Molecular identification and characterization of soybean mosaic virus isolates from Serbia

Summary: Certified and non-certified seed of the soybean cultivar Dukat showing symptoms of mottling were collected at the preharvest stage from soybean fields in the location Tamiš (South Banat District, Serbia) in 2021. To identify the presence of soybean mosaic virus (SMV), a total of 10 samples, five of each type of seed, were selected and tested by RT-PCR using primers SMVCPF/SMVCPR, which amplify a fragment of the coat protein (CP) gene. The molecular assay confirmed the presence of SMV in all tested seed samples. For further analyses, the CP and P1 genome regions of two selected isolates (DS-21 and NDS-21, representing certified and non-certified seed, respectively) were partially sequenced. The Serbian SMV isolates shared very low nucleotide diversity and were mutually closely related based on the CP (99.8% nt/99.2% aa identities) and P1 (99.7% nt/ 99.6% aa identities) regions. Phylogenetic analyses based on both genomic regions showed that the two Serbian SMV isolates originating from certified and non-certified seed belonged to one cluster, and consequently to the same SMV subpopulation. Since this is the first molecular characterization of SMV in Serbia, further research should be focused on collecting samples from different locations and of different varieties, and on determining the variability of SMV population in Serbia based on whole-genome analysis. This research will contribute to better understanding of the epidemiology of this pathogen with the ultimate goal of developing and implementing appropriate control measures


Introduction
Soybean [Glycine max (L.) Merrill], family Fabaceae, is an annual self-pollinated diploid legume native to East Asia.It is widely grown for its edible bean that has numerous uses and represents the most important source of edible oil and proteins.The area planted with soybean in Serbia has increased drastically (from 131 thousand hectares in 2005 to 230 thousand in 2019) over the past fifteen years, and the tendency is for more and more land being dedicated to soybean cultivation (www.stat.gov.rs).In 2018, Serbia had a record high soybean production season with an average yield of 3.3 metric tons per hectare, while total production reached approximately 760,000 tons, 50% higher than the 2017 figure (www.stat.gov.rs).This ranks Serbia as the largest soybean producer in the Western Balkans with more than half of the regional output of soybean.
Soybean mosaic virus (SMV) is a member of the genus Potyvirus (family Potyviridae) and it is the most common and prevalent viral pathogen of soybean worldwide.The virus causes deterioration in seed quality, reduction in seedling viability and vigor, seed coat mottling, flower abortion, and reduction in pod set, seed number and seed size (Gunduz et al., 2004;Hobbs et al., 2003;Mandhare & Gawade, 2010) and represents a major threat to soybean industry (Malapi-Nelson et al., 2009).It can cause serious yield losses from 35% to 50%, which may turn out to be as fatal as 100% in severe outbreaks (Arif & Hassan, 2002;Liao et al., 2002).Moreover, SMV-infected seed can lead to an increased protein and decreased oil.
Symptoms induced by SMV include severe mosaic, mottling, rugosity and necrosis on leaves of many soybean varieties (ICTVdB Management, 2006;Bashar, 2015).Plants infected at the early stage of growing show shortened internodes (Chen & Choi, 2006), while infected seeds have a radial brown stripe (Hill et al., 1987).Furthermore, the interaction between SMV strains and infected soybean cultivars results in a variety of symptoms (Seo et al., 2009) and may induce more severe damage when soybean is mix-infected with other viruses due to synergistic effects (Chen & Choi, 2006;Malapi-Nelson et al., 2009).SMV can be transmitted to soybean plants via infected seeds (1-68%) or by any of over 30 aphid species in a non-persistent manner (Hill, 1999;Hill et al., 2001;Balgude et al., 2012).SMV is transmitted through seed in about 50 plant species in the families Fabaceae, Amaranthaceae, Chenopodiaceae, Passifloraceae, Schropulariaceae and Solanaceae (Hill, 1999).
Several classification systems for grouping SMV isolates have been proposed independently around the world.Numerous SMV isolates have been classified into seven distinct strains (G1 to G7) with two more added in the United States later, G7A and C14, based on symptoms developed on various resistant soybean cultivars (Cho & Goodman, 1979;Buzzell & Tu, 1984;Chen & Choi, 2008).This classification system is by far the most recognized and widely used.Similarly, five strains (A-E) have been identified in Japan, while SMV isolates from South Korea have been classified into 11 groups: G1-G7, SMV-N, G5H, G7a, and G7H (Seo et al., 2009;Li et al., 2010).In China, 22 strains (SC1-SC22) have been established based on geographical regions of their origin and specific soybean responses (Takahashi et al., 1980;Wang et al., 2002;Guo et al., 2005;Li et al., 2010).The relationship between G strains in the United States and SC strains in China has not been fully clarified yet.
Over the last several years, numerous studies have been conducted in order to unravel the phylogenetic relationship among virus isolates, constructing the first SMV phylogenetic trees for the full-length genome or for its individual gene (P1, HC-Pro and CP) sequences, which allowed arranging SMV strains and isolates into distinct phylogenetic groups and subgroups (Domier et al., 2003;Choi et al., 2005;Chen & Choi, 2006;Sherepitko et al., 2011;Jeżewska et al., 2015).However, a unique and official way of grouping of SMV isolates has not yet been established.
Despite the economic importance of soybean in Serbia, the disease caused by SMV or other soybean-infecting viruses was not surveyed again after its first record in 1964 (Nikolić & Stakić, 1964).
In the present study, certified and non-certified soybean seed stocks were surveyed for the presence of SMV.Two SMV isolates from seed stocks in the South Banat region of Serbia were molecularly identified and characterized.We analyzed their phylogenetic relationships based on CP (coat protein) sequences which are involved in cell-to-cell and long-distance movement, as well as in transmission by aphids and seed, and on P1 (protein 1 gene), which has a role in symptom development and host adaptation (Hajimorad et al., 2018).Serbian SMV isolates were compared with other known isolates or strains in order to trace their origin.The information obtained from this study will be used as a starting point for further research of the SMV population in Serbia.

Seed sampling
Certified and non-certified soybean seed with symptoms of mottling were collected during the 2021 growing season from soybean fields at the preharvest stage of the cultivar Dukat in the location Tamiš (South Banat District, Serbia) to test seed for the presence of soybean mosaic virus (SMV).Samples of symptomatic seed were stored at temperatures of 22-25°C before use for RNA extraction.

RT-PCR detection and sequence analysis based on the CP gene
In order to identify the presence of SMV, total RNAs were extracted from 100 mg of symptomatic soybean seeds of 10 collected samples, five samples each of certified and non-certified seed, using a cetyltrimethylammonium bromide (CTAB) protocol (Li et al., 2008), and they were subjected to RT-PCR.RT-PCR was performed using the One-Step RT-PCR Kit (Qiagen GmbH, Germany) with the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) and the primer pair SMVCPF/SMVCPR (Jeżewska et al., 2015), previously described to amplify the partial sequence of coat protein (CP) gene.Tissue sample from healthy soybean seeds and RNase-free water were used as negative controls.The RT-PCR was performed in a thermal cycler (Biometra, T-1 Thermocycler) and the reaction mixture included 5 µl of 5x Qiagen OneStep RT-PCR Buffer, 1 µl of each of four 400 µM dNTPs, 1.5 µl of each 0.6 µM primer, 1 µl of RT-PCR enzyme mix and 1 µl of extracted RNAs.The volume of mixture was adjusted to 25 µl using RNase-free water.Reverse transcription was performed at 50°C for 30 min and initial PCR denaturation step at 95°C for 15 min, followed by 35 cycles consisting of a denaturation step of 30 s at 94°C, primer annealing for 30 s at 55°C, and extension for 1 min at 72°C.The final extension was performed at 72°C for 10 min.PCR products were analyzed by 1% agarose gel electrophoresis, stained with Midori Green Advance DNA Strain (Nippon Genetics Europe GmBH, Germany) and visualized under a UV transilluminator.The size of fragments was determined by comparison with MassRuler TM DNA ladder, Mix (Fermentas Life Sciences GmbH, Lithuania).
The amplified products derived from two isolates (DS-21 and NDS-21, from certified and noncertified seeds, respectively) were purified with QIAquick PCR Purification Kit (Qiagen) and sent for sequencing in both directions on an automated sequencer (Microgen-Europe BV) using the same primers as for amplification.The nucleotide sequences of amplification products were deposited in the National Center of Biotechnology Information (NCBI) GenBank database, and assigned accession numbers (Table 1).All generated sequences were compared with each other, and with previously reported isolates available in the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST/) by calculating nucleotide (nt) and deduced amino acid (aa) identities using the ClustalW program (Thompson et al., 1994) implemented in MEGA7 software (Kumar et al., 2016).A p-distance model was applied for nucleotide (nt) and deduced amino acid (aa) sequence analyses to calculated possible variation among Serbian isolates.

RT-PCR amplification and sequence analysis based on the P1 gene
Further characterization of the isolates DS-21 and NDS-21 was performed using RT-PCR with specific primers SMVP1F/SMVP1R (Sherepitko et al., 2011) previously described to amplify the protein 1 (P1) gene.The components of reaction mix were identical to those described for amplification of the CP gene.Cycling parameters were the following: reverse transcription at 50°C for 30 min and initial PCR denaturation step at 95°C for 15 min, 40 cycles consisting of a denaturation step of 30 s at 94°C, primer annealing for 30 min at 55°C, and extension for 1 min at 68°C.Sequencing and sequence analyses were performed as described above.

Phylogenetic analyses
Phylogenetic trees based on the CP and P1 sequences were constructed using Neighbor-Joining algorithm implemented in MEGA7 program with 1000 bootstrap replicates, and bootstrap values <50% were omitted.All sequences were trimmed to the length of the shortest fragments, 767 bp for P1 and 448 bp for the CP gene.The P1 and CP gene sequences of watermelon mosaic virus (WMV) served as an outgroup (GenBank Acc.No. AY437609) for construction of a corresponding phylogenetic tree.

Seed symptoms
In order to examine a virus infection observed over the 2021 growing season, samples of seed (Figure 1) showing symptoms of mottling were collected from a breeding plot on the location Tamiš (South Banat District) at the preharvest stage of Dukat soybean cultivar, and tested for the presence of soybean mosaic virus.According to Rupe & Luttrell (2008), Langham & Strunk (2019) and Suarez et al. (2021), symptoms such as seed mottling, the most common of which has been named "hilum bleeding" for being caused by hilum color spreading towards seed coat, are often associated with SMV-infected seed, but SMV infection could also reflect in smaller, less vigorous but symptomless seeds or beans.Seed infection can be very high, depending on soybean cultivar and virus strain, varying between 5 and 75% (Hill, 1999).So, the presence and high frequency of SMV in soybean crops in Serbia were not entirely surprising considering that SMV is one of the most common and serious virus diseases in soybean worldwide according to many authors (Li et al., 2010;Cui et al., 2011;Suarez et al., 2021).

Detection and identification SMV isolates based on CP gene sequences
The presence of SMV was confirmed by target cDNA amplification of fragments of predicted size of 469 bp using the specific pair of primers SMVCPF/SMVCPR (Sherepitko et al., 2011).The pair of primers used in this study for amplification of the partial CP gene proved to be successful for routine molecular detection of SMV in diagnostic laboratories (Wang & Ghabrial, 2002;Sherepitko et al., 2011;Jeżewska et al., 2015).The identities of RT-PCR products derived from the isolates DS-21 and NDS-21, representing partial nucleotide sequences of the CP gene, were confirmed by BLAST analyses which revealed high nt similarity of two Serbian SMV isolates with the respective virus sequences of the CP region available in GenBank.The isolate DS-21 showed the highest nt identity of 100% with seven (AH008459, MW822167-68, ON013906, KF297335, KF135489-90), while NDS-21 demonstrated the same identity with eight (MW464638, KF135467-69, KF135471 and KF135479-81) A B SMV isolates.The sequences generated in this study were also compared with each other in order to characterize their genetic variability, and multiple nucleotide and deduced amino acid sequence comparisons implicated a high nt similarity of 99.8% (99.2% aa identities) for this region in the SMV population in soybean in Serbia.

Detection and identification SMV isolates based on P1 gene sequences
The SMVP1F/SMVP1R primer pair (Jeżewska et al., 2015) was also used for further molecular characterization, amplifying a 756 bp fragment covering a part of the P1 gene of two selected isolates.This part of the SMV genome was chosen because P1 is consider to be the most variable and informative protein, which is why it is important for identification and characterization of SMV strains, as well as for determining the evolutionary relationship of Serbian isolates with those from other parts of the world (Suarez et al., 2021).After purification, the RT-PCR products derived from the P1 genomic region were also directly sequenced in both directions using the same primers as in RT-PCR, and were submitted to the GenBank database (Table 1).BLAST results revealed that P1 sequences of the two Serbian SMV isolates, DS-21 and NDS-21, had high nt similarity of 99.74% with Chinese SMV isolate HLJSB001 (GenBank ACC.No. KX834323) and Iranian SMV isolate Ar16 (KF135468), respectively.Sequences obtained in this study based on the P1 region were also compared with each other in order to characterize the genetic variability among them, and multiple nucleotide and deduced amino acid sequence comparison implicated a high nt similarity of 99.7% (99.6% aa identities).

Molecular characterization of Serbian SMV isolates
A neighbor-joining tree (Figure 2) of 34 SMV isolates, reconstructed using a partial nucleotide sequence of the CP gene, showed that both selected Serbian isolates, DS-21 and NDS-21, were grouped into the same cluster with isolates from Iran (Ar13, Ar33, Go11 and Lo3) and Poland (M).Also, the nt and aa sequence alignments of the central part of the CP region was analyzed by MEGA 7 to compare the Serbian isolates with other known SMV strains used for phylogenetic analyses.These alignments showed that the CP nucleotide sequence of the DS-21 and NDS-21 isolates shared 91.5 to 100% identity (74 to 100% aa identity) with the sequences of other SMV isolates (Table 2).The isolate DS-21 shared maximum nt and aa similarity of 100% with isolates from Iran (Ar33 and Lo3), while NDS-21 showed 99.8% similarity (99.2% aa identity) with Iranian isolates (Ar33, Go11 and Lo3) (Таble 2).These data indicated that most nucleotide substitutions in this part of the CP region were synonymous, which is not surprising because the regulation of viral RNA amplification, and the requirement to assemble stable virions, impose intense purifying selection pressures on the CP sequences (Urcuqui-Inchima et al., 2001;Domier et al., 2003;Sherepitko et al., 2011).The neighbor-joining tree (Figure 3) was contracted based on partial sequences of the P1 gene of the SMV isolates determined in this study and the selected sequences of 42 previously characterized SMV isolates were retrieved from the GenBank database (Table 1).The phylogenetic tree revealed the grouping of Serbian isolates into one cluster with Iranian (Ar33 and Lo3), Ukrainian (UA1Gr) and Polish (M) isolates, supported by high bootstrap value.The alignment of the partial P1 nt sequences showed that the similarities of the SMV isolates used in phylogenetic analysis varied from 87.8 to 99.9% (81.7 to 99.6%).The isolate DS-21 showed maximum nt identity of 99.6% (99.6% aa identity) with the isolate M from Poland and UA1Gr from Ukraine, while the isolate NDS-21 showed identity with isolate Ar16 from Iran (Table 2).The P1 protein is known to be the least conserved region of the entire polyprotein of potyvirus and to have higher level of aa tolerated variability compered to CP-coding regions with more non-synonymous substitutions (Urcuqui-Inchima et al., 2001;Choi et al., 2005;Maroof et al., 2008;Seo et al., 2009;Sherepitko et al., 2011).Topology of the tree constructed using nucleotide sequences of the part of SMV capsid protein and P1 gene was very similar to that recently described by Jeżewska et al. (2015).The results of our phylogenetic analysis, consistent with data of many other authors (Sherepitko et al., 2011;Seo et al., 2009), indicate that there is not a clear correlation between genetic differentiation and geographical origin of SMV isolates, which may be due to recombination events between various SMV strains (Gagarinova et al., 2008;Seo et al., 2009).Also, previous studies of SMV populations revealed no clear relationships between the phylogeny of the isolates and their pathogenicity on various soybean cultivars (Seo et al., 2009).The same authors suggested that the main reasons for phylogenetic inconsistency in geographical clustering were the fact that not much genetic drift has occurred in the SMV population, and human trading activities involving infected soybean seed.This is the first molecular characterization of SMV isolates in Serbia.Our findings show that commercial soybean seeds, both non-certified and certified, are not SMV-free.The transmission of SMV via infected seed is very important in the epidemiology of the disease as it enables constant maintenance of inoculum and efficient spread of SMV through international seed trade (Koning et al., 2003;Domier et al., 2007;Rupe & Luttrell, 2008;Bashar, 2015;Langham & Strunk, 2019).Although our investigation included only one location, it presents a basis for further characterization of SMV genetic population in Serbia, and such information should be taken into account considering that the virus caused epiphytotic disease and reduction in seed quality, as well as a decrease in soybean production in Serbia and other European countries over the past two decades.Therefore, further research in Serbia should focus on collecting samples from different locations and varieties and on determining the variability of SMV population in Serbia based on whole-genome analysis.

Conclusion
Over the last fifteen years, the area planted with soybeans [Glycine max (L.) Merrill] has increased drastically in Serbia.Soybean mosaic virus (SMV) is the most common and prevalent viral pathogen of soybean worldwide which poses a major threat to the soybean industry.It causes serious yield losses of 35% to 50%, reaching as much as 100% in severe outbreaks (Arif & Hassan, 2002;Liao et al., 2002).This study presents the first molecular identification and characterization of two SMV isolates from seed stocks in the South Banat region of Serbia.Serbian SMV isolates showed very low nucleotide diversity and were closely related based on the CP and P1 region, and phylogenetic analyses indicated that despite their different origin, from certified and non-certified seed, they belong to the same SMV subpopulation.The information obtained in this study will be used as a starting point for further research of the SMV population in Serbia.

Figure 1 .
Figure 1.Seed mottling on certified (A) and non-certified (B) soybean seeds

Figure 2 .
Figure 2. Neighbor-joining tree based on nucleotide sequences of 36 SMV CP gene isolates using French WMV-FR isolate as the outgroup sequence.The phylogram was generated with MEGA 7.0 using the p-distance model.Bootstrap analysis was performed with 1000 replicates and bootstrap values (>50%) are shown next to relevant branches.Serbian SMV isolates are marked blue and bolded

Figure 3 .
Figure 3. Neighbor-joining tree based on nucleotide sequences of 44 SMV P1 gene isolates using a French WMV-FR isolate as the outgroup sequence.The phylogram was generated with MEGA 7.0 using the p-distance model.Bootstrap analysis was performed with 1000 replicates and bootstrap values (>50%) are shown next to relevant branches.Serbian SMV isolates are marked blue and bolded

Table 1 .
Sequences of P1 and CP genes of soybean mosaic virus isolates available in the GenBank database were used for phylogenetic analyses a Isolate from this study; b Reference source included no host name

Table 2 .
Identity (%) of P1gene and CP gene nucleotide and amino acid sequences between Serbian SMV isolates and virus strains used for phylogenetic analyses a Corresponding sequences not available in GenBank database