Expression profiling of the CHS8, CHI1A, IFS2, and CHR genes in black *soybean seed [Glycine max (L). Merr.] of F4 generation

Black soybean [Glycine max (L.) Merr.] produces isoflavones as secondary metabolites, which have many benefits for human health and plant defense system. Expression profiling can guide potential work in functional genomics of the isoflavone biosynthesis pathway. Previous studies showed the vital role of the CHS8, CHI1A, and IFS2 genes in isoflavone biosynthesis. However, expression profiling of these genes in the local black soybean varieties is still limited. This study investigated the gene expression levels of the CHS8, CHI1A, IFS2, and CHR genes in local varieties, namely, UP106 (high isoflavone) and UP122 (low isoflavone) and its progenies, i.e., UP106xUP122 and UP122xUP106. Relative gene expression profiling was conducted on the basis of Reverse Transcriptase Polymerase Chain Reaction (RT‐PCR) with ACT2/7 as a housekeeping gene. As a result, the expression level of CHS8 in UP122 was lower than that in UP106. No significant difference in the expression level of CHI1A was observed in all samples. The expression levels of CHS8 and CHI1A in both progenies were higher than that in the parental line, whereas the expression levels of IFS2 in both progenies were lower than that in the parental line. CHS8 and IFS2 expression from UP106xUP122 was higher than that from UP122xUP106, whereas CHI1A expression from UP122xUP106 was higher than that from UP106xUP122. CHR showed a high expression in the reciprocal cross; however, this expression did not exceed from UP106. In conclusion, the crossing between parental lines did not affect the gene expression level in the isoflavone biosynthesis pathway.


Introduction
Black soybeans contain carbohydrates, lipids, and a high protein content, which reaches approximately 70%, as well as vitamins and minerals (Dajanta et al. 2013; Lam et al. 2010. In Indonesia, black soybeans have contributed significantly to the food industry, e.g., as the raw material for the soy sauce industry. In various countries, such as China, Japan, and Korea, black soybeans have also been used as a drug for hundreds of years to detox, prevent in flammation, and improve the quality of red blood cells (Xu and Chang 2008).
Black soybeans are known as a plant that has isoflavone levels 100 times higher than other legumes (Xu and Chang 2008). Isoflavones have an essential role in the defense system against biotic and abiotic stresses and in the process of forming root nodules as a result of the interaction between soybean plants and microbes (Akada et al. 1993; Christie et al. 1996. In fact, isoflavones pro vide many benefits for human health, such as reducing the risk of cardiovascular disease, osteoporosis, menopausal disorders, and cancer (Dhaubhadel 2011; Dixon 2004. Many researchers have investigated the isoflavone biosynthesis pathway, which is constructed by a variety of enzymatic reactions. The enzymes involved in isoflavone biosynthesis are genetically encoded through the transcrip tion and translation processes. Isoflavonoids are synthe sized through the phenylpropanoid pathway. The precur sor of the synthesis is Lphenylalanine. First, the amine group of Lphenylalanine is released. Then, cinnamic acid is produced through the Phenylalanine AmmoniaLyase (PAL) enzyme activity. In the subsequent phase reaction, cinnamate 4hydroxylase and 4coumarate CoA ligase convert cinnamic acid into pcoumaroyl CoA (Dhaubhadel 2011). Previous studies showed that CHR, CHI1A, and IFS2 are genes that play an important role in isoflavone biosynthesis (Yu et al. 2003). CHR and CHI1A are spe cific genes that are only found in legume plants. CHR gene expression plays a role in determining the composi tion ratio of isoflavones (daidzein/genistein/glycitein). In creased gene expression of CHR can increase the produc tion of the isoflavone daidzein (Yu et al. 2003). CHI1A expression plays an important role in the synthesis of type II chalcone isomerase that functions at the branching point between the production of flavonoids and isoflavonoids. IFS gene is one of the key genes in the synthesis of isoflavones. Previous research showed that the insertion of soybean IFS gene into nonlegume crops successfully alter the metabolism of nonlegume crops to produce the isoflavones genistein and daidzein (Jung et al. 2000).
The expression of genes related to the synthesis of isoflavones provides important information on the pro cess of metabolic engineering of isoflavone in plants. However, information on gene expression profiles is still limited because the expression of isoflavone genes is strongly influenced by cultivar and geographical envi ronment (Cheng et al. 2016). Thus, to manipulate the metabolism of isoflavones in Indonesian black soybean, research on the expression of genes related to the synthe sis of isoflavones in black soybean needs to be conducted. The expression profiles of genes related to the synthesis of isoflavones can be used to determine the function of genomic information in the isoflavone biosynthesis path way (Dhaubhadel et al. 2007). This research was con ducted to analyze the gene expression profiles of CHS, CHR, CHI1A, and IFS2 in the seeds of two local vari eties of black soybeans (i.e., UP106 and UP122) and its crossbreeding result at the F4 generation using Reverse TranscriptasePolymerase Chain Reaction (RTPCR).

Plant and seed samples
The black soybean (Glycine max L. Merr) seeds used were local varieties, i.e., UP106 and UP122 obtained from the Laboratory of Plant Breeding, Faculty of Agriculture, Pad jadjaran University (through Dr. Agung Karuniawan) and eight F4 genotypes as crossbreeding result. Plants were grown in the area of Tanjungsari, Sumedang (864 m above sea level) in February 2019. Pods were harvested when the plant has reached the R5 phase because isoflavone accu mulation in soybeans occurs during the seed maturation stage (Dhaubhadel 2011). Pods were collected, cleaned, and drained, wrapped using aluminum foil with the label, frozen in liquid nitrogen, and stored in −80°C.

Phenotypic measurement
In this study, phenotypic variables were measured and used as supporting data. The phenotypic variables in cluded plant height, number of branches, number of pods per plant, number of seeds per plant, the weight of seeds per plant, and the weight of 100 grains.

RNA isolation and quantitative and qualitative analyses
Total RNA samples of black soybean seeds were isolated following the protocol of GeneJET Plant RNA Purifica tion Mini Kit (Thermo Fisher Scientific™). RNA was an alyzed using electrophoresis and spectrophotometry to de termine the quality and quantity.

cDNA synthesis
Approximately one µg of RNA was used for cDNA syn thesis following the protocol of GoScript™ Reverse Tran scriptase (Promega Corporation).

Reverse transcriptase PCR and data analysis
The primer sequences used for CHS8, CHI1A, IFS2, and CHR were obtained from a previous study (Gutierrez Gonzalez et al. 2010), with ACT2/7 as a housekeeping gene. The cDNA samples were amplified using the PCR reagent of the DreamTaq Green PCR Master Mix Kit (Thermo Fisher Scientific™) and the DNA engine thermal cycler Veriti™ 96Well Thermal Cycler (Thermo Fisher Scientific™). PCR mixtures were prepared using 1 µL cDNA, 0.5 µL forward and reverse primers, 5 µL Dream Taq Green PCR Master Mix, and up to 10 µL nuclease free water. PCR products were visualized in 2% agarose gel stained with FloroSafe DNA Stain. Electrophoresis products were visualized by the Gel Doc™ EZ Imager. The thickness of the bands obtained from agarose gel elec trophoresis visualization was semiquantitatively analyzed using the ImageJ software (Sheffield 2008). Then, the ex pression values were tested using oneway analysis of vari ance (F3,8, p<0.001) with IBM SPSS20 software.

Phenotypic profile
Parental lines and their offspring showed variations in phe notypic profiles, except for plant height. In terms of plant height, all lines were nearly the same ( Figure 3). However, the F4(122x106)e strain was higher than the two parental lines.
Variations in phenotypic profiles (Figure 1 and 2) can be observed from the numbers of pods and seeds per plant (Figure 4), seed weight per plant, and weight of 100 seeds ( Figure 5). The two parental strains showed nearly the same numbers of pods and seeds per plant. The cross line generation showed higher numbers of pods and seeds per plant than the two parental lines, except for the F4(122x106)e line. Variations in seed weight per plant and weight of 100 seeds were observed among the lines. The cross line generation showed heavier seed weight per plant and weight of 100 seeds than the two parental lines, except for the F4(122x106)e line.
Based on the yield character, most of the offspring generations showed better yield than the two parental lines. This finding indicates that the yield character ex hibited transgressive segregation. The F4(122x106)e line showed the best yield.

Relative expression of CHS8 gene
The total relative gene expression shown in Figure 6 indi cated that the four samples affected the relative gene ex pression level of CHS8 (pvalue < 0.05). UP106xUP122 and UP122xUP106 exhibited significant differences in CHS8 relative gene expression in the isoflavone biosyn thesis pathway. The CHS gene was expressed higher at the initial development stage of the seed, i.e., during the R5 phase, than that at the other stages (Fritze et al. 1991). The results showed a difference in the relative gene ex pression level of CHS in the four samples because of the differences in the enhancer elements of the CHS gene pro moter, which could increase gene expression. Transcrip tion factors that bind to different enhancer elements pro duced different CHS gene expression levels (Fritze et al. 1991).

Relative expression of CHI1A gene
The CHI1A gene expression level was not significant in the four samples (pvalue > 0.05).
The high est expression levels were exhibited by UP106xUP122, UP106, UP122xUP106, and UP122. UP106xUP122 and UP122xUP106 did not exhibit a significant difference; thus, the reciprocal F4 generation had the same poten tial in expressing the CHI1A gene. Consistent with the findings of GutierrezGonzalez et al. (2010), we observed that CHI1A gene expression was not significant in the four samples. Moreover, the CHI1A gene showed stable ex pression levels during the generative growth phases, i.e., R5-R7, indicating that CHI ensures the stable synthesis of liquiritigenin and naringenin in all four genotypes.

Relative expression of IFS2 gene
IFS2 showed differential expression in the four samples (pvalue < 0.05). The IFS2 gene expression levels ex hibited by UP106xUP122 and UP122xUP106 were higher than those by the two parents. Thus, both UP106xUP122 and UP122xUP106 crosses exhibited significant differ ences, which, in this case, the group of tillers resulting from the crossing of UP106xUP122 had a higher poten tial level of relative gene expression than the group of off spring resulting from the crossing of UP122xUP106.

Relative expression of CHR gene
CHR showed the highest ratio in UP106, UP106xUP122, UP122xUP106, and UP122. The ratios in UP106xUP122 and UP122xUP106 were higher than that in UP122 but did not exceed that in UP106. The crossing of UP106xUP122 and UP122xUP106 had a significant effect (pvalue < 0.05) on the CHR gene expression level. Moreover, UP106xUP122 had a higher potential level of relative gene expression than the reciprocal UP122xUP106.
Particularly for legume plants, the chalcone reductase enzyme encoded by the CHR gene is needed for the pro duction of chalcone isoliquiritigenin. The CHR enzyme will compete with the CHI enzyme as a determinant of the relative changes in total genistein and daidzein formed by the division of the intermediate pathway from genistein to the branch of daidzein through increased CHR activity (GutierrezGonzalez et al. 2010).

Summary of Genes Expression Profile
The overall profile of relative gene expression in CHS8, CHI1A, IFS2, and CHR showed a positive correlation be tween parents and its F4 generation. The F4 generation UP106xUP122 has a higher potential for gene expres sion than its reciprocal because UP106 that acts as the fe male parent in the cross has a higher potential for gene expression than UP122. By contrast, the F4 generation UP1226xUP106 has a lower potential for gene expression than its reciprocal because UP122 that acts as the female parent in the cross has a lower potential for gene expres sion than UP106.
Genetic variation formed in the F4 generation, resulted from the crossing UP106xUP122, was caused by genetic recombination between the parents. Genetic recombina tion is the exchange of genetic material between individ uals reproduced sexually that produce recombinant off spring in segregation generation. Genetic recombination occurs through a cellular process during meiosis known as crossing over (Acquaah 2012).  Crossing over is the exchange of genetic material be tween homologous chromosomes that produce recombi nant chromosomes in sexual reproduction, and this event is one of the final stages of genetic recombination that oc curs in the pachytene prophase I stage of meiosis in synap tic processes. When chromosomal crossover occurs, ho mologous chromosomes pair up with each other and ex change segments of their respective genetic material to form recombinant chromosomes. Chromosomal crossover is an essential part of the normal chromosome segregation process that occurs during meiosis (Snustad and Simmons 2012).
The F4 generation has a complex and diverse mix of traits. Variations in cytoplasmic genomes can affect the yield character of the F4 generation phenotype (Park et al. 2006; Zeyl et al. 2005. Genetic variations in the cyto plasm and nucleus affect the interactions that control ge netic variation in the F4 generation, which is the result of different combinations of its parents, i.e., UP106xUP122 and its reciprocal UP122xUP106 (Juenger et al. 2010; McKay et al. 2003. According to Chiari et al. (2006), the effect of variations in the cytoplasm needs to be consid ered in inheritance studies because it will affect the con tribution of parental lines to their offspring. Six types of isoflavones show the presence of cytoplasmic effects (Chiari et al. 2006).  Each genotype group produces different gene expres sion levels in CHS8, CHI1A, IFS2, CHR, and F6H3, (Fig  ure 6). Differential expression can be attributed to the reg ulation of gene expression at the transcription level of each genotype. Cheng et al. (2016) showed that the genotypes were the most critical factors in determining the form and content of isoflavones. Each genotype has a different ge netic variation (polymorphism) because of the presence of singlenucleotide polymorphisms (SNPs) that distinguish the genetic information of one individual from that of an other individual. SNPs have an influence on the gene ex pression level (Stranger et al. 2007; Veyrieras et al. 2008. SNPs on the alleles of each genotype will cause an allele specific expression, in which two alleles are different or more genes are expressed at different levels. Mutations cause changes in the regulatory element (there is at least one regulator mutation) that appears in cisacting regions (Marioni et al. 2008). Transcriptional regulatory elements in cisacting regions contain recognition sites where tran scription factors attach as transacting elements, which en hance or suppress transcription (Maston et al. 2006).

Conclusions
In conclusion, CHS8, CHI1A, and IFS2 showed differ ential expression in the parental lines and reciprocal F4 hybrids. UP106 (high isoflavone) showed higher expres sion of CHS8 than UP122 (low isoflavone). By con trast, UP122 (low isoflavone) showed higher expression of IFS2 than UP106 (high isoflavone). The crossing between parental lines did not affect the gene expression level.