Genome-wide Identification, Classification and Expression Analysis of the Mildew Resistance Locus O ( MLO ) Gene Family in Sweet Orange ( Citrus sinensis )

ABSTRACT

Mildew resistance Locus O (MLO), a gene family specific to plants, plays significant roles in the resistance to powdery mildew (PM) and response to a variety of abiotic stresses, plant growth and development. Despite their importance as barley, rice, wheat, few studies are reported in dicots except Arabidopsis; no global analysis has been performed in the burgeoning model fruit plant sweet orange (Citrus sinensis). The recent release of the genome sequences of C. sinensis provides an opportunity to conduct a comprehensive overview the evolution and features of the MLO gene family in sweet orange. In this study, amount to 14 members of the Citrus sinensis MLO gene (CisMLO) family according to their gene structures, conserved motifs, and similitude among their presumptive Arabidopsis and rice orthologs were identified in silico. Based on these analyses, all CisMLOs were grouped into six clades and expanded partly due to one tandem duplication and two segmental duplication events. Survey of their chromosomal distributions uncovered that 14 CisMLOs are localized across 6 chromosomes. Multiple-sequence alignments showed that 11 of them shared seven highly conserved transmembrane domains (TMs), while all of the sweet orange MLO proteins except CisMLO4/14 had a calmodulin-binding domain for MLO function. Expression analysis demonstrated that the MLO gene family has a diverse tissue-specific expression profiles in the sweet orange development and plays potential critical roles in stress responses. These findings will facilitate further studies of evolutionary pattern and biological functions of MLO genes in sweet orange.

Key words: Citrus sinensis; MLO gene family; powdery mildew; abiotic stress

INTRODUCTION

Sweet orange is an important fruit tree widely cultivated in China. However, powdery mildew (PM) pathogen Acrosporium tingitaninum Carter is a major fungal threat to citrus production in areas of south and southwest China [¹]. Agricultural average productivity losses caused by PM have been estimated that up to 10-30% for citrus [²]. Chemical prevention of PM has become the majority of fungicides traditionally utilized in the agricultural environment [³]. Consequently, the development of new citrus cultivars with enhanced genetic resistance to PM would be widely conceived as a valuable tactic to minimize agricultural costs and to deal with public interest in human health and global environment [⁴].

The plant-specific Mildew resistance Locus O (MLO) gene family comprising several transmembrane domains (TMs), topologically redolent of G-protein coupled receptors in metazoan [⁵]. Particular homologs of the MLO family serve as perceptivity roles towards the PM fungi. Furthermore, their deactivation, through silencing or loss-of-function mutations, has been associated with a specific form of PM resistance, interpreted as mlo resistance [⁶]. This is connected with the increasement of exocytosis defense pathways at plant-pathogen interaction sites, which are perceived to result in the prevention of fungal penetration into host plants [⁷]. Originally discovered in Hordeum vulgare (barley), mlo immunity has also been found to exist in several plant species later, particularly Arabidopsis [⁸], tomato [^9,10]), pea [¹¹], pepper [¹²] and bread wheat [¹³]. This consequently led to the breeding protocol founded upon the methodical inactivation of MLO susceptibility genes across planted varieties infected by the PM fungi [^6,9,14]. Proof of formalization of this approach has been recently conducted by the experiment, showing the introduction of PM resistance in Triticum aestivum (bread wheat) following targeted mutagenesis of 3 MLO homoeoalleles by the CRISPR-Cas9 and TALEN system [¹³]. Similarly, knockdown through RNA interference of MdMLO19 and VvMLO7 in combination with VvMLO6 and VvMLO11 decreased PM severity in apple [¹⁵] and grapevine [¹⁶]. Rather than the majority of hereditary sources of PM resistance, experimental results clearly show that mlo resistance is not limited to special fungal strains and is supremely durable. For instance, loss-of-function mutations of barley MLO endow resistance to PM fungus Blumeria graminis f. sp. hordei, and was successfully applied in barley breeding nearly 40 years ago [¹⁷]. Similarly, knockdown through RNA interference of PhMLO1 in petunia increases resistance to PM [¹⁸]. Furthermore, PM resistance of pea er1 initialing from the loss of function of PsMLO1 was first reported since 1948 and is the sole resistance source internationally used for breeding objectives [^11,19].

Interestingly, when displayed in the evolutionary trees of MLO protein family, all MLO isoforms from dicots experimentally shown to be necessary for PM immunity group in the clade V in the scientific documents [^8-10,20-22]. Furthermore, clade IV harbors all featured monocot PM perceptivity proteins [²³]. These show that phylogenetic studies on MLO proteins may predict factor candidates for PM perceptivity. Previous research has also identified a calmodulin-binding domain (CaMBD) and two conserved regions (I and II) characteristic of barley MLO orthologs that modulate PM infection in the C-terminus of MLO proteins [^24,25]. Peptide domain I is situated at about 15 to 20 residues downstream of the CaMBD and is featured by conserved residues of threonine and serine. Peptide domain II is situated at the distal end of the C-terminus and harbors the consensus tetra-peptide sequence D/E-F-S/T-F.

Apart from the distinguished function of MLOs in response to pathogen defense, MLO proteins are involved in a variety of processes of plant development. Such as, Arabidopsis AtMLO4/11 are together functioned in root architecture regulation [²⁶], and AtMLO7 takes part in pollen tube reception at the female gametophyte in Arabidopsis [²⁷]. In rice, OsMLO12 is involved in pollen hydration through interaction with calmodulin in the cytosol [²⁸]. However, SlMLO1 mutant showed no abnormal phenotype in tomato [¹⁰]. Thus, further studies are needed to characterize the functions of the MLO homologs.

MLO genes are only identified in plants and moss. Currently, a list of MLO isoforms variable between 14 and 21 has been identified in most diploid species with completed genome sequences, such as, Arabidopsis (15 genes) ([^5,26]), tomato (17 genes) [²⁹] grapevine (17 genes) ([^21,22]), peach (16 genes) [³⁰], strawberry (17 genes) [³¹], apple (21 genes) [³¹] and cucumber (14 genes) [^32,33]. However, the largest MLO protein members (39 genes) were found so far in soybean [^34,35] recently. To date, MLO like gene in sweet orange has not been reported, a comprehensive analysis of the whole MLO family in this species is necessary.

The newly available sweet orange genome from the Citrus sinensis Annotation Project (CAP; http://citrus.hzau.edu.cn/orange/) [³⁶] was exploited to provide a broad overview of the member number, structural characteristics and phylogenetic relationship of the MLO gene family. In this study, we attempt to identify the members of the MLO-like gene family in the C. sinensis genome. Moreover, structural characteristics of exon/intron, phylogenetic relationship, and distribution on chromosomes of these genes were characterized and analyzed, and these have been used to perform a comprehensive study to assess the possible functions of MLO genes in growth and development process of sweet orange. The results provide a solid foundation for functional genomic characterizations of the MLO gene family of sweet orange in the future.

MATERIAL AND METHODS

Retrieve MLO genes in sweet orange

The sequences of 15 Arabidopsis [^26,37] and 12 rice [³⁸] MLO proteins were extracted from the TIGR database (http://rice.plantbiology.msu.edu) and the TAIR database (https://www.arabidopsis.org/browse/genefamily/mlo.jsp). To identify all the MLO genes in sweet orange, BLASTP searches were conducted in the sweet orange genome database (http://citrus.hzau.edu.cn/cgi-bin/orange/blast) [³⁶], and NCBI database (http://www.ncbi.nlm.nih.gov/) with the Arabidopsis and rice MLO proteins as queries, using default parameters. We selected sequence as a candidate CisMLO protein if it satisfied with E<10^-10. And then, the full-length cDNA sequences of CisMLO genes were extracted from the sweet orange genome database. Due to the variation in MLO sequences, we additionally retrieved the sweet orange sequencing database to get more detailed information of the potential CisMLO genes using the keyword “MLO”. Finally, all putative non-overlapping MLO protein sequences were submitted to the SMART (http://smart.embl-heidelberg.de/) and Pfam database (http://pfam.sanger.ac.uk/) to verify the existence of the MLO domain (PF03094), respectively. Information about full-length coding and the amino acid sequence was also received by BLAST program for each MLO gene from CAP. NCBI ORF (open reading frame) finder (https://www.ncbi.nlm.nih.gov/orffinder) was applied to uncover the presumptive ORFs. The exon/intron gene organization of the 14 CisMLO genes were determined by aligning the cDNA sequences against their corresponding gDNA sequences based on Gene Structure Display Server (GSDS) 2.0 (http://gsds.cbi.pku.edu.cn/) [³⁹].

Sequence alignment and phylogenetic analysis of the MLO genes

Selected sequences were aligned using program ClustalX 2.1 [⁴⁰]. The phylogenetic tree was constructed using Molecular Evolutionary Genetics Analysis version (MEGA) Version 6.06 software [⁴¹] by employing the neighbor-joining (NJ) method with 1000 replicates for statistical reliability. All the other parameters were taken to the default settings.

TMs, chromosomal locations and gene duplication analysis of CisMLO genes

The presence and number of TMs were predicted using the online software TOPCONS (http://topcons.cbr.su.se) [⁴²]. Molecular weight and the isoelectric point of the deduced polypeptides were calculated using ExPasy (http://web.expasy.org/compute_pi). The subcellular location was predicted with the WoLF PSORT programs (http://www.genscript.com/wolf-psort.html) and CELLO v.2.5 (http://cello.life.nctu.edu.tw) [⁴³]. To identify the locations of MLO genes on sweet orange chromosomes, the positions of these genes were retrieved from the sweet orange genome data downloaded from the CAP database (http://citrus.hzau.edu.cn/orange/) [⁴⁴]. The 14 sweet orange MLO genes were mapped to the chromosomes using MapChart [⁴⁵]. Moreover, the gene duplication events of CisMLOs were also defined according to the criteria: (i) the alignment was covered >70% of the longer gene; (ii) the aligned region had an identity >70%; (iii) only one duplication event was counted for tightly linked genes. A block of duplications was defined if more than one gene was involved in the duplication [⁴⁶]. The data of non-synonymous substitutions per non-synonymous site (Ka) and synonymous substitution per synonymous site (Ks) of duplicated genes were analyzed using DnaSP 5.10 [⁴⁷]. The ratio of Ka/Ks between paralogs was calculated to define the pattern of selection. To trace the dates of duplication events, the Ks value was translated into duplication time in million years ago using a synonymous mutation rate of λ substitutions per synonymous site per year. The duplication time (T) were calculated by the equation T = Ks/2λ × 10^-6, for sweet orange, the λ = 2.53 × 10^-9 substitutions/synonymous site/year [⁴⁸].

Protein motifs and structure analysis

Analysis for conserved functional motifs in the MLO proteins was performed using MEME Suite Version 4.10.2 (http://meme-suite.org/tools/meme) [⁴⁹]. The settings were: any number of repetitions of a single motif, the optimum width of a motif with 6 to 25 amino acids, and the maximum number of motifs up to 35 amino acids. The program of MAST was used to retrieve conserved motifs in protein databases. The details of amino acid sequence logo of 15 most frequent motifs were illustrated in Supplementary Fig. S1.

Identification and Analysis of the Promoter Regions of CisMLOS

To explore probable stress-related cis-elements in promoter sequences of the 14 CisMLO genes, 1500-base-pair of gDNA sequence upstream of the transcriptional start site was extracted from the C. sinensis genome database (http://citrus.hzau.edu.cn/cgi-bin/orange/) [³⁶]. The PlantCARE website (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) [^50,51] program was applied to analyze the CisMLO gene promoter sequence and identify the cis-elements. Regulatory elements existed in the database and proven to participate in the regulation of gene expression during the course of pathogen infection and under abiotic stress conditions were chosen for further investigation.

Expression Analysis of sweet orange MLO Genes

To investigate the expression profiles of all CisMLOs in four distinct tissues, i.e., callus, flower, leaf and fruit, the normalized RNA-Seq RPKM (reads per kilobase exon model per million mapped reads) values of all CisMLOs were downloaded from the online database of the CAP [⁴⁴], and imagined by the heat maps with transformed log₂ values using MultiExperiment Viewer (MeV) Version 4.9.0 software (http://www.tm4.org/mev.html) [⁵²].

RESULTS

Identification, chromosomal locations and evolutionary analysis of the CisMLO genes

To verify all potential genes of the MLO family in the sweet orange genome, we used BLASTP and the keywords “MLO” to search the CAP (http://citrus.hzau.edu.cn/orange/) [⁴⁴] with the 15 A. thaliana and 12 rice MLO proteins as a retrieve. Adopt this method, 14 potential MLO genes, named CisMLO1 to CisMLO14, were identified. The information of MLO family genes in sweet orange, including locus name, gene localization, amino acid length, Arabidopsis and rice orthologs, are listed in Table 1. The CisMLO proteins are rich in leucine, which consists of 9.3-13.5% of the total number of amino acid residues, a similar result was reported in 2012 [³⁵]. It has been documented that proteins rich in leucine residues responsible for the mechanisms of defense response [^53,54].

The genomic locations of CisMLOs were illuminated on chromosomes of sweet orange available at CAP [⁴⁴]. CisMLOs were widely scattered among all the chromosomes except chromosomes 4, 6, and 9. The spread of the genes was irregular: chromosome 3 harbors four MLOs, chromosome 1 has three MLOs, two genes on chromosome 5, 7 and 8, and one gene is localized to chromosome 2. Interestingly, most MLO-like genes are distributed among the genome, but in four cases two genes are located close together, i.e., the pairs CisMLO2/3, CisMLO7/8, CisMLO9/10, and CisMLO13/14 (Fig. 1). With the exception of CisMLO9/10, the genes in other three cases of tight linkage, grouped in different clades, were not tandem repeats (Fig. 1).

To identify the evolutionary relationships among candidate CisMLO proteins and cluster them within the established clades, multiple sequence alignments and phylogenetic analysis of these CisMLOs as well as MLOs of Arabidopsis and rice were further conducted (Fig. 2 A; Supplementary Table S1). 14 CisMLOs were clustered into six clades in light of the classification standards of MLO proteins in Arabidopsis [⁵]. In detail, CisMLO1, 2, and 6, which were clustered with three AtMLOs and two OsMLOs, belonged to clade I (Fig. 2 A). Clade II, the largest grouping of sweet orange MLO proteins, was comprised of CisMLO5, 9, 10, and 11, clustered with three Arabidopsis and six rice MLO-like proteins (Fig. 2 B). The clade III included CisMLO3 and 8. CisMLO7 belonged to clade IV, which was formerly reported to contain only monocot homologs acting as PM susceptibility factors [⁵], similar to that observed in the dicot grapevine (VvMLO14) [²¹] and soybean [³⁴]. CisMLO4, 12, and 13 fell into clade V along with Arabidopsis MLO homologs formerly linked to PM responses. CisMLO14 and AtMLO3 belonged to clade VI [²¹] (Fig. 2 A). By phylogenetic analysis, the MLO genes from sweet orange, Arabidopsis thaliana, and rice divided into six groups, indicating that these genes may have evolved either before (clades I-IV) or after (clades V and VI) the divergence of monocots-dicots.

Analysis for the TMs and conserved motif structures in the CisMLO genes

In addition, the TMs of all CisMLOs were investigated. Supplementary Fig. S2 showed the alignment of sweet orange MLO proteins. Results demonstrated that all MLO proteins except CisMLO6/10/11 had seven TMs and were highly conservative, which are considered to be remarkable features of MLO gene family [⁵]. Previous research has also identified a calmodulin-binding domain (CaMBD) and two conserved regions (I and II) that modulate PM infection in the C-terminus of MLO proteins [^24,25]. Peptide domain I is situated at about 15 to 20 residues downstream of the CaMBD and is featured by conserved residues of threonine and serine. Peptide domain II is situated at the distal end of the C-terminus and harbors the consensus tetra-peptide sequence D/E-F-S/T-F. As regard to this tetra-peptide sequence, CisMLO3/7 contain D-F-S-F, which is same as HvMLO and AtMLO2/6, while CisMLO12/13 contain different sequences (E-F-S-F). All clade V of sweet orange MLO proteins except CisMLO4 had these two peptide domains, which had potential conservative functions (Supplementary Fig. S2).

Moreover, the exon/intron structures of all CisMLOs and the conserved motifs of their coding proteins were investigated. Various gene structures were analyzed by comparing the predicted CDS with the genomic sequence of CisMLO genes in the CAP database (Fig. 3 B). Conserved motifs of sweet orange MLO proteins were further analyzed with MEME version 4.10.2 online tools (http://meme-suite.org/tools/meme) [⁴⁹]. The results showed 15 conserved motifs in sweet orange MLO proteins, and their lengths were 15 (motif 14), 21 (4, 5, 12, 13 and 15) and 25 (1-3, 6-11) amino acids (Fig. 3 C). The distributions of these conserved motifs in the CisMLOs were further analyzed. As demonstrated in Supplementary Fig. S2, five CisMLOs, i.e., CisMLO3/7/8/12/13, contained all 15 conserved motifs. CisMLO6/9/10/11/14 were the genes with the most deletions, containing only 12 conserved motifs. 8 out of 15 conserved motifs (motifs 1, 2, 4-6, 8, 13 and 14) were situated within the MLO domain (Fig. 3 C) for all members of the sweet orange MLO family.

Duplications of CisMLO genes

Next, to estimate the dates of the duplication events, we used the DnaSP 5.10 [⁴⁷] to calculate Ks and Ka. The estimated dates of duplication events were measured using Ks. The results of analysis between tandem and segmental duplication events are showed in Table 2. The segmental duplications of CisMLO3/8 and CisMLO10/11 (Fig. 1) in sweet orange originated from 142.29 million years ago (Mya) (Ks = 0.72) to 158.10 Mya (Ks = 0.80), with the mean being 150.20 Mya (Ks = 0.76). The Ks of tandem duplication of the CisMLO9/10 (Fig. 1) was 0.55, dating the tandem duplication event at 108.70 Mya. The above divergence time estimations indicated that all the duplications of MLO genes in sweet orange occurred after the monocots/dicots split (approximately 200 Mya) [⁵⁵]. The Ka/Ks ratio of one tandem duplication pair and two segmental duplication pairs were all more than 0.3, showing that palpable functional tributary might have happened after duplication blocks. All these data indicate that the sweet orange MLO family tandem/segmental expansion can be traced back to recent duplication blocks. Furthermore, these MLO genes maintained their function after duplication. It should arouse our interest to discover whether the few sequence differences result in a functional variety, or these genes are originally redundant in function.

In silico stress-responsive cis -element analysis of CisMLO genes

To discover potential stress-related cis-elements in the promoter regions of the CisMLO genes, 1.5 kb upstream promoter sequences of the CisMLO genes were submitted to program PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). Various biotic stress-responsive regulatory elements, for example, pathogen-related cis-elements TC/AT-rich repeat and W box/Box-W1 elements [⁵⁶] (Fig. 4; Supplementary Table S2), were observed. Many abiotic stress mediated cis-elements were also identified, containing wound response element (WUN motif), drought stress-related MYB binding site (MBS), heat shock element (HSE), and low temperature-responsive element (LTR) (Fig. 4). Moreover, many elements probably involved in reaction to hormones. For instance, auxin (TGA-element and AuxRE), abscisic acid (ABA) (ABRE), Ethylene (ET) (ERE), methyl jasmonate (MeJA) (CGTCA and TGACG motifs), gibberellin (GARE motif and TATC box) and salicylic acid (SA) (TCA-element), were detected as well (Fig. 4). All CisMLOs had at least one stress or hormone-responsive cis-element, in addition, 3 of the 14 CisMLO genes (CisMLO4/5/11) contained four different hormone-responsive elements each. Furthermore, 11 genes harbored pathogen elicitor response elements, and among the different motifs, the W box/Box-W1 motif was found in 6 genes; AT-rich repeat motif was detected in 3 genes; however, TC-rich repeat motif was present in 11 genes. Moreover, CisMLO1/6/8/11/13 had four different abiotic stress response elements each. Finally, 10 CisMLO genes contained more than 10 biotic and/or abiotic stress- or hormonal-response cis-elements, indicating these CisMLOs might variously participate in these pathways.

Analysis of the expression of CisMLO genes based on the sweet orange EST database

To predict the function of CisMLOs, RNA-seq data were downloaded from C. sinensis genome database [^36,44] and the expression patterns of CisMLOs in various tissues (callus, leaf, flower, and fruit) were analyzed and demonstrated in Fig. 5. A hierarchical classification was conducted according to the expression data of each CisMLOs in four different tissues. Based on the hierarchical classification results, we clustered genes of each clade to three assigned expression pattern groups (Fig. 5 A-C). The group A is the smallest with only three members, CisMLO10/12/13, showed prevalent higher expression level in callus as compared to other tissues. Whereas the members of group B, CisMLO4/5/7/8/9, showed no significant difference expression between tissues. The group C containing genes, CisMLO1/2/3/6/11/14, showed relatively higher expression level in flower, leaf, fruit tissues than in callus. It should be noted that CisMLO3 and CisMLO11 were highly expressed in all samples analyzed, with the highest expression in flower (Fig. 5). Most importantly, the different expression patterns of CisMLOs among the four tissues showed that these genes might participate in a broad range of biological processes in citrus, and the interesting challenge for the future is to determine the specific functions in conjunction with individual CisMLO genes in sweet orange.

DISCUSSION

We retrieved MLO-like genes in the sweet orange genome via an in silico approach and identified 14 potential CisMLOs, unevenly situated across 6 of the 9 chromosomes (Table 1; Fig. 1). By phylogenetic analysis, the MLO genes from sweet orange and other plant species were divided into six clades, some of which contained members from A. thaliana and rice (Fig. 2), indicating that these genes may have evolved after the divergence of monocots/dicots. Most CisMLO homologs seemed to be widely scattered within the sweet orange genomes (Fig. 1), illustrating segmental duplication as the predominant evolutionary pattern for the citrus MLO-like gene family. However, we also found a case of adjacent homologs (CisMLO9/10) (Fig. 1), which are consistent with the data of tandem duplication events. The results above indicate that the sweet orange MLO family tandem or segmental expansion can be traced back to recent duplication blocks. A similar conclusion was made and supported by Jiwan et al. [³⁰] and Acevedo-Garcia et al. [²⁰], respectively.

A massive amount of former researches demonstrated that phylogenetic analysis enables functional prediction of many novel genes because orthologs usually keep the same biological role during the course of evolution. For example, phylogenetic analyses of the ARF and NAC families of citrus with their corresponding orthologs with clear functions in Arabidopsis resulted in an approximately complete match between sequence homology and expression profiles or biological roles [^57,58]. And then, a phylogenetic analysis associating sweet orange, Arabidopsis and rice MLO protein family would not only help illuminate the evolutionary relationships of MLO proteins but would also grant the insight to the potential biological roles of the CisMLO proteins derived from the clustering clades currently documented in Arabidopsis and rice (Table 1; Fig. 2).

In the 14 sweet orange MLO proteins, CisMLO1/2/6 assembled with Arabidopsis AtMLO4/11/14 into clade I (Fig. 2), AtMLO4/11 referred to the regulation of root thigmomorphogenesis in Arabidopsis, however, the closely paralogous AtMLO14 gene does not participate in this plant growth process, since no obvious variant phenotype was displayed in the AtMLO14 mutant [^26,59]. CisMLO3/8 were assembled in clade III together with the five Arabidopsis proteins AtMLO5/7/8/9/10 representing the functional clade with AtMLO7 protein responsible for similar functions of pollen tube perception of the female germ cell in Arabidopsis [²⁷]. Importantly, three homologs, CisMLO4/12/13, were found to code for predicted proteins that assembled in clade V (Fig. 3). All dicot MLO homologs in clade V have experimentally proven hitherto to be necessary for PM pathogenesis, which thus provided a critical role for identification the candidate genes in sweet orange with similar functions in PM susceptibility [⁵]. Consequently, the clustering in clade V is the main selection criterion for MLO genes participated in PM pathogenesis, which provides an initial important clue for the reduction in the number of candidates for further functional experiment design. Interestingly, one homolog from sweet orange, CisMLO7, was found to group in clade IV, which includes all monocot MLO proteins modulating PM infection [⁵] (Fig. 2). Consistent with this finding, VvMLO14 in the dicot species grapevine, also assembles in clade IV [²¹]. Finally, clade VI, containing AtMLO3 and CisMLO14, which was considered as a single divergent lineage in former studies [²¹], more detailed functions of AtMLO3 and CisMLO14 will be revealed by the further investigation. These results strongly indicated the existence of sweet orange-specific MLO gene that was either lost in Arabidopsis and rice or acquired in the sweet orange lineages after an evolutionary splitting event from the most recent common ancestor. These results further indicated that functional prediction based on evolutionary analysis might be convenient for us to efficaciously choose candidate genes, which could then be given priority to the functional investigation in future.

Further study revealed that most of the CisMLOs promoter contains a variety of biotic- (W box / Box-W1, TC-rich repeat and AT-rich repeat), abiotic- (HSE, LTR, MBS, WUN motif), and hormonal- (ABRE, CGTCA/TGACG motif, AuxRE, TGA-element, ERE, TCA-element, CGTCA/TGACG motif, TATC box, GARE motif) responsive related cis-regulatory elements (Supplementary Table S2). Therefore, CisMLOs may play roles as an important cross-talk node in both stress and hormone signaling processes. The MLO genes are reported to take part in programmed cell death, and in the PM pathogen responses [^12,24], the biotrophic oomycete pathogenHyaloperonospora arabidopsidis, as well as the hemibiotrophic bacterial pathogens, for example, Pseudomonas syringae pv. tomato and Xanthomonas campestris pv. vesicatoria [^60,61]; wounding [^21,37,62]; phytohormone [^24,26,63] and abiotic stresses [^63,64]. As a rule, however, the paralogs in MLO family, as illuminated by their clustering pattern (Fig. 3 A), did not show similarities in their cis-elements (Fig. 4). For instance, the paralogous members of CisMLO3, 7 and 10 shared no pathogen-responsive elements (Fig. 4). The paralogs in CisMLO family, therefore, harbor the MLO domain, but their biological roles seem to have expanded during the course of evolution. These findings also suggest that CisMLO4/12/13 likely play roles in PM responses similar to that of their counterparts in other plant species, additional experiments are necessary to explore the accurate roles of these genes, harboring many pathogen-responsive elements, in response to biotic stress stimuli, especially in the emerging commercially important citrus.

CONCLUSION

Here, systemic investigations including chromosomal distribution, phylogenetic relationship, domain composition, gene structure, evolutionary features, expression patterns, and the cis-elements of all members of the sweet orange MLO family were conducted. Homology search using AtMLOs as query sequences verified a total of 14 CisMLO genes clustered into 6 clades and located unevenly across 6 of the 9 C. sinensis chromosomes. Comparative studies revealed that 3 paralogous pairs were amplified by tandem or segmental duplication types. In addition, the gene specific promoter cis-elements analysis indicated that these proteins might take part in various functions, such as response to biotic/abiotic stresses and developmental processes, several of them are probably related to the responses to fungal pathogens. A further in-depth discussion of the expression patterns based on the sweet orange EST database provided insights into the probable functional divergence among members of the MLO family. Although the distinct roles of sweet orange MLO genes remain largely unclear, and many novel researches are necessary to elucidate their exact functions, the findings from this study may contribute valuable information for future breeding PM resistance in citrus, and isolation and functional validation of the individual biological processes-linked CisMLO genes.

ACKNOWLEDGMENTS

This research was supported by the Doctoral Scientific Research and Development Fund of Yangtze University (801190010130 and 801190010131).

REFERENCES

• ¹ Cai M, Yi G, Peng C. Color atlas of citrus pest and disease (in Chinese). China Agriculture Press, Beijing, China, 2011.
• ² Yi L. The occurence regularity and control techniques of citrus powdery mildew (in Chinese). Southwest Hortic. 2001; 29(1): 7
• ³ Hewitt HG. Fungicides in crop protection. CAB International, Wallingford, 1998.
• ⁴ Russell PE. A century of fungicide evolution. J Agr Sci. 2005; 143(1): 11-25.
• ⁵ Devoto A, Hartmann HA, Piffanelli P, Elliott C, Simmons C, Taramino G, et al. Molecular phylogeny and evolution of the plant-specific seven-transmembrane MLO family. J Mol Evol. 2003; 56(1): 77-88.
• ⁶ Pavan S, Jacobsen E, Visser RGF, Bai Y. Loss of susceptibility as a novel breeding strategy for durable and broad-spectrum resistance. Mol Breed. 2010; 25(1): 1-12.
• ⁷ Assaad FF, Qiu J-L, Youngs H, Ehrhardt D, Zimmerli L, Kalde M, et al. The PEN1 syntaxin defines a novel cellular compartment upon fungal attack and is required for the timely assembly of papillae. Mol Biol Cell. 2004; 15(11): 5118-5129.
• ⁸ Consonni C, Humphry ME, Hartmann HA, Livaja M, Durner J, Westphal L, et al. Conserved requirement for a plant host cell protein in powdery mildew pathogenesis. Nat Genet. 2006; 38(6): 716-720.
• ⁹ Pavan S, Schiavulli A, Appiano M, Marcotrigiano A, Cillo F, Visser RF, et al. Pea powdery mildew er1 resistance is associated to loss-of-function mutations at a MLO homologous locus. Theor Appl Genet. 2011; 123(8): 1425-1431.
• ¹⁰ Bai Y, Pavan S, Zheng Z, Zappel NF, Reinstädler A, Lotti C, et al. Naturally occurring broad-spectrum powdery mildew resistance in a Central American tomato accession is caused by loss of mlo function. Mol Plant Microbe In. 2008; 21(1): 30-39.
• ¹¹ Humphry M, Reinstaedler A, Ivanov S, Bisseling T, Panstruga R. Durable broad-spectrum powdery mildew resistance in pea er1 plants is conferred by natural loss-of-function mutations in PsMLO1. Mol Plant Pathol. 2011; 12(9): 866-878.
• ¹² Zheng Z, Nonomura T, Appiano M, Pavan S, Matsuda Y, Toyoda H, et al. Loss of function in Mlo orthologs reduces susceptibility of pepper and tomato to powdery mildew disease caused by Leveillula taurica. PLoS One. 2013; 8(7): e70723.
• ¹³ Wang Y, Cheng X, Shan Q, Zhang Y, Liu J, Gao C, et al. Simultaneous editing of three homoeoalleles in hexaploid bread wheat confers heritable resistance to powdery mildew. Nat Biotechnol. 2014; 32(9): 947-951.
• ¹⁴ Dangl JL, Horvath DM, Staskawicz BJ. Pivoting the plant immune system from dissection to deployment. Science. 2013; 341(6147): 746-751.
• ¹⁵ Pessina S, Angeli D, Martens S, Visser RGF, Bai Y, Salamini F, et al. The knock-down of the expression of MdMLO19 reduces susceptibility to powdery mildew (Podosphaera leucotricha) in apple (Malus domestica). Plant Biotechnol J. 2016; in press, doi:10.1111/pbi.12562.
  » https://doi.org/10.1111/pbi.12562
• ¹⁶ Pessina S, Lenzi L, Perazzolli M, Campa M, Dalla Costa L, Urso S, et al. Knockdown of MLO genes reduces susceptibility to powdery mildew in grapevine. Hortic Res. 2016; 3: 16016.
• ¹⁷ Lyngkjær MF, Newton AC, Atzema JL, Baker SJ. The barley mlo-gene: an important powdery mildew resistance source. Agronomie. 2000; 20(7): 745-756.
• ¹⁸ Jiang P, Chen Y, Wilde HD. Reduction of MLO1 expression in petunia increases resistance to powdery mildew. Sci Hortic. 2016; 201: 225-229.
• ¹⁹ Appiano M, Pavan S, Catalano D, Zheng Z, Bracuto V, Lotti C, et al. Identification of candidate MLO powdery mildew susceptibility genes in cultivated Solanaceae and functional characterization of tobacco NtMLO1. Transgenic Res. 2015; 24(5): 847-858.
• ²⁰ Acevedo-Garcia J, Kusch S, Panstruga R. Magical mystery tour: MLO proteins in plant immunity and beyond. New Phytol. 2014; 204(2): 273-281.
• ²¹ Feechan A, Jermakow AM, Torregrosa L, Panstruga R, Dry IB. Identification of grapevine MLO gene candidates involved in susceptibility to powdery mildew. Funct Plant Biol. 2008; 35(12): 1255-1266.
• ²² Winterhagen P, Howard SF, Qiu W, Kovács LG. Transcriptional up-regulation of grapevine MLO genes in response to powdery mildew infection. Am J Enol Viticult. 2008; 59(2): 159-168.
• ²³ Reinstädler A, Müller J, Czembor J, Piffanelli P, Panstruga R. Novel induced mlo mutant alleles in combination with site-directed mutagenesis reveal functionally important domains in the heptahelical barley Mlo protein. BMC Plant Biol. 2010; 10(1): 31.
• ²⁴ Kim MC, Panstruga R, Elliott C, Muller J, Devoto A, Yoon HW, et al. Calmodulin interacts with MLO protein to regulate defence against mildew in barley. Nature. 2002; 416(6879): 447-451.
• ²⁵ Panstruga R. Discovery of novel conserved peptide domains by ortholog comparison within plant multi-protein families. Plant Mol Biol. 2005; 59(3): 485-500.
• ²⁶ Chen Z, Noir S, Kwaaitaal M, Hartmann HA, Wu M-J, Mudgil Y, et al. Two seven-transmembrane domain MILDEW RESISTANCE LOCUS O proteins cofunction in Arabidopsis root thigmomorphogenesis. Plant Cell. 2009; 21(7): 1972-1991.
• ²⁷ Kessler SA, Shimosato-Asano H, Keinath NF, Wuest SE, Ingram G, Panstruga R, et al. Conserved molecular components for pollen tube reception and fungal invasion. Science. 2010; 330(6006): 968-971.
• ²⁸ Yi J, An S, An G. OsMLO12, encoding seven transmembrane proteins, is involved with pollen hydration in rice. Plant Reprod. 2014; 27(4): 169-180.
• ²⁹ Chen Y, Wang Y, Zhang H. Genome-wide analysis of the mildew resistance locus o (MLO) gene family in tomato (Solanum lycopersicum L.). Plant Omics. 2014; 7(2): 87-93.
• ³⁰ Jiwan D, Roalson EH, Main D, Dhingra A. Antisense expression of peach mildew resistance locus O (PpMlo1) gene confers cross-species resistance to powdery mildew in Fragaria x ananassa. Transgenic Res . 2013; 22(6): 1119-1131.
• ³¹ Pessina S, Pavan S, Catalano D, Gallotta A, Visser R, Bai Y, et al. Characterization of the MLO gene family in Rosaceae and gene expression analysis in Malus domestica. BMC Genomics. 2014; 15(1): 618.
• ³² Schouten H, Krauskopf J, Visser RF, Bai Y. Identification of candidate genes required for susceptibility to powdery or downy mildew in cucumber. Euphytica. 2014; 200(3): 475-486.
• ³³ Zhou S, Jing Z, Shi J. Genome-wide identification, characterization, and expression analysis of the MLO gene family in Cucumis sativus. Genet Mol Res. 2013; 12(4): 6565-6578.
• ³⁴ Deshmukh R, Singh VK, Singh BD. Comparative phylogenetic analysis of genome-wide Mlo gene family members from Glycine max and Arabidopsis thaliana. Mol Genet Genomics. 2014; 289(3): 345-359.
• ³⁵ Singh V, Singh A, Chand R, Singh B. Genome wide analysis of disease resistance Mlo gene family in sorghum [Sorghum bicolor (L.) Moench]. J Plant Genom. 2012; 2(1): 18-27.
• ³⁶ Xu Q, Chen L-L, Ruan X, Chen D, Zhu A, Chen C, et al. The draft genome of sweet orange (Citrus sinensis). Nat Genet. 2013; 45(1): 59-66.
• ³⁷ Chen Z, Hartmann HA, Wu M-J, Friedman E, Chen J-G, Pulley M, et al. Expression analysis of the AtMLO gene family encoding plant-specific seven-transmembrane domain proteins. Plant Mol Biol . 2006; 60(4): 583-597.
• ³⁸ Liu Q, Zhu H. Molecular evolution of the MLO gene family in Oryza sativa and their functional divergence. Gene. 2008; 409(1-2): 1-10.
• ³⁹ Hu B, Jin J, Guo A-Y, Zhang H, Luo J, Gao G. GSDS 2.0: an upgraded gene feature visualization server. Bioinformatics. 2015; 31(8): 1296-1297.
• ⁴⁰ Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. Clustal W and Clustal X version 2.0. Bioinformatics . 2007; 23(21): 2947-2948.
• ⁴¹ Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol Biol Evol. 2013; 30(12): 2725-2729.
• ⁴² Tsirigos KD, Peters C, Shu N, Käll L, Elofsson A. The TOPCONS web server for consensus prediction of membrane protein topology and signal peptides. Nucleic Acids Res. 2015; 43(W1): W401-W407.
• ⁴³ Yu C-S, Chen Y-C, Lu C-H, Hwang J-K. Prediction of protein subcellular localization. Proteins. 2006; 64(3): 643-651.
• ⁴⁴ Wang J, Chen D, Lei Y, Chang J-W, Hao B-H, Xing F, et al. Citrus sinensis annotation project (CAP): a comprehensive database for sweet orange genome. PLoS One . 2014; 9(1): e87723.
• ⁴⁵ Voorrips RE. MapChart: software for the graphical presentation of linkage maps and QTLs. J Hered. 2002; 93(1): 77-78.
• ⁴⁶ Gu Z, Cavalcanti A, Chen F-C, Bouman P, Li W-H. Extent of gene duplication in the genomes of Drosophila, nematode, and yeast. Mol Biol and Evol. 2002; 19(3): 256-262.
• ⁴⁷ Librado P, Rozas J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics . 2009; 25(11): 1451-1452.
• ⁴⁸ Zhu AD. Citrus genome evolution and transcriptomic studies on postharvest fruits (PhD thesis in Chinese). Wuhan: Huazhong Agricultural University; 2012. p. 45.
• ⁴⁹ Bailey TL, Williams N, Misleh C, Li WW. MEME: discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res . 2006; 34(suppl 2): W369-W373.
• ⁵⁰ Lescot M, Déhais P, Thijs G, Marchal K, Moreau Y, Van de Peer Y, et al. PlantCARE, a database of plant cis-acting regulatory elements and a portal to tools for in silico analysis of promoter sequences. Nucleic Acids Res . 2002; 30(1): 325-327.
• ⁵¹ Rombauts S, Déhais P, Van Montagu M, Rouzé P. PlantCARE, a plant cis-acting regulatory element database. Nucleic Acids Res . 1999; 27(1): 295-296.
• ⁵² Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, Howe EA, et al. TM4 microarray software suite. Methods Enzymol. 2006; 411: 134-193.
• ⁵³ Jung EH, Jung HW, Lee SC, Han SW, Heu S, Hwang BK. Identification of a novel pathogen-induced gene encoding a leucine-rich repeat protein expressed in phloem cells of Capsicum annuum. Biochim Biophys Acta. 2004; 1676(3): 211-222.
• ⁵⁴ Torii KU. Leucine-rich repeat receptor kinases in plants: structure, function, and signal transduction pathways. Int Rev Cytol. 2004; 234: 1-46.
• ⁵⁵ Wikström N, Savolainen V, Chase MW. Evolution of the angiosperms: calibrating the family tree. Proc Biol Sci. 2001; 268(1482): 2211-2220.
• ⁵⁶ Pontier D, Balagué C, Bezombes-Marion I, Tronchet M, Deslandes L, Roby D. Identification of a novel pathogen-responsive element in the promoter of the tobacco gene HSR203J, a molecular marker of the hypersensitive response. Plant J. 2001; 26(5): 495-507.
• ⁵⁷ Li S-B, OuYang W-Z, Hou X-J, Xie L-L, Hu C-G, Zhang J-Z. Genome-wide identification, isolation and expression analysis of auxin response factor (ARF) gene family in sweet orange (Citrus sinensis). Front Plant Sci. 2015; 6: 119.
• ⁵⁸ de Oliveira T, Cidade L, Gesteira A, Coelho Filho M, Soares Filho W, Costa MC. Analysis of the NAC transcription factor gene family in citrus reveals a novel member involved in multiple abiotic stress responses. Tree Genet Genome. 2011; 7(6): 1123-1134.
• ⁵⁹ Bidzinski P, Noir S, Shahi S, Reinstädler A, Gratkowska DM, Panstruga R. Physiological characterization and genetic modifiers of aberrant root thigmomorphogenesis in mutants of Arabidopsis thaliana MILDEW LOCUS O genes. Plant Cell Environ. 2014; 37(12): 2738-2753.
• ⁶⁰ Kim DS, Choi HW, Hwang BK. Pepper mildew resistance locus O interacts with pepper calmodulin and suppresses Xanthomonas AvrBsT-triggered cell death and defense responses. Planta. 2014; 240(4): 827-839.
• ⁶¹ Kim DS, Hwang BK. The pepper MLO gene, CaMLO2, is involved in the susceptibility cell-death response and bacterial and oomycete proliferation. Plant J . 2012; 72(5): 843-855.
• ⁶² Feechan A, Jermakow AM, Dry IB. Grapevine MLO candidates required for powdery mildew pathogenicity? Plant Signal Behav. 2009; 4(6): 522-523.
• ⁶³ Shen Q, Zhao J, Du C, Xiang Y, Cao J, Qin X. Genome-scale identification of MLO domain-containing genes in soybean (Glycine max L. Merr.). Genes Genet Syst. 2012; 87(2): 89-98.
• ⁶⁴ Piffanelli P, Zhou F, Casais C, Orme J, Jarosch B, Schaffrath U, et al. The barley MLO modulator of defense and cell death is responsive to biotic and abiotic stress stimuli. Plant Physiol. 2002; 129(3): 1076-1085.

Publication Dates

• Publication in this collection
  2017

History

• Received
  05 May 2016
• Accepted
  21 July 2016