ZafA‐mediated regulation of zinc homeostasis is required for virulence in the plant pathogen Fusarium oxysporum

Summary

During infection, soilborne fungal pathogens face limiting conditions of different metal ions, including zinc. The role of zinc homeostasis in fungal pathogenicity on plants remains poorly understood. Here it is shown that the transcription factor ZafA, orthologous to Saccharomyces cerevisiae Zap1, functions as a key regulator of zinc homeostasis and virulence in Fusarium oxysporum, a cross‐kingdom pathogen that causes vascular wilt on more than 100 plant species and opportunistic infections in humans. Expression of zafA is induced under zinc‐limiting conditions and repressed by zinc. Interestingly, zafA is markedly up‐regulated during early stages of plant infection, suggesting that F. oxysporum must cope with limited availability of zinc. Deletion of zafA results in deactivation of high‐affinity zinc transporters, leading to impaired growth under zinc deficiency. Fusarium oxysporum strains lacking ZafA are reduced in their capability to invade and kill tomato plants and the non‐vertebrate animal model Galleria mellonella. Collectively, the results indicate that ZafA‐mediated adaptation to zinc deficiency is required for full virulence of F. oxysporum on plant and animal hosts.

Introduction

Soilborne fungal pathogens are ubiquitous, extremely persistent, and very difficult to keep at bay, producing enormous losses in field and greenhouse crops. Agricultural practices, like crop rotation, resistance breeding and fungicide application, are insufficient to prevent root diseases (Haas and Defago, 2005). Fusarium oxysporum is one of the most important soilborne fungal pathogens (Dean et al., 2012), causing vascular wilt disease in more than 100 different plant species (Armstrong and Armstrong, 1981). Additionally, F. oxysporum isolates also produce opportunistic infections in humans of varying severity depending on the immune status of the patient (Nucci and Anaissie, 2007). Fusarium oxysporum f. sp. lycopersici FGSC 9935 (FOL 4287) is a fully sequenced strain (Ma et al., 2010) able to infect and kill both tomato plants and immunosuppressed mice (Ortoneda et al., 2004). Thus, this isolate represents an exceptional model for the study of the genetic basis of cross‐kingdom pathogenicity in fungi.

Zinc is an essential metal ion that plays crucial roles in numerous cellular processes as a structural or catalytic component of many different enzymes (Andreini et al., 2008; Auld, 2009). On the other hand, its excess is highly toxic to the cell. Although this metal is usually abundant, it is frequently found in insoluble forms (Broadley et al., 2007; Gerwien et al., 2018), therefore organisms require fine‐tuned mechanisms to regulate the balance between uptake, storage and use of this microelement. Zap1, the major regulator of the adaptation to zinc‐limiting conditions in the model organism Saccharomyces cerevisiae (Eide, 2009; Zhao and Eide, 1997), was the first zinc‐responsive transcription factor identified in fungi. Since then, functional orthologues of Zap1 have been characterized in Candida albicans, Cryptococcus gattii and Aspergillus fumigatus (Wilson and Bird, 2016), the latter called ZafA (Moreno et al., 2007). In addition, interestingly, it has been demonstrated in A. fumigatus that ZafA‐mediated adaptation to zinc‐replete conditions is influenced by iron concentration through the regulation of alternative transcription units of zafA and the basal amount of this transcription factor (Vicentefranqueira et al., 2019).

Several genome‐wide transcriptional analyses have shown that ZafA/Zap1 regulates the expression of many genes in response to zinc deficiency (Lyons et al., 2000; North et al., 2012; Vicentefranqueira et al., 2018), including high‐affinity membrane zinc transporters. Consequently, inactivation of ZafA/Zap1 results in deficient growth under zinc‐limiting conditions (Herbig et al., 2005; Moreno et al., 2007; Zhao and Eide, 1997). Importantly, since zinc availability is extremely limited within mammalian hosts (Maret, 2013), ZafA/Zap1‐mediated adaptation to zinc deficiency is essential for virulence in several fungal human pathogens (Gerwien et al., 2018). Although ZafA/Zap1 is conserved throughout the fungal kingdom, its function in pathogenicity on plants has not been explored so far. Here, the role of ZafA in zinc homeostasis and the infection process of F. oxysporum was studied. It is shown that ZafA is a major regulator of the transcriptional response to zinc limitation and provide the first evidence for its relevance during fungal infection of plants. Experimental procedures are provided in Supporting Information Text S1.

A BLASTP search in FungiDB using ZafA from A. fumigatus (Afu1g10080) as bait identified a single F. oxysporum ZafA orthologue (FOXG_00370). Manual inspection of the ClustalW alignment of both proteins identified two N‐terminal putative activating domains with very low identity in F. oxysporum ZafA, as well as six C‐terminal zinc fingers (ZnF) with 45% identity and a nuclear localization signal in ZnF number six (Fig. S1). Importantly, these features are characteristic of zinc‐binding transcription factors regulating the adaptive response to zinc deficiency in different fungi (Vicentefranqueira et al., 2018). Next, transcript levels of F. oxysporum zafA, as well as of zrfA (FOXG_03983) and zrfB (FOXG_10454), the orthologues of two high‐affinity membrane zinc transporters in A. fumigatus (Vicentefranqueira et al., 2005), were compared in steady‐state zinc‐limiting (–Zn) and zinc‐replete (+Zn) conditions. As expected, expression of zafA, zrfA and zrfB was significantly higher in –Zn compared to +Zn (Fig. 1A). To further analyse the regulation of this set of genes, the transcript levels during short‐term adaptation from zinc‐limiting to zinc‐replete conditions (sZn) were measured. Interestingly, most of the zafA, zrfA and zrfB mRNAs were lost within 15 min after adding zinc (Fig. 1B). Collectively, these results indicate that transcription and transcript stability of zafA, and of zrfA and zrfB, are regulated by zinc in F. oxysporum.

Figure 1.

Zinc regulates zafA, zrfA and zrfB transcript levels in Fusarium oxysporum. (A) Quantitative real‐time reverse transcription (RT)‐PCR performed in the wild‐type strain germinated for 14 h at 28 °C in potato dextrose broth (PDB) and then transferred for 6 h to 0 µM ZnSO₄ (–Zn) and 500 µM ZnSO₄ (+Zn) minimal medium (MM) (upper panel). Transcript levels of the indicated genes are expressed relative to those obtained in –Zn (lower panel). (B) Quantitative real‐time RT‐PCR performed in the wild‐type strain germinated for 14 h at 28 °C in PDB, transferred for 6 h to 0 µM ZnSO₄ (–Zn) MM and then supplemented with 100 µM ZnSO₄ (sZn) for the indicated time periods (upper panel). Transcript levels of the indicated genes are expressed relative to those obtained in –Zn (lower panel). Bars represent standard deviations from two independent biological experiments with three technical replicates each.

To study the role of ZafA in F. oxysporum, several zafAΔ strains were generated by replacing the entire FOXG_00370 coding sequence with the hygromycin B resistance gene (Fig. S2). To determine whether FOXG_00370 is responsible for adaptation to zinc‐limiting conditions, the different strains were cultured on solid minimal medium (MM) without or with different ZnSO₄ concentrations. The wild‐type strain was able to grow in all tested conditions, although growth was weaker in the absence of zinc. In contrast, zafAΔ was unable to grow under zinc depletion or in the presence of low concentrations of ZnSO₄. On the other hand, zafAΔ grew like the wild‐type strain in the presence of high concentrations of zinc (Fig. 2A). Reintroduction of an intact zafA allele into zafAΔ, yielding the complemented strain zafAΔ^C (Fig. S2), fully restored wild‐type growth (Fig. 2A).

Figure 2.

Loss of ZafA impairs growth of Fusarium oxysporum under zinc‐limiting conditions. (A) Growth of the indicated fungal strains on minimal medium (MM) plates supplemented with the specified concentrations of ZnSO₄. Cultures were grown for 2 days at 28 °C. (B) Quantitative real‐time reverse transcription‐PCR performed in the indicated strains germinated for 14 h at 28 °C in potato dextrose broth and then transferred for 6 h to 0 µM ZnSO₄ MM. Transcript levels of the indicated genes are expressed relative to those obtained in the wild‐type strain. Bars represent standard deviations from two independent biological experiments with three technical replicates each. wt, wild type.

In A. fumigatus, transcriptional induction of the high‐affinity membrane zinc transporters zrfA and zrfB under zinc‐limiting conditions depends on ZafA (Moreno et al., 2007; Vicentefranqueira et al., 2018). Here in F. oxysporum, activation of zrfA and zrfB expression by zinc deficiency was completely abolished in zafAΔ and restored in the complemented strain (Fig. 2B). Consequently, ZafA is responsible for the adaptation to zinc‐limiting conditions in F. oxysporum through transcriptional activation of membrane zinc transporters.

It was noted that transcription of F. oxysporum zafA was highly induced during the early stages of plant infection, suggesting the presence of zinc‐limiting conditions (Fig. 3A), therefore the role of zafAΔ in virulence was tested. Tomato plants inoculated with conidia of the F. oxysporum wild‐type or zafAΔ^C strains showed progressive wilt symptoms and usually died before 20 days post‐inoculation (dpi). In contrast, plants inoculated with zafAΔ showed a significant delay in the development of disease symptoms and plant mortality (Fig. 3B). Moreover, the amount of fungal biomass in roots and stems was markedly reduced in plants infected with zafAΔ in comparison to those inoculated with the wild‐type or complemented strains (Fig. 3C). To determine whether or not ZapA is required for invasive growth of F. oxysporum on living host plant tissue, tomato fruits were inoculated by puncturing the epidermis with a sterile pipette tip and injecting a microconidial suspension into the fruit tissue. zafAΔ was reduced in its ability to colonize and rot tomato fruits compared to the wild‐type or zafAΔ^C strains (Fig. 4). Importantly, addition of 100 µM ZnSO₄ to the conidial suspension during inoculation of tomato fruits partially rescued invasive growth of zafAΔ (Fig. 4). Taken together, these results demonstrate that ZafA‐mediated adaptation to zinc limitation is required for full virulence of F. oxysporum on tomato plants and for efficient invasive growth and maceration of living plant tissue.

Figure 3.

ZafA is required for full virulence of Fusarium oxysporum on tomato plants. (A) Quantitative real‐time reverse transcription‐PCR performed in the wild‐type strain germinated for 14 h at 28 °C in potato dextrose broth and then transferred for 6 h to 0 µM ZnSO₄ (–Zn) minimal medium or from inoculated tomato roots at 4 days post‐inoculation (dpi). Transcript levels of zafA are expressed relative to those obtained in –Zn. Bars represent standard deviations from two independent biological experiments with three technical replicates each. (B) Groups of 20 tomato plants (cv. Monika) were inoculated with a suspension of freshly obtained microconidia of the indicated fungal strains. Percentage survival was plotted for 25 days. Data shown are from one representative experiment. Experiments were performed three times with comparable results. ***P < 0.0005. (C) Quantitative real‐time PCR was used to measure the relative amount of fungal DNA in total genomic DNA extracted from tomato roots (left panel) and stems (right panel) 4 and 7 dpi with the indicated fungal strains. Amplification levels are expressed relative to those obtained from plants infected with F. oxysporum wild‐type strain 4 dpi. Bars represent standard deviations from two independent biological experiments with three technical replicates each. *P < 0.05, **P < 0.005. wt, wild type.

Figure 4.

ZafA is required for invasive growth of Fusarium oxysporum on living host plant tissue. Tomato fruits (cv. Daniela) were inoculated with a suspension of freshly obtained microconidia of the indicated fungal strains supplemented or not with a 100 µM ZnSO₄ solution. Fruits were maintained for 3 days at 28 °C and 100% humidity.

ZafA/Zap1‐mediated regulation of zinc homeostasis was previously shown to be essential for virulence in several fungal human pathogens, such as A. fumigatus (Moreno et al., 2007), Candida dubliniensis (Bottcher et al., 2015) or Cryptococcus gattii (Schneider Rde et al., 2012). Because F. oxysporum also produces opportunistic infections in humans (Nucci and Anaissie, 2007), the invertebrate model Galleria mellonella was used for virulence studies in animal hosts. Inoculation with conidia of the wild‐type or zafAΔ^C strains resulted in the killing of all larvae 10 dpi, while animals inoculated with zafAΔ showed significantly delayed mortality (Fig. S3). Thus, ZafA‐mediated adaptation to zinc deficiency is required for virulence of F. oxysporum on an animal host.

Zinc is an essential metal ion. Although relatively abundant on earth, it has a low bioavailability, but can also become toxic in excess, therefore organisms have developed efficient strategies to maintain zinc homeostasis. Fungi, which are mostly composed of saprophytes thriving on dead organic material, obtain zinc from water, soil or organic complexes. In solution, zinc remains in its oxidized Zn²⁺ form and, unlike Fe²⁺ or Cu²⁺, is redox‐stable. However, insoluble zinc represents >90% in the soil (Broadley et al., 2007). Approximately, 10 000 fungal species have evolved a plant‐parasitic lifestyle (Agrios, 1997), whereas about 400 have been reported as pathogens of mammals (De Hoog and Guarro, 1995). These species must obtain zinc from the living host and, although zinc concentrations in eukaryotic cells are often high, free Zn²⁺ is extremely limited (Maret, 2013). In fungi, zinc deficiency activates expression of genes required for zinc uptake while zinc storage mechanisms are repressed (Rutherford and Bird, 2004; Vicentefranqueira et al., 2018). Here it was hypothesized that the soilborne pathogen F. oxysporum must face zinc limitation during both the saprophytic and pathogenic stages of its lifecycle. The study shows that ZafA, a conserved regulator of fungal zinc homeostasis, is required for adaptation of F. oxysporum to zinc‐limiting conditions, and that loss of ZafA affects both fungal growth as well as virulence on plant and animal hosts.

The data show that ZafA is essential for growth of F. oxysporum under zinc‐limiting conditions but not under zinc sufficiency or zinc excess. These results are in line with those reported in S. cerevisiae (Herbig et al., 2005; Zhao and Eide, 1997) and A. fumigatus (Moreno et al., 2007). The specific role during zinc limitation is supported by the ZafA‐dependent transcriptional activation of the high‐affinity membrane zinc transporters zrfA and zrfB, as previously reported in A. fumigatus (Moreno et al., 2007). In fungi, excess zinc is stored in the vacuole (Cai et al., 2018; Eide, 2009). Unlike iron homeostasis, where the single transcription factor HapX regulates adaptation to both iron‐limiting and iron excess conditions (Gsaller et al., 2014), ZafA is not involved in zinc detoxification. In A. fumigatus, transcription of the main zinc vacuolar transporter zrcA is repressed by ZafA during zinc deficiency (Vicentefranqueira et al., 2018) and activated by AceA, the major copper detoxification transcription factor (Wiemann et al., 2017) during zinc excess (Cai et al., 2018). Collectively, the results establish a conserved function of ZafA in activation of zinc uptake and highlight the essential role of this transcription factor in fungal adaptation to zinc deficiency.

Because the bioavailability of free Zn²⁺ is very low in mammalian cells or tissues (Maret, 2013), ZafA/Zap1‐mediated adaptation to zinc deficiency is required for virulence in different fungal human pathogens (Bottcher et al., 2015; Moreno et al., 2007; Schneider Rde et al., 2012). However, the relevance of this mechanism in pathogenicity on plants has not been examined so far. This study found that transcription of zafA increased dramatically during early stages of plant infection, indicating that, like in mammalian cells, free Zn²⁺ is extremely limited. It is known that soluble Zn²⁺ is not abundant in soils and therefore plants possess efficient mechanisms for uptake and storage of this micronutrient (Broadley et al., 2007). This study found that F. oxysporum zafAΔ is significantly delayed in its capacity to invade and cause mortality in tomato plants. Moreover, zafAΔ was unable to efficiently colonize living plant tissue, unless an excess of exogenous ZnSO₄ was added. Therefore, ZafA contributes to survival and proliferation of the fungus in the plant host through transcriptional reprogramming under severe zinc limitation. This provides the first evidence for a role of ZafA in virulence of a plant pathogen. Targeting zinc homeostatic mechanisms may therefore represent a new promising strategy for the control of fungal phytopathogens.

Supporting information

Fig. S1 Amino acid sequence alignment of Fusarium oxysporum and Aspergillus fumigatus ZafA. The predicted F. oxysporum zafA gene product (FOXG_00370) was aligned with A. fumigatus ZafA (Afu1g10080). Conserved and similar residues are shaded in black and grey, respectively. Putative activated domains are boxed in blue. CWCH2‐ and C2H2‐type zinc fingers are boxed in green and red, respectively. The predicted nuclear localization signal within the last zinc finger is boxed in purple.

Fig. S2 Targeted replacement of the Fusarium oxysporum zapA gene. (A) Physical maps of the F. oxysporum zafA locus and the split‐marker gene replacement constructs obtained by fusion PCR. Relative positions of restriction sites and PCR primers are indicated. hyg, hygromycin resistance gene. (B) PCR amplification of genomic DNA of the wild‐type strain (wt) and three independent transformants, using primers zafA‐5′‐F and zafA‐5′‐R. (C) Southern blot analysis. gDNA of the wt and three independent transformants was treated with BamHI, separated on a 0.7% agarose gel, transferred to a nylon membrane and hybridized with the DNA probe indicated in A. (D) PCR amplification of gDNA of the wt, zafAΔ and two independent complemented transformants, using primers zafA‐qPCR‐F and zafA‐qPCR‐R.

Fig. S3 ZafA is a virulence factor of Fusarium oxysporum in the invertebrate animal model Galleria mellonella. Groups of 20 G. mellonella larvae were infected with freshly obtained microconidia of the indicated fungal strains. Percentage survival was plotted for 15 days. Data shown are from one representative experiment. Experiments were performed three times with comparable results. **P < 0.005.

Text S1 Experimental procedures. Fungal isolates and culture conditions. Nucleic acid manipulations and quantitative real‐time reverse transcription‐PCR analysis. Primers used in this study. Fungal strains. Infection assays. Sequence alignment. References.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.