prove useful in the
future. The use of retroviral vectors that can express multiple
gene products is under investigation in our laboratories. Third,
as a convenient source of APCs, we have used a mixed population of spleen cells, including macrophages, dendritic cells,
and B cells. Now that new methods of obtaining relatively large
populations of dendritic cells are becoming available, we are
evaluating the feasibility of using dendritic cells as APCs in
these experiments. Ultimately, it should be possible to develop
the guided missile strategy described here for the treatment of
autoimmune diseases.
SPECIFIC IMMUNOTHERAPY BY GENE TRANSFER TO APC
The Journal of Immunology
32. Nair, S. K., D. Boczkowski, M. Morse, R. I. Cumming, H. K. Lyerly, and
E. Gilboa. 1998. Induction of primary carcinoembryonic antigen (CEA)-specific
cytotoxic T lymphocytes in vitro using human dendritic cells transfected with
RNA. Nat. Biotechnol. 16:364.
33. Nagata, S., and P. Golstein. 1995. The Fas death factor. Science 267:1449.
34. Suda, T., T. Takahashi, P. Golstein, and S. Nagata. 1993. Molecular cloning and
expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell
75:1169.
35. Suda, T., T. Okazaki, Y. Naito, T. Yokota, N. Arai, S. Ozaki, K. Nakao, and
S. Nagata. 1995. Expression of the Fas ligand in cells of T cell lineage. J. Immunol. 154:3806.
36. Zhang, H. G., D. Liu, Y. Heike, P. Yang, Z. Wang, X. Wang, D. T. Curiel, T. Zhou,
and J. D. Mountz. 1998. Induction of specific T-cell tolerance by adenovirus-transfected, Fas ligand-producing antigen presenting cells. Nat. Biotechnol. 16:1045.
4779
37. Zhang, H. G., X. Su, D. Liu, W. Liu, P. Yang, Z. Wang, C. K. Edwards,
H. Bluethmann, J. D. Mountz, and T. Zhou. 1999. Induction of specific T cell tolerance by Fas ligand-expressing antigen-presenting cells. J. Immunol. 162:1423.
38. Oertli, D., W. R. Marti, J. A. Norton, and K. Tsung. 1996. Non-replicating recombinant vaccinia virus encoding murine B-7 molecules elicits effective costimulation of naive CD4⫹ splenocytes in vitro. J. Gen. Virol. 77:3121.
39. Wong, B., J. Arron, and Y. Choi. 1997. T cell receptor signals enhance susceptibility to Fas-mediated apoptosis. J. Exp. Med. 186:1939.
40. Lenardo, M., K. M. Chan, F. Hornung, H. McFarland, R. Siegel, J. Wang, and
L. Zheng. 1999. Mature T lymphocyte apoptosis—immune regulation in a dynamic
and unpredictable antigenic environment. Annu. Rev. Immunol. 17:221. J. M.
41. Wu, J.-M., B. Wu, A. Miagkov, R. N. Adams, and D. B. Drachman. Specific
immunotherapy of experimental myasthenia gravis in vitro: the “guided missile”
strategy. Cell Immunol. In press.
Downloaded from http://www.jimmunol.org/ by guest on July 28, 2017