Catalase (CAT), which converts H₂O₂ to H₂O and O₂, is a common antioxidant enzyme in almost all organisms and plays an important role in the protection of plants against abiotic stresses and oxidative stresses (Zhou et al., 2018). CAT mainly scavenges hydrogen peroxide in free radicals and non-radical reactive oxygen species, thus maintaining a certain balance between the production and scavenging of ROS and maintaining the stability in plants (Sofo et al., 2015). The appropriate amount of H₂O₂ in an organism can act as a second messenger involved in cellular signaling pathways, causing activation of immune cells, cell proliferation, cellular senescence and apoptosis (Dash et al., 2012). On the contrary, excess ROS is capable of damaging various components in biological cells, such as proteins, lipids, DNA, etc. (Umeno et al., 2017). Therefore, CAT as a scavenger plays an extremely vital role in plants. The CATs gene family that encodes catalase in plants is a small family, which has been reported in different species. Barley (Hordeum vulgare L.) (Kendall et al., 1983) has 2 CATs genes, whereas there are three CATs genes in Arabidopsis (Arabidopsis thaliana) (Du et al., 2008), tobacco (Nicotiana L.) (Willekens et al., 1994) and rice (Oryza sativa L.) (Willekens et al., 2014), and four CATs genes in cucumbers (Cucumis sativus L.) (Hu et al., 2016) and Cotton (Gossypium spp.) (Wang et al., 2019). In cotton, the CAT proteins of two different varieties both contained a catalase core domain and a catalase immune response domain (Wang et al., 2019).

In ginseng, various expression profiles of PgCat1 was found in the leaves, stems and roots of the seedlings (Purev et al., 2010). Different stresses such as heavy metals, plant hormones, osmotic agents, high light, and abiotic stress can all induce the expression of PgCAT1(Purev et al., 2010). The mRNA transcripts of CAT1 and CAT2 were only detected in non-senescent leaves of tobacco, but the expression of CAT3 was detected in both non-senescent leaves and senescent leaves (Niewiadomska et al., 2009). Overexpression of ScCAT1enhanced the growth of recombinant E. coli under the stress of CuCl₂, CdCl₂ and NaCl, indicating that ScCAT1 could improve the tolerance of recombinant E. coli (Su et al., 2014). In rice seedlings, CATA, CATB and CATC genes are highly expressed in leaf sheaths, roots and leaves, respectively (Iwamoto et al., 2013). In Arabidopsis, the expression of CAT genes have obvious tissue specificity. CAT1and CAT2 are mainly expressed in leaves and siliques, while CAT3 is mainly expressed in stems and roots. The expression of CAT2 and CAT3 was found to be controlled by circadian rhythms; CAT2 can be activated by drought stress, while CAT3 can be activated by abscisic acid, oxidative treatments and senescence (Du et al., 2008).

Maize (Zea mays L.) is a widely grown food crop and an important energy crop and fodder that plays an important role in the national economy. Abiotic stresses such as salt (Iwamoto et al., 2013), drought (Du et al., 2008), low temperature (Hackenberg et al., 2013) and high temperature (Li et al., 2013) are the main factors affecting its yield. For example, abiotic stress and other osmotic stress will cause ROS accumulation, which will affect plant growth and yield (Zou et al., 2015). In this study, the ZmCATs and AtCATs gene families were identified and bioinformatics analysis was performed. The protein characteristics, evolutionary relationships, structural domains, secondary and tertiary structures of the ZmCATs and AtCATs gene families and the expression levels at different developmental stages were predicted and analyzed. This study will provide a theoretical basis and lay a solid foundation for the further cloning and identification of the biological functions of ZmCATs and AtCATs.

1 Results and Analysis

1.1 Genome-wide identification of the ZmCATs and AtCATs gene families

In order to retrieve catalase (CAT) genes from the maize genome, the Arabidopsis accession numbers in published articles were used to query the CATs protein sequence, and three maize CATs genes were retrieved in the NCBI database and the maize genome database. The characteristics of ZmCATs and AtCATs are shown in Table 1. The three ZmCATs gene family members are named ZmCAT1~3 according to their chromosomal distribution. The coding region of ZmCATs is between 1628~2500 bp, the number of amino acids of the protein is between 492~704 aa, the molecular weight is between 56~80 kDa, and the isoelectric point is between 6.45~9.33. The coding region of AtCATs ranged from 1900 to 2023 bp, the amino acid number of the protein was 492 aa, the molecular weight was 56.69 to 56.93 kDa, and the predicted isoelectric points were 6.63 to 7.31, respectively. The proteins encoded by both ZmCATs and AtCATs genes are hydrophobic proteins. Chromosome mapping indicated that the three ZmCATs genes were located on three of the ten chromosomes of maize, while the three AtCAT genes are located on two of the five chromosomes of Arabidopsis (Figure 1). The prediction of subcellular localization showed that 3 ZmCATs and 3 AtCATs proteins were localized in different subcellular localizations, which may play different functions in different positions.

1.2 Phylogenetic tree of ZmCATs and AtCATs gene families

The CAT protein sequences of Arabidopsis thaliana, wheat, rice, Sorghum, and upland cotton were retrieved from online databases and compared with ZmCATs. The phylogenetic tree was constructed using MEGA 7.0 software. The results show that subfamily I contains all members of the maize CAT family and one member of the Arabidopsis family (AtCAT3) as well as SbCATs, OsCATs, and TaCATs. Subfamily II is composed of two members of the Arabidopsis family (AtCAT1, AtCAT2) and members of the GhCAT family (Figure 2). In Figure 2, the CATs of monocotyledonous plants and dicotyledonous plants converge into different branches, and the relationship between Arabidopsis AtCAT3 and monocotyledonous CAT is closer than that of dicotyledonous plants. Most Arabidopsis CAT and its upland cotton homologues form phylogenetic branches with higher values, which indicates that they have a high degree of sequence homology. Similarly, maize CATs have a high degree of sequence homology with Sorghum, indicating their close kinship.

1.3 Structure and functional domain analysis of ZmCATs and AtCATs 

The online analysis software GSDS was used to analyze the structure of maize and Arabidopsis CAT family genes. The number of exons in ZmCATs is between 4 to 8, whereas the number of exons in AtCATs is between 6 and 7, respectively (Figure 3A). In order to further verify the characteristics of the CATs family of maize and Arabidopsis, a conserved domain analysis was conducted. The results showed that the two species have a high degree of similarity in the domains, and they both contain a catalase core, the structural domain (Catalase, Pfam: PF00199) and the catalase-related immune response domain (Catalase-rel, Pfam: PF06628) (Figure 3B). It shows that these two domains are necessary for CAT protein to function.

1.4 Analysis of the secondary and tertiary structure of ZmCATs and AtCATs 

The prediction of the secondary structure of ZmCATs and AtCATs family proteins shows that the proportion of α-helical amino acids in ZmCATs is between 26.42% and 27.87%, the proportion of extended chain amino acids is between 14.43% and 15.34%, the proportion of β-sheet amino acids is generally between 5.40% and 5.65%, and the proportion of random curly amino acids in ZmCATs is generally between 51.41% and 53.66%, respectively. The proportion of α-helical amino acids in AtCATs is between 26.02% and 26.42%, the proportion of extended chain amino acids is between 15.45% and 15.65%, the proportion of β-sheet amino acids is between 4.885 and 5.49%, the random coil amino acids of AtCATs, and the quantity ratio is between 52.44% and 53.25%, respectively. It can be seen that both ZmCATs and AtCATs proteins have α-helices and random coils as their main structures (Table 2; Figure 4A). Predictive analysis of the tertiary structure of ZmCATs and AtCATs protein showed that the tertiary structure of ZmCAT3 and AtCAT1 protein is similar, and the tertiary structure of ZmCAT1, ZmCAT2, AtCAT2, AtCAT3 protein is similar (Figure 4B).

1.5 Expression analysis of ZmCATs and AtCATs genes in different tissues and developmental stages

In order to determine the expression profiles of CAT genes in maize and Arabidopsis in different tissues and developmental stages, the maize GDB database was searched and expression analysis was performed. As shown in Figure 5, ZmCATs genes are expressed in different tissues, and the expression levels of ZmCATs are different in each developmental stage of vegetative growth and reproductive growth. The expression levels of the same ZmCAT in different developmental stages are also different. Compared with ZmCAT1 and ZmCAT2, ZmCAT3 has a higher expression level in various tissues in the reproductive and vegetative growth phases. The expression level of ZmCAT2 in various tissues during the reproductive growth stage and the vegetative growth stage is higher. Compared with ZmCAT2 and ZmCAT3, the expression level of ZmCAT1 in various tissues during the reproductive growth period is lower. AtCATs genes are expressed in different tissues. The expression level of AtCAT1 in mature pollen is higher than that in other growth and developmental stages, and the expression level of AtCAT1 in the developmental stages other than mature pollen is lower than that of other AtCATsgenes. Compared with AtCAT3 and AtCAT2, AtCAT1 has a higher expression level in the growth and development stage of Arabidopsis.

1.6 Expression analysis of ZmCATs gene under abiotic stress

Under high temperature stress, the expression of ZmCAT1 and ZmCAT3 genes in the ZmCATs family shows an up-regulated expression pattern, and the ZmCAT2 gene shows a down-regulated expression pattern. Under cold stress, the up-regulated expression of ZmCAT2gene is greater than the down-regulated expression of ZmCAT1 and ZmCAT3 genes. Under salt and ultraviolet irradiation stress, ZmCAT3 gene is up-regulated, but the magnitude of up-regulation is not obvious (Figure 6).

2 Discussion

Catalase (CAT) plays an important role in plant growth and development, adversity stress response, oxidative senescence and other physiological processes. Its activity is affected by various biological and abiotic factors, such as light, temperature, high salt, drought, and plant hormones and pathogenic microorganisms (Gondim et al., 2012; Luna et al., 2005). Catalase (CAT), peroxidase (POD) and superoxide dismutase (SOD) are effective scavengers of ROS, which can prevent cell damage (He et al., 2005).

In this study, the chromosome location map showed that AtCAT1 and AtCAT3 are localized on the same chromosome, but they are distributed in different subfamilies in the evolutionary tree. The reason why AtCAT1 and AtCAT2 are in the same subfamily may be that their sequences at the nucleotide and amino acid levels are more similar to each other than AtCAT3. Although they are not on the same chromosome, they remain tightly linked to promote co-regulation (Frugoli et al., 1996). In Hu L's study of cucumber CATs genes, the phylogenetic tree analysis of cucumber CATs revealed that Arabidopsis CATs were divided into three subfamilies, AtCAT1-3 were in different subfamilies, and the relationship between Arabidopsis and cucumber CAT families is relatively close (Hu et al., 2015). This study is based on the classification of CATs from monocotyledonous plants and divided them into two subfamilies. The reason for the obvious difference in this study may be related to the different members of the evolutionary tree and the different ways of classification.

In recent years, researchers have successively cloned related CAT genes from a variety of plants, and studied the functions and expression characteristics of these genes (Li et al., 2015). Chen Shanshan predicted and analyzed the secondary structure of sugarcane S-CAT protein, and found that in the secondary structure of the protein, random coils accounted for the highest proportion of 60.77%, and the proportions of α-helix and extended strands were relatively small, accounting for 17.07% and 22.15%, respectively (Chen et al, 2012). It is consistent with the results of this study on the secondary structure prediction and analysis of Arabidopsis and maize CAT protein. It is speculated that random coils are the largest structural element in the secondary structure of CAT proteins, and α-helices and extended chains are dispersed throughout the protein.

DuYanyan's research showed that different stresses may trigger different signal transduction pathways and activate the transcription of different CAT genes (Du et al., 2008). Plant catalase plays a variety of roles in germination, photorespiration, resistance to oxidative stress, and may mediate signal transduction involving H₂O₂ as a second messenger (McClung, 1997; Yang et al., 1997). This conclusion is reflected in the expression patterns of maize CAT genes under different abiotic stresses in this study. In Frugoli's research, Arabidopsis CAT2 and CAT3 showed high expression patterns in leaves (Joo et al., 2014). In this study, the FPKM expression level of the CAT genes screened in the phyzome database showed that only Arabidopsis CAT3 showed high expression levels in the leaves. In rice, OsCATA, OsCATB and OsCATC participate in the environmental stress response, the regulation of root growth regulation and photorespiration ROS level or dynamic balance, and the overexpression of OsCATA and OsCATC could enhance rice drought-tolerance (Leung et al., 2018).

This study explored the evolutionary relationship of CAT genes in monocotyledonous and dicotyledonous plants, and found that it is relatively conservative in the evolution of different species. The distribution patterns of CAT gene structures, exons/introns and conserved motifs are relatively similar in maize and Arabidopsis. The gene expression profile showed that the CAT genes of the two species were specifically expressed in different tissues. At the same time, it was found that the expression patterns of the maize CAT genes were different under different abiotic stresses. Particularly, the expression of the CAT genes was obviously induced by temperature stress, which indicated that the CAT genes may play important roles in the regulation of response to temperature stress, while the specific regulation mechanism needs further study.

3 Materials and Methods

3.1 Identification of CATs gene family members

Arabidopsis was searched from the Plant Genome Database (https://phytozome.jgi.doe.gov) and NCBI database (https://www. ncbi. nlm.nih.gov) based on the previous study (Alam et al., 2018). Protein sequences of AtCATs were retrieved from the Plant Genome Database (https://phytozome.jgi.doe.gov), Maize Genome Database (https://www.maizegdb.org/) and NCBI database (https://www.ncbi.nlm.nih. gov). Blastp screening was performed, and the initial screening was followed by sequence comparison with DNAMAN to remove redundancy and obtain the maize (Zea mays L.) CAT gene family members. Non-redundant proteins were validated using the NCBI, SMART (http://smart.embl-heidelberg.de/) and Pfam (http://pfam.xfam.org/) databases.

3.2 Basic characteristics of CATs in maize and Arabidopsis

The accession numbers, coding lengths and amino acid numbers of ZmCATs genes were obtained from the Maize Sequence database (https://www.maizegdb.org/). The molecular formula, molecular weight, isoelectric point and hydrophilicity of AtCATs and ZmCATswere obtained using the Expasy website (https://web.expasy.org/ protparam/). Subcellular localization analysis was performed using WolF PSORT Prediction (https://wolfpsort.hgc.jp/).

3.3 Protein evolutionary analysis of CATs

The CATs protein of Sorghum, Arabidopsis, rice, corn, wheat and upland cotton were analyzed by MEGA6.0 software. The Neighbor-Joining (NJ) method was used to construct a phylogenetic tree, and Bootstrap was set to 1000 for testing. In order to analyze the genetic relationship and phylogenetic relationship between the CATs of different species.

3.4 Analysis of gene structure and protein domains of CATs in maize and Arabidopsis

GSDS (http://gsds.cbi.pku.edu.cn/) was used to analyze the gene structure of ZmCATs and AtCATs. The conserved domains were analyzed through the SMART website (http://smart.embl-heidelberg.de/) and IBS 1.0.2 software. 

3.5 Advanced structural prediction of CATs protein in maize and Arabidopsis

The protein secondary structure of ZmCATs and AtCATs was predicted through the online website (https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_server.html). The tertiary structure of ZmCATs and AtCATs was predicted using the swiss website (https://swissmodel.expasy.org/). The predicted protein secondary structures were combined with the phylogenetic tree constructed by MEGA6.0 into a graph.

3.6 Expression analysis of CATs genes in maize and Arabidopsis at different tissues and development stages

Expression analysis of annotated ZmCATs was performed using the maize GDB (https://www.maizegdb.org/) database, and expression analysis of annotated AtCATs was performed via the Plant Genome Database (https://phytozome.jgi.doe.gov), and HemI software was used for clustering the heat map.

3.7 Expression analysis of maize CATs under abiotic stresses

The transcriptome database of maize under various abiotic stresses was downloaded from NCBI (https://www.ncbi.nlm.nih.gov/sra), including hot (the SAR number: SRR1238715, SRR1819196 and SRR1819198), cold (SRR1238717, SRR1819204 and SRR1819205), salt (SRR1238719) and UV (SRR1238720) treatment conditions (Makarevitch et al., 2015). Maize seedlings was treated under 24 natural light conditions until two leaves and one heart stage, and the treatment for cold was set at 5°C for 16 h, heat treatment at 50°C for 4 h, salt treatment at 300 mmol for 20 h and UV treatment for 2 h. The differentially expressed maize CAT genes were screened and the data were compiled and plotted using Excel software.

Authors’ contributions

GSN is the experiment designer and executor of the experiment, completing data analysis and writing the first draft of the paper; ZJJ and WYL participated in the experimental design and analysis of the experimental results; SWJ and LJX participated in the revision of the paper; DD and LM participated in the inspection and revision of the paper format; XJY is the creator and person in charge of the project, directing experimental design, data analysis, thesis writing and revision. All authors read and approved the final manuscript.

Acknowledgements

This research was jointly funded by the Heilongjiang Provincial Natural Foundation of China Key Project (ZD2020C007) and Heilongjiang Bayi Land Reclamation University's Three Horizontal and Three Vertical Support Program (TDJH201902). 

References

Alam N.B., and Ghosh A., 2018, Comprehensive analysis and transcript profiling of Arabidopsis thaliana and Oryza sativa catalase gene family suggests their specific roles in development and stress responses, Plant Physiology and Biochemistry, 123: 54-64

https://doi.org/10.1016/j.plaphy.2017.11.018

PMid:29223068

Chen S.S., Guo J.L., Li G.Y., Que Y.X., and Xu L.P., 2012, Electronic cloning and characterization of CAT gene from Saccharum officinarum using bioinformatics tools, Shengwu Xingxixue (China Journal of Bioinformatics), 10(1): 65-70

Dash B., and Phillips T.D., 2012, Molecular characterization of a catalase from Hydra vulgaris, Gene, 501(2): 144-152

https://doi.org/10.1016/j.gene.2012.04.015

PMid:22521743 PMCid:PMC3387816

Du Y.Y., Wang P.C., Chen J., and Song C.P., 2008, Comprehensive functional analysis of the catalase gene family in Arabidopsis thaliana, Journal of Integrative Plant Biology, 50(10): 1318-1326

https://doi.org/10.1111/j.1744-7909.2008.00741.x

PMid:19017119

Frugoli J.A., Zhong H.H., and Nuccio M.L., 1996, Catalase is encoded by a multigene family in Arabidopsis thaliana (L.) Heynh, Plant Physiology, 112(1): 327-336

https://doi.org/10.1104/pp.112.1.327

Hackenberg T., Juul T., Auzina A., Gwizdz S., Malolepszy A., Van Der Kelen K., Dam S., Bressendorff S., Lorentzen A., Roepstorff P., Nielsen K.L., Jørgensen J.E., Hofius D., Van Breusegem F., Petersen M., and Andersen S.U., 2013, Catalase and NO CATALASE ACTIVITY1 promote autophagy-dependent cell death in Arabidopsis, The Plant Cell, 25(11): 4616-4626

https://doi.org/10.1105/tpc.113.117192

He J.H., Li C.F., Liang Y.L., and Zhang Y.Q., 2005, Reactive oxygen species and signaling transduction in plant cell, Henan Nongye Kexue (Journal of Henan Agricultural Sciences), 34(8): 18-20

Hu L.F., Yang Y.G., Jiang L.W., and Liu S.Q., 2016, The catalase gene family in cucumber: genome-wide identification and organization, Genetics and Molecular Biology, 39(3): 408-415

https://doi.org/10.1590/1678-4685-GMB-2015-0192

PMid:27560990 PMCid:PMC5004828

Iwamoto M., Higo H., and Higo K., 2000, Differential diurnal expression of rice catalase genes: the 5′-flanking region of CatA is not sufficient for circadian control, Plant Science, 151(1): 39-46

https://doi.org/10.1016/S0168-9452(99)00194-6

Joo J., Lee Y.H., and Song S.I., 2014, Rice CatA, CatB, and CatC are involved in environmental stress response, root growth, and photorespiration, respectively, Journal of Plant Biology, 57(6): 375-382

https://doi.org/10.1007/s12374-014-0383-8

Kendall A.C., Keys A.J., Turner J.C., Lea P.J., and Miflin B.J., 1983, The isolation and characterisation of a catalase-deficient mutant of barley (Hordeum vulgare L.), Planta, 159(6): 505-511

https://doi.org/10.1007/BF00409139

PMid:24258326

Li J., Liu J., and Wang G., 2015, A chaperone function of NO CATALASE ACTIVITY1 is required to maintain catalase activity and for multiple stress responses in Arabidopsis, The Plant Cell, 27(3): 908-925

https://doi.org/10.1105/tpc.114.135095

PMid:25700484 PMCid:PMC4558663

Li Y., Chen L.C., Mu J.Y., and Zuo J.R., 2013, Lesion simulating disease1 interacts with catalases to regulate hypersensitive cell death in Arabidopsis, Plant Physiology, 163(2): 1059-1070

https://doi.org/10.1104/pp.113.225805

PMid:23958864 PMCid:PMC3793025

Makarevitch I., Waters A.J., and West P.T., 2015, Transposable elements contribute to activation of maize genes in response to abiotic stress, PLoS Genetics, 11(1): e1004915

https://doi.org/10.1371/journal.pgen.1004915

PMid:25569788 PMCid:PMC4287451

McClung C.R., 1997, Regulation of catalases in Arabidopsis, Free Radical Biology and Medicine, 23(3): 489-496

https://doi.org/10.1016/S0891-5849(97)00109-3

Niewiadomska E., Polzien L., Desel C., Rozpadek P., Miszalski Z., and Krupinska K., 2009, Spatial patterns of senescence and development-dependent distribution of reactive oxygen species in tobacco (Nicotiana tabacum) leaves, Journal of Plant Physiology, 166(10): 1057-1068.

https://doi.org/10.1016/j.jplph.2008.12.014

PMid:19261356

Purev M., Kim Y J, Kim M K, Kim M.K., Pulla R.K., and Yang D.C., 2010, Isolation of a novel catalase (Cat1) gene from Panax ginseng and analysis of the response of this gene to various stresses, Plant Physiology and Biochemistry, 48(6): 451-460

https://doi.org/10.1016/j.plaphy.2010.02.005

PMid:20347322

Sofo A., Scopa A., Nuzzaci M., and Antonella V., 2015, Ascorbate peroxidase and catalase activities and their genetic regulation in plants subjected to drought and salinity stresses, International Journal of Molecular Sciences, 16(6): 13561-13578

https://doi.org/10.3390/ijms160613561

PMid:26075872 PMCid:PMC4490509

Su Y.C., Guo J.L., Ling H., Chen S.S., Wang S.S., Xu L.P., Allan A.C., and Que Y.X., 2014, Isolation of a novel Peroxisomal Catalase gene from sugarcane, which is responsive to biotic and abiotic stresses, PLoS One, 9(1): e84426

Wang W., Cheng Y., Chen D., Liu D., Hu M., Dong J., Zhang X., Song L., and Shen F., 2019, The catalase gene family in cotton: Genome-wide characterization and bioinformatics analysis, Cells, 8(2): 86

https://doi.org/10.3390/cells8020086

PMid:30682777 PMCid:PMC6406514

Willekens H., Villarroel R., Van Montagu M., Inzé D., and Van Camp W., 1994, Molecular identification of catalases from Nicotiana plumbaginifolia (L.). Febs Letters, 352(1): 79-83

https://doi.org/10.1016/0014-5793(94)00923-6

Yang Y., Shah J., and Klessig D.F., 1997, Signal perception and transduction in plant defense responses, Genes & Development, 11(13): 1621-1639

https://doi.org/10.1101/gad.11.13.1621

PMid:9224713

Yu D.L., and Wang C.L., 2016, Progresses in the research of catalase, Zhongguo Zuzi Huaxue yu Xibao Huaxue Zazhi (Chinese Journal of Histochemistry and Cytochemistry), 25(2): 189-194

Zhou Y.B., Liu C., Tang D.Y., Yan L., Wang D., Yang Y.Z., Gui J.S., Zhao X.Y., Li L.G., Tang X.D., Yu F., Li J.L., Liu L.L., Zhu Y.H., Lin J.Z., and Liu X.M., 2018, The receptor-like cytoplasmic kinase STRK1 phosphorylates and activates CatC, thereby regulating H₂O₂ homeostasis and improving salt tolerance in rice, The Plant Cell, 30(5): 1100-1118

https://doi.org/10.1105/tpc.17.01000

PMid:29581216 PMCid:PMC6002193

Zou J.J., Li X.D., Ratnasekera D., Wang C., Liu W.X., Song L.F., Zhang W.Z., and Wu W.H., 2015, Arabidopsis calcium-dependent protein KINASE8 and CATALASE3 function in abscisic acid-mediated signaling and H₂O₂ homeostasis in stomatal guard cells under drought stress, The Plant Cell, 27(5): 1445-1460

https://doi.org/10.1105/tpc.15.00144

PMid:25966761 PMCid:PMC4456645