Short day length-induced decrease of cesium uptake without altering potassium uptake manner in poplar

Short day length-induced alteration of potassium (K) localization in perennial trees is believed to be a mechanism for surviving and adapting to severe winters. To investigate the relationship between cesium (Cs) and K localizations, a model tree poplar, hybrid aspen T89, was employed. Under short day length conditions, the amount of 137Cs absorbed through the root and translocated to the root was drastically reduced, but 42K was not. Potassium uptake from the rhizosphere is mediated mainly by KUP/HAK/KT and CNGC transporters. In poplar, however, these genes were constantly expressed under short-day conditions except for a slight increase in the expression a KUP/HAK/KT gene six weeks after the onset of the short-day treatment. These results indicated that the suppression of 137Cs uptake was triggered by short day length but not regulated by competitive Cs+ and K+ transport. We hypothesize that there are separately regulated Cs+ and K+ transport systems in poplar.

In 2011, the Fukushima Daiichi Nuclear Power Plant accident released a large amount of radionuclides into the environment. Due to its long half-life, radioactive cesium ( 137 Cs) was considered the main contaminant. To estimate the transfer process of 137 Cs in the terrestrial environment, we focused on its behavior in woody plants because the transfer process within the forest ecosystem is much slower than it is in other areas 1 . After the forest contamination by 137 Cs, depositions to tree canopies, leaf-and/or bark uptake, acropetal branch translocation, etc. were energetically investigated [2][3][4] . Little is known about the physiology of Cs transfer and distribution within trees. Cesium chemically resembles potassium (K) but it is not an essential nutrient for plant growth. Avery reported that Cs + inhibits the inward-K + channels in the plasma membrane and is therefore considered toxic to plants 5 . In general, rhizosphere Cs + consists mostly of stable 133 Cs and its concentration is < 200 μ M, which is not toxic to plants 6 . Cesium uptake and translocation within the plant body are thought to be mediated by K + transport systems 6 . Arabidopsis HAK5 (AtHAK5) is a K + uptake transporter in roots and is up-regulated under K + deficiency 7,8 . The AtHAK5 T-DNA insertion mutant, athak5, showed a significantly decreased K content and a tolerance to 300 μ M Cs + treatment under low K conditions 9,10 . AtCNGC2 demonstrated cation transport activity using transgenic transfected human embryonic kidney cells. AtCNGC1 is the candidate gene for Cs + uptake and was identified by quantitative trait locus analysis in Arabidopsis 11,12 . In this study, we investigated 137 Cs and 42 K localizations using a model tree, poplar, under both long-and short photoperiods. To estimate the 137 Cs retention time within the forest ecosystem, the Cs content of each tree organ must be determined under seasonal conditions since Cs is circulated via root uptake, translocation, and leaf abscission. Poplar is a perennial deciduous tree with a characteristic seasonal cycle of growth and dormancy. The phase shift from growth to dormancy is a winter adaptation 13 . The transition of meristems into dormant buds is crucial to protect them against hazardous frosts. Woody plants shift their growth stage when they perceive changes in photoperiod and temperature 14 . The initiation of cold acclimation under short day length increases endogenous abscisic acid levels and reduces endogenous gibberellic acid levels [15][16][17] . In beech tree (Fagus sylvatica L.) leaf senescence, leaf K content decreases before shedding and the recovered K is deposited in the stem cortex and pith 18 . Japanese native poplar (Populus maximowiczii) also showed a decrease in leaf K concentration following dormant bud formation 19 . An increase in K + concentration in xylem sap was observed during the winter season in field-grown Populus nigra 20 . These behaviors imply the existence of re-translocation mechanisms for K, and it is assumed that the potassium is transported to the organs that require it once it is resorbed. Potassium is transported through various systems within the plant body 21 . Epstein et al. showed that K + absorption in barley roots is mediated by both a high-affinity-and a low-affinity biphasic transport process 22 . The high-affinity transport system (HATS) is up-regulated by a decrease in external K + concentration. The low-affinity transport system (LATS), however, operates even when there is sufficient external K + 23 . Potassium ion uptake by the root symplast via HATS is mediated by the KUP/HAK/KT transporter family. There are thirteen such transporters in Arabidopsis 24,25 and twenty-seven in rice 26 . Non-selective cation transport mechanisms such as voltage-independent cation channels (VICC) are categorized as LATS. In Arabidopsis, AtCNGC encodes VICC type channels. Twenty types of AtCNGC are present in the Arabidopsis genome.
Based on the above, it is assumed that the expression and function of K + permeable transporters also regulate Cs + translocation in various plant species and situations. Therefore, we investigated the relationship between the change in K localization induced by short day length and the behavior of Cs absorbed from the rhizosphere. To this end, 137 Cs and 42 K accumulations and gene expression patterns of major K + transporters were analyzed using a model tree poplar, hybrid aspen T89.

Results
Amount of 137 Cs in shoots was down-regulated under short-day conditions. Under a controlled growth cycle system in Populus alba L., the shift from long-day (LD) to short-day (SD) conditions decreased phosphate in the lower leaves 27 . This change suggests the existence of mechanisms for the re-translocation of phosphate from older-to younger (upper) leaves in response to with seasonal changes. Furukawa et al. indicated Ca 2+ transport from root to shoot in Populus maximowiczii is also regulated by the shift from LD to SD 19 . Based on these facts, the uptake of Cs + within the root and its behavior within the plant body in LD and SD conditions were compared. To measure seasonal variations in Cs + uptake, a 137 Cs + solution was added to the growth media under LD3, LD9, and SD6 conditions (see Methods). Figure 1A shows the localization of 137 Cs by root absorption under LD3, LD9, and SD6 conditions. In the LD3 plants, 137 Cs was localized entirely and the radiation intensity around the apex was highest there. The LD9 plants were the same age as the SD6 plants and showed the same 137 Cs behavior as the LD3 plants. In the SD6 plants, 137 Cs was localized mainly in the stem and root and the total 137 Cs was lower than that for the other plants. In LD3, LD9, and especially SD6 plants, all the nodes showed high amounts of 137 Cs. The quantity of 137 Cs in the shoots of SD6 plants was about 36.3% and 23.6% lower than that in the LD3 and LD9 shoots, respectively (Fig. 1B). On the other hand, the amount of 137 Cs in the roots was similar for all three conditions. Cesium-137 accumulated mainly (48.8%) in the leaves under LD3 conditions (Fig. 1C). In LD9 conditions, 137 Cs also accumulated to a large extent in the leaves (42.5%). Nevertheless, under SD6 conditions, the leaf 137 Cs content was 32.1%, and organs containing the most 137 Cs were the stems (39.7%). For the shoot apices, 137 Cs levels were lower in SD6 plants than they were in LD3 and LD9, but the difference was not significant. The concentrations of 137 Cs were highest in the apices of the LD3 and LD9 plants (Fig. 1D). Nevertheless, the decreases in the 137 Cs concentrations in the apices and the leaves under SD6 conditions were significant, and the transition to SD suppressed Cs transport into the apices and the leaves.
Potassium-42 uptake was constant under LD and SD conditions. Based on the 137 Cs uptake activity assays, it was expected that the amount of 42 K absorbed through the root would also be down-regulated by the transition to SD. Poplar roots were treated with exogenous 42 K and the amounts of 42 K in the shoots and roots under LD3, SD2, SD4, and SD6 conditions were measured after 24 h incubation (Fig. 2). No difference was found in the amount of 42 K in the roots among four conditions. The amount of 42 K in the shoots at the early stage of SD was almost equivalent to that in the shoots at LD3. In contrast, the amount of 42 K in the SD6 plant was slightly higher than it was in the other conditions, but the difference was not significant. These data suggest that the demand for K in the rhizosphere neither increased nor decreased by the transition to SD in the poplar for up to six weeks.
A comparison of Figs 1B and 2 indicates that 137 Cs accumulation significantly decreased under SD6 condition, but 42 K accumulation remained almost constant through the SD transition. This fact indicates that the decrease in Cs levels in the shoot is not explained by a simple reduction in transpiration rate under SD conditions and that important K + uptake systems in poplar might be independently regulating Cs accumulations in it. It is also implied that the transporter responsible for Cs + uptake in poplar might have only limited involvement in K + uptake since no decrease in K accumulation was observed when Cs accumulation was low.
PttHAK-like1 was up-regulated by transition to short-day. We investigated under SD conditions the expression patterns of some candidate genes related to K + transport. We focused on the KUP/HAK/KT family K + transporters and a cyclic-nucleotide-gated channel (CNGC) type K + channels. As for KUP/HAK/KT family genes, we concentrated on three genes in poplar. One of these was Populus tremula K + uptake transporter 1 or PtKUP1 (Accession number, AJ299422; POPTR_0003s13370) which was identified in hybrid aspen 28 . PtKUP1 was used in a complementation test with a K + -uptake-deficient E. coli mutant. The addition of Cs + to the culture media strongly inhibited the growth of E. coli expressing PtKUP1 28 . We also looked at two KUP/HAK/KT family transporters resembling AtHAK5. Populus trichocarpa, whose genome was elucidated in 2006 29 , has nine AtHAK5 homolog genes in its genome, including PtKUP1. To evaluate the similarity between these putative K + transporters, we constructed a phylogenetic tree of these genes and AtHAK5 homologs reported to be involved in K + and Cs + transport. We used their amino acid sequences (Fig. S1A). An AtHAK5 homolog in barley, HvHAK1, also up-regulated its expression under K deficiency, and transgenic yeast expressing the HvHAK1 gene showed an enhanced growth rate 30 . Like HvHAK1, rice OsHAK5 also exhibits a high homology to AtHAK5 31 . The poplar  gene POPTR_0010s10450 had the highest amino acid sequence homology with AtHAK5. POPTR_0001s00580 had the second highest. We identified POPTR_0010s10450 and POPTR_0001s00580 orthologues in the hybrid aspen T89 and named them PttHAK-like1 and PttHAK-like2, respectively.
CNGC (cyclic-nucleotide-gated channel) may be a non-selective K + channel which mediates K + uptake by the root symplast 21 . In Arabidopsis, the CNGC channel AtCNGC2 shows K + permeability 32 . A quantitative trait locus analysis indicated that AtCNGC1 is associated with shoot K and Cs concentrations in Arabidopsis 12,33 . In P. trichocarpa, nine genes were selected as AtCNGC1 homologs based on their amino acid sequences (see Supplementary Fig. S1B). Proteome BLAST analysis showed that POPTR_0012s01690 and POPTR_0015s02090 scored significantly higher than did others and the orthologues in hybrid aspen T89 were named as PttCNGC1-like1 and PttCNGC1-like2, respectively.
To determine which K + uptake related gene is the most abundantly expressed among these five transporters, we evaluated the expression level of each gene under LD3 conditions. In poplar roots, no obvious differences were found in the expression levels of the genes selected (Fig. 3A). There may be redundancy in the expression of these K + influx transporters under LD3 conditions. During the transition to the SD conditions, the expression of PtKUP1 did not significantly change (Fig. 3B). PttHAK-like1 showed steady expression until the transition to SD4 conditions and was up-regulated by about 1.5-fold under SD6 conditions (Fig. 3C). PttHAK-like2 expression tended to decrease in SD2 and SD4 plants but maintained statistically steady-state transcription levels through SD transition (Fig. 3D). The expressions of PttCNGC1-like1, PttCNGC1-like2 were also relatively constant under SD conditions (Fig. 3E,F).
There was a small but statistically significant increase in the expression level of PttHAK-like1 under SD6 conditions but the amount of K absorbed through the root did not change with the transition to SD (Figs 2 and 3C). This inconsistency may be accounted for by the low elevation level of PttHAK-like1 expression and the significant differences in the amino acid sequence of the poplar genes. To confirm, we sequenced the entire PttHAK-like1 gene from the hybrid aspen T89. AtHAK5 and POPTR_0010s10450 in P. trichocarpa and PttHAK-like1 in hybrid aspen T89 and their alignment are shown in Supplementary Fig. S2. The AtHAK5 amino acid sequence showed 44.5% homology to POPTR_0010s10450 and 44.2% to PttHAK-like1. For the poplar, PttHAK-like1 and POPTR_0010s10450 shared 97.8% homology. The GEGGTFALY domain (AtHAK5-type transporters) is important for K + and Cs + selectivity 34 . Of these three genes, the GEGGTFALY domain was completely conserved and, consequently, there is no obvious explanation for the functional divergence.
Despite the steady 42 K uptake manner through seasonal transitions, Cs accumulation activity was down-regulated under SD6 conditions. Therefore, the Cs + and K + transport systems are probably separately regulated in poplar.

Discussion
Potassium is one of the most abundant essential plant nutrients. It is required for metabolism, photosynthesis, the tricarboxylic acid (TCA) cycle, glycolysis, and amino acid biosynthesis 35 . Maintaining enough K within the plant body is therefore quite important. For example, in K-deficient sunflowers, the carbon flux into the TCA cycle decreased due to changes in carbon distribution 36 . Potassium deficiency also inhibited sugar translocation in several plants 37 . Thus, the amount of K is closely tied to processes that maintain homeostasis in plants such as charge balance, pH regulation, and osmotic potential 35 . Potassium is the dominant solute in the xylem-and phloem saps of Ricinus communis and the circulation of K is required for plant growth and development 38 .
In this study, the relative amounts of 42 K accumulated were compared over seasonal transitions. It was found that 42 K accumulation remained constant until SD6. Dormant buds formed up to four weeks after the onset of the short-day treatment (data not shown); therefore, K re-translocation should have already started at SD6. Nevertheless, the results showed that 42 K accumulation from root uptake and the expression of genes related to the root K + uptake were almost constant (Figs 2 and 3B-F). It has been reported that the induction of AtHAK5 was enhanced by K + deficiency 7,8 or by Cs + applications when there was sufficient K + 39 . Therefore, the slight increase in PttHAK-like1 expression under SD6 might be a response to K + starvation during the long growth period.
Despite the constant K accumulation pattern under SD conditions, Cs accumulation drastically decreased in SD6 plants (Fig. 1A,B). Cesium ion uptake and translocation are considered to be regulated by the plant K + transport system but no down-regulation in the genes related to K + uptake was identified during SD transition (Fig. 3B-F). It is known that plant mineral uptake mechanisms are regulated by protein activity level as well as gene expression 6 . Further nutrient transport activity analysis is necessary but these results suggest the possible existence of a novel uptake transporter which carries Cs + much more efficiently than it does K + .
Since a decrease in Cs was observed only in the shoot (Fig. 1B), attention should be given to the transporters involved in Cs + transfer between the root cells adjacent to the xylem and the xylem vessels themselves. Potassium-42 translocation from the root to the shoot was not affected by the transition to SD (Fig. 2). Therefore, K + xylem loading might not be down-regulated, and there could be a Cs + re-uptake pathway from the xylem sap to the root cells. This type of regulation was hypothesized for the Zn 2+ transport system adjacent to root xylem vessels, and it may serve to keep shoot Zn 2+ concentrations below toxic levels 40 . This mechanism would only be plausible if Cs + specific transporters exist near the root xylem vessels-and these have not yet been found.
We did not find the K + uptake transporter which was obviously up-or down regulated in the transition to SD. We used Arabidopsis eFP Browser (http://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi), which provides detailed information about the Arabidopsis gene expression site and various gene induction factors such as organic-and inorganic stressors. Expressions of the homologous Arabidopsis genes AtKUP1, AtHAK5, and AtCNGC1 are not changed by short-day treatment or by exogenous abscisic acid. These expression patterns are consistent with our results and imply that short day length-induced regulation of K + uptake is also unnecessary in poplar.
Scientific RepoRts | 6:38360 | DOI: 10.1038/srep38360 Unlike K accumulation, Cs accumulation did not remain constant, but drastically changed by day length transition. Therefore, the Cs + uptake and translocation mechanisms differ from those for K + in poplar. It was not determined why Cs accumulation was down-regulated but K accumulation was constant under the transition to SD conditions. It is clear, however, that Cs accumulation was affected by photoperiod.  41,42 . The purity of the 42 K + was verified from the gamma-ray spectra emitted by the test solutions using a germanium detector (GEM-type, ORTEC, USA). The decay of the 42 K + spectral peak (1525 keV) was monitored for 7 d as described in Kobayashi et al. 43 . The half-lives of the test solutions were measured with a liquid scintillation counter (LSC-6100, Hitachi Aloka Medical, Japan) and were theoretically identical to the actual half-life of 42 K. Plants grown under LD and SD conditions were incubated with radioisotopes under the same photoperiods. Incubation times were 48 h and 24 h for the 137 Cs and 42 K experiments, respectively. Shoots and roots were separated and dried for 3 d at 50 °C. In the 137 Cs assay using SD6 plants, plants were cut into four parts: apex (shoot apex and top three leaves), leaf (remaining leaves and petioles), stem, and root. To measure 137 Cs and 42 K radioactivity, the gamma counters AccuFLEX γ 7001 (Hitachi Aloka Medical, Japan) and ARC-300 (Hitachi Aloka Medical, Japan) were used, respectively. The details of handling and measuring 137 Cs and 42 K were described in Kobayashi et al. 43 . Cesium-137 distribution was also investigated autoradiographically with a laser imaging scanner (FLA-9500, GE Healthcare, UK) in LD3, LD9, and SD6 plants. Significant differences between 42 K and 137 Cs quantities for each organ type and under each photoperiod were evaluated using one-way ANOVA.