Patent Application
20070172504
The present invention generally relates to methods for in vivo engineering of cell surfaces by displaying one or more exogenous immunostimulatory molecules thereon, for the purposes of stimulating an immune response against the cells. Compositions for effecting the engineering also are provided.
Approaches directed to boosting a host's immune response to address diseases and conditions characterized by a deficiency in immunity or resolvable by a more aggressive immune response have been extensively described. An exemplary such disease or condition is cancer. Tumors are targeted by the immune system because they express tumor associated antigens which are either mutated or over/aberrantly expressed self-proteins, or proteins derived from oncogenic viruses. Onset of cancer is believed to result from a deficiency in the immune surveillance role of adaptive immunity mediated by T cells.
Approaching cancer treatment with surgery, chemotherapy and radiotherapy is commonly used but these approaches are appreciated to lack tumor specificity, resulting in adverse side effects and less than satisfactory clinical responses. Accordingly, methods of boosting the immune response to cancer by specifically directing that response to the target cancerous cells without significant, detrimental effects on normal cells would offer distinct advantages over traditional cancer therapy.
Cancer vaccines which include antigens from the cancer have attracted particular interest because of the promise of specificity, safety, efficacy and the long-term memory of the immune system that may prevent recurrence of the cancer. However, this approach may be undermined if the cancer cells no longer express the particular antigen and/or if the tumor is able to suppress or tolerize the immune response.
Vaccines based on the genetic manipulation of autologous tumor cells to express immunological molecules of interest represent an attractive therapeutic modality for cancer since tumor cells can potentially express any tumor associated antigen and, as such, can activate a wide-variety of immune responses for effective therapy. For example, genetically modified tumor cells expressing CD80 have been reported to induce vigorous anti-tumor immune responses with potent therapeutic efficacy. Ramarathinam et al., 1994, J. Exp. Med., 179:1205-14; Townsend & Allison, Science, 259:368-70; see also, Lane, 1997, Ann. N.Y. Acad. Sci., 815:392-400; and Van Gool et al., 1996, Immunol. Rev., 153:47-83. Genetic vaccines are believed to be less susceptible to various immunoevasive mechanisms, such as antigenic shift and tolerance to immunodominant epitopes, which can develop during the course of tumor progression. Autologous tumor cell-based vaccines proposed to date generally require surgical removal of tumors, ex vivo genetic manipulation of the tumors to express immunological molecules of interest, and reintroduction of the manipulated cells into patients by vaccination. Several hurdles are associated with this approach, such as the need for a considerable tumor mass to prepare the vaccine, difficulties associated with the genetic manipulation of primary tumor cells, and the very long amount of time (weeks to months) required for vaccine preparation for gene therapy based approaches.
The present invention generally relates to methods for effecting in vivo engineering of cell surfaces, such as tumor cell surfaces, to compositions for effecting the in vivo engineering, and to engineered cell surfaces.
In accordance with one embodiment, the invention provides a method of engineering the surface of a cell to display an immune co-stimulatory polypeptide. The method comprises contacting a cell surface with (a) a bifunctional molecule comprising a first member of a binding pair and a molecule that binds to the surface of a cell and (b) a first immune co-stimulatory moiety comprising an immune co-stimulatory polypeptide and a second member of the binding pair. In accordance with the method, the immune co-stimulatory polypeptide is displayed on the cell surface via binding between the first and second members of the binding pair. In accordance with one embodiment, the method is effected in vitro. In accordance with another embodiment, the method is effected in vivo.
In one embodiment, the cell surface is a tumor cell surface. In another embodiment, the immune co-stimulatory polypeptide is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL.
In yet another embodiment, the first member of the binding pair comprises biotin and the second member of the binding pair comprises avidin or streptavidin, such as core streptavidin. In one embodiment, the bifunctional molecule comprises Sulfo-NHS-LC-biotin.
In accordance with another embodiment, the invention provides a chimeric protein comprising an immune co-stimulatory and a member of a binding pair, such as avidin or streptavidin, such as core streptavidin. In one embodiment, the immune co-stimulatory polypeptide is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL. The invention also provides a pharmaceutical composition comprising the chimeric protein and a pharmaceutically acceptable carrier.
In accordance with another embodiment, the invention provides a method of engineering the surface of a cell to display one or more immune co-stimulatory polypeptides. The method comprises administering to an individual (a) a bifunctional molecule comprising a first member of a binding pair and a molecule that binds to the surface of a cell; (b) a first immune co-stimulatory moiety comprising a first immune co-stimulatory polypeptide and a second member of the binding pair; and, optionally, (c) a second immune co-stimulatory moiety comprising a second immune co-stimulatory polypeptide and a second member of the binding pair. In accordance with the method, the first and optional second immune co-stimulatory polypeptides are displayed on the cell surface via binding between the first and second members of the binding pair.
In one embodiment, the cell surface is a tumor cell surface. In another embodiment, at least one of the immune co-stimulatory polypeptides is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL. In another embodiment, the first and second immune co-stimulatory polypeptides are selected from the group consisting CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L.
In yet another embodiment, the first member of the binding pair comprises biotin and the second member of the binding pair comprises avidin or streptavidin, such as core streptavidin. In one embodiment, the bifunctional molecule comprises Sulfo-NHS-LC-biotin.
In one embodiment, the first member of the binding pair is administered via intratumoral injection. In another embodiment, at least one of the first and second immune co-stimulatory moieties are administered via intratumoral injection, or by the same or different route as the first member of the binding pair.
In another embodiment at least about 5% of cells initially positive for administered first or second immune co-stimulatory moiety remain positive at about 2 days post-injection.
Another aspect of the invention provides a composition comprising a first fusion protein comprising a first immune co-stimulatory polypeptide and a member of a binding pair and a second fusion protein comprising a second immune co-stimulatory polypeptide and the same member of a binding pair. In one embodiment, at least one of the immune co-stimulatory polypeptides is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL. In another embodiment, the first and second immune co-stimulatory polypeptides are selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L.
Another aspect of the invention provides a combination comprising a first composition comprising a first fusion protein comprising a first immune co-stimulatory polypeptide and a member of a binding pair and a second composition comprising a second fusion protein comprising a second immune co-stimulatory polypeptide and the same member of a binding pair. In one embodiment, at least one of the immune co-stimulatory polypeptides is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL. In another embodiment, the first and second immune co-stimulatory polypeptides are selected from the group consisting CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L.
In accordance with another embodiment, the invention provides a method of reducing the size of a tumor or inhibiting tumor growth, comprising administering to an individual (a) a bifunctional molecule comprising a first member of a binding pair and a molecule that binds to the surface of a tumor cell of the tumor; (b) a first immune co-stimulatory moiety comprising a first immune co-stimulatory polypeptide and a second member of the binding pair; and, optionally, (c) a second immune co-stimulatory moiety comprising a second immune co-stimulatory polypeptide and a second member of the binding pair. In accordance with this embodiment, the first and optional second immune co-stimulatory polypeptides are displayed on the tumor cell surface via binding between the first and second members of the binding pair.
In one embodiment, at least one of the immune co-stimulatory polypeptides is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL. In another embodiment, the first and second immune co-stimulatory polypeptides are selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD3OL, LIGHT, BAFF, APRIL, CD80 and CD40L.
In accordance with one embodiment, the first member of the binding pair comprises biotin and the second member of the binding pair comprises avidin or streptavidin, such as core streptavidin. In one embodiment, the bifunctional molecule comprises Sulfo-NHS-LC-biotin.
The present invention provides methods and compositions for the in vivo engineering of cell surfaces, such as tumor cell surfaces, with one or more immune co-stimulatory polypeptides. The methods, compositions and engineered cells are useful, for example, to stimulate an immune response against the cells. When the engineered cell surfaces are tumor cell surfaces, the methods, compositions and engineered cells are useful for stimulating the patient's immune response against the cancer, thereby providing a therapeutic benefit. Thus, the invention also provides methods for cancer immunotherapy, including methods of reducing tumor size and methods of inhibiting the growth of tumor cells.
While not wishing to be bound by theory, it is believed that tumor cells engineered in vivo as described herein acquire enhanced APC functions. By this approach, the direct presentation of antigens by tumor cells engineered to display one or more immune co-stimulatory polypeptides to T cells will lead to T cell activation, differentiation, and acquisition of effector functions within the tumor microenvironment and will override various limitations faced by other approaches such as inefficient presentation of exogenous antigens via a class I pathway and trafficking of antigen activated T cells into the tumor site.
Use of an in vivo modified tumor cell displaying two or more immune co-stimulatory polypeptides has the additional advantage (compared to a single tumor antigen vaccine) of presenting a whole array of tumor associated antigens to naïve and memory T cells, thereby activating diverse immune responses for effective therapy. Such diverse activation enables such a tumor vaccine to overcome tolerance by anergy or clonal deletion which might result from a single tumor antigen vaccine that presents an immunodominant epitope.
The engineering of tumors in vivo to display one or more, or combinations of two or more, immune co-stimulatory polypeptides, such as at least one of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL, or at least two of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L, enhances receptor-ligand interactions on the surface of the tumor cells and T cells. These interactions do not need to persist indefinitely. Thus, the temporary display of immune co-stimulatory polypeptides on a tumor cell surface achieved as described herein is expected to enhance a patient's immune response against the tumor cells, and provide effective immunotherapy.
For the purposes of the present application, the following terms have these definitions:
As used herein “a” or “an” means one or more, unless specifically indicated to mean only one.
“Administration” as used herein encompasses all suitable means of providing a substance to a patient. Common routes include oral, sublingual, transmucosal, transdermal, rectal, vaginal, subcutaneous, intramuscular, intravenous, intra-arterial, intrathecal, via catheter, via implant etc. In some embodiments, a composition is administered near or directly to the tumor, such as by direct injection into the tumor or injection into the blood such as when the tumor is a tumor of the blood.
“Antigen” herein is used herein without limitation. Antigens include proteins, lipids, sugars, nucleic acids, chemical moeties, and other moeties that induce an immune response. Antigens include proteins, which may or may not be modified, such as by glycosylation or methylation, that are cyclized or bound to lipids, for example. Antigens associated with an infectious agent or disease include antigens that are part of the infectious agent, such as envelope proteins, capsid proteins, surface proteins, toxins, cell walls, antigenic lipids, and the like. Other antigens may be expressed only in the presence of the host. Other suitable antigens may, in some embodiments, include antigens of the host, including those that are induced, modified or otherwise overexpressed as a marker of infection or disease. All such antigens that are derived from, or associated with, an infectious agent, an infection, a condition or disease, are suitable for use in the present invention. Also suitable for use as an “antigen” in accordance with the present invention are peptides comprising antigenic portions of full-length proteins, such as peptides comprising a portion of a protein that induces an immune response, such as an immunogenic epitope. For example, suitable antigens may include synthetic peptides that induce an immune response.
“Binding pair” refers to two molecules which interact with each other through any of a variety of molecular forces including, for example, ionic, covalent, hydrophobic, van der Waals, and hydrogen bonding, so that the pair have the property of binding specifically to each other. Specific binding means that the binding pair members exhibit binding to each other under conditions where they do not bind to another molecule. Examples of binding pairs are biotin-avidin, hormone-receptor, receptor-ligand, enzyme-substrate, lgG-protein A, antigen-antibody, and the like.
“Cell surface” is used in accordance with its normal meaning in the art, and thus includes the outside of the cell which is accessible to binding by proteins and other molecules.
“Engineered cell surface” refers to a cell surface which is modified by binding an exogenous molecule such as an immune co-stimulatory polypeptide to the cell surface. Binding of the exogenous molecule to the cell surface can be effected via any means known in the art such as by covalent bonding, hydrogen bonding, receptor-ligand interactions, and antibody-antigen interactions.
“Modified cell surface” means a cell surface to which an exogenous molecule has been bound.
“Patient” as used herein includes any vertebrate animal, including equine, ovine, caprine, bovine, porcine, avian, canine, feline and primate species. In one embodiment, the patient is human. A person of ordinary skill in the art will recognize that particular immune co-stimulatory molecules, signaling molecules, cell markers, cell types, infectious agents etc., discussed with reference to one species, may have corresponding analogues in different species, and that such analogues, and their use in corresponding and related species, are encompassed by the present invention.
“Tumor” as used herein includes solid and non solid tumors (such as leukemia); and different stages of tumor development from pre-cancerous lesions and benign tumors, to cancerous, malignant and metastatic tumors.
In general terms, a cell surface is engineered in vitro or in vivo to display one or more immune co-stimulatory polypeptides by a method that comprises contacting a cell surface with, or administering to an individual containing said cell, (a) a bifunctional molecule comprising a first member of a binding pair and a moiety that links the bifunctional molecule to the cell; (b) a first immune co-stimulatory moiety comprising a first immune co-stimulatory polypeptide and a second member of the binding pair; and, optionally, (c) a second immune co-stimulatory moiety comprising a second immune co-stimulatory polypeptide and a second member of the binding pair. The first and optional second immune co-stimulatory polypeptides are displayed on the cell surface via binding between the first and second members of the binding pair. In one particular aspect of the invention, the cell surface is engineered to display at least one of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL, or at least two of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L.
In accordance with the invention, a cell surface can be engineered with a combination of two, three, four, five, or more immune co-stimulatory polypeptides by administering a plurality of immune co-stimulatory moieties, each comprising an immune co-stimulatory polypeptide and a second member of the binding pair. By engineering cells with combinations of two or more immune co-stimulatory polypeptides, the present invention achieves greater therapeutic benefit than when only single immune co-stimulatory polypeptides are displayed. Moreover, it is believed that cell surfaces can be engineered with a plurality of immune co-stimulatory polypeptides without impeding cellular functions. Nevertheless, the invention includes embodiments where cells are engineered with one immune co-stimulatory polypeptide, such as CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, and APRIL.
Immune Co-Stimulatory Polypeptides
Cell surfaces may be engineered in vivo with any of a variety of immune co-stimulatory polypeptides or co-stimulatory fragments thereof that are known in the art. By “immune co-stimulatory polypeptide” is meant a polypeptide that increases an individual's immune response against a pathogen or tumor. Exemplary immune co-stimulatory polypeptides include, without limitation, LIGHT, CD80 (B7-1), CD86 (B7-2), B7-H2 ICOS, CD94 (KP43), CD40L (CD154), ICAM-1 (CD54), ICAM-2, ICAM-3, SLAM (CD150), HAS (CD24), 4-1BB (CDw137), 4-1BBL (CDw137L), OX40L, CD28, CD40 (BP50), CD25 (IL-2Rα), Lymphotoxin (LTα or LTβ), TNF, Fas-L, GITR (activation-inducible TNRF), GITR Ligand, CD11a (αL integrin), CD11b (αM integrin), L-selectin (CD62L), CD69 (very early activation antigen), and CD70 (CD27L). See e.g., Watts & DeBenedette, 1999, Curr. Opin. Immunol., 11:286-93. Unless specified herein as “full-length,” reference herein to an immune co-stimulatory polypeptide encompasses the full-length polypeptide as well as fragments or portions thereof that exhibit immune co-stimulatory function, including, but not limited to those fragments and portions specifically identified below. Thus, for example, reference to a 4-1BBL polypeptide connotes a polypeptide comprising a fragment or portion of full-length 4-1BBL that exhibits immune co-stimulatory function, such as the extracellular domain of 4-1BBL or the full-length 4-1BBL protein.
Examples of representative nucleic acid sequences and the encoded immune co-stimulatory polypeptides include those shown in GenBank Accession Nos. AB029155 (murine LIGHT); NM—172014 (human TNFSF14 mRNA transcript variant 2); NM—003807 (human TNFSF14 mRNA transcript variant 1); NM—005191 (human CD80 mRNA); NM—009855 (mouse CD80 mRNA); NM—214087 Sus scrofa CD80 mRNA); NM—009404 (murine Tnfsf9 mRNA); NM—003811 (human TNFSF9 mRNA); NM—181384 (Rattus norvegicus Tnfsf9 mRNA); BAA88559 (murine LIGHT protein); Q9QYH9 (murine TNFSF14 membrane bound protein and soluble protein); AAH18058 (human TNFSF14 protein); NP—005182 (human CD80 protein); NP—033985 (murine CD80 protein); NP—037058 (Rattus norvegicus CD80 protein); NP—003802 (human TNFSF9 protein); NP—033430 (mouse TNFSF9 protein); NP—852049 (Rattus norvegicus TNFSF9 protein); NM—012967 (Rattus norvegicus ICAM-1 mRNA); X69711 (human ICAM-1 mRNA); X52264 (murine ICAM-1 mRNA); X69819 (human ICAM-3 mRNA); AF296283 (murine ICAM-4 mRNA); NM—021181 (human SLAMF7 mRNA); NM—033438 (human SLAMF9 mRNA); NM—029612 (murine SLAMF9 mRNA); NM—144539 (murine SLAMF7 mRNA); L13944 (murine CD18 gene); X53586 (human integrin α6 mRNA); X68742 (human integrin α mRNA); J04145 (Human neutrophil adherence receptor alpha-M subunit mRNA); AJ246000 (human leucocyte adhesion receptor, L-selectin mRNA); AY367061 (human L-selectin mRNA, partial cds); Y13636 (murine CD70 mRNA); NM—001252 (human TNFSF7 mRNA); BC000725 (human TNFSF7 mRNA (cDNA clone MGC:1597 IMAGE:3506629), complete cds); X69397 (human CD24 gene and complete CDS); NM—013230 (human CD24 mRNA); NM—012752 (Rattus norvegicus CD24 mRNA); Y00137 (murine tumor necrosis factor-beta (lymphotoxin) gene); X02911 (human tumor necrosis factor-beta (lymphotoxin) gene); D00102 (human lymphotoxin mRNA, complete CDS); X01393 (human lymphotoxin mRNA); and A06316 ((human lymphotoxin mRNA). Other nucleic acid sequences encoding the same or other immune co-stimulatory polypeptides and/or amino acid sequences of co-stimulatory polypeptides can be found, for example, by searching the publicly available GenBank database (available, for example, at ncbi.nlm.nih.gov on the World Wide Web).
Co-stimulatory polypeptides are involved in the natural interaction between naïve T cells and antigen presenting cells, which results in their reciprocal activation and prompts the expression of various cell surface ligands and receptors, and soluble proteins that contribute to the initiation, maintenance, and long-term memory of the immune response. At least three signals are required for the initial activation of naïve T cells. Signal 1 is generated by interactions between a T cell receptor (TCR) and a nominal peptide presented by major histocompatibility complex (MHC) molecules on the surface of professional APC, such as dendrtic cells (DC). Signal 2 can be mediated by several different molecules and is important to a sustained immune response. Signal 3 is transduced via cytokines elaborated by activated T cells and APC and is important to the maintenance of effector immune responses.
Tumors have developed various counter-mechanisms to evade immune mediated destruction. These counter-mechanisms include: (i) a lack of Signal 1, which may arise from the inefficient display of MHC/tumor antigen bimolecular complexes on tumor cells, defects in the transduction of this signal, and/or expression of major histocompatibility complex homologues, MIC, that inhibit NK cells expressing NKG2 inhibitory receptors; (ii) an absence of Signal 2, which may originate from the lack of co-stimulatory molecules on tumor cells or tumor expression of co-inhibitory molecules; (iii) tumor-mediated suppression of immune responses through, for example, the secretion of anti-inflammatory molecules, induction of anergy in tumor-reactive T cells, physical elimination of effector T cells via apoptosis, or induction of Treg; and (iv) regulation of immunity by the tumor stroma. Many of these counter-mechanisms operate simultaneously, particularly in patients with large tumor burdens.
CD80 (also known as B7.1, CD28LG, LAB7) and CD86 (also known as B7.2, CD28LG2, LAB72) are exemplary co-stimulatory polypeptides, both of which bind to the CD28/CTLA4 co-receptor expressed by T cells. CD80 contains 288 amino acids (33048 Da). See Freeman et al. J. Immunol. 143 (8), 2714-2722 (1989). The full amino acid sequence of human B7.1 can be found under accession no. P33681 in the Swiss-Prot database. B7.1 is a type I glycoprotein with residues 1-34 forming a secretion signal, residues 35-242 forming a potential extraceulluar domain, residues 243-263 forming a potential transmembrane domain, and residues 264-288 forming a potential cytoplasmic domain. Thus the mature B7.1 molecule without its secretion signal sequence represents amino acids 35-288. The nucleotide sequence in humans encoding B7.1 can be found in GenBank accession no. NM—005191.
Residues 35-242 of B7.1 or fragments thereof that can bind to its cognate receptor CD28 can be linked or expressed as a fusion protein with a binding pair member for use in accordance with the present invention. For example,
B7.2 contains 329 amino acids (37696 Da). See Freeman et al. Science 262 (5135), 909-911 (1993). The full amino acid sequence of human B7.2 can be found under accession no. P42081 in the Swiss-Prot database. B7.2 is a type I glycoprotein with residues 1-23 forming a secretion signal, residues 24-247 forming a potential extraceulluar domain, residues 248-268 forming a potential transmembrane domain, and residues 269-329 forming a potential cytoplasmic domain. Thus, the mature B7.2 molecule without its secretion signal sequence represents amino acids 24-329. The nucleotide sequence in humans encoding B7.2 can be found in GenBank accession no. NM—175862.
Residues 24-247 of B7.2 or fragments thereof that can bind to its cognate receptor CD28, can be linked or expressed as a fusion with a binding pair member for use in accordance with the present invention. For example,
B7.2 is usually not expressed on resting B cells and is expressed at low levels on peripheral blood monocytes (PBC) and DC. Its expression, however, is upregulated on B cells and other APC such as macrophages and DC following activation. In contrast, CD86 is constitutively expressed on PBC and DC and more rapidly upregulated on B cells. T cell receptor (TCR) interaction with the MHC/peptide complex on APC allows for simultaneous engagement of CD80/86 with CD28 on the T cell, which leads to tyrosine phosphorylation of the lipid kinase phosphotidylinositol 3-kinase, which in turn initiates a series of intracellular events that result in the induction of IL-2 gene expression, cell proliferation, and differentiation into effector function. Signal 2 may further augment a productive immune response by preventing cell death through the regulation of antiapoptotic genes, such as Bcl-xL. Tumor cells engineered in vivo to display CD80 and/or CD86 as described herein have the ability to induce vigorous anti-tumor immune responses with potent therapeutic efficacy.
The ligand of 4-1BB (4-1BB-L) and LIGHT are further examples of preferred immune co-stimulatory polypeptides useful for cell surface engineering in vivo as described herein. Following the initial stages of immune activation, “secondary” receptor/ligand pairs such as 4-1BBL/4-1BB and LIGHT/HVEM become upregulated on the surface of T cells and APC. These receptor/ligand pairs are involved in the maintenance of post initial activation events, immune homeostasis, and generation of immunological memory.
4-1BBL (also known as 4-BB-L, 4-BB ligand, TNFSF9, ILA ligand) contains 254 amino acids (26624 Da). See Alderson et al. Eur J Immunol. 1994 September; 24(9):2219-27. The full amino acid sequence of human 4-1BBL can be found under accession no. P41273 in the Swiss-Prot database. 4-1BBL is a type II glycoprotein with residues 1-28 forming a potential cytoplasmic domain, residues 29-49 forming a single predicted transmembrane domain, residues 50-254 forming a potential extraceulluar domain, and residues 35-41 representing a poly-Leu stretch. The nucleotide sequence in humans encoding the 4-1BBL can be found in GenBank accession no. NM—003811.
4-1BBL is expressed by activated antigen presenting cells including activated B cells, macrophages, and DC, 2-3 days following activation. 4-1BBL can activate T cells, including CD8+ T cells and DC, for a more robust immune response against tumors. 4-1BB (also known as CD137), which is the receptor for 4-1BBL, is expressed on the surface of activated CD4+ and CD8+ T cells, on natural killer cells, monocytes, and resting DC. Residues 50-254 of 4-1BBL or fragments thereof that can bind to its cognate receptor 4-1BB, can be linked or expressed as a fusion with a binding pair member for use in accordance with the present invention. For example,
One advantage of the present methods is that cells, such as tumor cells, engineered in vivo to display 4-1BBL can activate the 4-1BB cognate receptor on T cells, resulting in several important immune-stimulatory effects. One effect is the transduction of Signal 2 to CD8+ T cells in a CD28-independent manner, which stimulates the T cells to produce cytokines, to expand, and acquire effector functions. Another effect of 4-1BB/4-1BBL interaction is activation of monocytes and DC which results in the synthesis and release of cytokines. Yet another effect of 4-1BB signaling is the promotion of T cell survival and establishment of long-term immunological memory by preventing activation-induced cell death (AICD). Still another effect of 4-1BB/4-1BBL interaction is the selective production of type I cytokines, such as IL-2, IFN-γ, and TNF-α from T cells, DCs and macrophages, which act upon type 1 effector T cells important to tumor eradication. Such T cells include CD4+ CD25+ T regulatory (Treg) cells, which constitutively express 4-1BB receptor. Another advantage of the present methods is that cells, such as tumor cells, engineered to display 4-1BBL can activate T cells, particularly CD8+ T cells and DC for a more robust immune response against, for example, tumors.
The LIGHT polypeptide (also known as TNFS 14, HVEM-L, LTg, TR2) is a TNF superfamily member which is homologous to lymphotoxin. See Mauri et al. Immunity 8 (1), 21-30 (1998). The full amino acid sequence of human LIGHT can be found under accession no. 043557 in the Swiss-Prot database. LIGHT contains 240 amino acids (26351 Da) and is a type II glycoprotein with residues 1-37 forming a potential cytoplasmic domain, residues 38-58 forming a single predicted transmembrane domain, and residues 59-240 forming a potential extraceulluar domain. A cleavage site involves residues 82-83. The nucleotide sequence in humans encoding LIGHT can be found in GenBank accession no. NM—172014.
Residues 59-240 of LIGHT or fragments thereof that can bind to its cognate receptor HVEM, LTβR or TR6, can be linked or expressed as a fusion with a binding pair member for use in accordance with the present invention. For example,
LIGHT is primarily expressed on activated T cells, NK cells, and immature dendritic cells, and serves to regulate various aspects of immune responses. LIGHT is synthesized as a membrane-bound protein, but its cell-surface expression is regulated by several posttranslational mechanisms. LIGHT is cleaved from the cell surface by matrix metalloproteinases within minutes of its expression and accumulates as a soluble molecule (isoform 1; represents approximately residues 83-240; Swiss-Prot O43557-1). The cell surface cytoplasmic segment represents isoform 2 (Swiss-Prot O43557-2). Additionally, various cell types store LIGHT in vesicles and excrete them upon activation by various physiological stimuli. Although the role of the soluble form of LIGHT is not well characterized, it may serve as a negative feedback loop to inhibit the function of the membrane-bound form by competing for HVEM and LTβR.
LIGHT interacts with three different receptors: (1) herpesvirus entry mediator (HVEM) on T cells, (2) LTβR which is expressed primarily on epithelial and stromal cells, and (3) the soluble decoy receptor 3 on various cells. These interactions endow LIGHT with different functions. Interaction with LTβR on stromal cells is associated with the production of various cytokines/chemokines, lymph node (LN) organogenesis, and restoration of secondary lymphoid structures. On the other hand, interaction of LIGHT with HVEM receptor on lymphocytes results in activation and production of cytokines, dominated by IFN-γ and GM-CSF. In this context, the LIGHT/HVEM axis appears to deliver costimulatory signals associated with the activation of Th1 type responses which play critical roles in tumor eradication.
LIGHT plays a role in lymphoid organogenesis and in the generation of Th1 type responses. See e.g., Yang et al., 2002, J. Biol. Regul. Homeost. Agents, 16:206-10; Schneider et al., 2004, Immunol. Rev., 202:49-66.
The effect of LIGHT has been shown in different tumor models both in vitro and in vivo. Chronic lymphocytic leukemic cells transduced by herpes simplex virus amplicon expressing LIGHT have been reported to enhance T cell proliferation in mixed lymphocyte reactions. Over-expression of LIGHT on MDA-MB-231 human breast cancer cells has been shown to suppress tumor growth. Transfection of LIGHT into different cancer cell lines stimulate ICAM-1 expression in these cells. The presence of ICAM-1 is believed to be beneficial as it enables effective signaling to produce antitumor activity in tumor cells. Another important function of LIGHT, besides T cell activation, is its ability to transduce signals through LTβR, which plays an important role in the development of secondary lymphoid structures mediated through the induction of chemokine expression as well as adhesion molecules in stromal cells. The interaction of LIGHT with LTβR on stromal cells regulates the expression of CCL21, which control the homing of naïve T cells to lymphoid tissues.
One advantage of embodiments of the invention where tumor cells are engineered to display LIGHT is the ability of LIGHT to stimulate lymphoid organogenesis and support the generation of Th1 type responses. Another advantage is the ability of LIGHT to stimulate immune responses against tumors and activate the tumor stroma to further augment these responses.
The stroma serves as a physical barrier to prevent lymphocyte infiltration into the tumor site. The stroma also inhibits lymphocyte activation within the tumor microenvironment. This may be due to the lack of costimulatory signals needed for T cell activation and/or the presence of various immunoinhibitory soluble mediators, such as TGF-β and IL-10, that are synthesized and secreted by both stromal fibroblasts and tumor cells. The stroma promotes immunological ignorance by confining tumor cells to the tumor site, thereby preventing them from trafficking to the regional lymph nodes.
Tumor stromal cells also express various immunological receptors, such as LLTβR, that can be exploited for the enhancement of anti-tumor immunity in accordance with the present invention.
OX40L is expressed by dendritic cells and other APC and binds to OX40 which is present on activated T cells. OX40L contains 183 amino acids (21950 Da). See Miura et al. Mol. Cell. Biol. 11:1313-1325 (1991). The full amino acid sequence of OX40L can be found under accession no. P23510 in the Swiss-Prot database. OX40L is a type II glycoprotein with a cytoplasmic domain at residues 1-23, a transmembrane domain at residues 24-50 and an extracellular domain at residues 51-183. The nucleotide sequence of OX40L is 3510 bp, with the coding sequence being 157-708 (see Genbank accession no. NM—003326.2). Residues 51-183, or fragments thereof of OX40L that can bind to its cognate receptor OX40, can be linked or expressed as a C-terminal fusion to a binding pair member for use in accordance with the present invention.
CD40L is expressed by activated T cells and also exists as an extracellular soluble form which derives from the membrane form by proteolytic processing. CD40L (a.k.a. TNFSF5) contains 261 amino acids (29350 Da). See Villinger et al. Immunogenetics 53:315-328 (2001). The full amino acid sequence of CD40L can be found under accession no. Q9BDN3. CD40L is a type II glycoprotein with a cytoplasmic domain at residues 1-22, a transmembrane domain at residues 23-43 and an extracellular domain at residues 44-261. The nucleotide sequence of CD40L is 1834 bp, with the coding sequence being 73-858 (see Genbank accession no. NM—000074). Residues 44-261, or fragments thereof of CD40L that can bind to its cognate receptor CD40, can be linked or expressed as an N-terminal fusion to a binding pair member for use in accordance with the present invention.
PD-L1 is expressed on activated T and B cells, dendritic cells, keratinocytes and monocytes. PD-L1 (a.k.a. B7-H; B7H1; PDL1; PDCD1L1) contains 290 amino acids (33275 Da). See Dong et al. Nat. Med. 5: 1365-1369 (1999). The full amino acid sequence of PD-L1 can be found under accession no. Q9NZQ7 in the Swiss-Prot database. PD-L1 contains 290 amino acids of which 18 amino acids at the N terminus represent the signal sequence. The extracellular domain is located at amino acids 19-238, a transmembrane domain is located at resides 239-259 and a cytoplasmic domain is located at residues 260-290. The nucleotide sequence of PD-L1 (1553 bp) is available in public databases (see Genbank accession no. NM—014143) (coding sequence is 53-925). Isoforms of PD-L1 exist by way of alternative splicing. The extracellular domain or fragments thereof of PD-L1 that can bind to its cognate receptor PDCD 1, can be linked or expressed as an N-terminal fusion to a binding pair member for use in accordance with the present invention.
GL50 isoform 1 is widely expressed (brain, heart, kidney, liver, lung, pancreas, placenta, skeletal muscle, bone marrow, colon, ovary, prostate, testis, lymph nodes, leukocytes, spleen, thymus and tonsil); GL50 isoform 2 (swissprot O75144) is expressed in lymph nodes, leukocytes and spleen and on activated monocytes and dendritic cells. GL50 (a.k.a. B7-H2; B7H2; B7RP-1; B7RP1; ICOS-L; ICOSLG; KIAA0653; and LICOS) contains 290 amino acids (33275 Da). See Wang et al. Blood 96:2808-2813 (2000). The full amino acid sequence of GL50 can be found under accession no. O75144 in the Swiss-Prot database. GL50 contains 302 amino acids of which 18 amino acids at the N terminus represent the signal sequence. The extracellular domain is located at amino acids 19-256, a transmembrane domain is located at resides 257-277 and a cytoplasmic domain is located at residues 278-302. The nucleotide sequence of GL50 (3239 bp) is available in public databases (see Genbank accession no. NM—015259) (coding region representing 135-1043). Isoforms of GL50 exist by way of alternative splicing. The extracellular domain or fragments thereof of GL50 that can bind to its cognate receptor ICOS, can be linked or expressed as an N-terminal fusion to a binding pair member for use in accordance with the present invention.
Other immune co-stimulatory polypeptide can be used in accordance with the invention. For example US 2003/0219419 (the entire contents of which are incorporated herein by reference in their entirety) describes IL-2-CSA fusion proteins, and CSA-CD40L fusion proteins that are useful in the present invention.
Table 1 summarizes various costimulatory molecules and their receptors.
In summary, exemplary immune co-stimulatory polypeptides useful in accordance with the present invention include the following.
*it is a TNF member
** these are B cell related
Displaying Immune Co-Stimulatory Polypeptides
One or more immune co-stimulatory polypeptides can be displayed on cells, such as tumor cells, engineered in vitro or in vivo as described herein. When a combination of polypeptides is used, the specific combination employed can be chosen to bring different immune mechanisms into play. An example of specific combinations contemplated by the present invention include combinations comprising two or more of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L.
In one particular aspect of the invention, the combination comprises at least one of 4-1BBL or LIGHT, such as 4-1BBL and LIGHT, and 4-1BBL, LIGHT and CD80. Although not wishing to be bound by any theory, it is believed that cells, such as tumor cells, made to display these particular combinations of co-stimulatory molecules will have one or more of the following effects: (i) LIGHT displayed on tumor cells will engage LTβR on stromal cells in the tumor, resulting in activation of stromal cells and secretion of various lymphocyte-specific chemokines; (ii) the released lymphokines will recruit naïve T cells and DC into the tumor site and (iii) recruited T cells and DC will become activated within the tumor microenvironment by the engagement of LIGHT, CD80, and 4-1BBL on engineered tumor cells with their respective receptors, HVEM (the second LIGHT receptor), CD28, and 4-1BB expressed on T cells and DC (4-1BB). This and other effects will generate a robust, systemic anti-tumor immunity with therapeutic implications.
In accordance with one specific embodiment, a cell, such as a tumor cell, is engineered with one or more of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L, such as with a combination of two or more of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L. In one specific embodiment, a cell, such as a tumor cell, is engineered with each of LIGHT, CD80, and 4-1BBL. These embodiments offer all of the advantages discussed above and provides a unique approach that not only targets immune activation, but also modulates tumor stroma to potentiate immune responses. Thus, it is believed that engineering cells, such as tumor cells, in vivo with a combination of two or more of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L offers the possibility of converting, for example, tumors into pseudo secondary lymphoid structures for the generation of effective intratumoral and systemic anti-tumor immune responses that may eradicate existing tumors and also safeguard against recurrences.
Binding Pairs
An exemplary binding pair is biotin and streptavidin (SA) or avidin. SA or avidin fragments which retain substanatial binding activity for biotin, such as at least 50% or more of the binding affinity of native SA or avidin, respectively, also may be used. Such fragments include “core streptavidin” (“CSA”) a truncated version of the full-length streptavidin polypeptide which may include streptavidin residues 13-138, 14-138, 13-139 and 14-139. See, e.g., Pahler et al., 1987, J Biol. Chem., 262:13933-37. Other truncated forms of streptavidin and avidin that retain strong binding to biotin also may be used. See, e.g. Sano et al., J Biol Chem. 1995 Nov. 24;270(47):28204-9 (describing core streptavidin variants 16-133 and 14-138) (U.S. Pat. No. 6,022,951). Mutants of streptavidin and core forms of strepavidin which retain substantial biotin binding activity or increased biotin binding activity also may be used. See Chilcoti et al., Proc Natl Acad Sci USA. 1995 Feb. 28;92(5):1754-8; Reznik et al., Nat Biotechnol. 1996 August;14(8):1007-11. For example, mutants with reduced immunogenicity, such as mutants mutated by site-directed mutagenesis to remove potential T cell epitopes or lymphocyte epitopes, can be used. See Meyer et al., Protein Sci. 2001 10: 491-503. Likewise, mutants of avidin and core forms of avidin which retain substantial biotin binding activity or increased biotin binding activity also may be used. See Hiller et al., J. Biochem. (1991) 278: 573-85; Livnah et al. Proc Natl Acad Sci USA (0: 5076-80 (1993). For convenience, in the instant description, the terms “avidin” and “streptavidin” as used herein are intended to encompass biotin-binding fragments, mutants and core forms of these binding pair members. Avidin and streptavidin are available from commercial suppliers. Moreover, the nucleic acid sequences encoding streptavidin and avidin and the streptavidin and avidin amino acid sequences can be found, for example, in GenBank Accession Nos. X65082; X03591; NM—205320; X05343; Z21611; and Z21554.
As used herein “biotin” includes biotin-containing moieties that are able to bind to surfaces, such as cell surfaces (including tumor cell surfaces), such as NHS-biotin and EZ-Link™ Sulfo-NHS-LC-Biotin (Pierce). Such protein reactive forms of biotin are available commercially.
The interaction between biotin and its binding partner, avidin or streptavidin, offers several advantages in the context of the present invention. For example, biotin has an extremely high affinity for both streptavidin (1013 M−1) and avidin (1015 M−1). This embodiment also is advantageous because immune co-stimulatory moieties comprising streptavidin or avidin can be preferentially localized to the surface of any biotinylated cell, tissue, or organ. Targeting at desired levels can be achieved in a rapid (less than about 2 hours), efficient (about 100% of the targeted cells), and durable (t1/2=from days to weeks) manner without compromising relevant functions of either the immune co-stimulatory polypeptide or the targeted cell. Additionally, both streptavidin and avidin are tetrameric polypeptides that each bind four molecules of biotin. Immune co-stimulatory moieties comprising streptavidin or avidin therefore have a tendency to form tetramers and higher structures. As a result, they can cross-link their corresponding immune cell receptors for more potent signal transduction, such as through aggregation of receptors. Also, cell surfaces can be engineered with a plurality of immune co-stimulatory polypeptides without impeding cellular functions.
Those skilled in the art will recognize that other mechanisms (e.g., other conjugation methods using, for example, other linking moieties or chemical or genetic cross-linking) can be used to provide higher-order structures of immune co-stimulatory molecules, such as conjugates comprising dimers, trimers, tetramers and higher-order multimers of immune co-stimulatory molecules, which also will exhibit advantageous properties. Such conjugates are included within the scope of this invention.
Immune Co-Stimulatory Moieties
An immune co-stimulatory moiety comprising an immune co-stimulatory polypeptide and a second member of a binding pair can be made by methods well known in the art. For example, the polypeptide and second member can be covalently bound to each other or conjugated to each other directly or through a linker. In accordance with one embodiment of the invention, the immune co-stimulatory moiety is a fusion protein comprising an immune co-stimulatory polypeptide and a second member of a binding pair. Fusion proteins can be made by any of a number of different methods known in the art. For example, one or more of the component polypeptides of the fusion proteins can be chemically synthesized or can be generated using well known recombinant nucleic acid technology. (As used herein, “nucleic acid”refers to RNA or DNA.) Nucleic acid sequences useful in the present invention can be obtained using, for example, the polymerase chain reaction (PCR). Various PCR methods are described, for example, in PCR Primer: A Laboratory Manual, Dieffenbach 7 Dveksler, Eds., Cold Spring Harbor Laboratory Press, 1995.
In accordance with one embodiment, the immune co-stimulatory polypeptide is bound via its C-terminus to the N-terminus of the second member of the binding pair. For example, the invention includes CD80-CSA fusion proteins, where the CD80 moiety is bound via its C-terminal to the N-terminal of CSA. In accordance with another embodiment, the immune co-stimulatory polypeptide is bound via its N-terminus to the C-terminus of the second member of the binding pair. For example, the invention includes CSA-4-1BBL, CSA-LIGHT, CSA-CD40L, and CSA-OX40L fusion proteins, where the CSA moiety is bound via its C-terminal to the N-terminal of the immune co-stimulatory moiety. The immune co-stimulatory polypeptide may be directly bound to second member of the binding pair or may be bound via one or more linking moieties, such as one or more linking polypeptides.
Nucleic acids and polypeptides comprising a fragment of an immune co-stimulatory polypeptide and/or a fragment of a binding pair member are useful in the present invention, as long as the fragment retains the activity of the referent full-length polypeptide. Thus, the immune co-stimulatory fragment should retain its immune co-stimulatory activity (e.g., retain its ability to bind its receptor or ligand), and the binding member fragment should retain its ability to bind with its binding partner. Fragments can be screened for retained activity by methods that are routine in the art, including those exemplified in the examples below. Exemplary fragments of immune co-stimulatory polypeptides are set forth above.
The immune co-stimulatory moiety may include a linker such as a peptide linker between the second binding pair member and the immune co-stimulatory polypeptide. The linker length and composition may be chosen to enhance the activity of either functional end of the moiety. The linker is generally from about 3 to about 15 amino acids long, more preferably about 5 to about 10 amino acids long, however, longer or shorter linkers may be used or the linker may be dispensed with entirely. Flexible linkers (e.g. (Gly4Ser)3) such as have been used to connect heavy and light chains of a single chain antibody may be used in this regard. See Huston, et al. (1988) Proc. Nat. Acad. Sci. USA, 85:5879-5883; U.S. Pat. Nos. 5,091,513, 5,132,405, 4,956,778; 5,258,498, and 5,482,858. Other linkers are FENDAQAPKS or LQNDAQAPKS. One or more domains of and immunoglobulin Fc region (e.g CH1, CH2 and/or CH3) also may be used as a linker.
Nucleic acids and polypeptides that are modified, varied, or mutated also are useful in the present invention, as long as they retain the activity of the referent nucleic acid or polypeptide. For example, nucleic acid and polypeptide sequences suitable for use in the present invention can have at least about 80% sequence identity (including at least 80% sequence identity) to a referent nucleic acid or polypeptide, i.e., to a nucleic acid encoding a known immune co-stimulatory polypeptide or binding pair member. In some embodiments, the nucleic acid sequence or polypeptide has at least about 85%, at least about 90%, at least about 95%, or at least about 99% sequence identity to the referent nucleic acid or polypeptide.
The invention encompasses nucleic acids with base changes that are “silent,” in that they encode the same amino acid (i.e. degenerate nucleic acid sequences). The invention also encompasses nucleic acids that encode polypeptides with conservative amino acid substitutions, and such polypeptides. Conservative amino acid substitutions (for example, substituting one hydrophobic residue with a different hydrophobic residue) are well known in the art and can be effected, e.g., by point mutations and the like. The suitability of a given modified sequence, variant or mutant can be confirmed using receptor binding and/or biological screening methods that are known in the art, such as those discussed above with reference to fragments.
As used herein, “% sequence identity” is calculated by determining the number of matched positions in aligned nucleic acid or polypeptide sequences, dividing the number of matched positions by the total number of aligned nucleotides or amino acids, respectively, and multiplying by 100. A matched position refers to a position in which identical nucleotides or amino acids occur at the same position in the aligned sequences. The total number of aligned nucleotides or amino acids refers to the minimum number of nucleotides or amino acids that are necessary to align the second sequence, and does not include alignment (e.g., forced alignment) with non-homologous sequences, such as those that may be fused at the N-terminal or C-terminal of the sequence of interest (i.e., the sequence encoding the immune co-stimulatory polypeptide or binding pair member). The total number of aligned nucleotides or amino acids may correspond to the entire coding sequence or may correspond to fragments of the full-length sequence as defined herein.
Sequences can be aligned using the using the algorithm described by Altschul et al. (1997, Nucleic Acids Res., 25:3389-3402) as incorporated into the BLAST (basic local alignment search tool) programs, available at ncbi.nlm.nih.gov on the World Wide Web. BLAST searches or alignments can be performed to determine percent sequence identity between a nucleic acid molecule (the “query sequence”) and any other sequence or portion thereof using the Altschul algorithm. BLASTN can be used to align and compare the identity between nucleic acid sequences, while BLASTP can be used to align and compare the identity between amino acid sequences. When utilizing BLAST programs to calculate the percent identity between a nucleic acid sequence encoding a therapeutic polypeptide and another sequence, the default parameters of the respective programs can be used including the default for gap penalty.
Nucleic acids of the present invention may be detected by methods such as Southern or Northern blot analysis (i.e., hybridization), PCR, or in situ hybridization analysis. Polypeptides are typically detected by immunocytochemistry in transfected cell lines or by sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis followed by Coomassie Blue-staining or Western blot analysis using antibodies (monoclonal or polyclonal) that have specific binding affinity for the particular polypeptide. Many of these methods are discussed in detail in Sambrook et al. (1989, Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y).
Nucleic acid sequences encoding an immune co-stimulatory polypeptide and the second member of the binding pair can be operably linked to one another in a construct using conventional molecular biology techniques. See, for example, Molecular Cloning: A Laboratory Manual (Sambrook et al., 2001, 2nd Ed., Cold Spring Harbor Laboratory Press) or Short Protocols in Molecular Biology (Ausubel et al., 2002, 5th Ed., Current Protocols). Constructs suitable for use in these methods are commercially available and used routinely in the art. Constructs can include elements necessary for expression such as promoter sequences, regulatory elements such as enhancer sequences, and response elements and/or inducible elements that modulate expression of a nucleic acid sequence. As used herein, “operably linked” refers to (i) positioning of a promoter and/or other regulatory element(s) relative to a nucleic acid sequence in such a way as to direct or regulate expression of the nucleic acid; and/or (ii) positioning the nucleic acid encoding the immune co-stimulatory polypeptide and the nucleic acid encoding the second member of the binding pair, such that the coding sequences are “in frame,” i.e., such that the construct encodes a fusion protein comprising the immune co-stimulatory polypeptide and the second member of the binding pair.
A construct can be propagated or expressed to generate a polypeptide in a host cell by methods known in the art. As used herein, the term “host” or “host cell” is meant to include not only prokaryotes, such as E. coli, but also eukaryotes, such as yeast, insect, plant and animal cells. Animal cells include, for example, COS cells and HeLa cells. A host cell can be transformed or transfected with a DNA molecule (e.g., a construct) using any of the techniques commonly known to those of ordinary skill in this art, such as calcium phosphate or lithium acetate precipitation, electroporation, lipofection and particle bombardment. Host cells containing a vector of the present invention may be used for purposes such as propagating the vector, producing a nucleic acid (e.g., DNA or RNA), expressing an immune co-stimulatory polypeptide or fragments thereof, or expressing a fusion protein, as described above.
In vivo Cell Surface Engineering
In accordance with the present invention, cells in vivo are modified to display the immune co-stimulatory polypeptides. As discussed, this may be achieved by administering to the individual a bifunctional molecule comprising a first member of a binding pair and a molecule that can bind to the surface of a cell, such as a tumor cell, and one or more co-stimulatory moieties (each comprising a second member of the binding pair and at least one immune co-stimulatory polypeptide). The bifunctional molecule is designed such that the first member of the binding pair substantially retains its affinity for the second member of the binding pair after the bifunctional molecule has bound to the cell surface via the cell surface binding portion of the bifunctional molecule. In one specific embodiment, the bifunctional molecule is a form of biotin that can conjugate to proteins on the surface of cells in vivo, such as NHS-Biotin. When the bifunctional molecule comprising the first member comprises biotin as the molecule that can bind to the cell surface, it can be localized to the cell surface by methods exemplified below or by other methods, such as those described in WO 02/02751.
In some embodiments, there may be a linker or other such spacer located between the first member of the binding pair and the portion of the bifunctional molecule that is responsible for binding to the cell surface. The length and composition of such linker can be chosen to maximize the activities of the two ends of the bifunctional molecule as is well known in the art.
The bifunctional molecule (comprising the first binding pair member) and the one or more co-stimulatory moieties can be administered at substantially the same time or at different times. The first binding pair member usually will be administered before any of the one or more immune co-stimulatory moieties. For example, this time period may vary from one to a few hours, to from one day to a few days, to from one week or longer. In one specific embodiment, an immune co-stimulatory moiety is administered at least 16 hours after the bifunctional molecule, such as about 16 hours after, about 24 hours after, about 32 hours after, about 36 hours after, or about 48 hours after. In another embodiment, an immune co-stimulatory moiety is administered at least about 2 hours after the bifunctional molecule, including about 2 hours after, about 4 hours after, or about 8 hours after. However, the two administrations may be given at substantially the same time or with the first binding pair member administered after the one or more co-stimulatory moieties are administered. Alternatively, the first binding pair member and the immune co-stimultatory moieties may be administered as a single composition. When engineering with multiple immune co-stimulatory moieties, these moieties may be combined as a single composition or administered as separate compositions given at substantially the same time or sequentially (at different times).
The first member of the binding pair and/or the co-stimultatory moieties can be administered to the patient systemically or locally, such as by intravenous, peritoneal, or subcutaneous injection. In one embodiment, one or more of the composition(s) are administered locally via direct injection into the tumor site, such as by intratumoral injection. In another embodiment one or more of the compositions are administered by different routes. For example, or one or more composition can be administered locally and one or more can be administered systemically.
Any cell surface can be engineered in accordance with the invention. In one embodiment, the cell surface is a tumor cell surface. Representative tumor cells for which this invention is useful include, without limitation, carcinomas, which may be derived from any of various body organs including lung, liver, breast, skin, bladder, stomach, colon, pancreas, and the like. Carcinomas may include adenocarcinoma, which develop in an organ or gland, and squamous cell carcinoma, which originate in the squamous epithelium. Other cancers that can be treated include sarcomas, such as osteosarcoma or osteogenic sarcoma (bone), chondrosarcoma (cartilage), leiomyosarcoma (smooth muscle), rhabdomyosarcoma (skeletal muscle), mesothelial sarcoma or mesothelioma (membranous lining of body cavities), fibrosarcoma (fibrous tissue), angiosarcoma or hemangioendothelioma (blood vessels), liposarcoma (adipose tissue), glioma or astrocytoma (neurogenic connective tissue found in the brain), myxosarcoma (primitive embryonic connective tissue), an esenchymous or mixed mesodermal tumor (mixed connective tissue types). In addition myelomas, leukemias, and lymphomas are also susceptible to treatment.
In various embodiments, an immune response is elicited against specific tumor associated antigens (TAA). A number of TAAs associated with specific tumor types have been identified. These include human telomerase reverse transcriptase (hTERT), survivin, MAGE-1, MAGE-3, human chorionic gonadotropin, carcinoembryonic antigen, alpha fetoprotein, pancreatic oncofetal antigen, MUC-1, CA 125, CA 15-3, CA 19-9, CA 549, CA 195, prostate-specific antigens; prostate-specific membrane antigen, Her2/neu, gp-100, mutant K-ras proteins, mutant p53, truncated epidermal growth factor receptor, chimeric protein p210BCR-ABL; E7 protein of human papilloma virus, and EBNA3 protein of Epstein-Barr virus. Any of these antigens, antigenic fragments thereof, and mixtures of antigens and/or fragments can be used in accordance with the invention to generate or enhance a patient's anti-tumor immune response. Table 5 lists some exemplary TAAs and diseases associated with such TAAs.
Additional human TAAs recognized by T-cells may be found in, for example, Novellino et al. “A listing of human tumor antigens recognized by T cells: March 2004 update” Cancer Immunology and Immunotherapy, 54: 187-207 (2005) which is incorporated by reference herein. Many animal TAAs corresponding to animal correllaries of these diseases, and to other animal diseases, are known in the art and also included within the scope of the invention.
Cell surfaces that have been modified ex vivo and placed back into the patient (by for example, local injection) also can be engineered in vivo in accordance with the methods herein Thus, for example, individuals who are administered compositions to engineer tumor cells surfaces as described herein also may be administerd tumor cells engineered ex vivo, as has been previously described. See WO 02/02751.
As shown in the examples below, cell surfaces such as tumor cell surfaces may remain biotinylated for one week or longer. Thus, the cell surfaces can be engineered with immune co-stimulatory polypeptides up to one week or longer after initial biotinylation and/or after initial engineering. For example, after biotinylation, one or more immune co-stimulatory-streptavidin fusion proteins can be administered to display the immune co-stimulatory polypeptide on the biotin-modified tumor cell surfaces. Then, after a period of time, for example, from several hours to several days to a week or more, further engineering can be achieved by administering one or more immune co-stimulatory-streptavidin fusion proteins (the same as or different from the one(s) previously administered) which to display the one or more immune co-stimulatory polypeptides on the biotin-modified tumor cell surfaces. Additionally or alternatively, a second biotinylation step can be carried out, followed by the administration of one or more immune co-stimulatory-streptavidin fusion proteins.
Sequential engineering with the same or different immune co-stimulatory polypeptides can offer improved therapeutic efficacy. For example, cell surfaces such as tumor cell surfaces can be engineered with immune co-stimulatory polypeptide(s) that play a role early in the immune response followed by engineering with immune co-stimulatory polypeptide(s) that play a role later in the immune response. The time period between sequential administrations can vary from substantially none, to from one to a few hours, to from one day to a few days, to from one week or longer.
In one embodiment, cell surfaces, such as tumor cell surfaces, are first engineered with immune co-stimulatory molecules that bind to constitutive receptors, such as CD80, LIGHT, and CD40L, and subsequently engineered with immune co-stimulatory molecules that bind to inducible receptors, such as 4-1BBL and OX40L. For example, tumor cell surfaces may be engineered with CD80 and/or LIGHT, and then subsequently, such as 2-3 days or longer afterwards, engineered with 4-1BBL.
Table 6 provides a listing of costimulatory molecules that are constitutive and inducible.
Cell surfaces engineered in accordance with the present invention retain one or more immune co-stimulatory polypeptides for a period of time that is temporary, but not transient. For example, in vivo engineered cell surfaces may retain one or more immune co-stimulatory polypeptides for at least 24 hours, at least 36 hours at least 48 hours, at least 60 hours, at least 72 hours, at least 4 days, at least 5 days, or longer.
In one embodiment, at least about 5% of cells initially positive for immune co-stimulatory polypeptide remain positive at about 24 hours post-administration. In another embodiment, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 24 hours post-administration. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 36 hours post-administration. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 48 hours post-administration. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 60 hours post-administration. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 72 hours post-administration. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 4 days post-administration.
In one embodiment, at least about 5% of cells remain positive for administered immune co-stimulatory polypeptide at about 5 days post-administration. In another embodiment, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 5 days post-administration. For example, at 5 days post-injection, at least 5%, at least 10%, at least 15%, at least 20% or at least 25% of cells remain positive for administered immune co-stimulatory polypeptide. In another embodiment, at least about 5%, at least about 10%, at least about 15%, at least about 20% or at least about 25% of cells remain positive for administered immune co-stimulatory polypeptide at about 7 days post-administration, or at about 10 days post-administration. For example, at 7 or 10 days post-injection, at least 5%, at least 10%, at least 15%, at least 20% or at least 25% of cells remain positive for administered immune co-stimulatory polypeptide.
Cancer Immunotherapy
The invention also provides methods for cancer immunotherapy, including methods of reducing tumor size and methods of inhibiting the growth of tumor cells. In accordance with these methods, tumor cell surfaces are engineered with one or more immune co-stimulatory molecules as described above.
Efficacy of immunotherapy can be assessed by determining the decrease in tumor cell proliferation and/or tumor size. The number of tumor cells is not static and reflects both the number of cells undergoing cell division and the number of cells dying (e.g., by apoptosis). Increasing an individual's immune response against tumor cells may inhibit proliferation of the cells. Proliferation of tumor cells as used herein refers to an increase in the number of tumor cells (in vitro or in vivo) over a given period of time (e.g., hours, days, weeks, or months). Inhibiting the proliferation of tumor cells can be measured by a decrease in the rate of increase in tumor cell number, a complete loss of tumor cells, or any decrease in proliferation therebetween. A decrease in the size of a solid tumor is an indication of an inhibition of proliferation of tumor cells.
A lack of immune stimulation by engineered cells could be due to the immune co-stimulatory fusion proteins physically blocking the interaction of T cells with MHC molecules expressed on tumor cells, or due to the presence of a suboptimal number of co-stimulatory polypeptides on the cell surface. Those skilled in the art can vary the amount of immune co-stimulatory polypeptide targeted to the modified cell surfaces in order to optimize immunostimulatory activity.
Compositions and Combinations Thereof
The invention also provides compositions and combinations of compositions for use in the in vivo engineering methods described herein.
For example, the invention provides compositions comprising immune co-stimulatory moieties, including fusion proteins comprising immune co-stimulatory polypeptides and binding pair members. In one embodiment, the invention provides a composition comprising a first soluble fusion protein comprising a first immune co-stimulatory polypeptide and a second member of a binding pair. In one particular embodiment, the immune co-stimulatory polypeptide comprises CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, or APRIL. In another embodiment, In one the immune co-stimulatory polypeptide comprises CD80 or CD40L.
In another embodiment, the invention provides a composition comprising a first soluble fusion protein comprising a first immune co-stimulatory polypeptide and a second member of a binding pair and a second soluble fusion protein comprising a second immune co-stimulatory polypeptide and a second member of a binding pair. At least one of the first or second immune co-stimulatory polypeptide can be CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 or CD40L, and the second member of the binding pair can be streptavidin. In one embodiment, the composition comprises one or more soluble fusion proteins comprising a sequence selected from the group consisting of SEQ ID NO:2, SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7 and SEQ ID NO:9. In another embodiment, the invention provides a composition comprising a soluble fusion protein comprising LIGHT and the second member of a binding pair, a soluble fusion protein comprising CD80 and the second member of a binding pair, and a soluble fusion protein comprising 4-1BBL and the second member of a binding pair. The compositions also can be provided with instructions for administration to effect in vivo engineering of cell surfaces.
The compositions of the present invention optionally may comprise a pharmaceutically acceptable carrier. A pharmaceutically acceptable carrier is a material that can be used as a vehicle for the composition because the material is inert or otherwise medically acceptable, as well as compatible with the active agent(s), in the context of administration. A pharmaceutically acceptable carrier can contain conventional pharmaceutical additives like diluents and preservatives.
The invention also provides combinations comprising a first composition comprising a first soluble fusion protein comprising a first immune co-stimulatory polypeptide and a second member of a binding pair, and a second composition comprising a second soluble fusion protein comprising a second immune co-stimulatory polypeptide and a second member of a binding pair. In one embodiment, the second member of the binding pair is streptavidin. In one embodiment, at least the first or second immune co-stimulatory polypeptide is selected from the group consisting of CD86, ICOSL (including B7h, B7-H2, B7RP-1, GL-50 and LICOS), PD-L1, PD-L2, B7-H3, B7-H4, OX40L, 4-1BBL, CD27L, CD30L, LIGHT, BAFF, APRIL, CD80 and CD40L. In another embodiment at least the first or second soluble fusion protein comprises a sequence selected from the group consisting of SEQ ID NO:2, SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7 and SEQ ID NO:9. In another embodiment, the combination comprises a composition comprising a soluble fusion protein comprising LIGHT and the second member of a binding pair, a composition comprising a soluble fusion protein comprising CD80 and the second member of a binding pair, and composition comprising a soluble fusion protein comprising 4-1BBL and the second member of a binding pair. Each composition of the combination can be provided in a separate container. The combinations also can be provided with instructions for simultaneous or sequential administration of the compositions to effect in vivo engineering of cell surfaces.
The following examples illustrate the invention in more detail, and are not intended to limit the scope of the invention in any respect.
Experimental Methods
Tumor models used include the A20 tumor model and the Lewis lung carcinoma (LLC) cell line is used as a transplantable murine lung solid tumor model. The choice of LLC partly stems from the fact that lung cancer is the leading cause of cancer death among men and women in the U.S. LLC solid tumors can be established in mice and engineered in vivo for the display of immune co-stimulatory polypeptides by, for example, intratumoral injection of a biotin moiety and then a combination of two or more immune co-stimulatory-streptavidin fusion proteins, either simultaneously or sequentially. Tumors can be surgically removed and analyzed for the presence of cell surface biotin and immune co-stimulatory polypeptides to demonstrate that tumor cell surfaces can be engineered in vivo with a combination of immune co-stimulatory polypeptides to improve a patient's immune response against cancer.
Animals. Adult inbred C57BL/6 and BALB/c mice are purchased from Jackson Laboratories (Bar Harbor, Me.) or bred in the University of Louisville animal colony.
RNA preparation. Cells or tissues of interest are harvested and immediately homogenized in Trizol (InVitrogen, San Diego, Calif.). Total RNA is isolated according to the manufacturer's protocol. The pure RNA is dissolved in diethylpyrocarbonate-treated water, supplemented with RNAse inhibitor, dispensed into small aliquots, and stored at −70° C. before use. Each RNA aliquot is thawed only once and the unused portion is discarded to eliminate variations associated with RNA instability as a result of frequent freezing and thawing.
Cloning and expression of fusion proteins. Human CD80 and mouse 4-1BBL were cloned and expressed in an insect expression system. Singh et al., 2003, Cancer Res., 63:4067-73. For the cloning of LIGHT, mouse splenocytes are activated for 3 hrs with 10 ng/mL PMA and 0.5 μM ionomycin and processed for total RNA extraction. Id. Two mg of total RNA is converted into first strand cDNA and then amplified using primers specific for the extracellular portions of LIGHT that lacks the metalloproteinase cleavage in PCR as described in Yu et al., 2004, Nat. Immunol., 5:141-9. After sequence confirmation, the gene is subcloned C-terminus to core streptavidin (SA) in the Drosophila pMT/BiP/V5-His expression vector (Invitrogen, San Diego, Calif.) using standard molecular methods.
Transfection. Transient transfection of Drosophila S2 cells is performed with 20 μg plasmid DNA using Calcium Phosphate Transfection kit as previously described in Yolcu et al., 2002, Immunity, 17:795-808. Expression in S2 cells is induced by adding copper sulfate to the medium to a final concentration of 500 μM, 1 to 4 days after transfection. Stable transfectants are established using cotransfection with pCoHYGRO vector at 20:1 (vector: interest) ratio under similar experimental conditions as described for transient transfection. After 3 days, transfectants are maintained in the presence of 300 μg/ml hygromycin with culture medium changed every 3-5 days. Stable transfectants then are cloned at 0.3 cell/well in 96-titer plates for the identification of high expressors. The expression of these genes is determined using antibodies specific for 6×His-tag (QIAGEN, Valencia, Calif.) in ELISA, immunoprecipitation, and Western blots.
Purification of recombinant proteins. Recombinant fusion proteins are expressed with the 6-His-tag that is utilized for large-scale purification using nickel-nitrilotriacetic acid (Ni-NTA) immobilized metal affinity chromatography (QIAGEN). Briefly, stable transfectants are grown in serum-free medium and induced for 24 hr with copper sulfate. Culture medium is mixed with 2× Native Binding buffer in a 1:1 ratio and loaded onto columns containing Ni-NTA agarose beads. After washing the column extensively with the wash buffer (50 mM NaH2PO4 pH 8.0; 300 mM NaCl; 20 mM imidazole), fusion proteins are eluted with elution buffer (50 mM NaH2PO4 pH 8.0; 300 mM NaCl; 250 mM imidazole). The purity of the isolated proteins is determined by SDS-polyacrylamide gel electrophoresis. Protein concentration is determined using Bradford assay according to the manufacturer's instructions (BioRad, Hercules, Calif.).
ELISA. CCL21 expression by A20 or LLC tumor stromal cells is determined by ELISA. Briefly, tumors are re-sected and mechanically processed into single cell suspension. These cells are incubated with various concentrations of CSA-LIGHT (1-5000 ng/ml) for various time periods (2-5 days). Culture supernatants are harvested and tested for the presence of CCL21 using commercially available anti-mouse CCL21 Abs (R&D Systems Inc. Minneapolis, Minn.). Yang et al., 2004, Clin. Cancer Res., 10:2891-901. Cultures without CSA-LIGHT or with CSA control protein and those incubated with plate-bound LIGHT serve as negative and positive controls, respectively.
Engineering of cell surfaces. LLC or A20 tumor cells are washed with ice-cold PBS and incubated at 2×107 cells/ml in PBS, pH 8.0, containing 5-15 μM EZ-Link™ Sulfo-NHS-LC-Biotin (Pierce) at room temperature for 30 min. Cells are washed extensively and incubated at 1×106 cell/ml with 100 ng/purified immune co-stimulatory-streptavidin fusion protein for 45 min at 4° C. After extensive washing with PBS, cells are processed for flow cytometric analysis to assess the cells engineered with the immune co-stimulatory fusion proteins. Such cells also are used for functional studies in vitro. Cells are irradiated at 100 Gy prior to being used in in vitro studies.
Mixed lymphocyte tumor reactions (MLTRs). Lymph node (LN) cells are collected from animals of interest, processed into single cell suspension, resuspended in complete mixed lymphocyte reaction (MLR) medium (RPMI-1640 supplemented with 10 mM HEPES, 100 U/ml penicillin and 100 μg/ml streptomycin, 274 μM L-arginine, and 5% pooled mouse serum), and used as responders against irradiated (100 Gy) tumor cells from various genetic lines having antigenic disparities. After 3-5 days of incubation, cultures are pulsed with 1 μCi/well of [3H]thymidine (NEN Life Sciences Products, Boston, Mass.) for the last 18 hours of the culture period and harvested for the assessment of cell proliferation. Unmodified cells and cells decorated with control CSA protein serve as controls.
Cytotoxic T lymphocyte (CTL) assay. Effector cells (5×106) from LNs of animals of interest are generated in culture for 5 days with the same number of irradiated (100 cGy) tumor cells without decoration in complete MLTR medium supplemented with 60 IU/ml of IL-2. Effector cells are harvested using Lympholyte®-M (Accurate Chemical and Scientific Corporation, Westbury, N.Y.) and used against target cells at different effector:target (E:T) ratios using the JAM assay. See Matzinger, 1991, J. Immunol. Methods, 145:185-92.
In vivo tumor engineering and immunotherapy. B6 or BALB/c mice are challenged subcutaneously in the flank with a lethal dose (e.g., 1×105− 1×106) of viable LLC or A20 tumor cells, respectively. Once the tumor size reaches 6-8 mm in diameter, tumors are injected with 300 μl of biotin and then with immune co-stimulatory-streptavidin fusion proteins. Controls include unmodified tumors and tumor modified with CSA protein. The diameter of the tumors is measured by caliper measurement and expressed as the average of the longest diameter and the perpendicular diameter±standard error as measured three times a week. Animals are euthanized when the tumor size reaches approximately 20 mm in diameter.
Competitive RT-PCR for cytokine expression. Total RNA is prepared from samples of interest and 2 μg of total RNA is used in a 20-μl reaction mixture for 1st strand cDNA synthesis using oligo(dT)18 Mhoyan et al., 1997, Transplantation, 64:1665-70; Shirwan et al., 1998, Transplantation, 66:1802-9. Two μl of this reaction mixture is amplified using IL-12, IL-10, and IFN-γ, TGF-β, and CCL21-specific 5′ and 3′ primers. The PCR conditions are; denaturation at 94° C. for 45 sec, annealing at 60° C. for 60 sec, and extension at 72° C. for 90 sec for 25-38 cycles. Each reaction mixture also includes known quantities of a plasmid vector containing an insert with sequences for primer pairs for each cytokine. This serves as a competitive template to estimate the copy number of transcripts for each cytokine in a given sample. The PCR products are fractionated in 1% agarose, visualized by EtBr staining, and analyzed using the BioRad GelDoc system. The intensity of bands corresponding to the competitor and test is estimated by densitometric scanning and used to estimate the total numbers of cytokine-specific transcripts/μg total RNA. The expression of mouse HPRT is used to determine the quantity of RNA and also to normalize the amount of cDNA used for various cytokine analyses. Complete PCR mixtures without cDNA serve as negative controls.
Flow cytometry. For binding of immune co-stimulatory fusion proteins to their receptors, mouse splenocytes are isolated using Lympholyte®-M (Accurate Chemical and Scientific Corporation, Westbury, N.Y.) and incubated with various concentrations (10-1000 ng/ml) of the fusion proteins followed by staining with fluorescein-labeled antibodies against each molecule in combination with antibodies to CD4, CD8, and CD11c in flow cytometry. Cells incubated with recombinant core SA serve as negative controls. For general purposes, multicolor flow cytometric analyses is performed by first titrating the primary and secondary antibodies of interest and then using the optimum concentrations in flow cytometry using standard procedures. See Mhoyan et al., 1997, Transplantation, 64:1665-70. Isotype-matched antibodies serve as negative controls. Samples are run on a FACS Calibur or Vantage Sorter (Becton Dickinson). Analysis is performed using FlowJo or CellQuest software.
Statistics. The effect of experimental treatments on tumor survival is estimated using Kaplan-Meier curves. The differences in survival between different groups is assessed using the log-rank test (generalized Savage/Mantel Cox). Procedures involving the comparison of experimental data from groups of individual animals have the equality of variance examined using the F test (two groups) or Levene's test (multiple groups). If variances are not equal, log transformations are performed. When normally distributed sample means are to be compared, the Student's t test (two groups) or the Newman-Keuls test (multiple groups) is used. When the data is not normally distributed, the Mann-Whitney U test (two groups) or the Kruskal-Wallis test (multiple groups) is used. Statistical significance is defined as P<0.05.
Results herein demonstrate that (1) tumor and primary cells can effectively be biotinylated under physiological conditions and biotin persists on the cell surface for weeks; (2) biotinylated cells can be engineered with several fusion proteins that then persist on the cell surface in vivo for extended periods of time (days to weeks); (3) subcutaneous injection of live LLC or A20 tumor cells causes solid tumors in syngeneic mice. The data show that tumors can be engineered in vivo for the display of CD80.
A. Cell Surface Modification
A series of studies was performed using EZ-Link™ Sulfo-NHS-LC-Biotin derivative from Pierce Biotechnology, Inc. (Rockford, Ill.) to establish biotinylation conditions that do not compromise the viability and functions of the cells. Biotinylation in the range of 5-50 μM satisfied these criteria and provided an effective platform for the optimum display of exogenous proteins. Yolcu et al., 2002, Immunity, 17:795-808.
To assess the kinetics of biotin persistence on the cell surface in vivo, rat splenocytes were biotinylated in vitro (5-15 μM), labeled with the intracellular dye carboxyfluorescein diacetate succinimidyl ester (CFSE), and injected intravenously into syngeneic animals. Splenocytes were harvested at various times post-injection and CFSE positive cells were analyzed by flow cytometry for the presence of biotin on the cell surface using streptavidin-allophycocyanin (strepavidin-APC). The t1/2 of biotin on the cell surface in vivo was found to be greater than 15 days. See
Persistance of biotin on cell membrane was evaluated for splenocytes labeled with 15 μM biotin and cultured for various durations in the absence or presence of ConA. The cells were harvested and analyzed using flow cytometry for streptavidin reactive biotin. Biotin was present on the cell surface for extended periods of time with t1/2 of >20 and 18 days for resting (
A similar degree of biotin persistence was also observed using rat aortic endothelial cells. The kinetics of biotin cell surface stability for endothelial cells was biphasic with an initial estimated t1/2 of 7 days, followed by a longer t1/2 of more than 20 days.
Biotin persistence also was demonstrated using A20 cells. One million A20 cells were injected subcutaneously into the right flank of naïve Balb/c mice. Tumor growth was monitored and when the tumor size reached 6-8 mm diameter, different doses of biotin solution (0.5 mM, 1 mM and 5 mM in 300 μl PBS) were injected intratumorally. Three animals per group were euthanized at day 1, day 3, day 5 and day 7 after biotin injection. Tumors were surgically removed and mechanically processed into single cell suspension. One million cells were stained with streptavidin-APC and unbiotinylated tumor cells were used as negative control. Cells were analyzed using flow cytometry and CellQuest software. Biotin solution was prepared as 5 mM stock freshly on day 0 by dissolving 0.83 mg EZ-Link Sulfo-NHS-LC-Biotin (Mol. weight. 556.59, Pierce) per 300 μl PBS. Biotin stock was diluted (1:5 and 1:10) with PBS for lower concentrations.
The results are depicted in
These data demonstrate that a bifunctional molecule comprising biotin as the first binding pair member can persist on the cell surface for weeks or months in vivo and thus can be used to complex and, therefore, display on cell surfaces, fusion proteins comprising streptavidin or a functional fragment thereof linked to immune co-stimulatory polypeptides for an extended period of time.
B. Ex vivo Cell Surface Engineering
Immune co-stimulatory fusion proteins can be co-displayed on the surface of tumor cells that have been labeled with biotin ex vivo.
One million LLC cells were biotinylated using 15 μM of EZ-Link™ Sulfo-NHS-LC-Biotin derivative from Pierce Biotechnology, Inc. (Rockford, Ill.) in PBS at room temperature for 20-30 minutes. Biotin was then removed by centrifugation, cells were washed once, transferred into a clean second tube, washed 1-2 more times, and then contacted with CD80-CSA and 4-1BBL-CSA fusion proteins at 100 ng/protein/106 cells in PBS for 30-45 minutes on ice. Cells were then washed extensively and used for staining with saturating concentrations of various fluorochrome-conjugated antibodies to CD80 and 4-1BBL in PBS by incubation on ice for 30-45 minutes. Cells were then washed extensively with FACS medium and used for flow cytometric analysis. Both fusion proteins were co-displayed on the surface of tumor cells at comparable levels.
A20 cells were biotinylated (5 μM) and then contacted with CD80-CSA and 4-1BBL-CSA fusion proteins at 100 ng/protein/106 cells, following procedures similar to those set forth above. Both fusion proteins were co-displayed on the surface of tumor cells at comparable levels (
C. In vivo Kinetics of Protein Display On Ex vivo Engineered Tumors
The kinetics of persistence of the exogenous proteins on the cell surface was determined in vivo. Splenocytes were modified with biotin (15 μM), engineered with an CD80-CSA fusion protein (100 ng/1×106 cells), labeled with CFSE, and injected intravenously into syngeneic animals. Splenocytes and LN cells were harvested various days post-injection and analyzed for the presence of fusion protein on CFSE positive cells. CD80-CSA persisted on the surface of splenocytes with a t1/2 of >10 days.
These data demonstrate that tumor cells engineered with immune co-stimulatory fusion proteins can retain those proteins on the cell surface in vivo for weeks without any detectable toxicity to the engineered cell or the host patient (or test animal).
D. High Display Level Achieved By Ex vivo Engineering
To estimate the number of CD80-CSA molecules displayed on the cell as a function of various protein concentrations, A20 B lymphoma cells were biotinylated with 5 μM biotin and engineered with CD80 (100 ng/106 cells) in duplicate in 100 μL saline at 4° C. for 30 min. Cells were stained with phycoerythrin (PE) conjugated anti-human CD80 antibody. The Quantum-27 MESF kit from Bangs Laboratories was used to estimate the number of CD80-CSA molecules per cell. Molecules of equivalent soluble fluorochrome (MESF) for the samples were calculated according to the calibration curve constructed by five MESF beads with varying MESF values (0, 727, 2737, 10110, 41738) using Cell Quest analysis program (Becton Dickinson Biosciences). The data for each sample was converted into MESF values.
There was a direct correlation between the number of CD80-CSA molecules on the cell surface and the amounts of the protein used up to 100 ng. At this protein concentration, which was used for cancer vaccines in pre-clinical studies (see below), there is an average of 118,000 molecules per cell, which should be contrasted with 6,000-12,000 CD80 molecules per activated B cells or DC under physiological conditions, see Bergwelt-Baildon et al., 2002, Blood, 99:3319-25.
E. Auto/Alloreactive Responses Induced By CD80-SA Engineered Cells
To test whether CD80-CSA is functional and generates T cell immune responses, MLTR assays were performed using primary tumor cells and an established tumor cell line displaying CD80-SA. The human tumor cell line (HEC-1-A, endometrial cancer) engineered with CD80-CSA generated potent proliferative responses in peripheral blood lymphocytes from healthy volunteers. The proliferative response was CD80-CSA specific since non-engineered tumor cells or biotinylated cells treated with S2 culture supernatant elicited only a minimal proliferative response. The observed proliferative response was associated more with CD8+ T cells than CD4+ T cells.
F. Immunomodulation by 4-1BBL Engineering
CD4+ and CD8+ T cells were FACS-sorted from spleens and lymph nodes of naïve C57B/6 (H2-Kb) mice and were stained with carboxyfluorescein diacetate succinimidylester (CFSE). Briefly, cells were washed with PBS, incubated in 4 ml of 2.5 μM CFSE/1×106 cells for 7 minutes at room temperature. Cells were then incubated in two volumes of FBS for 1 min, and washed 2 times with PBS. A20 B lymphoma cells (H2-Kd) were washed in PBS and biotinylated using freshly prepared EZ-Link Sulfo-NHS-LC-Biotin in PBS (1 ml solution per 50×106 cells) for 30 min at room temperature. After washing with PBS, cells were irradiated at 10000 cGy. Biotinylated and irradiated A20 cells were incubated with CSA-4-1BBL (100 ng/1×106cells) or control protein CSA (38 ng/1×106 cells) in 5 ml tubes for 30 minutes in a cold room. After washing the cells extensively, some of the cells were stained with fluorochrome conjugated antibodies against 4-1BBL and streptavidin to detect the decoration, or with fluorochrome conjugated streptavidin to detect biotinylation. CFSE labeled sorted CD4+ or CD8+ T cells (1×105/well were co-cultured with irradiated A20 cells, irradiated/biotinylated A20 cells or irradiated/biotinylated/decorated A20 cells in a 4:1 ratio for 4 days in a 96-well plate.
Proliferation of T cells were determined using flow cytometry after staining cells with fluorochrome conjugated anti-CD4 or anti-CD8 antibodies. There was a vigorous proliferative response against A20 cells decorated with 4-1BBL (52.6%) as compared with controls decorated with CSA (8.6%) (
These data demonstrate that the use of4-1BBL in accordance with the present invention achieves effective immunomodulation because engineering cells with 4-1BBL boosts Teff functions.
G. In vivo Engineering of Solid Tumors
One million live A20 B lymphoma cells were injected subcutaneously to the right flank of naïve syngeneic Balb/c mice. When the tumor size reached 7-9 mm in diameter, 300 μL of 1 mM EZ-Link Sulfo-NHS-LC-Biotin in PBS was injected into the tumor, followed 24 hrs later by intratumoral injection of 40 μg of CD80-CSA fusion protein in 150 μL PBS. Tumor cells were harvested and analyzed 24 hrs after protein injection. Biotin levels were detected by streptavidin-APC and fusion protein levels were detected by an antibody against CD80. Cells from untreated tumors and biotinylated tumors were used as negative controls for biotin and CD80, respectively.
98% of tumor cells were positive for biotin 48 hrs after the injection of biotin and 81% were positive for CD80 24 hrs after the injection of CD80 fusion protein.
These results indicate that tumors can be engineered in vivo with immune co-stimulatory polypeptides that will persist for an extended period of time on the surface of tumor cells in vivo.
A similar protocol was followed to evaluate the effect of the timing of the CD80-CSA administration subsequent to biotinylation. 1×106 A20 cells were injected subcutaneously into the right back flank of naïve BALB/c mice. Tumor growth was monitored and measured using calipers three times a week. When the tumor size reached 7-9 mm average diameter, 300 μl of 5 mM biotin solution (EZ-Link Sulfo-NHS-LC-Biotin in PBS) was injected into the tumors. At 2, 4, 8 or 16 hours after biotin injection, biotinylation was assessed as described above and 50 μg of CD80-SA was administered by intratumoral injection. (Some animals were left untreated as negative controls.) Animals were euthanized 16 hours after CD80-SA injection. Tumors were processed into single cell suspensions, and fusion protein levels were detected by an antibody against CD80 using flow cytometry. The results (
This Example describes the engineering of tumor surfaces in vivo for effective and durable display of a combination of two or more immune co-stimulatory polypeptides, such as two or more of LIGHT, CD80, 4-1BBL, CD40L, OX40L, PD-L1 and GL50 polypeptides.
A. Biotinylation
An effectivce biotin dose for the biotinylation of tumor cells in vivo and the kinetics of biotin persistence on the surface of such tumor cells can be determined. An effective level of biotin conjugation in vivo can be defined as (i) biotinylation of >50% of tumor cells; (ii) biotin persistence on the surface of at least 50% of the tumor cells initially positive for biotin 5 days post-biotinylation, wherein the mean fluorescence intensity of the biotin positive cells is >70; and (iii) no detectable toxicity for either the tumor or the patient.
Doses of biotin that satisfy these criteria should permit display of >100,000 immune co-stimulatory fusion protein molecules per tumor cell immediately after biotinylation and >12,000 molecules per cell on day 5 post-biotinylation. The latter number is based on the physiological expression of CD80 on the surface of activated APCs. See Bergwelt-Baildon et al., 2002, Blood, 99:3319-25.
Inasmuch as T cell responses (including anti-tumor immune responses) take place 24-72 hrs after the initial antigenic challenge and not all tumor cells need to express co-stimulatory molecules to generate effective anti-tumor responses, the biotinylation criterion set forth above, when used in accordance with the present invention, is expected to achieve a potent anti-tumor immune response. For example, the 5 day and 50% remaining biotin-positive tumor cells parameters are expected to ensure that engineered tumor cell surfaces display the immune co-stimulatory polypeptides beyond the 5-day period and, if needed, may be re-engineered with additional immune co-stimulatory polypeptides without further biotinylation, as described above.
The extent of biotinylation may be measured in tumors following intratumoral injection with, for example, 100-200 μL of various concentrations (e.g., 0, 0.1, 0.5, 1.0, 5.0, 10.0, 50.0 mM) of EZ-Link™ Sulfo-NHS-LC-Biotin. (Larger amounts of solution can be used depending on tumor size.) Tumors are then harvested at various time points post-injection, prepared into single cell suspensions, and analyzed by flow cytometry for the presence of biotin (using, for example, SA-APC) on tumor cells, tumor infiltrating host cells such as T cells (CD3), B cells (CD19), and DC (CD11c).
Peripheral blood cells and tumor draining LNs also may be analyzed for the presence of biotin positive cells, which, if present, could be due to biotin leakage and decoration of peripheral cells and/or migration of biotinylated cells from the tumors into the periphery. Patients are monitored for survival and tumor growth, respectively, to determine whether the tested biotin doses are associated with any detectable toxicity.
Increasing the dose of biotin will increase the level of biotin displayed on the surface of tumor cells (and host cells present within the tumor). Similarly, higher doses of administered biotin result in longer kinetics of biotin persistence on targeted cells. Given that the cell surface kinetics of biotin is mostly regulated by the normal turnover of cell surface molecules to which biotin is conjugated as well as the dilution of biotin as a function of cell division, slower kinetics of biotin turnover will be observed for cells that are quiescent, such as DC.
B. Characterization of Immune Co-Stimulatory-CSA Fusion Polypeptides
Fusion polypeptides comprising an immune co-stimulatory polypeptide and streptavidin are generated, characterized functionally and structurally in vitro, and purified in large quantities, as described in more detail below.
A number of fusion polypeptides comprising an immune co-stimulatory polypeptide and CSA have been cloned and expressed. The sequences of these and other exemplary fusion polypeptides and their nucleotide sequences are set forth in
LIGHT is cloned from activated (PMA and ionomycin) mouse splenocytes and positioned C-terminus to core streptavidin in the DES expression vector. See Yolcu et al., 2002, Immunity, 17:795-808; Askenasy et al., 2003, Circulation, 107:41-7; Singh et al., 2003, Cancer Res., 63:4067-73. Drosophila S2 cells are transfected with 20 μg of plasmid DNA using the Calcium Phosphate Transfection kit according to the manufacturer's instructions (Invitrogen, San Diego, Calif.) to establish stable transfectants. Expression of fusion proteins is induced with 0.5 mM copper sulfate. Fusion proteins are purified in large quantities using Ni-NTA columns and the 6×His-tag.
Purified fusion proteins are tested for binding to their respective receptors on T cells using antibodies against mouse 4-1BBL and LIGHT molecules and flow cytometry as discussed in Example 1 above. Inasmuch as the expression of 4-1BB is inducible and HVEM is constitutively expressed on T cells (Granger & Rickert, 2003, Cytokine Growth Factor Rev., 14:289-96), splenocytes left unactivated or activated for 3 days with ConA are used. Activated and unactivated splenocytes are isolated using Lympholyte®-M (Accurate Chemical and Scientific Corporation, Westbury, N.Y.) and incubated with various concentrations (10-1000 ng/ml) of the fusion proteins. Following incubation, cells are stained with fluorescein-labeled antibodies against the respective fusion protein as well as with a combination of antibodies against various cell surface molecules, such as CD4 and CD8, for flow cytometry analysis. Cells incubated with recombinant core SA serve as negative controls.
The ability of LIGHT to activate tumor fibroblasts in vitro can be examined in an animal model as follows. Briefly, B6 mice are inoculated subcutaneously with 1×105 live syngenic LLC tumor cells, and tumors are surgically re-sected after reaching 15 mm in diameter. Tumors are mechanically processed into single cell suspensions and cultured in the presence of varying amounts of LIGHT protein (1-5000 ng/ml) for various periods of time (2-5 days). Culture supernatants are harvested and tested for the presence of CCL21 chemokine by ELISA as discussed in Example 1. Cultures without LIGHT or with CSA control protein, and those incubated with plate-bound LIGHT, serve as negative and positive controls, respectively.
Immune co-stimulatory fusion polypeptides displayed on the surface of LLC cells can be tested for ability to function as co-stimulatory molecules for immune activation. In an animal model, LLC cells minimally express MHC class I molecules and fail to express detectable levels of class II antigens or co-stimulatory molecules (e.g., LIGHT, CD80, or 4-1BBL). As such, LLC cells are unable to stimulate allogeneic T cells. A mixed lymphocyte reaction (MLR) assay is used to test whether fusion proteins can convert tumor cells into professional APCs for the generation of a productive T cell response. Briefly, tumor cells (1×106) are biotinylated using EZ-Link™ Sulfo-NHS-LC-Biotin (5 μM) in 1 ml of PBS for 30 min at room temperature. After several washes with PBS, biotinylated cells are incubated with immune co-stimulatory-CSA fusion polypeptides (50-100 ng/protein/106 cells) either individually or in various combinations for 45 min at 4° C. After washing several times with PBS, the presence of chimeric molecules on the cell surface is determined by flow cytometry using appropriate fluorescein-conjugated monoclonal antibodies. These cells are irradiated at 100 Gy and used as stimulator at 1:1 ratio with lymphocytes harvested from naïve syngeneic (B6) or allogeneic (BALB/c) animals. Cultures are pulsed for 18 hrs with 3[H] thymidine, and then harvested to measure cell proliferation as assessed by DNA-incorporated radioactivity. Cells are harvested on days 3, 4, and 5 post-culture to determine the kinetics of proliferation. Controls include responders alone, stimulators alone, and responders co-incubated with tumor cells decorated with control SA.
The fusion polypeptide containing the LIGHT molecule can stimulate tumor stromal cells for the secretion of CCL21 and, when displayed on the surface of tumor cells, can generate effective T cell proliferative responses in vitro.
C. Engineering with Combinations of Co-Stimulatory Polypeptides
Immune co-stimulatory polypeptides such as LIGHT, CD80 4-1BBL, CD40L, OX40L, PD-L1 and GL50 can function together in an additive and potentially a synergistic manner. This means that the observed proliferative effect is greater when tumor cells are engineered with a combination of two or more immune co-stimulatory polypeptides that when only one of the immune co-stimulatory polypeptides is displayed on the cell surface.
D. Evaluation of Combination In vivo Engineering
The following may be used to determine the level and persistence of chimeric LIGHT, CD80, 4-1BBL, CD40L, OX40L, PD-L1 or GL50 proteins on the surface of tumor cells engineered in vivo. These studies also may be used to examine whether proteins can be displayed on the surface of tumor cells in combinations and/or sequentially, and determine their cell surface turnover kinetics.
A biotin concentration that modifies >95% of tumor cells and allows for the persistence of biotin on at least 50% of the tumor cells for 5 days is used for in vivo tumor engineering. Various time points post biotin injection are assessed for engineering with immune co-stimulatory fusion protein because free biotin within the tumor microenvironment may compete with cell surface biotin for binding to fusion proteins. It was observed that biotinylation with I mM biotin followed by injection of 40 μg of CD80 24 hrs later was effective in displaying proteins on >80% of biotin positive cells. For example, two doses of immune co-stimulatory polypeptide fusion protein (20 and 200 μg/tumor) is tested to determine the shortest time duration after biotinylation at which the proteins can be administered into the tumor without a major effect on its display efficiency on tumor cells. This time point is used throughout the rest of the studies for the intratumoral injection of fusion proteins post-biotinylation.
Biotinylated tumors are injected with various concentrations of the immune co-stimulatory fusion polypeptides, singly, in combinations, or sequentially. Tumors are harvested on various days post-protein injection, prepared into single cell suspension, and analyzed in multicolor flow cytometry using fluorescent-conjugated antibodies against each protein. Seven animals/per group are used. Animals are closely monitored for survival and tumor growth.
Higher doses of fusion protein are expected to result in a higher level of protein display on the surface of biotinylated tumor cell surfaces and to longer kinetics of protein persistence, as long as biotin does not become saturated. Once the biotin sites are saturated, these two parameters will plateau. Given that the cell surface kinetics of biotin is mostly regulated by the normal turnover of cell surface molecules to which biotin is conjugated and dilution of biotin as a function of cell division, fusion protein levels are expected to vary. In addition, the kinetics of cell surface turnover are expected to be partly protein-dependent based on our previously published studies.
It should be remembered that tumor shrinkage may be observed as a result of anti-tumor immunity elicited by the immune co-stimulatory proteins, which may complicate long-term studies and accurate assessment of protein turnover kinetics. This effect may be amplified when all three proteins are co-displayed simultaneously or sequentially.
E. Therapeutic Efficacy of In vivo Engineered Tumor Cells
The therapeutic efficacy of tumor cells engineered in vivo with combinations of two or more immune co-stimulatory polypeptides, such as LIGHT, CD80, 4-1BBL, CD40L, OX40L, PD-L1 and GL50 is assessed as follows.
As set forth above, the codisplay of LIGHT, CD80, 4-1BBL, CD40L, OX40L, PD-L1 and/or GL50 within the tumor will not only stimulate the tumor stroma to secrete various chemokines required for the recruitment of lymphocytes to the tumor site, but also convert tumor cells into competent APC that can directly activate tumor infiltrating T cells by providing costimulatory signals. This can be shown using the LLC solid lung tumor model, which is chosen for its aggressive growth, metastatic behavior, and its lack of an optimum level of costimulatory molecules.
Animals are monitored for several immunological parameters, with an emphasis on CD8+ T cells, chemokines, and Th1 cytokine responses. These responses are correlated to the efficacy of immunotherapy.
Cancer cells frequently lack or manipulate co-stimulatory signals for the purpose of tumor evasion. In addition, the tumor stroma contributes to tumor progression by serving as a physical and immunological barrier. However, the tumor stroma expresses various immunological receptors such as LTβR. Engagement of LIGHT with LTβR on tumor stromal cells stimulates fibroblasts within the tumor stroma to synthesize and secrete various chemokines, such as CCL21, Mig, and IP 10. Therefore, the co-display of LIGHT, CD80, 4-1BBL, CD40L, OX40L, PD-L1 and/or GL50 within the tumor stimulates the tumor stroma to secrete various chemokines required for the recruitment of lymphocytes to the tumor site, and also converts tumor cells into competent APC that can directly activate tumor infiltrating T cells by providing co-stimulatory signals. The combination of one or more immune co-stimulatory polypeptides used to engineer tumor cells can modulate the tumor stroma and convert its immunosuppressive function to an immunostimulatory one.
Mechanistic studies are performed to identify immune parameters associated with effective therapy. The mechanistic studies include (1) the phenotyping of T cells infiltrating tumors using CD4, CD8, and CD25 markers; and (2) analysis of intratumoral IFN-γ and IL-12 as signature cytokines for ThI responses, IL-10 and TGF-β as immune inhibitory cytokines, and CCL21 chemokine.
The CCL21 chemokine is synthesized as a result of LIGHT engagement with LTβR on tumor stromal cells. Briefly, tumors are re-sected from animals subjected to protein display regimens that result in effective immunotherapy on various days (3, 6, 9) after in vivo engineering with immune co-stimulatory fusion proteins and mechanically processed into single cell suspensions. A portion of the cells are used for total RNA extraction while the remaining cells are stained with different fluorescent-conjugated antibodies against CD4, CD8, and CD25, and analyzed in multicolor flow cytometry.
Cytokine analysis is performed using the cytokine-specific primers in RT-PCR as described in Example 1 above. In addition, CD8+ T cell cytotoxic responses against tumor cells are examined. Briefly, tumor draining LN cells are separately harvested from animals with successful and unsuccessful immunotherapy, and co-incubated with irradiated LLC cells in the presence of IL-2 for 5 days. Effector cells are harvested using Ficoll gradients and tested against live LLC cells at various effector to target ratios in a standard cytotoxicity assay. Singh et al., 2003, Cancer Res., 63:4067-73. Tumors from animals that failed immunotherapy are subjected to similar analyses. A20 B cell lymphoma cells from BALB/c are used as third party targets to determine the antigen specificity of the killing response.
Engineering tumor cell surfaces with LIGHT alone may facilitate the recruitment of naïve T cells and DC into the tumor site and induce the activation and differentiation of these cells to gain effector functions. However, the co-display of LIGHT, CD80 and 4-1BBL is expected to provide more effective tumor immunotherapy because of the synergistic effect of these molecules in generating Th I responses and the ability of 4-1BBL to activate DC for the synthesis and elaboration of Th1 type cytokines. Furthermore, 4-1BBL can also contribute to long-term memory and effector function of CD8+ T cells for a long-lasting protection.
The complete eradication of tumors may require repeated introduction of the immune co-stimulatory polypeptides into the tumor. This can be accomplished within the first week of biotinylation without re-biotinylation of the tumors. At later time periods, tumors can be re-biotinylated and then decorated with the fusion proteins again.
Failure to eradicate tumors may be due to the induction of antibodies against the fusion protein, in particular, antibodies against the SA portion of the fusion protein. Direct injection of the immune co-stimulatory-SA fusion proteins into tumors is likely to be less sensitive to antibody-blocking compared with systemic injections. This is consistent with previously published results with a B lymphoma model. See Singh et al., 2003, Cancer Res., 63:4067-73, which demonstrated that tumor cells manipulated ex vivo to display CD80 molecule can be used as efficient tumor vaccines.
Alternatively, immune co-stimulatory fusion proteins can be generated with avidin instead of streptavidin. Avidin does not share extensive homology at the protein level with streptavidin. Therefore, fusion proteins containing either streptavidin or avidin could be used interchangeably to minimize neutralizing antibodies. Additionally or alternatively, as descirbied above, modified forms of avidin and streptavidin with reduced immunogenicity, such as forms modified by removal of lymphocyte epitopes, can be used in accordance with the invention.
High numbers of intratumoral CD4+ and CD8+ T cells with activated phenotypes (CD25+ ) and high levels of IFN-γ, IL-12, and CCL21 transcripts are observed in animals with effective immunotherapy. In marked contrast, animals that fail therapy are expected to show only minimal lymphocyte infiltration within the tumors and express high levels of IL-10 and TGF-β. Similarly, effective therapy is expected to generate an effective CD8+ T cell killing response against tumors.
While the invention has been described and exemplified in sufficient detail for those skilled in this art to make and use it, various alternatives, modifications, and improvements should be apparent without departing from the spirit and scope of the invention. The examples provided herein are representative of preferred embodiments, are exemplary, and are not intended as limitations on the scope of the invention. Modifications therein and other uses will occur to those skilled in the art. These modifications are encompassed within the spirit of the invention and are defined by the scope of the claims.
It will be readily apparent to a person skilled in the art that varying substitutions and modifications may be made to the invention disclosed herein without departing from the scope and spirit of the invention.
All patents and publications mentioned in the specification are indicative of the levels of those of ordinary skill in the art to which the invention pertains. All patents and publications are herein incorporated by reference to the same extent as if each individual publication was specifically and individually indicated to be incorporated by reference.
The invention illustratively described herein suitably may be practiced in the absence of any element or elements, limitation or limitations which is not specifically disclosed herein. Thus, for example, in each instance herein any of the terms “comprising”, “consisting essentially of” and “consisting of” may be replaced with either of the other two terms. The terms and expressions which have been employed are used as terms of description and not of limitation, and there is no intention that in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by preferred embodiments and optional features, modification and variation of the concepts herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention as defined by the appended claims.
Other embodiments are set forth within the following claims.
The present application claims the benefit of the filing dates of the following U.S. provisional applications: 60/748,177 (filed Dec. 8, 2005); 60/771,179 (filed Feb. 6, 2006); 60/799,643 (filed May 12, 2006); and 60/863,173 (filed Oct. 27, 2006). Each of the foregoing applications is incorporated by reference herein in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 60748177 | Dec 2005 | US | |
| 60771179 | Feb 2006 | US | |
| 60799643 | May 2006 | US | |
| 60863173 | Oct 2006 | US |
