fgcdc14 regulates cytokinesis, morphogenesis, and ...xulab.org/files/pdf/lic-2015-mm.pdf ·...

FgCDC14 regulates cytokinesis, morphogenesis, andpathogenesis in Fusarium graminearum

Chaohui Li,1 Michael Melesse,2 Shijie Zhang,1,3

ChaoFeng Hao,1 Chenfang Wang,1

Hongchang Zhang,1 Mark C. Hall,3* andJin-Rong Xu1,3*1NWAFU-PU Joint research Center, State KeyLaboratory of Crop Stress Biology for Arid Areas,College of Plant Protection, Northwest A&F University,Yangling, Shaanxi 712100, China.Departments of 2Biochemistry and 3Botany and PlantPathology, Purdue University, West Lafayette, IN 47907,USA.

Summary

Members of Cdc14 phosphatases are common inanimals and fungi, but absent in plants. Although itsorthologs are conserved in plant pathogenic fungi,their functions during infection are not clear. In thisstudy, we showed that the CDC14 ortholog is impor-tant for pathogenesis and morphogenesis in Fusariumgraminearum. FgCDC14 is required for normal celldivision and septum formation and FgCdc14 pos-sesses phosphatase activity with specificity fora subset of Cdk-type phosphorylation sites. TheFgcdc14 mutant was reduced in growth, conidiation,and ascospore formation. It was defective inascosporogenesis and pathogenesis. Septation inFgcdc14 was reduced and hyphal compartments con-tained multiple nuclei, indicating defects in the coor-dination between nuclear division and cytokinesis.Interestingly, foot cells of mutant conidia often differ-entiated into conidiogenous cells, resulting in the pro-duction of inter-connected conidia. In the interphase,FgCdc14-GFP localized to the nucleus and spindle-pole-body. Taken together, our results indicate thatCdc14 phosphatase functions in cell division andseptum formation in F. graminearum, likely by coun-teracting Cdk phosphorylation, and is required forplant infection.

Introduction

Fusarium graminearum is the causal agent of head blight(FHB) or scab of wheat and barley, a plant disease withgreat impact on agriculture and food safety (Bai andShaner, 2004; Goswami and Kistler, 2004). The primaryinoculum of this disease is ascospores released fromperithecia produced by the pathogen that overwinters oninfested plant debris (Markell and Francl, 2003; Goswamiand Kistler, 2004). Plant infection is initiated whenascospores land on flowering wheat or barley headsthat are susceptible to F. graminearum during anthesis(Goswami and Kistler, 2004). Under favorable environ-mental conditions, outbreaks of FHB cause severe yieldlosses and reduce grain quality. In addition, F. gramine-arum produces toxic secondary metabolites, includingdeoxynivalenol (DON) that is a potent inhibitor of proteinsynthesis in eukaryotic organisms (De Walle et al., 2010).DON is also a potent phytotoxin and the TRI5 trichodienesynthase gene is the first virulence factor gene character-ized by molecular studies (Proctor et al., 1995). The tri5deletion mutant is capable of colonizing the inoculatedflorets, but fails to spread via the rachis to other spikeletson the same spikes in susceptible cultivars (Bai et al.,2002).

In eukaryotic organisms, reversible phosphorylation ofproteins by protein kinases and phosphatases is known toregulate various growth and developmental processes. Inan earlier study to functionally characterize the F. gramine-arum kinome, 42 protein kinase genes were shown to beimportant for plant infection (Wang et al., 2011). For 22 ofthese, including several components of the well-conservedcAMP signaling and mitogen-activated protein kinasepathways, gene deletion mutants were either non-pathogenic or caused FHB symptoms only on the inocu-lated florets (Hou et al., 2002, Jenczmionka et al., 2003,Urban et al., 2003, Van Thuat et al., 2012). Interestingly,unlike the model yeasts and filamentous fungi, F. gramine-arum contains two Cdc2 kinase orthologs that have distinctfunctions in vegetative and infectious hyphae (Liu et al.,2015), suggesting differences in cell cycle regulationbetween saprophytic and pathogenic growth. In dimorphicpathogens such as the human pathogen Candida albicansand corn smut fungus Ustilago maydis, it also has beenshown that cell cycle regulation plays an important role in

Accepted 6 August, 2015. *For correspondence. E-mail [email protected]; Tel. 765-496-6918; Fax 765-494-0363 or [email protected]; Tel. 765-494-0714; Fax 765-494-7897.

Molecular Microbiology (2015) 98(4), 770–786 ■ doi:10.1111/mmi.13157First published online 10 September 2015

© 2015 John Wiley & Sons Ltd

mailto:[email protected]



pathogenesis and morphogenesis although they have onlyone Cdc2 ortholog (Zheng et al., 2004; Perez-Martin et al.,2006; Perez-Martin and Castillo-Lluva, 2008; Wilson andHube, 2010). In C. albicans, a hyphal-specific cyclin Hgc1(hypha-specific G1 cyclin) is required for cell cycle progres-sion in hyphae and virulence (Zheng et al., 2004; Wilsonand Hube, 2010). In the rice blast fungus Magnaportheoryzae, one round of mitosis in germ tubes is required forappressorium formation (Caracuel-Rios and Talbot, 2007;Saunders et al., 2010), indicating that proper cell cycleregulation may be important for infection-related morpho-genesis in plant pathogenic fungi.

In F. graminearum, only one of the CDC2 orthologs isimportant for ascosporogenesis (Liu et al., 2015). There-fore, cell cycle progression may be also different betweenvegetative growth and sexual production. In addition toFgCDC2A, 45 protein kinase genes have been shown tobe important for ascospore production or release inF. graminearum (Wang et al., 2011). One of them, FgKin1,was found by further studies to be important for regulatingascospore germination and localization of Tub1 beta-tubulin (Luo et al., 2014). Deletion of FgKIN1 resulted inthe accumulation of Tub1 in the nucleolus in F. gramine-arum. FgKin1 is a member of the microtubule affinity-regulating protein kinase (MARK) family that are involvedin various cellular functions and conserved in fungi andanimals, but absent in plants (Drewes et al., 1998; Tassanand Le Goff, 2004).

Interestingly, the Cdc14 protein phosphatases also areconserved from yeast to humans, but are absent in plants(Kerk et al., 2008; Mocciaro and Schiebel, 2010).Although they are related evolutionarily to the dual-specificity group of the protein tyrosine phosphatases,Cdc14 enzymes generally function as antagonists ofmitotic cyclin-dependent kinases (CDKs) in the regulationof mitotic events (Queralt and Uhlmann, 2008; Mocciaroand Schiebel, 2010). Mitosis and cytokinesis are essentialcellular processes for cell division and proliferation ineukaryotic organisms. Successful completion of mitosisdepends on the temporal and spatial coordination of manyprocesses, including chromosome segregation, spindledisassembly, and cytokinesis. In the budding yeast Sac-charomyces cerevisiae, CDC14 is an essential gene(Wan et al., 1992). Temperature sensitive cdc14 mutantcells arrest in late mitosis with elongated mitotic spindlesand segregated chromosomes but fail to undergo cytoki-nesis (Wood and Hartwell, 1982). In Schizosaccharomy-ces pombe, the CDC14 ortholog clp1 is not an essentialgene. Although it is not required for mitotic exit, the Clp1protein may function together with the septation initiationnetwork (SIN) in coordinating cytokinesis with nucleardivision (Trautmann et al., 2001). In Aspergillus nidulans,the Ancdc14 deletion mutant also is viable and grows anddevelops normally (Son and Osmani, 2009). In the ento-

mopathogenic fungus Beauveria bassiana, BbCdc14 isnot essential, but is important for asexual development,stress responses and virulence. Inactivation of BbCDC14results in the formation of multinucleate cells and defectsin conidiation and pathogenesis (Wang et al., 2013). Inthe human pathogen C. albicans, CaCDC14 is not essen-tial for vegetative growth, but affects late cell cycle eventsand morphogenesis. Deletion of CaCDC14 has no effecton growth rate, but results in defects in cell separation,mitotic exit, and morphogenesis (Clemente-Blanco et al.,2006).

Orthologs of Cdc14 are well conserved in plant patho-genic fungi. However, none of them have been functionallycharacterized. In this study, we aimed to determine thefunctions of FgCDC14 in development and pathogenesisin F. graminearum. The Fgcdc14 deletion mutant was gen-erated and characterized for its defects in growth, sexualand asexual reproduction, and pathogenesis. Our resultsalso showed that FgCDC14 is important for the coordina-tion between nuclear division and septum formation inF. graminearum, likely by counteracting Cdk phosphoryla-tion during vegetative growth, cellular differentiation, andplant infection.

Results

FgCDC14 is important for growth and conidiogenesis

Despite their specificity for phospho(p)-Ser-Pro Cdksites, Cdc14 enzymes are evolutionarily and mechanis-tically related to the protein tyrosine phosphatases withthe HCX5R catalytic site sequence (Mocciaro andSchiebel, 2010). The FgCDC14 gene (FGSG_00543) isorthologous to CDC14 of S. cerevisiae and clp1 ofS. pombe and it has the same HCKAGLGR catalyticsequence. To determine its function in F. graminearum,we generated the Fgcdc14 mutant by the split-markerapproach (Catlett et al., 2003). Putative Fgcdc14 dele-tion mutants were identified by PCR. Three Fgcdc14mutants, M1, M2 and M3 (Table 1) were confirmed bySouthern blot analysis (Fig. S1) and they had the samephenotype described later.

The Fgcdc14 mutant was reduced approximately 55% ingrowth rate (Table 2) and formed colonies with limitedaerial hyphae on 5 × YEG plates (Fig. 1A), indicating thatthe Fgcdc14 mutant was defective in vegetative growth.The Fgcdc14 mutant was also reduced over 90% in conidi-ation compared with the wild type (Table 2). Whereas thewild type mainly produced conidia on clusters of phialides(Wang et al., 2011), the Fgcdc14 mutant formed muchfewer, sparsely distributed phialides (Fig. 1B). Interest-ingly, unlike wild-type conidia that are arrested in growth,the foot cells of conidia formed by the Fgcdc14 mutantoften grew further and functioned as conidiogenous cells to

Functions of Cdc14 in Fusarium graminearum 771

© 2015 John Wiley & Sons Ltd, Molecular Microbiology, 98, 770–786

produce additional conidia (Fig. 1C). Occasionally, the tipcompartment of conidia also differentiated and producedadditional conidia. In total, 12.6% of conidia formed by themutant were inter-connected.

Deletion of FgCDC14 affects septum formation anddistribution of nuclei

To determine whether deletion of FgCDC14 affects cytoki-nesis, we transformed an H1-GFP fusion construct intoPH-1 and the Fgcdc14 mutant M1. The wild-type conidiahad four to six septa and a single nucleus in each com-partment (Fig. 2A). In the Fgcdc14 mutant, 77.5% of theconidia had three or fewer unevenly distributed septa(Fig. 2A). Whereas the end compartments normally had

one nucleus, the majority (97.6%) of the internal compart-ments had two or more nuclei in Fgcdc14 conidia.

In comparison with the wild type, the Fgcdc14 mutantrarely formed septa in germ tubes (Fig. S2) and vegeta-tive hyphae (Fig. 2B), further indicating the importance ofFgCDC14 in septation in F. graminearum. Unlike the wild-type hyphae, nuclei were distributed unevenly in theFgcdc14 hyphae (Fig. 2B). We also examined nucleusdistribution in the phialides formed in CMC cultures. Thewild type had one nucleus, but the Fgcdc14 mutant hadtwo or more nuclei in each phialide (Fig. S3). Theseresults indicated that deletion of FgCDC14 affectedmitotic division and cytokinesis (septation). It appears thatthe Fgcdc14 mutant is defective in mitotic exit and nucleicontinue to divide in the absence of cytokinesis.

Table 1. Wild-type and mutant strains of Fusarium graminearum used in this study.

Strains Brief descriptions Reference

PH-1 Wild-type Cuomo et al., 2007M1 Fgcdc14 deletion mutant of PH-1 This studyM2 Fgcdc14 deletion mutant of PH-1 This studyM3 Fgcdc14 deletion mutant of PH-1 This studyWE6 Fgswe1 deletion mutant of PH-1 Wang et al. 2011C1 Fgcdc14/FgCDC14-GFP transformant of M1a This studyC2 Fgcdc14/FgCDC14-GFP transformant of M1 This studyPD8 Transformant of M1 expressing FgCDC14C342S-GFP This studyPD14 Transformant of M1 expressing FgCDC14C342S-GFP This studyHP6 Transformant of PH-1 expressing H1-GFP Luo et al., 2014HM1 Transformant of M1 expressing H1-GFP This studyHM2 Transformant of M1 expressing H1-GFP This studyHC1 H1-mCherry & FgCDC14-GFP transformant of M1 This studyHC2 H1-mCherry & FgCDC14-GFP transformant of M1 This studyAP2 Transformant of PH-1 expressing LifeAct-GFP This studyAP3 Transformant of PH-1 expressing LifeAct-GFP This studyAM1 Transformant of M1 expressing LifeAct-GFP This studyAM2 Transformant of M1 expressing LifeAct-GFP This studyT3-CG1 TUB3-mCherry & FgCDC14-GFP transformant of M1 This studyT3-CG2 TUB3-mCherry & FgCDC14-GFP transformant of M1 This study

a. All the GFP and mCherry fusion constructs were integrated ectopically in the F. graminearum genome.

Table 2. Defects of the Fgcdc14 mutant in growth, conidiation, plant infection, and DON accumulation.

StrainsGrowth rate(mm/d)a

Conidiation(105/ml)b Disease indexc

DON (ppm)d

Wheat Rice

PH-1 (WT) 9.9 ± 0.2Ae 31.4 ± 1.8A 11 ± 1.6A 1372.1 ± 265.1A 2295.6 ± 696.1A

M1 (Fgcdc14) 4.5 ± 0.4B 0.6 ± 0.1C 0.5 ± 0.5B 0.0B 64.0 ± 32.1B

C1 (Fgcdc14/FgCDC14-GFP)

9.7 ± 0.1A 14.8 ± 1.7B 10 ± 1.2A 1178.2 ± 160.8A 2865.2 ± 159.4A

PD14 (Fgcdc14/FgCDC14C342S)

3.3 ± 0.3C 0.7 ± 0.3C 0.4 ± 0.5B 0.0B 55.3 ± 20.0B

a. Daily extension in colony radius on 5 × YEG plates.b. Conidiation in 5-day-old CMC cultures.c. Disease index was rated by the averagenumber of symptomatic spikelets per head 14 dpi.d. DON production was assayed with infected wheat kernels and rice grain cultures.e. Meansand standard deviations were calculated from three independent measurements. Data were analyzed with the protected Fisher’s least significantdifference test. Different uppercase letters marked significant difference (P = 0.05).

772 C. Li et al. ■


Actin ring constriction, but not cortex actin was affectedin the Fgcdc14 mutant

Because actin plays a crucial role in cytokinesis in fungi(Berepiki et al., 2011) and the Fgcdc14 mutant wasdefective in cytokinesis, we transformed the LifeAct-GFPconstruct into PH-1 and the Fgcdc14 mutant. In theresulting transformants, localization of LifeAct-GFP to thecortex of hyphal tips was similar between the wild-typeand mutant strains (Fig. 3A). However, the localization ofLifeAct-GFP to the septum was defective in the Fgcdc14mutant. Time-lapse imaging showed that a mesh ofF-actin cables and patches named the septal actin tangle(SAT) (Heilig et al., 2014) was formed at the septationsite in both PH-1 and Fgcdc14 mutant (Fig. 3B).Whereas the SAT coalesced to form the cytokinetic acto-myosin ring (CAR) that constricts to form a septum in thewild type, the Fgcdc14 mutant was defective in the coa-lescence of SAT and constriction of CAR (Fig. 3B).These results indicate that deletion of FgCDC14 had noeffects on the localization of actin to the cortex of hyphal

tips, but impacted the formation and function of the con-tractile actin ring during septation, consistent with theobserved septation defects.

The Fgcdc14 mutant is defective in ascosporogenesis

The Fgcdc14 deletion mutant exhibited significantlyreduced ascospore formation and failed to produceascospore cirrhi (Fig. 4A). On selfing plates, peritheciaproduced by Fgcdc14 were normal in morphology andsize compared with those of the wild type (Fig. 4A).However, it appeared that most of the ascogenous tissueswere defective in further development in 14-day-oldFgcdc14 perithecia (Fig. 4B). In the wild type, mature ascicontained eight ascospores that had four uni-nucleatecompartments. In the Fgcdc14 mutant, mature asci wererarely observed (Fig. 4B). Ascospores formed by Fgcdc14were morphologically abnormal, had fewer than four com-

Fig. 1. Growth and morphogenesis defects of the Fgcdc14mutant.A. Three day-old 5 × YEG cultures of the wide type (PH-1),Fgcdc14 mutant (M1), and Fgcdc14/FgCDC14 complementedtransformant (C1).B. Phialides and conidia produced by PH-1, M1, and C1 in4-day-old CMC cultures. DC, developing conidia; GF, germinatingfoot cells; P, phialides. C. Morphologies of conidia from PH-1 andM1. Bars = 10 μm.

Fig. 2. Defects of the Fgcdc14 mutant in nucleus division andseptum formation.A. Conidia harvested from transformants of PH-1 (HP6) and theFgcdc14 mutant (HM1) expressing the H1–GFP construct werestained with Calcofluor white (CFW) and examined by differentialinterference contrast (DIC) or epifluorescence microscopy.B. Hyphae of transformants HP6 and HM1 were stained with CFW.The Fgcdc14 mutant was defective in septum formation. Septawere marked with arrows. Bars = 10 μm.



partments, and often contained more than one nucleusper compartment (Fig. 4C). Therefore, deletion ofFgCDC14 affected both meiotic and post-meiotic nucleardivisions and septation in developing ascospores.

FgCDC14 is important for plant infection

In infection assays with flowering wheat heads, the wild-type PH-1 caused head blight symptoms in the inocu-lated florets and spread to other spikelets. On wheatheads inoculated with the Fgcdc14 mutant, the inocu-lated florets developed limited symptoms 14 days post-inoculation (dpi) and infection failed to spread to nearbyspikelets (Fig. 5A). The average disease indexes of theFgcdc14 mutant and PH-1 were 0.5 and 13.0, respec-tively (Table 2). In infection assays with corn silks, theFgcdc14 mutant also caused only limited discoloration at

Fig. 3. Localization of LifeAct-GFP in the wild type and theFgcdc14 deletion mutant.A. Hyphal tip regions of the transformants of PH-1 (AP2) and theFgcdc14 mutant (AM1) showing similar cortical actin patches.B. Time-lapse images showing the defects in the constriction andcompletion of the actin ring during septation. Bars = 5 μm.

Fig. 4. Defects of the Fgcdc14 mutant in sexual reproduction.A. Three-week-old mating cultures of the wild-type (PH-1), Fgcdc14mutant (M1), and Fgcdc14/FgCDC14 complemented transformant(C1). Cirrhi (masses of ascospores) formed on the top of peritheciaare marked with arrows. Bar = 500 μm.B. Perithecia from mating cultures of PH-1, M1, and C1 wereexamined for ascosporogenous tissues 10-days post-fertilization.The Fgcdc14 mutant failed to form asci with eight ascospores.Bar = 20 μm.C. Ascospores from transformants of PH-1 (HP6) and Fgcdc14(HM1) expressing the H1–GFP construct were stained with CFWand examined by DIC or epifluorescence microscopy. Bar = 10 μm.

Fig. 5. Infection assays with flowering wheat heads and cornsilks.A. Flowering wheat heads were drop-inoculated with conidia of thewide type (PH-1), Fgcdc14 mutant (M1), or Fgcdc14/FgCDC14complemented transformant (C1). Black dots mark the inoculationsites. Photographs were taken 14 days post-inoculation (dpi).B. Corn silks were inoculated with culture blocks of the same set ofstrains and examined at 5 dpi.C. Colonization of lemma tissues by PH-1 and mutant M1 48 hpi.Infectious hyphae were marked with arrows. Bar = 20 μm.D. Rachises next to the spikelets inoculated by PH-1 or M1 wereexamined 120 hpi. Infectious hyphae formed by PH-1 in planttissues were marked with arrows. Bar = 20 μm.

774 C. Li et al. ■


the inoculation sites. Under the same conditions, exten-sive discoloration was observed in corn silks inoculatedwith PH-1 (Fig. 5B), confirming that FgCDC14 is impor-tant for virulence in F. graminearum.

When examined by light microscopy of thick sections,extensive hyphal growth was observed in lemma tissuesof wheat heads inoculated with PH-1 by 48 h post-inoculation (hpi). In contrast, only limited fungal growthwas observed in the epidermal cells in samples inocu-lated with the Fgcdc14 mutant (Fig. 5C). Because theFgcdc14 mutant was defective in spreading in inoculatedwheat heads, we also examined fungal growth in therachis. By 120 hpi, PH-1 produced extensive infectioushyphae in the vascular tissues of the rachis. In wheatheads inoculated with the Fgcdc14 mutant, we failed toobserve fungal hyphae in the rachis below or above theinoculated spikelet (Fig. 5D). Therefore, FgCDC14 mustbe essential for infectious hyphae to spread from inocu-lated florets to the rachis and colonization of vasculartissues.

We also assayed DON production in the infested wheatkernels and rice grain cultures because of its importancein plant infection. In comparison with the wild type andcomplementation strains, the Fgcdc14 mutant was signifi-cantly reduced in DON production in both assays(Table 2), suggesting that deletion of FgCDC14 affectsDON biosynthesis in F. graminearum.

Localization of FgCdc14 to the nucleus and MTOC

For complementation assays, an FgCDC14-GFP fusionconstruct was generated and transformed into theFgcdc14 mutant strain M1. The resulting complementedtransformant C1 was normal in growth, conidiogenesis(Fig. 1), ascospore production (Fig. 4), and plant infec-tion (Fig. 5). These results indicate that deletion ofFgCDC14 is directly responsible for the defects of thedeletion mutant.

In S. cerevisiae and S. pombe, the Cdc14 phos-phatase is sequestered in the nucleolus during inter-phase and released throughout the nucleus andcytoplasm during mitosis (Shou et al., 1999; Visintinet al., 1999; Cueille et al., 2001). When examined for itssubcellular localization, FgCdc14-GFP fusion was dif-fused throughout the nucleus although GFP signalsappeared to be enhanced at the spindle pole body(SPB) in over 99% of the nuclei examined (Fig. 6A). Toconfirm this observation, we co-transformed the TUB3-mCherry and FgCDC14-GFP fusion constructs into theFgcdc14 mutant M1. Tub3 is a gamma-tubulin that local-izes to the SPB in F. graminearum (Luo et al., 2014). Inthe resulting transformants, co-localization of Tub3-mCherry with FgCdc14-GFP to the SPB was observed(Fig. 6B).

FgCdc14 is a protein phosphatase with substratespecificity similar to S. cerevisiae Cdc14

ScCdc14 has been linked to the reversal of Cdk phos-phorylation (minimal consensus Ser/Thr-Pro) at the endof mitosis (Visintin et al., 1998). Cdc14 enzymes exhibita strict preference for the pSer-Pro subset of Cdk sites,both in vitro and in vivo (Eissler et al., 2011; 2014;Bremmer et al., 2012). In vitro, ScCdc14 also exhibits astrong requirement for Lys/Arg at the +3 position relativeto pSer (Bremmer et al., 2012). The most potent in vitroScCdc14 substrates contain at least one additional basicamino acid in the immediate vicinity of the +3 position.Alignment of the FgCdc14 and ScCdc14 proteinsequences suggests that this specificity should be highlyconserved in F. graminearum (Fig. 7A). The ScCdc14amino acids predicted to be responsible for pSer selec-tivity, +1 Pro and +3 Lys/Arg dependence, and prefer-ence for additional basic amino acids (Gray et al., 2003;Eissler et al., 2014), are all identical in FgCdc14. To

Fig. 6. Subcellular localization of the FgCdc14–GFP fusionprotein.A. Freshly harvested 12 h germlings of the FgCDC14-GFPH1-mCherry transformant HC1 were examined by DIC andepifluorescence microscopy. Bar = 10 μm.B. Hyphae of the FgCDC14-GFP TUB3–mCherry transformantT3-CG1 were examined for the localization of Tub3–mCherry andFgCdc14–GFP fusion proteins. Bar = 10 μm.



776 C. Li et al. ■


verify that FgCdc14 possesses phosphatase activity andto characterize its substrate specificity, we purifiedrecombinant full length FgCdc14 and the catalyticdomain comprising amino acids 1–433 (Bremmer et al.,2012) (hereafter called FgCdc14cat) (Fig. 7B). FgCdc14preparations catalyzed the hydrolysis of the generalphosphatase substrate p-nitrophenyl phosphate (datanot shown) and dephosphorylation of a phosphopeptidesubstrate containing a consensus Cdk phosphorylationsite (Fig. 7C) with specific activities comparable withScCdc14. As expected, this activity was inhibited bysodium tungstate, a potent inhibitor of protein tyrosinephosphatase family members, including Cdc14 (Tayloret al., 1997). Specific activity of FgCdc14cat was roughlysixfold lower than the full length protein in these assays(data not shown), suggesting that the non-conservedC-terminal tail contributes in some way to catalysis, as itdoes in ScCdc14 (Taylor et al., 1997).

We used three series of synthetic phosphopeptide vari-ants based on sequences from known ScCdc14 substrates(Bremmer et al., 2012) in steady-state kinetic assays toempirically define the substrate preference of purifiedFgCdc14 and FgCdc14cat (Fig. 7D–F; Fig. S4; Table 3).Two of the peptide sequences represented optimalScCdc14 sites containing the minimal consensus pSer-Pro-x-Lys with additional basic residues at +4 and +5,whereas the third sequence contained only the minimalconsensus pSer-Pro-x-Lys. Similar to ScCdc14, the cata-lytic efficiency of FgCdc14 toward peptides with a pThr wasup to three orders of magnitude lower than for identicalpeptides containing pSer. A pTyr peptide was also ineffi-ciently dephosphorylated. Like ScCdc14, high activitytoward pSer substrates was strongly dependent on a +1Pro and a +3 basic amino acid. The FgCdc14 enzymesexhibited a greater preference for Lys over Arg at the +3position than previously observed for ScCdc14 (Eissleret al., 2014). Also like ScCdc14, basic residues at +2, +4and +5 further enhanced catalytic efficiency of theFgCdc14 enzymes toward substrates containing theminimal consensus sequence pSer-Pro-x-Lys/Arg. Resi-

dues N-terminal to the pSer in the −2 and −1 positions didnot appear to contribute significantly to catalytic efficiency.Despite the lower specific activity of FgCdc14cat, its sub-strate specificity was the same as full length FgCdc14(Fig. S4). Collectively, these results indicate that theFgCdc14 consensus substrate recognition motif (Fig. 7F)is essentially identical to that of ScCdc14 (Eissler et al.,2014).

Phosphatase activity is essential for the function but notsubcellular localization of FgCdc14

To determine whether the phosphatase activity is essen-tial for FgCdc14 function and localization, we generatedthe FgCDC14C342S-GFP allele and transformed it into theFgcdc14 mutant. The C342 residue of FgCdc14 isequivalent to the catalytic C286 of S. pombe Clp1 andC283 of S. cerevisiae Cdc14, which are essential for thephosphatase activity (Taylor et al., 1997; Wolfe andGould, 2004; Wolfe et al., 2006). The resulting Fgcdc14/FgCDC14C342S-GFP transformants PD8 and PD14(Table 1) had similar defects in growth rate (Fig. 8A),conidiogenesis (Fig. 8B), and virulence (Fig. 8C) com-pared with the original Fgcdc14 mutant. When examinedby epifluorescence microscopy, FgCdc14C342S-GFPfusion proteins still localized to the nucleus and SPB(Fig. 8D). Therefore, the phosphatase activity is essen-tial for the FgCdc14 function, but dispensable for its sub-cellular localization.

Predicted biologic targets of FgCdc14

The strict intrinsic substrate specificity of ScCdc14 hasallowed novel substrates to be identified using motifsearching algorithms (Eissler et al., 2014). BecauseFgCdc14 exhibits similar substrate specificity we usedthe PatMatch algorithm (Yan et al., 2005) to identifyF. graminearum proteins enriched in consensusFgCdc14 recognition motifs that are conserved in itsorthologs from other Fusarium and Sordariomycetes

Fig. 7. FgCdc14 exhibits strong selectivity for a subset of Cdk-type phosphorylation sites in vitro.A. Alignment of ScCdc14 and FgCdc14 sequences. Amino acids important for catalysis and substrate recognition in ScCdc14 are highlighted.The defining catalytic site motif (H/V)Cx5R(S/T) of the protein tyrosine phosphatase family is underlined in black.B. Recombinant GST-FgCdc14 (upper panel, 96 kDa) and 6His-FgCdc14cat (lower panel, 52 kDa) analyzed by SDS-PAGE and Coomassieblue staining. M, molecular weight standards.C. Activities of GST-FgCdc14 and GST-ScCdc14 assayed with 100 μM of the series 1 phosphopeptide from Table 3. Where indicated, sodiumtungstate was included at 100 μM.D. Reaction velocity was measured as a function of substrate concentration for series 2 peptide variants (Table 3). Data are the average ofthree trials. For peptides exhibiting enzyme saturation, lines were generated by fitting data with the Michaelis-Menten equation. For lowefficiency peptide substrates lines were generated by linear regression and were used to calculate apparent kcat/KM (Table 3). Black squaresrepresent the unaltered peptide sequence shown above the plot.E–F. Apparent kcat/KM values obtained from initial velocity measurements at low concentrations (below KM) of series 1 (panel E) and 3 (panelF) phosphopeptide substrates. The first bar represents the original peptide sequence shown above the plot and the amino acid substitutions ineach of the other peptides are labeled on the x axis (position numbers relative to pS are shown above the sequence). Data are the average ofthree independent experiments and error bars are standard deviations.G. The optimal substrate motif for FgCdc14 based on the phosphopeptide kinetic assays.



species. The candidate FgCdc14 substrates and theirfunction annotations are listed in Table S1. Gene ontol-ogy analysis with their S. cerevisiae orthologs revealedsignificant enrichment for proteins involved in processesrelated to mitosis, microtubule functions, chromosomesegregation, and cell division, which is consistent withknown Cdc14 functions in other organisms (Table S2).Some candidates, such as the ortholog of chromosomalpassenger complex subunit INCENP (yeast Sli15) andthe kinetochore component Ask1, are well-characterizedCdc14 substrates. The IQGAP and anillin orthologsinvolved in cytokinesis were also predicted as ScCdc14substrates using similar search criteria (Eissler et al.,2014).

We used a yeast two-hybrid assay to test for physicalinteractions between an FgCdc14 bait construct and preyconstructs of two substrate candidates, the protein kinasesFgDbf2 and FgWee1. The bait and prey constructs wereverified by DNA sequencing and co-transformed in pairsinto yeast strain AH109. The resulting Trp + Leu + trans-formants were able to grow on the SD-Trp-Leu-His reporter

plates and had LacZ activities (Fig. 9), indicating thatFgCdc14 interacts with FgDbf2 and FgWee1. Theseresults support our substrate prediction strategy andsuggest that these two protein kinases are likelyphysiologic targets of the FgCdc14 phosphatase inF. graminearum.

Deletion of FgCDC14 results in reduced Tyr15phosphorylation levels of Cdc2A/2B

In S. pombe, Clp1 phosphatase restrains the G2/M transi-tion by promoting the activity of Wee1 kinase, suppressingCdc25 phosphatase, and thereby protecting inhibitoryphosphorylation of Cdk1 at Tyr15 (Cueille et al., 2001;Trautmann et al., 2001; Wolfe and Gould, 2004; Lu et al.,2012). Similarly, in humans silencing of Cdc14A results inreduced expression level of Wee1 and phosphorylation ofCdk1 at Tyr15, suggesting that Cdc14A antagonizes Cdk1activation and mitotic entry (Ovejero et al., 2012). InF. graminearum, phosphorylation of Cdc2A and Cdc2B,two Cdc2 orthologs with similar molecular weight were

Table 3. Catalytic efficiency (kcat/KM) measurements of purified FgCdc14.

SeriesPhosphopeptidesequence

Sequencevariation

FgCdc14apparentkcat/Km (M−1S−1)

FgCdc14cat

apparentkcat/Km (M−1S−1)

S1 MIpSPSKKRTI 3674 ± 189 214 ± 7MIpTPSKKRTI pT 5 ± 1 0.29 ± 0.03MIpYPSKKRTI pY 74 ± 2 18.2 ± 0.6MIpSASKKRTI +1A 21 ± 2 12.4 ± 0.2MIpSPSAKRTI +3A 48 ± 19 11.5 ± 0.4MIpSPSRKRTI +3R 524 ± 9 150 ± 8GDpSPSKKRTI −2G, −1D 2313 ± 92 165 ± 10MIpSPSKAATI +4A, +5A 920 ± 17 183 ± 4MIpSPSRAATI +3R, +4A, +5A 115 ± 5 22 ± 1MIpSPSNAATI +3N, +4A, +5A 8 ± 1 2.5 ± 0.3

S2 VKGNELRpSPSKRRSQI 2143 ± 171 NDVKGNELRpTPSKRRSQI pT 4.1 ± 0.4 NDVKGNELRpSPSARRSQI +3A 56 ± 2 NDVKGNELRpSPSRRRSQI +3R 98 ± 25 NDVKGNELRpSPSRAASQI +3R, +4A, +5A 71 ± 22 NDVKGNELRpSPSKARSQI +4A 951 ± 209 NDVKGNELRpSPSKRASQI +5A 1040 ± 209 NDVKGNELRpSPSKAASQI +4A, +5A 317 ± 35 NDVKGNELRpSPSAAASQI +3A, +4A, +5A 19 ± 1 ND

S3 LLpSPGKQFRQ 219 ± 6 73 ± 2LLpTPGKQFRQ pT 4.4 ± 0.4 1.1 ± 0.1LLpSAGKQFRQ +1A 4.5 ± 0.3 0.97 ± 0.02LLpSPKKQFRQ +2K 2414 ± 75 311 ± 12LLpSPGAQFRQ +3A 9.9 ± 0.4 2.9 ± 0.1LLpSPGRQFRQ +3R 17 ± 1 7.4 ± 0.2GDpSPGKQFRQ −2G, −1D 174 ± 3 65 ± 1LLpSPGKKFRQ +4K 1007 ± 37 208 ± 8LLpSPGKQRRQ +5R 213 ± 8 74 ± 3

Catalytic efficiency (kcat/KM) toward phosphopeptide substrates was estimated from initial velocity measurements at low substrate concentrations(below KM). Data are the average of 3 independent experiments with standard deviations. Underlined text indicates the positions altered from theoriginal sequence in each peptide variant and the sequence variation column shows the nomenclature for each variant used in Fig. 6D–F andFig. S4. The phosphorylated amino acid is preceded by ‘p’ and shown in bold.

778 C. Li et al. ■


detectable with an anti-Tyr15 phosphorylation antibody(Fig. 10). In comparison with the complemented transfor-mant and PH-1, the Fgcdc14 deletion mutant was reducedin the phosphorylation level of Cdc2A/2B (Fig. 10). TheFgcdc14/FgCDC14C342S-GFP transformant PD14 also wasreduced in the Tyr15 phosphorylation level compared withPH-1 or the complemented transformant C1 (Fig. 10).

Reduction in the FgWee1 activity in the Fgcdc14 mutantmay affect the inhibitory phosphorylation of Cdc2A/2Band maintenance of the interphase state in phialides,ascospores and conidia.

Discussion

The CDC14 phosphatase genes are conserved in fungiand animals, but absent in higher plants, similar to theMARKs that are functionally related to microtubules(Drewes et al., 1998; Tassan and Le Goff, 2004). It istempting to speculate that the lack of MARK proteinkinases and Cdc14 phosphatases are functionally relatedto flagellum functions in motile cells. Whereas higherplants lack motile cells or flagella, chytridiomycetes, likelyancestral fungi, produce zoospores with basal bodies andflagella of typical 9 + 2 microtubule arrangement (Mitchell,2007). Basal bodies are microtubule organizing centersfor flagella, while SPBs or centrioles are microtubule-organizing centers for the mitotic spindle. Although theassociation of Cdc14 with basal bodies or motile cells havenot been documented in Chytridiomycetes, ZebrafishCdc14A1 phosphatase localizes to the base of the cilium,possibly in part at the basal body (Clement et al., 2012) andPiCdc14 from the oomycete Phytophthora infestans accu-

Fig. 8. The C342 mutation affected the function but notlocalization of FgCdc14.A. Two-day-old 5 × YEG cultures of the wide type (PH-1), Fgcdc14mutant (M1), Fgcdc14/FgCDC14 complemented transformant (C1),and Fgcdc14/FgCDC14C342S-GFP transformant (PD14).B. Conidia formed by transformant PD14 in 4-day-old CMCcultures.C. Flowering wheat heads and corn silks inoculated with PH-1 andtransformant PD14 were examined 14 and 6 dpi, respectively.D. Hyphae of transformant PD14 were examined by DIC andepifluorescence microscopy. Bar = 10 μm.

Fig. 9. Assays for the interaction of FgCdc14 with FgDbf2 andFgWee1 by yeast two-hybrid. Different cell concentrations(cells ml−1) of the yeast transformants expressing the labeled baitand prey constructs were assayed for growth on SD-Leu-Trp-Hisplates. Positive and negative controls were provided in the BDMatchmaker library construct kit. On the right, the same set ofyeast transformants was assayed for β-galactosidase activities.

Fig. 10. Assay for the Tyr15 phosphorylation level of Cdc2proteins.A. Western blots of total proteins isolated from the wild type (PH-1),Fgcdc14 mutant (M1), Fgcdc14/FgCDC14-GFP transformant (C1),Fgcdc14/FgCDC14C342S-GFP transformant (PD14), and the Fgwee1deletion mutant (WE6) were detected with the anti-phosphor Tyr15(pY15) and anti-PSTEIRE (Cdc2) antibodies. Detection with ananti-H3 antibody was the control for protein loading (bottom).B. The relative intensity of Tyr15 phosphorylation of Cdc2A/2B indifferent mutants in comparison to that of the wild type (arbitrarilyset to 1 for PH-1) after normalization with the Cdc2A/2B bandsdetected with the anti-PSTEIRE antibody. The intensities ofWestern blot bands were quantified with program ImageJ.



mulates at the basal body during zoospore formation(Ah-Fong and Judelson, 2011). PiCdc14 binds microtu-bules in vitro and forms insoluble complexes in vivo. It ispossible that one of the ancestral functions of Cdc14phosphatases, and possibly MARKs, are related to basalbody functions.

In F. graminearum, the localization of FgCdc14 islargely non-nucleolar during interphase, which is similar toPiCdc14 (Ah-Fong and Judelson, 2011), but different fromits ortholog in S. cerevisiae and S. pombe (Shou et al.,1999; Visintin et al., 1999; Cueille et al., 2001). It is pos-sible that subcellular localization of Cdc14 phosphatasesdiffer between yeasts and filamentous fungi. In C. albi-cans, CaCdc14p-YFP begins to accumulate both in thenucleolar staining (Nop1 signal) and DAPI-staining regionof the nucleus at the G1-S transition. At the end of mitosis,CaCdc14p is degraded instead of being sequestered intothe nucleolus (Clemente-Blanco et al., 2006). In plantpathogenic fungi, particularly hemibiotrophic pathogens,the Cdc14 orthologs may play important roles in infection-related morphogenesis and development of invasivehyphae that differ in morphology and physiology fromvegetative hyphae.

Although budding yeast Cdc14 and fission yeast Clp1are well characterized, Cdc14 orthologs have not beenfunctionally or biochemically characterized in plant patho-genic fungi. In F. graminearum, the Fgcdc14 mutanthad pleiotropic defects in hyphal growth, conidiation,ascospore formation, septation and plant infection. Giventhat proper regulation of cell cycle progression is impor-tant for fungal development and pathogenesis in fungalpathogens (Pérez-Martín et al., 2006), the observed phe-notypes are consistent with known cell cycle regulatoryfunctions for Cdc14 in budding and fission yeasts. InB. bassiana, BbCdc14 is also important for asexual repro-duction and insect infection (Wang et al., 2013). In C. albi-cans, CaCdc14p activity is required for true hyphal growthand invasion (Clemente-Blanco et al., 2006).

Interestingly, the tip compartment of Fgcdc14 conidiawere uni-nucleate and sometimes continued to grow anddevelop new conidia, suggesting defects in mitotic exitand restraining cell division. Earlier publication hasshown that fresh conidia often germinate from the endcompartments in F. graminearum (Harris, 2005). InC. albicans, deletion of CaCDC14 results in a significantdelay in the degradation of Clb2p during mitotic exit andCaCdc14p also plays an essential function in the acti-vation of the Ace2p transcriptional program required forcell separation (Clemente-Blanco et al., 2006). In fissionyeast, Clp1 regulates the G2/M transition and mitotic exitby dephosphorylating and suppressing activity of theCdc25 phosphatase that is involved in the activation ofCdk1 (Wolfe and Gould, 2004; De Wulf et al., 2009).Cells lacking Clp1 enter mitosis precociously, whereas

cells overexpressing Clp1 delay mitotic entry (Trautmannet al., 2001; 2004). It is possible that further growth fromthe tip compartments of conidia is related to the direct orindirect roles of FgCdc14 in the de-phosphorylation ofFgCdc25 and kinase activities of the Cdc2 orthologs (Liuet al., 2015) in F. graminearum.

The enzymatic specificity of Cdc14 enzymes isremarkably well conserved during evolution. It has beenreported that ScCdc14 and human Cdc14A and Bpossess the unusual ability to strictly distinguishbetween serine and threonine as the phoshoamino acidwithin Cdk-type phosphorylation sites (Bremmer et al.,2012). FgCdc14 clearly has this same phosphoserinespecificity. Thus, although structurally and mechanisti-cally related to protein tyrosine phosphatases, Cdc14evolved into a serine-selective phosphatase in the verydistant past of eukaryotic evolution. Additional strongspecificity determinants in ScCdc14 include +1 Pro, +3Lys/Arg (preferentially Lys), and additional Lys/Arg resi-dues in the vicinity of +3 (Eissler et al., 2014). FgCdc14also required these same specificity determinants forhigh activity, and was even more selective for Lys overArg at the +3 position.

Because of the conserved enzymatic properties ofFgCdc14 and ScCdc14, it is likely that FgCdc14 alsofunctions by selectively dephosphorylating specific Cdkphosphorylation sites in F. graminearum. Using anapproach similar to that used previously to predictScCdc14 substrates (Eissler et al., 2014), we identified56 putative targets of FgCdc14 (Table S1). Two pre-dicted targets of FgCdc14 are the protein kinasesFGSG_10228 (FgWee1) and FGSG_08635 (FgDbf2).FgCdc14 interacted with both FgWee1 and FgDbf2 in ayeast two-hybrid assay, providing support that they couldbe direct substrates. Orthologs of FgWEE1 and FgDBF2in S. cerevisiae are functionally related to mitosis andcytokinesis, respectively (Harvey and Kellogg, 2003;Mohl et al., 2009; Lianga et al., 2013). Cdc14 has beenlinked to the regulation of Wee1 orthologs in buddingyeast and humans (Ovejero et al., 2012; Raspelli et al.,2015). In F. graminearum, the Fgdbf2 and Fgwee1deletion mutants had reduced growth rate, conidiationand virulence. Both mutants produced morphologicallynormal perithecia that contained aborted ascogenoustissues and defective ascospores. In addition, theFgwee1 mutant was reduced in septation and rarely pro-duced septa in hyphae (Wang et al., 2011). The similar-ity of their phenotypes to that of the Fgcdc14 mutantsuggests that FgWee1 and FgDbf2 kinases may be alsofunctionally related to FgCdc14 in this important plantpathogenic fungus.

One of the most striking cellular defects in theFgcdc14 mutant was the absence of regular septationand normal distribution of nuclei. Multinucleate compart-

780 C. Li et al. ■


ments in ascospores and asexual spores, and vegeta-tive hyphae were observed in F. graminearum. Thissuggests that FgCdc14 plays important roles in regulat-ing septation and the coordination of cytokinesis withnuclear division. Actin plays a crucial role in cytokinesisin fungi (Berepiki et al., 2011) and the Fgcdc14 mutantwas defective in the formation of an actin ring duringseptation, which may be directly related to its defects inseptation. Interestingly, the Fgcdc14 mutant phenotypesare similar to those of associated with loss of type IImyosin Myo2 (Song et al., 2013), a molecular motor thatgenerates force for cytokinesis by interacting with actinfilaments in the contractile ring (CR). In S. cerevisiae,some myo1 mutant cells appeared to form chains orwere aberrantly large and/or multinucleate (Bi et al.,1998). In S. pombe, the myo2 mutant is also defective incell separation, resulting in the formation of septated,elongated cells with multiple nuclei (Kitayama et al.,1997). The recruitment of Myo2 to the CR is partiallydependent on anillin-related Mid1, and dephosphoryla-tion of Myo2, possibly by Clp1, has been implicated inmediating a Mid1-Myo2 interaction (Motegi et al., 2004;Clifford et al., 2008). Our results showed that function ofthe CR is abnormal in the Fgcdc14 mutant, likely result-ing in the septation defects and possibly the productionof inter-connected conidia and irregular nuclear distribu-tion in vegetative hyphae. Aberrant interconnectedmacro-conidia produced by the myo2 mutant andFgcdc14 mutants are reminiscent of the chains of cellsproduced by S. cerevisiae and S. pombe mutantslacking the type II myosin (Kitayama et al., 1997; Biet al., 1998; Song et al., 2013).

In S. pombe, the SIN plays a role in a cytokinesischeckpoint that inhibits further cell cycle progression untilcytokinesis is complete. Clp1 may function together withthe SIN in coordinating cytokinesis with the nuclear-division cycle (Trautmann et al., 2001). The F. gramine-arum genome has orthologs of all the SIN components.Therefore, like Clp1, FgCdc14 is likely required for theinitiation of cytokinesis and cytokinesis checkpoint forcompletion of cytokinesis. Although the direct FgCdc14targets responsible for the CR and septation defectsremain unknown, several proteins with functions relatedto septation, cytokinesis and actin/CR function areenriched in optimal Cdc14 substrate sites (Table S1).These include orthologs of budding/fission yeast IQGAPprotein Iqg1/Rng2, the annilin-related Bud4/Mid2, theformin Bni1/Rid1, the MEN/SIN scaffold Nud1/Cdc11, andthe septation factor and CR component Hof1/Imp2(Table S1). Cdc11 and Bni1 are known Cdc14 substratesin S. pombe and S. cerevisiae, respectively (Bloom et al.,2011; Chen et al., 2013). It will be important to determinethe functions of these genes and their relationship withFgCdc14 in F. graminearum.

Experimental procedures

Strains of F. graminearum and culture conditions

The wild-type strain PH-1 (Cuomo et al., 2007) and mutantsof F. graminearum generated in this study (Table 1) wereroutinely cultured at 25°C on PDA plates. Growth rate on5 × YEG (0.5% yeast extract, 1% glucose) plates, conidiationin carboxymethylcellulose (CMC) medium (Cappellini andPeterson, 1965), and sexual reproduction on carrot agarplates were assayed as previously described (Wang et al.,2011; Zheng et al., 2012). For DNA and protein extraction,vegetative hyphae were harvested from liquid YEPD (1%yeast extract, 2% peptone and 2% glucose). Protoplastpreparation and fungal transformation were performed asdescribed (Proctor et al., 1995; Hou et al., 2002). For trans-formation, hygromycin B (CalBiochem, La Jolla, CA, USA)and geneticin (Sigma-Aldrich, St. Louis, MO, USA) wereadded to the final concentration at 300 and 350 μg/ml,respectively.

Generation of the Fgcdc14 deletion mutants

The FgCDC14 gene replacement construct was generatedwith the split-marker approach (Catlett et al., 2003). The847-bp upstream and 703-bp downstream fragments ofFgCDC14 were amplified with primer pairs CDC14/1F-CDC14/2R and CDC14/3F-CDC14/4R (Table S3), respec-tively. The resulting PCR products were connected to thehygromycin phosphotransferase (hph) fragments amplifiedwith primers HY/R-YG/F or HYG/F-HYG/R by overlappingPCR and transformed into protoplasts of PH-1 as described(Zhou et al., 2010). After transformation of PH-1 protoplasts,hygromycin-resistant transformants were identified by PCRwith primers CDC14/5F and CDC14/6R (Table S3) andfurther characterized by Southern blot analysis (Fig. S1).

Generation of the FgCDC14-GFP andFgCDC14C342S-GFP transformants

To generate the FgCDC14-GFP construct by gap repair(Zhou et al., 2011), the entire FgCDC14 gene, including itspromoter region, was amplified with primer CDC14-CM/F andCDC14-CM/R (Table S3) and transformed with XhoI-digestedpFL2 (Zhou et al., 2011) into yeast strain XK1-25 asdescribed (Bruno et al., 2004). The resulting fusion constructrescued from Trp + yeast transformants was confirmed bysequencing analysis and transformed into the Fgcdc14 dele-tion mutant M1 (Table 1). Geneticin-resistant transformantsexpressing the FgCDC14-GFP construct were identified byPCR and examined for GFP signals. The same gap repairapproach was used to generate the FgCDC14C342S-GFP con-struct by amplifying FgCDC14 with primers PD/F and PD/R(Table S3) to introduce the C342S mutation. Transformantsof mutant M1 expressing the FgCDC14C342S-GFP constructwere identified by PCR and examined for GFP signals byepifluorescence microscopy.

Generation of the LifeAct-GFP transformants

The first 17 amino acid residues (MGVADLIKKFESISKEE) ofyeast Abp140 named LifeAct is an F-actin marker for higher



eukaryotes (Riedl et al., 2008). The LifeAct-Gly (10)-GFP con-struct used in M. oryzae (Zhou et al., 2012) was transformedinto protoplasts of PH-1 and the Fgcdc14 mutant M1. Trans-formants expressing the LifeAct-GFP construct were identifiedby PCR and confirmed by examination by for GFP signals.

4, 6-Diamidino-2-phenylindole (DAPI) andcalcofluor staining

Freshly harvested conidia and hyphae of F. graminearumwere first fixed for 30 min in 3.7% formaldehyde and 0.2%Triton X-100 in 50 mM PBS buffer (pH 7.0). After staining with20 μg/ml Calcofluor White (CFW) and 20 μg/ml DAPI, sam-ples of conidia or germlings were examined with an OlympusBX53 fluorescence microscope (Olympus, Tokyo, Japan). Toassay for defects in sexual reproduction, perithecia harvestedfrom 10-day-old mating plates were examined for the produc-tion of asci and ascospores after DAPI and CFW staining.

Plant infection and DON assays

Conidia harvested from 5-day-old CMC cultures werere-suspended to 2.0 × 105 conidia ml−1 in sterile distilled water.Flowering wheat heads of cultivar Xiaoyan 22 were inoculatedwith 10 μl of conidium suspensions at the fifth spikelet from thebase of the inflorescence as described (Gale et al., 2007; Dinget al., 2009). Spikelets with typical head blight symptoms ineach head were examined 14 dpi to estimate the diseaseindex as described (Gale et al., 2007; Ding et al., 2009) andinfected wheat kernels were harvested for DON assays(Bluhm et al., 2007). Infection assays with corn silks wereconducted as described (Seong et al., 2005). DON productionin rice grain cultures was assayed as described (Seo et al.,1996; Bluhm et al., 2007). All the infection and DON produc-tion assays were repeated at least three times.

Microscopic observations

Glumes and lemmas were collected from inoculated spikeletsand fixed with 4% (vol/vol) glutaraldehyde in 0.1 M phosphatebuffer (pH 6.8) overnight at 4°C. For light microscope exami-nation, infected wheat lemma and rachis tissues were fixed,dehydrated, and embedded in Spurr resin as described(Kang et al., 2008; Zheng et al., 2012). Thick sections (1 μm)were stained with 0.5% (wt/vol) toluidine blue and examinedwith an Olympus BX-53 microscope. At least three independ-ent biologic replicates were examined for the wild-type andmutant strains.

Purification of FgCdc14

The FgCDC14 ORF and its catalytic domain (codons 1–433,FgCDC14cat) were amplified from the first-strand cDNA byPCR, cloned into the Gateway® entry vector pENTR/D-TOPO [Thermo Fisher Scientific (Waltham, MA, USA)], veri-fied by DNA sequencing, and then transferred to thedestination vector pDEST15 (FgCDC14) and pDEST17(FgCDC14cat) for expression as an N-terminal glutathioneS-transferase (GST) fusion or an N-terminal 6x histidinefusion, respectively, in Escherichia coli. GST-FgCdc14

expression in BL21 (DE3) cells was induced with 0.4 mMisopropyl β-D-thiogalactopyranoside for 16 h at 25°C. The6His-FgCdc14cat fusion was induced in BL21 AI cells (LifeTechnologies) with 0.2% L-arabinose.

For GST-FgCdc14 purification, bacteria were lysed in 5 cellpellet volumes of buffer A (25 mM Tris-HCl pH 7.5, 500 mMNaCl, 2 mM ethylenediaminetetraacetic acid (EDTA), 0.1%Triton X-100, and 0.1% 2-mercaptoethanol) supplementedwith 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 μM pep-statin, and 10 μM leupeptin. Cells were lysed for 30 min onice with 1 mg ml−1 lysozyme and then sonicated to reduceviscosity. After centrifugation at 35,000 g for 30 min at 4°C,the resulting soluble extract was incubated with 1 ml glu-tathione agarose resin [EMD Biosciences (San Diego, CA,USA)] pre-equilibrated with buffer A for 1 h at 4oC on arocking platform. Agarose resin was collected by centrifuga-tion (700 g, 2 min, 4°C) and then washed three times with10 ml buffer A with rocking for 5 min at 4°C. GST-FgCdc14was eluted by incubating the resin several times with 1 mlbuffer A supplemented with 10 mM reduced glutathione at4°C for 5 min followed by centrifugation.

For 6His-FgCdc14cat purification, bacteria were lysed onice for 30 min in 10 cell pellet volumes of buffer B (25 mMTris-HCl pH 7.5, 500 mM NaCl) supplemented with 0.1%Triton X-100, 10 mM imidazole, 1 mM PMSF, 1 μM pepstatin,10 μM leupeptin, 1 mg/ml lysozyme, and 25 units/ml Univer-sal Nuclease [Thermo Fisher (Grand Island, NY, USA)].Lysate was clarified as described earlier, and the solubleextract was incubated with 1 ml HisPur nickel-NTA resin (LifeTechnologies) for 10 min at 4°C with gentle agitation. Afterwashing sequentially with 15 ml buffer B containing 10 mM,25 mM, and finally 40 mM imidazole, 6His-FgCdc14cat proteinwas eluted with five 1 ml aliquots of buffer B containing200 mM imidazole, collecting 0.5 ml fractions. For both pro-teins, elution fractions containing high protein concentrationwere pooled and dialyzed in 1L of storage buffer (25 mMTris-HCl pH 7.5, 300 mM NaCl, 2 mM EDTA, 0.1%2-mercaptoethanol, 40% glycerol) overnight at 4°C. Theresulting recombinant proteins were analyzed by SDS-PAGEand stored in small aliquots at −80oC. ScCdc14 was purifiedas described (Bremmer et al., 2012).

Phosphopeptide preparation and enzyme activity assay

Phosphopeptide substrates (Table 3) were synthesized byGenscript Inc. (Piscataway, NJ, USA) in crude form and thendesalted with Sep-Pak C18 cartridges [Waters Corporation(Milford, MA, USA)], dried by lyophilization, and reconstitutedin water. The concentrations of phosphopeptide stocks weremeasured with an ashing procedure and malachite green-ammonium molybdate dye as described previously (Bussand Stull, 1983). Phosphatase reactions (50 μl) containingrecombinant FgCdc14 (75 nM, or more for poor substrates)or FgCdc14cat (600 nM, or more for poor substrates), 25 mMHEPES pH7.5, 0.1% 2-mercaptoethanol, 1 mM EDTA,150 mM NaCl, and varying concentrations of phosphopeptidesubstrates were incubated for 30 min at 24oC and thenstopped by addition of 100 μl BIOMOL GreenTM reagent(ENZO Life Sciences) and transferred to a microplate.Absorbance at 620 nm was measured on a microplate readerand the amount of phosphate released was calculated from a

782 C. Li et al. ■


standard curve generated with sodium phosphate underidentical solution conditions.

Prediction of FgCdc14 substrates

The PatMatch algorithm (Yan et al., 2005) was used tosearch the predicted proteins of F. graminearum PH-1(Cuomo et al., 2007) for consensus recognition sites ofCdc14 (minimal Ser-Pro-X-Lys or Ser-Pro-X-Arg with at leastone additional basic amino acid at the +2, +4, or +5 positions).For each candidate, multiple sequence alignments were gen-erated with orthologs from two additional Fusarium speciesand three additional species of Sordariomycetes from thegenera Magnaporthe, Colletotrichum and Neurospora (andoccasionally Blumeria if an ortholog was not identified fromone of the other genera in BLAST searches). Entries wereremoved from the list of putative FgCdc14 substrates iforthologous proteins were not identified outside the genusFusarium or if the entry did not have at least two sites con-served in at least one of the species outsider the genusFusarium. Enrichment of functional annotations among can-didate substrates was determined using DAVID Bioinformat-ics Resources 6.7 (Huang et al., 2009).

Yeast two-hybrid assays

Protein–protein interactions were assayed with the Match-maker yeast two-hybrid system (Clontech, Mountain View,CA, USA). The FgCDC14 ORF was amplified from first-strand cDNA of PH-1 and cloned into pGBK7 as the baitconstruct. For the FgDBF2, FgWEE1 and FgCDC25 genes,their ORFs were amplified and cloned into pGADT7 as theprey constructs. The resulting bait and prey vectors wereco-transformed into the yeast strain AH109 as described(Zhao et al., 2005). The Leu + and Trp + transformants wereisolated and assayed for growth on SD-Trp-Leu-His mediumand galactosidase activities with filter lift assays as described(Park et al., 2006). The positive and negative controls wereprovided in the Matchmaker Library Construction & Screen-ing Kit (Clontech).

Phosphorylation assays with Cdc2A and Cdc2B

Total proteins were isolated from 12 h germlings grown inliquid YEPD, separated on a 10% SDS-PAGE gel, and trans-ferred to nitrocellulose membranes as described (Brunoet al., 2004). For Western blot analysis, the expression ofCdc2A and Cdc2B was detected with the Cdc2 p34(PSTAIRE) antibody [Santa Cruz Biotechnology, Inc. (Dallas,TX, USA)]. The inhibitory phosphorylation level of Cdc2 pro-teins was detected with the Phospho-Cdc2 (Tyr15) antibody(Cell Signaling Technology, Danvers, MA, USA). The proteinsamples were also detected with anti-histone H3 antibody(Abcam, Cambridge, MA, USA) as a loading reference. Theintensities of western blot bands were quantified withprogram ImageJ (Schindelin et al., 2015). The relative inten-sity of Tyr15 phosphorylation of Cdc2A/2B in differentmutants was estimated in comparison with that of the wildtype (arbitrarily set to 1) after normalization with theCdc2A/2B bands detected with the anti-PSTEIRE antibody.

Acknowledgements

We thank Shulin Cao, Cong Jiang, Tao Yin and Xuli Gao forassistance with yeast two-hybrid assays and DON productionassays. We also thank Drs. Huiquan Liu, Qiaojun Jin, andGuanghui Wang for fruitful discussions. This work was sup-ported by grants from the National Major Project of Breedingfor New Transgenic Organisms (2012ZX08009003), theNature Science Foundation of China (No. 31271989; No.31201464) and US Wheat and Barley Scab Initiative.

References

Ah-Fong, A.M., and Judelson, H.S. (2011) New role forCdc14 phosphatase: localization to basal bodies in theoomycete Phytophthora and its evolutionary coinheritancewith eukaryotic flagella. PLoS ONE 6: e16725.

Bai, G.-H., Desjardins, A., and Plattner, R. (2002)Deoxynivalenol-nonproducing Fusarium graminearumcauses initial infection, but does not cause Disease Spreadin wheat spikes. Mycopathologia 153: 91–98.

Bai, G.H., and Shaner, G. (2004) Management and resist-ance in wheat and barley to Fusarium head blight. AnnuRev Phytopathol 42: 135–161.

Berepiki, A., Lichius, A., and Read, N.D. (2011) Actin organi-zation and dynamics in filamentous fungi. Nat Rev Micro-biol 9: 876–887.

Bi, E., Maddox, P., Lew, D.J., Salmon, E.D., McMillan, J.N.,Yeh, E., and Pringle, J.R. (1998) Involvement of an acto-myosin contractile ring in Saccharomyces cerevisiaecytokinesis. J Cell Biol 142: 1301–1312.

Bloom, J., Cristea, I.M., Procko, A.L., Lubkov, V., Chait, B.T.,Snyder, M., and Cross, F.R. (2011) Global analysis ofCdc14 phosphatase reveals diverse roles in mitotic pro-cesses. J Biol Chem 286: 5434–5445.

Bluhm, B.H., Zhao, X., Flaherty, J.E., Xu, J.R., and Dunkle,L.D. (2007) RAS2 regulates growth and pathogenesis inFusarium graminearum. Mol Plant Microbe Interact 20:627–636.

Bremmer, S.C., Hall, H., Martinez, J.S., Eissler, C.L.,Hinrichsen, T.H., Rossie, S., et al. (2012) Cdc14 phos-phatases preferentially dephosphorylate a subset of cyclin-dependent kinase (Cdk) sites containing phosphoserine. JBiol Chem 287: 1662–1669.

Bruno, K.S., Tenjo, F., Li, L., Hamer, J.E., and Xu, J.R. (2004)Cellular localization and role of kinase activity of PMK1 inMagnaporthe grisea. Eukaryot Cell 3: 1525–1532.

Buss, J.E., and Stull, J.T. (1983) Measurement of chemicalphosphate in proteins. Methods Enzymol 99: 7–14.

Cappellini, R.A., and Peterson, J.L. (1965) Macroconidiumformation in submerged cultures by a non-sporulatingstrain of Gibberella zeae. Mycologia 57: 962–966.

Caracuel-Rios, Z., and Talbot, N.J. (2007) Cellular differen-tiation and host invasion by the rice blast fungus Magna-porthe grisea. Curr Opin Microbiol 10: 339–345.

Catlett, N., Lee, B.-N., Yoder, O., and Turgeon, B.G. (2003)Split-marker recombination for efficient targeted deletion offungal genes. Fungal Genet Newsl 9–11.

Chen, J.-S., Broadus, M.R., McLean, J.R., Feoktistova, A.,Ren, L., and Gould, K.L. (2013) Comprehensive proteomicsanalysis reveals new substrates and regulators of the fission



yeast Clp1/Cdc14 phosphatase. Mol Cell Proteomics 12:1074–1086.

Clement, A., Solnica-Krezel, L., and Gould, K.L. (2012) Func-tional redundancy between Cdc14 phosphatases inzebrafish ciliogenesis. Dev Dyn 241: 1911–1921.

Clemente-Blanco, A., González-Novo, A., Machín, F.,Caballero-Lima, D., Aragón, L., Sánchez, M., et al. (2006)The Cdc14p phosphatase affects late cell-cycle events andmorphogenesis in Candida albicans. J Cell Sci 119: 1130–1143.

Clifford, D.M., Wolfe, B.A., Roberts-Galbraith, R.H.,McDonald, W.H., Yates, J.R., 3rd, and Gould, K.L. (2008)The Clp1/Cdc14 phosphatase contributes to the robust-ness of cytokinesis by association with anillin-related Mid1.J Cell Biol 181: 79–88.

Cueille, N., Salimova, E., Esteban, V., Blanco, M., Moreno,S., Bueno, A., and Simanis, V. (2001) Flp1, a fission yeastorthologue of the S. cerevisiae CDC14 gene, is notrequired for cyclin degradation or rum1p stabilisation at theend of mitosis. J Cell Sci 114: 2649–2664.

Cuomo, C.A., Gueldener, U., Xu, J.R., Trail, F., Turgeon,B.G., Di Pietro, A., et al. (2007) The Fusarium gramine-arum genome reveals a link between localized polymor-phism and pathogen specialization. Science 317: 1400–1402.

De Walle, J.V., Sergent, T., Piront, N., Toussaint, O.,Schneider, Y.J., and Larondelle, Y. (2010) Deoxynivalenolaffects in vitro intestinal epithelial cell barrier integritythrough inhibition of protein synthesis. Toxicol Appl Phar-macol 245: 291–298.

De Wulf, P., Montani, F., and Visintin, R. (2009) Protein phos-phatases take the mitotic stage. Curr Opin Cell Biol 21:806–815.

Ding, S.L., Mehrabi, R., Koten, C., Kang, Z.S., Wei, Y.D.,Seong, K.Y., et al. (2009) Transducin beta-like gene FTL1is essential for pathogenesis in Fusarium graminearum.Eukaryot Cell 8: 867–876.

Drewes, G., Ebneth, A., and Mandelkow, E.M. (1998) MAPs,MARKs and microtubule dynamics. Trends Biochem Sci23: 307–311.

Eissler, C.L., Bremmer, S.C., Martinez, J.S., Parker, L.L.,Charbonneau, H., and Hall, M.C. (2011) A general strategyfor studying multisite protein phosphorylation using label-free selected reaction monitoring mass spectrometry. AnalBiochem 418: 267–275.

Eissler, C.L., Mazon, G., Powers, B.L., Savinov, S.N.,Symington, L.S., and Hall, M.C. (2014) The Cdk/Cdc14module controls activation of the Yen1 holliday junctionresolvase to promote genome stability. Mol Cell 54: 80–93.

Gale, L.R., Ward, T.J., Balmas, V., and Kistler, H.C. (2007)Population subdivision of Fusarium graminearum sensustricto in the upper midwestern United States. Phytopathol-ogy 97: 1434–1439.

Goswami, R.S., and Kistler, H.C. (2004) Heading for disaster:Fusarium graminearum on cereal crops. Mol Plant Pathol5: 515–525.

Gray, C.H., Good, V.M., Tonks, N.K., and Barford, D. (2003)The structure of the cell cycle protein Cdc14 reveals aproline-directed protein phosphatase. EMBO J 22: 3524–3535.

Harris, S.D. (2005) Morphogenesis in germinating Fusariumgraminearum macroconidia. Mycologia 97: 880–887.

Harvey, S.L., and Kellogg, D.R. (2003) Conservation ofmechanisms controlling entry into mitosis: budding yeastWee1 delays entry into mitosis and is required for cell sizecontrol. Curr Biol 13: 264–275.

Heilig, Y., Dettmann, A., Mourino-Perez, R.R., Schmitt, K.,Valerius, O., and Seiler, S. (2014) Proper actin ring forma-tion and septum constriction requires coordinated regula-tion of SIN and MOR pathways through the germinal centrekinase MST-1. PLoS Genet 10: e1004306.

Hou, Z.M., Xue, C.Y., Peng, Y.L., Katan, T., Kistler, H.C., andXu, J.R. (2002) A mitogen-activated protein kinase gene(MGV1) in Fusarium graminearum is required for femalefertility, heterokaryon formation, and plant infection. MolPlant Microbe Interact 15: 1119–1127.

Huang, D.W., Sherman, B.T., and Lempicki, R.A. (2009)Systematic and integrative analysis of large gene listsusing DAVID bioinformatics resources. Nat Protoc 4:44–57.

Jenczmionka, N.J., Maier, F.J., Losch, A.P., and Schafer, W.(2003) Mating, conidiation and pathogenicity of Fusariumgraminearum, the main causal agent of the head-blightdisease of wheat, are regulated by the MAP kinase gpmk1.Curr Genet 43: 87–95.

Kang, Z.S., Buchenauer, H., Huang, L.L., Han, Q.M., andZhang, H.C. (2008) Cytological and immunocytochemicalstudies on responses of wheat spikes of the resistantChinese cv. Sumai 3 and the susceptible cv. Xiaoyan 22 toinfection by Fusarium graminearum. Eur J Plant Pathol120: 383–396.

Kerk, D., Templeton, G., and Moorhead, G.B. (2008) Evolu-tionary radiation pattern of novel protein phosphatasesrevealed by analysis of protein data from the completelysequenced genomes of humans, green algae, and higherplants. Plant Physiol 146: 351–367.

Kitayama, C., Sugimoto, A., and Yamamoto, M. (1997) TypeII myosin heavy chain encoded by the myo2 gene com-poses the contractile ring during cytokinesis in Schizosac-charomyces pombe. J Cell Biol 137: 1309–1319.

Lianga, N., Williams, E.C., Kennedy, E.K., Dore, C., Pilon, S.,Girard, S.L., et al. (2013) A Wee1 checkpoint inhibits ana-phase onset. J Cell Biol 201: 843–862.

Liu, H., Zhang, S., Ma, J., Dai, Y., Li, C., Lyu, X., et al. (2015)Two Cdc2 kinase genes with distinct functions in vegetativeand infectious hyphae in Fusarium graminearum. PLoSPathog 11: e1004913.

Lu, L.X., Domingo-Sananes, M.R., Huzarska, M., Novak, B.,and Gould, K.L. (2012) Multisite phosphoregulation ofCdc25 activity refines the mitotic entrance and exitswitches. Proc Natl Acad Sci USA 109: 9899–9904.

Luo, Y., Zhang, H., Qi, L., Zhang, S., Zhou, X., Zhang, Y., andXu, J.R. (2014) FgKin1 kinase localizes to the septal poreand plays a role in hyphal growth, ascospore germination,pathogenesis, and localization of Tub1 beta-tubulins inFusarium graminearum. New Phytol 204: 943–954.

Markell, S.G., and Francl, L.J. (2003) Fusarium head blightinoculum: species prevalence and Gibberella zeae sporetype. Plant Dis 87: 814–820.

Mitchell, D. (2007) The evolution of eukaryotic cilia and fla-gella as motile and sensory organelles. In Eukaryotic Mem-

784 C. Li et al. ■


branes and Cytoskeleton. Jékely, G. (ed.). New York, NY:Springer, pp. 130–140.

Mocciaro, A., and Schiebel, E. (2010) Cdc14: a highly con-served family of phosphatases with non-conserved func-tions? J Cell Sci 123: 2867–2876.

Mohl, D.A., Huddleston, M.J., Collingwood, T.S., Annan,R.S., and Deshaies, R.J. (2009) Dbf2-Mob1 drives relocali-zation of protein phosphatase Cdc14 to the cytoplasmduring exit from mitosis. J Cell Biol 184: 527–539.

Motegi, F., Mishra, M., Balasubramanian, M.K., and Mabuchi,I. (2004) Myosin-II reorganization during mitosis is con-trolled temporally by its dephosphorylation and spatially byMid1 in fission yeast. J Cell Biol 165: 685–695.

Ovejero, S., Ayala, P., Bueno, A., and Sacristan, M.P. (2012)Human Cdc14A regulates Wee1 stability by counteractingCDK-mediated phosphorylation. Mol Biol Cell 23: 4515–4525.

Park, G., Xue, C., Zhao, X., Kim, Y., Orbach, M., and Xu, J.-R.(2006) Multiple upstream signals converge on the adaptorprotein Mst50 in Magnaporthe grisea. Plant Cell 18: 2822–2835.

Pérez-Martín, J., Castillo-Lluva, S., Sgarlata, C., Flor-Parra,I., Mielnichuk, N., Torreblanca, J., and Carbó, N. (2006)Pathocycles: Ustilago maydis as a model to study therelationships between cell cycle and virulence in patho-genic fungi. Mol Genet Genomics 276: 211–229.

Perez-Martin, J., and Castillo-Lluva, S. (2008) Connectionsbetween polar growth and cell cycle arrest during theinduction of the virulence program in the phytopathogenicfungus Ustilago maydis. Plant Signal Behav 3: 480–481.

Perez-Martin, J., Castillo-Lluva, S., Sgarlata, C., Flor-Parra,I., Mielnichuk, N., Torreblanca, J., and Carbo, N. (2006)Pathocycles: Ustilago maydis as a model to study therelationships between cell cycle and virulence in patho-genic fungi. Mol Genet Genom 276: 211–229.

Proctor, R.H., Hohn, T.M., and McCormick, S.P. (1995)Reduced virulence of Gibberella zeae caused by disruptionof a trichothecene toxin biosynthetic gene. Mol PlantMicrobe Interact 8: 593–601.

Queralt, E., and Uhlmann, F. (2008) Cdk-counteracting phos-phatases unlock mitotic exit. Curr Opin Cell Biol 20: 661–668.

Raspelli, E., Cassani, C., Chiroli, E., and Fraschini, R. (2015)Budding yeast Swe1 is involved in the control of mitoticspindle elongation and is regulated by Cdc14 phosphataseduring mitosis. J Biol Chem 290: 1–12.

Riedl, J., Crevenna, A.H., Kessenbrock, K., Yu, J.H.,Neukirchen, D., Bista, M., et al. (2008) Lifeact: a versatilemarker to visualize F-actin. Nat Methods 5: 605–607.

Saunders, D.G.O., Aves, S.J., and Talbot, N.J. (2010) Cellcycle-mediated regulation of plant infection by the rice blastfungus. Plant Cell 22: 497–507.

Schindelin, J., Rueden, C.T., Hiner, M.C., and Eliceiri, K.W.(2015) The ImageJ ecosystem: an open platform for bio-medical image analysis. Mol Reprod Dev 82: 518–529.

Seo, J.-A., Kim, J.-C., Lee, D.-H., and Lee, Y.-W. (1996)Variation in 8-ketotrichothecenes and zearalenone produc-tion by Fusarium graminearum isolates from corn andbarley in Korea. Mycopathologia 134: 31–37.

Seong, K., Hou, Z.M., Tracy, M., Kistler, H.C., and Xu, J.R.(2005) Random insertional mutagenesis identifies genes

associated with virulence in the wheat scab fungusFusarium graminearum. Phytopathology 95: 744–750.

Shou, W., Seol, J.H., Shevchenko, A., Baskerville, C.,Moazed, D., Chen, Z.S., et al. (1999) Exit from mitosis istriggered by Tem1-dependent release of the protein phos-phatase Cdc14 from nucleolar RENT complex. Cell 97:233–244.

Son, S., and Osmani, S.A. (2009) Analysis of all proteinphosphatase genes in Aspergillus nidulans identifies a newmitotic regulator, Fcp1. Eukaryot Cell 8: 573–585.

Song, B., Li, H.P., Zhang, J.B., Wang, J.H., Gong, A.D., Song,X.S., et al. (2013) Type II myosin gene in Fusariumgraminearum is required for septation, development,mycotoxin biosynthesis and pathogenicity. Fungal GenetBiol 54: 60–70.

Tassan, J.P., and Le Goff, X. (2004) An overview of theKIN1/PAR-1/MARK kinase family. Biol Cell 96: 193–199.

Taylor, G.S., Liu, Y., Baskerville, C., and Charbonneau, H.(1997) The activity of Cdc14p, an oligomeric dual specific-ity protein phosphatase from Saccharomyces cerevisiae, isrequired for cell cycle progression. J Biol Chem 272:24054–24063.

Trautmann, S., Wolfe, B.A., Jorgensen, P., Tyers, M., Gould,K.L., and McCollum, D. (2001) Fission yeast Clp1p phos-phatase regulates G2/M transition and coordination ofcytokinesis with cell cycle progression. Curr Biol 11: 931–940.

Trautmann, S., Rajagopalan, S., and McCollum, D. (2004)The S. pombe Cdc14-like phosphatase Clp1p regulateschromosome biorientation and interacts with Aurorakinase. Dev Cell 7: 755–762.

Urban, M., Mott, E., Farley, T., and Hammond-Kosack, K.(2003) The Fusarium graminearum MAP1 gene is essen-tial for pathogenicity and development of perithecia. MolPlant Pathol 4: 347–359.

Van Thuat, N., Schafer, W., and Bormann, J. (2012) Thestress-activated protein kinase FgOS-2 is a key regulator inthe life cycle of the cereal pathogen Fusarium gramine-arum. Mol Plant Microbe Interact 25: 1142–1156.

Visintin, R., Craig, K., Hwang, E.S., Prinz, S., Tyers, M., andAmon, A. (1998) The phosphatase Cdc14 triggers mitoticexit by reversal of Cdk-dependent phosphorylation. MolCell 2: 709–718.

Visintin, R., Hwang, E.S., and Amon, A. (1999) Cfi1 preventspremature exit from mitosis by anchoring Cdc14 phos-phatase in the nucleolus. Nature 398: 818–823.

Wan, J., Xu, H., and Grunstein, M. (1992) CDC14 of Saccha-romyces cerevisiae. Cloning, sequence analysis, and tran-scription during the cell cycle. J Biol Chem 267: 11274–11280.

Wang, C., Zhang, S., Hou, R., Zhao, Z., Zheng, Q., Xu, Q.,et al. (2011) Functional analysis of the kinome of the wheatscab fungus Fusarium graminearum. PLoS Pathog 7:e1002460.

Wang, J., Liu, J., Hu, Y., Ying, S.-H., and Feng, M.-G. (2013)Cytokinesis-required Cdc14 is a signaling hub of asexualdevelopment and multi-stress tolerance in Beauveriabassiana. Sci Rep 3: doi:10.1038/srep03086.

Wilson, D., and Hube, B. (2010) Hgc1 mediates dynamicCandida albicans-endothelium adhesion events during cir-culation. Eukaryot Cell 9: 278–287.



Wolfe, B.A., and Gould, K.L. (2004) Fission yeast Clp1pphosphatase affects G2/M transition and mitotic exitthrough Cdc25p inactivation. EMBO J 23: 919–929.

Wolfe, B.A., McDonald, W.H., Yates, J.R., 3rd, and Gould,K.L. (2006) Phospho-regulation of the Cdc14/Clp1 phos-phatase delays late mitotic events in S. pombe. Dev Cell11: 423–430.

Wood, J.S., and Hartwell, L.H. (1982) A dependent pathwayof gene functions leading to chromosome segregation inSaccharomyces cerevisiae. J Cell Biol 94: 718–726.

Yan, T., Yoo, D., Berardini, T.Z., Mueller, L.A., Weems, D.C.,Weng, S., et al. (2005) PatMatch: a program for findingpatterns in peptide and nucleotide sequences. NucleicAcids Res 33: W262–W266.

Zhao, X.H., Kim, Y., Park, G., and Xu, J.R. (2005) A mitogen-activated protein kinase cascade regulating infection-related morphogenesis in Magnaporthe grisea. Plant Cell17: 1317–1329.

Zheng, D., Zhang, S., Zhou, X., Wang, C., Xiang, P.,Zheng, Q., and Xu, J.-R. (2012) The FgHOG1 pathwayregulates hyphal growth, stress responses, and plant

infection in Fusarium graminearum. PLoS ONE 7:e49495.

Zheng, X.D., Wang, Y.M., and Wang, Y. (2004) Hgc1, a novelhypha-specific G1 cyclin-related protein regulates Candidaalbicans hyphal morphogenesis. EMBO J 23: 1845–1856.

Zhou, X., Li, G., and Xu, J.R. (2011) Efficient approaches forgenerating GFP fusion and epitope-tagging constructs infilamentous fungi. Methods Mol Biol 722: 199–212.

Zhou, X., Zhang, H., Li, G., Shaw, B., and Xu, J.-R. (2012)The cyclase-associated protein Cap1 is important forproper regulation of infection-related morphogenesis inMagnaporthe oryzae. PLoS Pathog 8: e1002911.

Zhou, X.Y., Heyer, C., Choi, Y.E., Mehrabi, R., and Xu, J.R.(2010) The CID1 cyclin C-like gene is important for plantinfection in Fusarium graminearum. Fungal Genet Biol 47:143–151.

Supporting information

Additional supporting information may be found in the onlineversion of this article at the publisher’s web-site.

786 C. Li et al. ■


fgcdc14 regulates cytokinesis, morphogenesis, and ...xulab.org/files/pdf/lic-2015-mm.pdf ·...

Documents