**4.3 Occupancy of infected indigenous bradyrhizobia carrying** *hupSL* **genes**

Eighty-seven indigenous soybean-nodulating bradyrhizobial isolates belonging to the cluster *B*. *diazoefficiens* USDA 110T were investigated for the presence of *hupSL* genes. PCR templates of the test isolates, obtained as described in Section 4.2, were used. The PCR amplification for *hupSL* was performed using KAPA *Taq*® Extra Hot Start ReadyMix with dye (Kapa Biosystems, USA) and the *hupSL* primer set (hupS-F261: 5′-TCGAACAGGCGTTGTAAGTG-3′, hupS-R830: 5′-TCGACTACGACGACACCATC-3′, hupL-F962:5′-TCGGGCAGATAGACCATTTC-3′ and hupL-R1632: 5′-GGGATCGAAGTGATCCTGAA-3′). The PCR cycle comprised a pre-run at 95°C for 3 min, denaturation at 95°C for 15 s, annealing at 55°C for 15 s, and extension at 72°C for 1 min. This temperature control sequence was repeated for 30 cycles, followed by a final run at 72°C for 1 min. The PCR products were electrophoresed on a 2% agarose gel to confirm amplification.

### **4.4 Occupancy of soybean-nodulating bradyrhizobia carrying** *hupSL* **genes**

The occupancy rate of indigenous bradyrhizobia infecting each soybean cultivar is presented in **Table 5**. Based on the fragment patterns obtained from PCR–RFLP analysis, the indigenous isolates with the similar patterns as the five reference strains, namely *B*. *japonicum* USDA 6T ; *B*. *diazoefficiens* USDA 110T ; and *B*. *elkanii* USDA 46, 76T , and 94 were defined as Bj6, Bd110, Be46, Be76, and Be94, respectively. Since cluster Bd110 includes isolates carrying the *hupSL* genes, such as *B*. *diazoefficiens* USDA 110T and 122 [16], the occupancy rate of these isolates was also determined. PCR analysis targeting the *hupSL* genes revealed amplicons exhibiting zero to two bands. The isolates exhibiting two amplification products corresponding to *hupS* and *hupL* were defined as *hupS*<sup>+</sup> *L*+ , those exhibiting a single amplification product corresponding to *hupL* were defined as *hupS*<sup>−</sup> *L*+ , and those exhibiting no amplification products were defined as *hupS*<sup>−</sup> *L*− . In B × F − E, Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ were the most dominant (62.5%), followed by Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ (12.5%) and Bj6 isolates (12.5%). In B × F − M, Bj6 isolates were the most dominant (70.8%), followed by Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ (16.7%). In B × F − L, Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ (45.8%) were the most dominant, followed by Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ (25.0%). In "Enrei," Bj6 isolates were the most dominant (83.4%), followed by Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ (8.3%) or *hupS*<sup>+</sup> *L*+ (8.3%). In "Sachiyutaka," Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ (39.1%) were the most dominant, followed by Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ (34.8%). In "Fukuyutaka," Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ (37.5%) were the most dominant, followed by Bj6 (25.0%) isolates. "Sachiyutaka" and "Fukuyutaka" with the *Rj*4 genotype tended to present a higher occupancy rate of Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ . In contrast, soybean lines with the *Rj*2*Rj*3*Rj*4 genotype tended to present a lower occupancy rate of Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ . These results may be explained by the effect of the presence of *Rj*2. *Rj*2 restricts *B*. *diazoefficiens* USDA 122 [32]. Indigenous bradyrhizobial isolates, such as *B*. *diazoefficiens* USDA 110 T , which are not restricted by the *Rj*2 gene, can infect *Rj* gene-accumulated soybean lines. To solve this problem, the occupancy rate of inocula carrying the *hupSL* genes may be improved by inoculating *Rj* geneaccumulated soybean with *B*. *diazoefficiens* USDA 110T during cultivation. To test this hypothesis, we are currently investigating the effect of *B*. *diazoefficiens* USDA


*Bj6, Bd110, Be46, Be76, and Be94 showed RFLP patterns similar to those of B. japonicum USDA 6T , B. diazoefficiens USDA 110T , B. elkanii USDA 46, B. elkanii USDA 76T , and B. elkanii USDA 94, respectively. hupS− L− , hupS− L+ , and hupS+ L+ indicate isolates carrying or not carrying the hupS and/or hupL genes.*

### **Table 5.**

*Occupancy rate (%) of indigenous soybean-nodulating bradyrhizobia in each soybean cultivar in 2017.*

*Breeding of* Rj *Gene-Accumulated Soybean Genotypes and Their Availability for Improving… DOI: http://dx.doi.org/10.5772/intechopen.102833*

110T inoculation on the growth and yield of various soybean genotypes, including *Rj* gene-accumulated ones.

### **4.5 Correlation between the occupancy rate of indigenous bradyrhizobial strains and yield components of soybean**

Correlation analysis was used to evaluate the association between the occupancy rate of indigenous bradyrhizobial strains and yield components. Correlation coefficients were computed based on data obtained from the measurement of yield components and occupancy rate of indigenous soybean-nodulating bradyrhizobia. The R package "psych" was used to compute and plot the correlations. Additionally, the significance of the correlations was tested using the "cor.test" function in R.

The results of correlation analysis between the occupancy rate of indigenous bradyrhizobial strains and yield components of soybean are presented in **Figure 3**. The correlation coefficients of the occupancy rate of Bj6 isolates with plant height, node number, shoot dry weight, pod number, seed number, 100-seed weight, and yield were − 0.30, −0.04, 0.09, −0.13, −0.15, 0.41, and − 0.16, respectively. The correlation

### **Figure 3.**

*Correlation coefficient between the occupancy rate of Bd110 isolates carrying the hupSL gene and yield components. The correlation coefficients were computed based on data in Tables 4 and 5 (n = 6). A, B, C, D, E, F, G, H, I, J, and K indicate the occupancy rate (%) of Bj6, occupancy rate (%) of Bd110 exhibiting hupS<sup>−</sup> L− , occupancy rate (%) of Bd110 exhibiting hupS<sup>−</sup> L+ , occupancy rate (%) of Bd110 exhibiting hupS<sup>+</sup> L+ , plant height, node number, shoot dry weight, pod number, seed number, 100-seed weight, and yield, respectively. \*p < 0.01, \*\*p < 0.001, respectively.*

**Figure 4.**

*Correlation between the occupancy rate of Bradyrhizobium diazoefficiens USDA 110T and shoot dry weight of soybean. The values were obtained based on the growth investigation of soybean inoculated with B. diazoefficiens USDA 110T , B. japonicum USDA 6T , B. japonicum USDA 123, and Bradyrhizobium elkanii USDA 31 at the same bacterial density (106 cells mL−1) and then cultivated for 5 weeks. Values are presented as the mean of three replicates.*

coefficients of the occupancy rate of Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*− with plant height, node number, shoot dry weight, pod number, seed number, 100-seed weight, and yield were 0.36, 0.68, 0.32, 0.25, 0.29, −0.18, and 0.33, respectively. The correlation coefficients of the occupancy rate of Bd110 isolates exhibiting *hupS*<sup>−</sup> *L*+ with plant height, node number, shoot dry weight, pod number, seed number, 100-seed weight, and yield were 0.25, −0.26, −0.56, −0.18, −0.22, −0.41, and − 0.25, respectively. The correlation coefficients of the occupancy rate of Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ with plant height, node number, shoot dry weight, pod number, seed number, 100 seed weight, and yield were − 0.02, 0.25, 0.40, 0.42, 0.51, 0.05, and 0.61, respectively. Among correlations between the occupancy rate of each indigenous bradyrhizobial strain and soybean yield, Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ presented a strong positive correlation (*r* = 0.61), albeit without significant differences. Additionally, in another experiment, which revealed different results from the present findings, showed that the occupancy rate of *B*. *diazoefficiens* USDA 110T was correlated with the shoot dry weight of soybean (**Figure 4**). This result was obtained in the correlation analysis between the growth and the occupancy rate of *B*. *diazoefficiens* USDA 110T , *B*. *japonicum* USDA 6T and 123, and *B*. *elkanii* USDA 31 in soybean inoculated at the same bacterial density (106 cells mL−1) and cultivated for 5 weeks. Although this experiment was set in a greenhouse using cultivation pots, a significant positive correlation between the occupancy rate of USDA 110 and shoot dry weight of soybean plants was noted (*r* = 0.86, *p* = 0.03). A positive correlation between soybean growth and yield has been reported in previous study [59–62]. In the present study, yield was positively correlated with plant height (*r* = 0.35) and shoot dry weight (*r* = 0.75), albeit without significant differences (**Figure 3**). Therefore, enhancing the infection rate of bradyrhizobial strains, such as *B*. *diazoefficiens* USDA 110T carrying the *hupSL* genes, may promote the growth of soybean and consequently increase yield.

### **5. Conclusion**

In the present chapter, we described the breeding and selection processes, shoot growth, yield components, and infection tendency of useful bradyrhizobial strains carrying the *hupSL* genes of *Rj* gene-accumulated soybean lines exhibiting the

*Breeding of* Rj *Gene-Accumulated Soybean Genotypes and Their Availability for Improving… DOI: http://dx.doi.org/10.5772/intechopen.102833*

*Rj*2*Rj*3*Rj*4 genotype, obtained by crossing the Japanese soybean cultivars "Bonminori" (*Rj*2*Rj*3) and "Fukuyutaka" (*Rj*4).

First, we selected eight lines exhibiting the characteristics of the *Rj*2*Rj*2*Rj*4*Rj*4 genotype from 153 lines of F3 seeds following inoculation with *B*. *japonicum* Is−1 and *B*. *japonicum* Is−34 (**Table 1**). These eight lines were grown for several years, and three groups (B × F − E, B × F − M, and B × F − L) differing in terms of the flowering and ripening periods by approximately a week each were detected, which were cultivated further. Second, to investigate yield components, three *Rj* gene-accumulated soybean lines (B × F − E, B × F − M, and B × F − L) of F10 or F11 plant and three soybean cultivars ("Enrei," "Sachiyutaka," and "Fukuyutaka") were cultivated in the 2016 and 2017 growing seasons. The yield of B × F − M was equivalent to that of "Sachiyutaka," and this genotype likely possesses a greater yield potential than the parent soybean cultivar "Fukuyutaka," among the *Rj* gene-accumulated soybean lines. However, the 100-seed weight of B × F − M was lower than that of "Sachiyutaka" and "Fukuyutaka." Therefore, backcrossing with these cultivars is expected to produce soybean cultivars with larger seeds and higher yield potential. Third, to assess the occupancy rate of infected indigenous soybean-nodulating bradyrhizobia carrying the *hupSL* genes, we collected nodules from soybean roots and performed PCR-RFLP analysis of the 16S–23S rRNA gene ITS region. Furthermore, 87 indigenous soybean-nodulating bradyrhizobial strains belonging to the *B*. *diazoefficiens* USDA 110T cluster were investigated for the presence of the *hupSL* genes using PCR. The occupancy rate of Bd110 isolates carrying the *hupSL* genes tended to be lower in the *Rj* gene-accumulated soybean lines than in "Sachiyutaka" and "Fukuyutaka." In addition, among the *Rj* gene-accumulated soybean lines, B × F − L presented the highest occupancy rate of Bd110 isolates carrying the *hupSL* genes. Based on these results, during the cultivation of *Rj* gene-accumulated soybean, the occupancy rate of inocula carrying the *hupSL* genes can be improved by inoculating *B*. *diazoefficiens* USDA 110T , which is not restricted by the *Rj*2 gene.

Finally, to evaluate the association between the occupancy rate of indigenous bradyrhizobial strains and yield components of soybean, correlation analysis was performed. Correlation coefficients of the occupancy rate of Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ with plant height, node number, shoot dry weight, pod number, seed number, 100-seed weight, and yield were − 0.02, 0.25, 0.40, 0.42, 0.51, 0.05, and 0.61, respectively, and the occupancy rate of Bd110 isolates exhibiting *hupS*<sup>+</sup> *L*+ was strongly and positively correlated with yield components (*r* = 0.61), albeit without significant differences. Furthermore, soybean yield was positively correlated with plant height (*r* = 0.35) and shoot dry weight (*r* = 0.75), albeit without significant differences. Therefore, enhancing the infection rate of bradyrhizobial strains, such as *B*. *diazoefficiens* USDA 110T carrying the *hupSL* genes, may promote the growth of soybean and consequently increase its yield. In the future, we intend to conduct further inoculation tests with useful strains, such as *B*. *diazoefficiens* USDA 110T carrying the *hupSL* genes, to evaluate in greater detail the availability of *Rj* gene-accumulated soybean lines with the *Rj*2*Rj*3*Rj*4 genotype for improving productivity.

### **Acknowledgements**

The authors thank the members of the laboratories of Shimane University and University of Miyazaki involved in the present study. Additionally, the authors thank the technical staff of the Agricultural Science Section, Education and Research Center for Biological Resources, Faculty of Life and Environmental Science, Shimane University for their support in managing soybean cultivation. The authors also thank the Faculty of Life and Environmental Sciences at Shimane University for financial support to publish this chapter.
