The importance of shedding of membrane proteins for cytokine biology.

ARTICLE

INTRODUCTION

Many cellular membrane proteins are subjected to limited proteolysis giving rise to soluble forms consisting of the entire extracellular domains of the proteins. This process is referred to as shedding. The field has been reviewed in ref. [1-3]. Importantly, soluble forms of receptors for growth factors and cytokines retain their ability to bind their ligands. Moreover, many cytokines are synthesized as transmembrane precursor proteins which have to be cleaved in order to reach into the circulation [4]. Therefore it seems that shedding may influence the biology of the cognate growth factors and cytokines [2, 5].

Shedding of membrane proteins is strongly induced by phorbol esters which are known to stimulate protein kinase C [2-4, 6, 7]. In addition, it has been shown that shedding of several membrane proteins can be stimulated by bacterial toxins [8]. More recently, it was demonstrated that shedding of the IL-6R is induced by the acute phase protein C reactive protein [9]. The identity of the responsible proteases has been addressed using hydroxamate based substances, which are known to be potent inhibitors of metalloproteases [10-12]. As an example, Figure 1 shows that the shedding of the membrane proteins TNF-alpha, SCF, and the type II IL-1R is induced by PMA and inhibited by the hydroxamate compound TAPI [13]. Subsequently, it became evident that virtually all shedding events could be blocked using similar concentrations of hydroxamates. Despite the fact that the cleavage sites of the membrane proteins show no sequence similarities, the common features of the shedding processes indicated that the proteases involved were similar if not identical (Table 1) [2, 3].

As discussed in detail below, soluble receptors for growth factors and cytokines due to their ability to bind their ligands may compete with their membrane bound counterparts for ligand binding. Soluble cytokine receptors have been shown to act as carrier proteins contributing to plasma stability of growth factors and cytokines. Moreover, the availability of some cytokines in the circulation depends on limited proteolysis from their transmembrane precursors. In this article we will review the available data on shedding proteases and address the physiological consequences of shedding processes for cytokine biology.

CLONING OF TACE

Tumor necrosis factor alpha (TNF-alpha) is a cytokine that contributes to a variety of inflammatory disease states, such as sepsis and arthritis. The protein exists as a membrane-bound precursor of an apparent molecular mass of about 26 kDa which can be processed by a TNF-alpha converting enzyme (TACE), to generate secreted 17 kDa mature TNF-alpha [14]. Soluble TNF-alpha is released by cleavage of the precursor at the Ala76-Val77 bond. The release of TNF-alpha from cells in vitro and in vivo can be specifically blocked by hydroxamate based metalloprotease inhibitors [10-12]. In consequence, mice were protected against an otherwise lethal dose of endotoxin by hydroxamate pre-treatment [10].

Cell lysates of lipopolysaccharide (LPS) stimulated human monocytic THP-1 cells were found to contain a proteolytic activity that processed a 12-residue peptide of pro-TNF-alpha (COOH-SPLAQAVRSSSR-NH₂) at the physiological Ala76-Val77 cleavage site [10, 15]. Furthermore, this peptide cleavage was hydroxamate sensitive. Since the characteristics of the physiological TNF-alpha release and the proteolytic activity in THP-1 cells were identical with regard to substrate specificity and inhibition profile, Black et al. [16] used THP-1 cells as a source for the isolation of TACE. The enzyme was purified from plasma membranes of approximately 10¹¹ THP-1 cells by classic protein purification methods, monitoring the enrichment of enzymatic activity by a TNF-alpha-peptide cleavage assay. A 85 kDa protein in activity-containing fractions was incubated with endo-LYS-C to generate peptides for microsequencing. The sequence information obtained by Edman degradation was used to design degenerate oligonucleotides primers for polymerase chain reaction (PCR) on cDNA derived from mRNA of LPS-stimulated THP-1 cells. A PCR product representing a fragment of the TACE cDNA was used as probe to isolate a full length TACE cDNA from a
lambda-library of the human epithelial cell line KB [16].

The TACE cDNA has an open reading frame of 2,472 bp encoding 824 amino acids (Figure 2) [16]. The sequence of one of the Lys-C-peptides which contained a zinc-binding domain similar to those found in several metalloproteases, and the aspartic acid residue following the third conserved histidine (Asp 416, Figure 2), indicated that TACE belonged to the adamalysin family of metzincin metalloproteases [17] (see below). Using recombinant TACE protein, it could be shown that the enzyme processed metabolically labeled pro-TNF-alpha at the physiological Ala76-Val77 cleavage site. Furthermore, Black et al. found that inactivation of the TACE gene by homologous recombination resulted in a 80-90% reduction of TNF-alpha release from T cells of TACE^­/­ mice [16]. At the same time, increased surface expression of pro-TNF-alpha was observed. The residual TNF-alpha release by TACE deficient T-cells could not be inhibited by a hydroxamate based metalloprotease inhibitor.

Simultaneously, Moss et al. [18] purified and cloned a TACE cDNA using a TNF-alpha peptide cleavage assay to monitor TACE activity. This group purified TACE from several kg of porcine spleen because this tissue has an abundance of immunocompetent cells and because of the similarities between the porcine and human TNF-alpha cleavage sites. Their purification scheme included an affinity step at a hydroxamate column. A partial porcine cDNA clone was used to isolate human TACE from a monocyte cDNA library.

TACE BELONGS TO THE ADAM FAMILY

TACE belongs to a family of membrane proteins which show high homology to soluble zinc-dependent metalloproteases present in snake venom [17, 19, 20]. The snake venom toxins have been named adamalysins after adamalysin II, an enzyme from the rattlesnake Crotalus adamanteus [21]. The mammalian enzymes were grouped into "A Disintegrin and Metalloprotease family" (ADAM family) (TACE is ADAM 17) [19, 20]. The domain structure of the extracellular part of the mammalian ADAMs and the adamalysins is virtually identical. As shown in Figure 2, in TACE (ADAM 17) a 17 amino acid signal peptide is followed by a prodomain of about 200 amino acids which has to be removed for activation of the enzyme [18]. The amino acid sequence "RVKR" at the end of the prodomain resembles the consensus sequence of furin proteases [22] and is most likely used to release this domain by limited proteolysis. A probable cysteine switch at Cys 184 [16] (Figure 2) is believed to fold across the active site with the thiol group coordinating (and thereby blocking) the catalytically essential zinc ion [17].

The catalytic domain of TACE contains a sequence (Figure 2; amino acids 405-416) corresponding to the "HEXXHXXGXXH(D) motif" [17] which is highly conserved in zinc-dependent metalloproteases. The three histidine residues are involved in binding the catalytically essential zinc ion, the aspartate residue following the third histidine is found in all adamalysins. The function of the disintegrin and EGF-like domain is not known (see below). In contrast to the snake venom metalloproteases, TACE contains a transmembrane region and a cytoplasmic tail of about 100 amino acids.

Recently, the crystal structure of the catalytic domain of TACE has been solved [23]. A close look at the topology of this domain revealed TACE to be a typical zinc-dependent metalloprotease with closest similarities to the catalytic domain of snake venom metalloproteases such as adamalysin II. About 175 out of 259 residues of the catalytic domain of TACE superimpose with the equivalent region of adamalysin, where the catalytic domain contains 201 residues. However, with 259 residues the catalytic domain of TACE is much longer than that of most other adamalysins/ADAMs. The additional amino acids give rise to two surface exposed loops in the central part of the catalytic domain; this is a feature TACE might only share with ADAM 10, the most closely related member of the ADAM family [23, 24].

The first proteolytic activity assigned to ADAM 10 was cleavage of myelin basic protein [25, 26], which is not a physiological substrate. Recently, in an attempt to purify and clone the TNF-alpha convertase, ADAM 10 was isolated from human THP-1 cells [27] and bovine spleen [28]. It is yet not known if ADAM 10 is involved in pro-TNF-alpha processing under physiological conditions. Results obtained with TACE deficient mice argue against the involvment of ADAM 10 in pro-TNF-alpha cleavage. TACE^­/­ mice release 80-90% less TNF-alpha than corresponding wild type animals [16, 29]. Additionally, mice reconstituted with a TACE^­/­ hematopoietic system also failed to release TNF-alpha [20].

OTHER ADAMs

The cloning of TACE revealed the molecular organisation of the protease (Figure 2). In addition, it became clear that TACE belonged to the still growing family of ADAM proteins the members of witch are involved in different physiological functions. A compilation of all known members of the ADAM family is shown in Table 2.

Besides the adamalysins the "metzincin" superfamily of zinc-dependent metalloproteases [17] includes three further distinct subfamilies. Astacins are named after the prototypical protease astacin which functions as a major collagenolytic enzyme in the digestive tract of the crayfish Astacus astacus. Matrix metalloproteases (MMPs) are involved in the regulated reconstruction of the extracellular matrix during development and differentiation. Serralysins are proteolytic enzymes which are secreted by various pathogenic bacteria of the genera Serratia, Pseudomonas and Erwinia. Interestingly, Serratia marcescens metalloprotease was found to release biologically active IL-6R from human cells [30]. Cleavage occurred at a different site than the one used by an endogenous metalloprotease/ADAM after phorbol ester or bacterial toxin treatment of cells [8, 30].

Members of the ADAM family are characterized by the presence of metalloprotease and disintegrin domains [31, 32]. ADAM proteins are implicated in such diverse physiological function as sperm-egg fusion, myoblast fusion, neural development and ectodomain shedding [31, 32]. There is evidence that the metalloprotease and disintegrin domains have independent functions; many ADAM family members have a metalloprotease domain in which the catalytic site consensus sequence HEXXH is not present, indicating lack of catalytic activity (Table 2). Moreover, in the case of fertilin, the noncatalytic metalloprotease domain is removed during sperm maturation. Other members like TACE or ADAM 10 are clearly dependent on the presence of the catalytic metalloprotease domain in their physiological function [20, 32].

So far it is unclear whether the metalloprotease domain is involved in integrin binding or whether the disintegrin domain plays a functional role in ectodomain shedding. Some recent examples, however, point to the importance of both, metalloprotease and disintegrin domains. The ADAM protein KUZ (the drosophila homologue of ADAM 10) which plays a role in regulating neural cell fate in drosophila has been shown to be inhibited by a truncated version of KUZ which lacked the metalloprotease domain. This truncated version of KUZ acted as a dominant negative inhibitor indicating binding to its substrate via the disintegrin domain [33]. Interestingly, substrates of KUZ seem to be components of the Notch signaling pathway [33, 34]. In the case of the protease ADAM 12 it was recently shown directly that the disintegrin domain is an active cell adhesion domain [35]. These data seem to implicate that there is functional interplay between the metalloprotease domain responsible for cleaving of membrane proteins and the disintegrin domain responsible for establishing cell-cell contact.

MECHANISM OF SHEDDING

Little is known about the regulation of the various shedding events which require the proteolytic activity of ADAMs. Knowledge of the structure of TNF-alpha and the catalytic domain of TACE allowed preliminary computer docking experiments [23]. The schematic model (Figure 3) derived from these experiments showed the interaction of the two membrane proteins TNF-alpha and TACE. The active site of the catalytic domain of TACE contacts the cleavage site region of one molecule of the pro-TNF-alpha trimer. Preferential cleavage at the Ala76-Val77 bond by TACE can partly be explained by favorable interactions in the active site vicinity. Additionally, an interaction of the catalytic domain of TACE and the cone [36] of the pro-TNF-alpha trimer is postulated. The disintegrin domain (which has been shown to interact with alphavbeta3 integrins in case of ADAM 15 [37]) seems not to be involved in the substrate-enzyme interaction hypothesized for pro-TNF-alpha and TACE. Studies of the proteolytic processing of pro-TGF-alpha [38] and angiotensin converting enzyme [39] support the idea that both, the specific interaction between the substrate cleavage site and the active site of the enzyme and the interaction of the sheddase with cleavage site distal regions of the substrate are important for shedding.

Arribas et al. showed that the proteolytic cleavage site of pro-TGF-alpha could be transferred to the otherwise uncleavable protein betaglycan, thereby rendering this protein susceptible to cleavage. For angiotensin converting enzyme it had been shown that the distal extracellular domain is sufficient for recognition and cleavage by the involved sheddase and that the transfer of this domain onto the protein CD4 renders this otherwise not proteolytically processed transmembrane protein susceptible to shedding [39].

So far, a direct interaction of the intracellular tails of ADAMs and their substrates has not been found. Nevertheless, binding of cytosolic proteins to the intracellular part of both groups of membrane proteins has been described. The calcium regulatory protein calmodulin was shown to be co-precipitated with L-selectin and might be involved in the regulation of shedding of this membrane protein [40]. Inhibitors, which block calmodulin binding to the cytoplasmic tail of L-selectin, accelerate the release of L-selectin from cells. Most likely, calmodulin is bound to L-selectin in resting cells. The removal of calmodulin during cell activation resulted in L-selectin shedding. The release of calmodulin from the cytoplasmic tail of L-selectin might induce a conformational change that exposes the extracellular cleavage site to a constitutively active ADAM, e.g. TACE.

Using the yeast two-hybrid system, an interaction between the cytoplasmic portion of TACE and mitotic arrest deficient 2 (MAD2), a protein involved in cell cycle control has been demonstrated [41]. The functional significance of the finding remains to be elucidated. TACE might play a role in control of cell cycle progression. Alternatively, MAD2 could also have cell cycle independent functions. With the same approach a novel MAD2 related protein was identified as an interacting protein of the cytoplasmic tail of ADAM 9 [41] a protease involved in processing of heparin-binding EGF-like growth factor (HB-EGF) [42]. Protein kinase Cdelta also binds to the intracellular part of ADAM 9, overexpression of this kinase led to increased shedding of the ectodomain of proEGF [42].

As mentioned above, shedding can be typically stimulated by phorbol ester, e.g. PMA. A recent study by Gechtman et al. demonstrated that PMA-induced shedding of a membrane-anchored member of the EGF family in CHO cells is regulated by the Raf/mitogen-activated protein kinase (MAPK) pathway [43]. Inhibitors of the MAPK pathway not only blocked the phosphorylation of MAPK but also the release of EGF from CHO cells. Furthermore, it was shown that phorbol ester stimulation alone was not sufficient to induce shedding of EGF. When CHO cells were kept in suspension, PMA treatment still led to phosphorylation of MAPK but no cleavage of proEGF was observed [43]. Shedding was restored when PMA treated cells were allowed to attach and spread on fibronectin coated culture dishes [43]. How cell-spreading contributes to EGF shedding is not known. Appropriate organization of the cytoskeleton of the cell might be needed to position the shedding machinery and substrates at the cell surface. Interestingly, pore forming bacterial toxins which disturb membrane integrity and cause rapid loss of membrane lipid asymmetry [44, 45] were found to induce massive shedding of IL-6R and CD14 from transfected cells and human monocytes [8].

CONSEQUENCES OF SHEDDING FOR CYTOKINE BIOLOGY

Shedding of membrane proteins can have two different physiologic consequences. First, the cell from which the protein is shed looses the membrane protein. If the membrane protein is a receptor, extensive shedding results in loss of responsiveness through this receptor system. Secondly, the once membrane associated protein can be distributed throughout the circulation. If the membrane bound protein is a cytokine, this leads to systemic availability of this cytokine. As detailed below, the presence of a soluble receptor can lead to agonistic or antagonistic biologic activities.

The possible consequences of shedding of membrane bound cytokines is illustrated in Figure 4. As shown in panel A, membrane bound cytokines can stimulate cytokine receptors on adjacent cells restricting the effect of the cytokine to cell to cell or paracrine action. Shedding of the cytokine by a membrane bound protease leads to the release of soluble cytokine molecules which now are spread through the circulation (Figure 4B). In the case of TNF-alpha it was shown that the membrane form of this cytokine preferentially binds and activates the type II TNF-R and thereby can lead to qualitatively different TNF responses such as rendering resistant tumor cells sensitive to TNF-mediated cytotoxicity [46]. These results indicate that, in the case of TNF-alpha, membrane bound and soluble forms of the cytokine can have distinct biologic functions. These findings were further corroborated by a transgenic mouse model. The group of Kollias constructed transgenic mice expressing a form of TNF-alpha which lacked the cleavage site and which therefore was not subjected to cleavage. These mice were crossed on the background of TNF-alpha^­/­ mice. Consequently, the only form of TNF-alpha in these mice was membrane bound TNF-alpha. Using this transgenic mouse model, it was shown in vivo that in the absence of soluble TNF-alpha, transmembrane TNF is by itself sufficient to mediate the pathogenesis of arthritis. These results indicate that blocking the activities of both soluble and transmembrane TNF may be required to effectively neutralize the pathogenic potential of this cytokine in arthritis [47]. More recently, it was demonstrated that the cytotoxic effects induced by the type II TNF-R stimulated by endogenous membrane-anchored TNF-alpha were mediated via stimulation of the endogenous production of TNF-alpha and subsequent autotropic or paratropic stimulation of type I TNF-R. These studies illustrate the complexity of this cytokine system which used both soluble and membrane bound cytokine components [48].

Since mice lacking TNF-alpha or TNF receptors were viable [49] it was surprising that TACE^­/­ mice were not [29]. This finding immediately indicated that TACE activity was also required for other processes. One possibility was that the release of membrane proteins other than TNF-alpha was affected. The TACE^­/­ mice phenotypically presented with abnormalities of the hair follicles and open eyes at birth [29]. This phenotype closely resembled the phenotype found in TGF-alpha^­/­ mice [50, 51]. One possible interpretation of these results was that cytokines and growth factors other than TNF-alpha were substrates for TACE and that the lack of these cytokines and growth factors in soluble form would be the reason for the severe phenotype of the TACE^­/­ mice. This result not only suggested the possible importance of TACE for the shedding of several substrates, but it provided evidence for the requirement of shedding of members of the TGF-alpha family for their own biologic function [29, 50, 51].

IL-1 binds to two types of receptors on the cell membrane, of which only type I (IL-1RI) transduces signals together with the recently identified coreceptor IL-1 receptor accessory protein (IL-1RAcP). The type II (IL-1RII) functions as a decoy receptor without participating in IL-1 signaling. It has recently been proposed that upon IL-1 binding the IL-1RII can recruit IL-1RAcP into a nonfunctional trimeric complex and thus modulate IL-1 signaling by sequestering the coreceptor molecule from the signaling IL-1RI. In this mechanism of coreceptor competition, the ratio between IL-1RII and IL-1RI is a central factor in determining the cellular responsiveness to IL-1 [52]. Interestingly, the extracellular domain of the type II IL-1 receptor is released from many cells (Figure 1) by a metalloprotease [53] and can function as a specific inhibitor of IL-1 activity. The soluble IL-1R retains its affinity for the ligand IL-1. However, upon shedding the affinity of the soluble receptor for the naturally occuring IL-1 receptor antagonist (IL-1Ra) drops by a factor of 2,000 when compared with the cell surface receptor [54]. Therefore, the type II sIL-1R inhibits IL-1 by competing for the interaction of mature IL-1 with the type I IL-1 receptor. In addition, type II sIL-1R does not interfere with inhibition of IL-1 signaling mediated by the IL-1RA [54]. Since IL-1 is involved in many inflammatory processes, the control of IL-1 activity by the type II sIL-1R is important for the understanding of these processes. Though the metalloprotease involved in the release of the type II sIL-1R has not been yet formally identified [53], recent preliminary evidence would suggest that TACE might be involved in the shedding mechanism.

Recent results indicate that also serine proteases secreted by activated neutrophils like elastase and cathepsin G may contribute to shedding of IL-2R and IL-6R [55]. High intracerebral protease concentrations were found in neurotrauma patients and the protease concentrations correlated with levels of sIL-2R and sIL-6R. Interestingly, elastase showed preference for the IL-2R whereas cathepsin G selectively used the IL-6R as a substrate. Purified proteases at concentrations found in cerebral fluids were able to generate soluble forms of IL-2R and IL-6R which showed intact ligand binding activity [55].

Possible consequences of shedding of receptors for growth factors and cytokines are shown in Figure 4. Most soluble receptors compete with their membrane bound counterparts for their ligands thereby acting as competitive antagonists. Receptors for IL-1, IL-2, IL-3, IL-4, IL-5, IL-7, IL-9, IL-10, IL-13 and IL-15 belong to the group of antagonistic soluble receptors. Moreover, many of these soluble receptors have been shown to prolong the plasma half life of cytokines and therefore are regarded as carrier proteins of their ligands (Figure 4C) [1-3, 56, 57]. In contrast, many soluble receptors for IL-6 type cytokines act as agonists. As schematically shown in Figure 4D, IL-6 can bind to its soluble receptor and this complex can bind to and activate the gp130 signal transducing protein thereby initiating signaling. Interestingly, a cell which expresses gp130 but no membrane bound IL-6R would not be able to respond to IL-6. Therefore, in concept, a cell which releases a soluble receptor renders a second cell type responsive to the cytokine. This principle has been named "transsignaling" [2]. So far, it has been formally demonstrated that IL-6, IL-11, and CNTF can activate target cells via transsignaling. Recently, many cells in the body have been identified to require soluble cytokine receptors to respond to cytokines of the IL-6 family. Such cells include hematopoietic progenitor cells [58, 59], endothelial cells [60], smooth muscle cells [61] and neural cells [62].

Recent work with primary sympathetic neurons [63] and with human primary smooth muscle cells [61] has revealed a new principle of paracrine cytokine signaling between cells. Both, neurons and smooth muscle cells secrete IL-6 but since they express no IL-6R on the cell surface they do not respond to the IL-6 they secrete. This IL-6, however, is fully active on other cells. Both cell types can be stimulated by the combination of IL-6 and sIL-6R or by a fusion protein consisting of IL-6 and sIL-6R [64]. It turned out that in the presence of recombinant sIL-6R both neurons and smooth muscle cells respond to the IL-6 they were secreting themselves. Therefore, it seems likely that cells which produce IL-6 are dependent on neighboring cells for their supply of sIL-6R. Good candidates for cells which produce sIL-6R are tissue macrophages and neutrophils which have recently been shown to generate sIL-6R in response to the acute phase protein C reactive protein [9].

OUTLOOK

It is now clear that the vast majority of membrane proteins can be cleaved by members of the ADAM protease family. As outlined above, shedding of cytokines or cytokine receptors can result in agonistic or antagonistic signals. Membrane bound cytokines can act on a cell to cell basis. Upon shedding, such cytokines are distributed through the circulation and therefore act systemically. Cytokine signals can be blocked by soluble cytokine receptors. As in the case of the IL-6 cytokine family, soluble cytokine receptors can participate in the stimulation of target cells. The large number of members of the ADAM family suggests that there may be subtle substrate specificities of shedding proteases leading to differential shedding of membrane proteins. Moreover, the regulation of expression of ADAM members is poorly understood. Finally, the activity of ADAM proteases apparently is regulated at the posttranslational level. These aspects have to be worked out before the complex interplay of membrane and soluble proteins can be fully appreciated. It should be kept in mind that blockage of TNF-alpha processing can save animals from LPS induced septic shock [10] and that the molecular understanding of the protease which releases IL-1RII could be a key point in therapeutically blocking an inflammatory reaction [65]. Moreover, shedding of L-selectin has been shown to affect leukocyte rolling and subsequent transmigration [40, 66]. Therefore, shedding proteases could define new targets for pharmacological intervention for both acute and chronic inflammatory diseases.

ABBREVIATIONS

ADAM, a disintegrin and metalloprotease; IL, interleukin; MAPK, mitogen-activated protein kinase; R, receptor; s, soluble; SCF, stem cell factor; TACE, TNF-alpha converting enzyme; TAPI, TNF-alpha protease inhibitor; TNF, tumor necrosis factor.

CONCLUSION

REFERENCES

1. Ehlers M R, Riordan J F. 1991. Membrane proteins with soluble counterparts: role of proteolysis in the release of transmembrane proteins. Biochemistry 30: 10065.

2. Rose-John S, Heinrich P C. 1994. Soluble receptors for cytokines and growth factors: generation and biological function. Biochem. J. 300: 281.

3. Hooper N M, Karran E H, Turner A J. 1997. Membrane protein secretases. Biochem. J. 321: 265.

4. Massague J, Pandiella A. 1993. Membrane-anchored growth factors. Annu. Rev. Biochem. 62: 515.

5. Heaney M L, Golde D W. 1996. Soluble cytokine receptors. Blood 87: 847.

6. Müllberg J, Schooltink H, Stoyan T, Heinrich P C, Rose-John S. 1992. Protein kinase C activity is rate limiting for shedding of the interleukin-6 receptor. Biochem. Biophys. Res. Commun. 189: 794.

7. Müllberg J, Schooltink H, Stoyan T, Gunther M, Graeve L, Buse G, Mackiewicz A, Heinrich P C, Rose-John S. 1993. The soluble interleukin-6 receptor is generated by shedding. Eur. J. Immunol. 23: 473.

8. Walev I, Vollmer P, Palmer M, Bhakdi S, Rose-John S. 1996. Pore-forming toxins trigger shedding of receptors for interleukin 6 and lipopolysaccharide. Proc. Natl. Acad. Sci. USA 93: 7882.

9. Jones S A, Novick D, Horiuchi S, Yamamoto N, Szalai A J, Fuller G M. 1999. C-reactive protein: a physiological activator of interleukin 6 receptor shedding. J. Exp. Med. 189: 599.

10. Mohler K M, Sleath P R, Fitzner J N, Cerretti D P, Alderson M, Kerwar S S, Torrance D S, Otten Evans C, Greenstreet T, Weerawarna K, Kronheim S R, Petersen M, Gerhart M, Kozlosky C J, March C J, Black R A. 1994. Protection against a lethal dose of endotoxin by an inhibitor of tumour necrosis factor processing. Nature 370: 218.

11. Gearing A J, Beckett P, Christodoulou M, Churchill M, Clements J, Davidson A H, Drummond A H, Galloway W A, Gilbert R, Gordon J L, Leber T M, Mangan M, Miller K, Nayee P, Owen K, Patel S, Thomas W, Wells G, Wood L M, Woolley K. 1994. Processing of tumour necrosis factor-alpha precursor by metalloproteinases. Nature 370: 555.

12. McGeehan G M, Becherer J D, Bast Jr R C, Boyer C M, Champion B, Connolly K M, Conway J G, Furdon P, Karp S, Kidao S, McElroy A B, Nichols J, Pryzwanszky M, Schoenen F, Sedut L, Truesdale A, Verghese M, Warner J, Ways J P. 1994. Regulation of tumour necrosis factor-alpha processing by a metalloproteinase inhibitor. Nature 370: 558.

13. Müllberg J, Rauch C T, Wolfson M F, Castner B, Fitzner J N, Otten Evans C, Mohler K M, Cosman D, Black R A. 1997. Further evidence for a common mechanism for shedding of cell surface proteins. FEBS Lett. 401: 235.

14. Aggarwal B B, Kohr W J, Hass P E, Moffat B, Spencer S A, Henzel W J, Bringman T S, Nedwin G E, Goeddel D V, Harkins R N. 1985. Human tumor necrosis factor. Production, purification, and characterization. J. Biol. Chem. 260: 2345.

15. Black R A, Durie F H, Otten Evans C, Miller R, Slack J L, Lynch D H, Castner B, Mohler K M, Gerhart M, Johnson R S, Itoh Y, Okada Y, Nagase H. 1996. Relaxed specificity of matrix metalloproteinases (MMPS) and TIMP insensitivity of tumor necrosis factor-alpha (TNF-alpha) production suggest the major TNF-alpha converting enzyme is not an MMP. Biochem. Biophys. Res. Commun. 225: 400.

16. Black R A, Rauch C T, Kozlosky C J, Peschon J J, Slack J L, Wolfson M F, Castner B J, Stocking K L, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley K A, Gerhart M, Davis R, Fitzner J N, Johnson R S, Paxton R J, March C J, Cerretti D P. 1997. A metalloproteinase disintegrin that releases tumour-necrosis factor-alpha from cells. Nature 385: 729.

17. Stöcker W, Grams F, Baumann U, Reinemer P, Gomis-Ruth F X, McKay D B, Bode W. 1995. The metzincins-topological and sequential relations between the astacins, adamalysins, serralysins, and matrixins (collagenases) define a superfamily of zinc-peptidases. Protein Sci. 4: 823.

18. Moss M L, Jin S L, Milla M E, Bickett D M, Burkhart W, Carter H L, Chen W J, Clay W C, Didsbury J R, Hassler D, Hoffman C R, Kost T A, Lambert M H, Leesnitzer M A, McCauley P, McGeehan G, Mitchell J, Moyer M, Pahel G, Rocque W, Overton L K, Schoenen F, Seaton T, Su J L, Warner J, Willard D, Becherer J D. 1997. Cloning of a disintegrin metalloproteinase that processes precursor tumour-necrosis factor-alpha. Nature 385: 733.

19. Wolfsberg T G, Primakoff P, Myles D G, White J M. 1995. ADAM, a novel family of membrane proteins containing A Disintegrin And Metalloprotease domain: multipotential functions in cell-cell and cell-matrix interactions. J. Cell Biol. 131: 275.

20. Black R A, White J M. 1998. ADAMs: focus on the protease domain. Curr. Opin. Cell Biol. 10: 654.

21. Gomis-Ruth F X, Meyer E F, Kress L F, Politi V. 1998. Structures of adamalysin II with peptidic inhibitors. Implications for the design of tumor necrosis factor-alpha convertase inhibitors. Protein Sci. 7: 283.

22. Barr P J. 1991. Mammalian subtilisins: the long-sought dibasic processing endoproteases. Cell 66: 1.

23. Maskos K, Fernandez Catalan C, Huber R, Bourenkov G P, Bartunik H, Ellestad G A, Reddy P, Wolfson M F, Rauch C T, Castner B J, Davis R, Clarke H R, Petersen M, Fitzner J N, Cerretti D P, March C J, Paxton R J, Black R A, Bode W. 1998. Crystal structure of the catalytic domain of human tumor necrosis factor-alpha-converting enzyme. Proc. Natl. Acad. Sci. USA 95: 3408.

24. Cerretti D P, DuBose R F, Black R A, Nelson N. 1999. Isolation of two novel metalloproteinase-disintegrin (ADAM) cDNAs that show testis-specific gene expression. Biochem. Biophys. Res. Commun. 263: 810.

25. Chantry A, Gregson N A, Glynn P. 1989. A novel metalloproteinase associated with brain myelin membranes. Isolation and characterization. J. Biol. Chem. 264: 21603.

26. Howard L, Lu X, Mitchell S, Griffiths S, Glynn P. 1996. Molecular cloning of MADM: a catalytically active mammalian disintegrin-metalloprotease expressed in various cell types. Biochem. J. 317: 45.

27. Rosendahl M S, Ko S C, Long D L, Brewer M T, Rosenzweig B, Hedl E, Anderson L, Pyle S M, Moreland J, Meyers M A, Kohno T, Lyons D, Lichenstein H S. 1997. Identification and characterization of a pro-tumor necrosis factor-alpha-processing enzyme from the ADAM family of zinc metalloproteases. J. Biol. Chem. 272: 24588.

28. Lunn C A, Fan X, Dalie B, Miller K, Zavodny P J, Narula S K, Lundell D. 1997. Purification of ADAM 10 from bovine spleen as a TNF-alpha convertase. FEBS Lett. 400: 333.

29. Peschon J J, Slack J L, Reddy P, Stocking K L, Sunnarborg S W, Lee D C, Russell W E, Castner B J, Johnson R S, Fitzner J N, Boyce R W, Nelson N, Kozlosky C J, Wolfson M F, Rauch C T, Cerretti D P, Paxton R J, March C J, Black R A. 1998. An essential role for ectodomain shedding in mammalian development. Science 282: 1281.

30. Vollmer P, Walev I, Rose-John S, Bhakdi S. 1996. Novel pathogenic mechanism of microbial metalloproteinases: liberation of membrane-anchored molecules in biologically active form exemplified by studies with the human interleukin-6 receptor. Infect. Immun. 64: 3646.

31. Wolfsberg T G, White J M. 1996. ADAMs in fertilization and development. Dev. Biol. 180: 389.

32. Blobel C P. 1997. Metalloprotease-disintegrins: links to cell adhesion and cleavage of TNF-alpha and Notch. Cell 90: 589.

33. Pan D, Rubin G M. 1997. Kuzbanian controls proteolytic processing of Notch and mediates lateral inhibition during Drosophila and vertebrate neurogenesis. Cell 90: 271.

34. Qi H, Rand M D, Wu X, Sestan N, Wang W, Rakic P, Xu T, Artavanis-Tsakonas S. 1999. Processing of the notch ligand delta by the metalloprotease Kuzbanian. Science 283: 91.

35. Zolkiewska A. 1999. Disintegrin-like/cysteine-rich region of ADAM 12 is an active cell adhesion domain. Exp Cell Res. 252: 423.

36. Jones E Y, Stuart D I, Walker N P. 1989. Structure of tumour necrosis factor. Nature 338: 225.

37. Zhang X P, Kamata T, Yokoyama K, Puzon-McLaughlin W, Takada Y. 1998. Specific interaction of the recombinant disintegrin-like domain of MDC-15 (metargidin, ADAM-15) with integrin alphavbeta3. J. Biol. Chem. 273: 7345.

38. Arribas J, Lopez-Casillas F, Massague J. 1997. Role of the juxtamembrane domains of the transforming growth factor-alpha precursor and the beta-amyloid precursor protein in regulated ectodomain shedding. J. Biol. Chem. 272: 17160.

39. Sadhukhan R, Sen G C, Ramchandran R, Sen I. 1998. The distal ectodomain of angiotensin-converting enzyme regulates its cleavage-secretion from the cell surface. Proc. Natl. Acad. Sci. USA 95: 138.

40. Kahn J, Walcheck B, Migaki G I, Jutila M A, Kishimoto T K. 1998. Calmodulin regulates L-selectin adhesion molecule expression and function through a protease-dependent mechanism. Cell 92: 809.

41. Nelson K K, Schlondorff J, Blobel C P. 1999. Evidence for an interaction of the metalloprotease-disintegrin tumour necrosis factor alpha convertase (TACE) with mitotic arrest deficient 2 (MAD2), and of the metalloprotease-disintegrin MDC9 with a novel MAD2-related protein, MAD2beta. Biochem. J. 343: 673.

42. Izumi Y, Hirata M, Hasuwa H, Iwamoto R, Umata T, Miyado K, Tamai Y, Kurisaki T, Sehara-Fujisawa A, Ohno S, Mekada E. 1998. A metalloprotease-disintegrin, MDC9/meltrin-gamma/ADAM9 and PKCdelta are involved in TPA-induced ectodomain shedding of membrane-anchored heparin-binding EGF-like growth factor. EMBO J. 17: 7260.

43. Gechtman Z, Alonso J L A, Raab G, Ingber D E, Klagsbrun M. 1999. The shedding of membrane-anchored heparin-binding epidermal-like growth factor is regulated by the Raf/Mitogen-Activated Protein Kinase cascade and by cell adhesion and spreading. J. Biol. Chem. 274: 28828.

44. Bhakdi S, Tranum-Jensen J. 1987. Damage to mammalian cells by proteins that form transmembrane pores. Rev. Physiol. Biochem. Pharmacol. 107: 147.

45. Bhakdi S, Bayley H, Valeva A, Walev I, Walker B, Kehoe M, Palmer M. 1996. Staphylococcal alpha-toxin, streptolysin O, and Escherichia coli hemolysin: prototypes of pore-forming bacterial cytolysins. Arch. Microbiol. 165: 73.

46. Grell M, Douni E, Wajant H, Löhden M, Clauss M, Maxeiner B, Georgopoulos S, Lesslauer W, Kollias G, Pfitzenmaier K, Scheurich P. 1995. The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor. Cell 83: 793.

47. Alexopoulou L, Pasparakis M, Kollias G. 1997. A murine transmembrane tumor necrosis factor (TNF) transgene induces arthritis by cooperative p55/p75 TNF receptor signaling. Eur. J. Immunol. 27: 2588.

48. Grell M, Zimmermann G, Gottfried E, Chen C M, Grunwald U, Huang D C, Wu Lee Y H, Dürkop H, Engelmann H, Scheurich P, Wajant H, Strasser A. 1999. Induction of cell death by tumour necrosis factor (TNF) receptor 2, CD40 and CD30: a role for TNF-R1 activation by endogenous membrane-anchored TNF. EMBO J. 18: 3034.

49. Eugster H P, Muller M, Karrer U, Car B D, Schnyder B, Eng V M, Woerly G, Le Hir M, di Padova F, Aguet M, Zinkernagel R, Bluethmann H, Ryffel B. 1996. Multiple immune abnormalities in tumor necrosis factor and lymphotoxin-alpha double-deficient mice. Int. Immunol. 8: 23.

50. Luetteke N C, Qiu T H, Peiffer R L, Oliver P, Smithies O, Lee D C. 1993. TGF-alpha deficiency results in hair follicle and eye abnormalities in targeted and waved-1 mice. Cell 73: 263.

51. Mann G B, Fowler K J, Gabriel A, Nice E C, Williams R L, Dunn A R. 1993. Mice with a null mutation of the TGF-alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inflammation. Cell 73: 249.

52. Lang D, Knop J, Wesche H, Raffetseder U, Kurrle R, Boraschi D, Martin M U. 1998. The type II IL-1 receptor interacts with the IL-1 receptor accessory protein: a novel mechanism of regulation of IL-1 responsiveness. J. Immunol. 161: 6871.

53. Orlando S, Sironi M, Bianchi G, Drummond A H, Boraschi D, Yabes D, Mantovani A. 1997. Role of metalloproteases in the release of the IL-1 type II decoy receptor. J. Biol. Chem. 272: 31764.

54. Symons J A, Young P R, Duff G W. 1995. Soluble type II interleukin-1 (IL-1) receptor binds and blocks processing of IL-1beta precursor and loses affinity for IL-1 receptor antagonist. Proc. Natl. Acad. Sci. USA 92: 1714.

55. Bank U, Reinhold D, Schneemilch C, Kunz D, Synowitz H J, Ansorge S. 1999. Selective proteolytic cleavage of IL-2 receptor and IL-6 receptor ligand binding chains by neutrophil-derived serine proteases at foci of inflammation. J. Interferon Cytokine Res. 19: 1277.

56. Fernandez-Botran R. 1991. Soluble cytokine receptors: their role in immunoregulation. FASEB J. 5: 2567.

57. Fernandez-Botran R, Vitetta E S. 1991. Evidence that natural murine soluble interleukin 4 receptors may act as transport proteins. J. Exp. Med. 174: 673.

58. Sui X, Tsuji K, Tanaka R, Tajima S, Muraoka K, Ebihara Y, Ikebuchi K, Yasukawa K, Taga T, Kishimoto T, Nakahata T. 1995. gp130 and c-Kit signalings synergize for ex vivo expansion of human primitive hemopoietic progenitor cells. Proc. Natl. Acad. Sci. USA 92: 2859.

59. Peters M, Schirmacher P, Goldschmitt J, Odenthal M, Peschel C, Dienes H P, Fattori E, Ciliberto G, Meyer zum Büschenfelde K H, Rose-John S. 1997. Extramedullary expansion of hematopoietic progenitor cells in IL-6/sIL-6R double transgenic mice. J. Exp. Med. 185: 755.

60. Romano M, Sironi M, Toniatti C, Polentarutti N, Fruscella P, Ghezzi P, Faggioni R, Luini W, van Hinsbergh V, Sozzani S, Bussolino F, Poli V, Ciliberto G, Mantovani A. 1997. Role of IL-6 and its soluble receptor in induction of chemokines and leukocyte recruitment. Immunity 6: 315.

61. Klouche M, Bhakdi S, Hemmes M, Rose-John S. 1999. Novel path of activation of primary human smooth muscle cells: upregulation of gp130 creates an autocrine activation loop by IL-6 and its soluble receptor. J. Immunol. 163: 4583.

62. März P, Otten U, Rose-John S. 1999. Neuronal activities of IL-6 type cytokines often depend on soluble cytokine receptors. Eur. J. Neurosci. 11: 2995.

63. März P, Cheng J-C, Gadient R A, Patterson P, Stoyan T, Otten U, Rose-John S. 1998. Sympathetic neurons can produce and respond to interleukin-6. Proc. Natl. Acad. Sci. USA 95: 3251.

64. Fischer M, Goldschmitt J, Peschel C, Kallen K J, Brakenhoff J P J, Wollmer A, Grötzinger J, Rose-John S. 1997. A designer cytokine with high activity on human hematopoietic progenitor cells. Nature Biotechnol. 15: 142.

65. Boraschi D, Cifone M G, Falk W, Flad H D, Tagliabue A, Martin M U. 1998. Cytokines in inflammation. Joint Workshop of the Deutsche Gesellschaft für Immunologie and the Gruppo di Cooperazione in Immunologia. Eur. Cytokine Netw. 9: 205.

66. Walchek B, Kahn J, Fisher J M, Wang B B, Fisk R S, Payan D G, Feehan C, Betageri R, Darlak K, Spatola S F, Kishimoto T K. 1996. Neutrophil rolling altered by inhibition of
L-selectin shedding in vitro. Nature 380: 720.

67. Müllberg J, Oberthur W, Lottspeich F, Mehl E, Dittrich E, Graeve L, Heinrich P C, Rose-John S. 1994. The soluble human IL-6 receptor. Mutational characterization of the proteolytic cleavage site. J. Immunol. 152: 4958.

68. Müllberg J, Durie F H, Otten Evans C, Alderson M R, Rose-John S, Cosman D, Black R A, Mohler K M. 1995. A metalloprotease inhibitor blocks shedding of the IL-6 receptor and the p60 TNF receptor. J. Immunol. 155: 5198.

69. Arribas J, Coodly L, Vollmer P, Kishimoto T K, Rose-John S, Massague J. 1996. Diverse cell surface protein ectodomains are shed by a system sensitive to metalloprotease inhibitors. J. Biol. Chem. 271: 11376.

70. Himmler A, Maurer-Fogy I, Kronke M, Scheurich P, Pfizenmaier K, Lantz M, Olsson I, Hauptmann R, Stratowa C, Adolf G R. 1990. Molecular cloning and expression of human and rat tumor necrosis factor receptor chain (p60) and its soluble derivative, tumor necrosis factor-binding protein. DNA Cell Biol. 9: 705.

71. Crowe P D, Walter B N, Mohler K M, Otten Evans C, Black R A, Ware C F. 1995. A metalloprotease inhibitor blocks shedding of the 80-kD TNF receptor and TNF processing in T lymphocytes. J. Exp. Med. 181: 1205.

72. Migaki G I, Kahn J, Kishimoto T K. 1995. Mutational analysis of the membrane-proximal cleavage site of L-selectin: relaxed sequence specificity surrounding the cleavage site. J. Exp. Med. 182: 549.

73. Higaki J, Quon D, Zhong Z, Cordell B. 1995. Inhibition of beta-amyloid formation identifies proteolytic precursors and subcellular site of catabolism. Neuron 14: 651.

74. Lammich S, Kojro E, Postina R, Gilbert S, Pfeiffer R, Jasionowski M, Haass C, Fahrenholz F. 1999. Constitutive and regulated alpha-secretase cleavage of Alzheimer's amyloid precursor protein by a disintegrin metalloprotease. Proc. Natl. Acad. Sci. USA 96: 3922.

75. Ehlers M R, Schwager S L, Scholle R R, Manji G A, Brandt W F, Riordan J F. 1996. Proteolytic release of membrane-bound angiotensin-converting enzyme: role of the juxtamembrane stalk sequence. Biochemistry 35: 9549.

76. Huang E, Nocka K, Beier D R, Chu T Y, Buck J, Lahm H W, Wellner D, Leder P, Besmer P. 1990. The hematopoietic growth factor KL is encoded by the Sl locus and is the ligand of the c-kit receptor, the gene product of the W locus. Cell 63: 225.

77. Derynck R, Roberts A B, Winkler M E, Chen E Y, Goeddel D V. 1984. Human transforming growth factor-alpha: precursor structure and expression in E. coli. Cell 38: 287.

78. Tanaka M, Itai T, Adachi M, Nagata S. 1998. Downregulation of Fas ligand by shedding. Nature Med. 4: 31.

79. Kojro E, Fahrenholz F. 1995. Ligand-induced cleavage of the V2 vasopressin receptor by a plasma membrane metalloproteinase. J. Biol. Chem. 270: 6476.

80. Arribas J, Massague J. 1995. Transforming growth factor-alpha and beta-amyloid precursor protein share a secretory mechanism. J. Cell Biol. 124: 433.

81. Merlos-Suarez A, Fernandez-Larrea J, Reddy P, Baselga J, Arribas J. 1998. Pro-tumor necrosis factor-alpha processing activity is tightly controlled by a component that does not affect notch processing. J. Biol. Chem. 273: 24955.