Quantitative assessment of enzyme specificity in vivo: P2 recognition by Kex2 protease defined in a genetic system

Abstract

The specificity of the yeast proprotein-processing Kex2 protease was examined in vivo by using a sensitive, quantitative assay. A truncated prepro-alpha-factor gene encoding an alpha-factor precursor with a single alpha-factor repeat was constructed with restriction sites for cassette mutagenesis flanking the single Kex2 cleavage site (-SLDKR downward arrowEAEA-). All of the 19 substitutions for the Lys (P2) residue in the cleavage site were made. The wild-type and mutant precursors were expressed in a yeast strain lacking the chromosomal genes encoding Kex2 and prepro-alpha-factor. Cleavage of the 20 sites by Kex2, expressed at the wild-type level, was assessed by using a quantitative-mating assay with an effective range greater than six orders of magnitude. All substitutions for Lys at P2 decreased mating, from 2-fold for Arg to >10(6)-fold for Trp. Eviction of the Kex2-encoding plasmid indicated that cleavage of mutant sites by other cellular proteases was not a complicating factor. Mating efficiencies of strains expressing the mutant precursors correlated well with the specificity (kcat/KM) of purified Kex2 for comparable model peptide substrates, validating the in vivo approach as a quantitative method. The results support the conclusion that KM, which is heavily influenced by the nature of the P2 residue, is a major determinant of cleavage efficiency in vivo. P2 preference followed the rank order: Lys > Arg > Thr > Pro > Glu > Ile > Ser > Ala > Asn > Val > Cys > AsP > Gln > Gly > His > Met > Leu > Tyr > Phe > Trp.

Figure 1

Schematic structures of α-factor precursors and overview of the genetic system. (A) Signal peptidase cleavage of the major form of prepro-α-factor (165 residues), which is encoded by the MFα1 gene, results in production of 146 residue pro-α-factor (18). Kex2 protease cleaves pro-α-factor between the glycosylated “pro” domain and the first α-factor repeat carboxyl to the sequence Ser-Leu-Asp-Lys-Arg↓ and at three sites between the α-factor repeats carboxyl to the sequence Pro-Met-Tyr-Lys-Arg↓ (19). Production of mature α-factor requires exoproteolysis by Kex1 carboxypeptidase to remove C-terminal Lys and Arg residues from the internal repeats (20) and by Ste13 dipeptidyl aminopeptidase to remove N-terminal GluAla and AspAla dipeptides from each repeat (21). (B) The product of the MFα1–100 gene contains a single α-factor repeat. Production of mature α-factor requires cleavage by Kex2 at a single site. (C) Schematic of yeast strain ABY08 carrying plasmids YCpG-MFα1–100 and pAL5-KEX2. The Xba I and HindIII sites in YCpG-MFα1–100 were used for cassette mutagenesis of the P₂ Lys (Lys⁸⁴). An important feature of this system is that cleavage of both wild-type and mutant sites must occur in the correct “phase,” i.e., immediately following the P₁ Arg. This is because production of mature α-factor requires removal of two Glu-Ala dipeptides by the Ste13 peptidase.

Figure 2

(A) Plot of relative mating efficiency vs. surface area of the P₂ side chain. Total solvent-accessible side chain surface areas for all residues except Gly were as described by Creighton (39). For Gly, total solvent-accessible side chain surface was estimated as half the surface area of a hydrogen atom. (B) Plot of relative mating efficiency vs. accessible nonpolar surface area. Total solvent-accessible nonpolar side chain surface areas for all residues except Gly were as described by Creighton (39). For Gly, total solvent accessible nonpolar side chain surface was estimated as one-half the surface area of a hydrogen atom.

Figure 3

Expression of mutant α-factor precursors. Yeast strain ABY08 was transformed with pBM258 (control plasmid or “−”), YCpG-MFα1–100 (K), and with plasmids carrying P₂-substituted mutant alleles of MFα1–100 indicated using the single letter amino acid code. The amber termination codon in place of Lys84 is labeled “amb.” Enzymatically deglycosylated proteins from culture media were separated by SDS/PAGE, blotted, and probed with anti-(pro-α-factor) antiserum as described in Materials and Methods. Mfα1–120p with Cys at P₂ formed an intermolecular disulfide and was only partially digested (shown) with Endoglycosidase H unless reduced with β-mercaptoethanol before deglycosylation (data not shown).

Figure 4

In vitro vs. in vivo cleavage efficiency. (A) Peptidyl-MCA substrates. Relative efficiency of the peptidyl-MCA substrates is correlated with the in vivo-mating efficiency. Relative cleavage efficiency is defined as the ratio of k_cat/K_M for a peptidyl-MCA substrate with the indicated residue at P₂ to the k_cat/K_M for a peptidyl-MCA substrate with Lys at P₂. Data for k_cat/K_M of purified secreted, soluble Kex2 protease with peptidyl-MCA substrates are from Brenner and Fuller (7), Rockwell, et al. (9), Rockwell and Fuller (10), and N. Rockwell and R.S.F. (unpublished results). P₂ residues of substrates compared are indicated in the single letter code (β represents nor-leucine). A substrate with nor-leucine at P₂ was compared with mating efficiency with Met at P₂ in vivo (YCpG-MFα1–110). In certain cases (Arg, Thr, Ala, and Gly), two synthetic substrates were available for comparison. The dashed line represents a line fit for direct proportionality between log (k_cat/K_M) and the log of the mating efficiency (correlation coefficient, r = 0.886; slope = 0.75). (B) Internally quenched fluorogenic peptide substrates. A correlation plot of relative efficiency of cleavage of the internally quenched fluorogenic peptide substrates (k_cat/K_M) v. relative mating efficiency for corresponding P₂ substitutions. Relative cleavage efficiency is defined as the ratio of k_cat/K_M for an internally quenched fluorogenic peptide substrate with the indicated residue at P₂ to the k_cat/K_M for an internally quenched fluorogenic peptide substrate with Lys at P₂. Data for k_cat/K_M of purified secreted, soluble Kex2 protease with internally quenched fluorogenic peptide substrates are from ref. . P₂ residues of substrates compared are indicated in the single letter code. The internally quenched fluorogenic peptides were of the sequence Arg-GlE-norLeu-Tyr-Xaa-Arg-↓-Glu-Ala-Glu-Ala-LyD-Arg, in which “GlE” represents glutamyl-EDANS, [EDANS, 5-(2-aminoethyl-amino)-naphthalene-1-sulfonic acid], “Xaa” represents the particular P₂ substitution, and “LyD” represents lysyl-DABSYL (DABCYL, 4-(4-dimethylaminophenyl)-azobenzoyl) (41, 42). This sequence was based on the Kex2 cleavage site between the α-factor repeats in pro-α-factor. A substrate with nor-leucine (X) at P₂ was compared with mating efficiency with Met at P₂ in vivo (YCpG-MFα1–110).