| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Holland Laboratory, Department of Immunology, American Red Cross, Rockville, MD 20855; and
Laboratory of Immunology, National Eye Institute, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Based on the tolerogenic properties of Ig carriers combined with the efficacy of B cell Ag presentation for unresponsiveness, we previously demonstrated the induction of tolerance using LPS-activated B cells retrovirally transduced with an IgG-Ag fusion protein (3, 4). These transduced B cells can efficiently express multiple antigenic epitopes presented in a tolerogenic manner requiring class II MHC on the B cells (4, 5). This protocol led to epitope-specific protection not only in naive, but also in already primed, recipients (3, 5, 6). Furthermore, the clinical potential of this form of gene therapy was first demonstrated in an experimental model of uveitis, a cell-mediated autoimmune disease (6). In this report, we extend our novel gene transfer approach to the treatment of two additional, widely used Th1-mediated autoimmune disease models: EAE, an inducible model of MS, and a spontaneous diabetes in nonobese diabetic (NOD) mice. We also examine whether this approach would cause immune deviation toward an allergic phenotype in the recipient mice, and whether it could be effective after disease symptoms appeared.
Several retroviral constructs that display myelin basic protein (MBP), glutamic acid decarboxylase (GAD) or insulin (B9–23) at the N terminus of murine IgG1 H chain were generated and were used to transduce LPS-stimulated B cell blasts. These B cells were then injected into primed recipients. The results demonstrated that our gene therapy approach to induce expression of Ag-IgG fusion proteins not only protected mice from ongoing EAE and diabetes, but also prevented allergic reactions in sensitized mice. These findings expand the clinical potential of our approach to EAE and diabetes and reiterate that the risk of horror autotoxicus associated with Th2 immune deviation (2) might not occur after such therapy, providing further evidence of the power of this tolerogenic treatment.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Retroviral constructs inserted with MBP-IgG, GAD-IgG, and
B9-23-IgG and control retroviral constructs, p1-102-IgG and OVA-IgG.
Molecular cloning of these retroviral vectors was similar to those
described previously (3, 4, 6). Briefly, for MBP-IgG, a
mouse MBP cDNA encoding exons 1, 2, 3, 4, 5, and 7 was amplified by PCR
from pHF45 plasmid, a gift of Dr. L. Hudson (National Institutes of
Health, Bethesda, MD). The PCR product was ligated into pGEM-T
vector and the fidelity of DNA polymerase was confirmed by DNA
sequencing. Subsequently, mouse MBP cDNA was subcloned into the
BSSK-IgG plasmid, a murine IgG1 H chain cassette plasmid
containing NotI and XhoI cloning sites at the N
terminus of the IgG1 H chain (a gift of Dr. R. Tisch, University of
North Carolina, Chapel Hill, NC). The BSSK-IgG-OVA and
BSSK-IgG-GAD constructs were similarly engineered. dsDNA encoding
insulin B9-23 was generated by annealing single-strand complementary
synthetic nucleotides, and then inserted at
NotI/XhoI sites in BSSK-IgG plasmid.
BSSK-IgG-MBP, BSSK-IgG-OVA, BSSK-IgG-B9-23, and BSSK-IgG-GAD were later
digested with SalI to extract the IgG-Ag cDNA. These IgG
cDNA fragments were then cloned into a SalI-digested MBAE
retroviral vector (3). Successful clones with correct
orientation were used to transfect packaging cell lines. The expression
of IgG-Ag fusion protein was driven by the 
-actin promoter/enhancer,
and a neomycin selection marker was included in the MBAE vector and
driven by viral long-terminal repeat.
The GPE86 packaging cell line was transfected with MBAE IgG-constructs
described above, and then selected under neomycin (G418, 0.6 mg/ml in
active form), as described previously (3, 4). High titer
clones (
105–106
neomycin-resistant NIH 3T3 CFU/ml) free from replication competent
virus were stored in liquid nitrogen and freshly thawed for each
experiment. The virus producer cell lines used for the experiments
described in this paper were named GPE.MBP/9, GPE.GAD, GPE.OVA,
GPE.p1-102, and GPE.InsulinB9-23.
Gene transfer protocols. The transfection of splenic lymphocytes for gene therapy was performed as previously described (3, 4, 5). Briefly, splenic B cells were stimulated with 20–50 µg/ml bacterial LPS (Escherichia coli 055:B5; Sigma-Aldrich, St. Louis, MO) overnight, and recultured (4 x 106/ml, 5 ml cultures) with viral packaging cells (GPE.MBP/9; GPE.GAD, GPE.OVA, or GPE.InsulinB9-23) in the presence of 6 µg/ml polybrene and LPS for an additional 24 h. The virally infected cells were washed and injected i.p. into syngeneic mice at designated intervals after the transfer of primed cells, except as noted.
RT-PCR and methods to ascertain the efficiency of gene therapy.
In preliminary experiments, detection of MBP, GAD, or OVA transcripts
in transduced cell lines or splenic tissues harvested from the mice
receiving the transduced LPS B cell blasts was confirmed for each
respective sequence by RT-PCR. Primers were designed to amplify MBP-,
GAD- or OVA-specific sequences. One microgram of total RNA from splenic
tissue was reverse transcribed (two rounds) with avian myeloblastosis
virus reverse transcriptase, dNTPs, and oligo(dT) and random
hexamer primers (Invitrogen cDNA cycle kit; San Diego, CA). The
resultant cDNA was amplified with specific 5' and 3' primers and
Taq DNA polymerase (Roche Molecular Biochemicals, Palo Alto,
CA). PCR conditions were 1 min at 94°C, 2 min at 55°C, and 3 min at
72°C for 30 cycles. Amplified DNA products were loaded onto 1%
agarose gels. -Actin RT-PCR using commercially available primers
(Stratagene, La Jolla, CA) was similarly performed to confirm the
integrity of RNA samples and served as a loading control. Note that we
have previously demonstrated that the construct and presumably the
transduced B cells persisted for a least 4–8 mo based on the PCR
signal and tolerance experiments (Refs. 4 and
6 and R. K. Agarwal and R. Caspi, unpublished
observations). The efficiency of transfection was estimated to be
3–5% in our original experiments (3). The expression of
Ag-IgG in the serum of mice receiving gene therapy was further verified
using a modified 5-iodo-4-hydroxy-3-nitrophenylacetyl (NIP)-binding
ELISA to detect the ability of the IgG to bind to NIP hapten, as
described previously (3, 4). However, the amount of
NIP-binding activity did not correlate with the degree of tolerance
induction or clinical efficacy, as we reported earlier (3, 4, 6). In other studies, CFSE-labeled B cell blasts actually
proliferated and were detectable for at least 30 days (M. Litzinger and
D. W. Scott, manuscript in preparation).
Immunological challenge and methods
Antigens.
Bovine myelin basic protein (bMBP) was purchased from
Sigma-Aldrich. Peptides MBP-Ac1-17 (MASQKRPSQRSKYLATA),
MBP60-80 (HTRTTHYGSLPQKSQHGRTQ), and myelin oligodendrocyte
glycoprotein 35-55 (MEVGWYRSPFSRVVHLYRNGK) were synthesized at
the Molecular Biology Core of the Holland Laboratory (Rockville, MD).
Proteolipid protein (PLP) 139-151 (HCLGKWLGHPDKF) was synthesized by
Sigma-Genosys (The Woodlands, TX). In additional experiments, the
bacteriophage 
cI immunodominant peptide, p12–26, was used for
immunization as described earlier (3, 4).
Mice. Female PLxSJL F1 (PLSJL), B10.PL, C57BL/6, and NOD mice were purchased from The Jackson Laboratory (Bar Harbor, ME) at 5–6 wk of age; gld (Fas ligand (FasL)-) mice on a BALB/c background and control BALB/c mice were generously provided by Dr. W. Davidson, Holland Laboratory. All animals were housed in pathogen-free, microisolator cages at the animal facilities of the Holland Laboratory. All animal procedures were approved by the Animal Care and Use Review Committee at the Holland Laboratory of the American Red Cross.
Treatment for tolerance induction. In the EAE experiments, tolerance was induced with a single i.p. injection of 1–2 x 107 gene-transferred LPS-stimulated spleen or purified B cells diluted in 0.5 ml of PBS. Except as noted, injections of tolerogenic B cells were done within 6 h after the i.v. injection of MBP-reactive lymph node (LN) T cells, which had been restimulated in vitro with MBP, to transfer disease. Recipients were analyzed for disease, expression of recombinant retrovirus, as well as immune tolerance to MBP for 4–6 wk after EAE induction. In NOD mice, tolerogenic treatment was performed at three different time points in separate experiments (starting at 7, 10, and 14 wk of age), using 1 x 107 LPS-stimulated spleen cells/mouse. To confirm these results, in one replicate experiment, groups of NOD mice were started on the tolerogenic gene therapy protocol at 7 or 14 wk of age. Groups of 5–10 mice were used in each replicate experiment.
Induction of EAE. To efficiently induce EAE with increased disease incidence, we adopted a combination of passive and active EAE induction protocols, because the incidence of EAE in our colony was low under standard immunization (M. Melo, unpublished observations). Briefly, 8- to 12-wk-old donor mice were primed with 100 µg of bMBP and MBPpAc1-17 emulsified in an equal volume of CFA (Bacto-Difco, West Molesey, Surrey, U.K.) at the base of the tail. After 15–30 days, LN T cells were harvested and cultured with bMBP (or MBP Ac1-17) and IL-2. In most experiments, spleen cells from the same primed animals were removed and used as sources of tolerogenic B cells, as described above, to accurately mimic a clinical situation. After 4–5 days, recipient mice were boosted with 200 µg of bMBP plus 100 µg of pAc1-17 emulsified in CFA. Pertussis toxin (List Biological Laboratories, Campbell, CA; 200 ng) was given i.p. in 0.3 ml at the time of immunization and again one day later. Groups of 5–10 mice were monitored for disease daily and EAE scored on a standard basis as follows: I = flaccid tail; II = partial paralysis of one or two limbs or flaccid tail with ataxia; III = total paralysis of two hind limbs; IV = quadriparalysis; moribund, V = death. Paralyzed mice were afforded easier access to food and water. In the experiments testing specificity and bystander suppression, disease was induced with a combination of bMBP emulsified with PLP-p139–151 or with the PLP peptide alone.
Lymphocyte proliferation and lymphokine measurements.
To assay T cell responsiveness, animals receiving passive EAE transfer
and subsequent active immunization were euthanized at 12 days after
s.c. immunization with bMBP/CFA, and the LN and spleen were removed.
Cells were cultured in X-Vivo medium (Life Technologies, Grand Island,
NY) supplemented with 2 x 10-5 2-ME
(Sigma-Aldrich) and stimulated with 3, 10, or 30 µM concentrations of
MBP or peptides. Proliferation was assayed by the addition of 1
µCi/well of [3H]thymidine (ICN
Pharmaceuticals, Irvine, CA) for the last 14–18 h of a 3–4 day
culture. The results were expressed as the stimulation index (cpm with
Ag/cpm with medium alone). For cytokine measurements, LN or
spleen cells were cultured with MBP or MBP peptides for 24–36 h in a
24-well plate. Cell supernatants were collected and assayed for
IFN-
, IL-2 and IL-4 using standard capture ELISA as described
elsewhere (4, 6).
Evaluation of disease in NOD mice
For treatment of NOD mice, B cell blasts from age- and sex-matched syngeneic donors were transduced to express IgG-GAD or IgG-Insulin B9-23, and then injected into 7-, 10-, or 14-wk-old female NOD recipients. Serum glucose levels were measured by standard methods in blood taken from the retro-orbital plexus. Mice were also evaluated daily for clinical signs of diabetes (e.g., weight loss) and mortality rate.
Statistical analysis
The Student t test (for paralysis scores and glucose levels) was used to evaluate significance for each time point. The statistical differences between cumulative incidence and incidence density of the groups was also calculated. Values of p < 0.05 were considered significant, whereas p < 0.07 was considered clinically significant. (see Ref. 7 for discussion). Details for a given set of experiments are presented later in the legends and text.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Induction of specific tolerance to ameliorate autoimmune disease is desirable, especially if it can be achieved after the appearance of overt symptoms. One approach to induce tolerance is the delivery and presentation of the self Ag on an Ig carrier (3, 8, 9). We previously demonstrated that gene therapy via retroviral expression of IgG-fusion protein leads to tolerogenic B cell presentation of the endogenously produced autoantigen (4). To extend our gene therapy approach in animal models of autoimmune diseases, we developed a retroviral construct-encoding MBP autoantigen at the N terminus of murine IgG1 H chain for the treatment of EAE, an inducible animal model of MS. LPS-stimulated B cell blasts transduced with MBP-IgG construct were injected into susceptible mice to determine the efficacy of this approach on the course of ongoing EAE.
In preliminary experiments, i.v. injection of 8 x
106 MBP-IgG transduced B cells 7 days before
disease induction was sufficient to protect PLxSJL mice from EAE (data
not shown). In another Ag model system, p12–26 of repressor
protein, we were able to induce tolerance with as low as 2 x
106 peptide-IgG expressing B cells (M. El-Amine
and D. W. Scott, unpublished data.) Because our goal is to arrest
or reverse ongoing disease in a simulated clinical setting, B cell
blasts from syngeneic bMBP-primed animals (exhibiting disease symptoms)
were used to deliver the gene therapy. Injection of LPS B cell blasts
expressing the murine MBP-fusion protein significantly protected
PLxSJL (four experiments) and C57BL/6 mice (one experiment) from EAE
even when the tolerogenic treatment was initiated after mice had
received cells from MBP/CFA-immunized mice. Fig. 1
is a representative experiment in
PLxSJL mice that received LN T cells from donors primed 15 days
earlier with MBP in CFA. As expected, the transfer of the MBP-primed LN
T cells induced EAE and 100% mortality in all control mice receiving
LPS blasts transduced with OVA-IgG construct. However, in contrast,
both the average scores (Fig. 1), the prevalence (40%), and mortality
rate (0%) were drastically diminished in the experimental group
receiving MBP-IgG-transduced LPS blasts.
|
. In this experiment, we used LN and
spleen cells from PLxSJL F1 mice with ongoing
EAE to transfer disease to syngeneic recipients (5 x
106 cells/mouse). Recipient mice were further
boosted with MBP/CFA plus pertussis toxin 4 days later. Ten days after
active immunization (which corresponds to day 3 after disease onset in
Fig. 2), each mouse received 107 primed B cell
blasts transduced with either a control construct (OVA-IgG) or the
experimental construct (MBP-IgG). At the time of the tolerogenic
treatment, 60% of the animals in each group were already showing
signs of EAE (Fig. 2B). The donors of these tolerogenic and
control B cell blasts came from mice primed in a similar way as the
recipient animals. That is, we were treating ongoing disease with
splenic B cells from animals that were already symptomatic. The
experiment was carefully designed so that each subgroup of mice had the
same number of sick animals and the same average score. Disease was
monitored daily. Fig. 2A shows the average daily scores of a
group of animals in which EAE was induced with MBP. Note that the
difference between the average score of the control subgroup
(GPE.OVA-treated) vs the experimental subgroup (GPE.MBP/9-treated) was
not significant up to day 8. At day 9, the p value was 0.001
and it remained significantly different (p <
0.05) at all subsequent time points, except for day 32. It is also
remarkable that at the end of the experiment, there was no animal
showing signs of EAE in the experimental group (Fig. 2B).
Note that the disease prevalence in control, mock-construct-treated
mice is reduced slightly due to spontaneous remission in a subset
of mice.
|
). There also was no effect on IFN-
production by spleen or LN cells and no IL-4 was detectable in any
groups (data not shown; M. Melo, R. Agarwal, and D. Scott,
manuscript in preparation).
|
To investigate possible mechanisms of our gene therapy approach,
several approaches tested the role of suppression, as well as FasL, in
tolerance via this gene therapy protocol. First, we coimmunized one
group of animals with bMBP/PLPp139 and another group with PLPp139 alone
to test for specificity of tolerance and bystander suppression. Primed
cells were transferred to syngeneic recipients, which then received
MBP-IgG-transduced B cells. The results in Fig. 4 demonstrate that the gene therapy with
MBP-IgG is specific in that no significant effect on PLP-induced
disease is observed with PLP peptide induction (Fig. 4A), or
with a mixture of PLP 139–151 plus MBP (Fig. 4B).
Importantly, these results also suggest that bystander suppression is
not involved in tolerance in this model, although this may require
further study.
|
p12-26 immunodominant
peptide and a p12-26-IgG vector to test for tolerance in BALB/c
congenic mice. B cells from gld or +/+ mice were stimulated
with LPS and then transduced with a p12-26-IgG construct used in our
earlier experiments. We verified that FasL was up-regulated in normal B
cells but not in gld B cells (M. El-Amine, unpublished
observations). The results in Fig. 5
indicate that tolerance to p12-26 was induced with transduced wild-type
B, but not with gld B, cells. These data suggest that
up-regulation of FasL on target B cells by LPS (10) and
Fas-FasL interactions are required for effective tolerance in this
model. However, these data do not imply a deletion mechanism for
tolerance because studies of the fate of TCR transgenic mice do
not show a loss of specific T cells, and lck-driven
Bcl-2 transgenic mice are not resistant to tolerance
induction (M. Litzinger and D. W. Scott, manuscript in
preparation; R. K. Agarwal and R. Caspi, unpublished observations)
as would be expected if deletion via apoptosis were involved.
|
Previously, we demonstrated that our gene therapy approach was
effective in an experimental model of uveitis, experimental autoimmune
uveoretinitis (6). Because both EAE and
experimental autoimmune uveoretinitis involve deliberate immunization
with pathogenic Ags or peptides, we extended our model to a spontaneous
autoimmune disease, and therefore tested the efficacy of gene therapy
in the NOD model of diabetes. NOD mice develop an immune response to
GAD and peri-insulitis by 3–7 wk of age, although islet infiltration
is usually not observed until 10–12 wk of age. Approximately 80% of
female NOD mice will then develop overt diabetes by 4–5 mo of age.
Therefore, we initiated the treatment in these NOD mice at three
different time points: 7, 10, and 14 wk of age. We used two tolerogenic
constructs coding for IgG-fusion proteins that contain either
full-length GAD or an immunodominant insulin peptide (Insulin-B9–23).
As shown in Fig. 6, gene therapy using
both constructs protected mice from disease, especially when treatment
was initiated at 7 wk. For example, when glucose levels were measured
at week 18 just before fatalities occurred, the difference between the
controls and the GAD-IgG treated mice was highly significant
(p < 0.02). The difference in disease incidence
between GAD-IgG and control irrelevant construct ( p1–102-IgG) was
clinically significant (p < 0.07; cf 7).
When treatment was initiated at 10 wk, there was still protection both
in terms of diabetes incidence and prolonged survival (data not shown).
However, if treatment is delayed until 14 wk of life, there was no
protection, presumably due to the extent of permanent islet damage that
has occurred before the initiation of treatment.
|
Oral, i.v., and intranasal administration of soluble autoantigens
or their peptides have been reported to ameliorate autoimmune diseases
through a mechanism involving immune deviation toward Th2. However,
this kind of "split tolerance" has recently been reported to
increase the risk of an allergic response to the tolerogen
(2). To determine whether the mechanism of our gene
therapy for tolerance involved immune deviation that would carry an
increased risk for anaphylaxis, we examined the effect of our approach
on anaphylaxis to MBP in B10.PL mice to test efficacy in a related
EAE-susceptible strain. After passive transfer of the MBP-reactive
cells, mice were boosted with MBP/CFA and after 20 days, the animals
were injected with 100 µg of bMBP i.v. As shown in Fig. 7, the anaphylactic shock response was
observed in all control animals (OVA-IgG) but overall scores and
incidence of fatal anaphylaxis were reduced in the experimental group
(MBP-IgG); pathological examination confirmed the diagnosis of shock in
the lungs of mice dying after the i.v. challenge (data not shown). Ten
minutes after i.v. challenge, all animals in the control group had
anaphylaxis, but three-eighths of the animals in the MBP-IgG-treated
group did not have any signs of shock. Furthermore, after 1 h,
fatal outcome occurred in five-eighths of the animals in the control
group but only in two-eighths of the experimental mice. Thus, these
results suggested that our gene transfer protocol did not cause immune
deviation toward an allergic shock, but actually ameliorated
anaphylaxis to MBP in contrast to peptide therapy.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because one of the most tolerogenic APCs is the B cell, and one of the most tolerogenic carriers is IgG (11, 12, 13, 14), our laboratory has developed protocols to gene transfer IgG-fusion proteins into B lymphocytes that are then used to express self IgG-fusion protein in vivo. Previously, we found that in primed mice, the expression of a model Ag in B cells as part of an IgG scaffold down-regulated both the Th1 and Th2 arms of the immunologic response (3, 6). We also found that MHC expression by the Ag-presenting B cells is required for tolerance (5), and that the IgG scaffold is important for down-regulation of the secondary immune response and maintenance of tolerance (4). Recently, the clinical potential of this therapeutic modality was shown in a mouse model of uveitis (6), a Th1-mediated disease that targets the retina. In this study, we extend our approach to two other autoimmune diseases, EAE and diabetes in mice (both Th1-mediated diseases) and to one murine model of anaphylactic shock (a Th2 disease). Thus, we demonstrate the general applicability of our approach and extend it to additional mouse strains.
A myriad of tolerogenic protocols designed to turn off pathological
immune responses have been reported in the last decade (8, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29). These involve i.v. (18), intrathymic
(19), i.p. (20), or mucosal (15, 16, 17, 21) administration of proteins and peptides, altered peptide
ligands (22), and also i.v. injection of fixed APCs
coupled with protein Ags (28, 29). Our goal was not to
compare these tolerogenic therapies with our ex vivo gene therapy
approach. Rather, we simply wished to establish its efficacy and
safety, and begin to explore its mechanisms. Miller et al.
(28) have previously proposed that specific peripheral
immune tolerance induced by the i.v. administration of
chemically-fixed splenocytes could be due to the induction of
Ag-specific regulatory T cells. However, it has also been proposed that
i.v. injection of MBP-coupled splenocytes induces tolerance by either
anergy or deletion (29). In our model, we have obtained no
data suggesting the involvement of regulatory cells (6) or
of bystander suppression or a Th1
Th2 shift (3, 6).
Thus, improvement in EAE symptoms correlated with reduced LN
proliferation, a small decrease in IL-2 production, and no change in
IFN-, IL-4, or IL-10 production (M. Melo, manuscript in
preparation). This dissociation of the effect on disease and
immunologic responses in vitro is not totally unexpected because modest
effects on in vitro parameters of responsiveness were previously
reported (6).
Interestingly, we found that B cells from gld mice were not tolerogenic in this protocol, suggesting that expression of FasL (as well as MHC class II) was required on the tolerogenic B cell APC. However, further data suggest that FasL-mediated deletion of target T cells is not involved because tolerance is equally effective in Bcl-2 transgenics (R. K. Agarwal, unpublished observations) and no loss of TCR transgenic T cells occurs (Litzinger and D. W. Scott, manuscript in preparation) in this gene therapy approach to tolerance.
Because of feasibility, injection, feeding, or nasal administration of autoantigens are very convenient ways to induce systemic tolerance. However, the therapeutic potential of each one of these approaches has not yet been proven in clinical practice (23, 24). Furthermore, in animal models, tolerance induction for ongoing immune responses or immune deviation is not only difficult to achieve in all strains and ages, but also can have deleterious consequences (2, 25). Altered peptide ligands are elegant and have potential, but these require knowledge of the precise autoantigenic epitope for a given MHC haplotype, and therefore cannot be used in every patient. Alternative protocols are clearly needed to stop ongoing pathological autoreactivity by sustaining tolerance in the largest number of patients without causing side effects.
Because our major goal is to treat ongoing disease, we used as B cell
donors mice that had been primed earlier and that had already shown
signs of EAE. Indeed, in most experiments, we transduced primed B cells
(as sources of tolerogenic APC) from the same MBP-primed animal that
served as a donor for LN T cells to transfer EAE. This was to simulate
the clinical situation in which autologous B cells from the patient to
be treated would be the ideal delivering cells for the gene therapy. In
this study, we showed that even using a potent encephalitogenic
induction protocol, disease arrest was still possible. The experiment
in Fig. 1
shows that PLxSJL mice can be protected after induction with
cells from donors actively immunized 15 days earlier. In other
experiments, we could protect mice from EAE transferred with T cells
from mice primed for 30 days (30), during which time
epitope spreading could have occurred. The experiment shown in Fig. 2
shows that we can actually arrest ongoing disease. After inducing
disease with cells from ongoing EAE and waiting until 60% of the
recipients show signs of disease, we can still significantly reduce
disease scores (Fig. 2A) and prevalence (Fig. 2B). This provides evidence that protection can be achieved
after the pathogenic T cells have migrated to the local sites of the
lesions in the CNS. Although protection was achieved in this case, it
is not possible to conclude that we are affecting determinant spreading
(26). Further studies are needed to evaluate the response
to minor Ags. The ability to track both the responding T cells with
MBP-specific T cells, as well as to define the precise location of the
tolerogenic B cells using CFSE labeling, will aid in understanding the
mechanisms involved in this tolerogenic therapy.
In mice, EAE is more easily induced in strains such as SJL, PL, and B10.PL; however, it is now recognized that most, if not all, "resistant" strains have the machinery for mounting a pathogenic immune response to myelin Ags. Induction in EAE-resistant mice requires in vitro stimulation of myelin-reactive T cells to break their in vivo unresponsiveness to MBP/CFA immunization. Such in vitro manipulations provide another opportunity to test the efficiency of new therapeutic modalities to stop the effector phase of EAE in strains with different immunodominant epitopes. Using the IgG-MBP construct, we tested the gene therapy on EAE induced in B6 mice with MBPp60-80. Our data confirm the efficacy of this approach also in this strain (data not shown).
In NOD mice, B cell blasts transduced to express IgG-GAD or IgG-Insulin B9-23 were able to delay diabetes onset even when administrated into 7- to 10-wk-old mice, a time when there is already evidence of peri-insulitis. Furthermore, in these studies, tolerogenic B cells came from similarly aged donors. This indicates that primed NOD B cells can be made to be tolerogenic. After 14 wk, no protective effect was observed. This could be explained by the speed in which the insulitis occurs in female NOD mice. In humans, the prediabetic stage can last for years, and it is postulated that during this period, insulitis is already taking place. Probably because of the small amount of tissue composing the mouse pancreas, the prediabetic stage lasts for only a few weeks in NOD mice. Because insulitis can be detected as early as 3–4 wk of age, by week 14, the inflammatory process could have already consumed all the islets in these mice. Hopefully, we may be able to achieve therapeutic effects with this protocol at later stages of disease when combined with islet transplantation.
The GAD-IgG retroviral construct we used was more effective than the insulin peptide-IgG construct. We suggest that this may reflect the cascade of responses to different epitopes that occurs in NOD animals. Thus, it has been postulated that the autoimmune cascade of determinant spreading in NOD mice starts with GAD (26, 27). This would explain the efficacy of this construct, because we are blocking the spreading of the response upstream in the cascade, although we cannot eliminate expression levels as an explanation.
In summary, our findings confirm the results in previous publications showing that this gene therapy approach induces tolerance at both the Th1 and Th2 levels. Because there were no signs of immune deviation, we conclude that our gene therapy does not increase (and actually can decrease) the risk of anaphylaxis in these disease models. The fact that the in vivo protection from disease was consistently demonstrated in different disease models and MHC haplotypes suggests potential general usefulness of this gene therapy approach in the diverse genetic background of the human population. This, together with our earlier data in the uveitis model (6), provides proof of principle for using this approach in a clinical setting.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 J.Q. and M.E.-A. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. David W. Scott, Department of Immunology, Holland Laboratory of the American Red Cross, 15601 Crabbs Branch Way, Rockville, MD 20855. E-mail address: scottd{at}usa.redcross.org
4 Abbreviations used in this paper: MS, multiple sclerosis; EAE, experimental allergic encephalomyelitis; NOD, nonobese diabetic; MBP, myelin basic protein; GAD, glutamic acid decarboxylase; NIP, 5-iodo-4-hydroxy-3-nitrophenylacetyl; bMBP, bovine MBP; FasL, Fas ligand; LN, lymph node; PLP, proteolipid protein.
Received for publication November 11, 2001. Accepted for publication February 21, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Frommer, T. J. A. J. Heinen, F. T. Wunderlich, N. Yogev, T. Buch, A. Roers, E. Bettelli, W. Muller, S. M. Anderton, and A. Waisman Tolerance without Clonal Expansion: Self-Antigen-Expressing B Cells Program Self-Reactive T Cells for Future Deletion J. Immunol., October 15, 2008; 181(8): 5748 - 5759. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Dalai, S. Mirshahidi, A. Morrot, F. Zavala, and S. Sadegh-Nasseri Anergy in Memory CD4+ T Cells Is Induced by B Cells J. Immunol., September 1, 2008; 181(5): 3221 - 3231. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Su, G. Carey, M. Maric, and D. W. Scott B Cells Induce Tolerance by Presenting Endogenous Peptide-IgG on MHC Class II Molecules via an IFN-{gamma}-Inducible Lysosomal Thiol Reductase-Dependent Pathway J. Immunol., July 15, 2008; 181(2): 1153 - 1160. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Werner-Klein, C. Dresch, P. Marconi, and T. Brocker Transcriptional Targeting of B Cells for Induction of Peripheral CD8 T Cell Tolerance J. Immunol., June 15, 2007; 178(12): 7738 - 7746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. T. Litzinger, Y. Su, T. C. Lei, N. Soukhareva, and D. W. Scott Mechanisms of Gene Therapy for Tolerance: B7 Signaling Is Required for Peptide-IgG Gene-Transferred Tolerance Induction J. Immunol., July 15, 2005; 175(2): 780 - 787. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-M. R. Saint-Remy and M. G. Jacquemin Tolerance induction: a job for B cells? Blood, June 15, 2005; 105(12): 4551 - 4551. [Full Text] [PDF]
|
|
|
|
|
|
T. C. Lei and D. W. Scott Induction of tolerance to factor VIII inhibitors by gene therapy with immunodominant A2 and C2 domains presented by B cells as Ig fusion proteins Blood, June 15, 2005; 105(12): 4865 - 4870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Takacs, C. Du Roure, S. Nabarro, N. Dillon, J. H. McVey, Z. Webster, A. MacNeil, I. Bartok, C. Higgins, D. Gray, et al. The regulated long-term delivery of therapeutic proteins by using antigen-specific B lymphocytes PNAS, November 16, 2004; 101(46): 16298 - 16303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-C. Chen, A. Rivera, J. P. Dougherty, and Y. Ron Complete protection from relapsing experimental autoimmune encephalomyelitis induced by syngeneic B cells expressing the autoantigen Blood, June 15, 2004; 103(12): 4616 - 4618. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V C Kyttaris, Y-T Juang, and G C Tsokos Gene therapy in systemic lupus erythematosus Lupus, May 1, 2004; 13(5): 353 - 358. [Abstract] [PDF]
|
|
|
|
 |
|
 M. El-Amine, J. A. Hinshaw, and D. W. Scott In vivo induction of tolerance by an Ig peptide is not affected by the deletion of FcR or a mutated IgG Fc fragment Int. Immunol., July 1, 2002; 14(7): 761 - 766. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
2 for
overall significance, the cumulative incidence, the incidence density,
and the difference in the severity scores, by day, were compared.
