After amplification, the PCR reactions were separated on agarose gels, the product bands excised, purified with innuPREP DOUBLEpure Kit (Analytik Jena AG, Jena, Germany) and digested with BsaI. The respective empty entry modules (∼ 100 ng) were also cut with this enzyme, usually in the same tube (1 h, 37°C). The digestion was purified with the above mentioned kit and ligated with T4 DNA ligase (1 h room temperature, overnight 4°C). After heat-inactivation (10 min, 70°C) the reaction was transformed via heat shock into ccdB sensitive E. coli strains (Mach1™-T1R, DH5α, XL1-Blue MR). Transformants were checked by colony PCR, plasmid DNA was isolated from positive clones and checked by sequencing and test digestion.
If internal BsaI recognition sites were present in the module sequence, they were removed by nucleotide substitution. For protein-coding sequences, silent mutations were chosen. In promoter and terminator sequences, the nucleotides to be changed were selected at random, but for later constructs we switched to always replace the first guanine by a cytosine. For simplicity, we used scar-free BsaI-cloning to create the substitutions. Primers were designed on both sides of the internal BsaI recognition sites introducing the mismatch and flanked on their 5′-ends by external BsaI recognition sites. The overhangs generated by the external BsaI cut were designed to be part of the gene-specific sequence and being different from the module specific overhangs.
Shorter modules were assembled as oligonucleotide duplices created from overlapping primers with unpaired 5′-overhangs complementary to the module specific overhangs. The oligonucleotides (10 µM or 100 µM) were mixed in equimolar ratios with each other, soused with boiling water, allowed to cool slowly down to room temperature and then ligated into BsaI digested and purified entry vector.