U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Madame Curie Bioscience Database [Internet]. Austin (TX): Landes Bioscience; 2000-2013.

Cover of Madame Curie Bioscience Database

Madame Curie Bioscience Database [Internet].

Show details

Targeting Antigen-Specific T Cells for Gene Therapy of Autoimmune Disease

and .

One of the most exciting advances in the field of gene therapy in recent years is the establishment of the antigen-specific T cell as a vector for the delivery of genetically-derived treatment in vivo. In contrast with traditional applications of gene therapy, the unique versatility, specificity and memory of the T cell affords the researcher or clinician the ability to apply a broad range of tactics in the genetic treatment of disease. The T cell may be modified to deliver therapeutic products or regenerative products to sites of inflammation and tissue destruction. In addition, the T cell may be altered to modulate cellular interactions or to correct its own genetic defects to ameliorate disease. These genetic modification strategies as they relate to the treatment of autoimmune disease in experimental animal models will be the focus of this chapter, with particular emphasis on the analogs of multiple sclerosis (MS), insulin-dependent diabetes mellitus (IDDM) and rheumatoid arthritis (RA).


Gene therapy, in its simplest form, can be represented by the replacement of a single missing or defective gene to correct a monogenic disorder. Human diseases of this nature, such as adenosine deaminase (ADA) deficiency or cystic fibrosis (CF), have been obvious and attractive targets for experimental gene therapy since the first human clinical trials began on ADA patients in 1990. In a broader sense, gene therapy can be considered not only the replacement of a defective endogenous gene, but can also incorporate the addition of foreign or modified genes to alter biological function. Thus, diseases which have multigenic, complex or unknown underlying pathologies, as most autoimmune diseases do, can also be candidates for gene therapy by exploiting endogenous biological control pathways or by creating new ones. These strategies can be categorized into four general groups: modification of target tissue; delivery of therapeutic product(s); delivery of regenerative product(s); and alteration of cellular interactions.

Although the field has met with some success, ADA remains the only human disease to date which has been effectively “cured” by gene therapy. Much of the initial enthusiasm surrounding early gene therapy experiments has led to disappointment, forcing a reevaluation of existing treatment strategies as they relate to established clinical goals. These goals of an ideal gene therapy design can be stated quite simply: delivery of the therapy should be appropriately targeted; the expression of the product should be long-term; and most importantly, it should be well-regulated. Any novel gene therapy scheme should be evaluated based on these three principles.

One of the most exciting and perhaps most promising strategies which has emerged in recent years is the use of antigen-specific T cells as delivery vectors for gene therapy. The T cell is an ideal vector for many reasons. First, it can accomplish all three goals of an ideal gene therapy design. The major hallmarks of immunity are specificity and memory. By nature, a T cell is uniquely suited to home to a specific antigen nearly anywhere in the body, providing exquisite targeting ability. T cells are also long-lived, thus providing potential for long-term expression of transgene(s). By incorporating an appropriate inducible promoter to drive expression of the transgene, tight regulation can be achieved. Moreover, T cells have been successfully employed in all four general gene therapy strategies either directly or indirectly. It is only in a small number of cases in which the T cell itself contains the genetic defect, as in ADA, does the T cell become the target tissue modified. The latter three categories, encompassing the delivery of therapeutic products, regenerative products and the alteration of cellular interactions provide countless possibilities for exploiting the potential of antigen-specific T cells in the treatment of autoimmune disease. These experimental applications will be the focus of this chapter.

Genetic Modification of T Cells

One misconception about the T cell is that it is extremely resistant to genetic modification. While it is true that T cells prove relatively more difficult than other types of cells, efficient DNA uptake can be achieved through a variety of techniques. A summary of these techniques appears in (Table 1). All approaches fall into two broad categories: nonviral (usually plasmid) DNA uptake, termed “transfection,” and viral-assisted DNA uptake, termed “transduction”. Each approach has associated pros and cons. Generally, nonviral methods tend to be less efficient but very stable and with fewer side effects, while viral methods are highly efficient but come at the cost of significant drawbacks including potentially serious side effects.1

Table 1. Viral and non-viral methods used to genetically modify T cells.

Table 1

Viral and non-viral methods used to genetically modify T cells.

Viral vectors, nevertheless, dominate the field of gene therapy. All viral vectors are engineered to be replication-deficient so as to minimize risk of infection to the host. Retroviral vectors are highly effective, but are limited in that they can only transduce actively dividing cells. Interestingly, Costa et al2 have exploited this “defect” to specifically transduce only rare populations of antigen-specific T cells which are dividing in response to cognate antigen. In contrast, adenoviral vectors have the capability of transducing both actively dividing and nondividing cells. However, they, as well as retroviral vectors, are subject to transient gene expression, gene silencing and positional effects. They also tend to induce a strong immune response in the host, a caveat of all virally-mediated therapies. While the majority of viral gene therapy experiments have been carried out using retroviral and adenoviral vectors, vectors based on adeno-associated virus (AAV), herpes simplex virus (HSV) and cytomegalovirus (CMV) have also been used. One of the most promising novel vectors is based on lentivirus. Lentiviral vectors, like adenoviral vectors, incorporate into both dividing and nondividing cells, but they are not prone to gene silencing and do not elicit a strong host immune response. In addition, being derived from HIV, they are naturally well-suited for infection of T cells. On the other hand, the fact that they are HIV-derived raises the most serious threat of all viral-based therapies: the return of replication competence. In the case of HIV, this event would obviously be catastrophic; however, even “safe” viruses such as adenovirus could wreak havoc in an immunocomprimised or seriously ill patient. Notwithstanding this possibility, there is still the risk that vector integration could cause malignant transformations of the host's tissues or cause unexpected complications related to the condition being treated. Still, viral-mediated gene transfer remains the leading technology in the field today.

Another approach which is being actively explored is the use of stem cells for gene therapy. Stem cells come in a variety of forms, but of particular interest to the immunologist is the hematopoetic stem cell (HSC). HSCs are capable of repopulating the entire hematopoetic compartment, including the immune system, thus making them perfect candidates for gene therapy strategies.3 By targeting long-term progenitor cells with the appropriate promoter, expression of the transgene of interest can be limited to a particular lineage. Whereas plasmid DNA and viral DNA can be deployed in vivo or ex vivo, depending on the therapeutic approach, HSCs are almost exclusively modified ex vivo and returned to host to repopulate the hematopoetic system. They have proven effective as vectors in the genetic treatment of human ADA-SCID.4

Finally, a continuous source of genetically-modified T cells can be created by producing a transgenic animal. By microinjection of DNA into a fertilized egg, all the cells of the resulting animal can potentially express the transgene. T cells can then be harvested from that animal, or alternatively, gene expression can be limited to T cells alone via a T cell-specific promoter such as lck or CD4, thus allowing direct experimentation. Once established, a transgenic line becomes an invaluable tool; however, generation of a stable transgenic line is replete with pitfalls. Complications can arise from deleterious or lethal genes. The gene may not express or may not penetrate all tissues, resulting in a “mosaic” animal. The gene may fail to incorporate into the germ line, or may simply fail to transmit to offspring. Despite these limitations, transgenic animals remain an attractive tool to investigators.

In summary, these four general methods for T cell genetic modification have been applied in countless studies across a wide variety of animal models of human autoimmune disease. Practically all available autoimmune models have been studied; however, by far the most published work has been on the “big three:” experimental autoimmune encephalomyelitis (EAE), nonobese diabetic (NOD) mice and collagen-induced arthritis (CIA). These models correspond to the human diseases multiple sclerosis (MS), insulin-dependent diabetes mellitus (IDDM) and rheumatoid arthritis (RA), respectively. These models will be the primary focus for the animal studies reviewed in this chapter.

Animal Studies: Therapeutic Products

As previously stated, gene therapy may be viewed as the addition of normal, foreign or modified genes to alter biological function in order to treat a disease. The simplest application of this principle is to deliver a therapeutic product to the appropriate site. Conventional medical treatment of disease usually involves systemic administration of drugs, and this approach invariably produces unwanted side effects. In the case of autoimmune diseases, treatment protocols typically entail general immunosuppression, which can involve systemic toxicity and increased risk of infections and malignancies. An antigen-specific T cell mediated gene therapy approach can circumvent these risks by delivering a therapeutic product encoded by a gene directly to the target tissues in a controlled manner. The delivered product can be encoded by a modified or foreign gene, but more often than not, normal signaling molecules are used to exploit native biological pathways already in place. A key example of this approach is the use of immunomodulatory cytokines or their antagonists to mediate the autoimmune activity, usually toward an anti-inflammatory Th2 type response.

One of the early studies using this method involved treatment of EAE with the Th2 cytokine interleukin-4 (IL-4). EAE is a widely-studied animal model of the human disease MS.5 As in MS, EAE pathology includes inflammatory lesions of the CNS with perivascular infiltrates. Progression of disease leads to loss of myelin accompanied by paralysis and disability. EAE can be experimentally induced by immunizing with whole proteins or synthetic peptides of myelin, such as proteolipid protein (PLP), myelin basic protein (MBP) or myelin oligodendrocyte glycoprotein (MOG). EAE can also be adoptively transferred using CD4+ Th1 cells from an immunized animal. Shaw et al6 transduced a T cell hybridoma specific for the peptide MBP 87-99 with a viral vector which constitutively expresses IL-4. The authors used a hybridoma due to their inability to transduce primary T cell lines at that time. IL-4 was chosen as the therapeutic product because it is a Th2 cytokine which inhibits Th1 induction and macrophage activation. By targeting the cells to a myelin protein, the therapy would be naturally directed to active CNS lesions. Transduced cells were injected into EAE hosts ten days after active immunization with MBP 87-99, but before first clinical signs of disease. The results showed that the treatment ameliorated the disease, and was antigen-specific in nature. However, the treatment regimen was not conducted after clinical manifestation of disease, which would have more accurately mimicked an application in human MS treatment. Moreover, no attempt was made to regulate the expression of IL-4. However, the most critical flaw was the use of hybridoma cells as the delivery vector. As these cells give rise to malignant tumors, this approach has no practical human therapeutic value. Despite these shortcomings, important proof-of-principle was established.

Similarly, our laboratory targeted EAE with the Th2 cytokine interleukin-10 (IL-10).7 However, in contrast to the previously described study, our construct was a nonviral plasmid which we used to transfect normal T cells specific for PLP 139-151 using a DMSO/Polybrene method. Additionally, regulation was achieved using an interleukin-2 (IL-2) promoter which is normally active upon antigen engagement and is capable of driving high-level gene expression. IL-10, like IL-4, is an anti-inflammatory Th2 cytokine. Its biological functions include downregulation of MHC II expression on antigen-presenting cells (APCs), inhibition of T cell costimulatory pathways and inhibition of interferon-gamma (IFNγ) secretion by Th1 cells. These cells proved to be therapeutic when injected into EAE host mice both before and after onset of clinical symptoms, thus truly mimicking an application in human MS therapy. These cells were shown to be antigen-specific, inducible and of normal phenotype. In addition, by avoiding a viral vector, viral-associated problems were precluded.

This type of cytokine gene therapy has also been effectively demonstrated in NOD mice. The NOD mouse provides a useful model for human IDDM, also known as Type I diabetes mellitus, which is characterized by autoimmune destruction of pancreatic islet β-cells.8 The NOD mouse spontaneously develops diabetes preceded by insulitis—an inflammation and infiltration of T cells into the islets of Langerhans. Disease can also be adoptively transferred to a susceptible host with T cells from a diabetic NOD mouse. The Th1 cytokines interleukin-1 (IL-1), tumor necrosis factor-alpha (TNFα), TNFβ, and IFNγ have all been implicated in disease pathology. Thus, strategies have surrounded either blocking the actions of these cytokines, or counteracting their effects with Th2 cytokine administration.

Moritani et al9 generated islet-specific Th1 clones from autoreactive islet infiltrates of NOD mice and transduced these clones with a viral vector that constitutively expresses IL-10. The transduced cells were adoptively transferred into NOD mice at a 1:1 wild-type:transduced ratio. Under these conditions, a reduced incidence and severity of disease was observed in the treated animals. RT-PCR analysis showed reduced levels of IFNγ mRNA and increased levels of IL-10 mRNA in recipient islets. However, no regulatory control over the IL-10 was incorporated into the design, and the therapy was administered at the induction of disease, not after clinical onset.

Yamamoto et al10 similarly used a viral vector to constitutively deliver IL-4 to treat diabetes in NOD mice. However, rather than targeting cells responding to a particular antigen, this study targeted T cells of a particular subtype, those bearing CD62L. The CD62L marker was chosen because it is present on CD4+ regulatory cells which inhibit disease, but not present on CD4+ or CD8+ diabetogenic effector cells. The fact that both CD62L+ and CD62L- subsets are present during all stages of disease suggests that a natural yet abortive inhibitory mechanism exists, and with facilitation, could prove therapeutic. While transduced CD62L- cells remained pathogenic, the immunoregulatory ability of CD62L+ cells was greatly enhanced. In mixed cotransfer studies, the immunoregulatory cells inhibited disease induction in hosts. Although once again no attempt was made to regulate IL-4 production, nor was any post-onset therapy attempted, this study illustrates that subset targeting may improve existing gene therapy designs, and more importantly, that natural anti-autoimmune mechanisms can be enhanced and exploited to ameliorate disease.

Cytokine gene therapy has also been successfully applied in the CIA model. This model closely mimics human RA, characterized by synovitis and destruction of articular tissue. Evidence suggests that the disease is mediated by CD4+ T cells.11 Th1 cells secreting IFNγ are critical in disease development, whereas Th2 cytokines such as IL-4 or IL-10 are protective. Bessis et al12 showed that transfected CHO fibroblasts expressing IL-4 or IL-13 can ameliorate disease. Like the EAE model, CIA can be induced either by active immunization with antigen (usually collagen type II, abbreviated CII) or by adoptive transfer of immunogenic cells from an immunized donor. Chernajovsky et al13 transduced such immunogenic splenocytes from CII-primed mice with a viral construct designed to constitutively express transforming growth factor-beta-1 (TGFβ1). TGFβ1 exhibits multiple immunosuppressive effects and is known to correlate with recovery and/or protection in autoimmune diseases.14 When injected into susceptible hosts, the transduced cells not only failed to confer disease, but proved to be therapeutic even when transferred into mice with established disease. Like the Moritani NOD study, antigen-targeting was provided by the native immunogenic repertoire. Likewise, no regulation of expression was attempted. Importantly, though, this study demonstrates a post-onset therapy potential, a critical prerequisite to human therapy.

Although most cytokine gene therapy designs involve direct secretion of an anti-inflammatory Th2 cytokine to inhibit disease, secretion of a Th1 cytokine antagonist can be equally as potent. A fine example of this aim is the use of interleukin-12 (IL-12) p40 dimer to block the action of the Th1 cytokine IL-12. In vivo, IL-12 exists as a heterodimer comprised of distinct subunits known as p40 and p35. Whereas the p40 subunit is responsible for the binding action of the molecule, the p35 subunit is necessary for proper signal transduction; without it, the p40 monomer or homodimer acts as an IL-12 antagonist by binding receptor without signaling. In heterodimer form, IL-12 serves as a powerful Th1 cytokine which induces IFNγ production as well as other inflammatory cytokines and may play a critical role in RA pathology; thus interference with its action provides a possible therapeutic tactic. Nakajima et al15 transduced CII-specific T cell hybridomas as well as primary T cell lines with a viral construct encoding IL-12 p40 along with an IRES (Internal Ribosome Entry Site) sequence and a luminescent reporter gene. The IRES sequence allows coexpression of two distinct genes from a single mRNA transcript. In this case, patterns of p40 expression could be visualized via the reporter gene. Transduced cells were injected during disease induction with CII immunization, but prior to first clinical signs. Disease was clearly ameliorated and only CII-specific cells had a therapeutic effect. Additionally, the homing behavior of the transferred cells to affected joints was demonstrated using real-time bioluminescent imaging. However, again no regulatory control over expression was imposed, the therapy was begun prior to onset of disease, and although transduced primary T cells were effective in treating the disease, they were markedly less effective than the transduced hybridomas, whose tumorigenic properties render them unusable as vectors for human therapy. This study does however illustrate the potential for evaluating antigen-specific therapies by using imaging techniques to elucidate local delivery patterns in vivo.

On a final note, antigen-specific T cell delivery of therapeutic products may not necessarily be restricted to traditional autoimmune diseases; this type of strategy has also been examined in transplant rejection models. While autoimmune diseases can be distilled into a “self” versus “nonself” paradigm, a transplanted organ physiologically becomes part of the “self,” and thus immune rejection may in a sense constitute autoimmunity. A longstanding goal of transplantation is to manipulate T cells to accept the foreign tissue as “self.” To date, this remains a daunting task. However, a clever study by Hammer et al16 takes advantage of the natural tendency for T cells to home and migrate into allografts to potentially deliver therapeutic product(s) to minimize rejection. In this study, allo-specific T cells were transduced ex vivo with a viral construct constitutively expressing enhanced green fluorescent protein (EGFP) and transferred back into host rats with various graft types. The EGFP reporter was detected at high levels in the transplants of rats with allografts, but not in syngeneic or third-party grafts. This demonstrated the efficacy and antigen-specificity of the design. While no actual therapeutic payload was delivered, proof-of-principle of this approach was established.

Animal Studies: Regenerative Products

Regenerative product delivery is a natural extension and progression from therapeutic product delivery. The obvious first step in combating an autoimmune disease is to stop the immune onslaught. However, this may not necessarily represent a “cure.” Because tissue destruction is frequently involved in disease pathology, tissue regeneration may be necessary to exact a complete recovery. This may include repair of damaged tissue, generation of new or artificial tissue, or both. In light of recent studies, it has become clear that during autoimmunity self-suppression and/or tissue regeneration to some degree does occur. Affected organs are not merely passive targets of autoimmune disease, but actively resist using endogenous systems. 14 If autoimmune disease is a failure of these natural systems to completely harness self-reactivity and repair damaged tissues, then obviously restoring or facilitating these abilities would provide an incredibly powerful tool for clinical treatment. Surprisingly, though, much of the research in the field has focused only on halting the autoimmune process, with little attention given to regeneration. In RA it is virtually unexplored. In IDDM, focus is mostly on creation of de novo non-β-cell insulin secretion. Only in CNS disorders has much experimental progress been made, probably due to the immense importance of the tissue and its relative inability to heal itself.

When considering regenerative therapy of the CNS, several outmoded dogmas must be discarded. Firstly, CNS tissues do possess an innate albeit weak ability to heal. This ability can be greatly augmented by the introduction of native neural growth factors via gene therapy. Grill et al showed that fibroblasts engineered to secrete nerve growth factor (NGF) caused axon regrowth not only in acute, but in chronic CNS injury.17 Secondly, the CNS has long been regarded an system of “immune privilege;” that immune cells are naturally excluded from entering the CNS where evolution has determined that they would do more harm than good. While the blood-brain barrier (BBB) does ordinarily hold the immune system at bay, T cells and macrophages can and do cross it during infection and injury,18 and Schwartz et al19 have shown that the native immune response following CNS injury can actually be beneficial. Therefore, a T cell-mediated gene therapy should be both possible and advantageous. Flügel et al20 have shown proof-of-principle of this approach by using T cells to deliver NGF to the CNS. Thirdly, while it has long been known that MS/EAE pathology involves damage to the myelin sheath surrounding the axon,21 only in recent years has it been show that the axon itself is also a target and consequently suffers severe damage.22 In support of the view that intrinsic but abortive repair systems in the CNS actively participate, Mathisen et al23 showed evidence of autologous partial remyelination in EAE animals which directly corresponded to recovery from disease. Therefore, in MS/EAE applications, products which aim to regenerate myelin and/or axons have been logical choices for investigators.

Our laboratory chose platelet-derived growth factor-A (PDGF-A) as the regenerative effector to genetically treat EAE.24 PDGF-A is important in the development of the oligodendrocyte—the myelinating cell of the CNS and primary target in EAE and MS. PDGF-A stimulates the proliferation, migration, differentiation and survival of oligodendrocyte precursors. Our construct, like our previous IL-10 delivery vector, was a nonviral plasmid which incorporated the antigen-inducible IL-2 promoter to regulate expression of the transgene. And similar to the prior study, normal T cells targeted to the myelin epitope PLP 139-151 were transfected using a DMSO/Polybrene method. Cells were fluorescently labeled with PKH26 for tracking purposes and injected into actively immunized EAE mice three days after clinical onset of disease. Bioassays revealed studies that the transfected T cells did indeed produce biologically active PDGF-A and cell tracking showed that they in fact migrated to the CNS. Ongoing disease was significantly ameliorated, demonstrating a true therapeutic effect. Not only does this study illustrate an effective treatment strategy for degenerative CNS diseases, but it also shows proof-of-principle that T cells are capable of expressing fully active “nonclassical” cytokines.

While regenerative therapy may be applied secondarily to halting the autoimmune process, the two aims could potentially be overlapped to provide a synergistic therapy. In fact, the two may even be one in the same. One such case exists in the actions of NGF. In the CNS, NGF plays a pivotal role in the survival and differentiation of select neuronal populations. In addition, it exhibits immunomodulatory effects such as suppression of MHC II inducibilty in microglia and stimulation of memory B cells and Th2 responses. Flügel et al20 tested NGF gene therapy in the EAE model. In that study, MBP-specific T cell lines were transduced with a viral construct encoding NGF. These cells were adoptively transferred alone or cotransferred with wild-type MBP-specific immunogenic T cells. The transduced cells alone were incapable of causing disease, and more importantly, suppressed disease in cotransfer experiments. Histological examination of the CNS revealed that the therapy also decreased infiltration of inflammatory cells. This property was also confirmed with an in vitro BBB model. These data suggested that NGF, acting through its receptor, p75, hindered the ability of monocytes/macrophages to cross the BBB, thus providing a therapeutic effect. While this study made no attempt at regulation of expression and inhibited induction of EAE rather than treating ongoing disease, it demonstrated the potential of both NGF and the targeting of the BBB in autoimmune CNS therapy. Disappointingly, though, the authors made no attempt to explore the regenerative possibilities of the therapy.

Animal Studies: Alteration of Cellular Interactions

Of the four general categories of T cell gene therapy, this one is by far the most diverse, and exists because these studies do not conform nicely into previous categories, yet they may encompass elements of one or more simultaneously. In a generic sense, delivery of a cytokine may itself be considered an alteration of cellular interactions; however for this discussion we will primarily be concerned with genetic manipulations of intercellular functions such as: modulation of signaling pathways; apoptosis induction; “vetoing;” inhibition of epitope spreading; tolerance induction; tolerance reversal; and specificity programming.

One study in which an attempt was made to interrupt a Th1 signaling pathway was conducted by Chen et al in the CIA model.25 In this study, a CIA-susceptible transgenic mouse line expressing a hybrid IL-2/IL-4 receptor on its T cells was generated. In response to IL-2 secretion, which normally occurs in the T cell upon antigen engagement, the receptor transduces a Th2-type IL-4 signal instead of its native Th1-type IL-2 signal. Other than this type-2 response to antigen engagement, the T cells were phenotypically normal. Since Th2 responses are generally regarded as beneficial in autoimmune diseases, including RA/CIA, it was hypothesized that these animals would be protected from disease upon priming with CII. Surprisingly, the converse turned out to be true; the disease was greatly exacerbated with a greater incidence, accelerated onset and increased severity versus controls. Although the antigen-specific proliferation of these transgenic cells was normal, and their cytokine profiles were typical of the anticipated Th2 phenotype, histological examination of the arthritic joints revealed a substantial recruitment of eosinophils. Because eosinophils are capable of tissue damage and the recruitment of additional inflammatory cells, they may represent a local pathological mediator. This finding suggests that a Th2-type response may play a role in disease pathogenesis, at least in this model, and raises an important caveat for all Th2-based therapies. This interpretation has been supported by a number of other recent studies which show Th2 pathogenesis in EAE and NOD models.26.27

Another strategy for curtailing the propagation of potentially harmful inflammatory events lies in the blockade of T cell costimulatory signals. In order to become activated, a T cell not only needs to have presentation of antigen via MHC to its T cell receptor (TCR) in the context of an antigen-presenting cell (APC), but also requires a second signal mediated through the additional binding of receptors and ligands between these two cells. In the absence of this second signal, T cells will undergo anergy or apoptosis. This system is believed to be a natural failsafe mechanism to discourage naïve T cells from responding to self. As such, it provides a logical basis for an exogenous therapy. The T cell surface receptors CD28 and CD40L and their respective APC binding partners B7/CTLA4 and CD40 have been shown to be important in costimulation. Matsui et al targeted these molecules with antagonists delivered by adenoviral vectors.28 These vectors produce IgG fusion proteins of either CTLA4 or CD40, which inhibit binding of the native ligands. The model system tested was experimental autoimmune myocarditis (EAM), an analog of human myocarditis which is induced in animals by immunization with cardiac myosin. Intravenous injections of the vector(s) were administered either at induction of disease with the priming antigen or two weeks later, following clinical onset. Treatment at induction completely inhibited disease, and treatment post-onset was also remarkably effective. This represents an important finding, for it suggests a relevant clinical application in human myocarditis.

A strategically different approach to halting the autoimmune response relies not merely on downregulation of the responding cells, but seeks to actively kill them. One method of accomplishing this is to exploit the natural pathways of apoptosis, or programmed cell death. Apoptosis can be triggered by means of the Fas receptor and Fas ligand (FasL) circuit. Upon engagement of Fas, FasL induces a death signal in the target cell. This mechanism is necessary in many biological systems to sustain normal function, and is believed to play an important role in maintaining self-tolerance. Fas is expressed constitutively on most tissues and is normally upregulated during inflammation, including on the activated synovial cells and infiltrating leukocytes responsible for the pathology of CIA/RA. Despite this inherent upregulation of Fas, FasL expression in the inflamed joint remains low. Zhang et al hypothesized that upregulation of FasL at the arthritic site would counteract the massive infiltration of inflammatory cells which mediate the disease and would thus reverse its course.29 This group devised an adenoviral vector which constitutively expresses FasL and injected it directly into the joints of mice with CII-induced CIA three days after onset of disease. This therapy proved effective in ameliorating the ongoing disease. The beneficial effect was directly attributed to FasL, as the effect was nullified when a Fas-blocking agent was introduced. However, as the authors point out, the adenoviral vector was relatively short-lived due to its clearance by the host's immune system. This would obviate the need to incorporate any sort of regulatory mechanism into the vector design, although an effective clinical therapy regimen would involve multiple intra-articular injections over time. This would, however, be preferable to direct injections of even shorter-lived nongenetically-derived therapeutic agents into the joints.

An additional method for the direct elimination of disease-mediating cells builds upon yet another intrinsic mechanism for harnessing self-reactivity. It has been proposed that a natural mechanism exists in which one CD8+ T cell may present a self-peptide/MHC complex to an autoreactive CD8+ T cell and eliminate or regulate it. This process has been described as “vetoing.” Thus, the creation of a veto-like cell through genetic modification could prove therapeutic in autoimmune disease. This type of targeting of the autoimmune effector cell is precisely what was performed in an elegant EAE study by Jyothi et al.30 This study was conducted using a transgenic mouse that expresses a chimeric receptor on all its T cells. The hybrid receptor consists of an MHC II molecule complexed with a self-epitope of the CNS, MBP 89-101. The MHC is subsequently linked to the cytoplasmic activation domains of TCR. Thus, engagement of this cell with a TCR specific for its “bait,” the self-peptide, activates it. The type of activation that results depends on the phenotype of the cell possessing the chimeric receptor. For the purposes of this study, the transgenic cells were differentiated into cytotoxic lymphocytes (CTLs) thus enabling cytotoxic destruction of the target autoreactive population. These cells were transferred into hosts primed for EAE with MBP 89-101 either at the time of immunization or after onset of clinical disease. In both cases, the transferred cells greatly reduced disease severity. Even after the specificity of the autoimmune repertoire had “spread” to additional myelin self-epitopes, these neo-autoreactive cells were also suppressed. This may be due to the fact that the treatment also caused a Th2 phenotype shift in the remaining MBP-specific cells, either by guiding the development or expansion of the Th2 cells, or by selectively killing only Th1-type responders. In any case, this Th2 shift, which was evidenced by increased IL-4 production, may have afforded additional protection from disease beyond the targeted destruction of the Th1-type priming repertoire. As this approach proved effective in treating ongoing disease, this study illustrates a novel method of targeting autoimmunity that has great potential clinical relevance. Because it can be adapted to a variety of situations by altering the effector cell type, antigen and signal transduced, this technique may yield even further applications.

The concept of epitope spreading, briefly alluded to in the previous study, warrants a thorough discussion as it has great implications for all antigen-specific therapies, and can itself be targeted for arresting autoimmune disease. Epitope spreading can be defined as acquired neo-autoreactivity to epitopes not initially involved in disease. This can be attributed to the endogenous self-priming that occurs when previously sequestered self-epitopes enter the inflammatory milieu as a result of tissue breakdown. This phenomenon has been demonstrated in EAE and MS and may provide an underlying mechanism for the relapses and remissions exhibited in both.31 As the response to the initial epitope wanes during remission, response to a new self-epitope may be acquired resulting in relapse. The pattern of epitope recognition during EAE in the SWXJ mouse when primed with PLP 139-151 exhibits a predictable and invariable sequence: PLP 139-151 → PLP 249-273 → MBP 87-99 → PLP 173-198.32 The fact that this sequence of acquired self-reactivity is predicable suggests that clinical intervention in this cascade may provide a basis for therapy. Our laboratory explored this possibility by injecting the previously described genetically-engineered T cells that secrete IL-10 in response to antigen engagement.33 However in this case, the cells generated were specific for the antigen MBP 87-99, which is a spreading epitope identified in the predictable cascade, rather than the priming antigen, PLP 139-151. Treatment was initiated two days after onset of clinical symptoms. As predicted, these cells dramatically ameliorated disease, while transfected T cells specific for a nonself antigen or a nonspreading myelin epitope did not. Moreover, it was observed that the source of IL-10 eventually shifted from the transferred cells to the native T cell population, suggesting that transferred cells had induced host-derived protection. Although this study provides proof-of-principle that preemptive targeting of the epitope spreading cascade in established disease is therapeutic, deployment of this type of therapy in human MS may prove challenging. Firstly, the priming event cannot be predicted and the priming antigen is not known. Secondly, whereas a predictable cascade can be elucidated in a genetically pure mouse strain, this is not the case in the genetically diverse human population. It is not clear whether this hurdle can be overcome in order to translate this technique into human therapy. In any case, the pathologic process of epitope spreading presents an immunologic moving target that may complicate antigen-specific therapies; therefore any such proposed treatment must address this issue.

Another important native immunological system which can be advantageously manipulated through gene therapy is that of tolerance. Tolerance, simply stated, is failure to respond to an antigen. Tolerance to self is critical in maintaining normal function; if tolerance to self-antigens is lost, devastating autoimmune disease can result. Restoration of this pathologic loss of self-tolerance has long been a goal of immunology. Conventional approaches include such means as oral tolerance, wherein the patient is literally “fed” the antigen, or altered peptide ligands, modified versions of the antigen designed to coax the immune system back to a nonreactive state. While these and other conventional attempts at tolerance have revealed much about the immune system, none has proven worthy as a treatment for human autoimmune disease. Gene therapy provides a novel and hopeful strategy for bringing about this aim. One of the simplest yet paradoxically least understood ways to induce genetic tolerance is through the use of DNA vaccination. A DNA vaccine consists simply of a naked plasmid which carries the cDNA encoding the protein of choice.34 When transferred into a host, usually by intramuscular injection, one of three outcomes is possible: the plasmid can be processed by host cells to merely produce the intended protein; the protein can further be presented to the immune system to induce anergy, or a tolerant state; or presentation could result in activation of the immune system, hence the term “vaccination.” The wide variety of potential outcomes presents a puzzle but seems to depend largely on the plasmid vector and whether or not it contains immunogenic bacterial CpG motifs which generally result in activation. Despite much research in this area, the mechanism of DNA vaccination has yet to be fully elucidated. Notwithstanding, many experiments applying this strategy to autoimmune disease models have been conducted. This has led to even greater confusion, since conflicting reports have been published treating the same disease using plasmids expressing the same antigen. For example, Ruiz et al35 showed EAE amelioration while Tsunoda et al36 showed enhanced EAE. Both groups vaccinated with the PLP 139-151 determinant. It is obvious that more study is necessary before any human therapies are considered.

A further and perhaps more predictable method for genetic tolerance induction targets the APC. Chen et al37 transduced B cells in such a way as to produce antigen-specific tolerance. This group constructed a retroviral vector expressing PLP 100-154 fused to a lysosomal targeting sequence to ensure proper association with MHC II. These B cells, when injected into naïve hosts, present this self-antigen to T cells in the absence of costimulatory molecules. As discussed earlier, this second signal is necessary for activation and without it T cells enter apoptosis or anergy, thus effectively tolerized. One to two weeks following transfer, mice were challenged with PLP 139-151, an immunogenic peptide traversed by construct. The tolerized mice fared better clinically than controls; the majority were protected from disease induction. Subsequent assays revealed antigen-specific T cell nonreactivity and decreased IL-2 production. However, this treatment regimen failed to protect a substantial portion of mice, and it while it may have proven preventative, post-onset therapy was not attempted.

Agarwal et al38 conducted a similar study in the experimental autoimmune uveitis model (EAU) using an IgG-coupled antigen. EAU is an experimental model for the human retinal disease uveitis that is induced with interphotoreceptor retinoid-binding protein (IRBP). In this study, IRBP was fused to the heavy chain of IgG1 because it has tolerogenic properties known to result in long-term suppression of the antigen-specific immune response. Like the Chen study, transduced B cells were transferred to naïve hosts, and immunization with the priming antigen occurred 10 days later. Likewise, these mice fared much better than controls in disease outcome. However in this case, post-immunization therapy was also attempted. Treatment begun seven days following priming also resulted in disease amelioration, although a much more intense regimen was required. While encouraging, it is not clear whether this type of approach would be effective in human patients with established or chronic disease, especially in light of epitope spreading. An effective clinical application would have to include every possible epitope of every possible protein target involved in that disease. Moreover, while evidence suggests that B cells are long-lived, only short-term effects were examined in these studies. Due to the problems inherent with viral vectors, the benefits noted may in fact be ephemeral.

Although it may seem counterintuitive, reversal of tolerance may also be beneficial in combating autoimmunity. In this case, by breaking tolerance to an inflammatory mediator such as a Th1 cytokine, the host mounts an immune response against a critical link in an ongoing autoimmune condition. The result: autoimmunity versus autoimmunity. There is some evidence to suggest that this process occurs naturally as one of the many intrinsic countermeasures against autoimmunity. And like so many other native mechanisms, gene therapy may enhance or reinstate an inadequate natural response. Wildbaum et al took advantage of this opportunity by targeting TNFα for tolerance reversal in the adjuvant-induced arthritis (AA) model.39 AA is an alternative to the CIA model of human RA that is induced with an injection of complete Freund's adjuvant (CFA). TNFα was chosen as the target because of its potent inflammatory effects and its implication in autoimmune pathologies, particularly in RA and MS. In this study DNA vaccination was performed using a plasmid encoding TNFα and containing the immunogenic CpG motif. Disease was induced with CFA before or after DNA vaccination, and in both cases the treatment was remarkably effective. In addition, anti-TNFα antibodies were produced that surprisingly appeared to “respond” appropriately in their production, reflecting the disease state. This suggests an enhancement of a native, controlled anti-autoimmune response. Most importantly however, this study represents the critical minority of approaches in which a significant amelioration of ongoing disease was achieved. This mode of anti-TNFα therapy has also been proven effective in the EAE model.40 Wildbaum et al also used this approach to effectively modulate a therapeutic Th1 → Th2 shift.41 By targeting IFNγ-inducible protein-10 (IP-10), a chemokine which drives naïve T cells to a Th1 phenotype, they were able to break tolerance and induce an anti-IP-10 response in the hosts. This in turn mediated a Th2 phenotype shift which prevented EAE induction and ameliorated ongoing EAE.

As we have seen, there is an enormous array of approaches which may be undertaken in harnessing the power of the T cell to counter autoimmune disease. One final adjunct to these studies involves the genetic modification of the most essential T cell function—specificity. This inherent characteristic is what makes antigen-specific T cell therapy so attractive. It is in fact what gives a T cell its identity. And yet this quality is determined solely by the two chains of the TCR, alpha and beta. Kessels et al sought to introduce a genetically engineered TCR to T cells in order to exogenously “lock in” the target.42 In this study, the genes encoding both chains of a TCR specific for an antigen shared by a particular tumor and influenza strain were inserted into a retroviral vector. Mouse splenocytes were transduced with the construct ex vivo and reinfused into donors. After two days, mice were challenged with the cognate influenza strain or control strain. The mice exhibited fully antigen-specific responses to the virus in vivo. The transduced T cells also proliferated normally then subsided appropriately following viral clearance. Moreover, this therapy was also effective against established tumors bearing the target antigen. Side effects, including autoimmune, were minimal. However, the viral-based treatment itself is vulnerable to immune attack; and since TCR heterodimers of endogenous/exogenous origin are possible, unpredictable and potentially harmful effects could result. Despite these concerns, this study illustrates a novel tactic with immense potential. Although designed primarily as a method to rapidly counter infections or tumors, it could certainly be deployed in antigen-specific autoimmune therapies. For instance, designer TCRs could be used in conjunction with therapeutic payloads to direct genetically-modified T cells to their appropriate targets. This example illustrates one of a myriad of ways that practically any of the approaches discussed in this chapter may be combined in a synergistic fashion.

Future Directions

It is clear from the current state of research in the field of gene therapy that great progress has been made; however in order to advance this science one must take stock of the innumerable near-misses and failures. These will no doubt aid in the design of the next generation of gene therapies which may ultimately hold the key to reversing human autoimmune disease. Several critical points should be noted when devising new strategies. Firstly, the therapy must prove effective in treating ongoing disease. A therapy which is only effective in preventing induction of an autoimmune disease is of little clinical value. Secondly, the possible deleterious effects of the therapeutic payload must be considered. The systemic, pleiotropic, long-term and unknown effects of any agent or biological manipulation need to be evaluated. While a particular therapeutic agent itself may not present any danger, simply by altering homeostatic balance other harmful effects may indirectly result. The delivery vector may also complicate the effects of the therapy; Croxford et al showed conflicting therapeutic outcomes of IL-10 in EAE when delivered via different vectors and modes.43 Additionally, when cellular delivery vectors are used, cell type is influential. Morita et al showed disease regression in CIA when dendritic cells (DCs) were transfected, but no effect was observed when T cells or fibroblasts were used.44 Thirdly, the delivery vector or process itself may be directly harmful to the host. Infectious viral vectors or transformation of host tissue are genuine dangers. Moreover, since so many vectors are designed with no regulatory elements in mind, one must ask if constitutive delivery of any agent is prudent. It can be assumed that it would not be beneficial to have continuous therapy after the disease or harmful condition were ameliorated. Lastly, the therapy must be designed to account for the dynamic processes of disease. These include epitope spreading, relapsing/ remitting, autologous anti-autoimmune responses and other transient processes inherent in the disease course.

Current gene therapy strategies may be improved upon by the lessons learned from prior studies. First of all, better regulation and delivery of products can be achieved. Constitutive promoters may be replaced with inducible and/or tissue-specific ones. Viral delivery vectors may also be enhanced. One such recent advance is the self-inactivating (SIN) vector. A commonly observed problem associated with traditional viral vectors is that of “promoter interference,” a condition which occurs when a regulatory promoter is inserted between the viral LTR promoter elements causing hindrance of one promoter or the other. This typically manifests as low gene expression with high viral titers, or vice versa. SIN vectors overcome this problem by neutralizing the 3' LTR after integration, leaving only the regulatory promoter active. Retroviral SIN vectors are commercially available45 and lentiviral versions are currently being developed.3 While SIN vectors offer a great advantage over traditional viral vectors, still greater efficiencies have been reported using hybrid vectors. Zhao-Emonet et al46 eliminated the 3' LTR element altogether and replaced it with a CD4 minimal promoter/enhancer. Thus, this construct allows T cell-specific gene expression while maintaining high viral titers.

Furthermore, to improve regulation of gene expression an added level of control can be imposed on the delivery vector. Several commercially available systems allow gene expression to be turned on or off in vivo simply with the addition of an exogenous substance, such as doxycyline or rapamycin, which can be injected or simply added to the diet. The cre/lox system offers even greater flexibility, as the cre recombinase which triggers the reaction can either be added exogenously or transcribed from another vector (or even from the same vector) under the appropriate regulatory control. The cre recombinase reacts with a pair of specific target sites, termed “loxP,” splicing out the intervening sequence. The result can be gene activation, inactivation, or a combination of both as one gene is interrupted while another is simultaneously rejoined.47 With careful design genes can even be made to self-splice.48 This event can be temporally and spatially controlled via appropriate promoter selection, allowing such a feat as normal development of a cre/lox transgenic animal before splicing occurs. While this system offers tremendous flexibility, unlike the aforementioned systems the effects are irreversible. Although this is generally considered a drawback, permanence in some designs may be desirable.

Another level of control which may be imposed is the “suicide gene,” a genetic kill switch which can terminate runaway therapies or those which have simply run their course. One of the most commonly used is the herpes simplex virus thymidine kinase (HSV-TK) gene. Under normal circumstances it is harmless, but upon administration of ganciclovir it kills dividing host cells from within.49 This technique has been used experimentally and clinically to arrest the graft-versus-host disease (GVHD) that sometimes occurs following an allogenic bone marrow graft.50,51 Its potential in killing alloreactive T cells in early post-transplant has also been shown.52

Finally, current therapy systems may be improved upon by enhancing expression of the delivered product(s). The addition of a reporter gene, either alone or in tandem with an effector gene via an IRES sequence, can allow spatial and temporal evaluation of expression patterns in vivo.3,15 Intron sequences can be incorporated into the design to enhance gene expression, and the addition of insulator elements can greatly reduce the effects of gene silencing.

While familiar strategies continue to improve, new strategies will undoubtedly emerge. New knowledge will breed new therapies. Better understanding of disease mechanisms and pathologies, along with better understanding of the mechanisms and functions of biological agents will certainly lead researchers down new paths. One historical example of this is a clinical study of MS conducted in 1987 in which patients were treated with injections of IFNγ.53 Seven of the eighteen participants suffered severe relapses; the study was immediately halted. While it may seem counterintuitive to use an inflammatory Th1 cytokine to treat an autoimmune disease, the contemporary belief was that the pathology of MS involved an IFNγ deficiency. Additionally the disease was widely viewed as viral in nature; IFNγ also exhibits strong anti-viral properties. Its cousins, IFNα and IFNβ had already been shown to be therapeutic in MS. Therefore, based on the information available at the time, IFNγ was a rational choice. Only in retrospect does it seem otherwise, and in a similar fashion some of the therapies chosen today may seem irrational when looking back. While so many anti-autoimmune therapies are based upon Th2 mediators, one must ask if this type of cytokine skewing is prudent. The Th2 phenotype is pathological in allergy and also in the autoimmune disease systemic lupus erythematosis (SLE). It may also play a role in the pathologies of diseases classically attributed to Th1 such as CIA/RA.25 In addition, native biologic countermeasures or redundant pathways may eventually compensate an artificial cytokine shift, thus negating the effect. In any case, future research will illuminate mistakes of the past and present and allow more appropriate choices of therapeutic agents.

New delivery vectors will also undoubtedly come into play. Science has harnessed the power of infectious viruses and rendered them replication-deficient, self-inactivating workhorses that do our bidding. We've even made an ally of HIV. What's next? The next generation of delivery vehicles may not be viral vectors or plasmids. Currently, human artificial chromosomes are being developed which are capable of expressing large genes in a stable, long-term and regulated manner with a complete absence of side effects.54 These have the distinct advantage of being a completely natural and maintainable structure in human cells. To date, proof-of-principle gene replacement therapy has been shown in vitro, but has not been tested in humans. There are many chromosomal elements that are not fully understood, particularly the centromere, which may require additional research. Furthermore, since synthetic chromosomes typically occupy several megabases of DNA, a formidable system is required for in vivo delivery.

Another possibility for gene delivery in vivo lies in gene activated matrices (GAMs). This technology consists of a matrix of solid material which can serve as a platform for the direct delivery of plasmid or viral DNA at the appropriate site.55 Genes expressed could, for example, mediate tissue repair while the matrix serves as a scaffold for new cells. This method could be applied to autoimmune diseases with a single focus, such as IDDM, or may be applied in other ways to diseases with multiple foci. For example, one strategy might involve treatment of MS by creating a synthetic GAM “thymus” which would clonally delete self-reactive cells with FasL.

This example again invokes the potential power of combination therapies. A hypothetical multi-angled therapy approach is illustrated in Figure 1. As previously mentioned, multiple gene therapies may be combined to create a synergistic effect. Coexpression of multiple therapeutic products may exhibit this. Ko et al showed synergy in suppressing diabetes in NOD mice with coadministration of IL-4 and IL-10 in dual vectors.56 Vectors could potentially coexpress immunomodulatory and regenerative products (or products which elicit both effects) to shut down the autoimmune response and promote healing. In a parallel therapy, target tissue may be genetically modified to induce immunoregulation, regeneration and/or apoptosis of invading immune cells. Although perhaps many years away, a novel strategy might involve transducing T cells ex vivo with a library of vectors encoding antigen-specific TCRs recognizing all potential target-tissue epitopes while coexpressing therapeutic proteins in an antigen-inducible manner. Alternately or in addition, vector libraries encoding all potential self-epitopes of target tissue may be delivered to B cells to tolerize the patient against future attacks. These strategies may also benefit from combination with traditional nongenetic medicine. This type of holistic approach may ultimately provide the maximum benefit.

Figure 1. A schematic depiction of a hypothetical multi-pronged T cell gene therapy design.

Figure 1

A schematic depiction of a hypothetical multi-pronged T cell gene therapy design. In this representative autoimmune attack on the CNS, the target tissue (the oligodendrocyte, top) has been genetically modified to resist the autoreactive T cell (left) (more...)

Finally, the nature of autoimmunity itself is continually questioned yielding better and more useful philosophies while outmoded dogmas wither. Even the “self ” versus “nonself ” paradigm, a cornerstone of immunology for many decades, has been challenged. Matzinger has proposed it be replaced with a “danger” hypothesis, which purports that the immune system disregards what is self or nonself and merely responds appropriately to that which poses a threat to the body.57 Similarly, the Th1/Th2 paradigm has been challenged. Many reports have surfaced of nonTh2 cells which appear to regulate Th1 cells. These include Tr1, Th3, natural killer (NK) T cells and autoimmune related regulatory T cells (ART).58 Furthermore, autoimmunity is increasingly being viewed as a potentially beneficial process. Autoimmunity occurs as an innate and propitious mechanism in healthy individuals, performing critical functions such as eliminating senescent red blood cells from the body. It can play an advantageous role in CNS injury19 and may participate in the native counter-response to harmful autoimmunity.40 It has recently been demonstrated that adoptive induction of autoimmunity directed against melanocyte self-antigens resulted in sustained regression of metastatic melanomas in human patients. 59 While these patients subsequently suffered significant destruction of melanocytes, this type of exogenously-induced autoimmunity may provide a useful therapy for tissue-specific tumors of organs or tissues not essential for survival.


The preceding studies illustrate in animal models that antigen-specific T cells may indeed be effective when engaged either directly or indirectly in the treatment of human autoimmune diseases and perhaps in diseases of other modalities as well. T cells have capably met our three established goals of gene therapy, exhibiting precise targeting, long-term expression, and strict regulation of therapy delivery. They have been deployed successfully in animal models using the four general strategies outlined, delivering therapeutic products and regenerative products, participating in auspicious cellular interactions, and have themselves had genetic defects repaired to reverse disease pathology.60 By exploiting their innate homing ability, antigen-specific T cells can be rendered “guided missiles” capable of delivering virtually any payload in a regulated manner to practically any target. While it is tempting to strive to duplicate in human clinical trials the successes apparent in animal studies, caution must be exercised. The adverse side effects associated with the therapy, including its route, delivery mode and genetic product must be considered. Long-term effects must also be evaluated, and curtailment potential should be engineered into the design. Proper correlation between animal models and human diseases should be determined, as these models do not mimic their human counterparts in every circumstance. Likewise, the human population is not homogeneic as experimental mouse strains are. Thorough understandings of human diseases and their pathologies must be achieved in order to predict and prevent negative outcomes. Lastly, ethical consideration must be given to any proposed human gene therapy.61 Could the therapy do more harm than good? Could it be fatal? What would the impact on future generations be if the germ line were altered? These are some of the many important questions which must be addressed before human gene therapy moves into the mainstream of clinical medicine.

Although the treatment of human monogenic disorders with gene therapy is now a reality, complex autoimmune disorders will not be as easily conquered. However, the emerging view of the T cell as a tool for the therapy of autoimmune and other diseases will no doubt lead to effective clinical therapies with the potential to cure grave and devastating illnesses. It may be well into the distant future that an “off-the-shelf ” genetic remedy is available for the treatment of patients; that goal may never be realized. Instead, genetic therapies may need to be custom-tailored to the individual. Nevertheless, advances in scientific and medical technology may eventually endow us with the ability to provide custom genetic therapy to those in need as easily as we deliver traditional medicine.


Schenborn ET. 2000. Transfection technologies. In: Tymms MJ, ed. Methods in Molecular Biology. vol 130. Totowa: Humana Press Inc; pp. 91–102. [PubMed: 10589424]
Costa GL, Benson JM, Seroogy CM. et al. Targeting rare populations of murine antigen-specific T lymphocytes by retroviral transduction for potential application in gene therapy for autoimmune disease. J Immunol. 2000;164(7):3581–3590. [PubMed: 10725713]
Cui Y, Golob J, Kelleher E. et al. Targeting transgene expression to antigen-presenting cells derived from lentivirus-transduced engrafting human hematopoetic stem/progenitor cells. Blood. 2002;99(2):399–408. [PubMed: 11781219]
Aiuti A, Slavin S, Memet A. et al. Correction of ADA-SCID by stem cell gene therapy combined with nonmyeloablative conditioning. Science. 2002;296(5577):2410–2413. [PubMed: 12089448]
Martin R, McFarland HF. Immunological aspects of experimental allergic encephalomyelitis and multiple sclerosis. Crit Rev Clin Lab Sci. 1995;32(2):121–182. [PubMed: 7598789]
Shaw MK, Lorens JB, Dhawan A. et al. Local delivery of interleukin 4 by retrovirus-transduced T lymphocytes ameliorates experimental autoimmune encephalomyelitis. J Exp Med. 1997;185(9):1711–1714. [PMC free article: PMC2196296] [PubMed: 9151908]
Mathisen PM, Yu M, Johnson JM. et al. Treatment of experimental autoimmune encephalomyelitis with genetically modified memory T cells. J Exp Med. 1997;186(1):159–164. [PMC free article: PMC2198970] [PubMed: 9207010]
Atkinson MA, Leiter EH. The NOD mouse model of type 1 diabetes: As good as it gets? Nat Med 1999; 5(6):601-604. [PMC free article: PMC193678] [PubMed: 10371488]
Moritani M, Yoshimoto K, Ii S. et al. Prevention of adoptively transferred diabetes in nonobese diabetic mice with IL-10-transduced islet-specific Th1 lymphocytes: A gene therapy model for autoimmune diabetes. J Clin Invest. 1996;98(8):1851–1859. [PMC free article: PMC507625] [PubMed: 8878437]
Yamamoto AM, Chernajovsky Y, Lepault F. et al. The activity of immunoregulatory T cells mediating active tolerance is potentiated in nonobese diabetic mice by an IL-4-based retroviral gene therapy. J Immunol. 2001;166:4973–4980. [PubMed: 11290776]
Myers LK, Rosloneic EF, Cremer MA. et al. Collagen-induced arthritis, an animal model of autoimmunity. Life Sci. 1997;61(19):1861–1878. [PubMed: 9364191]
Bessis N, Boissier M-C, Ferrara P. et al. Attenuation of collagen-induced arthritis in mice by treatment with vector cells engineered to secrete interleukin-13. Eur J Immunol. 1996;26:2399–2403. [PubMed: 8898952]
Chernajovsky Y, Adams G, Triantaphyllopoulos K. et al. Pathogenic lymphoid cells engineered to express TGF β1 ameliorate disease in a collagen-induced arthritis model. Gen Ther. 1997;4(6):553–559. [PubMed: 9231071]
Prud'homme GJ, Piccirillo CA. The inhibitory effects of transforming growth factor-beta-1 (TGF-β1) in autoimmune diseases. J Autoimmun. 2000;14:23–42. [PubMed: 10648114]
Nakajima A, Seroogy CM, Sandora MR. et al. Antigen-specific T cell-mediated gene therapy in collagen-induced arthritis. J Clin Invest. 2001;107(10):1293–1301. [PMC free article: PMC209299] [PubMed: 11375419]
Hammer MH, Schrö G, Risch K. et al. Antigen-dependent transgene expression in kidney transplantation: A novel approach using gene-engineered T lymphocytes. J Am Soc Nephrol. 2002;13:511–518. [PubMed: 11805182]
Grill RJ, Blesch A, Tuszynski MH. Robust growth of chronically injured spinal cord axons induced by grafts of genetically modified NGF-secreting cells. Exp Neurol. 1997;148:444–452. [PubMed: 9417824]
Staddon JM, Rubin L. Cell adhesion, cell junctions and the blood-brain barrier. Curr Opin Neurobiol. 1996;6(5):622–627. [PubMed: 8937826]
Schwartz M, Moalem G. Beneficial immune activity after CNS injury: Prospects for vaccination. J Neuroimmunol. 2001;113:185–192. [PubMed: 11164901]
Flü A, Matsumuro K, Neumann H. et al. Anti-inflammatory activity of nerve growth factor in experimental autoimmune encephalomyelitis: Inhibition of monocyte transendothelial migration. Eur J Immunol. 2001;31:11–22. [PubMed: 11169433]
Prineas JW, Graham JS. Multiple sclerosis: Capping of surface immunoglobulin G on macrophages engaged in myelin breakdown. Ann Neurol. 1981;10:149–158. [PubMed: 7025748]
Trapp BD, Peterson J, Ransohoff RM. et al. Axonal transection in the lesions of multiple sclerosis. N Engl J Med. 1998;338(5):278–285. [PubMed: 9445407]
Mathisen PM, Johnson JM, Kawzcak JA. et al. Differential DM20 gene expression distinguishes two distinct patterns of spontaneous recovery from murine autoimmune encephalomyelitis. J Neurosci Res. 2001;64:542–551. [PubMed: 11391709]
Mathisen PM, Yu M, Yin L. et al. Th2 cells expressing transgene PDGF-A serve as vectors for gene therapy in autoimmune demyelinating disease. J Autoimmun. 1999;13:31–38. [PubMed: 10441165]
Chen Y, Rosloniec E, Goral MI. et al. Redirection of T cell effector function in vivo and enhanced collagen-induced arthritis mediated by an IL-2Rα/IL-4Rβ chimeric receptor transgene. J Immunol. 2001;166:4163–4169. [PubMed: 11238667]
Lafaille JJ, Van de Keere AL, Hsu AL. et al. Myelin basic protein-specific T helper 2 (Th2) cells cause experimental autoimmune encephalomyelitis in immunodeficient hosts rather than protect them from disease. J Exp Med. 1997;186:307–312. [PMC free article: PMC2198987] [PubMed: 9221760]
Pakala SV, Kurrer MO, Katz JD. T helper 2 (Th2) T cells induce acute pancreatitis and diabetes in immune-compromised nonobese diabetic (NOD) mice. J Exp Med. 1997;186(2):299–306. [PMC free article: PMC2198973] [PubMed: 9221759]
Matsui Y, Inobe M, Okamoto H. et al. Blockade of T cell costimulatory signals using adenovirus vectors prevents both the induction and the progression of experimental autoimmune myocarditis. J Mol Cell Cardiol. 2002;34:279–295. [PubMed: 11945021]
Zhang H, Yang Y, Horton J. et al. Amelioration of collagen-induced arthritis by CD95 (Apo-1/ Fas)-ligand gene transfer. J Clin Invest. 1997;100(8):1951–1957. [PMC free article: PMC508384] [PubMed: 9329958]
Jyothi MD, Flavell R, Geiger TL. Targeting autoantigen-specific T cells and suppression of autoimmune encephalomyelitis with receptor-modified T lymphocytes. Nat Biotechnol. 2002;20(12):1215–1220. [PubMed: 12426577]
Tuohy VK, Yu M, Yin L. et al. The epitope spreading cascade during progression of experimental autoimmune encephalomyelitis and multiple sclerosis. Immunol Rev. 1998;164:93–100. [PubMed: 9795767]
Yu M, Johnson JM, Tuohy VK. A predictable sequential determinant spreading cascade invariably accompanies progression of experimental autoimmune encephalomyelitis: A basis for peptide-specific therapy after onset of clinical disease. J Exp Med. 1996;183:1777–1788. [PMC free article: PMC2192533] [PubMed: 8666934]
Yin L, Yu M, Edling AE. et al. Preemptive targeting of the epitope spreading cascade with genetically modified regulatory T cells during autoimmune demyelinating disease. J Immunol. 2001;167:6105–6112. [PubMed: 11714769]
Garren H, Steinman L. DNA vaccination in the treatment of autoimmune disease. In: Fathman CG, ed. Biologic and gene therapy of autoimmune disease. Basel: Karger. 2000:203–216.
Ruiz PJ, Garren H, Ruiz IU. et al. Suppressive immunization with DNA encoding a self-peptide prevents autoimmune disease: Modulation of T cell costimulation. J Immunol. 1999;162:3336–3341. [PubMed: 10092787]
Tsunoda I, Kuang LQ, Tolley ND. et al. Enhancement of experimental autoimmune encephalomyelitis (EAE) by DNA immunization with myelin proteolipid protein (PLP) plasmid DNA. J Neuropathol Exp Neurol. 1998;57:758–767. [PubMed: 9720491]
Chen C-C, Rivera A, Ron N. et al. A gene therapy approach for treating T-cell-mediated autoimmune diseases. Blood. 2001;97(4):886–894. [PubMed: 11159513]
Agarwal RK, Kang Y, Zambidis E. et al. Retroviral gene therapy with an immunoglobulin-antigen fusion construct protects from experimental autoimmune uveitis. J Clin Invest. 2000;106(2):245–252. [PMC free article: PMC517488] [PubMed: 10903340]
Wildbaum G, Youssef S, Karin N. A targeted DNA vaccine augments the natural immune response to self TNF-α and suppresses ongoing adjuvant arthritis. J Immunol. 2000;165:5860–5866. [PubMed: 11067946]
Wildbaum G, Karin N. Augmentation of natural immunity to a pro-inflammatory cytokine (TNF-alpha) by targeted DNA vaccine confers long-lasting resistance to experimental autoimmune encephalomyelitis. Gene Ther. 1999;6(6):1128–1138. [PubMed: 10455416]
Wildbaum G, Netzer N, Karin N. Plasmid DNA encoding IFN-γ-inducible protein 10 redirects antigen-specific T cell polarization and suppresses experimental autoimmune encephalomyelitis. J Immunol. 2002;168:5885–5892. [PubMed: 12023393]
Kessels HWHG, Wolkers MC, van den Boom MD. et al. Immunotherapy through TCR gene transfer. Nat Immunol. 2001;2(10):957–961. [PubMed: 11577349]
Croxford JL, Feldmann M, Chernajovsky Y. et al. Different therapeutic outcomes in experimental allergic encephalomyelitis dependent upon the mode of delivery of IL-10: A comparison of the effects of protein, adenoviral or retroviral IL-10 delivery into the central nervous system. J Immunol. 2001;166:4124–4130. [PubMed: 11238662]
Morita Y, Yang J, Gupta R. et al. Dendritic cells genetically engineered to express IL-4 inhibit murine collagen-induced arthritis. J Clin Invest. 2001;107:1275–1284. [PMC free article: PMC209294] [PubMed: 11375417]
Julius MA, Yan Q, Zhang Z. et al. Q vectors, bicistronic vectors for gene transfer. Biotechniques. 2000;28(4):702–708. [PubMed: 10769748]
Zhao-Emonet JC, Marodon G, Pioche-Durieu C. et al. T cell-specific expression from Mo-MLV retroviral vectors containing a CD4 mini-promoter/enhancer. J Gene Med. 2000;2:416–425. [PubMed: 11199262]
Kaczmarczyk SJ, Green JE. A single vector containing modified cre recombinase and LOX recombination sequences for inducible tissue-specific amplification of gene expression. Nucleic Acids Res. 2001;29(12):e56. [PMC free article: PMC55755] [PubMed: 11410679]
Bunting M, Bernstein KE, Greer JM. et al. Targeting genes for self-excision in the germ line. Genes Dev. 1999;13:1524–1528. [PMC free article: PMC316811] [PubMed: 10385621]
Mullen CA. Metabolic suicide genes in gene therapy. Pharmacol Ther. 1994;63:199–207. [PubMed: 7809180]
Cohen JL, Boyer O, Salomon B. et al. Prevention of graft-versus-host disease in mice using a suicide gene expressed in T lymphocytes. Blood. 1997;89(12):4636–4645. [PubMed: 9192790]
Bordignon C, Bonini C, Verzeletti S. et al. Transfer of the HSV-tk gene into donor peripheral blood lymphocytes for in vivo modulation of donor anti-tumor immunity after allogeneic bone marrow transplantation. Hum Gen Ther. 1995;6(6):813–819. [PubMed: 7548281]
Thomas-Vaslin V, Bellier B, Cohen JL. et al. Prolonged allograft survival through conditional and specific ablation of alloreactive T cells expressing a suicide gene. Transplantation. 2000;69(10):2002–2003. [PubMed: 10852616]
Panitch HS, Hirsch RL, Haley AS. et al. Exacerbations of multiple sclerosis in patients treated with gamma interferon. Lancet. 1987;1(8538):893–895. [PubMed: 2882294]
Willard HF. Artificial chromosomes coming to life. Science. 2000;290(5495):1308–1309. [PubMed: 11185406]
Chandler LA, Doukas J, Gonzalez AM. et al. FGF2-targeted adenovirus encoding platelet-derived growth factor-B enhances de novo tissue formation. Mol Ther. 2000;2(2):153–160. [PubMed: 10947943]
Ko KS, Lee M, Koh JJ. et al. Combined administration of plasmids encoding IL-4 and IL-10 prevents the development of autoimmune diabetes in nonobese diabetic mice. Mol Ther. 2001;4(4):313–316. [PubMed: 11592833]
Matzinger P. Tolerance, danger, and the extended family. Annu Rev Immunol. 1994;12:991–1045. [PubMed: 8011301]
Bach J-F. NonTh2 regulatory T-cell control of Th1 autoimmunity. Scand J Immunol. 2001;54:21–29. [PubMed: 11439144]
Dudley ME, Wunderlich JR, Robbins PF. et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science. 2002;298(5594):850–854. [PMC free article: PMC1764179] [PubMed: 12242449]
Cavazzana-Calvo M, Hacien-Bey S, de Saint Basile G. et al. Gene therapy of human severe combined immunodeficiency (SCID)-X1 disease. Science. 2000;288(5466):669–672. [PubMed: 10784449]
Sugarman J. Ethical considerations in leaping from bench to bedside. Science. 1999;285(5436):2071–2072. [PubMed: 10523204]
Copyright © 2000-2013, Landes Bioscience.
Bookshelf ID: NBK5986


  • PubReader
  • Print View
  • Cite this Page

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...