For full randomization of two contiguous sites, as well as two sites in close proximity, partially overlapped primers were used. In a case study, two active-site residues (V136/L141) of halohydrin dehalogenase from Arthrobacter sp. AD2 (HheA) were randomized. For this case, eight partially overlapping oligonucleotides were synthesized independently (four forward primers carrying mutagenized site 136: 5′-AGAAGCCGXXXGCCTACAA-3′, where XXX represents NDT, VMA, ATG, and TGG; four reverse primers carrying the mutagenized site 141: 5′-TTGCCCCXXXAGGAGGTGAT-3′, where XXX represents AHN, TKB, CAT, and CCA). In the case study to randomize two contiguous sites, two active-site residues (P135/F136) of halohydrin dehalogenase from Agrobacterium radiobacter AD1 (HheC) were selected. For this case, 16 mutagenic oligonucleotides (mutagenic oligonucleotides: 5′-TCTGCAACGXXXXXX-GGGCCTTG-3′, where XXX represents NDT, VMA, ATG, and TGG; there are 16 different combinations of the four degenerate codons in total) and one partially overlapping nonmutagenic primer (5′-GGTAGAAAGCTCCTT-CCAAGGC-3′) were synthesized independently. The recombinant expression vector pBADHheC containing the wild-type hheC gene (accession no: AF397296) was used as a template. These two PCRs were performed under the same conditions as described above.Small-intelligent library construction
After amplification, the reaction mixture was digested with DpnI (New England Biolabs) at 37°C for 1 h to remove the parental template. For one-site randomization, 2 µL DpnI-digested mixture were used to transform Escherichia coli TOP10 chemical competent cells, while 10 µL were used for transformations when two sites were randomized. The transformed culture was plated on LB agar plates supplemented with 100 µg/mL ampicillin, resulting in libraries with a few hundred to thousands of colonies for the above two cases. Sequence analysis was performed by Invitrogen.DC-Analyzer
The ultimate goal of the DC-Analyzer software developed in this work is to assist in degenerate primer design according to the user-defined randomization scheme utilized for the construction of small-intelligent libraries. In DC-Analyzer, the user can randomly choose the desired amino acids one by one. For other specific requests regarding the type of amino acids, such as randomizing one site with amino acids bearing a polar side chain (R, N, D, Q, E, H, K, S, and T) or a hydrophobic one (A, G, H, I, L, M, F, W, Y, and V), the user can directly click a radio button to toggle a group of the desired amino acids. For example, for encoding one site with amino acids bearing a polar side chain or a hydrophobic one, DC-Analyzer finds that VAM/MGC/ACG and YMC/GBC/ATK/TGG are the best schemes for the above two randomizations, respectively. The software is available from: http://cobi.uestc.edu.cn/resource/dc_analyzer/view.Results and discussion Outline of the small-intelligent strategy
The small-intelligent strategy comprises the two steps of degenerate codon design and PCR-based randomization (Figure 1). To construct a small-intelligent library with the prerequisite of no amino acid biases, the design of an optimal set of degenerate codons for each target site is critical and should obey the following criteria: (i) degenerate codon sets should comprise minimized codon variants that encode all user-desired amino acids with an equal probability of occurrence in the constructed library and (ii) stop codons and the rare codons of E. coli should be eliminated. For this, 53 out of 61 sense codons that make up the genetic codons were used in degenerate codon design to encode 20 amino acids by eliminating eight rare codons of E. coli (CGA, CGG, AGA, AGG for Arg; CUA for Leu; AUA for Ile; GGA for Gly, and CCC for Pro). As exemplified in the full randomization of one mutation site with 20 amino acids, a set of four degenerate codons of NDT (encoding 12 amino acids: N, S, I, H, R, L, Y, C, F, D, G, and V), VMA (encoding 6 amino acids: E, A, Q, P, K, and T), ATG (M), and TGG (W) were designed using DC-Analyzer according to the above criteria. Four primers with the above degenerate codons at the target site were synthesized independently and mixed at a ratio of 12:6:1:1 according to the numbers of encoded amino acids. The resulting primer mixture was termed small-intelligent primer and used for the subsequent PCR process. Although degenerate codons have been used in the construction of some restricted mutagenesis libraries (18), this is, to the best of our knowledge, the first report to use combinatorial degenerate primers for full randomization with just one codon per amino acid.