Orthogonal combinatorial mutagenesis: a codon-level combinatorial mutagenesis method useful for low multiplicity and amino acid-scanning protocols.

We describe here a method to generate combinatorial libraries of oligonucleotides mutated at the codon-level, with control of the mutagenesis rate so as to create predictable binomial distributions of mutants. The method allows enrichment of the libraries with single, double or larger multiplicity of amino acid replacements by appropriate choice of the mutagenesis rate, depending on the concentration of synthetic precursors. The method makes use of two sets of deoxynucleoside-phosphoramidites bearing orthogonal protecting groups [4,4'-dimethoxytrityl (DMT) and 9-fluorenylmethoxycarbonyl (Fmoc)] in the 5' hydroxyl. These phosphoramidites are divergently combined during automated synthesis in such a way that wild-type codons are assembled with commercial DMT-deoxynucleoside-methyl-phosphoramidites while mutant codons are assembled with Fmoc-deoxynucleoside-methyl-phosphoramidites in an NNG/C fashion in a single synthesis column. This method is easily automated and suitable for low mutagenesis rates and large windows, such as those required for directed evolution and alanine scanning. Through the assembly of three oligonucleotide libraries at different mutagenesis rates, followed by cloning at the polylinker region of plasmid pUC18 and sequencing of 129 clones, we concluded that the method performs essentially as intended.

Temperature-sensitive mutants of the EcoRI endonuclease.

The EcoRI endonuclease is an important recombinant DNA tool and a paradigm of sequence-specific DNA-protein interactions. We have isolated temperature-sensitive (TS) EcoRI endonuclease mutants (R56Q, G78D, P90S, V97I, R105K, M157I, C218Y, A235E, M255I, T261I and L263F) and characterized activity in vivo and in vitro. Although the majority were TS for function in vivo, all of the mutant enzymes were stably expressed and largely soluble at both 30 degrees C and 42 degrees C in vivo and none of the mutants was found to be TS in vitro. These findings suggest that these mutations may affect folding of the enzyme at elevated temperature in vivo. Both non-conservative and conservative substitutions occurred but were not correlated with severity of the mutation. Of the 12 residues identified, 11 are conserved between EcoRI and the isoschizomer RsrI (which shares 50% identity), a further indication that these residues are critical for EcoRI structure and function. Inspection of the 2.8 A resolution X-ray crystal structure of the wild-type EcoRI endonuclease-DNA complex revealed that: (1) the TS mutations cluster in one half of the globular enzyme; (2) several of the substituted residues interact with each other; (3) most mutations would be predicted to disrupt local structures; (4) two mutations may affect the dimer interface (G78D and A235E); (5) one mutation (P90S) occurred in a residue that is part of, or immediately adjacent to, the EcoRI active site and which is conserved in the distantly related EcoRV endonuclease. Finally, one class of mutants restricted phage in vivo and was active in vitro, whereas a second class did not restrict and was inactive in vitro. The two classes of mutants may differ in kinetic properties or cleavage mechanism. In summary, these mutations provide insights into EcoRI structure and function, and complement previous genetic, biochemical, and structural analyses.

Improving random mutagenesis by purification of the oligonucleotide variants.

An oligonucleotide-based mutagenesis method is presented where, contrary to most classical mutagenic approaches, preselection of the variants is performed at the oligonucleotide level to avoid cloning of non-desired sequences. The method relies on the generation of differentially phosphate-protected oligonucleotides. Protection of the phosphates is accomplished by substoichiometric incorporation of an Fmoc-protected and n-propyl-protected trinucleotide phosphoramidite during ordinary oligonucleotide assembly. Instead of the alkali-labile beta-cyanoethyl group introduced in ordinary assembly, the trinucleotide introduces the alkali-stable n-propyl group. As a result, single mutants carry three ionic phosphates less than the wild-type sequence, double mutants carry six ionic phosphates less and so on. This difference in ionic ratio enables separation of the variants by conventional polyacrilamide gel electrophoresis. In the exemplified library described herein, two sub-populations containing mainly triple and quadruple mutants were selected out of five possible sub-populations.

Combination of DMT-mononucleotide and Fmoc-trinucleotide phosphoramidites in oligonucleotide synthesis affords an automatable codon-level mutagenesis method.

BACKGROUND: Synthetic DNA has been used to introduce variability into protein-coding regions. In protocols that produce a few mutations per gene, the sampling of amino-acid sequence space is limited by the bias imposed by the genetic code. It has long been apparent that the incorporation of trinucleotides in the synthetic regime would circumvent this problem and significantly enhance the usefulness of the technique. RESULTS: A new method is described for the creation of codon-level degenerate oligodeoxyribonucleotides that combines conventional dimethoxytrityl (DMT) mononucleoside phosphoramidite chemistry with 9-fluorenylmethoxycarbonyl (Fmoc) trinucleotide phosphoramidites (whose synthesis is reported in the paper). The substoichiometric use of these Fmoc-trinucleotides in an automatable, solid-phase synthesis procedure afforded DNA fragments comprising the wild-type sequence and a controllable distribution of mutants within two- and three-codon stretches of DNA, within the multiple cloning site of the conventional cloning vector pUC19. CONCLUSIONS: DMT and Fmoc are compatible protecting groups in conventional oligonucleotide synthesis methods, resulting in controllable levels of codon-based mutagenesis.

A proposed architecture for the central domain of the bacterial enhancer-binding proteins based on secondary structure prediction and fold recognition.

The expression of genes transcribed by the RNA polymerase with the alternative sigma factor sigma 54 (E sigma 54) is absolutely dependent on activator proteins that bind to enhancer-like sites, located far upstream from the promoter. These unique prokaryotic proteins, known as enhancer-binding proteins (EBP), mediate open promoter complex formation in a reaction dependent on NTP hydrolysis. The best characterized proteins of this family of regulators are NtrC and NifA, which activate genes required for ammonia assimilation and nitrogen fixation, respectively. In a recent IRBM course (@ontiers of protein structure prediction," IRBM, Pomezia, Italy, 1995; see web site http://www.mrc-cpe.cam.uk/irbm-course95/), one of us (J.O.) participated in the elaboration of the proposal that the Central domain of the EBPs might adopt the classical mononucleotide-binding fold. This suggestion was based on the results of a new protein fold recognition algorithm (Map) and in the mapping of correlated mutations calculated for the sequence family on the same mononucleotide-binding fold topology. In this work, we present new data that support the previous conclusion. The results from a number of different secondary structure prediction programs suggest that the Central domain could adopt an alpha/beta topology. The fold recognition programs ProFIT 0.9, 3D PROFILE combined with secondary structure prediction, and 123D suggest a mononucleotide-binding fold topology for the Central domain amino acid sequence. Finally, and most importantly, three of five reported residue alterations that impair the Central domain. ATPase activity of the E sigma 54 activators are mapped to polypeptide regions that might be playing equivalent roles as those involved in nucleotide-binding in the mononucleotide-binding proteins. Furthermore, the known residue substitution that alter the function of the E sigma 54 activators, leaving intact the Central domain ATPase activity, are mapped on region proposed to play an equivalent role as the effector region of the GTPase superfamily.

Construction and characterization of new cloning vehicles. VII. Construction of plasmid pBR327par, a completely sequenced, stable derivative of pBR327 containing the par locus of pSC101.

In vitro recombinant DNA experiments, using plasmid pBR327 and a DNA fragment derived from plasmid pSC101 containing the par region, resulted in the construction of plasmid pBR327par. This new cloning vehicle has all the cloning properties of the parental plasmid, and is more stable than pBR327. Since the nucleotide sequence of the par region has been determined, this new vector is completely characterized. Some features of the sequence with possible functional significance are discussed.

Combinatorial mutagenesis of three major groove-contacting residues of EcoRI: single and double amino acid replacements retaining methyltransferase-sensitive activities.

A library of mutant ecoRIR genes encoding EcoRI restriction endonuclease was generated using trinucleotide blocks and a combination of recombinant DNA procedures, including primer extension and the polymerase chain reaction. Codons corresponding to three amino acids (E144, R145 and R200), previously implicated in the specific recognition of the DNA substrate, were combinatorially mutated so as to generate a library that potentially contains all 20(3) possible single, double and triple aa replacements, in a balanced distribution. Inspection of the phenotypes of Escherichia coli colonies bearing the mutant genes showed that several of them retained activities that were deleterious to the cells but were still protected by the EcoRI methyltransferase. These included new enzyme variants, including non-conservative single (Thr or Val for Glu144) and double (Val for Glu144 and Thr for Arg145) replacements.

Construction and characterization of new cloning vehicles. IV. Deletion derivatives of pBR322 and pBR325.

In vitro recombinant DNA experiments involving restriction endonuclease fragments derived from the plasmids pBR322 and pBR325 resulted in the construction of two new cloning vehicles. One of these plasmids, designated pBR327, was obtained after an EcoRII partial digestion of pBR322. The plasmid pBR327 confers resistance to tetracycline and ampicillin, contains 3273 base pairs (bp) and therefore is 1089 bp smaller than pBR322. The other newly constructed vector, which has been designated pBR328, confers resistance to chloramphenicol as well as the two former antibiotics. This plasmid contains unique HindIII, BamHI and SalI sites in the tetracycline resistance gene, unique PvuI and PstI sites in the ampicillin resistance gene and unique EcoRI, PvuII and BalI sites in the chloramphenicol resistance gene. The pBR328 plasmid contains approx. 4900 bp.

Saturation mutagenesis of His114 of EcoRI reveals relaxed-specificity mutants.

EcoRI recognizes and cleaves DNA at GAATTC sites and is one of the best characterized sequence-specific restriction endonucleases (ENases). In previous studies, an EcoRI mutant, which exhibited relaxed substrate specificity and cleaved both canonical and EcoRI star sites, was isolated. This mutant enzyme has Tyr instead of His114. Here, we subjected residue 114 of the EcoRI ENase to saturation mutagenesis. The resulting mutant enzymes were characterized both in vivo and in vitro, resulting in the identification of mutants with canonical (H114K, Q, D, I) or relaxed (H114Y, F, S, T) specificity, as well as one mutant with severely impaired activity (H114P). In the X-ray structure of an EcoRI-substrate complex, His114 is located between the catalytic and recognition regions of EcoRI and may directly contact the DNA phosphate backbone. Based on our genetic and biochemical findings and the X-ray structure, we propose that His114 participates in substrate recognition and catalysis, either directly, via protein-DNA interactions, or indirectly, by mediating conformational changes that trigger DNA cleavage in response to substrate recognition.

The use of synthetic oligodeoxyribonucleotides to produce specific deletions in the araBAD promoter of Escherichia coli B/r.

Two oligodeoxyribonucleotides were chemically synthesized and used to specifically mutate the regulatory region of the araBAD operon in Escherichia coli B/r. One oligodeoxyribonucleotide introduced a 3-bp deletion in the araC activator binding site, the other a 3-bp deletion in the CRP-cAMP binding site. The mutations were introduced onto an ara insert cloned in an M13 vector using the synthetic oligodeoxyribonucleotides as primers and the (+) strand of an M13 mp2::ara hybrid phage as a template in an in vitro polymerization reaction. Hybridizations using the original synthetic oligodeoxyribonucleotide as a radioactive probe identified phage containing the desired deletion. The mutant ara inserts were subcloned into a stable plasmid for functional analysis. Transcription studies performed on strains containing the mutant ara plasmids demonstrated that both mutations reduced the amount of araBA mRNA synthesized in the presence of L-arabinose.

Plasmid vector pBR322 and its special-purpose derivatives--a review.

The plasmid pBR322 was one of the first EK2 multipurpose cloning vectors to be designed and constructed (ten years ago) for the efficient cloning and selection of recombinant DNA molecules in Escherichia coli. This 4363-bp DNA molecule has been extensively used as a cloning vehicle because of its simplicity and the availability of its nucleotide sequence. The widespread use of pBR322 has prompted numerous studies into its molecular structure and function. These studies revealed two features that detract from the plasmid's effectiveness as a cloning vector: plasmid instability in the absence of selection and, the lack of a direct selection scheme for recombinant DNA molecules. Several vectors based on pBR322 have been constructed to overcome these limitations and to extend the vector's versatility to accommodate special cloning purposes. The objective of this review is to provide a survey of these derivative vectors and to summarize information currently available on pBR322.

Construction and characterization of new cloning vehicles. V. Mobilization and coding properties of pBR322 and several deletion derivatives including pBR327 and pBR328.

A DNA sequence essential for the R64drd11 + ColK-mediated conjugal transfer of pBR322 has been located in a 540 bp HaeIII fragment (HaeIII-2) between the vegetative origin of replication and the tetracycline resistance (Tcr) gene of this vector. The pBR322 derivatives pBR327 and pBR328 lack this DNA sequence and are not mobilized by conjugation. Two derivatives of pBR328 were constructed by re-inserting the HaeIII-2 fragment in both orientations into the chloramphenicol-resistance gene of the same vector. One orientation of the HaeIII-2 fragment permitted mobilization by conjugation while the opposite orientation prevented mobilization. Further examination of pBR322 and derivatives revealed that the region between the origin of replication and Tcr gene also plays a role in regulating plasmid copy number.

Did cyclodextrin glycosyltransferases evolve from alpha-amylases?

The hydrolytic enzymes, alpha-amylases, and the cyclodextrin glycosyltransferases (CGTases) are key enzymes in the depolymerization of starch. These two groups of enzymes are evolutionarily related. We propose that the transferase activity is likely to have evolved from an ancestral hydrolase. Sequence analysis provides support for this hypothesis. Consequently, we have conducted an experimental study to test the possible adaptive value for evolving a CGTase. We found that when an alpha-amylase and a CGTase are combined more glucose is generated from starch than would be expected from the independent action of either of these enzymes. Thus, we propose that the biological role of CGTases is to work in concert with alpha-amylases for the efficient saccharification of starch. This observation can be useful in industrial processes aimed at producing syrups with high contents of glucose or maltose.

Isolated domain II and III from the Bacillus thuringiensis Cry1Ab delta-endotoxin binds to lepidopteran midgut membranes.

The DNA fragment encoding Cry1Ab domain II-III (45.3 kDa) was cloned and expressed. Domain II-III is expressed in low yields. In vitro binding analysis to Manduca sexta and Trichoplusia ni larval midgut tissue sections demonstrated that domain II-III fragment bound along the microvilli of the midgut epithelium, indicating that this fragment retains binding functionality in the absence of domain I. Binding of domain II-III to the midgut brush border membrane proteins from T. ni larvae indicated that Cry1Ab toxin and domain II-III bind to the same 150 kDa protein. In contrast, in M. sexta membranes, Cry1Ab toxin binds to 200 and 120 kDa proteins, and domain II-III only binds to the 200 kDa protein. Finally, binding assays with isolated brush border membrane vesicles showed that the interaction of domain II-III with the membrane vesicles is highly reversible, supporting the proposition that the integration of domain I into the membrane could participate in the irreversible binding of the toxin. These studies confirm that this part of the toxin is involved in binding interactions and could be separated as a discrete fragment that conserves at least part of its functionality.

Escherichia coli TEM1 beta-lactamase in CTAB reverse micelles: exchange/diffusion-limited catalysis.

We report kinetic data of penicillin hydrolysis catalyzed by beta-lactamase entrapped in reverse micelles formed with cetyl trimethylammonium bromide (CTAB), n-octane, hexanol and aqueous buffer. The K(cat) of this diffusion-limited reaction can be improved in aqueous buffer by a factor of 1.1-1.2 just by increasing the phosphate buffer concentration from 50 to 100 mM. In reverse micelles, increasing the buffer concentration has little effect on K(cat) when the size of the empty micelle is below the size of the protein. However, in larger micelles, the effect is enhanced and the K(cat) improves several fold, changing the form of the curve of K(cat) versus Wo from bell-shaped to almost hyperbolic. The results indicate that micellar exchange and internal diffusion may limit the reaction in reverse micelles and provide further evidence that the form of the curve depends on other factors besides the relationship between the size of the enzyme and that of the empty reverse micelle.

The disulfide bridges of toxin 2 from the scorpion Centruroides noxius Hoffmann and its three-dimensional structure calculated using the coordinates of variant 3 from Centruroides sculpturatus.

The disulfide bridges of toxin 2 from the venom of the scorpion Centruroides noxius Hoffmann were found by amino acid sequence determination of fragments of native toxin, produced by enzymatic cleavage and separated by high-performance liquid chromatography (HPLC). They are: Cys12-Cys65, Cys16-Cys41, Cys25-Cys46 and Cys29-Cys48. The coordinates of the X-ray diffraction structure of toxin variant 3 of C. sculpturatus [(1980) Proc. Natl. Acad. Sci. USA 77, 6496-6500] were used to construct a three-dimensional model of toxin 2. All the amino acid replacements were easily accommodated, and the modeled structure reveals a clustered pattern of sequence variation, which may help to identify residues responsible for functional differences among toxins of mammals and insects.

Introducing transglycosylation activity in a liquefying alpha-amylase.

By mutating Ala-289 by Phe or Tyr in the Bacillus stearothermophilus alpha-amylase, we induced this enzyme to perform alcoholytic reactions, a function not present in the wild-type enzyme. This residue was selected from homology analysis with neopullulanase, where the residue has been implicated in the control of transglycosylation [Kuriki et al. (1996) J. Biol. Chem. 271, 17321-173291. We made some inferences about the importance of electrostatic and geometrical modifications in the active site environment of the amylase to explain the behavior of the modified enzyme.

Alcoholysis reactions from starch with alpha-amylases.

The ability of alpha-amylases from different sources to carry out reactions of alcoholysis was studied using methanol as substrate. It was found that while the enzymes from Aspergillus niger and Aspergillus oryzae, two well-studied saccharifying amylases, are capable of alcoholysis reactions, the classical bacterial liquefying alpha-amylases from Bacillus licheniformis and Bacillus stearothermophilus are not. The effect of starch and methanol concentration, temperature and pH on the synthesis of glucosides with alpha-amylase from A. niger was studied. Although methanol may inactivate alpha-amylase, a 90% substrate relative conversion can be obtained in 20% methanol at a high starch concentration (15% w/v) due to a stabilizing effect of starch on the enzyme. As the products of alcoholysis are a series of methyl-oligosaccharides, from methyl-glucoside to methyl-hexomaltoside, alcoholysis was indirectly quantified by high performance liquid chromatography analysis of the total methyl-glucoside produced after the addition of glucoamylase to the alpha-amylase reaction products. More alcoholysis was obtained from intact soluble starch than with maltodextrins or pre-hydrolyzed starch. The biotechnological implications of using starch as substrate for the production of alkyl-glucosides is analyzed in the context of these results.

Combinatorial libraries of proteins: analysis of efficiency of mutagenesis techniques.

A number of considerations are made on the efficiency of mutagenesis techniques used in protein engineering, particularly those that include a random component. The expected outcome of different protocols is analyzed using computer programs. Special emphasis is made on the effect that the degeneracy of the genetic code has on the bias of the representation of amino acid replacements. The consequences of using alternative methods is analyzed in terms of the likelihood of obtaining underrepresented amino acid substitutions in mutant libraries. A consideration is also made of the outcome of combinatorial mutagenesis experiments with regard to the size of the amino acid window and the multiplicity of replacements that could be sampled with different methods. Optimal mutagenesis rates for specific conditions could be derived from the presented data and the computer program made available with this paper.

Microbial sensor for new-generation cephalosporins based in a protein-engineered beta-lactamase.

A protein-engineered beta-lactamase, constructed by site-directed mutagenesis in Escherichia coli (E104M/G238S), and having broadened specificity, was able to degrade cephalosporins of first, second, and third generations. Manipulations of culture conditions allowed an increase in beta-lactamase specific activity by up to twofold. The resultant bacteria were used to construct an immersable whole-cell biosensor for the detection of new-generation cephalosporins. Cells were immobilized on agar membranes, which in turn were attached to the surface of a flat pH electrode, thus constituting a biosensor based on the detection of pH changes. The sensor was able to detect second- and third-generation cephalosporins: cefamandole (0.4-4 mM), cefotaxime (0.4-3.5 mM), and cefoperazone (0.3-1.85 mM). Response times were between 3.5 and 11 min, depending on the kind of cephalosporin tested. The biosensor was stable for at least 7 d, time during which up to 100 tests were performed.