Allelic variations at four major maturity E genes and transcriptional abundance of the E1 gene are associated with flowering time and maturity of soybean cultivars

Abstract

The time to flowering and maturity are ecologically and agronomically important traits for soybean landrace and cultivar adaptation. As a typical short-day crop, long day conditions in the high-latitude regions require soybean cultivars with photoperiod insensitivity that can mature before frost. Although the molecular basis of four major E loci (E1 to E4) have been deciphered, it is not quite clear whether, or to what degree, genetic variation and the expression level of the four E genes are associated with the time to flowering and maturity of soybean cultivars. In this study, we genotyped 180 cultivars at E1 to E4 genes, meanwhile, the time to flowering and maturity of those cultivars were investigated at six geographic locations in China from 2011 to 2012 and further confirmed in 2013. The percentages of recessive alleles at E1, E2, E3 and E4 loci were 38.34%, 84.45%, 36.33%, and 7.20%, respectively. Statistical analysis showed that allelic variations at each of four loci had a significant effect on flowering time as well as maturity. We classified the 180 cultivars into eight genotypic groups based on allelic variations of the four major E loci. The genetic group of e1-nf representing dysfunctional alleles at the E1 locus flowered earliest in all the geographic locations. In contrast, cultivars in the E1E2E3E4 group originated from the southern areas flowered very late or did not flower before frost at high latitude locations. The transcriptional abundance of functional E1 gene was significantly associated with flowering time. However, the ranges of time to flowering and maturity were quite large within some genotypic groups, implying the presence of some other unknown genetic factors that are involved in control of flowering time or maturity. Known genes (e.g. E3 and E4) and other unknown factors may function, at least partially, through regulation of the expression of the E1 gene.

Figure 1. Genotyping methods for E1 to E4 loci.

A–C: Genotyping of the E1 gene. Fragment was amplified by primer pair of TI-Fw and TI-Rv (A), subsequently subjected to TaqI digestion (B) and HinfI (C). The triangle represents the e1-nl type (lacking of fragment). The arrows in B and in C are representing e1-as and e1-fs genotypes, respectively. D: Genotyping of the E2 gene, fragment was amplified with primer pair of SoyGI_dCAP_Dra_fw and SoyGI_dCAP_Dra_rv, after digestion of DraI, the e2 was cut into two fragments, while E2 genotype remained uncut (arrows). E–G: Genotyping of the E3 gene. E: bands with three sizes were generated using the mixed primers. Band specific for E3-Mi and e3-tr were indicated by arrow and star, respectively. A 557 bp fragment (triangle) could be amplified either for E3-Ha or e3-Mo genotypes. F: PCR product yielded from primer pair E3_08094FW and E3_08417RV were digested with MseI. The E3-Ha genotype remained uncut, while e3-Mo allele (arrow) could be cut into 223 bp and 101 bp. G: specific determination of E3-fs type (arrow). CAPE primer pair E3-fsFW/E3-fsRV, restriction enzyme: AleI. H-J: Genotyping of the E4 gene. H: Using three mixed primers, a 837 fragment was specifically amplified from the E4 allele (arrow). I: CAPE primer pair e4-kam specific for e4-kam gene (arrow), enzyme AflII. J: CAPE primer pair e4-kes specific for e4-kes genotype (arrow), enzyme BspHI.

Figure 2. Geographic locations, daylength, and temperature of six experimental sites.

A: The geographic locations of the six experimental sites. B: the average day length (hr) between 2011 and 2012. C: The changes in temperature recorded in 2011. Since there was no temperature data available in Gongzhuling (43°53′ N, 124°84′E), we used the data from the neighboring city Changchun (43°88′ N, 125°35′ E) (60 Km apart) instead.

Figure 3. The correlation analyses of the time to flowering (R1) between 2011 and 2012.

A: Huaian; B: Mudanjiang.

Figure 4. The correlation between R1 and R3, R7 or R8 at Nanjing, Huaian, Gongzhuling and Mudanjiang locations, average of the two years of 2011 and 2012.

A: Correlation between R1 and R3 in Mudanjiang; B: Correlation between R1 and R7 in Mudanjiang. C: correlation between R1 and R3 in Gongzhuling; D: Correlation between R1 and R8 in Gongzhuling. E: Correlation between R1 and R3 in Huaian; F: Correlation between R1 and R8 in Huaian. G: Correlation between R1 and R3 in Nanjing; H: Correlation between R1 and R8 in Nanjing.

Figure 5. The phenotypic variations in R1 among different genotypic groups.

The phenotypic segregation is shown in box-plot format. The interquartile region, median, and range are indicated by the box, the bold horizontal line, and the vertical line, respectively. The 12 panels (A to L) represented 6 experimental locations in 2011 and 2012, respectively. A: Harbin in 2011; B: Harbin in 2012; C: Mudanjiang in 2011; D: Mudanjiang in 2012; E: Gongzhuling in 2011; F: Gongzhuling in 2012; G: Jinan in 2011; H: Jinan in 2012; I: Huaian in 2011; J: Huaian in 2012; K: Nanjing in 2011; L: Nanjing in 2012.

Figure 6. The phenotypic variations in R7 or R8 among different genotypic groups.

The phenotypic segregation is shown in box-plot format. The interquartile region, median, and range are indicated by the box, the bold horizontal line, and the vertical line, respectively. A: R8 at Mudanjiang in 2011; B: R8 at Mudanjiang in 2012; C: R7 at Gongzhuling in 2011; D: R7 at Gongzhuling in 2012; E: R7 at Huaian in 2011; F: R7 at Huaian in 2012; G: R7 at Nanjing in 2011; H: R7 at Nanjing in 2012.

Figure 7. Correlation coefficients between the transcript abundance of the E1 (A) or e1-as (B) genes and flowering time of cultivars grown at Harbin, in 2012.