Mapping and validation of molecular markers of genes Dt 1 and Dt 2 to determine the type of stem growth in soybean

The type of stem growth in soybean is a distinguishing feature of cultivars. The genes Dt1 and Dt2 affect the termination of the stem, and the types of growth are classified as determinate, semideterminate and indeterminate. Phenotypic characterization of the type of growth is complex and is occasionally erroneously described. The objective of this study was to map and validate molecular markers to classify the type of soybean growth to facilitate the description of cultivars and genotypic selection. Two populations were used for mapping and validation of molecular markers F2:3: T 117 (semi-determinate growth type) x Igra RA 518 RR (indeterminate growth type) and CD 235RR (determinate growth type) x Igra RA 518 RR. The study demonstrates that the association of the molecular marker with the gene GmTFL1b was efficient in the classification of soybean growth types. The marker sat_064 is connected to the Dt2 gene, which is located in the Liaison Group G of the consensus map of soybeans with a recombination frequency of 19.4%. Molecular markers for Dt1 and Dt2 genes efficiently described the genotypes of soybean stem growth and selection.


Introduction
The type of stem growth of soybean [Glycine max (L.) Merr.] influences several agronomic characteristics.Plants with a determined type of growth typically exhibit smaller heights with greater resistance to lodging, smaller insertion heights for the first pod and a greater number of branches per plant compared with cultivars of an undetermined growth type and similar maturity group.Lineages of determined types also present a reduced period of flowering compared with lineages of undetermined types (Bernard, 1972;Kilgore-Norquest & Sneller, 2000).
Two genes, Dt1 and Dt2, affect the ending of the stem in soybean (Bernard, 1972).A recessive allele, dt1, and a dominant allele, Dt2, cause the ending of the apical growth of the stem to proceed in a rapid fashion, which diminishes the height of the plant and the number of knots.Of these genes, dt1 exerts a much greater effect.Bernard (1972) observed this phenotype and described it as intermediary semidetermined, which is distinct from the undetermined and determined phenotypes in hybrid populations among the undetermined and determined lineages.These semi-determined plants segregated to undetermined and determined types, indicating that Dt1 acts as a partially dominant gene in the tested genetic base.
The type of growth was occasionally incorrectly classified (Bernard, 1972).This difficulty of selection -description and tendency to use undetermined and semi-determined cultivars with greater frequency in the improvement programsbroadens in a progressive manner.The type of growth is a differentiating characteristic of soybean cultivars and is a part of the minimum demands of UPOV (International Union for the Protection of New Varieties of Plants) as a descriptor of cultivars with means to protect individual property.In this regard, determined, semi-determined and undetermined growth types are used to describe cultivars (Serviço Nacional de Proteção de Cultivares [SNPC], 2012).
Furthermore, the assisted characterization by molecular markers (SAM) is a useful tool in the quick and efficient development of cultivars (Schuster & Oliveira, 2006), whereas the molecular base of the growth type was dissected by Liu et al. (2010).Based on the soybean's genome and mapping analysis, the Dt1 gene codes the GmTFL1b protein, and the type of growth of the soybean stem is determined by variants of this gene.The Dt1 gene was mapped with molecular markers in the GL L (Cregan et al., 1999;Liu et al., 2007;Tian et al., 2010), and the Dt2 gene was mapped in the GL G (Muehlbauer et al., 1989;Cregan et al., 1999).These regions are currently designated as chromosomes 19 and 18, respectively (Schumutz et al., 2010).
In the beginning of the 21 st century, 93% of protected soy cultivars in Brazil were of a determined growth type (Brasil, 2002).In the agricultural year of 2010/2011, among the ten soy cultivars that seeded, half presented the undetermined type of growth.Regarding the other half, which was located in the Southern region of the country, five additional seeded cultivars were classified as the undetermined growth type (Kleffmann & Partner, 2011).Facing this reality, Brazilian farmers are changing their preference of growth type and improvement programs that aim at attending to the demand to face the need for the selection, description and distinction of determined, semi-determined and undetermined types of growth.However, during the separation process of the three phenotypes, difficulty is encountered in the selection and description.
Given this difficulty, the present paper aims at mapping and validating molecular markers to classify soy according to its growth type to facilitate the description of cultivars and selection within a genetic improvement program.

Genetic material
For the mapping and validation of molecular markers, two F 2:3 populations were obtained according to the description below.
Population 2: the cultivar CD 235RR was used as a female parental; this cultivar possesses the genotype dt1dt1dt2dt2 (determined growth type).The male parental was the Igra RA 518 RR cultivar, which possesses the genotype Dt1Dt1dt2dt2 (undetermined growth type).
The hybridizations were made in a vegetation house in the summer of 2008/2009.However, the previous generation of the F 1 plants was made in a vegetation house in the autumn and winter of 2009, and the F 2 plants were planted in the field in the summer of 2009/2010.The F 2 plants were picked and threshed individually to obtain seeds (F 2:3 ).The seeds of each F 2 plant composed a F 2:3 family.These seeds were identified with the origin number of the F 2 plant.For the sowing of the F 2:3 families of each population, 60 seeds for each lineage were separated, and the remaining seeds were used for molecular analysis.

Phenotypic evaluation
The experiment was seeded in the 2010/2011 crop in the Coodetec -Agricultural Research Center Cooperative in Cascavel, Paraná State (Latitude S 24° 52' 56.9"; Longitude W 53° 32' 00.4" and altitude 690 m) in Purple Latosolic Distroferric soil.The F 2:3 lineages and cultivars were sowed in individual lines that were 4 m in length with 0.45 m spacing between the lines.For optimal results, fertilization and the control of weeds, plagues and diseases were performed according to the technical demands for the culture.Phenotypic evaluation of stem type of growth for the study and assessment of populations was performed during plant maturation (stage R8; Fehr, Caviness, Burmood, & Pennington, 1971).

DNA extraction
For the study, ten seeds of each F 2:3 family were used for the extraction of DNA and then ground to a fine powder.
DNA was extracted from the seeds of each F 2:3 cultivar and family according to the protocol described by McDonald, Elliot and Sweeney (1994) with some modifications (Schuster, Queiroz, Teixeira, Barros, & Moreira, 2004).Approximately 50 mg of seeds were placed in 1.5-ml micro tubes.Then, 500 μL of extraction plug, which contains 200 mM Tris HCl (pH 7.5), 25 mM EDTA, 288 mM NaCl and 0.5% of SDS, was added.For a more precise maceration, a 3-mm diameter glass sphere was added in each tube, and the tube was agitated in a grinder to infuse the plant tissue.After three min.of agitation, 500 μL of the extraction plug was added followed by centrifugation at 14,000 rpm for ten min.After this process, the supernatant was transferred to a new tube, and 10 μL of K proteinase (10 mg mL -1 ) was added for the removal of proteins.The samples were placed in a bath at 37ºC for 30 min.To precipitate the DNA, 500 μL of iced isopropanol was added.After two min. of rest, the samples were centrifuged for 15 min.at 14,000 rpm.The supernatant was discarded, and the precipitate was placed in ambient temperature for 15 min.for drying.For the elimination of RNA, the precipitate was resuspended in 300 μL of TE (Tris HCl pH 7.5 and 5 M EDTA), containing 40 μg μL -1 RNAse A (10 mg mL -1 ).The samples were placed in a bath at 37ºC for 30 min.DNA was once again precipitated by the addition of 500 μL of iced isopropanol.After two min., the tubes were centrifuged for 15 min.at 14,000 rpm.The supernatant was discarded, and the precipitate was placed in ambient temperature for 15 min.for drying.After this period, 300 μL of TE was added to resuspend the precipitate.
The DNA concentration in each sample was estimated using a spectrophotometer (Nanodrop1000) to assess absorbance at 260 nm; each absorbance unit corresponded to 50 μg mL -1 double-stranded DNA (Sambrook, Fritsch, & Maniatis, 1989).The DNA of each sample was diluted in TE to a working concentration of 5 ng μL -1 .
The quality of the DNA was evaluated in a 0.8% agarose gel containing 0.02 μL mL -1 ethidium bromide.

Dt1 gene amplification
For Dt1 gene amplification, we used the primers TFL1b-pro-F (5´-CCATGCTTAATCGGCATCACT-3´) and TFL1bpro-R (5´-GGTGGTGGCATAGTTTAATT-3´), which generate 410-bp (base pairs) fragments.These primers were designed to detect a single nucleotide polymorphism (SNP) in the promoter of the Dt1 gene (Liu et al., 2010).To validate the efficiency of this marker, plants with determined and undetermined growth types were used.The marker was subsequently assessed in the two CD 235RR x Igra RA 518 RR (145 lineages F 2:3 ) and T 117 x Igra RA 518 RR (90 lineages F 2:3 ) populations.
For polymerase chain reaction (PCR), the following reagents were added for a final volume of 20 μL: 2 mM MgCl 2 , plug 1x (50 μM Tris, 10 μM HCl), 300 μM dNTPs, 0.2 μM primer (sense and antisense), 1 unit of Taq DNA polymerase, and 50 ng of DNA.PCR was performed in a Veriti thermocycler (Applied Biosystems).The following program was used: 94ºC for 5 min.;45 cycles at 94ºC for 1 min., 55ºC for 1 min., 72ºC for 1 min.;and a final extension at 72ºC for 10 min.The products of the amplification were submitted to electrophoresis in a 2% agarose gel.The gel was stained with ethidium bromide, and the products were revealed under ultraviolet light in a Vilber Lourmat photo documentation device (Marne La valle, FR).
Next, 10 μL of the obtained PCR products were submitted to digestion with the NdeI enzyme (New England) according to the manufacturer's instructions.The reaction proceeded in a 37ºC bath for 3 hours and 50 min.and then submitted to 2% agarose gel electrophoresis.

Dt2 gene mapping
DNA was extracted from ten semi-determined growth type samples and ten undetermined growth type samples derived from the hybrids obtained from T 117 x Igra RA518 RR, all of which contained the Dt1 gene.The sample comprised 10 seeds from each F 2:3 lineage.For each type of growth, two DNA bulk preparations were made from five samples each and used as patterns for the T 117 (semi-determined Dt2Dt2) and Igra RA518 RR (undetermined dt2dt2) cultivars.
The primers that identified polymorphism among the bulk of DNA with contrasting growth types were considered as candidates and used individually for the amplification of the ten samples of each growth type that formed the bulks.
PCR was performed in a Veriti thermocycler (Applied Biosystems) programed follows: 94ºC for 3 min.;35 cycles at 94ºC for 30 s, 55ºC for 30 s, and 72ºC for 45 s; and a final extension at 72ºC for 20 min.The products of the amplification were dyed with a charging plug (bromophenol blue) and submitted to denaturation at 95ºC for 4 min.The fragments were separated by electrophoresis in a 6% denaturing polyacrylamide gel and dyed with 0.1% silver nitrate for fragment visualization.

Evaluation of the efficiency of selection with molecular marker for growth type in soybean
To validate the candidate marker, these methods were applied to the populations derived from the crossings between CD 235RR x Igra RA 518 RR and T 117 x Igra RA 518 RR.For the evaluation, 145 and 90 F 2:3 lineages of each crossing were used.The association of the marker with the phenotype was evaluated using the χ2 test with contingency tables containing phenotypic and molecular data.The analysis was conducted using the Genes Program (Cruz, 2013).
In the population derived from the crossing of CD 235RR x Igra RA 518 RR, the efficiency of selection with the Dt1 gene marker was evaluated once segregation occurred exclusively for this gene.In the population obtained from the T 117 x Igra RA 518 RR crossing, the efficiency of the Dt2 gene marker was evaluated for the population exclusively segregated for this gene.
With the data obtained from the crossing of T 117 x Igra RA 518 RR, the frequency of recombination of the marker was calculated using the GQMOL program (Cruz;Schuster, 2006).The parameters of the F 2 population were determined using co-dominant markers and an estimate method for maximum analytic verisimilitude with an LOD minimum score of 3.0 for the significance of re-combination.
Evaluation of cultivars using Dt1 and Dt2 gene markers DNA was extracted from 22 soy cultivars according to the method of McDonald et al. (1994) with modifications (Schuster et al., 2004).DNA was also extracted from samples of seeds from four individual plants of each cultivar that presented more than one pod in the apex of the stem, with the hypothesis that these plants possess the Dt2 gene or that they were heterozygote for the Dt1 gene.In total, 36 samples, which are identified in Table 1, amplified the Dt1 and Dt2 gene markers.The phenotypic information of stem growth type, which is presented in Table 1, is described by the products of the respective cultivars.The samples were amplified with markers for the Dt1 and Dt2 gene as described previously.

Results and discussion
To validate the selection with molecular markers for the Dt1 gene in soybean, the primers TFL1b-pro-F and TFL1b-pro-R (Liu et al., 2010) were used.These primers amplify a 410-bp fragment of genomic DNA in lineages with determined and undetermined growth.Only the fragments that amplify the Dt1 allele are digested into 217-and 193-bp fragments with NdeI enzyme.
Thus, we demonstrated that the evaluated plants (Figure 1  The molecular basis of the growth type was determined by Liu et al. (2010).The authors isolated two pea orthologs (Pisum sativum), GmTFL1a and GmTFL1b.Based on the soy genome and mapping analysis, it was suggested that the Dt1 gene encodes the GmTFL1b protein and that the type of soy stem growth is determined by variations in this gene.
The dt1dt1 genotype is responsible for the determined growth type, whereas the Dt1Dt1 genotype results in the undetermined growth type.Dt1dt1 causes a semi-determined phenotype (Bernard, 1972).
The contrasting bulks were amplified, and the marker (considered candidate) used in this assessment was sat_064, which results in polymorphisms between the bulks and consistency with the parenting results.
The sat_064 marker was used to individually amplify all the plants that formed the bulks (Figure 2).A possible segregation among the samples was observed.All the plants derived from crossing CD 235RR x Igra RA 518 RR and T 117 x Igra RA 518 RR were analyzed with molecular marker for Dt1 and Dt2.
The association test between the phenotypic and molecular data obtained from the CD 235RR x Igra RA 518 RR population were significant given that the phenotypic and molecular data are not independent (Table 2).However, various different results between the phenotypic and molecular evaluation were analyzed.This result was not expected; once the marker is used within the gene in the promoting area, there should be no risk for genetic recombination.In the field evaluations, some difficulty was encountered in distinguishing the plants according to the growth type based on the phenotype, which has led to erroneous classifications.In this case, we attribute the observed inconsistencies to errors in phenotype determination.Thus, the maker used for this analysis can be a very efficient tool to avoid such errors.(χ 2 = 128.88;GL = 4; P(%) = 0.001).
The obtainer of cultivar CD 235RR described its stem growth type as semi-determined (Cooperativa Central de Pesquisa Agrícola [Coodetec], 2011).In this case, this cultivar possessed the Dt1Dt1 Dt2Dt2 genotype.In addition, the crossing with 'Igra RA 518 RR' should not result in segregation of the Dt1 gene.As a result, determined lineages in the progeny of this crossing should not be observed.The evaluation of the progeny of the hybrid CD 235RR x Igra RA 518 RR revealed F 2 determined, undetermined and semidetermined plants (which segregated in the F 3 generation).Based on these results, we conclude that the cultivar CD 235RR presents the dt1dt1 genotype and a determined growth type.
In the progeny of the crossing T 117 x Igra RA 518 RR, we expected 100% of the plants to be homozygous for the Dt1 gene and both parental factors (T 117 x Igra RA 518 RR) present in this gene.All the analyzed plants exhibited the Dt1 gene in the molecular evaluation, which confirms the efficiency of the molecular marker used in this assessment.Moreover, no plants were classified as a determined type in the phenotypic analysis.
Two types of growth were expected in the offspring of the crossing T 117 x Igra RA 518 RR: undetermined and semi-determined.The T 117 cultivar harbors the Dt2 gene and an undetermined growth type, whereas the Igra RA 518 RR cultivar exhibits an undetermined type.
For cultivars of determined and semi-determined growth, the terminal gem transforms into a terminal inflorescence.In contrast, for cultivars of undetermined growth, this gem does not form, and the stem continues to develop, even after the beginning of the florescence (Müller, 1981).In general, the cultivars are characterized as a determined type and are rarely characterized as semi-determined.
The association analysis between the phenotypic and molecular data obtained from the T 117 x Igra RA 518 RR population was also highly significant, and the distribution of phenotypic and molecular data was not independent (Table 3).This result indicates that the marker sat_064 is associated with the Dt2 gene.The frequency of recombination obtained between the sat_064 marker and the Dt2 gene was 19.4% with a LOD score of 8.19.It is possible that this estimate was increased compared with the actual frequency of recombination between the gene and the marker due to the difficulty in classifying the plants phenotypically as semi-determined types.According to the work of Bernard (1972), the progeny tests occasionally indicated that semi-determined or undetermined plants were incorrectly classified.In Table 3, the increased number of mistakes between phenotypic and molecular data is observed in classes wherein the plants were phenotypically classified as semidetermined type and segregating.
To evaluate the efficiency of the markers, 22 soybean varieties were evaluated using the markers for the Dt1 and Dt2 genes.
As noted in Figure 3, all the materials corresponded to that described by the obtainers (Table 1) with regard to the genotype and the presence or absence of the Dt1 gene.The only exception was the cultivar FTS Campo Mourão RR, which was described by the obtainer as a semi-determined growth type; however, the Dt1 gene is not present this genotype.In Figure 4, all of the genotypes corresponded to that described by the obtainers (Table 1).The only exception is the cultivars FTS Campo Mourão RR, which the obtainer described as semi-determined growth; however, the Dt2 gene is not present.In addition, the Dt2 gene was not noted in the cultivar Anta 82 during molecular analysis.This result diverges from the obtainer's description of the cultivars NK 7059RR, BMX Apolo RR, BMX Potência RR and NA 5909RR.These cultivars were classified as semidetermined types, and the genotype should possess the Dt2 gene because they presented a plant with more than one pod in the apex of the stem.In the four analyses of the individual plant of each cultivar, the presence of the Dt2 gene was not confirmed, validating and confirming the information from the obtainers that indicated that these cultivars are homogeneous in regards to the absence of the Dt2 gene.
According to the obtained results, the cultivar FTS Campo Mourão RR has a dt1dt1dt2dt2 genotype and determined growth.This result is inconsistent with the obtainer's description of semi-determined growth and the Dt1Dt1Dt2Dt2 genotype.
'Anta 82,' which is described as having a semi-determined growth type, demonstrated the presence of the Dt1 gene and absence of the Dt2 gene through the use of molecular markers.This result indicates that this cultivar presents a genotype that confers a phenotype of undetermined growth.This result is substantiated by information obtained in 2011 and never published by the Coodetec.A hybrid of Anta 82 cultivar (a female parental and male parental CD 216 and M-SOY 6101 with undetermined growth type) resulted in F 2 populations wherein all of the plants presented an undetermined growth type, indicating that 'Anta 82' exhibits an undetermined growth type.

Conclusion
The association of molecular markers with the GmTFL1b gene is highly effective for the classification of soybean growth.
The sat_064 marker is associated with the Dt2 gene, which is located in Connection Group G of the joint soybean map and exhibits a recombination frequency of 19.4%.
The molecular markers for Dt1 and Dt2 genes are efficient in the description of genotypes with regard to soybean stem growth and can be used in the selection of the improvement program.
) exhibited the expected DNA profile.The cultivars T 117, Igra RA 518 RR, BMX Apolo RR, NK 7059RR and plants 1 through 6 exhibit the Dt1Dt1 genotype.The cultivars CD 214RR and CD 206RR exhibit the determined growth type and the dt1dt1 genotype.

Figure 1 .
Figure 1.Amplification with GmTFL1b-pro marker and digestion with NdeI enzyme.Plants with a semi-determined T 117 growth type and plants with an undetermined growth type: Igra RA 518 RR, BMX Apolo RR, and NK 7059 RR.Plants with a determined growth type: CD 214RR, CD 206RR.Plants 1 through 6 are obtained from a population that results from the crossing T 117 x Igra RA 518 RR, for which the phenotype is undetermined.M: 100-bp molecular weight marker.

Figure 2 .
Figure 2. Amplification with the sat_064 marker.M: 100-bp molecular weight marker; T1: Control 1 -Cultivar T 117, which exhibits a semi-determined growth type as conferred by the Dt2 gene; T2: Control 2 -Cultivar Igra RA 518 RR, which exhibits an undetermined growth type and does not possess the Dt2 gene; A: Plants with a semi-determined growth type derived from crossing T 117 x Igra RA 518 RR; B: Plants with an undetermined growth type derived from crossing T 117 x Igra RA 518 RR.

Figure 3 .
Figure 3. Amplification of DNA in soy cultures using the GmTFL1b-pro marker and digestion with NdeI enzyme.M: 100-bp molecular weight marker.

Figure 4 .
Figure 4. Amplification of soy cultivars' DNA with the sat_064 marker.M: 100-bp molecular weight marker.The Dt1 gene was absent from and the Dt2 gene was present in CD 201 and CD 219RR cultivars.This phenotype confers a determined growth type and confirms the description of the obtainer.OC 4-Iguaçu*5 and W20 are the parental stains of CD 201; OC 94-2062 and CO 2131 are the parental strains of CD 219RR.The progenitor CO 2131 possesses the CP4-EPSPS gene, which confers tolerance to glyphosate (RPRCB, 2005).This result is supported by unpublished information by Coodetec: the OC 94-2062 lineage was the originator of the CD 211 cultivar, which possesses the dt1dt1dt2dt2 genotype.The probable origin of the Dt2 gene in the CD 219RR cultivar is CO 2131, which configures itself in a cross between [OC 95-3585 X OC 95(4)-3355 X H 5566RR]. Here, OC 95-3585 and OC 95(4)-3355 are brother lineages of OC 95(4)-2422 and are responsible for the origin of the CD 201 cultivar.The origin of the Dt2 gene in CD 201 and CD 219RR cultivars likely involves the common ancestor Ocepar 4-Iguaçu.

Table 1 .
Cultivars of soy and their respective stem growth type phenotypes as described by the obtainers.

Table 2 .
Contingency table with molecular and phenotypic data obtained in the analysis of F 2:3 CD 235RR and Igra RA 518 RR populations.

Table 3 .
Contingency table with molecular and phenotypic data obtained in the analysis of the F 2:3 population obtained from the crossing of T 117 and Igra RA 518 RR.