Wenwen Song,Ruping Yng,Xiushi Yng,Shi Sun,Srinivs Ro Mentreddy,Bingjun Jing,Tingting Wu,Shiyn Tin,Enoh Spey,Cunxing Wu,Wensheng Hou,Guixing Ren,*,Tinfu Hn,*

aMOA Key Laboratory of Soybean Biology,Institute of Crop Sciences,Chinese Academy of Agricultural Sciences,Beijing 100081,China

bInstitute of Dryland Agriculture,Gansu Academy of Agricultural Sciences,Lanzhou 730070,Gansu,China

cDepartment of Biological and Environmental Sciences,Alabama A&M University,AL 35762,USA

Keywords:Soybean Geographic distribution Bioactive components Weather factors

ABSTRACT Bioactive components are partially responsible for the nutritional and health benefits of soybeans.Four major bioactive components:isoflavones,oligosaccharides,phospholipids,and saponins,were quantified in 763 soybean samples collected from widely distributed regions across China from 2010 to 2013.A majority of the tested bioactive components showed generally declining trends from the north(high latitude)to the south(low latitude).A positive relationship between total oligosaccharides(TO)and altitude was observed.Total isoflavones(TI),phospholipids(TP)and TO were negatively correlated with cumulative temperature above or equal to 15°C(AT15)and mean daily temperature(MDT),but positively correlated with diurnal temperature range(DTR)and hours of sunshine(HS).Total saponins(TS)were negatively correlated with MDT but positively correlated with rainfall(RF),whereas TO were negatively correlated with RF.Path-coefficient analysis showed that,besides genotype differences,temperature and HS during the reproductive period influenced TI and TP contents,while temperature and RF influenced TS and TO.The effects of weather factors on soybean bioactive components in diverse regions of China were characterized.These findings will be helpful in promoting soybean production for functional food purposes.

1.Introduction

Soybean[Glycine max(L.)Merr.]bioactive components such as isoflavones,oligosaccharides,phospholipids,and saponins offer nutritional and health benefits to both humans and animals[1–6].Isoflavones have been reported to be helpful in the prevention of cardiovascular diseases,cancer and obesity[6,7].Oligosaccharides are important polymeric carbohydrates that are believed to have the physiological functions of reducing cholesterol and alleviate constipation[8].Soluble sugars,especially sucrose,are the major sources of energy for fermentation and contribute to the sweetness of soymilk and tofu[9].Phospholipids are important constituents of the membrane bilayer,which plays a role in biological functions such as memory,associated with cellular signal transduction[4,10].Saponins are excellent emulsifiers and foaming agents that are reputed to show hypocholesterolemic and antitumorigenic activities[11,12].Soybean saponins showed an inhibitory effect on the infectivity and cytopathic activity of the human immunodeficiency virus(HIV)[13].

A b b r e v i a t i o n s T I t o t a l i s o f l a v o n e s T O t o t a l o l i g o s a c c h a r i d e s T P t o t a l p h o s p h o l i p i d s T S t o t a l s a p o n i n s A T 15 c u m u l a t i v e t e m p e r a t u r e a b o v e o r e q u a l t o 1 5°C M D T m e a n d a i l y t e m p e r a t u r e D T R d i u r n a l t e m p e r a t u r e r a n g e H S h o u r s o f s u n s h i n e R F r a i n f a l l N E S R N o r t h e a s t S p r i n g P l a n t i n g S u b r e g i o n N W S R N o r t h w e s t S p r i n g P l a n t i n g S u b r e g i o n H H H R H u a n g-H u a i-H a i R i v e r V a l l e y S u m m e r P l a n t i n g R e g i o n S M C R S o u t h M u l t i p l e C r o p p i n g R e g i o n C V c o e f f i c i e n t o f v a r i a t i o n M D m a l o n y l d a i d z i n M G m a l o n y l g e n i s t i n P I p h o s p h a t i d y l i n o s i t o l P C p h o s p h a t i d y l c h o l i n e P E p h o s p h a t i d y l e t h a n o l a m i n e U V B u l t r a v i o l e t r a d i a t i o n b

Contents of bioactive components in soybean seeds are not only genetically controlled but influenced by environmental conditions[14,15].In most previous studies,one or several bioactive component profiles in different soybean cultivars were evaluated for selecting genotypes with superior healthpromoting characteristics[14,16,17].Little is known about possible influences of environment on contents of bioactive componentsin soybean cultivars.Severalreportshave focused on the influence of planting time and location on soybean seed bioactive components.For example,it was reported that genotype,location and year of planting affected the isoflavone content of soybeans and planting year had a stronger effect on isoflavone content than location and genotype[14,17].The contributions of planting location and time to large variation of oligosaccharides in soybean seeds have been reported[18].Planting dates also showed consistent influence on saponin B concentrations in soybean[19].The effects of planting time and location on bioactive components in these studies may be attributed to differences in temperature or other weather factors during specific soybean growth stages.

Previous studies of the relationships between weather factors and soybean bioactive components have usually been conducted in pots under controlled environments or field conditions,with only a few factors considered or tested on a limited regionalscale [20–22].In recentyears,multienvironmental studies have been conducted for assessing the effects of different weather factors on the soybean quality.For example,Carrera and Dardanelli[23]used soybeans grown in 76 environments in Argentina to quantify the relationship between soybean TI,temperature and solar radiation under water-deficitconditions.Songetal.[24]evaluated the influence of weather factors on soybean protein and oil contentsandcomposition usingextensivehigh-density sampling in China[25].These studies revealed that multienvironmental field conditions were necessary for assessing environmentaleffectsonsoybeanseedcompositionto capture a wide range of weather variation.

Soybean has been consumed as food in China for many centuries,and increasing its nutritional value is a consistent target of breeders and producers[25].Soybean bioactive components such as isoflavones have been investigated only for soybean cultivars of certain regions in China[26].Wu et al.[27]identified the temporal-spatial trends of 35 nutritional andbioactivecomponentsof89widely-grownsoybean cultivars over a period of nine decades(1923–2007)in China,but cultivars used in the study were adapted mainly to northeast China.A full-scale investigation of the variation in bioactive components in Chinese soybean cultivars is desirable.Soybean is cultivated over a broad ecological region,across 33°of latitude and 57°of longitude in China[24].This wide distribution affords the opportunity to obtain a comprehensive understanding of the relationship between weather conditions and soybean composition.The objective of the present study was to quantify the main bioactive components in 763 soybean samples and evaluate the effects of weather factors on them.

2.Materials and methods

2.1.Soybean materials

A total of 763 soybean samples were collected from four soybean-producing regions:the Northeast Spring Planting Subregion (NESR),NorthwestSpringPlantingSubregion(NWSR),Huang-Huai-Hai River Valley Summer Planting Region(HHHR)and South Multiple Cropping Region(SMCR)[25]in 2010–2013.For collecting the samples,the soybean production area in China was divided into two-degree longitude–latitude cell grids and 30 soybean experimental stations of the China Agriculture Research System(CARS)were asked to collect samples randomly in the corresponding cell grid and an adjacent cell grid each year during 2010–2013.In each site(field),the soybeans were grown following standard local practices.Five spots were chosen for sample collection using a five-point sampling method.In each spot(nearly 1 m2),seeds of approximately 20 soybean plants were collected.All the seeds were pooled,and 500 g of the seeds were taken as a sample for bioactive component analysis.The sampling sites covered 28 of the 34 provinces(municipalities or autonomous regions)of China,with a geographical range from 19.71°N to 53.48°N and from 77.27°E to 134.32°E.The geographical conditions of each site and cultivar information are presented in Table 1.Among the 763 samples,269,187,101,and 206 samples werecollectedin 2010,2011,2012,and2013,respectively.Because the sampling site in each cell grid was randomized each year,the cumulative 763 sampling sites in four years were not repeated,so that only one sample was collected at each site.Given that some cultivars were widely grown,a total of 314 cultivars were sampled among the 763 samples.Complete information for the 763 samples,including geographic location,cultivar,sowing-season ecotype,sowing date,and plant density are presented in a supplementary file(Table S1).Seed samples were ground and passed through a 60-mesh sieve in a Cyclotec 1093 Sample 126 Mill(FOSS Tecator,Hoganas,Sweden)for bioactive component analysis.Isoflavones and oligosaccharides of all 763 samples collected from 2010 to 2013 were determined annually,whereas the 269 samples collected in 2010 were used for phospholipid and saponin determination.All samples were processed within onehalfyearafterharvestto preventdegradation of components.

2.2.Determination of bioactive components

Isoflavones were extracted following Sakthivelu et al.[3].Five isoflavone standards: daidzin, glycitin, genistin,malonyldaidzin(MD),and malonylgenistin(MG)(Indofine,Hillsborough,NJ,USA),were used to prepare calibration curves.Oligosaccharides were determined following Wu et al.[27].Three oligosaccharide standards:sucrose,raffinose and stachyose(Sigma Chemical Co.,St.Louis,MO,USA),were used for preparing calibration curves.Phospholipids were determined following Lee et al.[28]with some modifications.Three phospholipid standards:phosphatidylinositol(PI),phosphatidylcholine(PC),andphosphatidylethanolamine(PE)(Sigma)were used in the preparation of calibration curves.The saponin contents were determined by HPLC following Yang et al.[29].Five saponin standards:saponin I,saponin II,saponin αg,saponin βg,and saponin βα(ChromaDex,Irvine,CA,USA),were used to prepare calibration curves.

2.3.Weather data

Weather records for the soybean reproductive period for each sampling site were obtained from the closest national-level weather stations in the China Climate Data Sharing Service System(http://cdc.nmic.cn/home.do).The cumulative temperature above or equal to 15°C(AT15),mean daily temperature(MDT),diurnal temperature range(DTR),rainfall(RF),and hours of sunshine(HS)for each site were calculated separately following Song et al.[24].The weather conditions of the major soybean-planting regions are shown in Table 1.

2.4.Statistical analysis

All the data were analyzed using SPSS 19.0 statistical software(IBM Corp.,Armonk,NY,USA).Each of the bioactive components was determined in duplicate.The weather factors,the major traits TI,TO,TS,and most of their components,showed normal distributions.Some of the traits including daidzin,saponin II,saponin αg,and saponin βg exhibited an approximately normal distribution with a single central peak of frequency.Multiple comparisons of means were performed by Kruskal-Wallis test.Differences were considered significant when a P value was＜0.05.Pearson's correlation coefficients were calculated to determine the significance of the relationship between soybean bioactive components and weather factors.The coefficients were further analyzed by path coefficient analysis[30]to separates the coefficients into direct and indirect effects of weather factors on the bioactive components via alternative characters or pathways.Since samples collected from different years and locations were independent of one a other,correlation and pathway coefficient analysis were performed for multiple(four)years in addition to the analysis for individual years.

3.Results

3.1.Geographical differences in bioactive components

The means of total isoflavones(TI),oligosaccharides(TO),phospholipids(TP),and saponins(TS)of the 763 soybean samples were 2.17,90.74,18.12,and 4.30 mg g-1,respectively(Table 2).The coefficient of variation(CV)was high and around 30%for TI and TS,while it was low and＜20%for TO and TP.The contents of TI,TP and TS were all highest in NESR.HHHR showed the lowest TI and SMCR the lowest TP and TS.Samples collected from NWSR and SMCR had the highest and lowest TO contents,respectively.Among the isoflavone components,daidzin,glycitin,genistin,and MD were higher in NESR than in the other regions,showing similar trends to that of TI.Contents of sucrose and raffinose showed similartrends to that of TO,whereas stachyose showed different trends,with contents highest in SMCR but lowest in HHHR.For the phospholipid components determined,PI and PC displayed similar trends consistent with that of TP,whereas the content of PE was highest in HHHR and lowest in NESR.The contents of saponin I,saponin II,and saponin αg were highest in NWSR and lowest in HHHR,but soybeans grown in HHHR contained the highest contents of saponin βg and saponin βα among the four regions.

Table 1–Geographic characterization,cultivar information,and weather conditions during the reproductive phase of soybean for the four major soybean-producing regions in China.

Table 2–Statistical variation ofbioactive componentsofsoybean samples collected from fourregions inChina.

Table 3–Correlation coefficients for contents of soybean bioactivecomponentswith longitude,latitude,and altitude of sampling sitesa.

Contents of all of the bioactive components in our study,namely TI,TO,TP,and TS,correlated positively with latitude(Table 3),and showed an increasing trend from south to north of China.Except for TO,all the bioactive components showed positive correlations with longitude,indicating a general decreasing trend from the east to the west of China.TO was negatively correlated with longitude.Interestingly,a positive relationship between TO and altitude was observed.However,the other three bioactive components did not show such a correlation.

3.2.Correlations between weather factors and bioactive components

The correlations of TI and TO with weather factors were similar in different years,differing only in 2012(Table 4),and were not statistically significant in 2012.Accordingly,the overall results of the four years were used.It was concluded that TI,TO,and TP were negatively correlated with AT15,and MDT(P＜0.05)but positively correlated with DTR and HS(P＜0.01;Table 4)on a national scale.TO was negatively correlated with RF(P＜0.01).Neither TI nor TP showed a significant correlation with RF.TS was negatively correlated with MDT (P＜0.05)and positivelycorrelated with RF(P＜0.01),and showed no significant correlation with AT15,HS,and DTR.Similar to TI,daidzin and genistin were also negatively correlated with AT15,and MDT(P＜0.01)but positively correlated with HS and DTR(P＜0.01;Table 5).MD showed a positive correlation with HS and DTR(P＜0.05).Sucrose and raffinose showed a negative correlation with MDT and RF(P＜0.05),and a positive correlation with HS and DTR(P＜0.01;Table 5),which was consistent with TO.Raffinosecontentwasnegatively correlated with AT15(P＜0.05),whereas no such correlation was observed for sucrose.PC and PI showed a relationship that was similar to that between TP and weather factors.They were negatively correlatedwith AT15andMDT (P＜0.01)butpositivelycorrelated with DTR and HS(P＜0.05).In contrast,PE showed a positive correlation with AT15and MDT(P＜0.01)but a negative correlation with HS(P＜0.05).Consistent with TS,saponin I and saponin αg showed negative relationships with MDT(P＜0.05);inconsistently,they showed a significant negative correlation with RF(P＜0.01).saponin I,saponin II,and saponin αg were positively correlated with DTR and HS(P＜0.01),whereas TS showed no relationship with DTR and HS.Like TS,saponin βg and saponin βα showed significantly positive correlations with RF(P＜0.01);however,they were not correlated with AT15,MDT,or DTR.

Table 4–Correlation coefficients between total amounts ofbioactivecomponentsandweatherfactorsona national scale in Chinaa.

3.3.Direct and indirect effects of weather factors on bioactive components

Path-coefficient analysis was used to identify relationships among weather factors and assess their relative effects on soybean bioactive components(Table 6).Among the weather factors,MDT and HS showed dominant and direct effects on soybean TI content.MDT showed a negative effect,whereas HS showed a positive effect.AT15,DTR,and RF exerted effects through an association with MDT and HS.MDT,DTR,and RF showed direct effects on TO,and their order was as follows:DTR＞RF＞MDT.MDT and RFshowed negative effects,whereas DTR showed a positive effect.For TP,AT15and HS showed greater direct effects than other factors.AT15showed a negative effect,whereas HS showed a positive effect.Because TS showed correlations only with MDT and RF,thedirect and indirect effects of weather factors on TS were not analyzed.It was concluded that TI,TP,and TS contents were affected predominantly by temperature and HS,whereas TO was affected mainly by temperature and RF.

Table 5–Correlation coefficients between compositions of bioactive components in soybeans and weather factorsa.

3.4.Major weather factors affecting bioactive components

In view of the ecological diversity of different soybeanproducing areas in China,the regional variations of relationships between bioactive components and weather factors were analyzed(Table 7).In the NESR,TI showed a significant correlation with DTR and HS(P＜0.05).TO was negatively correlated with DTR(P＜0.01),but positively correlated with HS and RF(P＜0.01).TP showed negative correlations with AT15,MDT and RF(P＜0.01).TS was negatively correlated with DTR and HS(P＜0.01)but positively correlated with RF(P＜0.01).In the NWSR,TI was positively correlated with HS(P＜0.01)but negatively correlated with RF(P＜0.01).TO was negatively correlated with AT15,MDT and RF(P＜0.05)but positively correlated with DTR and HS(P＜0.05).TS showed a negative correlation with HS(P＜0.05).In the HHHR,TI showed a positive correlation with RF(P＜0.01).TO wasnegatively correlated with AT15,MDT and RF(P＜0.05),but positively correlated with DTR(P＜0.05).Neither TP nor TS showed significant relationships with weather factors.In the SMCR,the relationship between bioactive components and weather factors was weaker than in other regions.Only TI and TO showed negative correlations with MDT(P＜0.05).It could be concluded that the major limiting factors for majority of the bioactive components were DTR,HS,and RF in the NESR;HS and RF in the NWSR;RF in the HHHR;and MDT in the SMCR.

Table 6–Path coefficients of soybean bioactive components and weather factors.

Table 7–Correlations between soybean bioactive components and weather factors in four regionsa.

4.Discussion

4.1.Contents of soybean bioactive components in China

In the present study,the mean TI in Chinese soybean cultivars was similar to that of soybeans in America[31],but higher than that of soybeans from tropical countries such as Brazil and Ecuador[32,33].This finding could have been due to the genetic diversity and environmental influences among these regions.Soybeans grown in a high-latitude region with lower temperature may contain higher total isoflavones[34].Our results for sucrose and raffinose contents were similar to those reported by Hartwig et al.[35],Trugo et al.[36],and Hou et al.[9],but the stachyose content was lower than the reported values.Given that no significant correlations were found between soybean stachyose and weather factors,the genetic diversity of our sample may account for this difference.Among the saponin compositions,the mean saponin αg,saponin βg,and saponin βα contents were higher than those reported by Tsukamoto et al.[37].Such a difference could be due to the diversity in planting conditions and/or differences in methods used in the two studies.The total content and percentage composition of phospholipids were similar to those reported by Racicot and Handel[38].

4.2.Effects of temperature on bioactive components

The decreasing trend of the four bioactive components from north to south China was consistent with that of crude oil content in soybean[24,39].Temperature has been reported to be the key factor affecting soybean protein and oil content[24].Weather factors in the high altitude regions,such as low temperature and limited precipitation,could be beneficial to the formation and accumulation of bioactive components.High temperature during soybean seed development significantly reduced the concentration of isoflavones by downregulating key enzymes in isoflavone synthesis,including phenyl ammonia lyase,chalcone synthase,and chalcone reductase[37,40–42].

A negative correlation between sucrose and MDT was also reported by Kumer et al.[43].Similarly,Guo and Oosterhuis[44]found that low-temperature stress resulted in an increase in the accumulation of seed sucrose content in hydroponically grown soybean plants.Likewise,Wolf et al.[45]reported this correlation in a phytotron experiment.In the same study,stachyose showed a slightly negative correlation with temperature,and raffinose was unaffected by temperature.Gorecki et al.[46]observed a twofold increase in stachyose in lupine seeds maturing under low-temperature conditions.However,Kumer et al.[43]reported a nonsignificant effect of location on raffinose and stachyose contents.Our findings on the effects of temperature on raffinose and stachyose were in contrast to Wolf's and Gorecki's results[45,46].

Saponins have been reported to depend more strongly on genetic characteristics than on environmental effects[37,47].Information about the influence of environmental factors on soybean saponins is scarce.Seguin et al.[19]reported that soyasaponin B concentration was reduced by 11%in soybeans planted in mid-May compared to that of those planted in late June,suggesting that high-temperature stress during the soybean seed filling stages accounted for the large reductions[19].A general negative correlation between soybean saponins and temperature established in the present study was in accord with a finding of Seguin et al.[19].However,these authors also found that the response of soybean TS to high temperature was cultivar-pecific,indicating that temperature was not the key factor affecting soybean TS.

To our knowledge,very few studies have investigated the relationship between soybean phospholipids and weather factors.Given that phospholipids in soybeans are co-products of oil accumulation,the production of soybean phospholipids rises with soybean oil yield.The present study demonstrated significant negative correlations between TP and AT15,MDT,and DTR,consistent with a previous finding of the relationship between soybean oil content and weather factors[24].

4.3.Effects of rainfall and hours of sunshine on bioactive components

The nonsignificant correlation between TI and rainfall in our study shows that water effect on soybean TI is complicated.Swanson et al.[48]found that dry growing conditions increased TI synthesis owing to a decrease in the redox potential of plant cells.In contrast,irrigation dramatically enhanced isoflavone content by an unknown mechanism[5].A combined effect of temperature and soil moisture deficit on soybean isoflavone contentproposed by Caldwell et al.[40]and Lozovaya et al.[41]may explain the contradictory results of the previous studies.Carrera and Dardanelli[23]evaluated the combined effects of temperature and drought on TI content using multi-environment field trials and reported that TI decreased linearly with rising temperature and with increasing water deficit when precipitation was below 70 mm(indicating drought)during the seed filling period.The mean RF of all the sampling sites during the seed-filling period in our study was above 70 mm,but a threshold RF might be required for a relationship with TI in soybean.

Sucrose metabolism is crucial to seed development and particularly susceptible to drought stress[49].Liu et al.[21]showed that sucrose concentrations in flowers and pods were higher under drought conditions than in well-watered controls,implyingapossiblenegativecorrelation between sucrose and RF.This correlation was confirmed in our study.The sucroseaccumulationinduced bydrought incrop reproductive organs could be due partially to a low activity under drought conditions of acid invertase,which cleaves incoming sucrose into hexoses[50,51].

Positive effects of RF on saponin components have also been found in other species such as quinoa[52].There are few reports describing a relationship between saponin and precipitation.Reeti et al.[53]speculated that the higher saponin content in June-than in July-planted soybeans could be attributed to higher maximum daily temperature and RF during the flowering period.To our knowledge,the present report may be the first describing the relationship between soybean saponin and precipitation.The mechanisms responsible for this relationship invite further study.

Flavonoids are known to absorb strongly in the ultraviolet b(UVB)region of the solar spectrum,protecting plants from the effects of UVB radiation by acting as a simple sunscreen[54].Flavonoids with a higher level of B-ring hydroxylation were preferentially synthesized following UVB treatment[55].Lee et al.[56]showed that isoflavone content increased in soybean sprouts under light conditions.These findings are in agreement with the positive correlation between soybean TI contents and HS in our study.

Based on the present results,the following recommendations may proposed for improving soybean nutritional value in different regions.Long HS and large DTR resulted in more bioactive component accumulation in soybean seeds in the NESR and NWSR regions.Accordingly,the two regions can prioritize the production of functional soybeans rich in bioactive components for food uses.In view of the moderate contents of bioactive component contents and relatively high protein levels of soybean from the HHHR,this region is recommended for the production of raw material for soyfood processing.In the SCMR,bioactive component contents in soybean are low,but protein content is high.Thus,soybean production in this region can focus on vegetable soybean and localsoyfoodprocessing.Agronomicmeasuressuchas irrigation and sowing-date adjustment are also recommended to improve environmental conditions during the soybean flowering and pod filling stages.

5.Conclusions

Themajorbioactivecomponentsofsoybeans,namely isoflavones,oligosaccharides,phospholipids,and saponins,displayed marked geographicdistribution characteristics across China,suggesting that environment could influence these components.Analysis of the relationship between main environmental factors and bioactive components confirmed the effect of weather factors on the bioactive components.After analysis of the regional data,specific recommendations were proposed for the development of soybean production for functional food purposes in producing areas of China.

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2018.05.001.

Acknowledgments

We thank the China Agriculture Research System soybean experimental stations for collecting the soybean samples.We also thank Sancai Liu,Fang Liu,and Yan Li for technical support for composition analysis.This study was supported by the National Key Research and Development Program of China(2017YFD0101400),China Agriculture Research System(CARS-04)and Agricultural Science and Technology Innovation Project of the Chinese Academy of Agricultural Sciences.