**3.2 Identification and variation analysis of the responsible gene for the** *E3* **locus**

The line RIL1-146 was found to be heterozygous for the *FT3* locus. One other line, RIL6-22, showed segregation for growth habit. This trait is controlled by the *Dt1* locus and is linked to the *FT3* locus at a distance of about 25 cM. The segregating region of RIL6-22 included both the *Dt1* and the *FT3* loci. A single plant with a genotype of *dt1dt1 FT1FT1 ft2ft2 FT3ft3* was selected from RIL1-146, and 5plants with a genotype of *Dt1dt1 ft1ft1 ft2ft2 FT3ft3* were selected from RIL6-22 and designated as RHL1-146 and RHL6-22, respectively. From both progenies of these RHLs, two NILs, 1-146-*FT3* and –*ft3*, and 6-22-*FT3* and *–ft3* were selected. Using BSA analysis, a polymorphic AFLP marker, E6M22, was detected between the earlyflowering bulk and the late-flowering bulk derived from the progeny of RIL1-146. This marker was located at the LOD peak position of the *FT3* (Fig. 8).

Fig. 8. LOD scores for the *FT3* locus and heterozygous regions of RHLs. The location of the *FT3* locus and the segregating regions of two RHLs, 6-22 and 1-146 are shown. Solid line indicates the LOD scores calculated by composite interval mapping for the QTL (A). Shaded bars indicate the heterozygous regions of two RHLs (B).

A257 A169 GMS018 GM251 E16M23 GM017 E3M26 B124a B046a GM041 Satt156 GM267 Satt448 Satt166

LG L

*Dt1* Sat\_184 Satt664 A489 Satt229 GM043 E6M22 GM120a Satt513 Satt373

136.8cM

Fig. 8. LOD scores for the *FT3* locus and heterozygous regions of RHLs. The location of the *FT3* locus and the segregating regions of two RHLs, 6-22 and 1-146 are shown. Solid line indicates the LOD scores calculated by composite interval mapping for the QTL (A). Shaded

homozygous

**RHL6-22**

*dt1/dt1 Dt1/dt1 Dt1/Dt1*

homozygous

*FT3*

*Dt1*

homozygous

**RHL1-146**

homozygous

**3.2 Identification and variation analysis of the responsible gene for the** *E3* **locus**  The line RIL1-146 was found to be heterozygous for the *FT3* locus. One other line, RIL6-22, showed segregation for growth habit. This trait is controlled by the *Dt1* locus and is linked to the *FT3* locus at a distance of about 25 cM. The segregating region of RIL6-22 included both the *Dt1* and the *FT3* loci. A single plant with a genotype of *dt1dt1 FT1FT1 ft2ft2 FT3ft3* was selected from RIL1-146, and 5plants with a genotype of *Dt1dt1 ft1ft1 ft2ft2 FT3ft3* were selected from RIL6-22 and designated as RHL1-146 and RHL6-22, respectively. From both progenies of these RHLs, two NILs, 1-146-*FT3* and –*ft3*, and 6-22-*FT3* and *–ft3* were selected. Using BSA analysis, a polymorphic AFLP marker, E6M22, was detected between the earlyflowering bulk and the late-flowering bulk derived from the progeny of RIL1-146. This

marker was located at the LOD peak position of the *FT3* (Fig. 8).

A B

LOD score

8.0 2.04.06.0 0

Sat\_245 10cM

bars indicate the heterozygous regions of two RHLs (B).

Position 117.3cM LOD score 6.33

PVE 4.5%

Additive effect 2.4day

As a result of marker analysis, the heterozygous region in RHL1-146 extended for about 5 cM including the *FT3* locus. In contrast, the heterozygous region in RHL6-22 extended for about 40 cM including the *FT3* and *Dt1* loci. Two groups of NILs, NILs1-146 and NILs6-22, were used to develop the AFLP markers tightly linked to the *FT3* locus. Of all possible 4096 primer pairs, only six fragments showed constant polymorphism between the genotypes of *FT3*/*FT3* and *ft3*/*ft3* in NILs1-146 and NILs6-22. These polymorphic bands were excised from the gel, then sequenced, and converted to codominant SCAR markers. Several BAC and transformation-competent bacterial artificial chromosome (TAC) clones were screened using the SCAR markers. The nucleotide sequences of a BAC clone, GMJMiB242F01, and a TAC clone, GM\_TMiH\_H17D12, were determined. These BAC/TAC sequences were used to develop new PCR-based markers. A total of six DNA markers, including three AFLPderived markers (markers 1, 3, and 6) and three PCR-based markers developed from the BAC/TAC sequences (markers 2, 4, and 5) were used for fine mapping of the *FT3* locus (Table 2).

A population of 897 plants derived from seven RHL1-146 plants was used for precise mapping of the *FT3* locus. No recombination between these markers was found in 883 plants. The numbers of *FT3* homozygous late-flowering genotype (n=208) and heterozygous (n=441) and *ft3* homozygous early flowering genotypes (n=234) fitted a 1: 2: 1 segregation ratio. These results suggested the presence of a single QTL for flowering time within a small heterozygous region in RHL1-146. The additive effect and the dominant effect of this QTL were estimated to be 3.0 and 0.98 days, respectively. The ratio of genetic variance explained by the *FT3* locus accounted for 70.7 % of the total variance. On the other hand, 14 plants showed recombination between these markers (Fig. 9) and the recombination points were determined by the genotype of markers 2-5. The *FT3* genotypes in each recombinant completely coincided with the genotypes of marker 3 that originated from the closest AFLP marker E6M22 to the LOD peak position (Fig. 8). Moreover, recombination points occurred on both sides of marker 3 and corresponded to both sides of the TAC clone, GM\_TMiH\_H17D12. These results suggested that the gene responsible for the *FT3* locus was restricted to the physical region covered by GM\_TMiH\_H17D12 (Fig. 9).


Table 2. List of DNA markers used for fine mapping of the *FT3* locus.

Positional Cloning of the Responsible Genes for Maturity Loci *E1*, *E2* and *E3* in Soybean 63

similar to that of the non-long-terminal-repeat (LTR) retrotransposon reverse transcriptase element, but did not resemble the Ty1/copia or Ty1/gypsy sequences in the *e4* allele (Liu et al., 2008). Moreover, this inserted sequence showed a similar short sequence on both sides of the inserted position. To collect allelic information about *GmPhyA3*, the genes from Harosoy and Harosoy-*e3* were also isolated and designated as *GmPhyA3*-*E3* and *GmPhyA3*-*e3*, respectively. While a large retrotransposon-like insertion sequence was observed in *GmPhyA3-E3*, similar to that in *GmPhyA3*-Mo, the amino acid sequences encoded by *GmPhyA3*-Mi and –*E3* were identical (Fig. 10). On the other hand, a large deletion of 13.33 Kb at a position after the third exon was detected in *GmPhyA3*-*e3* (Fig. 10). Additionally, one mutant (*GmPhyA3*-mut), with a 40-bp deletion in the middle of the first exon of the *GmPhyA3* gene was screened from the mutant libraries of Bay by TILLING (Fig. 10). The

> IACGMAAARIASKDILDQDRSHTASEIRWGGAKHEPGERDDGRRV IACGMAAA(TQPQKSDGVVQSMSLVKGMMVGGCIQDHHSRLSLKL\*)

GAATAGCTTCCAAAGATATACTTTTCTGGTTTCGGTCTCA

Amino acid substitution

Gene structure Red light sensitivity

Normal

Normal

Less

Less

Less

40bp deletion

Fig. 10. Variation of gene structure of *GmPhyA3* and red light sensitivity. Open boxes, shaded boxes, and horizontal lines indicate exons, UTRs, and introns, respectively. The deleted region detected in Harosoy-*e3* is denoted by a dotted line. The deleted region in the

middle part of the first exon of the mutant is shown at the bottom of the figure. The sequence of 40-bp deletion and the corresponding translated amino acid sequence in the wild-type plant are displayed. As a result of the deletion, a stop codon following the 36

For allelism test among the *E3*, *FT3*, and *ft3* alleles, two population from crosses between Harosoy (*Dt1Dt1 e1e1 e2e2 E3E3*) and 6-22-*FT3* (*Dt1Dt1 ft1ft1 ft2ft2 FT3FT3*) and 6-22-*ft3* (*Dt1Dt1 ft1ft1 ft2ft2 ft3ft3*) were developed. Genetic analysis revealed that only the crossing population of Harosoy and 6-22-*ft3* showed a significant difference in genetic effect on flowering time. This indicated that the *E3* and *FT3* alleles had the same effect. The large insertion-like retrotransposon observed in *GmPhyA3*-*E3* and –Mo therefore might have no effect on the phenotype, whereas the one-amino-acid substitution observed in the *GmPhyA*-

amino acids at the deletion site appears in the mutant.

Mo might have weakened the effect of the *FT3* allele.

sequence of *GmPhyA3* from Bay was identical to that of *GmPhyA3*-*E3*.

*GmPhyA3-Mi* (E3Mi)

*GmPhyA3-E3* (E3Ha)

*GmPhyA3-Mo* (e3Mo)

*GmPhyA3-e3* (e3T)

*GmPhyA3-mut*

*GmPhyA3-mut* Wild Type

Fig. 9. Fine mapping of the *FT3* locus. The genotypes of each recombinant are shown in the left panel. Misuzudaizu homozygous, Moshidou Gong 503 homozygous and heterozygous genotypes are indicated by solid, open and meshed boxes, respectively. The phenotypic segregation in the progenies of each recombinant was shown in the right panel. The interquartile region, median, and range are indicated by a box, bold vertical line, and horizontal line, respectively.

A total of 11 genes were predicted in the sequence of GM\_TMiH\_H17D12. Previous studies had suggested that the *FT3* locus may be identical to the maturity locus *E3* (Yamanaka et al., 2001) and that the *E3* gene which showed a large effect on flowering time under FLD conditions had some association with a photoreceptor (Cober et al., 1996b). Considering these findings, one gene highly similar to that encoding phytochrome A was considered to be the gene responsible for the *FT3* locus. To confirm this assumption, differences in this gene between the parental lines were investigated. This phytochorome gene was referred to as *GmPhyA3*, since two other phytochrome A genes had been previously designated as *GmPhyA1* and *GmPhyA2* by Liu et al. (2008). *GmPhyA3* obtained from Misuzudaizu (*GmPhyA3*-Mi) was found to encode a protein composed of 1130 amino acids. A BLAST search found that *GmPhyA3*-Mi displayed normal features of phytochrome A, including a chromophre-attached domain, two PAS domains, and a histidine kinase domain as conserved domains. Compared to *GmPhyA3*-Mi, the *GmPhyA3* gene of Moshidou Gong 503 (*GmPhyA3*-Mo) showed a large insertion in the fourth intron and one SNP for a nonsynonymous amino acid substitution (glycine to arginine) in the third exon (Fig. 10). This SNP corresponded to the polymorphism detected by the AFLP marker E6M22. The inserted sequence was 2.5 Kb in length and a part of this sequence was found to be highly

Fig. 9. Fine mapping of the *FT3* locus. The genotypes of each recombinant are shown in the left panel. Misuzudaizu homozygous, Moshidou Gong 503 homozygous and heterozygous genotypes are indicated by solid, open and meshed boxes, respectively. The phenotypic segregation in the progenies of each recombinant was shown in the right panel. The interquartile region, median, and range are indicated by a box, bold vertical line, and

A total of 11 genes were predicted in the sequence of GM\_TMiH\_H17D12. Previous studies had suggested that the *FT3* locus may be identical to the maturity locus *E3* (Yamanaka et al., 2001) and that the *E3* gene which showed a large effect on flowering time under FLD conditions had some association with a photoreceptor (Cober et al., 1996b). Considering these findings, one gene highly similar to that encoding phytochrome A was considered to be the gene responsible for the *FT3* locus. To confirm this assumption, differences in this gene between the parental lines were investigated. This phytochorome gene was referred to as *GmPhyA3*, since two other phytochrome A genes had been previously designated as *GmPhyA1* and *GmPhyA2* by Liu et al. (2008). *GmPhyA3* obtained from Misuzudaizu (*GmPhyA3*-Mi) was found to encode a protein composed of 1130 amino acids. A BLAST search found that *GmPhyA3*-Mi displayed normal features of phytochrome A, including a chromophre-attached domain, two PAS domains, and a histidine kinase domain as conserved domains. Compared to *GmPhyA3*-Mi, the *GmPhyA3* gene of Moshidou Gong 503 (*GmPhyA3*-Mo) showed a large insertion in the fourth intron and one SNP for a nonsynonymous amino acid substitution (glycine to arginine) in the third exon (Fig. 10). This SNP corresponded to the polymorphism detected by the AFLP marker E6M22. The inserted sequence was 2.5 Kb in length and a part of this sequence was found to be highly

60 65 70

Flowering segregation in progeny

Heterozygous allele

Misuzudaizu homozygous allele Moshidou Gong 503 homozygous allele

Days to flowering

03009 03027 04043\_rec 04110\_rec 04232\_rec 04241\_rec 04253\_rec 04277\_rec 04343\_rec 04351\_rec 04478\_rec 04531\_rec 04564\_rec 04733\_rec 04783\_rec 04814\_rec

horizontal line, respectively.

DNA marker

Physical contigs

Genetic mapping

Marker1 Marker2 Marker3 Marker4 Marker5 Marker6

*FT3 (E3)* locus

GM\_TMiH\_H17D12 100Kbp

GMJMiB242F01

Satt229 E54/56M59 E41M34 E18M23 E6M22 E30M47 GM120a

similar to that of the non-long-terminal-repeat (LTR) retrotransposon reverse transcriptase element, but did not resemble the Ty1/copia or Ty1/gypsy sequences in the *e4* allele (Liu et al., 2008). Moreover, this inserted sequence showed a similar short sequence on both sides of the inserted position. To collect allelic information about *GmPhyA3*, the genes from Harosoy and Harosoy-*e3* were also isolated and designated as *GmPhyA3*-*E3* and *GmPhyA3*-*e3*, respectively. While a large retrotransposon-like insertion sequence was observed in *GmPhyA3-E3*, similar to that in *GmPhyA3*-Mo, the amino acid sequences encoded by *GmPhyA3*-Mi and –*E3* were identical (Fig. 10). On the other hand, a large deletion of 13.33 Kb at a position after the third exon was detected in *GmPhyA3*-*e3* (Fig. 10). Additionally, one mutant (*GmPhyA3*-mut), with a 40-bp deletion in the middle of the first exon of the *GmPhyA3* gene was screened from the mutant libraries of Bay by TILLING (Fig. 10). The sequence of *GmPhyA3* from Bay was identical to that of *GmPhyA3*-*E3*.

Fig. 10. Variation of gene structure of *GmPhyA3* and red light sensitivity. Open boxes, shaded boxes, and horizontal lines indicate exons, UTRs, and introns, respectively. The deleted region detected in Harosoy-*e3* is denoted by a dotted line. The deleted region in the middle part of the first exon of the mutant is shown at the bottom of the figure. The sequence of 40-bp deletion and the corresponding translated amino acid sequence in the wild-type plant are displayed. As a result of the deletion, a stop codon following the 36 amino acids at the deletion site appears in the mutant.

For allelism test among the *E3*, *FT3*, and *ft3* alleles, two population from crosses between Harosoy (*Dt1Dt1 e1e1 e2e2 E3E3*) and 6-22-*FT3* (*Dt1Dt1 ft1ft1 ft2ft2 FT3FT3*) and 6-22-*ft3* (*Dt1Dt1 ft1ft1 ft2ft2 ft3ft3*) were developed. Genetic analysis revealed that only the crossing population of Harosoy and 6-22-*ft3* showed a significant difference in genetic effect on flowering time. This indicated that the *E3* and *FT3* alleles had the same effect. The large insertion-like retrotransposon observed in *GmPhyA3*-*E3* and –Mo therefore might have no effect on the phenotype, whereas the one-amino-acid substitution observed in the *GmPhyA*-Mo might have weakened the effect of the *FT3* allele.

Positional Cloning of the Responsible Genes for Maturity Loci *E1*, *E2* and *E3* in Soybean 65

approximately 500 bases on the fragments. The PCR products were sequenced and alignment of the sequences was carried out. Days from sowing to the first flowering and alleles at the *E3* locus of 30 accessions are listed in Table 5. The results showed that E3Mi and e3T types were abundant, followed by the E3Ha type, while the e3Mo seldom occurred. No other type has been detected so far. The E3Ha type was detected in the accessions from China and North America. The latest flowering group harbored the E3Mi type, while the earliest flowering group, the e3T type. There was no clear relationship between the flowering time and the alleles at the *E3* locus in the other groups, because the flowering time

> SNP : G → A Arg → Gly

E3\_05879FW E3\_08417RV

E3\_06355RV

E3\_05879FW E3\_08417RV

E3\_06355RV

E3\_03384FW E3\_03552RV

the reverse primer for E3f4 was different from that for other alleles.

E3f1 E3f2 E3f3 E3f4 (E3Mi )

Fig. 12. Variation of *GmPhyA3* gene structure with the position and orientation of primers for PCR walking. The PCR products (E3fi, E3f2, E3f3 and E3f4) are shown at the bottom of the figure. As the e3T type lacked a portion of the third intron and the downstream region,

Insertion : 2.6 kb

E3f4 (E3Ha / e3Mo )

E3f4 (e3T )

Deletion : 13.3 kb

E3\_08115FW e3T\_3544RV

E3\_08115FW E3\_09908RV

E3\_08115FW E3\_09908RV

depends on the combination of alleles at many loci.

Table 4. Anchor primers for sequence analysis at the *E3* locus.

E3\_03384FW E3\_03552RV

E3\_00527FW

E3\_00527FW

E3Ha

E3Mi

e3T

e3Mo

Since Cober's study (1966b) indicated that the *E3* allele exerted a large effect under FLD, the sensitivity to FLD conditions between the three NILs ( Harosoy and –*e3*, 6-22-*FT3* and –*ft3*, 1-146-*FT3* and –*ft3*) and the mutant line for the *GmPhyA3* gene were evaluated. While the flowering days of each line varied because of their different genetic backgrounds, the effect of the *E3*/*FT3* allele was enhanced under FLD conditions in all the NILs. The mutant line with *GmPhyA3*-mut flowered 15 days earlier than the original variety Bay under extended mercury-vapor lamp with high red/far-red (R/FR) conditions like FLD.

These results strongly suggest that *GmPhyA3* is the gene responsible for the locus *E3*/*FT3*. We designated the type of gene structure of *GmPhyA3*-Mi, *GmPhyA3*-*E3*, *GmPhyA3*-Mo and *GmPhyA3*-*e3* as E3Mi, E3Ha, e3Mo and e3T, respectively, hereafter. Distribution of these alleles was investigated using several cultivars and lines covering all the maturity groups in Japan. Three primer pairs were designed for discrimination among E3Mi, E3Ha/e3Mo and e3T. The sequences of these primers are shown in Table 3 and the positions of these primers are indicated in Fig. 11. The e3Mo type was distinguished from E3Ha type by *Mse1* digestion of a PCR product using specific primers, E3\_07666FW and E3\_08417RV. PCR products or digested fragments were separated by 1% agarose gel electrophoresis. Among the 80 accessions randomly selected from Genebank of the National Institute of Agrobiological Sciences (NIAS) in Japan, the E3Mi and e3T types were equally abundant, while the E3Ha and e3Mo types seldom occurred.


Table 3. The DNA markers for genotype analysis of the *E3* locus.

Fig. 11. Variation of *GmPhyA3* gene structure with the position and orientation of PCR primers.

The *E3* region were amplified with four pairs using the total DNA of 30 varieties, and four PCR products were designated as E3f1 to E3f4 (Table 4). The positions of these primers and PCR fragments are indicated in Fig. 12. Sequencing primers were constructed at intervals of

Since Cober's study (1966b) indicated that the *E3* allele exerted a large effect under FLD, the sensitivity to FLD conditions between the three NILs ( Harosoy and –*e3*, 6-22-*FT3* and –*ft3*, 1-146-*FT3* and –*ft3*) and the mutant line for the *GmPhyA3* gene were evaluated. While the flowering days of each line varied because of their different genetic backgrounds, the effect of the *E3*/*FT3* allele was enhanced under FLD conditions in all the NILs. The mutant line with *GmPhyA3*-mut flowered 15 days earlier than the original variety Bay under extended

These results strongly suggest that *GmPhyA3* is the gene responsible for the locus *E3*/*FT3*. We designated the type of gene structure of *GmPhyA3*-Mi, *GmPhyA3*-*E3*, *GmPhyA3*-Mo and *GmPhyA3*-*e3* as E3Mi, E3Ha, e3Mo and e3T, respectively, hereafter. Distribution of these alleles was investigated using several cultivars and lines covering all the maturity groups in Japan. Three primer pairs were designed for discrimination among E3Mi, E3Ha/e3Mo and e3T. The sequences of these primers are shown in Table 3 and the positions of these primers are indicated in Fig. 11. The e3Mo type was distinguished from E3Ha type by *Mse1* digestion of a PCR product using specific primers, E3\_07666FW and E3\_08417RV. PCR products or digested fragments were separated by 1% agarose gel electrophoresis. Among the 80 accessions randomly selected from Genebank of the National Institute of Agrobiological Sciences (NIAS) in Japan, the E3Mi and e3T types were equally abundant,

Insertion : 2.6 kb

E3Ha\_1000RV

E3\_08557FW e3T\_0716RV

E3\_08557FW

E3\_08557FW E3Ha\_1000RV

E3\_08557FW

Deletion : 13.3 kb

E3\_09908RV

mercury-vapor lamp with high red/far-red (R/FR) conditions like FLD.

while the E3Ha and e3Mo types seldom occurred.

 : Target region for CAPS analysis : Primer position of STS markers

primers.

Table 3. The DNA markers for genotype analysis of the *E3* locus.

SNP : G → A Arg → Gly

E3Ha

E3Mi

e3T

Fig. 11. Variation of *GmPhyA3* gene structure with the position and orientation of PCR

The *E3* region were amplified with four pairs using the total DNA of 30 varieties, and four PCR products were designated as E3f1 to E3f4 (Table 4). The positions of these primers and PCR fragments are indicated in Fig. 12. Sequencing primers were constructed at intervals of

e3Mo

approximately 500 bases on the fragments. The PCR products were sequenced and alignment of the sequences was carried out. Days from sowing to the first flowering and alleles at the *E3* locus of 30 accessions are listed in Table 5. The results showed that E3Mi and e3T types were abundant, followed by the E3Ha type, while the e3Mo seldom occurred. No other type has been detected so far. The E3Ha type was detected in the accessions from China and North America. The latest flowering group harbored the E3Mi type, while the earliest flowering group, the e3T type. There was no clear relationship between the flowering time and the alleles at the *E3* locus in the other groups, because the flowering time depends on the combination of alleles at many loci.


Fig. 12. Variation of *GmPhyA3* gene structure with the position and orientation of primers for PCR walking. The PCR products (E3fi, E3f2, E3f3 and E3f4) are shown at the bottom of the figure. As the e3T type lacked a portion of the third intron and the downstream region, the reverse primer for E3f4 was different from that for other alleles.

Positional Cloning of the Responsible Genes for Maturity Loci *E1*, *E2* and *E3* in Soybean 67

mapping was conducted by using more than 13,000 plants with a protocol for large-scale genotyping of soybean seeds (Fig. 2) and a candidate gene was identified (Xia et al.,

Fig. 13. Graphical genotype of RHL1-156. Solid bars and bars with slanted lines represent Misuzudaizu and Moshidou Gong 503 homozygous segments, respectively. Open bars

represent unclassified segments. Putative location of each QTL is circled.

LOD score LG C2

Dominance effect: 4.0days

Position: Satt658-E46M9A (0.5cM/1.2cM) Additive effect: 9.7days LOD score: 53.2 PVE: 73.4%

60 40 20 0

A121 GM047 GM031 E59M27B E43M11A A122 A059 cyc2Gm Satt432 Satt281 Satt520 Satt422 A655B E44M13B A655c E35M13a

K262 E38M8D E38M8B A063 A635 Satt363 E35M6c Satt286 Satt277 E47M5A *<sup>T</sup>* Satt365 Satt658

AG36 Satt100 Satt134 E45M11a E60M28H A538a A748 GM065 GM156 E43M5D

*FT1* (*E1*)

10 cM

GM100 K455 A676 Satt357

Fig. 14. QTL analysis for the *FT1* locus. LOD scores calculated by interval mapping are shown in the left panel. Close-up of the *FT1* region is highlighted in the right panel.

GM133 GmN93 A109 Satt489 Satt289

*T* (sf3'H) Satt365 Satt658 E46M9A GM169

*FT1* (*E1*)

0.8cM 1.2cM 1.2cM 0.6cM

A703A GM118 GM024a GM224a

unpublished ).


Table 5. Days from sowing to the first flowering and alleles at the *E3* locus. These accessions were sown on June 10, 2008 at the NIAS

### **3.3 Toward the positional cloning of the** *E1* **gene**

Among the the 156 RILs, a single line was identified as being heterozygous around the *FT1* locus (approximately 17 cM) based on the genotypes of the DNA markers, and was named RHL1-156 (Fig. 13). A population of 1,006 plants derived from RHL1-156 was used for fine mapping of the *FT1* locus. The *FT1* locus mapped between tightly linked DNA markers, Satt365 and GM169 (Fig. 14).

As it was difficult to find AFLP markers around this region in this population, we used mapping populations derived from a cross between Harosoy-*E1* (*E1E1 e2e2 E3E3*) and Harosoy (*e1e1 e2e2 E3E3*). The *E1* locus was mapped proximate to Satt557 between Satt365 and Satt289 using the F2 population (117 plants). In a F2:4 population (mixed progenies from F2 heterozygotes at Satt557 locus) with 1,442 individuals, seven recombinants were identified between Satt365 and Satt289. The flowering genotypes for each recombinant are confirmed by the progeny segregation pattern. With these recombinants, we were able to delimit the *E1* region to approximately 289 kb between markers A and E5 (Fig. 15). No recombination was found between markers S8 and Satt557, despite a physical distance of 133 kb. Because more than 40 genes were identified in the 289 kb region, more intense fine

Table 5. Days from sowing to the first flowering and alleles at the *E3* locus. These accessions

Among the the 156 RILs, a single line was identified as being heterozygous around the *FT1* locus (approximately 17 cM) based on the genotypes of the DNA markers, and was named RHL1-156 (Fig. 13). A population of 1,006 plants derived from RHL1-156 was used for fine mapping of the *FT1* locus. The *FT1* locus mapped between tightly linked DNA markers,

As it was difficult to find AFLP markers around this region in this population, we used mapping populations derived from a cross between Harosoy-*E1* (*E1E1 e2e2 E3E3*) and Harosoy (*e1e1 e2e2 E3E3*). The *E1* locus was mapped proximate to Satt557 between Satt365 and Satt289 using the F2 population (117 plants). In a F2:4 population (mixed progenies from F2 heterozygotes at Satt557 locus) with 1,442 individuals, seven recombinants were identified between Satt365 and Satt289. The flowering genotypes for each recombinant are confirmed by the progeny segregation pattern. With these recombinants, we were able to delimit the *E1* region to approximately 289 kb between markers A and E5 (Fig. 15). No recombination was found between markers S8 and Satt557, despite a physical distance of 133 kb. Because more than 40 genes were identified in the 289 kb region, more intense fine

were sown on June 10, 2008 at the NIAS

Satt365 and GM169 (Fig. 14).

**3.3 Toward the positional cloning of the** *E1* **gene** 

mapping was conducted by using more than 13,000 plants with a protocol for large-scale genotyping of soybean seeds (Fig. 2) and a candidate gene was identified (Xia et al., unpublished ).

Fig. 13. Graphical genotype of RHL1-156. Solid bars and bars with slanted lines represent Misuzudaizu and Moshidou Gong 503 homozygous segments, respectively. Open bars represent unclassified segments. Putative location of each QTL is circled.

Fig. 14. QTL analysis for the *FT1* locus. LOD scores calculated by interval mapping are shown in the left panel. Close-up of the *FT1* region is highlighted in the right panel.

Positional Cloning of the Responsible Genes for Maturity Loci *E1*, *E2* and *E3* in Soybean 69

photoperiodic regulation system, and the *GmFT5a* expression was also possibly regulated by

*GI* have the conserved function of controlling the expression of the *FT* gene in *Arabidopsis*, rice and pea (Hayama et al., 2003; Mizoguchi et al., 2005; Hecht et al., 2007). We analyzed the expression of *GmFT2a* and *GmFT5a* at 9:00 a.m. 4 weeks after sowing under natural daylength conditions using *E2* (*FT2*) NILs in which photoperiod changed from LD to SD. A clear association between the *GmFT2a* expression level and early flowering phenotype was observed in both NILs. However, there was no significant difference in the *GmFT5a* expression levels between these NILs. These results suggested that *GmGIa* probably controlled flowering time through the regulation of *GmFT2a*. The recessive alleles of the *E2* (*FT2*) locus were perhaps unable to suppress *GmFT2a* expression and resulted in the early

There are strong interaction among the effects of *E1* (*FT1*) and *E2* (*FT2*), *E1* (*FT1*) and *E3* (*FT3*) (Yamanaka et al. 2000; Watanabe et al. 2004). The *e3e3* recessive homozygote can initiate flowering under R-enriched LD, but the *e3e3* genotype is necessary for plants with *e4* mutant allele to flower under FR-enriched LD. In the mapping population with *e3* background, photoperiodic insensitivity could occur in either genotypes of *e1E4*, *E1e4* or *e1e4* (Abe et al., 2003). These results suggest that *E1*, *E2*, *E3* and *E4* might concurrently mediate photoperiodic flowering in a shared pathway. The expression of the candidate gene for the *E1* locus was found to be repressed under SD. Under SD conditions, *E3*/*E4*-mediated photoperiodic regulation system up-regulates the expression of *GmFT2a* and *GmFT5a* possibly through the repression of the *E1* gene (Fig. 16). The *E2* locus also might control the

*GmFT2a/GmFT5a* **Unidentified factor X**

?

We successfully identified the responsible genes for the *E1*, *E2* and *E3* by positional cloning strategy and proposed a tentative flowering time gene network in soybean based on interaction of these genes. We used RHLs derived from RIL for fine mapping a single QTL effectively. An RHL harbors a heterozygous region where the target QTL is located and a homozygous background in most other regions of the genome. Novel DNA markers tightly linked to the locus were developed based on AFLP between the NILs of the locus derived from an RHL. A large-scale population derived from RHLs was used to locate the target

? ? ?

*E1* **or** 

Relationship via expression level

Genetic Interaction

photoperiod-independent system in LD.

*GmFT2a* expression through the *E1* gene.

*E2* **(***GmGIa***)** *E3* **(***GmPhyA3***)** *E4 (GmPhyA2***)**

(Soybean florigen genes)

Fig. 16. A putative network of flowering time genes in soybean.

flowering phenotype.

**5. Conclusion** 

Fig. 15. Fine mapping of the *FT1/E1* locus. *E1* homozygous, *e1* homozygous and heterozygous genotypes are shown by solid, open and meshed boxes, respectively. The *FT1* genotype of each recombinant was identified by progeny test.

The *FT1* locus was genetically mapped into the semi-central domain of linkage group C2 (Fig. 3) and was included in the pericentromeric region of chromosome 06 (http://www.phytosome.net/). In the heterochromatic regions, the ratio of physical to genetic distance is 3.5Mb/cM in comparison of 197 Kb/cM in euchromatic regions (Schmutz et al., 2010). The responsible gene for *FT1*/*E1* locus is characterized by relatively lower mRNA abundance. In fact, no EST data of the *FT1*/*E1* gene could be retrieved from public databases. The gene encodes a novel small protein and is unique in the sense of no apparent orthologs in model plants *Arabidopsis* or rice. We are analyzing the ligands of this protein and the interaction with DNA sequences.
