Factors affecting the cesium transfer factor to shiitake (Lentinula edodes) cultivated in sawdust medium

The transfer factor (TF) of radioactive cesium-137 (Cs) to shiitake (Lentinula edodes) cultivated on bed logs varies greatly. Therefore, the present study investigated which factors affect the TF using stable cesium-133 (Cs) and sawdust medium with 5% rice bran as a model, which had similar Cs TFs to bed-log cultivation. It was found that the Cs concentration and nutrient concentration (represented by the nitrogen concentration) concerned with the TF in the model sawdust medium. In addition, the TFs calculated using total Cs and Cs concentrations differed in both bed-log cultivation and the model sawdust medium cultivation, while the TFs calculated using exchangeable Cs and Cs concentrations were the same in sawdust medium cultivation, indicating that exchangeable Cs in the medium is the source of Cs for the fruiting body and the former difference was due to the presence of other chemical speciation of Cs that could not be absorbed. One purpose of the TF on the mushroom farm is to determine the fruiting body Cs concentration at the start of bed-log cultivation, therefore the prediction method of TF are discussed considering the future changes of Cs concentrations in trees.


Introduction
The Fukushima Daiichi Nuclear Power Plant was damaged by the Great East Japan Earthquake in March 2011, resulting in radioactive materials being released into the environment and contaminating a large area of eastern Japan. These radioactive materials included an estimated 1.5 × 10 16 Bq of radioactive cesium-137 ( 137 Cs), which has a half-life of approximately 30.1 years [1].
Radioactive materials are absorbed by agricultural products, so an upper limit of radioactivity in general foods has been set at 100 Bq/kg to prevent health damage in humans [2]. Fortunately, the transfer factor (TF) of 137 Cs to plants is generally low [3], so the effect of 137 Cs contamination on plants is not currently a serious issue. However, mushrooms are reported to accumulate high concentrations of 137 Cs [4], so the radioactive contamination of mushrooms has been a problem since the nuclear power plant accident occurred due to the long half-life of this material. To address this, provisional reference indices of 50 and 200 Bq/kg were set for mushroom cultivation in bed log and sawdust medium, respectively [5], based on 137 Cs TFs to the fruiting body of 2.0 and 0.5, respectively [6].
Most of the mushrooms on the market are cultivated in sawdust medium. The limit in sawdust medium is higher than that in bed log and the radioactive concentration in sawdust medium can be easily controlled within this limit, so the effect of 137 Cs contamination on mushrooms cultivated in sawdust medium are not currently a serious issue. Most shiitake (Lentinula edodes) are also cultivated in sawdust media and the remainder in bed logs. The bed-log cultivation does not require special facilities, so they are a valuable source of income in hilly and mountainous areas [7]. However, the radioactive concentration in bed log cannot be easily controlled, which needs to be addressed.
The provisional reference index for bed-log cultivation was established based on the frequency distribution of the bed log 137 Cs TF immediately after the nuclear power plant accident [6]. The bark of bed logs was mainly contaminated by the radioactive fallout immediately after the nuclear power plant accident, but the numbers of bed logs and trees that are directly contaminated has decreased over time. On the other hand, 137 Cs in the soil is being accumulating in trees via root absorption [8], resulting in the location of 137 Cs contamination in bed logs shifting from the bark to the wood. Consequently, the 137 Cs TF at the time of the accident is likely to be different from the current 137 Cs TF since nutrients are mainly supplied to the developing fruiting body of shiitake from the wood, including the inner bark, during bed-log cultivation [9].
Logs have an inhomogeneous physical structure and chemical composition, so the TF can vary widely in a single bed log. By contrast, sawdust medium is a homogeneous culture medium, allowing the effects of different factors on the TF to be easily assessed. However, 137 Cs TFs greatly differ between bed-log cultivation and sawdust medium cultivation. Therefore, to use sawdust medium as a model for bed logs, it is necessary to first establish a sawdust medium cultivation condition that can obtain a similar 137 Cs TF to bed-log cultivation.
TFs have been evaluated by measuring 137 Cs concentrations. However, the TF can vary greatly even among bed logs obtained from the same forest [10]. Moreover, a concentration of 50 Bq/kg of 137 Cs is equivalent to 1.56 × 10 -11 g/kg based on the half-life, and the 137 Cs TF will vary with only a small difference in 137 Cs concentration. By contrast, 133 Cs has almost the same chemical properties as 137 Cs, but has already reached a steady state in nature and is present at a concentration of approximately 30 μg/kg in bed logs [11]. Consequently, the 133 Cs TF could be used to accurately estimate the 137 Cs TF. However, while 137 Cs adheres to the bark due to fallout and is also absorbed inside the tree through root absorption, 133 Cs is mainly absorbed through the roots. Cs is fixed to organic matter [12], so the available concentrations of these nuclides will differ due to their different chemical speciation, which may result in different TFs. There was a significant positive correlation between the 137 Cs concentration in konara oak (Quercus serrata) shoot and the exchangeable 137 Cs (ex 137 Cs) concentration in the soil, which is extracted with a neutral 1 mol/L ammonium acetate solution [10]. Therefore, it is considered necessary to study the difference in behavior of exCs for both nuclides.
The ultimate goal of this research was to re-evaluate the 137 Cs TF to shiitake cultivated in bed logs. To achieve this, a sawdust medium cultivation condition that gave a similar TF to that of bed-log cultivation was first established using 133 Cs. The factors that affect the TF were then clarified by cultivating shiitake in the model sawdust medium. Finally, a method for predicting the fruiting body 137 Cs concentration in bed-log cultivation was discussed.

Strains
Two commercial shiitake strains were purchased, which are referred to as Strain 1 and Strain 2, respectively, hereafter in consideration of reputational damage. Strain 1 and Strain 2 are mainly used for sawdust medium cultivation and bed-log cultivation, respectively.

Bed-log cultivation for estimating the sampling position of bed log for accurate TF measurement
Konara oak bed logs that were approximately 11 cm in diameter and 90 cm long and had been inoculated with Strain 2 approximately 2 years previously were purchased from a grower. A hole with a diameter of 1 cm was made in the wood (depth = 3.2 cm) in the longitudinal center of each bed log, and antibiotic assay filter paper (49005010; ADVANTEC Co., Ltd., Tokyo, Japan) impregnated with a CsCl solution containing 1.0 mg 133 Cs was placed in the bottom of the hole. The hole was then sealed with Japanese beech (Fagus crenata) sawdust. Three bed logs were prepared for the experiment. The bed logs were soaked in tap water for approximately 16 h, and maintained at 15 °C and a relative humidity of ≥ 90% under fluorescent lamp irradiation for fruiting body development. Each fruiting body was harvested when the cap was approximately 80% open, and the point on the bed log where the fruiting body developed was recorded. Because the stipes of shiitake cultivated in bed-log cultivation are cut off before the ingredients are analyzed [13], the stipe was discarded and only the cap was sliced, dried at 105 °C, and milled with a Millser (IFM-620DG; Iwatani Co., Tokyo, Japan). The concentration of 133 Cs in each cap was determined by subtracting 0.59 mg/kg from the measured value, which equated to the cap 133 Cs concentration in an untreated bed log. The cap 133 Cs concentrations at the sampling locations are shown as a contour graph using SigmaPlot version 14.0 (Systat Software, Inc., California, USA), with the circumferential direction of the bed log aligned along the x-axis, the axial direction of the bed log aligned along the y-axis, and the perforated portion is represented by (0, 0).

Bed-log cultivation for estimating the 137 Cs TF and the 133 Cs TF
The same kind of konara oak bed logs were purchased from the same grower as described above, and the fruiting bodies were developed and harvested, the point on the bed log where the fruiting body developed was recorded, and the caps were treated in the same manner as described above. Once fruiting body development was completed, each bed log was cut radially 3 cm axially from the recorded point, and the resulting disc was then cut axially along the line connecting the center of the bed log and the place 5 cm circumferentially away from the recorded point (Fig. 1). The bark and wood were separated from the resulting wedge-shaped blocks and dried at 105 °C. The wood was then crushed with a cutting mill (P-15; Fritsch Japan Co., Ltd., Kanagawa, Japan) and both the bark and wood were milled separately with the Millser.

Sawdust medium cultivation for estimating the effect of medium 133 Cs concentrations on the cap 133 Cs concentrations
Commercially available Japanese beech sawdust and rice bran were used as the sawdust medium. Sawdust and rice bran were mixed at a ratio of 75:25 (w/w), and the sawdust medium 133 Cs concentration was adjusted to 5 levels [control, 0.50, 0.10, 0.15, and 0.20 mg/kg (w/w)] using CsCl. After adjusting the water content to 65% (w/w), a 1 kg mixture was packed into a plastic bag and autoclaved at 121 °C for 90 min. Three sawdust media were prepared for each CsCl concentration. The sterilized sawdust media were inoculated with the mycelia of the both strains that had been cultivated for 2 weeks in 40 mL SMY liquid medium [10 g/L sucrose, 10 g/L malt extracts (Kyokuto Pharmaceutical Industrial Co., Ltd., Tokyo, Japan), and 4 g/L yeast extracts (Oxoid Ltd., Hampshire, England)] and cultured at 22 °C and 70% relative humidity in the dark for 16 weeks [14]. The fruiting bodies that developed were then treated in the same manner as described above for bed-log cultivation. Once fruiting body development was completed, the sawdust medium was dried at 105 °C, crushed with a pulverizer (V-360; HORAI Co., Ltd., Osaka, Japan), and milled with the Millser.

Sawdust medium cultivation for estimating the effect of the nutrient concentration on TF
Japanese beech sawdust and rice bran were mixed at 95:5, 90:10, 85:15, 80:20, 75:25, and 70:30 (w/w), and the 133 Cs concentrations in these 6 media were adjusted to 0.20 mg/kg (w/w) using CsCl. The cultivation was performed in the same manner as described above for sawdust medium cultivation.

Sawdust medium cultivation for estimating the effect of the cultivation period on TF
Japanese beech sawdust and rice bran were mixed at a ratio of 95:5 (w/w), and the 133 Cs concentrations in the resulting sawdust media were adjusted to 0.20 mg/kg (w/w) using CsCl. The cultivation was performed in the same manner as described above for sawdust medium cultivation, but cultivation periods were set for 5 levels (16,20,24,28, and 32 weeks).

Sawdust medium cultivation for estimating the differences between the 137 Cs TF and 133 Cs TF, and the differences between strains
Radioactive contaminated sawdust and rice bran were mixed at a ratio of 95:5 (w/w). The radioactive sawdust was collected from a contaminated site in Fukushima prefecture in July 2011 and consisted of mainly hardwood (unknown species). The cultivation was performed in the same manner as described above for sawdust medium cultivation, but the cultivation period was 28 weeks.

Elemental analysis
The total 137 Cs (to 137 Cs) concentrations in the bark and wood samples were measured by filling 20-mL vials (6000477; PerkinElmer Japan Co., Ltd., Kanagawa, Japan) with each sample and measuring the to 137 Cs concentration twice per sample for 1 h each using a gamma counter (2480 WIZARD 2 ; PerkinElmer Japan Co., Ltd.). The to 137 Cs concentration in the sawdust medium was measured by placing each sample in a 0.7-L Marinelli beaker or U-8 container, depending on the sample volume, and using a high-purity germanium detector (GEM20-70; SEIKO EG&G Co., Ltd., Tokyo, Japan). The cap 137 Cs concentrations in the bedlog cultivation were measured by mixing each sample cap with a cap of known 137 Cs concentration (6.3 Bq/ kg) to give a total amount of approximately 9 g, because each sample volume was too small to measure with the gamma counter. The 137 Cs concentration in the resulting mixture was then measured in the same manner as described above using the gamma counter and the cap 137 Cs concentration was calculated from its weight ratio in the mixed sample. The cap 137 Cs concentrations in the sawdust medium cultivation were measured by enclosing each sample in a U-8 container and using the high-purity germanium detector, as described above.
To determine the ex 137 Cs concentration in the sawdust medium, 10 mL of 1 mol/L ammonium acetate solution adjusted to pH 7.0 with acetic acid or ammonia solution was added to approximately 1 g of each sawdust sample. The mixture was then swirled for 16 h with a rotating incubator (RT-50; TAITEC Corp., Saitama, Japan) and centrifuged at 5000 × g for 10 min. The resulting supernatant was filtered through a 0.50-μm hydrophilic polytetrafluoroethylene (PTFE) membrane filter, and the ex 137 Cs concentration was measured three times per sample for 2 h each using the gamma counter, as described above.
To measure the total 133 Cs (to 133 Cs) concentration in the all samples, 14 mL of concentrated nitric acid (Ultrapure-100; KANTO CHEMICAL Co., Inc., Tokyo, Japan) was added to approximately 0.5 g of sample in the lower container of an ECO-PRE (OD-98-100; ACTAC Co., Kanagawa, Japan), and 6 mL of 5% nitric acid was added to the trap part. Digestion was then carried out by repeating nine cycles of 210 °C for 20 min and 100 °C for 25 min. The resulting solution was filtered through the same PTFE membrane filter, and the to 133 Cs concentration was measured using an inductively coupled plasma mass spectrometer (ICP-MS) (7700x; Agilent Technologies International Japan Ltd., Tokyo, Japan). The cap cultivated in the bed-log cultivation was mixed with supplemental sample for measuring the cap 137 Cs concentration, and the supplemental cap 133 Cs concentration was 0.32 mg/kg, so the cap 133 Cs concentration in the bed-log cultivation was calculated as described above.
To measure the exchangeable 133 Cs (ex 133 Cs) concentration in the sawdust medium, the ex 133 Cs was extracted as described above for ex 137 Cs, and measured as described above using ICP-MS.
The nitrogen (N) concentration in the sawdust medium was measured with an NC analyzer (Sumigraph NC-22F; Sumica Chemical Analysis Service Ltd., Osaka, Japan) using approximately 20 mg sample.

Calculation
The sawdust medium Cs concentrations were changed depending on the cultivation stages, so the TFs based on the medium Cs concentration at the start of cultivation (TF SC ), before fruiting body development (TF BD ), and after fruiting body development (TF AD ) were calculated as follows: (1) Fruiting body Cs concentration Medium CS concentration at the start of cultivation , Fruiting body Cs concentration Medium Cs concentration after fruiting body development .
Since there were two kinds of Cs concentration (toCs concentration and exCs concentration) and two kinds of Cs ( 137 Cs and 133 Cs), they were reported as to 137 Cs TF, to 133 Cs TF, ex 137 Cs TF, and ex 133 Cs TF, respectively. The amount of fixed Cs (fxCs) was calculated as follows: The exCs solubilization ratio was calculated as follows: where fxCs SC and fxCs AD were the fxCs amounts at the start of cultivation and after harvesting fruiting body, (4) Fixed Cs amount = total Cs amount − exchangeable Cs amount.
respectively. The wet weight-based TF was calculated as follows: because the water content of the culture medium is specified as 12% [15]. The significance level was set to 0.05.

Relationship between the fruiting body Cs concentration and the distance from the source of cesium
A sufficient number of fruiting bodies required to draw contour graphs was only obtained from one of the three bed logs examined. The cap 133 Cs concentrations at locations where fruiting bodies developed (n = 16) are shown in Fig. 2. A mountain-shaped contour graph with vertices near (0, 0) was observed. It can clearly be seen that the cap 133 Cs concentration decreased with increasing distance from the point where 133 Cs was applied. In addition, the fruiting body absorbed 133 Cs from a distance of about 20 cm in an axial direction and about 10 cm in a circumferential direction.

Both nuclides to Cs TF AD in bed-log cultivation
Relationships between the cap Cs concentration and bed log Cs concentration and toCs TF AD in the bed-log cultivation are shown in Table 1. There was no significant correlation between the cap 137 Cs concentration and the bark to 137 Cs concentration [Spearman's rank order correlation (ρ) = 0.14, P = 0.37, n = 43]. However, there was a significant positive correlation between the cap 137 Cs concentration and the wood to 137 Cs concentration (ρ = 0.41, P < 0.01, n = 43). In contrast, there was no significant correlation between the cap 133 Cs concentration and the wood to 133 Cs concentration (ρ = −0.09, P = 0. 56, n = 43). Furthermore, when comparing the toCs TF AD of the wood, the to 133 Cs TF AD was significantly higher than the to 137 Cs TF AD (t-test: P < 0.01; Table 1).
Units of the 137 Cs and 133 Cs concentration were Bq/ kg and mg/kg, respectively. The toCs TF AD was referred

Effect of the nutrient concentration in the sawdust medium on the to 133 Cs TF AD
There were significant negative correlations between the to 133 Cs TF AD and the sawdust medium N concentrations for both strains (Strain 1 and 2: ρ = − 1.00, P < 0.01; Fig. 4). When the N concentration of the sawdust medium was 2.6 g/kg (rice bran addition ratio = 5%), the 133 Cs TF AD for Strain 1 and 2 with were 18.8 ± 0.7 and 20.4 ± 2.3, respectively. The water contents of the fruiting bodies for Strain 1 and 2 with the same sawdust media adjusted an N concentration of 2.6 g/kg were 93.9 ± 0.2 and 90.5 ± 2.5%, respectively. Therefore, the to 133 Cs TF AD for Strain 1 and 2 on a wet weight basis were 1.9 ± 0.5 and 2.2 ± 0.4, respectively (Eq. 6).

Effect of the cultivation period in the model sawdust medium on to 133 Cs TF AD
No fruiting bodies were obtained from Strain 2 with cultivation periods of 16 to 20 weeks, so only data for cultivation periods of 24-32 weeks were included in the    analysis. The relationships between to 133 Cs TF AD , cultivation period, cap yield, and cap number for Strain 1 and 2 are shown in Table 2. There was a significant negative correlation between the to 133 Cs TF AD and the cultivation period and between the to 133 Cs TF AD and the cap yield for Strain 1. However, there was no significant difference among the to 133 Cs TF AD for Strain 1 (one-way analysis of variance, P = 0.15). Moreover, there was no significant correlation between the cultivation period and the cap yield for Strain 1 (P = 0.08). Furthermore, there was no significant correlation among the to 133 Cs TF AD , the cultivation period, the cap yield, and the cap number for Strain 2 ( Table 2). to 133 Cs TF AD was referred to Eq. 3. The numbers indicated the Spearman's rank order correlations (n = 3 sawdust medium bags per strain). * indicated that the P value was below the significance level (0.05).

Differences in TFs between 137 Cs and 133 Cs in the model sawdust medium
There were significant differences between the to 137 Cs TF SC and to 133 Cs TF SC for both strains (t-test: P < 0.01 for all contrasts; Table 3, a and b) and between ex 137 Cs TF SC and ex 133 Cs TF SC for both strains (t-test: P < 0.01 for all contrasts; Table 3, c and d). There was also a significant difference between the to 137 Cs TF BD and to 133 Cs TF BD for both strains (t-test: P < 0.01 for all contrasts; Table 3, e and f ) and between the to 137 Cs TF AD and to 133 Cs TF AD for both strains (t-test: P < 0.01 for all contrasts; Table 3, i and j). However, there was no significant difference between the ex 137 Cs TF BD and the ex 133 Cs TF BD for either strain (t-test: Strain 1: P = 0.57, Table 3, g; Strain 2: P = 0.40, Table 3, h) and between the ex 137 Cs TF AD and the ex 133 Cs TF AD for either strain (t-test: Strain 1: P = 0.62, Table 3, k; Strain 2: P = 0.61; Table 3, l). Table 3 shows that the toCs and exCs TF SC of both nuclides are higher than the toCs and exCs TF BD of both nuclides and the toCs and exCs TF AD of both nuclides, indicating that Cs concentrations of both nuclides and both types in the model sawdust media increased during cultivation. The medium weights of Strains 1 and 2 decreased by 41.8 ± 1.5 and 57.2 ± 0.4%, respectively, from the start of cultivation until harvesting the fruiting bodies. Furthermore, the amount of fx 137 Cs in the medium decreased by 18.5% for Strain 1 and 18.0% for Strain 2 from the start of cultivation until all the fruiting bodies were harvested, and the amount of fx 133 Cs also decreased, albeit by a lesser amount (3.3% and 8.2%, respectively) ( Table 4).

Differences in TF BD between the strains
The ex 137 Cs TF BD for Strain 1 was significantly lower than that for Strain 2 (t-test, P = 0.03, Table 3, m). In contrast, there was no significant difference in the ex 133 Cs TF BD between the strains (t-test, P = 0.07, Table 3, n). Therefore, clear strain differences were not found between the strains. Furthermore, there was no significant difference in the ex 137 Cs solubilization ratio between the strains (t-test, P = 0.76, Table 4, a), nor was there a significant difference in the ex 133 Cs solubilization ratio (t-test, P = 0.22, Table 4, b).

Model sawdust medium
When the sawdust medium N concentration was 2.6 g/ kg (rice bran addition ratio = 5%), the to 133 Cs TF AD for Strain 1 and 2 on wet weight bases were close to the TFs that led to the provisional reference index value of 2.0 in bed-log cultivation (Fig. 4) [5]. In addition, the to 133 Cs TF AD for Strains 2 in sawdust medium cultivation at an N concentration of 2.6 g/kg was almost the same as the to 133 Cs TF AD for Strains 2 in bed-log cultivation (Fig. 4, Table 1). These results indicated that the sawdust medium with an N concentration adjusted to 2.6 g/ kg could be used as a model for bed logs with respect to

Table 4 Mass balance (existence rate, %) of 137 Cs and 133 Cs in the model sawdust medium
The fixed Cs (fxCs) amount and exchangeable Cs (exCs) solubilization ratio were referred to Eqs. 4 and 5. The numbers in parentheses indicated the exCs solubilization ratio and its standard deviation. *Total value did not match due to the rounding of values. The same letters indicated the comparison groups and there was no significant difference between them according to t-test TF. A 133 Cs concentration in the sawdust medium up to about 0.20 mg/kg had no effect on cap yield. Therefore, the 133 Cs concentration of the model sawdust medium was set to 0.20 mg/kg to clarify the behavior of TF. There was no significant relationship between the cultivation period and to 133 Cs TF AD during the tested period in the sawdust medium cultivation (Table 2). However, the yield of Strain 1 tended to increase with an increasing cultivation period ( Table 2). Based on these findings, the cultivation period was set to 28 weeks for harvesting a sufficient yield of fruiting bodies. By the way, since Strain 1 and 2 are mainly used for sawdust medium cultivation and bedlog cultivation, respectively, it is likely that Strain 2 took longer to develop fruiting bodies than Strain 1 in the sawdust medium cultivation.

Factors affecting the TF
There were significant positive linear correlations between the cap 133 Cs concentrations and the sawdust medium 133 Cs concentrations (Fig. 3). A similar relationship has been reported for oyster mushroom (Pleurotus ostreatus) [14], then it is considered that the cap Cs concentration is proportional to the Cs concentration in the medium if the Cs concentration in the medium is uniform. On the other hand, the to 137 Cs concentration at the point where fruiting bodies developed were varied (Table 1), which indicated that the distribution of 137 Cs concentration in a single bed log also varied. Therefore, the cap 133 Cs concentration in bed-log cultivation is considered to be affected by the concentration of the 133 Cs source and the distance from the source within about 20 cm in an axial direction and about 10 cm in a circumferential direction (Fig. 2). A change in enzyme activity has also been observed in bed log about 17 cm in an axial direction from the point where the fruiting body developed [9], indicating that not only nutrients, but also Cs is absorbed at a considerable distance from the point where the fruiting body developed. The bed log Cs concentration at the point where the fruiting body developed would have the greatest influence on the cap Cs concentration, but the conventional method of calculating TF uses the 137 Cs concentration of the entire bed log [15]. Therefore, the toCs TF AD was measured using the bed log Cs concentration at the point where fruiting body developed. There was a significant positive correlation between the cap 137 Cs concentration and the wood to 137 Cs concentration, but no significant correlation between the cap 137 Cs concentration and the bark to 137 Cs concentration, although the bark to 137 Cs concentration showed as much variation as the wood to 137 Cs concentration, as indicated by their similar coefficients of variation (CVs) ( Table 1). On the other hand, one reason for the lack of significant correlation between the cap 133 Cs concentration and the wood 133 Cs concentration may be because the bed log 133 Cs concentrations were not sufficiently varied to make a significant correlation since the CV is 0.22 (Table 1). Therefore, bark to 137 Cs is considered to have less effect on to 137 Cs TF AD than wood to 137 Cs. Furthermore, a clear linear relationship has not been obtained even between the cap 137 Cs concentration and the wood 137 Cs concentration since the ρ is 0.41 (Table 1). These results indicated that there are other factors affecting the toCs TF AD besides Cs concentration. Many factors will affect the toCs TF AD in bed-log cultivation, making it difficult to elucidate the most important ones. Therefore, these factors were examined using the model sawdust medium.
By varying the nutrient contents of the sawdust media, the difference in to 133 Cs TF AD between bed-log cultivation and sawdust medium cultivation is thought to be caused mainly by the difference in nutrient concentration of the substrate (Fig. 3). However, rice bran contains various nutrients [16], so it is unclear which component affected the TF AD . N is the most consumed nutrient component in sawdust medium [17], so the nutrient concentration was represented as N concentration. Since kunugi oak (Q. acutissima) is estimated to have a higher N concentration than konara oak [18], the use of kunugi oak may reduce the to 137 Cs TF AD .
There was a significant negative correlation between the to 133 Cs TF AD and the cap yield for Strain 1 but not for Strain 2 ( Table 2). As the yield of the cap changes, the amount of Cs transferred to the cap changes, and as a result, the cap Cs concentration is likely to change. How the cap Cs concentration changes is thought to depend on the nature of the strain.
The strain difference is considered to be caused by the difference in the amount of exCs transferred from the medium to the fruiting body and the difference in the amount of dissolved Cs which is fixed in the medium during cultivation. I speculate that the former corresponds to exCs TF BD and the latter corresponds to exCs solubilization ratio. However, there was no clear difference in the ex 137 Cs TF BD and ex 133 Cs TF BD between the strains (Table 3). Moreover, there was no significant difference in the ex 137 Cs and the ex 133 Cs solubilization ratios between the strains (Table 4). From these results, no strain difference was evident between the strains. In this experiment, two kinds of strains for sawdust medium cultivation and bed-log cultivation were used in order to obtain strain differences, but it is presumed that the differences between the strains were not large enough to obtain clear significant differences. However, the details of this are unknown, so further study is required.
On the other hand, the lack of significant differences in the ex 133 Cs solubilization ratio between the strains may be due to the large standard deviation ( Table 4). The 133 Cs present in the tree is absorbed from the soil by root absorption, and the absorbed Cs may be bound to intracellular lignin and other substances, some of which may be surrounded by various substances such as cellulose and hemicellulose, and therefore many types of enzymes may be required to solubilize 133 Cs. The diversities of these enzyme activities among strains have been shown [19,20], and because Cs is not an essential element for shiitake, it is hypothesized that 133 Cs was not solubilized at a constant rate. The 137 Cs concentration in trees is increasing [21], and the variations in ex 137 Cs solubilization rate may further vary cap 137 Cs concentration.

Cs concentration leading to accurate TF
There were significant differences between the to 137 Cs TF AD and to 133 Cs TF AD in the bed-log cultivation (Table 1) and the model sawdust medium cultivation (Table 3), but no difference between ex 137 Cs TF AD and ex 133 Cs TF AD in the model sawdust medium cultivation ( Table 3). Mechanism of Cs migration from the bed log to the fruiting body has been proposed to be intracellular transport through mycelium and the extracellular transport via the inter-hyphal cavity [22]. In either case, ionic Cs is migrated from the bed log to the fruiting body, suggesting that exCs acts as a Cs supply source for the fruiting body. Soil organic matter contains readily soluble and sparingly soluble Cs [12,23], and the sawdust medium also contained exCs and fxCs, which corresponds to soluble and sparingly soluble Cs, respectively (Table 4). However, the TF is usually calculated using toCs TF [15], so it is considered appropriate to use exCs concentration rather than toCs concentration for calculating the accurate TF.
There are two methods for calculating TF in bed-log cultivation, one is TF SC and the other is TF AD [15]. However, there was difference between TF SC and TF AD in the model sawdust medium cultivation ( Table 3). The sparingly soluble Cs is presumed to be bound to lignin or tannin [12], which can be degraded by wood-rotting shiitake [24]. The fxCs concentrations of both nuclides decreased in the sawdust medium cultivation (Table 4), which indicated that exCs concentrations for both were increasing from the start of the cultivation to the fruiting body development, along with the decrease in the model sawdust medium weight. Therefore, TF AD is considered more appropriate than TF SC . The to 137 Cs TF AD of wood was over 10 (Table 1), and the cap 137 Cs existence rates in Strain 1 and 2 were 9.0 and 11.2%, respectively, in the model sawdust medium cultivation (Table 4). If the TF AD is sufficiently small, the Cs concentration in the sawdust medium can be regarded as constant, but the TF AD of shiitake was large, and the amount of Cs in the sawdust medium after fruiting body development was reduced as compared with that before fruiting body development (Table 4). In order to obtain an accurate TF, it is considered necessary to consider the amount of Cs taken up by the fruiting body from the medium. Therefore, TF BD is considered more appropriate than TF AD .
When calculating the mass balance of sawdust medium, unknown 137 Cs and 133 Cs accounted for a few percent of the to 137 Cs and to 133 Cs, respectively, after harvesting fruiting bodies (Table 4). Only caps were analyzed in this experiment, therefore it is possible that the unknown Cs amount was equivalent to the amount of Cs in the stipe. However, 137 Cs is mainly accumulated around the edge of the cap in the fruiting body [25], and it is unlikely that the stipe would contain 26%-93% Cs of the cap Cs. However, the details of this are unknown, so further study is required.

Calculation method of TF in bed-log cultivation
What is desired on the mushroom farm is to predict the fruiting body 137 Cs concentration from the to 137 Cs concentration of the log at the start of cultivation. For the purpose, it is necessary to find the relationship between to 137 Cs TF SC and ex 137 Cs TF BD . However, it is difficult to obtain accurate exCs concentrations because the Cs concentrations in the current bed logs vary even within a single bed log. Furthermore, ex 137 Cs TF BD in bed-log cultivation is considered to be affected by the nutrient concentration as well as the sawdust medium. Nutrients such as N, potassium (K), and phosphorus have been reported to be likely to be deficient during bed-log cultivation [18], and 137 CsTF in wild mushrooms has been reported to decrease with increasing concentrations of exchangeable K in the soil [26]. It is necessary to clarify the effects of the concentrations of these nutrients on ex 137 Cs TF BD for clarifying the relationship between the to 137 Cs TF SC and ex 137 Cs TF BD . However, even if the relationship between the concentrations of these nutrients and ex 137 Cs TF BD becomes clear, the concentrations of these nutrients in the logs vary with the time of tree felling [27,28]. The upper limit of to 137 Cs concentration in logs capable of producing less than 100 Bq/kg of fruiting bodies may be different from the current value when taking into account the variation in the concentration of these nutrients in the log. In addition, it seems practically impossible to measure these nutrient concentrations for each lot of logs and determine the suitability of each lot of logs.
As more time passes since the nuclear power plant accident, fewer directly contaminated trees will be used as bed logs, with newly planted and sprouted coppice trees increasingly being used in their place. However, the