Supplementary information

Experimental materials

The fifty-two wild accessions of L. japonicus used in this study were Gifu, MG20, MG2, MG3, MG4, MG5, MG8, MG10, MG16, MG19, MG23, MG29, MG34, MG38, MG39, MG40, MG44, MG45, MG46, MG48, MG49, MG51, MG52, MG57, MG62, MG71, MG72, MG73, MG74, MG75, MG76, MG77, MG80, MG81, MG82, MG83, MG84, MG85, MG87, MG88, MG90, MG91, MG92, MG93, MG94, MG95, MG97, MG98, MG99, MG100, MG107 and MG117. Those wild accessions were provided by the National Bioresource Project (Lotus japonicus, Glycine max).

Primers

Sequences of primers used in this study are listed below.

Primer A; 5’-ACCTTCCCCCATGGGCAGGGTA-3’

Primer B; 5’-TGTTGCAAACCCAAGCTCCAGG-3’

Primer C; 5’-CAGCTTGAGGACAAGCTGAGTC-3’

Primer 1; 5’-GAGTTCAGTTTGGGTCAACCTC-3’

Primer 2; 5’-CCAGAACCACTTCTGTACCTC-3’

Primer 3; 5’-GGAGTATTCTTAAACGGGTTGC-3’

Primers used for detection of each individual LORE2 copy identified in Gifu and MG20. All primers were combined with primer C.

LORE2A; 5’- ACGTGGGAAGCCTCGTTCATGG -3’

LORE2B-1; 5’- TGATAGACTCTCACTACAAGTC-3’

LORE2B-2; 5’-GGATAGTTTTCATTGGAATTATC-3’

LORE2B-3; 5’- TAATCCAACATTATAGTGAACG-3’

LORE2B-4; 5’-AATCACAAGTGATTGGCTTAGC-3’

LORE2B-5; 5’-TAGTGCAGCAACAGTAGCAGTC-3’

LORE2B-6; 5’- TATCATCCATTGGGTATACACG -3’

LORE2B-7; 5’- AGCAGTTTCAGTTATGCATTAC -3’

LORE2B (TM0007-1); 5’- ACAGAGGTGCAACTCATACCAC-3’

LORE2B (TM0029); 5’-AAGCTAGAATTAGTCATCATGG-3’

Supplemental Fig. 1 Detection of newly transposed LORE2 copies in nfr5-2 and symrk-cac41.5

(A) Genomic Southern blot analysis detecting Spe I fragments of LORE2 in wild-type Gifu, nfr5-2, and symrk-cac41.5. The fragment representing a transposed LORE2 in nfr5-2 was marked with an asterisk. (B) SSAP analysis amplifying 3’ flanking regions of LORE2. Detected bands representing transposed LORE2 copies in nfr5-2 and symrk-cac41.5 are marked with asterisks. Sequencing of the three fragments revealed that the band indicated as “SymRK” originated from the LORE2B (cac41.5) insertion. Sequences of the another two fragments, indicated as “1” and “2”, were not identical to NFR5 or flanking sequences of identified preexisting LORE2 copies in Gifu. (C) PCR amplification of 3’ flanking regions of newly transposed LORE2 copies detected as fragment 1 and 2 in (B). PCR reactions were conducted using genome DNA of wild-type Gifu, nfr5-2, and symrk-cac41.5 and primer C together with nfr5-2 fl primer: (5’- TGCATCCCAGGGTAATACCAAG -3’ ), designed from fragment 1, or 41.5 fl primer (5’- AACCTCAGGTTCCAGAATCACC -3’), designed from fragment 2. Since each primer pair amplified a unique fragment from nfr5-2 or symrk-cac41.5, fragments 1 and 2 in (B) were confirmed to originate from newly transposed LORE2 copies in those two mutants other than LORE2A (nfr5-2) and LORE2B (cac41.5), respectively. WT: wild-type Gifu, nfr5-2: nfr5-2, cac41.5: symrk-cac41.5.

Supplemental Fig. 2 Structure of LORE1

Schematic representation of LORE1a, one of the nine LORE1 members identified in Gifu (Madsen et al. 2005). Boxes with triangles represent long terminal repeats (LTRs). Position of the DNA probe used in the Southern and Northern bolot analyses (Fig. 5) is represented as a bar. Vertical lines with H and X indicate the recognition sites of Hind III and Xba I, respectively, and numbers in the brackets show their positions in base pairs in LORE1a.