Modulation of juxtamembrane cleavage ("shedding") of angiotensin-converting enzyme by stalk glycosylation: evidence for an alternative shedding protease.

The role of juxtamembrane stalk glycosylation in modulating stalk cleavage and shedding of membrane proteins remains unresolved, despite reports that proteins expressed in glycosylation-deficient cells undergo accelerated proteolysis. We have constructed stalk glycosylation mutants of angiotensin-converting enzyme (ACE), a type I ectoprotein that is vigorously shed when expressed in Chinese hamster ovary cells. Surprisingly, stalk glycosylation did not significantly inhibit release. Introduction of an N-linked glycan directly adjacent to the native stalk cleavage site resulted in a 13-residue, proximal displacement of the cleavage site, from the Arg-626/Ser-627 to the Phe-640/Leu-641 bond. Substitution of the wild-type stalk with a Ser-/Thr-rich sequence known to be heavily O-glycosylated produced a mutant (ACE-JGL) in which this chimeric stalk was partially O-glycosylated; incomplete glycosylation may have been due to membrane proximity. Relative to levels of cell-associated ACE-JGL, rates of basal, unstimulated release of ACE-JGL were enhanced compared with wild-type ACE. ACE-JGL was cleaved at an Ala/Thr bond, 14 residues from the membrane. Notably, phorbol ester stimulation and TAPI (a peptide hydroxamate) inhibition of release-universal characteristics of regulated ectodomain shedding-were significantly blunted for ACE-JGL, as was a formerly undescribed transient stimulation of ACE release by 3, 4-dichloroisocoumarin. These data indicate that (1) stalk glycosylation modulates but does not inhibit ectodomain shedding; and (2) a Ser-/Thr-rich, O-glycosylated stalk directs cleavage, at least in part, by an alternative shedding protease, which may resemble an activity recently described in TNF-alpha convertase null cells [Buxbaum, J. D., et al. (1998) J. Biol. Chem. 273, 27765-27767].

Phorbol ester-induced juxtamembrane cleavage of angiotensin-converting enzyme is not inhibited by a stalk containing intrachain disulfides.

Specialized proteases, referred to as sheddases, secretases, or membrane-protein-solubilizing proteases (MPSPs), solubilize the extracellular domains of diverse membrane proteins by catalyzing a specific cleavage in the juxtamembrane stalk regions of such proteins. A representative MPSP (tumor necrosis factor-alpha convertase) was cloned recently and shown to be a disintegrin metalloprotease that is inhibited by peptide hydroxamates including the compound TAPI. Substrate determinants that specify cleavage by MPSPs remain incompletely characterized, but may include the physicochemical properties of the stalk or unidentified recognition motifs in the stalk or the extracellular domain. We constructed a mutant angiotensin-converting enzyme (ACE) in which the stalk has been replaced with an epidermal growth factor (EGF)-like domain (ACE-JMEGF), to test the hypothesis that MPSP cleavage requires an open, comparatively unfolded or extended stalk. Wild-type ACE is a type I transmembrane (TM) ectoprotein that is efficiently solubilized by a typical MPSP activity. We found that ACE-JMEGF was solubilized inefficiently and accumulated in a cell-associated form on transfected Chinese hamster ovary (CHO) cells; cleavage was stimulated by phorbol ester and inhibited by TAPI, features typical of MPSP activity. Determination of the C-terminus of soluble ACE-JMEGF revealed that, surprisingly, cleavage occurred at a Gly-Phe bond between the fifth and sixth cysteines within the third disulfide loop of the EGF-like domain. Reduction of intact CHO cells with tributylphosphine resulted in the rapid release of ACE-JMEGF (but not wild-type ACE) into the medium, suggesting that a proportion of membrane-bound ACE-JMEGF is cleaved but remains cell-associated via disulfide tethering. The mechanism for the release of ACE-JMEGF in the absence of chemical reduction is unclear. We conclude that the presence of a compact, disulfide-bridged domain does not per se inhibit cleavage by an MPSP activity, but ectodomain release is prevented by disulfide tethering to the TM domain.

Proteolytic release of membrane proteins: studies on a membrane-protein-solubilizing activity in CHO cells.

Diverse membrane proteins are solubilized by a specific proteolytic cleavage in the stalk sequence adjacent to the membrane anchor, with release of the extracellular domain. Examples are the amyloid precursor protein, membrane-bound growth factors and angiotensin-converting enzyme (ACE). The identities and characteristics of the responsible proteases remain elusive. We have studied this process in Chinese hamster ovary (CHO) cells stably expressing wild-type ACE (WT-ACE) or juxtamembrane (stalk) deletion or chimaera mutants. Determination of the C termini (i.e. the cleavage sites) of released, soluble wild-type and mutant ACE by MALDI-TOF mass spectrometry indicated that the membrane-protein-solubilizing protease (MPSP) in CHO cells is not constrained by a particular cleavage site motif or by a specific distance from the membrane, but instead may position itself with respect to the putative proximal, folded extracellular domain adjacent to the stalk. Nevertheless, kinetic analyses of release rates indicated that a minimum distance from the membrane must be preserved. Interestingly, soluble full-length (anchor-plus) WT-ACE incubated with fractions of, or intact, CHO cells was not cleaved. In all cases, release was stimulated by a media change or by the addition of phorbol ester, with rate enhancements of 5- and 50-fold, respectively, for WT-ACE. The phorbol ester effect was abolished by staurosporine, a protein kinase C (PKC) inhibitor. We propose that the CHO cell MPSP that solubilizes ACE: (1) only cleaves proteins embedded in a membrane; (2) requires an accessible stalk and cleaves at a minimum distance from both the membrane and proximal extracellular domain; (3) positions itself primarily with respect to the proximal extracellular domain and (4) is regulated in part by a PKC-dependent mechanism.

Nonopsonic binding of Mycobacterium tuberculosis to human complement receptor type 3 expressed in Chinese hamster ovary cells.

Nonopsonic invasion of mononuclear phagocytes by Mycobacterium tuberculosis is likely important in the establishment of a primary infection in the lung. M. tuberculosis binds to a variety of phagocyte receptors, of which the mannose receptor and complement receptor type 3 (CR3) may support nonopsonic binding. CR3, a beta2 integrin, is a target for diverse intracellular pathogens, but its role in nonopsonic binding remains uncertain. We have examined the binding of M. tuberculosis H37Rv to human CR3 heterologously expressed in Chinese hamster ovary (CHO) cells, thereby circumventing the problems of competing receptors and endogenously synthesized complement, which are inherent in studies with mononuclear phagocytes. The surface expression of CD11b and CD18 was assessed by immunofluorescence, immunobead binding, flow cytometry, and immunoprecipitation with anti-CD11b and anti-CD18 monoclonal antibodies (MAbs). The functional activity of the surface-expressed CD11b/CD18 (CR3) heterodimer was confirmed by rosetting with C3bi-coated microspheres. We found that M. tuberculosis bound four- to fivefold more avidly to CR3-expressing CHO cells than to wild-type cells and, importantly, that this binding was at similar levels in the presence of fresh or heat-inactivated human or bovine serum or no serum. In contrast, Mycobacterium smegmatis bound poorly to CR3-expressing CHO cells in the absence of serum, but after opsonization in serum, binding was comparable to that of M. tuberculosis. The binding of M. tuberculosis to the transfected CHO cells was CR3 specific, as it was inhibited by anti-CR3 MAbs, particularly the anti-CD11b MAbs LM2/1 (I domain epitope) and OKM1 (C-terminal epitope), neither of which inhibit C3bi binding. MAb 2LPM19c, which recognizes the C3bi-binding site on CD11b, had little or no effect on M. tuberculosis binding. The converse was found for the binding of opsonized M. smegmatis, which was strongly inhibited by 2LPM19c but unaffected by LM2/1 or OKM1. CR3-specific binding was also evidenced by the failure of M. tuberculosis to bind to CHO cells transfected with an irrelevant surface protein (angiotensin-converting enzyme) in the presence or absence of serum. We conclude that the binding of M. tuberculosis H37Rv to CR3 expressed in CHO cells is predominantly nonopsonic and that the organism likely expresses a ligand that binds directly to CR3.

Proteolytic release of membrane-bound angiotensin-converting enzyme: role of the juxtamembrane stalk sequence.

Many structurally and functionally diverse membrane proteins are solubilized by a specific proteolytic cleavage in the stalk sequence adjacent to the membrane anchor, with release of the extracellular domain. Examples are the amyloid precursor protein, membrane-bound growth factors, and angiotensin-converting enzyme (ACE). The identities and characteristics of the responsible proteases remain elusive. We have studied this process in Chinese hamster ovary (CHO) cells stably expressing wild-type ACE (WT-ACE; human testis isozyme) or one of four juxtamembrane (stalk) mutants containing either deletions of 17, 24, and 47 residues (ACE-JM delta 17, -JM delta 24, and -JM delta 47, respectively) or a substitution of 26 stalk residues with a 20-residue sequence from the stalk of the low-density lipoprotein receptor (ACE-JMLDL). The C termini of released, soluble WT-ACE and ACE-JM delta 17 and -JMLDL were determined by MALDI-TOF mass spectrometry analyses of C-terminal peptides generated by CNBr cleavage. Observed masses of 4264 (WT-ACE) and 4269 (ACE-JM delta 17) are in good agreement with an expected mass of 4262 for the C-terminal CNBr peptide ending at Arg-627, indicating cleavage at the Arg-627/Ser-628 bond in both WT-ACE and ACE-JM delta 17, at distances of 24 and 10 residues from the membrane, respectively. Data for ACE-JM delta 24 are also consistent with cleavage at or near Arg-627. For ACE-JMLDL, in which the native cleavage site is absent, observed masses of 4372 and 4542 are in close agreement with expected masses of 4371 and 4542 for peptides ending at Ala-628 and Gly-630, respectively, indicating cleavages at 17 or 15 residues from the membrane. These data indicate that the membrane-protein-solubilizing protease (MPSP) in CHO cells is not constrained by a particular cleavage site motif or by a specific distance from the membrane but instead may position itself with respect to the putative proximal, folded extracellular domain adjacent to the stalk. Nevertheless, cleavage at a distance of 10 residues from the membrane is more favorable, as ACE-JM delta 17 is cleaved 12-fold faster than WT-ACE. In contrast, ACE-JM delta 24 is released 17-fold slower, suggesting that a minimum distance from the membrane must be preserved. This is supported by results with the ACE-JM delta 47 mutant, which is membrane-bound but not cleaved, likely because the entire stalk has been deleted. Finally, soluble full-length (anchor-plus) WT-ACE is not cleaved when incubated with various CHO cell fractions or intact CHO cells. On the basis of these and other data, we propose that the CHO cell MPSP that solubilizes ACE (1) only cleaves proteins embedded in a membrane; (2) requires an accessible stalk and cleaves at a minimum distance from both the membrane and proximal extracellular domain; (3) positions itself primarily with respect to the proximal extracellular domain; and (4) may have a weak preference for cleavage at Arg/Lys-X bonds.

Proteolytic release of human angiotensin-converting enzyme expressed in Chinese hamster ovary cells is enhanced by phorbol ester.

Membrane-bound angiotensin-converting enzyme (ACE) expressed in Chinese hamster ovary (CHO) cells is proteolytically released in soluble form into the medium. We find that this release is stimulated up to 50-fold by phorbol-12,13-dibutyrate and also by the addition of fresh, serum-containing media. Concomitant with the enhanced release is a marked decrease in levels of membrane-bound ACE, down to 7% of resting levels in the case of phorbol ester stimulation. Staurosporine, a protein kinase C (PKC) inhibitor, abolishes the phorbol ester effect. Kinetic analysis of the stimulated release rate indicates that it is first order, likely due to substrate depletion; calculated half times, t1/2, are 174 +/- 12 min and 40 +/- 6 min for the media-change and phorbol ester stimulated rates, respectively. Thus, release of membrane-bound ACE in CHO cells is regulated, in part, by a PKC-dependent mechanism.

Nucleotide sequence and analysis of the Vibrio alginolyticus sucrose uptake-encoding region.

The nucleotide sequence of the Vibrio alginolyticus sucrose uptake-encoding region was determined, and contained two genes, scrA and scrK. The scrA gene encodes an enzyme IISucrose (EIIScr) protein of the phosphoenolpyruvate dependent phosphotransferase system and the scrK gene encodes a fructokinase. The deduced amino acid (aa) sequence for the V. alginolyticus EIIScr protein was homologous with the EIIScr proteins from Streptococcus mutans, Salmonella typhimurium (pUR400 system) and Bacillus subtilis. The deduced aa sequence for the V. alginolyticus fructokinase was homologous with the Escherichia coli enzymes, 6-phosphofructokinase (isoenzyme 2) and ribokinase. Transposon phoA mutagenesis experiments indicated that the EIIScr protein was a membrane-bound protein with a region that extended into the periplasm.

Expression and regulation of a Bacteroides fragilis sucrose utilization system cloned in Escherichia coli.

A Bacteroides fragilis strain isolated from human feces was the source of chromosomal DNA in the construction of plasmid pBS100. The cloned 6-kilobase insert of plasmid pBS100 conferred a sucrose positivity phenotype on transformed cells of Escherichia coli JA221. E. coli JA221(pBS100) cells were able to utilize sucrose as the sole source of carbon because of the presence of sucrase enzyme and sucrose uptake activities. Sucrase activity was inducible in B. fragilis but constitutive in E. coli JA221(pBS100) cells. In sucrose-minimal medium, both B. fragilis and E. coli JA221(pBS100) produced intracellular and extracellular sucrase activities throughout the growth cycle. Osmotic shock experiments performed on E. coli JA221(pBS100) indicated that up to 55% of the sucrase activity was localized in the periplasmic space, 30% was in the cytoplasm, and the remaining 15% was in the cell-free extracellular supernatant fluid. B. fragilis and E. coli JA221(pBS100) actively transported sucrose. Sucrose uptake was induced by sucrose in B. fragilis, whereas the uptake activity in E. coli JA221(pBS100) was constitutive. E. coli JA221(pBS100) appeared to transport sucrose by a phosphotransferase-independent system. B. fragilis transported sucrose only under strictly anaerobic conditions. No uptake activity was detected under aerobic conditions with or without addition of catalase.

Nucleotide sequence and analysis of the Vibrio alginolyticus sucrase gene (scrB).

The nucleotide sequence of a 2.119-kb DNA fragment containing the Vibrio alginolyticus sucrase gene (scrB) was determined. The complete sequence (484 aa residues) of the sucrase was deduced and homology was detected between the sucrase enzymes from V. alginolyticus and the Gram-positive bacteria Bacillus subtilis and Streptococcus mutans. In Escherichia coli cells the cloned V. alginolyticus sucrase is translocated to the periplasm. Transposon phoA mutagenesis experiments strongly suggested that V. alginolyticus sucrase in E. coli is not exported across the cytoplasmic membrane by means of a typical signal sequence.

Expression and regulation of a Vibrio alginolyticus sucrose utilization system cloned in Escherichia coli.

A halotolerant collagenolytic Vibrio alginolyticus strain isolated from salted hides had intracellular sucrase activity and did not secret sucrase into the medium. The strain actively transported sucrose by a sucrose-inducible, Na+-independent process. A 10.4-kilobase DNA fragment of V. alginolyticus DNA was cloned into Escherichia coli. The recombinant E. coli(pVS100) could utilize sucrose as a sole carbon source. In contrast to V. alginolyticus, the recombinant E. coli produced both intra- and extracellular sucrase activities. Up to 20% of the total sucrase activity was in the supernatant. Sucrase synthesis in E. coli(pVS100) was inducible and was subject to glucose repression, which was relieved by cyclic AMP. Sucrose was actively transported by a sucrose-inducible, Na+-independent system in E. coli(pVS100). Sucrose uptake was inhibited by the addition of a proton conductor. The maximum velocity and apparent Km values of sucrose uptake for the V. alginolyticus strain and E. coli(pVS100) were 130 nmol/mg of protein per min and 50 microM and 6 nmol/mg of protein per min and 275 microM, respectively.