Calcium-dependent protein kinases (CDPKs) as crucial sensors of calcium concentration changes play important roles in responding to abiotic and biotic stresses. In this study, 55 BrrCDPK genes, which were phylogenetically clustered into four subfamilies, were identified. Chromosome locations indicated that the CDPK family in turnip expanded by segmental duplication and genome rearrangement. Moreover, gene expression profiles showed that different BrrCDPKs were expressed in specific tissues or stages. Transcript levels of BrrCDPKs indicated that they were involved in abiotic and biotic stresses and that paralogs exhibited functional divergence. Additionally, we identified 15 Rboh genes in turnip; the results of yeast two-hybrid analysis suggested that BrrRbohD1 interacted only with BrrCDPK10 and that BrrRbohD2 interacted with BrrCDPK4/7/9/10/17/22/23. Most of the genes play an important role in pst DC3000 defense by regulating the accumulation of H2O2 and stomatal closure. Our study may provide an important foundation for future functional analysis of BrrCDPKs and reveal further biological roles.
Calcium, a universal second messenger, plays an important role in the signal transduction pathways that were developed to survive developmental and stress stimuli in eukaryotes1,2. In plants, Ca2+ changes are sensed and decoded by Ca2+ sensors or Ca2+-binding proteins that transduce signatures into a series of downstream effects3,4. Four classes of sophisticated Ca2+ sensors or Ca2+-binding proteins exist, including calmodulins (CaM), calmodulin-like proteins, calcineurin B-like proteins, and calcium-dependent protein kinases (CDPK)5,6,7,8. CDPKs are unique in having both protein kinase and calmodulin-like domains in a single polypeptide that directly result in Ca2+-binding and Ca2+-stimulated kinase activities without independent CaM9,10.
CDPKs forms large multigene families; 34, 31, 26, 35, 25, and 50 CDPKs have been identified in Arabidopsis thaliana 11, rice12, wheat13, maize14, canola15, and soybean16, respectively. A typical CDPK protein contains a variable N-terminal domain, a catalytic Ser/Thr protein kinase domain, an autoinhibitory domain, and a CaM domain5,11. The variable N-terminal domain may be important for substrate recognition17 and usually has palmitoylation or myristoylation sites that are related to membrane localization9. The protein kinase domain is the catalytic domain containing an adenosine triphosphate binding site and lies adjacent to the autoinhibitory junction domain. The consecutive calmodulin-like domain always contains four EF hands for Ca2+ binding and is the reason why CDPKs rely on Ca2+ and not calmodulin11.
Many studies have reported that CDPKs are involved in plant responses to biotic and abiotic stresses and signal the transduction of hormones11,12,18,19. In Arabidopsis, AtCDPK1 and AtCDPK2 mRNA expression were rapidly induced by drought and high salt20. AtCDPK4/5/6/11 regulated microbe-associated molecular patterns (MAMPs) or pathogen-associated molecular patterns that triggered immunity and phosphorylated WRKY transcription factors to regulate gene expression21. Moreover, by working together with their substrates22, CDPKs have diverse functions in carbon and nitrogen metabolism, phospholipid synthesis, defense responses, ion and water transport, cytoskeleton organization, transcription, and hormone responses23. Many transcription factors, such as abscisic acid (ABA)-responsive element-binding factor (ABF), repression of shoot growth (RSG), and heat shock factor B2a, are confirmed substrates of CDPK. AtDi19-2, a drought-induced protein family member, was strongly phosphorylated by AtCPK1624. AtCPK32 regulated ABA-responsive gene expression via ABF425. NtCDPK1 interacted with NtRSG and phosphorylated Ser-114 of NtRSG26. Additionally, Rboh was a substrate of CDPK27. In Arabidopsis, AtCDPK1/2/4/11 phosphorylated Rbohs to induce the production of reactive oxygen species (ROS)28 that played a positive role in resistance to various pathogens by regulating the initiation of programmed cell death or by promoting the salicylic acid (SA) signaling pathway28,29. Furthermore, NtCDPK2/3 and StCDPK4/5 were biochemically activated and participated in the induction of early defense responses by phosphorylating RbohB after exposure to biotic stress30,31.
Turnip (Brassica rapa var. rapa) is a subspecies of B. rapa 32 and an important traditional crop plant in the Tibet Plateau. The tuberous root of turnip is used as food, forage, and medicine. It has been used for human consumption worldwide33. Turnip can also grow well in cold regions34. CDPK genes are promising candidates for modification of plant stress tolerance. However, few studies on turnip are available. Thus, we initiated a study to isolate CDPKs from turnip. In this study, we determined the number of CDPK genes in turnip and analyzed phylogeny, chromosome locations, divergence time, expression profiles, and transcriptional levels. Furthermore, we studied the relationship between BrrCDPKs and BrrRbohs using leaf inoculation with pst DC3000. Our results provide important information on the evolutionary history and the biological functions of the turnip CDPK family.
Identification of the CDPK gene family in turnip
To find all CDPKs in turnip, a genome-wide analysis of the CDPK gene family was performed with the assistance of the Turnip Genome Database in the JBrowse website. A total of 72 proteins with a protein kinase domain and at least one EF-hand domain were identified. Up to 17 of them that were similar to CDPK-related proteins, calcium/calmodulin-dependent proteins, and calcium and calcium/calmodulin-dependent protein kinases were eliminated; the remaining proteins were considered to be BrrCDPKs. We then named the remaining genes BrrCDPK1 to BrrCDPK55 according to the position of the genes on chromosomes (Table 1). Additionally, we obtained detailed gene information. Of these 55 genes, 52 BrrCDPK genes contained four EF-hand domains, with BrrCDPK5, BrrCDPK25, and BrrCDPK45 having only three, one, and three EF-hands, respectively. The BrrCDPKs identified in our study ranged in molecular mass from 38.45 kDa to 106.2 kDa, and all contained the typical CDPK structure. Furthermore, 38 CDPKs contained myristoylation sites and 47 of the 55 BrrCDPK proteins possessed palmitoylation sites at their N-termini (Table 1). CDPK proteins possessing myristoylation and palmitoylation motifs are related to their subcellular localization and substrate specificity5,35.
Phylogenetic relationships and exon–intron structure analysis
To investigate the phylogenetic relationship of CDPK genes, a neighbor-joining tree was constructed by CDPK protein sequences from turnip (Fig. 1). According to the phylogenetic tree, all CDPKs from turnip were divided into four major subfamilies based on the tree topology (Groups 1, 2, 3, and 4); this agreed with the classification of Arabidopsis CDPKs 11. As exon/intron structures and the types and numbers of introns can demonstrate the evolutionary history of some gene families36, the exon/intron structure of all 55 BrrCDPKs was analyzed to gain further insight. As Fig. 1 shows, most of the members of Group 1 had five to six introns except BrrCDPK17, BrrCDPK32, BrrCDPK36, and BrrCDPK37, whereas those in Group 3 contained six to seven introns. Additionally, five to seven and 10 to 12 introns were found in Groups 2 and 4, respectively. In general, BrrCDPKs that clustered in the same subfamily showed similar exon/intron structures, indicating their close evolutionary relationships and suggesting that gene family expansion occurred via ancient paralogs or multiple origins of gene ancestry.
Chromosomal distributions of BrrCDPKs
To understand the genomic distribution of the predicted BrrCDPKs, the DNA sequences were used to search the turnip genome database. As seen in Fig. 2, chromosome 8 was the only one of the 10 chromosomes that carried no BrrCDPK genes, while only chromosome 3 had 11 BrrCDPK genes, the maximum number found on one chromosome. Meanwhile, chromosomes 4 and 7 contained four BrrCDPKs, chromosomes 1 and 6 carried five BrrCDPKs, chromosomes 2 and 5 contained six BrrCDPK genes, and chromosomes 9 and 10 had seven BrrCDPK genes. Interestingly, all BrrCDPKs were located in front of the chromosomes and most adjacent genes were arranged densely. Moreover, a large number of segmental duplications were found between genes and chromosomes. For instance, BrrCDPK12 was close to BrrCDPK13 on chromosome 3 and BrrCDPK54 was next to BrrCDPK55 on chromosome 10. Furthermore, BrrCDPK12 and BrrCDPK55 were paralogs, as were BrrCDPK13 and BrrCDPK54.
Evolutionary history of the turnip CDPK gene family
To estimate the divergence time of turnip CDPKs, synonymous (Ks) and nonsynonymous (Ka) mutation rates were measured between predicted orthologous CDPK genes. We used a divergence rate of 1.5 × 10−8 mutations per Ks site per year37 to estimate the divergence time of 21 pairs of CDPK orthologs. We found that the estimated time was between 5.740 and 28.343 MYA (Table 2). The divergence time of six BrrCDPK orthologous genes (BrrCDPK2 and BrrCDPK30; BrrCDPK8 and BrrCDPK53; and BrrCDPK16 and BrrCDPK31) was not included in the next calculation because the estimated separation time of these genes (28.35, 28.3433, and 27.0867 MYA) was much earlier than the known speciation time for Brassica and Arabidopsis (14.5–20.4 MYA)38. The orthologs of BrrCDPK5 and BrrCDPK45 had no Ks value, indicating that they did not show synonymous substitution. Thus, we cannot estimate divergence time. The average time of the remaining orthologs was calculated at ~11 MYA with a standard deviation of 3 MYA. Additionally, the Ka/Ks (ω) value of all the predicted CDPK paralogs that had average values of 0.076 was less than one, suggesting that the 20 pairs of CDPK proteins were under strong purifying selection pressure.
Development-related expression profiles of BrrCDPK genes
Transcriptional profiles are typically closely related to a gene function. To investigate the link between evolutionary and functional divergence of 55 BrrCDPK genes in turnip development and growth, the expression levels of turnip CDPK genes were analyzed in four different tissues (root, stem, leaf, and flower) by quantitative real-time polymerase chain reaction (qRT-PCR) and using publicly available qRT-PCR expression profiles of three root development stages39. A heatmap was created using Genesis 126.96.36.199 software. As shown in Fig. 3, most BrrCDPK genes were specifically expressed in the root, leaf, and flower, whereas only two BrrCDPKs (41 and 47) were highly expressed at the stem. Additionally, BrrCDPK46 was highly expressed in all organs except the stem. The root-specific CDPKs can be targets for the breeding of flourishing roots and waterlogging-tolerant cultivars because of the need for tuberous root and turnip’s vulnerability to flooding.
To analyze gene expression in different stages of tuberous root development, a heat map of different root development stages was made using the transcriptional data generated by Jingjuan Li39. Samples were collected on day 18 (the early stage before cortex splitting, ES), day 28 (the cortex splitting stage, CSS), and day 64 (the stage of root thickening, RTS) after sowing, respectively. Additionally, every stage had two independent biological replicates. The samples were labeled as ES1, CSS1, and RTS1 for the first biological replicate, and ES2, CSS2, and RTS2 for the second replicate. As Fig. 3 shows, most BrrCDPK genes were expressed at the early stage before cortex splitting (ES, day 18); six genes (BrrCDPK5, 11, 20, 21, 22, and 25) were not found in the transcriptional data because of their low expression levels in the root. Additionally, 19 BrrCDPKs were expressed in both replicates of ES, three genes (BrrCDPK35, 36, and 39) in CSS, and six in RTS.
To observe the functional divergence between homologous genes, the expression levels of different tissues and stages were analyzed. As shown in Fig. 3, BrrCDPK4 and BrrCDPK36, BrrCDPK7 and BrrCDPK51, BrrCDPK13 and BrrCDPK54, and BrrCDPK34 and BrrCDPK52 were expressed in the same tissues. However, other paralogs were differentially expressed in different tissues. For example, BrrCDPK1 was expressed in the root and flower. Its homologous gene BrrCDPK21 exhibited high expression in the root and leaf, similar to paralogs at different root stages. Orthologs such as BrrCDPK8 and BrrCDPK53, BrrCDPK17 and BrrCDPK32, and BrrCDPK33 and BrrCDPK48 were highly expressed at ES, and two paralogous gene pairs (BrrCDPK35 and BrrCDPK43 and BrrCDPK39, and BrrCDPK47) had relatively high expression at ES and CSS. However, some orthologs exhibited rather different patterns of expression. For instance, BrrCDPK12 and BrrCDPK55 were expressed at RTS and ES, respectively. Overall, some homologous genes were involved in development and growth of the same organs or at the same stages, whereas some showed functional divergence and were involved in different areas of development and growth.
Responses to abiotic and biotic stresses
Plants are regularly challenged by abiotic and biotic stresses. While CDPK may be involved in responding to abiotic and biotic stimuli37,40, information related to CDPK involvement in stress responses in turnip is limited. To gain insight into the functions of BrrCDPKs in abiotic and biotic stress responses, the expression levels of 55 BrrCDPKs were examined by qRT-PCR. Seedlings were harvested after being exposed to cold, salt, drought, ABA, pst DC3000, 1-aminocyclopropane-1-carboxylic acid (ACC; the precursor of ET), jasmonic acid (JA), and salicylic acid (SA). As shown in Fig. 4, each BrrCDPK gene typically responded to several treatments, with each relevant treatment altering expression of multiple BrrCDPK genes. For example, BrrCDPK1 responded to seven treatments (cold, salt, ABA, pst DC3000, ACC, JA, and SA), and eight genes (BrrCDPK1, 4, 24, 33, 38, 39, 42, and 47) were induced after 1 h of cold treatment. We also calculated the number of significantly expressed (>2.5 fold) genes responding to every treatment. We found that BrrCDPK38 and BrrCDPK47 responded to all abiotic treatments. BrrCDPK38 and BrrCDPK42 were significantly expressed with pst DC3000, ACC, JA, and SA treatments. Interestingly, BrrCDPK38 was upregulated with all eight treatments, suggesting that the function of BrrCDPK38 was very wide. Generally, 39 BrrCDPK genes responded to pst DC3000 treatment, which was the maximum number of upregulated genes for any treatment, suggesting that most BrrCDPKs were involved in responding to pathogens. However, the cold and JA treatments only resulted in the upregulation of 19 BrrCDPKs. Additionally, we analyzed the expression of homologous genes in the eight treatments. Three pairs (BrrCDPK13 and BrrCDPK54, BrrCDPK26 and BrrCDPK27, and BrrCDPK38 and BrrCDPK42) were showed consistent expression, with both members of each gene pair upregulated or downregulated together. Among them, BrrCDPK26 and BrrCDPK27 were downregulated in all treatments, whereas BrrCDPK38 and BrrCDPK42 were upregulated under all conditions. The remaining paralogous genes showed different expression levels with at least one stress treatment; BrrCDPK35 and BrrCDPK43 responded to the seven treatments differently and only had similar expression patterns with JA treatment. BrrCDPK12 and BrrCDPK55 were expressed differently under salt, ABA, pst DC 3000, ACC, JA, and SA stress (Fig. 4). The remaining orthologs responded to one to five treatments differently. These results further suggested that homologous genes in turnip showed functional divergence.
Interactions of BrrCDPKs with BrrRbohs
The Rboh gene family is one of the substrates of CDPK27,30. To date, no studies have reported the relationship of CDPKs and Rbohs in turnip. To investigate the relationship of BrrCDPKs and BrrRbohs, we selected 15 BrrRboh genes in turnip (Fig. 5A). Of these, we cloned seven BrrRboh genes, including BrrRbohC3, BrrRbohD1, BrrRbohD2, BrrRbohE2, BrrRbohF, BrrRbohG1, and BrrRbohH. After sequence confirmation, the correct CDPKs and Rbohs were subcloned into pGBKT7 (BD) and pGADT7 (AD) vectors, respectively. According to the results of the yeast two-hybrid assay, BrrRbohD1 interacted only with BrrCDPK10, and BrrRbohD2 interacted with BrrCDPK4/7/9/10/17/22/23 (Fig. 5B). Furthermore, Fig. 5C shows that BrrRbohD1 and BrrRbohD2 were upregulated with ACC, JA, SA, and DC3000 treatment, with BrrRbohD1 upregulated by 5-, 2-, 7-, and 3-fold, respectively. BrrRbohD2 was up-regulated 2-, 3-, 4-, and 2-fold, respectively.
H2O2 level in guard cells and stomatal aperture of turnip treated by pst DC3000
Interestingly, most CDPKs were upregulated under pst DC3000 treatment, and BrrRbohD1/D2 were clearly involved in pst DC3000 defense. Additionally, rapid generation of H2O2 during the oxidative burst has been recognized as a central component of plant defense responses to pathogens41,42,43. Moreover, a ROS burst is involved in stomatal closure associated with defense and cell death responses44,45. Therefore, we analyzed the H2O2 level in guard cells and stomatal apertures at 0, 1, and 3 h after treatment with pst DC3000. As Fig. 6A shows, the level of H2O2 at both 1 and 3 h was approximately thrice that at 0 h. We also observed the state of stomata and stomatal apertures. As shown in Fig. 6B, the stomata were opening at 0 h and almost closing at 1 and 3 h. The concentration of H2O2 was much higher at 1 and 3 h than 0 h. Additionally, stomatal aperture size at 0 h was almost thrice that at 1 and 3 h (Fig. 6C).
CDPK genes have been found in plant genomes, such as Arabidopsis 11, rice46, wheat47, cassava48, and grape49. Moreover, CDPKs play important roles in biotic and abiotic stresses. In this study, we identified 55 CDPKs in turnip through a genome-wide database search based on conserved domains and similarities to known CDPKs. Only 34, 31, 26, 35, and 25 CDPKs have been identified in Arabidopsis, rice, wheat, maize, and canola, respectively, much fewer than the number found here in turnip11,12,13,14,15. Similar to turnip, however, the CDPK gene family in soybean has been expanded to 50 members16. A phylogenetic tree was constructed that divided the turnip CDPKs into four major subfamilies that agreed with the subgroups of Arabidopsis 11. However, every subfamily in turnip had more members than that in Arabidopsis. Even the simple subgroup had five members, more than the three found in Arabidopsis 11. Additionally, 15 pairs of orthologous genes in turnip were located on two duplicated segments. This indicated that they were derived from segmental duplication because most paralogous genes were not physically adjacent to each other and were instead located on different chromosomes, such as BrrCDPK12 and BrrCDPK55. This result indicated that segmental duplication and genome rearrangement occurred during turnip evolution50. This phenomenon may be the cause of the expansion of the CDPK gene family in turnip.
Gene divergence is a significant process in the evolution of novel functions51. To investigate the functional divergence of BrrCDPKs, we analyzed the expression of 21 pairs of orthologous genes. In this study, all the Ka/Ks values for 21 homologous genes were less than one, indicating a functional constraint with purifying selection on the genes52. In different tissues, only four pairs were expressed in the same tissues, whereas members of other pairs were expressed in different tissues (Fig. 3). Three pairs were expressed at the same stage of root development. Others were expressed at different stages or not expressed in the root, similar to the patterns observed in other plant species such as cucumber53, soybean16, and pepper54. Plant growth, development, and response to environmental stresses are regulated at the transcriptional level55,56,57. We, therefore, analyzed the expression of BrrCDPKs under different stress conditions. Our study provided evidence that most BrrCDPKs had remarkable responses to both abiotic and biotic stress, which was consistent with the results obtained in Arabidopsis 11,20. Many CDPKs were able to respond to several stresses; BrrCDPK38 responded to eight treatments. By contrast, multiple CDPK genes could be affected by one stress treatment. For instance, 39 genes were upregulated by pst DC3000 stress. Additionally, nine pairs were differentially expressed with more than four different stress treatments, whereas three pairs remained the same. Interestingly, one pair, BrrCDPK26 and BrrCDPK27, was downregulated in all treatments, whereas the BrrCDPK38 and BrrCDPK42 pair was upregulated, indicating that these two pairs showed no functional divergence. Through the analysis of these results, we found that different BrrCDPK genes were specifically expressed in different tissues and developmental stages and responded to different stresses, showing the functional diversity of BrrCDPKs. Additionally, some homologous genes showed functional divergence, with their responses to different stresses potentially extending novel functions51.
Many studies have shown that some CDPK genes are involved in anti-pathogen responses; Rbohs, one of the substrates of CDPK, played an important role in plant immunity27,30,58,59,60,61. StCDPK5 also interacted with StRbohC27. Studying the interactions between BrrCDPKs and BrrRbohs, we found that BrrRbohD1 only interacted with BrrCDPK10, whereas BrrRbohD2 interacted with BrrCDPK4/7/9/10/17/22/23 through the yeast two-hybrid assay. This result confirmed that BrrCDPKs interacted with BrrRbohs. Moreover, according to our data, we found that most CDPKs were upregulated under pst DC3000 treatment. Multiple BrrCDPK genes also responded to phytohormones such as ACC, JA, and SA that participate in biotic stress signaling pathways62,63. To identify the function of BrrRbohD1/2 in defense against pst DC3000, we determined that both were upregulated under ACC, JA, SA, and pst DC3000 stress, indicating that they are involved in resistance to pathogens and that BrrRbohD1 was redundant with the dominant BrrRbohD2 gene. Additionally, BrrCDPK4/10/17 were significantly upregulated under pst DC3000 stress where the expression of each increased by almost 2-fold. We, therefore, hypothesized that BrrCDPK4/10/17 interacted with BrrRbohD1/D2 to facilitate resistance to pst DC3000.
The accumulation of H2O2 is one of the earliest events in the hypersensitive response (HR)41 by which pathogens are recognized. Additionally, H2O2 is rapidly produced after pathogen infection41. Previous studies have demonstrated that AtCDPK1 and AtRbohD produced greatly increased ROS to protect against infection with pst DC300060. CDPKs may also function upstream of ROS production30,64 and may regulate RbohD to accumulate ROS. As Fig. 6A shows, the H2O2 content at 1 and 3 h was almost thrice that at 0 h suggesting that H2O2 accumulated after pst DC3000 inoculation. Additionally, the stomata are a natural entry site for harmful microbes, with many microbes invading plant tissues through the stomata. For example, pst DC3000 enters plants via open stomata on Arabidopsis leaves65. Similarly, stomatal closure can prevent disease incidence; leaf inoculation with pst DC3000 decreased stomatal aperture within 1 to 2 h of inoculation65. H2O2 functions in both MAMP- and ABA-induced stomatal closure66. As shown in Fig. 6B, the stomata were almost closed within 1 and 3 h of inoculation and H2O2 was found in the guard cells. Moreover, as seen in Fig. 6C, the stomatal apertures at 1 and 3 h were smaller than at 0 h. We, therefore, concluded that H2O2 can be induced by inoculation with pst DC3000 and that increased H2O2 can regulate stomatal closure to prevent pathogen entry. Previously, MAMP and ABA have been shown to trigger ROS production by activating Rbohs, especially RbohD67,68. Furthermore, CDPK signaling is known to occur upstream of H2O2 69, and MAMP-triggered ROS production has been shown to regulate stomatal closure21. Consistent with these earlier results, we hypothesize that pst DC3000-activated BrrCDPK4/10/17 activated BrrRbohD1/D2 to produce H2O2 that induced stomatal closure to prevent pst DC3000 infection.
At present, few studies of the CDPK family have been reported in turnip and the function of CDPK family members remain to be clarified. In our study, 55 BrrCDPK genes were identified that were phylogenetically clustered into four subfamilies. Chromosome locations indicated that the turnip CDPK gene family exhibited expansion by segmental duplication and genome rearrangement. Moreover, gene expression profiles showed that different BrrCDPKs were expressed in specific tissues or stages. The transcript levels of BrrCDPKs showed that they were involved in responses to abiotic and biotic stress and that paralogs showed functional divergence. Subsequently, our data revealed that BrrRbohD1 only interacted with BrrCDPK10 and that BrrRbohD2 interacted with BrrCDPK4/7/9/10/17/22/23. Additionally, most of these were involved in defense against pst DC3000. We also observed that the accumulation of H2O2 and stomatal closure were induced by inoculation with pst DC3000. We predicted that BrrCDPK4/10/17 regulated BrrRbohD1/D2 to increase the level of H2O2 and regulate stomatal closure to assist in disease resistance. These results may lay the foundation for further studies of turnip CDPK genes and provide candidate genes for the development of plant immunity.
Identification of the CDPK gene family in turnip
To identify CDPK genes in turnip, we obtained genes from the Turnip Genome Database in JBrowse website (http://www.bioinformatics.nl/brassica/index.html?data=bras_tp%2Fdata&loc=A01%3A11418033.17127994&tracks=DNA%2CGenes%2CTranscripts&highlight=). The protein domains and functional sites were examined using the domain analysis program ps_scan.pl70. Protein sequences, including protein kinase domains (PS50011) and EF-hand calcium-binding domains (PS50222), were extracted and used to search against the GenBank nonredundant (Nr) protein database. Then, we excluded the sequences that were CDPK-related proteins, calcium/calmodulin-dependent proteins, and calcium and calcium/calmodulin-dependent protein kinases. The remaining proteins were regarded as BrrCDPKs. The number of EF hands was predicted by SMART (http://smart.embl-heidelberg.de/)71. The molecular weight, theoretical pI, and grand average of hydropathicity were predicted by the ProtParam tool of ExPaSy (http://web.expasy.org/protparam/)72. Myristoylation and palmitoylation sites and N-terminal acylation were predicted using PlantsP (http://plantsp.genomics.purdue.edu/myrist.html), CSS-Palm 3.073, and NetAcet 1.074 software, respectively.
Sequence alignment and phylogenetic analyses
Fifty-five full-length turnip CDPK protein sequences were aligned via the MAFFT 7.0 program and then used to construct a phylogenic tree via the neighbor-joining method of MEGA7.0 software75. Bootstrap values were estimated to assess the relative support for each branch with 1,000 replicates. Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/) was used to analyze the gene exon/intron structure76.
Genomic distribution and synteny analysis
BrrCDPKs were mapped on chromosomes by confirming their detailed chromosomal position supplied in the Turnip Genome Database. The segmental and tandem duplication regions were obtained using the MCscanX software77. For synteny analysis, synteny blocks of the turnip gene were visualized using Circis (http://circos.ca/).
Divergence time estimation
Synonymous (Ks) and nonsynonymous (Ka) substitution rates were calculated using the codeml program of PAML478. The divergence time (T) of turnip CDPK gene pairs was calculated using the formula T = Ks/2r, where r represents the divergence rate of 1.5 × 10−8 mutations per Ks site per year.
Plant growth, and sample treatments
Seeds of turnip were germinated in sterile culture dishes with wet filter paper for 24 h in the dark at 23 °C. The seedlings were grown in perlite and Hoagland’s nutrient solution under controlled conditions (28 °C day/25 °C night cycle, 200 mmol photons m−2 s−1 light intensity, and 75−80% relative humidity). After 4 w of germination, the seedlings were subjected to different treatments. For cold treatment, plants were exposed to 4 °C. For salt and drought stress, 250 mM NaCl or 20% (w/v) polyethylene glycol was sprayed on both sides of each leaf. After 1 h, the leaves were cut and frozen in liquid nitrogen immediately, and stored at −80 °C. For JA, SA, ET, and ABA treatments, 7.5 g/L JA and 1 mM SA, ACC, or ABA was sprayed on the leaves, and sampled after 30 min16. Seedlings were soaked in Pst DC3000 (Pseudomonas syringae pv. Tomato DC3000) bacterium solution for 10 min and the leaves were harvested after 1 h.
Development-related expression profiles
The roots, stems, leaves, and flowers of mature plants were sampled from the greenhouse using scissors and stored at −80 °C until use for tissue-specific expression analysis.
For the three root developmental stages, the expression levels from RNA-Seq data were obtained from previous research39. Samples were collected on day 18 (the early stage before cortex splitting, ES), day 28 (the cortex splitting stage, CSS), and day 64 (the stage of root thickening, RTS) after sowing. Additionally, each stage included two independent biological replicates.
RNA extraction and Analysis of quantitative real-time PCR (qRT-PCR)
Total RNA was isolated from turnip samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. RNA quality was determined using a NanoDrop ND1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Then, DNA-free total RNA (5 μg) was used for first-strand cDNA synthesis by Superscript III reverse transcriptase (Invitrogen) following the manufacturer’s instructions. The primers were designed by Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/index.cgi?LINK_LOC = BlastHome), offering the following parameters: 150–200 bp of PCR product size, Nr database, 57–63 °C primer melting temperatures, and B. rapa (taxid:3711) for ‘Organism’. The turnip tublin-β gene was used as the control (Table S1). Each sample included three biological and technical replicates. PCR reactions were performed under the following conditions: 40 cycles of 5 s at 95 °C, 15 s at 60 °C, and 34 s at 72 °C using FastStart Universal SYBR Green Master (Rox, Roche, Indianapolis, IN, USA).
Yeast two-hybrid analysis
The full-length BrrCDPKs were cloned and subcloned into pGBKT7 (BD) vectors. BrrRboh gene sequences were also obtained from the Turnip Genome Database in JBrowse website. Then, BrrRbohs were cloned and subcloned into pGADT7 (AD) vectors. The primers are shown in Table S2. The plasmids of CDPK and RBOH were transformed into yeast strain Y2H and Y187 separately. After being plated on three kinds of media, namely, SD-leucine-tryptophan (SD-LW), SD-leucine-tryptophan-histidine (SD-LWH), and SD-leucine-tryptophan-histidine-adenine (LWHA), the yeast colonies were grown at 28 °C for 2–5 d before being photographed.
The transformants were cultivated in yeast extract peptone dextrose media and were then diluted to 1/10, 1/100, and 1/1000. Two microliters of the serially diluted yeast cells was spotted on SD-LW, SD-LWH, and SD-LWHA media and grown at 28 °C for 2–5 d before being photographed.
Analysis of H2O2 concentrations and stomatal apertures
Seedlings were soaked in pst DC3000 bacterium solution for 10 min and the leaves were harvested after 1 and 3 h. Leaves were sampled before treatment as a control. The measurement of H2O2 concentrations was performed following the manufacturer’s instructions for the H2O2 assay kit.
The leaf abaxial epidermis was stained for 10 min by H2O2 probes (H2DCF-DA, 50 µmol/L)79, and a laser-scanning confocal microscope was used to observe the state of stomas and the concentrations of H2O2 in guard cells. The fluorescent light was excited with the 488-nm laser line and emission was captured using a 505–530 band-pass filter. Then, we calculated the stomatal aperture. The measurement of stomatal apertures was performed using an eyepiece micrometer. The size of a stomatal aperture is the maximum distance between the inside of two guard cells. For each time point, we analyzed three leaves, with 30 stomas evaluated randomly per leaf.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported financially by the Major Projects of National Natural Science Foundation of China (No. 31590823) and the National Natural Science Foundation of China (No. 31601999). We thank Emma Tacken, PhD, from Liwen Bianji, Edanz Group China (www.liwenbianji.cn/ac), for editing the English text of a draft of this manuscript.