Characterization of Callose Deposition and Analysis of the Callose Synthase Gene Family of Brassica napus in Response to Leptosphaeria maculans

Callose plays a critical role in different biological processes including development as well as in the response to multiple biotic and abiotic stresses. In this study, we characterized the callose deposition in cotyledons of different Brassica napus varieties post-inoculated with different Leptosphaeria maculans isolates. Further, members of the callose synthase gene were identified from the whole genome of B. napus using the 12 Arabidopsis thaniana callose synthase protein sequences, and were then classified into three groups based on their phylogenetic relationships. Chromosomal location and duplication patterns indicated uneven distribution and segmental duplication patterns of BnCalS genes in the B. napus genome. Subsequently, gene structures, conserved domains analysis, and protein properties were analyzed for BnCalS genes. In addition, 12 B. napus orthologs of the AtCalS were selected for investigating the tissue expression pattern, indicating diverse expression patterns for these BnCalS genes. Responses of the selected 12 orthologs and all the BnCalS genes were characterized in the different types (AvrLm1-Rlm1, AvrLm4-Rlm4, AvrLepR1-LepR1) of B. napus–L. maculans interactions and B. napus-Leptosphaeria biglobosa interactions, implying their potential roles in response to Leptosphaeria infection.

Callose is a polysaccharide existing widely in the cell walls of a variety of higher plants. Its major composition is β-1,3-glucan. Synthesis and degradation of callose are involved in a variety of processes in plant development as well as in responses to multiple biotic and abiotic stresses [17]. Callose has been proposed to play crucial roles in another development. Its synthesis and degradation are tightly controlled during and following meiosis [18]. Callose is believed to be deposited as a temporary callose wall between the primary cell wall and the plasma membrane, preventing microsporocytes fusion and cohesion as well as maintaining the microsporocyte morphology and shielding microsporocytes were inoculated with D4 and D6 isolates, respectively. Westar + D4 and Westar + D6 indicate cotyledons of Westar were inoculated with D4 and D6 isolates, respectively. One, 3, 7 and 11 dpi indicate 1, 3, 7 and 11 days post-inoculation. Red arrows indicate the pycnidia produced by L. maculans. Three replicates were performed for each treatment and showed a similar pattern. Scale bars = 100 μm.

Chromosomal Localization and Duplication of BnCalS Gene Family in B. napus
Thirty-two BnCalS genes were mapped onto 15 chromosomes in B. napus ( Figure 3, Table S1). BnCalS genes were unevenly distributed throughout the chromosomes. Chromosome A09, C05, and C09 contained the maximum number of CalS genes (five), while some chromosomes contained no CalS genes, such as A04, A06, A08, C01 and C07 ( Figure 3 and Table S1). Two chromosomes (A05 and A10) contained three CalS genes ( Figure 3 and Table S1) and chromosome C04 contained two Several gene duplication modes including whole-genome duplication (WGD) or segmental duplication, tandem duplication, and rearrangements were considered to drive the evolution of protein-coding gene families at the gene and chromosomal level [34]. The contribution of the CalS gene family to the expansion of the B. napus genome was investigated using MCscanX software suite [35]. According to a whole genome analysis of gene duplications, two types of duplication modes, dispersed and WGD/segmental, were detected among 32 BnCalS genes (Table S1). Twenty-seven BnCalS genes were produced by WGD, with five others produced by dispersed duplications (Table  S1). The collinear relationships of the duplicated pairs in the BnCalS gene family in B. napus were investigated using MCscanX and are shown in Figure 4. In total, we identified 30 paralog pairs that shared a higher identity according to their protein sequences ( Figure 4). Enormous synteny blocks were also identified between A and C subgenomes or between different chromosomes of A or C, and all the 30 paralog pairs were located within synteny blocks on chromosomes ( Figure 4). Several gene duplication modes including whole-genome duplication (WGD) or segmental duplication, tandem duplication, and rearrangements were considered to drive the evolution of protein-coding gene families at the gene and chromosomal level [34]. The contribution of the CalS gene family to the expansion of the B. napus genome was investigated using MCscanX software suite [35]. According to a whole genome analysis of gene duplications, two types of duplication modes, dispersed and WGD/segmental, were detected among 32 BnCalS genes (Table S1). Twenty-seven BnCalS genes were produced by WGD, with five others produced by dispersed duplications (Table S1). The collinear relationships of the duplicated pairs in the BnCalS gene family in B. napus were investigated using MCscanX and are shown in Figure 4. In total, we identified 30 paralog pairs that shared a higher identity according to their protein sequences ( Figure 4). Enormous synteny blocks were also identified between A and C subgenomes or between different chromosomes of A or C, and all the 30 paralog pairs were located within synteny blocks on chromosomes ( Figure 4).

Gene Structure and Conserved Domain Analysis of BnCalS Genes
The open reading frame (ORF) length of the 32 BnCalS genes ranged from 1809 to 6912 bp, with an average of 4977 bp (Table 1). The number of exons in the BnCalS gene family varied from one to 55, with most genes (29 of 32, 71.9%) having more than 30 exons ( Figure 5 and Table S1). The gene pairs among the BnCalS gene family shared a high similarity in terms of exon numbers and protein/ORF length ( Figure 5 and Table S1). For example, the gene pairs BnA01.Cals: BnC03.Cals had 50 and 50 exons, respectively, and gene pair BnA09.Cals.d: BnC09.Cals.d had 3 and 3 exons, respectively ( Figure 5 and Table S1). The conserved domains in BnCalS gene family were investigated using Batch CD-Search. Nine conserved domains were discovered in the 32 CalS genes, and the Glucan_synthase superfamily domain was observed to be the most common domain (30 of 32) in the BnCalS proteins, followed by FKS1_dom1 domain, which was found to be in 27 BnCalS proteins ( Figure 6). However, the domains DUF1084 superfamily and UPF0139 superfamily were observed in only one protein, BnC06.CalS and BnA03.CalS, respectively ( Figure 6).

Gene Structure and Conserved Domain Analysis of BnCalS Genes
The open reading frame (ORF) length of the 32 BnCalS genes ranged from 1809 to 6912 bp, with an average of 4977 bp (Table 1). The number of exons in the BnCalS gene family varied from one to 55, with most genes (29 of 32, 71.9%) having more than 30 exons ( Figure 5 and Table S1). The gene pairs among the BnCalS gene family shared a high similarity in terms of exon numbers and protein/ORF length ( Figure 5 and Table S1). For example, the gene pairs BnA01.Cals: BnC03.Cals had 50 and 50 exons, respectively, and gene pair BnA09.Cals.d: BnC09.Cals.d had 3 and 3 exons, respectively ( Figure 5 and Table S1). The conserved domains in BnCalS gene family were investigated using Batch CD-Search. Nine conserved domains were discovered in the 32 CalS genes, and the Glucan_synthase superfamily domain was observed to be the most common domain (30 of 32) in the BnCalS proteins, followed by FKS1_dom1 domain, which was found to be in 27 BnCalS proteins ( Figure 6). However, the domains DUF1084 superfamily and UPF0139 superfamily were observed in only one protein, BnC06.CalS and BnA03.CalS, respectively (

Expression Analysis of the BnCalS Gene Family in Different Tissues
To gain a better understanding of the functions of BnCalS genes in plant development, we selected 12 BnCalS genes showing the highest similarity with 12 AtCalS genes and analyzed their expression patterns in six tissues including root, stem, leaf, flower, silique, and bud at three different developmental stages (<1 mm, 1-2 mm, >2 mm) using quantitative RT-PCR (qPCR) ( Table S2)

Expression Analysis of the BnCalS Gene Family in Different Tissues
To gain a better understanding of the functions of BnCalS genes in plant development, we selected 12 BnCalS genes showing the highest similarity with 12 AtCalS genes and analyzed their expression patterns in six tissues including root, stem, leaf, flower, silique, and bud at three different developmental stages (<1 mm, 1-2 mm, >2 mm) using quantitative RT-PCR (qPCR) ( Table S2) (Figure 7). In the three bud stages, BnA02.CalS had a higher expression level across all three bud stages, and BnC05.CalS.a was mainly expressed in the <1 mm stage, while BnC09.CalS.b showed higher expression levels in the >2 mm stage (Figure 7). Due to the low expression level detected here, we performed another cycle of qPCR using another known internal gene GAPDH (BnaC05g09880D) to validate the expression of BnCalS genes in different tissues. Most of these genes showed very similar expression patterns, as in Figure 7, except that expression levels increased a little in most genes ( Figures S3 and S4).

Expression Analysis of BnCalS Genes in Different R-Gene-Containing Varieties Infected by L. maculans Isolates with Different Avr Genes
Callose synthesis was reported to be involved in defense against different kinds of pathogens [36][37][38][39][40]. To gain insight into the roles of BnCalS genes in the responses of B. napus against L. maculans, 12 BnCalS with higher similarity to 12 AtCalS genes were selected for determining expression levels in varieties of MT29 (containing Rlm1 and Rlm9, inoculated with D6), Jet Neuf (containing Rlm4, inoculated with D4), and Westar (containing no R gene, inoculated with D4 or D6). One-week-old cotyledons from each cultivar were punctured prior to inoculation. As shown in Figures S1 and S2, the interactions of MT29 with D6, and Jet Neuf with D4, were defined as incompatible interactions, while interactions of Westar with D6 or D4 were indicated as compatible interactions. The expression of the 12 selected BnCalS genes showed diverse patterns (Figure 8). There were two genes, BnC04.CalS.b and BnA05.CalS.b, the expressions of which were shown to be suppressed under four treatments (Figure 8). Expression of the twelve selected genes appeared to decrease or not be altered in Jet Neuf after inoculation with D4 isolate (Figure 8) To further explore BnCalS gene expression in other types of B. napus-L. maculans pathosystems, such as the AvrLepR1-LepR1 interaction or the B. napus-L. biglobosa pathosystem, we analyzed the expression level of these BnCalS genes under different types of pathosystems based on published RNA-seq datasets [41,42]. For analyzing the expression of BnCalS during the AvrLepR1-LepR1 interaction, one-week-old seedlings of cultivars DF78 and Westar inoculated with of D3 isolates were harvested at different time points for RNA purification and subsequently RNA sequencing [42]. As shown in Figure 9  Tukey's test (ANOVA) were performed to determine the significant difference between each treatment. The statistically significant differences were defined as p < 0.05 and are marked with different letters. Error bars denote standard error.

Differential Expression Analysis of BnCalS Genes in the Interaction of B. napus with Two Leptosphaeria Species: L. maculans and L. biglobosa
Leptosphaeria biglobosa is closely related with L. maculans, both belonging to the Leptosphaeria genus [1, 43,44]. As L. biglobosa is less aggressive and causes lower losses to oilseed rape production compared with L. maculans, it is essential to investigate the host response differences to these two species. We utilized the published RNA sequencing data [41] to differentiate the expression difference of BnCalS in Westar with inoculation by L. maculans and L. biglobosa. As stated by Lowe et al. [41], 10-day-old cotyledons of Westar were inoculated with L. maculans isolate IBCN18 and L. biglobosa isolate 06J154 and collected for RNA sequencing at 7 and 11 dpi, respectively. As shown in Figure 10, various kinds of expression patterns were indicated. The first class were those having ultralow or no expression levels in the interaction of B. napus with both Leptosphaeria species, and those genes including BnA03.CalS, BnA09.CalS.d, BnC03.CalS, BnC05.CalS.c, BnC09.CalS.d, and BnUn.CalS. The second class were those whose expression was repressed in the interaction of Westar with both

Differential Expression Analysis of BnCalS Genes in the Interaction of B. napus with Two Leptosphaeria Species: L. maculans and L. biglobosa
Leptosphaeria biglobosa is closely related with L. maculans, both belonging to the Leptosphaeria genus [1, 43,44]. As L. biglobosa is less aggressive and causes lower losses to oilseed rape production compared with L. maculans, it is essential to investigate the host response differences to these two species. We utilized the published RNA sequencing data [41] to differentiate the expression difference of BnCalS in Westar with inoculation by L. maculans and L. biglobosa. As stated by Lowe et al. [41], 10-day-old cotyledons of Westar were inoculated with L. maculans isolate IBCN18 and L. biglobosa isolate 06J154 and collected for RNA sequencing at 7 and 11 dpi, respectively. As shown in Figure 10 species, including BnA02. CalS, BnA10.CalS.c, BnA05.CalS.b, BnC02.CalS, BnC09.CalS.e, BnA09.CalS.e,  BnC08.CalS, BnC05.CalS.b, BnA10.CalS.b, BnC04.CalS.b, and BnC06.CalS. The third class included BnA05. CalS.a, BnA05.CalS.c, BnC05.CalS.d, BnC04.CalS.a, and BnC09.CalS.a, in which the expression level was not altered in Westar inoculated with L. biglobosa but showed an elevated level in Westar inoculated with L. maculans. The fourth class was those which showed increased expression level in both kinds of interactions, and included BnA07. CalS, BnC05.CalS.e, BnA09.CalS.a, BnC09.CalS.a,  BnC05.CalS.a, BnA09.CalS.b, BnC09.CalS.b, and BnA10.CalS.a. The fifth class included BnA09.CalS.c and BnC09.CalS.c, in which the expression was not altered in the Westar-L. biglobosa interaction but was repressed in the Westar-L. maculans interaction.

Discussion
Callose is suggested to have diverse functions during the process of plant development as well as in responses to environmental biotic or abiotic stimulus [17,22]. Blackleg is one of the major diseases around the world, and callose deposition plays an important role in the interaction of B. napus and L. maculans [42], the causal agent of blackleg. Here, we characterize the deposition of callose during the whole infection process of blackleg in B. napus using two different R-Avr interaction types (Rlm1-AvrLm1, Rlm4-AvrLm4). This contributed to knowledge of callose deposition during blackleg infection. During the incompatible interaction, pyknotic callose structure circled the infection site, which can be beneficial in enhancing the resistance of host to pathogen by preventing the extension of hyphae [45]. In the compatible interaction, unconsolidated distribution of callose around the infection site starting from 3 dpi may cause the B. napus to be more susceptible to L. maculans. However, callose production was continuous in compatible interactions, and higher after 7 dpi, which may suggest that callose production is a sign for responses of B. napus to L. maculans, and R genes could effectively modulate the callose deposition to form a compact callose wall around the infection site. Furthermore, behavior of callose deposition displays a very similar pattern between different types of R-Avr interactions, which might suggest that they share a similar regulation mechanism of callose deposition.
Callose is synthesized by callose synthases in plants, and 12 callose synthases genes were proposed in Arabidopsis [22]. Some of the 12 AtCalS genes were investigated for their functions in plant development [20,24,46,47] and defense against biotic or abiotic stress [28,45]. However, little is known about the CalS gene family in B. napus.
In this study, 32 CalS genes were identified in B. napus, which is almost three times the number of Arabidopsis CalS genes. This may suggest that genome duplication occurred in the evolution of B. napus, in accordance with the fact that B. napus is an allotetraploid species with widespread genome duplication and merging events [32]. B. napus contains the same number of genes as the total number of B. rapa and B. oleracea, which are two ancestral species of B. napus [31], indicating that CalS genes may be derived from B. rapa and B. oleracea CalS genes. Combing phylogeny with gene structures and protein properties, it was found that most of the genes in the same groups showed similar exon numbers and protein pI values, such as the group B members having 1-4 exons with pI values around 9, while most members of group A had approximately 50 exons with pI values about 8. This indicated that group classification based on the phylogenetic tree was supported by gene structure and protein character. Five gene duplication modes including whole-genome duplication (WGD) or segmental duplication, tandem duplication, and rearrangements were considered as the major driving forces for the evolution of gene families at the gene [34]. In addition, segmental gene duplication is one of the main reasons for maintenance of gene families [48]. Thirty-two BnCalS genes were unevenly distributed on 15 of 19 chromosomes, and 27 of them were found to be produced by segmental duplication, indicating that segmental duplication played an indispensable role in the evolution of the CalS gene family in B. napus. Based on a constructed phylogenetic tree, the 32 BnCalS proteins were classified into three groups, which showed similar patterns as reported previously [49]. In addition, the Glucan_synthase family or Glucan_synthase domain was detected in all the BnCalS gene family, indicating that these CalS proteins may have glucan synthase activity, which has the main role in callose synthesis.
Expression patterns of CalS genes in different tissues have not been reported for many plants. For this study, we selected the 12 BnCalS genes showing the most similarity with the 12 AtCalS genes for analyzing expression in seven different tissues. The qPCR results from two reference genes showed a small difference in the expression level, which might be mainly due to relatively stable expression but not absolutely constant expression of these reference genes. Homologs from different species usually share a similar function in the biological process. BnC09.CalS.b showed a relatively higher level in open flowers and buds, especially in buds with diameters larger than 2mm, which is recognized as the microspore stage [50]. This expression pattern is in accordance with that of AtCalS5, however, BnC09.CalS.b was not expressed in roots, leaves, stems, and silique, which is totally different from AtClaS5. For this case, we infer that different paralogs of BnC09.CalS.b may be expressed in these tissues to ensure the normal synthesis of callose, or that the other CalS family members may help to achieve normal formation of callose in these tissues. Callose is ultra-important for pollen development and entry into the ovary [17,20,24,51], and results from the present study also suggest that most of the CalS genes display a higher expression level in the open flower and bud development process, which indicate that these CalS genes play key roles in the process of pollen development and pollen grain germination. In addition, the 12 tested CalS showed different levels in different tissues, suggesting the complex expression profiles of the CalS gene family members in plant development stages, and that different members may be responsible for the synthesis of callose in different tissues and development stages.
Callose is considered to provide mechanical support to the cell wall against environmental stresses, and its synthesis can be altered by the infection of pathogens [45]. Callose was also found to be involved in the B. napus-L. maculans interaction [42]. Comparing the expression of CalS genes in different B. napus varieties inoculated with different isolates, it was found that most of the BnCalS genes were repressed in the varieties, suggesting that L. maculans may have mechanisms to regulate the expression of CalS genes. There are some BnCalS genes that displayed different expression patterns in response to different types of variety-isolate interactions; for instance, the expression of BnC05.CalS.a was suppressed in both DF78 and Westar inoculated with D3 isolate, while its expression was inhibited in Jet Neuf but activated in Westar when they were both inoculated with the D4 isolate. Such BnCalS genes like BnC05.CalS.a may be modulated by the interaction of specific R and Avr genes. Expression of some BnCalS genes, such as BnC09.CalS.b, were repressed in the susceptible variety of Westar but were induced in resistant varieties DF78 and MT29, which may contribute to resistance conferred by R genes. There are some genes whose expression peaked earlier in incompatible interactions of plant-pathogens, contributing to the prompt response to L. maculans infection, such as BnA03.CalS and BnA09.CalS.d in the interactions of DF78 and D3. Based on these results, we concluded that BnCalS plays a key role in the response of B. napus against L. maculans infection through mediating the synthesis of callose. In addition, more up-regulated BnCalS genes were found in the interaction of B. napus-L. maculans, indicating a stronger defense response. All up-regulated BnCalS genes found in the interaction of B. napus-L. biglobosa are also shared with up-regulated genes in the interaction of B. napus-L. maculans. This may indicate that these shared BnCalS genes are in the common signaling pathway for defense against Leptosphaeria species.

Plant and Fungal Materials
The susceptible B. napus cultivar Westar (no R gene) was inoculated with L. maculans isolate D4 or D6, while the B. napus lines Jet Neuf (Rlm4) and MT29 (Rlm1 and Rlm9) were inoculated with L. maculans isolates D4 and D6, respectively. The D4 isolate contained seven avirulent genes AvrLm4, AvrLm5, AvrLm6, AvrLm7, AvrLm8, AvrLepR1, and AvrLepR2, and the D6 isolate contained five avirulent genes AvrLm1, AvrLm5, AvrLm6, AvrLm8, and AvrLmS [52]. B. napus seeds were germinated in soil and grown for one week, with a photoperiod of 16 h/8 h (16 • C dark, 22 • C light). The L. maculans inoculation was performed as described by Zhang et al. [52] and Liban, et al. [53]. Sterilized distilled water was used as mock control. Samples were collected and flash-frozen in liquid nitrogen. Cotyledons inoculated with L. maculans isolate or mock were sampled at the 1st day, 3rd day, 7th day, and 11th day after inoculation. Roots of Westar seedlings were used for root tissue, and stems of two-month-old Westar plants were used for stem tissue. For leaf tissue collection, the fifth leaves of 8-leaf-stage Westar plants were sampled. Buds with different diameters (<1 mm, 1-2 mm, and >2 mm) were collected from Westar plants and used for different stages of buds. Flowers were sampled the same day they bloomed. Silique samples were collected after flowering for two weeks.

Identification of CalS in B. napus, B. rapa, and B. oleracea
The 12 AtCalS protein sequences from the A. thaliana genome (http://www.arabidopsis.org/) were used as queries to identify CalS genes in B. napus, B. rapa, and B. oleracea via a BLASTP search [55]. The BLASTP searches used default parameters, with E-value less than 1 × 10 −20 and the score set at more than 1000. In addition, obtained BnCalS sequences were used as queries in BLASTP against the A. thaliana protein database for confirmation. The sequences of obtained proteins were analyzed using the Batch CD-Search (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) to ensure target proteins contained glucan_synthase or glucan_synthase family domains. The identified CalS protein sequences of B. napus, B. rapa, B. oleracea, and A. thaliana were used for multiple sequence alignments using the ClustalW program. Phylogenetic relationships were established using alignment results with MEGA 6.0 [56] based on the neighbor-joining (NJ) method using a boot strap replication of 1000.

Chromosomal Location and Orthologous Identification of CalS
Chromosomal position information for BnCalS genes was retrieved from relevant generic feature format (GFF) files, which were downloaded from Ensembl plants. These BnCalS genes were denominated based on their position on the chromosomes. Gene duplication types and collinearity relationships were detected using MCscanX. BLASTP was performed to do self-self-comparison using B. napus protein sequences with the E-value under 1 × 10 −10 . Program detect_collinearity_within_gene_families.pl in MCscanX was used to detect collinearity within the BnCalS gene family [35]. Five duplication types including WGD/segmental duplication events were detected for classifying origins of the duplicate genes of the B. napus genome using the program incorporated in MCScanX, and duplication information of the BnCalS gene family was extracted. The duplications within the BnCalS gene family were displayed using the program circle_plotter of MCScanX.

Gene Structures and Protein Conserved Domains Analysis of the BnCalS Gene Family
The information of exon-intron structures for BnCalS genes was extracted from GFF files and displayed using TBtools software [57]. Conserved domains of the BnCalS proteins were investigated with the Batch CDD tool (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) with default parameters used and displayed by TBtools.

RNA Purification and Quantitative RT-PCR Analysis
Total RNA from different tissues or cotyledons with different treatments was purified with PureLink™ Plant RNA Reagent (Invitrogen, Carlsbad, CA, USA). The purified RNA was treated with the TURBO DNA-free™ Kit (Invitrogen) for 30 minutes to degrade genomic DNA. Following the manufacturer's instructions, the first-strand cDNA was synthesized using 1 µg of the total RNA with the 1st cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA). For the qPCR assay, cDNA was diluted 1:100 with ddH 2 O, and 4.2 µL of cDNA plus 5 µL of SYBR Green I Master Mix (Clontech, Palo Alto, CA, USA) and 0.4 µL of each primer (10 mM) were used for the PCR reaction. PCRs were performed using a CFX96 Real Time Instrument (Bio-Rad, Hercules, CA, USA), and the amplification programs were set at the following conditions: 95 • C for 1 min, 40 cycles of 95 • C for 10 s, 60 • C for 30 s, followed by 95 • C for 10 min. Melting curve analysis was conducted by increasing 0.5 • C at 5 s/step from 65 to 95 • C to estimate the specificity of product. The relative gene expression levels of the target genes were calculated using the 2 −∆∆Ct method with the B. napus Actin gene (AF111812.1) as a reference. All the primers used for qPCR were deposited in Table S2.

Transcriptome Analysis of BnCalS Genes Based on Published RNA-Seq Data
To investigate the expression patterns of BnCalS genes in the AvrLepR1-LepR1 interaction system, Sequence Read Archive (SRA) files were downloaded from Gene Expression Omnibus (GEO) RNA-seq datasets (GSE77723). All reads were cleaned using trimmomatic 0.36 [58] and aligned to the transcriptome of B. napus 'Darmor-bzh' with Bowtie2 [59]. Gene expression quantification was performed using a software package called RSEM [60]. The transcripts per kilobase of exon model per million mapped reads (TPM) values were extracted from the expression quantification data, after which heatmaps were constructed from relative TPM values using HemI software (1.0) [61].