Progressin GrowthFactorResearch,Vol. 3, pp. 159-175,1991 Printed in Great Britain.

0955-2235l'91$0.00+ .50 1991PergamonPressplc

TRANSFORMING GROWTH FACTOR-,// AND THE IMMUNE SYSTEM F r a n c i s W . R u s c e t t i * a n d M i c h a e l A . Palladino~" *Laboratory of Molecular lmmunoregulation Biological Response Modifiers Program NCI-Frederick Cancer Research and Development Center Frederick, MD 21702-1201, U.S.A. tDepartment of Cell Biology Genentech, Inc., S. San Francisco, CA 94080, U.S.A.

It is now apparent that the transformhlg growth factor fl (TGF-fl) family of proteins has potent hmntmoregulatory properties rangh~g front effects oll the growth and differentiation of prhnitive stem cells to the differentiated fiawtions of hnmune effector cells. Although most reports have described the hnmunosuppressive activities of TGF-fl, recent evidence supports the concept that TG F-fl can have both hthibitory and sthmdatory actions on these systems. Recently, it has been found that TGF-fl can have atttocrflle as well as paracrhw effects on the immune system, hldicath~g that hnmune cells can activate the htactive secreted form of TGF-fl. Fttrthermore, TGF-fl has differential hltracelhdar effects on cell surface receptor modtdation, tyroshze phosphorylation, and ck'tokhle gene transcription as well as cell-mediated cytotoxicity, hnportantly, the admhffstration of TGF-fl has proven beneficial hz several anhnal disease models such as septic shock, allograft rejection, and atttohnmunio'. Moreover, the hwreased levels of TGF-flfo,md hl several disease states associated with hnmunosuppression such as different forms of malignancy, chronic degenerative diseases, and AIDS hnplicate the hzvolvement of TGFfl ht the pathogenesis o f some diseases. Ulthnately, well designed clinical trials will determhw whether the excithtg potential of TGF-fl can be used to treat or prevent disease. Keyaords: Immunoregulation, transforming growth factor fl, receptor modulation, cytokines, hematopoiesis, leukemic cell growth. INTRODUCTION A l t h o u g h transforming g r o w t h factor-beta 1 ( T G F - f l l ) was originally named for its ability to induce a reversible phenotypic transformation o f normal rat fibroblasts, it is n o w apparent that the T G F - f l family o f proteins provides important regulatory signals for m a n y cell processes [1, 2]. In several species, the conservation o f

Acknowledgements--The authors are grateful to Drs Joost Oppenheim and Jeff Rossio for critical review of this manuscript. 159

160

F.W. Ruscetti and M.A. Palladino

the genes encoding the various TGF-fl molecules, the sequence homology between the various TGF-p proteins as well as the wide range of cell types that express TGF-/3 receptors support the fundamental role of these molecules in regulating basic biologic processes. Now that multiple members ofthe TGF-fl family ofproteins are available in highly purified forms through the use of recombinant DNA technologies, the hi vivo activities of these molecules can be defined. Recently, TGF-fl was demonstrated to regulate immune cell activities hz vitro. This wide range of immunoregulatory activities suggested that TGF-p may regulate immune functions h~ vivo. In this review, we will discuss recent data on the hi vitro and h~ vivo immunoregulatory properties of recombinant human TGF-fll (rHuTGF-fll) and rHuTGF-fl2. Although we are in the early stages of understanding and controlling the immunoregulatory properties of TGF-fl(s) hi vivo, the data suggest that they constitute an important class of immunosuppressive molecules with potential therapeutic utility. The TGF-fl Family: Structure and Function

Multiple species of TGF-fl exist which are encoded by different genes. In 1985 the cDNA for TGF-fll was cloned by Derynck et aL [3]. Seyedin et aL isolated a polypeptide factor from bovine bone that induced type II collagen synthesis [4]. The factor, termed TGF-fl2, was cloned and exhibited 71% amino acid sequence identity to the mature active form of TGF-fl and had functional similarity to it [5, 6]. A third TGF-fl, designated TGF-fl3, showed 75% and 80% identity in the mature polypeptide region with TGF-fll and TGF-fl2, respectively, and demonstrated functional similarities to the other TGF-fl species [7]. Mammals can express all three types of TGF-fl. In addition, mRNA for TGF-fl4 has been detected only in chicken chondrocytes and for TGF-fl5 in the frog, Xenopus laevis [8, 9]. All TGF-fls consist of a disulfide-linked homodimer. Each chain of 112 amino acids is synthesized as the C-terminal domain of a precursor of approximately 390 amino acids. The glycosylated precursor contains a hydrophobic signal sequence for translocation across the endoplasmic reticulum. The cleavage site is a sequence of four basic amino acids immediately preceding the active monomer. The degree ofhomology between all five members varies from a low of 64% (TGF-fll versus TGF-fl4) to 82% (TGF-fl2 versus TGF-fl4). Among the three mammalian TGFfls, each form is individually highly conserved (>97%). The fact that these genes have the same genomic organization (7 exons) suggest that they arose by gene duplication. The five TGF-fls currently identified belong to a larger gene family that shares structural similarity, especially in the C-terminal cysteine-rich portion of the protein where seven out of nine cysteines are conserved in all family members. They are functionally related in their ability to regulate cellular growth and differentiation. Members of this family include two forms ofinhibin, three forms ofactivin, Miillerian inhibitory substance, the transcript of the decapentaplegic gene complex of the fruitfly Drosophila, several bone morphogenic proteins, and the putative products of the Vgl gene in the frog and mammalian Vgrl. The abilities of these family members to alter immune functions hi vitro are outlined in Table 1. The actions of the activins and inhibins on the immune system are much more limited than the TGF-fls. Activins stimulate erythropoiesis indirectly through activation ofcytokine production by T cells [10]. Inhibins inhibit only those functions stimulated by activins. Whether the ability to

hnmunoregulation by TGF-fl

161

TABLE 1. lmmunoregulatorypropertiesof TFG-flfamilymembers. TGF-fl familymember

Immunoregulatory

TGF-fll TGF-fl2 TGF-fl3 TGF-fl4 TGF-fl5 Activin Vgl Mfillerian inhibitory substance Decapentaplegicgene complexof Drosophilaembryo Bone morphogenicprogein

Yes* Yes* Yes* NTt Yes:~ Yesw NT NT NT NT

*As determined by the ability to inhibit the proliferativeresponse of IL-2 dependent cytotoxic T cell lines in vitro, and inhibit growth and cytokine production of fresh cells. "l'Nottested. ~Unpublished observations. w 25% inhibition of IL-2-dependent cytotoxie T cell proliferationat 1pg/ml.TGF-fll showssimilarinhibitionat approximately100 pg/ml. Also, production ofcytokines from activatedT cells [10]. regulate the immune system at some level is a property of all members of this family remains to be determined. The existence o f so many TGF-fl family members and their evolutionary conservation suggests that there are specific functions for different family members. Indeed, an exclusive role for TGF-fl2 in mesoderm induction has been reported [11]. Numerous cell types in culture express one or multiple forms of TGF-fl, at least at the m R N A level. Most immune effector cells, including CD4 § cells, CD8 § B cells, large granular lymphocytes (LGL), lymphocyte activated killer (LAK) cells, and macrophages produce T G F - f l l . We have also established that lymphoid cells make TGF-fl3 but not TGF-fl2.

TGF-fl Binding Protehzs Use of cross-linking reagents with radiolabeled TGF-fls has shown that binding of TGF-fl is mediated by several coexisting membrane proteins. Two glycoproteins of 53 kDa (receptor I) and 73-100 kDa (receptor II) and a membrane proteoglycan, (now called betaglycan) of 280-350 kDa are present on most cell types [12]. Receptors I and II bind T G F - f l at higher affinities (5-50 pM) than the betaglycan (30-300 pM). In receptor competition assays, the order of decreasing relative affinities is TGF-fll > TGF-fl3, and > TGF-fl2 with a 10- to 20-fold range. Receptor II shows similar, but not identical affinities. Its variation in size is due to differences in glycosylation. The absence of the betaglycan on rat skeletal cells, some hematopoietic cells, and vascular endothelial cells suggest that it is not involved in T G F - f l signaling. Most mammalian cells show similar sensitivity to various forms of TGF-fl, whether the cells coexpress receptors I and II alone or with the betaglycan. The presence of type I or type II receptors alone have rarely been found by themselves on transformed cell lines. The first member of the T G F - f l family for which a receptor has been cloned is activin, which contains a predicted transmembrane serine kinase [13]. Recently, using the

162

F. IV. Ruscetti and ilf.A. Palladino

expression cloning techniques, TGF-fl type II and type III receptor genes have been independently isolated by Weinberl and Massagu6 and their colleagues. Primary hematopoietic and lymphoid cells possess as few as 100-200 receptors/cell which can markedly increase during active proliferation or differentiation [14]. Hematopoietic progenitor and lymphoid cells have predominantly type I receptors while macrophages can have also high levels of the betaglycan. As with many other systems, the ligand-receptor complex is rapidly internalized by the cell and these complexes fuse with lysosomes. It has recently been suggested that the rate of reappearance of the type I and II receptors on the cell surface following interaction with TGF-fl plays a role in the potency with which TGF-fl inhibits cell growth [15]. Extracelhdar Control o f TGF-fl Activity

Since most cell types produce at least one type of TGF-fl and have receptors on their cell surface for TGF-fl, the activity of TGF-fl must be tightly controlled. Unique among cytokines, TGF-fl is secreted from cells, or released from platelets, as an inactive or latent complex which is unable to bind to its receptor. The complex consists of a mature dimer plus two proregion peptides disulfide linked to each other and to a glycoprotein of 125 kDa of unknown function. However, studies with recombinant materials indicate that association of the mature dimer with the proregion is sufficient to keep the molecule inactive. This 'latent' recombinant form also occurs naturally, produced by bone osteoclasts. The mechanism(s) that activates TGF-fl hi vivo is not known, but is probably enzyme mediated. Endothelial cell cultures can activate the latent complex, but only when in contact with vascular pericytes [16]. This interaction appears to be mediated through the mannose-6 phosphate receptor [17] to which the proregion of TGF-fl can bind. In addition, there is evidence that the proteolytic action of plasmin or cathepsin D may be involved in removing the proregion. Once released from the latent complex, active TGF-fl can bind to various extracellular components. Clearance of TGF-fl by binding to cz2-macroglobulin is extremely rapid, ( < 3 min) [18, 19]. Thus, the availability of TGF-fl is brief and probably restricted to the local microenvironment. The role of binding to extracellular matrix materials such as betaglycan as a clearance system, or a protected long-term reservoir for sustained release, is uncertain. TGF-fl as an hmmmoregulator In Vitro

The ability of TGF-fl to elicit many cellular responses, particularly those of an opposing nature has stimulated great interest. In particular, its ability to stimulate and inhibit cell proliferation of closely related cell types has been intensively studied. TGFfl in general acts as a stimulant of connective tissue cells, but has antiproliferative, as well as differentiative, effects on most other cell types. The cell association and tissue distribution of TGF-fll suggests that this factor has a role in hematopoiesis and lymphopoiesis. With monospecific antibodies to TGF-fll and immunoperoxidase staining methods, TGF-fl was found to be associated with hematopoietic and lymphopoietic cell populations in the fetal liver, bone marrow, and thymus. Within the thymus, TGF-fll is specifically localized within the medullary reticuloepithelial cells. In contrast, cortical accessory cells, the thymic capsule, and the thymocytes are not labeled by these antibodies. Thus, it is not surprising that TGF-fl

hnmunoregulation b)' TGF-fl

163

TABLE 2. Immunoregulatory properties of TGF-p: in iitro.

hnmunosuppression Human lymphocyte antigens-DR and Fc receptor expression Cytokine production: IFN-:y, TNF-~, IL-Ifl, ILo6, GM-CSF Thymocyte proliferation T cell/B cell proliferation lgG, IgM production IL-2 receptor p55 expression Cytotoxic T cell generation/function LAK cell activation/function NK cell activation/function Generation of cytotoxic macrophages Macrophage H202 production Neutrophil adhesion to endothelium Hematopoiesis

hnmt#tostittlulalor)' IgA production Cytokine production; TNF-c~, IL-lc~, IL-lfl, TGF-ct, TGF-fl, IL-6, contra IL-! Macrophage and neutrophil chemotaxis Granulopoiesis

has both immunostimulatory and immunosuppressive activities (Table 2). Also, an inhibitory role of TGF-fl in lymphocyte [14] and myeloid cell development is now evident [20, 21]. Recent studies have shown that TGF fll and TGF-fl2 are equally effective at suppressing the proliferative response of murine thymocytes to ILl, IL-2, IL-4, ILo6, IL-7, TNF-c~ and PHA. The inhibitory activity is dose dependent over the concentration range 0.4-100 pM. The half-maximal inhibitory concentrations for both forms of TGF-fl is approximately 5 pM. IL-I induced thymocyte growth is among the most sensitive to inhibition by TGF-fll and can be used as an assay for TGF-fl [14]. The suppressive activity of TGF-fl can be overcome by adding high concentrations of exogenous IL-2 to the thymocyte cultures, suggesting that TGF-fl inhibits thymocyte proliferation by preventing the production of IL-2. The localization of "FGF-fll within medullary reticuloepithelial cells suggests a role for TGF-fl in regulating clonal expansion of the developing T cell. TGF-fll and TGF-fl2 inhibit mitogen-induced human peripheral blood T cell proliferation. The proliferative response is inhibited (75-90%) in a dose-dependent fashion over the concentration range 1-100 pM. Kehrl et al. [22] have shown that TGFfll also inhibits IL-2 mRNA expression by mitogen-activated tonsillar T cells. Further, peripheral T cells are induced to express TGF-fl mRNA shortly (2 h) after mitogen stimulation. Secreted TGF-fl, however, is not detected in the culture supernatant until 2-3 days after mitogen activation, suggesting that the TGF-fl has a late effect and may function in an autocrine/paracrine manner to limit the extent o f T cell clonal expansion [22]. As in the case of thymocyte proliferation, the inhibitory effect on the growth of activated T cells can be overcome by increasing the concentration of IL-2. Recently, this phenomenon has also been seen in the proliferation of natural killer (NK) [23], lymphokine activated killer [24], and cytotoxic T cells [25] suggesting that the number of positive versus negative signals determines the growth response of lymphoid cells. Although the precise functions of IL-2 and TGF-fl produced during lymphoid activation are not defined, the two cytokines regulate in an opposing manner many

164

F. IV. Ruscetti and M . A . Palladino

TABLE3. Effectsof TGF-flon lymphoidcytokinemediator/production.

TNF-~ IFN-), IL-6 IL-8 PFP MIP-Iu TGF-fl Contra-IL-I IL-I~ GM-CSF IL-2Rcr IL-2Rfl

NK Cell

Tcell

,1, 1 1 ND ,t ND T ND ND ND ~. NE

1 1 11 ], ,t 1 T T IT ,l J, NE

ND = not determined;NE = no effect.

important immune functions. Many of the effects stimulated by IL-2 as well as TNF-2 are inverse of those exhibited by TGF-fl (Table 3). The kinetics of IL-2 and TGF-fl production during lymphocyte activation provides some insight into the possible immunoregulatory roles of these molecules. IL-2 can be detected as early as 2-4 h after T cell activation and peak levels are reached by 18-24 h [22]. While the production of TGF-fl m R N A can also be detected early after T cell activation, secreted proteins are not seen until day 2-3 and peak levels are reached at a later time, usually by day 5 when proliferation and cytokine production have significantly decreased [22]. It is therefore tempting to speculate that the TGF-fl produced during normal T cell activation could dampen T cell responses in an autocrine/paracrine fashion and act as an endogenous immunosuppressive agent. It should be noted that the TGF-fl is produced by lymphocytes in a latent or inactive form. The precise mechanism(s) of how lymphocytes/monocytes convert latent TGF-fl to the active form is not known. Further investigation of the effect of TGF-fl on T cell proliferation yielded some unexpected results [26]. IL-I is a costimulus for T cell proliferation, along with T cell receptor cross-linking agents or lectins. Bristol et al. [26] demonstrated that IL-1 o~, in the absence of a T cell receptor stimulus, induces active TGF-fl, production by nonproliferating T cells. Production of active TGF-fl protein by T cells in response to IL- 1~xwas established by: (i) growth inhibition of the TGF-fl-responsive CCL 64 target cell, (ii) neutralization of inhibitory activity in supernatants with anti-TGF-fl monoclonal antibodies (MAb), and (iii) competition ofsupernatants with ~25I-TGF-fl binding to CCL 64 membranes. Thus, although it is an incomplete mitogenic signal, IL-1 ~ by itselfinduced the production of a negative modulator for T cell proliferation. In contrast, when concanavalin A and IL-1 were combined to stimulate T cell growth, active TGF-fl was not detected in supernatants during the first 3 days. Examination of the mechanism of IL- 1c~induction of TGF-fl showed that IL-I c~did not increase TGFfll message over constitutive levels, nor did it induce transcription of TGF-fl2 message. Cycloheximide inhibited IL-l~-induced expression of TGF-fl by T cells. Consequently, IL-1 o~may regulate TGF-fl expression in T cells through translational

hmmmoregulation by TGF-fl

165

and/or post-translational mechanisms. IL- 1 regulation ofactive TGF-fl production by resting T cells may be a mechanism for controlling the growth response o f T cells and why a second or a costimulus is needed for growth. In addition to profound effects on the proliferation of immune cells, TGF-/3 markedly inhibits generation of effector cell function. The generation of cytotoxic T cells, CD3-NK cell activity, and LAK activity are all inhibited. Ochoa and coworkers have found that TGF-/~ is involved at several levels in the generation of LAK activity from human peripheral blood T cells [27]. CD4 + and CD8 § T cells do not develop significant LAK activity when cultured as undepleted peripheral blood lymphocytes with IL-2 or activated with a T cell stimulus such as OKT3 MAb. The possibility that a T cell regulatory mechanism prevented the development of LAK activity by CD4 § and CD8 + cells in these cultures was tested by depleting CD8 § or CD4 + cells from peripheral blood lymphocytes prior to stimulation with OKT3 and IL-2. Under these conditions, the remaining CD4 + and CD8 § cells were able to generate significant nonmajor histocompatibility complex restricted lysis of NK resistant tumor targets. This regulatory signal had to be present throughout the culture to prevent the development of lytic function by T cells. The finding that T cells removed from the peripheral blood lymphocyte cultures are then able to generate LAK activity, suggests that inhibition of the CD4 § or CD8 + mediated LAK activity acts at the level of the activated cell and not the precursor cell. TGF-fll, when added to CD4 § or CD8 § depleted cultures, was able to inhibit LAK activity in a dose-dependent manner. Surprisingly, LAK activity was completely inhibited by 3 ng/ml of TGF-fl in the presence of 300 units of IL-2; conditions where there is no effect on the proliferation of LAK cells [27]. Thus, cytotoxic effector cell function is much more sensitive to TGF-fl inhibition than is cell growth. TGF-/~I may play a role in the generation of LAK activity by T cells, as well as LGL, and consequently in the development (or lack of it) of a therapeutic response in cancer immunotherapy. The mechanism of this TGF-p inhibition of CD8 § LAK has also been studied. Smyth et al. [28] showed that TGF-fl suppresses the ability of CTL to produce poreforming protein (PFP), a major cytolytic protein in the granules of eytotoxic CD8 § T cells. The inhibitory effects of TGF-fl on both CD8 § T cell PFP mRNA expression and LAK activity were reversed by removal of TGF-/3 from the culture. Lymphokines, adhesion]recognition molecules, and activated IL-2 receptor-~ chain expression previously implicated in the lytic mechanism ofcytotoxic lymphocytes were either not expressed in the presence or not regulated by TGF-fl. In addition, independent of effector-target cell binding, lectin or heteroconjugated antibody-dependent cellular cytotoxicity of IL-2/OKT-3 MAb activated CD8 § T cells was inhibited by prior incubation with TGF-fl. TGF-p also inhibited the rapid activation induced expression of PFP mRNA and cytotoxic potential in resting T cells, thereby indicating that the effect of TGF-I3 was independent o f T cell proliferation. TGF-p inhibition ofCD8 + T cell PFP mRNA expression and cytotoxic potential was TGF-j3 dose-dependent. However, a variety of activation stimuli (including IL-2, IL-6, and OKT-3 MAb) were all similarly inhibited by TGF-/3. Therefore, TGF-/3 may be an important general regulator of CD8 + T cell cytotoxic function. As exemplified by the effect of TGF-I3 on PFP production, many of the effects of TGF-fl on the immune system could be indirect, by influencing the production or activity of other immunoregulatory molecules. TGF-j3 increases the production of some cytokines while it inhibits the production ofothers (Table 3). While it is generally

F.W. Ruscetti and M.A. Palladino

166

agreed that TGF-fl can increase the steady state levels of mRNA synthesis for IL-I ~, IL-lfl, TNF-~, as well as itself in monocytes, most reports have demonstrated that TGF-/3 can suppress lipopolysaccharide, but not phorbol 12-myristate-13-acetate (PMA) induced production of TNF-~, IL-l~z, and IL-lfl by monocytes [29, 30]. In addition, TGF-/3 can inhibit phytohemagglutin induced synthesis of interferon (IFN)-?' IL-2, IL-3, granulocyte-macrophage colony stimulating factor (GM-CSF), and TNF-~ [31]. More recently, it was demonstrated that in PHA activated peripheral blood mononuclear cells (PBMC) cultures treated with the TGF-fl specific MAb, IFN-?' and TNF-~ levels are also significantly enhanced [32]. In support ofthe complex interactions between TGF-fl and cytokine production are two recent reports [33, 34] that TGF-fl can enhance the production of IL-6 by PBMC yet inhibit IL-I induced IL-6 production in monocytes. In addition, TGF-fl can enhance the production of the IL-1R antagonist suggesting that the immunosuppressive effects ofTGF-fl hz vitro are in part due to the enhanced production of this protein [35]. TGF-fl ht htflammation

Monocytes/macrophages play a central role in many cellular reactions, particularly in inflammatory responses and wound repair processes. The positive and negative regulation is illustrated by their interaction with TNF-~ and TGF-fl which display both inhibitory and stimulatory effects (Table 4). In regard to macrophage functions, TGF-fl is basically an anti-inflammatory molecule (36). TGF-fl can deactivate macrophages by reducing their capacity to release H202, their cytotoxic activity, their expression ofclass II and Fc~R2 (CD23) expression as well as their production of TNFceand IL- 1, thus perhaps imposing negative feedback signals on the stimulatory signals provided by TNF-~ and other cytokines [37]. TABLE 4. Comparative effects of cytokines on maerophage activities.

TNF-c~ TGF-p tL-l~, IL-f: IL-6 TNF-~ TGF-p Cytotoxicactivity H20zproduction Chemotaxis Class II major histocompatibilitycomplex CD23 expression

t ]" T T t T T T T

.IT IT ,~ T ,L 1 1

Macrophages, in the capacity of antigen presenting cells, play a central role in activating T cells to respond to antigen. Critical to the interaction of antigen with T cells is the expression on macrophages of class II major histocompatibility complex antigens, which can be up-regulated by IFN-7 and TNF-~. TGF-fl inhibits IFN-y induced class II expression on both human peripheral blood derived macrophages and on HS294T melanoma cells [38]. TGF-fl and cyclosporin A share the ability to counteract both in vivo and hz vitro the augmenting effects of IL-1, TNF-cq IFN-y,

hnmunoregulation by TGF-fl

167

IL-3, and GM-CSF but not IL-2 or IL-6, on the expression of ia antigens on epidermal Langerhans cells. Thus, TGF-fl plays important roles in coordinating the immune response to antigen and pathogens through down-regulating class II expression. An important consideration is whether there are autocrine as well as paracrine effects of TGF-fl on the immune system. Using recombinant DNA technology, Wallick et al. [39] developed Chinese hamster cell lines that synthesize latent TGF-fl. h2 vitro, the latent material secreted by these cells inhibited cytotoxic T lymphocyte generation as well as exogenously added active TGF-fl. hz vivo, this material suppressed splenic NK activity as well as active TGF-fl. However, only latent TGF-fl was found in the serum of such mice indicating that the activation of TGF-fl hi vivo was a local event. These results support the work of Lucas et al. [40] who showed that MAb to active TGF-fl could enhance IL-2 proliferation of human T cells and generation of LAK cells and Kehrl et al. [41] who observed that similar antibodies could increase B cell proliferation and immunoglobulin synthesis. It appears that endogenous latent material can be activated by normal physiological mechanisms, and then can down-regulate the immune response in an autocrine manner. The processes by which latent TGF-fl becomes active during the immune response are not known. Previous work had suggested that macrophages have the ability to secrete latent and active TGF-fl [42]. The fact that macrophages as well as osteoclasts have acidic microenvironments, and possess sialidase activity further implicate the macrophage as an activator of latent TGF-fl. Adherent human macrophages stimulated by IFN-~, [43], but not by iipopolysaccharide [39], were able to activate latent recombinant TGF-fl. It may be that efficient activation requires cooperating cell types such as the smooth muscle and pericytes to activate latent TGFfl [16]. I N VIVO EFFECTS OF TGF-fl Endotoxht hlduced TNF-ce Production

Despite aggressive antibiotic therapy, systemic bacterial infections can progress rapidly to uncontrollable sepsis leading to irreversible cardiovascular collapse and multiple organ failures. TNF-cr cachectin, a hormone produced by macrophages in response to bacterial lipopolysaccharide, has been shown to be a key mediator ofseptic shock, while antibodies to TNF-~, in contrast, are protective [44]. It has been shown that TGF-fl can oppose TNF-~ effects hi vitro [45]. Therefore, the ability of TGF-fl to modulate LPS induced TNF-~ production Fischer rats were treated intraperitoneally with 10/2g of either rHuTGF-fll, rHuTGF-fl2 or diluent control and 72 h later, pathogenic E. coli (1-3 x 109) were administered intraperitoneaily. Mortality at 72 h after infection in untreated controls was approximately 80% (19 of 24) compared to approximately 50% (16 of 35) in the rHuTGF-fl-treated animals. In other experiments, circulating levels of TNF-ar were significantly suppressed in rats given rHuTGF-fll/rHuTGF-fl2 two hours before E. coli compared to untreated controls [46]. These data demonstrate that h~ vivo, rHuTGF-fll rHuTGF-fl2 pretreatment can reduce cytokine production after lethal E. colichallenge, as well as reduce the mortality observed with septic shock and suggest that TGF-fl may offer a potential new form of therapy for life-threatening infections, although the shock mechanism is unknown.

168

F, IV. Ruscetti and M . A . Palladino

Allograft Rejection

As the ability to induce organ-specific tolerance in many animal transplantation models has unfortunately not been seen clinically, the transplantation of human organs is often palliative, not curative therapy. Because ofthis requirement for chronic immunosuppressive therapy, there is an urgent need for more effective, more specific, and less toxic immunosuppressive agents. The hi vitro data on the immunosuppressive activities of TGF-fl warranted testing as an immunosuppressive agent to delay or prevent organ rejection. Therefore, in our second series ofhz vivo studies, we investigated the ability of TGFfl to suppress various immune functions and to prolong the survival of neonatal BALB/c (H-2 k) mouse hearts transplanted into the subcutaneous skin of the ears of adult C3H/KM mice (H-2d). This model has been shown to be useful in characterizing the immunosuppressive activities of newly identified molecules [47]. Normal C3H/KM mice were injected intraperitoneally with various doses of rHuTGF-fll. Spleen cells removed at various times after the last rHuTGF-fll injection showed significantly suppressed NIL activity against YAC-I target cells and reduced ability to respond to Concanavalin A [41]. In the second series of studies, groups of C3H/KM recipients of BALB/c heart allografts were treated intraperitoneally with I/tg rHuTGF-fll on days 1-13. Heart graft survival evaluated on days 10-14 after transplant was significantly prolonged compared to the saline-treated control group. Similar organ graft survival on day 14 after daily treatment with cyclosporin A required a dose of 6 mg/kg. In addition, rHuTGF-fll suppressed the T cell-mediated lymph node enlargement occurring in a host versus graft-popliteal lymph node assay [39]. The results demonstrate that rHuTGF-fll is a potent immunosuppressant h~ vivo and suggest the possible clinical usefulness ofthis agent in the inhibition oforgan rejection. A summary of the hi vivo effects of TGF-fl is presented in Table 5. TABLE 5. lmmunoregulatory properties of TGF-fl: in wiro. Inhibition

Cytokine production; TNF-~, CSF NK activity T cell responses; cytotoxic T cell generation, Concanavalin A responsiveness E. coli-induced septic shock Allograft rejection: tteart, skin Hematopoiesis Sthnulation

Granulopoiesis Cytokine production: Contra-IL-I

Mechanism of Action of TGF-fl in hnmune Effector Cells

The mechanisms by which TGF-fl acts as a potent inhibitor of the growth and functions of lymphoid and hematopoietic progenitor cells are not known. Cell proliferation depends not only on the presence of growth factors, but also on the development of specific receptor-signal transducing complexes. We, therefore,

hnnmnoregulation by TGF-fl

169

investigated whether the inhibitory actions of TGF-fl could be mediated by inhibition of growth factor receptors. TGF-fl inhibited the constitutive level of IL-1R expressed on several murine lymphoid and myeloid progenitor cell lines, as well as IL-IR expression induced by IL-3 on normal murine and human bone marrow cells [48]. Furthermore, treatment of bone marrow progenitor cells with TGF-fl concomitantly inhibited the ability of IL-I to promote HPP colony formation and also blocked IL-1 induced IL-2 production by EL 4 6.1 cells. These findings provide the first evidence that the inhibitory action of TGF-fl on the growth and functional activities of hematopoietic and T cells is associated with a reduction in the cell surface receptor expression for IL-I. Since TGF-fl can induce the production ofcontra-IL-1 [35] which should suppress IL-I functions, there are probably several pathways which contribute to TGF-fl inhibition of IL-I function. However, studies by Ortaldo et al. [49] on the mechanism on TGF-fl inhibition of CD3-NK proliferation and function showed that receptor down-modulation is not universal for all immune cells' receptors. CD3 LGL express constitutive levels of functional IL-2Rfl. TGF-fl inhibited several IL-2Rfl-mediated events in LGL, including IL-2-induced NK and LAK activities, IFN-~' gene expression and secretion, and IL-2R~ expression. TGF-fl inhibited these functions in a dose-dependent and reversible manner. In contrast, TGF-fl had little effect on LGL I L-2Rfl expression, nor were TGF-flR induced by IL-2. Studies were performed to examine binding and internalization of radiolabeled IL-2. These experiments demonstrated that the rapid binding and internalization of ~25I-IL-2 was not altered in CD3- LGL pretreated with TGF-fl. These internalization studies indicated that the TGF-fl inhibition represented postreceptor binding events in NK cells. Further studies were initiated to examine signaling events in CD3- LGL. When IL-2-induced tyrosine phosphorylation events were examined, significant inhibition was seen in TGF-fl pretreated cells. Specifically, a p97 tyrosine kinase associated with IL-2Rfl had decreased phosphorylation in the presence of TGF-fl. The ability of TGF-fl to also inhibit IL-2 induction of LGL IL2R~z and IFN-7 mRNA expression was consistent with the hypothesis that posttranscriptional mechanisms were unlikely to be effected by TGF-fl. Collectively, these data indicated that TGF-fl inhibited IL-2-induced CD3- LGL functions and suggested that TGF-fl inhibition in this system, occurs either at the level of specific tyrosine phosphorylation and/or IL-2-indueed transcriptional control factors. In examining the effects of TGF-fl on lymphoid cell cycle progression, it is clear that in both mitogen activation [50] and IL-2 mediated growth [51], TGF-fl prevents entry into S phase. It induces a specific block late in Glb which allows for the accumulation of considerable mRNA throughout Gla [50]. Studies on the IL-2 and IL-4 induced growth ofmurine HT-2 and CT-6 T cell lines suggests a rapid (1-2 h) transcriptional inhibition ofc-myc RNA plays a role in the antiproliferative effect of TGF-fl [51]. This observation supports similar work by Moses and coworkers showing that regulation of c-myc through control of the RB gene was central in TGF-fl inhibition of keratinocyte growth [52].

Role of TGF-fl in Pathogenesis of Diseases Related to Autoinununity Recently, there have been reports describing the protective effects of TGF-fl in animal models of autoimmunity. Experimental autoimmune encephalomyelitis is a chronic autoimmune disease characterized by inflammation and demyelination in the

170

F. IV. Ruscetti and M.A. Palladino

central nervous system. Studies by Kuruvilla et al. and Racke et al. have demonstrated respectively that TGF-fl can significantly reduce the induction and severity of experimental autoimmune encephalomyelitis in mice induced either by immunization with myelin basic protein in complete Freunds adjuvant or by the transfer of myelin basic protein specific T cell lines [53, 54]. Recent studies have also demonstrated that systemically administered TGF-fl can reduce the incidence ofcollagen type II induced arthritis in mice and streptococcal induced arthritis in rats [55]. Rheumatoid synovial fluids contain significant amounts of TGF-fl [56]. However, whether the TGF-fl produced constitutively in the joint contributes to joint destruction and synovial fibroblast hyperplasia, or functions to suppress further joint destruction by the immune system needs further study. In contrast to the protective effects of TGF-fl, TNF-cz, IL-1, and IFN-~' have been shown to enhance the severity ofcollagen type II induced arthritis [55]. It is tempting to speculate that one mechanism for the development of autoimmune diseases is the aberrant production and activation of locally released TGF-fl. Whether TGF-fl can be utilized clinically to treat the human equivalent of these diseases awaits further studies. Development o f Lymphomas

Since escape from negative regulators such as TGF-fl could play a role in the growth of neoplastic cells, we examined the effects of TGF-fl on lymphoid leukemic cells. TGF-fl exerts profound inhibitory effects on a number of cell types, including normal B and T lymphocytes. In contrast, we have found a number of lymphoid tumor cell lines to be insensitive to the antiproliferative effects of TGF-fll and TGF-fl2 [57]. Binding and cross-linking with radio-iodinated TGF-pl demonstrated either low or absent expression ofall three TGF-fl receptor species on three B cell tumor lines, but T cell and non-T, non-B tumors expressed large numbers of receptors. Treatment of the B cell lines with PMA induced the expression of TGF-fl receptors and inhibited proliferation in all three lines in a dose- and time-dependent manner. The cell lines constitutively produced TGF-fl mRNA and released small amounts oflatent TGF-fl; however, PMA induced increased expression of TGF-p mRNA and, more importantly, release of active TGF-fl. A neutralizing antibody to TGF-fl was able to reverse the PMAinduced growth inhibition, and addition of exogenous TGF-fl reversed the effects of the neutralizing antibody [59]. Thus, TGF-fl can inhibit human lymphoma cell growth ht vitro through an autocrine mechanism. Some lymphoma cells appear to have escaped from TGF-fl negative regulation by failing to express functional TGF-fl receptors and/or failing to secrete active TGF-fl. One mechanism by which PMA acts to inhibit lymphoma cell growth is by inducing the expression of TGF-fl receptors and the secretion of active TGF-fl thereby re-establishing an autocrine growth inhibitory loop. However, recent studies have shown that Epstein-Barr virus-transformed B cells and Burkitt's lymphoma cells can be either TGF-fl sensitive or insensitive without regard for the presence or absence of TGF-receptors on the cell surface [58]. Also, all HTLV-I transformed T cells tested could bind TGF-fl but could not be inhibited by TGF-fl suggesting that there are several ways to escape regulation by TGF-fl. hnmunosuppression Durhlg Tumorigenesis

In addition to the ability of tumor cells to escape negative growth regulation by

171

bnmunoregulation by TGF-fl

TGF-fl, there is mounting evidence that tumor induced immunosupporession is mediated through TGF-fl and that cancer immunotherapy can be adversely affected by TGF-fl. Clinically, patients with glioblastomas which secrete active TGF-fl2 have a generalized immunosuppression [59]. Several studies have previously shown that immunogenic tumors made to secrete TGF-fl escape immune surveillance, and grow progressively in the host [60]. It is clear that in tumor bearing animals, there is marked immunosuppression. Tada et al. [61] have shown that tumors can make active TGF-fl detectable in a local environment, and the suppression o f C D 4 + helper activity can be completely overcome by the addition of neutralizing antibodies to TGF-fl. This provides a mechanism for the previously observed selective inhibition ofCD4 § T cells in tumor bearing animals. Despite the ability to have LAK cells infiltrating the tumors of many patients [61], these cells do not display much cytotoxic activity to the tumor cells which they can kill hz vitro. The discovery that cytotoxic activity of LAK cells is much more sensitive to TGF-fl inhibition than LAK cell proliferation [27] gives credence to the idea that cancer immunotherapy might be improved by overcoming the action of TGF-fl. Retroviral-htduced bnmunosuppression

HTLV-I is the etiologic agent for acute T cell leukemia and tropical spastic paraparesis. Cellular and humoral immune responses, including helper and suppressor T cell function, NK cell function and B cell Ig Synthesis, are markedly depressed in acute T cell leukemia patients. Kim and coworkers [62] have shown that HTLV-I tax gene product will stimulate the TGF-fll promoter through AP-1 binding sites. In addition, fresh leukemic cells from acute T cell leukemia patients constitutively release high levels of TGF-fll into the culture media, suggesting that TGF-fl might be responsible for some of the immunosuppression seen in these patients. In AIDS, HIV infection results in profound defects in cellular and humoral immunity. As in acute T cell leukemia patients, HIV-infected AIDS patients showed increased titers ofTGF-fl RNA and protein in PBMC cultures [63]. These patients had impaired CD4 § T cell responses to PPD. These responses could be partially overcome with a neutralizing antibody to TGF-fl. These results suggested that overproduction of TGF-fl in AIDS patients could contribute to the immunosuppression. As in tumor bearing animals [61], a selective loss o f T helper cell activity mediated by TGF-fl has recently been shown in cultures of human PBMC from normal [64] and HIV-infected individuals [63]. Overproduction of TGF-fl by HIV-infected monocytes, which also secrete products which stimulate astrocytes to produce TGF-fl, has implicated TGF-fl in the pathogenic processes responsible for AIDS-related central nervous system dysfunction [65]. It has been speculated that inhibitors of TGF-fl function might be useful in the treatment of AIDS, particularly to improve helper T cell function. Due to the fact that TGF-fl may either suppress or stimulate HIV production, as well as prevent the down-regulation by TGF-fl of HIV-stimulating effects of TNF-c~, IL-6, and GM-CSF [66], this therapy would seem to be unpredictable and scientifically unwarranted. CONCLUSION In summary, the regulation of immune responses is a complex interplay between the

172

F. IV. Ruscetti and M . A . Palladino

p r o d u c t i o n a n d action of i m m u n o s t i m u l a t o r y a n d i m m u n o s u p p r e s s i v e agents. We have a t t e m p t e d to illustrate how o n e such family o f factors, the T G F - f l s , is a n i m p o r t a n t m e d i a t o r o f i m m u n e reactions. T h r o u g h a d d i t i o n a l e x p e r i m e n t a t i o n in vivo a n d hi vitro, a better u n d e r s t a n d i n g will be gained to guide in the use o f T G F - f l a n d a n t a g o n i s t s to T G F - f l in i m m u n o l o g i c a l l y based therapies. At this m o m e n t , it is fair to say that this cytokine has exciting potential for the t r e a t m e n t or p r e v e n t i o n o f some diseases. REFERENCES i. Sporn MB, Roberts AB, Wakefield LM, de Crombrugghe B. Some recent advances in the chemistry and biology of transforming growth factor-beta. J Cell Biol. 1987; 195: 1039-1045. 2. Massagu6J. The transforming growth factor-il family. Annu Rev Cell Biol. 1990;597-641. 3. Derynck R, Jarrett JA, Chen EY, Eaton DH, BellJR, Assoian RK, Roberts AB, Sporn MB, Goeddel DV. Human transforming growth factor-il complementary DNA sequenceand expression in normal and transformed cells. Nature 1985; 316: 701-705. 4. SeyedinSM, Thomas TC, Thompson AY, Rosen DM, Piez KA. Purification and characterization of two cartilage-inducing factors from bovine demineralized bone. Proc Natl Acad Sci USA. 1985; 82: 2267-227 !. 5. Madisen L, Webb NR, Rose TM, Marquardt H, Ikeda T, Twardzik D, Seyedin S, Purchio AF. Transforming growth factor-il2: cDNA cloning and sequence analysis. DNA 1988; 7: !-8. 6. De Martin R, Haendler B, Hofer-Warbinek R, Gaugitsch H, Wrann M, SchlfisenerH, Seifert JM, Bodmer S, Fontana A, Hofer E. Complementary DNA for human glioblastoma-derived T cell suppressor factor, a novel member of the transforming growth factor-fl gene family. EMBO J. 1987;6: 3673-3677. 7. Derynck R, Lindquist PB, Lee A, Wen D, Tamm J, Graycar JL, Rhee L, Mason AJ, Miller DA, Coffey RJ, Moses HL, Chen EY. A new type of transforming growth factor-il, TGF-Il3. EMBO J. 1988;7: 3737-3743. 8. Jakowlew SB, Dillard PJ, Kondaiah P, Sporn MB, Roberts AB. Complementary deoxyribonucleic acid cloning of a novel transforming growth factor-il messenger ribonucleic acid from chick embryo chondrocytes. Mol Endocrinol. 1988; 2: 747-755. 9. Kondaiah P, Sands MJ, Smith, et aL Identification of a novel transforming growth factor-fl (TGF-fl5) mRNA in Xenopus laevis. J Biol Chem. 1990;265: 1089-1093. 10. BroxmeyerH, Lu L, Cooper S, Schwall R, Rason A, Nikolics K. Selectiveand indirect modulation of human multipotential and erythroid hematopoietic progenitor cell proliferation by recombinant human activin and inhibition. Proc Natl Acad Sci USA. 1988;85: 9052-9056. Ii. Rosa F, Roberts AB, Danielpour D, Dart LL, Sporn MB, Dawid I. Mesoderm induction in amphibians: the role of TGF-fl2-1ikefactors. Science 1988; 239: 783-785. 12. Cheifetz S, Weatherbee JA, Tsang MLS, Anderson JK, Mole JE, Lucas R, Massagu~ J. The transforming growth factor-il system, a complex pattern of cross reactive ligands and receptors. Cell 1987; 48: 409-415. 13. MatthewsLS, Vale WW. Expressioncloning of an activin receptor, a predicted transmembrane serine kinase. Cell 1991;65: 973-982. 14. Ellingsworth LR, Nakayama D, Segarini P, Dasch J, Carrillo P, Waegell W. Transforming growth factor-ils are equipotent growth inhibitors of interleukin-l-induced thymocyte proliferation. Cell Immunol. 1988; 114: 41-54. 15. Birchenall-Roberts MC, Falk LA, Kasper J, Keller J, Faltynek CR, Rusceni FW. Differential expression of transforming growth factor-ill (TGF-fll) receptors in murine myeloid cell lines transformed with oncogenes. J Biol Chem. 1991; 266: 9617-9621. 16. Antonelli-Orlidge A, Saunders KB, Smith SR, D'Amore PA. An activated form of transforming growth factor il is produced by cocultures of endothelial cells and pericytes. Proc Natl Acad Sci USA. 1989; 86: 454~4548. 17. Dennis PA, Rifkin DB. Cellular activation of latent transforming growth factor ,8 requires binding to the cation-independent mannose 6-phosphate/insulin-like growth factor type II receptor. Proc Natl Acad Sci USA. 1991; 88: 580-584.

bnmunoregulation by TGF-fl

173

18. O'Connor-McCourt MD, Wakefield LM. Latent transforming growth factor-ft in serum. A specific complex with a2-macroglobulin. J Biol Chem. 1987; 262: 14090-14099. 19. LaMarre J, Hayes MA, Wollenberg GK, Hussaini I, Hall SW, Gonias SL. An q2-macroglobulin receptor-dependent mechanism for the plasma clearance of transforming growth factor-ill in mice. J Clin Invest. 1991; 87: 39-44. 20. Kelter JR, Mantel C, Sing GK, EIIingsworth LR, Ruscetti SK, Ruscetti FW. Transforming growth factor fll selectively regulates early murine hematopoietic progenitors and inhibits the growth of IL-3 dependent myeloid leukemia cell lines. J Exp Med. 1988; 168: 737-750. 21. Sing GK, Keller JR, Ellingsworth LR, Ruscetti FW. Transforming growth factor ft selectively inhibits normal and leukemic human bone marrow cell growth in vitro. Blood 1988; 72:1504--1511. 22. Kehrl JH, Wakefield LM, Roberts AB, Jakowlew S, Alvarez-Mon M, Derynck R, Sporn MB, Fauci AS. Production of transforming growth factor fl by human T lymphocytes and its potential role in the regulation ofT ceil growth. J Exp Med. 1986; 136: 1037-1050. 23. Rook A, Kehrl H, Wakefield LM, Roberts AB, Sporn MB, Burlington DB, Lane HC, Fauci AS. Effects of transforming growth factor ft on the functions of natural killer cells: depressed cytolytic activity and blunting of interferon responsiveness. J Immunol. 1986; 136: 3916-3920. 24. Grimm EA, Crump WL, Durett A, Hestler JP, Lagoo-Deenadalayan S. TGF-beta inhibits the in vitro induction of lymphokine-activated killing activity. Cancer lmmunol Immunother. 1988; 27: 53-58. 25. Ranges GE, Figari IS, Espevik T, Palladino MA. Inhibition of cytotoxic T cell development by transforming growth factor ,6 and reversal by recombinant tumor necrosis factor ct. J Exp Med. 1987; 166: 991-998. 26. Bristol LA, Ruscetti FW, Brody DT, Durum SK. IL-lct induces expression of active transforming growth factor-fl in nonproliferating T cells via a post-transcriptional mechanism. J Immunol. 1990; 145:4108-4114. 27. Geller RL, Smyth M J, Strobl SL, Bach FH, Ruscetti FW, Longo DL, Ochoa AC. Generation of lymphokine-activated killer activity in T cells. J Immunol. 1991; 146: 3280-3288. 28. Smyth M J, Strohl SL, Young HA, Ortaldo JR, Ochoa AC. Regulation oflymphokine-activated killer activity and pore-forming protein gene expression in human peripheral blood CD8 ~ T lymphocytes. J Immunol. 1991; 146: 3289-3297. 29. Chantry D, Turner M, Abney E, Feldmann M. Modulation ofcytokine production by transforming growth factor-f. J Immunol. 1989; 142: 4295--4300. 30. Espevik T, Figari IS, Shalaby MR, Lackides GA, Lewis GD, Shepard HM, Palladino MA. Inhibition of cytokine production by cyclosporin A and transforming growth factor,8. J Exp Med. 1987; 166:571-576. 31. Nedwin GE, Svedersky LP, Bringman TS, Palladino MA, Goeddel DV. Effect of interleukin 2, interferon-?',and mitogens on the production of tumor necrosis factors a: and ft. J lmmunol. 1985; 135: 2492-2497. 32. Lucas C, Figari IS, Reynolds C, Palladino MA. Transforming growth factor-beta I endogenously produced by lymphocytes is a potent regulator of tumor necrosis factor-alpha production. Seventh International Conference on AIDS, 1991. 33. Turner M, Chantry D, Feldmann M. Transforming growth factor-fl induces the production of interleukin 6 by human peripheral blood mononuclear cells. Cytokine 1990; 2:211-216. 34. Musso T, Espinoza-Delgado I, Pulkki K, Gusella GL, Longo DL, Varesio L. Transforming growth factor ft downregulates interleukin (IL-1)-induced IL-6 production by human monocytes. Blood 1990; 76: 2466-2469. 35. Turner M, Chantry D, Katsikis P, Berger A, Brennan FM, Feldmann M. Induction ofthe interleukin I receptor antagonist protein by transforming growth factor-ft. Eur J Immunol. 1991; 21: 1635-1639. 36. Wahl SM, Hung DA, Wakefield LM, McCartney-Francis N, Wahl LM, Roberts AB, Sporn MB. Transforming growth factor type ft induces monocyte chemotaxis and growth factor production. Proc Natl Acad Sci USA. 1987, 84: 5788-5792. 37. Tsunawaki S, Sporn M, Ding A, Nathan C. Deactivation of macrophages by transforming growth factor-ft. Nature 1988; 334: 260-262. 38. Czarniecki CW, Chiu HH, Wong GHW, McCabe SM, Palladino MA. Transforming growth factor-ft~ modulates the expression of class II histocompatibility antigens on human cells. J Immunol. i 988; 140: 4217-4223. 39. Wallick SC, Figari IS, Morris RE, Levinson AD, Palladino MA. Immunoregulatory role of transforming growth factor ,8 (TGF-ft) in development of killer cells: comparison of active and latent TGF-flt. J Exp Med. 1990; 172: 1777-1784.

174

40.

41. 42.

43.

44. 45. 46.

47.

48.

49.

50.

51.

52. 53.

54.

55. 56. 57.

58.

59.

60.

F. IV. Ruscetti and M . A . Palladino

Lucas C, Mora-Worms M, Fendly BM, Bald LN, Figari IS, Palladino MA. A monoclonal antibodybased enzyme-linked immunosorbent assay to study the autocrine production of transforming growth factor Bl during lymphocyte activation. J lmmunol. 1990; 145: 1415-1422. Kehrl JH, Taylor AS, Delsing A, Roberts AB, Sporn MA, Fauci AS. Further studies of the role of transforming growth factor-B in human B cell function. J lmmunol. 1989; 143: 1868-1874. Assoian RK, Fleurdelys BE, Stevenson HC, Miller PJ, Madtes DK, Raines EW, Ross R, Sporn MB. Expression and secretion of type fl transforming growth factor by activated human macrophages. Biochemistry 1987; 84: 6020-6024. Twardzik E, Mikovits J, Ranchalis J, Purchio A, Ellingsworth L, Ruscetti F. y interferon induced activation of latent transforming growth factor fl by human monocytes. In: Piez K, Sporn M, eds. Transforming growth factor Bs: chemistry, biology and therapeutics. Ann N Y Acad Sci. 1990; 593: 237-248. Beutler B, Krochin N, Milsark lW, Leudke C, Cerami A. Contl:ol ofcachectin (tumor necrosis factor) synthesis: mechanisms of endotoxin resistance. Science 1986; 232: 977-980. Palladino M, Morris RE, Starnes HF, Levinson AD. The transforming growth factor-betas; a new family ofimmunoregulatory molecules. Ann N Y Acad Sci. 1990; 593: 181-187. Starnes HF, Pearce MK, Tewari A, Yim JH, Zou J-C, Abrams .IS. Anti-lL-6 monoclonal antibodies protect against lethal Escherichia coli infection and lethal tumor necrosis factor--a challenge in mice. J Immunol. 1990; 145: 4185-4191. Babany G, Morris RE, Babany I, Kates RE. Evaluation of the hi vivo dose-response relationship of immunosuppressive drugs using a mouse heart transplant model: application of cyclosporin. J Pharmacol Exp Ther. 1988; 244: 259-264. Dubois C, Ruscetti F, Palaszynski E, Falk L, Oppenheim J, Keller J. Down modulation of the interleukin-I receptor by transforming growth factor B: produced mechanism of inhibition of IL-I action. J Exp Med. 1990; 172: 737-744. Ortaldo JR, Mason AT, O'Shea J J, Smyth M J, Falk LA, Kennedy ICS, Longo DL, Ruscctti FW. Mechanistic studies of transforming growth factor- B inhibition of I L-2-dependent activation of CD3large granular lymphocyte functions. J Immunol. 1991; 146: 3791-3798. Smeland EB, Blomhoff HK, Holte It, Ruud E, Beiske K, Funderud S, Godal T, Ohlsson R. Transforming growth factor type B (TGF-B) inhibits G, to S transition, but not activation of human B lymphocytes. Exp Cell Res. 1987; 171: 213-222. Ruegemer J J, Ho SN, Augustine JA, Schlager JW, Bell MP, McKean DJ, Abraham RT. Regulatory effects of transforming growth factor- B on 1L-2- and IL-4-dependent T cell-cycle progression. J Immunol. 1990; 144: 1767-1776. Moses HL, Yang EY, Pietenpol JA. TGF- B stimulation and inhibition of cell proliferation: new mechanistic insights. Cell 1990; 63: 245-247. Kuruvilla AP, Shah R, Hochwald GM, Liggitt HD, Palladino MA, Thorbecke GJ. Protective effect of transforming growth factor pl on experimental autoimmune diseases in mice. Proc Natl Acad Sci USA. 1991; 88: 2918-2921. Raeke MK, Dhib-Jalbut S, Cannella B, Albert PS, Raine CS, McFarlin DE. Prevention and treatment of chronic relapsing experimental allergic encephalomyelitis by transforming growth factor-ill. J Immunol. 1991; 146: 3012-3017. Shah R, Kuruvilla AP, Palladino MA Jr, Liggitt HD, Hardison AM, Thorbecke GJ. Transforming growth factor ,6'1 protects against collagen type 11 induced arthritis. Arthritis Rheum. 1991: in press. Brandes ME, Allen JB, Ogawa Y, Wahl SM. Transforming growth factor fll suppresses acute and chronic arthritis in experimental animals. J Clin Invest. 1991; 87:1109-1113. Sing GK, Ruscetti FW, Beckwith M, Keller JR, Ellingsworth L, Urba WJ, Longo DL. Growth inhibition of a human lymphoma cell line: induction of a transforming growth factor-B-medlated autocrine negative loop by phorbol myristate acetate. Cell Growth Differentiation 1990; 1: 549-557. Altiok A, Bejarano MT, Ruscetti F, Altiok E, Klein G, Klein E. Effect of transforming growth factorbeta I and -beta 2 on the proliferation of Burkitt lymphoma and lymphoblastoid cell lines. Growth Factors 1991; 4:117-128. Miescher S, Whiteside TL, Carrel S, yon Fliedner V. Functional properties of tumor infiltrating and blood lymphocytes in patients with solid tumors: effects of tumor cells and their supernatants on proliferative responses oflymphocytes. J lmmunol. 1986; 136: 1899. Torre-Amione G, Beauchamp RD, Koeppen H, Park BH, Schreiber H, Moses HL, Rowley DA. A highly immunogenic tumor transfected with a murine transforming growth factor type B~ eDNA

hnmunoregulation by TGF-fl

175

escapes immune surveillance. Proc Nail Acad Sci USA. 1990; 87: 1486. Tada T, Ohzeki S, Utsumi K, Takiuchi It, Muramatsu M, Li X-F, Shimizu J, Fujiwara H, Hamaoka T. Transforming growth factor-fl-induced inhibition of Tcell function. J Immunol. 1991; 146:1077-1082. 62. Kim S-J, Kehrl JH, Burton J, Tendler CL, Jeang K-T, Danielpour D, Thevenin C, Kim DY, Sporn MB, Roberts AB. Transactivation of the TGF-fll gene by HTLV-I tax: a potential mechanism for the increased production of TGF-fll in adult T-cell leukemia. J Virol. 1991: in press. 63. Kekow J, Wachsman W, McCutchan JA, Cronin M, Carson DA, Lotz M. Transforming growth factor fl and noncytopathic mechanisms of immunodeficiency in human immunodeficiency virus infection. Proc Natl Acad Sci USA. 1990; 87: 8321-8325. 64. Shearer GM, Clerici M. Early T-helper cell defects in HIV infection. AIDS 1991; 5: 245-253. 65. Wahl SM, Alien JB, McCartney-Francis N, Morganti-Kossmann MC, Kossmann T, Ellingsworth L, Mai UEH, Mergenhagen SE, Orenstein JM. Macrophage- and astrocyte-derived transforming growth factor 13as a mediator of central nervous system dysfunction in acquired immune deficiency syndrome. J Exp Med. 1991; 173: 981-991. 66. Roy S, Wainberg MA. Role of the mononuclear phagocyte system in the development of acquired immunodeficiency syndrome. J Leukoc Biol. 1988; 43: 91. 61.