Effect of drought stress or soil pH on cesium accumulation in Napier grass

Radio-cesium (Cs) decontamination efficiency by plants is supposedly affected by environmental conditions. To elucidate the factors influencing the unstable 137Cs-decontamination efficiency, we first examined the influence of drought or soil pH on Cs accumulation using cesium-133 (133Cs) using Napier grass (Pennisetum purpureum Schum.) grown under pot conditions. The experiment was performed on three different conditions with 150 µM 133Cs applied to soil: low pH (L-pH, pH = 5.6 ± 0.0), near-neutral pH (N-pH, pH = 6.6 ± 0.1), and the combination of low pH and drought stress (Drought). Drought stress had prominent negative effects on plant height, aboveground dry matter weight (DWabove), 133Cs concentrations in aboveground or root (Csabove or Csroot) parts, or 133Cs total content in the part aboveground (Cs-T). 133Cs concentration and total content in Drought conditions were reduced by 44.1% and 60.0% aboveground and 63.6% and 66.0% in root parts compared with counterpart normal soil moisture conditions (L-pH), respectively. Conversely, there were no significant effects of soil pH on Csabove, Csroot, or Cs-T in L-pH and N-pH conditions. Chlorophyll fluorescence parameters (Fv/Fm, Fv/F0) and the temperature in leaves were negatively affected by drought stress or soil pH conditions. From our results, drought strongly influenced plant growth and Cs accumulation in plants compared with soil acidity. Therefore, soil moisture appears to be a significant factor in maintaining 137Cs-decontamination efficiency by Napier grass.


Introduction
Soil radio-cesium (Cs) contamination is an environmental problem due to not only intensive agricultural practices but also the uncontrolled industrialization and unsustainable urbanization in Fukushima Prefecture of Japan, mostly consequence of the Fukushima Daiichi Nuclear Power Plant accident in 2011. Decontamination is urgently required because Cs supposes a high threat to soil quality, food safety, and human health. Among soil decontamination methods, phytoextraction by plant cultivation is an option for removing Cs from the soil. Although phytoextraction has a relatively low decontamination efficiency compared to topsoil stripping, it can decontaminate the subsoil. Therefore, methods to improve and/or maintain decontamination efficiency by increasing Cs uptake from soil to plants are required (Wu et al., 2009). Previously, we attempted to increase 137 Cs decontamination efficiency using various effective methods in Cs-polluted agricultural land. However, its effectiveness on 137 Cs decontamination was unstable even in the same experimental site and often affected by natural environmental factors (Kang et al., 2021). Several factors, such as the soil's physicochemical properties, Cs bioavailability, and climatic conditions are closely related with plant decontamination efficiency (Patra et al., 2020). Particularly, Cs availability and plant uptake is heavily influenced by soil characteristics (Massas et al., 2010) because plant roots mediate Cs uptake from the soil. Most existing evidence suggest a correlation between Cs uptake and soil pH, clay content, soil potassium (K)-status, and clay minerals (Crout et al., 2003;Skarlou et al., 1996).
Among the several soil factors related to plant growth or Cs uptake, soil pH directly or indirectly affects radio-Cs uptake efficiency through the plant roots. Many studies have demonstrated increased radio-Cs accumulation in plants with decreasing soil pH, such as in Dactylis glomerata, Phalaris arundinacea, Lolium perenne, Anthoxanthum odoratum (Ogura et al., 2014), Quercus serrata (Kanasashi et al., 2020), Raphanus sativus, Cucumis sativus, Glycine max (Massas et al., 2010), R. sativus, C. sativus, Triticum aestivum (Skarlou et al., 1996), and Vigna mungo (Win et al., 2017). Soil acidification can lead to Cs desorption from soils, thus becoming more bioavailable in the environment; however, the effect of pH on Cs desorption depends on soil type (Giannakopoulou et al., 2007). On the other hand, the opposite effects or lack thereof have been found in Riccia fluitans (Heredia et al., 2002). Similarly, although soil pH (range, 3.9-8.4) has been reported to not affect radio-Cs transfer, soil-to-plant Cs transfer varied in different soil types, being lower in clay or loam than in sand (Crout et al., 2003;Van Bergeijk et al., 1992).
Another factor in soil related to plant growth or Cs uptake is soil moisture, which is closely related to plant growth or gas exchanges. Trace elements become concentrated in the remaining soil solution due to higher evapotranspiration resulting in higher uptake (Ehlken & Kirchner, 1996). Previously, we suggested that 137 Cs decontamination efficiency by Napier grass was strongly affected by fluctuating environmental conditions, particularly precipitation, during the cultivation period (Kang et al., 2021). Drought stress led to reduced 137 Cs concentration in V. mungo (Win et al., 2015(Win et al., , 2017, but there was no difference in the association with Cs concentration in grasses such as Agropyron spicatum, Leymus cinereus, Agropyron cristatum, and Bromus tectorum (Cook et al., 2009).
According to previous reports, the soil-to-plant transfer factor of radio-Cs is well correlated with that of stable Cs on several plant species such as Phragmites australis, Helianthus annuus, Populus simonii (Soudek et al., 2004(Soudek et al., , 2006, and Oryza sativa (Tsukada et al., 2002). Thus, in the present study, we used stable Cs to facilitate experimental studies on radio-Cs absorption and mobility in the environment. This study aims to investigate the influence of drought or soil pH on plant growth or 133 Cs uptake to facilitate future improvements in long-term phytoextraction in 137 Cscontaminated agricultural lands in the Fukushima Prefecture. We hypothesized that differences in soil pH or moisture affected plant growth and/or 133 Cs accumulation in Napier grass.

Materials and methods
Plant materials, growth conditions, and experimental design A common Napier grass variety (P. purpureum var. Merkeron) was used for the experiments. The experiment was conducted in a rainout greenhouse in Goshogawara, Aomori Prefecture (40.5494 N, 140.2743E), in northern Japan between May 31 and July 21, 2021 (51 days). There were three different growth conditions with nine individual plants (replicates) per condition: low pH (L-pH, pH = 5.6 ± 0.0), near-neutral pH (N-pH, pH = 6.6 ± 0.1), and those combination of low pH and drought stress (Drought), for a total of 27 pots. Fourweek-old nursery plants were transferred into a 1/2000a Wagner pots containing 10 kg of the dried commercial soil (Table 1) on May 31, 2021. Chemical fertilizers were applied at a rate of 3.2 g N, 3.3 g P 2 O 5 , and 0.7 g K 2 O (20% of the recommended amount of potassium fertilizer) per pot as a basal dressing, whereas no fertilizer was applied as a top dressing until the harvesting was done. One Napier grass plant was assigned to each pot in a completely randomized design. Maximum and minimum temperatures in the greenhouse were measured using a data logger (Temperature and Humidity USB Datalogger DL171, AS ONE Co. Ltd., Osaka, Japan) throughout the experiment (Fig. 1).

Treatments
To prevent root damage, 14 days before transplanting, the soil pH was adjusted using calcium hydroxide [Ca(OH) 2 ] to produce N-pH at a rate of 4 g [Ca(OH) 2 ] per 1 kg dried commercial soil and thoroughly mixed (Table 1) according to a soil pH correction method based on a buffer curve (Wang et al., 2015). Cesium (atomic weight 133 Cs) from cesium chloride (CsCl) was concentrated at 150 µM in 2 L water and applied to each pot prior to transplanting. Drought stress was imposed during the tillering stage from 21 to 51 days 11.9 ± 0.4

Fig. 1
Air temperature during plant growth. DAT, days after transplanting after transplanting (DAT) and controlled by the watering time of pots into normal moisture (L-pH and N-pH) and moderate drought stress (Drought) conditions ( Fig. 2). Before imposing drought stress at 21 DAT, all pots were equally watered to ensure similar soil moisture in all pots (total, 5000 mL water per pot). Soil tension at 20-cm depth was continuously measured using a tension meter (DIK-8333, Daikirika Co. Ltd., Japan) in the pots during the experimental period. Pots were watered when soil tension reached the target value: from − 20 to − 30 (ideally, − 25) kPa for normal moisture and − 70 to − 80 (ideally, − 75) kPa for drought stress conditions (Fig. 2).

Measurement of chlorophyll fluorescence and SPAD value
Chlorophyll fluorescence (ChlF) was measured in the fully expanded upper leaves from nine individual plants per treatment at 31, 41, and 51 DAT (M 1 , M 2 , and M 3 ) at greenhouse temperature with a portable chlorophyll fluorometer (OS-30p, Opti-Science Inc., USA). Plants were adapted to dark condition for 30 min before the measurement of ChlF parameters. Running the system according to manufacturer's instructions, the following leaf measurements were performed: maximal fluorescence (F m ), maximal variable fluorescence (F v = F m -F 0 ), and maximal quantum efficiency of photosystem (PS) II photochemistry (F v /F m ). The potential activity of PS II (F v /F 0 ) was computed according to a previous report (Banks, 2017). After calculating ChlF, the SPAD value was measured in the same leaf from the nine individual plants per condition using SPAD (SPAD-502, Minolta Co., Ltd., Osaka, Japan) at M 1 , M 2 , and M 3 ; the SPAD value indicates chlorophyll (Chl) content or leaf color.
Measurements of agronomic traits, leaf temperature, and final soil moisture content Before harvesting at 51 DAT, plant height (PH) and tiller number (TN) were measured in nine individual plants per treatment; then, the leaf temperature (T leaf ) was measured in fully expanded upper leaves using an infrared thermometer (AD-5611A, A&D Co., Ltd., Japan). After harvesting the aboveground part, the final soil moisture content (SMC) in the pots was measured at soil depth of 5 cm, 10 cm, and 15 cm using a soil moisture meter (PMS-714, Lutron Electronic Enterprise Co., Ltd., Taiwan). 133 Cs content analysis in aboveground and root parts The aboveground and root parts of nine individual plants per treatment were harvested at 51 DAT and dried in an oven at 80 °C for 72 h, and the dry matter of both the aboveground (DW above ) and root (DW root ) parts weighed. The ratio of aboveground and root parts (T/R ratio) was computed based on the dry matter weight. To determine 133 Cs concentration within each plant, 0.2 g each of dried aboveground and root samples were digested in 10 mL HNO 3 (nitric acid, Kanto chemical, Japan) at 110 °C for 13 h using Values of soil tension are presented as mean ± standard error (n = 9 individual pots replicates). M1, M2, M3, timing of chlorophyll fluorescence (ChlF) and SPAD measurements at 31, 41, and 51 DAT, respectively. DAT, days after transplanting an aluminum block bath (DTU-2BN; Taitec Co., Saitama, Japan). After cooling, samples were diluted with ultra-pure water and 3% HNO 3 , and supernatants obtained, wherein the 133 Cs concentration was measured using an inductively coupled plasma-mass spectrophotometer (ICP-MS, Agilent 8800, Agilent Technologies Co., Ltd., Santa Clara, CA, USA) according to a previously reported protocol (Yang et al., 2016).

Statistical analysis
Nine plants or soil replicates in three conditions were measured for ChlF, SPAD value, PH, TN, DW above and DW root , T leaf , 133 Cs content, or final SMC in the pots. Data were analyzed using Tukey's multiplerange test to determine the significance of any differences between mean values, using KaleidaGraph (ver. 4.1, Synergy Software) software. Data for all traits were subjected to analysis of variance (ANOVA) using the XLSTAT software 2022 version (Addinsoft, New York, USA). The correlation coefficients among agronomic traits and 133 Cs content were determined using simple linear correlations, and the results were analyzed using the Pearson's correlation coefficient (r) (Triola, 2015).

Soil and air temperature
The physical and chemical characteristics of the commercial soil are presented in Table 1. The exchangeable potassium and magnesium, cation exchange capacity, and humus content were high. The soil belonged to loam. Mean maximum, minimum, and average air temperatures during the experimental period were 40.5 ± 0.9 °C, 16.5 ± 0.4 °C, and 25.9 ± 0.5 °C, respectively (Fig. 1).
Chlorophyll fluorescence, SPAD, leaf temperature, and soil moisture ChlF or SPAD values differed according to conditions or measuring time (Table 2). Overall, F v /F m , and F v /F 0 in L-pH were significantly or relatively higher than in both N-pH and Drought conditions at M 2 (measurement at 41 DAT) and M 3 (measurement at 51 DAT), but both parameters were not significantly letters and ns indicate no significant differences between the conditions at the 5% level according to Tukey's multiple-range test. *, **, and *** indicates a significant correlation at a 5%, 1%, and 0.1% level, respectively; ns indicates no significant difference at a 5% level . An ANOVA revealed significant effects of soil pH (T 1 ) or drought stress (T 2 ) on ChlF at M 2 and M 3 rather than at M 1 . Both values in the L-pH condition were greater than in the other two conditions. According to the ANOVA, F v /F m (P < 0.001 at M 2 , P < 0.01 at M 3 ) and F v /F 0 (P < 0.01 at M 2 and M 3 ) were significantly affected by T 1 . Similarly, F v /F m (P < 0.01 at M 2 , P < 0.05 at M 3 ) and F v /F 0 (P < 0.05 at M 2 and M 3 ) were significantly affected by T 2 . From the ANOVA analysis, an effect of soil pH × drought stress interactions (T 1 × T 2 ) was found for all ChlF parameters at M 2 and M 3 , however, SPAD value was not affected. Soil pH and drought stress were associated with the T leaf of Napier grass and SMC in pots (Table 2). T leaf significantly increased by 3.5 °C in N-pH or 6.3 °C in Drought conditions compared with L-pH conditions. The F-ratios obtained with the ANOVA were 18.758 (P < 0.001) for T 1 , 46.578 (P < 0.001) for T 2 , and 22.290 (P < 0.001) for T 1 × T 2 .
The soil moisture at three different soil-depths at the end of the experiment varied among conditions (Table 2). There was a significant difference between conditions at all depths, with soil moisture levels following the order N-pH > L-pH > Drought conditions.

Agronomic traits
The PH, DW above , and T/R ratio significantly differed among conditions (Table 3). A significant difference in agronomic traits was observed between non-drought (L-pH) and drought-stressed (Drought) conditions. On the other hand, a significant difference between L-pH and N-pH was found only for DW above . According to ANOVA's results, PH (P < 0.05), DW above (P < 0.001), and T/R ratio (P < 0.001) were significantly affected by T 2 ; DW above (P < 0.05) and T/R ratio (P < 0.01) were also affected by T 1 .

Concentration and total 133 Cs content
Similar to agronomic traits, a significant difference on 133 Cs concentration was observed in the presence of drought (Table 3). The 133 Cs concentration in aboveground (Cs above ) and root (Cs root ) parts in drought conditions were significantly lower than those of other conditions (L-pH or N-pH). Despite the higher 133 Cs concentration in N-pH than in L-pH conditions, there was no statistically significant difference. An ANOVA revealed that Cs above (P < 0.001) and Cs root (P < 0.001) were significantly affected by T 2 . Similarly, the total 133 Cs content (Cs-T) in the aboveground or root parts of Napier grass in nondrought (L-pH, N-pH) conditions was significantly higher than in Drought conditions. However, there was no significant difference in Cs-T between L-pH and N-pH conditions (Fig. 3). Table 4 shows the correlation coefficients among the agronomic traits and 133 Cs content in Napier grass. Table 3 Comparisons of plant height (PH), tiller number (TN) per plant, aboveground or root dry matter weight (DW above or DW root ), aboveground and root ratio (T/R ratio), and 133 Cs con-centrations of aboveground (Cs above ) or root (Cs root

) parts in Napier grass at 51 DAT (M 3 )
Values indicate the mean ± standard error (n = 9 individual plants). The same letters and ns indicate no significant differences between the conditions at a 5% level according to Tukey's multiple-range test. *, **, and *** indicates a significant correlation at a 5%, 1%, and 0.1% level, respectively; ns indicates no significant difference at 5% level A significantly positive correlation between Cs above and Cs-T (P < 0.001 in L-pH and N-pH, P < 0.01 in Drought) was observed in Napier grass under all conditions. In addition, there was a significantly positive correlation between DW above and Cs above (P < 0.05) and between DW above and Cs-T (P < 0.001) in N-pH conditions. However, both Cs above and Cs-T in L-pH conditions were negatively correlated with DW above . PH and TN were positively correlated with DW above in all conditions; in Drought conditions, they were also positively correlated with Cs above and Cs-T. On the other hand, the PH in the N-pH condition was correlated with Cs above and Cs-T, but not TN.

Discussion
In this study, we investigated the effect of soil pH or drought stress on 133 Cs accumulation by Napier grass in pots. Overall, our results demonstrated that plant growth and 133 Cs uptake were strongly affected by drought stress rather than by soil acidity.

Influence of soil pH or drought stress on plant growth
Plant growth was negatively affected by drought stress or soil pH (Table 3). Dry matter production, particularly in DW above , was significantly reduced in N-pH or Drought conditions compared with L-pH conditions and caused by PH inhibition. The PH reduction was more pronounced in T 2 than in T 1 . Conversely, Matsuo et al. (2002) found significantly lower dry matter yield of Napier grass with decreasing soil pH. In this study, the T/R ratio was significantly reduced with a large reduction of DW above and a small reduction in DW root under N-pH or Drought conditions. Similar to our results, soil pH and drought stress had significant (P < 0.05) negative effects on total leaf area and biomass in V. mungo growing in different soil types (Win et al., 2017). In addition, drought stress led to significantly negative effects on plant growth rate (P ≤ 0.001) in Juglans species (Liu et al., 2019) and biomass (P < 0.05) in Amaranthus hybridus L. (Vargas-Ortiz et al., 2021). Chlorophyll fluorescence can provide an insight into the ability of a plant to tolerate environmental stresses and the extent to which stresses has damaged the photosynthetic apparatus (Maxwell & Johnson, 2000). From our results, F v /F m and F v /F 0 decreased with drought stress or N-pH conditions, a significant decline with longer drought stress or growing period (M 2 and M 3 ) ( Table 2). Similar results showing that drought stress triggers the inhibition of PS II activity, decreasing the photosynthesis rate, and reducing aboveground DW have been reported for O. sativa (Auler et al., 2021), Carica papaya L. (Ruas et al., 2022), andA. hybridus L. (Vargas-Ortiz et al., 2021). In addition, a significant decrease in Chl content, F v /F m , and F v /F 0 under drought-stressed conditions was reported in H. vulgare (Li et al., 2006) and Phaseolus vulgaris L. (Mathobo et al., 2017). On the other hand, in a recent report, no significant F v /F m differences were observed in a Miscanthus × giganteus hybrid at different soil pH, probably because Fig. 3 Total 133 Cs content in aboveground or root parts of Napier grass. Values indicate the mean ± standard error (n = 9 individual plants). The same letters indicate no significant differences between conditions at a 5% level according to Tukey's multiple-range test Miscanthus species can grow in a broad range of soil pH (Tomaškin et al., 2021). In this study, however, the ChlF parameters in N-pH conditions significantly decreased compared to L-pH conditions, even though Napier grass can be cultivated between approximately pH 5.0-7.3. (Matsuo et al., 2002). In this regard, it is necessary to investigate how the soil pH affected ChlF parameters. We could not find any significant effect on SPAD of drought stress or soil pH in this study. Anjum et al. (2011) described that the reduction or lack of change in leaf Chl content due to drought stress depends on its duration or severity. In a recent study, changes in Chl content in C. papaya leaves by drought stress did not affect either sensitivity or tolerance to photoinhibition (Ruas et al., 2022).
Stomatal closure by water limiting conditions leads to reduced gas exchange (CO 2 or H 2 O) and increased T leaf , in turn reducing net photosynthetic CO 2 assimilation and transpiration rates (Kitao et al., 2021;Ndlovu et al., 2021). In this study, T leaf in fully expanded upper leaves grown under Drought conditions was significantly higher than that in L-pH conditions (Table 2). According to a recent study, T leaf and photochemical efficiency in C. papaya are similarly influenced by soil water restriction (Ruas et al., 2022), and electron transport is mainly used by photosynthesis and photorespiration (Kitao et al., 2021). Based on these evidence, the decreased F v /F m and F v /F 0 may relate to high T leaf due to stomatal closure in Napier grass. On the other hand, a significantly high T leaf was observed in plants grown under N-pH compared to L-pH conditions, even when Napier grass can be cultivated in a wide range of soil pH (Matsuo et al., 2002). Further research is needed to investigate how soil pH relates to the increase in T leaf .

Influence of soil pH or drought stress on 133 Cs accumulation
Drought stress produces a differential response in terms of efficiency in the uptake of various mineral elements. However, mineral uptake is reportedly compromised due to reduced transpiration pull after stomatal closure under drought stress, leading to mineral deficiencies (Ndlovu et al., 2021). In this study, both the 133 Cs concentration and Cs-T in aboveground and root parts were strongly influenced by drought stress, but not much by soil pH (Table 3 and Fig. 3). Similarly, 137 Cs

Cs content in Napier grass
Each correlation analysis used data from pot replication (n = 9 individual plants). *, **, and ***: significant correlation at 5%, 1%, and 0.1% levels, respectively, according to the Pearson's correlation coefficient (r) Condition concentrations in aboveground and root parts significantly decreased with drought stress in V. mungo (Win et al., 2015). Therefore, we consider that drought stress severely inhibited Cs uptake into the root from the soil and/or Cs transport from root to shoot compared with soil pH. Many studies have reported on the influence of soil pH on Cs accumulation; however, these effects are inconsistent in different plant species. For example, Cs uptake increases with decreased soil pH in R. sativus, C. sativus, H. annuus, G. max, T. aestivum, P. vulgaris (Massas et al., 2010;Skarlou et al., 1996), several grasses, legumes, forb (Ogura et al., 2014), Quercus serrata (Kanasashi et al., 2020), whereas no effect of changing pH on Cs uptake was found in grasses (Van Bergeijk et al., 1992) or R. fluitans in K + -sufficient conditions (Heredia et al., 2002). This may indicate that soil pH could indirectly affect the plant Cs concentration. With respect to the soil pH, Ogura et al. (2014) suggested a higher radio-Cs uptake by the roots under low soil pH. The increase in radio-Cs activity concentration due to decreasing soil pH might be due to increasing H + concentration which in turn decreases the potassium exchange capacity in the soil (Kanasashi et al., 2020). Similarly, more Cs is dissolved at acid pH levels due to greater competition with other cations for available sorption sites (Giannakopoulou et al., 2007).

Relationship between agronomic traits and Cs uptake
The correlations among the agronomic traits and 133 Cs content in Napier grass varied with growing conditions (Table 4). Consistently, Cs-T is closely correlated with Cs above (P < 0.001 in L-pH and N-pH, P < 0.01 in Drought) under all conditions. DW above , an important factor determining Cs-T, was negatively correlated with Cs above in L-pH or Drought conditions. In this regard, we have previously reported that a decrease in 137 Cs concentration with increasing DW was occurred in plants due to dilution effect (Kang et al., 2017). On the other hand, DW above in N-pH conditions showed a positive relationship between Cs above and Cs-T, probably because the slightly higher Cs concentration at N-pH than L-pH conditions resulted from a significant DW above reduction at N-pH conditions. Further research is needed to investigate how soil pH relates to the increase in Cs above .

Conclusion
We evaluated the effect of drought or soil pH on plant growth and 133 Cs accumulation in pot conditions. Our results suggest that: (1) There was no significant effect of soil acidity on 133 Cs accumulation in Napier grass.
(2) Chlorophyll fluorescence in leaves is influenced by drought stress or soil acidity. (3) Plant growth and 133 Cs accumulation is more affected by drought stress than by soil acidity.
Further studies are warranted to investigate the influence of soil types at various pH ranges pH and drought stress on Cs plant accumulation.