https://orcid.org/0000-0002-1665-6229
Figures
Abstract
Protein transporters move essential metabolites across membranes in all living organisms. Downy mildew causing plant pathogens are biotrophic oomycetes that transport essential nutrients from their hosts to grow. Little is known about the functions and gene expression levels of membrane transporters produced by downy mildew causing pathogens during infection of their hosts. Approximately 170–190 nonredundant transporter genes were identified in the genomes of Peronospora belbahrii, Peronospora effusa, and Peronospora tabacina, which are specialized pathogens of basil, spinach, and tobacco, respectively. The largest groups of transporter genes in each species belonged to the major facilitator superfamily, mitochondrial carriers (MC), and the drug/metabolite transporter group. Gene expression of putative Peronospora transporters was measured using RNA sequencing data at two time points following inoculation onto leaves of their hosts. There were 16 transporter genes, seven of which were MCs, expressed in each Peronospora species that were among the top 45 most highly expressed transporter genes 5–7 days after inoculation. Gene transcripts encoding the ADP/ATP translocase and the mitochondrial phosphate carrier protein were the most abundant mRNAs detected in each Peronospora species. This study found a number of Peronospora genes that are likely critical for pathogenesis and which might serve as future targets for control of these devastating plant pathogens.
Citation: Johnson ET, Lyon R, Zaitlin D, Khan AB, Jairajpuri MA (2023) A comparison of transporter gene expression in three species of Peronospora plant pathogens during host infection. PLoS ONE 18(6): e0285685. https://doi.org/10.1371/journal.pone.0285685
Editor: Hernâni Gerós, Universidade do Minho, PORTUGAL
Received: September 13, 2022; Accepted: April 28, 2023; Published: June 1, 2023
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All raw sequencing data are deposited in the National Center for Biotechnology Information under accession code PRJNA852505 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA852505). Published data sets with NCBI accession codes PRJNA603356 and PRJNA647030 were used.
Funding: Funding for this work was provided to EJ and RL using United States Department of Agriculture, Agricultural Research Service in-house project 5010-22410-017-00-D. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The genus Peronospora contains many plant pathogens that cause damage to crops and ornamentals. There were approximately 400 described Peronospora species in the late 20th century [1], but one study estimated that there could be 3,000–5,000 species of Peronospora worldwide [2]. All species of Peronospora are obligate biotrophs, which means they are completely dependent on the living tissues of their host plant for nutrition. Several of the most economically relevant Peronospora species have been studied in detail. For example, Peronospora effusa causes a serious disease of cultivated spinach. The pathogen possibly grows systemically in the plant and can be transmitted via the seed [3]. Synthetic fungicides, in combination with host resistance genes, can manage P. effusa [4–6], but this is not an option for organic spinach production. In addition, 17 races of P. effusa have been identified as of 2018. [7]. The genomes of P. effusa races 13 and 14 have been sequenced and described [7]. Another problem species is Peronospora belbahrii, which grows on sweet basil. The disease was first identified in Switzerland in 2001 [8] but can be found worldwide now, most likely due to the dissemination of infected seeds [9]. P. belbahrii causes yellowing of infected leaves, and then sporangia grow out of stomata on the undersides of leaves 7–10 days after inoculation [10,11]. Synthetic fungicides are the primary means to control downy mildew disease in basil [12–15]. Control of downy mildew disease in basil was achieved in several cases through the action of two or more genes, but the mechanism(s) of host resistance is undetermined [16–18]. The genome of a German isolate of P. belbahrii was sequenced and described [19]. Another downy mildew disease in many parts of the world is tobacco blue mold disease, caused by Peronospora tabacina, that can result in substantial financial losses [2]. P. tabacina can induce lesions on leaves, and there is additional evidence that the pathogen can grow systemically [20,21]. This tobacco pathogen is controlled by synthetic fungicides at present, but resistance in the pathogen has been documented [22,23]. Genomes from two German isolates of P. tabacina were sequenced several years ago [24].
Analyses of the five Peronospora genomes indicated that there were approximately 8,000 gene models in P. tabacina and P. effusa whereas there were slightly more than 9,000 gene models in P. belbahrii [7]. It was determined that 4,148 gene models were unique to one or the other of the two sequenced P. tabacina isolates, which differed in their sensitivity to the fungicide metalaxyl-M [7,24]. In addition, 3,095 genes were unique to P. tabacina compared to other oomycetes, although P. belbahrii was not included in this analysis [7]. The mitochondrial sequences of the two P. tabacina isolates had sequence variations, including seven single nucleotide polymorphisms, three indels, and a difference in the copy number of a repeated sequence [24]. In contrast, the genome sequences of the two P. effusa isolates were more similar to each other, with only 1,415 genes unique to one isolate or the other, and the P. effusa mitochondrial genome sequences were identical [7]. Analysis of P. effusa gene models indicated that 1,807 were unique compared to other oomycetes, although P. belbahrii was not included in the comparison [7]. Although extensive comparison of the P. belbahrii genome with other oomycete genomes was not done, hierarchical clustering analysis of metabolic networks from 11 oomycetes determined that the network of P. belbahrii clustered with oomycete networks from obligate biotrophs (including Albugo laibachii, Plasmopara halstedii, and Hyaloperonospora arabidopsidis) whereas the hemibiotrophs, primarily Phytophthora species, grouped together in a different clade [19]. In addition, the hierarchical clustering analysis indicated that the metabolic networks in obligate biotrophs are generally smaller than the metabolic networks of hemibiotrophs [19]. A similar pattern of clustering of the metabolic networks of obligate plant pathogens was found when principal component analysis of metabolic networks was performed with the genomes of 42 oomycetes, six of which were obligate biotrophs (A. laibachii, H. arabidopsidis, P. belbahrii, P. effusa, P. halstedii, and Plasmopara viticola) [25]. The losses of metabolic genes in the obligate plant pathogens were typically in the same sets of metabolic enzymes despite the fact they were classified into four different genera [25].
Obligate biotrophs cannot be cultured on artificial medium in the laboratory. Analysis of the first genome sequence of a downy mildew causing pathogen (Hyaloperonospora arabidopsidis) indicated that it lacked the ability to reduce nitrate and sulfite [26]. This same scenario was also found in one isolate of P. tabacina, and one isolate of Plasmopara halstedii, a downy mildew causing pathogen of sunflower [7]. Interestingly, one isolate of P. tabacina only lacked nitrate reductase, and not sulfite reductase; this was also true in two isolates of P. effusa [7]. These gene deficiencies strongly suggest that the host plant is a sufficient source of nitrogen and sulfate [27]. Additional analyses indicated that the genomes of P. effusa and P. tabacina have reduced numbers of genes that encode proteins for carbohydrate, calcium, flagella-motor, phytopathology, and transporter activities compared to the genomes of Phytophthora species [7]. The reduction in the numbers of genes encoding different transporters suggests that Peronospora species are limited in the host substrates that can be transported into hyphae or sporangia, but the expression of genes encoding transporters, as well as the possible substrates of the transporters, has been only documented for a select a group of transporters. The aim of this study was to identify all the transporter genes in three Peronospora species and determine which of these genes are well expressed during infection of their host. This study of Peronospora transporter gene expression could be valuable in formulating crop protection products in the future, and contributes to a better understanding of the key genes involved in oomycete nutrient acquisition.
Materials and methods
Plant materials and propagation of the P. tabacina pathogen
The pathogen P. tabacina was maintained on tobacco variety KY14 as previously described [28]. Briefly, plants were routinely inoculated with P. tabacina by spraying them with a sporangia suspension of ~100,000/ml. The inoculated plants were placed in large plastic totes, with the lids sealed, which were kept in an isolation chamber maintained at 21°C. The plants were removed from the totes and placed on shelves in the isolation chamber the next day and grown for 6 days at 21°C with a 12-hour photoperiod. Leaves showing signs of infection were harvested and sealed in a large baggie or a plastic box with moist paper towels and placed in the dark in the same chamber for 16 hours to allow the pathogen to sporulate. The sporangia were brushed off the undersides of the leaves with a camelhair brush (~1 inch width) into a flat container (Pyrex baking dish) containing highly purified deionized water. The sporangia were used directly for re-inoculation or were collected by filtration on a 5-micron membrane (Millipore-Sigma, Burlington, MA) and washed several times prior to resuspension in water and counting with a hemacytometer. The washed sporangia were used in experiments where it was important to know the concentration.
Inoculation of tobacco leaves with P. tabacina
Five nonflowering KY14 tobacco plants, approximately four months old, were inoculated in at least 30 places per plant (four-five leaves per plant) by infiltration of 50–100 μl of P. tabacina sporangia (100,000 per ml) using a 1 ml syringe. The plants were sampled at 2 and 5 days post-inoculation (DPI); only samples from diseased plants were processed. Eight tissue samples per plant were excised using a 2-cm diameter cork borer from each inoculation site. The circular disks were immediately frozen in liquid nitrogen and stored at -80°C.
RNA extraction and transcriptome sequencing of tobacco leaves
RNA was extracted from frozen leaves as previously described [29]. Briefly, leaf tissue was powdered with a mortar and pestle using liquid nitrogen. The powdered tissue was placed in a 1.5 ml centrifuge tube that contained one ml of TRIzol (Thermo Fisher Scientific, Waltham, MA). RNA was purified from the solution using the TRIzol Plus RNA Purification Kit (Thermo Fisher Scientific). The RNA was incubated with DNAse (Qiagen, Hilden, Germany) to remove any trace genomic DNA and purified using the GeneJET RNA Cleanup Kit (Thermo Fisher Scientific). The RNA samples (2 DPI, three samples, 5 DPI, four samples) were shipped to Azenta Life Sciences (South Plainfield, NJ) for transcriptome sequencing. Quality control analysis of all the samples was performed by Azenta Life Sciences using TapeStation (Agilent Technologies, Palo Alto, CA), which found that the RIN values were >4.0 and the DV200 scores (percentage of RNA fragments >200 base pairs in length) were >70% (S1 Table), which indicated that the quality and integrity of the RNA samples were sufficient for transcriptome sequencing [30].
Mapping of tobacco transcriptomes to the P. tabacina genome
The raw reads were trimmed and mapped using the "map reads to reference" tool in QIAGEN CLC Genomics Workbench version 22.0.1. Alignment of the reads to the P. tabacina genome (GenBank: GCA_002099245 [24]) was performed using default settings. In other analyses, a single gene served as the reference for read mapping of transcriptomes using the same tool with default settings.
Identification of putative transporter genes
Sequences of putative transporter proteins from the P. effusa genome, which were organized into 19 different groups [31], were downloaded from the InterPro website [32]. This set of transporter proteins is not comprehensive because it depends on high quality genome annotation. The CD-HIT suite [33] was utilized to identify highly homologous proteins in this set; a protein was removed from the set if it had 90% identity to another protein, and the longest protein (of a homologous pair or group) was retained. The gene sequences of the putative P. effusa transporters were used to find the homologous genes in P. belbahrii (using BLAST searches of the genome sequence, GCA_902712285.1 [19], on CLC Genomics Workbench) and P. tabacina (using BLAST searches on the NCBI website). The protein sequences were screened for a transporter motif using the Conserved Domain Database from NCBI and hmmscan from EMBL-EBI; only proteins that had a transporter motif score of 10−5 were retained in the set. All the motif scores are available in S2 Table.
Gene expression analysis
A gene expression value (TPM, transcripts per million) was calculated for each putative transporter gene and seven housekeeping genes (S1 File) in each transcriptome sample. Transcriptome read files from susceptible KY14 burley tobacco infected with P. tabacina at 2 and 5 DPI (described above), susceptible spinach cultivar ‘Viroflay’ infected with P. effusa at 2 and 7 DPI [34], and susceptible basil infected with P. belbahrii at 3 and 6 DPI [29] were included in the expression analysis. The data from each transcriptome were normalized using the trimmed mean of M-values normalization method [35].
Phylogenetic and protein analysis
Protein sequences were aligned and phylogenetic trees were constructed using MEGA version 10 [36]. Protein alignments and a percent identity matrix of transporters was calculated using Clustal Omega [37]. The number of transmembrane helices in proteins was estimated using several software programs including HMMTOP [38], deep TMHMM [39], and SPLIT 4.0 [40]. The TM-align algorithm [41] was utilized to compare two modeled protein structures. The EMBOSS needle alignment algorithm [42], which is available on the EMBL- EBI server [43], was used for comparing two protein sequences.
Ab Initio modelling of protein sequences
I-TASSER (Iterative Threading Assembly Refinement; https://zhanglab.ccmb.med.umich.edu/I-TASSER/) was used to predict protein structure and function. I-TASSER creates a 3D model of a protein using an ab initio modelling method [44]. The C-score is a confidence score used by I-TASSER to estimate the superiority of projected models; it is normally in the range of -5 to 2, with a higher number indicating a model with high confidence [45,46].
Results and discussion
Identification of putative transporter genes in three Peronospora species
This study focused on the identification of transporter genes primarily involved in providing nutrition for Peronospora pathogens and excluded the ABC transporters and ion channels. Nearly 200 putative transporter genes were identified in the pathogen P. effusa (Tables 1 and S2). Fewer transporter genes were identified in P. belbahrii and P. tabacina for similar reasons. In a few cases, stop codons could be found in the P. belbahrii or P. tabacina genes that were homologous to those from P. effusa (S2 Table). In other cases, no homologous genes could be found in P. belbahrii or P. tabacina. Lastly, no conserved transporter motifs were identified in the protein encoded by the gene, or the motif identified lacked statistical support.
The Major Facilitator Superfamily (MFS) of transporters had the highest number of gene members in all three Peronospora species compared to the other transporter families (Table 1). The MFS family is also the largest transporter family in other plant pathogens such as Phytophthora infestans, Pythium ultimum, and Magnaporthe oryzae [31]. It should be noted that the Folate-Biopterin, Glycoside-Pentoside-Hexuronide:Cation Symporter, and Proton-dependent Oligopeptide Transporter families are specialized MFS transporters. There were also a number of gene members from the Mitochondrial Carrier and Drug/Metabolite Transporter families identified in the genomes of the three Peronospora species.
Differential expression of Peronospora transporter genes during interactions between oomycete pathogens and their susceptible hosts
Expression of all putative transporter genes in P. effusa was analyzed in spinach leaves during susceptible host infection (S3 Table). The TPM levels of the transporter genes of P. effusa were quite low at 2 DPI; the proportion of total RNA mapped to the P. effusa genome was <1% in these leaf samples [34]. Only one transporter gene, RMX67221, had enhanced expression at 2 DPI, although it was ~50% the level of the tubulin B housekeeping gene (Table 2). Many more transporter genes were expressed by P. effusa at 7 DPI (S3 Table); this finding is likely due to more growth of the pathogen in the plants between 2 DPI and 7 DPI because 25 to 48% of the total sequence reads were mapped to the P. effusa genome in these RNA samples [34]. Fifty percent of the top 50 most highly expressed genes had TPM values >5,000 and included the housekeeping genes (S3 Table). The class of transporters that was most frequently found in the top 50 most highly expressed genes encoded MFS transporters, followed by the Mitochondrial Carrier (MC) transporters (Fig 1A). In addition, the TPM levels of two MFS, two MC and one Amino Acid/Auxin Permease (AAAP) gene were close to the same TPM levels of the housekeeping genes beta tubulin and pyruvate dehydrogenase alpha subunit in P. effusa at 7 DPI (Table 2).
The panels show the distribution of the top 50 most highly expressed genes into various transporter classes in each pathosystem at the listed time point; refer to Table 1 for the meaning of the abbreviations. HK stands for housekeeping gene. A, P. effusa 7 DPI; B, P. belbahrii 3 DPI; C, P. belbahrii 6 DPI; D, P. tabacina 2 DPI; E, P. tabacina 5 DPI. DPI, days post inoculation.
The expression rates of transporter genes analyzed for P. effusa were also measured in RNA samples extracted from basil leaves infected by P. belbahrii [29], as detailed in S4 Table; approximately 17% and 32% of the total reads mapped to the P. belbahrii genome in the 3 and 6 DPI replicates, respectively [29]. Slightly less than 50% of the top 50 most highly expressed genes had TPM values >5,000 at 3 DPI whereas 60% of the top 50 most highly expressed genes were >5,000 TPM at 6 DPI (S4 Table). All of the P. belbahrii housekeeping genes had TPM values of least 4,000 in all of the 3 and 6 DPI samples (S4 Table). The expression levels of P. belbahrii actin and RMX67221 exceeded 30,000 TPM at both 3 and 6 DPI (S4 Table). The transporter class that was most frequently found in the top 50 most highly expressed P. belbahrii gene transcripts analyzed were MC transporters (Fig 1B and 1C). Some of the most highly expressed P. belbahrii gene transcripts are listed in Table 2. Three different P. belbahrii MC transporter genes were highly expressed at 3 DPI, especially expression levels of RMX67221. At 6 DPI, the RMX67221 TPM levels were still very high, at approximately the same level as actin; the mRNA expression levels of two other MC genes were close to those of the housekeeping genes beta tubulin and pyruvate dehydrogenase alpha at this time point. In addition, an AAAP, MFS, and a Choline Transporter-Like (CTL) transporter gene had substantial mRNA levels in P. belbahrii at 6 DPI.
Lastly, the transporter genes were analyzed from RNA isolated from tobacco leaves infected by P. tabacina (S5 Table). The mean mapping rates of the total reads to the P. tabacina genome were 20% ± 6 (SD) for the 2 DPI samples (N = 3), and 50% ± 5 for the 5 DPI samples (N = 4). The expression levels of actin and RMX67221 exceeded 30,000 TPM in P. tabacina at both 2 and 5 DPI (S5 Table). Approximately 34% and 52% of the top 50 most highly expressed gene transcripts in P. tabacina had TPM values >5,000 TPM at 2 and 5 DPI, respectively (S5 Table). All of the P. tabacina housekeeping genes had TPM values of at least 3,000 in all the samples (S5 Table). As we found for P. belbahrii, the MC transporter genes of P. tabacina were the predominant transporter class among the top 50 most highly expressed genes analyzed in P. tabacina (Fig 1D and 1E). Some of the most well-expressed (where the expression level was 10,000 TPM or greater at 2 or 5 DPI) P. tabacina genes are given in Table 2. Another MC transporter gene, RMX65655, was also well expressed at both 2 and 5 DPI. A variety of other P. tabacina transporter genes were transcribed at TPM levels of 10,000 or higher at 5 DPI.
Comparisons among the top 45 most highly expressed transporter genes (not including housekeeping genes) were made between the three Peronospora species at the later dates following inoculation (5–7 DPI, Fig 2A). Sixteen transporter genes were well-expressed in each organism at these later dates following inoculation timepoint. Of the 16 transporter genes in common, seven were MCs, as shown in Fig 2B. In addition, there were four MFS transporter genes that were well-expressed in all three Peronospora pathogens (Fig 2C). In each species, 11–18 transporter genes were expressed exclusively among the top 45 most highly expressed genes (Fig 2A) which may reflect the diversity of metabolites available to the pathogen in each plant host or may be due to variations in metabolism in each pathogen. A comparison of 91 KEGG [47] metabolic pathways from 42 oomycete proteomes, including P. effusa and P. belbahrii, was recently published [25]. The "pan-pathways" were defined as those pathways where each reaction was identified in at least one species [25]. Pathway coverage was then calculated by dividing the number of reactions of each species by the total number of reactions in the pan-pathway [25]. The pathway coverage of P. effusa and P. belbahrii was similar for the majority of the 91 metabolic pathways, but there were notable differences; for example, isoquinoline alkaloid biosynthesis pathway coverage was 14% for P. belbahrii and 43% for P. effusa; lipopolysaccharide biosynthesis pathway coverage was complete (100%) for P. belbahrii but only 10% for P. effuse [25]. These differences in pathogen metabolism may require unique transporters for metabolite import or export. It is also interesting to note that a handful of transporter genes in Plasmopara viticola, a downy mildew causing pathogen of grapevine, were under positive selection [48]. Much more research is necessary to identify essential metabolites imported by these three Peronospora pathogens from their hosts.
A, the top 45 most highly expressed transporter genes; B, the number of MC transporters among the top 45 most highly expressed transporter genes; C, the number of MFS transporters among the top 45 most highly expressed transporter genes. DPI, days post inoculation. bel, P. belbahrii; effusa, P. effusa; tab, P. tabacina.
Possible functions for the highly expressed Peronospora transporter genes
The putative functions of the 16 transporter genes that were well-expressed in all three Peronospora organisms were identified from the NCBI annotation (Table 3). In the next few paragraphs, the putative functions of the 16 transporter genes will be described and examples from other organisms discussed.
The RMX67221 gene encodes a putative ADP/ATP translocase that is necessary for energy production in the mitochondria of all organisms and is one of the most abundant proteins in the inner mitochondrial membrane [49]. The protein encoded by RMX67221 in all three Peronospora species contains the consensus motif RRRMMM, which is typical of all ADP/ATP translocases [50]. The RMX65655 gene encodes a putative mitochondrial phosphate carrier protein, and like ADP/ATP translocases, it is essential for energy production from the mitochondria [51]. The protein encoded by RMX65655 is quite similar to the phosphate carrier in Arabidopsis thaliana (BLASTP E value was 2e-97). The results of a previous study indicated that both ADP/ATP translocase and phosphate carrier protein levels were the most abundant proteins in the inner mitochondrial membrane in A. thaliana [52].
RMX67457 codes for a putative mitochondrial GTP/GDP carrier protein. The yeast gene orthologous to RMX67457, GGC1, is important for maintenance of mitochondrial DNA synthesis in yeast cells [53]. RMX67668 encodes a putative citrate/oxoglutarate carrier protein; the yeast orthologous protein transports citrate and oxoglutarate, and was also shown to transport oxaloacetate, succinate and fumarate [54], suggesting that the best substrate for the Peronospora RMX67668 proteins should be tested using in vitro techniques. The top BLASTP hit for RMX63170 indicated that this gene encodes a putative ADP/ATP carrier protein, otherwise known as an ADP/ATP translocase. However, lower scoring BLASTP hits suggested that RMX63170 encodes for a putative peroxisomal adenine nucleotide transporter. Phylogenetic analysis showed that the predicted RMX63170 proteins from the three Peronospora species cluster in a clade with other known peroxisomal adenine nucleotide transporters, not in the clade containing the ADP/ATP carrier proteins (Fig 3). The peroxisomal adenine nucleotide transporter moves ATP from the cytosol into the peroxisome in exchange for AMP; the ATP is important for the activation of fatty acids for β-oxidation that occurs in the peroxisome [55].
The maximum likelihood tree was constructed using the LG model. The percentages of replicate trees in which the associated proteins clustered together in the bootstrap test (1,000 replicates) are shown next to the nodes. The proteins in the blue box are putative ADP/ATP translocases; the proteins in the black box are putative peroxisomal nucleotide transporters; the proteins directly left of the green line are the outgroup, which are putative thiamine pyrophosphate (TPP) transporters. The genus of the organism from which the protein originated and the Genbank number is given at the tip of each branch; a ‘_’ follows each XP or NP. The scale bar indicates the number of protein substitutions per site.
The RMX66310 gene encodes a putative solute carrier family 25 member 40; the substrates for the orthologous proteins from human and fruit fly have yet to be identified, but the genes are expressed in nerve tissue mitochondria in both of these organisms [56]. RMX69262 encodes a putative 2-oxoglutarate/malate carrier protein, also known as the mitochondrial dicarboxylate-tricarboxylate carrier (DTC) protein [57]. The RMX69262 protein from P. effusa shares 38% identity with A. thaliana DTC (At5g19760, BLASTP E value = 2e-67), which specifically transports protonated citrate and unprotonated malate [57]; DTCs can also transport oxaloacetate, oxoglutarate, isocitrate cis-aconitate, and trans-aconitate [57].
RMX65758 encodes a putative acetyl-coenzyme A transporter protein that is localized to the endoplasmic reticulum (ER) in mammalian cells [58]. The acetyl-coenzyme A is used to acetylate membrane proteins in the ER, and defects in the acetyl-coenzyme A transporter have been linked to neurodegenerative disease in humans [59]. Surprisingly, the P. effusa acetyl-coenzyme A transporter has 47% amino acid identity with the human protein, known as AT-1. Carbohydrate transporters are part of the MFS superfamily. Three of the four MFS genes listed in Table 3 were predicted to be carbohydrate (specifically glucose) transporters. One of the most highly expressed MFS transporter genes in P. belbahrii was RMX67547 (Table 2), which was also among the top 50 most highly expressed transporter genes in P. effusa at 7 DPI; however, the RMX67547 gene was not expressed in P. tabacina at either 2 or 5 DPI.
The protein encoded by RMX64359 contains a putative Hedgehog/Intein (Hint) domain, but other high scoring annotations included “Calponin homology domain-containing proteinoline transporter”, metal transporter CNNM4 (magnesium transporter, also includes CNNM2), and Pns1p (the only choline transporter-like protein (CTL) in Saccharomyces cerevisiae [60]). In addition, the RMX68742 gene potentially codes for the Pns1p protein. The Hint domain in the hedgehog protein and the calponin homology domain do not exhibit properties of a membrane transporter [61,62]. We performed a phylogenetic analysis of a sequence alignment that included several putative magnesium transporters and many putative CTLs (Fig 4). All of the Peronospora proteins encoded by RMX64359- and RMX68742-type genes were more related to the clade containing Pns1p than they were to the magnesium transporters or the clade containing Ctl1 from the fission yeast Schizosaccharomyces pombe, which is required for autophagosome formation [63]. However, one study found that S. cerevisiae Pns1p does not actually transport choline [64], and its function is unknown [63]. In another study, two mammalian CTL-like proteins transported ethanolamine rather than choline [65]. This indicates that the CTL family of transporters may have a more diverse range of substrates than previously thought, and that the substrate(s) of the Peronospora CTLs will need to be identified in the future.
The maximum likelihood tree was constructed using the WAG+F model. The percentages of replicate trees in which the associated proteins clustered together in the bootstrap test (1,000 replicates) are shown adjacent to the nodes. The proteins directly to the left of the black line are the magnesium transporters, whereas the Ctl1 and Pns1p proteins are noted with arrows. The genus of the organism from which the protein originated and the Genbank number is given at the tip of each branch; a ‘_’ follows each XP or NP. The scale bar indicates the number of protein substitutions per site.
The RMX67305 gene encodes a putative NIPA-like protein, which is a selective magnesium transporter in mammals [66,67]. Magnesium, along with calcium, contributes to cross linking of the middle lamella in plants that makes this structure more resistant to the pectolytic enzymes produced by plant pathogens [68]. The Peronospora oomycete pathogens might actively remove magnesium from the plant middle lamella during infection by increasing the expression levels of their NIPA-like transporter-encoding genes. RQM14033 encodes a putative C4-dicarboxylate transport protein which has substantial homology (BLASTP E value = 9e-78) to the C4-dicarboxylate transporter from Bacillus subtilis, which is part of the DctA family of secondary transporters in bacteria [69]. The in vitro substrate specificity of the B. subtilis DctA transporter is limited to succinate, malate, fumarate, and oxaloacetate, which are C4-dicarboxylates of the Krebs cycle [70]. Similar in vitro experiments will need to be completed in the future to determine the substrate specificity of each Peronospora DctA-like transporter.
The RMX68746 gene encodes a putative UAA family transporter that is a nucleotide sugar transporter (NST); these transporters are critical for the movement of nucleotide sugars from the cytosol to the Golgi apparatus [71]. A phylogenetic analysis of 257 NST proteins showed that the substrate specificity of any NST if can be inferred from its primary sequence [71]. A BLASTP analysis indicated that the proteins encoded by the three Peronospora RMX68746 genes are part of the solute carrier family 35 (member B1 isoform 1), which corresponded to clade F of the NST superfamily [71]. The proteins encoded by the three Peronospora RMX68746 genes clustered with all 15 clade F proteins [71] in a phylogenetic tree (S1 Fig). Clade F proteins transport uridine-5’-diphosphate (UDP)-galactose, UDP-glucose, and adenosine 3’-phospho 5’-phosphosulfate [71]. This proposed substrate specificity of the Peronospora proteins encoded by the RMX68746 genes should be experimentally verified in the future.
In-depth transporter protein analysis
The proteins encoded by the 16 transporter genes that were highly expressed in the three Peronospora organisms were examined in more detail (Table 4). Pairwise comparisons of each species with each other indicated a high degree of amino acid identity. Most of the Peronospora MC proteins were close to 300 amino acids in length, which is typical for these transporters [72]. The exception was protein A0A3M6VF52 from P. effusa, which was 586 amino acids long. Alignment of A0A3M6VF52 with the homologous proteins of P. belbahrii and P. tabacina showed that the homologous regions of the three proteins are in the C-terminus of A0A3M6VF52 (see S2 File). Mapping of the reads in the four P. effusa transcriptomes (7 DPI) to the P. effusa RMX65655 gene revealed that the reads primarily localized to the 3’ end of A0A3M6VF52, which suggests that the P. effusa RMX65655 gene is not correctly annotated. The P. effusa RMX66310 gene encodes a 97-amino acid protein, but a protein from Genbank (UIZ22858.1) is 355 amino acids in length and is likely to be the authentic protein, because it shared 90% identity with Ptab2_001070 from P. tabacina.
There was substantial variability in amino acid sequence length among the three proteins (putative glucose transporters) in each Peronospora species encoded by the RQM16367 genes. GLUTs or glucose transporters should contain 12 membrane-spanning helices [73]. A0A3R7W609 was predicted to contain 11, 14, and 11 transmembrane helices whereas PBEL_00785 was predicted to contain 11, 12, and 9 transmembrane helices according to the transmembrane helix predictor programs HMMTOP [38], deep TMHMM [39] and SPLIT 4.0 [40], respectively. The variability in these predictor programs indicates that more detailed protein structural analysis needs to be performed. However, it is likely that the actual coding sequence for Ptab2_015719, the protein from P. tabacina, is incomplete.
It should be noted that the genes Ptab2_000964, Ptab2_000106, and Ptab2_009561 all contained heterozygous nucleotide sites. The encoded amino acids at these heterozygous sites were arbitrarily chosen to be a glycine (G) so that it would not match the amino acid at that position in P. belbahrii or P. effusa; if adding a G resulted in a perfect match at the P. tabacina heterozygous site among all three Peronospora species, a different amino acid was randomly added at that site to purposefully cause mismatches among the three Peronospora proteins. Even with these intentional mismatches included in the proteins encoded by genes with heterozygous nucleotide sites, the transporter proteins in the three Peronospora were highly similar (75–92% identical).
Protein modelling was completed for the ADP/ATP translocase, P. effusa A0A3M6VMF9, the mitochondrial phosphate carrier protein, P. belbahrii PBEL_07973, and one glucose transporter, P. effusa A0A3R7W609 (details of the models are in S3 File). These specific proteins were chosen for modelling because of their high levels of expression and presumed important roles in all three Peronospora species during host infection. Each Peronospora protein model was aligned to a protein model with the same putative function in a mammal (Fig 5). The ADP/ATP translocase pair had 58% sequence identity and a TM-score of 0.69 if normalized to the length of the P. effusa protein; TM-scores range from 0 to 1, and scores above 0.5 indicate that the proteins have similar folding patterns [74]. The mitochondrial phosphate carrier and glucose transporter pairs had 48% and 31% sequence identity and TM-scores of 0.77 and 0.83, respectively (both TM-scores were normalized to the length of the Peronospora protein).
A, A0A3M6VMF9 from P. effusa (green) and 1OKC (PDB) from Bos taurus (cyan); B, PBEL_07973 from P. belbahrii (cyan) and NP_002626.1 from Homo sapiens (green); C, A0A3R7W609 from P. effusa (cyan) and XP_011511389.1 from Homo sapiens (green).
ADP/ATP translocases have been extensively studied. Yeast and bovine ADP/ATP translocases share almost 50% amino acid sequence identity [50]. The bovine protein used in the structural comparison shown in Fig 5A, 1OKC, was crystallized with an inhibitor, carboxyatractyloside [50]. P. effusa A0A3M6VMF9 and 1OKC share 67% amino acid identity as determined by the EMBOSS needle alignment algorithm. The RRRMMM motif in the bovine ADP/ATP translocase (which is also present in the three Peronospora translocases) was postulated to be a two-way switch which regulates the nucleotide binding stoichiometry [50].
The mammalian phosphate carrier, also known as SLC25A3, has two isoforms, A and B, that differ by 13 amino acids between residues 54 and 80 [75]. The structure of SLC25A3 isoform B from humans was used for comparison with P. belbahrii PBEL_07973 (Fig 5B). Bovine SLC25A3 isoform B was shown to have a phosphate transport rate that was ~3-fold higher than the phosphate transport rate of bovine SLC25A3 isoform A in liposomes [76]. Human SLC25A3-A and SLC25A3-B transported phosphate in Lactococcus lactis, as the transformed cells were unable to grow in 1.6 mM arsenate, a toxic mimetic of phosphate [75]. Surprisingly, human SLC25A3-A and SLC25A3-B also exhibited the ability to transport copper, because the L. lactis transformants were unable to grow in the presence of 100 μM silver, which is an indirect method to demonstrate copper transport [75]. The PIC2 protein transports phosphate and copper in Saccharomyces cerevisiae, but MIR1 can only transport phosphate in this organism; the different substrate specificities of these two mitochondrial carriers suggests that they evolved from an ancient gene duplication [77]. There are PIC2-like and MIR1-like proteins present throughout eukaryotic lineages [77]. Based on their clustering pattern in a phylogenetic tree (S2 Fig), the Peronospora phosphate carriers identified in this study (PBEL_07973, A0A3M6VF52, and Ptab2_000589; Table 4) are PIC2-like proteins and could possibly transport both phosphate and copper. The P. effusa genome is not likely to contain any MIR1-like proteins because the top BLASTP hit using several oomycete MIR1-like proteins as queries at the NCBI server was the PIC2-like protein A0A3M6VF52 (data not shown); however, more complete annotation of the Peronospora genomes may uncover some MIR1-like proteins.
In the last structural comparison (Fig 5C), the human glucose transporter GLUT2 (XP_011511389.1) and P. effusa A0A3R7W609 were found to share 24% amino acid sequence identity using the EMBOSS needle alignment algorithm [42] at the following Internet server [43]. GLUT2 can also transport galactose, mannose, fructose, and glucosamine [78]. This broad substrate specificity of GLUT2 suggests that substrate testing should be performed for all the Peronospora carbohydrate transporters in the future.
Importance of transporters in pathogen nutrition, virulence, and defense
The majority of the highly expressed transporters in this study (Fig 1) contribute to the movement of essential molecules across the inner mitochondrial membrane [79]. In addition, we found that many MFS transporter genes were well expressed during infection. Some Peronospora MFS transporters likely contribute to the movement of carbohydrate(s) from the host to the pathogen, which is vital for pathogen growth within the plant, but other Peronospora MFS transporters could be involved in the movement of plant defensive products, or man-made fungicides, out of the pathogen. An MFS transporter in Botrytis cinerea promotes tolerance to isothiocyanate, a breakdown product of glucosinolates, which are natural defense compounds produced by species in the Brassicaceae [80]. The authors of the previously mentioned study on B. cinerea speculated that the large number of MFS transporters in that pathogen are utilized to counteract antimicrobial compounds produced by plants during infection. There are also some reports that microbial MFS transporters secrete pathogenicity factors or toxins into plant tissues [81], and these MFS transporters may also protect the pathogens themselves from deleterious levels of their own toxins [82,83]. On another note, some eukaryotic organisms develop multidrug resistance, as well as fungicide resistance, through the expression of ABC and MFS transporters that export the fungicide out of the cell [84]. For example, mutation of an MFS transporter in Alternaria alternata increased the sensitivity of the mutant to several fungicides [85]. Disruption of MgMfs1 in Mycosphaerella graminicola increased the mutant’s sensitivity to several strobilurin fungicides [86]. The substrate specificities of the many Peronospora MFS transporters need to be identified to determine whether each transporter plays role in pathogen protection or pathogenicity.
This study identified the genes that encode several Peronospora transporters that are likely to be crucial for growth and colonization of their plant host. On the one hand, whereas it is relatively straightforward to characterize the gene expression patterns of pathogen transporter genes during infection, the bottleneck in further knowledge of these membrane proteins will be the identification of transporter substrates, which is usually done with artificial membranes, or in yeast, and recombinant expression of the transporter protein. On the other hand, the identification of these crucial transporter genes in Peronospora pathogens could be helpful for designing crop protection products against pathogens. The addition of small double-stranded RNA molecules targeting a downy mildew pathogen’s cellulose synthase gene to the spores of Hyaloperonospora arabidopsidis inhibited infection of its host plant [87]. It is conceivable that downy mildew pathogen infection of basil, spinach, or tobacco could be inhibited by targeting one of the 16 well-expressed transporter genes identified in this study. The use of gene silencing RNA molecules for plant protection is still relatively untested in agricultural settings, but the concept is being examined for control of several fungal and oomycete plant pathogens [88,89]. New control strategies need to be developed in the near future because downy mildew causing pathogens can evade current control technologies over time.
Supporting information
S1 Fig. Phylogenetic tree of Clade F NST proteins.
The neighbor-joining phylogenetic tree was constructed using the JTT model. The percentage of replicate trees in which the associated proteins clustered together in the bootstrap test (1000 replicates) are posted next to the branches. The F and I proteins, listed with their Uniprot accession numbers, formed two separate clades in the tree. The bar indicates the number of protein substitutions per site.
https://doi.org/10.1371/journal.pone.0285685.s001
(PPTX)
S2 Fig. Phylogenetic tree of PIC2-like and MIR-like proteins.
The maximum likelihood tree was constructed using the LG model. The percentage of replicate trees in which the associated proteins clustered together in the bootstrap test (1000 replicates) are posted next to the branches. The proteins in the red box are putative PIC2-like; the proteins in the blue box are putative MIR1-like transporters. The genus of the organism from which the protein originated and the Genbank number is listed on each branch; a ‘_’ follows each XP or NP. The bar indicates the number of protein substitutions per site.
https://doi.org/10.1371/journal.pone.0285685.s002
(PPTX)
S1 Table. Quality metrics for tobacco leaf transcriptome samples infected with P. tabacina.
https://doi.org/10.1371/journal.pone.0285685.s003
(XLSX)
S2 Table. Motif scores for all the putative transporters identified in the three Peronospora species.
https://doi.org/10.1371/journal.pone.0285685.s004
(XLSX)
S3 Table. Gene expression values for P. effusa.
https://doi.org/10.1371/journal.pone.0285685.s005
(XLSX)
S4 Table. Gene expression values for P. belbahrii.
https://doi.org/10.1371/journal.pone.0285685.s006
(XLSX)
S5 Table. Gene expression values for P. tabacina.
https://doi.org/10.1371/journal.pone.0285685.s007
(XLSX)
S1 File. DNA sequences of housekeeping genes of all three Peronospora species.
https://doi.org/10.1371/journal.pone.0285685.s008
(TXT)
S2 File. Multiple sequence alignment of the phosphate carrier proteins in the three Peronospora species.
https://doi.org/10.1371/journal.pone.0285685.s009
(DOCX)
S3 File. Confidence tests for the modeled transporter proteins.
https://doi.org/10.1371/journal.pone.0285685.s010
(DOCX)
Acknowledgments
We thank Mark Doehring for excellent technical assistance and Shyam Kandel for reviewing the draft manuscript. The mention of firm (company) names or trade products does not imply that they are endorsed or recommended by the USDA over other firms (companies) or similar products not mentioned. USDA is an equal opportunity provider and employer.
References
- 1. Constantinescu O. An annotated list of Peronospora names. Thunbergia. 1991; 15: 1–110.
- 2. Thines M, Choi Y-J. Evolution, diversity, and taxonomy of the Peronosporaceae, with focus on the genus Peronospora. Phytopathology. 2016; 106: 6–18.
- 3. Kandel SL, Mou B, Shishkoff N, Shi A, Subbarao KV, Klosterman SJ. Spinach downy mildew: Advances in our understanding of the disease cycle and prospects for disease management. Plant Dis. 2019; 103: 791–803. pmid:30939071
- 4. Correll J, Bluhm B, Feng C, Lamour K, Du Toit L, Koike S. Spinach: better management of downy mildew and white rust through genomics. Eur J Plant Pathol. 2011; 129: 193–205.
- 5. Correll J, Feng C, Matheron M, Porchas M, Koike S. Evaluation of spinach varieties for downy mildew resistance. Plant Dis Manag Rep. 2015; 10: V107.
- 6. Klosterman S. Spinach downy mildew–Threat, prevention and control. Prog Crop Consult. 2016; 1: 12–15.
- 7. Fletcher K, Klosterman SJ, Derevnina L, Martin F, Bertier LD, Koike S, et al. Comparative genomics of downy mildews reveals potential adaptations to biotrophy. BMC Genomics. 2018; 19: 1–23.
- 8. Belbahri L, Calmin G, Pawlowski J, Lefort F. Phylogenetic analysis and Real Time PCR detection of a presumbably undescribed Peronospora species on sweet basil and sage. Mycol Res. 2005; 109: 1276–1287.
- 9. Farahani-Kofoet RD, Römer P, Grosch R. Systemic spread of downy mildew in basil plants and detection of the pathogen in seed and plant samples. Mycol Prog. 2012; 11: 961–966.
- 10. Koroch AR, Villani TS, Pyne RM, Simon JE. Rapid staining method to detect and identify downy mildew (Peronospora belbahrii) in basil. Appl Plant Sci. 2013; 1: 1300032. pmid:25202569
- 11. Farahani-Kofoet RD, Römer P, Grosch R. Selecting basil genotypes with resistance against downy mildew. Sci Hortic. 2014; 179: 248–255.
- 12. McGrath MT, Sexton Z. Evaluation of fungicides for downy mildew in sweet basil, 2016. Plant Dis Manag Rep. 2017; 11: V032.
- 13. Homa K, Barney WP, Ward DL, Wyenandt CA, Simon JE. Evaluation of fungicides for the control of Peronospora belbahrii on sweet basil in New Jersey. Plant Dis. 2014; 98: 1561–1566.
- 14. Raid R, Vital J, Hartman D. Evaluation of fungicide drench treatments for control of basil downy mildew, 2016. Plant Dis Manag Rep. 2017; 11: V031.
- 15. Zhang G, Babadoost M, De Young A, Johnson ET, Schisler DA. Evaluation of selected fungicide application regimes and biotic agents for the management of basil downy mildew. HortTechnology. 2018; 28: 822–829.
- 16. Ben-Naim Y, Falach L, Cohen Y. Transfer of downy mildew resistance from wild basil (Ocimum americanum) to sweet basil (O. basilicum). Phytopathology. 2018; 108: 114–123. pmid:29083273
- 17. Pyne R, Honig J, Vaiciunas J, Koroch A, Wyenandt C, Bonos S, et al. A first linkage map and downy mildew resistance QTL discovery for sweet basil (Ocimum basilicum) facilitated by double digestion restriction site associated DNA sequencing (ddRADseq). PloS One. 2017; 12: e0184319. pmid:28922359
- 18. Ben-Naim Y, Weitman M. Joint action of Pb1 and Pb2 provide dominant complementary resistance against new races of Peronospora belbahrii (Basil Downy Mildew). Phytopathology. 2021.
- 19. Thines M, Sharma R, Rodenburg SY, Gogleva A, Judelson HS, Xia X, et al. The genome of Peronospora belbahrii reveals high heterozygosity, a low number of canonical effectors, and TC-rich promoters. Mol Plant-Microbe Interact. 2020; 33: 742–753. pmid:32237964
- 20. Blanco-Meneses M, Ristaino JB. Detection and quantification of Peronospora tabacina using a real-time polymerase chain reaction assay. Plant Dis. 2011; 95: 673–682.
- 21. Heist E, Nesmith W, Schardl C. Interactions of Peronospora tabacina with roots of Nicotiana spp. in gnotobiotic associations. Phytopathology. 2002; 92: 400–405.
- 22. LaMondia J. Reduced sensitivity of Peronospora tabacina, causal agent of tobacco blue mold, to dimethomorph fungicide in Connecticut. Tobacco Sci. 2013; 50: 19–24.
- 23. Zipper R, Hammer TR, Spring O. PCR-based monitoring of recent isolates of tobacco blue mold from Europe reveals the presence of two genetically distinct phenotypes differing in fungicide sensitivity. Eur J Plant Pathol. 2009; 123: 367–375.
- 24. Derevnina L, Chin-Wo-Reyes S, Martin F, Wood K, Froenicke L, Spring O, et al. Genome sequence and architecture of the tobacco downy mildew pathogen Peronospora tabacina. Mol Plant-Microbe Interact. 2015; 28: 1198–1215.
- 25. Rodenburg SY, de Ridder D, Govers F, Seidl MF. Oomycete metabolism is highly dynamic and reflects lifestyle adaptations. BioRxiv. 2020–02.
- 26. Baxter L, Tripathy S, Ishaque N, Boot N, Cabral A, Kemen E, et al. Signatures of adaptation to obligate biotrophy in the Hyaloperonospora arabidopsidis genome. Science. 2010; 330: 1549–1551.
- 27. Judelson HS. Metabolic diversity and novelties in the oomycetes. Annu Rev Microbiol. 2017; 71: 21–39. pmid:28504899
- 28. Heist E, Zaitlin D, Funnell DL, Nesmith W, Schardl C. Necrotic lesion resistance induced by Peronospora tabacina on leaves of Nicotiana obtusifolia. Phytopathology. 2004; 94: 1178–1188.
- 29. Johnson ET, Kim H-S, Tian M, Dudai N, Tal O, Gonda I. Dual transcriptional analysis of Ocimum basilicum and Peronospora belbahrii in susceptible interactions. Plant Gene. 2022; 29: 100350.
- 30. Matsubara T, Soh J, Morita M, Uwabo T, Tomida S, Fujiwara T, et al. DV200 index for assessing RNA integrity in next-generation sequencing. BioMed Res Int. 2020; 2020. pmid:32185225
- 31. Abrahamian M, Ah-Fong AM, Davis C, Andreeva K, Judelson HS. Gene expression and silencing studies in Phytophthora infestans reveal infection-specific nutrient transporters and a role for the nitrate reductase pathway in plant pathogenesis. PLoS Pathog. 2016; 12: e1006097.
- 32. Blum M, Chang H-Y, Chuguransky S, Grego T, Kandasaamy S, Mitchell A, et al. The InterPro protein families and domains database: 20 years on. Nucleic Acids Res. 2021; 49: D344–D354. pmid:33156333
- 33. Huang Y, Niu B, Gao Y, Fu L, Li W. CD-HIT Suite: a web server for clustering and comparing biological sequences. Bioinformatics. 2010; 26: 680–682. pmid:20053844
- 34. Kandel SL, Hulse-Kemp AM, Stoffel K, Koike ST, Shi A, Mou B, et al. Transcriptional analyses of differential cultivars during resistant and susceptible interactions with Peronospora effusa, the causal agent of spinach downy mildew. Sci Rep. 2020; 10: 1–13.
- 35. Robinson MD, Oshlack A. A scaling normalization method for differential expression analysis of RNA-seq data. Genome Biol. 2010; 11: 1–9. pmid:20196867
- 36. Kumar S, Stecher G, Li M, Knyaz C, Tamura K. MEGA X: molecular evolutionary genetics analysis across computing platforms. Mol Biol Evol. 2018; 35: 1547–1549. pmid:29722887
- 37. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high‐quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011; 7: 539. pmid:21988835
- 38. Tusnady GE, Simon I. The HMMTOP transmembrane topology prediction server. Bioinformatics. 2001; 17: 849–850. pmid:11590105
- 39. Hallgren J, Tsirigos KD, Pedersen MD, Armenteros JJA, Marcatili P, Nielsen H, et al. DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks. bioRxiv. 2022.
- 40. Juretić D, Zoranić L, Zucić D. Basic charge clusters and predictions of membrane protein topology. J Chem Inf Comput Sci. 2002; 42: 620–632. pmid:12086524
- 41. Zhang Y, Skolnick J. TM-align: a protein structure alignment algorithm based on the TM-score. Nucleic Acids Res. 2005; 33: 2302–2309. pmid:15849316
- 42. Needleman SB, Wunsch CD. A general method applicable to the search for similarities in the amino acid sequence of two proteins. J Mol Biol. 1970; 48: 443–453. pmid:5420325
- 43. Madeira F, Pearce M, Tivey A, Basutkar P, Lee J, Edbali O, et al. Search and sequence analysis tools services from EMBL-EBI in 2022. Nucleic Acids Res. 2022. pmid:35412617
- 44. Zhang Y. I-TASSER server for protein 3D structure prediction. BMC bioinformatics. 2008; 9: 1–8.
- 45. Roy A, Kucukural A, Zhang Y. I-TASSER: a unified platform for automated protein structure and function prediction. Nat Protoc. 2010; 5: 725–738. pmid:20360767
- 46. Yang J, Yan R, Roy A, Xu D, Poisson J, Zhang Y. The I-TASSER Suite: protein structure and function prediction. Nat Methods. 2015; 12: 7–8. pmid:25549265
- 47. Kanehisa M, Sato Y, Kawashima M. KEGG mapping tools for uncovering hidden features in biological data. Protein Sci. 2022; 31: 47–53. pmid:34423492
- 48. Dussert Y, Mazet ID, Couture C, Gouzy J, Piron M-C, Kuchly C, et al. A high-quality grapevine downy mildew genome assembly reveals rapidly evolving and lineage-specific putative host adaptation genes. Genome Biol Evol. 2019; 11: 954–969. pmid:30847481
- 49. Kunji ER, Aleksandrova A, King MS, Majd H, Ashton VL, Cerson E, et al. The transport mechanism of the mitochondrial ADP/ATP carrier. Biochim Biophys Acta Mol Cell Res. 2016; 1863: 2379–2393.
- 50. Pebay-Peyroula E, Dahout-Gonzalez C, Kahn R, Trézéguet V, Lauquin GJ-M, Brandolin G. Structure of mitochondrial ADP/ATP carrier in complex with carboxyatractyloside. Nature. 2003; 426: 39–44. pmid:14603310
- 51. Seifert EL, Ligeti E, Mayr JA, Sondheimer N, Hajnóczky G. The mitochondrial phosphate carrier: role in oxidative metabolism, calcium handling and mitochondrial disease. Biochem Biophys Res Commun. 2015; 464: 369–375. pmid:26091567
- 52. Fuchs P, Rugen N, Carrie C, Elsässer M, Finkemeier I, Giese J, et al. Single organelle function and organization as estimated from Arabidopsis mitochondrial proteomics. Plant J. 2020; 101: 420–441. pmid:31520498
- 53. Vozza A, Blanco E, Palmieri L, Palmieri F. Identification of the mitochondrial GTP/GDP transporter in Saccharomyces cerevisiae. J Biol Chem. 2004; 279: 20850–20857.
- 54. Castegna A, Scarcia P, Agrimi G, Palmieri L, Rottensteiner H, Spera I, et al. Identification and functional characterization of a novel mitochondrial carrier for citrate and oxoglutarate in Saccharomyces cerevisiae. J Biol Chem. 2010; 285: 17359–17370.
- 55. Traba J, Satrústegui J, Del Arco A. Adenine nucleotide transporters in organelles: novel genes and functions. Cell Mol Life Sci. 2011; 68: 1183–1206. pmid:21207102
- 56. Curcio R, Lunetti P, Zara V, Ferramosca A, Marra F, Fiermonte G, et al. Drosophila melanogaster mitochondrial carriers: similarities and differences with the human carriers. Int J Mol Sci. 2020; 21: 6052.
- 57. Picault N, Palmieri L, Pisano I, Hodges M, Palmieri F. Identification of a novel transporter for dicarboxylates and tricarboxylates in plant mitochondria: bacterial expression, reconstitution, functional characterization, and tissue distribution. J Biol Chem. 2002; 277: 24204–24211.
- 58. Jonas MC, Pehar M, Puglielli L. AT-1 is the ER membrane acetyl-CoA transporter and is essential for cell viability. J Cell Sci. 2010; 123: 3378–3388. pmid:20826464
- 59. Hirabayashi Y, Kanamori A, Nomura KH, Nomura K. The acetyl-CoA transporter family SLC33. Pflügers Arch. 2004; 447: 760–762. pmid:12739170
- 60. Traiffort E, Ruat M, O’Regan S, Meunier F-M. Molecular characterization of the family of choline transporter‐like proteins and their splice variants. J Neurochem. 2005; 92: 1116–1125. pmid:15715662
- 61. Ingham PW, Nakano Y, Seger C. Mechanisms and functions of Hedgehog signalling across the metazoa. Nat Rev Genet. 2011; 12: 393–406. pmid:21502959
- 62. Bañuelos S, Saraste M, Carugo KD. Structural comparisons of calponin homology domains: implications for actin binding. Structure. 1998; 6: 1419–1431. pmid:9817844
- 63. Sun L-L, Li M, Suo F, Liu X-M, Shen E-Z, Yang B, et al. Global analysis of fission yeast mating genes reveals new autophagy factors. PLoS Genet. 2013; 9: e1003715. pmid:23950735
- 64. Zufferey R, Santiago TC, Brachet V, Ben Mamoun C. Reexamining the role of choline transporter-like (Ctlp) proteins in choline transport. Neurochemical research. 2004; 29: 461–467. pmid:15002745
- 65. Taylor A, Grapentine S, Ichhpuniani J, Bakovic M. Choline transporter-like proteins 1 and 2 are newly identified plasma membrane and mitochondrial ethanolamine transporters. J Biol Chem. 2021; 296. pmid:33789160
- 66. Goytain A, Hines RM, El-Husseini A, Quamme GA. NIPA1 (SPG6), the basis for autosomal dominant form of hereditary spastic paraplegia, encodes a functional Mg2+ transporter. J Biol Chem. 2007; 282: 8060–8068.
- 67. Goytain A, Hines RM, Quamme GA. Functional characterization of NIPA2, a selective Mg2+ transporter. Am J Physiol-Cell Physiol. 2008; 295: C944–C953. pmid:18667602
- 68. Huber DM, Jones JB. The role of magnesium in plant disease. Plant Soil. 2013; 368: 73–85.
- 69. Janausch I, Zientz E, Tran Q, Kröger A, Unden G. C4-dicarboxylate carriers and sensors in bacteria. Biochim Biophys Acta Bioenerg. 2002; 1553: 39–56. pmid:11803016
- 70. Groeneveld M, Detert Oude Weme RG, Duurkens RH, Slotboom DJ. Biochemical characterization of the C4-dicarboxylate transporter DctA from Bacillus subtilis. J Bacteriol. 2010; 192: 2900–2907.
- 71. Orellana A, Moraga C, Araya M, Moreno A. Overview of nucleotide sugar transporter gene family functions across multiple species. J Mol Biol. 2016; 428: 3150–3165. pmid:27261257
- 72. Palmieri F, Monné M. Discoveries, metabolic roles and diseases of mitochondrial carriers: A review. Biochim Biophys Acta Mol Cell Res. 2016; 1863: 2362–2378. pmid:26968366
- 73. Castillo J, Crespo D, Capilla E, Díaz M, Chauvigné F, Cerdà J, et al. Evolutionary structural and functional conservation of an ortholog of the GLUT2 glucose transporter gene (SLC2A2) in zebrafish. Am J Physiol Regul Integr Comp Physiol. 2009; 297: R1570–R1581. pmid:19776250
- 74. Xu J, Zhang Y. How significant is a protein structure similarity with TM-score = 0.5? Bioinformatics. 2010; 26: 889–895. pmid:20164152
- 75. Boulet A, Vest KE, Maynard MK, Gammon MG, Russell AC, Mathews AT, et al. The mammalian phosphate carrier SLC25A3 is a mitochondrial copper transporter required for cytochrome c oxidase biogenesis. J Biol Chem. 2018; 293: 1887–1896. pmid:29237729
- 76. Fiermonte G, Dolce V, Palmieri F. Expression in Escherichia coli, functional characterization, and tissue distribution of isoforms A and B of the phosphate carrier from bovine mitochondria. J Biol Chem. 1998; 273: 22782–22787.
- 77. Zhu X, Boulet A, Buckley KM, Phillips CB, Gammon MG, Oldfather LE, et al. Mitochondrial copper and phosphate transporter specificity was defined early in the evolution of eukaryotes. Elife. 2021; 10. pmid:33591272
- 78. Chadt A, Al-Hasani H. Glucose transporters in adipose tissue, liver, and skeletal muscle in metabolic health and disease. Pflügers Arch. 2020; 472: 1273–1298. pmid:32591906
- 79. Ruprecht JJ, Kunji ER. The SLC25 mitochondrial carrier family: structure and mechanism. Trends Biochem Sci. 2020; 45: 244–258. pmid:31787485
- 80. Vela-Corcía D, Aditya Srivastava D, Dafa-Berger A, Rotem N, Barda O, Levy M. MFS transporter from Botrytis cinerea provides tolerance to glucosinolate-breakdown products and is required for pathogenicity. Nature communications. 2019; 10: 1–11.
- 81. Del Sorbo G, Schoonbeek H-j, De Waard MA. Fungal transporters involved in efflux of natural toxic compounds and fungicides. Fungal Genet Biol. 2000; 30: 1–15. pmid:10955904
- 82. Alexander N, McCormick S, Hohn T. TRI12, a trichothecene efflux pump from Fusarium sporotrichioides: gene isolation and expression in yeast. Mol Gen Genet. 1999; 261: 977–984.
- 83. Wang Q, Chen D, Wu M, Zhu J, Jiang C, Xu J-R, et al. MFS transporters and GABA metabolism are involved in the self-defense against DON in Fusarium graminearum. Front Plant Sci. 2018; 9: 438.
- 84. Gulshan K, Moye-Rowley WS. Multidrug resistance in fungi. Eukaryot Cell. 2007; 6: 1933–1942. pmid:17873085
- 85. Lin H-C, Yu P-L, Chen L-H, Tsai H-C, Chung K-R. A major facilitator superfamily transporter regulated by the stress-responsive transcription factor Yap1 is required for resistance to fungicides, xenobiotics, and oxidants and full virulence in Alternaria alternata. Front Microbiol. 2018; 9: 2229. pmid:30279684
- 86. Roohparvar R, De Waard MA, Kema GH, Zwiers L-H. MgMfs1, a major facilitator superfamily transporter from the fungal wheat pathogen Mycosphaerella graminicola, is a strong protectant against natural toxic compounds and fungicides. Fungal Genet Biol. 2007; 44: 378–388. pmid:17107817
- 87. Bilir Ö, Telli O, Norman C, Budak H, Hong Y, Tör M. Small RNA inhibits infection by downy mildew pathogen Hyaloperonospora arabidopsidis. Mol Plant Pathol. 2019; 20: 1523–1534.
- 88. Qiao L, Lan C, Capriotti L, Ah‐Fong A, Nino Sanchez J, Hamby R, et al. Spray‐induced gene silencing for disease control is dependent on the efficiency of pathogen RNA uptake. Plant Biotechnol J. 2021; 19: 1756–1768. pmid:33774895
- 89. Cheng W, Lin M, Chu M, Xiang G, Guo J, Jiang Y, et al. RNAi-based gene silencing of RXLR effectors protects plants against the oomycete pathogen Phytophthora capsici. Mol Plant-Microbe Interact. 2022; 35: 440–449.