- Email: [email protected]
Theoretical model of the effect of potassium on the uptake of radiocesium by rice
Journal of Environmental Radioactivity 138 (2014) 122e131
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Theoretical model of the effect of potassium on the uptake of radiocesium by rice Shigeto Fujimura a, *, Junko Ishikawa b, Yuuki Sakuma a, Takashi Saito a, Mutsuto Sato a, Kunio Yoshioka a a b
Fukushima Agricultural Technology Centre, 116 Shimonakamichi, Takakura-Aza, Hiwada-machi, Koriyama, Fukushima 963-0531, Japan NARO Institute of Crop Science, Tsukuba, Ibaraki 305-8518, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 March 2014 Received in revised form 26 August 2014 Accepted 26 August 2014 Available online
After the accident at the Fukushima Dai-ichi Nuclear Power Plant owned by Tokyo Electric Power Company on 11 March 2011, potassium was applied to fields in the Tohoku and Kanto areas of Japan to reduce radiocesium uptake by crops. Despite the intense studies relating to the effect of potassium application on availability of radiocesium in the soil, physiological changes of radiocesium uptake by crops in response to Kþ concentration around roots remains elusive. In the present study, we developed physiological models describing the effect of Kþ on the uptake of radiocesium by rice. Two Csþ:Kþ competition models were evaluated using a wide range of data obtained from pot and field experiments: the model assuming a uniformity in the gene expression of Kþ transporter (Model I) and the model assuming the increase in the gene expression of Kþ transporter in response to Kþ concentration below threshold (Model II). The root-mean-square deviation between the measured and estimated values was larger in Model I than in Model II. Residuals were positively correlated with Kþ in Model I but showed no deflection in Model II. These results indicate that Model II explains the effect of Kþ on the uptake of radiocesium better than Model I. Model II may provide the appropriate countermeasures in inhibiting the transfer of radiocesium from soil to crop. The effect of changes in the variables in Model II on the relationship between available Kþ in soil and 137Cs uptake by plant was simulated. An increase in available 137Csþ in soil enhanced the response of 137Cs uptake to Kþ. The effects of MichaeliseMenten constant for Csþ were the inverse of the 137Csþ effect. The effect of MichaeliseMenten constant for Kþ showed the same tendency as that of 137Csþ, but the effect was much less than that of 137Csþ. An increase in the threshold of Kþ below which the gene expression of Kþ transporter increases enhanced the response of 137Cs uptake to Kþ in the high-Kþ range. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Competition Model Potassium transporter Radiocesium Rice Uptake
1. Introduction A large amount of radionuclides was released into the environment after the accident at the Fukushima Dai-ichi Nuclear Power Plant owned by Tokyo Electric Power Company on 11 March 2011. Radiocesium (134Cs and 137Cs) in agricultural products has been intensively monitored because radiocesium has a relatively long half-life and a high biological availability, and consumption of food contaminated with radiocesium is the principal route of
* Corresponding author. NARO Tohoku Agricultural Research Center, 50 Harajukuminami, Arai, Fukushima-shi, Fukushima 960-2156, Japan. Tel.: þ81 24 563 4101; fax: þ81 24 593 2155. E-mail addresses: [email protected] (S. Fujimura), [email protected] (J. Ishikawa), [email protected] (Y. Sakuma), saito_takashi_ [email protected] (T. Saito), [email protected] (M. Sato), [email protected] (K. Yoshioka). http://dx.doi.org/10.1016/j.jenvrad.2014.08.017 0265-931X/© 2014 Elsevier Ltd. All rights reserved.
internal exposure to radiocesium (Zhu and Smolders, 2000). The Japanese Ministry of Health, Labor and Welfare has established provisional regulation values for radioactive substances in food (500 Bq kg1 for grains) in March 2011 and a new limit (100 Bq kg1 for grains) in April 2012. Elevated radiocesium concentrations exceeding the Japanese provisional regulation value were found in brown rice (unpolished rice) from some paddy fields in the northern part of Fukushima Prefecture in 2011 (Saito et al., 2012). On the basis of previous research (Ehlken and Kirchner, 2002; Fujimura et al., 2013; Saito et al., 2012; Tsumura et al., 1984), large amounts of potassium (K) were applied to paddy fields in the Tohoku and Kanto areas of Japan, including Fukushima Prefecture, in 2012 and 2013 to reduce radiocesium levels in future harvests of brown rice. Theoretical models describing the effect of Kþ on radiocesium uptake by crops may suggest the appropriate amount of K to apply to radiocesium-contaminated areas to efficiently reduce radiocesium levels in crops in such areas.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Availability of radiocesium for plants is strongly influenced by soil properties such as clay content, and K and ammonium statuses (Absalom et al., 2001; Sawhney, 1972). Cesium (Cs) is strongly adsorbed by clay minerals, notably illite (Sawhney, 1966; Tsumura et al., 1984). Since Kþ and ammonium ion are significant competitors for adsorption of Csþ in the clay minerals, increasing in Kþ and ammonium ion in soil results to increase in dissociate radiocesium (Absalom et al., 2001; Sawhney, 1972), which is available for plants. Models to describe the availability of radiocesium in soil were developed by Absalom et al. (1999, 2001) and Smolders et al. (1997). These models dynamically estimated the availability of radiocesium in soil, but transfer of radiocesium from soil solution to plant was only empirically estimated. Because Cs is a group I alkali metal with chemical properties similar to those of K, plant roots take up Csþ through Kþ transport pathways (Qi et al., 2008). A high-affinity and a low-affinity system have been proposed for the transport of Kþ from soil to root (Zhu and Smolders, 2000). The high-affinity system operating at low external Kþ concentration (often below 0.3 mM) can transport Csþ efficiently whereas Csþ permeates only slowly in the low-affinity system operating at high external Kþ concentration (above 0.5e1 mM) (Zhu and Smolders, 2000). This indicated that the competitive effect of Kþ on the uptake of Csþ was important at low external Kþ concentration. Shaw and Bell (1991) developed a physiological model for the solution-to-root uptake of radiocesium in wheat, focusing on the Csþ:Kþ competition in wheat seedlings. Their model accurately described Csþ uptake by wheat roots during the 15 minuteexperiment. In the present work, we developed models describing the effect of Kþ concentration in soil on radiocesium uptake by rice plants through the cultivation period (from transplanting to harvest) and on the radiocesium concentration in brown rice on the basis of the Csþ:Kþ competition model developed by Shaw and Bell (1991). We developed two models hypothesizing that the Kþ transport system involves transport of Csþ from soil to plant. The velocity of Csþ uptake is dependent on the gene expression of Kþ transporters in the roots. Because the gene expression is greatly enhanced (several-fold) by low Kþ concentration in the root medium (Ma et al., 2012), one of two models integrated the change in the gene expression of Kþ transporter into the Csþ:Kþ competition model. The gene expression of Kþ transporter was not considered in the other model. Both models were evaluated by using data obtained from pot and field experiments.
VCs ¼
123
VmaxðCsÞ ½Cs ½Cs þ KmðCsÞ
(1)
where VCs is the velocity of Csþ uptake by plant, Vmax(Cs) is the maximum velocity of Csþ uptake, [Cs] is the concentration of Csþ available for plants, and Km(Cs) is the MichaeliseMenten constant for Csþ. The transport of Csþ through the system is affected by inhibitor þ (K ). The K0 m(Cs) (Km(Cs) under competitive inhibition by Kþ) is defined as. 0
KmðCsÞ ¼ KmðCsÞ
½K 1þ KmðKÞ
! (2)
where [K] is the concentration of Kþ available for plants, and Km(K) is the MichaeliseMenten constant for Kþ. As described by Shaw and Bell (1991), the velocity of Csþ uptake inhibited by Kþ is
VmaxðCsÞ ½Cs . ½Cs þ KmðCsÞ 1 þ ½K KmðKÞ h i , !! VmaxðCsÞ Cs KmðKÞ ½Cs ¼ ½K þ KmðKÞ 1 þ KmðCsÞ KmðCsÞ
VCs ¼
(3)
Variables such as [Cs], [K], Km(Cs), and Km(K) are independent, whereas Vmax(Cs) is dependent on the gene expression of Kþ transporter. Expression of Kþ transporter is up-regulated by Kþ deficiency (Ma et al., 2012). This means that Vmax(Cs) increases under Kþ deficiency, although Ma et al. (2012) did not fully elucidate the quantitative relationship between [K] and the gene expression of the Kþ transporter. In the case of NOe 3 in Arabidopsis, gene expression of high-affinity nitrate transporter is induced linearly with decreasing NOe 3 concentration of culture solution below the threshold (Zhuo et al., 1999). On the basis of the result for the nitrate transporter, the gene expression of Kþ transporter is assumed to be linearly induced by the decrease in [K] below the threshold. The gene expression of Kþ transporter is.
E¼b
ð½K Klimit Þ
(4)
E ¼ að½K Klimit Þ þ b
ð½K < Klimit Þ
(5)
where E is the gene expression of Kþ transporter, Klimit is the threshold of [K] below which the gene expression of Kþ transporter increases, a is a proportionality constant, and b is the gene expression of Kþ transporter above Klimit. Because Vmax(Cs) is positively related to E, Vmax(Cs) is
2. Materials and methods 2.1. Model development The mechanism for Csþ uptake by plants is based on MichaeliseMenten kinetics without inhibitors:
VmaxðCsÞ ¼ gE ¼ bg
ð½K Klimit Þ
(6)
Table 1 Effect of applied K on exchangeable Kþ content in soil, dry matter production, and uptake of137Cs in the pot experiments for brown rice. Applied K (g per pot)
Exchangeable Kþ content (cmolc kg1)
0.00 0.08 0.17 0.33 0.50 P value
0.038a 0.041b 0.046bc 0.059bc 0.090c <0.001
137
Dry weight (g per pot) DWbrown 13.3b 15.1ab 14.6ab 15.6a 15.8a 0.01
rice
DWtotal
DWbrown
26.7b 29.8ab 28.7ab 30.2ab 30.3a 0.04
0.50 0.51 0.51 0.52 0.52
rice/DWtotal
[137Cs]brown
Cs activity (Bq per pot)
Abrown
rice
3.1a 2.2b 1.3c 0.6d 0.2d <0.001
Values followed by the same letters (aee) are not different at the 5% level by the Tukey HSD test.
ACs-137
[P137Cs]
7.3a 6.1b 4.6c 2.1d 0.8e <0.001
0.43 0.36 0.28 0.27 0.21
(Abrown rice/A
137
Cs)
220a 130b 83c 34d 10d <0.001
rice
(Bq kg1)
124
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Fig. 1. Relationships between exchangeable Kþ content in soil and total activity of 137Cs in plant (panels a and b) and concentration of 137Cs in brown rice (panels c and d) in the pot experiments. The points correspond to data for individual pots. Lines indicate values calculated by Models I and II.
VmaxðCsÞ ¼ gE ¼ agð½K Klimit Þ þ bg
ð½K < Klimit Þ
Because the velocity of 137Csþ uptake by plants (V137Cs) is dependent on the ratio of [137Cs] to [Cs], V137Cs is.
(7)
where g is a proportionality constant. The VCs is found by substituting Eqs. (6) and (7) into Eq. (3):
VCs ¼
h i , bg Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½Cs 1þ KmðCsÞ
V137 Cs ¼
!!
½137 Cs VCs ½137 Cs þ ½* Cs
(12)
The V137Cs is found by substituting Eqs. (10) and (11) into Eq. (12):
ð½K Klimit Þ (8)
VCs ¼
h i , ð agð½K Klimit Þ þ bgÞ Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½Cs 1þ KmðCsÞ
The [137Cs] (the concentration of 137Csþ available for plants) plus [*Cs] (sum of the concentrations of Csþ available for plants other than 137Cs þ) is substituted for [Cs] in Eqs. (8) and (9).
VCs ¼
VCs ¼
h i h i , bg 137 Cs þ * Cs KmðKÞ KmðCsÞ ð agð½K Klimit Þ þ bgÞ
h
½K þ KmðKÞ
137 Cs
KmðCsÞ
i
þ
h
i
* Cs
½137 Cs þ ½* Cs 1þ KmðCsÞ KmðKÞ
!! ð½K < Klimit Þ
(9)
The values of Km(Cs) are the same for stable Csþ and radioactive Cs because the affinity of the Kþ transporter is similar toward both þ
!!
, ½K þ KmðKÞ
ð½K Klimit Þ
½137 Cs þ ½* Cs 1þ KmðCsÞ
(10) !! ð½K < Klimit Þ
(11)
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
V137 Cs ¼
V137 Cs ¼
h i , bg 137 Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ 1 þ
½137 Cs þ ½* Cs KmðCsÞ
h i , ð agð½K Klimit Þ þ bgÞ 137 Cs KmðKÞ KmðCsÞ
!! ð½K Klimit Þ
½K þ KmðKÞ 1 þ
types of Csþ. The Km(Cs) for Csþ uptake by roots of wheat seedlings is 5.8 mM (Shaw and Bell, 1991), a concentration equivalent to a 137Csþ activity concentration of 2.6 MBq L1, which is much higher than the expected [137Cs] in soil contaminated as a result of a nuclear power plant accident or by routine releases of radiocesium from a nuclear power plant. Therefore, V137Cs is approximated as. Here, the effect of the concentration of available Kþ on 137Cs uptake is considered because the main focus of this study is to
V137 Cs ¼
V137 Cs ¼
h i , bg 137 Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½* Cs 1þ KmðCsÞ
h i , ð agð½K Klimit Þ þ bgÞ 137 Cs KmðKÞ KmðCsÞ
a ½K þ b
V137 Cs ¼
c d ½K þ b
ð½K Klimit Þ
½K þ KmðKÞ
(18)
bg½137 CsKmðKÞ KmðCsÞ
ð½K < Klimit Þ
(14)
(19)
where A137Cs is the total activity of 137Cs uptake during the whole cultivation period, and d is a proportionality constant. Thus, A137Cs is a function of V137Cs. The concentration of
137
Cs in brown rice is.
(15) !! ð½K < Klimit Þ
½137 Csbrown rice ¼ ¼
(16)
P137 Cs DWbrown rice
P137 Cs DWbrown rice
A137 Cs (20)
dtV137 Cs
where [137Cs]brown rice is the concentration of 137Cs in brown rice, P137Cs is the ratio of the total activity of 137Cs in brown rice to the total activity of 137Cs uptake during the cultivation period, and DWbrown rice is the dry weight of brown rice. The [137Cs]brown rice is a function of V137Cs when P137Cs and DWbrown rice are assumed to be independent from the variables in Eqs. (15) and (16). 2.2. Models tested
!
b ¼ KmðKÞ 1 þ
½* Cs KmðCsÞ
c ¼ g aKmðKÞ
½* Cs 1þ KmðCsÞ
d¼
!!
A137 Cs ¼ dV137 Cs
½* Cs 1þ KmðCsÞ
where a, b, c, and d are parameters as follows:
a¼
½137 Cs þ ½* Cs KmðCsÞ
ð½K Klimit Þ
(17)
ð½K < Klimit Þ
(13)
!!
clarify the quantitative effect of K application on 137Cs uptake. Equations (15) and (16) are simply rearranged as a function of [K].
V137 Cs ¼
125
!
! þ aKlimit þ b
. ½137 CsKmðKÞ KmðCsÞ
ag ½137 CsKmðKÞ KmðCsÞ
The foregoing equations were established to obtain V137Cs at each stage of the cultivation period. If the total 137Csþ uptake during the whole cultivation period is assumed to be proportional to a V137Cs at an arbitrary stage, thus.
We tested two models: Model I assuming a uniformity in the gene expression of Kþ transporter [i.e., Eq. (17)] and Model II assuming the increase in the gene expression of Kþ transporter in response to Kþ concentration below Klimit [i.e., Eqs. (17) and (18) with Klimit]. Parameters in the models were restricted above zero and estimated by nonlinear least-squares analyses of data collected in the pot and field experiments described below using the solver tool in Excel 2013. The exchangeable Kþ content in soil was used as [K] because exchangeable Kþ is the main source of available Kþ for plants and thus is an effective index to evaluate the amount of available Kþ. The [K] could change during cultivation owing to Kþ uptake by plant, runoff from the paddy field, irrigation water inflow, and so on. To test the models, we used the exchangeable Kþ content in soil sampled at harvest as a representative value of [K] to analyze the
126
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
effect of [K] on A137Cs and [137Cs]brown rice because the data of exchangeable Kþ content in soil was only available at harvest in the pot experiment and the field experiment in 2011. To test the models, we expressed goodness of fit as the rootmean-square deviation (RMSD):
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u n u1 X ðESTi MSRi Þ2 RMSD ¼ t n i¼1 where MSRi and ESTi are the measured and estimated values, respectively, and n is the number of data sets. The RMSD is often used as a measure of the accuracy of a model (Kobayashi and Salam, 2000; Shimono et al., 2005). Residuals were calculated and the relationship between exchangeable Kþ in soil and residuals was tested by Spearman's rank correlation coefficient.
48 h at 105 C and 80 C, respectively, in a ventilated oven. The 137Cs concentrations in the plants were determined with germanium detectors (GC4020-7500SL-2002CSL, GC3520-7500SL-2002CSL and GC3020-7500SL-2002CS, Canberra, Meriden, Connecticut, USA). The measurement results were decay-corrected to 25 June 2012. The total activity of 137Cs in the whole plant (excluding the root) was defined as the sum of the activities of 137Cs in the brown rice, husk, and culm and leaf. After rice harvesting, soil was dried to determine the exchangeable Kþ content in soil. Exchangeable Kþ was extracted using the semimicro-Schollenberger method (Schollenberger and Simon, 1945), and Kþ concentrations were measured with an atomic absorption spectrophotometer (AA280FS, Agilent Technologies Japan, Ltd., Tokyo). The data were statistically analyzed using ANOVA followed by Tukey's HSD test as a completely randomized design.
2.3. Pot experiments
2.4. Field experiments
Pot experiments were conducted in a glasshouse at the Fukushima Agricultural Technology Centre, Fukushima, Japan, using the rice cultivar Kochiminori (Oryza sativa L. cv. Kochiminori). Five seedlings at the third leaf stage were transplanted to a Wagner pot (200 cm2, 20-cm height) on 16 April 2012. Each pot was filled with 3.5 kg of soil collected from a paddy field, where the field experiments mentioned below were conducted, in Obama/Nihonmatsu, Fukushima Prefecture, in October 2011. The soil was classified as Stagnic Anthrosols in WRB (FAO, 2006; Saito et al., 2012) and its particle size distribution was 15.3% clay, 14.2% silt and 70.5% sand (Fujimura et al., 2013). The concentration of 137Cs in the soil was around 2000 Bq kg1 and the pH was 5.4. Fertilizer was applied as basal dressing (0.1 g each of nitrogen (N) and phosphorus (P) per pot) and top dressing (0.1 g of N per pot). The rates of K application were 0, 0.08, 0.17, 0.33, and 0.50 g per pot. Three pots were used for each K treatment. Fertilizer including K was incorporated into the soil before transplanting. Ammonium sulfate, super phosphate of lime and potassium chloride were used for N, P and K, respectively. Rice plants were harvested at maturity (25 June 2012). The above-ground part of each plant was divided into brown rice, husk, and culm and leaf. The brown rice and other parts were dried for
Field experiments were conducted in three paddy fields in Obama/Nihonmatsu, Fukushima Prefecture, in 2011 and 2012 using the rice cultivar Koshihikari (Oryza sativa L. cv. Koshihikari). The same three paddy fields were used in both years. Details of the experiments were previously reported (Saito et al., 2012; Sakuma and Sato, 2014), and data for sites 3, 4, and 5 in Saito et al. (2012) and for the experiments 1 and 2 in Sakuma and Sato (2014) were used. Briefly, fertilizer was applied at a rate of 5.2 g of N, 5.6 g of P, and 3.3 g of K per m2 before transplanting, and at a rate of 0.5 g of P and 1.0 g of K per m2 through topdressing on 25 July 2011. Rice seedlings were transplanted on 15 May 2011. Brown rice and soil cores sampled at a depth of 0e15 cm were collected from 15 plots on 2 October 2011. In the field experiments of 2012, the same rate of fertilizer was applied before transplanting as in the field experiments of 2011, plus additional fertilizer (K and/or zeolitic fertilizer) was applied before transplanting. Rice seedlings were transplanted on 20 May 2012. Brown rice and soil cores sampled at a depth of 0e15 cm were collected from 21 plots on 27 September 2012. Soil solution (Soil Water Sampler, DIK-8392, Daiki Rika Kogyo Co., Ltd.) was collected five times (12 June, 21 June, 4 July, 31 July and 30 August) in 2012. It was collected from four plots for each paddy
Fig. 2. Relationships between exchangeable Kþ content in soil and residuals of 137Cs uptake by plant in the pot experiments. Residuals were calculated by subtracting the measured value from the predicted value. The points correspond to data for individual pots.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
127
Fig. 3. Relationships between exchangeable Kþ content in soil and concentration of 137Cs in brown rice in the field experiments. The points correspond to data for individual plots. Lines indicate values calculated by Models I and II.
field with the exception on 31 July when it was collected from three plots for two paddy fields and four plots for one paddy field. The concentration of 137Cs in brown rice was determined with germanium detectors. The results were decay-corrected to 1 October 2012. Exchangeable Kþ content in soil was measured by the same method as that used in the pot experiments. The Kþ concentration in soil solution was measured with the atomic absorption spectrophotometer.
19% and decreased A137Cs and P137Cs by 89% and 51%, respectively. As a result, the value of [137Cs]brown rice was 95% lower for the experiment with 0.5 g of applied K per pot than for the experiment with no applied K.
3.2. Relationship between uptake of 137Cs and exchangeable Kþ content in soil in the pot experiments
3. Results 3.1. Effect of application of K on dry matter production and uptake by rice
137
Cs
A wide range of data was obtained from the pot experiments to estimate the parameters in the models. Exchangeable Kþ content in soil at harvest, A137Cs, and [137Cs]brown rice were in the range of 0.038e0.090 cmolc kg1, 0.8e7.3 Bq per pot, and 10e220 Bq kg1, respectively (Table 1). Application of K significantly increased DWbrown rice and total dry weight (DWtotal), though the DWbrown rice/DWtotal ratio was similar for all K treatments. The activity of 137Cs in brown rice (Abrown rice) and A137Cs were significantly decreased by application of K. The ratio of the activity of 137Cs in brown rice to the total activity of 137Cs in the whole plant (P137Cs) was also decreased by application of K. As a result, [137Cs]brown rice was significantly decreased by application of K. Compared to the experiment with no applied K, application of 0.5 g of K per pot increased DWbrown rice by
The A137Cs and [137Cs]brown rice were negatively correlated with [K] in the pot experiments (Fig. 1). Parameters in Model I and Model II were determined by the least square fitting using results of the pot experiments. The values of Klimit calculated from parameters a, b, c, and d in Model II were 0.070 cmolc kg1 from A137Cs (Figs. 1b) and 0.068 cmolc kg1 from [137Cs]brown rice (Fig. 1d). The RMSDs of A137Cs in Fig. 1a and b were 1.4 Bq per pot in Model I and 0.96 Bq per pot in Model II, while those of [137Cs]brown rice in Fig. 1c and d were 53 Bq kg1 in Model I and 37 Bq kg1 in Model II. Residuals in Model I were negative in the low-[K] range and positive in the high-[K] range (Fig. 2). Although the residuals in Model II tended to be larger in the low-[K] range than in the high-[K] range, they evenly distributed negative and positive side. Spearman's rank correlation coefficient for the relationship between exchangeable Kþ in soil and residuals was significant in both A137Cs and [137Cs]brown rice in Model I (p ¼ 0.002 and p ¼ 0.008, respectively), but not significant in A137Cs nor [137Cs]brown rice in Model II (p ¼ 0.55 and p ¼ 0.94, respectively).
128
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Fig. 4. Relationships between exchangeable Kþ content in soil and residuals of concentration of 137Cs in brown rice in the field experiments. Residuals were calculated by subtracting the measured value from the predicted value. The points correspond to data for individual plots.
3.3. Relationship between uptake of 137Cs and exchangeable Kþ content in soil in the field experiments The range and geometric mean of [137Cs]brown rice were 36e170 Bq kg1 and 81 Bq kg1, respectively, in 2011, and 3.0e54 Bq kg1 and 14 Bq kg1, respectively, in 2012 in the field experiments (Fig. 3). The [137Cs]brown rice was significantly higher in 2011 than in 2012 (p < 0.001 by t test). The range and geometric mean of exchangeable Kþ content in soil were 0.049e0.103 cmolc kg1 and 0.078 molc kg1, respectively, in 2011, and 0.049e0.178 cmolc kg1 and 0.093 cmolc kg1, respectively, in 2012. The difference between the exchangeable Kþ content in soil in 2011 and that in 2012 was statistically significant (p ¼ 0.04 by t test). The [137Cs]brown rice and [K] were negatively correlated, and the response of [137Cs]brown rice to [K] tended to be greater in 2011 than in 2012. Parameters in Model I and Model II were determined by the least square fitting using those data. For Model II, a single equation [Eq. (18)] best fit the data for the field experiments carried out in 2011, suggesting that most of the [K] was below Klimit in 2011. The calculated value of Klimit was 0.078 cmolc kg1 for the experiments in 2012. The RMSDs for Models I and II were 27 and 25 Bq kg1, respectively, in 2011, and 8.3 and 4.6 Bq kg1, respectively, in 2012. The relationship between [K] and residuals were not clear in year 2011 in both models (Fig. 4). In year 2012, the relationship between [K] and residuals showed the same tendency as that in the pot experiments (Fig. 2). Spearman's rank correlation coefficient for the
Table 2 Concentration of Kþ in soil solution in the field experiment in 2012. Concentration of Kþ in soil solution (mM)
Mean ± SD Highest Lowest Number of samples
12 June
21 June
4 July
31 July
30 August
86 ± 29 146 48 12
81 ± 30 147 50 12
61 ± 23 111 29 12
7±6 19 2 10
7±4 13 3 12
relationship between exchangeable Kþ in soil and residuals was significant at 10% level in Model I in year 2012. Table 2 showed Kþ concentration in soil solution collected in 2012. It was in the rage of 48e147 mM at 12 and 21 June (23 and 32 d after transplanting) and declined less than 20 mM by 31 July (72 d after transplanting). 3.4. Simulation of the effect of changes in the variables on the relationship between exchangeable Kþ content in soil and total 137Cs uptake in plant Among the variables ([137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g) in Model II, [137Cs] and [*Cs] are mainly dependent on soil conditions, whereas the other variables are dependent on environmental conditions and plant characteristics. We calculated the effect of changing the value of each variable on the relationship between A137Cs and [K]. Simulations were performed using the values of the parameters (a, b, c, and d) determined from the pot experiments (Fig. 1). The value of the variable of interest was changed 0.25-, 1-, and 4-fold, and values of the parameters were changed simultaneously. Since the effects of changing the values of [*Cs] and Km(Cs) on the relationship between A137Cs and [K] are dependent on the relative value of [*Cs] to Km(Cs), these effects were calculated for three cases; the values of [*Cs] are 0.1-, 1- and 10-fold of Km(Cs). Simulated lines for each set of parameters are shown for [137Cs], Km(K), Klimit, a, b, and g in Fig. 5 and for [*Cs] and Km(Cs) in Fig. 6. An increase in [137Cs] or g enhanced the response of A137Cs to [K]. The effects of [*Cs] and Km(Cs) were the inverse of the [137Cs] and g effects. In addition, the effect of [*Cs] was much less when the value of [*Cs] was 0.1-fold of Km(K) than those when the values of [*Cs] was 1- and 10-fold of Km(K). The effect of Km(K) showed the same tendency as those of [137Cs] and g, but the effect on A137Cs was much less than those of [137Cs] and g. An increase in Klimit enhanced the response of A137Cs to [K] in the high-[K] range. An increase in a did not affect A137Cs above Klimit but enhanced the response of A137Cs to [K] below Klimit. An increase in b enhanced the response of A137Cs to [K] above Klimit and increased A137Cs below Klimit.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
129
Fig. 5. Effect of change in [137Cs], Km(K), Klimit, a, b, and g on the relationship between exchangeable Kþ content in soil and total activity of 137Cs in plant. Simulations were performed using the values of a, b, c, and d determined from the pot experiments (Fig. 1).
Fig. 6. Effect of change in [*Cs] and Km(Cs) on the relationship between exchangeable Kþ content in soil and total activity of 137Cs in plant. The effects were calculated for three cases; the values of [*Cs] are 0.1-, 1- and 10-fold of Km(Cs). The ratio of [*Cs] to Km(Cs) at 1-fold of each parameter is shown in each subfigure. Simulations were performed using the values of a, b, c, and d determined from the pot experiments (Fig. 1).
130
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
4. Discussion
4.2. Effects of changes in the variables on 137Csþ uptake in Model II
4.1. Effectiveness of the model to explain the relationship between 137 Cs uptake and exchangeable Kþ content in soil
The effect of the variables ([137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g) in Model II on the response of 137Cs uptake to [K] is shown in Figs. 5 and 6. Figs. 5 and 6 indicate that smaller values of Km(K), Klimit, a, b, and g, and larger values of Km(Cs), are desirable to decrease contamination of food by radiocesium. Suitable values of Km(K), Km(Cs), Klimit, a, b, and g were not obtained in the present study. Further study is needed to clarify the effects of environmental conditions on the variables and the variations of the physiological variables (e.x. Km(K) and Km(Cs)) between species and cultivars.
Competitive effect of Kþ on the uptake of Csþ was important at low external Kþ concentration (often below 0.3 mM) when the high-affinity system operates to transport Kþ and Csþ (Zhu and Smolders, 2000). The Kþ concentration in soil solution was below 0.3 mM through the growth in the field experiment in 2012 (Table 2). In the pot experiment applying 0.33 g K per pot, the Kþ concentration was 0.6 mM and <0.1 mM at 9 d and 21 d after transplanting, respectively (unpublished data). These results suggested that the high-affinity system operated to transport Kþ and Csþ in the present study. Two Csþ:Kþ competition models were compared in this study: Model I, which assumes a uniformity in the gene expression of Kþ transporter; and Model II, which assumes the increase in the gene expression of Kþ transporter below Klimit. Model II could explain the effect of [K] on A137Cs and [137Cs]brown rice more precisely than Model I. The RMSD was lower in Model II than in Model I both in the pot experiments and in the field experiments (Figs. 1 and 3). A positive relationship between residuals and [K] was evident in Model I especially in the pot experiments. Thus, uptake of 137Cs was underestimated in the low-[K] range and overestimated in the high[K] range in Model I. There was no deflection of residuals in Model II, although residuals tended to be larger in the low-[K] range. We assumed the increase in the gene expression of high-affinity Kþ transporter below Klimit in Model II based on the study in Ma et al. (2012). They reported that the expression of high-affinity Kþ transporter is up-regulated by Kþ deficiency (Ma et al., 2012). The quantitative relationship between [K] and the gene expression of the high-affinity Kþ transporter should be studied in future. In the present study, some assumptions were used to develop the Model II including the gene expression of high-affinity Kþ transporter. The total 137Csþ uptake during the whole cultivation period was assumed to be proportional to a V137Cs at an arbitrary stage in the models. The changes in variables in the model during the growth period could affect the total 137Csþ uptake. In addition, we used the exchangeable Kþ content at the harvest as the representative value of those during the growth. In the result of field experiment in 2012, the exchangeable Kþ content in the soil changed during the growth stage (Sakuma and Sato, 2014). The exchangeable Kþ content at harvest was correlated significantly and positively with those at the other stages, but the ratio of exchangeable Kþ contents between different plots were not constant throughout growth period. The changes in the exchangeable Kþ content during growth could affect the total 137Csþ uptake during the whole cultivation period. Thus, these assumptions should be verified in future work. Ongoing trials are investigating the effect of changes in the variables during the growth stage on the 137 þ Cs uptake, and examining the effect of changes in [K] during growth stages on the total 137Csþ uptake to improve Model II developed in the present study. Equation (20) indicates that [137Cs]brown rice depends on A137Cs, 137 Cs. The pot experiments indicated that DWbrown rice, and P [137Cs]brown rice was mainly dependent on A137Cs; however, the effect of [K] on DWbrown rice and P137Cs should be considered. The effect of [K] on P137Cs was greater than that on DWbrown rice; K applied at 0.5 g per pot decreased the value of P137Cs to around half that with no K application (Table 1). This result suggests that the value of [137Cs]brown rice in the models may be slightly underestimated in the low-[K] range. The effect of K application on P137Cs should be examined in detail to predict realistic values of the concentration of 137Cs in brown rice.
4.3. Factor for the difference in [137Cs]brown
rice
between years
The values of [137Cs]brown rice obtained in the field experiments of 2011 were significantly different from those of 2012 (Fig. 3). The [137Cs]brown rice was higher in 2011 than in 2012, even at similar [K], suggesting that [K] solely cannot explain the difference in the [137Cs]brown rice between years and the other factors were also responsible for the difference. In the Model II developed in the present study, [137Cs]brown rice depends on [137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g. Variables other than [137Cs] and [*Cs] could be affected by environmental conditions such as soil pH and temperature or by plant cultivar. The most of [*Cs] consisted of stable Cs. The concentration of stable Cs available for plant uptake and soil pH in 2011 were similar to that in 2012 because the same paddy fields were used for the experiments carried out during these years. There were no cultivar effects, because the same cultivar was used in both years. Seasonal average air temperature, lowest monthly mean air temperature, and highest monthly mean air temperature from May to September were 21.5, 15.8, and 24.8 C in 2011, respectively, and 21.5, 15.7, and 26.6 C in 2012, respectively (temperatures recorded at the Nihonmatsu AMeDAS site of the Japan Meteorological Agency, located 20 km from the paddy fields). These data suggest that the effect of inter annual variations in temperature would be minimal and that the values of Km(K), Km(Cs), Klimit, a, b, and g for 2011 would be similar to those for 2012. Thus the difference in the concentration of 137Cs in brown rice was mainly dependent on [137Cs]. It is well known that Csþ is irreversibly fixed to clay minerals, and the activity of exchangeable Csþ in soil decreases with time, resulting in a decrease in the uptake of Csþ by plant with time (Ehlken and Kirchner, 2002; Tsumura et al., 1984). 5. Conclusions We have shown that the model assuming the increase in the gene expression of Kþ transporter below Klimit can predict the total activity of 137Cs uptake by rice plants and the 137Cs concentration in brown rice using variables such as kinetic constants for uptake of Csþ and Kþ, available concentration of Csþ and Kþ in soil, and variables relating to the gene expression of Kþ transporter. Further study is needed to determine suitable values for these variables and verify the assumptions in developing the model. Acknowledgments This study was partially supported by grants from ZEN-NOH (National Federation of Agricultural Cooperative Associations) and Japan's Ministry of Agriculture, Forestry and Fisheries (“Development of Radioactive Materials Removal and Reduction Technology for Forests and Farmland”). We thank S. Ohkoshi, K. Iwabuchi, M. Saito, T. Nemoto, M. Sato, M. Akiyama, and T Yaginuma (Fukushima Agricultural Technology Centre) for collecting data, T. Shinano
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
(NARO Tohoku Agricultural Research Center), S. Fujita and Y. Suzuki (Fukushima Agricultural Technology Centre) for discussion and ELSS, inc. for providing language help. References Absalom, J.P., Young, S.D., Crout, N.M.J., Sanchez, A., Wright, S.M., Smolders, E., Nisbet, A.F., Gillett, A.G., 2001. Predicting the transfer of radiocaesium from organic soils to plants using soil characteristics. J. Environ. Radioact. 52, 31e43. Absalom, J.P., Young, S.D., Crout, N.M.J., Nisbet, A.F., Woodman, R.F.M., Smolders, E., Gillett, A.G., 1999. Predicting soil to plant transfer of radiocesium using soil characteristics. Environ. Sci. Technol. 33, 1218e1223. Ehlken, S., Kirchner, G., 2002. Environmental processes affecting plant root uptake of radioactive trace elements and variability of transfer factor data: a review. J. Environ. Radioact. 58, 97e112. FAO, 2006. World Reference Base for Soil Resources 2006. In: World Soil Resour. Rep., vol. 103, pp. 1e128. Fujimura, S., Yoshioka, K., Saito, T., Sato, M., Sato, M., Sakuma, Y., Muramatsu, Y., 2013. Effects of applying potassium, zeolite and vermiculite on the radiocesium uptake by rice plants grown in paddy field soils collected from Fukushima Prefecture. Plant Prod. Sci. 16, 166e170. Kobayashi, K., Salam, M.U., 2000. Comparing simulated and measured values using mean squared deviation and its components. Agron. J. 92, 345e352. Ma, T.-L., Wu, W.-H., Wang, Y., 2012. Transcriptome analysis of rice root responses to potassium deficiency. BMC Plant Biol. 12, 161. Qi, Z., Hampton, C.R., Shin, R., Barkla, B.J., White, P.J., Schachtman, D.P., 2008. The high affinity Kþ transporter AtHAK5 plays a physiological role in planta at very low Kþ concentrations and provides a caesium uptake pathway in Arabidopsis. J. Exp. Bot. 59, 595e607.
131
Saito, T., Ohkoshi, S., Fujimura, S., Iwabuchi, K., Saito, M., Nemoto, T., Sato, M., Sato, M., Yoshioka, K., Tsukada, H., 2012. Effect of potassium application on root uptake of radiocesium in rice. In: Proceedings of International Symposium on Environmental Monitoring and Dose Estimation of Residents after Accident of TEPCO's Fukushima Daiichi Nuclear Power Stations. Kyoto University Research Reactor Institute, Osaka, pp. 165e169. Sakuma, Y., Sato, M., 2014. Inhibition of radiocesium absorption into rice plant by application of zeolite and potassium materials. In: Bulletin of the Fukushima Agricultural Technology Centre Radioactive Substances Special Issue, pp. 37e40 (in Japanese). Sawhney, B.L., 1966. Kinetics of cesium sorption by clay minerals. Soil. Sci. Soc. Am. J. 30, 565e569. Sawhney, B.L., 1972. Selective sorption and fixation of cations by clay minerals: a review. Clays Clay Min. 20, 93e100. Schollenberger, C.J., Simon, R.H., 1945. Determination of exchange capacity and exchangeable bases in soil-ammonium acetate method. Soil. Sci. 59, 13e24. Shaw, G., Bell, J.N.B., 1991. Competitive effects of potassium and ammonium on caesium uptake kinetics in wheat. J. Environ. Radioact. 13, 283e296. Shimono, H., Hasegawa, T., Moriyama, M., Fujimura, S., Nagata, T., 2005. Modeling spikelet sterility induced by low temperature in rice. Agron. J. 97, 1524e1536. Smolders, E., Van den Brande, K., Merckx, R., 1997. Concentrations of 137Cs and K in soil solution predict the plant availability of 137Cs in soils. Environ. Sci. Technol. 31, 3432e3438. Tsumura, A., Komamura, M., Kobayashi, H., 1984. Behavior of radioactive Sr and Cs in soils and soil-plant systems. Bull. Natl. Inst. Agric. Sci. Ser. B 36, 57e113. Zhu, Y.-G., Smolders, E., 2000. Plant uptake of radiocaesium: a review of mechanisms, regulation and application. J. Exp. Bot. 51, 1635e1645. Zhuo, D., Okamoto, M., Vidmar, J.J., Glass, A.D.M., 1999. Regulation of a putative high-affinity nitrate transporter (Nrt2;1At) in roots of Arabidopsis thaliana. Plant J. 17, 563e568.
Contents lists available at ScienceDirect
Journal of Environmental Radioactivity journal homepage: www.elsevier.com/locate/jenvrad
Theoretical model of the effect of potassium on the uptake of radiocesium by rice Shigeto Fujimura a, *, Junko Ishikawa b, Yuuki Sakuma a, Takashi Saito a, Mutsuto Sato a, Kunio Yoshioka a a b
Fukushima Agricultural Technology Centre, 116 Shimonakamichi, Takakura-Aza, Hiwada-machi, Koriyama, Fukushima 963-0531, Japan NARO Institute of Crop Science, Tsukuba, Ibaraki 305-8518, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 12 March 2014 Received in revised form 26 August 2014 Accepted 26 August 2014 Available online
After the accident at the Fukushima Dai-ichi Nuclear Power Plant owned by Tokyo Electric Power Company on 11 March 2011, potassium was applied to fields in the Tohoku and Kanto areas of Japan to reduce radiocesium uptake by crops. Despite the intense studies relating to the effect of potassium application on availability of radiocesium in the soil, physiological changes of radiocesium uptake by crops in response to Kþ concentration around roots remains elusive. In the present study, we developed physiological models describing the effect of Kþ on the uptake of radiocesium by rice. Two Csþ:Kþ competition models were evaluated using a wide range of data obtained from pot and field experiments: the model assuming a uniformity in the gene expression of Kþ transporter (Model I) and the model assuming the increase in the gene expression of Kþ transporter in response to Kþ concentration below threshold (Model II). The root-mean-square deviation between the measured and estimated values was larger in Model I than in Model II. Residuals were positively correlated with Kþ in Model I but showed no deflection in Model II. These results indicate that Model II explains the effect of Kþ on the uptake of radiocesium better than Model I. Model II may provide the appropriate countermeasures in inhibiting the transfer of radiocesium from soil to crop. The effect of changes in the variables in Model II on the relationship between available Kþ in soil and 137Cs uptake by plant was simulated. An increase in available 137Csþ in soil enhanced the response of 137Cs uptake to Kþ. The effects of MichaeliseMenten constant for Csþ were the inverse of the 137Csþ effect. The effect of MichaeliseMenten constant for Kþ showed the same tendency as that of 137Csþ, but the effect was much less than that of 137Csþ. An increase in the threshold of Kþ below which the gene expression of Kþ transporter increases enhanced the response of 137Cs uptake to Kþ in the high-Kþ range. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Competition Model Potassium transporter Radiocesium Rice Uptake
1. Introduction A large amount of radionuclides was released into the environment after the accident at the Fukushima Dai-ichi Nuclear Power Plant owned by Tokyo Electric Power Company on 11 March 2011. Radiocesium (134Cs and 137Cs) in agricultural products has been intensively monitored because radiocesium has a relatively long half-life and a high biological availability, and consumption of food contaminated with radiocesium is the principal route of
* Corresponding author. NARO Tohoku Agricultural Research Center, 50 Harajukuminami, Arai, Fukushima-shi, Fukushima 960-2156, Japan. Tel.: þ81 24 563 4101; fax: þ81 24 593 2155. E-mail addresses: [email protected] (S. Fujimura), [email protected] (J. Ishikawa), [email protected] (Y. Sakuma), saito_takashi_ [email protected] (T. Saito), [email protected] (M. Sato), [email protected] (K. Yoshioka). http://dx.doi.org/10.1016/j.jenvrad.2014.08.017 0265-931X/© 2014 Elsevier Ltd. All rights reserved.
internal exposure to radiocesium (Zhu and Smolders, 2000). The Japanese Ministry of Health, Labor and Welfare has established provisional regulation values for radioactive substances in food (500 Bq kg1 for grains) in March 2011 and a new limit (100 Bq kg1 for grains) in April 2012. Elevated radiocesium concentrations exceeding the Japanese provisional regulation value were found in brown rice (unpolished rice) from some paddy fields in the northern part of Fukushima Prefecture in 2011 (Saito et al., 2012). On the basis of previous research (Ehlken and Kirchner, 2002; Fujimura et al., 2013; Saito et al., 2012; Tsumura et al., 1984), large amounts of potassium (K) were applied to paddy fields in the Tohoku and Kanto areas of Japan, including Fukushima Prefecture, in 2012 and 2013 to reduce radiocesium levels in future harvests of brown rice. Theoretical models describing the effect of Kþ on radiocesium uptake by crops may suggest the appropriate amount of K to apply to radiocesium-contaminated areas to efficiently reduce radiocesium levels in crops in such areas.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Availability of radiocesium for plants is strongly influenced by soil properties such as clay content, and K and ammonium statuses (Absalom et al., 2001; Sawhney, 1972). Cesium (Cs) is strongly adsorbed by clay minerals, notably illite (Sawhney, 1966; Tsumura et al., 1984). Since Kþ and ammonium ion are significant competitors for adsorption of Csþ in the clay minerals, increasing in Kþ and ammonium ion in soil results to increase in dissociate radiocesium (Absalom et al., 2001; Sawhney, 1972), which is available for plants. Models to describe the availability of radiocesium in soil were developed by Absalom et al. (1999, 2001) and Smolders et al. (1997). These models dynamically estimated the availability of radiocesium in soil, but transfer of radiocesium from soil solution to plant was only empirically estimated. Because Cs is a group I alkali metal with chemical properties similar to those of K, plant roots take up Csþ through Kþ transport pathways (Qi et al., 2008). A high-affinity and a low-affinity system have been proposed for the transport of Kþ from soil to root (Zhu and Smolders, 2000). The high-affinity system operating at low external Kþ concentration (often below 0.3 mM) can transport Csþ efficiently whereas Csþ permeates only slowly in the low-affinity system operating at high external Kþ concentration (above 0.5e1 mM) (Zhu and Smolders, 2000). This indicated that the competitive effect of Kþ on the uptake of Csþ was important at low external Kþ concentration. Shaw and Bell (1991) developed a physiological model for the solution-to-root uptake of radiocesium in wheat, focusing on the Csþ:Kþ competition in wheat seedlings. Their model accurately described Csþ uptake by wheat roots during the 15 minuteexperiment. In the present work, we developed models describing the effect of Kþ concentration in soil on radiocesium uptake by rice plants through the cultivation period (from transplanting to harvest) and on the radiocesium concentration in brown rice on the basis of the Csþ:Kþ competition model developed by Shaw and Bell (1991). We developed two models hypothesizing that the Kþ transport system involves transport of Csþ from soil to plant. The velocity of Csþ uptake is dependent on the gene expression of Kþ transporters in the roots. Because the gene expression is greatly enhanced (several-fold) by low Kþ concentration in the root medium (Ma et al., 2012), one of two models integrated the change in the gene expression of Kþ transporter into the Csþ:Kþ competition model. The gene expression of Kþ transporter was not considered in the other model. Both models were evaluated by using data obtained from pot and field experiments.
VCs ¼
123
VmaxðCsÞ ½Cs ½Cs þ KmðCsÞ
(1)
where VCs is the velocity of Csþ uptake by plant, Vmax(Cs) is the maximum velocity of Csþ uptake, [Cs] is the concentration of Csþ available for plants, and Km(Cs) is the MichaeliseMenten constant for Csþ. The transport of Csþ through the system is affected by inhibitor þ (K ). The K0 m(Cs) (Km(Cs) under competitive inhibition by Kþ) is defined as. 0
KmðCsÞ ¼ KmðCsÞ
½K 1þ KmðKÞ
! (2)
where [K] is the concentration of Kþ available for plants, and Km(K) is the MichaeliseMenten constant for Kþ. As described by Shaw and Bell (1991), the velocity of Csþ uptake inhibited by Kþ is
VmaxðCsÞ ½Cs . ½Cs þ KmðCsÞ 1 þ ½K KmðKÞ h i , !! VmaxðCsÞ Cs KmðKÞ ½Cs ¼ ½K þ KmðKÞ 1 þ KmðCsÞ KmðCsÞ
VCs ¼
(3)
Variables such as [Cs], [K], Km(Cs), and Km(K) are independent, whereas Vmax(Cs) is dependent on the gene expression of Kþ transporter. Expression of Kþ transporter is up-regulated by Kþ deficiency (Ma et al., 2012). This means that Vmax(Cs) increases under Kþ deficiency, although Ma et al. (2012) did not fully elucidate the quantitative relationship between [K] and the gene expression of the Kþ transporter. In the case of NOe 3 in Arabidopsis, gene expression of high-affinity nitrate transporter is induced linearly with decreasing NOe 3 concentration of culture solution below the threshold (Zhuo et al., 1999). On the basis of the result for the nitrate transporter, the gene expression of Kþ transporter is assumed to be linearly induced by the decrease in [K] below the threshold. The gene expression of Kþ transporter is.
E¼b
ð½K Klimit Þ
(4)
E ¼ að½K Klimit Þ þ b
ð½K < Klimit Þ
(5)
where E is the gene expression of Kþ transporter, Klimit is the threshold of [K] below which the gene expression of Kþ transporter increases, a is a proportionality constant, and b is the gene expression of Kþ transporter above Klimit. Because Vmax(Cs) is positively related to E, Vmax(Cs) is
2. Materials and methods 2.1. Model development The mechanism for Csþ uptake by plants is based on MichaeliseMenten kinetics without inhibitors:
VmaxðCsÞ ¼ gE ¼ bg
ð½K Klimit Þ
(6)
Table 1 Effect of applied K on exchangeable Kþ content in soil, dry matter production, and uptake of137Cs in the pot experiments for brown rice. Applied K (g per pot)
Exchangeable Kþ content (cmolc kg1)
0.00 0.08 0.17 0.33 0.50 P value
0.038a 0.041b 0.046bc 0.059bc 0.090c <0.001
137
Dry weight (g per pot) DWbrown 13.3b 15.1ab 14.6ab 15.6a 15.8a 0.01
rice
DWtotal
DWbrown
26.7b 29.8ab 28.7ab 30.2ab 30.3a 0.04
0.50 0.51 0.51 0.52 0.52
rice/DWtotal
[137Cs]brown
Cs activity (Bq per pot)
Abrown
rice
3.1a 2.2b 1.3c 0.6d 0.2d <0.001
Values followed by the same letters (aee) are not different at the 5% level by the Tukey HSD test.
ACs-137
[P137Cs]
7.3a 6.1b 4.6c 2.1d 0.8e <0.001
0.43 0.36 0.28 0.27 0.21
(Abrown rice/A
137
Cs)
220a 130b 83c 34d 10d <0.001
rice
(Bq kg1)
124
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Fig. 1. Relationships between exchangeable Kþ content in soil and total activity of 137Cs in plant (panels a and b) and concentration of 137Cs in brown rice (panels c and d) in the pot experiments. The points correspond to data for individual pots. Lines indicate values calculated by Models I and II.
VmaxðCsÞ ¼ gE ¼ agð½K Klimit Þ þ bg
ð½K < Klimit Þ
Because the velocity of 137Csþ uptake by plants (V137Cs) is dependent on the ratio of [137Cs] to [Cs], V137Cs is.
(7)
where g is a proportionality constant. The VCs is found by substituting Eqs. (6) and (7) into Eq. (3):
VCs ¼
h i , bg Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½Cs 1þ KmðCsÞ
V137 Cs ¼
!!
½137 Cs VCs ½137 Cs þ ½* Cs
(12)
The V137Cs is found by substituting Eqs. (10) and (11) into Eq. (12):
ð½K Klimit Þ (8)
VCs ¼
h i , ð agð½K Klimit Þ þ bgÞ Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½Cs 1þ KmðCsÞ
The [137Cs] (the concentration of 137Csþ available for plants) plus [*Cs] (sum of the concentrations of Csþ available for plants other than 137Cs þ) is substituted for [Cs] in Eqs. (8) and (9).
VCs ¼
VCs ¼
h i h i , bg 137 Cs þ * Cs KmðKÞ KmðCsÞ ð agð½K Klimit Þ þ bgÞ
h
½K þ KmðKÞ
137 Cs
KmðCsÞ
i
þ
h
i
* Cs
½137 Cs þ ½* Cs 1þ KmðCsÞ KmðKÞ
!! ð½K < Klimit Þ
(9)
The values of Km(Cs) are the same for stable Csþ and radioactive Cs because the affinity of the Kþ transporter is similar toward both þ
!!
, ½K þ KmðKÞ
ð½K Klimit Þ
½137 Cs þ ½* Cs 1þ KmðCsÞ
(10) !! ð½K < Klimit Þ
(11)
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
V137 Cs ¼
V137 Cs ¼
h i , bg 137 Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ 1 þ
½137 Cs þ ½* Cs KmðCsÞ
h i , ð agð½K Klimit Þ þ bgÞ 137 Cs KmðKÞ KmðCsÞ
!! ð½K Klimit Þ
½K þ KmðKÞ 1 þ
types of Csþ. The Km(Cs) for Csþ uptake by roots of wheat seedlings is 5.8 mM (Shaw and Bell, 1991), a concentration equivalent to a 137Csþ activity concentration of 2.6 MBq L1, which is much higher than the expected [137Cs] in soil contaminated as a result of a nuclear power plant accident or by routine releases of radiocesium from a nuclear power plant. Therefore, V137Cs is approximated as. Here, the effect of the concentration of available Kþ on 137Cs uptake is considered because the main focus of this study is to
V137 Cs ¼
V137 Cs ¼
h i , bg 137 Cs KmðKÞ KmðCsÞ
½K þ KmðKÞ
½* Cs 1þ KmðCsÞ
h i , ð agð½K Klimit Þ þ bgÞ 137 Cs KmðKÞ KmðCsÞ
a ½K þ b
V137 Cs ¼
c d ½K þ b
ð½K Klimit Þ
½K þ KmðKÞ
(18)
bg½137 CsKmðKÞ KmðCsÞ
ð½K < Klimit Þ
(14)
(19)
where A137Cs is the total activity of 137Cs uptake during the whole cultivation period, and d is a proportionality constant. Thus, A137Cs is a function of V137Cs. The concentration of
137
Cs in brown rice is.
(15) !! ð½K < Klimit Þ
½137 Csbrown rice ¼ ¼
(16)
P137 Cs DWbrown rice
P137 Cs DWbrown rice
A137 Cs (20)
dtV137 Cs
where [137Cs]brown rice is the concentration of 137Cs in brown rice, P137Cs is the ratio of the total activity of 137Cs in brown rice to the total activity of 137Cs uptake during the cultivation period, and DWbrown rice is the dry weight of brown rice. The [137Cs]brown rice is a function of V137Cs when P137Cs and DWbrown rice are assumed to be independent from the variables in Eqs. (15) and (16). 2.2. Models tested
!
b ¼ KmðKÞ 1 þ
½* Cs KmðCsÞ
c ¼ g aKmðKÞ
½* Cs 1þ KmðCsÞ
d¼
!!
A137 Cs ¼ dV137 Cs
½* Cs 1þ KmðCsÞ
where a, b, c, and d are parameters as follows:
a¼
½137 Cs þ ½* Cs KmðCsÞ
ð½K Klimit Þ
(17)
ð½K < Klimit Þ
(13)
!!
clarify the quantitative effect of K application on 137Cs uptake. Equations (15) and (16) are simply rearranged as a function of [K].
V137 Cs ¼
125
!
! þ aKlimit þ b
. ½137 CsKmðKÞ KmðCsÞ
ag ½137 CsKmðKÞ KmðCsÞ
The foregoing equations were established to obtain V137Cs at each stage of the cultivation period. If the total 137Csþ uptake during the whole cultivation period is assumed to be proportional to a V137Cs at an arbitrary stage, thus.
We tested two models: Model I assuming a uniformity in the gene expression of Kþ transporter [i.e., Eq. (17)] and Model II assuming the increase in the gene expression of Kþ transporter in response to Kþ concentration below Klimit [i.e., Eqs. (17) and (18) with Klimit]. Parameters in the models were restricted above zero and estimated by nonlinear least-squares analyses of data collected in the pot and field experiments described below using the solver tool in Excel 2013. The exchangeable Kþ content in soil was used as [K] because exchangeable Kþ is the main source of available Kþ for plants and thus is an effective index to evaluate the amount of available Kþ. The [K] could change during cultivation owing to Kþ uptake by plant, runoff from the paddy field, irrigation water inflow, and so on. To test the models, we used the exchangeable Kþ content in soil sampled at harvest as a representative value of [K] to analyze the
126
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
effect of [K] on A137Cs and [137Cs]brown rice because the data of exchangeable Kþ content in soil was only available at harvest in the pot experiment and the field experiment in 2011. To test the models, we expressed goodness of fit as the rootmean-square deviation (RMSD):
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u n u1 X ðESTi MSRi Þ2 RMSD ¼ t n i¼1 where MSRi and ESTi are the measured and estimated values, respectively, and n is the number of data sets. The RMSD is often used as a measure of the accuracy of a model (Kobayashi and Salam, 2000; Shimono et al., 2005). Residuals were calculated and the relationship between exchangeable Kþ in soil and residuals was tested by Spearman's rank correlation coefficient.
48 h at 105 C and 80 C, respectively, in a ventilated oven. The 137Cs concentrations in the plants were determined with germanium detectors (GC4020-7500SL-2002CSL, GC3520-7500SL-2002CSL and GC3020-7500SL-2002CS, Canberra, Meriden, Connecticut, USA). The measurement results were decay-corrected to 25 June 2012. The total activity of 137Cs in the whole plant (excluding the root) was defined as the sum of the activities of 137Cs in the brown rice, husk, and culm and leaf. After rice harvesting, soil was dried to determine the exchangeable Kþ content in soil. Exchangeable Kþ was extracted using the semimicro-Schollenberger method (Schollenberger and Simon, 1945), and Kþ concentrations were measured with an atomic absorption spectrophotometer (AA280FS, Agilent Technologies Japan, Ltd., Tokyo). The data were statistically analyzed using ANOVA followed by Tukey's HSD test as a completely randomized design.
2.3. Pot experiments
2.4. Field experiments
Pot experiments were conducted in a glasshouse at the Fukushima Agricultural Technology Centre, Fukushima, Japan, using the rice cultivar Kochiminori (Oryza sativa L. cv. Kochiminori). Five seedlings at the third leaf stage were transplanted to a Wagner pot (200 cm2, 20-cm height) on 16 April 2012. Each pot was filled with 3.5 kg of soil collected from a paddy field, where the field experiments mentioned below were conducted, in Obama/Nihonmatsu, Fukushima Prefecture, in October 2011. The soil was classified as Stagnic Anthrosols in WRB (FAO, 2006; Saito et al., 2012) and its particle size distribution was 15.3% clay, 14.2% silt and 70.5% sand (Fujimura et al., 2013). The concentration of 137Cs in the soil was around 2000 Bq kg1 and the pH was 5.4. Fertilizer was applied as basal dressing (0.1 g each of nitrogen (N) and phosphorus (P) per pot) and top dressing (0.1 g of N per pot). The rates of K application were 0, 0.08, 0.17, 0.33, and 0.50 g per pot. Three pots were used for each K treatment. Fertilizer including K was incorporated into the soil before transplanting. Ammonium sulfate, super phosphate of lime and potassium chloride were used for N, P and K, respectively. Rice plants were harvested at maturity (25 June 2012). The above-ground part of each plant was divided into brown rice, husk, and culm and leaf. The brown rice and other parts were dried for
Field experiments were conducted in three paddy fields in Obama/Nihonmatsu, Fukushima Prefecture, in 2011 and 2012 using the rice cultivar Koshihikari (Oryza sativa L. cv. Koshihikari). The same three paddy fields were used in both years. Details of the experiments were previously reported (Saito et al., 2012; Sakuma and Sato, 2014), and data for sites 3, 4, and 5 in Saito et al. (2012) and for the experiments 1 and 2 in Sakuma and Sato (2014) were used. Briefly, fertilizer was applied at a rate of 5.2 g of N, 5.6 g of P, and 3.3 g of K per m2 before transplanting, and at a rate of 0.5 g of P and 1.0 g of K per m2 through topdressing on 25 July 2011. Rice seedlings were transplanted on 15 May 2011. Brown rice and soil cores sampled at a depth of 0e15 cm were collected from 15 plots on 2 October 2011. In the field experiments of 2012, the same rate of fertilizer was applied before transplanting as in the field experiments of 2011, plus additional fertilizer (K and/or zeolitic fertilizer) was applied before transplanting. Rice seedlings were transplanted on 20 May 2012. Brown rice and soil cores sampled at a depth of 0e15 cm were collected from 21 plots on 27 September 2012. Soil solution (Soil Water Sampler, DIK-8392, Daiki Rika Kogyo Co., Ltd.) was collected five times (12 June, 21 June, 4 July, 31 July and 30 August) in 2012. It was collected from four plots for each paddy
Fig. 2. Relationships between exchangeable Kþ content in soil and residuals of 137Cs uptake by plant in the pot experiments. Residuals were calculated by subtracting the measured value from the predicted value. The points correspond to data for individual pots.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
127
Fig. 3. Relationships between exchangeable Kþ content in soil and concentration of 137Cs in brown rice in the field experiments. The points correspond to data for individual plots. Lines indicate values calculated by Models I and II.
field with the exception on 31 July when it was collected from three plots for two paddy fields and four plots for one paddy field. The concentration of 137Cs in brown rice was determined with germanium detectors. The results were decay-corrected to 1 October 2012. Exchangeable Kþ content in soil was measured by the same method as that used in the pot experiments. The Kþ concentration in soil solution was measured with the atomic absorption spectrophotometer.
19% and decreased A137Cs and P137Cs by 89% and 51%, respectively. As a result, the value of [137Cs]brown rice was 95% lower for the experiment with 0.5 g of applied K per pot than for the experiment with no applied K.
3.2. Relationship between uptake of 137Cs and exchangeable Kþ content in soil in the pot experiments
3. Results 3.1. Effect of application of K on dry matter production and uptake by rice
137
Cs
A wide range of data was obtained from the pot experiments to estimate the parameters in the models. Exchangeable Kþ content in soil at harvest, A137Cs, and [137Cs]brown rice were in the range of 0.038e0.090 cmolc kg1, 0.8e7.3 Bq per pot, and 10e220 Bq kg1, respectively (Table 1). Application of K significantly increased DWbrown rice and total dry weight (DWtotal), though the DWbrown rice/DWtotal ratio was similar for all K treatments. The activity of 137Cs in brown rice (Abrown rice) and A137Cs were significantly decreased by application of K. The ratio of the activity of 137Cs in brown rice to the total activity of 137Cs in the whole plant (P137Cs) was also decreased by application of K. As a result, [137Cs]brown rice was significantly decreased by application of K. Compared to the experiment with no applied K, application of 0.5 g of K per pot increased DWbrown rice by
The A137Cs and [137Cs]brown rice were negatively correlated with [K] in the pot experiments (Fig. 1). Parameters in Model I and Model II were determined by the least square fitting using results of the pot experiments. The values of Klimit calculated from parameters a, b, c, and d in Model II were 0.070 cmolc kg1 from A137Cs (Figs. 1b) and 0.068 cmolc kg1 from [137Cs]brown rice (Fig. 1d). The RMSDs of A137Cs in Fig. 1a and b were 1.4 Bq per pot in Model I and 0.96 Bq per pot in Model II, while those of [137Cs]brown rice in Fig. 1c and d were 53 Bq kg1 in Model I and 37 Bq kg1 in Model II. Residuals in Model I were negative in the low-[K] range and positive in the high-[K] range (Fig. 2). Although the residuals in Model II tended to be larger in the low-[K] range than in the high-[K] range, they evenly distributed negative and positive side. Spearman's rank correlation coefficient for the relationship between exchangeable Kþ in soil and residuals was significant in both A137Cs and [137Cs]brown rice in Model I (p ¼ 0.002 and p ¼ 0.008, respectively), but not significant in A137Cs nor [137Cs]brown rice in Model II (p ¼ 0.55 and p ¼ 0.94, respectively).
128
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
Fig. 4. Relationships between exchangeable Kþ content in soil and residuals of concentration of 137Cs in brown rice in the field experiments. Residuals were calculated by subtracting the measured value from the predicted value. The points correspond to data for individual plots.
3.3. Relationship between uptake of 137Cs and exchangeable Kþ content in soil in the field experiments The range and geometric mean of [137Cs]brown rice were 36e170 Bq kg1 and 81 Bq kg1, respectively, in 2011, and 3.0e54 Bq kg1 and 14 Bq kg1, respectively, in 2012 in the field experiments (Fig. 3). The [137Cs]brown rice was significantly higher in 2011 than in 2012 (p < 0.001 by t test). The range and geometric mean of exchangeable Kþ content in soil were 0.049e0.103 cmolc kg1 and 0.078 molc kg1, respectively, in 2011, and 0.049e0.178 cmolc kg1 and 0.093 cmolc kg1, respectively, in 2012. The difference between the exchangeable Kþ content in soil in 2011 and that in 2012 was statistically significant (p ¼ 0.04 by t test). The [137Cs]brown rice and [K] were negatively correlated, and the response of [137Cs]brown rice to [K] tended to be greater in 2011 than in 2012. Parameters in Model I and Model II were determined by the least square fitting using those data. For Model II, a single equation [Eq. (18)] best fit the data for the field experiments carried out in 2011, suggesting that most of the [K] was below Klimit in 2011. The calculated value of Klimit was 0.078 cmolc kg1 for the experiments in 2012. The RMSDs for Models I and II were 27 and 25 Bq kg1, respectively, in 2011, and 8.3 and 4.6 Bq kg1, respectively, in 2012. The relationship between [K] and residuals were not clear in year 2011 in both models (Fig. 4). In year 2012, the relationship between [K] and residuals showed the same tendency as that in the pot experiments (Fig. 2). Spearman's rank correlation coefficient for the
Table 2 Concentration of Kþ in soil solution in the field experiment in 2012. Concentration of Kþ in soil solution (mM)
Mean ± SD Highest Lowest Number of samples
12 June
21 June
4 July
31 July
30 August
86 ± 29 146 48 12
81 ± 30 147 50 12
61 ± 23 111 29 12
7±6 19 2 10
7±4 13 3 12
relationship between exchangeable Kþ in soil and residuals was significant at 10% level in Model I in year 2012. Table 2 showed Kþ concentration in soil solution collected in 2012. It was in the rage of 48e147 mM at 12 and 21 June (23 and 32 d after transplanting) and declined less than 20 mM by 31 July (72 d after transplanting). 3.4. Simulation of the effect of changes in the variables on the relationship between exchangeable Kþ content in soil and total 137Cs uptake in plant Among the variables ([137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g) in Model II, [137Cs] and [*Cs] are mainly dependent on soil conditions, whereas the other variables are dependent on environmental conditions and plant characteristics. We calculated the effect of changing the value of each variable on the relationship between A137Cs and [K]. Simulations were performed using the values of the parameters (a, b, c, and d) determined from the pot experiments (Fig. 1). The value of the variable of interest was changed 0.25-, 1-, and 4-fold, and values of the parameters were changed simultaneously. Since the effects of changing the values of [*Cs] and Km(Cs) on the relationship between A137Cs and [K] are dependent on the relative value of [*Cs] to Km(Cs), these effects were calculated for three cases; the values of [*Cs] are 0.1-, 1- and 10-fold of Km(Cs). Simulated lines for each set of parameters are shown for [137Cs], Km(K), Klimit, a, b, and g in Fig. 5 and for [*Cs] and Km(Cs) in Fig. 6. An increase in [137Cs] or g enhanced the response of A137Cs to [K]. The effects of [*Cs] and Km(Cs) were the inverse of the [137Cs] and g effects. In addition, the effect of [*Cs] was much less when the value of [*Cs] was 0.1-fold of Km(K) than those when the values of [*Cs] was 1- and 10-fold of Km(K). The effect of Km(K) showed the same tendency as those of [137Cs] and g, but the effect on A137Cs was much less than those of [137Cs] and g. An increase in Klimit enhanced the response of A137Cs to [K] in the high-[K] range. An increase in a did not affect A137Cs above Klimit but enhanced the response of A137Cs to [K] below Klimit. An increase in b enhanced the response of A137Cs to [K] above Klimit and increased A137Cs below Klimit.
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
129
Fig. 5. Effect of change in [137Cs], Km(K), Klimit, a, b, and g on the relationship between exchangeable Kþ content in soil and total activity of 137Cs in plant. Simulations were performed using the values of a, b, c, and d determined from the pot experiments (Fig. 1).
Fig. 6. Effect of change in [*Cs] and Km(Cs) on the relationship between exchangeable Kþ content in soil and total activity of 137Cs in plant. The effects were calculated for three cases; the values of [*Cs] are 0.1-, 1- and 10-fold of Km(Cs). The ratio of [*Cs] to Km(Cs) at 1-fold of each parameter is shown in each subfigure. Simulations were performed using the values of a, b, c, and d determined from the pot experiments (Fig. 1).
130
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
4. Discussion
4.2. Effects of changes in the variables on 137Csþ uptake in Model II
4.1. Effectiveness of the model to explain the relationship between 137 Cs uptake and exchangeable Kþ content in soil
The effect of the variables ([137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g) in Model II on the response of 137Cs uptake to [K] is shown in Figs. 5 and 6. Figs. 5 and 6 indicate that smaller values of Km(K), Klimit, a, b, and g, and larger values of Km(Cs), are desirable to decrease contamination of food by radiocesium. Suitable values of Km(K), Km(Cs), Klimit, a, b, and g were not obtained in the present study. Further study is needed to clarify the effects of environmental conditions on the variables and the variations of the physiological variables (e.x. Km(K) and Km(Cs)) between species and cultivars.
Competitive effect of Kþ on the uptake of Csþ was important at low external Kþ concentration (often below 0.3 mM) when the high-affinity system operates to transport Kþ and Csþ (Zhu and Smolders, 2000). The Kþ concentration in soil solution was below 0.3 mM through the growth in the field experiment in 2012 (Table 2). In the pot experiment applying 0.33 g K per pot, the Kþ concentration was 0.6 mM and <0.1 mM at 9 d and 21 d after transplanting, respectively (unpublished data). These results suggested that the high-affinity system operated to transport Kþ and Csþ in the present study. Two Csþ:Kþ competition models were compared in this study: Model I, which assumes a uniformity in the gene expression of Kþ transporter; and Model II, which assumes the increase in the gene expression of Kþ transporter below Klimit. Model II could explain the effect of [K] on A137Cs and [137Cs]brown rice more precisely than Model I. The RMSD was lower in Model II than in Model I both in the pot experiments and in the field experiments (Figs. 1 and 3). A positive relationship between residuals and [K] was evident in Model I especially in the pot experiments. Thus, uptake of 137Cs was underestimated in the low-[K] range and overestimated in the high[K] range in Model I. There was no deflection of residuals in Model II, although residuals tended to be larger in the low-[K] range. We assumed the increase in the gene expression of high-affinity Kþ transporter below Klimit in Model II based on the study in Ma et al. (2012). They reported that the expression of high-affinity Kþ transporter is up-regulated by Kþ deficiency (Ma et al., 2012). The quantitative relationship between [K] and the gene expression of the high-affinity Kþ transporter should be studied in future. In the present study, some assumptions were used to develop the Model II including the gene expression of high-affinity Kþ transporter. The total 137Csþ uptake during the whole cultivation period was assumed to be proportional to a V137Cs at an arbitrary stage in the models. The changes in variables in the model during the growth period could affect the total 137Csþ uptake. In addition, we used the exchangeable Kþ content at the harvest as the representative value of those during the growth. In the result of field experiment in 2012, the exchangeable Kþ content in the soil changed during the growth stage (Sakuma and Sato, 2014). The exchangeable Kþ content at harvest was correlated significantly and positively with those at the other stages, but the ratio of exchangeable Kþ contents between different plots were not constant throughout growth period. The changes in the exchangeable Kþ content during growth could affect the total 137Csþ uptake during the whole cultivation period. Thus, these assumptions should be verified in future work. Ongoing trials are investigating the effect of changes in the variables during the growth stage on the 137 þ Cs uptake, and examining the effect of changes in [K] during growth stages on the total 137Csþ uptake to improve Model II developed in the present study. Equation (20) indicates that [137Cs]brown rice depends on A137Cs, 137 Cs. The pot experiments indicated that DWbrown rice, and P [137Cs]brown rice was mainly dependent on A137Cs; however, the effect of [K] on DWbrown rice and P137Cs should be considered. The effect of [K] on P137Cs was greater than that on DWbrown rice; K applied at 0.5 g per pot decreased the value of P137Cs to around half that with no K application (Table 1). This result suggests that the value of [137Cs]brown rice in the models may be slightly underestimated in the low-[K] range. The effect of K application on P137Cs should be examined in detail to predict realistic values of the concentration of 137Cs in brown rice.
4.3. Factor for the difference in [137Cs]brown
rice
between years
The values of [137Cs]brown rice obtained in the field experiments of 2011 were significantly different from those of 2012 (Fig. 3). The [137Cs]brown rice was higher in 2011 than in 2012, even at similar [K], suggesting that [K] solely cannot explain the difference in the [137Cs]brown rice between years and the other factors were also responsible for the difference. In the Model II developed in the present study, [137Cs]brown rice depends on [137Cs], [*Cs], Km(K), Km(Cs), Klimit, a, b, and g. Variables other than [137Cs] and [*Cs] could be affected by environmental conditions such as soil pH and temperature or by plant cultivar. The most of [*Cs] consisted of stable Cs. The concentration of stable Cs available for plant uptake and soil pH in 2011 were similar to that in 2012 because the same paddy fields were used for the experiments carried out during these years. There were no cultivar effects, because the same cultivar was used in both years. Seasonal average air temperature, lowest monthly mean air temperature, and highest monthly mean air temperature from May to September were 21.5, 15.8, and 24.8 C in 2011, respectively, and 21.5, 15.7, and 26.6 C in 2012, respectively (temperatures recorded at the Nihonmatsu AMeDAS site of the Japan Meteorological Agency, located 20 km from the paddy fields). These data suggest that the effect of inter annual variations in temperature would be minimal and that the values of Km(K), Km(Cs), Klimit, a, b, and g for 2011 would be similar to those for 2012. Thus the difference in the concentration of 137Cs in brown rice was mainly dependent on [137Cs]. It is well known that Csþ is irreversibly fixed to clay minerals, and the activity of exchangeable Csþ in soil decreases with time, resulting in a decrease in the uptake of Csþ by plant with time (Ehlken and Kirchner, 2002; Tsumura et al., 1984). 5. Conclusions We have shown that the model assuming the increase in the gene expression of Kþ transporter below Klimit can predict the total activity of 137Cs uptake by rice plants and the 137Cs concentration in brown rice using variables such as kinetic constants for uptake of Csþ and Kþ, available concentration of Csþ and Kþ in soil, and variables relating to the gene expression of Kþ transporter. Further study is needed to determine suitable values for these variables and verify the assumptions in developing the model. Acknowledgments This study was partially supported by grants from ZEN-NOH (National Federation of Agricultural Cooperative Associations) and Japan's Ministry of Agriculture, Forestry and Fisheries (“Development of Radioactive Materials Removal and Reduction Technology for Forests and Farmland”). We thank S. Ohkoshi, K. Iwabuchi, M. Saito, T. Nemoto, M. Sato, M. Akiyama, and T Yaginuma (Fukushima Agricultural Technology Centre) for collecting data, T. Shinano
S. Fujimura et al. / Journal of Environmental Radioactivity 138 (2014) 122e131
(NARO Tohoku Agricultural Research Center), S. Fujita and Y. Suzuki (Fukushima Agricultural Technology Centre) for discussion and ELSS, inc. for providing language help. References Absalom, J.P., Young, S.D., Crout, N.M.J., Sanchez, A., Wright, S.M., Smolders, E., Nisbet, A.F., Gillett, A.G., 2001. Predicting the transfer of radiocaesium from organic soils to plants using soil characteristics. J. Environ. Radioact. 52, 31e43. Absalom, J.P., Young, S.D., Crout, N.M.J., Nisbet, A.F., Woodman, R.F.M., Smolders, E., Gillett, A.G., 1999. Predicting soil to plant transfer of radiocesium using soil characteristics. Environ. Sci. Technol. 33, 1218e1223. Ehlken, S., Kirchner, G., 2002. Environmental processes affecting plant root uptake of radioactive trace elements and variability of transfer factor data: a review. J. Environ. Radioact. 58, 97e112. FAO, 2006. World Reference Base for Soil Resources 2006. In: World Soil Resour. Rep., vol. 103, pp. 1e128. Fujimura, S., Yoshioka, K., Saito, T., Sato, M., Sato, M., Sakuma, Y., Muramatsu, Y., 2013. Effects of applying potassium, zeolite and vermiculite on the radiocesium uptake by rice plants grown in paddy field soils collected from Fukushima Prefecture. Plant Prod. Sci. 16, 166e170. Kobayashi, K., Salam, M.U., 2000. Comparing simulated and measured values using mean squared deviation and its components. Agron. J. 92, 345e352. Ma, T.-L., Wu, W.-H., Wang, Y., 2012. Transcriptome analysis of rice root responses to potassium deficiency. BMC Plant Biol. 12, 161. Qi, Z., Hampton, C.R., Shin, R., Barkla, B.J., White, P.J., Schachtman, D.P., 2008. The high affinity Kþ transporter AtHAK5 plays a physiological role in planta at very low Kþ concentrations and provides a caesium uptake pathway in Arabidopsis. J. Exp. Bot. 59, 595e607.
131
Saito, T., Ohkoshi, S., Fujimura, S., Iwabuchi, K., Saito, M., Nemoto, T., Sato, M., Sato, M., Yoshioka, K., Tsukada, H., 2012. Effect of potassium application on root uptake of radiocesium in rice. In: Proceedings of International Symposium on Environmental Monitoring and Dose Estimation of Residents after Accident of TEPCO's Fukushima Daiichi Nuclear Power Stations. Kyoto University Research Reactor Institute, Osaka, pp. 165e169. Sakuma, Y., Sato, M., 2014. Inhibition of radiocesium absorption into rice plant by application of zeolite and potassium materials. In: Bulletin of the Fukushima Agricultural Technology Centre Radioactive Substances Special Issue, pp. 37e40 (in Japanese). Sawhney, B.L., 1966. Kinetics of cesium sorption by clay minerals. Soil. Sci. Soc. Am. J. 30, 565e569. Sawhney, B.L., 1972. Selective sorption and fixation of cations by clay minerals: a review. Clays Clay Min. 20, 93e100. Schollenberger, C.J., Simon, R.H., 1945. Determination of exchange capacity and exchangeable bases in soil-ammonium acetate method. Soil. Sci. 59, 13e24. Shaw, G., Bell, J.N.B., 1991. Competitive effects of potassium and ammonium on caesium uptake kinetics in wheat. J. Environ. Radioact. 13, 283e296. Shimono, H., Hasegawa, T., Moriyama, M., Fujimura, S., Nagata, T., 2005. Modeling spikelet sterility induced by low temperature in rice. Agron. J. 97, 1524e1536. Smolders, E., Van den Brande, K., Merckx, R., 1997. Concentrations of 137Cs and K in soil solution predict the plant availability of 137Cs in soils. Environ. Sci. Technol. 31, 3432e3438. Tsumura, A., Komamura, M., Kobayashi, H., 1984. Behavior of radioactive Sr and Cs in soils and soil-plant systems. Bull. Natl. Inst. Agric. Sci. Ser. B 36, 57e113. Zhu, Y.-G., Smolders, E., 2000. Plant uptake of radiocaesium: a review of mechanisms, regulation and application. J. Exp. Bot. 51, 1635e1645. Zhuo, D., Okamoto, M., Vidmar, J.J., Glass, A.D.M., 1999. Regulation of a putative high-affinity nitrate transporter (Nrt2;1At) in roots of Arabidopsis thaliana. Plant J. 17, 563e568.