Fusarium culmorum: causal agent of foot and root rot and head blight on wheat

Summary

Fusarium culmorum is a ubiquitous soil‐borne fungus able to cause foot and root rot and Fusarium head blight on different small‐grain cereals, in particular wheat and barley. It causes significant yield and quality losses and results in contamination of the grain with mycotoxins. This review summarizes recent research activities related to F. culmorum, including studies into its population diversity, mycotoxin biosynthesis, mechanisms of pathogenesis and resistance, the development of diagnostic tools and preliminary genome sequence surveys. We also propose potential research areas that may expand our basic understanding of the wheat–F. culmorum interaction and assist in the management of the disease caused by this pathogen.

Taxonomy

Fusarium culmorum (W.G. Smith) Sacc. Kingdom Fungi; Phylum Ascomycota; Subphylum Pezizomycotina; Class Sordariomycetes; Subclass Hypocreomycetidae; Order Hypocreales; Family Nectriaceae; Genus Fusarium.

Disease symptoms

Foot and root rot (also known as Fusarium crown rot): seedling blight with death of the plant before or after emergence; brown discoloration on roots and coleoptiles of the infected seedlings; brown discoloration on subcrown internodes and on the first two/three internodes of the main stem; tiller abortion; formation of whiteheads with shrivelled white grains; Fusarium head blight: prematurely bleached spikelets or blighting of the entire head, which remains empty or contains shrunken dark kernels.

Identification and detection

Morphological identification is based on the shape of the macroconidia formed on sporodochia on carnation leaf agar. The conidiophores are branched monophialides, short and wide. The macroconidia are relatively short and stout with an apical cell blunt or slightly papillate; the basal cell is foot‐shaped or just notched. Macroconidia are thick‐walled and curved, usually 3–5 septate, and mostly measuring 30–50 × 5.0–7.5 μm. Microconidia are absent. Oval to globose chlamydospores are formed, intercalary in the hyphae, solitary, in chains or in clumps; they are also formed from macroconidia. The colony grows very rapidly (1.6–2.2 cm/day) on potato dextrose agar (PDA) at the optimum temperature of 25 °C. The mycelium on PDA is floccose, whitish, light yellow or red. The pigment on the reverse plate on PDA varies from greyish‐rose, carmine red or burgundy. A wide array of polymerase chain reaction (PCR) and real‐time PCR tools, as well as complementary methods, which are summarised in the first two tables, have been developed for the detection and/or quantification of F. culmorum in culture and in naturally infected plant tissue.

Host range

Fusarium culmorum has a wide range of host plants, mainly cereals, such as wheat, barley, oats, rye, corn, sorghum and various grasses. In addition, it has been isolated from sugar beet, flax, carnation, bean, pea, asparagus, red clover, hop, leeks, Norway spruce, strawberry and potato tuber. Fusarium culmorum has also been associated with dermatitis on marram grass planters in the Netherlands, although its role as a causal agent of skin lesions appears questionable. It is also isolated as a symbiont able to confer resistance to abiotic stress, and has been proposed as a potential biocontrol agent to control the aquatic weed Hydrilla spp.

Useful websites

http://isolate.fusariumdb.org/; http://sppadbase.ipp.cnr.it/; http://www.broad.mit.edu/annotation/genome/fusarium_group/MultiHome.html; http://www.fgsc.net/Fusarium/fushome.htm; http://plantpath.psu.edu/facilities/fusarium‐research‐center; http://www.phi‐base.org/; http://www.uniprot.org/; http://www.cabi.org/; http://www.indexfungorum.org/

Introduction

Fusarium culmorum (W.G. Smith) Sacc. is a ubiquitous soil‐borne fungus with a highly competitive saprophytic capability. As a facultative parasite, it is able to cause foot and root rot (FRR) and Fusarium head blight (FHB) on different small‐grain cereals, in particular wheat and barley. Fusarium culmorum is also known as a post‐harvest pathogen, especially on freshly harvested grain that has not been dried or stored properly (Aldred and Magan, 2004; Eifler et al., 2011; Lowe et al., 2012; Magan et al., 2003, 2010). Together with F. graminearum Schwabe (teleomorph Gibberella zeae) and F. pseudograminearum O'Donnell and Aoki (teleomorph Gibberella coronicola), F. culmorum has been reported as one of the main pathogens of wheat worldwide (Burgess et al., 2001; Goswami and Kistler, 2004; Hogg et al., 2010; Kosiak et al., 2003; Miedaner et al., 2008; Treikale et al., 2010; Wagacha and Muthomi, 2007; Wang et al., 2006).

Yield and quality losses are particularly important when F. culmorum induces FHB, which develops from infection at anthesis and spreads until grain harvest, causing grain contamination with mycotoxins, such as type B trichothecenes, zearalenone and fusarins (Hope et al., 2005; Jennings et al., 2004; Kammoun et al., 2010; Lacey et al., 1999; Placinta et al., 1999; Rohweder et al., 2011; Visconti and Pascale, 2010). The sesquiterpene epoxide trichothecenes are considered to be the most bioactive compounds produced by F. culmorum. These mycotoxins are able to inhibit eukaryotic protein synthesis (Wei and McLaughlin, 1974) and cause toxicoses in humans or animals consuming contaminated food or feed (Sudakin, 2003). They have also been reported to induce apoptosis (Desmond et al., 2008; Yang et al., 2000) and play an important role as virulence factors (Bai et al., 2002; Desjardins et al., 1996, 2000; Harris et al., 1999; Jansen et al., 2005; Maier et al., 2006; McCormick, 2003; Proctor et al., 1995, 2002; Scherm et al., 2011; Ward et al., 2008; Zhang et al., 2010).

The purpose of this profile is to provide an overview of the recent research activities related to F. culmorum, including those on population diversity, mycotoxin biosynthesis, mechanisms of pathogenesis and resistance, the development of diagnostic tools and preliminary genome sequence surveys (see Tables 1 and 2, respectively, for a list of PCR‐based and non PCR‐based approaches to discriminate and detect F. culmorum). We also propose potential research areas that may expand our basic understanding of the wheat–F. culmorum interaction and ultimately assist in the management of the different facies of the disease caused by this pathogen.

Table 1.

Summary of published primers used for species and chemotype determination in Fusarium culmorum

Identification of	Primers and probes (5′ → 3′)	Target DNA	PCR technique	Reference
Species (F. culmorum and F. graminearum)	FcF CAAAAGCTTCCCGAGTGTGTC FcR GGCGAAGGTTCAAGGATGAC	Unknown	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Doohan et al. (1998)
Species (not able to distinguish from F. cerealis)	FculC561fwd CACCGTCATTGGTATGTTGTCACT FculC614rev CGGGAGCGTCTGATAGTCG	ef1‐α	Real‐time PCR	Nicolaisen et al. (2009)
Species (together with F cerealis and F. graminearum)	FIP‐hyd5 GCACAGCACTGGGAAGTGCGAGAAGCGACAGGCCTACA BIP‐hyd5 TGGGTGTTGCTGACCTCGACGGGGCTGTTCATGTTAGCT B3‐hyd4 GACAGCGCTGAAGTTGTC LoopB‐hyd5 CCGTAAGTACTCGAGTCTG LoopF‐hyd5 GTAGAGGCCACTGCAAGG F3‐hyd5 CTTGGAGCCGTTGTCTCTG	Hyd 5	LAMP PCR	Denschlag et al. (2012)
Species (together with F. crockwellense)	CRO‐C fwd CTCAGTGTCCACCGCGTTGCGTAGTGT CRO‐C rev AAGCAGGAAACAGAAACCCTTTCC	RAPD fragment	Conventional PCR	Yoder and Christianson (1998)
Species (together with F. graminearum)	CUL‐A fwd TTTCAGCGGGCAACTTTGGGTAGA CUL‐A rev AAGCTGAAATACGCGGTTGATAGG	RAPD fragment	Conventional PCR	Yoder and Christianson (1998)
Species	C51END fwd AACTGAATTGATCGCAAGC C51END rev CCCTTCTTACGCCAATCTC	Unknown	Real‐time PCR	Covarelli et al. (2012)
Species	OPT18F470 GATGCCAGACCAAGACGAAG OPT18R470 GATGCCAGACGCACTAAGAT	SCAR	Conventional PCR Real‐time PCR*	Baturo‐Ciesniewska and Suchorzynska (2011); Brandfass and Karlovsky 2006; Schilling et al. (1996)
Species	Fc92s1 forward TTCACTAGATCGTCCGGCAG Fc92s1 reverse GAGCCCTCCAAGCGAGAAG	Unknown	Real‐time PCR	Leisova et al. (2006)
Species	Fc01F ATGGTGAACTCGTCGTGGC Fc01R CCCTTCTTACGCCAATCTCG	RAPD fragment	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Nicholson et al. (1998)
Species	Fcg17F TCGATATACCGTGCGATTTCC Fcg17R TACAGACACCGTCAGGGGG	RAPD fragment	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Nicholson et al. (1998)
Species	175F TTTTAGTGGAACTTCTGAGTAT 430R AGTGCAGCAGGACTGCAGC	ITS region	Fluorescent‐labelled PCR‐based assay	Mishra et al. (2003)
Species	culmorum MGB‐R GAACGCTGCCCTCAAGCTT culmorum MGB‐F TCACCCAAGACGGGAATGA Probe CACTTGGATATATTTCC	Genomic DNA	Real‐time PCR (TaqMan)	Waalwijk et al. (2004)
Type B trichothecene producers	Fcu‐F GACTATCATTATGCTTGCGAGAG Fgc‐R CTCTCATATACCCTCCG	IGS region	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Jurado et al. (2005)
15‐ADON subchemotype	Tri3F971 CATCATACTCGCTCTGCTG Tri3R1679 TT(AG)TAGTTTGCATCATT(AG)TAG	TRI3	Conventional PCR	Pasquali et al. (2011); Quarta et al. (2006)
3‐ADON subchemotype	Tri3F1325 GCATTGGCTAACACATGA Tri3R1679 TT(AG)TAGTTTGCATCATT(AG)TAG	TRI3	Conventional PCR	Pasquali et al. (2011); Quarta et al. (2006)
Nivalenol subchemotype	Tri7F340 ATCGTGTACAAGGTTTACG Tri7R965 TTCAAGTAACGTTCGACAAT	TRI7	Conventional PCR	Pasquali et al. (2011); Quarta et al. (2005)
High‐deoxynivalenol‐producing strains	N1‐2 CTTGTTAAGCTAAGCGTTTT N1‐2R AACCCCTTTCCTATGTGTTA	TRI6/TRI5 intergenic region	Conventional PCR	Bakan et al. (2002)
Low‐deoxynivalenol‐producing strains	4056 ATCCCTCAAAAACTGCCGCT 3551 ACTTTCCCACCGAGTATTTC	TRI6/TRI5 intergenic region	Conventional PCR	Bakan et al. (2002)
Species (together with F. graminearum and F. pseudograminearum)	Gzeae87T forward CGCATCGAGAATTTGCA Gzeae87T reverse TGGCGAGGCTGAGCAAAG Gzeae87T probe 6FAM‐TGCTTACAACAAGGCTGCCCACCA‐TAMRA	TRI5	Real‐time PCR (TaqMan)	Strausbaugh et al. (2005)
Deoxynivalenol‐producing isolates (F. graminearum and F. culmorum)	22F AATATGGAAAACGGAGTTCATCTACA 122R ATTGCCGGTGCCTGAAAGT	TRI6‐TRI5 intergenic region	Real‐time PCR (SYBR Green I)	Terzi et al. (2007)
PKS13‐containing strains (F. culmorum and F. graminearum)	ZEA‐F CTGAGAAATATCGCTACACTACCGAC ZEA‐R CCCACTCAGGTTGATTTTCGTC	PKS13	Conventional PCR/Real‐time PCR (SYBR Green I)	Atoui et al. (2012)
Deoxynivalenol‐producing strain	Tri7F TGCGTGGCAATATCTTCTTCTA Tri7DON GTGCTAATATTGTGCTAATATTGTGC	TRI7	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Chandler et al. (2003)
Deoxynivalenol‐producing strain	Tri13F CATCATGAGACTTGTKCRAGTTTGGGC Tri13DONR GCTAGATCGATTGTTGCATTGAG	TRI13	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Chandler et al. (2003)
Nivalenol‐producing strain	Tri7F TGCGTGGCAATATCTTCTTCTA Tri7NIV TGTGGAAGCCGCAGA	TRI7	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Chandler et al. (2003)
Nivalenol‐producing strain	Tri13NIVF CCAAATCCGAAAACCGCA Tri13R TTGAAAGCTCCAATGTCGTG	TRI13	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Chandler et al. (2003)
3‐ADON‐producing strain	Tri303F GATGGCCGCAAGTGGA Tri303R GCCGGACTGCCCTATTG	TRI3	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Jennings et al. (2004)
Trichothecene producer	Tox5‐1 GCTGCTCATCACTTTGCTCAG Tox5‐2 CTGATCTGGTCACGCTCATC	TRI5	Conventional PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Niessen and Vogel (1998)
Nivalenol‐producing strain	12NF TCTCCTCGTTGTATCTGG 12CON CATGAGCATGGTGATGTC	TRI12	Conventional PCR	Pasquali et al. (2011); Ward et al. (2002)
Deoxynivalenol‐producing strain	12‐3F CTTTGGCAAGCCCGTGCA 12CON CATGAGCATGGTGATGTC	TRI12	Conventional PCR	Pasquali et al. (2011); Ward et al. (2002)
Nivalenol‐producing strain	Tri13P1 CTCSACCGCATCGAAGASTCTC Tri13P2 GAASGTCGCARGACCTTGTTTC	TRI13	Conventional PCR	Pasquali et al. (2011); Wang et al. (2008)
3‐ADON‐producing strains	3ADONf AACATGATCGGTGAGGTATCGA 3ADONr CCATGGCGCTGGGAGTT	TRI12	Real‐time PCR	Nielsen et al. (2012)
Nivalenol‐producing strain	NIVf GCCCATATTCGCGACAATGT NIVr GGCGAACTGATGAGTAACAAAACC	TRI12	Real‐time PCR	Nielsen et al. (2012)
3‐ADON‐producing strains (F. culmorum and F. graminearum)	3ADON fwd CATGCGGGACTTTGATCGAT 3ADON rev TTTGTCCGCTTTCTTTCTATCATAAA 3ADON probe FAM‐CTCACCGATCATGTTC‐MGB	TRI12	Taqman real‐time PCR	Kulik (2011)
Nivalenol‐producing strain (F. culmorum and F. graminearum)	NIV fwd TCGCCAGTCTCTGCATGAAG NIV rev CCTTATCCGCTTTCTTTCTATCATAAA NIV probe FAM‐CTGATCATGTCCCGCATC‐MGB	TRI12	Taqman real‐time PCR	Kulik (2011)
Zearalenone producer	PKS4‐PS.1 GTGGGCTTCGCTAGACCGTGAGTT PKS4‐PS.2 ATGCCCTGATGAAGAGTTTGA	PKS4	Real‐time PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Lysøe et al. (2006)
Zearalenone producer	F1 CGTCTTCGAGAAGATGACAT R1 TGTTCTGCAAGCACTCCGA	PKS4	PCR	Baturo‐Ciesniewska and Suchorzynska (2011); Meng et al. (2010)

ADON, acetylated deoxynivalenol; PCR, polymerase chain reaction; RAPD, random amplification of polymorphic DNA; SCAR, sequence characterized amplified region.

Table 2.

Non‐polymerase chain reaction (PCR)‐based approaches to discriminate and detect Fusarium culmorum

Technology employed	Reference
Surface plasmon resonance (SPR) sensor based on DNA hybridization	Zezza et al. (2006)
Luminex assay to discriminate Fusarium species and chemotypes	Ward et al. (2008)
DNA microarray for detection and identification of 14 Fusarium species	Kristensen et al. (2007)
Spore shape discrimination analysed by computerized algorithms	Dubos et al. (2012)
Metabolomic analysis and monitoring of the metabolic activity	Lowe et al. (2010)
Electronic nose for discriminating species infecting grains	Eifler et al. (2011)
Quick matrix‐assisted laser desorption/ionization (MALDI) linear time‐of‐flight mass spectrometry analysis of fungal spores	Kemptner et al. (2009)

Disease Symptoms

Fusarium culmorum causes two distinct diseases on wheat: FRR and FHB, also known as ear blight or scab. FRR symptoms vary depending on the time of infection: if the fungus attacks at the early stage, just after sowing, pre‐ and post‐emergence seedling death occurs, with brown discoloration on the coleoptiles, roots and the pseudostem; if the infection starts later in the season, brown lesions appear on the first two or three internodes of the main stem and tiller abortion occurs (Fig. 1B). In the presence of high humidity, a reddish‐pink discoloration is often evident on the nodes caused by the presence of sporulating mycelium (Fig. 1C). The presence of whiteheads with shrivelled grain—or no grain at all—is easily observed when the wheat is still immature (Fig. 1D,E). Infected plants are more prone to lodging. FHB symptoms include partial head blighting, with the appearance of one or more prematurely bleached spikelets, or blighting of the entire head, which is easily observed when wheat has not yet reached the ripening stage (Fig. 2A,B). Initially, infected spikelets show light‐brown, water‐soaked spots on the glumes, which then become dark brown. Infected spikelets remain empty or contain shrunken grey/brown kernels. Browning on the rachilla and the rachis can be observed and, under favourable conditions, the fungus may infect the stem below the head, inducing a brown/purplish discoloration (Fig. 2C). Pink to orange sporodochia may be evident at the base of the spikelets or between the glumes and lemmas, if the environmental conditions are particularly humid (Fig. 2D,F).

Figure 1.

Foot and root rot (FRR) symptoms: (A) macroconidia; (B) browning on the stem base; (C) reddish‐pink discoloration on the basal nodes; (D,E) presence of whiteheads.

Figure 2.

Fusarium head blight (FHB) symptoms: (A,B) head blight symptoms; (C) brown/purplish discoloration below head; (D–F) orange sporodochia on spikelets.

Epidemiology

Fusarium culmorum has been traditionally reported as the incitant of FHB in northern, central and western Europe (Muthomi et al., 2000;de Nijs et al., 1997; Parry et al., 1995). However, recently, in northern Europe, a change is being observed in the frequency of isolation, and F. culmorum is seldom reported compared with F. graminearum. This progressive switch may be explained by the widespread use of feed maize as a rotation crop with wheat in northern Europe, with consequent F. graminearum inoculum build‐up in the soil. It is noteworthy that F. culmorum is occasionally isolated from maize crops and maize kernels, but never as the main pathogen (Logrieco et al., 2002; Scauflaire et al., 2011; Van Asselt et al., 2012). Other reasons for the transition from F. culmorum to F. graminearum may be related to the gradual adaptation of F. graminearum to colder climates as a result of genome plasticity (Lysøe et al., 2011; Raffaele and Kamoun, 2012) or to the rise in average temperatures caused by climate change (Jennings et al., 2004; Waalwijk et al., 2003; West et al., 2012; Xu et al., 2005). However, in Luxembourg, following the year 2011 with hardly any precipitation in May, 90% of the blighted spikes were infected by F. culmorum, whereas only 10% were infected by F. graminearum, suggesting a role of climatic conditions in driving the prevalence of each species, reversing drastically the previous species distribution (Giraud et al., 2010).

Contrary to early reports from colder areas in central and northern Europe, F. culmorum is now frequently reported as the main agent of FHB in the Mediterranean region, and particularly in years characterized by wet conditions during the phenological phases of flowering and kernel filling (Corazza et al., 2002; Fakhfakh et al., 2011; Kammoun et al., 2010; Pancaldi et al., 2010). The greater incidence of FHB caused by F. culmorum in these areas is correlated with its presence as the main cause of FRR, a disease that is particularly severe on durum wheat in southern Italy and North Africa.

Key factors in the development of FRR are the previous crop, residue management, nitrogen fertilization, plant density and the environmental conditions. Conidial germination and germ tube extension on sterile and unsterile wheat straw leaf sheaths were significantly higher relative to other crop residue colonizers, such as Gliocladium, Trichoderma and Penicillium spp., when tested at different water potential × temperature (Magan, 1988). Therefore, wheat monoculture and/or rotation with another cereal crop (such as barley, triticale, rye, spelt, oat or corn) boosts the inoculum and, consequently, the chances of increasing FRR severity: although cereals are not equally sensitive to F. culmorum, all may contribute to maintain inoculum survival in the soil. High nitrogen fertilization rates and high sowing density are believed to increase the incidence of FRR: increased leaf index and transpiration rates and the reduction of plant water potential induce water stress and, consequently, a higher sensitivity to the pathogen (Davis et al., 2009; Papendick and Cook, 1974).

FRR by F. culmorum is severe when wheat is grown in warm areas, where the host plant is more subject to water stress (Bateman, 1993; Cariddi and Catalano, 1990; Chekali et al., 2010; Colhoun et al., 1968; Inglis and Cook, 1986; Papendick and Cook, 1974; Parry, 1990; Prew et al., 1995). Drought conditions increase the susceptibility of the plant rather than the virulence of the fungus. However, FHB occurs preferentially when the pathogen is present at the soil level, and the weather is moist and warm, with frequent rains between flowering and kernel filling stages (Bateman, 2005). Rain is an essential determinant of FHB infection, as demonstrated experimentally on wheat crops receiving overhead irrigation (Strausbaugh and Maloy, 1986). The macroconidia that are found in soil on crop residues reach the ear by rain splash, wind or insects, attaining distances of up to 60 cm vertically and 1 m horizontally (Jenkinson and Parry, 1994; Parry et al., 1995; Rossi et al., 2002). Compared with F. graminearum, F. culmorum does not produce ascospores, being unable to differentiate sexual perithecia. From an epidemiological standpoint, this is paramount, given the crucial role of wind‐borne ascospores in the spread of FHB caused by the former species (Markell and Francl, 2003).

Once the inoculum reaches the ear, humidity and temperature in the crop microclimate play a critical role: it takes at least 24 h of moisture with temperatures above 15 °C, with an optimum of 25 °C, to allow infection (Doohan et al., 2003; Parry et al., 1995). Nonetheless, among the species causing FHB, F. culmorum has the smallest need for the presence of high relative humidity to infect wheat (Klix et al., 2008; Rossi et al., 2001).

Population Diversity and Mycotoxin Production

The perfect stage (teleomorph) of F. culmorum is not known, even though transcribed mating type genes have been identified in this species. Only one MAT idiomorph (MAT1‐1 or MAT1‐2) has been reported so far, postulating heterothallism (Kerényi et al., 2004; Mishra et al., 2003; Obanor et al., 2010; Tòth et al., 2004). It is noteworthy that, among a vast majority of isolates from Turkey carrying either the MAT‐1 or MAT‐2 sequence, Çepni et al. (2012) were recently able to identify two F. culmorum isolates that carried both sequences.

The genetic variability of F. culmorum in different geographical areas suggests that genetic exchange occurs or has occurred in the past, as the population structure is not clonal (Miedaner et al., 2001; Mishra et al., 2003; Tòth et al., 2004).

Population studies carried out within restricted geographical areas, or even at the single field level, have reported a wide genetic variability, whereas relatively modest differences have been detected among populations obtained from different climatic regions (Gargouri et al., 2003; Nicholson et al., 1993). A high level of diversity has also been found recently in F. culmorum isolates from Turkey by intergenic spacer‐restriction fragment length polymorphism (IGS‐RFLP) analysis, further confirming the wide genetic variability associated with FRR disease (Çepni et al., 2012). A phylogenetic study conducted with over 100 isolates of F. culmorum from Australia, West Asia, North Africa and Europe identified three to four distinct groups or lineages. However, no correlation was found between lineages and their geographical origin, with the exception of one cluster including isolates from a single area (Obanor et al., 2010).

Two chemotypes have been described in F. culmorum: chemotype I, which produces deoxynivalenol (DON) and/or its acetylated derivatives (3‐ADON, 15‐ADON), and chemotype II, which produces nivalenol (NIV) and/or fusarenone‐X (FUS), NIV being 10 times more toxic than DON (Minervini et al., 2004). DNA sequence variation in the coding region of the trichothecene biosynthetic gene TRI8 was found in Fusarium spp., including F. culmorum, indicating that differential activity of the Tri8 protein (i.e. deacetylation of the trichothecene biosynthetic intermediate 3,15‐diacetyldeoxynivalenol at carbon 15 versus carbon 3 to yield 3‐ADON or 15‐ADON, respectively) determines the 3‐ADON and 15‐ADON subchemotypes in Fusarium (Alexander et al., 2011).

Studies on F. culmorum chemotypes are less frequent than those focusing on F. graminearum, but it is possible to trace their distribution in some geographical areas (Table 3).

Table 3.

Distribution of Fusarium culmorum chemotypes: country, chemotyping method used, number of isolates analysed, main finding and bibliographic reference

Country	Chemotyping method used	Number of isolates analysed	Main finding	Reference
Europe	Chemical	42	∼84% DON producers, ∼16% NIV producers	Gang et al. (1998)
Germany	Chemical	27	∼60% NIV producers, ∼40% DON producers	Muthomi et al. (2000)
Norway	Chemical	23	Mostly 3‐ADON producers, two NIV producers	Langseth et al. (2001)
France	Genetic and chemical	60	58% NIV producers, 42% DON producers	Bakan et al. (2001, 2001,2002)
Denmark, Germany, Austria	Chemical	102	1995 sampling: ∼90% DON producers, ∼10% NIV producers	Hestbjerg et al. (2002)
The Netherlands	Genetic	85	2000–2001 sampling: mostly NIV producers	Waalwijk et al. (2003)
Worldwide (Australia, Canada, Israel, Hungary, Germany, Denmark, the Netherlands, Morocco)	Genetic and chemical	37	19% NIV producers, 81% 3‐ADON producers	Tòth et al. (2004)
UK	Genetic	157	DON producers are prevalent, but NIV producers are distributed consistently	Jennings et al. (2004)
Europe (Spain, Italy, Poland, Norway, the Netherlands, France, Finland, former Yugoslavia)	Genetic	55	∼20% NIV producers, ∼80% 3‐ADON producers	Quarta et al. (2005)
Belgium	Genetic	128	In 2007 (95%) and in 2008 (88%) NIV producers are the most diffused	Audenaert et al. (2009)
Luxembourg	Genetic and chemical	175	3‐ADON and NIV producers are evenly distributed Chemotyping is useful to predict toxin content Chemical analysis confirms genetic chemotyping	Pasquali et al. (2010)
Tunisia	Genetic and chemical	100	Mostly 3‐ADON producers, 2% NIV producers Chemical analysis confirms genetic chemotyping	Kammoun et al. (2010)
Poland	Genetic	68	6% NIV producers, 94% 3‐ADON producers	Baturo‐Ciesniewska and Suchorzynska (2011)
Turkey	Genetic	21	100% 3‐ADON producers	Yörük and Albayrak (2012)

ADON, acetylated deoxynivalenol; DON, deoxynivalenol; NIV, nivalenol.

The link between the presence of the pathogen and its toxins (in this case, type B trichothecenes) is often complicated by the complexity of toxin induction and pathogen adaptation. Although F. culmorum has been reported to be one of the main fungal species associated with diseased wheat in warmer regions, such as Turkey (Tunalı et al., 2006), Tunisia (Kammoun et al., 2010), Australia and New Zealand (Lauren et al., 1992), no clear data on its role in toxin accumulation are evident. Moreover, although this species was the most prevalent in 2009 in the central region of Poland, the level of toxin contamination reported in the grains was very low, and no direct correlation between fungal contamination and toxin accumulation could be found (Chelkowski et al., 2012). The identification of the chemotype may provide insight into the toxigenic potential of F. culmorum isolates. For example, the presence of F. culmorum with the NIV subchemotype has been linked to the accumulation of NIV in wheat harvested in Luxembourg during 2007 and 2008 (Pasquali et al., 2010), confirming the findings obtained in a within‐field comparison experiment described by Xu et al. (2008). Similar results pinpointing a role of F. culmorum in the accumulation of NIV have been reported in a recent screening of historical Danish seed samples by real‐time PCR (Nielsen et al., 2012).

Host–Pathogen Interaction

Although a wide array of information on F. culmorum pathogenesis can be inferred from reports using F. graminearum as the species of interest, in the present review, we have attempted to limit references to related Fusarium species only when absolutely necessary. Fusarium culmorum remains viable as mycelium in crop residues left on the ground surface, and can survive in soil for 2–4 years by forming chlamydospores (Bateman et al., 1998; Cook, 1980; Inglis and Cook, 1986). When the seed germinates, the fungus penetrates through the lesions that are formed during primary root emergence, and then progresses towards the culm. Alternatively, it penetrates through the stomata at the insertion point of the basal leaf sheath towards the stem. The colonization follows, initially, an intercellular apoplastic pathway between cells of the epidermis and cortex; subsequently, the fungus progresses intracellularly in the symplast to complete colonization of the tissues (Beccari et al., 2011; Covarelli et al., 2012; Pettitt and Parry, 2001). The fungus may then grow further along the stem, although it is usually limited to the first basal internodes. The symptoms of basal browning may occur prior to the presence of the fungus in these portions, as a result of the plant response to infection (Beccari et al., 2011; Covarelli et al., 2012).

FHB infection occurs between flowering and the soft dough stage (GS 65–85; Zadoks' scale modified by Tottman and Makepeace, 1979), the phases between flowering and the milk stage (GS 65–77) being the most favourable for the infection by F. culmorum (Lacey et al., 1999). Once the macroconidia arrive onto the ear, they germinate rapidly and the fungus penetrates into host tissues, either directly through the stomata, or through the floret mouth or crevices formed between the palea and lemma, and then progresses inter‐ and intracellularly and reaches the endosperm within 12–24 h. Betaine and choline, which are contained in the anthers, stimulate the growth of conidial germ tubes towards the head surface (Strange et al., 1974, 1978). Similar to other FHB pathogens, F. culmorum may have an initial brief biotrophic phase within plant tissues, but then shifts to a necrotrophic stage through the production of trichothecenes and cell wall‐degrading enzymes (CWDEs; Bushnell et al., 2003).

The infection process by F. culmorum is strongly influenced by temperature, humidity, carbon and nitrogen availability, as well as the ability of the specific strain to produce mycotoxins that may confer a higher aggressiveness by inhibiting the defence response by the plant. Key factors for its growth are temperature and water availability (water activity a _w; Magan et al., 2006). Schmidt‐Heydt et al. (2011) compared the effect of a _w × temperature of one isolate of F. culmorum and F. graminearum on growth, F. culmorum showing an optimum at 30 °C and 0.98a _w, whereas its minimum limit for growth was 15 °C over 0.88–0.995a _w. Germination of F. culmorum macroconidia is restricted to a minimum of 0.86a _w, but is functional over a wide temperature range from 5 to 35 °C (Magan et al., 2006). Fusarium culmorum hydrolytic enzymes are produced over the same broad temperature range, allowing the rapid utilization of nutritional resources (Magan and Lynch, 1986).

Mycotoxin biosynthesis is mainly influenced by temperature and moisture (Homdork et al., 2000; Tanaka et al., 1988). Studies with F. culmorum and F. graminearum isolates from Spain (Llorens et al., 2004) showed that both fungi require high humidity (>0.90a _w) to support trichothecene production, with optimum temperatures of 25–28 °C for DON, 20 °C for NIV and a minimum of 15 °C for 3‐ADON. Fusarium culmorum demonstrated a significantly higher mycotoxigenic rate (up to five times higher for type B trichothecenes) than F. graminearum, and the toxin biosynthesis could not be correlated with mycelial growth (Llorens et al., 2004; Lori et al., 1999).

Trichothecene production, which is driven by the expression of the TRI5 gene encoding the key biosynthesis enzyme trichodiene synthase, can be observed as early as 36 h post‐inoculation during the colonization of wheat spikelets (Beccari et al., 2011; Kang and Buchenauer, 2002). The ability of aggressive strains of F. culmorum to infect wheat is related to their ability to produce larger amounts of DON in culture or in infected tissues (Hestbjerg et al., 2002; Manka et al., 1985; Scherm et al., 2011), although correlation is not always linear (Gang et al., 1998). Similar to F. graminearum, trichothecene mycotoxins produced by F. culmorum are essential for the spread of the disease by inhibiting defence mechanisms activated by the plant (Wagacha and Muthomi, 2007). Following inoculation of the stem base of soft wheat seedlings with F. culmorum, Covarelli et al. (2012) demonstrated the translocation of DON to the head, even though the fungus was unable to grow systemically beyond the third node. This finding suggests that FRR may represent an additional potential source of grain contamination, providing an explanation for previous reports on the presence of DON in grain harvested in the field, even in the absence of detectable fungus (Xu et al., 2008).

Different plant compounds involved in host–pathogen interactions are able to interfere with mycotoxin production within plant tissue (Boutigny et al., 2008). On infection, plant cells respond with a hypersensitive reaction by the generation of reactive oxygen species (ROS), such as H₂O₂ and superoxide. The strong oxidative properties of H₂O₂ modulate trichothecene biosynthesis (Ponts et al., 2006; Sweeney and Dobson, 1999), leading to increased expression of TRI genes (Ochiai et al., 2007; Ponts et al., 2007). In vitro production of DON and ADON by F. culmorum chemotype I isolates was enhanced after H₂O₂ treatment, whereas NIV and FUS production by chemotype II isolates was reduced (Ponts et al., 2009). Differences in the efficiencies of detoxification have been described in F. culmorum isolates of the two chemotypes. Usually, chemotype I isolates exposed to oxidative stress react with an increase in catalase activity, resulting in a higher H₂O₂‐destroying capacity (Ponts et al., 2009).

Typical growth patterns of F. culmorum are accompanied by a pH increase during infection (Lamour and Marchant, 1977), followed by increased extracellular enzyme expression activity and DON production. The role of CWDEs as virulence factors in F. culmorum has been investigated extensively (Cooper et al., 1988; Hestbjerg et al., 2002; Miedaner et al., 1997; Tunalı et al., 2012; Wang et al., 2006). The production of CWDEs able to hydrolyse cellulose, xylan and pectin of the plant cell wall (PCW) allows F. culmorum to invade host tissues within 3–4 days (Kang and Buchenauer, 2002). These alterations may occur even before the presence of fungal hyphae within the host tissues, suggesting an apoplastic movement of these enzymes (Kang and Buchenauer, 2000a, 2000b).

Fusarium culmorum creates the conditions for maximum activity of its pectin lyases (PNLs) and other depolymerizing enzymes by raising the apoplastic pH from 6 to 7.3. When grown with pectin as the sole carbon source, F. culmorum modulates the pH to more alkaline conditions, favouring significantly PNL production and repressing polygalacturonase (PG) expression, which has an activity window at the very initial stages of infection. This pH change triggers the synthesis of additional ‘weapons’, such as subtilisin and trypsin‐like enzymes, which are relevant in this colonization phase (Aleandri et al., 2007; Pekkarinen and Jones, 2002; Pekkarinen et al., 2002). In vivo, F. culmorum attacks an arabinoxylan‐rich cell wall (constituting up to 40% of its components) of graminaceous crops, and produces much more xylanases than other pathogens (Bëlien et al., 2006; Carpita, 1996; Hatsch et al., 2006). Moreover, effective hydrolysis of PCW requires the synergistic action of several CWDEs that have been found to be expressed and to act in complexes (Alfonso et al., 1995; Collins et al., 2005; Jaroszuk‐Scisel et al., 2011). The activities of seven CWDEs (glucanases, chitinases, xylanases, endo‐ and exocellulases, pectinases, PGs) have been traced in cultures of F. culmorum grown on fungal cell walls (FCWs) or PCW as carbon source, with glucanases, chitinases, xylanases and pectinases revealing a significantly higher activity. Replacement of FCW by PCW triggers an increase in PG activity, underlining their role in the initial phase of host cell wall attack (Jaroszuk‐Scisel and Kurek, 2012). Fusarium culmorum cultures with FCW as the only carbon source enhance their acid glucanase and chitinase repertoire, whereas PCW‐based cultures produce high concentrations of xylanases, as also documented for Fusarium‐infected barley (Jaroszuk‐Scisel and Kurek, 2012; Schwarz et al., 2002). Differences in the disease induction and tissue colonization between pathogenic and nonpathogenic isolates of F. culmorum have also been related to their different CWDE efficiencies (Jaroszuk‐Scisel and Kurek, 2012) and to their ability to induce local and systemic defence responses, i.e. cell wall thickening or oxidative burst (Jaroszuk‐Scisel et al., 2008; Martinez et al., 2000).

On infection with an F. culmorum spore suspension, wheat seeds and seedlings express several pathogenesis‐related (PR) proteins, including glucanases (PR1, PR2), chitinase (PR3), peroxidase (POX) and the PR protein Wheatwin1‐2 (PR4) (Aleandri et al., 2008; Bertini et al., 2003; Caruso et al., 1999). In in vitro experiments, stimulation of wheat seeds with different chemical inducers, such as salicylic acid (SA) and jasmonic acid (JA), or by mechanical damage through wounding, was followed in each case by an increase in PR4 expression, indicating its regulation by these pathways (Bertini et al., 2003). Fusarium culmorum‐infected wheat roots, instead, underwent increased expression of defence‐associated genes in leaf sheaths which had not yet been in contact with the fungus, indicating the role of a systemic response in FRR (Beccari et al., 2011).

Effective and persistent resistance in the host plant can be induced by low‐molecular‐mass molecules able to restrict fungal growth in the different tissue layers or by the inhibition of fungal CWDEs. In wheat, xylanase‐specific inhibitors, such as TAXI (Goesaert et al., 2003), XIP (Juge et al., 2004), thaumatin‐like XI (TLXI; Fierens et al., 2007) and PG‐inhibiting proteins (PGIPs; Di Matteo et al., 2003; Ferrari et al., 2012) have been described. Transgenic wheat plants expressing the bean PvPGIP2 gene in their flowers showed significantly reduced symptoms in F. graminearum‐incited FHB (Ferrari et al., 2012). Pectin methyl‐esterification influences plant resistance, as PCW becomes less susceptible to fungal pectinases and endopolygalacturonases. The level of esterification in the PCW is controlled by a pectin methyl‐esterase inhibitor (PMEI), supposed to confer resistance to the plant when demethylation is effectively inhibited. Wheat transgenic lines expressing AcPMEI from Actinidia chinensis showed reduced pectin methyl‐esterase (PME) activity, and hence high pectin methylation levels and significantly reduced disease symptoms following inoculation with F. graminearum (Volpi et al., 2011). Recently, three PMEI genes have been identified and characterized in wheat (Rocchi et al., 2012), opening up new perspectives in the development of transgenic wheat lines potentially resistant to different Fusarium species, including F. culmorum.

Plants are able to chemically transform trichothecenes by their degradation or detoxification, or to reduce their accumulation by the inhibition of biosynthesis through the activity of endogenous compounds (Alabouvette et al., 2009; Bollina and Kushalappa, 2011; Boutigny et al., 2010; Yoshinari et al., 2008). Glycosylation represents the main plant‐driven chemical transformation of mycotoxins in response to Fusarium attack (Karlovsky, 2011). In the naturally FHB‐resistant wheat cultivar Sumai3, genetic mapping has revealed that the ability to detoxify DON by a DON glucosyltransferase colocalizes with a major quantitative trait locus (QTL) for FHB resistance (Lemmens et al., 2005). Transgenic Arabidopsis thaliana expressing a barley UDP‐glucosyltransferase exhibited resistance to DON (Shin et al., 2012). Although several studies have been devoted to the selection of plant glycosylases, this does not appear to be an efficient strategy to control mycotoxin production, because of the possibility that glycosyl‐protected mycotoxins may be re‐converted into the original toxic form by hydrolysis in the digestive tract or during food/feed processing (the so‐called ‘masked’ mycotoxins).

Some secondary plant metabolites, present in larger amounts in FHB‐resistant plants, have been shown to inhibit fungal growth in vitro and/or mycotoxin production by Fusarium spp. These are phenolic and polyphenolic compounds belonging to the benzoic and cinnamic acids, furanocoumarins, phenylpropanoids, chromenes and flavones (Bakan et al., 2003; Boutigny et al., 2010; Mellon et al., 2012; Ojala et al., 2000; Takahashi‐Ando et al., 2008; Wu et al., 2008). Most are constituents of PCW: in response to infection, plants release phenols from the cell wall in order to limit the pathogen spread by reinforcing plant structural components. Some dialkyl resorcinols and coumarins manifest antifungal activity against F. culmorum (Ojala et al., 2000; Pohanka et al., 2006). Moreover, phenols present anti‐oxidant and/or radical scavenging activities (Kim et al., 2006). Therefore, defence mechanisms triggered in the plant in response to pathogenic oxidative processes involve the production of these secondary metabolites that can interfere in different ways with trichothecene biosynthesis.

Options for Control

The multiple factors influencing fungal growth and trichothecene production by F. culmorum require the application of an integrated pest management approach, combining genetic, agronomic, chemical and biological control measures.

The growth of susceptible wheat varieties does not only increase the severity of FHB, but also the fungal biomass, with a consequent increase in the amount of toxins present in the harvested grain (Blandino et al., 2012; Snijders and Krechting, 1992; Tòth et al., 2008). The adoption of wheat cultivars showing resistance to primary infection and to the spread of the disease would be the ideal strategy. Unfortunately, there are no highly resistant wheat cultivars (Pereyra et al., 2004; Wisniewska and Kowalczyk, 2005). Nonetheless, extensive effort has been devoted to map the QTLs associated with FHB resistance in wheat (see, for example, Häberle et al., 2009; Schmolke et al., 2008). Genotypes bearing resistance to FHB have been reported and it is encouraging that resistance of a given genotype is not specific to a single Fusarium species, but can be extended to all the causative agents of this disease (Mesterhazy et al., 2005; Miedaner et al., 2012).

Being a typical seed‐borne pathogen, F. culmorum survives on or within the infected seed, which remains the main cause of pre‐ or post‐emergence seedling death, and contributes to increase the inoculum potential in the soil. Consequently, ploughing should be preferred to direct sowing or minimum tillage practices, which favour inoculum survival (Blandino et al., 2012; Dill‐Macky and Jones, 2000; Miller et al., 1998; Teich and Nelson, 1984). Similarly, crop rotation with noncereal host crop intermediates, such as legumes, alfalfa and Brassicaceae, may reduce the incidence of disease (Kurowski et al., 2011; Parry et al., 1995). The use of healthy seed coated with fungicides represents a most efficient means of control, but is usually limited to the early stages of the wheat cycle, as fungicides do not maintain their efficiency over a longer period. To improve the slow release of the delivered compound, a tebuconazole–β‐cyclodextrin inclusion complex has been proposed for the control of FRR during the early stages of durum wheat growth (Balmas et al., 2006).

Several fungicides, mainly belonging to the azole (bromuconazole, cyproconazole, metconazole, prochloraz, propiconazole, prothioconazole and tebuconazole) and strobin (azoxystrobin) classes, have been shown to control the disease by up to 70% in the field and to reduce the amount of mycotoxins in kernels; this is particularly evident under low disease pressure or on wheat genotypes possessing moderate resistance (Chala et al., 2003; Jones, 2000; Menniti et al., 2003; Paul et al., 2008). However, an increase in mycotoxin content in the kernel can occur when fungicides are applied at sublethal concentration or if they differ in their activity against distinct Fusarium pathogens (Covarelli et al., 2004; Gardiner et al., 2009; Gareis and Ceynowa, 1994; Haidukowski et al., 2005; Hysek et al., 2005; Matthies and Buchenauer, 2000; Matthies et al., 1999; Ochiai et al., 2007; Simpson et al., 2001; Stack, 2000). Moreover, the prolonged use of molecules sharing the same mode of action may induce a selective pressure on the pathogenic fungal populations, enabling the selection of resistance traits. Resistance to trifloxystrobin (a complex III respiration inhibitor) and isopyrazam (a complex II respiration inhibitor) has been reported recently on two isolates within two different chemotypes (Pasquali et al., submitted). These results have been confirmed on a larger set of isolates collected in Luxembourg (M. Beyer, Centre de Recherche—Gabriel Lippmann, Belvaux, Luxembourg , personal communication), suggesting that, as in the case of F. graminearum, these resistance traits are of natural origin (Dubos et al., 2011, 2013).

An alternative approach to minimize the risk of resistance among fungal populations relies on the use of new molecules, based on the structure of natural and natural‐like inhibitors, able to counteract the pathogenic and mycotoxigenic potential of natural populations of Fusarium, rather than acting on their saprophytic phase, or capable of stimulating natural resistance responses by the host plant. Essential oils of plant origin and some natural monoterpenes, considered as ‘Generally Recognized As Safe’ (GRAS) chemicals (safe for food use), have both inhibitory effects against mycotoxin biosynthesis and fungicide activity (Dambolena et al., 2008; Ellouze et al., 2012; Yaguchi et al., 2009). In particular, extracts from malva, chamomile and citrus manifest fungistatic activity against F. culmorum (Ellouze et al., 2012; Magro et al., 2006).

A specific and powerful inhibitory activity has been demonstrated by phenolic and polyphenolic natural compounds (Bakan et al., 2003; Boutigny et al., 2010; Desjardins et al., 1988; Takahashi‐Ando et al., 2008). The most abundant phenols extracted from maize kernel pericarp and wheat bran are trans‐ferulic acid and the corresponding dehydrodimers (DFAs), namely dehydrodiferulates (Bily et al., 2003; Boutigny et al., 2008; Kim et al., 2006). Hydroxycinnamic acids are known to be major components of the primary cell wall of cereals (Bakan et al., 2003). These compounds are ester bound to the C5 hydroxyl of the arabinosyl side chain of cell wall arabinoxylan chains. The feruloyl residues, predominant species, can also be dimerized under an oxidative coupling mediated by POXs, form cross‐links or dehydrodimers of ferulic acid, and then lead to a reinforcement of the primary wall of the plant.

A phenolic fraction rich in these phenolic acids manifested a drastic reduction on in vitro DON and ADON biosynthesis by F. culmorum (Boutigny et al., 2010). Although the mechanism remains unclear, it is reasonable to hypothesize that these compounds, mainly DFAs, interfere with in vitro cell wall degradation by fungal hydrolases. The activity of fungal esterases, overexpressed during growth on host tissues, can release free forms of ferulic ester from cell wall tissues (Balcerzak et al., 2012; Jaroszuk‐Scisel et al., 2011). Once released, free ferulate may inhibit the ability of Fusarium to produce mycotoxins. One of the DFAs present in the phenolic acid mixture, 8,5′‐benzofuran dimer, shows the same inhibitory activity of ferulic acid against F. culmorum, although a synergism of the phenolic acid mixture may play a crucial role in the inhibition of mycotoxins (Boutigny et al., 2010).

The X‐ray crystal structure of trichodiene synthase, purified from F. sporotrichioides and complexed with Mg²⁺(three ions)‐inorganic pyrophosphate (PPi), provides critical details regarding the molecular recognition of PPi, giving further insights into the trichothecene pathway, and therefore on the possibility of using external ligands able to interfere with mycotoxin production (Rynkiewicz et al., 2001; Vedula et al., 2008). The combination of bioprospecting and computational studies offers a useful way to select and investigate new natural and natural‐like mycotoxin inhibitors and fungicides against Fusarium. A collection of natural and natural‐like phenols and dimers was recently correlated with their ability to inhibit in vitro 3‐ADON and DON in F. culmorum and to interact with the trichodiene synthase crystal structure (G. Delogu, Istituto CNR di Chimica Biomolecolare, Sassari, Italy, unpublished data).

The susceptibility of the model plant A. thaliana to both F. graminearum and F. culmorum infection (Urban et al., 2002) has opened up new possibilities of developing high‐throughput experimental approaches to select new protecting compounds. Working with F. graminearum, Schreiber et al. (2011) identified small molecules, such as sulphamethoxazole and the indole alkaloid gramine, that protect Arabidopsis seedlings from infection. The same chemicals reduced significantly the severity of F. graminearum infection in wheat (Schreiber et al., 2011).

The integration of biological control approaches may offer an effective support to F. culmorum management on wheat and other cereals. The flag leaf and ripening ear surfaces of wheat are colonized by a panoply of micro‐organisms whose numbers may vary with plant growth stage and environmental conditions (Magan and Lacey, 1986). The application of natural antagonists to the crop residues or directly onto plant organs by spray or by seed dressing achieved reduced severity of FRR or FHB by F. culmorum on wheat, and the contamination of grain with mycotoxins (Table 4).

Table 4.

Biological control agents developed to control Fusarium culmorum infection on wheat

Antagonist	Target disease	Application method	Reference
Chaetomium sp. Idriella bolleyi Gliocladium roseum	FRR	Seed coating (field)	Knudsen et al. (1995)
Alternaria alternata Botrytis cinerea Cladosporium herbarum	FHB	Spray at ear emergence complete or anthesis complete (glasshouse)	Liggitt et al. (1997)
Trichoderma harzianum	FRR	Seed coating (field)	Michalikova and Michrina (1997)
Trichoderma harzianum Trichoderma atroviride Trichoderma longibrachiatum Gliocladium roseum Penicillium frequentans	FRR, FHB	Seed coating (field)	Roberti et al. (2000)
Gliocladium roseum (Clonostachys rosea)	FRR FRR	Seed coating (field) Seed coating (in vitro)	Jensen et al. (2000); Roberti et al. (2008)
Phoma betae	FHB	Spray at early anthesis (glasshouse)	Diamond and Cooke (2003)
Pseudomonas fluorescens Pantoea agglomerans	FRR	Seed coating (glasshouse and field)	Johansson et al. (2003)
Fusarium equiseti	FHB	Spray at anthesis (field)	Dawson et al. (2004)
Bacillus mycoides	FRR	Seed coating (microplot)	Czaban et al. (2004)
Different filamentous fungi and yeasts	FRR, FHB	Wheat straw (in vitro)	Luongo et al. (2005)
Pseudomonas fluorescens Pseudomonas frederiksbergensis	FHB	Spray at mid‐anthesis (glasshouse and field)	Khan and Doohan (2009); Petti et al. (2008)
Streptomyces sp.	FRR	Seed coating (glasshouse)	Orakci et al. (2010)
Bacillus subtilis	FRR	Seed coating (glasshouse)	Khezri et al. (2011)
Trichoderma gamsii	FRR, FHB	Wheat haulms and rice kernels (in vitro)	Matarese et al. (2012)

Functional Genomics

The F. culmorum genome is largely unknown. On analysis of the National Center for Biotechnology Information (NCBI) database for proteins associated with F. culmorum, 189 hits were returned on 15 November 2012. Annotated proteins include elongation factor 1α, a putative reductase, the RNA polymerase II, a phosphate permease, a putative regulatory protein used for phylogenetic analysis (Ward et al., 2002) and genes of the TRI cluster, involved in the synthesis of trichothecenes, also used for phylogenetic studies. Other F. culmorum annotated proteins include an ABC transporter (Skov et al., 2004), the trichodiene synthase used for RNA silencing experiments (Scherm et al., 2011), three putative allergenic proteins (Hoff et al., 2003), hydrophobin precursors involved in gushing (Stübner et al., 2010) and further proteins involved in the foam effect in beers (Zapf et al., 2007), and a fragment of a polyketide synthase essential in zearalenone biosynthesis (Atoui et al., 2012). Other genes have also been cloned in F. culmorum whilst studying the production of secondary metabolites, such as the nonribosomal peptide synthetase NPS2 able to synthesize ferricrocin (Tobiasen et al., 2007). Proteinases have also been isolated from F. culmorum (Levleva et al., 2006).

Functional characterization of the genes involved in the pathogenic process in F. culmorum is even more limited. Genetic transformation of the fungus is well established (Doohan et al., 1998), but the lack of a full genome has limited the functional analysis of genes to a few examples. Scherm et al. (2011) demonstrated that RNAi silencing as a functional approach is working in F. culmorum. Silencing of the zinc finger transcription factor TRI6, using inverted repeat transgenes, led to significantly decreased expression rates of the trichodiene synthase encoding gene TRI5 and, consequently, to a decline in DON production. Hence, trichothecene production of F. culmorum is tightly related to its aggressiveness and virulence in determining the symptoms of FRR on wheat (Scherm et al., 2011).

A second gene shown to play a role in pathogenesis is an ABC transporter, FcABC1, supposed to confer resistance to defensive compounds produced by the plant during the head infection process in wheat (Skov et al., 2004). The FcABC1 deletion mutant was unaltered in its physiology, but showed up to 98% reduced aggressiveness compared with the wild‐type strain, suggesting that the ability to excrete secondary plant metabolites allows F. culmorum to overcome the inhibition of host tissue invasion (Skov et al., 2004).

An F. culmorum topoisomerase I gene (top1) was found by a random plasmid insertional mutagenesis approach in F. graminearum and deleted in F. culmorum (Baldwin et al., 2010). The deletion mutant showed a complete block of conidia production as a result of its inability to regulate the transcriptional changes required for perithecial development. Furthermore, the mutant showed a significantly reduced virulence in wheat ear infection with low ability to colonize tissues after penetration (Baldwin et al., 2010).

The role of the gene FcStuA, a stuA orthologue protein with an APSES domain sharing 98.5% homology to the FgStuA transcription factor (FGSG10129) of F. graminearum (Lysøe et al., 2011), was recently determined by the functional characterization of deletion mutants. FcStuA was found to completely control pathogenicity and to reduce significantly (but not by blocking as in F. graminearum) DON production in F. culmorum mutants, together with a strong impairment of conidiation and significant morphological changes (M. Pasquali, Centre de Recherche—Gabriel Lippmann, Belvaux, Luxembourg, personal communication).

Given the very limited number of genes described to be involved in the pathogenic process in F. culmorum, further instruments and approaches are needed to explore the pathogenic arsenal of the fungus. A forward genetic tool based on a transposon insertion screening in the genome of F. culmorum (Spanu et al., 2012) did not lead to the identification of FRR PR genes, but allowed the isolation of partial sequences of aurofusarin genes and other genes involved in oxidative stress resistance, and the partial mapping of this unknown genome by the generation of more than 50 000 bp of F. culmorum sequence.

The availability of genomes would facilitate targeted functional genomics studies that, at the moment, are based on the similarities of genes with F. graminearum (Baldwin et al., 2010), but this cannot explore genes that are peculiar to F. culmorum (Spanu et al., 2012).

It is quite opportune that two F. culmorum genome sequencing programmes are on their way to being released. The first involves F. culmorum isolate FcUK99 (NRRL 54111; FGSC 10436), recovered from an infected wheat ear in the UK in 1998 (Baldwin et al., 2010). This isolate is fully pathogenic on wheat ears, tomato fruits and Arabidopsis floral tissue, and produces DON and 3‐ADON. By 454 sequencing, a 13.4× coverage of the F. culmorum isolate FcUK99 genome has been generated. In addition, four normalized cDNA libraries have been Illumina sequenced to give a transcriptome coverage of 100× (6 Gb of data). The F. culmorum genome size is estimated to be 39 Mbp, i.e. slightly larger than F. graminearum. In addition, the draft genomes of a further three F. culmorum isolates with different biological properties have been generated by sequencing with Illumina technology using 100‐bp pair‐end reads (M. Urban, J. Antoniw, N. Hall and K. E. Hammond‐Kosack, Wheat Pathogenomics, Plant Biology and Crop Sciences Department, Rothamsted Research, Harpenden, Herts, UK, personal communication).

As part of a larger programme of sequencing of the genomes of cereal Fusarium pathogens causing crown rot disease using Illumina paired‐end sequencing (see Gardiner et al., 2012), Donald Gardiner and John Manners at the Commonwealth Scientific and Industrial Research Organization (CSIRO, Clayton, Vic., Australia), together with Bioplatforms Australia (Sydney, NSW, Australia), have obtained sequence information for another isolate of F. culmorum, obtained from infected crown tissue of a wheat plant grown in Western Australia. Genome coverage will be >30‐fold and sequence information will be made publicly available early in 2013 on an Australian‐based website, and ultimately published on the NCBI site (J. M. Manners, CSIRO, Clayton, Vic., Australia, personal communication).

Future Challenges

Although it is not yet regarded as a ‘model system’, the F. culmorum–wheat interaction presents several features allowing it to be considered as a tractable model for investigation. Sequencing data permit a comparison of F. culmorum with other species whose genome information has already been released. One of the future challenges of genomics research will be to identify the peculiarities of this species involved in environmental adaptation and toxigenic and pathogenic potential compared with the closely related Fusarium spp. Many fundamental questions remain open. Has F. culmorum indeed lost its sexual cycle? What favours the shift in the F. culmorum/F. graminearum ratio in cereals? What is the role of nonpathogenic populations of F. culmorum in conferring adaptation to their host plants and how do saprophytic strains differ from pathogenic strains? Knowledge on the F. culmorum chemotype distribution worldwide may help us to better understand how chemotypes can be favoured by certain agroclimatological conditions. Given the general lack of information on the chemotype from the Southern Hemisphere and from worldwide populations of F. culmorum, it would be worth studying the chemotype distribution in relation to the host and to the disease phases (i.e. FHB or FRR), and comparing this with isolates obtained from undisturbed soils, in order to decipher the role of the chemotype in the presence versus absence of agricultural selection environments.

Finally, the identification of new natural and natural‐like molecules inhibiting trichothecene biosynthesis by F. culmorum, without affecting its vegetative growth, presents a vast array of practical applications. The bioavailability of inhibiting molecules and the evidence that exposure in vitro to different concentrations may result in opposite effects (i.e. inhibition versus enhancement of trichothecene production; G. Delogu, unpublished data) may prompt the development of new ecofriendly formulations to reduce the risk of these compounds being strongly affected by environmental conditions when applied in the field.