REDIRECTED, GENETICALLY-ENGINEERED T REGULATORY CELLS AND THEIR USE IN SUPPRESSION OF AUTOIMMUNE AND INFLAMMATORY DISEASE
A redirected Treg cell is endowed with specificity toward a selected target antigen or ligand. The cell contains a chimeric receptor polypeptide that is expressed in a single, continuous chain, with an extracellular recognition region displayed on the surface of the cell, a transmembrane region and an intracellular signaling region. The extracellular recognition region is specific for the selected target antigen or ligand. The intracellular signaling region includes a combination of T-cell signaling polypeptide moieties, which combination, upon binding of the extracellular recognition region to the selected target antigen or ligand, triggers activation of the redirected Treg cells to cause suppression of T-cell mediated immunity. Such redirected Treg cells may be used to suppress undesired activity of T effector cells thereby mediating an immune or inflammatory response. They are particularly useful in treating T effector cell-mediated diseases, such as inflammatory bowel disease, transplant rejection and GVH disease.
Latest YEDA RESEARCH AND DEVELOPMENT CO. LTD. Patents:
1. Field of the Invention
The invention in the field of immunology and medicine relates to genetic modification of T regulatory cells with chimeric receptors with antibody-type specificity, and the use of such cells to suppress the action of T effector cells and treat any of a number of diseases and conditions in which such suppression is beneficial, primarily autoimmune and inflammatory diseases such as inflammatory bowel diseases (IBD), organ-specific autoimmune diseases, allograft rejection and Graft-vs. Host disease.
2. Description of the Background Art
Regulatory T-Cells (Tregs)
One line of research that led to discovery of Treg cells was the observation that thymectomy of mice of certain susceptible strains on postnatal day 3 results in a spectrum of organ-specific autoimmune effects, which were preventable by “reconstitution” of these animals early in life with normal adult lymphocytes (Asano M et al., J Exp Med 1996; 184:387-96). The effectors and suppressors of autoimmunity in this model of multi-organ autoimmunity were CD4+T-cells. It was subsequently shown that the regulatory CD4+Tregs that prevent disease coexpressed CD25. CD4+CD25+ Tregs represent 5-10% of total peripheral CD4+ cells in mice and 3-6% of total peripheral blood CD4+T-cells in humans (Jonuleit H et al., J Exp Med 2001; 193:1285-94).
Over the past decade, CD4+CD25+ Tregs have been studied for their function in autoimmune disease. CD4+CD25+ Tregs suppressed disease induced by cloned autoantigen-specific T effector cells (Sufi-Payer, E et al., J. Immunol., 1998; 160:1212-18). The CD4+CD25+ Treg cells appeared to be members of a unique lineage of regulatory T-cells. These authors noted that, although the target antigen(s) and mechanism of action of the CD4+CD25+T-cells remained to be determined, they likely played an important role in modulating other autoimmune diseases that are mediated by activation of self-reactive T-cells. Tregs prevent organ-specific autoimmune diseases including autoimmune thyroiditis, autoimmune gastritis, insulitis and arthritis.
More recently, it was discovered that Tregs express a transcription factor known as Foxp3 intracellularly. The absence of the transcription factor called Scurfin (also forkhead box P3) and encoded by the gene Foxp3 was known to cause a rapidly fatal lymphoproliferative disease, similar to that seen in mice lacking cytolytic T lymphocyte-associated antigen 4 (CTLA-4). Khattri R et al. (Nat Immunol. 2003; 4:337-42) showed that Foxp3 was highly expressed by Treg cells and was associated with their activity and phenotype. Foxp3-deficient mice lacked Treg cells whereas mice that overexpressed Foxp3 possess more Treg cells. Tregs constitutively express the transcription factor Foxp3 and the inhibitory costimulatory molecule CTLA-4 (Chen W et al., J Exp Med 1998; 188:1849-57.
Foxp3 is believed to act through negative transcriptional regulation of cytokine genes, including IL2, IL4 and IFN-γ (Kasprowicz D J et al., J Immunol. 2003; 171:1216-23, 2003), though many aspects of Foxp3 activity and regulation of its expression remain obscure.
Loser K et al., Gene Ther 2005; 12:1294-304, generated Tregs in vitro by infecting naïve CD4+CD25−T-cells with a retrovirus encoding Foxp3. Foxp3-infected T-cells were similar to naturally occurring Treg cells as evidenced by surface marker expression and function. These authors investigated the effects of Foxp3-infected T-cells on contact hypersensitivity responses mediated by T effector cells by injecting into sensitized mice Foxp3- or control virus-infected T-cells. Only injection of Foxp3-infected T-cells significantly inhibited CHS compared to controls, indicating that Foxp3-infected T-cells are suppressive in vivo. The authors then used Foxp3-infected T-cells to treat autoimmune-prone CD40L transgenic (Tg) mice, which develop a severe systemic autoimmune disease including autoreactive T-cells and autoantibodies. Injection of Foxp3-infected T-cells into these mice inhibited the ongoing development of autoimmune dermatitis and activation of cytotoxic CD8+T-cells. This treatment also reduced serum concentrations of antinuclear antibodies, which was paralleled with reduced renal immunoglobulin depositions and increased kidney function. The authors concluded that newly in vitro-generated regulatory T-cells can be used to treat inflammatory and ongoing autoimmune disorders successfully.
Suri-Payer E and Fritzsching B (Springer Semin Immunopathol. 2006; 28:3-16) recently summarized evidence for a role for Treg in suppression of innate and adaptive immune responses in experimental models of autoimmunity including arthritis, colitis, diabetes, autoimmune encephalomyelitis, lupus, gastritis, oophoritis, prostatitis, and thyroiditis. A common observation from such studies is that Tregs are activated in an antigen-specific manner, but exert their suppressive function in an antigen-independent manner, mainly by producing and secreting suppressive cytokines such as IL-10 and TGF-β. Tregs can suppress “conventional” T-cells in vitro by direct cell contact. It is appreciated, however, that down-regulation of antigen-presenting cell (APC) function, such as that of dendritic cells, and attenuation of secretion of inhibitory cytokines such as IL-10 and TGF-β might be important for Treg function in vivo. The final outcome of autoimmunity vs. tolerance depends on the balance between stimulatory signals to T effector cells and inhibitory signals from Treg. Whereas earlier studies analyzed the capacity of Tregs to prevent onset of autoimmune disease, more recent reports indicate successful treatment of ongoing disease.
In vivo, adoptive transfer of Tregs achieved the following effects:
(1) suppressed development of autoimmunity;
(2) suppressed acute rejection of transplanted solid organs; and
(3) suppressed anti-tumor immunity,
A review of Sakaguchi S et al., Immunol Rev. 2006; 212:8-27, noted that naturally arising CD25+CD4+ Treg cells play key roles in the maintenance of immunologic self-tolerance and negative control of a variety of physiological and pathological immune responses. Most of these cells are produced by the normal thymus as a functionally mature T cell subpopulation. Natural Tregs specifically express Foxp3, a transcription factor that plays a critical role in their development and function. Complete depletion of Foxp3-expressing natural Tregs (whether they are CD25+ or CD25−) activated even weak or rare self-reactive T effector cell clones, inducing severe and widespread autoimmune/inflammatory diseases. Natural Tregs are highly dependent on exogenously provided interleukin (IL)-2 for their survival in the periphery. In addition to Foxp3 and IL-2/IL-2 receptor, a deficiency or functional alteration of other molecules (expressed by T-cells or non-T-cells), may affect the development or function of Tregs, self-reactive T effector cells, or both, and consequently tip the balance between the two populations in the periphery toward autoimmunity. Thus, Tregs suppress the activity of T effector cells that are a major cause of antigen-specific autoimmune inflammatory disorders. Tregs induce anergy and promote suppression by a process that involves both cell-cell contact, and probably more importantly, by their secretion of TGF-β and IL-10.
Sakaguchi et al., supra, stated that elucidation of the molecular and cellular basis of Treg-mediated active maintenance of self-tolerance will facilitate (1) our understanding of the pathogenesis of autoimmune disease and (2) development of novel methods of autoimmune disease prevention and treatment.
The present invention is directed to one such novel approach to prevent or treat autoimmune disease and related immune/inflammatory conditions by imposing Treg-mediated control over T effector cells.
Engineering of Chimeric Receptors with Antibody Specificity into T-Cells
Efforts to confer antibody-like specificity to T lymphocytes arose as a response to certain basic discoveries and the failure to convert them to therapeutic success. Human tumor-associated antigens have been shown to exist as peptides associated with major histocompatibility complex (MHC) proteins. The lack of success in vaccinating tumor patients with a large variety of tumor-associated antigen vaccines were thus frustrating (Rosenberg S A et al., Nat Med 2004; 10:909-15). Passive vaccination with anti-tumor antibodies (Abs), tumor infiltrating lymphocytes (TILs), or lymphokine activated killer (LAK) cells showed very limited effectiveness (primarily for vascular tumors). More recent successes with the adoptive transfer of TILs into stage 1V melanoma patients ((Dudley M E, et al., J Immunother 2001; 24:363-73; Dudley M E, et al., Science 2002; 298:850-4) were still burdened by the fact that the results were limited to a few cancers and to individuals from whom it is possible to derive specific TILs.
To overcome such limitations in adoptive cellular immunotherapy of cancer, and in order to develop an approach that was not restricted to individual patients or limited to a specific form of cancer, one of the present inventors and his colleagues developed the “T-body” approach (Gross G et al., Proc Natl Acad Sci USA, 1989; 86:10024-8; Gross G and Eshhar, Z, FASEB J 1992; 6:3370-8), See also the following patent publications by Eshhar and colleagues, all of which are incorporated by reference in their entirety: US Pat Pub 2002-0137697, U.S. Pat. No. 5,912,172, U.S. Pat. No. 5,906,936, International Patent publications WO00/31239, WO97/15669, WO95/14710 and WO93/19163. In these approaches, a chimeric receptor (CR), was made, for example, by fusing the variable portion of an antibody, such as an anti-tumor monoclonal antibody (mAb), to a lymphocyte intracellular triggering domain, so as to be expressed by the T-cell into which the gene has been transfected as the extracellular domain of that T-cell triggering molecule. Following the expression of such CR genes in immune effector cells (T-cells and natural killer (NK) cells), the resulting engineered cells (nicknamed “T-bodies”) recognized their tumor targets and efficiently killed them. Thus, the chimeric immune receptor confers redirected antigenic specificity coupled to direct, MHC-independent triggering of cellular activation in response to binding of pre-defined target antigen.
Originally, the heterodimeric CR configuration comprised the two T cell receptor (TCR) α and β chains in which each pair of TCR variable domains (Vα and Vβ) was replaced with a pair of VH and VL domain derived from a selected antibody. These two Ab-derived coding sequences were co-transfected into T cell lines and were found to confer antibody specificity (Gross et al., 1989, supra). Thus, T-cells could be activated to effector function such as cell killing or cytotoxic activity against an immunologically specific target in a manner that was MHC independent (and thereby non-restricted) (Gross et al., 1992, supra).
In “second generation” CR's, the single chain configuration of the CR was further manipulated to obtain a configuration that would be useful particularly for cancer or antiviral therapy. Here, a single chain Fv (scFv) of an antibody was linked to transmembrane and cytoplasmic domains of lymphocyte triggering moieties such as the TCR/CD3 complex-associated ζ chain, or the Fc receptor γ chain (Eshhar Z et al., Proc Natl Acad Sci USA 1993; 90:720-4). This single chain configuration, which combined antibody recognition and T cell signaling in a single continuous protein, was a modular structure with functional domains that are simple to manipulate, and could be readily expressed in human lymphocytes using retrovirus-based vectors (Eshhar Z et al., J Imm Meth 2001; 248:67-76). Other such receptors designed by some of the present inventors and colleagues, and discussed in more detail below, are referred to as tripartite chimeric receptors (TpCR) that also include a costimulatory domain(s) (e.g. CD28, 4-1BB).
With time, it has emerged that redirection of the specificity of T effector cells using single chain CRs has become a valid therapeutic option for cancer. Many investigators have adapted this “T-body” approach to endow T-cells with various specificities and functions (Gross & Eshhar, supra; Willemsen R A et al., Hum Immunol 2003; 64:56-68; Baxevanis C N et al., Cancer Immunol Immunother 2004; 53:893-903). For a recent review, see: Eshhar, Z. “The T-Body Approach Redirecting T Cells with Antibody Specificity,” in Therapeutic Antibodies. Handbook of Experimental Pharmacology 181, Chemajovsky & Nissim (eds.), Springer-Verlag, 2008, pp. 329-342.
In the present invention, the inventors have conceived expanded uses of such CR's for the treatment of undesired immune/inflammatory conditions such as autoimmune diseases (with a particular initial emphasis on inflammatory bowel disease, IBD) and graft rejection.
Chronic Inflammatory Disease and Animal Models
Inflammatory conditions, particularly chronic inflammatory diseases, are of particular importance in clinical medicine. These diseases, caused by actions of the immune system, involve inappropriate or excessive activation of certain T-cells, expression of regulatory cytokines and chemokines, loss of immune tolerance, and the like. Examples of autoimmune and/or chronic inflammatory diseases are multiple sclerosis, inflammatory bowel diseases (IBD), joint diseases such as rheumatoid arthritis, and systemic lupus erythematosus. Some of these diseases are rather organ/tissue-specific as follows: intestine (Crohn's Disease), skin (psoriasis), myelinated nerves (multiple sclerosis or MS), pancreatic islet or β cells (insulin dependent diabetes mellitus (IDDM) or Type I Diabetes), salivary glands (Sjogren's disease), skeletal muscle (myasthenia gravis), the thyroid (Hashimoto's thyroiditis; Graves' Disease), the anterior chamber of the eye (uveitis), joint tissue (rheumatoid arthritis), and various cardiovascular diseases.
Inflammatory bowel disease (IBD) is a collective term used to describe two intestinal disorders whose etiology is not completely understood: Crohn's disease and ulcerative colitis. IBD occurs worldwide and afflicts several million people (0.3% of people in Western countries), and its incidence is on the rise (Tsironi E, et al., Am J Gastroenterol. 2004; 99:1749-55). The course and prognosis of IBD varies widely. Onset of IBD is predominant in young adulthood and presents typically with diarrhea, abdominal pain, and fever; anemia and weight loss are also common signs. Between 10% and 15% of people with IBD require surgery over a ten year period. Patients with IBD are also at increased risk for the development of intestinal cancer. These diseases are accompanied by a high frequency of psychological symptoms, including anxiety and depression.
Although the pathogenesis of this common T cell-mediated disorder remains uncertain, it is believed to result from loss of tolerance in the intestinal immune system due to the presence of constant antigenic stimulation provided by the very large numbers of resident bacteria (Podolsky D K, N Engl J Med 2002; 347:417-29). Unfortunately, new therapies for IBD are few, and both diagnosis and treatment have been hampered by a lack of detailed knowledge of the etiology. A combination of genetic factors, exogenous triggers and endogenous microflora can contribute to the immune-mediated damage of intestinal mucosa. Bacteria have been implicated in initiation and progression of Crohn's Disease since intestinal inflammation frequently responds to antibiotics. Common intestinal colonists and novel pathogens have been implicated, either because of direct detection or disease-associated anti-microbial immune responses. In many genetically susceptible animal models of chronic colitis, luminal microorganisms appear to be a necessary cofactor for disease.
The initiating step in autoimmune disease pathology is often obscure in humans where the diseases are largely sporadic, and symptoms may appear years after the first pathogenic T cell is activated. It has therefore been difficult to design effective therapies to block induction of disease. In contrast, there are common features in many of the later stages of these diseases. Inflammation at the disease site/target organ is typically present, caused by the release of inflammatory, also termed “proinflammatory,” cytokines (e.g. TNF-α and interferons) by T-cells and by other cells that contribute to the activation steps and effector pathways of immune/inflammatory processes. These cells include (among others) macrophages, dendritic cells and their precursors, B lymphocytes and plasma cells and NK cells (including NKT-cells). These reactions often involve destruction of “target” cells and tissue damage.
Studies using murine models of experimental chronic inflammation are helping to define nature of the immunological dysregulation that initiates inflammation and leads to destruction of specific end organs as well as for testing therapies. See, for example, Mombaerts et al. Cell, 1993; 75:274-82; Tarrant et al., 2998; J Immunol, 161:122-7; Powrie et al., Immunity, 1994; 1:553-62; Hong et al., J Immunol, 1999; 162:7480-91; Horak, Clin Immunol Immunopathol, 1995, 76(3 Pt 2):5172-173; Ehrhardt et al. J Immunol, 1997; 158:566-73; Davidson et al., J Immunol, 1998; 161:3143-9; Kuhn et al. Cell, 1993. 75:263-74; Neurath et al., J Exp Med, 1995. 182:1281-90. W. Strober, 2002; Annu. Rev. Immunol. 20:495-54 reviews mucosal models of inflammation, and is incorporated by reference in its entirety. One hallmark of the better of these models is that the histopathology and pathophysiology resembles that of the parallel human conditions, further enhancing the models' utility in testing novel treatment strategies. In the case of IBD this development has not been uniform. Most emphasis has been placed on modulation of immune mechanisms (Blumberg R S et al., Curr Opin Immunol. 1999; 11:648-56; Strober et al., supra) and recently of the enteric flora (Sartor R B, Curr Opin Gastroenterol. 2001; 4:324-330).
Bhan A K et al., 1999 Immunol Rev 169:195-207 reviewed studies of colitis in transgenic (Tg) and knockout (KO) animal models for mucosal inflammation in IBD. Genetics and the environment, particularly the normal enteric flora, were factors in the development of mucosal inflammation, as stated above. Normal mucosal homeostasis was disrupted by cytokine imbalance, abrogation of oral tolerance, breach of epithelial barriers, and loss of immunoregulatory cells. Some but not all immunodeficiencies, in the appropriate setting, led to colitis. CD4+T effector cells have been identified as the pathogenic lymphocytes in colitis, and can mediate inflammation by either the Th1 or the Th2 pathway. The Th1 pathway dominates in most colitis models (and human Crohn's Disease). In contrast, in the colitis observed in mice, the T cell receptor (TCR) a chain knockout mice (TCRα KO mice) shared many features of ulcerative colitis including the dominance of Th2 pathway in colon inflammation. Such models are important for the development of therapeutic strategies to treat IBD. In a later review, the same group (Mizoguchi A et al., 2003, Inflamm Bowel Dis. 9:246-259) noted that exaggerated immune responses to normal enteric microflora are involved in the initiation and perpetuation of chronic intestinal inflammation. A major pathway involves development of “acquired” immune responses by the interactions of CD4+TCRαβ+T-cells with antigen-presenting cells (APC), particularly dendritic cells. CD4+CD25+ Treg cells attenuated activated T cell responses.
The progression from the acute to the chronic phase of IBD has not been well characterized in animal models and cannot be easily evaluated in patients. Spencer D M et al. Gastroenterol. 2002; 122:94-105 reported changes in the mucosal immune response over time in experimental colitis. Severity of colitis, body mass, stool consistency and blood content, serum amyloid A, and tissue histology were examined in mice deficient in interleukin-10 (IL-10) over 35 weeks. The corresponding production of IL-12, IL-18, IFNγ, TNFα, IL-4, and IL-13 by lamina propria mononuclear cells in the inflamed intestine was measured. Administration of a neutralizing anti-IL-12 monoclonal antibody (mAb) at distinct times during disease progression permitted evaluation of the therapeutic potential of this agent. Lamina propria mononuclear cells from mice with early disease synthesized progressively more IL-12 and IFNγ, whereas production of both cytokines declined dramatically and returned to pre-disease levels in the late phase. Consistent with this pattern, the neutralizing anti-IL-12 reversed early, but not late, disease. In contrast, IL-4 and IL-13 production increased progressively from pre- to early to late disease. It was concluded that colitis developing in IL-10-deficient mice evolves into two distinct phases. IL-12 plays a pivotal role in early colitis, whereas other immune mechanisms, presumably mediated by IL-4 and IL-13, predominate in late disease to sustain chronic inflammation.
IL-10 and Chronic Inflammatory Disease
It has been known for some years that IL-10 affects the growth and differentiation of many hemopoietic cell types in vitro and is a particularly potent suppressor of macrophage and T cell functions. These observation were based in part from use of IL-10-deficient (knockout, KO) mutant mice by gene targeting (Kuhn R et al., Cell 1993; 75:263-74). In these mice, lymphocyte development and antibody responses are normal, but most animals are growth retarded, anemic and suffer from chronic enterocolitis. Alterations in the intestine include extensive mucosal hyperplasia, inflammation, and aberrant epithelial expression of major histocompatibility complex (MHC) class II molecules. In contrast, IL-10 KO mutants kept under specific pathogen-free conditions, develop only localized inflammation (limited to the proximal colon). It was concluded that (1) bowel inflammation in these mutants originated from uncontrolled immune responses stimulated by enteric antigens and (2) IL-10 is an essential (negative) regulator in the intestinal tract.
In a study validating this IL-10 KO mouse model of colitis, T. Scheinin, T et al. (Clin Exp Immunol 2003; 133:38-43) emphasized that a valuable animal model must respond to existing therapy in a way that resembles the response of human disease. Since refractory Crohn's Disease responded well to anti-TNFα antibody therapy, the investigators examined responses of IL-10 KO mice to anti-TNFα therapy, using a new scoring system similar to the Crohn's Disease “Activity Index” in humans. Stool samples were tested for cytokines and the findings compared with histology. Results showed that anti-TNF antibody therapy starting at 4 weeks markedly ameliorated the disease (as judged by the clinical score or by gut histology). A marked diminution of inflammatory cytokines in stool samples was noted, adding a further accurate measure of clinical improvement. The authors concluded that this model is useful for evaluating other therapeutic modalities of relevance to Crohn's Disease.
Treg Cells and Inflammatory Bowel Disease
One of the commonly used animal models of IBD involves adoptive transfer of CD45RBhiCD4+T-cells into SCID mice, leading to the development of massive colon mononuclear cell infiltrates, epithelial cell hyperplasia and ulceration (Thornton A M, et al., J Immunol 2000; 164(1):183-90). Cotransfer of large numbers of CD4+CD25+ Tregs prevented the development of colitis or cured established colitis, an affect that required signaling through CTLA-4 (Read S, et al., J Exp Med 2000; 192:295-302; 2000; Morrissey P J, et al., J Exp Med 1993; 178:237-44). Even after the development of immune-mediated colitis, adoptive transfer of 106 CD4+CD25+ cells caused significant improvement of intestinal inflammation (Fantini M C et al., Gut 2006; 55:671-80; Mottet C, et al., J Immunol 2003; 170(8):3939-43; Uraushihara K, et al., J Immunol 2003; 171:708-16).
In humans suffering from IBD, peripheral regulatory CD4+CD25+ cells retain suppressive activity. However, in contrast to other intestinal inflammatory disorders, the number of these regulatory cells decreases in peripheral blood during active inflammation and only slightly increases in intestinal lesions (Maul J, et al., Gastroenterology 2005; 128(7):1868-78). This aberration suggests that Treg homing defects, as well as dysregulated in situ activation contribute to the pathogenesis of IBD
There is therefore a recognized need in the art to find modalities to suppress autoimmune/inflammatory reactions and diseases, including but not limited to IBD, as well as to suppress rejection of organ and tissue grafts and prevent Graft vs. Host (GVH) disease. The present invention provides a novel approach, that of redirecting Treg cells, as a means to recruit Tregs to sites of inflammation, and activate them to suppress such immune/inflammatory reactions and protect against, alleviate and even cure such disease as IBD.
SUMMARY OF THE INVENTIONThe present invention is based on the inventors' conception that CR-mediated redirection and activation of Treg cells at sites of inflammation results in suppression of inflammatory conditions, commonly part of organ-specific autoimmune disease and exemplified herein as inflammation in the colon in experimental IBD. The inventors have further conceived of using these cells to overcome rejection of mismatched cells and tissues by T effector cells that arise in transplant recipients or to inhibit the pathogenic action of transplanted immunocompetent cells in the case of GVH disease.
The invention relies on the inventors' innovative T-body approach that has thus far proven useful for immunotherapy of cancer (and is currently in phase I/II clinical trials). The invention provides a new approach to the exploitation of Treg cells for amelioration of pathologic and undesired immune responses, particularly immunotherapy of autoimmune and inflammatory conditions, including various organ-predominant autoimmune diseases, and other pathologic or undesirable immune responses such as graft rejection and graft vs. host disease.
According to this invention, Treg cells are endowed with CRs that are specific for a selected target antigen or ligand. Such modification causes activation of redirected Tregs at sites of inflammation to suppress the proinflammatory effector-type immune responses. Based on the present inventors' (and their colleagues') results with redirecting antitumor effector lymphocytes, it is expected that Tregs, endowed with predefined specificity, will migrate/home to and accumulate in, a targeted site, such as the inflamed colon, where they will suppress disease-mediating T effector cells. To avoid the necessity of migration or homing to the targeted site, the Tregs may, where possible, be administered directly at or to such site, where they will become activated and suppress disease-mediating T effector cells.
Such redirected Treg cells, also referred to as “T-bodies,” are Treg cells that have been genetically engineered to express a CR, preferably a tripartite chimeric receptor (TpCR) that is made of a single chain extracellular recognition unit, a transmembrane region, and an intracellular signaling region.
The extracellular recognition region is specific for a selected target antigen or ligand and may preferably be a single chain antibody variable (scFv) region or another ligand that is capable of binding to the target antigen or ligand. The extracellular recognition region preferably does not comprise an MHC protein extracellular domain. The redirected Tregs of the present invention are sometimes referred to herein as “T-bodies” despite the fact that the extracellular recognition region is not necessarily an antibody domain. Thus, this term is not intended to be limited to Tregs with antibody-like specificity, but also includes Tregs with ligand-receptor-like specificity or otherwise.
A flexible spacer region may be present between the extracellular recognition region and the transmembrane region. Such flexible spacer is preferably an immunoglobulin (Ig)-like hinge, such as any hinge region derived from the Ig superfamily.
The intracellular region includes a combination of T-cell signaling polypeptide moieties, fused in tandem, which combination of moieties, upon binding of the extracellular recognition region to the selected target antigen or ligand, triggers activation of the Treg cells to cause suppression of T-cell mediated immunity. The T-cell signaling moieties preferably include one or more cytoplasmic domains of a costimulatory molecule (e.g., CD28) and a cytoplasmic T-cell stimulatory domain, e.g., of FcRγ or a CD3-ζ chain. The redirected Treg cells become specifically activated, upon binding of the extracellular recognition region of the CR to its target antigen or ligand, in a manner that is preferably (1) not restricted by, or dependent upon, the binding of the target antigen or ligand to an MHC, nor is it otherwise dependent in any way on the MHC (HL-A) haplotype of the recipient and (2) independent of engagement of costimulatory ligand(s) on a target cell.
A preferred target disease of this invention is an IBD such as ulcerative colitis, in which the present methods, as used successfully in an animal model, will permit Treg cells to reach bowel lesions in IBD patients and become efficiently activated at the inflammation site. The present invention results in site-specific Treg accumulation, ultimately resulting in CR-mediated, antigen-specific activation that results in the production of suppressive cytokines which in turn suppress effector autoimmune T-cells in an antigen-nonspecific manner, leading to alleviation of symptoms and thereby treating the disease.
The present therapeutic approach has several unique advantages. In contrast to other immunotherapeutic models, it involves T-bodies redirected with a CR, and preferably a TpCR, that combines antibody/antigen or ligand/receptor recognition with stimulatory and costimulatory motifs. Thus, T-bodies can be fully activated in a way that is not restricted by the MHC and is independent of a requirement for costimulation. The second advantage of the present invention stems from the fact that, although Treg activation is antigen-dependent, the suppressive action of these cells is antigen-, TCR-, and MHC-independent. By exploiting this property, one can construct a chimeric receptor that is specific for one or more tissue-associated antigens rather than requiring specificity for an unknown number of yet undefined autoimmune disease-specific antigens. Expression of such chimeric receptors in Tregs redirects these cells and their activation to the appropriate target tissue (in a preferred embodiment, the colon) so that they are activated in an antigen-specific manner, where their potent suppressive effects take place without a need for further recognition of disease-associated-antigens (the “bystander effect”). By using specifically-activated Tregs, many fewer cells are required to treat autoimmune inflammatory conditions; such as IBD, or allograft-associated reactions in patients than would have been possible prior to this invention when much higher numbers of non-specific Tregs would have been needed.
The present inventors have constructed strains of transgenic (Tg) mice whose T-cells and natural killer (NK) cells express an antigen-specific TpCR. For exemplification of the invention, the inventors selected the trinitrophenyl (TNP) hapten as the specific target of the TpCRs. TNP specific Tregs isolated from these Tg mice suppressed TNP-specific effector T-cells in vitro and in vivo and were able to suppress trinitrobenzenesulfonic acid (TNBS)-induced colitis in mice. In this embodiment, the target antigen for Treg and the pathogenic antigen—the hapten TNP—are the same. In another example, TNP-specific Tregs suppressed oxazolone-induced colitis in mice in which a low dose of TNP was introduced into the colon together with the oxazolone challenge. However, the TNP-specific Tregs had no effect on the oxazolone-induced colitis in the absence of TNP introduction. This establishes the “bystander” effect of the present invention, i.e., that the target antigen need not be the pathogenic antigen, as long as the redirected Tregs are activated in the vicinity of the pathogenic or undesired immune response.
One distinct advantage of the present invention is that it provides cells and methods that permit antigen-specific activation and antigen-nonspecific action of Treg cells used to suppress effector T cell responses (and treat consequent pathologies) in a way that does not require identity between the ligand (e.g., the antigen) recognized by the TpCR (e.g., by its target recognition portion) and the ligand/antigen that plays a pathogenic role in the disease process. Thus, the antigen that is pathogenic does not have to be recognized by the T effector cells being suppressed, and, indeed, may be unrelated to the disease or condition being treated. Thus, the invention exploits the “bystander” effect. As long as the Treg is in the correct vicinity where T effector cells are located and mediating their undesired effects, the redirected Tregs of the present invention can be triggered or activated at that location to release of suppressive cytokines (e.g., IL-10 and TGF-β), that will result in suppression of any “bystander” effector T-cells, and by this mechanism, quell an ongoing inflammatory/autoimmune response.
The unique characteristics of the Tg system used in the present examples enables evaluation of the suppressive effect of antigen-specific Tregs in IBD both in vitro and in vivo. Different means are used to induce Tregs, allowing those of skill in the art to select the optimal method for generating efficient numbers of antigen-specific redirected Tregs for a desired antigen or disease/condition. According to this invention, redirected human Tregs constitute an effective cell-based therapeutic modality for IBD or ulcerative colitis and, more broadly, for any T effector cell-mediated disease or condition.
To overcome the scarcity of antigen-specific Tregs, the present invention includes methods to induce these cells using cytokines (e.g. TGF-β) or by expression of transgenes (e.g. encoding the Foxp3 transcription factor) that will, together with the TpCR, allow antigen-specific Treg expansion.
According to the present invention, human Treg cells, derived from either the subject with the autoimmune/inflammatory disease or condition to be treated, or from an HLA-matched healthy donor (or a universal cell that is not recognized by the recipient's immune system), are endowed with antigen/ligand-specificity, by transduction with the antigen/ligand-specific TpCR as disclosed herein. Alternatively, the cells being endowed with antigen-specificity are the entire T-cell population and the nucleic acid construct including the sequence encoding the TpCR further includes a Foxp3 transgene that is present as an independently transcribed cistron. In another such alternative, a Foxp3 transgene is separately transfected into the T-cell population, to turn the T-cells into Treg cells. Examples provided below include studies using murine colonoscopy, in vivo imaging and immunofluorescence, and provide the basis for a novel cell-based therapeutic modality for IBD, and, by extension, for other pathologic and undesired immune responses mediated by antigen specific T effector cells.
Various embodiments of the invention are described more specifically below.
-
- The present invention is directed to a redirected regulatory T lymphocyte (Treg cell) endowed with specificity toward a selected target antigen or ligand, which cell comprises a chimeric nucleic acid that encodes a chimeric receptor (CR) polypeptide that comprises, expressed in a single, continuous chain, an extracellular recognition region, a transmembrane region and an intracellular signaling region, and is expressed in the Treg cell so as to display the extracellular region on the cell surface, wherein
- (a) the extracellular recognition region of the chimeric receptor is specific for the selected target antigen or ligand, and does not comprise an MHC protein extracellular domain; and
- (b) the intracellular region comprises a combination of T-cell signaling polypeptide moieties which combination of moieties, upon binding of the extracellular recognition region to the selected target antigen or ligand, triggers activation of the Treg cells to cause suppression of T-cell mediated immunity.
In preferred embodiments of the present invention, the extracellular recognition region is an antibody-derived scFv domain that is specific for a selected antigen. In another preferred embodiment, the extracellular recognition region is a member of a ligand-receptor pair, which is specific for the other member of that pair.
Preferably, the extracellular recognition region is linked to the transmembrane region through a flexible spacer, which, more preferably, is a hinge from a molecule of the immunoglobulin family.
The intracellular signaling region preferably includes a signaling moiety from a chain of an antigen-specific T-cell receptor, which more preferably is one having a polypeptide region comprising an immunoreceptor tyrosine-based activation motif (ITAM). Non-limiting examples of antigen-specific T-cell receptors are chains of the TCR/CD3 complex, a TCR α, β, γ or δ chain, and the γ chain of an Ig Fc receptor (FcRγ). The chain of an antigen-specific T-cell receptor is preferably the CD3/ζ chain or an FcRγ subunit.
The intracellular signaling region further preferably includes a signaling moiety of a costimulatory-receptor protein of a T-cell. The costimulatory-receptor protein is preferably selected from CD28, OX40, CD40L (gp39), 4-1BB and PD-1 (or preferably the human form or homolog of these costimulatory molecules). Most preferred among these is CD28 or 4-1BB. In another preferred embodiment, the intracellular signaling region includes more than one of the costimulatory-receptor protein signaling moieties. For example, the combination of T-cell signaling polypeptide moieties in the intracellular signaling region may include both CD28 and 4-1BB. In a particularly preferred embodiment, the extracellular hinge and transmembrane regions of CD28 are used as the extracellular hinge and transmembrane regions of the chimeric receptor.
The intracellular signaling region may also include a signaling moiety from a cytokine receptor of a T-cell, such as the IL-2 receptor or the TGF-β receptor. The latter may help to induce the T-cell containing the chimeric receptor to adopt the characteristics of a Treg cell.
The intracellular region may also include a signal-transducing enzyme that (a) is an enzyme in the signal transduction pathway of an antigen-specific T-cell receptor or (b) is an enzyme with corresponding specificity and activity as the enzyme of (a), derived from a non-T-cell lymphocyte. Such enzyme is preferably a kinase, such as the Syk kinase.
The chimeric nucleic acid encoding the CR may also include a nucleotide sequence that encodes Foxp3 arranged such that Foxp3 is expressed by the Treg cell independently of the chimeric receptor. In other words, the chimeric nucleic acid may be bicistronic such that the Foxp3 transgene is present as an independently transcribed cistron.
The target antigen or ligand is preferably one that is present or expressed at a site or target tissue of an immune or inflammatory response mediated by effector T-cells. The autoimmune or inflammatory response may comprise an autoimmune response or disease, an allograft or xenograft response or rejection, or graft vs. host (GVH) disease. In an alternative preferred embodiment, the target antigen or ligand may be an autoantigen or an antigen that is cross-reactive with an autoantigen, i.e., is also bound by an antibody that is specific to the autoantigen. The autoantigen may be a pathogenic antigen in the pathophysiology of the autoimmune disease.
The antigen is not necessarily an autoantigen, but can be, for example, an antigen that is part of the bacterial flora, such as LPS derived from the bacteria native to the colon.
The autoimmune disease or graft response and the antigen/ligand or antigens/ligands against which the Treg is specific is preferably selected from the following group:
- (a) inflammatory bowel disease (IBD), wherein the antigen or ligand is one that is expressed in diseased colon or ileum;
- (b) rheumatoid arthritis, wherein the antigen or ligand is an epitope of collagen or an antigen present in joints;
- (c) Type I diabetes mellitus or autoimmune insulitis, wherein the antigen or ligand is a pancreatic β cell antigen;
- (d) multiple sclerosis, wherein the antigen or ligand is, for example, a myelin basic protein (MBP) antigen or MOG-1 or MOG2-2, or a neuronal antigen.
- (e) autoimmune thyroiditis, wherein the antigen or ligand is a thyroid antigen;
- (f) autoimmune gastritis, wherein the antigen or ligand is a gastric antigen;
- (g) autoimmune uveitis or uveoretinitis, wherein the antigen or ligand is S-antigen or another uveal or retinal antigen
- (h) autoimmune orchitis, wherein the antigen or ligand is a testicular antigen;
- (i)) autoimmune oophoritis, wherein the antigen or ligand is an ovarian antigen;
- (j) psoriasis, wherein the antigen or ligand is a keratinocyte antigen or another antigen present in dermis or epidermis;
- (k) vitiligo, where the antigen or ligand is a melanocyte antigen such as melanin or tyrosinase;
- (l) autoimmune prostatitis, wherein the antigen or ligand is a prostate antigen;
- (m) any undesired immune response, wherein the antigen or ligand is an activation antigen or other antigen expressed on T effector cells present at the site of the undesired response;
- (n) tissue rejection, wherein the antigen or ligand is the MHC specific to the transplanted tissue; and
- (o) an inflammatory condition, wherein the antigen or ligand is one that is expressed on nonlymphoid cells of the hemopoietic lineage that participate in inflammation.
Most preferred is a Treg cell that is able to act and suppress IBD or ulcerative colitis, and may be specific for an antigen associated with IBD such as carcinoembryonic antigen (CEA) or an antigen of intestinal bacterial flora such as bacterial lipopolysaccharide (LPS) or a component thereof, preferably a Lipid A component.
The Treg may be specific for an activation antigen expressed on T effector cells such as CD69 or CD107a. The Treg may be specific for an antigen expressed on a dendritic cell, macrophage/monocyte, granulocyte or eosinophil present at the inflammation site.
In a preferred embodiment, the above Treg cell is specific for an antigen that is introduced exogenously to a subject to the site or target tissue of the immune or inflammatory response, either before, concomitantly with, or after administration of the Treg cell.
The above Treg cell preferably is one that expresses CD4 or CD8, along with CD25 on its surface and expresses the Foxp3 transcription factor intracellularly. The Foxp3 transcription factor may be expressed in the cell endogenously (i.e., from the cells' own Foxp3 gene); this expression is enhanced by exposure of cells to TGF-β or another cytokine that induces Foxp3 expression and induces a Treg phenotype in T-cells. In another embodiment of the above Treg cell, Foxp3 is expressed from a nucleic acid that has been introduced into the cell exogenously (i.e., transduced) as a recombinant nucleic acid expression construct encoding Foxp3 and regulating its expression. The above Treg may be obtained from a mammalian subject prior to introduction of the chimeric nucleic acid and prior to stimulation that induces Foxp3 expression or prior to transducing the exogenous Foxp3-encoding construct. The chimeric nucleic acid encoding the chimeric receptor and the nucleic acid construct encoding Foxp3 may be co-transduced into the cell. In one embodiment, co-transduction is achieved using a bicistronic vector that includes, in a single vector, a sequence of (i) the chimeric nucleic acid encoding the chimeric receptor and (ii) the nucleic acid construct encoding Foxp3, under the control of a common (or separate) promoter and regulatory sequences.
The above Treg cells may be enriched or purified from a mixed population of lymphocytes or T-cells on the basis of the Treg cells' expression of CD4 (or CD8) and CD25 and/or Foxp3. The cell may be subjected to the following treatment:
- (a) exposure ex vivo of:
- (i) peripheral blood mononuclear cells,
- (ii) peripheral blood lymphocytes,
- (iii) T-cells enriched or purified from (i) or (ii), or
- (iv) a subset of T-cells enriched or purified from (iii); to an amount of TGF-β or other Treg-inducing cytokine or agent that is effective to convert T-cells to a Treg phenotype and to induce expression of Foxp3; and
- (b) optionally, culturing and expanding the exposed cells of step (a).
Preferred Treg cells comprise the above cell that has been transduced with an expression vector encoding Foxp3.
Also provided herein is an immunoregulatory pharmaceutical composition for suppressing a T effector cell-mediated immune/inflammatory response or treating a T effector cell-mediated immune/inflammatory disease or condition, comprising a redirected Treg as described above and a pharmaceutically and immunologically acceptable carrier, excipient or diluent.
This invention is also directed to a method for producing the above redirected Treg that expresses a chimeric receptor as described. This method preferably comprises:
- (a) obtaining from a subject and, optionally, enriching or isolating and propagating, a population of lymphocytes or T-cells;
- (b) inducing the Treg phenotype in these lymphocytes by suitably stimulating or activating the cells by exposure to TGF-β or another cytokine or agent that induces Foxp3 expression and a Treg phenotype;
- (c) before or after step (b), transducing the cells ex vivo with an expression vector encoding the chimeric receptor to be expressed in the Treg; and
- (d) optionally, growing or expanding in vitro the cells obtained as above.
In another embodiment, this method comprises:
- (a) obtaining from a subject and, optionally, enriching or isolating and propagating, a population of lymphocytes or T-cells;
- (b) transducing the cells ex vivo with a vector encoding the chimeric receptor;
- (c) before, after, or concomitantly with step (b), transducing the cells ex vivo with a recombinant nucleic acid expression construct encoding Foxp3; and
- (d) optionally, growing or expanding in vitro the cells obtained as above.
This invention further is directed to a method of suppressing undesired activity of T effector cells in mediating an immune or inflammatory response, comprising delivering to a population of T effector cells to be suppressed (or to a site where such T effector cells are present) an amount/number of redirected Tregs as above, effective to suppress activity of the T effector cells.
Also intended is a method of suppressing undesired activity of T effector cells as indicated above, which method comprises delivering to a population of T effector cells to be suppressed (or to a site where such T effector cells are present) an amount/number of redirected Tregs produced according to the above methods that are effective for suppressing the T effector cell activity.
This delivering is preferably in vivo. The redirected Treg cells are delivered by injection or infusion to a subject in whom the T effector cell activity is to be suppressed, preferably by a route selected from intravenous, intramuscular, subcutaneous, intraperitoneal, intra-articular, intrathecal, intraluminal, intracerebroventricularly, rectal, and topical. In one embodiment, the Treg cells are delivered regionally or locally to a site of inflammation.
The above method is intended for use in situations wherein the T effector cells mediate an autoimmune inflammatory response or disorder, rejection of a transplant or GVH disease.
In one embodiment, the method for treating or ameliorating symptoms of a disease or condition in a subject that is mediated by undesired activity of T effector cells comprises administering to the subject in need thereof an effective amount/number of Treg cells as described above, or a pharmaceutical composition described above, wherein the target recognition domain of the redirected Treg cells is specific for an antigen/ligand present in the subject in the vicinity of the T effector cells so that, upon recognizing and binding the antigen, these redirected Treg cells are activated to secrete suppressive cytokines that suppress the T effector cells in an antigen-nonspecific manner. As noted above, the Treg cell activation occurs in a manner that is not restricted by the MHC and does not require costimulation by a ligand for the costimulatory signaling protein.
Also included is a method for treating or ameliorating symptoms of a disease or condition in a subject that is mediated by undesired activity of T effector cells, the method comprising: (a) producing redirected Treg cells using the production methods described above; (b) administering to the subject in need thereof an effective amount/number of these Treg cells, thereby treating or ameliorating symptoms of the disease or condition.
Stated more generally, the invention is directed to a method for suppressing a T effector cell-mediated immune/inflammatory process in a subject in need thereof, comprising administering to the subject an effective amount/number of redirected Treg cells that express on their surface an antigen-specific chimeric receptor that includes portions that activate Treg cells upon contact with the antigen for which the receptor is specific, the antigen being one that is present in the vicinity of the immune/inflammatory activity. The disease or condition to be treated or ameliorated is preferably: (a) IBD; (b) rheumatoid arthritis; (c) Type I diabetes mellitus or autoimmune insulitis; (d) multiple sclerosis; (e) thyroiditis; (f) gastritis; (g) uveitis or uveoretinitis; (h) orchitis; (i)) oophoritis; (j) psoriasis; (k) prostatitis; (l) encephalomyelitis; (m) vitiligo; (n) rejection of a mismatched cell, tissue or organ graft; or (o) GVH disease.
The present method is used to inhibit the rejection of transplanted cells, tissue, or an organ (alto- or xeno-) that is, for example, mismatched for a major and/or one or more minor histocompatibility antigens. In the case of GVH disease, the recipient generally has received a transplant of allogeneic, semi-allogeneic or non-MHC-mismatched bone marrow cells or enriched or isolated hematopoietic stem cells that are responsible for mediating pathogenic effects.
The present invention is further directed to the novel chimeric DNA that can be used to produce the redirected T-cells described above, as well as to the chimeric receptor protein encoded thereby. Such chimeric DNA comprises:
a first DNA segment encoding an extracellular recognition region specific for a selected target antigen or ligand, which does not comprise an MHC protein extracellular domain, the selected target antigen or ligand being one that is present or expressed at a site or target tissue of a pathogenic or undesired immune response mediated by effector T-cells;
a second DNA segment encoding a transmembrane region; and
a third DNA segment encoding an intracellular signaling region comprising a combination of T-cell signaling polypeptide moieties, which combination of moieties, upon transfection of the chimeric DNA into a regulatory T lymphocyte (Treg cell) and binding of the extracellular recognition region to the selected target antigen or ligand thereof, triggers activation of the Treg cells to cause suppression of T-cell mediated immunity, which chimeric DNA, upon transfection into a Treg cell, expresses the extracellular recognition region, the transmembrane region and the intracellular signaling region in one single, continuous chain on the surface of the transfected cell such that the transfected Treg is triggered to activate and cause suppression of T-cell mediated immunity when the expressed extracellular recognition region binds to its selected target antigen or ligand.
-
- a. TNP-TPCR: extracellular recognition region comprised an scFv of a TNP-specific mAb.
- b. MD2-TPCR: extracellular recognition region comprised an LPS binding fragment or motif of the human MD2 protein, an LPS co-receptor (that interacts with TLR4 receptors). The fragment of MD2 corresponds to residues 120-132 of SEQ ID NO:5. The sequences of such chimeric nucleic acids used here are SEQ ID NO:8 and 9.
- c. CD14-TPCR: extracellular recognition region comprised an LPS binding fragment of the human CD14 protein, a known LPS receptor. The fragment of CD14 corresponds to residues 100-119 of SEQ ID NO:4. The sequences of such chimeric nucleic acids used here are SEQ ID NO:6 and 7.
- d. MD2-CD14-TPCR: the extracellular recognition region comprised both the MD2 fragment and the CD14 fragments described above. The sequences of such chimeric nucleic acids used here are SEQ ID NO:10 and 11.
Each of the constructs encoded as stimulatory and costimulatory moieties, tandemly linked sequences encoding CD28 and FcRγ. Results are shown as side-by-side light and fluorescence micrographs.
“Regulatory T lymphocyte” or “Treg cell” or “Treg,” as used in the present specification and claims are synonymous and are intended to have its standard definition as used in the art. Treg cells are a specialized subpopulation of T cells that act in a “regulatory” way to suppress activation of the immune system and thereby maintain immune system homeostasis and tolerance to self-antigens. Tregs have sometimes been referred to suppressor T-cells. Treg cells are characterized by expression of the forkhead family transcription factor Foxp3 (forkhead box p3). They may also express CD4 or CD8 surface proteins. They usually also express CD25. As used in the present specification and claims, and unless otherwise specified, Tregs include natural Tregs and induced or adaptive Tregs and Tregs that have been created using recombinant DNA technology. Naturally-occurring Treg cells (CD4+CD25+Foxp3+) arise like all other T cells in the thymus. In contrast, adaptive Treg cells (also known as Trl cells or Th3 cells) may originate during a normal immune response. Antigen-specific activation of human effector T-cells leads to inducible expression of Foxp3 in a subgroup of the activated effector cells, and this subgroup can develop a regulatory (Treg) phenotype. One way to induce Tregs is by prolonged exposure of T effector cells to TGF-β. T-cells may also be converted to Treg cells by transfection or transduction of the Foxp3 gene into a mixed population of T-cells. A T-cell that is caused to express Foxp3 adopts the Treg phenotype and such recombinant Tregs are also defined herein as “Tregs”.
“Redirected Treg” is intended to be a comprehensive term for Tregs carrying a chimeric receptor (CR) as described and claimed herein which confers on the cells the ability to bind to and be activated by a target antigen or ligand that is different from that to which a Treg population may have been previously specific (as controlled by its endogenous antigen-specific TCR). Redirected Tregs are “MHC-independent” and “non-MHC restricted” in the process of their activation and in their actions as they do not require association of a peptide derived from their target antigen or ligand with MHC in order to recognize it. However, for special purposes, it may be possible to design a redirected Treg that recognizes a specific epitope of an MHC molecule per se, e.g., functioning as a transplantation antigen. In such a case these redirected Tregs are still non-MHC restricted.
The term “selected target antigen or ligand” means a molecule to which the extracellular recognition region of the redirected Treg is intended to bind so as to activate that Treg. If the selected target is an antigen, then an antibody can be raised against it and the binding regions of such an antibody used to construct the extracellular recognition region of the redirected Treg. If the target molecule is a member of a receptor/ligand pair (defined below), then the other member of that pair can be used as part of the extracellular recognition region of the redirected Treg. Generally, in designing a redirected Treg for use according to this invention, the intended target tissue or site where the Treg is to be employed is first identified, and then, an antigen or ligand that is present on or near this intended target tissue or site is identified. An antibody or ligand/receptor that binds thereto is then identified, or, if necessary, created or constructed for use on the redirected Treg.
The term “ligand” as used herein, and particularly as part of the term “target antigen or ligand” or the term “receptor/ligand pair” refers to a molecule that is able to bind to and form a complex with another biomolecule to serve a biological purpose. Often the binding partner of a ligand is called a receptor so that the two binding partners are termed a “receptor/ligand pair.” For the purpose of the present specification and claims, the term “receptor,” when used in the sense of a “receptor/ligand pair,” has a broader meaning than, for example, a typical definition of a “receptor” as a protein in or on a cell that binds to a specific ligand. It is rather intended to mean any binding partner for a ligand. Either member of a binding pair can be considered the “ligand” while the other member is considered the “receptor.” Thus, a classical receptor may qualify as a “ligand” when used herein in the term “target antigen or ligand” as it is one member of a receptor/ligand binding pair. For example, IL-2 can be a ligand because it binds to and forms a complex with another biomolecule, i.e., an IL-2 receptor (IL-2R), to serve a biological purpose. However, IL-2R is also a “ligand” because it is a molecule that binds to and forms a complex with another biomolecule, i.e., IL-2, to serve a biological purpose. Thus, under the present usage, if IL-2R is considered a ligand in the IL2R/IL-2 binding pair, IL-2 may be considered the receptor, and vice versa.
A “chimeric receptor,” as used in the present specification and claims, is a recombinant polypeptide that includes an extracellular recognition region that is derived from one molecule and at least one intracellular signaling moiety that is derived from a different molecule. In that sense it is chimeric.
A “chimeric nucleic acid” is a recombinant polynucleotide that includes a sequence that encodes a chimeric receptor.
The terms “recombinant” or “recombinantly” when applied to a polynucleotide, polypeptide or cell means that the molecule or cell is made using genetic engineering techniques and would not exist but for the hand of man.
The term “T-cell signaling polypeptide moiety” means that portion of a molecule endogenous to a T-cell that mediates signaling. It may be a portion of a T-cell receptor molecule that mediates signaling, or a downstream signal-transducing enzyme or a portion thereof that mediates signaling, i.e., that has enzymatic activity.
The term “antibody-derived scFv domain” means a single-chain antibody in which the VL of a specific antibody is linked to the VH thereof by a flexible linker or spacer.
The term “an MHC protein extracellular domain” refers to the disclosure of Meal D J et al. (Proc Nall Acad Sci USA 2005; 102:11817-22), discussed below, and Jodi M D et al., Nat. Biotechnol., 2002, 20:1215-1220. These publications describe Treg cells redirected against T-cells in a murine system. They used the class II MHC Is α and Is β chains as extracellular regions of two separate chimeric receptors for use in a redirected Treg. The term MHC protein extracellular domain is defined so as to encompass what was used in the Meal et al. and Jodi et al. publications and any analogs or fractions thereof that would have been obvious to a person of ordinary skill in the art to substitute for such domains for the purpose disclosed by Meal et al. and by Jodi et al., i.e., to cause binding of the redirected Treg to a T lymphocyte specifically directed to a particular autoantigen.
The term “flexible spacer” means any flexible peptide moiety that will facilitate the functionality of the extracellular recognition region. When this region is not rigidly attached to the transmembrane region, but is allowed some degree of flexibility with respect to the cell membrane, the ability of the recognition region to recognize and bind to its target antigen or ligand is facilitated. Small neutral amino acids, such as glycine and serine, confer such flexibility. Examples are Gly4Ser and Gly4Ser3.
The “immunoglobulin superfamily” (Igs) means the large group of cell surface and soluble proteins that are involved in the recognition, binding, or adhesion processes of cells. Molecules are categorized as members of this superfamily based on shared structural features with immunoglobulins (also known as antibodies); they all possess a domain known as an immunoglobulin domain or fold. Members of the Igs include various cell surface antigen receptors, co-receptors and co-stimulatory molecules of the immune system, molecules involved in antigen presentation to lymphocytes, cell adhesion molecules and certain cytokine receptors. They are commonly, though not exclusively, associated with roles in the immune system.
The term “hinge” when referring to a region of a molecule of the Igs means the region between the CH1 and CH2 domains consisting of a small number of amino acids. The hinge is flexible and allows the binding region to move freely relative to the rest of the molecule. At the hinge region are the disulfide bridges which link the two dimers, creating the tetramer structural unit. Examples of such immunoglobulin hinge sequences may be found in U.S. Pat. No. 6,165,476, which is incorporated herein by reference.
The term “antigen-specific receptor of a T-cell” refers to a receptor that is found on a T-cell that is antigen-specific, i.e., naturally has an extracellular region that binds specifically to a particular antigen in preference to another. Examples of such antigen-specific receptors of a T-cell are the TCR α, β, γ or δ chains, the TCRαβ dimer and TCR dimer.
The term “TCR/CD3 complex” is sometimes called the “TCR complex.” CD3 is a protein complex composed of four chains in mammals (CD3γ, CD3δ and two CD38 chains), that associate with molecules known as the T cell receptor (TCR; see above) and with the ζ-chain and η-chain (as homo- or heterodimers) to generate an activation signal in T lymphocytes. The intracellular tails of these CD3 molecules contain a single conserved motif known as an “immunoreceptor tyrosine-based activation motif” or ITAM for short, which is essential for the signaling capacity of the TCR. The CD3-γ, -δ, and -ε chains and the ζ- and η-chains, also known as CD3-ζ and CD3-η chains, together with the TCR, form what is known as the T cell receptor complex.
The term “T-cell costimulatory-receptor protein” means a receptor of the T-cell that provides a costimulatory signal. During the activation of T cells, costimulation is often crucial to the generation of an effective immune response. T cells require two signals to become fully activated. A first, antigen-specific, signal is provided through the T cell receptor/CD3 complex. A second signal, the costimulatory signal, is antigen-nonspecific and is provided by costimulatory molecules expressed on the T cell membrane. Examples of T-cell costimulatory-receptor proteins are CD28, OX40, CD40L, 4-1BB and PD-1.
The present invention is based on the conception that regulatory T-cells (Treg) that have been modified to possess antibody-type antigen specificity, can be harnessed to suppress T effector cells function in vivo. The action of these Treg cells is mediated in an antigen-nonspecific manner, primarily by release of suppressive cytokines in the vicinity where the Tregs are activated or stimulated by an antigen recognized by their TpCR. Once activated, Tregs can suppress bystander T cell responses. Thus, transfer of these cells that have been engineered to express the CRs (as described herein) to a subject in whom it is desired to suppress a T effector cell response and its attendant or consequent inflammation, and their delivery to, and activation at, the site of such inflammatory activity, results in therapeutic effects.
Thus, preferred target diseases or conditions for this invention are autoimmune diseases, more preferably, organ-specific, T cell-mediated autoimmune diseases. Other examples of undesired immune responsiveness to be targeted herein are graft rejection of solid tissue and organ grafts as well as grafts of suspended cells (e.g. bone marrow (BM) transplants or hemopoietic stem cell (HSV) transplant). Another target disease is graft-vs-host disease (GVH) that is a common consequence of a mismatched BM or HSV transplant. An additional condition targeted by this invention is transplant rejection (e.g., of a mismatched kidney) where the recipient's immune effector cells reject the graft.
Because the Foxp3 transcription factor (a member of the forkhead family) appears to be essential for Treg development and function, and is a distinctive marker for these cells (along with CD4 and CD25), the present invention provides methods for producing Tregs, as well as providing the Tregs produced by those methods, that are based on induction of Foxp3 in T-cells in a process of driving cells along the pathway to Treg status.
In another embodiment, DNA encoding Foxp3 is transduced or transfected into T-cells using any suitable expression vector as a delivery vehicle in a process to drive these cells to become Tregs. Further support for this conception is found in reports that prevention of Foxp3 expression in vivo results in animals with a propensity for development of autoimmune and lymphoproliferative disorders (Sakaguchi S, et al., J Immunol 1995; 155:1151-64; Hori S et al., Science 2003; 299:1057-61; Khattri R, et al., supra; Fontenot J D et al., Nat Immunol. 2003; 4:330-6.).
The starting population can be total PBL, T-cells that have been enriched or isolated from PBL, or CD4+T-cells that have been enriched or isolated from such T-cells (either expressing CD25 or not). These cells can be redirected by transducing the TpCR prior to, concomitantly with, or after transducing DNA encoding Foxp3 DNA, preferably in the form of a Foxp3 expression vector. Walker M R. et al., 2005, Proc Natl Acad Sci USA. 102:4103-8, have shown that antigen-specific human CD4+CD25+ Treg cells can be generated de novo from CD4+CD25− cells. The advantage of the present invention over the approach described by Walker et al. is that the antigen-specificity and Treg activation requirements are independent of the MHC. This important improvement makes isolation and activation of antigen-specific Tregs simpler and allows for therapeutic methods (described below) in which the antigen can be conveniently administered together with the transferred Treg cells to a desired site, such as an inflammatory site, exemplified by the colon in IBD.
Naturally-Occurring vs. Inducible Tregs
“Classical” naturally-occurring Tregs are thymus-derived, express high levels of Foxp3 and suppress activation of effector lymphocytes. Antigen-specific activation of human effector T-cells leads to inducible expression of Foxp3 in a subgroup of the activated effector cells, which subgroup can develop a regulatory (Treg) phenotype. These induced regulatory T-cells can suppress (independently of cell contact) freshly isolated effector cells (Walker M R., et al., 2005, supra; Walker M R., et al., 2003, J Clin Invest. 112:1437-43). In mice, both in vitro and in vivo induction of Tregs is achieved by prolonged exposure of effector cells to TGF-β (Wan Y R, et al., 2005, Proc Natl Acad Sci USA. 102:5126-31; Mantini M C, et al., 2004, J Immunol. 72:5149-53; Mantini et al., 2006, supra). This small, peripherally generated population of inducible Tregs are believed to play a central role in regulating and containing ongoing immune responses just as the lack of Treg induction is associated with a propensity for autoimmunity.
Genetic Manipulation of CD4+CD25+T Regulatory CellsAccording to the present invention, approaches that specifically redirect regulatory T-cells to suppress the activity of pathological T-cells are beneficial in inflammatory conditions by facilitating localization of Tregs to inflammatory sites and their specific activation by inflammation-associated antigens. When specifically activated in inflammatory lesions, such Tregs are expected to attenuate inflammatory disease by suppressing pathogenic effector T lymphocytes in an antigen-nonspecific, MHC-unrestricted, manner.
The MHC-independent activation and action of Treg cells according to the present invention is an important advantage. Such action is contrasted with the report of Meal D J et al. (Proc Natl Acad Sci USA 2005; 102:11817-22) of a study of experimental allergic encephalomyelitis (EAE) which described CD4+25+ Treg cells redirected against myelin basic protein (MBP) epitope 89-101-reactive T cells by a CR that included the MBP epitope linked to MHC class II protein. By enforcing the interaction between a Treg cell and the autoreactive T cells directed against MBP epitope 89-101, the Treg activity is antigen-specifically focused against the autoreactive T-cells. Such a model requires some degree of MHC-dependency as a single CR can only have domains of a single MHC and thus can only be used for patients with that HLA characteristic.
In contrast, the Treg cells of the present invention act to suppress pathogenic T effector cells in an MHC-independent manner, making them more advantageous for treating autoimmune/inflammatory conditions because they can target common target antigens shared among many individuals. In the model of Meal et al., supra, Tregs acted by MHC- and antigen-restricted engagement. As such these Tregs, which express the ligand that is recognized by the TCR of the autoreactive T-cells, are stimulated by such interaction and suppress the effector cells. However, as a clinical approach this suffers from the disadvantage that it would require full donor-recipient MHC compatibility in the human population in which MHC (HLA) diversity is substantial. Moreover, such an approach would be limited to suppression of clones that are autoreactive against a single, recognizable peptide epitope in the context of defined MHC-II (HLA-DR). An additional significant advantage of the present invention is that it overcomes the requirement that the pathogenic antigen be known. Indeed, disease-associated antigens in a large number of human autoimmune disorders, including human IBD, are not yet known and may be multiple in number.
The “T-Body” Approach of the Present InventionThe “T-body” approach was designed by one of the present inventors and his colleagues as a novel modality for specific redirection and activation of effector T lymphocytes towards pre-defined targets, mostly those associated with neoplastic processes (e.g., Pin thus JHU et al., J Clin Invest 2004; 114:1774-81) and infectious diseases (Bitton N, et al., Curr Top Microbiol Immunol 2001; 260:271-300). The T-body approach was intended to overcome the relative inaccessibility of antibodies to certain sites (such as solid tumors) and the general ineffectiveness of tumor-infiltrating lymphocytes to combat solid tumors by combining into one effector cell population the properties of the humoral and cellular arms of the immune system (Gross G et al., 1989; supra Eshhar Z, et al., Br J Cancer Suppl, 1990; 10:27-9).
The preferred T-body chimeric receptors comprise a ligand binding portion, preferably (1) a single chain antibody variable region (scFv) directed against a disease-associated antigen, linked to (2) an optional extracellular spacer and a transmembrane region and (3) one or more intracytoplasmic moieties of T cell costimulatory and stimulator/signaling molecules. Such CRs as initially developed enable non-MHC restricted, specific antibody-type recognition, homing and penetration of neoplastic tissues. Within the target tissues, antigen-specific activation of chimeric-receptor bearing T effector cells enabled T cell-mediated destruction of tumor cells either by direct cytotoxicity or by induction of a local inflammatory response.
While the scFv domain is the preferred recognition unit of the present invention, in other embodiments, it may be substituted by another structure that serves as a targeting ligand (or ligand binding partner) that will facilitate bringing the Treg cells expressing the CR to a selected site or a selected antigen. Capon and colleagues have disclosed a number of CRs of this sort, such as one where a ligand binding partner polypeptide is fused at its C-terminus to the N-terminus of an immunoglobulin constant region. See, for example, Roberts M R. et al., Blood 1994; 84:2878-89; Ashkenazi A et al., Int Rev Immunol. 1993; 10:219-27; Chamow S M et al., Int J Cancer Suppl. 1992; 7:69-72. See also U.S. Pat. Nos. 6,710,169; 6,407,221; 6,406,697; 6,319,494; 6,117,655; 6,103,521; 6,077,947; 5,741,899; 5,714,147; 5,514,582; 5,455,165; 5,428,130; 5,359,046; 5,336,603; 5,225,538; and 5,116,964. All of these documents are incorporated by reference in their entirety.
The CR polypeptide of the present invention is characterized broadly as comprising (1) an extracellular portion or domain capable of binding to a ligand (such as a target antigen) in a non-MHC restricted manner, (2) an optional extracellular spacer and a transmembrane domain and (3) a cytoplasmic region (one or more domains) capable of activating an intracellular signaling pathway.
Examples of preferred T cell CRs comprise a first binding domain, a preferred example of which is an extracellular scFv fragment derived from a monoclonal antibody (mAb) specific for a selected antigen. The foregoing domain is fused to a spacer domain (preferably a hinge domain of the Ig family that provides spacing and flexibility), a transmembrane domain, a costimulatory region, for example parts of a CD28 molecule, and a further intracellular signaling moiety for T-cells. Examples of the latter include a TCR/CD3 complex-associated ζ chain or η chain, or an ITAM-containing cytoplasmic region such as the γ chain of an Ig Fc receptor (FcRγ). An ITAM is an “immunoreceptor tyrosine-based activation motif; for reviews, see Humphrey M B et al., Immunol Rev. 2005 December; 208:50-65; Pitcher L A et al., Trends Immunol. 2003; 24:554-60; Isakov N, Receptors Channels. 1998; 5:243-53; Daeron M, Annu Rev Immunol. 1997; 15:203-34; Isakov N, J Leukoc Biol. 1997, 61:6-16; Cambier J C, J Immunol. 1995; 155:3281-5; Flaswinkel H et al., Semin Immunol. 1995; 7:21-7, all of which are incorporated by reference in their entirety. It is also possible to use the intracellular portions of TCR α, β, γ or δ receptor molecules in the CR for this purpose. The signaling moiety of a cytokine receptor may also be present in the chimeric receptor chain for use in the present invention. For example, adding the signaling portion of the IL-2 receptor will cause the Treg cell to further act as if it had been subjected to external IL-2 upon binding of the extracellular targeting domain to the selected target antigen or ligand. Furthermore, adding the signaling moiety of the TGFβ receptor will induce a T effector cell to become a Treg cell and thus this may also be a useful addition to the chimeric receptor chain of the present invention. Such CRs expressed on T-cells are known to be functional and, upon exposure to antigen, promote cytokine production (and, when expressed on appropriate effector cell type in the prior art, promoted lysis of antigen-bearing target cells (Stancovski I, et al., J Immunol 1993; 151:6577-82)).
An early configuration of a scFv-based CR comprised an extra-cellular recognition domain and an intracellular signaling moiety. Full activation of such T-bodies through the CR required either pre-stimulation of the T-body or activation of a costimulatory pathway by exposure to CD08/CD86 (B7)-bearing antigen presenting cells (APCs).
The creation of a tripartite chimeric receptor (TpCR) by one of the present inventors and by others (Pule M A, et al., Mol. Ther. 2005; 12:933-41)), in which the signaling domain of the costimulatory CD28 molecule was added to the cytoplasmic domain of the CR, enabled antigen-mediated activation of both the stimulatory and costimulatory signaling pathways independent of B7-CD28 interactions (Eshhar et al., 2001, supra). This approach facilitates full activation of scFv-expressing lymphocytes, resulting in improved effects (in the case of T effector cells, improved anti-tumor; Pin thus JHU et al., supra).
Another useful intracellular signaling domain for the present invention is all or part of the cytoplasmic domain of a phosphotyrosine kinase (e.g., a molecule of the Syk family) which is fused to the CR. See, for example, Eshhar Z & Fitzer-Attas C J, Adv Drug Deliv Rev. 1998; 31:171-82; Fitzer-Attas C J et al., J Immunol. 1998; 160:145-54.; and Eshhar Z et al., Springer Semin Immunopathol. 1996; 18(2):199-209. Use of such a signaling moiety bypasses membrane-proximal signaling events that are often defective in T-cells of subjects with acute or chronic inflammation or cancer.
Costimulatory Domains/Regions and Signals in the Tripartite Chimeric ReceptorRetroviral-mediated expression of CRs in T-cells in general requires T cell activation which activation is commonly achieved by combined use of anti-CD3 and anti-CD28 antibodies. Such pre-activation was sufficient to prime the T-cells to respond to a signal mediated through the CR upon interaction with the antigen for which the CR is specific—both in vitro and in vivo (e.g., Schwartz R H; Annu Rev Immunol 2003; 21:305-34). A costimulatory signal is advantageous for optimal and sustained T cell function and antigen-driven re-activation, even by targets that often lack ligands for costimulatory molecules.
Antigen stimulation alone of CRs that lack a structure or mechanism for costimulatory signaling is generally inadequate to activate resting or naïve lymphocytes (Brocker T et al., J Exp Med 1995; 181:1653-9). Thus, in the absence of costimulatory signaling by CD28, resting T lymphocytes typically undergo anergy or apoptosis (Boussiotis V A et al., Immunol Rev 1996; 153:5-26). For further discussion of CD28 and its interactions with B7, see also, L. Chen (ed.) The B7-CD28 Family Molecules, Landes Bioscience, 2003, which is incorporated by reference in its entirety.
To overcome these limitations in the CR's used in the present invention, the first (recognition) domain, preferably an scFv domain, is linked through an Ig hinge spacer and transmembrane segments to the intracellular segment of a costimulatory signaling molecule, preferably CD28, and then to an intracellular activation region, such as from the CD3ζ chain or the FcRγ chain. Co-expression of two CRs, each with the same scFv, the first linked to CD3ζ and the second to CD28, was found to provide the requisite stimulatory and costimulatory signals for T cell activation (Beecham E J et al., J Immunother 2000; 23:631-42).
Thus, in one preferred embodiment herein, an extracellular recognition site, preferably an antibody-based recognition site such as an scFv, is linked to a CD28 intracellular domain “in series” and further linked to the intracellular signaling region of the TCR complex ζ chain. Such a construct was 20-fold more potent in stimulating IL-2 production upon exposure to solid phase antigen (compared with transfectants expressing CR's lacking the CD28 domain (Finney H M et al., J Immunol 1998; 161:2791-7)). Intracellularly, this domain in the CR binds the p85 subunit of phosphatidylinositol 3′-kinase.
One of the present inventors designed a novel tripartite CR composed of a scFv recognition moiety fused to the non-ligand binding part of the extracellular domain (ECD) of CD28, the entire transmembrane and intracellular domains of CD28, and the intracellular stimulatory domain of FcRγ (“scFv-CD28-γ”) (Eshhar et al., 2001, supra). Human PBL transduced with a nucleic acid construct encoding this CR were specifically stimulated to produce IL-2. Activation was dependent on CD28 costimulatory activity.
The present inventors' laboratory has generated several lines of Tg mice expressing CRs under control of T cell-specific regulatory sequences. T lymphocytes from unprimed, naïve mice that are Tg for the scFv-CD28-γ TpCR manifested potent responses (proliferation, IL-2 secretion, and rescue from apoptosis) upon stimulation solely by the cognate antigen in immobilized form (Friedmann-Morvinski D et al., supra).
According to the present invention, molecules other than, or in addition to, CD28 are exploited to provide costimulatory signals when included in the present CR configuration. Preferred examples of these are the members of the “inducible co-stimulator” (ICOS) family, including OX40 (CD134), CD40 ligand (CD40L, CD154), PD-1 (“programmed death receptor-1), and 4-1BB (CD137). Each of these ligand/receptor pairs possess distinct functions that differ according to the nature of the stimulus and the “antigenic history” of the T-cells on which they are expressed. For example, CD28 signaling is accompanied by induction of ICOS, which, in turn, co-stimulates CD4+T cell activation. The engagement of OX40 (studied in the context of tumor-specific adoptive immunotherapy) improved survival and anti-metastatic actions of T effector cells by CD4+T helper cell-dependent mechanism (Weinberg A D, Trends Immunol 2002; 23:102-9). Activation of OX40 promotes expression of anti-apoptotic proteins Bcl-XL and Bcl-2 and, accordingly, enhances the survival and hence the number of antigen-specific CD4+ T-cells, resulting in strong antigen-specific CD4+T cell memory. Engagement of 4-1BB (CD137) costimulatory receptor with its ligand, 4-1BBL, increased TCR-induced proliferation, survival, and cytokine production in both CD4+ and CD8+T-cells (Cheuk A T et al., Cancer Gene Ther 2004; 11:215-26). Cell survival was associated with increased expression of the anti-apoptotic genes bcl-XL and bfl-1. In general, the interacting ligand/receptor pair 4-1BB/4-1BBL acts to amplify existing costimulatory signals, particularly those emanating from CD28 (Guinn B A et al., J Immunol 1999; 162:5003-10). Human CD4+T-cells express PD-1 and its ligands, PD-L1 and PD-L2, upon activation. Antibodies to the receptor can be agonists or antagonists of the apoptotic pathway. PD-1 engagement can promote ICOS- or CD28-mediated costimulation. (e.g., Bennett F et al., J Immunol. 2003; 170:711-8.
The activity of costimulatory domains of CD28, ICOS, OX40 (CD134), and 4-1BB (CD137) in CRs is also known in human CD4+ and CD8+T-cells (Finney H M et al., J Immunol 2004; 172:104-13). In that study, the tripartite genes were electroporated into cells to avoid pre-activation of the cells. When CR-bearing T-cells were stimulated by their specific antigen (CD33), cytokine release and cytotoxic activity were dramatically enhanced compared to cells in which the CRs lacked costimulatory signaling structures. Inclusion of the 4-1BB signaling domain as the costimulatory moiety in a TpCR on human T-cells with specificity against the CD19 antigen (anti-CD19-1BB-ζ) led to potent cytotoxicity against CD19-bearing acute lymphoblastic leukemia target cells in vitro (Imai C, et al., 2004; 18:676-84).
While the present invention includes the use of an intracellular domain or part of any of these costimulatory sequences in the CR, it is not certain that signaling evoked by these molecules has practical advantages over use of the CD28 costimulatory sequences alone. So, even though the performance of CD28 appears thus far to be quite satisfactory both in vitro and in vivo, the present invention includes within its scope the use of additional or alternative costimulatory systems to CD28 for generating Treg cells that perform optimally in suppressing T effector cells and treating autoimmune/inflammatory and other conditions as described herein. 4-1BB has been used successfully as an alternative to CD28 in T-bodies. See Zhang et al., J. Immunol., 2007; 179:4910-4918.
Transfer of Redirected Tregs to Recipient SubjectsUse of transferred T-cells in vivo in adoptive therapy requires that transferred cells survive, overcome the host's homeostatic control mechanisms that may serve to hinder the acceptance of these cells, and migrate to (home to, or traffic to) and accumulate or localize at, the desired target site(s).
The immune system utilizes internal stimuli to regulate the total size of lymphocyte pools. The total number of peripheral T lymphocytes remains fairly constant, despite production of new cells, turnover of existing cells, and clonal expansion of antigen-specific cells during an immune response (Jameson S C. Nat Rev Immunol 2002; 2:547-56.). These “internal stimuli,” include cytokines and self-peptide-MHC ligands for the TCR. At least two general mechanisms are believed to be responsible for homeostatic effects of bystander T-cells in limiting proliferation: (1) inhibition by physical T cell-T cell interactions; and/or (2) competition for limited “resources” (e.g., IL-7 and access to APCs with suitable self-MHC ligands). The most prominent cytokines in this process are those that signal through receptors containing a common γ chain, termed collectively “γC cytokines.” These include IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21. Homeostatic control of naïve T cell expansion (examined in vitro) is supported by IL-4, IL-7, IL15 and IL21 through the CD28 transmembrane region, whereas only IL-7 appears to be required in vivo (Jameson, supra).
Lymphodepletion or “lymphoablation” is preferably performed to condition a recipient of the transduced Tregs of the present invention. Any method known in the art may be used, for example, irradiation, treatment with certain antimetabolites such as fludarabine, etc. Such treatments have been used in conjunction with adoptive T cell therapy in other contexts. Lymphodepletion in vivo performed as a precursor to adoptive cell transfer is known to boost antitumor immunotherapeutic activity in mice and in humans (as studied particularly with autologous, tumor-reactive T effector cells; Klebanoff C A et al., Trends Immunol 2005; 26:111-7). In clinical trials, objective response rates of 50% were seen in patients with solid metastatic tumors who had first been subjected to lymphodepletion. The mechanisms that are believed to underlie such effects include: the elimination of cellular cytokine ‘sinks’ for homeostatic γC-cytokines (such as IL-7, IL-15 and possibly IL-21 (which serve to activate and expand effector T-cells)), induction of tumor cell apoptosis and necrosis in conjunction with APC activation, and, most important for the present invention, the impairment of CD4+CD25+ Treg cells that suppress T effector cells.
As noted, treatment with homeostatic cytokines may be used to maintain the Treg populations in the recipient.
The present inventors' group found that activation of T-cells in general and T-bodies in particular (such as that required during the ex vivo manipulations to express the CR with certain vectors) down regulated expression of the chemokine receptor CXCR4, thereby impairing T cell homing in response to the chemokine SDF-1, for example. SDF-1 is a chemoattractant for T-cells that express the CXCR4 (Bleul C C et al., J Exp Med 1996; 184:1101-9; Beider K et al., Blood 2003; 102:1951-8). Using ErbB2-specific human T-bodies; these investigators showed that this homing is an essential step for the T effector cells to act in vivo, measured as inhibition of advanced prostate cancer progression and even cure (Pin thus et al., supra). Based on the foregoing knowledge, according to the present invention, redirected Treg cells must either home/migrate to the desired target sites or be administered to such sites.
Persistence of Responses of TpCR-Bearing T-CellsA key factor for success of adoptively transferred T cell therapy (which thus far has been examined most thoroughly with T effector cells in cancer) is maintenance of the transduced T-cells' (effector) function. In one embodiment of the present invention, it is desired to maintain the function of Tregs that have been administered to perform a suppressive function. In another embodiment, it may be preferred that the Tregs act in shorter “bursts” to curtail a more acute (rather than a chronic) T effector response.
Because lymphocytes found in tumor patients include CD4+CD25+ Treg cells that suppress T effector cells (Wang H Y et al., Immunity 2004; 20:107-18; Curiel T J, et al., Nat Med 2004; 10:942-9), such “endogenous” suppressive activity must be overcome to optimize the action of redirected T effector cells. In the present invention, the objective is the converse: redirected Treg cells are administered to a subject in need thereof to quell or otherwise inhibit immune/inflammatory responses that characterize autoimmune conditions, transplant rejection, etc.
Examples of Clinical Trials Using Redirected T-CellsWhile clinical trials using Tregs in accordance with the present invention have not yet been carried out, a number trials using redirected, CR-bearing T effector cells are described below. Advantage may be taken of various lessons learned in those trials in practicing the present invention.
In a Phase I trial in HIV infected subjects, autologous lymphocytes bearing a CD4-ζ CR were administered (Mitsuyasu R T, et al., 2000, Blood 96:785-93). Out of 24 patients, 11 also received concurrent IL-2 infusions for 5 days. The treatment was well tolerated. In some patients, a transient decrease of the viral load was observed in plasma and rectal mucosa (the tissue reservoir for HIV). All subjects tested negative for replication-competent retrovirus (the delivery vector) for up to 1 year after infusion.
Cell Genesys, Inc. conducted phase I clinical trials in colorectal cancer patients using an anti-TAG72-ζ CR made from the humanized CC49 mAb (Warren R et al., In: 7th International Conference on Gene Therapy of Cancer; 1998).
The group of Junghans tested 24 doses of CR-bearing lymphocytes the antigen-specificity of which was directed to CEA in colorectal patients. Up to 1011 cells/patient were given. The treatment was adequately tolerated (Junghans R et al., Proc Am Assoc Can Res, 2000, 41:543).
Hwu and co-workers at the National Cancer Institute conducted a phase I clinical trial in ovarian cancer patients using T-bodies expressing a CR directed against the MoV18-murine anti-folate-binding protein. Large doses of the modified cells were infused into patients together with controlled administration of IL-2. No adverse side effects were reported. Neutralizing antibodies specific to murine MoV18 mAb determinants were found in the sera of several patients (Kershaw M H et al., Clin Canc Res, 2006; 12:6106-15.
A Phase I clinical trial in renal cell cancer (RCC) employed autologous G250-specific genetically modified T lymphocytes (Lamers C H J et al., Daniel den Hoed Cancer News, 2004, 2:8-10). Infusions of these cells were clinically well-tolerated. After 4-5 infusions, patients began to develop liver enzyme abnormalities, a finding explained by the reactivity of the infused T-cells with G250L expressed on bile duct epithelium, albeit at low levels. Treatment was thus limited to only low doses of CR-expressing T-bodies. The results showed in any case that the redirected T-cells did exert CR-dictated functions in vivo.
Two other Phase I clinical trials have been initiated though their results have not yet been reported to the best of the inventors' knowledge. One Phase I trial treated neuroblastoma patients with PBLs and Epstein Barr virus-specific CTLs, both expressing GD-2 specific chimeric T cell receptors (Brenner M K. World wide web URL clinicaltrials.gov/ct/show/NCT00085930, 2005). The other trial employs genetically modified CD20-specific CD8+CTLs for relapsed follicular lymphoma (Wang J, et al., Mol Ther 2004; 9:577-86
The present inventors recognize that certain events may interfere with the efficacy of the therapy using Treg cells expressing CR's in vivo in humans, for example:
-
- (1) formation of neutralizing anti-idiotypic antibodies directed to an idiotope of the scFv part of the CR that could reduced the life-span or effectiveness of the Tregs;
- (2) the low proportion of engineered cells that eventually reached the targeted sites and;
- (3) the potential damage to healthy tissue that expresses the targeted antigen.
The use of Tregs according to the present invention has a much lower risk of (3). As described herein, direct administration of Tregs to sites of inflammation should overcome the limitation of (1)-(3). Adjustment of dose regimens (number of cells, frequency of administration) using routine clinical considerations are expected to limit the impact of the above limiting factors.
According to the present invention, an effective amount of redirected Treg cells are administered to a subject. Preferred carriers for the Treg cells are phosphate buffer, preferably 0.01-0.1M, more preferably 0.05M, or 0.8% saline. Acceptable diluents or carriers for various routes of administration are well-known.
While individual needs vary, determination of optimal ranges of effective amounts of a given cell type for a particular disease or condition is within the skill of the art. The dosage administered will be dependent upon the age, health, and weight of the recipient, kind of concurrent treatment, if any, frequency of treatment, and the nature of the effect desired. Determination of the effective amounts can readily be made empirically by those of ordinary skill in the art without undue experimentation.
Typical dosages are between about 106 and about 1011 Treg cells per injection or infusion, more preferably, about 107 to about 1010 cells. If an antigen is to be administered with the cells (or separately, but to a site where it is intended to activate the cells), a dose of about 0.01 to 100 mg/kg/body preferably, 0.1 to 50 mg/kg/body wt is preferred.
An effective amount of Treg cells is that needed to induce a measurable change, generally a decrease, in the severity of any measurable symptom of the disease, preferably more than one symptom, and most preferably, would result in cessation of symptoms and cure of the disease or condition. For example, without limiting the invention, the above decrease may be at least about 10%, more preferably at least about 20%, more preferably at least about 30%, even more preferably, at least about 40%, and more preferably, at least about 50%, 60%, 70%, 80%, 90%, 95%, or 99%. It is within the skill of the clinical arts to determine when such therapeutic goals have been achieved, and to adjust the dose or frequency of administration accordingly, or to cease further treatment.
The Treg cells of the invention may be given once, or on multiple occasions, via a single or multiple routes. The cells may be administered daily, or preferably on alternate days, preferably weekly or biweekly. Administration can range over an interval of several days to weeks, or even months or years. The frequency and duration of administration can be determined empirically, or based on the clinical history and experience of the subject.
The cellular compositions of the present invention can be administered by any of a number of means and routes known in the art. Administration is preferably parenteral. Preferred routes include, intravenous, intramuscular, subcutaneous, intraperitoneal, intra-articular, intracerebroventricular, intraluminal (preferably into the lumen of the ileum or colon), rectal or the topical route. Also included is the “intrathecal” route, which is intended to encompass injection, infusion or instillation directly into a cavity or space (thecum) surrounding an organ or body region in which an undesired immune/inflammatory response is occurring. Such spaces include the pleural space, peritoneum, subarachnoid space or dural space, or pericardial space. The generic term for administration into a sheath encasing an organ is termed “intrathecal (see, for example, definition in Dorland's Medical Dictionary 29th Edition, WB Saunders (2000) and Stedman's Medical Dictionary, 27th Edition, Lippincott, Williams & Wilkins (2000)) as meaning “within a sheath.” As used herein, this term is intended to be broader than a more commonly used definition which is limited to intracranial spaces.
The compositions, methods, and products of this invention are applicable to human and veterinary uses. The preferred subject is a human.
Transgenic Mice Expressing TNP-Specific Chimeric ReceptorsSeveral Tg strains of mice that express the TNP-specific TpCR, that were recently produced by the present inventors and their colleagues (Friedmann-Morvinski D, 2005) are described herein. These mice are the source of TNP-specific T effector and T regulatory cells and are used as experimental animals in which the induction of colitis is evaluated using the ‘classical’ reactive hapten, TNBS. As a control for these CR-bearing cells, cells from erbB-2-specific TpCR Tg mice that were produced in the present inventors' laboratory are used as they express a CR specific for an irrelevant antigen.
All mature T-cells and NK cells in these Tg mice express the scFv-CD28-FcRγ construct. Naive Tg T-cells can be fully activated by plastic-immobilized TNP without the need for pre-sensitization. (Friedmann-Morvinski D, et al., supra). Results in the Examples herein show that splenic CD4+CD25+ Tregs isolated from such mice specifically suppress proliferation and cytokine secretion by TNP-specific effector T-cells. Moreover, these Tregs are responsible for the delayed development and attenuation of TNBS-induced colitis in these animals. Of importance is the fact that the level of Tregs in the periphery of the TNP-specific TpCR-expressing strains is higher than in wild-type (WT) mice and that the Tregs do not require pre-activation to exhibit their suppressive activity in vivo. This is believed to result from the cross-reactivity of the SP6 mAb, from which the scFv of the TpCR was derived.
Delivery of DNA Encoding the CR into T-Cells
Genetic modification of human peripheral T-cells is achieved in one embodiment using retroviral vectors (Eshhar Z, et al., 2001, supra). As a non-limiting example, the pBullet vector is used, into which the CR-encoding cDNA (Weijtens M E, et al., 1998) is introduced. A bicistronic expression construct is used in which the TpCR and eGFP cDNA are expressed under control of the LTR. This serves to generate a packaging cell based on PG13 that is being used to pseudotype the retroviral vector with the gibbon ape leukemia virus (GALV). Flow cytometric sorting is done on the basis of eGFP expression, and packaging cells producing high-titer virions are selected to achieve high transduction efficacy. To transduce human lymphocytes from healthy donors, lymphocytes are activated in culture with plate-bound anti-CD3 and anti-CD28 mAbs (or using commercial microbeads coated with these antibodies; e.g. from Invitrogen, Miltenyi Biotec, Inc.,) and are transferred to plates coated with Retronectin™ (fibronectin fragment CH-296) plates together with fresh supernatants taken from the packaging cells. At the end of the process that takes 5-8 days, cells are propagated in the presence of IL-2 and then harvested and used. Following this ex vivo procedure, 45-70% of the cells are positive for CR (and GFP) expression.
Useful additional reagents are anti-idiotypic antibodies against idiotopes of the scFv of the TpCR. These enable direct labeling and visualization of TpCR on cell membranes. Such antibodies against the SP6 scFv (exemplified below) have been made and used by the present inventors.
An annotated nucleotide sequence (SEQ ID NO:1) and amino acid sequence (SEQ ID NO:2) of the TNP-specific TpCR used herein is shown in
A preferred sequence that excludes the scFv above, and that can be linked to any other appropriate ligand binding region, preferably a different scFv specific for another antigen, is that defined by the above sequences beginning at the CD28 region. Thus, a preferred nucleotide coding sequence is nucleotides 2203-2523 of SEQ ID NO:1 and amino acids 260-367 of SEQ ID NO:2. Additional nucleotides comprising a 5′ restriction site, and amino acids “inadvertently” encoded thereby, may also be included in a preferred sequence. Additional coding sequence added at the 3′ end of 2203-2523 of SEQ ID NO:1, or additional amino acids encoded thereby and added to at the C-terminus of 260-367 of SEQ ID NO:2, may be present, provided that they permit the encoded sequence, as expressed on the redirected Tregs, to function as a TpCR in ways described herein. Those skilled in the art of cloning and recombinant DNA technology will understand how to modify theses sequences to achieve the desired objective without undue experimentation.
Expression vectors comprising the foregoing sequences are also used in the present invention, in the production of redirected, TpCR-expressing Tregs.
Generation and Expression of TpCR and Foxp3-GFP Fusion Gene and its ExpressionRedirected T-cells are “converted” to Tregs by causing them to express both Foxp3 transcription factor and the antigen-specific TpCR. Such manipulation permits production of large numbers of Tregs for evaluation and therapeutic use. Successful co-transduction or co-expression is tested by including a Foxp3-GFP fusion gene in the same construct as a TpCR to express both in the same cells. This approach is particularly useful when the starting cell populations are human PBL in which Tregs constitute only about 3-5% of CD4+T-cells. This avoids the complications of another approach, also within the scope of the invention, in which large scale Treg propagation is required for effective transduction with retroviral vectors. Moreover it will simplify the isolation of the Tregs and assessment of their fate in vivo.
In one non-limiting example, messenger RNA (mRNA) for Foxp3 is cloned from purified Tregs using PCR. Foxp3 cDNA is cloned into an eGFP Clontech plasmid to create a Foxp3-GFP fusion protein. The fusion protein is cloned into the pBullet vector containing TpCR inserted after an IRES to create a bicistronic expression vector. Both a Foxp3-GFP single gene retroviral vector and a bicistronic TpCR—IRES−Foxp3-GFP double gene retroviral vector are transduced into isolated CD4+CD25− human peripheral blood T-cells following their activation with anti-CD3 and anti-CD28 antibodies. The resulting cells are tested for expression of the three genes by FACS using (1) antiidiotypic antibodies specific for the scFv idiotype, or anti-hinge region antibodies and (2) intracellular GFP and Foxp3 by staining fixed cells with primary antibodies specific for Foxp3 (Alexis Biochemicals, Lausanne, Switzerland).
In another embodiment, sequential expression protocols are used (first TpCR and then Foxp3-GFP genes) or co-expression protocols. Once the genes are expressed, relatively large number of Tregs can be obtained and separated using cell sorter (FACSaria fluorescence-activated cell sorting (Becton Dickinson, Mountain View, Calif.), sorting for GFP and TpCR co-expressing cells.
The Foxp3 construct may be in the form of a bicistronic vector that includes DNA encoding a reporter molecule such as a fluorescent protein. Suitable reporter molecules are well-known in the art and include fluorescent, chemiluminescent or chromogenic proteins, for example Green fluorescent protein (GFP) or enhanced yellow fluorescent protein (EYFP) or a fluorescent homologue thereof, firefly luciferase protein (encoded by the Luc gene) the enzymes chloramphenicol acetyl-transferase (CAT), or bacterial LacZ, (β-galactosidase) or the thymidine kinase gene (encoded by the HSV1 TK gene. GFP and EYFP are detected by fluorimetry or fluorescence histochemistry; enzymes are detected by use of a chromogenic substrate that is converted into a colored product which can be used in histochemical colorimetric detection of enzymatic activity. Luciferase is measured by activation of luciferin which emits light at a known wavelength. Reporter molecules may be detected in vivo by non-invasive detection techniques such as fluorescence optical imaging (FOI), bioluminescence optical imaging (BOI), cooled charged coupled device (CCD) camera optical imaging (CCOI) and positron emission tomography (PET).
Infection of human CD4+CD25− T lymphocytes with retroviral vectors carrying the Foxp3 gene was shown to convert these cells into ones with a Treg phenotype (Walker et al., 2005, supra; Wan et al., 2005, supra).
Any method for introducing DNA into a cell and expressing it may be used in the present invention, including, but not limited to vectors such as retroviral or lentiviral vectors, electroporation, lipofection, and the like.
The functionality of redirected Tregs can be determined using co-culture tests as described in the Examples. If APCs are to be used in such tests, a preferred source is irradiated monocytes. The antigen is loaded into irradiated human APCs which will present it to T effectors and Tregs. In the case of antigens such as CEA, human colon carcinoma cells stably transfected with the CEA epitope may be used. In such coculture tests, one may detect specific activation of TpCR-bearing Tregs through the TpCR. Treg activation is assessed by examining these cells' action on T effector cell (1) proliferation and (2) cytokine secretion profile, focusing on IL2, IL4, IL10, IFN-γ and TGF-β (using commercial ELISA kits, e.g., Ready-Set Go ELISA kit, Ebioscience CA). It is preferred to assay TGF-β and/or IL-10 as an indication of the cells' Treg phenotype.
In a preferred embodiment, the present invention redirects Tregs to sites of colonic inflammation, by introducing into such cells CRs with antibody-type specificity. In sites of inflammation the redirected Tregs are activated to suppress IBD-associated immune response. Tregs endowed with predefined specificity migrate and home to inflamed sites in the colon where they undergo activation and, as a result, suppress T effector cells that mediate the disease processes.
The present redirected Tregs represent a novel form of the ‘T-bodies’ discussed above and are employed as a novel therapeutic modality in IBD. These T-bodies are T-cells that have been genetically engineered to express TpCR in which an antibody variable region is the recognition unit linked to T-cell costimulatory and stimulatory domains that enable specific activation of these T-cells but in a manner that is MHC independent and not MHC-restricted. Based on previous studies using tumor models described above, these redirected Tregs are tested in murine models of IBD models.
An important aspect of this invention is the inventors' conception that, in the context of treating IBD, the colon-associated antigen(s) to which the T-bodies are redirected and targeted are not necessarily the pathogenic autoantigens recognized by the autoaggressive T effector cells. Thus, this invention can exploit the phenomenon of “bystander” reactivity—where the presence of the relevant antigens at the sites of the inflammatory reactions serve to attract and “hold” or localize the redirected Tregs, permitting them to be activated and to exert their suppressive effects in a paracrine manner—acting on target effector cells in the vicinity irrespective of differences in the T effector cells' and Treg cells' antigen specificity.
CEA and LPS-Colonic Antigens as Targets for Redirected Human Tregs in IBDAdvantage was taken of a hapten-specific IBD model that is based on specificity to the hapten TNP to study the suppressive effects of Tregs. In human disease, other antigens that are expressed in intestinal or colonic tissue either normally or in the relevant disease state are preferred targets. The include carcinoembryonic antigen, CEA, and bacterial floral antigens such as lipopolysaccharide, LPS.
Human IBD is idiopathic to the extent that pathogenic antigen(s) remain unknown. Lack of knowledge of the antigen would appear to be an obstacle to implementing the T-bodies clinically. Nevertheless, according to the present invention, there is no requirement that a pathogenic antigen must also be the target antigen for Treg redirection and activation. Treg activation is indeed antigen-specific and thus depends on TCRs, or in the present Tregs, on antibody-based specificity, associated with costimulation together with the activation/mediated by the intracellular signaling moieties of the present constructs. However, once the Tregs are activated, their suppressive action is antigen-independent, and is carried out by secretion of suppressive cytokines (e.g., TGF-β and IL-10) even after the activating antigen has been eliminated. Thus, inducing colonic Treg activation by any local colon-associated antigen will promote potent Treg activation and proliferation, while the action of these cells in inhibiting local inflammatory processes proceeds independently of antigen. CEA is significantly over-expressed in diseased colon tissue in patients with active ulcerative colitis compared to normal individuals and to patients with quiescent IBD (Smithson J E et al., J Pathol. 1996; 180:146-51; Pavelic Z P et al., Anticancer Res. 1991; 11:1671-5). This enhanced tissue expression of CEA was independent of dysplastic changes and is a result of the mucosal reaction to the inflammatory process itself. Thus, CEA is a preferred candidate for Treg TpCR targeting in active ulcerative colitis.
A second candidate antigen (or “non-antigen” ligand) to which Tregs may be redirected is endotoxin or LPS, derived from the outer membrane of Gram-negative bacteria resident in the colon. In one embodiment, the antibody-like part (scFv) the CR's extracellular recognition region may be derived from an anti-LPS antibody, such as the mAb produced by the hybridoma with ATCC Accession No. HB9081. The nucleotide sequence of an scFv made from this mAb is shown as an annotation in
Several types of non-antibody LPS receptors are known in the art. CD14 (SEQ ID NO:4) is a class of LPS receptor that is a GPI-anchored 356 aa glycoprotein. It contains a 19aa signal peptide, an extracellular domain which contain 11 leucine-rich repeat (LRR) domains, 4 N-glycosylation sites and an unknown number of O-glycosylation sites. At least 2 soluble forms of CD14 have been described, one retains GPI and is released from the cell surface which results in an approximately 48 kDa molecule and the other is released prior to the addition of the GPI anchor resulting in a higher molecular weight (>48 kDa).
While LPS interacts with CD14, CD14 is not capable of initiating a transmembrane activation signal because it is a glycosylphosphatidylinositol (GPI)-anchored protein. Thus, LPS must interact with a transmembrane receptor(s) that is responsible for signal transduction. LPS is recognized by the toll-like receptor TLR4 and MD-2 (SEQ ID NO:5; human), a molecule associated with the extracellular domain of TLR4. CD14 greatly enhances the formation of LPS-TLR4-MD-2 complexes, apparently by LPS loading onto TLR4-MD-2 but not in the interaction itself between LPS and TLR4-MD-2. (Akashi S, et al., J. Exp. Med. 198:1035-42 (2003)).
Interaction of LPS with MD-2 in a TLR4-MD-2 complex triggers an intracellular signal transduction cascade that leads to the production and release of proinflammatory cytokines, particularly TNF-α (Dauphinee S M et al., 2006, Lab. Invest. 86, 9-22). Patients with IBD show increased colon and serum levels of endotoxin, LBP, CD14, and MD-2 (Pastor Rojo O, et al., 2006, Inflamm Bowel Dis., December 19 (epub); Amati L et al., Curr Pharm Des. 2003; 9:1937-45; Cario E et al., J Immunol. 2006; 176:4258-66). This change correlates with disease activity, and proinflammatory cytokine levels return to normal after treatment.
A motif of human MD-2, for example, from amino acids 119-132 (14 residues) of SEQ ID NO: ______ can substitute for MD-2 in MD-2-TLR4 complex binding to the lipid A moiety of LPS, which (Mancek M et al., Biochem Biophys Res Comm 2002; 292: 880-5; Kobayashi M et al., J Immunol. 2006; 176:6211-8).
Thus, in one preferred TpCR of the present invention, the extracellular recognition region comprises, in place of an antibody-like structure (e.g., an scFv), a receptor that binds to a ligand that is not acting as an “antigen.” A preferred ligand in the present invention is LPS. Thus, the extracellular recognition region may comprise any of the following receptor structures:
-
- (a) CD14 (SEQ ID NO:4),
- (b) an LPS-binding motif of CD14, such as residues 100-119 of SEQ ID NO:4,
- (c) full length MD-2 (SEQ ID NO:5),
- (d) an LPS-binding motif of MD-2 (residues 120-132 of SEQ ID NO:5),
- (e) a combination of a CD14 and MD-2 or
- (f) a combination of a CD14-motif and an MD-2 motif (as is encoded by the relevant segment of the chimeric nucleic acid of SEQ ID NO:10.
Any of these constructs, when displayed on a Treg surface, will allow the redirected Treg to bind to, and be activated by LPS molecules, for example, at colon inflammatory sites, and thereby exert their suppressive activities in that vicinity. Again, this is an example of receptor-ligand binding/recognition that is not “antibody-like” but nevertheless permits the TpCR to act in accordance with this invention and activate Tregs in an antigen-nonspecific (and MHC-independent) manner.
The present invention includes an embodiment in which redirected Tregs bearing a TpCR are designed to be specific for an antigen, referred to herein as “AgX,” that may have no inherent relationship with the tissue being targeted or the disease being treated. In this embodiment, the Tregs specific for AgX are activated specifically in a selected site by administering them together with AgX to that site. The site is one where T effector cells are situated and active, where the ongoing inflammation is to be suppressed. The AgX-specific antibody-like receptor of the Tregs will recognize AgX without a need for antigen presentation, MHC, etc., and the linked signaling moieties on the TpCR will serve to activate the Tregs to release inhibitory cytokines at that site. This process will lead to nonspecific suppression of the ongoing T effector cell and inflammatory activity.
The methods and compositions described herein are useful for any of a number of autoimmune diseases which involve undesired effector T-cells activity as an underlying cause or as a consequence of the pathophysiology. Such diseases include, but are not limited to, IBD, rheumatoid arthritis, Type I diabetes, multiple sclerosis, autoimmune thyroiditis, autoimmune uveoretinitis, autoimmune orchitis, autoimmune insulitis, autoimmune oophoritis, psoriasis, autoimmune polymyositis and the like. See, for example, Theofilopoulos, A., In: Stites, D P et al., eds., Basic and Clinical Immunology, Lange Medical Publications, Los Altos, Calif., 1988)).
Having now generally described the invention, the same will be more readily understood through reference to the following examples which are provided by way of illustration, and are not intended to be limiting of the present invention, unless specified.
Example I Materials and MethodsThe following materials and methods are used in various of the Examples that follow, as well as in carrying out certain embodiments of the invention.
Cell Fractionation and IsolationCD4+CD25+ Tregs were purified from splenic lymphocytes or peripheral blood mononuclear cell populations using several methods. One method utilized magnetic bead separation (MACS). Spleens are mashed gently into HBSS/5% FCS to prepare single cell suspensions.
CD4+T-cells were purified by negative selection by incubation with biotin-conjugated CD4 MACS beads (Miltenyi Biotec, Inc., Auburn, Calif.). Further purification of CD4+CD25+ cells was conducted by incubation with phycoerythrin (PE)-conjugated anti-CD25 antibodies or anti-CD45RBhigh, followed by incubation with anti-PE microbeads (Miltenyi Biotec, Inc., Auburn, Calif.). Magnetic separation was conducted using magnetic columns according to manufacturer's instructions. For highly-purified (>99%) Treg and effector T lymphocyte subpopulation, high-speed cell sorting is be applied, using BD FACSaria (®) cell-sorting system (BD Bioscience)
Lamina propria lymphocytes from colon were isolated as previously described (Han X et al., Gastroenterology. 2005; 129:185-203). Briefly, colonic mucosa was dissected, followed by incubation with Ca2+—Mg2+-free Hanks' balanced salt (HBSS) solution containing 1 mM dithiothreitol (Sigma-Aldrich, St. Louis, Mo.) for 30 min to remove mucus, and then serially incubated twice times in medium containing 0.75 mM EDTA (Sigma-Aldrich) for 60 min at each incubation. The supernatants from these incubations containing epithelium and intraepithelial lymphocyte population are discarded, and the residual fragments pooled and treated with 2 mg/mL collagenase A (Worthington Biomedical, Freehold, N.J.) and 0.01% DNase (Worthington) in humidified air at 37° C. for 2 hours. The cells are then be pelleted twice through a 40% isotonic Percoll solution, after which they are purified further by Ficoll-Hypaque density gradient centrifugation (40%/75%).
In Vitro Induction of TregsNaturally-occurring Tregs are thymus derived, express high levels of Foxp3 forkhead transcription factor and suppress activation of effector lymphocytes. It has been discovered that antigen-specific activation of human effector T-cells may induce expression of Foxp3 in a subgroup of the activated effector cells, which in turn develop a regulatory phenotype. These induced regulatory T-cells were shown to be capable of cell-contact-dependent suppression of freshly isolated effector cells (Walker et al., 2003, supra). In mice, prolonged exposure of effector cells to TGF-β induces Tregs both in vitro and in vivo (Fantini et al., J Immunol. 2004 and 2006, supra). This small, peripherally generated population of inducible Tregs may be central in regulation and containment of ongoing immune response, while the inability to induce such Tregs may be responsible for a propensity to develop autoimmunity.
To test whether such induction occurred after stimulation of T effector cells through the TpCR, wildtype, TNP-Tg, Erbb2-Tg and TNP-CD28A-Tg T effector cells were isolated by FACS sorting and cultured for 7 days in the presence of either (1) anti CD3 Ab, (2) murine TGF-β, (3) mAb to TNP, (4) anti CD3 Ab+TGF-β, or (5) anti TNP Ab+TGF-β. Induction of Foxp3 in cells “developing” from these effector T-cells was assessed after seven days of culture using intracellular Foxp3 staining
Antigen-specific activation of human effector T-cells leads to inducible expression of Foxp3 in a subgroup of activated effector cells, which in turn develop regulatory phenotype. These induced regulatory T-cells are capable of cell-contact-dependent suppression of freshly isolated effector cells. In mice, both in vitro and in vivo induction of Tregs can be achieved with prolonged exposure of effector cells to TGF-β (Fantini et al., 2004, 2006, supra). The present inventors adopted this technology to induce murine redirected Tregs from redirected effector T-cells (see
Animals
Several mouse strains were used in the studies described below and are used in various other embodiments of the invention. These include transgenic mouse lines that specifically expresses anti-TNP or anti-Erb B2 TpCRs (bearing CD28-FcR signaling chains) under the control of a CD2 promoter, as well as a transgenic mouse line expressing human CEA (Saha A et al., Immunology 2006, 118:483-496)
All transgenic mice were back-crossed to Balb/c. Balb/c wild-type mice serve routinely as controls and recipients of adoptively transferred cells.
One cell-transfer colitis model is used in immune deficient Rag−/− and SCID mice.
All invasive procedures were and are conducted under Ketamine and Xylazine general aesthesia (127.5 and 4.5 mg/kg, respectively). Subcutaneous (S.C.) injections are conducted under local anesthesia with 10% Xylocalne spray.
Colitis Induction and Assessment:To induce TNP hapten-mediated colitis mice were sensitized with 15 μl of the haptenating agent 2,4,6-trinitrobenzenesulfonic acid (TNBS, Sigma-Aldrich) at a concentration of 2.5% v/v in 50% ethanol by skin painting on day 1. On day 8, 150 μl of 1% TNBS in 50% ethanol was administered intrarectally via a 3.5 F catheter under general anesthesia. OXA-induced colitis was induced by sensitizing mice with oxazolone (4-ethoxymethylene-2-phenyl-2-oxazolin-5-one; Sigma-Aldrich) at a concentration of 3% v/v in 100% ethanol by skin painting on day 1, followed by intrarectal administration of 150 μl at a concentration of 1% v/v in 50% ethanol on day 8.
In one preferred cell transfer colitis models, CD45RBhigh (naïve) T-cells are transferred to immune deficient mice from syngeneic background (Powrie F et al., J Exp Med. 1994; 179:589-600. This model of mucosal inflammation allows separating T effector and Treg cell function within an inflammatory site.
In all models, colitis is assessed following induction using the following parameters: degree of colon ulcerations, intestinal and peritoneal adhesions, wall thickness, and degree of mucosal edema. Each parameter is graded on a scale from 0 (completely normal) to 4 (most severe) by two experienced, blinded observers. For histological evaluation of inflammation, distal colon tissue (last 10 cm) is removed and fixed in 10% formaldehyde. Five paraffin sections from each mouse are stained with hematoxylin-eosin using standard techniques. The degree of inflammation is graded semiquantitatively on microscopic cross sections of the colon from 0 to 4 as follows: Grade 0: Normal with no signs of inflammation; Grade 1: very low level of leukocyte infiltration; Grade 2: Low level of leukocyte infiltration; and Grade 3: High level of infiltration with high vascular density, and bowel wall thickening; Grade 4: Transmural infiltrates with loss of goblet cells, high vascular density, wall thickening, and disruption of normal bowel architecture.
Murine ColonoscopyFor continuous monitoring of colitis pathology, a newly-developed, high resolution mouse video endoscopic system has been used Becker C et al., Gut. 2005; 54:950-4. The experimental endoscopy system (from Karl Storz, Tuttlingen, Germany) consists of a miniature endoscope (1.9 mm outer diameter), a xenon light source, a triple chip camera, and an air pump. Parameters for grading of colitis include bowel wall thickening, granularity, fecal consistency, fibrin deposition and vascular pattern. Whole colon methylene blue chromoendoscopy staining is used, when appropriate, to visualize crypt pattern. A 3fr. Flexible biopsy forceps is used for biopsy-taking. Biopsies are either placed in formalin for paraffin embedding, sectioning and subsequent immunohistochemistry, frozen in liquid nitrogen for cryosections, or obtained and used for RNA isolation. A typical yield of a biopsy specimen is approximately 2 μg RNA
In Vivo Imaging:To follow migration (also referred to as homing or trafficking) of redirected Tregs in mice, a whole body CCD camera (IVIS® 100 Series Imaging System, Xenogen, Alameda Calif.). was used. Redirected Tregs were labeled with the near-infrared (NIR) lipophilic carbocyanine dye 1,1′-dioctadecyl-3,3,3′, 3′-tetramethylindotricarbocyanine iodide (DiR, Invitrogen, USA). This dye has absorption and fluorescence maxima at 750 and 782 nm, respectively, enables the safe direct labeling of membranes of human lymphoid cells with very low light absorption and autofluorescence levels in living tissues (Miller M J et al., Proc Natl Acad Sci USA, 2003; 100:2604-9; Kalchenko V et al., submitted for publication, 2007). Additional in vivo visualization of Tregs labeled with carboxy fluorescein diacetate succinimide ester (CFSE) at colonic mucosa was performed by intrarectal insertion of a 300 and 650 μm diameter confocal microendoscope (Cell Vizio, MKT, Paris, France). This unique modality, previously untested in colitis models, allows repeated in vivo assessment of homing of CFSE-labeled redirected Tregs to the most inner layers of colon tissue following induction of inflammation.
Determination of Colon Cytokine LevelsColon mRNA expression of selected cytokines is determined to allow assessment of redirected Treg effects on local intestinal immune response. in particular, levels of pro-inflammatory (TNFα and IFNγ) and anti-inflammatory cytokines (TGFβ and IL10), as well as levels of the TH1 transcription factor Tbet and the TH2 transcription factor GATA-3. Colon cytokine levels are assessed by measuring mRNA expression and protein levels.
Samples for mRNA isolation are removed from colons of mice using in vivo colonoscopy or during sacrifice. Total RNA is isolated and processed and cDNA produced by RT-PCR. In all experiments, mice are divided into the following groups: naive mice, colitis-induced mice, and colitis-induced mice adoptively transferred with Tregs (naturally occurring, induced, or redirected, see detailed adoptive transfer experiments herein). The following sets of oligonucleotides and amplification conditions are used:
The relative mRNA expression compared to the housekeeping GAPDH is assessed using NIH image software and averaged from mice in each group.
IL-10 and IFN-γ protein expression levels in colon tissue are quantified by a cytofluorimetry-based ELISA system. In brief, whole proteins are isolated from colon specimens in the absence of detergent. Proteins (100 μg) are immediately used for cytokine determination according to manufacturer's instructions.
Foxp3 Immunohistochemistry of Colon Samples:Foxp3 immunofluorescence is performed to estimate in situ the targeting of Treg to diseased colon, using TSA Cy3 and a fluorescence microscope (Olympus). In brief, cryosections are fixed in cold acetone for 10 minutes, followed by sequential incubation with methanol, avidin/biotin (Vector Laboratories, CA), and protein blocking reagent to eliminate nonspecific background staining. Slides are then incubated overnight with primary antibodies specific for Foxp3 (e.g., from Alexis Biochemicals, Lausanne, Switzerland). Subsequently, slides are incubated for 30 minutes at room temperature with biotinylated secondary antibodies, and treated with streptavidin-horseradish peroxidase and stained with Tyramide (Cy3 or FITC). Before examination, nuclei are counterstained with Hoechst 3342 (Molecular Probes, Ohio).
Example I Phenotypic Characterization of TNP-Specific TregsThe inventors have produced transgenic (Tg) mice expressing a TNP-specific tripartite chimeric receptor (TpCR) that serve as a source of redirected Treg cells specific for the trinitrophenyl (TNP) hapten. This hapten has served as a “classical” antigen for years in studying both antibodies and T cell-mediated immunity. A chemically reactive form of this hapten, TNBS, is a contact sensitizing agent that induces and evokes delayed-type hypersensitivity (DTH) responses as well as inducing colitis in animals, as described herein.
Generation of TNP-specific Tregs was achieved by the creation of Tg mice that express TNP-specific TpCR that comprises an scFv from the TNP-specific mAb Sp6 mAb linked to a truncated CD28 molecule which was inserted between the scFv and the cytoplasmic part of the FcRγ chain (abbreviated as y herein (see
For the truncated form of CD28 (TpCR/CD28,
For expression of TpCR in T-cells of Tg mice, a construct comprising an anti-TNP (Sp6-derived scFv-CD28-γ was cloned into a human CD2 promoter/enhancer minigene-based vector. Tg mice were generated at the Weizmann Institute's Department for Veterinary Resources by pronuclear microinjection of (BALB/c×C57BL/6)F1 fertilized eggs derived from hyperovulated donor females. Founder mice were screened by PCR of DNA from tail samples. Several founder strains were obtained that express high level of the TpCR on their cell surfaces. These were backcrossed for more than nine generations to either BALB/c or C57BL/6 mice to obtain MHC-homogeneous mice.
The studies below describe the characterization of various Treg subpopulations in the different strains of TNP-CR transgenic mice, and the expression of TpCR on these Tregs.
Example II Isolation of Trees in which TNP-Specific TpCR are Highly ExpressedTregs were isolated using double magnetic bead separation (Miltenyi Biotech) or by fluorescent cell sorting in which fluorescently labeled CD4+CD25+ cells were sorted using the FACSARIA cell sorting system.
Treg expression of TNP-specific TpCR was assessed by containing cells for Foxp3 (considered the “gold standard” marker of Tregs) and PE-labeled mAb specific for TNP antibody (generated in the inventors' laboratory). Controls included groups stained with the appropriate isotype controls. As is shown in
Peripheral lymphocytes from the spleen as well as gut-associated lymphocytes from the lamina propria of the colon were stained. As shown in
To resolve what may have appeared to be an inconsistency between the highly elevated Foxp3+Treg population in TNP-Tg mice and the modestly elevated CD4+CD25+ Treg population in these mice, effector CD4+CD25− cells were isolated by cell sorting to a level of 99% purity. Isolated cells were stained for Foxp3 (
Induction of TNBS colitis results in further elevation in splenic (
A key prerequisite for the utility of Tregs expressing TNP-specific TpCR in the treatment of autoimmunity is verification of their regulatory activity, namely an ability to suppress T effector cell proliferation in a dose-dependent manner. Also examined was whether such Treg activation occurs as a result of TpCR signaling, and whether it was indeed independent of CD28-B7 interaction. A series of coculture experiments examined Tregs from the different Tg strains, as is outlined below.
Example VI Tregs Bearing the TNP-Specific Chimeric Receptor Specifically Suppressed the Activity of T Effector CellsTo characterize whether TNP-Tg Tregs retained their anergic properties, CD4+CD25+ Treg cells and CD4+CD25− T effector cells from different Tg mouse founders (anti-TNP, anti-Erb-b2 control and wildtype (WT) mice) were purified from bulk splenocytes. 105 cells were incubated in vitro for 24 h, 48 h or 72 hrs (
All effector cell populations showed significantly increased proliferation and IL2 secretion following non-specific stimulation with anti-CD3+ anti-CD28 Abs. Specific stimulation by FyG-TNP resulted in proliferation and IL2 secretion by T effector cells bearing TNP-chimeric receptor, but not by such T-cells from WT or anti-Erb-b2 Tg mice. In contrast, Tregs from wildtype mice, TNP-chimeric receptor Tg mice and Erb-b2 Tg mice retained their anergic properties: they did not undergo measurable proliferation or IL2 secretion when subjected to the non-specific stimulus or specific Ag.
To characterize whether polyclonal activation could trigger the suppressive action of TNP-Tg Tregs, these Tregs were cocultured in 96-well microplates (0.2 ml) with irradiated antigen presenting cells (APCs) and T effector cells (CD4+CD25−) at 1:1 ratios. Cells in these culture were activated either by (1) immobilized antigen “mimic” (anti-CD3+ anti-CD28) or (2) soluble Concanavalin A (ConA). T cell proliferation was measured as Thymidine uptake and IL2 secretion was measured as growth of cells of the IL-2-dependent CTLL-2 cell line (XTT staining).
To study the antigen-specific Treg stimulation through the TpCR, coculture experiments were done in which TNP-loaded APCs provided the Ag presentation (
(1) marked proliferation of TNP-Tg but not of wildtype or ErbB2-Tg effector T-cells in the absence of Tregs; and
(2) activation of TNP-Tg, but not of WT or Erb-b2-Tg Tregs, manifest as suppression of effector cell proliferation by TNP-specific Tregs only.
These results proved the antigen-specific manner of activation and function of TNP specific TpCR Tregs cells in response to the antigen, TNP.
Co-culture of varying ratios of TNP-specific Tregs and TNP-specific effector T-cells (
Studies supporting the existence of the bystander effects were carried out. Colitis was induced in mice as above using OXA as described in Example I. Adoptive transfer of TNP-specific Tregs alone did not protect these animals from colitis. However, in the presence of trace amounts of TNP applied to the colon, animals were protected from this OXA-induced colitis.
Example VIII Suppressive Activity of TpCR-Redirected Tregs is Independent of Costimulatory ReceptorsTo assess the role of costimulatory signaling in the above TpCR-Tg model, coculture experiments as above were performed using as APC's (a) TNP-modified P815 cells, a cell line that does not express B7, or (b) TNP-loaded genetically modified P815 cells stably expressing the B7 gene (
TNBS is a potent inducer of T-cell responses such as DTH/contact sensitization. This reactive hapten also induces autoimmune colitis when applied to the colon of pre-sensitized mice. To determine whether TpCR-bearing Tregs could suppress autoimmunity, the acute TNBS-mediated colitis model was employed. Intra-rectal administration of TNBS leads to its binding to colon proteins, rendering these modified proteins immunogenic so that they elicited a T cell mediated immune response. The suppressive effect of endogenous or exogenously transferred Tregs on autoimmune inflammatory disease was tested in this model. A different hapten, oxazolone (OXA) with similar sensitizing properties and which induces experimental colitis was used as a specificity control in vivo.
Example X Transgenic Mice Whose Entire Treg Population Expresses the Chimeric Anti-TNP Receptor are Resistant to TNBS-Induced ColitisTNP hapten-mediated colitis was induced in Tg and WT mice by first sensitizing the animals with 150 μl of the 2,4,6-trinitrobenzenesulfonic acid (TNBS, Sigma-Aldrich) at a concentration of 2.5% in 50% ethanol painted on the skin on day 1. On day 8, the antigen was administered rectally (150 μl of 1% TNBS in 50% ethanol; high dose colitis). WT mice developed severe colitis within 2-5 days of rectal TNBS administration (
To produce mortality curves the above experiments were repeated with lower doses of TNBS (75 μl of 1% TNBS in 50% ethanol). Similar differences in colitis severity and in mortality were noted (
Evidence of antigen specificity of the protection from hapten-mediated colitis came from studies of OXA-induced colitis. As shown in
Naturally-occurring Tregs are thymus derived, express high levels of Foxp3 and suppress activation of effector lymphocytes. Antigen-specific activation of human effector T-cells may induce expression of Foxp3 in a subgroup of the activated effector cells, which in turn develop a regulatory phenotype. These induced regulatory T-cells were shown to be capable of cell-contact-dependent suppression of freshly isolated effector cells (Walker et al., 2003, supra). In mice, it has been demonstrated that prolonged exposure of effector cells to TGF-β induces Tregs both in vitro and in vivo (Fantini et al., 2004, 2006, supra). This small, peripherally generated population of inducible Tregs may be central in regulation and containment of ongoing immune response, while the inability to induce such Tregs may be responsible for a propensity to develop autoimmunity.
To test whether such induction occurred after stimulation of T effector cells through the TpCR, wildtype, TNP-Tg, Erbb2-Tg and TNP-CD28null-Tg T effector cells were isolated by FACS sorting and cultured for 7 days in the presence of either (1) anti CD3 Ab, (2) murine TGFβ, (3_mAb to TNP, (4) anti-CD3 Ab+TGF-β, or (5) anti-TNP Ab+TGF-β. Induction of Foxp3 in cells “developing” from these effector T-cells was assessed after seven days of culture using intracellular Foxp3 staining (
At time 0 to the time of T effector cell sorting, no Foxp3 staining was noted. A week of stimulation with anti-CD3+TGF-β, but not with TNP+TGF-β, resulted in a 2-fold increase in Foxp3+ cells in wildtype, ErbB2-tg and TNP-CD28null-Tg T effector cells. In contrast, a dramatic 30-fold increase in Foxp3+ cells was observed in TNP-Tg effector cells following exposure to TNP+TGF-β. Interestingly, no Foxp3 induction was noted in TNP-Tg T effector cells following incubation with anti CD3 Ab or with TGF-β, probably due to significant attenuation of CD3 expression in TNP-Tg T-cells (Morvinsky-Friedman et al, in press).
The foregoing results demonstrated that antigen-specific stimulation through the TpCR in the presence of TGF-β, led to induction of Ag-specific Tregs from T effector cells, further contributing to Treg expansion. This change was dependent both on the antigen-specificity of the Ab recognition unit of the TpCR and the intra-cytoplasmic CD28 signaling moiety.
According to the present invention, induction of Tregs in this manner permits the generation of large populations of TpCR-bearing Tregs that can be used in cell-based therapy of autoimmunity.
Example XII Adoptive Transfer of TNP-Tg Tregs to WT Mice with TNBS Colitis Ameliorates Symptoms and Improved SurvivalStudies were conducted to establish that TNP-Tg Tregs are responsible for the resistance of TNP-Tg mice to TNBS colitis and to evaluate their therapeutic capacity in autoimmunity. Wildtype, TNP-Tg and Erb-b2-Tg Tregs were isolated and administered in varying numbers to wildtype mice a day after induction of TNBS colitis. As was previously described, adoptive transfer of large numbers of Tregs of any origin (>2×105) caused nonspecific attenuation of TNBS colitis. This is believed to result from the presence of a sufficiently large population of pre-activated Tregs that can exert their suppressive activity in the absence of antigen stimulation or specificity. In contrast, adoptive transfer of smaller numbers (5×104) of TNP-Tg Tregs but not of wildtype or Erbb2-Tg Tregs, prolonged survival (
Studies were done to garner additional support for the role of TNP-Tg Tregs in attenuating TNBS colitis by showing that these cells indeed localize to inflamed colon tissue. WT and TNP-Tg Tregs were isolated and stained with the fluorescent intracellular dye, carboxyfluorescein diacetate succinimidyl ester (CFSE). Following staining, 106 Tregs were administered intraperitoneally (ip) to control WT mice or to WT mice in which TNBS colitis had been induced 12 hours earlier. Sixteen hours after this treatment, mice were sacrificed and lamina propria lymphocytes isolated from their colons. The protocol used was described above to isolate lamina propria lymphocytes. Thereafter, the cells were examined for the presence of CFSE-positive cells by FACS analysis. As shown in
An important aspect of understanding the role of Tregs for adoptive therapy of autoimmune inflammation of the Treg in diseased organs, where they are expected to exert their suppressive effects. To demonstrate TNP-Tg-Treg localization, WT and TNP-Tg Tregs were labeled with CFSE and transferred to WT mice 24 hours following induction of TNBS colitis. While very small numbers of CFSE-labeled WT Tregs were observed in cell extracted from colonic lamina propria of naïve or TNBS colitis-induced mice, a nine-fold increase in TNP-Tg Tregs was noted (
To study the kinetics of TNP-Tg Treg localization of in the living animal, the IVIS® 100 Series Imaging System was employed (Xenogen, Alameda Calif.). Wildtype and TNP-Tg Tregs, 1.5×106, labeled with the near-infrared lipophilic carbocyanine dye 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (DiR, Invitrogen USA), were administered ip to WT mice with or without TNBS colitis, who were monitored daily with the IVIS whole body CCCD camera (
To determine whether TNP-Tg Tregs reach the inner colonic mucosal layer, the location where most of TNBS-induced mucosal damage takes place, the Cell Vizio confocal microendo-scopy system was employed (Cell Vizio, MKT, Paris, France). An intrarectally-inserted 650 μm diameter confocal microendoscope enabled visualization of CFSE-labeled cells in up to a 150 μm bowel wall thickness (
In contrast to hapten-mediated colitis, in which the eliciting antigen is predefined, the disease-causing antigen in inflammatory bowel disease (IBD) is unknown. To enable application of the ‘T-body’ approach in IBD, naïve TPCR-bearing Tregs were tested to determine whether they can be activated by a predetermined ‘bystander’ colon- or colitis-associated antigen, to perform their antigen-nonspecific suppressive action. To this end, WT and TNP-Tg mice were pre-sensitized to oxazolone only. A mixture of oxazolone and low doses of TNBS were introduced intrarectally. As is shown in
In utter contrast, TNP-Tg mice administered TNBS+oxazolone featured normal-appearing colonic mucosa with scattered areas of mild colitis (
Notably, this “bystander” protective effect also occurred when TNP-Tg Tregs were adoptively transferred to oxazolone-presensitized wild-type mice which were intrarectally boosted with a mixture of oxazolone and low doses of TNBS (
An experiment was conducted to verify that the Foxp3 can be expressed following transduction of A273 cells with retroviral vector constructs designed to transduce Treg cells. A fused gene was generated that included eGFP sequences encoding green fluorescent protein (referred to as eGFP or GFP). This was engineered as a bicistronic construct with the GFP sequence alone or linked with a Foxp3-encoding sequence (after an IRES) into vectors that comprised a chimeric receptor construct with the following extracellular recognition regions: See description of
The results are shown in
All the control group expressed the eGFP in their cytoplasm only. In contrast, in cells that were transduced with the eGfP-Foxp3-fusion constructs, the nuclei were fluorescent (appearing as bright nuclear images) due to the transport to and expression of the Foxp3 transcription factor in the nuclei.
In another experiment not shown here, expression of chimeric receptor made of the full length MD2 protein (SEQ ID NO:5) or the CD14 protein (SEQ ID NO:4) was confirmed by the ability of transduced cells, which expressed the extracellular region of the CR on their surface, to bind the ligand of MD2 and CD14, bacterial LPS, which was provided in biotinylated form and revealed by secondary binding of fluorescent avidin.
Example XVII Vectors Comprising Chimeric Receptors with LPS-Binding Extracellular RegionsNucleic acid constructs and vectors that encode extracellular regions that comprise an anti-LPS antibody domain (e.g., SEQ ID NO:3 or the scFv-coding portion thereof) have been made and others can be made. Such vectors express extracellular polypeptide domains that are shown to bind LPS, for example in an assay using biotinylated LPS and detectably labeled (e.g., fluorescently labeled) avidin. See also Example XVI above.
Nucleic acid constructs and vectors that encode extracellular regions that comprise a LPS-binding nonantibody polypeptide have been made (e.g., SEQ ID NO:6-11, 13 and 14). Such constructs include bicistronic ones that also comprise Foxp3. Other such constructs can be made using the method described above along with methods well-known in the art. Such constructs (such as SEQ ID NO: 13 and 14) include those encoding full length CD14 (SEQ ID NO:4) or MD2 (SEQ ID NO:5) protein, and constructs encoding LPS-binding motifs therefrom (such as SEQ ID NO:6-9) and combinations (such as SEQ ID NO:10 and 11). The constructs that are made include those with CD28-FcRγ intracellular stimulatory/costimulatory regions and those that utilize others of the type disclosed herein.
Treg cells are redirected as described herein using the above constructs, including those that have been made and tested and those that can be made.
Such Treg cells are administered into subjects suffering from IBD, such as ulcerative colitis. Treg cells are administered in numbers in accordance with the above examples, or in numbers that are readily determined to be effective by those skilled in the art using only routine experimentation, and via routes of administration as exemplified above and disclosed throughout this document. These redirected Treg cells that express an LPS binding antibody region or another LPS-binding moiety on their surface (CD14, MD2, fragments thereof, or combinations of these) as part of their CR's are able to reduce the symptoms, intensity, severity and duration of the IBD in the subject to a significant degree compared to untreated control subjects or control subjects administered with control Tregs. (Such control Treg cells are ones not transduced to express the present CR's, or those redirected to be specific to antigens or ligands not present at the site of the IBD.) Introduction of LPS-related molecules or epitopes that bind to these same extracellular receptors on the redirected Treg cells to the sites of administration (and/or expected action) of the redirected Treg cells further facilitates their therapeutic activity.
The references cited above are all incorporated by reference herein, whether specifically incorporated or not.
Having now fully described this invention, it will be appreciated by those skilled in the art that the same can be performed within a wide range of equivalent parameters, concentrations, and conditions without departing from the spirit and scope of the invention and without undue experimentation.
Claims
1. A redirected regulatory T lymphocyte (Treg cell) endowed with specificity toward a selected target antigen or ligand, which cell comprises a chimeric nucleic acid that encodes a chimeric receptor polypeptide that comprises, expressed in a single, continuous chain, an extracellular recognition region, a transmembrane region and an intracellular signaling region, and is expressed in the redirected Treg cell so as to display the extracellular region on the cell surface, wherein
- (a) said extracellular recognition region of said chimeric receptor is specific for the selected target antigen or ligand; and
- (b) said intracellular signaling region comprises a combination of T-cell signaling polypeptide moieties, which combination of moieties, upon binding of the extracellular recognition region to the selected target antigen or ligand, triggers activation of the redirected Treg cells to cause suppression of T-cell mediated immunity.
2. A redirected Treg cell in accordance with claim 1, wherein said extracellular recognition region comprises an antibody-derived scFv domain that is specific for a selected antigen.
3. A redirected Treg cell in accordance with claim 1, wherein said extracellular recognition region comprises a member of a ligand-receptor pair, which is specific for the other member of that pair.
4. A redirected Treg cell in accordance with claim 1, wherein said extracellular recognition region does not comprise an MHC protein extracellular domain.
5. A redirected Treg cell in accordance with claim 1, wherein said extracellular recognition region is linked to said transmembrane region through a flexible spacer.
6. A redirected Treg cell in accordance with claim 5, wherein said flexible spacer is a hinge from a molecule of the immunoglobulin superfamily.
7. A redirected Treg cell in accordance with claim 1, wherein said intracellular region comprises a signaling moiety from (i) a polypeptide chain of an antigen-specific receptor of a T-cell and/or (ii) a polypeptide chain of a receptor of a T-cell that has a region comprising an immunoreceptor tyrosine-based activation motif (ITAM).
8. A redirected Treg cell in accordance with claim 7, wherein said intracellular region comprises a signaling moiety from (ii).
9. A redirected Treg cell in accordance with claim 7, wherein said intracellular region comprises a signaling moiety from a polypeptide chain selected from the group consisting of a chain of the TCR/CD3 complex, and the γ chain of an Ig Fc receptor (FcRγ).
10. A redirected Treg cell in accordance with claim 9, wherein said polypeptide chain of an antigen-specific receptor of a T-cell is the CD3-ζ chain or the FcRγ chain.
11. A redirected Treg cell in accordance with claim 1, wherein said intracellular region includes a signaling moiety of a T cell costimulatory-receptor protein.
12. A redirected Treg cell in accordance with claim 11, wherein said costimulatory-receptor protein is at least one protein selected from the group consisting of CD28, OX40, CD40L, 4-1BB and PD-1.
13. A redirected Treg cell in accordance with claim 11, wherein said costimulatory-receptor protein is CD28.
14. A redirected Treg cell in accordance with claim 11, wherein said costimulatory-receptor protein is 4-1BB.
15. A redirected Treg cell in accordance with claim 1, wherein said intracellular region comprises at least two different ones of said costimulatory-receptor protein signaling moieties.
16. A redirected Treg cell in accordance with claim 15, wherein said intracellular region comprises an intracellular signaling moiety from CD28 and from 4-1BB.
17. A redirected Treg cell in accordance with claim 1, wherein said intracellular region comprises a signaling moiety from a T cell cytokine receptor.
18. A redirected Treg cell in accordance with claim 17, wherein said cytokine receptor is the IL-2 receptor.
19. A redirected Treg cell in accordance with claim 17, wherein said cytokine receptor is the TGF-β receptor.
20. A redirected Treg cell in accordance with claim 1, wherein said intracellular signaling region includes a signal-transducing enzyme that (a) is an enzyme in the signal transduction pathway of an antigen-specific receptor of a T-cell or (b) is an enzyme with corresponding specificity and activity as the enzyme of (a), derived from a non-T-cell lymphocyte.
21. A redirected Treg cell in accordance with claim 20, wherein said signal-transducing enzyme is a member of the Syk-kinase family.
22. A redirected Treg cell in accordance with claim 1, wherein said chimeric nucleic acid further includes a nucleotide sequence that will cause the redirected Treg to express Foxp3.
23. A redirected Treg cell in accordance with claim 1, wherein said selected target antigen or ligand is one that is present or expressed at a site or target tissue of an autoimmune or inflammatory response mediated by effector T-cells.
24. A redirected Treg cell in accordance with claim 23, wherein said autoimmune or inflammatory response comprises an autoimmune response or disease, an allograft or xenograft rejection, or graft-vs. host (GVH) disease.
25. A redirected Treg cell in accordance with claim 1, wherein said target antigen or ligand is an autoantigen or an antigen that is cross-reactive with an autoantigen.
26. A redirected Treg cell in accordance with claim 25, wherein the autoantigen is a pathogenic antigen in the pathophysiology of said autoimmune disease.
27. The redirected Treg cell of claim 1, wherein said autoimmune or inflammatory response and said target antigen or ligand are selected from the group consisting of:
- (a) inflammatory bowel disease (IBD), wherein said antigen or ligand is one that is expressed in diseased colon or ileum;
- (b) rheumatoid arthritis, wherein said antigen or ligand is an epitope of collagen or an antigen present in joints;
- (c) Type I diabetes mellitus or autoimmune insulitis, wherein said antigen or ligand is a pancreatic β cell antigen;
- (d) multiple sclerosis, wherein said antigen or ligand is a myelin basic protein antigen, MOG-1, MOG-2 or another neuronal antigen;
- (e) autoimmune thyroiditis, wherein said antigen or ligand is a thyroid antigen;
- (f) autoimmune gastritis, wherein said antigen or ligand is a gastric antigen;
- (g) autoimmune uveitis or uveoretinitis, wherein said antigen or ligand is S-antigen or another uveal or retinal antigen;
- (h) autoimmune orchitis, wherein said antigen or ligand is a testicular antigen;
- (i)) autoimmune oophoritis, wherein said antigen or ligand is an ovarian antigen;
- (j) psoriasis; wherein said antigen or ligand is a keratinocyte antigen or another dermal or epidermal antigen;
- (k) vitiligo, where said antigen or ligand is a melanocyte antigen;
- (l) autoimmune prostatitis, wherein said antigen or ligand is a prostate antigen;
- (m) any undesired immune response, wherein said antigen or ligand is an activation antigen expressed on T effector cells present at the site of the undesired immune response;
- (n) tissue rejection, wherein said antigen or ligand is the MHC molecule having the haplotype of the transplanted tissue, or a portion of that MHC molecule; and
- (o) an inflammatory condition, wherein said antigen or ligand is one that is expressed on nonlymphoid cells of the hemopoietic lineage that participate in inflammation.
28. A redirected Treg cell in accordance with claim 27, wherein said autoimmune disease is IBD and said antigen or ligand is carcinoembryonic antigen (CEA), or an antigen of intestinal bacterial flora.
29. A redirected Treg cell in accordance with claim 27, wherein the activation antigen of (m) expressed on T effector cells is CD69 or CD107a.
30. A redirected Treg cell in accordance with claim 27, wherein said target antigen or ligand of (o) is one that is expressed on a dendritic cell, macrophage/monocyte, granulocyte or eosinophil present at said inflammation site.
31. An immunoregulatory pharmaceutical composition for suppressing a T effector cell-mediated immune/inflammatory response or treating a T effector cell-mediated immune/inflammatory disease or condition, comprising:
- (a) a Treg cell according to claim 1, and
- (b) a pharmaceutically and immunologically acceptable carrier, excipient or diluent.
32. A method of producing the redirected Treg cell of claim 1, comprising:
- (a) obtaining from a subject and, optionally enriching or isolating and propagating, a population of lymphocytes or T-cells;
- (b) inducing a Treg cell phenotype in said cells by suitably stimulating or activating the cells by exposure to TGF-β or another cytokine that induces Foxp3 expression thereby inducing the Treg phenotype;
- (c) before or after step (b), transducing the cells ex vivo with an expression vector encoding said chimeric receptor to be expressed on said Treg cell; and
- (d) optionally, growing or expanding the cells obtained as above in vitro.
33. A method of producing the redirected Treg cell of claim 1, comprising:
- (a) obtaining from a subject and, optionally enriching or isolating and propagating, a population of lymphocytes or T-cells;
- (b) transducing the cells ex vivo with a vector encoding the chimeric receptor;
- (c) before, after, or concomitantly with step (b), transducing the cells ex vivo with a recombinant nucleic acid expression construct encoding Foxp3; and
- (d) optionally, growing or expanding the cells obtained as above in vitro.
34. A method of suppressing undesired activity of T effector cells in mediating an immune or inflammatory response, comprising delivering to a site of T effector cells to be suppressed an amount of Treg cells according to claim 1 effective to suppress said T effector cell activity.
35. The method of claim 34 wherein said Treg cells are administered systemically.
36. The method of claim 34 wherein the redirected Treg cells are administered regionally or locally to a site of inflammation.
37. The method of claim 34 wherein said undesired activity of T effector cells is an autoimmune inflammatory response or disorder, rejection of a transplant or GVH disease.
38. A method for treating or ameliorating symptoms of a disease or condition in a mammalian subject that is mediated by undesired activity of T effector cells, said method comprising administering to said subject in need thereof an effective amount of Treg cells in accordance with claim 1, wherein said target antigen or ligand is one that is present at the site of said undesired T effector cell activity.
39. The method of claim 38 wherein the redirected Treg cells are produced by obtaining a population of T-cells from the mammalian subject to be treated, transfecting said cells with said chimeric nucleic acid and causing the cells to express Foxp3 by
- (i) stimulating the transfected cells to induce Foxp3 expression, or
- (ii) introducing an exogenous Foxp3-encoding construct into the transfected cells.
40. The method of claim 38 wherein the redirected Treg cells are produced by obtaining a mixed population of T-cells from the mammalian subject to be treated, enriching or purifying Treg cells from the mixed population of T-cells on the basis of the Treg cells' expression of CD4 and CD25 and/or Foxp3, and transfecting said enriched or purified Treg cells with said chimeric nucleic acid.
41. The method of claim 40 wherein the redirected Treg cells are produced by:
- (a) obtaining (i) peripheral blood mononuclear cells, (ii) peripheral blood lymphocytes, (iii) T-cells enriched or purified from (i) or (ii), or (iv) a subset of T-cells enriched or purified from (iii);
- (b) exposing the cells obtained in (a), ex vivo, to an amount of TGF-β or other Treg-inducing cytokine or agent that is effective to induce expression of Foxp3 and convert T-cells to a Treg phenotype,
- (c) optionally, culturing and expanding said exposed cells of (a); and
- (d) before after or between said steps (a) and (b), transfecting said cells with said chimeric nucleic acid.
42. The method of claim 38 wherein said disease or condition comprises an autoimmune response or disease, an allograft or xenograft rejection, or graft-vs. host (GVH) disease.
43. The method of claim 38, wherein said disease or condition is IBD and said target antigen or ligand is CEA, or an antigen of intestinal bacterial flora.
44. A method of claim 38, wherein said target antigen or ligand is LPS.
45. A method of claim 44, wherein said extracellular recognition region comprises an LPS-binding domain of MD-2 or an LPS-binding domain of CD-14 or both an LPS-binding domain of MD-2 and an LPS binding domain of CD-14.
46. The method of claim 38, further including the step of, either before, concomitantly with, or after administration of the redirected Treg cells, introducing to a subject to the site or target tissue of said immune or inflammatory response, an exogenous antigen or ligand and said target antigen or ligand is said exogenous antigen or ligand.
47. A chimeric DNA molecule comprising:
- (a) a first DNA segment comprising a sequence encoding an extracellular recognition region specific for a selected target antigen or ligand, which region does not comprise an MHC protein extracellular domain, said selected target antigen or ligand being one that is present or expressed at a site or tissue of a pathogenic or undesired immune response mediated by effector T-cells;
- (b) a second DNA segment comprising a sequence encoding a transmembrane region; and
- (c) a third DNA segment comprising a sequence encoding an intracellular signaling region comprising a combination of T-cell signaling polypeptide moieties, which combination of moieties trigger activation of the Treg cells to cause suppression of T-cell mediated immunity when the chimeric DNA is transfected into a Treg cell and said extracellular recognition region binds to the selected target antigen or ligand,
- wherein, upon transfection of the chimeric DNA into a Treg cell, the Treg cell expresses said extracellular recognition region on its surface, said transmembrane region and said intracellular signaling region in one single, continuous chain.
48. A chimeric receptor polypeptide encoded by the chimeric DNA molecule of claim 47.
Type: Application
Filed: Jan 31, 2008
Publication Date: Jun 3, 2010
Applicant: YEDA RESEARCH AND DEVELOPMENT CO. LTD. (Rehovot)
Inventors: Zelig Eshhar (Rehovot), Eran Elinav (Jerusalem)
Application Number: 12/525,270
International Classification: A61K 35/12 (20060101); C12N 5/10 (20060101); C07H 21/04 (20060101); C07K 16/00 (20060101); A61P 29/00 (20060101); A61P 37/00 (20060101);
