The cinnamyl alcohol dehydrogenase gene family is involved in the response to Fusarium oxysporum in resistant and susceptible flax genotypes генотипов льна на заражение Fusarium oxysporum

Flax ( Linum usitatissimum L.) is used for the production of textile, oils, pharmaceuticals, and composite materials. Fusarium wilt, caused by the fungus Fusarium oxysporum f. sp. lini , is a very harmful disease that reduces flax production. Flax cultivars that are resistant to Fusarium wilt have been developed, and the genes that are involved in the host response to F. oxysporum have been identified. However, the mechanisms underlying resistance to this pathogen remain unclear. In the present study, we used transcriptome sequencing data obtained from susceptible and resistant flax genotypes grown under control conditions or F. oxysporum infection. Approximately 250 million reads, generated with an Illumina NextSeq instrument, were analyzed. After filtering to exclude the F. oxysporum transcriptome, the remaining reads were mapped to the L. usitatissimum genome and quantified. Then, the expression levels of cinnamyl alcohol dehydrogenase ( CAD ) family genes, which are known to be involved in the response to F. oxysporum , were evaluated in resistant and susceptible flax genotypes. Expression alterations in response to the pathogen were detected for all 13 examined CAD genes. The most significant differences in expression between control and infected plants were observed for CAD1B , CAD4A , CAD5A , and CAD5B , with strong upregulation of CAD1B , CAD5A , and CAD5B and strong downregulation of CAD4A . When plants were grown under the same conditions, the expression levels were similar in all studied flax genotypes for most CAD genes, and statistically significant differences in expression between resistant and susceptible genotypes were only observed for CAD1A . Our study indicates the strong involvement of CAD genes in flax response to F. oxysporum but brings no evidence of their role as resistance gene candidates. These findings contribute to the understanding of the mechanisms underlying the response of flax to F. oxysporum infection and the role of CAD genes in stress resistance.


Introduction
Flax (Linum usitatissimum L.) is an agricultural crop with numerous uses. High-quality fiber can be obtained from flax stems and is used for the production of textile and fiber-based materials for the healthcare, military, aerospace, and electronics industries (Costa et al., 2018). Flaxseed is also used in the production of pharmaceuticals, functional foods, and other products for human consumption, while linseed is used in paints, varnishes, and animal feed (Singh et al., 2011;Goyal et al., 2014). Fusarium oxysporum f. sp. lini is a harmful pathogen that reduces flax production and quality (Rashid, 2003). Flax genotypes showing resistance to Fusarium wilt have been identified, and cultivars with improved resistance have been bred (Diederichsen et al., 2008;Rozhmina et al., 2017). However, the molecular mechanisms underlying resistance to Fusarium wilt remain unclear, and the search for genes involved in the response to F. oxysporum is an area of active research. The involvement of pathogenesisrelated proteins in the response to F. oxysporum infection was demonstrated (Wrobel-Kwiatkowska et al., 2004;Wojtasik et al., 2014;Galindo-Gonzalez, Deyholos, 2016), and the roles of antioxidants, polyamines, and phenolic compounds in response to the pathogen were shown (Lorenc-Kukula et al., 2007Boba et al., 2011Boba et al., , 2016Zeitoun et al., 2014;Wojtasik et al., 2015). Moreover, F. oxysporum-infected flax plants show cell wall rearrangements (Wojtasik et al., 2015Boba et al., 2016).
Cinnamyl-alcohol dehydrogenases (CADs) are involved in the biosynthesis of lignin, which can function as a barrier against pathogens, and the role of CAD genes in the response to F. oxysporum was previously shown (Wrobel-Kwiatkowska et al., 2007;Preisner et al., 2014Preisner et al., , 2018. Plants with downregulated CAD showed reduced (Wrobel-Kwiatkowska et al., 2007) or slightly decreased  resistance to F. oxysporum. Sixteen CAD genes were identified in L. usitatissimum and their roles in plant growth and stress responses were examined in the Nike cultivar, which is relatively resistant to Fusarium infection (Preisner et al., 2018). In the present study, we evaluated the expression of CAD genes in resistant and susceptible flax genotypes under control conditions and F. oxysporum infection to determine the general trends in response to the pathogen and genotype-specific alterations in expression.

Materials and methods
Two F. oxysporum-susceptible flax cultivars (TOST and AP5), two resistant cultivars (3896 and Dakota), and two resistant BC 2 F 5 populations (3896 × АР5, recurrent parent АР5, and Dakota × АР5, recurrent parent АР5) were used in the present study. Seeds were obtained from the Institute for Flax (Torzhok, Russia) and sterilized, first in 70 % ethanol for 1 min and then in 1 % sodium hypochlorite for 20 min. The plants were grown in 15 ml glass tubes on Murashige -Skoog medium in a growth chamber at 22 °C under a 16/8 h day/night cycle for seven days. Then, half of the plants were inoculated with F. oxysporum (pathogenic isolate #39 from the phytopathogen collection of the Institute for Flax); the remaining uninoculated plants were used as controls. Forty-eight hours later, the root tips were collected and frozen in liquid nitrogen. In total, the material was obtained from 240 plants.
Total RNA was extracted from the pooled plants (10-12 plants each) using the RNeasy Plant Mini Kit (Qiagen, USA). We obtained 24 RNA samples from the TOST, AP5, 3896, Dakota, 3896 × АР5, and Dakota × АР5 plants under control and infection in duplicate. The quality and concentration of the isolated RNA were evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA) and Qubit 2.0 fluorometer (Life Technologies, USA). The TruSeq Stranded Total RNA Sample Prep Kit (Illumina, USA) was used to prepare the cDNA library. The libraries were sequenced on a Next-Seq500 sequencer (Illumina) with 80-nucleotide paired-end reads (SRP119227, Sequence Read Archive).
Reads were trimmed using Trimmomatic (Bolger et al., 2014) and filtered against the F. oxysporum reference genome as described in our previous study (Dmitriev et al., 2017). Then, the remaining reads were mapped to the L. usitatissimum genome (GenBank assembly: GCA_000224295.2) using STAR (Dobin et al., 2013) and quantified using BEDTools (Quinlan, Hall, 2010). The genome sites where the CAD family genes (CAD1A, CAD1B, CAD2A, CAD2B, CAD3A, CAD3B, CAD4A, CAD4B, CAD5A, CAD5B, CAD6, CAD7, and CAD8) are localized were identified in the latest L. usitatissimum genome assembly (GenBank assembly: GCA_000224295.2) using the data of Preisner et al. (2018). Namely, the sequences of the scaffold regions of the old L. usitatissimum genome assembly (GenBank: AFSQ00000000.1) that encode CAD tran scripts were mapped to the latest L. usitatissimum ge-nome assembly (GCA_000224295.2) and new coordinates were identified for the CAD transcripts. The counts per million (CPM) values were determined for 13 CAD genes in each cultivar and population under both control conditions and Fusarium infection, and then the log(CPM Fusarium/ CPM control) values were calculated for each cultivar and BC 2 F 5 population. This was performed using the equipment of the Genome Center of Engelhardt Institute of Molecular Biology (http://www.eimb.ru/rus/ckp/ccu_genome_c.php).

Results
In our previous study on the response of L. usitatissimum to F. oxysporum, we used RNA-Seq data for de novo transcriptome assembly and annotation and then quantified the expression levels of the identified transcripts (Dmitriev et al., 2017). Unfortunately, we had failed to identify a significant number of CAD family genes in our de novo transcriptome assembly. In the present study, we used an improved assembly of the L. usitatissimum genome (GCA_000224295.2), in which the scaffolds were mapped to specific chromosomes, as a reference for RNA-Seq read mapping, and CAD gene sequence data (Preisner et al., 2018) were used to identify the genome sites in which the CAD genes are located. This approach enabled us to evaluate the expression levels of all presently identified CAD genes in flax. Unique chromosome locations were determined for CAD1A, CAD1B, CAD2A, CAD2B, CAD3A, CAD3B, CAD4A, CAD4B, CAD5A, CAD5B, CAD6, and CAD7, while CAD8A, CAD8B, CAD8C, and CAD8D were mapped to the same region of chromosome Lu7. Therefore, we performed an expression analysis of CAD1A, CAD1B, CAD2A, CAD2B, CAD3A, CAD3B, CAD4A, CAD4B, CAD5A, CAD5B, CAD6, CAD7, and CAD8 genes.
The expression levels of the 13 CAD family genes in six flax cultivars and populations grown under control conditions or inoculated with a pathogenic isolate of F. oxysporum were evaluated based on RNA-Seq data. The results are shown in the Figure, which presents the CPM values for each studied gene in resistant (3896, Dakota, 3896 × АР5, and Dakota × АР5) and susceptible (TOST and AP5) genotypes under control con ditions and 48 h after inoculation with F. oxysporum in biological replicates. Under both conditions, the expression levels of CAD5A and CAD6 were the highest, and the expression levels of CAD2A, CAD3A, CAD3B, and CAD8 were the lowest. Compared to control conditions, statistically significant alterations in expression under Fusarium infection were observed for all 13 genes ( p < 0.01 for all genes except CAD2B, which was p < 0.05, Mann-Whitney test). The most significant differences in expression between the control and infected plants were observed for CAD1B, CAD4A, CAD5A, and CAD5B; CAD1B, CAD5A, and CAD5B showed strong upregulation under F. oxysporum infection, and CAD4A showed strong downregulation. Under the same conditions, the expression levels were similar in all studied flax genotypes for most CAD genes; however, some exceptions were observed. For example, the expression levels of CAD3A, CAD3B, CAD4A, and CAD4B in the infected plants of the 3896 × AP5 population were higher than those in other infected genotypes.
To identify genotype-specific expression alterations in response to F. oxysporum infection, the log(CPM Fusarium/ CPM control) values were calculated (see the Table). In all studied genotypes, the most significant (more than 3-fold in average) increases in expression were observed for CAD1B, CAD5A, and CAD5B, while the most significant decreases in expression were observed for CAD4A. For CAD2A, CAD3A, CAD3B, CAD4B, and CAD6, downregulation was observed in the majority of genotypes after F. oxysporum infection. For CAD2B, CAD7, and CAD8, the decrease in expression was only slight, and for some genotypes, no decrease was observed. Statistically significant differences in expression between resistant (3896, Dakota, 3896 × АР5, and Dakota × АР5) and susceptible (TOST and AP5) genotypes were only observed for the CAD1A gene ( p < 0.05, Mann-Whitney test).  Figure).

Discussion
Recent progress in molecular analysis has provided novel opportunities for plant studies (Kage et al., 2015;Poland, 2015). Using high-throughput sequencing, genome and transcriptome sequences for many plant species can be obtained in a short time period and used for further research (Varshney et al., 2009;He et al., 2014). The draft genome sequence for flax was obtained in 2012 (Wang et al., 2012), and in 2018, the genome assembly was improved and chromosome-scale pseudomolecules were obtained using BioNano genome optical mapping (You et al., 2018). Flax transcriptomes and small RNAs obtained from different tissues, development stages, and under biotic and abiotic stresses have been sequenced. Such data has enabled the identification of genes and miRNAs that are expressed in particular organs and in the definite time and could play key roles in plant development, as well as the discovery of genes and miRNAs with altered expression under unfavorable conditions that are likely involved in the response to stress (Yu et al., 2014(Yu et al., , 2016Melnikova et al., 2015Melnikova et al., , 2016Dmitriev et al., 2016Dmitriev et al., , 2017Dmitriev et al., , 2019Galindo-Gonzalez, Deyholos, 2016;Dash et al., 2017;Gorshkova et al., 2018;Zyablitsin et al., 2018;Gorshkov et al., 2019;Krasnov et al., 2019;Wu et al., 2019).
In the present study, we used RNA-Seq data from F. oxysporum-resistant (3896, Dakota, 3896 × АР5, and Dakota × АР5) and -susceptible (TOST and AP5) flax genotypes grown under control conditions or 48 h after inoculation with F. oxysporum to evaluate the expression of CAD1A, CAD1B, CAD2A, CAD2B, CAD3A, CAD3B, CAD4A, CAD4B, CAD5A, CAD5B, CAD6, CAD7, and CAD8 and identified the CAD genes that were involved in the response to the pathogen. The genes showed different expression changes after F. oxysporum infection: in most genotypes, CAD1A, CAD1B, CAD5A, and CAD5B were upregulated, while CAD2A, CAD2B, CAD3A, CAD3B, CAD4A, CAD4B, CAD6, CAD7, and CAD8 were downregulated. In the study by Preisner et al. (2018), decreased expression was observed for most CAD genes at 24 and 96 h after F. oxysporum infection in the flax cultivar Nike. In the present study, the greatest expression changes in infected plants were observed for CAD1B, CAD4A, CAD5A, and CAD5B; three of these genes were upregulated, while CAD4A was downregulated. Our data for CAD1B, CAD4A, CAD5A, and CAD5B are consistent with the results of the previous study at 24 h after infection (Preisner et al., 2018). However, the changes observed by us were more significant, with a 3-5-fold change for CAD1B, a 2-5-fold change for CAD4A, a 4-13-fold change for CAD5A, and a 4-12-fold change for CAD5B. Based on our results, we suggest that CAD1B, CAD4A, CAD5A, and CAD5B are the most involved in the response of flax to F. oxysporum.
Searching for genes with diverse expression alterations in resistant and susceptible genotypes under stress conditions is important for the identification of resistance genes. In our study, statistically significant differences in expression between resistant and susceptible genotypes in response to the pathogen were observed only for CAD1A. Therefore, this gene could be involved in resistance to F. oxysporum. However, the changes were not pronounced. Thus, further investigations are necessary.

Conclusion
We performed expression analysis of CAD family genes after F. oxysporum inoculation based on RNA-Seq data and identified genes with significant up-and down-regulation after pathogen infection. The results of the present study indicate the involvement of CAD genes in response to Fusarium infection, but their role as resistance genes in the studied cultivars and populations is questionable. Our data also contribute to the understanding of the role of CAD genes in stress response and resistance.