Barley varieties tolerant to waterlogged reduced soil show the better root growth in hypoxia

ABSTRACT Barley (Hordeum vulgare L.) growth is often inhibited under waterlogged fields. Under waterlogging, a higher C/N ratio in soil easily leads to low oxygen concentration and soil reduction. Although the tolerant varieties for waterlogging and soil reduction were previously reported, those root growth under hypoxia remains unknown. In this study, more than 90% of total leaf area in Deder 2, Hayakiso 2 and Akashinriki remained green even under waterlogged and reduced soil condition at seedling stage. The root elongation of Deder 2 and Hayakiso 2 was similar between aerobic and hypoxic conditions. In comparison, severe leaf necrosis was observed in Haruna Nijo and Houshun. Those root elongation was significantly inhibited under hypoxia, as compared with that under aerobic condition. In conclusion, it was considered that barley varieties tolerant to waterlogged reduced soil showed the better root growth in hypoxia in this study. Graphical abstract


Introduction
Waterlogging is a global problem for upland crop production, and damage by waterlogging is frequently reported worldwide: in Australia and Northern Europe (Byrne et al., 2022), Southeast Asia and Australia (Setter & Waters, 2003;Setter et al., 2009), and Japan (Hamachi et al., 1985).Waterlogging is caused not only by severe rainfall CONTACT  events but also by field topography, geology, agricultural ecosystems, and management.During waterlogging, the space between soil particles is filled with water and gas diffusion becomes extremely slow, leading to oxygen deficiency (Bailey-Serres & Colmer, 2014).Following soil oxygen deficiency after waterlogging, aerobic microbial growth becomes exponential (Johnson, 1967;Laanbroek, 1990); then, waterlogged soil becomes reduced (Fiedler et al., 2007;Takai, 1978).In rice -barley rotation in Japan, a total of 8,000 kg ha −1 of straw is cut and buried into the soils after rice paddy cultivation, to return soil organic carbon (Jin et al., 2020).Reduction of waterlogged soil during organic matter degradation is a serious problem for winter barley cultivation.Using model simulation, Mano and Takeda (2012) confirmed that the addition of starch as a carbon source exacerbates this problem at the barley seedling stage.
Leaf necrosis after waterlogging has been used as a seedling stage symptom to evaluate tolerance, and genetic diversity has been reported in waterlogging tolerance in barley based on field trials (Byrne et al., 2022;Li et al., 2008;Setter et al., 2009;Sundgren et al., 2018) and pot experiments (Mano & Takeda, 2012;Zeng et al., 2013).In a field trial, scoring of leaf greenness, grain yield, and biomass was compared between control and waterlogged fields over two years (Byrne et al., 2022).Minorimugi (six-rowed barley) was tolerant and Haruna Nijo (two-rowed barley) was sensitive to waterlogging during internode elongation (Hamachi et al., 1987(Hamachi et al., , 1988)).However, controlled conditions and repeatable trials are required for selecting cultivars for recommended lists and breeding (Byrne et al., 2022).In pot experiments, 14 varieties of barley were evaluated, and Deder 2, Harumaki Rokkakumugi, and Kikai Hadaka were found to have high waterlogging tolerance (Mano & Takeda, 2012;Takeda, 1989).An analysis of 11 varieties of barley showed that a waterlogging-tolerant variety had a well-formed root aerenchyma (Zhang et al., 2016).Using a population from a cross between the varieties Franklin (waterlogging sensitive) and YuYaoXiangTian Erleng (waterlogging tolerant), Broughton et al. (2015) detected one QTL related to root porosity under hypoxia.However, information on waterlogging tolerance of commercial varieties, including tolerance to root hypoxia, remains to be clarified.In this study, we firstly analyzed the status of oxygen concentration and redox potential under waterlogged and reduced soil.And then, we analyzed (i) tolerance to reduced soil due to waterlogging and (ii) inhibition of root elongation by hypoxia in a short experiment under controlled conditions in barley varieties including past and current commercial varieties that are important for breeding.

Plant materials
The barley varieties used in this study are listed in Table 1.Seeds were supplied from the Fukuoka Agriculture and Forestry Research Center and Okayama University.Seeds were produced in a field at the Fukuoka Agriculture and Forestry Research Center (33° 504′N, 130°569′E).
As described in Table 1, a total of 20 barley varieties including old, or current commercial varieties were evaluated under (i) waterlogged reduced soil.And total of 7 barley varieties which crossed populations were developed and evaluated under (ii) hydroponic cultivation.As the candidate lines, these seven barley varieties were used for the development of crossed populations to generate new varieties for barley malt.Seeds were imbibed and germinated for 1 day on a Petri dish at 20°C in the dark.In an experiment (i), Tolerance of barley seedlings to waterlogged reduced soil was evaluated in a greenhouse with natural daylight during the winter season.Temperature was maintained at 20°C by heating and opening and closing of the side curtain of the greenhouse.Seedlings were grown in containers (600 mm length, 360 mm width, 190 mm depth) filled with granular soil Kokuryu Baido (Seishin Sangyou Industry, Fukuoka, Japan; granule diameter 1.5-3.3mm) fertilized with 0.28 g kg −1 N, 0.28 g kg −1 P, and 0.28 g kg −1 K.At 2 days after sowing, soluble starch solution was added to the containers at 0.1% for moderately reduced soil or 0.2% (w/v) for severely reduced soil.Water level was maintained at 15-20 mm above the soil surface.The plants were harvested 14 days after sowing.After the start of waterlogging, redox potential (mV) was measured at the depth of 80-100 mm below the soil surface with an Eh meter (PRN-41, Fujiwara Scientific Co. Ltd., Tokyo, Japan).Oxygen concentration was measured with a microsensor (OX-N, Aarhus, Unisense).This vertical data was obtained at the same timing of sampling in the container without growing plants as negative control.Images of whole leaves were captured by a scanner (GT-X980, Seiko Epson Corp., Nagano, Japan).Leaf area was measured with a WinRHIZO root scanner (Regent Instruments Inc., Montreal, QC, Canada); green and yellow areas were distinguished automatically.In an experiment (ii), Tolerance of oxygen deficiency in hydroponic cultivation was evaluated in a growth chamber (PHC Corp., Tokyo, Japan) controlled at 20°C with a photoperiod of 14 h light (200-250 µmol quanta m −2 s −1 ) and 10 h dark.Basic nutrient solution was as described previously (Mae & Ohira, 1981;Obara et al., 2010), except that the pH was adjusted to 6.0.A total of 54 germinated seeds were sown on nets floated with the aid of polystyrene in a container (275 mm length, 131 mm width, 267 mm height) as described previously (Abiko & Obara, 2014).The plants were grown for 7 days under aerobic or hypoxic conditions.In the former case, air was bubbled through the nutrient solution using a fine-pore air stone and an air pump.In the latter case, the plants were grown in deoxygenated; the solution was deoxygenated by bubbling nitrogen instead of air as above stagnant solution containing 0.1% (w/v) agar (Wiengweera et al., 1997).Agar was added to restrict flow in hydroponic solution.

Measurements
Dissolved oxygen concentration was measured at 0 and 7 days after sowing with a dissolved oxygen meter (OM-51, Horiba, Kyoto, Japan).The oxygen concentrations were measured under the two conditions (Table 2).On

Statistical analyses
All data are represented as means ± standard deviation (SD).The t-tests and two-way analysis of variance (ANOVA) were performed in Bell Curve Excel statistics software (Social Survey Research Information Co., Ltd., Tokyo, Japan).

Soil environments
To understand the soil environment to which the evaluated seedlings were exposed, the vertical oxygen distribution from water surface into waterlogged soil was measured (Figure 1).Under waterlogged soil condition, dissolved oxygen concentration near the water surface was 251.3 ± 6.2 µmol L −1 , or 87.9% of oxygen concentration in air saturation, and gradually decreased to nearly zero at 25 mm below the soil surface.On the other hand, under reduced soil condition induced by carbon source, dissolved oxygen concentration near the water surface was 242.3 ± 1.6 µmol L −1 , or 84.8% of oxygen concentration in air saturation, and severely decreased to 9.6 µmol L −1 near the soil surface.Oxygen was severely limited below the soil surface, decreasing to 1.1% of oxygen concentration in air saturation (Figure 1).Redox potential has been evaluated during waterlogging in the field as one of the important indicators (Ponnamperuma, 1984;Setter & Waters, 2003).
A decrease in redox potential induced by carbon supply was detected during cultivation (Figure 2).The reduction pattern was clearly dependent on carbon source concentration.The value of redox potential at the beginning of cultivation (before waterlogging) was 482 mV.Under waterlogged condition, the redox potential was maintained at range from 483.3 mV to 637.7 mV.Under moderate reduced soil condition, the value of the redox potential ranged from −139.0 to 404.7 mV and was 137.0 mV on the final day of cultivation (Figure 2).Under more severely reduced condition which 0.2% starch was added, the redox potential was lower at each time point; it ranged from −282.0 to 124.7 mV and was 29.3 mV on the final day of cultivation.A decrease in redox potential in waterlogged soil by addition of a carbon source has been reported under controlled conditions (Mano & Takeda, 2012).In this study, we also observed a decrease in redox potential in waterlogged soil supplied with a carbon source.

Tolerance of 20 barley varieties to waterlogged reduced soil
Absolute values of green and total leaf area are shown in Figure 3.In waterlogged soil without reduction, total leaf area was larger than that under reduced soil condition and all of leaves showed green (Figure 3a).In moderately reduced soil, green area was below ca.15 cm 2 , except in Shunrei and Harumaki Rokkakumugi (Figure 3b).In severely reduced soil, green area was below ca. 10 cm 2 , except in Harumaki Rokkakumugi (Figure 3c).Low redox potential values lead to generation of the reduced forms of nutrient elements such as Mn 2+ , Fe 2+ , or S 2+ (Takai, 1978), which may inhibit plant growth (Setter et al., 2009).In this study, we also observed inhibition of plant growth caused by a decrease in redox potential under waterlogged and reduced soil condition.In moderately reduced soil, the ratio of green leaf area to total leaf area, which was used as a measure of tolerance, was above 80% in 15 of the 20 varieties and was larger in Hayakiso 2, Haruka Nijo, and Akashinriki (Figure 3b).In severely reduced soil, the ratio of green area was above 80% in 7 of the 20 varieties and was above 90% in Seijo 17, Deder 2, Hayakiso 2, and Akashinriki (Figure 3c).The high tolerance of Akashinriki, Hayakiso 2, and Ichibanboshi is newly reported in this study.Akashinriki is classified into the Japanese landraces and is used for barley grass juice and is grown for leaves in converted rice paddies in Japan (Table 1).Akashinriki may have genotypes adaptive to the Japanese environment that are not present in improved varieties, such as aluminium tolerance in the context of the crop evolution (Fujii et al., 2012).Although Akashinriki was evaluated the waterlogging tolerance as an indicator in this study, it is considered that it is not suitable for the breeding of the beer malt, because of six row and naked of hull type (Table 1).
Regarding Hayakiso 2, a population suitable for QTL analysis has been developed by crossing Hayakiso 2 and Haruna Nijo (Saisho et al., 2011).It was originally intended for vernalization studies, but could be used to map QTLs for waterlogging tolerance.Ichibanboshi is one of the leading varieties in northern Kyushu region, a naked barley.Harumaki Rokkakumugi and Deder 2 were tolerance (Figure 3).Tolerance of Harumaki Rokkakumugi and Deder 2 has been reported at the early growth stage (Mano & Takeda, 2012) and was confirmed in this study.Harusayaka showed slightly tolerance (Figure 3).This variety is currently cultivated as commercial varieties in northern Kyushu region in Japan (Table 1).Some varieties such as Morex were moderately tolerant to waterlogging.Morex was registered in the United States in 1978 and its genome sequence is used as a reference (Masher et al., 2013).Its high-quality malting traits are inherited in many commercial malting barley varieties.Houshun and Haruna Nijo were sensitive, with about half of the total leaf area injured (Figure 3b,c).Although Haruka Nijo was tolerant in moderately reduced soil, it was sensitive in severely reduced soil.

Oxygen concentration in aerobic and hypoxic conditions
Mean dissolved oxygen concentration under aerobic (control) condition was ca.8.0 mg L −1 before cultivation and slightly decreased to ca. 7.3 mg L −1 on day 7 of cultivation (Table 2).In contrast, under hypoxia, dissolved oxygen concentrations were low during cultivation (ca.0.1 mg L −1 at the start and end) (Table 2).

Responses under aerobic and hypoxic conditions in seven barley varieties
Maximum shoot length was significantly lower under hypoxia than under aerobic condition in five varieties and tended to be lower in Deder 2 and Seijo 17 (Figure 4a).Leaf area was significantly smaller under hypoxia than under aerobic condition (Figure 4b; p < 0.01), except for Harumaki Rokkakumugi, Seijo 17 and Deder 2. Deder 2 showed similar shoot phenotypes under hypoxic and aerobic conditions (Figure 4).Maximum shoot length and leaf area of Haruna Nijo, Houshun and Nishino Gold were significantly lower under hypoxia, compared that under aerobic condition.Maximum and total root length of Deder 2 and Hayakiso 2 was similar under both aerobic and hypoxic conditions (Figure 5a).Maximum root length of Seijo 17 was similar under both conditions, but total root length tend to be shorter under hypoxia, compared with aerobic condition.Maximum root length of Harumaki Rokkakumugi and Nishino Gold was similar under aerobic and hypoxic conditions (Figure 5a), but their total root length was significantly lower under hypoxia than under aerobic conditions (Figure 5b; p < 0.01).Haruna Nijo and Houshun had longer roots under aerobic condition (Figure 5a), but maximum and total root length of Haruna Nijo and Houshun were significantly lower under hypoxia than under aerobic conditions (Figure 5; p < 0.01).Maximum root length was significantly affected by cultivar and treatment, and there was cultivar × treatment interaction (Figure 5a; p < 0.001).In our result, it was confirmed that Deder 2 is tolerant of waterlogging and soil reduction (Mano & Takeda, 2012).Thus, it was clarified that the growth of Deder 2 in response to hypoxia was not different from that in response to aerobic condition.Hayakiso 2 was also tolerant of waterlogging, soil reduction and hypoxia (Figure 3 and 5), except for shoot growth under hypoxia (Figure 4; p < 0.01).
Especially, the maximum root length tended to be longer under hypoxia, compared with aerobic condition.The results for seedlings of Seijo 17 were inconclusive (Figures 4 and 5).However, Seijo 17 was waterlogging tolerant (Figure 3), and its yield is highly tolerant (Hamachi et al., 1987).In our hydroponic experiments, the results for seedlings of Nishino Gold were also inconclusive (Figures 4 and 5), although root elongation was intended to be inhibited.Nishino Gold was sensitive in the soil experiment (Figure 3) and slightly intolerant in a field experiment (Furusho et al., 1999;Itoh et al., 1987).Haruna Nijo and Houshun had large necrotic areas on leaves and their root development was significantly inhibited under hypoxia (Figure 3 and 5; p < 0.01).Haruna Nijo was released by Sapporo Breweries Ltd. in 1981 and its genome has been sequenced (Sato et al., 2016).It is known as a high-quality Japanese malting barley and has been used in the breeding of most barley varieties in Japan.However, Haruna Nijo was sensitive to waterlogging and soil reduction.Houshun was registered by Fukuoka Prefecture in 2002 (Table 1) and remains a commercial variety in northern Kyushu in Japan.Its waterlogging tolerance was low in this study, and it had severe leaf necrosis in reduced soil (Figure 3).Houshun was reportedly moderately or slightly waterlogging tolerant in a field experiment (Furusho et al., 1999).Our results regarding the sensitivity of shoot and root phenotypes to hypoxia disagree with a previous field trial (Furusho et al., 1999).Therefore, our results might suggest that root phenotypes under hypoxia mostly reflected shoot phenotypes under waterlogging and soil reduction.Their clear traits at the seedling stage suggest that this evaluation would also be useful in breeding barley varieties with high waterlogging tolerance.

Conclusion
In conclusion, we newly found that (i) a few varieties such as Hayakiso 2 and Akashinriki showed soil reduction tolerance to waterlogging under reducing conditions induced by the supply of a carbon source, and (ii) waterlogging tolerant varieties such as Deder 2 and Hayakiso 2 intended to have short roots (p < 0.01) and showed less inhibition of root elongation by hypoxia at the seedling stage.To the best of our knowledge, this is the first study to report that barley varieties tolerant of reduced soil have hypoxia-tolerant roots and that root length is an indicator of tolerance.For updating the recommended lists of waterlogging-tolerant varieties, repeatable trials under controlled conditions would be effective in breeding barley varieties with high waterlogging tolerance in the future.

Figure 3 .
Figure 3. Green and yellow leaf area in 20 barley varieties in (a) waterlogged condition, (b) moderately reduced soil and (c) severely reduced soil in an experiment (i).Values are mean ± SD (n = 5-10).Numbers in parentheses indicate relative green leaf area per total leaf area.

Figure 4 .
Figure 4. Shoot phenotypes of 7 varieties grown in hydroponic solution under aerobic or hypoxic conditions in an experiment (ii).(a) Maximum shoot length; (b) leaf area.Values are mean ± SD (n = 5-10).Asterisks indicate significant differences between the two conditions at **p < 0.01 (t-test).The same letters within each cultivar indicated no significance as determined by tukey-Kramer test at the 5% level.Uppercase and lowercase alphabet indicate comparison among cultivars under aerobic and hypoxia condition, respectively.ANOVA results: ** p < 0.01; *** p < 0.001; ns, not significant.

Figure 5 .
Figure 5. Root phenotypes of 7 varieties grown in hydroponic solution under aerobic or hypoxic conditions in an experiment (ii).(a) Maximum root length; (b) total root length.Values are mean ± SD (n = 5-10).Asterisks indicate significant differences between the two conditions at **p < 0.01 (t-test).The same letters within each cultivar indicated no significance as determined by tukey-Kramer test at the 5% level.Uppercase and lowercase alphabet indicate comparison among cultivars under aerobic and hypoxia condition, respectively.ANOVA results: **p < 0.01; ***p < 0.001; ns, not significant.

Table 1 .
Barley varieties evaluated in this study.

Table 2 .
Dissolved oxygen concentrations (mg L −1 ) in aerobic and hypoxia conditions in an experiment (ii).PRODUCTION SCIENCE the final day, the plants were harvested, and the shoot lengths and maximum root were measured with a ruler.Total root length and leaf area were calculated in WinRHIZO software (Regent Instruments Inc.).