A polyketide synthase gene, ACRTS2, is responsible for biosynthesis of host-selective ACR-toxin in the rough lemon pathotype of Alternaria alternata.

The rough lemon pathotype of Alternaria alternata produces host-selective ACR-toxin and causes Alternaria leaf spot disease of rough lemon (Citrus jambhiri). The structure of ACR-toxin I (MW = 496) consists of a polyketide with an α-dihydropyrone ring in a 19-carbon polyalcohol. Genes responsible for toxin production were localized to a 1.5-Mb chromosome in the genome of the rough lemon pathotype. Sequence analysis of this chromosome revealed an 8,338-bp open reading frame, ACRTS2, that was present only in the genomes of ACR-toxin-producing isolates. ACRTS2 is predicted to encode a putative polyketide synthase of 2,513 amino acids and belongs to the fungal reducing type I polyketide synthases. Typical polyketide functional domains were identified in the predicted amino acid sequence, including ß-ketoacyl synthase, acyl transferase, methyl transferase, dehydratase, ß-ketoreductase, and phosphopantetheine attachment site domains. Combined use of homologous recombination-mediated gene disruption and RNA silencing allowed examination of the functional role of multiple paralogs in ACR-toxin production. ACRTS2 was found to be essential for ACR-toxin production and pathogenicity of the rough lemon pathotype of A. alternata.

ACTTS3 encoding a polyketide synthase is essential for the biosynthesis of ACT-toxin and pathogenicity in the tangerine pathotype of Alternaria alternata.

The tangerine pathotype of Alternaria alternata produces host-selective ACT-toxin and causes Alternaria brown spot disease of tangerine and tangerine hybrids. Sequence analysis of a genomic BAC clone identified part of the ACT-toxin TOX (ACTT) gene cluster, and knockout experiments have implicated several open reading frames (ORF) contained within the cluster in the biosynthesis of ACT-toxin. One of the ORF, designated ACTTS3, encoding a putative polyketide synthase, was isolated by rapid amplification of cDNA ends and genomic/reverse transcription-polymerase chain reactions using the specific primers designed from the BAC sequences. The 7,374-bp ORF encodes a polyketide synthase with putative beta-ketoacyl synthase, acyltransferase, methyltransferase, beta-ketoacyl reductase, and phosphopantetheine attachment site domains. Genomic Southern blots demonstrated that ACTTS3 is present on the smallest chromosome in the tangerine pathotype of A. alternata, and the presence of ACTTS3 is highly correlated with ACT-toxin production and pathogenicity. Targeted gene disruption of two copies of ACTTS3 led to a complete loss of ACT-toxin production and pathogenicity. These results indicate that ACTTS3 is an essential gene for ACT-toxin biosynthesis in the tangerine pathotype of A. alternata and is required for pathogenicity of this fungus.

Function of genes encoding acyl-CoA synthetase and enoyl-CoA hydratase for host-selective act-toxin biosynthesis in the tangerine pathotype of Alternaria alternata.

The tangerine pathotype of Alternaria alternata produces host-selective ACT-toxin and causes Alternaria brown spot disease. Sequence analysis of a genomic cosmid clone identified a part of the ACTT gene cluster and implicated two genes, ACTT5 encoding an acyl-CoA synthetase and ACTT6 encoding an enoyl-CoA hydratase, in the biosynthesis of ACT-toxin. Genomic Southern blots demonstrated that both genes were present in tangerine pathotype isolates producing ACT-toxin and also in Japanese pear pathotype isolates producing AK-toxin and strawberry pathotype isolates producing AF-toxin. ACT-, AK-, and AF-toxins from these three pathotypes share a common 9,10-epoxy-8-hydroxy-9-methyl-decatrienoic acid moiety. Targeted gene disruption of two copies of ACTT5 significantly reduced ACT-toxin production and virulence. Targeted gene disruption of two copies of ACTT6 led to complete loss of ACT-toxin production and pathogenicity and a putative decatrienoic acid intermediate in ACT-toxin biosynthesis accumulated in mycelial mats. These results indicate that ACTT5 and ACTT6 are essential genes in ACT-toxin biosynthesis in the tangerine pathotype of A. alternata and both are required for full virulence of this fungus.

Functional analysis of a multicopy host-selective ACT-toxin biosynthesis gene in the tangerine pathotype of Alternaria alternata using RNA silencing.

Alternaria brown spot, caused by the tangerine pathotype of Alternaria alternata, is a serious disease of commercially important tangerines and their hybrids. The pathogen produces host-selective ACT toxin, and several genes (named ACTT) responsible for ACT-toxin biosynthesis have been identified. These genes have many paralogs, which are clustered on a small, conditionally dispensable chromosome, making it difficult to disrupt entire functional copies of ACTT genes using homologous recombination-mediated gene disruption. To overcome this problem, we attempted to use RNA silencing, which has never been employed in Alternaria spp., to knock down the functional copies of one ACTT gene with a single silencing event. ACTT2, which encodes a putative hydrolase and is present in multiple copies in the genome, was silenced by transforming the fungus with a plasmid construct expressing hairpin ACTT2 RNAs. The ACTT2 RNA-silenced transformant (S-7-24-2) completely lost ACTT2 transcripts and ACT-toxin production as well as pathogenicity. These results indicated that RNA silencing may be a useful technique for studying the role of ACTT genes responsible for host-selective toxin biosynthesis in A. alternata. Further, this technique may be broadly applicable to the analysis of many genes present in multiple copies in fungal genomes that are difficult to analyze using recombination-mediated knockdowns.

Overexpression of a Gene Encoding a Catabolite Repression Element in Alternaria citri Causes Severe Symptoms of Black Rot in Citrus Fruit.

ABSTRACT A gene (AcCreA) encoding a catabolite repression element (CreA) with (two zinc fingers of the Cys(2)His(2) type was isolated from the postharvest fungal pathogen Alternaria citri. The AcCreA overexpression mutant AcOEC2 of A. citri showed normal growth on pectin medium and on segments of peel or the juice sac area from citrus fruit. Production of endopolygalacturonase, an essential virulence factor of this pathogen, was similar in AcOEC2 and the wild type in pectin-containing media. However, addition of glucose to the medium showed that carbon catabolite repression of endopolygalacturonase gene (Acpg1) expression, as well as endopolygalacturonase production, was lost in AcOEC2. The wild-type strain of A. citri causes rot mainly in the central axis of citrus fruit without development of rotting in the juice sac area; however, AcOEC2 caused severe black rot symptoms in both the central axis and juice sac areas. These results indicate that AcCreA-mediated catabolite repression controls the virulence or infection of this pathogen, and that the wild-type A. citri does not cause symptoms in the juice sac area due to carbon catabolite repression by sugars in the juice of the juice sac area.

A Virulence-Reducing Mutation in the Postharvest Citrus Pathogen Alternaria citri.

ABSTRACT Alternaria citri causes Alternaria black rot, a postharvest fruit disease, on a broad range of citrus cultivars. We previously described that an endopolygalacturonase minus mutant of A. citri caused significantly less black rot in citrus fruit. To search for other essential factors causing symptoms in addition to endopolygalacturonase, a random mutation analysis of pathogenicity was performed using restriction enzyme-mediated integration. Three isolates among 1,694 transformants of A. citri had a loss in pathogenicity in a citrus peel assay, and one of these three mutants was a histidine auxotroph. Gene AcIGPD that encodes imidazole glycerol phosphate dehydratase, the sixth enzyme in the histidine biosynthetic pathway, was cloned, and the mutant containing the disrupted target gene, AcIGPD, caused less black rot.

An Isolate of Alternaria alternata That Is Pathogenic to Both Tangerines and Rough Lemon and Produces Two Host-Selective Toxins, ACT- and ACR-Toxins.

ABSTRACT Two different pathotypes of Alternaria alternata cause Alternaria brown spot of tangerines and Alternaria leaf spot of rough lemon. The former produces the host-selective ACT-toxin and the latter produces ACR-toxin. Both pathogens induce similar symptoms on leaves or young fruits of their respective hosts, but the host ranges of these pathogens are distinct and one pathogen can be easily distinguished from another by comparing host ranges. We isolated strain BC3-5-1-OS2A from a leaf spot on rough lemon in Florida, and this isolate is pathogenic on both cv. Iyokan tangor and rough lemon and also produces both ACT-toxin and ACR-toxin. Isolate BC3-5-1-OS2A carries both genomic regions, one of which was known only to be present in ACT-toxin producers and the other was known to exist only in ACR-toxin producers. Each of the genomic regions is present on distinct small chromosomes, one of 1.05 Mb and the other of 2.0 Mb. Alternaria species have no known sexual or parasexual cycle in nature and populations of A. alternata on citrus are clonal. Therefore, the ability to produce both toxins was not likely acquired through meiotic or mitotic recombination. We hypothesize that a dispensable chromosome carrying the gene cluster controlling biosynthesis of one of the host-selective toxins was transferred horizontally and rearranged by duplication or translocation in another isolate of the fungus carrying genes for biosynthesis of the other host-selective toxin.

Green Fluorescent Detection of Fungal Colonization and Endopolygalacturonase Gene Expression in the Interaction of Alternaria citri with Citrus.

ABSTRACT Alternaria citri, a postharvest pathogen, produces endopolygalacturonase (endoPG) and causes black rot on citrus fruit. We previously described that an endoPG-disrupted mutant of Alternaria citri was significantly reduced in its ability to macerate plant tissue and cause black rot symptoms on citrus. In order to investigate colonization of citrus fruit tissues by Alternaria citri, pTEFEGFP carrying a green fluorescent protein (GFP) gene was introduced into wild-type Alternaria citri and its endoPG-disrupted mutant (M60). Green fluorescence was observed in spores, germ tubes, appressoria, and infection hyphae of transformants G1 (derived from wild type) and GM4 (derived from M60). Hyphae of G1 but not GM4 vertically penetrated the peel, but the hyphae of both G1 and GM4 spread equally in the juice sac area of citrus fruit. Green fluorescence of Alternaria citri transformant EPG7 carrying a GFP gene under control of the endoPG gene promoter of Alternaria citri was induced by pectin in the peel during the infection stage, but repressed completely in the juice sac area, likely by carbon catabolite repression by sugars in the juice.

Worldwide Phylogeography of the Citrus Brown Spot Pathogen, Alternaria alternata.

ABSTRACT Sixty-five isolates of Alternaria alternata were sampled from brown spot lesions on tangerines and mandarins (Citrus reticulata) and tangerine x grapefruit (C. reticulata x C. paradisi) hybrids in the United States, Colombia, Australia, Turkey, South Africa, and Israel to investigate the worldwide phylogeography of the fungus. Genetic variation was scored at 15 putative random amplified polymorphic DNA (RAPD) loci and 465 bp of an endo-polygalacturonase (endo-PG) gene was sequenced for each isolate. Cluster analysis of RAPD genotypes revealed significant differentiation between United State and Colombia isolates and Turkey, South Africa, Israel, and Australia isolates. Sequencing of endo-PG revealed 21 variable sites when the outgroup A. gaisen (AK-toxin-producing pathogen of Japanese pear) was included and 13 variable sites among the sampled isolates. Nucleotide substitutions at 10 of 13 variable sites represented silent mutations when endo-PG was translated in frame. Eight distinct endo-PG haplotypes were found among the sampled isolates and estimation of a phylogeny with endo-PG sequence data revealed three clades, each with strong bootstrap support. The most basal clade (clade 1) was inferred based on its similarity to the outgroup A. gaisen and consisted exclusively of pathogenic isolates from the United States and Colombia. Clade 2 consisted of pathogenic and nonpathogenic isolates from the United States, Australia, South Africa, and Israel and clade 3 contained pathogenic and nonpathogenic isolates from Australia, South Africa, Israel, and Turkey. Quantitative estimates of virulence (disease incidence) were obtained for isolates from the United States, Colombia, South Africa, Israel, and Turkey by spray inoculating detached citrus leaves and counting the number of lesions 24 h after inoculation. Large differences in virulence were detected among isolates within each location and isolates from the United States were significantly more virulent than isolates from other locations. Several isolates from Colombia, South Africa, Israel, and Turkey had low virulence and 8% of all isolates were nonpathogenic. All but one of the nonpathogenic isolates were found in clade 2 of the endo-PG phylogeny, which also included the most highly virulent isolates sampled.

Cloning and nucleotide sequence of the mycodextranase gene from Streptomyces sp. J-13-3.

Mycodextranase (EC 3.2.1.61) is an alpha-glucanase that cleaves alpha-1,4-bonds of alternating alpha-1,3- and alpha-1,4-linked D-glucan (nigeran). The gene encoding mycodextranase from Streptomyces sp. J-13-3 was cloned by hybridization with a degenerate oligonucleotide probe from the amino-terminal amino acid sequence of the enzyme and its nucleotide structure was analyzed. The open reading frame consisted of 1,803 base pairs encoding a signal peptide of 60 amino acids and a mature protein of 540 amino acids with a calculated molecular weight of 56,078. The deduced amino acid sequence showed weak similality to a chitinase homolog from Streptomyces lividans and a chitinase from Xanthomonas sp.

Endopolygalacturonase is essential for citrus black rot caused by Alternaria citri but not brown spot caused by Alternaria alternata.

Alternaria citri, the cause of Alternaria black rot, and Alternaria alternata rough lemon pathotype, the cause of Alternaria brown spot, are morphologically indistinguishable pathogens of citrus: one causes rot by macerating tissues and the other causes necrotic spots by producing a host-selective toxin. To evaluate the role of endopolygalacturonase (endoPG) in pathogenicity of these two Alternaria spp. pathogens, their genes for endoPG were mutated by gene targeting. The endoPGs produced by these fungi have similar biochemical properties, and the genes are highly similar (99.6% nucleotide identity). The phenotypes of the mutants, however, are completely different. An endoPG mutant of A. citri was significantly reduced in its ability to cause black rot symptoms on citrus as well as in the maceration of potato tissue and could not colonize citrus peel segments. In contrast, an endoPG mutant of A. alternata was unchanged in pathogenicity. The results indicate that a cell wall-degrading enzyme can play different roles in the pathogenicity of fungal pathogens. The role of a cell wall-degrading enzyme depends upon the type of disease but not the taxonomy of the fungus.

Molecular cloning of a homologue of dad-1 gene in citrus: distinctive expression during fruit development.

A cDNA homologue to the human defender against apoptotic death gene (dad-1), which is involved in programmed cell death, was isolated from satsuma mandarin (Citrus unshiu Marc.) fruit. It (Citdad-1-1) was 345 bp long, with a deduced protein sequence of 115 amino acids. Southern hybridization suggests that dad-1-related sequences are present as a small gene family in the citrus genome. Expression of Citdad-1-1 was progressively down-regulated in leaves as they matured, but not in juice sac/segment epidermis (edible part) towards fruit ripening. The role of dad-1 during citrus development is also discussed.

Distribution and Characterization of AKT Homologs in the Tangerine Pathotype of Alternaria alternata.

ABSTRACT The tangerine pathotype of Alternaria alternata produces a host-selective toxin (HST), known as ACT-toxin, and causes Alternaria brown spot disease of citrus. The structure of ACT-toxin is closely related to AK- and AF-toxins, which are HSTs produced by the Japanese pear and strawberry pathotypes of A. alternata, respectively. AC-, AK-, and AF-toxins are chemically similar and share a 9,10-epoxy-8-hydroxy-9-methyl-decatrienoic acid moiety. Two genes controlling AK-toxin biosynthesis (AKT1 and AKT2) were recently cloned from the Japanese pear pathotype of A. alternata. Portions of these genes were used as heterologous probes in Southern blots, that detected homologs in 13 isolates of A. alternata tangerine pathotype from Minneola tangelo in Florida. Partial sequencing of the homologs in one of these isolates demonstrated high sequence similarity to AKT1 (89.8%) and to AKT2 (90.7%). AKT homologs were not detected in nine isolates of A. alternata from rough lemon, six isolates of nonpathogenic A. alternata, and one isolate of A. citri that causes citrus black rot. The presence of homologs in the Minneola isolates and not in the rough lemon isolates, nonpathogens or black rot isolates, correlates perfectly to pathogenicity on Iyo tangerine and ACT-toxin production. Functionality of the homologs was demonstrated by detection of transcripts using reverse transcription-polymerase chain reaction (RT-PCR) in total RNA of the tangerine pathotype of A. alternata. The high sequence similarity of AKT and AKT homologs in the tangerine patho-type, combined with the structural similarity of AK-toxin and ACT-toxin, may indicate that these homologs are involved in the biosynthesis of the decatrienoic acid moiety of ACT-toxin.

Density Gradient Study of Victorin-Binding Proteins in Oat (Avena sativa) Cells.

Victorin-binding proteins (VBPs) in oat (Avena sativa) cells were identified using native victorin and anti-victorin polyclonal antibodies. Homogenates of oat tissues were fractionated in continuous or discontinuous sucrose density gradients or with an aqueous two-phase method, and covalent binding sites of victorin were detected by western blotting. In a 20 to 45% (w/w) sucrose continuous density gradient, the 100-kD VBP was located in fractions of 37 to 44% sucrose, with a peak at 39% sucrose. Based on marker enzyme assays, plasma membranes peaked at 39 to 41% sucrose, mitochondria peaked at 41%, but Golgi and endoplasmic reticulum were in lower density fractions, peaking at 28 to 29% and 22 to 24% sucrose, respectively. The 100-kD VBP was not found in plasma membranes purified by the aqueous two-phase method or in mitochondria purified by discontinuous density gradient centrifugation. Victorin binding to 65- and 45-kD proteins was detected in all fractions in the continuous sucrose density gradients. The 65- and 45-kD proteins were both detected in purified plasma membranes, but only the 65-kD protein was detected in purified mitochondria. The subcellular location of VBPs was the same in sensitive and resistant oat cells.

Covalent binding sites of victorin in oat leaf tissues detected by anti-victorin polyclonal antibodies.

Polyclonal antibodies against victorin, the host-specific toxin produced by Cochliobolus victoriae, were raised in rabbits immunized with a victorin-bovine serum albumin conjugate. The antibodies were purified from serum by protein A column chromatography and characterized by indirect and direct enzymelinked immunosorbent assays (ELISA). The concentration of victorin that inhibited anti-victorin antibody binding by 50% was 10 nanograms per milliliter in an indirect ELISA. The lowest concentration of victorin detectable was 10 picograms per milliliter. In a direct ELISA, 25 nanograms per milliliter of victorin inhibited binding of victorin-horseradish peroxidase conjugate by 50%. In vivo and in vitro covalent binding of victorin to proteins in susceptible and resistant oat (Avena sativa) tissue was examined by western blotting assays using anti-victorin antibody and a second antibody conjugated with (125)I or alkaline phosphatase. In vivo binding of victorin to proteins of 100 and 45 kilodaltons was observed in both susceptible and resistant cultivars of oats. Victorin also bound in vitro to proteins of 100, 65, and 45 kilodaltons in both susceptible and resistant oats. The data indicate that victorin binds covalently to the same sites in susceptible and resistant genotypes of oats.

Host-Specific Effects of Toxin from the Rough Lemon Pathotype of Alternaria alternata on Mitochondria.

Host-specific toxin from the rough lemon (Citrus jambhiri Lush) pathotype of Alternaria alternata (ACR toxin) was tested for effects on mitochondria isolated from several citrus species. The toxin caused uncoupling of oxidative phosphorylation and changes in membrane potential in mitochondria from leaves of the susceptible host (rough lemon); the effects differed from those of carbonylcyanide-m-chlorophenylhydrazone, a typical protonophore. ACR toxin also inhibited malate oxidation, apparently because of lack of NAD(+) in the matrix. In contrast, the toxin had no effect on mitochondria from citrus species (Dancy tangerine and Emperor mandarin [Citrus reticulata Blanco], and grapefruit [Citrus paradisi Macf.]) that are not hosts of the fungus. The effects of the toxin on mitochondria from rough lemon are similar to the effects of a host-specific toxin from Helminthosporium maydis (HMT) on mitochondria from T-cytoplasm maize. Both ACR and HMT toxins are highly selective for the respective host plants. HMT toxin and methomyl had no effect (toxic or protective) on the activity of ACR toxin against mitochondria from rough lemon.

A simple method for preparing physiologically active mitochondria from plant leaves rich in oils and phenolics.

Amberlite XAD-7, a nonionic polyacrylate adsorbent, was found to be a very effective protectant for isolating mitochondria from tissues rich in oils and phenolics. Physiologically active, well-coupled mitochondria were successfully prepared from young green leaf tissues of citrus, apple, pear and tobacco.