1 Results section: the second line, ref should be cited by numbers? Please check and clarify.

Answer：Thank you very much for your suggestion. I have already checked and make sure that the reference have been cited by numbers.

2. Fig2, the figure legends are too simple and some are not clear. Please explain each item in more detail and these should be mentioned also in the text. Do not let reader to guess what you mean.

Answer：Thank you very much for your suggestion. I have carefully explained each item in the text and more details were added in the figure legends.

3. Fig3, the localization is not quite clear. Could the authors treat the cells (harboring OsPDR-GFP) with Co and then check the localization? It will be better if the authors have other methods to further prove the vacuolar localization of the OsPDR. Sometimes, the FM4-64 is also used to check plasma membrane localization of proteins.

Answer：Thank you very much for your suggestion. OsPDR-GFP expressed yeast cells with Co stress can better analyze the localization of OsPDR, and may provide a further evidence to understand the Co tolerance. We have observed the localization of OsPDR in yeast by laser confocal scanning microscope several times, and we think that OsPDR was located on the vacuolar membrane. We will further observe the subcellular localization of OsPDR protein under different concentration of Co in subsequent experiments. Thank you again for your suggestion.

4 It would be great if the authors could transform the gene into plant, e.g., rice or Arabidopsis to see what happens.

Answer：Thank you very much for your suggestion, the suggestion of investigating the gene functions in plants is very good. However, it will take a long time to get relevant results about the OsPDR transgenic plants due to the long cultivation period of plants and we will present the results in the future research. Thank you very much.

5. 3.2, first line, to investigate the transporter activity by subcellular localization. This statement is not accurate since the experiment only tells us where the protein is located but not its functions. Please clarify.

Answer：Thank you very much. I have already revised it in the text. The transporter function of OsPDR in yeast investigated was based on the study of the subcellular localization and metal tolerance experiments of OsPDR, and a large number of references were also needed to be cited to support the speculation. Thank you.

To Reviewer #2's comments

钴（Co）污染严重影响了粮食产量和人类健康，研究植物对Co污染响应的机理具有重要意义。本论文克隆了水稻中镉(Cd)响应的金属离子转运子OsPDR，通过异源转化酵母发现其对Co的耐受性提高，含量测定表明酵母体内积累了高含量的Co。进一步研究发现OsPDR蛋白定位于液泡膜上，推测OsPDR可能通过泵入高浓度的Co进入液泡膜，调控酵母体内的Co稳态。该论文为进一步研究OsPDR在植物中的功能奠定了基础。

该论文具有一定的学术水平和创新性，实验方法具有可靠性，结论合理，具有很高的潜在应用价值，为进一步研究植物对Co的响应机理奠定了基础。

我们衷心感谢评审专家对文章的认可，谢谢。

1　水稻RNA-seq表明OsPDR在Cd处理条件下其表达量能被明显上调，而将OsPDR异源转化至酵母中却发现酵母没有明显的Cd抗性，请解释。

答：十分感谢您的提问。首先OsPDR只是在Cd处理条件下其表达量有明显上调的表现，但不能证明过表达OsPDR的植物及酵母对Cd有显著的抗性，而且由于OsPDR是在酵母中异源表达并不是在水稻本体中进行表达，两种材料的结构及表达系统有不同，发挥作用方式也会有不同。在植物细胞中，转运蛋白的活性可能受到另外一些与酵母不同的蛋白因子的调节，基因的表达在时空上受到更加严格的调控，所以这就会造成它们功能上具有的一些与酵母不同的特异性，并且我们将OsPDR异源转化至酵母中并进行Cd处理，多次重复实验结果中并没有发现明显的Cd抗性，因此我认为这样的结果是可信的。

2 OsPDR的亚细胞定位除了FM4-64之外，还需要酵母体内液泡膜上的标记基因进一步确认。

答：非常感谢您的建议，您的建议对于我们很有帮助。利用酵母体内液泡膜上的标记基因确实能够进一步验证OsPDR的亚细胞定位。酵母是单细胞微生物，有细胞壁、细胞膜、细胞核、细胞质、液泡、线粒体等结构如图（酵母结构图），我们经过多次在激光共聚焦扫描显微镜下观察OsPDR在酵母中的定位，认为OsPDR定位于液泡膜上。我们在本文中利用FM4-64初步验证OsPDR在酵母中的定位情况主要以辅助后续OsPDR在酵母中异源表达后的金属抗性实验的研究，因为在酵母中发现OsPDR对于Co的抗性而未发现对Cd有明显抗性现象，我们下一步将利用酵母体内液泡膜上的标记基因来确认OsPDR在酵母中的亚细胞定位。

3 在高浓度Co处理条件下，OsPDR蛋白的亚细胞定位是否改变需要进一步观察。

答：非常感谢您的建议，您的建议对于本文水平的提高起着很大的作用。在高浓度Co处理条件下进一步观察OsPDR蛋白的亚细胞定位是一个非常好的建议，能够进一步更严谨地辅助解释后续在Co胁迫下OsPDR在酵母中异源表达对Co 的耐受性，我们会在后续实验中进一步观察高浓度Co处理条件下OsPDR蛋白的亚细胞定位情况，并且在后续水稻的研究中会分析OsPDR蛋白在水稻中的亚细胞定位以及高浓度Co处理条件下OsPDR蛋白在水稻中的亚细胞定位等，并辅以OsPDR过表达水稻中对于Co抗性功能的研究，如高浓度Co处理条件下，OsPDR过表达水稻的表型变化、生理生化情况以及OsPDR过表达水稻不同时期的Co含量的变化等等，我们目前主要希望通过金属耐受性实验来验证OsPDR在酵母中过表达是否表现出对Cd或者其他金属的抗性。谢谢您！

4. OsPDR系统分析中未标示各个基因所属的物种名信息，基因树的构建未设置合适的外类群。建议以Maximum-likelihood method重新构建系统发育树。

答：非常感谢您的建议，您的建议非常重要。我已选择合适的外类群并利用Maximum-Likelihood method重新构建了系统发育树，并且在Fig.1a的系统发育树中加入各个基因所属的物种名信息。

5 中文摘要中“通过RNA-Seq技术分析转录组”说法欠妥。

答：非常感谢您的建议。确实“通过RNA-Seq技术分析转录组”说法欠妥当，已在文中进行修改。修改后为“在通过RNA-Seq技术得到的镉响应转录组图谱中，一个镉响应金属离子转运蛋白OsPDR被鉴定出在用50 μM Cd处理24 h后，其在水稻(Oryza sativa ssp. japonica cv. Nipponbare)茎中的表达量显著上调”。

参考文献不一致，如J. Biol. Chem、Genetics and Molecular Biology和EMBO J。

答：十分感谢您的建议。我已经在文中将参考文献进行改正，谢谢您！

1 请将图3中覆盖在图片上方的文字去掉，或提供分层图文件；

答：十分感谢您的意见。我已将覆盖在图3中上方的文字去掉，谢谢您！

2 文献格式不符合要求。

请严格按照我刊征稿简则中的格式修改（尤其注意多作者的需写出3个姓名，姓前名后，名缩写,其他用“et al”）。可在我刊网站下载“Endnote样式文件及使用说明”。

答：非常感谢您的意见。我已将文献的格式进行查看并改正，谢谢！

(¹ College of Life Science, University of the Chinese Academy of Sciences, Beijing 101408, China

2 Southeast Asia Biodiversity Research Institute, Chinese Academy of Science, Yezin, Nay Pyi Taw 05282, Myanmar)

Email: hwang@ucas.ac.cn; tychai@ucas.ac.cn

Cobalt (Co) is an essential trace element for plant growth and is a component of vitamin B12. Low concentrations of Co can promote plant growth properly, but if the concentration of Co in the soil solution reaches 10 mg/L, it inhibits the growth and development of plants or even causes death, which would have a significant effect on human health^[1,2,3]. With the development of modern industry, an increasing number of cultivated rice fields are polluted by the heavy metal Co. Co, once entering the soil, is stored in the cultivated layer by adsorption, precipitation and complexation^[4,5]. Additionally, Co can be highly transported and is easily absorbed through the roots of plants and, in turn, accumulates in the vegetative and reproductive organs of plants, damaging not only the yield and quality of crops but also endangering human health by entering the human body through the food chain^[6,7].

The ATP-binding cassette (ABC) transporter subfamily, a family of membrane proteins with a strong metal transport function, is mainly located in the plasma membrane, endoplasmic omentum, mitochondrial membrane and vacuolar membrane^[8,9,10,11]. All of the members of this subfamily contain 4 or 6 highly hydrophobic transmembrane domains and ATP binding and/or nucleoside binding domains (NBDs) in the peripheral cytoplasm^[8]. The ABC transporter proteins in yeast can be classified into subfamilies, from ABCB to ABCG, according to their sequence similarity to mammalian NBDs^[12].

Recent research has revealed that ABC transporters can export harmful substances, extracellular toxins and targeted membrane modules, including substances that are related to resistance to biotic and abiotic stress in plants^[13]. The known ABC genes can be classified into three subfamilies in plants: multidrug resistance (MDR), multidrug resistance-associated protein (MRP) and pleiotropic drug resistance (PDR)^[14,15]. MRP genes play a role in many processes, such as detoxification and the transport of vacuolar flavonoid^[16,17]. The MRP and MDR subfamilies are mainly related to Cd transportation in plants^[18,19,20]. MRPs may be involved in the transport of the PC-Cd or GS-Cd complex on the vacuolar membrane^[21,22,23]. PDR proteins, which are the subject of this study, have been found in plants that can respond to abiotic and biotic stress^[24]. cDNA of SpTUR2 was cloned as the first PDR gene in Spirodella polyrhiza and encodes a PDR5-like ABC transporter^[25]. The expression of PDR genes is induced by pathogens, and PDR proteins can transport products to the cell surface^[15]. PDR proteins are widely found in the plant cell membrane and can transport many cell substrates, such as metal ions^[26,27]. Furthermore, studies have shown that the absence of some ABCG/PDR transporters on the plasma membrane of rice can increase its sensitivity to heavy metals^[28]. Several PDR genes have been found in plants. NtPDR3 was induced under iron-deficiency in Nicotiana tabacum^[29,30]. The expression of AtPDR8 in Arabidopsis thaliana increased, and the tolerance of transgenic plants with overexpression of AtPDR8 to Cd and Zn also increased^[31,32]. Further studies showed that the expression of OsABCG43/PDR5 had effects on the distribution of Cd in yeast cells^[33]. OsPDR9 can be induced by Cd and Zn in rice^[34]. Finally, the expression of AtPDR9 can detoxify the herbicide 2,4-D in Arabidopsis thaliana^[35]. However, there are few reports of PDR genes that are related to the transport of Co.

To analyze the function of OsPDRinvolved in heavy metal transport in yeast, confocal microscopy was used to observe the subcellular localization of OsPDR. Changes in the phenotype of yeast with OsPDR over-expressed were observed and the heavy metals content of OsPDR expressed in yeast was also measured to analyze the transport functions of OsPDR. The mechanism for regulating the Co content was defined, and this mechanism may underlie a theoretical foundation to support the breeding of rice with low Co accumulation in the future.

1 Materials and Methods

1.1 Plant materials and growth conditions

Wild-type rice (Oryza sativa cv. Nipponbare) was used as the experimental material for gene cloning. The rice seeds were germinated in sterile water in glass dishes and were kept at 37℃ for 2-3 days in the dark and then were transferred to 1/2 Hoagland’s solution that was renewed every 2 days^[36]. The materials were cultured in a greenhouse at 25℃ under a 16/8 h light/dark cycle^[37].

1.2 Gene cloning and plasmid construction

The OsPDR gene was obtained from the transcriptome profiles of cadmium-responsive metal ion transporters identified in rice^[38]. Total RNA was isolated from seedlings of wild-type rice (21-day-old) using RNAsio Plus (TaKaRa, Japan). The cDNA was synthesized by the HiScript 1st Strand cDNA Synthesis Kit (Vazyme, China). The PCR products were cloned into the pEASY^®-Blunt Zero Cloning Vector (Transgen biotech, China) and were used as templates. The full-length sequences of OsPDR, OsPDR-EGFP and EGFP werecloned into the pYES2 vector between HindIII and EcoRI restriction sites using the ClonExpress^® II TM One-Step Cloning Kit (Vazyme), and the primers are listed in Table .1

Table 1 List of primers used to amplify OsPDR sequences

Name	Sequence（5’-3’）	Vector
OsPDR-F	ATGCTCACTGGACCAGCAAC	pEASY®-Blunt-OsPDR
OsPDR-R	GAAAAGTTACCTTCCTGCAGC
pYES2-OsPDR-F	ACTATAGGGAATATTaagcttATGCTCACTGGACCAGCAAC	pYES2-OsPDR
pYES2-OsPDR-R	TGATGGATATCTGCAgaattcGAAAAGTTACCTTCCTGCAGC
pYES2-OsPDR-F	ACTATAGGGAATATTaagcttATGCTCACTGGACCAGCAAC	pYES2-OsPDR-EGFP
EGFP-OsPDR-R	CCCTTGCTCACTCTAGACATCCTTCCTGCAGCAGGTGCAA
OsPDR-EGFP-F	TTGCACCTGCTGCAGGAAGGATGTCTAGAGTGAGCAAGGG
EGFP-pYES2-R	TGATGGATATCTGCAgaattcTTACTTGTACAGCTCGTCCA
pYES2-EGFP-F	ACTATAGGGAATATTaagcttATGTCTAGAGTGAGCAAGGG	pYES2-EGFP
EGFP-pYES2-R	TGATGGATATCTGCAgaattcTTACTTGTACAGCTCGTCCA

Restriction sites are in bold letters.

1.3 Yeast strains and growth conditions

The yeast strains used in this study included BY4741 (his3D1, leu2D0, met15D0, ura3D0, mating type α) and YK44 mutant strains (ura3-52, his3-200, ZRC, Cdt1, amating type α), which were received from Euroscarf (Frankfurt, Germany). The LiOAc/PEG method was used for yeast transformation, and the positive clone was selected from synthetic dropout medium without uracil (SD-Ura)^[39]. The Yeast Extract Peptone Dextrose Medium (YPD) was used for the growth assay and glucose (Glu) in YPD medium was replaced by 2 % (w/v) galactose (Gala) (YPG) was used for inducing the expression of genes.

1.4 Subcellular localization in yeast cells by fluorescence microscopy

pYES2-EGFP and pYES2-OsPDR-EGFP were transformed intoBY4741 and YK44 and were used to observe the subcellular localization of OsPDR in yeast. Transformants were pre-cultured on SD-Ura medium and then on YPG for induction^[40]. The transformed yeast cells were selectively stained with FM4–64, a marker of vacuolar membranes^[41,42]. The yeast cells were incubated with 5 μM FM4-64 for 30 min at 30℃ in the dark, washed three times with YPG medium, cultured with 3 mL YPG for 2 h and then washed three times with 0.05 mol/L PBS (NaCl 137 mmol/L，KCl 2.7 mmol/L，Na₂HPO₄ 10 mmol/L, KH₂PO₄ 2 mmol/L, pH 7.4) before observation with a confocal laser scanning microscope (LSM 710 NLO; Carl Zeiss, Jena, Germany)^[43].

1.5 Metal tolerance experiments

For the complementation assays, pYES2-OsPDR and pYES2 plasmids were transformed into yeast BY4741 and yeast YK44 on YPD medium and YPG medium containing 500μM or 1mM CoCl₂ using the lithium acetate method^[44]. The transformed yeast cells were cultured overnight at 30℃ until reaching an OD₆₀₀ value of 0.5 and then were successively diluted to OD₆₀₀ values of 0.5, 0.05, 0.005 and 0.0005. Diluted yeast cultures (4.5 μL) were used for the drop-test experiments with metal solutions (500 μM and 1 mM ZnSO₄, 50 μM and 100 μM CdSO₄, 500 μM and 1 mM NiCl₂, and 500 μM and 1 mM CoCl₂) added to the solid medium. Three biological replicates were analyzed for all of the drop-test experiments.

1.6 ICP-MS experiments

To detect the metal content in yeast cells，OsPDR expressed in the yeast strains BY4741 and YK44 were pre-cultured on SD-Ura medium at 30℃ for 30 h and grownfor 12 h on the induction medium to induced the expression of OsPDR by a dilution of culture broth (1:1000). The OD₆₀₀ was diluted to 0.8 and subsequently incubated on the induction medium supplemented with 150 μM CoCl₂ and 0 μM CoCl₂ (control) for 48 h. Ethylenediamine tetra-acetic acid (EDTA, 10 mM) was used to absorb metal ions bound to the precipitate after centrifugation, rinsed with deionized water three times, then stored at 50°C for 3 days. The concentration changes of Co in yeast were determined after microwave digestion (Milestone, Italy) by inductively coupled plasma mass spectrometry (ICP-MS).

2 Results

2.1 Isolation and phylogenetic analysis of the OsPDR

Cadmium-upregulated transcriptome profiles were identified in rice roots and shoots under Cd treatment by RNA-Seq analysis^[38]. We focused on one gene that was highly responsive under Cd stress identified by the Cufflinks program^[38]. The open reading frame of OsPDR (chr07: 20207865..20213557) was 900 bp and analyzed using DNASTAR Lasergene v7.1 software^[45]. OsPDR can be classified as a member of the ABCG/PDR subfamily through a phylogenetic analysis (Fig. 1a). The OsPDR gene was 77% identical to TuPDR5 at the amino acid level using sequence alignment and encoded a deduced protein of 299 amino acids (Fig. 1b)^[25]. OsPDR has 3 transmembrane domains (TMs) predicted by TMHMM with an extracellular N terminus (http://www.cbs.dtu.dk/services/TMHMM/) (Fig. 1c,d)^[46,47]. SWISS-MODEL provides the information of the model, such as the oligomeric state, ligands and cofactors. The modelling process and reliability of the model was estimated by the global quality estimation score (GMQE:0.55) and the local composite scoring function (QMEAN. :-5.23). The combined quality estimate showed that the resulting GMQE was 0.20, which combines the QMEAN with the GMQE acquired from the alignment between target and template. The tertiary structure of OsPDR credible predicted by the SWISS MODEL was generated based on the template of the ATP-binding cassette subfamily G member 2 (ABCG2), which has 33% sequence similarity, 23.29% sequence identity with ABCG2 and the coverage was 0.98 (Fig. 2).

Fig. 1 Sequence features of OsPDR.

a A phylogenetic tree of OsPDR was speculated by the Maximum-Likelihood method and analyzed with MEGA6. ABC(ATP-binding cassette transporter), ABCG(ATP-binding cassette subfamily G), PDR(pleiotropic drug resistance) b Amino acid sequence alignment between OsPDR and TuPDR5. The alignment was conducted by ClustalW. White letters in black indicate identical amino acid residues. The predicted transmembrane domains of OsPDR are denoted by TM1-TM3. c Transmembrane domains predicted by TMHMM. d Topological analysis of OsPDRby SACS HMMTOP.

Fig. 2 Tertiary structure of OsPDR predicted by the SWISS MODEL (https://www.swissmodel.expasy.org/).

a Template of the ATP-binding cassette subfamily G member 2(ABCG2). b The model of OsPDR. c Modelling results. An analysis of the quality estimation information. The oligomeric structure (matching prediction) was a monomer without ligand and the model was built with ProMod3 Version 1.1.0. The score of GMQE was 0.55 and the score of QMEAN was -5.23. Local model quality estimates were presented as a per-reside plot by the QMEAN scoring function and the overall model quality based on global QMEAN estimates was provided as a Z-score, which the obtained values in relation to the global score caculated by a series of PDB structures with high-resolution. The combined quality estimate showed that the resulting GMQE (combines the QMEAN with the GMQE obtained from the alignment between target and template) was 0.20. The sequence identity was 23.29% and the sequence similarity was 33% between OsPDR and the template ABCG2.

2.2 OsPDRis localized to the vacuolar membrane in yeast

To confirm the subcellular localization of OsPDR, the OsPDR-EGFP fusion protein and EGFP protein were heterologously expressed in BY4741 yeast respectively. For the EGFP protein expressed in BY4741 yeast, green fluorescence was observed throughout nearly all of the cell, indicating that the protein is localized in the cytoplasmic matrix^[48,49]. To further verify the subcellular localization of the fusion protein, the lipophilic dye FM4–64 was used to specifically stain the vacuolar membrane. Merging the GFP and FM4–64 images demonstrated that the EGFP-tagged OsPDR localized to the vacuolar membrane when analyzed in the stationary phase (Fig. 3b).

Fig. 3 Subcellular localization of OsPDR.

a Schematic representation of pYES2-OsPDR-EGFP.bThe localization of EGFP (a1–a4) and OsPDR-EGFP fusion proteins (b1–b4). pYES2-EGFP and pYES2-OsPDR-EGFP expressed in the yeast strain BY4741. pYES2-EGFP and pYES2-OsPDR-EGFP expressed in yeast were cultured in SD-Ura until they reached the stationary phase. The cells were then induced in YPG medium overnight.From left to right, the figure shows the bright-field images, green fluorescence signals, FM4-64 fluorescence signals and the merged images with FM4-64 in red and GFP in green for pYSE2-EGFP or pYSE2-OsPDR-EGFP. Scale bar= 5 μm.

2.3 Overexpression of OsPDR in yeast confers Co tolerance

The pYES2 and pYES2-OsPDR plasmids were transformed into the wild-type yeast strain BY4741 and the yeast mutant YK44 that is sensitive to Zn, Cd, Co and Ni. Compared with pYES2 transformed into yeast, pYES2-OsPDR expressed in BY4741 and YK44 grew more efficiently in the presence of 500 μM CoCl₂ and 1 mM CoCl₂ plus Gala. The growth of pYES2 and pYES2-OsPDR in BY4741 and YK44 on the YPD medium was uniform (Fig. 4b and 4c). However, there were no apparent phenotypic differences between pYES2 and pYES2-OsPDR in BY4741 and YK44 on the medium after a series of drop-test experiments were carried out with different concentrations of Zn, Cd and Ni. These results indicate that the expression of OsPDR in yeast BY4741 and YK44 can enhance the tolerance to heavy metal Co and may transport Co into vacuoles together with the result of vacuolar membrane localization.

Fig. 4 Co tolerance of OsPDR over-expressed in BY4741 (b) and YK44 (sensitive to Zn/Cd/Co/Ni ) (c).

a Schematic representation of pYES2-OsPDR.b and c Sensitivity of yeast mutants on solid medium containing heavy metals. BY4741 and YK44 were transformed with pYES2 or pYES2-OsPDR and pYES2-OsPDR-transformed BY4741 and YK44 grows more efficiently than the pYES2-transformed BY4741 and YK44 (control) in the presence of 500 μM CoCl₂ and 1 mM CoCl₂. The YPG medium supplemented with 500 μM CoCl₂ and 1 mM CoCl₂ or no additional metal (control) was uesd for and incubated for 3–5 days at 30℃.

2.4 The overexpression of OsPDR affected the content of Co in yeast

The Co content of pYES2 and pYES2-OsPDR transformed into BY4741 and YK44 strains was measured by ICP-MS. When OsPDR was expressed in BY4741 and YK44, the Co content was increased compared to that found in the pYES2-only transformants treated with 150 μM CoCl₂ (Fig. 5).

The Co content did not change significantly between pYES2 and pYES2-OsPDR-transformed yeast of BY4741 and YK44 under the condition of no additional metals. The content of Co in pYES2-OsPDR-transformed yeast of BY4741 was significantly increased compared to that of pYES2-only transformed yeast of BY4741 after 150 μM CoCl₂ treatment (Fig. 5a), and the same results were obtained in yeast Yk44 (Fig. 5b). The pYES2-OsPDR-transformed yeast in BY4741 and YK44 can be more tolerant to Co compared with the empty vector pYES2 expressed in yeast BY4741 and YK44. The same results were obtained in yeast strains BY4741 and YK44. The result is consistent with the results of the metal tolerance experiments in yeast, which further proved that heavy metal Co can be stored in vacuoles where can isolate Co from causing damage to the yeast cells (Fig. 4).

Fig. 5 ICP-MS measurement of the Co content in yeast BY4741 and YK44 with pYES2-transformed and pYES2-OsPDR-transformed.

The content of Co in BY4741 cells grown in the presence of 150 μM Co or no additional metal added to the YPG culture medium. b Content of Co in YK44 cells grown in the presence of 150 μM Co or no additional metal added to the YPG culture medium. The error bars represent the SD of three replicates (* p<0.05, ** p<0.01).

3 Discussion

3.1 OsPDR can enhance Co tolerance in yeast

Previous studies have shown that the expression of OsPDR is upregulated in shoots of rice (43-fold) with 24-h Cd exposure and exhibits no difference in the roots between 1 and 24 h. The gene for PDR5 (the homolog of yeast OsPDR) cloned from Saccharomyces cerevisiae is linked to the excretion of cytotoxic metabolites^[25]. Here, we analyzed the function of OsPDR in yeast. In this study,we found no Cd sensitivity in yeast expressing OsPDR, however, we found that heterologous expression of OsPDR in yeast BY4741 and YK44 can result in increasing tolerance to Co when we tested the transport ability of OsPDR (Fig. 4). The functions of vacuoles are diverse and can involve the following points: maintaining osmotic pressure, participating in and regulating the accumulation and transportation of intracellular substances, maintaining the stability of cells and their internal environment and isolating wastes which have detoxifying effects. These data suggest that OsPDR is a vacuolar membrane-localized Co transporter that can pump Co into vacuoles for storage. Our results in yeast differ from a previous study that showed that OsPDR was highly responsive to cadmium exposure in the rice shoot. This inconsistency may be largely due to the heterologous expression of plant genes in yeast that can yield a difference from the functions of genes in rice plants. The reason that only OsPDR expressed in yeast BY4741 and YK44 showed tolerance to Co with no effect on other metals such as Zn, Ni, or Cd is not clear. We hypothesize that the differences may be related to the different mechanisms for compartmentalizing metals and/or the toxicity of these metals toward yeast^[50].

3.2 Overexpression of OsPDR increased the Co content in yeast under Co stress

Empty vector-transformed yeast and OsPDR-transformed yeast showed higher Co contents for both BY4741 or YK44 when treated with 150 μM CoCl₂ compared with other metals. This difference may be related to the greater content of Co transported from the culture medium into cells. The content of Co was almost unchanged between yeast cells (BY4741 and YK44) expressing pYES2 or pYES2-OsPDR without any Co in the medium with Gala. However, with 150 μM CoCl₂ supplied in the medium, the Co content was higher in OsPDR-transformed yeast compared with empty vector-transformed yeast (Fig. 5), which means that the expression of OsPDR can be more tolerance to Co. The localization of EGFP-tagged OsPDR to the vacuolar membrane also supported that supposition (Fig. 3). This evidence strongly suggests that the metal tolerance ofyeast strains with transformed OsPDR might be due to the sequestration of Co into vacuoles through the activity of OsPDR.

Taken together, our results show that OsPDR can function as a Co influx pump that can sequester extra Co into vacuoles in yeast. To further understand the function of OsPDR,additional studies need to be carried out in plants.

Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant No. U1632111, Grant No. 61672489), and the Southeast Asia Biodiversity Research Institute, Chinese Academy of Sciences (Grant No. Y4ZK111B01), and the Chinese Academy of Sciences (Grant No. KJRH2015-001). We thank Dr. David Eide (University of Wisconsin-Madison) for the gifts of yeast strains used for complementation analysis and Euroscarf (Frankfurt, Germany) for providing the yeast strains BY4741 and YK44.

邮箱: hwang@ucas.ac.cn; tychai@ucas.ac.cn

Manuscript ID: 20180242Title: Heterologous expression of OsPDR enhances the tolerance to Co in yeast

Tian Authors:Siqi， Qiao Kun， Wang Fanhong， Liang Shuang， Wang Hong， Chai Tuanyao

References