• pathogenesis mechanism;
  • pathogenesis regulator;
  • necrotrophic fungus;
  • plant response to pathogen;
  • gene expression profile;
  • gene induction;
  • Alternaria brassicicola ;
  • Brassica oleracea


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Alternaria brassicicola is a successful saprophyte and necrotrophic plant pathogen. To identify molecular determinants of pathogenicity, we created non-pathogenic mutants of a transcription factor-encoding gene, AbPf2. The frequency and timing of germination and appressorium formation on host plants were similar between the non-pathogenic ∆abpf2 mutants and wild-type A. brassicicola. The mutants were also similar in vitro to wild-type A. brassicicola in terms of vegetative growth, conidium production, and responses to a phytoalexin, reactive oxygen species and osmolites. The hyphae of the mutants grew slowly but did not cause disease symptoms on the surface of host plants. Transcripts of the AbPf2 gene increased exponentially soon after wild-type conidia contacted their host plants . A small amount of AbPf2 protein, as monitored using GFP fusions, was present in young, mature conidia. The protein level decreased during saprophytic growth, but increased and was located primarily in fungal nuclei during pathogenesis. Levels of the proteins and transcripts sharply decreased following colonization of host tissues beyond the initial infection site. When expression of the transcription factor was induced in the wild-type during early pathogenesis, 106 fungal genes were also induced in the wild-type but not in the ∆abpf2 mutants. Notably, 33 of the 106 genes encoded secreted proteins, including eight putative effector proteins. Plants inoculated with ∆abpf2 mutants expressed higher levels of genes associated with photosynthesis, the pentose phosphate pathway and primary metabolism, but lower levels of defense-related genes. Our results suggest that AbPf2 is an important regulator of pathogenesis, but does not affect other cellular processes in A. brassicicola.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Some biotrophic plant-pathogenic fungi are obligate parasites of living plant tissue. They overcome host plant immunity by suppressing plant defenses (Buttner and Bonas, 2003; Glazebrook, 2005; Sexton and Howlett, 2006). In contrast, necrotrophic pathogens use nutrients derived from dead host plant tissue, or kill host tissue before colonizing it. They are facultative parasites, able to complete their life cycles either as saprophytes on dead organic matter or as pathogens in killed host plant tissue. It is unclear which gene or genes determine whether a facultative fungus is saprophytic or parasitic.

Both biotrophic and necrotrophic fungi respond to environmental conditions, such as nutrient deprivation, osmotic stress, defense metabolites and reactive oxygen species (ROS) in host plant tissue. The reactions of pathogenic fungi on host plants are of special interest to plant pathologists as the reactions are directly associated with the mechanisms of pathogenesis (Shlezinger et al., 2011; Williams et al., 2011). Several molecular sensing mechanisms in fungi lead to complex changes in their cell physiology and in infection structures that promote fungal pathogenesis (Hoch et al., 1987; Kulkarni et al., 2005; Kloda et al., 2008). The speed of the fungal response to its host plants at the gene level and the regulation of this response are two fascinating but challenging areas of study.

Alternaria brassicicola is a necrotrophic fungus that causes black spot disease of cultivated brassicas, such as cabbage (Brassica oleracea), canola (B. napus L. or B. rapa var.) and mustard (B. campestris L.). It is also pathogenic on Arabidopsis thaliana, and the Alternaria brassicicolaArabidopsis thaliana system is occasionally used to study host–pathogen interactions (Thomma et al., 1999; Oh et al., 2005). A conserved mitogen-activated protein kinase encoded by Amk1, and its downstream transcription factor encoded by AbSte12, are essential for pathogenesis (Cho et al., 2007, 2009). However, mutation of either Amk1, AbSte12 or their homologs in other fungi caused defects in other cellular processes in addition to the loss of pathogenicity (Caracuel et al., 2003; Cho et al., 2007, 2009; You and Chung, 2007; Zhao et al., 2007; Wong Sak Hoi and Dumas, 2010). Most transcription factors associated with pathogenesis in parasitic fungi are also linked to additional cellular processes (Kim et al., 2009; Lin et al., 2009; Guo et al., 2011; Joubert et al., 2011; Son et al., 2011; Wang et al., 2011; Chen et al., 2012). Because of the multiple functions of transcription factors, it is difficult to identify the pathogenicity genes regulated by them. In this study, we identified a transcription factor gene that was essential for initiating pathogenesis in A. brassicicola. Targeted mutants of the gene were non-pathogenic and grew on the surface of host tissue without causing disease symptoms. We tested three hypotheses: whether the transcription factor was important for fungal development, for neutralization of defense chemicals, or for aggressiveness towards the host plant.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Mutant strains of the AbPf2 gene

We replaced a gene (AB06533.1) with a hygromycin B (HygB) resistance cassette. The gene was predicted to encode 630 amino acids. It included a GAL4 (Zn2Cys6) fungal-specific DNA binding domain (Interpro ID: IPR001138, Pfam ID: PF00172) as predicted by Interpro (Bateman et al., 2004). Homologs of the gene are annotated as either a hypothetical protein or a C6 zinc finger domain protein in other fungal genomes. We named this gene AbPf2 (Alternaria brassicicola pathogenicity factor 2). Southern hybridization verified that the AbPf2 coding region was replaced by two copies of the HygB resistance cassette in one mutant (Δabpf2-5), and by one copy in six mutants (Figure S1b, left blots). We also produced a second set of mutants whose AbPf2 gene was replaced by a GFP-HygB resistance cassette. Southern hybridization verified modifications of Abpf2 gene, resulted in three replacement mutants (s4-s6) and three ectopic insertion mutants (s1-s3) (Figure S1b, right blots).

Loss of pathogenicity in the Δabpf2 mutants

Pathogenicity assays were initially performed using five strains of mutant (Δabpf2-1, -2, -4, -5 and -s5) on green cabbage (Brassica oleracea). All five strains were non-pathogenic on the leaves of 6-8-week-old plants. We performed further experiments with Δabpf2-2 and Δabpf2-s5, representing two groups of mutants. We also tested the pathogenicity of the mutants on wild-type A. thaliana, ecotype Col-0, and the A. thaliana mutant pad3. The pad3 mutant lacks the phytoalexin camalexin (Zhou et al., 1999). Inoculations with 1–2 × 103 conidia of wild-type A. brassicicola produced large lesions on the leaves of green cabbage (Figure 1a), and the leaves of the pad3 mutants were extensively colonized by 5 days post-inoculation (dpi) (Figure 1b). The wild-type killed whole plants of the Col-0 ecotype when they were drenched with a conidial suspension of 5 × 105/ml (Figure 1c). In contrast, none of the Δabpf2 mutants caused disease symptoms on healthy young leaves of green cabbage, the pad 3 mutant, or wild-type Arabidopsis (Figure 1). The Δabpf2 mutants occasionally caused mild disease symptoms on senescing leaves. In these rare cases, the symptoms were not apparent until 4–6 dpi, compared to 24 h post-inoculation (hpi) with wild-type inoculum. In addition, the small spots caused by the mutants did not expand beyond the initial infection site (Figure 1b). The mutants did colonize host leaves wounded before inoculation, but lesion expansion was so slow that they rarely grew beyond the wounds during the 6-day assay period (Figure 1a, center and right images).


Figure 1. Loss of pathogenicity in Δabpf2 mutants.

(a) Lesions on B. oleracea leaves at 5 days post-inoculation with 1000 conidia in 10 μl water.

(b) Lesions on the A. thaliana pad3 mutant (Col-0 background).

(c) Lesions on A. thaliana, ecotype Col-0.

(d) Restoration of virulence in two mutants by complementing them with the wild-type allele of the AbPf2 gene.

Abbreviations: wt, wild-type Alternaria brassicicola; Δabpf2, AbPf2 deletion mutant; c–2 and c–s5, complemented mutants Δabpf2–2:AbPf2 and Δabpf2–s5:AbPf2, respectively; injured, injured by scratching with a pipette tip; Col–0, A. thaliana ecotype Columbia–0.

Download figure to PowerPoint

Restoration of pathogenicity by complementation

Two strains of the non-pathogenic mutants Δabpf2-s5 and Δabpf2-2 were independently complemented using either single or multiple copies of the wild-type allele (Figure S1). Pathogenicity of the complemented mutants, Δabpf2-s5:AbPf2 and Δabpf2-2:AbPf2, was restored on green cabbage and A. thaliana (Figure 1). We quantified the virulence of the complemented mutants by comparing the diameters of lesions caused by the complemented mutants and wild-type A. brassicicola on green cabbage leaves. The lesion sizes were comparable (Figure 1d and Table 1). These results indicated that AbPf2 is essential for pathogenicity and full virulence.

Table 1. Restoration of pathogenicity and full virulence by complemented mutants
 Lesion size (mm)
  1. Numbers indicate mean diameters of lesions and their standard deviations on 6-week-old green cabbage plants.

Wild-type17.3 ± 4.9
Δabpf2-s50.0 ± 0.0
Δabpf2-s5:AbPf2-c1018.5 ± 3.6
Wild-type19.8 ± 3.7
Δabpf2-20.0 ± 0.0
Δabpf2-s5:AbPf2-c216.9 ± 3.9

Germination and appressorium formation

We tested whether the loss of pathogenicity in Δabpf2 was associated with developmental defects. Spore germination by the mutants and wild-type was approximately 100% at 3 hpi. There were no differences in either germination, vegetative growth (Figure 2) or spore production on potato dextrose agar (PDA) (Table S1). Germination rates for both the wild-type and the Δabpf2 mutants were also similar on host plant leaves. Germination was approximately 100% by 8 hpi on 6-8-week-old green cabbage, and 12 hpi on 5-6-week-old Arabidopsis (Figure 3a,e). The lengths of their germ tubes and hyphae were so variable that the differences in mean lengths were statistically insignificant at 12, 18 and 20 hpi (Table S2). A small, swollen structure (appressorium) occasionally formed at the tips of germ tubes at 8–12 hpi on green cabbage and 12–24 hpi on Arabidopsis. By 20 hpi, appressoria had formed on approximately 45% of the germ tubes produced by the mutants and the wild-type fungus on both host plants (Figure S2 and Table S3). In summary, the mutants germinated and formed appressoria similar to the wild-type on the surface of both host plants.


Figure 2. Pharmacological tests. Growth of wild-type and mutant on PDA containing 0.1 mm brassinin, 2 m sorbitol, 15 mm H2O2, 10 mm KO2, 0.7 m KCl or 0.5 m NaCl.

Download figure to PowerPoint


Figure 3. Hyphal growth of the ∆abpf2 mutant and wild-type Alternaria brassicicola on the leaves of green cabbage.

(a) Wild-type at 12 hpi

(b) wild-type at 24 hpi

(c) wild-type at 30 hpi

(d) ∆abpf2 at 12 hpi

(e) ∆abpf2 at 24 hpi

(f) ∆abpf2 at 30 hpi. Fungal tissues were stained with trypan blue.

(g) ∆abpf2 at 67 hpi; Several appressoria are indicated by arrows.

(h) Wild-type at 67 hpi. Aerial hyphae (A), their point of emergence (E), and hyphae within plant tissues (H) are shown.

Fungal tissues in (g) and (h) were stained with calcofluor white. Areal hyphae that have emerged from plant tissue appear bright white, while hyphae in the plant tissue are vague but visible.

Download figure to PowerPoint

Mutant growth on plant surfaces

We compared plant tissue at infection sites and the growth patterns of the ∆abpf2 mutants and wild-type A. brassicicola on green cabbage and Arabidopsis. Wild-type A. brassicicola created sunken infection sites on Arabidopsis leaves by approximately 48 hpi. At this time, both plant tissue and wild-type fungal hyphae were stained with trypan blue (Figure S3d). In contrast, the plant tissue infected with the ∆abpf2 mutants was not sunken nor was it stained by trypan blue (Figure S3h). However, the mutants continued growing on the leaves of Arabidopsis (Figure S3e–h) and green cabbage (Figure 3) during this time.

No changes in colony or conidial response to stressors

Plants produce ROS and phytoalexins in response to pathogen infection (VanEtten et al., 1995; Thomma et al., 1999; Torres et al., 2006; Ahuja et al., 2012). In addition, the infection court contains various osmolites. To indirectly evaluate the importance of the AbPf2 gene in managing stressors, we evaluated the effects of two oxygen stressors, the phytoalexin brassinin, and two osmolites on the vegetative growth of Δabpf2 mutants. Colony size and growth patterns of the Δabpf2-2 and Δabpf2-s5 mutants were comparable to those of their complemented mutants and the wild-type on PDA containing any of the tested chemicals (Figure 2).

AbPf2 expression during plant infection

We investigated the temporal expression pattern of AbPf2 transcripts in wild-type A. brassicicola during pathogenesis. The level of AbPf2 transcripts was lowest during saprophytic growth in necrotic tissue (late-stage colonization) and slightly higher in conidia harvested from PDA (Figure 4). Compared to conidia before inoculation, transcription levels increased over 30-fold (< 0.001) in pre-germination conidia at 4 hpi on both green cabbage and A. thaliana. These increased levels were maintained only until the initial infection site was colonized. Gene expression levels fell sharply by the time lesions had expanded beyond the initial infection site. This occurred at 24 hpi on green cabbage and 48 hpi on Arabidopsis. Gene expression remained low throughout lesion expansion and saprophytic growth in necrotic tissue. Expression of the gene increased again during conidiation at approximately 9 dpi on green cabbage leaves. The increase in gene expression was not obvious on Arabidopsis, where fewer conidia were produced (Figure 4b).


Figure 4. Expression of the AbPf2 gene during pathogenesis in the host plant green cabbage (a) and A. thaliana (Col-0) (b).

The y axes illustrate the relative quantity of the transcripts compared to the mean of two housekeeping genes: glyceraldehyde 3-phosphate dehydrogenase and elongation factor 1-α. The x axes show the number of hours post-inoculation. Error bars indicate standard deviation. Numbers below the charts: 1, before germination; 2, germination; 3, appressorium formation; 4, lesion expansion (early stage of saprophytic growth); 5, saprophytic growth; 6, extensive conidium formation.

Download figure to PowerPoint

Localization of AbPf2 protein during saprophytic growth

We monitored the expression and localization of AbPf2 protein using an AbPf2–GFP fusion protein. The mutant expressing the fusion protein was similar to wild-type A. brassicicola in terms of germination and vegetative growth on PDA. To reveal the nuclei in fungal cells, we used a fusion protein of mCherry and a nuclear localization signal (mCherry–NLS) (Khang et al., 2010). In hyphal tips growing on PDA, mCherry–NLS was moderately expressed, but the AbPf2–GFP protein was hard to detect (Figure 5a and Table 2). Unlike the growing hyphal tips, little AbPf2 protein accumulated in young mature conidia on 3-day-old plates (Table 2). When the conidia were inoculated on PDA, they germinated and began to grow in approximately 3 h. During this time, the GFP signal moved into the germ tube and became weaker as the germ tube grew (Figure 5c,d and Table 2).

Table 2. Green fluorescent protein signal strength indicating the relative amount of AbPf2–GFP protein in conidia, germ tubes and hyphae during growth on potato dextrose agar
 Hyphal tips in PDAConidia (0 hpi)Germ tubes (4hpi)Hyphae (24 hpi)
  1. a

    Numbers indicate fluorescence signal strength measured by pixels in the region of interest using the FV10-ASW 2.0 viewer software (Olympus:

  2. b

    P value: probability of statistical significance for the difference in mean signal strength between the indicated stage and the preceding stage.

  3. c

    % change: relative signal strength at the indicated stage compared to the preceding stage.

 P valueb0.
% changec466.919.2–28.1–69.8–3.273.6
 P value0.

Figure 5. AbPf2 protein expression and localization during germination on potato dextrose agar (PDA).

(b) Conidia before inoculation.

(c) Germinating conidium at 4 h after transfer.

(d) Hyphal growth at 24 h after transfer.

All images were acquired under the same scanning conditions. Green represents the AbPf2–GFP fusion protein, and pink represents nuclei labeled by mCherry fused to a nuclear localization signal (mCherry–NLS). Composites represent overlays of green, pink and light microscope images after sequential scanning.

Download figure to PowerPoint

Expression and localization of AbPf2 protein during pathogenesis

We also monitored AbPf2 protein expression and localization during the disease process by measuring GFP signal strength. When conidia of the AbPf2–GFP-expressing mutant were inoculated on A. thaliana, the GFP signal increased approximately twofold (< 2.3E-04) compared to the GFP level in the inoculum before germination (Figure 6a,b and Table 3). The GFP signal increased in both the cytoplasm and nuclei. After germination, a stage comparable to 8–12 hpi in Figure 4(b), the GFP signal increased even more in the germ tubes and the appressoria (Figure 6c,d). The GFP signal strength was approximately threefold higher in the nuclei than in the cytoplasm at the appressorium-forming stage (Table 3). When the inoculation site became sunken, but fungal hyphae had not grown beyond the initial site, the GFP signal was reduced over twofold and stayed mainly in nuclei of the hyphae (Figure 6e and Table 3). By the time hyphae had colonized dead plant tissue beyond the initial infection site (48 hpi, Figure 4b), the GFP signal had decreased further and was weakly detected in nuclei of the hyphae (Figure 6f and Table 3). AbPf2–GFP was also localized in the nuclei during early infection in green cabbage (Figure 7a). As a negative control, we monitored GFP alone, constitutively expressed under the control of the ToxA promoter (Ciuffetti et al., 1997; Lorang et al., 2001). The GFP was evenly distributed in the cytoplasm (Figure 7b,c) and did not co-localize with 4′,6′-diamidino-2-phenylindole (DAPI) in the nuclei (Figure 7c).

Table 3. Green fluorescent protein signal strength indicating the relative amount of AbPf2–GFP in the nuclei and cytoplasm of conidia and hyphae during growth on Arabidopsis thaliana
 Conidia (0 hpi)Before germinationGerm tube + appressoriumInfection at initial siteColonization of large area
  1. a

    Numbers indicate fluorescence signal strength measured by pixels in the region of interest using the FV10-ASW 2.0 viewer software.

  2. b

    P value: probability of statistical significance for the difference in mean signal strength between the indicated stage and the preceding stage.

  3. c

    % change: relative signal strength at the indicated stage compared to the preceding stage.

P valueb
% changec  111.6–6.196.820.5–59.0–28.2–87.5–41.5
P value
% change  128.0109.436.3–93.2–65.1141.9–95.4832.9

Figure 6. Confocal microscope images showing AbPf2–GFP fusion protein expression and localization in A. thaliana.

(a) Conidia before inoculation.

(b) Conidia near germination.

(c) Conidia immediately after germination.

(d) Conidium with a germ tube and an appressorium. The arrow indicates an appressorium at the tip of a germ tube. Arrowheads indicate autofluorescence of plant tissue.

(e) Invading hyphae in the plant tissue before disease spots expanded beyond the initial infection site.

(f) Conidia and colonizing hyphae in necrotic plant tissues.

Pink indicates mCherry–NLS in nuclei.

Download figure to PowerPoint


Figure 7. Localization of AbPf2–GFP fusion protein and GFP(green fluorescent protein).

All confocal images were acquired during host plant infection on 4-week-old green cabbage.

(a) Mutant strain expressing the AbPf2–GFP fusion protein under the control of the AbPf2 promoter.

(b) Mutant strain expressing (GFP) under the control of the ToxA promoter.

(c) Colonizing hyphae of the mutant strain expressing GFP under the control of ToxA promoter.

Download figure to PowerPoint

Gene expression changes in Δabpf2 mutants

To identify pathogenicity-associated genes regulated by the transcription factor, we compared gene expression profiles between the Δabpf2 mutant and wild-type A. brassicicola at two time points, 12 and 48 hpi. From tissue samples at 12 hpi, when AbPf2 transcripts reached their highest level (Figure 4), a total of 192 and 220.7 million sequence tags were produced for the wild-type and Δabpf2 mutant, respectively. Of these, 8.5 × 105 (0.55%) and 9.3 × 105 (0.53%) mapped to the genome of A. brassicicola, and 158.4 × 106 (82.5%) and 189.8 × 106 (86.0%) mapped to the genome of A. thaliana, respectively. We identified differentially expressed genes using fungal tags. Among 10 688 predicted genes in the A. brassicicola genome, 106 genes were expressed at a significantly lower level and 62 genes were expressed at a significantly higher level (more than twofold, < 0.05) in the Δabpf2 mutant compared to the wild-type (Figure 8 and Table S4). These 106 and 62 genes represented 0.99 and 0.58%, respectively, of the predicted genes in the current A. brassicicola genome. Although sequence tags of the AbPf2 gene were present in the ∆abpf2 mutants, all of them mapped to the 3′ side of the coding region that was part of the deletion construct (Figure S1). There were no full-length transcripts. Among the 106 genes expressed at lower levels, 33 encoded proteins with secretion signal peptides (Figure 8 and Table 4). They included genes encoding two pectate lyases, a necrosis-inducing factor, and eight small proteins with 88–147 amino acids.

Table 4. List of fungal genes differentially expressed in the Δabpf2 mutants compared to the wild-type at 12 hpi
  Number of tags at 12 hpiNumber of tags at 48 hpi
Protein IDGO annotationWild-typeabpf2log2 ratioWild-typeabpf2log2 ratio
  1. Asterisks indicate that the P value is statistically significant (< 0.05). (S), secretion protein.

  2. # NUM indicates an unlimited number generated by a 0 denominator.

AB04512Effector 4512 (S)4210713−2.611740−1.5
AB09024Effector 9024 (S)363613−8.1753*0*#NUM
AB04813GO:0030570 pectate lyase (S)324590−5.290#NUM
AB01151 2751623−2.1338116−1.5
AB01161Effector 1161 (S)2640802−1.715421.5
AB01332GO:0030570 pectate lyase (S)2473153−4.0129*0*#NUM
AB05127GO:0016846 carbon–sulfur lyase1794396−2.2896575−0.6
AB06363 1767814−1.122760−1.9
AB06895GO:0005840 ribosome1739821−1.12617917−1.5
AB02504GO:0005840 ribosome1482717−1.02035732−1.5
AB09632GO:0006508 proteolysis (S)147175−4.3731*28*−4.7
AB10586Effector 10586 (S)14070#NUM1271*0*#NUM
AB09633GO:0016787 hydrolase activity106983−3.7397*24*−4.1
AB07442 992375−1.4502190−1.4
AB05161Effector 5156 (S)988407−1.3534315−0.8
AB09034 906180−2.3737256−1.5
AB10561 770161−2.31125*290*−2.0
AB05512GO:0005215 transporter activity728347−1.11377*384*−1.8
AB10354GO:0005215 transporter activity712326−1.11166*190*−2.6
AB09033 694207−1.7192127−0.6
AB09035GO:0016787 hydrolase activity67593−2.9125*11*−3.6
AB08663GO:0004364 glutathione transferase activity (S)652171−1.9988*237*−2.1
AB08641 573123−2.2772*165*−2.2
AB01959GO:0018667 cyclohexanone monooxygenase activity; GO:0006118 electron transport512207−1.313477−0.8
AB04781 505167−1.6345187−0.9
AB08526GO:0016020 membrane (S)470124−1.9210151−0.5
AB05176 45874−2.613680−0.8
AB04295Effector 4295 GO:0003723 RNA binding (S)45261−2.938016−4.6
AB05160GO:0000293 ferric chelate reductase activity440173−1.3213161−0.4
AB06101GO:0000166 nucleotide binding; GO:0017111 nucleoside-triphosphatase activity438167−1.4358*78*−2.2
AB09460GO:0016215 CoA desaturase activity; GO:0018688 1,1,1-trichloro-2,2-bis-(4-chlorophenyl) ethane 2,3-dioxygenase activity415199−1.157610.1
AB05390GO:0004190 aspartic-type endopeptidase activity (S)40896−2.118595−1.0
AB07069 399190−1.1390252−0.6
AB02743GO:0008812 choline dehydrogenase activity397113−1.8969*209*−2.2
AB06548 37988−2.1326126−1.4
AB01405GO:0017111 nucleoside-triphosphatase activity36541−3.1141*19*−2.9
AB05415GO:0005351 sugar:hydrogen symporter activity36190−2.018200.2
AB05741GO:0005634 nucleus360166−1.12013−0.6
AB07340GO:0016491 oxidoreductase activity35026−3.795*8*−3.5
AB03651GO:0005215 transporter activity341103−1.7620*166*−1.9
AB07771GO:0016810 hydrolase activity, acting on carbon–nitrogen33926−3.725482−1.6
AB03515GO:0008869 galactonate dehydratase activity321119−1.4337140−1.3
AB01593GO:0005351 sugar:hydrogen symporter activity316101−1.61160*101*−3.5
AB06533 AbPf2 30866−2.213849−1.5
AB04118GO:0008843 endochitinase activity (S)30122−3.8127*10*−3.7
AB00783GO:0033754 indoleamine 2,3-dioxygenase activity29497−1.6743*146*−2.3
AB06341GO:0004086 carbamoyl-phosphate synthase activity283133−1.1203151−0.4
AB10029 28177−1.9412*109*−1.9

Figure 8. Hierarchical clustering of fungal RNA-seq data from plants at 24 and 48 hpi.

(a) Set of 168 genes that show differential expression patterns between the ∆abpf2 mutant and wild-type A. brassicicola at 12 hpi. The color key represents the log2 ratio of fragments per kilobase of exon model per million (FPKM). Red indicates higher expression levels and green indicates lower expression levels in the ∆abpf2 mutant than in wild-type A. brassicicola.

(b) Expanded view of genes expressed at higher levels in the mutants than the wild-type at 12 hpi.

(c) Expanded view of genes expressed at lower levels in the mutants at 12 hpi.

S' indicates putative secretion proteins predicted by hidden Markov models and signal P (, e) Comparisons of genes with lower and higher levels of expression at 12 and 48 hpi.

Download figure to PowerPoint

The pectate lyase genes in the wild-type were highly expressed (0.67% of total expression). Five of the eight small proteins contained more than six cysteine residues, which are considered important for formation of disulfide bonds in effector proteins (Luderer et al., 2002; Doehlemann et al., 2009). These proteins were expressed at high levels, representing up to 0.5% of the sequence tags in the wild-type but very few in the mutant (Table S4). Their small size and the presence of secretion signal peptides and frequent cysteine residues are characteristics consistent with other fungal effector proteins. The 62 genes expressed at lower levels in the mutant included seven transporters and several putative detoxifying enzymes, such as cyanide hydrolase, indolamine dioxygenase, glutamine aldoltransferease and lactamase. They also included four glycoside hydrolases, three bacterial rhodopsins, and a short protein containing eight cysteine residues and secretion signal peptides. Genes encoding enzymes such as superoxide dismutase, peroxidases, catalase, laccases and polyphenol oxidases were not differentially expressed at 12 hpi.

At 48 hpi, the level of AbPf2 transcripts had returned to almost the same level as before induction (Figure 4). At this time, a total of 119.6 × 106 and 100.2 × 106 reads were produced for the wild-type and the Δabpf2 mutant, respectively. Of these, 4.9 × 106 (4.1%) and 1.1 × 106 (1.1%) mapped to the genome of A. brassicicola, and 98.3 × 106 (82.5%) and 85.9 × 106 (85.7%) mapped to the genome of A. thaliana. Among the 10 688 fungal genes, a total of 252 genes were expressed at more than twofold lower levels (< 0.05) in the mutant than in the wild-type. Of these 252 genes, 40 were also found among the 106 genes expressed at lower levels at 12 hpi (Figure 8d). The expression level of these 40 genes had decreased sharply in wild-type A. brassicicola at 48 hpi. Furthermore, 525 genes were expressed at higher levels in the mutant than in wild-type A. brassicicola (Table S5). Of these genes with higher expression levels, only 32 were also expressed at higher levels in the mutant at 12 hpi (Figure 8e and Table 5). In addition, a total of 232 genes were expressed in the mutant at higher levels at 48 hpi than at 12 hpi. The differentially expressed genes were unique, with little overlap at 12 and 48 hpi (Figure S4).

Table 5. Number of differentially expressed fungal genes during plant infection
 Up-regulated genesDown-regulated genesTotal
ftf2 mutant 12 hpi/wild-type 12 hpi62106168
ftf2 mutant 48 hpi/wild-type 48 hpi525252777
Wild-type 48 hpi/wild-type 12 hpi243267510
ftf2 mutant 48 hpi/∆ftf2 mutant 12 hpi232107339

Expression pattern of selected genes during pathogenesis

We also surveyed the expression patterns of five genes in the wild-type and ∆abpf2-2 mutant during the disease cycle. They encoded two pectate lyases (AB01332.1 and AB04813.1), two effector proteins (AB04512.1 and AB09024.1) and an amidohydrolase (AB09632.1), and were expressed at lower levels in the ∆abpf2 mutant than in the wild-type at 12 hpi. In conidia, the expression levels of all five genes were low before inoculation, and then increased exponentially in the wild-type at 4 and 8 hpi (Figure 9a–e,i). Their expression was further increased at 12 hpi. The high expression level was maintained in the invading hyphae at 24 hpi, and then decreased rapidly in the colonizing hyphae at 48 hpi. At 48 hpi, expression of AbPf2 was at the same level as before induction (Figure 4). As controls for the quantitative real-time PCR analysis, we also examined three previously studied genes, a pectate lyase (AB05514.1), Cbh7 (AB06252.1) and a chymotrypsin gene (AB01734.1) (Cho et al., 2012). The expression patterns of these three genes were very different from those of the five genes and the AbPf2 gene (Figure 9f–h,j). In addition, the quantitative RT-PCR results were similar to the RNA-seq data at 12 and 48 hpi for all eight genes, although RNA-seq was slightly less sensitive than quantitative RT-PCR in some cases (Table S4).


Figure 9. Expression of eight genes during pathogenesis.

(a–h) Comparison of transcripts (∆Ct) of each gene in the ∆abpf2 mutant and wild-type Alternaria brassicicola. They were normalized by the mean value for two housekeeping genes: glyceraldehyde 3-phosphate dehydrogenase and elongation factor 1-α. The y axes indicate the relative quantity of the transcripts on the log2 scale. The x axes show the number of hours post-inoculation at which biological samples were collected.

Error bars indicate standard deviation. Amhy, amidohydrolase AB09633.1; EF4512, effector AB04512.1; PL4813, pectate lyase AB04813.1; EF9024, effector Ab09024.1; PL1332, pectate lyase AB01332.1; PL5514, pectate lyase AB05514.1; AbCbh7, cellobiohydrolase AB06252.1; Chymo, chymotrypsin AB01734.1.

Download figure to PowerPoint

Conserved sequence motifs shared among promoters of genes with lower levels of expression

We searched for common motifs in the putative promoter regions of genes that were expressed at significantly lower levels in the Δabpf2 mutant compared to the wild-type, and for which a full-length promoter was available. Fifty genes matched these criteria. The comparison revealed a conserved motif that was shared by the promoters of 25 of these genes (Figure 10). The occurrence of this motif was significantly greater than expected for random selection, as 1044 promoters contain this motif among a total of 8237 promoters (< 1 × 10−4). The motif contained a CGG sub-sequence, which is characteristic of a binding site for a fungal-specific transcription factor. The motif showed similarity to the previously described binding sites of several fungal-specific transcription factors in the JASPAR CORE fungi database (e.g. MA0429.1, = 6 × 10−3).


Figure 10. Putative transcription factor binding sites.

Over-represented sequence motif in promoters of genes with lower levels of expression in the ∆abpf2 mutants than in wild-type Alternaria brassicicola.

Download figure to PowerPoint

Plant responses

Differentially expressed plant genes were also identified by aligning the sequence tags to the Arabidopsis genome. A total of 1277 (3.4%) of the 34134 predicted genes in A. thaliana were expressed at significantly (< 0.05) higher levels, and 2117 genes (6.2%) were expressed at levels that were more than twofold lower in plants inoculated at 12 hpi with the Δabpf2 mutant compared to the wild-type (Table S6). The functional category most over-represented among the more highly expressed plant genes was chloroplast proteins (Table S7). They included genes associated with light harvesting and electron transport in photosystems I and II and with CO2 fixation. Many other genes expressed at higher levels and with unknown functions also encoded proteins that are located in chloroplasts. The over-represented groups among up-regulated genes also included genes associated with the pentose phosphate pathway, transcription, translation, cell differentiation, cell morphogenesis, meristem growth, cell tip growth and response to auxin. Plant genes with lower levels of expression and over-represented included many genes associated with defense against fungi. Their functions were related to a response to chitin, ethylene stimulus, wounding, jasmonic acid, hydrogen peroxide and oxidative stress, among others (Table S7). Interestingly, 21 genes encoding leucine-rich receptor (LRR) proteins were also included in this group (Table S8).


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Transcription factor AbPf2

Bioinformatics analysis by Interpro (Bateman et al., 2004) suggested that the AbPf2 protein was a transcription factor with a GAL4 fungal-specific DNA binding domain (Interpro ID: IPR001138, Pfam ID: PF00172). No obvious nuclear localization signal was predicted by PSORT (Horton et al., 2007). However, the AbPf2 protein accumulated in nuclei of the conidia and invading hyphae during early pathogenesis, as monitored using AbPf2–GFP fusion proteins (Figures 6 and 7). The data indicate that AbPf2 had entered the nuclei. The presence of a DNA binding domain predicted by Interpro (Bateman et al., 2004) and a conserved DNA-binding motif among co-expressed genes, and the nuclear localization during early pathogenesis suggest that AbPf2 is a transcription factor.

AbPf2 mutations affect only pathogenesis

Several transcription factor genes are associated with pathogenesis in other fungi. Mutation of these genes, or their downstream genes, had pleiotropic effects, including defects in pathogenesis (Kim et al., 2009; Guo et al., 2011; Son et al., 2011; Wang et al., 2011). These additional effects are of vegetative growth, conidium production or colony morphology. Other pathogenesis-associated genes in conserved signal pathways (Xu and Hamer, 1996; Lev et al., 1999; Xue et al., 2002) or primary metabolism (Seong et al., 2005; Oide et al., 2006; Lee et al., 2009) also affect multiple traits. In addition, an appressorium is required for early penetration and full virulence in several plant pathogenic fungi (Lev et al., 1999; Thines et al., 2000). In contrast, all strains of ∆abpf2 mutants were similar to the wild-type in terms of conidia production (Table S1), vegetative growth and colony morphology in vitro (Figure 2). The germination rates and the frequency of appressorium formation on host plant tissue were also similar (Figure S2 and Table S2). These results suggest that AbPf2 is dispensable for fungal development and primary metabolism in A. brassicicola, and is associated exclusively with pathogenesis.

The mutants' ability to overcome stressors

Pathogenic fungi must overcome diverse plant defense mechanisms and toxic metabolites to successfully infect their host plants. Two of the most prominent plant defenses are phytoalexins (VanEtten et al., 1995; Thomma et al., 1999; Ahuja et al., 2012) and reactive oxygen species (ROS) (Cessna et al., 2000; Mayer et al., 2001; Molina and Kahmann, 2007; Lin et al., 2009; Guo et al., 2011). Most brassicaceous plants produce phytoalexins, such as camalexin in A. thaliana and brassinin in cultivated brassicas (Pedras et al., 2004, 2011). The A. thaliana mutant pad3 does not produce camalexin and is very susceptible to A. brassicicola (Zhou et al., 1999). However, all ∆abpf2 mutants failed to infect the pad3 mutant (Figure 1). In addition, the phytoalexin brassinin, produced by B. oleracea, had a similar effect on vegetative growth of the ∆abpf2 mutants and wild-type A. brassicicola. Neither was there a difference in observed effects on the mutants or wild-type when exposed to KO2, H2O2, sorbitol, NaCl or KCl (Figure 4). ∆abpf2 mutants expressed similar amounts of transcripts of putative ROS-scavenging enzymes, such as superoxide dismutase, peroxidases, catalase, laccases, and polyphenol oxidases, as the wild-type during pathogenesis. Putative detoxification enzymes, such as cyanide hydrolase, indolamine dioxygenase, glutamine aldoltransferease and lactamase, were expressed at higher levels in the ∆abpf2 mutant (Table S6). These results suggest that AbPf2 is also dispensable for the detoxification of phytoalexins, unknown phytotoxins or reactive oxygen species, and for osmoregulation. In other words, the loss of pathogenicity is not due to the inability of the mutants to efficiently manage phytoalexins, phytotoxins and ROS produced by these host plants.

Functional importance of AbPf2 in pathogenesis

When the host plant was inoculated with wild-type A. brassicicola, the expression level of AbPf2 transcripts reached its peak at 12 hpi, and sharply decreased shortly thereafter (Figure 4). The signal strength of the green fluorescence from the AbPf2–GFP fusion protein also suggested that protein expression was similarly induced, and then decreased (Table 2). The protein was localized mainly in the nuclei during early pathogenesis (Figures 6 and 7). The expression level of 106 genes also showed a similar pattern of induction and decrease, with a slight delay compared to the AbPf2 transcripts. The nuclear localization of AbPf2 protein and the similar expression pattern of the transcription factor and 106 genes with a slight time delay suggested their relationship as regulator and regulated genes. The changes in gene expression profiles of the mutant when in contact with its host plants provided a clue to its loss of pathogenicity. Of 10 688 genes, 106 were expressed at lower levels in the ∆abpf2 mutant than in the wild-type during the early penetration stage (12 hpi). They included 13 genes encoding hydrolytic enzymes and eight genes encoding putative effector proteins. The hydrolytic enzyme-coding genes included two of 18 pectate lyase genes in the genome, AB04813.1 and AB01332.1. These two pectate lyase genes were exponentially induced in wild-type A. brassicicola during early infection (Figure 9). This contrasts with the expression of six pectate lyase genes (AB05514.1, AB00904.1, AB10322, AB06838.1, AB03608 and AB10575.1), which was highly induced during the late stage of plant infection (Srivastava et al., 2012). The basal secretome of other fungi contains polysaccharide-degrading enzymes and proteases (Girard et al., 2013). The reduced expression in the mutants of these two pectate lyases and other cellulases may have been detrimental to the initial penetration of host tissue due to inefficient digestion of plant cell walls. This reduction of cell wall-degrading enzymes by the mutant may have slowed down the colonization process once it was initiated, but not stopped it. Thus, the reduced expression of hydrolytic enzyme genes was insufficient to explain both the loss of pathogenicity and the failure of lesion expansion at wound sites or on senescent host tissue (Figure 1). The eight putative effector proteins may have had crucial roles in pathogenesis. Five of the eight genes were highly expressed in the wild-type but were only expressed at low levels in the ∆abpf2 mutant at 12 hpi (Table S4). The importance of effectors in the interactions between various host plants and their fungi and fungus-like oomycete pathogens has been established previously (Kale et al., 2010; Kale and Tyler, 2011). In compatible interactions, many effector proteins re-engineer host gene expression, causing a suppression of the plant's defenses (Doehlemann et al., 2009; Djamei et al., 2011). In the incompatible interactions between biotrophic pathogens and their host plants, effector proteins are usually recognized by host receptor proteins. Such small, secreted effector proteins associated with pathogenesis have been well documented for (hemi)biotrophic interactions, but with few examples in necrotrophic fungi (De Wit et al., 2009). Effectors from necrotrophic fungi mainly act as necrosis-inducing factors after interacting with cognate susceptibility proteins (Liu et al., 2012). We suspect that expression of the eight putative effectors, the necrosis-inducing factor and hydrolytic enzymes together contributed to make the host cells susceptible to wild-type A. brassicicola. Other necrotrophic fungi also encode a wide range of effectors in their genomes, suggesting that necrotrophic fungi also subtly manipulate their hosts during infection (Girard et al., 2013). Our primary interest is in investigating functions of the eight putative effectors in pathogenesis. Gene expression within the plant during this interaction suggests that they are effector proteins (see below).

Plant genes affected by AbPf2

Some plant genes were expressed at lower levels following a challenge by the ∆abpf2 mutants than when challenged by wild-type A. brassicicola. These included genes associated with host responses to chitin, jasmonic acids, ethylene, wounding, oxidative stress and fungi. These genes are important in the defense against necrotrophic fungi (McDowell and Dangl, 2000). They also included genes associated with lignin synthesis and transporter proteins. The lower level of expression of these genes in plants infected by the mutant than by the wild-type A. brassicicola suggested that the ∆abpf2 mutant did not activate, or only marginally activated, the host plant defense mechanisms against the fungus. Plants activate an initial defense after recognition of microbe-associated molecular patterns and a full defense response when triggered by effectors (Jones and Dangl, 2006). It is possible that the plants in our study did not sense, or neglected the presence of, the ∆abpf2 mutants growing on their surface (Figure 3). Alternatively, defense reactions may have been activated after microbe-associated pattern recognition, but were weak compared to the effector-triggered immune responses because the ∆abpf2 mutant did not secrete effectors. The eight putative effector genes identified in this study may be essential for effector-triggered immunity.

The plant genes expressed at higher levels in plants inoculated with ∆abpf2 mutants than in those inoculated with wild-type A. brassicicola were associated with primary metabolism. They included genes associated with transcription, translation, cell division, cell differentiation, and biosynthesis of membranes and cell walls. The differential expression of these genes suggested that wild-type A. brassicicola damaged the primary metabolism and tissue growth of the host plant, but the ∆abpf2 mutants did not. We excluded the possibility that the plant gene expression data were the result of complete arrest of growth and metabolism in the ∆abpf2 mutant for two reasons. First, mutant conidia germinated and grew on the host plant surface (Figure 3 and Figure S3). Second, many genes were expressed more highly in the ∆abpf2 mutant than in the wild-type A. brassicicola at 48 hpi (Table 5). Interestingly, plants infected by the mutant expressed genes associated with chloroplasts and photosynthesis to a higher level than in the wild-type. Infection of plants by the wild-type may have caused increased expression of host defense genes at the expense of photosynthesis. We prefer the alternative possibility that wild-type A. brassicicola manipulated its host plants to slow photosynthesis, as implied in other systems (Gohre et al., 2012). Photosynthesis was reduced when conidia of certain pathogens or fungal elicitors were applied to intact leaves (Govrin et al., 2006; Bolton, 2009). Also, the ToxA protein secreted by Stagonospora nodorum and Pyrenophora tritici-repentis interacted directly with a chloroplast protein and indirectly with another protein, Tsn1, that is involved in circadian rhythms and photosynthesis (Manning et al., 2007; Faris et al., 2010). The gene expression data suggest that wild-type A. brassicicola suppressed primary metabolism and photosynthesis during early infection, but the ∆abpf2 mutant did not.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Transformation, maintenance of fungal strains, pharmacological test, and confocal microscopy

Growth and maintenance of A. brassicicola Schweinitz & Wiltshire (American Type Culture Collection number ATCC96836) and its transformation, nucleic acid isolation, mutant purification and mutant verification by Southern hybridization were performed as described previously (Cho et al., 2009). Three probes were produced for Southern hybridization. Wild-type A. brassicicola and each of the mutant strains created during this study were purified by two rounds of single-spore isolation to obtain a uniform genetic background. Cultures were maintained as glycerol stock in separate tubes, with one tube used for each assay. Our method of capturing images of fungal tissue by confocal microscopy has been described previously (Srivastava et al., 2012). All primer information is provided in Table S9. The pharmacological tests were performed three times in the dark for 5 days at 25°C. The full-length AbPf2 gene with its 5′ side flanking region was amplified by PCR using primers P1 and P2. Its sequence was determined using primers P1–P16 (Table S9). The completed sequence was used to identify a predicted stop codon. The sequence data have been deposited in GenBank (accession number JQ899199).

Generation of replacement mutants for AbPf2

All transformation constructs used here were produced as described previously (Cho et al., 2009). We created AbPf2 deletion mutants by replacing the 891 nucleotides spanning the partial promoter (233 nucleotides) and partial protein-coding region (658 nucleotides) with a HygB resistance cassette (Figure S1). The replacement construct was produced using three sets of primers, P17/P18, P21/P22 and P19/P20 (Figure S5a), to amplify 1056, 951 and 1436 bp. The construct was transformed into wild-type A. brassicicola. We used two primers, P23 and P24, to create 476 bp Southern hybridization probes (Table S9).

Complementation of two Δabpf2 mutant strains

The Δabpf2-2 and Δabpf2-s5 mutants were complemented with the wild-type AbPf2 allele and its native promoter as described previously (Cho et al., 2012). Details are provided in Methods S1.

Mutant expressing the AbPf2–GFP fusion protein and mCherry–NLS

We produced a mutant strain that expressed a fusion protein of AbPf2–GFP under the control of the AbPf2 native promoter. To create a transformation construct, the AbPf2 coding region (1054 bp) and 3′ flanking region (448 bp) were amplified using primers P25/P26 and P27/P28, respectively (Figure S5c). Another set of primers, P29 and P30, was used to amplify the 2384 bp covering the coding regions of the GFP and the resistance cassette HygB. The final transformation constructs were produced by PCR amplification from a mixture of the three PCR products using primers P25 and P28. The construct was transformed into wild-type A. brassicicola. To investigate the localization of AbPf2 proteins in nuclei, we also created a construct expressing the mCherry–NLS fusion protein under the control of ToxA promoter. To make a transformation construct, mCherry–NLS was amplified using primers P35 and P36 from pBV579 (Khang et al., 2010). Another set of primers, P33 and P34, was used to amplify the ToxA promoter. The final transformation constructs were produced by PCR amplification from a mixture of the three PCR products using primers P33 and P36 (Figure S5d). The construct and the 2267 bp nourseothricin-resistant cassette were transformed into a strain expressing AbPf2–GFP. The nourseothricin-resistant transformants were purified twice by single-spore isolation.

Pathogenicity assays

We performed pathogenicity assays as described previously (Cho et al., 2009) with minor modifications. Either whole plants or detached leaves harvested from 5–8-week-old B. oleracea (green cabbage) were inoculated with 1–2 × 103 conidia in 10 μl water. Whole plants of the pad3 mutant (Zhou et al., 1999) were inoculated with 2000 conidia in 10 μl water. Whole plants of A. thaliana (Col-0) were inoculated by spraying to run-off with a concentration of 5 × 105 conidia/ml. We measured pathogenicity as the presence or absence of visible lesions at the inoculation sites after 5 dpi.

Examination of germ tubes and appressoria

To examine germ tubes and the formation of appressoria at the tips of germ tubes, trypan blue staining was performed as described previously (Srivastava et al., 2012).

Quantitative real-time PCR

Expression of the AbPf2 gene in wild-type A. brassicicola was measured by quantitative RT-PCR. We collected mature conidia before inoculation (0 hpi), conidia attached to the host plant (4 hpi), and conidia at approximately 100% germination (8 hpi on cabbage and 12 hpi on Arabidopsis). Other fungal tissues were collected at the stage of host-plant penetration (12 hpi on cabbage and 24 hpi on Arabidopsis), early hyphal colonization (24 hpi on cabbage and Arabidopsis), saprophytic growth on necrotic host tissues (48–72 hpi on cabbage and 48–114 hpi on Arabidopsis), and the conidiation stage (216 hpi on cabbage). All tissues collected from inoculation courts were a mixture of fungal and host plant tissues. RNA extraction, cDNA synthesis and quantitative RT-PCR reactions were performed as described previously (Srivastava et al., 2012). The relative amounts of the transcripts of AbPf2 and eight downstream genes were calculated as 2DCt using a threshold cycle (Ct), where ΔCt = (Ct(AbPf2)Ct(mean of two genes)). Two housekeeping genes, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and elongation factor 1-α (Ef1-α), were used for normalization. The relative amounts of transcript of each gene are presented as the difference in Ct values [∆Ct = (Ct(gene i)Ct(mean of Ef1-α and GAPDH))], which are equivalent to the relative amounts of transcripts on a log2 scale. The relative transcript amounts between the wild-type and the ∆abpf2 strain are presented as difference in ΔCt (ΔΔCt) between the mutant and the wild-type, where ΔΔCt = [(Ct(gene i)Ct(mean of Ef1-α and GAPDH))∆abpf2 (Ct(gene i)Ct(mean of Ef1-α and GAPDH))wild-type].

RNA-seq data generation, gene expression profile analysis, and identification of transcription factor binding sites

RNA-seq data (accession number GSE38984) were analyzed as described previously (Cho et al., 2012). Details are provided in Methods S2.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Wayne Borth, Anne Alvarez and John Hu for generously sharing their research equipment, Tina M. Carvalho for assisting with the confocal microscopy, Johnson Siu and Hui Trung for plant growth, and Fred Brooks for critical reading of the manuscript. We are especially grateful to Chang-Hyun Khang (Department of Plant Biology, Georgia University, GA) and Barbara Valent (Department of Plant Pathology, Kansas State University , KS) for sharing plasmid pBV579 encoding mCherry tagged with a nuclear localization signal. This research was supported by HATCH funds to Y.C., administered by the College of Tropical Agriculture and Human Resources, University of Hawaii at Manoa, Honolulu, HI. Analysis of RNA-seq data was supported by the Office of Science of the US Department of Energy under contract number DE-AC02-05CH11231.


  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Ahuja, I., Kissen, R. and Bones, A.M. (2012) Phytoalexins in defense against pathogens. Trends Plant Sci. 17, 7390.
  • Bateman, A., Coin, L., Durbin, R. et al. (2004) The Pfam protein families database. Nucleic Acids Res. 32, D138D141.
  • Bolton, M.D. (2009) Primary metabolism and plant defense – fuel for the fire. Mol. Plant–Microbe Interact. 22, 487497.
  • Buttner, D. and Bonas, U. (2003) Common infection strategies of plant and animal pathogenic bacteria. Curr. Opin. Plant Biol. 6, 312319.
  • Caracuel, Z., Roncero, M.I., Espeso, E.A., Gonzalez-Verdejo, C.I., Garcia-Maceira, F.I. and Di Pietro, A. (2003) The pH signaling transcription factor PacC controls virulence in the plant pathogen Fusarium oxysporum. Mol. Microbiol. 48, 765779.
  • Cessna, S.G., Sears, V.E., Dickman, M.B. and Low, P.S. (2000) Oxalic acid, a pathogenicity factor for Sclerotinia sclerotiorum, suppresses the oxidative burst of the host plant. Plant Cell, 12, 21912200.
  • Chen, L.H., Lin, C.H. and Chung, K.R. (2012) Roles for SKN7 response regulator in stress resistance, conidiation and virulence in the citrus pathogen Alternaria alternata. Fungal Genet. Biol. 49, 802813.
  • Cho, Y., Cramer, R.A. Jr, Kim, K.H., Davis, J., Mitchell, T.K., Figuli, P., Pryor, B.M., Lemasters, E. and Lawrence, C.B. (2007) The Fus3/Kss1 MAP kinase homolog Amk1 regulates the expression of genes encoding hydrolytic enzymes in Alternaria brassicicola. Fungal Genet. Biol. 44, 543553.
  • Cho, Y., Kim, K.-H., La Rota, M., Scott, D., Santopietro, G., Callihan, M. and Lawrence, C.B. (2009) Identification of virulence factors by high throughput targeted gene deletion of regulatory genes in Alternaria brassicicola. Mol. Microbiol. 72, 13161333.
  • Cho, Y., Srivastava, A., Ohm, R.A., Lawrence, C.B., Wang, K.H., Grigoriev, I.V. and Marahatta, S.P. (2012) Transcription factor Amr1 induces melanin biosynthesis and suppresses virulence in Alternaria brassicicola. PLoS Pathog. 8, e1002974.
  • Ciuffetti, L.M., Tuori, R.P. and Gaventa, J.M. (1997) A single gene encodes a selective toxin causal to the development of tan spot of wheat. Plant Cell, 9, 135144.
  • De Wit, P.J., Mehrabi, R., Van den Burg, H.A. and Stergiopoulos, I. (2009) Fungal effector proteins: past, present and future. Mol. Plant Pathol. 10, 735747.
  • Djamei, A., Schipper, K., Rabe, F. et al. (2011) Metabolic priming by a secreted fungal effector. Nature, 478, 395398.
  • Doehlemann, G., van der Linde, K., Assmann, D., Schwammbach, D., Hof, A., Mohanty, A., Jackson, D. and Kahmann, R. (2009) Pep1, a secreted effector protein of Ustilago maydis, is required for successful invasion of plant cells. PLoS Pathog. 5, e1000290.
  • Faris, J.D., Zhang, Z., Lu, H. et al. (2010) A unique wheat disease resistance-like gene governs effector-triggered susceptibility to necrotrophic pathogens. Proc. Natl Acad. Sci. USA, 107, 1354413549.
  • Girard, V., Dieryckx, C., Job, C. and Job, D. (2013) Secretomes: the fungal strike force. Proteomics, 13, 597608.
  • Glazebrook, J. (2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol. 43, 205227.
  • Gohre, V., Jones, A.M., Sklenar, J., Robatzek, S. and Weber, A.P. (2012) Molecular crosstalk between PAMP-triggered immunity and photosynthesis. Mol. Plant–Microbe Interact. 25, 10831092.
  • Govrin, E.M., Rachmilevitch, S., Tiwari, B.S., Solomon, M. and Levine, A. (2006) An elicitor from Botrytis cinerea induces the hypersensitive response in Arabidopsis thaliana and other plants and promotes the gray mold disease. Phytopathology, 96, 299307.
  • Guo, M., Chen, Y., Du, Y. et al. (2011) The bZIP transcription factor MoAP1 mediates the oxidative stress response and is critical for pathogenicity of the rice blast fungus Magnaporthe oryzae. PLoS Pathog. 7, e1001302.
  • Hoch, H.C., Staples, R.C., Whitehead, B., Comeau, J. and Wolf, E.D. (1987) Signaling for growth orientation and cell differentiation by surface topography in uromyces. Science, 235, 16591662.
  • Horton, P., Park, K.J., Obayashi, T., Fujita, N., Harada, H., Adams-Collier, C.J. and Nakai, K. (2007) WoLF PSORT: protein localization predictor. Nucleic Acids Res. 35, W585W587.
  • Jones, J.D. and Dangl, J.L. (2006) The plant immune system. Nature, 444, 323329.
  • Joubert, A., Simoneau, P., Campion, C., Bataille-Simoneau, N., Iacomi-Vasilescu, B., Poupard, P., Francois, J.M., Georgeault, S., Sellier, E. and Guillemette, T. (2011) Impact of the unfolded protein response on the pathogenicity of the necrotrophic fungus Alternaria brassicicola. Mol. Microbiol. 79, 13051324.
  • Kale, S.D. and Tyler, B.M. (2011) Entry of oomycete and fungal effectors into plant and animal host cells. Cell. Microbiol. 13, 18391848.
  • Kale, S.D., Gu, B., Capelluto, D.G. et al. (2010) External lipid PI3P mediates entry of eukaryotic pathogen effectors into plant and animal host cells. Cell, 142, 284295.
  • Khang, C.H., Berruyer, R., Giraldo, M.C., Kankanala, P., Park, S.Y., Czymmek, K., Kang, S. and Valent, B. (2010) Translocation of Magnaporthe oryzae effectors into rice cells and their subsequent cell-to-cell movement. Plant Cell, 22, 13881403.
  • Kim, S., Park, S.Y., Kim, K.S., Rho, H.S., Chi, M.H., Choi, J., Park, J., Kong, S., Goh, J. and Lee, Y.H. (2009) Homeobox transcription factors are required for conidiation and appressorium development in the rice blast fungus Magnaporthe oryzae. PLoS Genet. 5, e1000757.
  • Kloda, A., Petrov, E., Meyer, G.R., Nguyen, T., Hurst, A.C., Hool, L. and Martinac, B. (2008) Mechanosensitive channel of large conductance. Int. J. Biochem. Cell B, 40, 164169.
  • Kulkarni, R.D., Thon, M.R., Pan, H. and Dean, R.A. (2005) Novel G–protein-coupled receptor-like proteins in the plant pathogenic fungus Magnaporthe grisea. Genome Biol. 6, R24.
  • Lee, S.H., Han, Y.K., Yun, S.H. and Lee, Y.W. (2009) Roles of the glyoxylate and methylcitrate cycles in sexual development and virulence in the cereal pathogen Gibberella zeae. Eukaryotic Cell, 8, 11551164.
  • Lev, S., Sharon, A., Hadar, R., Ma, H. and Horwitz, B.A. (1999) A mitogen-activated protein kinase of the corn leaf pathogen Cochliobolus heterostrophus is involved in conidiation, appressorium formation, and pathogenicity: diverse roles for mitogen-activated protein kinase homologs in foliar pathogens. Proc. Natl Acad. Sci. USA, 96, 1354213547.
  • Lin, C.H., Yang, S.L. and Chung, K.R. (2009) The YAP1 homolog-mediated oxidative stress tolerance is crucial for pathogenicity of the necrotrophic fungus Alternaria alternata in citrus. Mol. Plant–Microbe Interact. 22, 942952.
  • Liu, Z., Zhang, Z., Faris, J.D. et al. (2012) The cysteine rich necrotrophic effector SnTox1 produced by Stagonospora nodorum triggers susceptibility of wheat lines harboring Snn1. PLoS Pathog. 8, e1002467.
  • Lorang, J.M., Tuori, R.P., Martinez, J.P. et al. (2001) Green fluorescent protein is lighting up fungal biology. Appl. Environ. Microbiol. 67, 19871994.
  • Luderer, R., De Kock, M.J., Dees, R.H., De Wit, P.J. and Joosten, M.H. (2002) Functional analysis of cysteine residues of ECP elicitor proteins of the fungal tomato pathogen Cladosporium fulvum. Mol. Plant Pathol. 3, 9195.
  • Manning, V.A., Hardison, L.K. and Ciuffetti, L.M. (2007) Ptr ToxA interacts with a chloroplast-localized protein. Mol. Plant–Microbe Interact. 20, 168177.
  • Mayer, A.M., Staples, R.C. and Gil–ad, N.L. (2001) Mechanisms of survival of necrotrophic fungal plant pathogens in hosts expressing the hypersensitive response. Phytochemistry, 58, 3341.
  • McDowell, J.M. and Dangl, J.L. (2000) Signal transduction in the plant immune response. Trends Biochem. Sci. 25, 7982.
  • Molina, L. and Kahmann, R. (2007) An Ustilago maydis gene involved in H2O2 detoxification is required for virulence. Plant Cell, 19, 22932309.
  • Oh, I.S., Park, A.R., Bae, M.S., Kwon, S.J., Kim, Y.S., Lee, J.E., Kang, N.Y., Lee, S., Cheong, H. and Park, O.K. (2005) Secretome analysis reveals an Arabidopsis lipase involved in defense against Alternaria brassicicola. Plant Cell, 17, 28322847.
  • Oide, S., Moeder, W., Krasnoff, S., Gibson, D., Haas, H., Yoshioka, K. and Turgeon, B.G. (2006) NPS6, encoding a nonribosomal peptide synthetase involved in siderophore-mediated iron metabolism, is a conserved virulence determinant of plant pathogenic ascomycetes. Plant Cell, 18, 28362853.
  • Pedras, M.S., Ahiahonu, P.W. and Hossain, M. (2004) Detoxification of the cruciferous phytoalexin brassinin in Sclerotinia sclerotiorum requires an inducible glucosyltransferase. Phytochemistry, 65, 26852694.
  • Pedras, M.S., Yaya, E.E. and Glawischnig, E. (2011) The phytoalexins from cultivated and wild crucifers: chemistry and biology. Nat. Prod. Rep. 28, 13811405.
  • Seong, K., Hou, Z., Tracy, M., Kistler, H.C. and Xu, J.R. (2005) Random insertional mutagenesis identifies genes associated with virulence in the wheat scab fungus Fusarium graminearum. Phytopathology, 95, 744750.
  • Sexton, A.C. and Howlett, B.J. (2006) Parallels in fungal pathogenesis on plant and animal hosts. Eukaryotic Cell, 5, 19411949.
  • Shlezinger, N., Minz, A., Gur, Y., Hatam, I., Dagdas, Y.F., Talbot, N.J. and Sharon, A. (2011) Anti-apoptotic machinery protects the necrotrophic fungus Botrytis cinerea from host-induced apoptotic-like cell death during plant infection. PLoS Pathog. 7, e1002185.
  • Son, H., Seo, Y.S., Min, K. et al. (2011) A phenome-based functional analysis of transcription factors in the cereal head blight fungus, Fusarium graminearum. PLoS Pathog. 7, e1002310.
  • Srivastava, A., Ohm, R.A., Oxiles, L., Brooks, F., Lawrence, C.B., Grigoriev, I.V. and Cho, Y. (2012) A zinc-finger-family transcription factor, AbVf19, is required for the induction of a gene subset important for virulence in Alternaria brassicicola. Mol. Plant–Microbe Interact. 25, 443452.
  • Thines, E., Weber, R.W. and Talbot, N.J. (2000) MAP kinase and protein kinase A-dependent mobilization of triacylglycerol and glycogen during appressorium turgor generation by Magnaporthe grisea. Plant Cell, 12, 17031718.
  • Thomma, B.P., Nelissen, I., Eggermont, K. and Broekaert, W.F. (1999) Deficiency in phytoalexin production causes enhanced susceptibility of Arabidopsis thaliana to the fungus Alternaria brassicicola. Plant J. 19, 163171.
  • Torres, M.A., Jones, J.D. and Dangl, J.L. (2006) Reactive oxygen species signaling in response to pathogens. Plant Physiol. 141, 373378.
  • VanEtten, H.D., Sandrock, R.W., Wasmann, C.C., Soby, S.D., McCluskey, K. and Wang, P. (1995) Detoxification of phytoanticipins and phytoalexins by phytopathogenic fungi. Can. J. Bot. 73, 518525.
  • Wang, Y., Liu, W., Hou, Z., Wang, C., Zhou, X., Jonkers, W., Ding, S., Kistler, H.C. and Xu, J.R. (2011) A novel transcriptional factor important for pathogenesis and ascosporogenesis in Fusarium graminearum. Mol. Plant–Microbe Interact. 24, 118128.
  • Williams, B., Kabbage, M., Kim, H.J., Britt, R. and Dickman, M.B. (2011) Tipping the balance: Sclerotinia sclerotiorum secreted oxalic acid suppresses host defenses by manipulating the host redox environment. PLoS Pathog. 7, e1002107.
  • Wong Sak Hoi, J. and Dumas, B. (2010) Ste12 and Ste12-like proteins, fungal transcription factors regulating development and pathogenicity. Eukaryotic Cell, 9, 480485.
  • Xu, J.R. and Hamer, J.E. (1996) MAP kinase and cAMP signaling regulate infection structure formation and pathogenic growth in the rice blast fungus Magnaporthe grisea. Genes Dev. 10, 26962706.
  • Xue, C., Park, G., Choi, W., Zheng, L., Dean, R.A. and Xu, J.R. (2002) Two novel fungal virulence genes specifically expressed in appressoria of the rice blast fungus. Plant Cell, 14, 21072119.
  • You, B.J. and Chung, K.R. (2007) Phenotypic characterization of mutants of the citrus pathogen Colletotrichum acutatum defective in a PacC-mediated pH regulatory pathway. FEMS Microbiol. Lett. 277, 107114.
  • Zhao, X., Mehrabi, R. and Xu, J.R. (2007) Mitogen-activated protein kinase pathways and fungal pathogenesis. Eukaryotic Cell, 6, 17011714.
  • Zhou, N., Tootle, T.L. and Glazebrook, J. (1999) Arabidopsis PAD3, a gene required for camalexin biosynthesis, encodes a putative cytochrome P450 monooxygenase. Plant Cell, 11, 24192428.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information
tpj12217-sup-0001-FigureS1.epsimage/eps2724KFigure S1. Deletion of the AbPf2 gene.
tpj12217-sup-0002-FigureS2.epsimage/eps7707KFigure S2. Germination and germ tube growth of wild-type Alternaria brassicicola and mutant conidia on green cabbage leaves.
tpj12217-sup-0003-FigureS3.epsimage/eps9358KFigure S3. Hyphal growth of the ∆abpf2 mutant and wild-type Alternaria brassicicola on the leaves of green cabbage.
tpj12217-sup-0004-FigureS4.epsimage/eps1166KFigure S4. Hierarchical clustering of fungal RNA-seq data.
tpj12217-sup-0005-FigureS5.epsimage/eps1245KFigure S5. Schematic diagram of the PCR strategy used to create all constructs.
tpj12217-sup-0006-TableS1.docWord document30KTable S1. Number of conidia produced by Δabpf2 and wild-type Alternaria brassicicola during saprophytic growth on potato dextrose agar.
tpj12217-sup-0007-TableS2.docWord document28KTable S2. Germ tube growth (μm) over time for the ∆abpf2 mutants and wild-type Alternaria brassicicola on the host plant B. oleracea.
tpj12217-sup-0008-TableS3.docWord document28KTable S3. Frequency of appressorium formation during early pathogenesis.
tpj12217-sup-0009-TableS4.xlsapplication/msexcel495KTable S4. List of fungal genes differentially expressed in the Δabpf2 mutants compared to the wild-type at 12 hpi.
tpj12217-sup-0010-TableS5.xlsapplication/msexcel232KTable S5. List of fungal genes differentially expressed in the Δabpf2 mutants compared to the wild-type at 48 hpi.
tpj12217-sup-0011-TableS6.xlsapplication/msexcel3226KTable S6. List of plant genes differentially expressed in plants inoculated with the Δabpf2 mutants compared to plants inoculated with wild-type Alternaria brassicicola at 12 hpi.
tpj12217-sup-0012-TableS7.xlsapplication/msexcel119KTable S7. Statistically over-represented functional annotation terms among plant genes differentially expressed at 12 hpi.
tpj12217-sup-0013-TableS8.xlsapplication/msexcel56KTable S8. List of genes with leucine-rich repeat domains.
tpj12217-sup-0014-TableS9.docWord document75KTable S9. List of primers used for quantitative real-time PCR, sequencing, or transformation constructs.
tpj12217-sup-0015-MethodS1.docWord document30KMethods S1. Complementation of two Δabpf2 mutant strains.
tpj12217-sup-0016-MethodS2.docWord document54KMethods S2. RNA-seq data generation, gene expression profile analysis, and identification of transcription factor binding sites.
tpj12217-sup-0017-FigureLegends.docWord document31K 

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.