We previously implicated tumor necrosis factor-␣ converting enzyme (TACE/ADAM17) in the processing of the integral membrane precursor to soluble transforming growth factor-␣ (TGF-␣), pro-TGF-␣. Here we examined TGF-␣ processing in a physiologically relevant cell model, primary keratinocytes, showing that cells lacking TACE activity shed dramatically less TGF-␣ as compared with wild-type cultures and that TGF-␣ cleavage was partially restored by infection of TACE-deficient cells with TACE-encoding adenovirus. Moreover, cotransfection of TACE-deficient fibroblasts with pro-TGF-␣ and TACE cDNAs increased shedding of mature TGF-␣ with concomitant conversion of cell-associated pro-TGF-␣ to a processed form. Purified TACE accurately cleaved pro-TGF-␣ in vitro at the N-terminal site and also cleaved a soluble form of pro-TGF-␣ containing only the ectodomain at the C-terminal site. In vitro, TACE accurately cleaved peptides corresponding to cleavage sites of several epidermal growth factor (EGF) family members, and transfection of TACE into TACEdeficient cells increased the shedding of amphiregulin and heparin-binding EGF (HB-EGF) proteins. Consistent with the hypothesis that TACE regulates EGF receptor (EGFR) ligand availability in vivo, mice heterozygous for Tace and homozygous for an impaired EGFR allele (wa-2) were born with open eyes significantly more often than Tace ؉/؉ Egfr wa-2/wa-2 counterparts. Collectively, these data support a broad role for TACE in the regulated shedding of EGFR ligands.

EGF family growth factors, including transforming growth factor-alpha (TGFalpha), amphiregulin (AR), and heparin-binding EGF (HB-EGF), are invariably expressed as transmembrane precursors that are cleaved at one or more sites in the extracellular domain to release soluble growth factor. Considerable attention has focused on the identification of proteases responsible for these processing events. We previously implicated tumor necrosis factor-alpha converting enzyme (TACE/ADAM17) in the generation of soluble TGFalpha from its transmembrane precursor, proTGFalpha. Here, we review our findings that primary keratinocytes from Tace(deltaZn/deltaZn) mice, which express a nonfunctional TACE, released dramatically lower levels of soluble TGFalpha compared to their normal counterparts, even though TGFalpha mRNA and cell-associated protein levels were similar in the two cell populations. Restoration of TACE activity in Tace(deltaZn/deltaZn) cells increased shedding of TGFalpha species, including the mature, 6-kDa protein. Further, exogenous TACE enzyme accurately cleaved the N-terminal processing site of proTGFalpha in cell lysates, as well as both physiologic sites of a soluble proTGFalpha ectodomain. TACE also accurately cleaved peptide substrates corresponding to the processing sites of several additional EGF family members, and restoration of TACE activity enhanced the shedding of soluble AR and HB-EGF proteins from Tace(deltaZn/deltaZn) cells. Finally, reduction of functional TACE gene dosage greatly exacerbated the open-eye defect of Egfr(wa-2/wa-2) newborns, which is regulated by redundant actions of several EGF family ligands. The implications of these results for the biology of the EGF family and TACE are discussed.

Epidermal growth factor (EGF) family ligands are derived by proteolytic cleavage of the ectodomains of integral membrane precursors. Previously, we established that tumor necrosis factor ␣-converting enzyme (TACE/ ADAM17) is a physiologic transforming growth factor-␣ (TGF-␣) sheddase, and we also demonstrated enhanced shedding of amphiregulin (AR) and heparin-binding (HB)-EGF upon restoration of TACE activity in TACEdeficient EC-2 fibroblasts. Here we extended these results by showing that purified soluble TACE cleaved single sites in the juxtamembrane stalks of mouse pro-HB-EGF and pro-AR ectodomains in vitro. For pro-HB-EGF, this site matched the C terminus of the purified human growth factor, and we speculate that the AR cleavage site is also physiologically relevant. In contrast, ADAM9 and -10, both implicated in HB-EGF shedding, failed to cleave the ectodomain or cleaved at a nonphysiologic site, respectively. Cotransfection of TACE in EC-2 cells enhanced phorbol myristate acetateinduced but not constitutive shedding of epiregulin and had no effect on betacellulin (BTC) processing. Additionally, soluble TACE did not cleave the juxtamembrane stalks of either pro-BTC or pro-epiregulin ectodomains in vitro. Substitution of the shorter pro-BTC juxtamembrane stalk or truncation of the pro-TGF-␣ stalk to match the pro-BTC length reduced TGF-␣ shedding from transfected cells to background levels, whereas substitution of the pro-BTC P2-P2 sequence reduced TGF-␣ shedding less dramatically. Conversely, substitution of the pro-TGF-␣ stalk or lengthening of the pro-BTC stalk, especially when combined with substitution of the pro-TGF-␣ P2-P2 sequence, markedly increased BTC shedding. These results indicate that efficient TACE cleavage is determined by a combination of stalk length and scissile bond sequence.

Transmembrane forms of neural cell adhesion molecule (NCAM140, NCAM180(1)) are key regulators of neuronal development. The extracellular domain of NCAM can occur as a soluble protein in normal brain, and its levels are elevated in neuropsychiatric disorders, such as schizophrenia; however the mechanism of ectodomain release is obscure. Ectodomain shedding of NCAM140, releasing a fragment of 115 kD, was found to be induced in NCAM-transfected L-fibroblasts by the tyrosine phosphatase inhibitor pervanadate, but not phorbol esters. Pervanadate-induced shedding was mediated by a disintegrin metalloprotease (ADAM), regulated by ERK1/2 MAP kinase. In primary cortical neurons, NCAM was shed at high levels, and the metalloprotease inhibitor GM6001 significantly increased NCAM-dependent neurite branching and outgrowth. Moreover, NCAM-dependent neurite outgrowth and branching were inhibited in neurons isolated from a transgenic mouse model of NCAM shedding. These results suggest that regulated metalloprotease-induced ectodomain shedding of NCAM down-regulates neurite branching and neurite outgrowth. Thus, increased levels of soluble NCAM in schizophrenic brain have the potential to impair neuronal connectivity.

Shedding of TNF-alpha requires a single cleavage event, whereas the ectodomain of proTGF-alpha is cleaved at N-proximal (N-terminal) and membrane proximal (C-terminal) sites to release mature TGF-alpha. Tumor necrosis factor-alpha converting enzyme (TACE) was shown to have a central role in the shedding of both factors. Here we show that cleavage of the proTGF-alpha C-terminal site, required for release of mature growth factor, is less sensitive to a panel of hydroxamates than TNF-alpha processing. Recombinant TACE cleaves TNF-alpha and N-terminal TGF-alpha peptides 50-fold more efficiently than the C-terminal TGF-alpha peptide. Moreover, fractionation of rat liver epithelial cell membranes yields two populations: one contains TACE and cleaves peptides corresponding to TNF-alpha and both proTGF-alpha processing sites, while the other lacks detectable TACE and cleaves only the C-terminal proTGF-alpha processing site. Activities in both fractions are inhibited by hydroxamates and EDTA but not by cysteine, aspartate, or serine protease inhibitors. Both membrane fractions also contain ADAM 10. ADAM 10 correctly cleaves peptides and a soluble form of precursor TGF-alpha (proTGFecto) at the N-terminal site but not the C-terminal site. However, the kinetics of N-terminal peptide cleavage by ADAM 10 are 90-fold less efficient than TACE. Our findings indicate that while TACE is an efficient proTGF-alpha N-terminal convertase, a different activity, distinguishable from TACE, exists that can process proTGF-alpha at the C-terminal site. A model that accounts for these findings and the requirement for TACE in TGF-alpha shedding is proposed.