The seamless point mutations in the large biosynthetic genes were created by CcdB counterselection-based recombineering reported previously^⁵⁰,⁵¹. This method mentioned two rounds of recombineering: the site of interest was first replaced by a counterselection cassette amp-ccdB in E. coli GBred-gyrA462; then a synthetic single-strand oligonucleotide containing a mutation of interest to replace this cassette under the counterselection of the toxicity marker inferred by the counter-selectable gene ccdB. All single mutation R148M, R148L, R148V, R148A, R148G of RzmA-Cs in rzmA (p15A-tnpA-km-rzmA and pET28a-rzmACs) and A149M, A149L, A149V, A149R, A149G of HolA-Cs in holA (p15A-tnpA-km-holA and pET28a-holACs) were introduced using this method and verified by sequencing, respectively. The double or multiple mutants including Q36G/R148G (Q36G), Q136A/R148G (Q136A), Y138A/R148G (Y138A), M143A/R148G (M143A), Q136A/Y138A/R148G (QYA), Y138A/M143A/R148G (YMA), and Q136A/Y138A/M143A/R148G (QYMA), H140V/R148A of RzmA-Cs in rzmA (p15A-tnpA-km-rzmA and pET28a-rzmACs) were also achieved using repeated counterselection recombineering. All primers used for point mutations listed in Supplementary Data ¹. The correct plasmids containing biosynthetic gene clusters were transferred into S. brevitalea DSM 7029 for heterologous expression and metabolic analysis, respectively. The Cs domain variants were transferred into E. coli BL21 for protein expression and purification. Full information for the construction of plasmids and related strains/mutants/primers were given in Supplementary Data ¹.