IMMOBILIZED TUMOR NECROSIS FACTOR-ALPHA MUTEINS FOR ENHANCING IMMUNE RESPONSE IN MAMMALS

FIELD OF THE INVENTION

This invention relates generally to the field of immunotherapy and, more specifically, to methods for enhancing host immune responses.

BACKGROUND OF THE INVENTION

The immune system of mammals has evolved to protect the host against the growth and proliferation of potentially deleterious agents. These agents include infectious microorganisms such as bacteria, viruses, fungi, and parasites which exist in the environment and which, upon introduction to the body of the host, can induce varied pathological conditions. Other pathological conditions may derive from agents not acquired from the environment, but rather which arise spontaneously within the body of the host. The best examples are the numerous malignancies known to occur in mammals. Ideally, the presence of these deleterious agents in a host triggers the mobilization of the immune system to effect the destruction of the agent and, thus, restore the sanctity of the host environment.

The destruction of pathogenic agents by the immune system involves a variety of effector mechanisms which can be grouped generally into two categories: innate and specific immunity. The first line of defense is mediated by the mechanisms of innate immunity. Innate immunity does not discriminate among the myriad agents that might gain entry into the host's body. Rather, it responds in a generalized manner that employs the inflammatory response, phagocytes, and plasma-borne components such as complement and interferons. In contrast, specific immunity does discriminate among pathogenic agents. Specific immunity is mediated by B and T lymphocytes and it serves, in large, part, to amplify and focus the effector mechanisms of innate immunity.

The elaboration of an effective immune response requires contributions from both innate and specific immune mechanisms. The function of each of these arms of the immune system individually, as well as their interaction with each other, is carefully coordinated, both in a temporal/spatial manner and in terms of the particular cell types that participate. This coordination results from the actions of a number of soluble immunostimulatory mediators or “immune system stimulators” (reviewed in, Trinchieri et al., J. Cell Biochem. 53:301-308 (1993)). Certain of these immune system stimulators initiate and perpetuate the inflammatory response and the attendant systemic sequelae. Examples of these include, but are not limited to, the proinflammatory mediators tumor necrosis factors α and β, interleukin-1, interleukin-6, interleukin-8, interferon-.gamma., and the chemokines RANTES, macrophage inflammatory proteins 1-α and 1-β and macrophage chemotactic and activating factor. Other immune system stimulators facilitate interactions between B and T lymphocytes of specific immunity. Examples of these include, but are not limited to, interleukin-2, interleukin-4, interleukin-5, interleukin-6, and interferon-.gamma. Still other immune system stimulators mediate bidirectional communication between specific immunity and innate immunity. Examples of these include, but are not limited to, interferon-7, interleukin-1, tumor necrosis factors α and β, and interleukin-12. All of these immune system stimulators exert their effects by binding the specific receptors on the surface of host cells, resulting in the delivery of intercellular signals that alter the function of the target cell. Cooperatively, these mediators stimulate the activation and proliferation of immune cells, recruit them to particular anatomical sites, and permit their collaboration in the elimination of the offending agent. The immune response induced in any individual is determined by the particular complement of immune system stimulators produced, and by the relative abundance of each.

In contrast to the immune system stimulators described above, the immune system has evolved other soluble mediators that serve to inhibit immune responses (reviewed in Arend, Adv. Int. Med. 40:365-394 (1995)). These “immune system inhibitors” provide the immune system with the ability to dampen responses in order to prevent the establishment of a chronic inflammatory state with the potential to damage the host's tissues. Regulation of host immune function by immune system inhibitors is accomplished through a variety of mechanisms as described below.

First, certain immune system inhibitors bind directly to immune system stimulators and, thus, prevent them from binding to plasma membrane receptors on host cells. Examples of these types of immune system inhibitors include, but are not limited to, the soluble receptors for tumor necrosis factors α and β, interferon-.gamma., interleukin-1, interleukin-2, interleukin-4, interleukin-6, and interleukin-7.

Second, certain immune system inhibitors antagonize the binding of immune system stimulators to their receptors. By way of example, interleukin-1 receptor antagonist is known to bind to the interleukin-1 membrane receptor. It does not deliver activation signals to the target cell but, by virtue of occupying the interleukin-1 membrane receptor, blocks the effects of interleukin-1.

Third, particular immune system inhibitors exert their effects by binding to receptors on host cells and signaling a decrease in their production of immune system stimulators. Examples include, but are not limited to, interferon-β, which decreases the production of two key proinflammatory mediators, tumor necrosis factor α and interleukin-1 (Coclet-Ninin et al., Eur. Cytokine Network 8:345-349 (1997)), and interleukin-10, which suppresses the development of cell-mediated immune responses by inhibiting the production of the immune system stimulator, interleukin-12 (D'Andrea et al., J. Exp. Med. 178:1041-1048 (1993)). In addition to decreasing the production of immune system stimulators, certain immune system inhibitors also enhance the production of other immune system inhibitors. By way of example, interferon-α_2binhibits interleukin-1 and tumor necrosis factor α production and increases the production of the corresponding immune system inhibitors, interleukin-1 receptor antagonist and soluble receptors for tumor necrosis factors α and β (Dinarello, Sem. in Oncol. 24(3 Suppl. 9):81-93 (1997).

Fourth, certain immune system inhibitors act directly on immune cells, inhibiting their proliferation and function, thereby decreasing the vigor of the immune response. By way of example, transforming growth factor-β inhibits a variety of immune cells and significantly limits inflammation and cell-mediated immune responses (reviewed in Letterio and Roberts, Ann. Rev. Immunol. 16:137-161 (1998)). Collectively, these various immunosuppressive mechanisms are intended to regulate the immune response, both quantitatively and qualitatively, to minimize the potential for collateral damage to the host's own tissues.

In addition to the inhibitors produced by the host's immune system for self-regulation, other immune system inhibitors are produced by infectious microorganisms. For example, many viruses produce molecules which are viral homologues of host immune system inhibitors (reviewed in Spriggs, Ann. Rev. Immunol. 14:101-130 (1996)). These include homologues of host complement inhibitors, interleukin-10, and soluble receptors for interleukin-1, tumor necrosis factors α and β, and interferons α, β and γ. Similarly, helminthic parasites produce homologues of host immune system inhibitors (reviewed in Riffkin et al., Immunol. Cell Biol. 74:564-574 (1996)), and several bacterial genera are known to produce immunosuppressive products (reviewed in, Reimann et al., Scand. J. Immunol. 31:543-546 (1990)). All of these immune system inhibitors serve to suppress the immune response during the initial stages of infection, to provide advantage to the microbe, and to enhance the virulence and chronicity of the infection.

A role for host-derived immune system inhibitors in chronic disease also has been established. In the majority of cases, this reflects a polarized T cell response during the initial infection, wherein the production of immunosuppressive mediators (i.e., interleukin-4, interleukin-10, and/or transforming growth factor-β) dominates over the production of immunostimulatory mediators (i.e., interleukin-2, interferon-γ, and/or tumor necrosis factor β) (reviewed in Lucey et al., Clin. Micro. Rev. 9:532-562 (1996)). Overproduction of immunosuppressive mediators of this type has been shown to produce chronic, non-healing pathologies in a number of medically important diseases. These include, but are not limited to, diseases resulting from infection with: 1) the parasites, Plasmodium falciparum (Sarthou et al., Infect. Immun. 65:3271-3276 (1997)), Trypanosoma cruzi (reviewed in Laucella et al., Revista Argentina de Microbiolgia 28:99-109 (1996)), Leishmania major (reviewed in Etges and Muller, J. Mol. Med. 76:372-390 (1998)), and certain helminths (Riffkin et al., supra); 2) the intracellular bacteria, Mycobacterium tuberculosis (Baliko et al., FEMS Immunol. Med. Micro. 22:199-204 (1998)), Mycobacterium avium (Bermudez and Champsi, Infect. Immun. 61:3093-3097 (1993)), Mycobacterium leprae (Sieling et al., J. Immunol. 150:5501-5510 (1993)), Mycobacterium bovis (Kaufmann et al., Ciba Fdn. Symp. 195:123-132 (1995)), Brucella abortus (Fernandes and Baldwin, Infect. Immun. 63:1130-1133 (1995)), and Listeria monocytogenes (Blauer et al., J. Interferon Cytokine Res. 15:105-114 (1995)), and, 3) intracellular fungus, Candida albicans (reviewed in Romani et al., Immunol. Res. 14:148-162 (1995)). The inability to spontaneously resolve infection is influenced by other host-derived immune system inhibitors as well. By way of example, interleukin-1 receptor antagonist and the soluble receptors for tumor necrosis factors α and β are produced in response to interleukin-1 and tumor necrosis factor α and/or β production driven by the presence of numerous infectious agents. Examples include, but are not limited to, infections by Plasmodium falciparum (Jakobsen et al., Infect. Immun. 66:1654-1659 (1998); Sarthou et al., supra), Mycobacterium tuberculosis (Balcewicz-Sablinska et al., J. Immunol. 161:2636-2641 (1998)), and Mycobacterium avium (Eriks and Emerson, Infect. Immun. 65:2100-2106 (1997)). In cases where the production of any of the aforementioned immune system inhibitors, either individually or in combination, dampens or otherwise alters immune responsiveness before the elimination of the pathogenic agent, a chronic infection may result.

In addition to this role in infectious disease, host-derived immune system inhibitors contribute also to chronic malignant disease. Compelling evidence is provided by studies of soluble tumor necrosis factor receptor Type I (sTNFRI) in cancer patients. Nanomolar concentrations of sTNFRI are synthesized by a variety of activated immune cells in cancer patients and, in many cases, by the tumors themselves (Aderka et al., Cancer Res. 51:5602-5607 (1991); Adolf and Apfler, J. Immunol. Meth. 143:127-136 (1991)). In addition, circulating sTNFRI levels often are elevated significantly in cancer patients (Aderka et al., supra; Kalmanti et al., Int. J. Hematol. 57:147-152 (1993); Elsasser-Beile et al., Tumor Biol. 15:17-24 (1994); Gadducci et al., Anticancer Res. 16:3125-3128 (1996); Digel et al., J. Clin. Invest. 89:1690-1693 (1992)), decline during remission and increase during advanced stages of tumor development (Aderka et al., supra; Kalmanti et al., supra; Elsasser-Beile et al., supra; Gadducci et al., supra) and, when present at high levels, correlate with poorer treatment outcomes (Aderka et al., supra). These observations suggest that sTNFRI aids tumor survival by inhibiting anti-tumor immune mechanisms which employ tumor necrosis factors α and/or β (TNF), and they argue favorably for the clinical manipulation of sTNFRI levels as a therapeutic strategy for cancer.

Direct evidence that the removal of immune system inhibitors provides clinical benefit derives from the evaluation of Ultrapheresis, a promising experimental cancer therapy (Lentz, J. Biol. Response Modif. 8:511-527 (1989); Lentz, Ther. Apheresis 3:40-49 (1999); Lentz, Jpn. J. Apheresis 16:107-114 (1997)). Ultrapheresis involves extracorporeal fractionation of plasma components by ultrafiltration. Ultrapheresis selectively removes plasma components within a defined molecular size range, and it has been shown to provide significant clinical advantage to patients presenting with a variety of tumor types. Ultrapheresis induces pronounced inflammation at tumor sites, often in less than one hour post-initiation. This rapidity suggests a role for preformed chemical and/or cellular mediators in the elaboration of this inflammatory response, and it reflects the removal of naturally occurring plasma inhibitors of that response. Indeed, immune system inhibitors of TNF α and β, interleukin-1, and interleukin-6 are removed by Ultrapheresis (Lentz, Ther. Apheresis 3:40-49 (1999)). Notably, the removal of sTNFRI has been correlated with the observed clinical responses (Lentz, Ther. Apheresis 3:40-49 (1999); Lentz, Jpn. J. Apheresis 16:107-114 (1997)).

Ultrapheresis is in direct contrast to more traditional approaches which have endeavored to boost immunity through the addition of immune system stimulators. Preeminent among these has been the infusion of supraphysiological levels of TNF (Sidhu and Bollon, Pharmacol. Ther. 57:79-128 (1993)); and of interleukin-2 (Maas et al., Cancer Immunol. Immunother. 36:141-148 (1993)), which indirectly stimulates the production of TNF. These therapies have enjoyed limited success (Sidhu and Bollon, supra, Maas et al., supra) due to the fact: 1) that at the levels employed they proved extremely toxic; and 2) that each increases the plasma levels of the immune system inhibitor, sTNFRI (Lantz et al., Cytokine 2:402-406 (1990); Miles et al., Brit. J. Cancer 66:1195-1199 (1992)). Together, these observations support the utility of Ultrapheresis as a biotherapeutic approach to cancer—one which involves the removal of immune system inhibitors, rather than the addition of immune system stimulators.

Although Ultrapheresis provides advantages over traditional therapeutic approaches, there are certain drawback that limit its clinical usefulness. Not only are immune system inhibitors removed by Ultrapheresis, but other plasma components, including beneficial ones, are removed since the discrimination between removed and retained plasma components is based solely on molecular size. An additional drawback to Ultrapheresis is the significant loss of circulatory volume during treatment, which must be offset by the infusion of replacement fluid. The most effective replacement fluid is an ultrafiltrate produced, in an identical manner, from the plasma of non-tumor bearing donors. A typical treatment regimen (15 treatments, each with the removal of approximately 7 liters of ultrafiltrate) requires over 200 liters of donor plasma for the production of replacement fluid. The chronic shortage of donor plasma, combined with the risks of infection by human immunodeficiency virus, hepatitis A, B, and C or other etiologic agents, represents a severe impediment to the widespread implementation of Ultrapheresis.

Because of the beneficial effects associated with the removal of immune system inhibitors, there exists a need for methods which can be used to specifically deplete those inhibitors from circulation. Such methods ideally should be specific and not remove other circulatory components, and they should not result in any significant loss of circulatory volume. The present invention satisfies these needs and provides related advantages as well.

SUMMARY OF THE INVENTION

The present invention provides a method for stimulating immune responses in a mammal through the depletion of immune system inhibitors such as soluble TNF receptors present in the circulation of the mammal. The depletion of immune system inhibitors such as soluble TNF receptors can be effected by removing biological fluids from the mammal and contacting these biological fluids with a binding partner capable of selectively binding to the targeted immune system inhibitor, for example, TNFα muteins.

Binding partners useful in these methods are TNF α muteins having specificity for soluble TNF receptors. Moreover, mixtures of TNF α muteins having specificity for one or more soluble TNF receptors can be used.

In a particularly useful embodiment, the binding partner, such as a TNF α mutein, is immobilized previously on a solid support to create an “absorbent matrix” (FIG. 1). The exposure of biological fluids to such an absorbent matrix will permit binding by the immune system inhibitor such as soluble TNF receptor, thus, effecting a decrease in its abundance in the biological fluids. The treated biological fluid can be returned to the patient. The total volume of biological fluid to be treated and the treatment rate are parameters individualized for each patient, guided by the induction of vigorous immune responses while minimizing toxicity. The solid support (i.e., inert medium) can be composed of any material useful for such purpose, including, for example, hollow fibers, cellulose-based fibers, synthetic fibers, flat or pleated membranes, silica-based particles, macroporous beads, and the like.

In another embodiment, the binding partner such as TNFα mutein can be mixed with the biological fluid in a “stirred reactor” (FIG. 2). The binding partner-immune system inhibitor complex then can be removed by mechanical or by chemical or biological means, and the altered biological fluid can be returned to the patient.

The present invention also provides apparatus incorporating either the absorbent matrix or the stirred reactor.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 schematically illustrates an “absorbent matrix” configuration of an embodiment of the invention. In this example, blood is removed from the patient and separated into a cellular and an acellular component, or factions thereof. The acellular component, or fractions thereof, is exposed to the absorbent matrix to effect the binding and, thus, depletion of a targeted immune system inhibitor such as soluble tumor necrosis factor (TNF) receptor. The altered acellular component, or fractions thereof, then is returned contemporaneously to the patient.

FIG. 2 schematically illustrates a “stirred reactor” configuration of an embodiment of the invention. In this example, blood is removed from the patient and separated into a cellular and an acellular component, or fractions thereof. A binding partner such as a TNF α mutein is added to the acellular component, or fractions thereof. Subsequently, the binding partner (TNF α mutein)/immune system inhibitor (soluble TNF receptor) complex is removed by mechanical or by chemical or biological means from the acellular component, or fractions thereof, and the altered biological fluid is returned contemporaneously to the patient.

FIG. 3A shows an alignment of TNF α sequences from various mammalian species (mouse, SEQ ID NO:10; rat, SEQ ID NO:11; rabbit, SEQ ID NO:12; cat, SEQ ID NO:13; dog, SEQ ID NO:14; sheep, SEQ ID NO:15; goat, SEQ ID NO:16; horse, SEQ ID NO:17; cow, SEQ ID NO:18; pig, SEQ ID NO:19; human, SEQ ID NO:2). The top sequence shows the conserved amino acids across the shown species (SEQ ID NO:1) (completely conserved or with one exception). Non-conserved amino acids are indicated by “.” (taken from Van Ostade et al., Prot. Eng. 7:5-22 (1994), which is incorporated herein by reference). FIG. 3B shows an alignment of the conserved TNF α sequence with human TNF α and six representative TNF α muteins, designated mutein 1 (SEQ ID NO:3), mutein 2 (SEQ ID NO:4), mutein 3 (SEQ ID NO:5), mutein 4 (SEQ ID NO:6), mutein 5 (SEQ ID NO:7), and mutein 6 (SEQ ID NO:8). The four muteins differ from the human sequence by single amino acid substitutions, indicated with bold and underline. FIG. 3C shows a representative consensus TNFα: sequence (SEQ ID NO:9).

FIG. 4 shows the presence of human TNFα and TNFα muteins 1, 2, 3 and 4 in periplasmic preparations of Escherichia coli transformed with the respective expression constructs.

FIG. 5 shows that TNF α muteins bind to sTNFRI. Wells of a microtiter plate were coated with TNF α, blocked, and incubated with sTNFRI either in the presence or absence of the inhibitors, TNF α and TNF α muteins 1, 2 and 4.

FIG. 6 shows the depletion of soluble TNF receptor I (sTNFRI) by immobilized TNF muteins. Muteins 1, 2 and 4 were immobilized on Sepharose™ 4B, and normal human plasma spiked with recombinant human sTNFRI was passed through columns of the immobilized muteins. Depletion of sTNFRI from the serum was measured by enzyme-linked immunosorbent assay (ELISA).

DETAILED DESCRIPTION OF THE DRAWINGS

The present invention provides methods to reduce the levels of immune system inhibitors such as soluble TNF receptors in the circulation of a host mammal, thereby potentiating an immune response capable of resolving a pathological condition or decreasing the severity of a pathological condition. By enhancing the magnitude of the host's immune response, the methods of the present invention avoid the problems associated with the repeated administration of chemotherapeutic agents which often have undesirable side effects, for example, chemotherapeutic agents used in treating cancer.

The methods of the present invention generally are accomplished by: (a) obtaining a biological fluid from a mammal having a pathological condition; (b) contacting the biological fluid with a TNFα mutein binding partner capable of selectively binding to a targeted immune system inhibitor such as soluble TNF receptor to produce an altered biological fluid having a reduced amount of the targeted immune system inhibitor; and, thereafter (c) administering the altered biological fluid to the mammal.

As used herein, the term “immune system stimulator” refers to soluble mediators that increase the magnitude of an immune response, or which encourage the development of particular immune mechanisms that are more effective in resolving a specific pathological condition. Examples of immune system stimulators include, but are not limited to, the proinflammatory mediators tumor necrosis factors α and β, interleukin-1, interleukin-2, interleukin-4, interleukin-5, interleukin-6, interleukin-8, interleukin-12, interferon-.gamma., interferon-7; and the chemokines RANTES, macrophage inflammatory proteins 1-α and 1-β and macrophage chemotactic and activating factor, as discussed above.

As used herein, the term “immune system inhibitor” refers to a soluble mediator that decreases the magnitude of an immune response, or which discourages the development of particular immune mechanisms that are more effective in resolving a specific pathological condition, or which encourages the development of particular immune mechanisms that are less effective in resolving a specific pathological condition. Examples of host-derived immune system inhibitors include interleukin-1 receptor antagonist, transforming growth factor-β, interleukin-4, interleukin-10, or the soluble receptors for interleukin-1, interleukin-2, interleukin-4, interleukin-6, interleukin-7, interferon-y and tumor necrosis factors α and β. In a particular embodiment of the present invention, the immune system inhibitor is soluble TNF receptor Type I (sTNFRI) or Type II (sTNFRII). Immune system inhibitors produced by microorganisms are also potential targets including, for example, soluble receptors for tumor necrosis factor α and β. As used herein, the term “targeted” immune system inhibitor refers to that inhibitor, or collection of inhibitors, which is to be removed from the biological fluid by a method of the invention, for example, sTNFRI and/or sTNFRII.

As used herein, the term “soluble TNF receptor” refers to a soluble form of a receptor for TNFα and TNFβ. Two forms of TNF receptor have been identified, type I receptor (TNFRI), also known as TNF-R55, and type II receptor (TNFRII), also known as TNF-R75, both of which are membrane proteins that bind to TNFα and TNFβ and mediate intracellular signaling. Both of these receptors also occur in a soluble form. The soluble form of TNF receptor functions as an immune system inhibitor, as discussed above. As used herein, a soluble TNF receptor includes at least one of the soluble forms of TNFRI and TNFRII or any other type of TNF receptor. It is understood that, in the methods of the invention, the methods can be used to remove one or both types of TNF receptor depending on whether the TNFα mutein or plurality of muteins used in the method binds to one or both types of receptors.

As used herein, the term “mammal” can be a human or a non-human animal, such as dog, cat, horse, cattle, pig, sheep, non-human primate, mouse, rat, rabbit, or other mammals, for example. The term “patient” is used synonymously with the term “mammal” in describing the invention.

As used herein, the term “pathological condition” refers to any condition where the persistence, within a host, of an agent, immunologically distinct from the host, is a component of or contributes to a disease state. Examples of such pathological conditions include, but are not limited to those resulting from persistent viral, bacterial, parasitic, and fungal infections, and cancer. Among individuals exhibiting such chronic diseases, those in whom the levels of immune system inhibitors are elevated are particularly suitable for the treatment of the invention. Plasma levels of immune system inhibitors can be determined using methods well known in the art (see, for example, Adolf and Apfler, supra, 1991). Those skilled in the art readily can determine pathological conditions that would benefit from the depletion of immune system inhibitors according to the present methods.

As used herein, the term “biological fluid” refers to a bodily fluid obtained from a mammal, for example, blood, including whole blood, plasma, serum, lymphatic fluid, or other types of bodily fluids. If desired, the biological fluid can be processed or fractionated, for example, to obtain an acellular component. As it relates to the present invention, the term “acellular biological fluid” refers to the acellular component of the circulatory system including plasma, serum, lymphatic fluid, or fractions thereof. The biological fluids can be removed from the mammal by any means known to those skilled in the art, including, for example, conventional apheresis methods (see, Apheresis: Principles and Practice, McLeod, Price, and Drew, eds., AABB Press, Bethesda, Md. (1997)). The amount of biological fluid to be extracted from a mammal at a given time will depend on a number of factors, including the age and weight of the host mammal and the volume required to achieve therapeutic benefit. As an initial guideline, one plasma volume (approximately 3-5 liters in an adult human) can be removed and, thereafter, depleted of the targeted immune system inhibitor according to the present methods.

As used herein, the term “selectively binds” means that a molecule binds to one type of target molecule, but not substantially to other types of molecules. The term “specifically binds” is used interchangeably herein with “selectively binds.”

As used herein, the term “binding partner” is intended to include any molecule chosen for its ability to selectively bind to the targeted immune system inhibitor. The binding partner can be one which naturally binds the targeted immune system inhibitor. For example, tumor necrosis factor α or β can be used as a binding partner for sTNFRI. Alternatively, other binding partners, chosen for their ability to selectively bind to the targeted immune system inhibitor, can be used. Those include fragments of the natural binding partner, polyclonal or monoclonal antibody preparations or fragments thereof, or synthetic peptides. In a particular embodiment of the present invention, the binding partner is a TNFα mutein.

As used herein, the term “TNFα mutein” refers to a TNFα variant having one or more amino acid substitutions relative to a parent sequence and retaining specific binding activity for a TNF receptor. Generally, the muteins of the present invention have a single amino acid substitution relative to a parent sequence. Exemplary TNFα muteins include the human TNFα muteins designated muteins 1, 2, 3, 4, 5 and 6 (see FIG. 3B), which are derived from human TNFα but have a single amino acid substitution relative to the wild type sequence, as discussed below. It is understood that analogous muteins of species other than human are similarly included, for example, muteins analogous to muteins 1, 2, 3, 4, 5 or 6 in the other mammalian species shown in FIG. 3A, or other mammalian species. These and other muteins, as described in more detail below, are included within the meaning of a TNFα mutein of the invention.

The present invention provides compositions and method for stimulating or enhancing an immune response in a mammal. The invention advantageously uses ligands that bind to immune system inhibitors to counterbalance the dampening effect of immune system inhibitors on the immune response. Such ligands, also referred to herein as “binding partners,” can be attached to a solid support to allow the removal of an immune system inhibitor from a biological fluid.

A binding partner particularly useful in the present invention is a ligand that binds with high affinity to an immune system inhibitor, for example, soluble TNF receptor and in particular sTNFRI. Another useful characteristic of a binding partner is a lack of direct toxicity. For example, a binding partner lacking TNF agonist activity is particularly useful. Generally, even when a ligand such as a binding partner is covalently bound to a solid support, a certain percentage of the bound ligand will leach from the support, for example, via chemical reactions that break down the covalent linkage or protease activity present in a biological fluid. In such a case, the ligand will leach into the biological fluid being processed and, thus, be returned to the patient. Therefore, it is advantageous to use a ligand that has affinity for an immune system inhibitor but has decreased ability to stimulate a biological response, that is, has decreased or low agonist activity. In this case, even if some of the ligand leaches into the processed biological fluid, the ligand would still exhibit low biological activity with respect to membrane receptor signaling when reintroduced into the patient.

Yet another useful characteristic of a binding partner is a lack of indirect toxicity, for example, immunogenicity. As discussed above, it is common for a bound ligand to leach from a matrix, resulting in the ligand being present in the processed biological fluid. Because the biological fluid is returned to the patient, this results in the introduction of a low level of the ligand to the patient. If the ligand is immunogenic, an immune response against the ligand can be stimulated, resulting in undesirable immune responses, particularly in a patient in which the process is being repeated. Therefore, a ligand having low immunogenicity would minimize any undesirable immune responses against the ligand. As disclosed herein, a particularly useful ligand to be used as a binding partner of the invention is derived from the same species as the patient being treated. For example, for treating a human, a human TNFα mutein can be used as the binding partner, which is expected to have low immunogenicity given the homology to the endogenous TNFα. Similarly, muteins derived from other mammalian species can be used in the respective species.

As disclosed herein, TNFα muteins are particularly useful binding partners in methods of the invention. A number of TNFα muteins have been previously described (see, for example, Van Ostade et al., Protein Eng. 7:5-22 (1994); Van Ostade et al., EBMO J. 10:827-836 (1991); Zhang et al., J. Biol. Chem. 267:24069-24075 (1992); Yamagishi et al., Protein Eng. 3:713-719 (1990), each of which is incorporated herein by reference). Specific exemplary muteins include the human TNFα muteins shown in FIG. 3B.

There are several advantages to using TNFα muteins as binding partners in the present invention. Although TNFα muteins can display lower binding activity for TNF receptors, some TNFα muteins bind only 5- to 17-fold less effectively than native TNFα. Such a binding affinity, albeit reduced relative to native TNFα, can still be an effective binding partner in the present invention (see Example 3). Another advantage of using TNFα muteins is that some exhibit decreased signaling through membrane receptors, for example, decreased cytotoxic activity or in vivo toxicity, relative to native TNFα. In particular, muteins 1, 2, 3, 4, 5 and 6 exhibit a 200- to 10,000-fold decrease in cytotoxicity (see below and Van Ostade, supra, 1994; Yamagishi et al., supra, 1990; Zhang et al., supra, 1992). Thus, even though the binding affinity is reduced 10- to 17-fold, there can be a 200- to 10,000-fold decrease in signaling through membrane receptors, for example, decreased cytotoxic activity or in vivo toxicity. As discussed above, such a reduced signaling through membrane receptors, for example, reduced cytotoxicity or in vivo toxicity, is advantageous in view of the potential leaching of the ligand from a matrix and introduction of low levels into a patient when an altered biological fluid is returned to the patient.

An additional advantage of using TNFα muteins is that they have a native structure. Because the muteins are highly homologous to the native TNFα sequence, these muteins can fold into a native structure that retains TNF receptor binding activity. Such a native structure means that the same amino acid residues are exposed on the surface of the molecule as in the native TNFα, except for possibly the mutant amino acid residue. Such a native folding means that the TNFα muteins should have little or no immunogenicity in the respective mammalian species.

As disclosed herein, particularly useful muteins are human muteins 1, 2, 3, 4, 5 and 6 (FIG. 3B) and the analogous muteins in other mammalian species. Mutein I is a single amino acid substitution relative to wild type human TNFα of Arg³¹with Pro (Zhang et al., supra, 1992). This mutein exhibits approximately 10-fold lower binding activity and approximately 10,000-fold lower cytotoxicity relative to native TNFα. Mutein 2 is a single amino acid substitution relative to wild type human TNFα of Asn³⁴with Tyr (Yamagishi et al., supra, 1990; Asn³²in the numbering system of Yamagishi et al.). This mutein exhibits approximately 5-fold lower binding activity and approximately 12,500-fold lower cytotoxicity relative to native TNFα. Mutein 3 is a single amino acid substitution relative to wild type human TNFα of Pro¹¹⁷with Leu (Yamagishi et al., supra, 1990; Pro 115 in the numbering system of Yamagishi et al.). This mutein exhibits approximately 12-fold lower binding activity and approximately 1400-fold lower cytotoxicity. Mutein 4 is a single amino acid substitution relative to wild type human TNFα of Ser¹⁴⁷with Tyr (Zhang et al., supra, 1992). This mutein exhibits approximately 14-fold lower binding activity and approximately 10,000-fold lower cytotoxicity relative to native TNFα. Mutein 5 is a single amino acid substitution relative to wild type human TNFα of Ser⁹⁵with Tyr (Zhang et al., supra, 1992). This mutein exhibits approximately 17-fold lower binding activity and approximately 200-fold lower cytotoxicity relative to native TNFα. Mutein 6 is a single amino acid substitution relative to wild type human TNFα of Tyr¹¹⁵with Phe (Zhang et al., supra, 1992). This mutein exhibits approximately 17-fold lower binding activity and approximately 3,300-fold lower cytotoxicity relative to native TNFα. As disclosed herein, it is understood that analogous muteins can be generated in other mammalian species by making the same amino acid substitutions in the analogous position of the respective species.

Although muteins 1, 2 and 4, as well as other TNFα muteins, were previously known and characterized with respect to binding the multivalent membrane receptor, it was previously unknown whether these TNFα muteins would bind to the monovalent soluble TNF receptors. As disclosed herein, the TNFα muteins bind with an affinity sufficient to deplete soluble TNF receptor from plasma (see Example 3). These results indicate that TNFα muteins can be an effective binding partner for depleting soluble TNF receptor from a biological fluid.

It is understood that TNFα muteins additional to the specific muteins exemplified herein can be used in methods of the invention. TNFα from various mammalian species show a high degree of amino acid identity (see FIGS. 3A and 3B, conserved sequence SEQ ID NO:1; Van Ostade et al., supra, 1994). As described by Van Ostade et al. (supra, 1994), a conserved TNFα amino acid sequence was identified across 11 mammalian species. The conserved amino acid residues are conserved across all 11 shown species or have only a single species showing variation at that position (see FIG. 3A and Van Ostade et al., supra, 1994). Thus, in one embodiment, the invention provides a TNFα mutein comprising the conserved sequence referenced as SEQ ID NO:1.

One skilled in the art can readily determine additional muteins suitable for use in methods of the invention. As discussed above, TNFα muteins having relatively high affinity for TNF receptors and decreased signaling through membrane receptors, for example, decreased cytotoxicity or in vivo toxicity, relative to native TNFα are particularly useful in methods of the invention. One skilled in the art can readily determine additional suitable TNFα muteins based on methods well known to those skilled in the art. Methods for introducing amino acid substitutions into a sequence are well known to those skilled in the art (Ausubel et al., Current Protocols in Molecular Biology (Supplement 56), John Wiley & Sons, New York (2001); Sambrook and Russel, Molecular Cloning: A Laboratory Manual, 3rd ed., Cold Spring Harbor Press, Cold Spring Harbor (2001); U.S. Pat. Nos. 5,264,563 and 5,523,388). Generation of TNFα muteins has been previously described (Van Ostade et al., supra, 1994; Van Ostade et al., supra, 1991; Zhang et al., supra, 1992; Yamagishi et al., supra, 1990). Furthermore, one skilled in the art can readily determine the binding and cytotoxicity and/or in vivo toxicity of candidate muteins to ascertain the suitability for use in a method of the invention (Van Ostade et al., supra, 1994; Van Ostade et al., supra, 1991; Zhang et al., supra, 1992; Yamagishi et al., supra, 1990).

Muteins of particular interest for use in methods of the present invention, in addition to having relatively high affinity for TNF receptors and reduced signaling through membrane receptors, for example, reduced cytotoxicity or in vivo toxicity, are those having amino acid substitutions in three regions of TNFα, region 1, amino acids 29-36, region 2, amino acids 84-91, and region 3, amino acids 143-149 (numbering as shown in FIG. 3A). Muteins 1, 2 and 4 are exemplary of muteins having single amino acid substitutions in these regions. Region 1 corresponds to amino acids 29-36, residues LNRRANAL (SEQ ID NO:18) of human TNFα. Region 2 corresponds to amino acids 84-91, residues AVSYQTKV (SEQ ID NO:19) of human TNFα. Region 3 corresponds to amino acids 143-149, residues DFAESG (SEQ ID NO:20) of human TNFα. In addition to the TNFα muteins disclosed herein, other TNFα muteins can be generated, for example, by introducing single amino acid substitutions in regions 1, 2 or 3 and screening for binding activity and cytotoxic activity and/or in vivo toxicity as disclosed herein (see also Van Ostade et al., supra, 1991; Zhang et al, supra, 1992; Yamagishi et al., supra, 1990). Methods for introducing amino acid substitutions at a particular amino acid residue or region are well known to those skilled in the art (see, for example, Van Ostade et al., supra, 1991; Zhang et al, supra, 1992; Yamagishi et al., supra, 1990; U.S. Pat. Nos. 5,264,563 and 5,523,388). For example, each of the other 19 amino acids relative to a native sequence can be introduced at each of the positions in regions 1, 2 and 3 and screened for binding activity and/or signaling activity, for example, cytotoxic activity or in vivo toxicity, to soluble and/or membrane bound TNF receptor. This would only require the generation of approximately 420 mutants (19 single amino acid substitutions at each of 22 positions in regions 1, 2 and 3), a number which can be readily generated and screened by well known methods. Those having desired characteristics as disclosed herein, for example, specific binding activity for soluble TNF receptor and reduced signaling through the membrane TNF receptor, can be selected as a TNFα mutein useful in methods of the invention.

The invention additionally provides a TNFα mutein having the consensus sequence of SEQ ID NO:9 (FIG. 3C). In one embodiment, a TNFα mutein comprises the consensus sequence SEQ ID NO:9, wherein X.sub.1 is an amino acid selected from Leu and Val; wherein X₂is a 2 or 3 amino acid peptide selected from GlnAsnSer, ArgAlaLeu, ArgThrPro, GlnAlaSer, and GlnThr; wherein X₃is an amino acid selected from Asp and Asn; wherein X₄is a 5 amino acid peptide selected from HisGlnValGluGlu, HisGlnAlaGluGlu, ProGlnValGluGly, ProGluAlaGluGly, LeuSerAlaProGly, IleSerAlaProGly, ProGlnAlaGluGly, IleAsnSerProGly, and ValLysAlaGluGly; wherein X₅is an amino acid selected from Glu, Gln and Arg; wherein X₆is a 4 amino acid peptide selected from LeuSerGlnArg, LeuSerArgArg, GlyAspSerTyr, LeuSerGlyArg, TrpAspSerTyr, GinSerGlyTyr, and LeuAsnArgArg; wherein X₇is an amino acid selected from Leu, Met, and Lys; wherein X₈is a two amino acid peptide selected from MetAsp, MetLys, ValGlu, ValLys, and ValGln; wherein X₉is an amino acid selected from Lys, Thr, Glu, and Arg; wherein X₁₀is an amino acid selected from Val, Lys, and Ile; wherein X₁₁is a 2 amino acid peptide selected from AlaAsp, SerAsp, ThrAsp, LeuAsp, AlaGlu, and SerGlu; wherein X₁₂is an amino acid selected from Lys, Ser, Thr, and Arg; wherein X₁₃is an amino acid selected from Gln and His; wherein X₁₄is a 4 or 5 amino acid peptide selected from AspValValLeu, AspTyrValLeu, SerTyrValLeu, ProProProVal, SerThrHisValLeu, SerThrProLeuPhe, SerThrHisValLeu, and SerThrAsnValPhe; wherein X₁₅is an amino acid selected from Val and Ile; wherein X₁₆is an amino acid selected from Phe, Ile, and Leu; wherein X₁₇is an amino acid selected from Ile and Val; wherein X₁₈is a 2 amino acid peptide selected from GlnGlu, ProAsn, GlnThr, and ProSer; wherein X₁₉is an amino acid selected from Leu and Ile; wherein X₂₀is a 3 amino acid peptide selected from ProLysAsp, HisArgGlu, GlnArgGlu, and HisThrGlu; wherein X₂₁is an amino acid selected from Gly, Glu, Gln, and Trp or is absent; wherein X₂₂is an amino acid selected from Leu, Pro, and Ala; wherein X₂₃is an amino acid selected from Leu and Gln; wherein X₂₄is an amino acid selected from Gly and Asp; wherein X₂₅is an amino acid selected from Gln, Leu, and Arg; wherein X₂₆is an amino acid selected from Ala and Thr; wherein X₂₇is an amino acid selected from Val and Ile; wherein X₂₈is an amino acid selected from Leu, Gin, and Arg; wherein X₂₉is an amino acid selected from Lys, Glu, Ala, Asn, and Asp; wherein X₃₀is an amino acid selected from Phe, Ile, Leu and Tyr; and wherein X₃₁is an amino acid selected from Val and Ile (see FIG. 3A; Van Ostade et al., supra, 1994). Such a consensus TNFα mutein is expected to exhibit binding activity for TNF receptor, and such activity can be readily determined by those skilled in the art using well known methods, as disclosed herein.

In addition to the variant positions described above, it is understood that a TNFα mutein can additionally include variant amino acids in the conserved sequence referenced as SEQ ID NO:1. As shown in FIG. 3A and as discussed above, the conserved TNFα sequence includes certain positions where one of the shown mammalian species differs from the other ten. For example, the conserved amino acid at position 2, Arg, is Leu in dog (FIG. 3A). Thus, a TNFα mutein can include a substitution of Leu at position 2 with the remainder of the conserved sequence referenced as SEQ ID NO:1. Similarly, substitutions of other “conserved” positions, where at least one of the species has an amino acid substitution relative to the conserved sequence, are included as TNFα muteins. For example, a TNFα mutein can have the corresponding substitution of mutein 1, that is, Arg³¹Pro and substitution in the conserved sequence in the variable positions, as described above represented by X, and/or substitution in a conserved position that varies in a single species. Furthermore, a TNFα mutein can include conservative amino acid substitutions relative to the conserved sequence or the sequence of a particular species of TNFα. Such TNFα muteins can be readily recognized by one skilled in the art based on the desired characteristics of a TNFα mutein, as disclosed herein.

Additionally, any of the TNFα muteins disclosed herein can be modified to include an N-terminal deletion. As discussed in Van Ostade (supra, 1994), short deletions at the N-terminus of TNFα retained activity, whereas deletion of the N-terminal 17 amino acids resulted in a loss of activity. Therefore, it is understood that a TNFα mutein of the invention also includes TNFα muteins having N-terminal deletions that retain activity. Such TNFα muteins can include, for example, an N-terminal deletion of 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acids. Furthermore, one skilled in the art can readily determine whether further N-terminal deletions can be incorporated into a TNFα mutein by making the deletion mutations and screening for desired characteristics, as disclosed herein.

The invention provides a variety of TNFα muteins, as disclosed herein. Generally, a particularly useful TNFα mutein of the invention has about 2-fold, about 3-fold, about 4-fold, about 5-fold, about 6-fold, about 7-fold, about 8-fold, about 9-fold, about 10-fold, about 11-fold, about 12-fold, about 13-fold, about 14-fold, about 15-fold, about 16-fold, about 17-fold, about 18-fold, about 19-fold, about 20-fold, about 25-fold, about 30-fold, or even higher fold reduced binding affinity for TNF receptors, particularly membrane bound TNF receptors, relative to native/wild type TNFα. Such reduced binding affinity can be, but is not necessarily, exhibited toward sTNFR. Also, a particularly useful TNFα mutein of the invention has about 10-fold, about 50-fold, about 100-fold, about 150-fold, about 200-fold, about 300-fold, about 500-fold, about 1000-fold, about 2000-fold, about 3000-fold, about 4000-fold, about 5000-fold, about 6000-fold, about 7000-fold, about 8000-fold, about 9000-fold, about 10,000-fold, about 20,000-fold, about 30,000-fold, about 50,000-fold, or even higher fold reduced signaling through the membrane receptors, for example, reduced cytoxicity or in vivo toxicity, relative to native/wild type TNFα. It is understood that a TNFα mutein can have reduced binding affinity and/or reduced cytoxicity, as discussed above and disclosed herein.

The invention provides a conjugate comprising a tumor necrosis factor α (TNFα) mutein attached to a substrate. In one embodiment, the TNFα mutein of the conjugate comprises the conserved sequence referenced as SEQ ID NO:1.

In another embodiment, the invention provides a conjugate where the TNFα mutein has the consensus sequence SEQ ID NO:9, wherein X₁is an amino acid selected from Leu and Val; wherein X₂is a 2 or 3 amino acid peptide selected from GlnAsnSer, ArgAlaLeu, ArgThrPro, GlnAlaSer, and GlnThr; wherein X₃is an amino acid selected from Asp and Asn; wherein X₄is a 5 amino acid peptide selected from HisGlnValGluGlu, HisGlnAlaGluGlu, ProGInValGluGly, ProGluAlaGluGly, LeuSerAlaProGly, IleSerAlaProGly, ProGlnAlaGluGly, IleAsnSerProGly, and ValLysAlaGluGly; wherein X₅is an amino acid selected from Glu, Gln and Arg; wherein X₆is a 4 amino acid peptide selected from LeuSerGlnArg, LeuSerArgArg, GlyAspSerTyr, LeuSerGlyArg, TrpAspSerTyr, GlnSerGlyTyr, and LeuAsnArgArg; wherein X₇is an amino acid selected from Leu, Met, and Lys; wherein X₈is a two amino acid peptide selected from MetAsp, MetLys, ValGlu, ValLys, and ValGln; wherein X₉is an amino acid selected from Lys, Thr, Glu, and Arg; wherein X₁₀is an amino acid selected from Val, Lys, and Ile; wherein X₁₁is a 2 amino acid peptide selected from AlaAsp, SerAsp, ThrAsp, LeuAsp, AlaGlu, and SerGlu; wherein X₁₂is an amino acid selected from Lys, Ser, Thr, and Arg; wherein X₁₃is an amino acid selected from Gln and His; wherein X₁₄is a 4 or 5 amino acid peptide selected from AspValValLeu, AspTyrValLeu, SerTyrValLeu, ProProProVal, SerThrHisValLeu, SerThrProLeuPhe, SerThrHisValLeu, and SerThrAsnValPhe; wherein X₁₅is an amino acid selected from Val and Ile; wherein X₁₆is an amino acid selected from Phe, Ile, and Leu; wherein X₁₇is an amino acid selected from Ile and Val; wherein X₁₈is a 2 amino acid peptide selected from GlnGlu, ProAsn, GlnThr, and ProSer; wherein X₁₉is an amino acid selected from Leu and Ile; wherein X₂₀is a 3 amino acid peptide selected from ProLysAsp, HisArgGlu, GlnArgGlu, and HisThrGlu; wherein X₂₁is an amino acid selected from Gly, Glu, Gln, and Trp or is absent; wherein X₂₂is an amino acid selected from Leu, Pro, and Ala; wherein X₂₃is an amino acid selected from Leu and Gln; wherein X₂₄is an amino acid selected from Gly and Asp; wherein X₂₅is an amino acid selected from Gln, Leu, and Arg; wherein X₂₆is an amino acid selected from Ala and Thr; wherein X₂₇is an amino acid selected from Val and Ile; wherein X₂₈is an amino acid selected from Leu, Gln, and Arg; wherein X₂₉is an amino acid selected from Lys, Glu, Ala, Asn, and Asp; wherein X₃₀is an amino acid selected from Phe, Ile, Leu and Tyr; and wherein X₃₁is an amino acid selected from Val and Ile.

In yet another embodiment, the invention provides a conjugate where the TNFα mutein has an amino acid substitution in a region of TNFα selected from region 1 amino acids 29-36, region 2 amino acids 84-91 and region 3 amino acids 143-149 of human TNFα (SEQ ID NO:2) or the analogous position of TNFα from another species.

In still another embodiment, the invention provides a conjugate where the TNFα mutein is selected from mutein 1 (SEQ ID NO:3), mutein 2 (SEQ ID NO:4), mutein 3 (SEQ ID NO:5), mutein 4 (SEQ ID NO:6), mutein 5 (SEQ ID NO:7) and mutein 6 (SEQ ID NO:8). In a particular embodiment, the invention provides a conjugate where the TNFα mutein is selected from mutein 1 (SEQ ID NO:3), mutein 2 (SEQ ID NO:4), and mutein 4 (SEQ ID NO:6). The TNFα mutein of the conjugate can be derived from a species selected, for example, from human, dog, cat, horse, sheep, goat, pig, cow, rabbit and rat.

The invention additionally provides a method of stimulating an immune response in a mammal having a pathological condition. The method can include the steps of obtaining a biological fluid from the mammal; contacting the biological fluid with a tumor necrosis factor α (TNFα) mutein having specific binding activity for a soluble tumor necrosis factor receptor (TNFR); removing the TNFα mutein bound to the soluble TNFR from the biological fluid to produce an altered biological fluid having a reduced amount of soluble TNFR; and administering the altered biological fluid to the mammal. The biological fluid can be, for example, blood, plasma, serum or lymphatic fluid, including whole blood. In one embodiment, a method using whole blood as the biological fluid can further include the step of separating the whole blood into a cellular component and an acellular component or a fraction of the acellular component, wherein the acellular or the fraction of the acellular component contains a soluble TNFR. The method can additionally include the step of combining the cellular component with the altered acellular component or altered fraction of the acellular component to produce altered whole blood, which is administered to the mammal as the altered biological fluid.

In a particular embodiment of a method of the invention, the TNFα mutein can have specific binding activity for a single type of soluble TNFR, for example sTNFRI or sTNFRII. Alternatively, the TNFα mutein can have specific binding activity for more than one type of soluble TNFR, for example, both sTNFRI and sTNFRII.

The present invention further relates to the use of various mixtures of binding partners. One mixture can be composed of multiple binding partners that selectively bind to a single targeted immune system inhibitor. Another mixture can be composed of multiple binding partners, each of which selectively binds to different targeted immune system inhibitors. Alternatively, the mixture can be composed of multiple binding partners that selectively bind to different targeted immune system inhibitors. For example, the mixture can contain more than one TNFα mutein. Furthermore, the multiple TNFα muteins can specifically bind to a single type of soluble TNF receptor or can bind to more than one type of TNF receptor, for example, sTNFRI and sTNFRII.

In another embodiment of a method of the invention, the biological fluid can be contacted with a plurality of TNFα muteins. In a particular embodiment, the plurality of TNFα muteins can have specific binding activity for a single type of soluble TNFR, for example, sTNFRI or sTNFRII. Alternatively, the plurality of TNFα muteins can have specific binding activity for more than one type of soluble TNFR, that is, sTNFRI and sTNFRII.

For certain embodiments in which it is desirable to increase the molecular weight of the binding partner/immune system inhibitor complex, the binding partner can be conjugated to a carrier. Examples of such carriers include, but are not limited to, proteins, complex carbohydrates, and synthetic polymers such as polyethylene glycol.

As used herein, “functionally active binding sites” of a binding partner refer to sites that are capable of binding to one or more targeted immune system inhibitors.

Methods for producing the various binding partners useful in the present invention are well known to those skilled in the art. Such methods include, for example, recombinant DNA and synthetic techniques, or a combination thereof. Binding partners such as TNFα muteins can be expressed in prokaryotic or eukaroytic cells, for example, mammalian, insect, yeast, and the like. If desired, codons can be changed to reflect any codon bias in a host species used for expression.

In one embodiment of the present methods, the binding partner such as a TNFα mutein is attached to an inert medium to form an absorbent matrix (FIG. 1). The TNFα mutein can be, for example, covalently attached to a substrate such as an inert medium. As used herein, the term “inert medium” is intended to include solid supports to which the binding partner(s) can be attached. Particularly useful supports are materials that are used for such purposes including, for example, cellulose-based hollow fibers, synthetic hollow fibers, silica-based particles, flat or pleated membranes, macroporous beads, agarose-based particles, and the like. The inert medium can be in the form of a bead, for example, a macroporous bead or a non-porous bead. Exemplary macroporous beads include, but are not limited to, naturally occurring materials such as agarose, cellulose, controlled pore glass, or synthetic materials such as polyacrylamide, cross-linked agarose (such as Trisacryl™, Sephacryl, and Ultrogel™), azlactone, polymethacrylate, polystyrene/divinylbenzene, and the like. Exemplary non-porous beads include, but are not limited to, silica, polystyrene, latex, and the like. Hollow fibers and membranes can also be composed of natural or synthetic materials. Exemplary natural materials include, but are not limited to, cellulose and modified cellulose, for example, cellulose diacetate or triacetate. Exemplary synthetic materials include, but are not limited to, polysulfone, polyvinyl, polyacetate, and the like. Such inert media can be obtained commercially or can be readily made by those skilled in the art. The binding partner can be attached to the inert medium by any means known to those skilled in the art including, for example, covalent conjugation. Alternatively, the binding partner can be associated with the inert matrix through high-affinity, non-covalent interaction with an additional molecule which has been covalently attached to the inert medium. For example, a biotinylated binding partner can interact with avidin or streptavidin previously conjugated to the inert medium.

The absorbent matrix thus produced can be contacted with a biological fluid, or a fraction thereof, through the use of an extracorporeal circuit. The development and use of extracorporal, absorbent matrices has been extensively reviewed (see Kessler, Blood Purification 11:150-157 (1993)).

In another embodiment, herein referred to as the “stirred reactor” (FIG. 2), the biological fluid is exposed to the binding partner such as a TNFα mutein in a mixing chamber and, thereafter, the binding partner/immune system inhibitor complex is removed by means known to those skilled in the art, including, for example, by mechanical or by chemical or biological separation methods. For example, a mechanical separation method can be used in cases where the binding partner, and therefore the binding partner/immune system inhibitor complex, represent the largest components of the treated biological fluid. In those cases, filtration can be used to retain the binding partner and immune system inhibitors associated therewith, while allowing all other components of the biological fluid to permeate through the fiber and, thus, to be returned to the patient. In an example of a chemical or biological separation method, the binding partner and immune system inhibitors associated therewith can be removed from the treated biological fluid through exposure to an absorbent matrix capable of specifically attaching to the binding partner. For example, a matrix constructed with antibodies reactive with a TNFα mutein can serve this purpose. Similarly, were biotin conjugated to the binding partner such as a TNFα mutein prior to its addition to the biological fluid, a matrix constructed with avidin or streptavidin could be used to deplete the binding partner and immune system inhibitors associated therewith from the treated fluid.

In a final step of the present methods, the treated or altered biological fluid, having a reduced amount of targeted immune system inhibitor such as soluble TNF receptor, is returned to the patient receiving treatment along with untreated fractions of the biological fluid, if any such fractions were produced during the treatment. The altered biological fluid can be administered to the mammal by any means known to those skilled in the art, including, for example, by infusion directly into the circulatory system. The altered biological fluid can be administered immediately after contact with the binding partner in a contemporaneous, extracorporeal circuit. In this circuit, the biological fluid is (a) collected, (b) separated into cellular and acellular components, if desired, (c) exposed to the binding partner, and if needed, separated from the binding partner bound to the targeted immune system inhibitor, (d) combined with the cellular component, if needed, and (e) readministered to the patient as altered biological fluid. Alternatively, the administration of the altered biological fluid can be delayed under appropriate storage conditions readily determined by those skilled in the art.

If desirable, the entire process can be repeated. Those skilled in the art can readily determine the benefits of repeated treatment by monitoring the clinical status of the patient, and correlating that status with the concentration(s) of the targeted immune system inhibitor(s) such as soluble TNFα receptor in circulation prior to, during, and after treatment.

The present invention further provides an apparatus for reducing the amount of a targeted immune system inhibitor such as soluble TNF receptor in a biological fluid. The apparatus is composed of: (a) a means for separating the biological fluid into a cellular component and an acellular component or fraction thereof; (b) an absorbent matrix having attached thereto a TNFα mutein or a stirred reactor as described above to produce an altered acellular component or fraction thereof; and (c) a means for combining the cellular fraction with the altered acellular component or fraction thereof. The apparatus is particularly useful for whole blood as the biological fluid in which the cellular component is separated either from whole plasma or a fraction thereof.

The means for initially fractionating the biological fluid into the cellular component and the acellular component, or a fraction thereof, and for recombining the cellular component with the acellular component, or fraction thereof, after treatment are known to those skilled in the art (see Apheresis: Principles and Practice, supra).

In a specific embodiment, the immune system inhibitor to be targeted is sTNFRI (Seckinger et al., J. Biol. Chem. 264:11966-11973 (1989); Gatanaga et al., Proc. Natl. Acad. Sci. USA 87:8781-8784 (1990)), a naturally occurring inhibitor of the pluripotent immune system stimulator, TNF. sTNFRI is produced by proteolytic cleavage, which liberates the extra-cellular domain of the membrane tumor necrosis factor receptor type I from its transmembrane and intracellular domains (Schall et al., Cell 61:361-370 (1990); Himmler et al., DNA and Cell Biol. 9:705-715 (1990)). sTNFRI retains the ability to bind to TNF with high affinity and, thus, to inhibit the binding of TNF to the membrane receptor on cell surfaces.

The levels of sTNFRI in biological fluids are increased in a variety of conditions which are characterized by an antecedent increase in TNF. These include bacterial, viral, and parasitic infections, and cancer as described above. In each of these disease states, the presence of the offending agent stimulates TNF production which stimulates a corresponding increase in sTNFRI production. sTNFRI production is intended to reduce localized, as well as systemic, toxicity associated with elevated TNF levels and to restore immunologic homeostasis.

In tumor bearing hosts, over-production of sTNFRI may profoundly affect the course of disease, considering the critical role of TNF in a variety of anti-tumor immune responses (reviewed in, Beutler and Cerami, Ann. Rev. Immunol. 7:625-655 (1989)). TNF directly induces tumor cell death by binding to the type I membrane-associated TNF receptor. Moreover, the death of vascular endothelial cells is induced by TNF binding. destroying the circulatory network serving the tumor and further contributing to tumor cell death. Critical roles for TNF in natural killer cell- and cytotoxic T lymphocyte-mediated cytolysis also have been documented. Inhibition of any or all of these effector mechanisms by sTNFRI has the potential to dramatically enhance tumor survival.

That sTNFRI promotes tumor survival, and that its removal enhances anti-tumor immunity, has been demonstrated. In an experimental mouse tumor model, sTNFRI production was found to protect transformed cells in vitro from the cytotoxic effects of TNF, and from cytolysis mediated by natural killer cells and cytotoxic T lymphocytes (Selinsky et al., Immunol. 94:88-93 (1998)). In addition, the secretion of sTNFRI by transformed cells has been shown to markedly enhance their tumorigenicity and persistence in vivo (Selinsky and Howell, Cell. Immunol. 200:81-87 (2000)). Moreover, removal of circulating sTNFRI has been found to provide clinical benefit to cancer patients, as demonstrated by human trials of Ultrapheresis as discussed above (Lentz, supra). These observations affirm the importance of this molecule in tumor survival and suggest the development of methods for more specific removal of sTNFRI as promising new avenues for cancer immunotherapy.

The following examples are intended to illustrate but not limit the invention.

EXAMPLE 1

Production, Purification, and Characterization of the Immune System Inhibitor, Human sTNFRI

The sTNFRI used in the present studies was produced recombinantly either in E. coli (R&D Systems; Minneapolis, Minn.) or in eukaryotic cell culture essentially as described (see U.S. Pat. No. 6,379,708, which is incorporated herein by reference). The construction of the eukaryotic expression plasmid, the methods for transforming and selecting cultured cells, and for assaying the production of sTNFRI by the transformed cells have been described (Selinsky et al., supra, 1998).

sTNFRI was detected and quantified in the present studies by capture ELISA (Selinsky et al., supra). In addition, the biological activity of recombinant sTNFRI, that is, its ability to bind TNF, was confirmed by ELISA. Assay plates were coated with human TNFα (Chemicon; Temecula, Calif.), blocked with bovine serum albumin, and sTNFRI, contained in culture supernatants as described above, was added. Bound sTNFRI was detected through the sequential addition of biotinylated-goat anti-human sTNFRI, alkaline phosphatase-conjugated streptavidin, and p-nitrophenylphosphate.

EXAMPLE 2

Production, Purification, and Characterization of TNFα Muteins

Briefly, TNFα muteins 1, 2, 3 and 4 were produced by expression of the respective cDNAs in E. coli. Genes encoding TNFα and TNFα muteins 1, 2, 3 and 4 were prepared using overlapping oligonucleotides having codons optimized for bacterial expression. Each of the coding sequences was fused in frame to that encoding the ompA leader to permit export of the recombinant polypeptides to the periplasm. Synthetic fragments were cloned into a pUC19 derivative immediately downstream of the lac Z promoter, and the resulting recombinant plasmids were introduced into E. coli. Recombinant bacteria were cultured to late-log, induced with isopropyl-β-D-thiogalactopyranoside (IPTG) for three hours, and harvested by centrifugation. Periplasmic fractions were prepared and tested by ELISA using polyclonal goat anti-human TNFα capture antibodies. After the addition of the diluted periplasms, bound TNFα and TNFα muteins 1, 2, 3 and 4 were detected by sequential addition of biotinylated polyclonal goat anti-human TNFα, streptavidin-alkaline phosphatase, and para-nitrophenyl phosphate (pNPP). TNFα and each of the TNFα muteins were detectable in the respective periplasms, though the level of TNFα mutein 3 only slightly exceeded the detection limit of the assay (FIG. 4).

The TNFα and TNFα mutein polypeptides 1, 2 and 4 were purified from periplasmic fractions by sequential chromatography on Q and S anion and cation exchange columns, respectively, essentially as described (Tavernier et al., J. Mol. Biol. 211:493-501 (1990)). The TNFα and TNFα mutein polypeptides were purified to >95% homogeneity as analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The gels revealed a 17 kDa band corresponding to TNFα or the muteins and a 34 kDa band, which was confirmed by Western blotting to be dimerized TNFα mutein.

The TNFα muteins were tested for their ability to bind to sTNFRI. Wells of a microtiter plate were coated with TNFα, blocked, and incubated with sTNFRI either in the presence or absence of the inhibitors, TNFα and TNFα muteins 1, 2 and 4. As shown in FIG. 5, TNFα muteins 1, 2 and 4 each bind to sTNFRI.

EXAMPLE 3

Depletion of the Immune System Inhibitor, sTNFRI, From Human Plasma Using TNFα Mutein Absorbent Matrices

The TNFα mutein absorbent matrices were produced and tested for their ability to deplete sTNFRI from human plasma. Briefly, purified TNFα muteins 1, 2 and 4 each was conjugated to cyanogen bromide (CNBr) Sepharose™ 4B at a density of 0.5 mg per ml of beads, and the remaining CNBr groups were quenched with ethanolamine. The resulting matrices were packed in individual column housings and washed extensively with phosphate buffered saline prior to use.

Normal human plasma was spiked (33% v/v) with culture supernatant containing recombinant human sTNFRI (see Example 1) to a final concentration of 8 nanograms per milliliter and passed through the respective columns at a flow rate of one milliliter of plasma per milliliter of resin per minute. An additional column, with no immobilized protein and quenched with ethanolamine, was included to control for non-specific depletion. One ml fractions were collected, and the relative levels of sTNFRI contained in the starting material and in the fractions were determined using a capture ELISA. To perform the capture ELISA, wells were coated with polyclonal goat anti-sTNFRI, and then were blocked with 2% BSA. Plasma samples were diluted 1:2. added to the wells, and sTNFRI therein was captured. Biotinylated polyclonal goat anti-sTNFRI was added, followed by streptavidin-alkaline phosphatase, and p-nitrophenylphosphate. Relative absorbance at 405 nm was used to estimate the depletion.

As shown in FIG. 6, all three of the immobilized TNFα muteins effectively depleted sTNFRI from human plasma, and the hierarchy observed in FIG. 5 again was manifested. The control matrix produced no reduction in sTNFRI levels, confirming the specificity of the depletion observed with the TNFα mutein matrices. Importantly, near quantitative depletion was achieved by TNFα muteins 1 and 4 at a flow rate that approximates that anticipated for use in a clinical setting.

Although the invention has been described with reference to the presently preferred embodiments, it should be understood that various modifications can be made without departing from the spirit of the invention. Accordingly, the invention is limited only by the following claims.

	Number	Date	Country
Parent	12378078	Feb 2009	US
Child	13957577		US
Parent	11234057	Sep 2005	US
Child	12378078		US

IMMOBILIZED TUMOR NECROSIS FACTOR-ALPHA MUTEINS FOR ENHANCING IMMUNE RESPONSE IN MAMMALS

Information

Publication Number

Date Filed

Date Published

Inventors

Original Assignees

CPC

International Classifications

Abstract

Description

Claims

RELATED APPLICATION

Continuations (2)