Host to a Stranger: Arabidopsis and Fusarium Ear Blight

Transcription

1 Review Host to a Stranger: Arabidopsis and Fusarium Ear Blight Helen C. Brewer 1 and Kim. E. Hammond-Kosack 1, * Fusarium ear blight (FEB) is a devastating fungal disease of cereal crops. Outbreaks are sporadic and current control strategies are severely limited. This review highlights the use of Arabidopsis to study plant FEB interactions. Use of this pathosystem has identified natural variation in Fusarium susceptibility in Arabidopsis, and native plant genes and signalling processes modulating the interaction. Recent breakthroughs include the identification of plant- and insect-derived small molecules which increase disease resistance, and the use of a host-induced gene silencing (HIGS) construct to silence an important Fusarium gene to prevent infection. Arabidopsis has also been used to study other fungi that cause cereal diseases. These findings offer the potential for translational research in cereals which could yield much-needed novel control strategies. FEB A Continuing Global Threat to Cereal Crops and Food Security FEB is a globally important fungal disease of wheat and other cereals (see Glossary). This floral disease is caused by several Ascomycete species of the genus Fusarium, primarily Fusarium graminearum and Fusarium culmorum. Outbreaks occur sporadically, favoured by warm, wet weather at crop anthesis. FEB epidemics can cause 50 70% loss of marketable grain. Climate change is predicted to positively impact on the frequency and severity of epidemics. In addition to reducing grain weight and quality, most causal species produce mycotoxins such as deoxynivalenol (DON), which are harmful to mammals [1 5] ( FEB control measures are currently very limited. Chemical control with fungicides is impeded by an early symptomless disease phase; applications must occur before symptoms appear [6]. This requires either prophylactic treatment, which may be costly and unnecessary, or advanced forecasting systems, which are not always accurate [7]. Furthermore, the dominant Fusarium species F. graminearum is inherently resistant to azole fungicides owing to the presence of multiple copies of the gene encoding the target enzyme CYP51, and exposure to some azoles may lead to increased DON accumulation [8,9]. Current control efforts focus heavily on marker-assisted breeding for genetic resistance traits. However, all sources of resistance currently identified in hexaploid wheat (Triticum aestivum) are quantitative trait locus (QTL)-based, and the identity of the underlying genes has not yet been reported. Furthermore, introgression of the most promising QTLs into elite wheat varieties has not to date resulted in high levels of FEB resistance [10 13]. Novel chemical and/or genetic control strategies for FEB are therefore urgently needed. This article will review the use of the model plant Arabidopsis as an experimental host for cereal-infecting Fusaria. This pathosystem Trends Arabidopsis provides a valuable model for the study of Fusarium ear blight, a devastating fungal disease of cereals. Arabidopsis defence against Fusarium graminearum is evidently mediated by SA signalling, with JA signalling facilitating early infection. This is less evident for F. culmorum infection. Transgenic expression and/or chemical application of several small antimicrobial peptides has been shown to enhance resistance to Fusarium in Arabidopsis and cereals. Host-induced gene silencing (HIGS) of pathogen genes is a promising tool for enhancing plant resistance to Fusarium and other pathogens. Arabidopsis is increasingly being used as a model system for research on fungal pathogens of cereals and to explore non-host immunity mechanisms, host defence responses, and/ or the role(s) of specific pathogen virulence determinants. 1 Plant Biology and Crop Science, Rothamsted Research, Harpenden AL5 2JQ, UK *Correspondence: kim.hammond-kosack@rothamsted.ac. uk (K.E. Hammond-Kosack). Trends in Plant Science, October 2015, Vol. 20, No Elsevier Ltd. All rights reserved.

2 has been used to study the genetic and molecular basis of plant Fusarium interactions, natural variation in Fusarium resistance, and novel transgenic and/or chemical approaches to limit infection which have the potential to provide much needed additional control strategies in crop species. The use of Arabidopsis as a model host to investigate other cereal pathogens is also highlighted. However, further work will be necessary to fully understand these pathosystems and to assess the translatability of findings into cereal crop protection strategies (see Outstanding Questions Box). Arabidopsis as a Model for FEB Infection The large, complex genome of wheat and other cereals has impeded functional genomics studies into host resistance. By contrast, the model plant Arabidopsis thaliana has a small, diploid, fully sequenced genome, which extends to a large number of different accessions [14 16]. This is supported by large collections of genetic mutants, and online resources such as annotated genome browsers that are linked to gene expression and pathway data [17 21]. Arabidopsis has been widely adopted to study the genetic basis of plant pathogen interaction outcomes, results from which have the potential to be translated into important crop traits [22,23]. This includes its use as a model for studying infection by cereal pathogens, including FEB-causing Fusarium isolates. It has been demonstrated that both F. culmorum and F. graminearum infect Arabidopsis floral and silique tissue, but that this infection does not spread to the main stem or leaf tissue, thus being analogous to FEB infection of cereals, which is predominantly limited to the ear/floral spike [24]. DON is also produced during Arabidopsis infection. Arabidopsis floral infection by FEBcausing Fusarium species was therefore put forward as a suitable model for studying plant defence signalling against the causal agents of FEB, without the need for complex genetic studies in wheat (Figure 1). Since the publication of this original study, several research groups have used the Fusarium Arabidopsis pathosystem to investigate the role of various defence-associated genes and signalling pathways in determining the outcome of this interaction, as well as variation in susceptibility among accessions. While some studies have used the originally published floral infection system and disease-scoring method [24], others have developed leaf and/or seedling pathosystems. Variation in Fusarium Susceptibility between Arabidopsis Ecotypes The original screen of 236 Arabidopsis accessions did not find any that were extremely resistant or susceptible to Fusarium floral infection [24]. However, this initial study identified that inoculation of the Landsberg erecta 0 (Ler-0) accession results in consistently severe floral infection with low variability, while infection of accession Columbia-0 (Col-0), which has a wild-type ERECTA gene, was highly variable. This difference was attributed to compaction of the Arabidopsis inflorescence caused by mutation of erecta, analogous to the compact wheat ear morphology of modern elite varieties, facilitating disease spread. This is corroborated by the recent finding that mutation of ERECTA increases floral susceptibility in Col-0 (H.C.B., PhD thesis, University of Exeter, 2015). A detached Arabidopsis leaf assay was developed that involves wounding and the application of exogenous DON (75 mm) to induce consistent F. graminearum infection in Arabidopsis rosette leaves [25]. Using this method, differences were found in susceptibility between Arabidopsis accessions; Bay-0, Kas-1, and Ler-0 were more susceptible to F. graminearum than were four distinct Columbia lines. These leaf results indicate that differences in susceptibility between Col- 0 and Ler-0 are not purely the result of erecta acting on the floral morphology. Indeed, QTL analysis on a Col-0 Ler-0 mapping population revealed that Fusarium susceptibility was associated with a major QTL on chromosome 4 that was not linked to the erecta mutation. Glossary Anthesis: the period of flower opening. Deoxynivalenol (DON): a toxic epoxy-sesquiterpenoid compound produced by some Fusarium fungi. DON detoxification (DON detox): involves glycosylation of DON to form a non-toxic compound. Ethylene (ET): a plant hormone involved in biotic defence responses and senescence. Fusarium ear blight (FEB): an important fungal disease of cereals, also known as head blight and head scab. Host-induced gene silencing (HIGS): silencing of pathogen genes by native or transgenically expressed RNAi from the host. Hypersensitive response (HR): characterised by localised rapid cell death that prevents pathogen growth. Jasmonic acid (JA): a plant hormone involved in biotic defence and development. Methyl jasmonate (MeJA): an elicitor of JA-mediated plant defence responses. Mycotoxins: toxic compounds produced by plant-infecting fungi. Non-host resistance: the innate mechanisms by which all variants of a plant species are resistant to all variants of a pathogen species. Programmed cell death (PCD): regulated death of plant cells that is required for the hypersensitive response (HR) and/or for a resistant response. Quantitative trait locus (QTL): a genomic region containing genes that correlates with variation in a phenotype (a trait). Resistance (R) gene: confers immunity to pathogen strains possessing a corresponding effector molecule. Salicylic acid (SA): a plant hormone involved in biotic defence and development. Thionins: a family of small proteins found in higher plants, some of which have toxic/antimicrobial properties. 652 Trends in Plant Science, October 2015, Vol. 20, No. 10

3 (A) (B) (D) (F) (C) (E) Figure 1. Fusarium Ear Blight Disease of Wheat and Arabidopsis. Field infection of wheat (A,F) compared to experimental infection of Arabidopsis floral (B,C) and silique (D,E) tissue. In both species, infection is predominantly limited to floral tissue (A,C). Silique infection (D) results in shrivelled Arabidopsis seeds and premature silique senescence (E) comparable to shrivelled and bleached infected wheat kernels (grains) (A, inset F) and premature ear senescence (A). Infection of Arabidopsis commences with colonisation of the anthers and petals (B), whereas in wheat infections occur via the anthers as well as the inner glumes, lemma and palea Rothamsted Research Ltd. All Rights Reserved. It is worth noting that this leaf study relied on significant manipulation of the Arabidopsis Fusarium interaction via wounding and exogenous DON application, and that detachment of the leaves might also impact on the plant defence response due to altered hormone levels (M. Grant, University of Exeter, personal communication). Arabidopsis Genes Screened for Effects on Fusarium Susceptibility Several studies have used gene deletion/insertion mutants and/or overexpression lines to study the role of specific defence-related genes on the interaction between wheat-infecting Fusarium strains and Arabidopsis floral and/or leaf tissue. The outcomes of these individual studies are summarised in Table 1 and detailed below. These studies predominantly assessed which already known plant defence hormones and signalling pathways are required for defence against Fusarium. It is generally understood that, in Arabidopsis, salicylic acid (SA)-mediated signalling is required for defence against biotrophs (pathogens that colonise and feed from living host tissue), whereas defence against necrotrophs (pathogens that kill the host cells and feed off of the dead tissue) relies on jasmonic acid (JA) and/or ethylene (ET) signalling, with antagonism between the two pathways [26 29]. However, there are many exceptions to this rule and questions about its validity in other plant species Trends in Plant Science, October 2015, Vol. 20, No

4 Table 1. Arabidopsis Mutants with Altered Defence Responses to Fusarium graminearum (Fg) and Fusarium culmorum (Fc) Strains a Genotype Gene function Pathogen Inoculation Assessment Disease Refs coi1 JA signalling Fc 98/11 Fg Z-3639 Fg DAOM Floral spray Leaf infiltration Seeding inoculation Disease score % Leaf infected Cotyledon infection b [39] [41] [59] cpr5 SA and JA downregulation Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [41] dmr1 Homoserine phosphorylation Fc 98/11 Floral and rosette leaf spray Disease score, infected leaves [60] eds1-1 SA-mediated defence Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score + [45] eds1-2 SA-mediated defence Fc 98/11 Floral spray Disease score Wt d [40] eds11 Unknown Fc 98/11 Floral spray Disease score, DON + c [39] ein2 ET signalling Fc 98/11 Fg UK1 Floral spray Detached leaf/flower Disease score, DON Disease severity, spore count + [39] [47] ERF1 OE JA/ET signalling Fc 98/11 Floral spray Disease score, DON Wt [39] esa1 Unknown Fc Floral spray Disease score + [35] eto1 ET regulation Fc 98/11 Floral spray Disease score, DON Wt [39] etr1 ET signalling Fc 98/11 Floral spray Disease score, DON Wt [39] GLK1 OE Chloroplast development Fg Z-3639 Fg DAOM jar1 JA signalling Fc 98/11 Fg Z-3639 Fg DAOM Leaf infiltration Seeding inoculation Floral spray Leaf infiltration, floral spray Seedling inoculation Imaging Cotyledon infection Disease score % Leaf infected, disease score Cotyledon infection jar1 npr1-1 See single mutants Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score ++ e [41] lox1 Lipoxygenase involved in JA signalling Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [46] lox5 Lipoxygenase involved in JA signalling Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [46] opr3 JA signalling Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [41] NahG OE f SA degradation Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score + [38,41] npr1-1 Central regulator Fc 98/11 Fg Z-3639 Floral spray Leaf infiltration, floral spray Disease score, DON % Leaf infected, disease score [38] [59] [39] [41] [59] + [39] [41] NPR1 OE Central regulator Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [41] pad4 SA-mediated defence Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score + [45] sag101 SA-mediated defence Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score + [45] sgt1b R gene-mediated defence Fc 98/11 Floral spray Disease score, DON [40] sid2 SA synthesis Fc 98/11 Fg Z-3639 Floral spray Leaf infiltration, floral spray Disease score % Leaf infected, disease score ssi2 SA downregulation Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [41] wrky18 NPR1-mediated defence Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score + [41] WRKY18 OE NPR1-mediated defence Fg Z-3639 Leaf infiltration, floral spray % Leaf infected, disease score [41] a This table summarises the effects of alterations in expression of innate Arabidopsis genes or modulation of innate processes transgenic and chemical approaches to disease control are excluded. Abbreviations: DON, deoxynivalenol; ET, ethylene; JA, jasmonic acid; OE, overexpression of a transgene; SA, salicylic acid; R, resistance; Wt, wild type. b Indicates less disease than wild type. c Indicates more disease than wild type. d Indicates equivalent disease to wild type. e Indicates increased disease in the jar1 npr1 double mutant compared to the npr1 single mutant. f Expression of a transgene. Wt + [39] [41] 654 Trends in Plant Science, October 2015, Vol. 20, No. 10

5 [30 32]. Furthermore, Fusarium is considered to have a hemi-biotrophic or 'switching' lifestyle in wheat, with asymptomatic colonisation of living tissue preceding necrotrophy [33], suggesting that multiple pathways could be required for, or indeed inhibit, defence. Fusarium is considered to exhibit an equivalent 'hemi-biotrophic' lifestyle in Arabidopsis based on hyphal colonisation of living tissue [24], although a symptomless phase has not been formally identified. The Arabidopsis mutant esa1 (enhanced susceptibility to Alternaria 1) carries a mutation in an unmapped gene, which increases susceptibility to several necrotrophs. The mutation does not alter susceptibility to biotrophs such as the oomycete Hyaloperonospora arabidopsidis. Further analysis revealed that esa1 has attenuated production of the antimicrobial compound camalexin as well as diminished JA- and ET-induced defence responses [34,35]. A screen of the esa1 mutant for floral susceptibility to several pathogens of the genus Fusarium (including the FEBcausing species F. graminearum and F. culmorum) showed that the esa1 mutant had increased floral susceptibility to F. culmorum. A similar but non-significant trend was observed for F. graminearum. This finding indicated that ESA1-mediated defence against necrotrophs is also involved in defence against cereal-infecting F. culmorum. The Arabidopsis gene GLK1 (Golden Like Kinase 1) is a transcriptional activator involved in chlorophyll biosynthesis [36,37]. The overexpression of GLK1 resulted in upregulation of several defence-related genes, including antimicrobial peptides, but downregulation of PR1 (PATHO- GENESIS RELATED 1), a marker for SA-mediated defence [38]. Arabidopsis plants overexpressing GLK1 supported less F. graminearum growth in leaves inoculated using a needleless syringe. The most extensive studies on responses of Arabidopsis mutants to wheat-infecting Fusarium have so far been reported by Cuzick et al. [39,40] and Makandar et al. [41]. The former used the floral spray assay devised by Urban et al. [24] and a F. culmorum inoculum, while the latter used a combination of the floral spray assay and leaf syringe infiltration, both with F. graminearum. Functional Evaluation of Genes Involved in SA Signalling Mutation of NPR1 (NON-EXPRESSOR OF PR1) [42], a gene required for activation of SA signalling and downregulation of JA signalling, resulted in increased Fusarium susceptibility in two independent studies [39,41]. In one study the authors attributed this to the requirement for a functional SA signalling pathway for defence against F. graminearum in view of the susceptibility profiles of other SA signalling mutants and transgenic lines (Table 1) [41]. For example, the SA induction mutant sid2-2, as well as transgenic lines expressing the bacterial SA hydroxylase NahG, which converts SA to catechol, showed increased F. graminearum susceptibility in both leaves and floral tissue. By contrast, the second study using F. culmorum did not observe a significant difference in floral infection levels between sid2-2 and wild-type plants, casting doubt over the role of SA signalling in resistance to F. culmorum and suggesting that the npr1-1 susceptibility phenotype might be due to alterations in JA, rather than SA, signalling [39]. Similarly, later studies found no effect of mutation of EDS1 (ENHANCED DISEASE SUSCEPTIBILITY 1) or EDS5 (ENHANCED DISEASE SUSCEPTI- BILITY 5) on F. culmorum floral susceptibility [40]. Both these genes are required for SA-mediated defence against biotrophs [43,44]. However, mutation of EDS1 or of its signalling partners PAD4 (PHYTOALEXIN DEFICIENT 4) and SAG101 (SENESCENCE ASSOCIATED GENE 101) has recently been shown to compromise F. graminearum resistance in Arabidopsis leaves and flowers [45]. Furthermore, two Arabidopsis lipoxygenase mutants, lox1 and lox5, have enhanced SA and diminished JA signalling during F. graminearum infection, and increased F. graminearum resistance. Expression of these genes is upregulated during F. graminearum infection, suggesting that the fungus may manipulate Arabidopsis defence signalling to facilitate infection [46]. Overall, there is strong evidence of an SA requirement for defence in Arabidopsis against F. graminearum which is less apparent for F. culmorum. High floral susceptibility to F. culmorum has also been reported for the Pseudomonas-susceptible mutant eds11 (ENHANCED DISEASE SUSCEPTIBILITY 11) Trends in Plant Science, October 2015, Vol. 20, No

6 [39,54]. The genomic location and function of EDS11 are unknown, but this locus does not appear to be required for SA- or JA-mediated defence responses. The role of EDS11 in defence is therefore unclear. Functional Evaluation of Genes Involved in JA/ET Signalling Contrasting conclusions have been drawn on the roles of JA and ET signalling in defence against Fusarium. One study found that the JA signalling mutants coi1 and jar1 were more resistant to F. culmorum floral infection, but attributed this to alterations in floral morphology, such as increased stem elongation and decreased pollen, that affect nutrient availability to the fungus [39]. Only one of four ET signalling mutants studied, ein2 (ethylene insensitive 2), had reduced F. culmorum floral resistance, and the evidence for a role of ET signalling is therefore inconclusive. By contrast, another study put forward evidence for a role of JA signalling in F. graminearum susceptibility, based on decreased leaf and floral infection of JA signalling mutants jar1 and opr3 (Table 1) [41]. However, jar1 has altered floral morphology, and opr3 is male sterile [39,47]. Therefore, this study indicates a role for JA during either initial infection and/or internal colonisation of Arabidopsis leaves, but the floral data are inconclusive. Interestingly, the jar1 npr1-1 double mutant was found to be more susceptible to floral and leaf infection by F. graminearum than either wild-type plants or the npr1-1 single mutant [41]. This would suggest that JAR1, although initially contributing to susceptibility, may play a role in resistance at some later stage in infection, or that JA-mediated defence may help to block infection in the absence of NPR1- mediated defence responses. Evidence for the former hypothesis was supported by the finding that application of methyl jasmonate (MeJA, an elicitor of JA-mediated plant defence responses) early during leaf infection enhanced susceptibility, whereas later application enhanced resistance. This correlates with the findings that JA signalling is induced in resistant wheat plants, but not susceptible plants, 12 hours after inoculation [48]. One study reported that ET signalling mutants are more resistant to F. graminearum infection, using the previously described DON-amended detached leaf assay [49]. However, only the ein2 mutant showed significantly less disease on detached flowers. This contrasts with the findings that the ein2 mutation enhances floral susceptibility to F. culmorum [39]. Ethylene signalling was also found to contribute to disease susceptibility of detached wheat leaves and ears. It has been postulated that ET signalling contributes to DON-induced host cell death, thereby facilitating infection [49]. However, these results conflict with findings on the role of ET signalling in Fusarium resistant wheat lines [50]. Analysis of the function of ET signalling genes in FEB resistance in wheat is ongoing using an RNA silencing approach (S.R. Scofield et al., Purdue University, personal communications). The Role of the SGT1b RAR1 HSP90 Resistance-Related Complex The SGT1 (SUPPRESSOR OF G2 ALLELE OF SKP1) and RAR1 (REQUIRED FOR MLA12 RESISTANCE 1) proteins form part of an intracellular complex with HSP90 (HEAT SHOCK PROTEIN 90) which is involved in disease-resistance protein stabilisation, the hypersensitive response (HR), and regulates multiple plant pathogen interactions [51,52]. It has been shown that mutation of SGT1b, but not of SGT1a, increased resistance to F. culmorum floral infection, while rar1 mutants were unaffected [40]. Mutation of RAR1 was also recently found not to alter resistance to F. graminearum in Arabidopsis leaves or flowers [45]. The reason for this remains unclear. Previous studies indicate that abolition of programmed cell death (PCD) by silencing of SGT1 was responsible for enhanced resistance to necrotrophs in Nicotiana benthamiana [53]; however, no differences in PCD between sgt1b mutant and wild-type plants were found [40]. Collectively, these findings present a complex and at times contradictory picture of the relative roles of various defence signalling pathways in resistance to Fusarium in Arabidopsis (Figure 2). 656 Trends in Plant Science, October 2015, Vol. 20, No. 10

7 Key Figure Findings Relating to Fusarium Ear Blight Resistance in Arabidopsis Leaf and Floral Tissue, Compared to Wheat Arabidopsis Wheat DMR1 ERECTA SGT1b ESA1 EDS11 NPR1 SA Defensin Camalexin Glucosinolates Cinnamic acids Homoserine f b s ms Thionins PGIPs AtPAD4 AtNPR1 Sulfamethoxazole Gramine PGIPs sp r DON detox SA (0-24h) JA /ET (34+h) ET^ TaNPR1s Lipoxygenases Flavonoids Cinnamic acids ET^ SA DON detox GLK1 p Early l Late Sulfamethoxazole (C) Gramine (C) Thanan JA Lipoxygenases Host induced gene silencing (FgCYP51)^ Figure 2. Positive and negative regulators act either directly upon defence or upon other processes, as indicated by black arrows (positive) and red lines with bars (negative). Key to symbols: *, symptomless growth; ^, data from detached tissue assays; (C), data from cotyledon assays. Unbroken lines denote innate processes studied via the analysis of gene mutation lines (Table 1) or by comparative analyses of resistant and susceptible lines; broken lines denote transgenic or chemically explored processes. White text indicates findings relating to both plant species. Black text indicates findings only relating to one species. Box colours denote proteins (orange), pathways and processes (blue), and chemicals (grey). Plant tissues and/or organs are in black text. Abbreviations: b, bud; f, flowers; l, leaf; ms, main stem; p, peduncle; r, rachis; s, silique; sp, spikelet Rothamsted Research Ltd. All Rights Reserved. The current evidence appears to be that initial colonisation and subsequent disease development are limited by SA-mediated defence signalling, whereas JA- and ET-mediated signalling facilitate early infection, particularly in the case of F. graminearum on Arabidopsis leaves and wheat ears. However, roles for both JA and ET in resistance to later stages of infection have been noted in both wheat and Arabidopsis. Evidence for antagonism between pathways during the Fusarium Arabidopsis interaction is lacking. Furthermore, the extent to which the requirement Trends in Plant Science, October 2015, Vol. 20, No

8 for specific defence responses is dependent on the infecting fungal species is not clear because few direct comparisons have been made between F. culmorum and F. graminearum under the same laboratory conditions. Transgenic and Chemical Approaches to Disease Control The Arabidopsis Fusarium pathosystem has been used to test several transgenic and/or chemical approaches to controlling FEB in wheat. Enhancing Fusarium Resistance by Transgene Expression Constitutive expression of polygalacturonase-inhibiting proteins (PGIPs) in Arabidopsis resulted in inhibition of Fusarium polygalacturonase activity (involved in cell wall degradation) and enhanced floral resistance to F. graminearum [55]. Transgenic wheat plants expressing the bean PGIP PvPGIP2 showed reduced FEB symptoms, indicating the translatability of this study. Transgenic Arabidopsis plants expressing a barley (Hordeum vulgare) UDP-glucosyltransferase, which detoxifies DON, were able to grow on plates supplemented with high levels of DON, whereas wild-type plants died [56]. The resistance of these transgenic lines to Fusarium infection was not assessed; however, DON detoxification is an important component of FEB resistance in wheat [57]. Previous studies had identified native Arabidopsis UDP-glucosyltransferases with DON detoxification activity. However, overexpression of these native genes, while conferring enhanced DON tolerance, also resulted in unwanted pleiotropic effects [58,59]. Candidate UDPglucosyltransferases from wheat have been identified and functionally analysed in Arabidopsis and yeast, but were not found to detoxify DON [60]. UDP-glucosyltransferases from Brachypodium distachyon, rice (Oryza sativa), and sorghum (Sorghum bicolor) have also been identified and were found to detoxify DON when expressed in yeast, although further work is needed to test their effects in planta [61]. Use of Small Molecules without In Vitro Antifungal Activity to Control Fusarium Infections A high-throughput cotyledon infection assay was used to screen for chemicals which inhibit F. graminearum growth solely in planta [62]. Two chemicals, sulfamethoxazole and gramine, inhibited infection of Arabidopsis and also FEB disease and DON accumulation in wheat ears, although neither showed in vitro antifungal activity. Similarly, an allelic series of dmr1 (downy mildew resistant 1) mutants demonstrated enhanced silique and leaf resistance to F. culmorum infection [63]. DMR1 encodes Arabidopsis homoserine kinase, and mutants accumulate homoserine otherwise destined for phosphorylation and subsequent threonine and methionine biosynthesis. Exogenous homoserine application was found to enhance resistance in both wild-type and dmr1 mutant plants, but did not show in vitro antifungal activity. This complemented previous findings that homoserine enhances resistance to biotrophic mildew pathogens, indicating that homoserine may directly or indirectly confer broadspectrum disease resistance [64,65]. A possible link to homoserine-lactone-induced resistance mechanisms has not been investigated; however, homoserine-induced resistance appears not to be dependent on many of the known Arabidopsis defence signalling pathways/genes [65]. Use of Small Molecules With In Vitro Antifungal Activity to Control Fusarium Infections Several studies using transgenic approaches have shown a role of secreted antimicrobial defensin-like compounds in Arabidopsis resistance to Fusarium. Expression of the Medicago trunculata defensin MtDef4.2 targeted both intra- and extracellularly decreased silique infection by F. graminearum following point wound inoculations, and lowered DON accumulation [66]. MtDef4.2 also inhibited in vitro Fusarium growth. Extracellular targeting of this defensin also reduced growth of the obligate biotrophic oomycete pathogen 658 Trends in Plant Science, October 2015, Vol. 20, No. 10

9 Hyaloperonospora arabidopsidis. Likewise, the insect defensin thanatin from the spiny soldier bug Podisus maculiventris was found to inhibit F. graminearum in vitro growth, as well as reducing leaf infection by F. graminearum and the necrotrophic fungus Botrytis cinerea when transgenically expressed in planta [67]. Thionins, a family of small proteins with antimicrobial properties, have been found to accumulate in the cell walls of wheat spikes following infection with F. culmorum [68]. The Arabidopsis thionin Thi2.4 has been found to inhibit in vitro Fusarium growth, and constitutive expression in Arabidopsis reduced leaf and floral infection by F. graminearum. Thi2.4 is predicted to interact with fungal fruiting body lectin (FFBL), a demonstrated virulence factor from F. graminearum [69]. Collectively, these studies provide proof of concept for the use of transgenic Arabidopsis plants expressing small antimicrobial peptides to control FEB. This has the potential to bypass the issue of the complexity of defence signalling pathways and their conflicting roles in defence against pathogens with different lifestyles. However, in most cases further studies are needed to evaluate the translatability of these findings into cereal systems. HIGS of Fusarium Genes The Arabidopsis Fusarium pathosystem has been used to demonstrate a potential role of HIGS in disease control. F. graminearum contains three copies of CYP51, the gene encoding sterol 14ademethylase, which is important for growth and virulence on wheat and Arabidopsis [8]. Expression of RNA complementary to the three F. graminearum CYP51 genes in Arabidopsis and barley resulted in their silencing in the fungus, inhibiting fungal growth and infection of detached Arabidopsis and barley leaves [70]. This is one of the first successful examples of induced trans-species gene silencing, which may provide a novel effective mechanism for disease control [71]. Comparing the Wheat and Arabidopsis Pathosystems While functional studies involving Arabidopsis- and cereal-infecting Fusaria have continued at pace, several wheat transcriptome, metabolite, and gene function studies have also been reported. The outcomes of these studies are compared in Figure 2. Strikingly, the endogenous defence responses against Fusarium floral infection in the two species appear to have similar features. Multiple transcriptional analyses comparing FEB resistant and susceptible wheat cultivars have demonstrated the involvement of several signalling pathways in defence. In the resistant cultivar Wangshuibai, SA signalling was initially induced within 6 hours after infection, followed by ET and then JA signalling after 12 hours [48]. SA signalling was delayed and JA/ET signalling was reduced in a susceptible mutant. Similarly, JA and ET signalling-related genes were upregulated during infection of the resistant cultivar Dream compared to the susceptible cultivar Lynx [48]. This complements findings that both SA- and JA-mediated defence signalling may be recruited for defence against Fusarium in Arabidopsis [41]. Conversely, ET signalling was shown to contribute to F. graminearum susceptibility in detached foliar and floral tissues of Arabidopsis and wheat [49]. Three wheat orthologues of Arabidopsis NPR1 have been identified, and two were shown to be upregulated in a resistant compared to a susceptible cultivar during infection [72]. This relates to findings that Arabidopsis NPR1 is required for defence against Fusarium [39,41]. Previous studies had reported that constitutive expression of the Arabidopsis AtNPR1 in wheat driven by the maize (Zea mays) Ubi1 promoter enhances floral defence against F. graminearum [73]. More recently, expression of the Arabidopsis SA signalling gene PAD4 in wheat has also been shown to enhance F. graminearum resistance [45]. Conversely, silencing of the wheat lipoxygenase TaLpx-1, a homologue of the Arabidopsis JA signalling genes LOX1 and LOX5, has been shown to enhance F. graminearum resistance, although the effect of this silencing on the different aspects of wheat defence signalling was not assessed [46]. Trends in Plant Science, October 2015, Vol. 20, No

10 Table 2. Comparison of Disease Symptom Development within Distinct Tissues of Different Plant Species Infected with FEB-Causing Fusaria Plant species Disease phenotype Refs Seedling death Leaf lesion Upper stem lesion Anther/flower infection Infected grain Arabidopsis + a /+ c [24,39,41] Wheat + / [1,6] Barley + / [1,67] Maize +? d + + [86,87] Brachypodium b [88] a Indicates disease. b No disease. c Disease that is sporadic/dependent on experimental manipulation. d Unknown phenotype. The apparent involvement of several signalling pathways in defence may reflect the complex infection lifestyle of Fusarium, with a brief symptomless phase in wheat that precedes necrotrophy [33]. However, further functional studies using mutant/transgenic lines, with particular focus on the JA and ET pathways, would be beneficial in confirming their relative contributions to resistance and susceptibility at different stages of infection. Synthesis of phenylpropanoid-derived secondary metabolites such as cinnamic acid and scopoletin appears to be associated with limitation of infection and colonisation in Arabidopsis and wheat, along with maize and barley [74 79] (J. Baker, J.L. Ward, and K.E.H-K., unpublished). Furthermore, in these species, host susceptibility is predominantly confined to floral tissues (Figure 1). These numerous similarities support the usefulness of Arabidopsis as a model for FEB disease. For completeness, the tissues and organs reported to be susceptible to FEBcausing Fusaria and the resulting disease phenotypes observed on wild-type Arabidopsis, wheat, barley, maize, and Brachypodium plants are compared in Table 2. Use of Arabidopsis to Study Other Cereal-Infecting Pathogens Since the development of the Arabidopsis FEB pathosystem, Arabidopsis has been tested for suitability as an experimental host for other fungi which cause cereal diseases. The Basidiomycete fungus Ustilago maydis, which causes maize smut disease, was shown to infect Arabidopsis seedlings, causing necrosis, stunting, and plant death. Strains which were avirulent on maize were not able to colonise Arabidopsis, suggesting similarity in infection mechanisms [80]. Follow-up transcriptomic analysis revealed that infection of both species induces similar transcriptomic changes in U. maydis, and that infection may be biotrophic or necrotrophic, depending on the ploidy of the pathogen strain [81]. Similarly, rice (Oryza sativa)- infecting strains of the Ascomycete fungus Magnaporthe oryzae, which cause leaf and neck blast disease, have been shown to infect Arabidopsis leaf tissue, causing necrotic lesions. However, M. oryzae mutants lacking key virulence genes for rice infection were still pathogenic for Arabidopsis. Therefore, the infection mechanisms differ between the two plant species [82]. For the Ascomycete cereal powdery mildew pathogens, Blumeria graminis f. sp. tritici and B. graminis f. sp. hordei, Arabidopsis is known not to be a host. Therefore, Arabidopsis has been used to study the determinants of non-host resistance. Genetic screens using this pathosystem have identified several genes required for non-host resistance to these pathogens, including the PEN (PENETRATION) gene suite and PLDd (PHOSPHOLIPASE Dd) [83 85]. 660 Trends in Plant Science, October 2015, Vol. 20, No. 10

11 Concluding Remarks FEB remains a devastating global cereal disease for which control mechanisms are severely limited. This review summarises the use of Arabidopsis over the past 13 years as a model host to explore wheat-infecting Fusarium species. Using this pathosystem, substantial insights have been gained into mechanisms of host defence and susceptibility to Fusarium in Arabidopsis and wheat. In addition, the pathosystem has recently been used as a proof of concept for several chemical and transgenic approaches to controlling FEB, most notably the overexpression of small antimicrobial peptides and HIGS/RNAi constructs. Arabidopsis is used to study infection by other cereal pathogens. Translation of these findings into cereals is urgently required to generate novel strategies for disease control. Acknowledgments The authors would like to thank the Rothamsted Visual Communications Unit (VCU) for access to photographs and Amy Dodd for production of plant illustrations. Stereo-microscopy was performed in the Bioimaging facility at Rothamsted Research. We also thank Martin Urban, Stefan Quarry, Simon Bishop, and Joanna Scales for comments on the manuscript. This review was written as part of a PhD Quota studentship (BB/F016824/1) funded by the BBSRC (Biotechnology and Biological Sciences Research Council, UK). Rothamsted Research receives grant-aided support from the BBSRC. K.H.K. receives additional support from within the BBSRC Institute Strategy Grant 20:20 Wheat (BB/J/00426X/1). References 1. Parry, D.W. et al. (1995) Fusarium ear blight (scab) in small-grain cereals a review. Plant Pathol. 44, Doohan, F.M. et al. (2003) Influence of climatic factors on Fusarium species pathogenic to cereals. Eur. J. Plant Pathol. 109, Pestka, J. (2010) Deoxynivalenol: mechanisms of action, human exposure, and toxicological relevance. Arch. Toxicol. 84, Rocha, O. et al. (2005) Effects of trichothecene mycotoxins on eukaryotic cells: a review. Food Addit. Contam. 22, Dean, R. et al. (2012) The top 10 fungal pathogens in molecular plant pathology. Mol. Plant Pathol. 13, Brown, N.A. et al. (2011) Characterisation of the Fusarium graminearum wheat floral interaction. J. Pathog. 2011, Shah, D.A. et al. (2013) Predicting Fusarium head blight epidemics with weather-driven pre- and post-anthesis logistic regression models. Phytopathology 103, Fan, J. et al. (2013) Characterization of the sterol 14a-demethylases of Fusarium graminearum identifies a novel genus-specific CYP51 function. New Phytol. 198, Audenaert, K. et al. (2010) Hydrogen peroxide induced by the fungicide prothioconazole triggers deoxynivalenol (DON) production by Fusarium graminearum. BMC Microbiol. 10, Zhou, M.P. et al. (2010) Saturation and mapping of a major Fusarium head blight resistance QTL on chromosome 3BS of Sumai 3 wheat. J. Appl. Genet. 51, Jayatilake, D.V. et al. (2011) A novel quantitative trait locus for Fusarium head blight resistance in chromosome 7A of wheat. Theor. Appl. Genet. 122, Buerstmayr, H. et al. (2009) QTL mapping and marker-assisted selection for Fusarium head blight resistance in wheat: a review. Plant Breed. 128, Cativelli, M. et al. (2013) A Fusarium head blight resistance quantitative trait locus on chromosome 7D of the spring wheat cultivar Catbird. Crop Sci. 53, Cao, J. et al. (2011) Whole-genome sequencing of multiple Arabidopsis thaliana populations. Nat. Genet. 43, Ossowski, S. et al. (2008) Sequencing of natural strains of Arabidopsis thaliana with short reads. Genome Res. 18, Arabidopsis Genome Initiative (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408, Hruz, T. et al. (2008) Genevestigator V3: a reference expression database for the meta-analysis of transcriptomes. Adv. Bioinformatics 2008, Kanehisa, M. et al. (2014) Data, information, knowledge and principle: back to metabolism in KEGG. Nucleic Acids Res. 42, Lamesch, P. et al. (2011) The Arabidopsis Information Resource (TAIR): improved gene annotation and new tools. Nucleic Acids Res. 40, Alonso, J.M. et al. (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301, Schmid, M. et al. (2005) A gene expression map of Arabidopsis thaliana development. Nat. Genet. 37, Piquerez, S.J.M. et al. (2014) Improving crop disease resistance: lessons from research on Arabidopsis and tomato. Front. Plant Sci. 5, Dangl, J.L. et al. (2013) Pivoting the plant immune system from dissection to deployment. Science 341, Urban, M. et al. (2002) Arabidopsis is susceptible to the cereal ear blight fungal pathogens Fusarium graminearum and Fusarium culmorum. Plant J. 32, Chen, X. et al. (2006) Characterization of Arabidopsis thaliana Fusarium graminearum interactions and identification of variation in resistance among ecotypes. Mol. Plant Pathol. 7, Glazebrook, J. (2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol. 43, Koornneef, A. and Pieterse, C.M.J. (2008) Cross talk in defense signaling. Plant Physiol. 146, Beckers, G.J.M. and Spoel, S.H. (2006) Fine-tuning plant defence signalling: salicylate versus jasmonate. Plant Biol. 8, Hammond-Kosack, K. and Parker, J. (2003) Deciphering plant pathogen communication: fresh perspectives for molecular resistance breeding. Curr. Opin. Biotechnol. 14, Leon-Reyes, A. et al. (2009) Ethylene modulates the role of NON- EXPRESSOR OF PATHOGENESIS-RELATED GENES 1 in cross talk between salicylate and jasmonate signaling. Plant Physiol. 149, Thaler, J.S. et al. (2004) The role of the jasmonate response in plant susceptibility to diverse pathogens with a range of lifestyles. Plant Physiol. 135, Truman, W. et al. (2007) Arabidopsis systemic immunity uses conserved defense signaling pathways and is mediated by jasmonates. Proc. Natl. Acad. Sci. U.S.A. 104, Brown, N.A. et al. (2010) The infection biology of Fusarium graminearum: defining the pathways of spikelet to spikelet colonisation in wheat ears. Fungal Biol. 114, Outstanding Questions In the various floral tissues, do the plant defence responses controlled by the signalling molecules SA, JA, and ET operate antagonistically or sequentially? What are the communication events occurring between Fusarium hyphae and plant cells before the DON mycotoxin levels become elevated? Why are leaves not naturally infected by cereal-infecting Fusarium species? By identifying the underlying mechanism (s), could floral tissues be modified to become more 'leaf-like', thereby reducing FEB incidence and severity? Could a forward-genetic screen using the highly-susceptible Arabidopsis ecotype Ler-0 be established to identify mutants that exhibited enhanced disease resistance in floral tissue? What is the basis for the differences in requirements for SA signalling in Arabidopsis for defence against F. graminearum compared to F. culmorum? There may be important differences between the infection lifestyles of these two closely related pathogens, but this has not so far been investigated. Do the innate and novel defence strategies identified against Fusarium infection extend to other cereal-infecting pathogens and insect pests? Or could enhancing resistance to FEB via these mechanisms lead to unforeseen reductions in resistance to other biotic threats? Trends in Plant Science, October 2015, Vol. 20, No

12 34. Tierens, K.F.M.J. et al. (2002) Esa1, an Arabidopsis mutant with enhanced susceptibility to a range of necrotrophic fungal pathogens, shows a distorted induction of defense responses by reactive oxygen generating compounds. Plant J. 29, Van Hemelrijck, W. et al. (2006) The Arabidopsis defense response mutant esa1 as a model to discover novel resistance traits against Fusarium diseases. Plant Sci. 171, Waters, M.T. et al. (2009) GLK transcription factors coordinate expression of the photosynthetic apparatus in Arabidopsis. Plant Cell 21, Waters, M.T. et al. (2008) GLK transcription factors regulate chloroplast development in a cell-autonomous manner. Plant J. 56, Savitch, L. et al. (2007) The GLK1 'regulon' encodes disease defense related proteins and confers resistance to Fusarium graminearum in Arabidopsis. Biochem. Biophys. Res. Commun. 359, Cuzick, A. et al. (2008) NPR1 and EDS11 contribute to host resistance against Fusarium culmorum in Arabidopsis buds and flowers. Mol. Plant Pathol. 9, Cuzick, A. et al. (2009) Lack of the plant signalling component SGT1b enhances disease resistance to Fusarium culmorum in Arabidopsis buds and flowers. New Phytol. 181, Makandar, R. et al. (2010) Involvement of salicylate and jasmonate signaling pathways in Arabidopsis interaction with Fusarium graminearum. Mol. Plant Microbe Interact. 23, Dong, X. (2004) NPR1, all things considered. Curr. Opin. Plant Biol. 7, Aarts, N. et al. (1998) Different requirements for EDS1 and NDR1 by disease resistance genes define at least two R gene-mediated signaling pathways in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A. 95, Serrano, M. et al. (2013) Export of salicylic acid from the chloroplast requires the multidrug and toxin extrusion-like transporter EDS5. Plant Physiol. 162, Makandar, R. et al. (2015) The combined action of ENHANCED DISEASE SUSCEPTIBILITY1, PHYTOALEXIN DEFICIENT4 and SENESCENCE-ASSOCIATED101 promotes salicylic acid-mediated defenses to limit Fusarium graminearum infection in Arabidopsis thaliana. Mol. Plant Microbe Interact. Published online April 27, Nalam, V.J. et al. (2015) Facilitation of Fusarium graminearum infection by 9-lipoxygenases in Arabidopsis and wheat. Mol. Plant Microbe Interact. Published online June 15, org/ /mpmi r 47. Sanders, P.M. et al. (2000) The Arabidopsis DELAYED DEHIS- CENCE 1 gene encodes an enzyme in the jasmonic acid synthesis pathway. Plant Cell 12, Ding, L. et al. (2011) Resistance to hemi-biotrophic F. graminearum infection is associated with coordinated and ordered expression of diverse defense signaling pathways. PLoS ONE 6, e Chen, X. et al. (2009) Fusarium graminearum exploits ethylene signalling to colonize dicotyledonous and monocotyledonous plants. New Phytol. 182, Gottwald, S. et al. (2012) Jasmonate and ethylene dependent defence gene expression and suppression of fungal virulence factors: two essential mechanisms of Fusarium head blight resistance in wheat? BMC Genomics 13, Azevedo, C. et al. (2006) Role of SGT1 in resistance protein accumulation in plant immunity. EMBO J. 25, Kadota, Y. et al. (2010) NLR sensors meet at the SGT1 HSP90 crossroad. Trends Biochem. Sci. 35, El Oirdi, M. and Bouarab, K. (2007) Plant signalling components EDS1 and SGT1 enhance disease caused by the necrotrophic pathogen Botrytis cinerea. New Phytol. 175, Glazebrook, J. et al. (1996) Isolation of Arabidopsis mutants with enhanced disease susceptibility by direct screening. Genetics 143, Ferrari, S. et al. (2012) Transgenic expression of polygalacturonase-inhibiting proteins in Arabidopsis and wheat increases resistance to the flower pathogen Fusarium graminearum. Plant Biol. 14 (Suppl. 1), Shin, S. et al. (2012) Transgenic Arabidopsis thaliana expressing a barley UDP-glucosyltransferase exhibit resistance to the mycotoxin deoxynivalenol. J. Exp. Bot. 63, Lemmens, M. et al. (2005) The ability to detoxify the mycotoxin deoxynivalenol colocalizes with a major quantitative trait locus for Fusarium head blight resistance in wheat. Mol. Plant Microbe Interact. 18, Poppenberger, B. et al. (2006) Heterologous expression of Arabidopsis UDP-glucosyltransferases in Saccharomyces cerevisiae for production of zearalenone-4-o-glucoside. Appl. Environ. Microbiol. 72, Poppenberger, B. et al. (2003) Detoxification of the Fusarium mycotoxin deoxynivalenol by a UDP-glucosyltransferase from Arabidopsis thaliana. J. Biol. Chem. 278, Schweiger, W. et al. (2010) Validation of a candidate deoxynivalenolinactivating UDP-glucosyltransferase from barley by heterologous expression in yeast. Mol. Plant Microbe Interact. 23, Schweiger, W. et al. (2013) Functional characterization of two clusters of Brachypodium distachyon UDP-glycosyltransferases encoding putative deoxynivalenol detoxification genes. Mol. Plant Microbe Interact. 26, Schreiber, K.J. et al. (2011) Found in translation: high-throughput chemical screening in Arabidopsis thaliana identifies small molecules that reduce Fusarium head blight disease in wheat. Mol. Plant Microbe Interact. 24, Brewer, H.C. et al. (2014) Mutations in the Arabidopsis homoserine kinase gene DMR1 confer enhanced resistance to Fusarium culmorum and F. graminearum. BMC Plant Biol. 14, Huibers, R.P. et al. (2013) Powdery mildew resistance in tomato by impairment of SlPMR4 and SlDMR1. PLoS ONE 8, e van Damme, M. et al. (2009) Downy mildew resistance in Arabidopsis by mutation of HOMOSERINE KINASE. Plant Cell 21, Kaur, J. et al. (2012) Subcellular targeting of an evolutionarily conserved plant defensin MtDef4.2 determines the outcome of plant pathogen interaction in transgenic Arabidopsis. Mol. Plant Pathol. 13, Koch, A. et al. (2012) The antimicrobial peptide thanatin reduces fungal infections in Arabidopsis. J. Phytopathol. 160, Kang, Z. and Buchenauer, H. (2003) Immunocytochemical localization of cell wall-bound thionins and hydroxyproline-rich glycoproteins in Fusarium culmorum-infected wheat spikes. J. Phytopathol. 151, Asano, T. et al. (2013) The secreted antifungal protein Thionin 2.4 in Arabidopsis thaliana suppresses the toxicity of a fungal fruit body lectin from Fusarium graminearum. PLoS Pathog. 9, e Koch, A. et al. (2013) Host-induced gene silencing of cytochrome P450 lanosterol C14a-demethylase-encoding genes confers strong resistance to Fusarium species. Proc. Natl. Acad. Sci. U. S.A. 110, Bailey, A.M. (2014) A silenced spring? Mol. Plant Pathol. 15, Yang, Z. et al. (2013) Cloning of NPR1-like genes and their response to Fusarium graminearum infection in wheat. Acta Agron. Sin. 39, Makandar, R. et al. (2006) Genetically engineered resistance to Fusarium head blight in wheat by expression of Arabidopsis NPR1. Mol. Plant Microbe Interact. 19, Bollina, V. et al. (2010) Mass spectrometry-based metabolomics application to identify quantitative resistance-related metabolites in barley against Fusarium head blight. Mol. Plant Pathol. 11, Bollina, V. et al. (2011) Identification of metabolites related to mechanisms of resistance in barley against Fusarium graminearum, based on mass spectrometry. Plant Mol. Biol. 77, Kumaraswamy, K.G. et al. (2011) Mass spectrometry based metabolomics to identify potential biomarkers for resistance in barley against Fusarium head blight (Fusarium graminearum). J. Chem. Ecol. 37, Gunnaiah, R. et al. (2012) Integrated metabolo-proteomic approach to decipher the mechanisms by which wheat QTL FHB1 contributes to resistance against Fusarium graminearum. PLoS ONE 7, e Trends in Plant Science, October 2015, Vol. 20, No. 10