1070
Biochemistry 1995,34, 1070-1075
ADP-Ribosylation of a&20 by the S1 Subunit and Deletion Peptides of S1 of Pertussis Toxin? Viviane Finck-Barbanqon and Joseph T. Barbieri” Department of Microbiology, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, Wisconsin 53226 Received March 8, 1994; Revised Manuscript Received November 9, 1994@
PT (rS1) and two carboxyl-terminal deletion peptides, C180 and C204, which comprise the amino-terminal 180 and 204 amino acids of S1, respectively, were analyzed for the ability to ADP-ribosylate a13C20,a synthetic peptide composed of the 20 carboxyl-terminal amino acids of the a subunit of the heterotrimeric G protein G13. Under linear velocity conditions, C180 ADPribosylated a13C20 at a 3-fold higher rate than either C204 or rS1. At variable NAD, rS1, C204, and C 180 ADP-ribosylated aI3C2Owith similar initial velocities which followed Michaelis-Menten kinetics. In contrast, at variable aI3C2O,rS 1, C204, and C 180 ADP-ribosylated aoC20 with different initial velocities. At variable a13C2O,C204- and rS 1-catalyzed ADP-ribosylation followed Michaelis-Menten kinetics, while the velocity curve generated by C180 diverged from Michaelis-Menten kinetics. The rates of initial velocity of C 180 did not fit the Lineweaver-Burk equation, but could be transformed into the Hill equation which yielded a Hill coefficient of 2. This predicted that C180 possessed cooperativity between the two substrate binding sites. Other experiments showed that C180 ADP-ribosylated aI3C2Oat 60% of the rate for the ADP-ribosylation of Gt. These data showed that the entire catalytic mechanism for ADP-ribosylation resides within the first 180 amino acids of S1 and that the carboxyl-terminal 55 residues of S1 allow the ADP-ribosylation of aI3C2Oto proceed via Michaelis-Menten kinetics. These data along with earlier studies (Krueger & Barbieri, 1993) were also consistent with the presence of two Gt protein binding sites within S1. One G protein binding site was located within the amino-terminal 180 residues of S1 and interacted with a region of Gta defined by a13C20. The second G protein binding site was defined by residues 195-204 of S1 and was located within Gta but outside the a13C20 binding site. This second interaction provides a greater kcat and lower KmG, for the ADP-ribosylation of the heterotrimeric Gt by S 1. ABSTRACT: Recombinant S1 subunit of
Pertussis toxin (PT,M, 105 890) is a major virulence determinant of Bordetella pertussis, the causative agent of whooping cough (Weiss & Falkow, 1984). PT is a member of the family of bacterial ADP-ribosylating enzymes which catalyze the transfer of the ADP-ribose moiety of NAD to specific eucaryotic target proteins (Katada et al., 1982; Ui, 1990). PT is composed of 6 proteins, designated S1 through S5, which are noncovalently associated in a 1:1:1:2:1 ratio, respectively (Tamura et al., 1982). PT follows the “A:B” model for exotoxin structure-function (Tamura et al., 1982; Ui, 1990), where the “A” domain (Sl) contains the ADPribosyltransferase activity, and the “B” domain (S2 through S5) binds to the cell surface receptors to allow passage of S1 across the cell membrane. S1 catalyzes the ADPribosylation of a conserved cysteine residue located 4 amino acids from the carboxyl terminus of the a subunits of a subset of heterotrimeric GTP-binding proteins, Gi, Go, and G, (Van Dop et al., 1984; West et al., 1985; Hsia et al., 1985; Bimbaumer et al., 1990), which uncouples signal transduction between the G protein and its G protein coupled receptor (Ui, 1990; Conklin & Boume, 1993). The operon encoding the PT subunits has been cloned and sequenced (Locht et a1.,1986; Nicosia et al., 1986), and the
genes encoding the individual subunits of PT have been expressed in Escherichia coli (Barbieri et al., 1987; Burnett et al., 1988; Locht et al., 1987; Nicosia et al., 1987). Biochemical studies showed that recombinant S 1 (rS 1) possessed similar kinetic properties as S1 (235 amino acids) purified from pertussis toxin (Barbieri et al., 1990). Analysis of deletion peptides of S 1 identified regions of S 1 that were responsible for catalysis. The C180 peptide which comprised residues 1- 180 catalyzed the NAD glycohydrolase reaction at the same rate as S 1, but ADP-ribosylated the heterotrimeric G protein, Gt, at 1% of the rate of S1 (Cortina & Barbieri, 1991). Kinetic studies showed that C180 possessed a higher KmGt and a lower k,, than S1 (Cortina et al., 1991). These studies also showed that S1 ADP-ribosylated Gt at a 17fold higher rate than the isolated Gta subunit, while C180 ADP-ribosylated Gt at a rate that was less than 2-fold higher than the isolated G,, subunit. These data indicated that the high affinity of rS1 for heterotrimeric Gt was due, either directly or indirectly, to interactions of the carboxyl-terminal 55 amino acid residues of S1 with G Q ~ Analysis . of deletion peptides of S1 showed that residues between 195 and 204 were responsible for the high affinity binding of S1 to G, (Cortina et al., 1991; Krueger & Barbieri, 1994).
This research was supported by Public Health Grants AI-30162 and AI-01087 (J.T.B.) from the National Institutes of Health. V.F.-B. is supported by a fellowship from the International Human Frontier Science Program. * Phone (414) 456-8412; fax (414) 266-8522. Abstract published in Advance ACS Abstracts, December 15, 1994.
Moss and co-workers (Watkins et al., 1985) reported that tryptic peptides of Gta were not ADP-ribosylated by PT, which suggested that PT recognized the conformation of Gta. Recently, Bimbaumer and co-workers (Graf et al., 1992) showed that peptides comprising the 10-20 carboxyl-
@
0006-2960/95/0434-1070$09.00/0 0 1995 American Chemical Society
ADP-Ribosylation of ai3C20 by S 1 and S 1 Deletion Peptides terminal residues of the a subunit of PT-sensitive G proteins, including Gi, were substrates for ADP-ribosylation by PT. In the present study, we measure the ability of S1 and two deletion peptides to ADP-ribosylate ai3C20, a synthetic peptide composed of the 20 carboxyl-terminal amino acids of the a subunit of the heterotrimeric G protein Gi3. These studies allowed the localization of several functions within the carboxyl-terminal 55 amino acids of S 1.
MATERIALS AND METHODS Materials
Construction of ptacslc, ptacC180, and ptdlC204 has been described previously (Barbieri et al., 1989; Krueger & Barbieri, 1994). [adenylate p h o ~ p h a t e - ~ ~ P ] Nwas A D purchased from Du Pont-New England Nuclear. Protein molecular mass markers were purchased from PharmaciaLKB Biotechnology Inc. Analysis of kinetic data was performed with the software "Enzfitter" (Elsevier-Biosoft, Cambridge, U.K.). Transducin (G,) was purified from bovine retina (J. A. Lawson Co., Lincoln, NE) as previously described (Cortina & Barbieri, 1991). Methods Synthesis of ai3C20Peptide. The 20 amino acid peptide
NH2-VFDAVTDVIIKNNLKECGLY-COOH(termed
~
3
C20) was synthesized at the Shared Protein-Nucleic Acid Facility at the Medical College of Wisconsin. The composition of ai3C20 was confirmed by amino acid and mass spectroscopic analysis. HPLC analysis showed the purity of the peptide to be 72%. The ai3C20 peptide was soluble in 0.1 N W O H and stored at -20 "C. PuriJication of rSI, Cl80, and C204. (A) rS1. Purification of rS 1 from E. coli has been described (Xu et al., 1994) with the indicated modifications. Overnight cultures of E. coli (ptacSlc) were diluted 1/50 into 2.4 L of 2 x L-Broth supplemented with 100 p g of ampicillidml and 0.5 mM ZnS04 and incubated for 6 h at 30 "C at 250 rpm. Cells were concentrated, washed, suspended in 20 mL of 25 mM Tris-HC1 (pH 7.6), and broken by two passes through a French press. French-pressed cells were centrifuged (3000g) for 5 min to pellet unbroken cells and then centrifuged at 45000g for 35 min to pellet insoluble material. This insoluble material was suspended in 20 mL of 25 mM TrisHCl (pH 7.6), 100 mM NaC1, and 2% Triton X-100 and centrifuged at 45000g for 35 min. This insoluble material was suspended in 25 mM Tris-HC1 (pH 7.6) and 1 M NaCl and centrifuged for 30 min at 45000g. The insoluble material was then suspended in 25 mM Tris (pH 7.6) containing 8 M urea overnight at 4 "C and centrifuged at 163000g for 2 h; the soluble fraction was chromatographed over a S-200 HR Sephacryl in 25 mM Tris-HC1 (pH 7.6) containing 4 M urea. Column fractions containing rS 1 were pooled and concentrated by ammonium sulfate precipitation (50% final concentration). Prior to analysis, an aliquot of rS1, stored as an ammonium sulfate precipitate, was pelleted by centrifugation and resuspended in 25 mM Tris-HC1 (pH 7.6) containing 1 M urea. (B) C180. Purification of the C180 peptide was as described previously (Barbieri et al., 1989). Briefly, 6 L of E. coli (ptacC180) was cultured for 6-8 h in the presence of 100 pg of ampicillidml and 0.5 mM ZnS04. The C180
-
Biochemistry, Vol. 34, No. 3, 1995 1071
peptide was purified from a periplasmic fraction via conventional chromatography and stored as an ammonium sulfate precipitate. (C) C204. Purification of C204 peptide has been described (Krueger & Barbieri, 1994). Briefly, 2.4 L of E. coli (ptdlC204) was grown for 6 h (ODm
