DNA Transposons

To generate Illumina-compatible sequencing libraries to link random DNA bar codes to transposon insertion sites, we first isolated genomic DNA from cell pellets of the mutant libraries with the DNeasy kit (Qiagen). Genomic DNA was quantified with a Qubit double-stranded DNA (dsDNA) HS (high sensitivity) assay kit (Invitrogen), and 1 µg was fragmented by ultrasonication to an average size of 300 bp with a Covaris S220 focused ultrasonicator. To remove DNA fragments of unwanted size, we performed a double size selection using AMPure XP beads (Beckman Coulter) according to the manufacturer’s instructions. The final fragmented and size-selected genomic DNA was quality assessed with a DNA1000 chip on an Agilent Bioanalyzer. Illumina library preparation involves a cascade of enzymatic reactions, each followed by a cleanup step with AMPure XP beads. Fragmentation generates genomic DNA templates with a mixture of blunt ends and 5′ and 3′ overhangs. End repair, A-tailing, and adapter ligation reactions were performed on the fragmented DNA using the NEBNext DNA Library preparation kit for Illumina (New England Biolabs), according to the manufacturer’s recommended protocols. For the adapter ligation, we used 0.5 µl of a 15 µM double-stranded Y adapter, prepared by annealing Mod2_TS_Univ (ACGCTCTTCCGATC*T) and Mod2_TruSeq (Phos-GATCGGAAGAGCACACGTCTGAACTCCAGTCA). In the preceding oligonucleotides, the asterisk and Phos represent phosphorothioate and 5′ phosphate modifications, respectively. To specifically amplify transposon insertion sites by PCR, we used the transposon-specific primer Nspacer_barseq_pHIMAR (ATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNNCGCCCTGCAGGGATGTCCACGAG), which contains a random hexamer and transposon-specific sequence on the 3′ end and an Illumina TruSeq sequence on the 5′ end, and one of 16 primers (see P7_MOD_TS_index primers in Table S2 in the supplemental material) containing the Illumina P7 end. For the transposon-insertion site enriching PCR, we used JumpStart Taq DNA polymerase (Sigma) in a 100-µl total volume with the following PCR program: 94°C for 2 min and 25 cycles of 94°C 30 s, 65°C for 20 s, and 72°C for 30 s, followed by a final extension at 72°C for 10 min. For the E. coli mutant library (KEIO_ML9), we replaced Nspacer_barseq_pHIMAR with Nspacer_barseq_universal (ATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNNGATGTCCACGAGGTCT). The final PCR product was purified using AMPure XP beads according to the manufacturer’s instructions, eluted in 25 µl of water, and quantified on an Agilent Bioanalyzer with a DNA1000 chip. For all five mutant libraries, we first sequenced the TnSeq libraries on an Illumina MiSeq to assess the quality of the mutant library. The P. inhibens, P. stutzeri, S. amazonensis, and S. oneidensis TnSeq libraries were paired end sequenced (2 × 150 bp) on an Illumina MiSeq using the MiSeq reagent kit v2 (300 cycles). For KEIO_ML9, we performed single-end sequencing (1 × 150 bp) with the MiSeq reagent kit v3 (150 cycles). Each TnSeq library was also sequenced on either the HiSeq 2000 or HiSeq 2500 system (Illumina) to map a greater fraction of the mutant population.

In addition to the genome sequence, 15 publicly available BAC sequences for common bean were also downloaded from GenBank for a total of 2.2 Mb of sequence, including from accessions DQ205649, DQ323045, FJ817289–FJ817291 and GU215957–GU215966. Transposon annotation was conducted using different methods according to the sequence structures and transposases of various transposons. To annotate LTR retrotransposons, the genome sequence was screened with LTR_Finder^{35 (link)} using default parameters, except that we set a 50-bp minimum LTR length and 50-bp minimum distance between LTRs. All predicted LTR retrotransposons were manually inspected to eliminate incorrectly predicted sequences, including tandem repeats, nested transposons, incomplete DNA transposons and other sequences. The internal sequences of LTR retrotransposons were used to perform BLASTX and/or BLASTP searches to define superfamilies: Ty1-copia, Ty3-gypsy or other. LINEs (long interspersed elements) were predicted on the basis of the non-LTR retrotransposase and polyA sequences. SINEs (short interspersed elements) were annotated with the polyA structure feature and combined with BLAST searches. To find DNA transposons, conserved domains for transposases from different reported superfamilies were used as queries to search the common bean genome. The matching sequences and flanking sequence (10 kb on each side) were extracted to conduct BLASTN searches to identify complete DNA transposons by terminal inverted repeats (TIRs) and target size duplication (TSD). Furthermore, MITEs-Hunter software^{36 (link)} was also used to identify DNA elements. The annotated transposons and two reported LTR retrotransposons, pva1-118d24-re-5 (FJ402927) and Tpv2-6 (AJ005762), were combined and used as a transposon library to screen the genome using RepeatMasker with default settings except that we used the 'nolow' option to avoid masking low-complexity DNA or simple repeats. Transposons were summarized according to names, subclasses and classes, and overlapping regions in the RepeatMasker output file were counted once (Supplementary Table 9).
To estimate the insertion times of LTR retrotransposons, the 5′ and 3′ LTRs for each full-length LTR retroelement were aligned and used to calculate the nucleotide divergence rate with the Kimura-2 parameter using MEGA 4. The insertion date (T) was estimated with the formula T = K/2r, where K is the average number of substitutions per aligned site and r is an average substitution rate. We used the average substitution rate of 1.3 × 10⁻⁸ substitutions per synonymous site per year^{55 (link)} to calibrate the insertion times.