EP0889954A1

EP0889954A1 - Treatment of antigen presenting cells to modulate antigen presenting cell function

Info

Publication number: EP0889954A1
Application number: EP97939745A
Authority: EP
Inventors: Stephen P. Brooks; Thomas B. Tomasi; Zale P. Bernstein
Original assignee: Health Research Inc
Current assignee: Health Research Inc
Priority date: 1996-09-03
Filing date: 1997-09-02
Publication date: 1999-01-13
Also published as: AU4176497A; WO1998010056A1

Abstract

Provided herein is the discovery of a novel mechanism by which the ability of antigen presenting cells to stimulate T-cell function is inhibited by the formation of immunosuppresive complexes comprising the antigen presenting cell membrane-associated βglycan and cytokine TGF-β. Also provided are methods for restoring T-cell stimulatory function of antigen presenting cells of an individual, having such function suppressed by βglycan-TGF-β complex formation, by either removing TGF-β from the cell surface of the antigen presenting cells, removing β-glycan or βglycan complexed to TGF-β from the cell surface of the antigen presenting cells, or by contacting the antigen presenting cells with one or more antigen presenting cell activating factors which overcome the suppression of the T-cell stimulatory function of antigen presenting cells.

Description

TREATMENT OF ANTIGEN PRESENTING CELLS TO MODULATE ANTIGEN PRESENTING CELL FUNCTION

This invention was made in part with government support under grant HD- 17013 awarded by the National Institutes of Health. The government has certain rights in the invention. This application is a nonprovisional application of earlier co-pending provisional application United States Serial No. 60/025,332, filed September 3, 1996, which is incorporated herein by reference.

FIELD OF THE INVENTION

The present invention relates to a novel mechanism by which the ability of antigen presenting cells (APCs) to stimulate T-cell function is inhibited. More particularly, the present invention is directed to methods and compositions for modulating the immunostimulatory function of APCs by regulating, on the surface of APCs, the formation or presentation of immuno- suppressive complexes comprising the membrane-associated proteoglycan βglycan and cytokine TGF-3.

BACKGROUND OF THE INVENTION

The present invention relates to antigen presenting cells and their membrane-associated βglycan; to pathological conditions in which the expression of transforming growth factor β (TGF-β) appears to be increased; and to the discovery illustrated herein, that in certain pathological conditions, TGF-/? binds to Sglycan on APCs thereby forming immunosuppressive complexes which suppress the T-cell stimulatory function of the APCs.

1. Antigen Presenting Cells

Antigen presenting cells (APCs) represent three populations of myeloid cells that are generated from a common CD34+ bone marrow precursor. The three populations include: dendritic cells, found in tissues such as skin (Langerhans' cells) , spleen, lymph node, and Peyer's patch, and in peripheral blood; monocytes found primarily in peripheral blood; and macrophages, resident in both tissues and peripheral blood. APCs are specialized accessory cells having functions that include presenting antigen in T-dependent immune responses, sensitization, and tolerance. Dendritic cells have been shown to be the primary allo-stimulating and antigen presenting cell in peripheral blood, and accounts for more than 80% of the T-cell stimulatory function in the mixed lymphocyte reaction. Additionally, dendritic cells, like macrophages, are effective in stimulating memory responses. The mixed lymphocyte reaction is an in vi tro assay in which irradiated antigen presenting cells from an individual (stimulator) are incubated with T lymphocytes from a genetically disparate individual (allogenic responder) resulting in the proliferation of the responder lymphocytes. A reduced (less than normal) activity observed in the mixed lymphocyte reaction may represent a defect in the antigen presenting function of the stimulator APCs or a defect in the ability of the responder T lymphocytes to proliferate.

There are several pathological conditions in which T-cell proliferation/ stimulation is suppressed, and in which the allo- stimulatory accessory cell function may be inhibited. These conditions include, but are not limited to inflammation, inflammatory diseases (e.g., inflammatory joint disease, Crohn's disease), graft versus host disease, cancer (e.g., Hodgkin's disease, sarcomas), and microbial infections (e.g., HIV infection) . For example, the antigen-specific immune response mediated by T- cells in HIV seropositive individuals is depressed or absent. This loss of T-cell stimulatory function of APCs (accessory cells, dendritic cells, and macrophages) , has been reported by some groups to accompany HIV infection and has been hypothesized to be one of the primary mechanism by which the virus induces the suppression of systemic immunity which defines AIDS. The reduction in the number of CD4+ T- cells, loss of recall antigen response, and the failure to properly respond to infectious disease, have all been linked to virally compromised accessory cell function. The defect in APCS remains to be elucidated (Mactonia et al . , 1992, Jnununolog 75:576-81). Many mechanisms have been proposed including altered signal transduction, changes in major histocompatibility antigens, alterations in antigen processing, or other T- cell dysfunction (Mactonia et al . , 1992, supra; Blauvelt et al., 1995, J. Immunol . 154:3506-15); and HIV infection and HIV-mediated depletion of the APCs (Mactonia et al . , 1990, Immunology 71:38-45; Knight et al., 1988, Immunol . Lett . 19:177-81). Identification and circumvention of the defect in the APCs could offer new therapeutic approaches (Mactonia et al . , 1992, supra) .

In another example, individuals with multiple sclerosis have T cells that proliferate as well as T cells from healthy individuals. However, monocytes

(APCs) from multiple sclerosis patients during active disease exhibit almost a complete inability to stimulate responder cell proliferation in the allo- or auto-mixed lymphocyte reaction. In contrast, monocytes from multiple sclerosis patients during the inactive stage of the disease exhibited some, but less than the normal range of, ability to stimulate responder cell proliferation in vi tro . This APC stimulatory defect appears to fluctuate in close association with the disease (Baxevanis et al . , 1987, Clin . exp . Immunol .

67:362-371). Thus, identification and circumvention of the defect in the APCs could offer new therapeutic approaches for multiple sclerosis.

2. Betaglycan (βqlvcan) and TGF- β βglycan, formerly known as the TGF-0 type III receptor, is a membrane-anchored proteoglycan (Lopez - Casillas et al . , 1994, J. Cell Biol . 124:557-568). The mammalian family of transforming growth factor β (TGF-01, TGF-/32, TGF-03) has been associated with the suppression of a number of immune reactions and are known to play a role in the regulation of T-cell function. TGF-/3 has been shown to inhibit the proliferation response of T- cells to various in vi tro activation signals (see, e.g., Ellingsworth et al . , 1988, Cell Immunol . 114:41-54). TGF-01. and TGF-β receptors, are also synthesized by T- cells in response to T-cell receptor mediated activation signals as part of a negative feedback mechanism that regulates T-cell proliferation. TGF-3 has also been well characterized as cell cycle inhibitors, and it has been suggested that TGF-0 inhibition of in vi tro T-cell proliferation is the result of cell cycle arrest (Ellingsworth et al., 1988, supra) . However, the mechanisms by which TGF-/? acts as a potent inhibitor of the growth and functions of lymphoid cells are not known (Ruscetti et al . , 1993,

Ann . N. Y. Acad. Sci . 685:488-500). Current theories are that TGF-β inhibition of the growth and function of lymphoid cells is associated with the T-cell surface receptor expression of, or activation via, growth factor receptors such as interleukin- 1 (IL-l) or IL-2 (Ruscetti et al., 1993, supra) .

TGF-3 mediated immunosuppression is believed to play a role in several pathological conditions. Tumors that actively secrete TGF-β can inhibit or suppress CD4+ helper T cell activity, wherein such suppression can be overcome by the addition of neutralizing antibodies to TGF-0 (Ruscetti et al . , 1993, supra). A similar selective loss of T helper cell activity, as observed during tumorigenesis, has also been observed in symptomatic HIV- infected individuals. Candidates proposed to be somehow involved in the suppression observed in HIV infection include IL-4, IL-10 and TGF-/? (Ruscetti et al . , 1993, supra) . Additionally, TGF-/?1 expression is increased in the active stage of Crohn's disease (Babyatsky et al . , 1996, Gastroenterolog 110:975-84) .

As described previously, (Kaname and Ruoslahti, 1996, Biochem . J. 315:815-20; Lόpez-Casillas et al . , 1994, supra) , the TGF-/? receptor complex on cells is made up of three component: /?glycan, TGF-/? receptor I (T/?R I) , and TGF-/? receptor II (T/?R II) which combine to form the heterodimeric TGF-/? transmembrane signaling receptor. TGF-β has been shown to bind directly to the T/?R II receptor subunit allowing this receptor/ligand dimer to complex with and phosphorylate the T/?R I subunit. The formation of this trinary complex can then transduce a signal via the serine/threonine kinase function of the cytoplasmic tails of the two receptor subunits .

/?glycan has a high affinity for binding all isotypes of TGF-β to its protein core, and is the most numerous TGF-/? binding molecule found on the surface of most cells that bind to TGF-/?, with a density as high as 2 to 5 x 10^s molecules per cell. It is noted, however, that many cell types that respond to TGF-/? (e.g., hematopoietic cells) do not appear to have detectable amounts of /?glycan (Kaname and Ruoslahti, 1996, supra) . Because of its high affinity for binding TGF-/? and its high expression on the cell surface, βglycan functions as the primary capture molecule for TGF-/? and concentrates the cytokine at the cell surface. TGF-β bound to the βglycan can be presented to the T0RII subunit resulting in the formation of /?glycan/TGF-/?/T/?R II signalling complexes and transcription from TGF- β responsive genes. TGF-/? forms long-term stable associations with the glycan molecule and its binding to the proteoglycan does not result in a down regulation of /?glycan expression on the cell membrane (Massague, 1985, Cancer Cells, 3:73). This permits a cell to "express" /?glycan/TGF-0 on its surface for prolonged periods of time in a configuration that makes the TGF- β available to bind to and signal through the TGF-/? signaling receptor.

Thus, there is a need to identify and overcome defects in APCs, and/or TGF-? mediated suppression of CD4+ helper T cell activity, observed in pathological conditions. Methods for overcoming such defects and/or suppression offers new therapeutic approaches for these pathologic conditions.

SUMMARY OF THE INVENTION Accordingly, it is a primary object of the present invention to provide methods directed to overcoming defects in or the loss of T-cell stimulatory function of

APCs in certain pathological conditions.

It is another object of the present invention to provide methods directed to overcoming TGF-/? mediated suppression of CD4+ helper T cell activity observed in certain pathological conditions.

It is a further object of the present invention to provide in vitro methods for overcoming defects in or the loss of T-cell stimulatory function of APCs, and/or overcoming TGF- β mediated suppression of CD4+ helper T cell activity, observed in certain pathological conditions .

It is a further object of the present invention to provide in vivo method for overcoming defects in or the loss of T-cell stimulatory function of APCs, and/or overcoming TGF-/? mediated suppression of CD4+ helper T cell activity, observed in certain pathological conditions .

The foregoing objects are achieved by providing methods, involving the novel use of one or more compositions, to treat an individual having a pathological condition, in which during the active stage of the condition occurs TGF-? mediated suppression or inhibition of antigen presenting cell function. The methods comprise one or more of the following therapies for restoring antigen presenting cell function: a means to remove TGF-/?:/?glycan complexes from the antigen presenting cells' surfaces; a means to remove βglycan from the antigen presenting cells' surfaces; administration of anti -TGF-/? antibody; administration of anti-?glycan antibody; administration of a combination of anti -TGF-/? antibody and anti-/?glycan antibody; and administration of one or more cytokines which function to stimulate proliferation or function of antigen presenting cells, thereby overcoming TGF-/? suppression of APC function.

These and further features and advantages of the invention will be better understood from the description of the preferred embodiments when considered in relation to the figures in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a bar graph illustrating the level of allo- stimulatory function of antigen presenting cells at different time periods of incubation with TGF-/?.

FIG. 2 is a bar graph illustrating the level of allo- stimulatory function in AIDS patients at different time periods following infusion with cytokine granulocyte/monocyte colony stimulating factor (GM-CSF) . DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

Definitions

The term "pharmaceutically acceptable carrier" is used herein, for purposes of the specification and claims, to mean a medium for facilitating the in vivo administration of a therapeutic compound. Such pharmaceutically acceptable carrier media are known to those skilled in the art to include buffered saline solutions, buffered carbohydrate solutions, liposomes (Phillips et al . , 1994, J. Immunother . Emphasis Tumor Immunol . 15:185-93), and the like.

The term "antigen presenting cells" is used herein, for purposes of the specification and claims, to mean accessory cells which function to stimulate CD4+ helper T cell activity, which are generated from a common CD34+ bone marrow progenitor and which express differentiated cell surface markers as known in the art (Szablocs et al . , 1996, Blood 87:4520-4530), including: dendritic cells found in tissues such as skin (Langerhans' cells), spleen, lymph node, and Peyer's patch, and in peripheral blood; monocytes; and macrophages .

The term "TGF-/?" is used herein, for purposes of the specification and claims, to mean either or collectively all, of the three mammalian isotypes including TGF-01, TGF-/?2, and TGF-/?3, as all three isotypes have been shown to suppress the APC function of stimulating CD4+ helper T cell activity.

The term "PBL" is used herein, for purposes of the specification and claims, to mean peripheral blood leukocytes contain various leukocyte subpopulations including antigen presenting cells, but which typically in the experimental conditions, neutrophils have been depleted. The present invention relates to the discovery of a novel mechanism by which the ability of antigen presenting cells to stimulate T-cell function is inhibited. More particularly, it is disclosed that in vitro, and in vivo in diseases where T-cell proliferation/stimulation is suppressed, antigen presenting cells contain immunosuppressive complexes comprising the membrane-associated proteoglycan βglycan and cytokine TGF-β . It will be apparent from the present invention that a uni-que mechanism for the suppression of T-cell stimulation is TGF-/? suppression of APC function. The TGF-/? suppression is not dependent upon a trans -signalling event mediated through the TGF-β receptor of the APC, nor is it due to the production of a soluble suppressive factor. As illustrated herein, the TGF-/? suppression of APC function can be reversed by mild trypsin digestion of the APC. Thus, it is apparent that TGF-/? mediated suppression where the cytokine forms immunosuppressive complexes with the membrane associated proteoglycan, /?glycan. While not intending to be bound by theory, these immunosuppressive complexes appear to be capable of suppressing T-cell proliferation when the TGF-/? is presented to the TGF- β signaling receptors on the T-cell. Thus, the compositions of the present invention, and the methods for delivering such compositions, are targeted to either (i) blocking or inhibiting the binding of TGF-/? to /?glycan on the cell surface of APCs (e.g., administration of anti-TGF-/? antibody, or of anti-/?glycan antibody, or of a combination of anti-TGF-/? antibody and anti-/?glycan antibody) ; (ii) removing βglycan on the cell surface of APCs, thereby preventing the formation of immunosuppressive complexes between TGF-/? and jSglycan; removing the TGF-/?: βglycan immunosuppressive complexes on the cell surface of APCs; (iii) administration of one or more cytokines which stimulate proliferation or function of antigen presenting cells, thereby overcoming TGF-/? suppression of APC function; (iv) blocking the interaction between TGF-/? and βglycan to prevent complexes from forming (e.g., by anti-/?glycan antibody); and (v) blocking the presentation of the complexed TGF-/? on the APC surface to the TGF-? receptor on T- cells (e.g., by administration of anti-TGF-/? antibody, of anti-/?glycan antibody, or of a combination of anti-TGF-/? antibody and anti-/?glycan antibody)

EXAMPLE 1

This example illustrates that (a) the /?glycan on APCs binds to TGF-/?; and (b) the APC function of stimulating CD4+ helper T cell activity is suppressed or inhibited by the binding of TGF-β to βglycan. The mechanism of TGF-/? suppression of APC function was characterized by using two well defined in vi tro assays: (1) the allo- stimulatory function of the dendritic cell component of spleen cells was assayed in a one-way mixed lymphocyte response assay; and (2) analyzed was the presentation of soluble protein antigen by purified peritoneal macrophage to antigen sensitized lymph node T- cells.

Spleen cells from BALB/c mice were harvested and washed using methods known in the art. In the mixed lymphocyte response assay, the stimulator cells were spleen cells isolated from BALB/c (H-2^d) mice, washed with phosphate-buffered saline, and irradiated at 3500 rads before being resuspended in complete cell culture media (RPMI 1640 with 10% FBS , penicillin/ streptomycin and 5 x 10^"5 M /3-mercaptoethanol) . Responder cells were isolated lymph node cells from CBA (H-2^b) mice that were washed with phosphate-buffered saline and resuspended in cell culture medium. The proliferation assay was carried out in 96 -well plates with a final volume of 200μl/well, and 1 x 10⁵ responder cells per well. The assay was incubated for five days at 37°C and pulsed for the last 18 hours with [³H] thymidine. Cells were harvested onto glass wool filters and the filters are counted in 2 ml of scintillation fluid using a tritium window (expressed in counts per minute or "cpm") .

As a control for background, one group of wells contained no spleen cells, but contained the responder cells. As a positive control, 2 x 10⁵ irradiated cells were used as stimulator cells and mixed with the responder cells. To assay the binding of TGF-/? to /?glycan and subsequent inhibition of APC function, spleen cells were first pre- treated with TGF- β for various periods of time, ranging from 30 minutes to 48 hours. For incubations of 24 hours and 48 hrs, 2 x IO⁷ spleen cells in 10 ml of cell culture medium were pre- treated with 2 ng/ml of TGF-/? at 37°C. For the two hour incubation, 1 x IO⁷ spleen cells/ml of cell culture medium were pre- reated with 10 ng TGF-? at 37°C. For the 30 minute incubation, 1 x 10⁷ spleen cells in 0.5 ml of cell culture medium were pre- treated at 37°C with 10 ng of TGF-/?. After pre- treatment , the stimulator cells were then washed four times in cell culture medium, and then irradiated with 3500 rads , before addition (10⁵ spleen cells per well) to the responder cells in the mixed lymphocyte response assay.

It was found that the incubation of APC with TGF-/? for various intervals of time, ranging from 30 minutes to 48 hours, will significantly suppress the T-cell stimulatory function of the APCs. While some variation in the suppression of APC function was seen between experiments, in this assay even the 30 minute exposure to TGF-/? was as suppressive as a 48 hour incubation. Further, in various experiments compared was the suppressive function of each of the three mammalian isotypes of TGF- β {βl , β2 and /?3) without finding any significant difference between the isotypes as to their ability to suppress APC function.

As shown in Table 1, when the stimulator spleen cells were incubated with TGF-/? for as short a period of time as 30 minutes, observed is a statistically significant (p<0.003) suppression of the spleen's antigen presenting cell function of stimulating allotypic lymph node cells (T- cells) . Since it is known in the art that the majority of allo- stimulatory function in the mixed lymphocyte response assay is due to dendritic cells, it may be concluded from this data that exposure of dendritic cells to TGF- β can suppress their T-cell stimulatory function.

Unless otherwise stated, the data is presented as cpm, which represents the mean of six wells for each experimental condition, or as stimulation index (SI) . SI is the fold increase in proliferation of T-cells in the presence of allo- stimulator cells in the mixed lymphocyte response assay, and is calculated by dividing the ³H-thymidine incorporation (cpm) of unstimulated cells into the ³H- thymidine incorporation (cpm) from stimulated cells. Data is analyzed by applying the two tailed Student's T test.

TABLE 1

Stimulator Cells mean cpm (SI)

No Spleen Cells 966

Control Spleen Cells 21,654 (22.4)

Spleen Cells + TGF-/J 1758 (3.4)

In another illustration using the mixed lymphocyte response assay, the stimulator cells were peritoneal macrophages isolated by adherence from BALB/c (H-2^d) mice. The isolated peritoneal macrophages were then washed with phosphate-buffered saline being resuspended in complete cell culture medium. CBA (H-2^b) mice were sensitized by injection of ovalbumin (OVA) in adjuvant into the hind foot pads, using methods known in the art (Krinzman et al . , 1996, Am. J. Physiol . 271:1476-83; Hogaboam et al . , 1996, Gastroenterology 110:1456-66). Responder cells were lymph node cells isolated from the ovalbumin- sensitized mice, that were then washed with phosphate-buffered saline and resuspended in cell culture medium. The proliferation assay was carried out in 96 -well plates with a final volume of 200μl/well, and 5 x 10⁵ responder cells per well, with 25μg ovalbumin per well, and with 8x IO⁴ of peritoneal macrophages per well. It is noted that ovalbumin- sensitized T- cells require the addition of APCs to proliferate in response to OVA. The assay was incubated for five days at 37°C and pulsed for the last 18 hours with ³H thymidine. Cells were harvested onto glass wool filters and the filters are counted in 2 ml of scintillation fluid using a tritium window (expressed in counts per minute or "cpm"). As a control for background, one group of wells contained peritoneal macrophages and responder cells (Table 2, "Con pMφ") . Another control comprised TGF-β treated peritoneal macrophages and responder cells without OVA (Table 2, "TGF-/3 treated pMø" ). As positive control, peritoneal macrophages and OVA were mixed with the responder cells (Table 2, "Con pMφ + OVA"). To assay the binding of TGF-/? to β glycan and subsequent inhibition of APC function, another group of wells contained peritoneal macrophages that were pre-treated with TGF-/? for 24 hours; OVA; and responder cells ("TGF- β treated pMφ + OVA") .

The results, as shown in Table 2, confirm the results illustrated in Table 1 herein. That is, there is a statistically significant (p<0.003) suppression of the peritoneal macrophage antigen presenting cell function of stimulating syngeneic lymph node cells (T- cells) . The data are the mean of four representative experiments. In fact, the results in Table 2 show that exposure of the macrophage to TGF-/? will almost completely ablate the APC function of these cells.

TABLE

Stimulator Cells mean cpm (SI)

Con pMφ 607

Con pMφ + OVA 4,675 (7.3)

TGF-/S Treated pM 771

TGF-β Treated p.Mφ + OVA 2,479 (2.8)

EXAMPLE 2 This example illustrates that (a) antigen presenting cells express /?glycan on their cell surface; (b) that TGF-/? can be enzymatically removed from the glycan on the cell surface of APCs; and (c) that removal of TGF-/? bound to /?glycan on the cell surface of APCs resulted in restoration of the APC function of stimulating CD4+ helper T cell activity. Using the methods illustrated in Example 1 herein, mouse spleen cells were pre- treated with TGF-/?. Before adding the TGF-/? treated spleen cells to the mixed lymphocyte response assay, the spleen cells were contacted and incubated with trypsin to remove the /?glycan/TGF-/? complex from the antigen presenting cells' membranes using a technique known in the art for removing TGF-/? from thymic cortical epithelial cells. Briefly, TGF-/? treated, irradiated mouse spleen cells were washed once in serum free cell culture medium, and then suspended in a preheated trypsin/EDTA solution (proprietary, and commercially available from Grand Island Biologies Co. ; GIBCO) at a concentration of 2 x IO⁷ cells in 10 ml. The suspension was incubated at 37°C for 10 minutes. The cells were then washed twice in culture medium to remove the trypsin, and then assayed for their allo-stimulatory function in an mixed lymphocyte response assay. In the mixed lymphocyte response assay, to responder lymph node cells (1 x 10⁵ cells/well) were added either 10⁵ spleen cells (positive control) ; 10⁵ spleen cells which had been pre-treated with 5 ng/ml TGF-/? for 24 hours, or 10⁵ spleen cells which had been pre- treated with 5 ng/ml TGF-/? for 24 hours and then incubated with trypsin/EDTA for 10 minutes. The values shown in Table 3 represent the mean of six wells; and the data is representative of six different experiments. As shown in Table 3, the trypsin treatment of TGF- β preincubated spleen cells resulted in the complete recovery of their suppressed allo-stimulatory function. The spleen cells had an SI of 12.2 that was suppressed to an SI of 4.6 following TGF-β incubation. The allo-stimulatory function of TGF-/? suppressed spleen cells increased to an SI of 14.9 after trypsinization.

TABLE 3

Stimulator Cells cpm (SI)

Spleen Cells 9,965 (12.2)

Spleen Cells + TGF-J 3,807 (4.6)

Spleen Cells + TGF-p^* 12,167 (14.9) + trypsin

To determine if βglycan is expressed on the APC cell surface, and to determine if the incubation of spleen cells has led to the formation of suppressive

/?glycan/TGF-/? complexes, a trypsinization protocol was used to remove /?glycan from the cell membrane. The digestion of cells with low concentrations of trypsin has been reported to selectively remove the /?glycan molecule (Cheifetz et al . , 1984, J. Biol . Chem. 263:16) from cell surfaces. In this experimental protocol, the selective nature of the trypsin/EDTA digestion is demonstrated by the fact that trypsinization of TGF-/? pre- treated accessory cells restores a T-cell stimulatory function that is dependent on the expression of numerous cell surface glycoproteins (MHC class II, B7-2, ICAM-l, CD40 and others) . To determine the effect of a 10 minute trypsin/EDTA digestion on the expression of T-cell stimulatory molecules, flow cytometery was used to quantitate the levels of I-A^d and B7-2 on spleen cells before and after trypsinization. Trypsin/EDTA digestion had no significant effect on the expression of I-A^d or of B7-2 by these cells.

EXAMPLE 3

This example illustrates that (a) TGF- β mediated suppression of APC function does not involve a soluble suppressive factor; and (b) TGF- β mediated suppression of APC function does not require a trans-signalling event on the antigen presenting cell. Using the methods illustrated in Example 1 herein, mouse spleen cells were prepared as stimulator cells for the mixed lymphocyte response assay. In this experimental protocol, the spleen cells are fixed in paraformaldehyde . However, since fixed cells are less effective in stimulating T- cell helper activity, the BALB/c spleen cells were cultured overnight in cell culture medium with 50μg/ml of lipopoly-saccharide (LPS) . LPS is known by those skilled in the art to amplify the antigen presenting function of APCs. The LPS treatment was intended to improve, for purposes of assaying by the mixed lymphocyte response, the detection of antigen presenting cell function displayed by fixed cells. The washed LPS- stimulated spleen cells were fixed in a solution of 0.4% paraformaldehyde in phosphate buffered saline for 5 minutes at room temperature. The reaction was stopped by adding an equal volume of 0.2 M lysine. Fixed cells were assayed for viability by exclusion of trypan blue dye, and were determined to be effectively killed by this fixation technique. The fixed cells were then washed three times in cell culture medium. The control fixed spleen cells were not treated with TGF-/?. The test group comprised fixed spleen cells that were incubated for 30 minutes with 10 ng of TGF-β as described above.

Fixed spleen cells cannot respond to signals transduced through their T/?R I/II signaling receptor. Thus, the results shown in Table 4 confirm that TGF-β mediated suppression of APC function does not require a trans -signalling event on the antigen presenting cell, and that TGF-/? mediated suppression of APC function does not involve a soluble suppressive factor, since the fixed cells still stimulated a vigorous allogeneic proliferative response with an SI of 7.5. The incubation of fixed cells with TGF-/? completely ablated their allo-stimulatory function, giving an SI of <1.0. This data is representative of five repeated experiments. Therefore, it is concluded from this experiment that the generation of soluble suppressive factors by TGF-/? treated accessory cells is not the mechanism of TGF-/? mediated suppression of APC function, nor does the mechanism require a trans-signalling event on the antigen presenting cell.

TABLE 4

Stimulator Cells cpm (SI) no stimulators 849

Fixed Spleen Cells 6,359 (7.5)

Fixed Spleen Cells + TGF-p" 583 (<1.0)

Additional experiments confirmed that the production of a soluble suppressive factor was not the mechanism of TGF-/? mediated suppression of APC function. Using reverse transcriptase-polymerase chain reaction technology (RT/PCR) to monitor the presence of cytokine RNA in TGF-/? treated macro-phages , detected was a significant increase in the production of IL-10 transcripts and TNFα transcripts. However, when neutralizing antibodies against these two cytokines were introduced into the antigen presentation assay (mixed lymphocyte response assay) , they had no significant effect on the recovery of APC function.

Further experiments confirm that production of a soluble suppressive factor was not the mechanism of

TGF-/? mediated suppression of APC function. Using the methods disclosed in Example 1, in a mixed lymphocyte response assay TGF-/? pre- reated spleen cells (5 x 10⁵ cells incubated with 10 ng TGF-? for 30 minutes at 37°C) were mixed with untreated (not treated with TGF- β ) spleen cells as stimulators for responder lymph node cells (1 x 10⁵ cells per well) . The mixture (1 x 10* cells per well) of TGF-/? pre- treated spleen cells and untreated cells was compared to control spleen cells (1 x 10⁵ cells per well) and TGF-/? pre- treated spleen cells (1 x IO⁵ cells per well) in the mixed lymphocyte response assay. As shown in Table 5 (data is representative of four experiments) , the presence of the TGF-/? pre- reated spleen cells did not inhibit the allo-stimulatory function of the untreated spleen cells, as would be expected if no soluble suppressor factor is present.

TABLE 5

Stimulator Cells cpm (SI)

10^s Control Spleen Cells 9 , 334 ( 7 . 5 )

10^s Spleen Cells + TGF-01 2 , 961 (2 . 3 )

10^s Control : 10^s + TGF- /3 7 , 773 ( 6 . 6 )

Additionally, conditioned media from TGF-/? treated spleen cells was mixed with untreated (control) spleen cells were used in a mixed lymphocyte response assay. There was no suppression of APC function observed when the conditioned media from the TGF-/? treated spleen cells was used with the control spleen cells as stimulators in this assay.

In another set of experiments, using TGF-/? pre- treated adherent macrophages as antigen presenting cells and lymph node T- cells as responding cells in a mixed lymphocyte response assay, there was no indication that contact with the TGF-/? treated accessory cells induced cell cycle arrest in the T- cells. In summarizing the results of the mixed lymphocyte response assays described above, TGF-/? mediated suppression of APC function neither involves a soluble suppressive factor, nor requires a trans -signalling event on the antigen presenting cell; nor suggests that the presence of TGF-/? pre- treated accessory cells is inducing irreversible anergy (e.g., the presence of T-cells which, following contact with TGF-/? suppressed APC, are incapable of responding in a typical manner to proliferation signals from unsuppressed AOCs) or cell cycle arrest or apoptosis of the T-cells. While not intending to be bound by theory, the results support a mechanism that relates to intercellular presentation between the antigen presenting cell the T-cell which is responsible for the suppression of allo-stimulatory function mediated by TGF-/? (e.g., the presentation of complexed TGF-/? on the surface of the APC to the TGF-/? receptor of the T-cell) .

EXAMPLE 4 This ex.ample illustrates that TGF-/? mediated suppression of APC allo-stimulatory function can be overcome by treatment with one or more agents that induce homeotypic aggregation or conjugation of the antigen presenting cells with T-cells. It is noted that T-cell activation requires the cognate interaction between accessory cells and T-cells to be maintained for upwards of 20 minutes (Vallitutti et al . , 1995, J. Exp . Med. 181:577-584), and is dependent upon the activation of LFA-1 on the T-cell to maintain such contact (Dus in et al., 1989, Nature 341:619-630).

In a mixed lymphocyte response assay, irradiated PBL from AIDS patients were used as allo- stimulator cells, and purified T-cells from an HIV-seronegative, healthy individual were used as responder cells.

Phorbol myristyl acetate (PMA) was added a concentration per well of either 0.1 μg/ml or 0.001 μg/ml to induce homeotypic aggregation (human T-cells will proliferate in response to PMA at concentrations of 1-10 μg/ml). The T-cells proliferated in response to the low concentrations of PMA and in the presence of PBLs from AIDS patients. In contrast, at these low concentrations, PMA alone did not stimulate T-cell proliferation. The results demonstrate that the addition of low concentrations of PMA, which fail to induce T-cell proliferation but can induce homeotypic aggregation, will overcome the suppressive effects of TGF-? on antigen presenting cells. It can be concluded from these results that an agent which functions to promote maintenance of the conjugation between the antigen presenting cell and T-cell can overcome the TGF-/? suppressive effects.

EXAMPLE 5 This example illustrates that treating APCs with one or more cytokines can overcome TGF-β mediated suppression of APC function. Using the methods illustrated in Example 1 herein, mouse spleen cells were prepared as stimulator cells for the mixed lymphocyte response assay. In this experimental protocol, GM-CSF was tested for its ability, in a mixed lymphocyte response assay, to restore the suppressed allo-stimulatory function of TGF-β treated spleen cells. Irradiated spleen cells (1 x 10⁷) in 0.5 ml were incubated with 10 ng TGF-/? for 30 minutes. In the mixed lymphocyte response assay, to responder lymph node cells (1 x 10⁵ cells/well) were added either IO⁵ TGF-/? treated spleen cells per well (Table 6, "spleen cells + TGF-/?" ); IO³ TGF-/? treated spleen cells per well to which had been added recombinant murine granulocyte/monocytecolony stimulating factor (rmGM-CSF) , at 2000 U/ml during the five day assay. As a background control, one group of wells contained responder cells only (Table 6, "no spleen cells") . The data shown in Table 6 is representative of five repeated experiments. As shown in Table 6, the treatment of TGF-/? preincubated spleen cells with GM-CSF resulted in near complete recovery of their suppressed allo-stimulatory function.

TABLE 6

Stimulator Cells cpm (SI) no Spleen Cells 3,786 (9.6)

Spleen Cells + TGF-/3 408 (1.1)

Spleen Cells + TGF-3 3,282 (8.3) + GM-CSF

Using the methods according to Example 1 herein, GM-CSF was analyzed to overcome the TGF/?-mediated suppression of antigen presentation of ovalbumin by purified peritoneal macrophages to ovalbumin- sensitized lymph node T-cells (2 x 10⁵ nylon wool enriched LN cells/well from OVA immunized BALB/c mice) . As a control for background, one group of wells contained peritoneal macrophages and responder T-cells (without OVA; Table 7, "pMφ - OVA"). A positive control comprised TGF-/? treated peritoneal macrophages and responder cells with OVA (Table 7, "pMφ + OVA") . To assay the binding of TGF-/? to β glycan and subsequent inhibition of APC function, another group of wells contained peritoneal macrophages that were pre- reated with TGF-/? (48 hours with 2 ng/ml TGF-?) ; OVA; and responder cells (Table 7, "pM-φ + TGF-/? + OVA"). To assess the ability of GM-CSF to overcome the TGF/?- mediated suppression of antigen presentation of ovalbumin by purified peritoneal macrophages to ovalbumin-sensitized lymph node T-cells, 5000 U/ml of recombinant mouse GM-CSF were added per well to the five day proliferation assay (Table 7, "pMφ + TGF-/? + OVA + GM-CSF"). The data shown in Table 7 is representative of four repeated experiments. As shown in Table 7, the treatment of TGF-/? pre-treated peritoneal macrophages with GM-CSF resulted in near complete recovery of their suppressed allo-stimulatory function.

TABLE 7

Collectively, these results indicate that pre- treatment of macrophages with TGF/? inhibits APC function, and that GM-CSF will restore the TGF-β suppressed T-cell stimulatory function of antigen presenting cells. Since GM-CSF appears not promote the proliferation of T-cells directly (stimulatory effect has been described primarily for granulocytes ; see, e.g., Demetri et al . , 1992, Semin . Oncol . 19:362-85), it would appear that the cytokine is modifying the function of the TGF- β suppressed accessory cells. EXAMPLE 6

This ex-ample further illustrates that antigen presenting cells express /?glycan on their cell surface. Peripheral blood leukocytes (PBL) from AIDs patients were separated from peripheral blood using separation media and centrifugation methods known in the art (His opaque™; Sigma Chemical Company) . The PBL were then incubated overnight on plastic culture dishes, and the non-adherent cells were further separated on a 13.5% metrizamide gradient by centrifugation for 10 minutes at 400 x g. Low density cells were plated for 20 minutes at 22 °C on culture dishes coated with 100 μg/ml human IgG to remove Fc- receptor (FcR) positive cells. The FcR negative cells were then incubated for 30 minutes at 4°C with mouse antibodies made against human CD2 , CD3 , CD16, and CD19 (commercially available) . The cells were then washed in phosphate buffered saline, and then absorbed onto magnetic beads crosslinked with anti-mouse IgG (commercially available) to remove contaminating T- cells, NK cells, and B- cells. The resultant cells were determined by morphology and phenotype (HLA-DR^hl) to be >95% dendritic cells. This enriched population of dendritic cells was adhered to alcian blue- coated coverslips, and fixed in 2% formalin. The fixed cells were first blocked with rabbit IgG (control antibody) in 1 mg/ml bovine serum albumin, and then incubated with either fluorescein isothiocyanate (FITC) -conjugated rabbit anti-human TGF-/?1 antibody (O'Conner et al . , 1987, J. Biol . Chem. 262:14090-14099), or FITC- conjugated rabbit anti -human /?glycan antibody, or FITC- conjugated isotype control antibody; and then analyzed for immunofluorescence using confocal laser scanning microscopy. The same procedures were repeated for dendritic cells isolated from an HIV seronegative individual as a control, and dendritic cells isolated from an HIV seronegative individual which were pre- treated with TGF-/Ϊ before incubation with FITC- conjugated rabbit anti-human TGF-/?1 antibody.

Immunofluorescence analysis showed the following:

(a) that the dendritic cells from AIDS patients failed to bind to the FITC- conjugated isotype control antibody;

(b) dendritic cells isolated from an HIV seronegative individual did not have detectable TGF-β on their surface; (c) dendritic cells isolated from an HIV seronegative individual which were pre- reated with TGF- β stained positive for membrane associated TGF-/?l; (d) dendritic cells isolated from AIDS patients (greater than 12 patients tested) stained positive for membrane associated TGF-/?1; and (e) dendritic cells isolated from AIDS patients (greater than 12 patients tested) stained positive for membrane associated /?glycan. Three dimensional image analysis confirmed that in the cells staining positive TGF-βl or for βglycan, the signal is associated with the cell membrane, and not the cytoplasm, of the dendritic cells.

EXAMPLE 7 This example illustrates that one cause of the loss of T-cell stimulatory function of antigen presenting cells (e.g., dendritic cells, monocytes, and macrophages) observed in AIDS patients is due to TGF-/? mediated immunosuppression. This is in contrast to some reports (Cameron et al . , 1992, Clin . Exp. Immunol . 88:226-236; Weissman et al . , 1995, Proc. Natl . Acad. Sci . USA 92:826-830) that APC enriched PBL from HIV+ patients retained their ability to stimulate allo- responsive T-cells. As illustrated in Examples 1-6 herein, in an experimental mouse model it was demonstrated that antigen presenting cells express jβglycan on their cell surface; that /?glycan on APCs binds to TGF-/?; that APC function of stimulating CD4+ helper T cell activity is suppressed or inhibited by the binding of TGF- β to /?glycan; that TGF-/? :/?glycan complexes can be enzymatically removed from the cell surface of APCs; that removal of TGF-/?:/?glycan complexes from the cell surface of APCs resulted in restoration of the APC function of stimulating CD4+ helper T cell activity; that TGF-/? mediated suppression of APC function does not involve a soluble suppressive factor; that TGF- β mediated suppression of APC function does not require a trans -signalling event on the antigen presenting cell; that TGF-/? mediated suppression of APC allo-stimulatory function can be overcome by treatment with one or more agents that induce homeotypic aggregation or conjugation of the antigen presenting cells with T-cells; and that treating APCs with one or more cytokines can overcome TGF-/? mediated suppression of APC function.

The in vi tro T-cell stimulatory function of antigen presenting cells in peripheral blood from AIDS patients was assayed in a one-way mixed lymphocyte response assay. All patients in this study were diagnosed as having stage IV disease (Centers for Disease Control criteria) with peripheral CD4 counts ranging from 4 to 360 CD4+ T- cell/mm³. Peripheral blood leukocytes ("PBL") were separated from peripheral blood using separation media and centrifugation methods known in the art

(Histopaque™; Sigma Chemical Company) . The stimulator PBL were irradiated (3,500 rads) , and then added to a 96 -well plate at a concentration of 1 X 10^s cells/well in cell culture medium. Responder lymphocytes, from a control HIV-1 sero-negative volunteer, were added to the assay in cell culture medium at a concentration of 1 X 10⁵ cells/well. The assay was then incubated at 37°C for 7 days. Responder T-cell proliferation is pulsed for the last 18 hours with ³H thy idine. Cells were harvested onto glass wool filters and the filters are counted in 2 ml of scintillation fluid using a tritium window (expressed in cpm) .

The antigen presenting cell function of PBL isolated from AIDS patients was then compared to PBL isolated from healthy individuals for allo-stimulatory function in the one-way mixed lymphocyte response assay. The stimulation index (SI) is the fold increase in proliferation of T-cells in the presence of allo- stimulator cells in the mixed lymphocyte response assay, and is calculated by dividing the ³H- thymidine incorporation (cpm) of unsti ulated cells into the ³H- thymidine incorporation (cpm) from stimulated cells. By assaying the allo-stimulatory function of PBL from 66 AIDS patients, the mean allo-stimulatory function SI of 1.6 ± 0.1 standard error of the mean. By comparison, PBL from seven healthy donors had a mean SI of 10.3 ± 2.2 standard error of the mean. Thus, there is a statistically significant suppression of allo-stimulatory function of PBL in AIDS patients. To assay the binding of TGF-/? to /?glycan on human antigen presenting cells, and subsequent inhibition of APC function, human PBL (1 x 10⁷ cells) from a healthy individual donor were incubated in 0.5 ml of cell culture medium with TGF-/? (10 ng) for varying times (either 30 minutes, 2 hours, 4 hours, 24 hours, or 48 hours) at 37°C. After pre- treatmen , the stimulator cells were then washed four times in cell culture medium, and then irradiated with 3500 rads , before addition (10⁵ spleen cells per well) to the human lymphocyte responder cells in the mixed lymphocyte response assay. The APC function of TGF-/? treated PBL was compared to that of untreated PBL from the same individual .

The results, illustrated in Fig. 1, indicate that when the stimulator PBL were incubated with TGF-? for as short a period of time as 30 minutes, observed is a statistically significant suppression of the PBL antigen presenting cell function of stimulating allotypic T- cells. The mean Stimulation Index (SI) decreased from a pre- incubation level of 9.3 to 3.5 following a 30 minute exposure to TGF-/?. A conclusion from these results is the novel discovery that in certain diseases (e.g., AIDS) , the function of antigen presenting cells is not lost, but rather masked by TGF-/?, and therefore recoverable. It is noted that PBL from AIDS patients reportedly secrete TGF-βl when cultured in vi tro (Lotz et al . , 1994, J. Cell Biol . , 124:365-371). Thus, it was necessary to determine if the lack of T-cell proliferation in response to PBL from AIDS patients in the in vi tro mixed lymphocyte response assay was due to the direct suppressive effect of de novo secreted TGF-/?1 on the responding T-cells. Using the methods illustrated herein, equal numbers of irradiated PBL from AIDS patients and irradiated PBL from uninfected controls were mixed before their addition as stimulator cells to a mixed lymphocyte response assay. Control antigen presenting cells maintained full allo-stimulatory function in the presence of PBL from AIDS. In other experiments, conditioned media from in vitro cultured PBL derived from AIDS patients showed no suppression of T-cell prolifera ion in a mixed lymphocyte response assay that was stimulated by normal PBL. Based on these results, the defect in allo-stimulatory function of antigen presenting cells from AIDS patients appears not to be the result of in vi tro production of any soluble suppressive factors.

EXAMPLE 8

This example illustrates that (a) antigen presenting cells in human PBL express /?glycan on their cell surface; (b) that βglycan and the TGF-/?:/?glycan complex can be enzymatically removed from the cell surface of these APCs; and (c) that removal of TGF- /?:/?glycan complex on the cell surface of these APCs resulted in restoration of the APC function of stimulating CD4+ helper T cell activity. To demonstrate that TGF-/? mediated suppression of APC function in human PBL is associated with the presence of /?glycan:TGF-/? complexes, a mild trypsin digestion was used to remove /?glycan:TGF-/?l complexes from the membranes of both TGF-βl pre- treated normal PBL and PBL from AIDS patients. Like other proteoglycans, /?glycan has a single accessible trypsin cleavage site in the membrane proximal region of its extra- cellular domain and is extremely sensitive to tryptic digestion. Using the methods illustrated in Example 7 herein, a mixed lymphocyte response assay was performed. Irradiated human PBL were incubated for 30 minutes in cell culture medium either in the absence (Table 8, "none") or the presence of 10 ng of TGF-/? for 30 minutes at 37°C (Table 8, "+ TGF-/?"). Also, a portion of the TGF-? treated PBL were incubated with a solution of trypsin/EDTA (GIBCO) for 10 minutes at 37°C before addition to the assay (Table 8, " + TGF-β + Trypsin"). In the mixed lymphocyte response assay, also analyzed was PBL from AIDS patients (Table 8, AIDS- "None"), and PBL which were incubated with trypsin/EDTA for 10 minutes at 37°C before addition to the assay (Table 8, AIDS- "Trypsin"). In the assay, 1X10⁵ washed, irradiated (3500 rad) stimulator cells/well were mixed with 1X10⁵ responder cells/well from a different donor as allo-responder cells. The assay was incubated for seven days and ³H- thymidine added for the final 18 hours. The values are the mean of six replicate wells for each of three experiments, and are expressed as the stimulation index (SI) with the standard error of the mean (± SEM) . TABLE 8

Trypsinization restores allo-stimulatory function of normal and AIDS PBL

Cells Treatment mean SI (± SEM)

CONTROL None 9.3 (0.6)

Normal + TGF/? 3.5 (0.9)

Normal + TGFjS + Trypsin 10.8 (3.3)

AIDS None 1.5 (0.2)

AIDS Trypsin 6.5 (1.8)

As shown in Table 8, trypsinization completely restored T-cell stimulatory function to the TGF-/? pre- treated PBL from the healthy donor and restored statistically significant allo-stimulatory function to PBL from AIDS patients (p<0.001). The 10 minute trypsinization of PBL resulted in an increase in the allo-stimulatory function of PBL from each of the 12 AIDS patients assayed.

EXAMPLE 9 This example further illustrates that in certain pathological conditions, TGF-/? mediated suppression of APC function in human PBL is associated with the presence of /?glycan:TGF-/? complexes. The role of TGF-/? in the suppression of antigen presenting cell function in advanced AIDS patients was confirmed herein by using antibodies which recognize epitopes on TGF-/?1 when it is complexed with glycan. The antibodies function to neutralize the intercellular interaction of the APCs with the responder T-cells (e.g., inhibit the presentation of the complexed TGF-β on the APC to the TGF-/? receptor of T-cells) . Additionally, it is demonstrated herein that antibodies which recognize epitopes on the βglycan component of /?glycan/TGF-/? complexes also function to neutralize the intercellular interaction of the APCs with the responder T-cells.

Using the methods illustrated in Example 8 herein, a mixed lymphocyte response assay was performed wherein PBL isolated from AIDs patients were compared for antigen presenting function without antibody treatment ("untreated"); in the presence of anti -TGF- /Si antibody (lOμl/ml included for the duration of the assay; R&D Systems) ; in the presence of control antibody (lOμl/ml of anti-human IgA included for the duration of the assay), or in the presence of rabbit anti-?glycan antiserum (lOμl/ml included for the duration of the assay). The results are illustrated in Table 9, wherein the stimulation index (SI) is the fold increase in proliferation of T-cells in the presence of allo- stimulator cells in the mixed lymphocyte response assay expressed with the standard error of the mean (± SEM) .

TABLE 9

Anti-TGF-/? and anti-/?glycan restore allo-stimulatory function of PBL from AIDS patients.

Treatment SI (± SEM) p value untreated 1.4 (0.2) - anti-TGF-/? Ab 9.8 (0.3) <0.001 untreated 1.4 (0.2) - anti -IgA <1.0 not significant untreated 1.9 (0.1) - anti-jSglycan Ab 12.7 (3.3) <0.01

As shown in Table 9, the inclusion of anti-TGF-/?l antibody in the mixed lymphocyte response assay resulted in statistically significant restoration of allo-stimulatory function to PBL from each of 15 AIDS patients assayed. The allo-stimulatory function of PBL from these patients increased significantly from a mean SI of 1.4 to a mean SI of 9.8 in the presence of anti-TGF- βl antibody. The addition of a control antibody, (anti-human IgA) failed to restore allo-stimulatory function to PBL from these patients.

Further confirming that the TGF-βl mediated suppression in these patients is the result of the presence of /?glycan/TGF-/?l complexes, inclusion of anti-/?glycan antibody resulted in statistically significant restoration of allo-stimulatory function to PBL from each of 6 AIDS patients assayed. The allo-stimulatory function of PBL from these patients increased significantly from a mean SI of 1.9 to a mean SI of 12.7 in the presence of anti-/?glycan antibody. Normal rabbit sera, used as a control antibody for the anti-/?glycan antibody experiment, had no significant effect on the allo-stimulatory function of PBL from AIDS patients. In additional control experiments, it was found that anti-TGF-/?l and anti-/?glycan antibodies were not directly mitogenic for the responder cells; and that the antibody- facilitated T-cell proliferation is inhibited by the addition of anti-HLA-DR antibodies, and therefore appears to be MHC class II restricted.

EXAMPLE 10

This example illustrates that treating APCs in human PBL with one or more cytokines can overcome TGF-/? mediated suppression of APC function. In previous Examples herein it is clear that dendritic cells in AIDS patients have the cytokine TGF-/? present on their cell membranes, thereby forming immunosuppressive complexes with the membrane proteoglycan βglycan. The enzymatic removal of, or blocking by antibodies of, these complexes restores T-cell stimulatory function to these APCs, indicating that accessory cell function is not lost. In this Example, illustrated is the in vi tro effect of GM-CSF on the suppressed accessory cell function of PBL from AIDS patients.

Using the methods illustrated in Examples 7-9 herein, PBL from AIDS patients were prepared as stimulator cells for assaying the in vi tro T-cell stimulatory function of APCs in a one-way mixed lymphocyte response assay. Briefly, the separated PBL were irradiated (3,500 rads) and then added to a 96-well plate at a concentration of 1 X 10 cell/well in cell culture medium. Responder peripheral blood leukocytes (PBL), from an HIV-1 volunteer, were added to the assay in CM at a concentration of 1 X 10⁵ cells/well. The assay was incubated at 37°C for 7 days, with responder T-cell proliferation being assayed by overnight ^[3,thymidine incorporation, and quantitation of the counts per minute from harvested cells. In some wells of the mixed lymphocyte response assay, recombinant human GM- CSF (rh GM-CSF) was added at a concentration of lOng/ml for the duration of the assay. In some experiments, irradiated PBL from AIDS patients were preincubated at 37°C for 30 minutes in lOng/ml rh GM-CSF, and then washed three times with culture medium before inclusion to the mixed lymphocyte response assay as stimulator cells. As shown in Table 10, in a mixed lymphocyte response assay, the presence of GM-CSF resulted in a statistically significant (p<0.001) restoration of T-cell stimulatory function to the antigen presenting function of PBL in each of 16 AIDS patient samples assayed. Allo-stimulatory function increased from a mean SI of 1.7, SEM ± 0.2, for untreated PBL, to a mean SI of 4.5, SEM ± 0.6 in the presence of rh GM-CSF (values for each patient represents the mean of six replicate wells) . TABLE 10

Patient Pretreatment + GM-CSF # cpm (SI) cpm (SI)

1 1,228 (2.2) 2,143 (3.9)

2 587 (1.1) 2,008 (3.7)

3 1,341 (2.5) 2,047 (3.8)

4 470 (1.6) 1,540 (5.4)

5 1,948 (3.6) 3,413 (6.3)

6 482 (0.9) 1,342 (2.5)

7 913 (1.7) 1,433 (2.6)

8 817 (1.5) 2,370 (4.4)

9 939 (3.3) 2,901 (10.2)

10 347 (1.2) 2,394 (8.4)

11 627 (2.2) 1,870 (6.6)

12 268 (1.0) 577 (2.2)

13 340 (1.3) 1,026 (2.7)

14 163 (0.3) 1,035 (1.8)

15 458 (0.8) 830 (1.5)

16 564 (2.2) 1,568 (6.2)

As shown in Table 11, pre-treating PBL with GM-CSF for 30 minutes before adding the PBL into a mixed lymphocyte response assay, resulted in a statistically significant (p<0.001) restoration of T-cell stimulatory function to the antigen presenting function of PBL in 7 of 9 AIDS patient samples assayed. Allo-stimulatory function increased from a mean SI of 1.6 for untreated PBL, to a mean SI of 4.7 for PBL pre- treated with rh GM-CSF.

TABLE 11

Patient Non- treated GM-CSF treated # cpm (SI) cpm (SI)

1 1728 (3.9) 3117 (7.2) 2 585 (1.3) 434 (1.0)

3 425 (1.0) 822 (1.8)

4 242 (1.2) 286 (1.4)

5 412 (2.0) 1,046 (5.0)

6 963 (1.8) 10,150 (19.4)

7 377 (0.7) 860 (1.6)

8 678 (1.3) 1,359 (2.3)

9 766 (1.5) 1,264 (2.5)

As demonstrated by the results in Tables 10 & 11, treatment with GM-CSF restores the in vi tro allo-stimulatory function of PBL from AIDS patients. Additional experiments were performed to confirm that GM-CSF directly affects the APCs, and has no direct mitogenic effect on the responder PBL in the mixed lymphocyte response assay. Table 12 shows the results of incubating rh GM-CSF (10 ng/ml) to PBL (1 x IO⁵ cells/well) from three normal controls in a seven day proliferation assay. The SI is calculated by dividing the cpm from wells with GM-CSF by the cpm from wells without GM-CSF. The values are the mean of six replicate wells. These results confirms that GM-CSF fails to induce the proliferation of normal control PBL.

TABLE 12

Control PBL with GM-CSF cpm (SI)

1 673 (2.7)

2 593 (2.7)

3 943 (2.8)

EXAMPLE 11 This example further illustrates that treating APCs in human PBL with one or more cytokines can overcome TGF-/S mediated suppression of APC function. In the previous Example herein, it was demonstrated in vi tro that GM-CSF restores T-cell stimulatory function of PBL, from AIDS patients, which had been previously suppressed by TGF-/?. GM-CSF signals through a transmembrane heterodimeric receptor composed of a cytokine specific a- chain and a common signal transducing β - chain (Sato et al., 1993, EMBO J. 12:4181). Interleukin -3 (IL-3) is another member of this family of cytokines which shares this common β - chain as a signal transducer. This family of cytokines which bind to their receptors that share a common beta subunit, wherein the subunit binds to Janus protein kinase 2 (Jak2), include GM-CSF, IL-3, and IL-5 (Sakai and Kraft, 1997, J. Biol . Chem. 272:12350-8). This family of cytokines may be classified as a group of antigen presenting cell activating factors. To determine whether other cytokines of this family can potentiate the T-cell stimulatory function of antigen presenting cells from AIDS patients, the effect of IL-3 on the allo-stimulatory function of PBL from AIDS patients was evaluated in a mixed lymphocyte response assay.

Using the methods illustrated in Examples 7-9 herein, PBL from AIDS patients were prepared as stimulator cells for assaying the in vi tro T-cell stimulatory function of APCs in a one-way mixed lymphocyte response assay. Briefly, the separated PBL were irradiated (3,500 rads) and then added to a 96-well plate at a concentration of 1 X 10⁵ cell/well in cell culture medium. Responder PBL, from an HIV-l negative volunteer, were added to the assay in CM at a concentration of 1 X 10^s cells/ well. The assay was incubated at 37°C for 7 days, with responder T-cell proliferation being assayed by overnight ^,3,thymidine incorporation, and quantitation of the counts per minute from harvested cells. In some wells of the mixed lymphocyte response assay, recombinant human IL-3 (rh IL-3) was added at a concentration of lOng/ml for the duration of the assay.

As shown in Table 13, in a mixed lymphocyte response assay, the presence of IL-3 resulted in a statistically significant (p<0.001) restoration of T-cell stimulatory function to the antigen presenting function of PBL in the group of nine AIDS patient samples assayed. The mean SI of 1.5 ± 0.3 (without IL-3 treatment) increased significantly to an SI of 5.5, SEM ± 1.1 in the presence of IL-3 (values are the mean of six replicate wells) . It is noted that the one non- responding patient's PBL did have an increased allo-stimulatory function in the mixed lymphocyte response assay when treated with GM-CSF.

TABLE 13

Patient No -treated IL-3 treated # cpm (SI) cpm (SI)

1 470 (1.6) 2,414 (8.5)

2 939 (3.3) 3,288 (11.6)

3 347 (1.2) 2,434 (8.6)

4 627 (2.2) 618 (2.2)

5 268 (1.0) 1,050 (4.1)

6 163 (0.3) 1,171 (2.1)

7 458 (0.8) 1,111 (2.0)

8 340 (1.3) 1,026 (4.1)

9 564 (2.2) 1,762 (6.9)

In summary, Examples 10 and 11 provide evidence that the family of cytokines which share a common beta subunit for their receptors (e.g., GM-CSF, or IL-3) will restore in vi tro significant T-cell stimulatory function to PBL from AIDS patients. Using the methods illustrated in this Example and Examples 7-9 herein, PBL from AIDS patients were tested as stimulator cells for assaying the in vi tro T-cell stimulatory function of APCs in a one-way mixed lymphocyte response assay, in the presence of or in the absence of PIXY321 (lOng/ml) for the duration of the assay. PIXY321 is an IL- 3/GM-CSF fusion protein which demonstrates enhanced biological activity over GM-CSF or IL-3 alone or in combination (Mignacca et al . , 1996, Blood 88:848-854; Gillis, 1991, Breast Cancer Res . Treat . 20Suppl: S3-9); and which in clinical phasel/II studies has been shown to increase the number of circulating hematopoietic progenitors and differentiated hematopoietic cells (Gheilmini et al . , 1996, Sr. J. Haemtol . 93:6-12). PIXY321 is also a potent stimulator of T-cell proliferation. As shown in Table 14, addition of PIXY321 to PBL from AIDS patients led to a statistically significant (p=0.0095) restoration of allo-stimulatory function in all five of the AIDS patient samples tested. Thus, PIXY321 may serve the dual role of restoring antigen presenting cell function and stimulating CD4+ T-cell regeneration.

TABLE 14

Cells Untreated; SI (±SEM) PIXY321; SI (±SEM)

AIDS PBL 1.1 (± 5.4) 17.7 (±7.9)

Using the mixed lymphocyte response assay, a number of antigen presenting cell activating factors (other than the cytokines described above) were individually tested for their ability to restore the allo-stimulatory function of PBL from AIDS patients. The results indicate antigen presenting cell activating factors can restore in vi tro antigen presenting cell function of PBL from AIDS patients or other patients exhibiting TGF-/? mediated suppression of allo-stimulatory function of

APCs. These antigen presenting cell activating factors may include, but are not limited to, CD40 ligand, fit- 3, B-7 family of cross-linking reagents, ICAM- l/ICAM-2 crosslinking reagents, MHC cross -linking reagents, and human monocyte chemotactic proteins (MCP) -2 and -3. Additionally, it is known that tumor necrosis factor alpha (TNF-α) strongly potentiates the biological effects of GM-CSF and IL-3 (Caux et al . , 1996, Blood 87:2376-85). Thus, therapeutically effective amounts of combinations of TNF-α and GM-CSF; TNF-α and IL-3; TNF-α, GM-CSF, and IL-3; and TNF-α and PIXY321, may be used to effect in vi tro or in vivo restoration of the allo- stimulatory function of PBL from AIDS patients or other patients exhibiting TGF-/? mediated suppression of allo- stimulatory function of APCs.

EXAMPLE 12 This example illustrates that treating APCs in human PBL with one or more cytokines can overcome TGF-/? mediated suppression of APC function in vivo . In vivo administration of such cytokines, or other antigen presenting cell activating factors, may effect significant restoration of T-cell stimulatory function to antigen presenting cells from most AIDS patients; and may be used as a component of comprehensive multi -modal treatment for AIDS patients (or other patients exhibiting TGF-/3 mediated suppression of allo- stimulatory function of APCs) designed to restore or improve immune function.

There are many reports of clinical use of GM-CSF in HIV+ patients and in HIV+ patients with AIDS -related malignancies. These reports teach that GM-CSF use is directed to correcting leukopenia and improving neutrophil function in these patients without, or in combination with, antiretroviral therapy with zidovudine (see, e.g., Miles, 1991, Cancer Invest . 9:229-38; Miles, 1992, AIDS Res . Hum. Retroviruses 8:1073-80; Mitsuyasu, 1991, .Rev. Infect . Dis . 13:979-84; Krown et al . , 1992, J. Clin . Oncol . 10: 1344-51; Hengge et al . , 1992, Clin . Invest . 70:922-6; and Hewitt et al . , 1993, Ar-tiiπicroJ . Agents Chemother. 37:512-22). However, it is known in the art that the biological effects of GM-CSF observed " in vi tro have not been striking or consistent in clinical trials" (Demetri and Ant an, 1992, Semin . Oncol . 19:362-85). Further, not described previously is a method for either the in vi tro or in vivo restoration of allo-stimulatory function of antigen presenting cells, in APCs having such function suppressed by /?glycan-TGF-/? complex formation (TGF- β mediated suppression of APC function) . Thus, it was necessary to conduct clinical trials to evaluate whether treating APCs in human PBL with one or more cytokines can overcome TGF-/? mediated suppression of APC function in vivo, as observed in vi tro.

In this clinical trial, a group of twelve AIDS patients were infused intravenously with 125 μg/m³ of recombinant human GM-CSF over a two hour period. This regiment of GM-CSF treatment has been shown to produce serum levels of GM-CSF as high as lOng/ml. To qualify for the study, patient volunteers had to conform to clinical guidelines that include 1) having CDC stage IV disease, 2) not receiving any other immunomodulating therapy at the time of trial, and 3) otherwise displaying reasonable good health.

Using the methods as previously described herein, the allo-stimulatory function of peripheral blood mononuclear cells was assayed in a mixed lymphocyte response assay before, and at various time points after, the completion of the single bolus infusion. Venous blood was drawn from the contralateral arm 1 hour, 5 hours and 24 hours after cytokine treatment. The allo- stimulator function of each patient was assayed against the PBL from a healthy individual (selected as being the best responder out of a group of four responders assayed) . The pre-infusion proliferative response was normalized to l-unit of response and the mean proliferation of the post infusion time points were represented as fold increase compared to pre- infusion proliferation.

The results of the clinical trial are shown in FIG. 2, wherein the level of allo-stimulatory function at 1 hour post -infusion is represented by D, the level of allo-stimulatory function at 5 hours post- infusion is represented by I, and the level of allo-stimulatory function at 24 hours post- infusion is represented by §.

Absence of a level at a specified time period for a particular patient indicates a specimen was not obtained for that time period. Of the 12 patients in the trial, only one patient failed to show an increase in allo-stimulatory function at one or more time points after GM-CSF infusion. A statistically significant recovery of antigen presenting cell function was seen in 9 of the AIDS patients. Two basic patterns of response to GM-CSF infusion were seen. As shown in FIG. 2, patients #2, #5, and #12 had an early response that had regressed by 24 hours; while patients #4, #7, #8, #-9, #10, and #11 had later maximal responses. One of these late responders, #7, showed the maximal recovery of antigen presenting cell function at 24 hours after the cytokine infusion.

It is noted that due to the short biological half life of GM-CSF (only 15-20 minutes) , a therapeutic infusion of GM-CSF will result in a brief period of high systemic cytokine concentration. Therefore, in view of the results shown in FIG. 2, antigen presenting cells (a) must be able to recover T-cell stimulatory function after only a limited exposure to GM-CSF, and (b) appear to maintain that function after the cytokine exceeds its serum half -life. The results shown in FIG. 2 demonstrate that a single infusion of GM-CSF can restore significant T-cell stimulatory function of APCs in most AIDS patients tested; and suggests that prolonged usage of cytokines (of the family of cytokines which share a common beta subunit for their receptors; e.g., GM-CSF, IL-3, etc.) should be considered for inclusion as part of comprehensive multi -modal treatment for AIDS patients designed to restore or improve immune function.

EXAMPLE 13

This example illustrates a first embodiment of the method according to the present invention. In one embodiment, provided is a method for restoring T-cell stimulatory function of APCs of an individual having such function suppressed by /?glycan-TGF-|3 complex formation, wherein the method comprises treating the APCs by the steps of:

(1) isolating a population of cells enriched for APCs from the individual; and

(2) contacting the enriched APC population in vi tro with a therapeutically effective amount of a composition selected from the group consisting of a means to remove /?-glycan or /Sglycan complexed to TGF-/? from the cell surface of the APCs, a means to prevent presentation of complexed TGF-β (complexed to _/Sglycan) , to TGF-/? receptors on T-cells, and one or more antigen presenting cell activating factors which function as co- stimulatory molecules on APCs, thereby resulting in overcoming the suppression of the T-cell stimulatory function of APCs.

This method may further comprise infusion (e.g., intravenously) of the treated APCs into back into the individual, wherein the treated cells are expected to migrate back into various body tissues (e.g., blood, spleen, lymph nodes, etc.). The means to remove /?-glycan or /?glycan complexed to TGF-/? from the cell surface of the APCs, involve contacting the cells with a trypsin/EDTA solution. The means to prevent presentation of complexed TGF-/? to TGF- β receptors on T-cells include the use of anti-TGF-/? antibodies or anti-/?glycan antibodies or a combination thereof. The one or more antigen presenting cell activating factors may include, but are not limited to, GM-CSF, IL-3, PIXY321, CD40 ligand, flt-3, B-7 family of cross-linking reagents, ICAM-l/ICAM-2 crosslinking reagents, MHC cross- linking reagents, MCP-2, MCP-3, TNF- a and GM-CSF, TNF-o. and IL-3, TNF-α? and PIXY321, and a combination thereof. For this method of treatment, the relative ratio and total amounts of the composition can be determined by techniques known to those skilled in the art, including the mixed lymphocyte response assay described in detail herein.

Isolating from an individual a population enriched in antigen presenting cells may be accomplished by any one of several techniques known in the art. As described in detail in Example 6 herein, and as further described previously (Macatonia et al . , 1992, Immunol . 75:576-581) peripheral blood leukocytes (PBL) can be separated from peripheral blood drawn from an individual. The PBL can be incubated overnight on plastic culture dishes, resulting in adherent cells and non-adherent cells. The adherent cells may be removed by gentle scraping, and represent a population of macrophages . The nonadherent cells are collected and then separated by gradient centrifugation, and low density cells are panned to remove FcR + macrophages, and antibody labelled non antigen presenting cells. Resultant cells may then be analyzed for morphology and phenotype (HLA-DR¹") to determine enrichment of a dendritic cell population. Similar panning techniques may be used for the enrichment of Langerhans cells (Koch et al., 1992, J. Invest . Derma tol . 99:803-807). Alternatively, using flow cytometric methods (Szabolcs et al., 1996, Blood 87:4520-4530, herein incorporated by reference) , a population enriched in macrophages (CD14+, HLA-DR+) and a population of dendritic cells (CD14-, HLA-DR¹") can be obtained.

EXAMPLE 14

Another embodiment of the method according to the present invention is a method for restoring T-cell stimulatory function of APCs in an individual having such function suppressed by /?glycan-TGF-/? complex formation, wherein the method comprises treating the APCs in vivo by administering to the individual a therapeutically effective amount of a composition comprising one or more antigen presenting cell activating factors which function as co- stimulatory molecules on APCs, thereby resulting in overcoming the suppression of the T-cell stimulatory function of APCs. Alternatively, the composition may comprise a therapeutically effective amount of antibody selected from the group consisting of an anti-TGF-/? antibody, an anti-/?-glycan antibody, or a combination thereof. The antibody functions to inhibit (blocking) the presentation of the complexed TGF-/? on the surface of the APC to the TGF-/? receptor present on the T-cell, and/or may function to prevent TGF-/3 :/?glycan complexes from forming.

The composition may be given to the individual in any one of several routes of administration known in the art; however, a preferred route of administration is infusion intravenously. The relative ratio and total amounts of the composition may be determined by a physician considering such factors as the age, weight, health, and possibly other clinical factors, of the individual. The composition may further comprise a pharmaceutically acceptable carrier medium known in the art for use in facilitating the administration of therapeutic compositions (e.g., saline solution, parenterally acceptable carbohydrate or protein solutions, liposomes, etc.). The one or more antigen presenting cell activating factors may include, but are not limited to, GM-CSF, IL-3, PIXY321, CD40 ligand, flt-3, B-7 family of cross-linking reagents, ICAM-l/ICAM-2 crosslinking reagents, MHC cross- linking reagents, MCP-2, MCP-3, TNF-o. and GM-CSF, TNF-α and IL- 3, TNF-α and PIXY321, and a combination thereof.

From the foregoing, it will be obvious to those skilled in the art that various modifications in the above-described methods, and compositions can be made without departing from the spirit and scope of the invention. Accordingly, the invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof . Present embodiments and examples, therefore, are to be considered in all respects as illustrative and not restrictive, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

Claims

What is claimed is:

1. A method for restoring T-cell stimulatory function of antigen presenting cells of an individual having such function suppressed by /?glycan-TGF-/? complex formation, wherein the method comprises treating the antigen presenting cells by the steps of:

(a) isolating a population of cells enriched for antigen presenting cells from the individual; and

(b) contacting the enriched antigen presenting cell population in vi tro with a therapeutically effective amount of a composition selected from the group consisting of a means to remove /?-glycan or /Sglycan complexed to TGF- β from the cell surface of the antigen presenting cells, a means to prevent presentation of complexed TGF-/? to TGF-/? receptors on T-cells, and one or more antigen presenting cell activating factors which function as co- stimulatory molecules on antigen presenting cells thereby resulting in overcoming the suppression of the T-cell stimulatory function of antigen presenting cells.

2. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 1, further comprising infusing the treated antigen presenting cells back into the individual.

3. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 1, wherein the antigen presenting cells are contacted in vi tro with a therapeutically effective amount of a composition consisting of a means to remove /3-glycan or _/Sglycan complexed to TGF-/? from the cell surface of the antigen presenting cells, and wherein such means comprises contacting the cells with a trypsin/EDTA solution.

4. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 1, wherein the antigen presenting cells are contacted in vi tro with a therapeutically effective amount of a composition selected from the group consisting of anti-TGF-/3 antibody, anti-/3glycan antibody, and a combination thereof.

5. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 1, wherein the antigen presenting cells are contacted in vi tro with a therapeutically effective amount of a composition comprising one or more antigen presenting cell activating factors, and wherein the antigen presenting cell activating factor is selected from the group consisting of GM-CSF, IL-3, PIXY321, CD40 ligand, flt-3, B-7 family of cross -linking reagents, ICAM-l/ICAM-2 crosslinking reagents, MHC cross -linking reagents, MCP-2, MCP-3, TNF-α and GM-CSF, TNF-α and IL- 3, TNF-Q! and PIXY321, and a combination thereof.

6. A method for restoring T-cell stimulatory function of antigen presenting cells in an individual having such function suppressed by /?glycan-TGF-/? complex formation, wherein the method comprises treating antigen presenting cells in vivo by administering to the individual a therapeutically effective amount of a composition comprising one or more antigen presenting cell activating factors which function as co- stimulatory molecules on antigen presenting cells, thereby resulting in overcoming the suppression of the T-cell stimulatory function of antigen presenting cells.

7. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 6, wherein the antigen presenting cell activating factor is selected from the group consisting of GM-CSF, IL-3, PIXY321, CD40 ligand, flt-3, B-7 family of cross-linking reagents, ICAM-l/ICAM-2 crosslinking reagents, MHC cross- linking reagents, MCP-2, MCP-3, TNF-α and GM-CSF, TNF-α and IL-3, TNF-α and PIXY321, and a combination thereof .

8. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 6, wherein the composition further comprises a pharmaceutically acceptable carrier medium.

9. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 6, wherein the composition is administered by intravenous infusion.

10. A method for restoring T-cell stimulatory function of antigen presenting cells in an individual having such function suppressed by /Sglycan -TGF-β complex formation, wherein the method comprises treating antigen presenting cells in vivo by administering to the individual a composition comprising a therapeutically effective amount of an antibody selected from the group consisting of anti-TGF-/? antibody, anti-/?glycan antibody, and a combination thereof .

11. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 10, wherein the composition further comprises a pharmaceutically acceptable carrier medium.

12. The method for restoring T-cell stimulatory function of antigen presenting cells according to claim 10, wherein the composition is administered by intravenous infusion.