Single Amino Acid Substitution Mutants Research Articles

Functional Analysis of the Propeptide of Subtilisin E as an Intramolecular Chaperone for Protein Folding: REFOLDING AND INHIBITORY ABILITIES OF PROPEPTIDE MUTANTS

The amino-terminal propeptide, consisting of 77 amino acid residues, is known to be required as an intramolecular chaperone to guide the folding of mature subtilisin E, a serine protease, into active mature enzyme. Many mutations within the pro-sequence have been shown to abolish the production of active subtilisin E (Kobayashi, T., and Inouye, M. (1992) J. Mol. Biol. 226, 931-933). Here we report characterization, refolding, and inhibitory abilities of six single amino acid substitution mutations (Ile-67-->Val, Ile-48-->Thr, Gly-44-->Asp, Lys-36-->Glu, Ala-30-->Thr, and Pro-15-->Leu) and a nonsense mutation (N59-mer) at the codon for Lys-18. These mutant propeptides were expressed in Escherichia coli using a T7 expression system and were purified to homogeneity. Surprisingly, Lys-36-->Glu, Ala-30-->Thr and Pro-15-->Leu were found to still function as a chaperone for in vitro refolding of denatured subtilisin BPN' with 60, 80, and 54% efficiency compared to the wild-type propeptide, respectively. The Ki values against subtilisin BPN' were 1.6 x 10(-9) M, and 2.1 x 10(-9) M, respectively. The Ki values against subtilisin BPN' were 1.6 x 10(-9) M, and 2.1 x 10(-9) M, respectively, almost identical to the Ki value exhibited by the wild-type propeptide (1.4 x 10(-9) M). In contrast, Ile-67-->Val and Gly-44-->Asp were able to refold denatured subtilisin BPN' with only 18 and13% efficiencies and had Ki values of 10 and 11 x 10(-9) M, respectively. The Ile-48-->Thr mutant propeptide was unable to refold denatured subtilisin BPN' and gave a 100-fold higher Ki (118 x 10(-9) M) than the wild-type propeptide. The N59-mer propeptide extending from Leu-19 to Met-78 was unable to function as a chaperone. Like the wild-type propeptide, none of the mutant propeptides had secondary structures as judged by their circular dichroism spectra. The present results demonstrate that the ability of the propeptide as a chaperone to refold the denatured protein is well correlated with its ability as a competitive inhibitor for the active enzyme. This supports the notion that the secondary and tertiary structures of the propeptide are identical or highly homologous between the renatured propeptide-subtilisin complex and the inhibitory complex formed between the propeptide and the active enzyme.

Genetic basis of bacteriophage HK97 prohead assembly.

We report studies to determine which bacteriophage genes are required for assembly of phage HK97 proheads and what roles they play. We identify the gene encoding the major capsid protein of phage HK97 and report its DNA sequence, together with the DNA sequences of the two genes immediately upstream from it. When the capsid protein is expressed from a plasmid in the absence of other phage-encoded proteins, it assembles, with good efficiency and accuracy into prohead-like structures composed of the unprocessed 42 kDa capsid protein. No separately encoded scaffolding protein is required for this assembly. If the 25 kDa product of the next gene upstream is co-expressed with the capsid protein, the prohead structures that are produced undergo the normal morphogenetic cleavage, which removes 102 amino acids from the N terminus of each subunit, leaving 31 kDa subunits. The 25 kDa protein is therefore probably a phage-encoded protease. The third gene, upstream from the protease gene, encodes the portal protein. Presence of the portal protein is not required for assembly of the capsid protein. Analysis of the phenotypes of four single amino acid-substitution mutants in the capsid-protein gene leads to several insights into the functions of the capsid protein and its interactions with the putative protease.

Genetic basis of bacteriophage HK97 prohead assembly

We report studies to determine which bacteriophage genes are required for assembly of phage HK97 proheads and what roles they play. We identify the gene encoding the major capsid protein of phage HK97 and report its DNA sequence, together with the DNA sequences of the two genes immediately upstream from it. When the capsid protein is expressed from a plasmid in the absence of other phage-encoded proteins, it assembles, with good efficiency and accuracy into prohead-like structures composed of the unprocessed 42 kDa capsid protein. No separately encoded scaffolding protein is required for this assembly. If the 25 kDa product of the next gene upstream is co-expressed with the capsid protein, the prohead structures that are produced undergo the normal morphogenetic cleavage, which removes 102 amino acids from the N terminus of each subunit, leaving 31 kDa subunits. The 25 kDa protein is therefore probably a phage-encoded protease. The third gene, upstream from the protease gene, encodes the portal protein. Presence of the portal protein is not required for assembly of the capsid protein. Analysis of the phenotypes of four single amino acid-substitution mutants in the capsid-protein gene leads to several insights into the functions of the capsid protein and its interactions with the putative protease.

Use of high and low level overexpression plasmids to test mutant alleles of the recF gene of Escherichia coli K-12 for partial activity.

We showed that sufficient overexpression of the wild-type recF gene interfered with three normal cell functions: (1) UV induction of transcription from the LexA-protein-repressed sulA promoter, (2) UV resistance and (3) cell viability at 42 degrees. To show this, we altered a low-level overexpressing recF+ plasmid with a set of structurally neutral mutations that increased the rate of expression of recF. The resulting high-level overexpressing plasmid interfered with UV induction of the sulA promoter, as did the low-level overexpressing plasmid. It also reduced UV resistance more than its low level progenitor and decreased viability at 42 degrees, an effect not seen with the low-level plasmid. We used the high-level plasmid to test four recF structural mutations for residual activity. The structural alleles consisted of an insertion mutation, two single amino acid substitution mutations and a double amino acid substitution mutation. On the Escherichia coli chromosome the three substitution mutations acted similarly to a recF deletion in reducing UV resistance in a recB21 recC22 sbcB15 sbcC201 genetic background. By this test, therefore, all three appeared to be null alleles. Measurements of conjugational recombination revealed, however, that the three substitution mutations may have residual activity. On the high-level overexpressing plasmid all three substitution mutations definitely showed partial activity. By contrast, the insertion mutation on the high-level overexpressing plasmid showed no partial activity and can be considered a true null mutation. One of the substitutions, recF143, showed a property attributable to a leaky mutation. Another substitution, recF4101, may block selectively two of the three interference phenotypes, thus allowing us to infer a mechanism for them.

Energetics of arginine-4 substitution mutants in the N-terminal cooperativity domain of T4 gene 32 protein.

Gene 32 protein (gp32) from bacteriophage T4 is a sequence-nonspecific single-strand (ss) nucleic acid binding protein which binds highly cooperatively to ss nucleic acids. The N-terminal "B" or basic domain (residues 1-21) is known to be required for highly cooperative binding by gp32 (where K(app) = K(int) omega, omega > or = 500), since its removal results in a protein which binds ss nucleic acids noncooperatively (omega = 1). In this paper, we probe the molecular details of cooperative binding by gp32 by physicochemical characterization of a set of four single amino acid substitution mutants of Arg4: Lys4 (R4K gp32), Gln4 (R4Q gp32), Thr4 (R4T gp32), and Gly4 (R4G gp32). The qualitative ranking of binding affinities to poly(A) is wild-type > or = R4K > R4Q > R4T > R4G > gp32-B (gp32 lacking the first 21 amino acids). The occluded site size is n(app) = 7.5 +/- 0.5 for all gp32s. Resolution of K(int) and omega for wild-type, R4K, R4Q, and R4T gp32s was estimated under conditions of low lattice saturation (v < or = 0.011) using multiple reverse fluorescence titrations collected at 10 mM Tris-HCl, pH 8.1, 20 degrees C, and a NaCl concentration where K(app) was (2-4) x 10(6) M-1 for each gp32 on the ribohomopolymer poly(A). Binding parameters for all gp32s were obtained directly or compared by conservative extrapolation of the [NaCl] dependence of K(app) to 0.20 M NaCl, 20 degrees C, pH 8.1. The magnitude of omega was then assumed not to vary with [NaCl] (shown for R4T gp32), allowing estimation of K(int) at 0.20 M NaCl. We find that R4K gp32 binds to poly(A) with an overall affinity (K(app)) which is 2-3-fold lower than wild-type gp32, while omega for each molecule seems indistinguishable (wild-type gp32, omega approximately 800-1300; R4K gp32, omega approximately 600-1200). Surprisingly, R4Q gp32 is characterized by an omega also not readily distinguishable from the wild-type and R4K proteins (omega approximately 800-4400), while K(app) is reduced about 10-fold. This mutant also shows a significantly reduced [NaCl] dependence of the binding to poly(A). R4T gp32 binds about 10-fold weaker than the Q mutant. It exhibits an omega ranging from 300 to 700 and a substantially reduced [NaCl] dependence (delta log K(int)/delta log [NaCl] = -1.4 from 0.10 to 0.20 M NaCl), indicative of significant perturbations in both K(int) and omega terms.(ABSTRACT TRUNCATED AT 400 WORDS)

Point mutagenesis of carboxyl-terminal amino acids of cholesteryl ester transfer protein. Opposite faces of an amphipathic helix important for cholesteryl ester transfer or for binding neutralizing antibody.

The cholesteryl ester transfer protein (CETP) mediates the transfer of neutral lipids between the plasma lipoproteins. A carboxyl-terminal sequence of CETP was recently shown to form the epitope of a neutralizing monoclonal antibody (TP2) and to be necessary for neutral lipid transfer activity. To determine the role of specific amino acids in the epitope and/or in lipid transfer activity, we made single amino acid substitution mutants between Pro446 and Ser476 by in vitro mutagenesis and expressed the mutants in mammalian cells. The binding of TP2 to CETP was abolished by mutations primarily of polar or charged amino acids that occurred periodically within the sequence between His466 and Leu475 and at amino acid Asp460; however, these mutants had well preserved cholesteryl ester (CE) transfer activity. By contrast, mutants of bulky hydrophobic amino acids in this region (particularly Leu475, Phe471, Leu468, Phe461, and Phe454) showed markedly decreased CE transfer specific activity, but essentially normal binding of TP2. The paradoxical effects on antibody binding and activity could be explained if amino acids determining monoclonal antibody binding and activity are disposed on opposite faces of an amphipathic helix between amino acids 465 and 476. This model was tested by substituting alanine residues for pairs of nonpolar amino acids which would be contiguous on a helix, resulting in low activity mutants equivalent to those produced by deletion of this region. We conclude that the hydrophobic face of a carboxyl-terminal helix of CETP is directly involved in the mechanism of CE transfer, and that TP2 inhibits activity by local sterical hindrance. The general hydrophobic character of this region, imparted by the bulky hydrophobic amino acids Leu and Phe, is important for normal CE transfer.

Simultaneous gain and loss of functions caused by a single amino acid substitution in the beta subunit of Escherichia coli RNA polymerase: suppression of nusA and rho mutations and conditional lethality.

Transcript elongation and termination in Escherichia coli is modulated, in part, by the nusA gene product, an acidic protein that interacts not only with RNA polymerase itself but also with ancillary factors, namely the host termination protein Rho and phage lambda antitermination protein, N. The E. coli nusA1 mutant fails to support lambda development due to a specific defect in N-mediated antitermination. Certain rifampicin-resistant (rifR) variants of the nusA1 host support lambda growth. We report here the isolation and pleiotropic properties of one such rifR mutant, ts8, resulting from a single amino acid substitution mutation in rpoB, the structural gene for polymerase beta subunit. ts8 is a recessive lethal mutation that blocks cell growth at 42 degrees. Pulse-labeling and analysis of newly synthesized proteins indicate that the mutant cell is proficient in RNA synthesis at high temperature. Apparently, ts8 causes a loss of some specialized function of RNA polymerase without a gross defect in general transcription activities. ts8 is an allele-specific suppressor of nusA1. It does not suppress nusAsal, nusB5 and nusE71 mutations nor does it bypass the requirement for a functional N gene and the nut site for antitermination and lambda growth. A mutation in the N gene, punA1, that restores lambda growth in the nusA1 mutant host but not in the nusAsal host, compensates for the nusAsal allele in the ts8 mutant. This combined effect of two allele-specific suppressors suggests that they enhance some aspect of polymerase-NusA-N interaction and function. ts8 suppresses the rho15 mutation, but not the rho112 mutation, indicating that it might render RNA polymerase susceptible to the action of a defective Rho protein. Marker rescue analysis has localized ts8 to a 910-bp internal segment of rpoB that encodes the Rif domain. By amplification, cloning and sequencing of this segment of the mutant chromosome we have determined that ts8 contains Phe in place of Ser522, caused by a C to T transition. By gene conversion, we have established that the simultaneous gain and loss of three functions of polymerase is caused by this single amino acid substitution. Clearly, a site in the beta subunit critical for the functioning of both termination and antitermination factors is altered by ts8. The alteration, we imagine, might make this site on polymerase receptive to some factors but repulsive to others.

Phenotypic selection and characterization of mutant alleles of a eukaryotic DNA topoisomerase I.

We have developed a simple, effective genetic screen for mutant alleles of eukaryotic DNA topoisomerase I that manifest severely depressed or complete loss of enzymatic function. The screen is based on the extreme toxicity of vaccinia topoisomerase expression in the Escherichia coli lysogen strain BL21(DE3) and is notable for its ease in distinguishing nonsense mutations (that result in truncated proteins) from missense mutations. The power of the method is evinced by our observation that 100% of the candidate alleles identified in the screen were ultimately found to have single-base changes at the DNA level that result in amino acid substitutions at the protein level. By mutagenizing plasmid DNA in vitro with hydroxylamine and applying this phenotypic screen, we have isolated five distinct single amino acid substitution mutants, each of which shows a biochemical phenotype, that is, greater than or equal to 90% reduction in specific DNA relaxing activity of the mutant protein relative to wild type. The amino acids thus implicated in topoisomerase function have identical or related counterparts at homologous positions in the topoisomerases from yeast and man. The same genetic screen has been applied to the selection of temperature-sensitive alleles of the vaccinia topoisomerase, leading to the isolation of two additional single-hit mutant alleles that display a temperature-sensitive growth phenotype in E. coli BL21(DE3). By broadening our mutagenesis procedures, we expect to generate a comprehensive map of vaccinia topoisomerase function and primary protein structure that should have direct application to eukaryotic cellular enzymes. Our methodology should be applicable to the selection of missense and conditional mutant alleles in other genes whose expression in bacteria is toxic.

Multiple functional domains of Tat, thetrans-activator of HIV-1, defined by mutational analysis

The tat gene of HIV-1 is a potent trans-activator of gene expression from the HIV long terminal repeat (LTR). To define the functionally important regions of the product of the tat gene (Tat) of HIV-1, deletion, linker insertion and single amino acid substitution mutants within the Tat coding region of strain SF2 were constructed. The effect of these mutations on trans-activation was assessed by measuring the expression of the bacterial chloramphenicol acetyltransferase (CAT) reporter gene linked to the HIV-LTR. These studies have revealed that four different domains of the protein that map within the N-terminal 56 amino acid region are essential for Tat function. In addition to the essential domains, an auxiliary domain that enhances the activity of the essential region has also been mapped between amino acid residues 58 and 66. One of the essential domains maps in the N-terminal 20 amino acid region. The other three essential domains are highly conserved among the various strains of HIV-1 and HIV-2 as well as simian immunodeficiency virus (SIV). Of the conserved domains, one contains seven Cys residues and single amino acid substitutions for several Cys residues indicate that they are essential for Tat function. The second conserved domain contains a Lys X Leu Gly Ile X Tyr motif in which the Lys residue is essential for trans-activation and the other residues are partially essential. The third conserved domain is strongly basic and appears to play a dual role. Mutants lacking this domain are deficient in trans-activation and in efficient targeting of Tat to the nucleus and nucleolus. The combination of the four essential domains and the auxiliary domain contribute to the near full activity observed with the 101 amino acid Tat protein.

Isolation and preliminary characterization of single amino acid substitution mutants of aspartate carbamoyltransferase.

In order to isolate functional Escherichia coli aspartate carbamoyltransferase (carbamoylphosphate:L-aspartate carbamoyltransferase, EC2.1.3.2) with single amino acid replacements, a series of pyrB nonsense mutants has been isolated. These nonsense mutants were induced by 2-aminopurine mutagenesis and selected by a combination of antibiotic treatments, direct enzyme assays, and suppressibility tests. Suppression of the pyrB nonsense mutation with various suppressors, which insert different amino acids, has resulted in the formation of a series of mutant aspartate carbamoyltransferases, each differing in one amino acid from the wild-type enzyme. After partial purification, kinetic studies revealed that some of the mutant enzymes had altered homotropic and heterotropic interactions. The mutants that had a tyrosine insert showed the most pronounced changes, followed by those with a serine insert. The mutants having a glutamine insert, howevr, were indistinguishable from the wild-type enzyme, supporting the conclusion that, because of the specificity of the mutagen, the glutamine insert had regenerated the wild-type enzyme.