We introduce a method of in vitro recombination or "DNA shuffling" to generate libraries of evolved enzymes. The approach relies on the ordering, trimming, and joining of randomly cleaved parental DNA fragments annealed to a transient polynucleotide scaffold. We generated chimeric libraries averaging 14.0 crossovers per gene, a several-fold higher level of recombination than observed for other methods. We also observed an unprecedented four crossovers per gene in regions of 10 or fewer bases of sequence identity. These properties allow generation of chimeras unavailable by other methods. We detected no unshuffled parental clones or duplicated "sibling" chimeras, and relatively few inactive clones. We demonstrated the method by molecular breeding of a monooxygenase for increased rate and extent of biodesulfurization on complex substrates, as well as for 20-fold faster conversion of a nonnatural substrate. This method represents a conceptually distinct and improved alternative to sexual PCR for gene family shuffling.

Pseudomonas putida utilizes the catBC operon for growth on benzoate as a sole carbon source. This operon is positively regulated by the CatR protein, which is encoded from a gene divergently oriented from the catBC operon. The catR gene encodes a 32.2-kilodalton polypeptide that binds to the catBC promoter region in the presence or absence of the inducer cis-cis-muconate, as shown by gel retardation studies. However, the inducer is required for transcriptional activation of the catBC operon. The catR promoter has been localized to a 385-base-pair fragment by using the broad-host-range promoter-probe vector pKT240. This fragment also contains the catBC promoter whose -35 site is separated by only 36 nucleotides from the predicted CatR translational start. Dot blot analysis suggests that CatR binding to this dual promoter-control region, in addition to inducing the catBC operon, may also regulate its own expression. Data from a computer homology search using the predicted amino acid sequence of CatR, deduced from the DNA sequence, showed CatR to be a member of a large class of procaryotic regulatory proteins designated the LysR family. Striking homology was seen between CatR and a putative regulatory protein, TfdS.

The 3-chlorocatechol operon ckcABD is central to the biodegradative pathway of 3-chlorobenzoate. The ckcR regulatory gene, which activates the ckABD operon, was cloned from the region immediately upstream of the operon and was shown to complement an insertion mutation for growth on 3-chlorobenzoate. ClcR activated the ckcA promoter, which controls expression of the clABD operon, in trans by 14-fold in an in vivo promoter probe assay in Pseudomonas putida when cells were incubated with 15 mM 3-chlorobenzoic acid. Specific binding of ClcR to the clcR-ckA intergenic promoter region was observed in a gel shift assay. Nucleotide sequence analysis of the ckcR gene predicts a polypeptide of 32.5 kDa, which was confirmed by using specific in vivo 35s labeling of the protein from a T7 promoter-controlled ATG fusion construct. ClcR shares high sequence identity with the LysR family of bacterial regulator proteins and has especially high homology to a subgroup of the family consisting of TcbR (57% amino acid sequence identity), TfdS, CatR, and CatM. ClcR was shown to autoregulate its own production in trans to 35% of unrepressed levels but partially relieved this autorepression under conditions that induced transcription at the ckcA promoter. Several considerations indicate that the ckcR-ckABD locus is most similar to the tcbR-tcbCDEF regulon.Modified ortho cleavage pathways are responsible for the biodegradation of a large array of chloroaromatic compounds and are widespread in soil bacterial communities. A well-characterized group of modified ortho cleavage pathway operons that encode functions for the dissimilation of catechol and chlorocatechols provides a model system for studying the regulation of evolutionarily related pathways. These include operons in degradative pathways for 3-chlorobenzoic acid (3Cba) from Pseudomonas putida, 1,2,4-trichlorobenzene from a Pseudomonas sp., 2,4-dichlorophenoxyacetic acid from Alcaligenes eutrophus, and benzoic acid from P. putida and from Acinetobacter calcoaceticus. In the above order, the relevant operons of the first three pathways include clcABD (15), tcbCDEF (50), and tfdCDEF (12), which convert 3-chlorocatechol, 3,4,6-trichlorocatechol, and 3,5-dichlorocatechol to maleylacetic acid, 5-chloromaleylacetic acid, and 2-chloromaleylacetic acid, respectively (12,14,44,45,52). The catechol-degradative genes in the benzoate pathways of P. putida andA. calcoaceticus are clustered in dissimilar operons. In P. putida (after the conversion of catechol to cis,cis-muconate by the catA gene product), the catBC operon encodes the conversion of the latter to 13-ketoadipate enol lactone, which is then transformed to 0-ketoadipate through the action of other loci (36).Catechol degradation in A. calcoaceticus is directed by the catA gene and the catBCEFD operon, which convert catechol through 13-ketoadipate to succinyl coenzyme A and acetyl coenzyme A (36). In nearly every case, the analogous reactions of the above five operons are catalyzed by homologous enzymes.Transcription of the catBCEFD, ...

There is great interest in engineering human growth factors as potential therapeutic agonists and antagonists. We approached this goal with a synthetic DNA recombination method. We aligned a pool of "top-strand" oligonucleotides incorporating polymorphisms from mammalian genes encoding epidermal growth factor (EGF) using multiple polymorphic "scaffold" oligonucleotides. Top strands were then linked by gap filling and ligation. This approach avoided heteroduplex annealing in the linkage of highly degenerate oligonucleotides and thus achieved completely random recombination. Cloned genes from a human-mouse chimeric library captured every possible permutation of the parental polymorphisms, creating an apparently complete recombined gene-family library, which has not been previously described. This library yielded a chimeric protein whose agonist activity was enhanced 123-fold. A second library from five mammalian EGF homologs possessed the highest reported recombination density (1 crossover per 12.4 bp). The five-homolog library yielded the strongest-binding hEGF variant yet reported. In addition, it contained strongly binding EGF variants with antagonist properties. Our less biased approach to DNA shuffling should be useful for the engineering of a wide variety of proteins.