Summary. The small GTPase CDC42 is ubiquitously expressed in eu-karyotes, where it participates in the regulation of the cytoskeleton and awide range of cellular processes, including cytokinesis, gene expression,cell cycle progression, apoptosis, and tumorigenesis. As very little isknown on the molecular level about mycorrhizal morphogenesis and de-velopment and these events depend on a tightly regulated reorganisationof the cytoskeleton network in filamentous fungi, we focused on themolecular characterisation of the cdc42 gene in Tuber borchii Vittad., anascomycetous hypogeous fungus forming ectomycorrhizae. The entiregene was isolated from a T. borchii cDNA library and Southern blotanalyses showed that only one copy of cdc42 is present in the T. borchiigenome. The predicted amino acid sequence is very similar to those ofother known small GTPases and the similar domain structures suggest asimilar function. Real-time PCR analyses revealed an increased expres-sion of Tbcdc42 during the phase preparative to the instauration of sym-biosis, in particular after stimulation with root exudate extracts.Immunolocalisation experiments revealed an accumulation of CDC42 inthe apical tips of the growing hyphae. When a constitutively active Tbcdc42mutant was expressed in Saccharomyces cerevisiae, morphological changestypical of pseudohyphal growth were observed. Our results suggest a fun-damental role of CDC42 in cell polarity development in T. borchii.

Keywords: Tuber borchii; Apical growth; GTPase CDC42; Immunolo-calisation.

Introduction

Small GTPases of the Rho family function as molecularswitches that tightly regulate diverse cellular functions.They are ubiquitously expressed in eukaryotes and aregrouped in three major subfamilies: Rho, Rac, and CDC42.The Rho GTPases were thought to be primarily involved

in the regulation of cytoskeleton organisation (Hall 1998)required for membrane trafficking, motility, adhesion andmorphogenesis (Kaibuchi et al. 1999, Rivero and Somesh2002). More recently, it has been shown that CDC42, Rac,and Rho are involved in the regulation of cellular processesas diverse as microtubule organisation, cytokinesis, geneexpression, cell cycle progression, apoptosis, and tumori-genesis (Johnson 1999, Jaffe and Hall 2005).

The multiple biological functions of the Rho-relatedproteins are mediated through a highly regulated GTP-binding/GTPase cycle. Three different classes of proteinsare required for the regulation: (1) guanine nucleotide ex-change factors (GEFs), which stimulate the GTP–GDPexchange reaction; (2) GTPase-activating proteins (GAPs),which stimulate the GTP-hydrolysing reaction; and (3)guanine nucleotide dissociation inhibitors (GDIs), whichantagonize the actions of GEFs and GAPs and regulatethe subcellular localisation and the cycling of GTPases be-tween membrane and cytosol (Hoffman et al. 2000, Riveroet al. 2002). The active GTP-bound GTPases interact witha myriad of effectors that relay upstream signals, inducinga number of downstream events, including rearrangementsof the actin cytoskeleton network and protein kinase-de-pendent induction of transcription (Hall 1998, Jaffe andHall 2005).

CDC42 was identified in Saccharomyces cerevisiae bycomplementation of a temperature-sensitive cdc42 mutant(Johnson and Pringle 1990, Johnson 1999). The cdc42 mu-tation blocked bud formation, and fluorescence microscopywith rhodamine-conjugated phalloidin highlighted the dis-ruption of the actin cytoskeleton. These peculiarities showed

Molecular characterisation of the small GTPase CDC42 in the ectomycorrhizal fungus Tuber borchii Vittad.

M. Menotta1, A. Amicucci1,*, G. Basili1, F. Rivero2, E. Polidori1, D. Sisti3, and V. Stocchi1

1 Istituto di Chimica Biologica “G. Fornaini’’, Università degli Studi di Urbino, Urbino2 Zentrum für Biochemie, Medizinische Fakultät, Universität zu Köln, Cologne3 Istituto e Orto Botanico “P. Scaramella’’, Università degli Studi di Urbino, Urbino

Received April 25, 2006; accepted November 23, 2006; published online August 30, 2007© Springer-Verlag 2007

* Correspondence and reprints: Istituto di Chimica Biologica “G. Fornaini’’,Università degli Studi di Urbino, Via Saffi 2, 61029 Urbino, Italy.E-mail: a.amicucci@uniurb.it

Protoplasma (2007) 231: 227–237DOI 10.1007/s00709-007-0254-y PROTOPLASMA

Printed in Austria

that CDC42 is involved in the organisation of the actin cy-toskeleton, necessary for polarised cell growth and budformation (Drubin 1991). In mammalian cells, a specificextracellular signal binds to a G-protein-coupled receptorand activates CDC42, which then directs the organisationof actin to form filopodia and peripheral actin microspikes(Schmidt and Hall 1998). Moreover, several recent studieshave considered interesting new roles for CDC42 in intra-cellular and membrane trafficking and in the regulation ofmammalian cell growth (Lowe and Kreis 1998, Eitzen et al.2001, Müller et al. 2001, Zhang et al. 2001, Nickel et al.2002, Cerione 2004).

Very little is known about the molecular mechanismsresponsible for the establishment of cell polarity, the main-tenance of hyphal growth, and mycorrhizal formation infilamentous fungi. As these events depend on a tightlyregulated reorganisation of the cytoskeleton network, wefocused our research in the present study on the small GT-Pase CDC42 in Tuber borchii Vittad., a hypogeous fila-mentous fungus that forms ectomycorrhizae whenassociated with a host plant and develops fruitbodies dur-ing its life cycle. Interest in T. borchii centres on its agro-forestry and commercial importance and on the availabilityof an in vitro system, T. borchii–Tilia americana L., thatprovides suitable starting material for molecular investiga-tions (Sisti et al. 1998, Zambonelli et al. 2002, Hall et al.2003, Amicucci et al. 2005).

In this investigation, the cdc42 gene of T. borchii wasisolated and molecularly characterised. The expression ofTbcdc42 was also analysed in the vegetative hyphae ofT. borchii and in an in vitro system in which the sym-bionts Tilia americana and T. borchii interact without di-rect contact. In addition, to determine whether T. borchiiCDC42 GTPase plays a key role in the tightly modulatedsignalling pathway upstream of cytoskeleton organisation,as it is reported to do in S. cerevisiae, the distribution ofCDC42 in the fungal hyphae of T. borchii was confirmedby indirect immunofluorescence microscopy. To gain in-sight into the role of CDC42, we expressed a CDC42 Q63Lmutant that is constitutively GTP-bound (activated) inS. cerevisiae.

Material and methods

Fungal strains and culture conditions

The Tuber borchii culture consisted of vegetative mycelia of T. borchii(strain 10 RA) grown on nylon membranes laid on solid MS/2 medium,pH 6.5, with 10 g of glucose per litre as a carbon source (Murashige andSkoog 1962). The culture was grown for 30 days at 24 °C and a photope-riod of 16 h light provided by cool white fluorescent lamps and 8 h dark(driver sample). The tester sample was prepared as follows. Tuber borchii

mycelia were grown under the above conditions but in the presence ofthe host plant T. americana, from which they were separated by a nylonmembrane. In order to obtain T. americana root exudates, the plantletswere inoculated alone into a mycorrhization medium in glass tubes con-taining peat moss (Molina 1979) under the same conditions describedabove, as reported by Sisti et al. (1998). After 30 days, the peat mossmixture was drawn from the tubes in sterile conditions and the liquidmedium was purified and filtered through Millipore 22 �m pore size fil-ters. Air-stimulated tester (TSA) mycelia samples were obtained bytreating T. borchii mycelia grown as described above with 200 �l ofT. americana root exudates after 25 days of cultivation. The stimulationwas performed by putting a glass capillary (volume, 20 �l) containingthe purified root exudates onto the nylon membrane for 5 days.

The S. cerevisiae INVsc1 strain (MATa his3-�1 leu2 trp1-289 ura3-52/MAT� leu2 trp1-289 ura3-52) was supplied by Invitrogen and main-tained and cultivated by standard procedures.

DNA and RNA isolation

Genomic DNA was isolated from 1-month-old cultures of T. borchiimycelia following the protocol described by Erland et al. (1994). TotalRNA was extracted from driver, tester, and TSA mycelial samples by aQiagen RNeasy kit according to the manufacturer’s instructions. A DNase(Ambion) digestion step was performed before all subsequent reactions.

Cloning of the Tbcdc42 gene

A 1 �g aliquot of DNase-treated (DNA-free, Ambion) total RNA fromfree-living mycelia was denatured at 70 °C for 2 min and then reversetranscribed in a 10 �l reaction mixture using random hexamers asprimers and the Moloney murine leukaemia virus PowerScript reversetranscriptase from Clontech (Palo Alto, Calif., U.S.A.). Two pairs of de-generate primers were deduced from the most conserved regions of amultiple alignment of fungal CDC42 proteins. The primers were synthe-sised by M-Medical Genenco (Florence, Italy): CDC42-1For, 5�-GTNTTY GAY AAY TAY GC-3�, and CDC42-4Rev, 5�-DAT NGC YTCRTC RAA NAC-3�; and CDC42-2For, 5�-GGN CAR GAR GAY TAYGA-3�, and CDC42-3Rev, 5�-RCA YTC NAC RTA YTT NAC-3�. Thedegenerate PCR primers were used in PCR experiments using T. borchiicDNA.

The 50 �l PCR reaction contained 1.5 mM MgCl2, 0.2 mM (each) de-oxynucleoside triphosphates, 100 pmol of each primer CDC42-1For andCDC42-4Rev, 200 ng of cDNA, 1 U of Taq polymerase and 1� PCRbuffer (Qiagen). The PCR was carried out by initial denaturation at 94 °Cfor 5 min, followed by 30 cycles of 94 °C for 1 min, 40 °C for 1 min, and72 °C for 1 min. A nested PCR was carried out with the CDC42-2Forand CDC42-3Rev internal primers on 1 �l of the first PCR product di-luted 1/50, using the same conditions as described above. A single PCRfragment (245 bp) was obtained, inserted into the pGEM vector II(Promega) and subsequently cloned using Escherichia coli XL1-Blue.After verification by sequencing, the cloned fragment was used as a ho-mologous probe for the screening of 3 � 105 �ZAP (Stratagene, La Jolla,Calif., U.S.A.) clones of a T. borchii mycelium cDNA library (Polidoriet al. 2002). Library screening, subcloning, and routine procedures wereperformed by standard protocols (Sambrook et al. 1989). DNA sequenceswere analysed by SeqPup version 0.9e running on Sun Java VM version1.4.1_02. Database searches of the DNA and protein sequences wereperformed with the BLAST programs (NCBI). Two pairs of primers com-plementary to the 3� and 5� ends of Tbcdc42 cDNA were used to amplifythe Tbcdc42 genomic sequence.

Southern analysis

Genomic DNA samples for gel blot analysis (10 �g each) were digestedwith the EcoRI, ScaI, and BamHI enzymes and then electrophoresed on

228 M. Menotta et al.: CDC42 and apical growth in Tuber borchii

a 0.8% agarose gel. The DNA samples were then blotted onto positivelycharged Hybond N� nylon membranes (Amersham BioScience) accord-ing to the manufacturer’s instructions, and hybridised at 65 °C overnightin phosphate buffer (Sambrook et al. 1989) with the 245 bp Tbcdc42 frag-ment, which was labelled with 32P using the RediPrime labelling kit(Amersham Life Science). The final posthybridisation wash was carriedout in 15 mM NaCl plus 1.5 mM trisodium citrate (0.1� SSC) and0–1% sodium dodecyl sulfate at 65 °C.

cDNA synthesis and quantitative real-time PCR

cDNA from driver, tester, and TSA mycelial samples were obtained asdescribed above. The 18S rRNA gene from T. borchii was used as a refer-ence (18S RT F, 5�-TGGTCCGGTCGGATCTT-3�; 18S RT R, 5�-CATTACGGCGGTCCTAGAAA-3�). Specific primers for Tbcdc42 (CDC RTF1, 5�-TGACCCAAAAGGGTCTCAAG-3�; CDC RT F2, 5�-CAGCCGTCAAGTATGTCGAA-3�; CDC RT R1, 5�-TGTAGTGTAGGGGGCTCCAG-3�) were designed to amplify under the same cycling conditions(95 °C for 10 min, followed by 50 cycles of 95 °C for 30 s and 60 °C for30 s), generating a product of 65 bp.

The PCR was performed in a Bio-Rad iCycler iQ multicolor real-timePCR detection system. Each 25 �l reaction consisted of 1 �l of 1/30 di-luted cDNA, 12.4 �l of 2� Quantitect SYBR Green PCR kit and 300 nM(each) primers. The specificity of the amplification products was confirmedby examining thermal denaturation plots and by sample separation in a2% DNA agarose gel. The amount of the target transcript was related tothat of the reference gene by the ��Ct method as described by Winer et al.(1999) (�Ct is the difference of the average Cttbcdc42 minus the averageCttb18S). Each sample was tested in triplicate, and the Kruskal–Wallis testwas applied to samples obtained from at least four independent experi-ments. Results were considered significant if P � 0.05. In the control ex-periments, we characterised the performance of the specific primer assaysover a range of tester template concentrations, from 1 : 5 to 1 : 500 dilu-tions of the starting cDNA. The PCR reaction efficiency was 100.2% forthe inner standard 18S (18S RT F–18S RT R primers) and 99% for theCDC RT F2–CDC RT R1 primers. From the dissociation curves ofthe amplified products, it was possible to verify the specificity of theprimers being used and the absence of primer dimers that could nega-tively affect the quantification processes performed by the SYBRGreen method.

Plasmid construction and yeast transformation

The S. cerevisiae expression plasmid pYES2 (Invitrogen Life Technol-ogy) containing the URA3 auxotrophic marker was used to expressTbcdc42 under the inducible promoter GAL1. Tbcdc42 cDNA was am-plified by PCR with PfuUltra polymerase (Stratagene) according to themanufacturer’s instructions. Primers were designed to include EcoRI andNotI restriction sites at the 5� and 3� ends, respectively. The PCR productwas first cloned into the pGEM T/A cloning vector, and after digestionwith EcoRI/NotI, Tbcdc42 was directionally inserted into pYES2.

The constitutively active mutant Q63L (Johnson 1999) was obtainedby site-directed mutagenesis (Quikchange, Stratagene) of the pYES2-Tbcdc42 vector using the primers 5�-GTTGTTTGATACCGCCGGGCTAGAAGATTACGATCGACTT-3� and 5�-AAGTCGATCGTAATCTTCTAGCCCGGCGGTATCAAACAAC-3�.

The chimeric TbCDC42 proteins fused with green-fluorescent protein,GFP-TbCDC42 and GFP-TbCDC42 Q63L, were obtained by direction-ally cloning the GFP cDNA fragment into the BamHI/EcoRI-digestedpYES2-Tbcdc42 vector.

INVsc1 yeast cells were grown to an OD600 of 0.8 in YPD medium(Sigma Aldrich Co) at 30 °C and then transformed with pYES2-Tbcdc42,pYES2-Tbcdc42-Q63L, pYES2-GFP-Tbcdc42, or pYES2-GFP-Tbcdc42-Q63L by chemical transformation (Sigma Aldrich Co) according to themanufacturer’s instructions. Transformants were grown on glucose or

galactose minimal medium lacking uracil at 30 °C. The negative INVsc1control carrying the empty vector pYES2 was prepared in the same way.

Western blotting

Tuber borchii total proteins were extracted by mechanical disruption ofmycelia in the presence of a lysis buffer provided with the PhosphoPro-tein purification kit (Qiagen) according to the manufacturer’s instructions.The protein concentration of the lysates was determined spectropho-tometrically. An equal amount of protein (15 �g) from every sample wasboiled in Laemmli buffer, separated by sodium dodecyl sulfate-polyacryl-amide gel electrophoresis, and then transferred to nitrocellulose mem-branes. The membranes were incubated with yeast CDC42 antibody(1:500) (Santa Cruz) overnight at 4 °C. An anti-rabbit peroxidase-conju-gated antibody was employed at a 1 : 5000 dilution for 1 h at room tem-perature. Immunoreactive bands were visualised by enhanced chemilu-minescence (Amersham Pharmacia Biotech) following the proceduresrecommended by the manufacturer.

In vitro actin polymerisation assay

The pYES2-GFP-Tbcdc42 and pYES2-GFP-Tbcdc42-Q63L were di-gested with HindIII and NotI and subsequently blunted and ligated into aHindIII- and NotI-digested p3xFLAG-CMV-14 vector (Sigma-Aldrich,St. Louis, Mo., U.S.A.) containing the human cytomegalovirus promoterregion. For expression in the lymphoma cell line U937, 20 �g of plasmidwere used to transiently transfect 3 � 106 cells in a cuvette (electrodegap, 4 mm) by electroporation using the Gene Pulser II (Bio-Rad), whichwas set to 300 V, 950 �F. The cells were cultivated and maintained bystandard procedures. After 24 h, cells were washed and lysed by sonica-tion. The lysate was incubated on a microscope slide cover glass with0.1 mM GTPS for 10 min at room temperature and subsequently fixedfor 30 min in 3.7% formaldehyde solution containing 0.4 �M of tetra-methylrhodamine isothiocyanate–phalloidin. After washing with phos-phate-buffered saline, the cover glass was mounted and the lysate wasobserved under a Leica DMLB fluorescence microscope equipped with aDC300F charge-coupled-device camera.

Indirect immunofluorescence microscopy

Immunocytochemical staining of T. borchii vegetative hyphae was per-formed by the methods developed by Raudaskoski et al. (1991) and Gorferet al. (2001) with slight modifications. We used the yeast CDC42 anti-body as previously described for Western blotting. Immunoreactive hy-phae were visualised with a secondary fluorescein-conjugated anti-rabbitantibody. The glass-mounted samples were analysed with a fluorescencemicroscope as described above.

Results

Cloning of the Tbcdc42 gene

Two pairs of degenerate primers based on conserved regionsof fungal CDC42 proteins were used in PCR to amplify a245 bp fragment from T. borchii cDNA. This product wasused to screen a cDNA library of the T. borchii vegetativemycelium and 7 positive clones were isolated. The longestclone, 1748 bp in length, was entirely sequenced and shownto contain an open reading frame of 582 bp encoding aprotein with an estimated mass of 22055 Da (Fig. 1). Thegene, named Tbcdc42, showed the highest homology with

M. Menotta et al.: CDC42 and apical growth in Tuber borchii 229

cdc42 of Schizosaccharomyces pombe. A 5� untranslatedregion (5�-UTR) of 256 bp and a 3�-UTR of 912 bp arepresent.

In Southern analysis, the Tbcdc42 DNA probe detectedonly one band in each digest, suggesting that Tbcdc42 is asingle-copy gene (Fig. 2).

The whole Tbcdc42 genomic sequence was subsequentlyobtained with two pairs of primers designed to bind to the3� and 5� ends of Tbcdc42 cDNA and with T. borchii ge-nomic DNA from mycelium as a target. The comparisonbetween the genomic and cDNA sequences allowed us todetect three introns in the coding region with lengths of

230 M. Menotta et al.: CDC42 and apical growth in Tuber borchii

Fig. 1. Nucleotide and deduced amino acidsequences of the Tbcdc42 gene (GenBankaccession number DQ402135). Deducedamino acid sequences are indicated abovethe nucleotide sequences. Introns are inlower case; introns inside the 3�- and 5�-UTR regions are marked by dash underlines.The stop codon is marked by a star. Up-stream of the ATG starting codon is a 3adenine (underlined), in agreement with theKozak sequence

72, 115, and 64 bp. In addition, an intron was detected inboth the 5�- and 3�-UTR regions. They were 325 and62 bp long, respectively. All the introns identified showed5� and 3� consensus splicing sites typical of filamentousfungi (Gurr et al. 1987) (Fig. 1). The analysis of codon us-age showed that 55 of the 61 sense codons for the 20amino acids are used. Furthermore, the codon AGN isused for arginine and serine, different from mycorrhizalfungi. UAA is used as a stop codon, as it is in mycorrhizalfungi (Gurr et al. 1987). In the sequence upstream fromthe ATG starting codon, we detected a consensus sequencecharacterised by the presence of an adenine in position 3(3 nucleotides before the ATG codon) corresponding to theKozak sequence (Fig. 1).

Comparison of the amino acid sequence with those ofother fungal species and representative animal CDC42 pro-teins revealed the five typical functional domains of thisprotein family associated with binding and hydrolysis ofguanine nucleotides (Fig. 3). These domains are highlyhomologous with those in mammals and yeasts. The de-

M. Menotta et al.: CDC42 and apical growth in Tuber borchii 231

Fig. 2. Southern blot analysis using a Tbcdc42-specific probe. Tuberborchii genomic DNA was digested with BamHI, ScaI and EcoRI re-striction enzymes. The blot was probed with a 245 bp long radioactivelylabelled probe

Fig. 3. Multiple alignment of CDC42 proteins from T. borchii and representative species. Sequences were aligned with Clustal W. Identical residuesare shown as dots. Secondary structural elements are indicated on top of the aligned sequences: the conserved domains that form the � helices, �

sheets, and the domain involved in the interaction with Rho-GDI. Lines above the sequences indicate amino acids composing the five GTP-bindinghydrolysis domains (I–V) and the Rho insert region (Rho). The three functional amino acids of the CDC42 family, Thr37, Gln63, and Cys189, aremarked with arrows. Proteins were from the following species (NCBI Entrez Protein accession number[s]): TbCdc42, Tuber borchii; AnCdc42, As-pergillus nidulans (AAF24514); ScCdc42 and ScRho1, Saccharomyces cerevisiae (AAB67416 and P06780); DmCdc42 and DmRho1, Drosophilamelanogaster (AAD43790 and AAF01186); HsCdc42 and HsRhoA, Homo sapiens (AAT70721 and AAV38673)

duced amino acid sequence of TbCDC42 also contains thetypical carboxyl-terminal consensus sequence, CAAX (C,cysteine; A, any aliphatic amino acid; and X, any aminoacid), and a lysine-rich polybasic region upstream of theCAAX domain important for intracellular localisation.The residue Thr37, which is involved in the coordinationof Mg2� cations in the interaction with Rho-GDI in com-petition with GEF, is present inside the typical domain I(switch I), and the residue Gln63, involved in GTP hy-drolysis, is present in switch II.

Expression analyses by real-time PCR

In order to evaluate Tbcdc42 expression in the vegetativemycelium in the absence (driver) or presence (tester) of thesymbiotic host or in the presence of root exudates (TSA),we set up a quantitative real-time PCR assay for the geneunder study. We designed a suitable primer pair with ahigh melting temperature, including a splice junction toavoid genomic amplification, hence nonspecific productswere not generated. The average Ct values for Tbcdc42were normalised against average Ct values for the 18SrRNA. The expression differences in the three tissueswere extrapolated using the �CT average of the driver asa calibrator (Fig. 4). In the presence of the plant (testersamples), the fungus showed a twofold increase in theamount of expressed Tbcdc42, while the stimulated sam-ples (TSA) showed an expression fivefold higher than thatof the control samples (driver). The differences among the�CT medians of the analysed samples were statisticallysignificant (Kruskal–Wallis test).

Actin polymerisation

The role of TbCDC42 in the dynamic reorganisation ofthe actin cytoskeleton was tested by an in vitro actin poly-

merisation reaction triggered by GTPS, using the lym-phoma cell line U937 as a heterologous expression system.GTPS is a nonhydrolysable nucleotide analogue. Uponbinding to GTPS, TbCDC42 remains blocked in its active

232 M. Menotta et al.: CDC42 and apical growth in Tuber borchii

Fig. 4. Real-time PCR quantification of Tbcdc42 in tester and TSA sam-ples compared with driver mycelia. The ��CT method was used as de-scribed in the Material and methods section

Fig. 5 A–C. Morphology and filamentous-actin distribution obtained byin vitro actin polymerisation visualised by GFP-Tbcdc42. GFP-Tbcdc42cDNA was cloned into p3xFLAG-CMV-14, transfected and expressed ina U937 cell line. A Actin filaments are stained with tetramethylrho-damine isothiocyanate–phalloidin. B GFP-TbCDC42 signal. C Imagesare overlapped to show the localisation of TbCDC42 (green) along thecell wall (red, actin)

form, activating the proteins involved in the actin polymeri-sation processes. The fluorescence microscopic observa-tion of a cellular lysate which was not cleared of cell debrisallowed us to identify cell membrane fragments rich inGFP-fused TbCDC42 near filaments of polymerised actin,as visualised with tetramethylrhodamine isothiocyanate–phalloidin (Fig. 5). The verification of this reaction invitro suggests that TbCDC42 is a GTPase involved in theassembly processes of the T. borchii cytoskeleton.

Mutagenesis and TbCDC42 expression in yeast cells

To confirm the role of TbCDC42 in cytoskeleton reorgani-sation, constitutively active TbCDC42 (Q63L) was ex-pressed in S. cerevisiae. Yeast cells were chosen for theseexperiments due to the fact that T. borchii, like other my-corrhizal fungi, has so far proved difficult to transform.The inserted mutation Q63L substitutes glutamine 63 witha leucine in the TbCDC42 GTPase domain, making theprotein constitutively active, i.e., unable to hydrolyseGTP, even in the presence of GAPs. Saccharomyces cere-visiae cells transformed with pYES2-Tbcdc42-Q63L

showed phenotypes different to those transformed withpYES2-Tbcdc42 (Fig. 6). The expression of the wild-typeTbCDC42 in S. cerevisiae did not change the cellular phe-notype of the cells (Fig. 6A, B), which maintained theirtypical shape: they were small and roundish, and someshowed the beginning of bud formation, an event that oc-curs in the phase before cell division (Etienne-Manneville2004). In contrast, the hyperactive mutated protein inducedstrong morphological modifications in the yeast cells(Fig. 6C, D). These cells became more voluminous, sel-domly had elongated buds, and showed stronger adhesionwith one another.

Immunolocalisation of TbCDC42 in vivo

To establish the specificity of a polyclonal anti-CDC42antibody for further use in immunolocalisation experi-ments, total protein extracts from driver, tester, and TSAT. borchii mycelial samples were analysed by Westernblotting. The immunoreactive band showed a single frag-ment with the correct estimated mass (Fig. 7A), confirm-ing the results of the expression analyses.

M. Menotta et al.: CDC42 and apical growth in Tuber borchii 233

Fig. 6 A–D. Expression of Tbcdc42 in S. cerevisiae cells. A Wild-type cells. B Morphology of pYES2-Tbcdc42-transformed wild-type cells. C and DMorphology of pYES2-Tbcdc42-Q63L-transformed cells. C Elongated budding sites (marked by arrow 1) and cells showing pseudohyphal growth(arrows 2) can be seen. D Pseudohyphal growth and cell aggregates (arrow 3) are visible

The distribution pattern of TbCDC42 in T. borchii hy-phae was observed by indirect immunofluorescence mi-croscopic investigation using the anti-CDC42 polyclonalantibody (Fig. 7B). Fluorescence was observed mainly inthe apical tip region of vegetative hyphae, at subapicalbranching sites, and also in the proximity of sites of septaformation.

Below the tip region, the signal was more diffuse. Incontrol samples, where the hyphae were only labelled withfluorescein-conjugated secondary antibodies, no strong lo-calised signal was observed in the hyphal tips and only aweak fluorescence in the hyphae, with some diffuse fluo-rescence seen in the subapical part of the hyphae labelledwith the antibody against CDC42.

Discussion

Very little is known about polarised tip growth in fila-mentous mycorrhizal fungi and about the molecularmechanisms at work in cytoskeleton reorganisation lead-

ing to morphological changes in cells and to motility ofthe presymbiotic and symbiotic hyphae. Polarised cellgrowth is an important aspect of the development ofmulticellular organisms. It requires the selection of spe-cific starting sites where a particular reorganisation ofthe actin cytoskeleton is localised. One of the main mod-ulators of cytoskeleton reorganisation during polarisedgrowth is the Rho GTPase CDC42. CDC42, like otherRho-like proteins, has a key role in signal transductionpathways that regulate gene expression and cell growth(Hall 1998, Hoffman et al. 2000). It has been demon-strated to be essential for pseudohyphal growth in S.cerevisiae (Mosch et al. 1996, 2001; Peter et al. 1996;Palecek et al. 2002).

In the pathogenic fungus Candida albicans, CDC42 reg-ulates the polar growth in response to hyphal growth in-duction in infection processes (Hazan et al. 2002).

The present study reports the first biochemical charac-terisation of CDC42 in T. borchii. To evaluate cdc42 regu-lation in T. borchii at the molecular level, the Tbcdc42gene was cloned and characterised. On the basis of thepredicted amino acid sequence and the virtual three-di-mensional structure, it was possible to highlight the char-acteristic domains of this protein. The results showedslight amino acid differences from other CDC42 proteinsthat do not seem to cause any functional changes. Thepredicted three-dimensional structure shows the presenceof active sites matching those of CDC42 in other eukary-otes, mainly fungi, present in databases. These data led usto hypothesise that CDC42 may have the same functionsin T. borchii that it has in other organisms (Johnson 1999,Cerione 2004), including fungi (Harris and Momany 2004,Boyce et al. 2005, Weber 2005, Mahlert et al. 2006).

The expression analyses showed that the fungus ex-pressed twice as much Tbcdc42 when the host plant waspresent and 5 times more when root exudates were addedto the medium. These data are in agreement with evidencethat the presence of the host plant, hence of root exudates,induces an increase in mycelium branching in the earlyphases of symbiosis instauration. In fact, as described byMenotta et al. (2004b), the vegetative hypha tends to branchin the presence of the host plant. Each branching eventgives rise to two apical tips and this is reflected in atwofold expression of Tbcdc42. Furthermore, the experi-mental evidence that the application of root exudates byair diffusion provokes a rapid increase in Tbcdc42 geneexpression is intriguing. We can hypothesise that com-pounds with low molecular weight previously found in themycorrhization medium (Menotta et al. 2004a) may actu-ally modulate gene expression in T. borchii. The over-

234 M. Menotta et al.: CDC42 and apical growth in Tuber borchii

Fig. 7. A Immunoblot analysis of TbCDC42 using polyclonal anti-CDC42 antibody. The anti-CDC42 antibody identified a single 21 kDaband in protein extracted from T. borchii vegetative hyphae (D), while astronger band was detected in hyphae in the presence of the plant (T) andin the TSA samples. Equal loading of protein among lanes was con-firmed with an actin antibody. B Localisation of TbCDC42 in T. borchiivegetative hyphae by indirect immunofluorescence microscopy with thepolyclonal antibody against CDC42. A strong fluorescence at the hyphaltips, the subapical branch originating sites (arrow 1) and at the sites ofseptum formation (arrow 2), and a diffuse fluorescence along the subapi-cal part of the hypha can be seen. Left panels, hyphae in transmittedlight; right panels, hyphae in fluorescence

expression of Tbcdc42 in tester and, in particular, in TSAsamples is consistent with what has been observed in theearly phase of mycorrhizal instauration, when the apicalpolarised growth increases, causing the mycelium to“walk” closer to and contact the plant roots. These resultsare also supported by the previous detection of differen-tially expressed genes involved in vesicular transport andcell wall construction (syntaxin-binding protein, asparticprotease, centractin-like protein) in the presymbiotic phase(Menotta et al. 2004b). In fact, it has been reported thatthe hyphal branching process is physiologically equivalentto cell proliferation in unicellular organisms like yeastcells (Buee et al. 2000), and every new hyphal tip is like anew life unit in which most of the cell machinery is found.

Moreover, the immunodetection of TbCDC42 in T.borchii vegetative hyphae allowed us to observe the pres-ence of TbCDC42 mainly at the apical level. These dataare consistent with the role of TbCDC42 in the promotionof apical growth in filamentous fungi and confirms its rolein the cytoskeleton reorganisation responsible for polarisedapical growth, as already shown by expression analyses.In addition, the presence of TbCDC42 in the branchingsites demonstrates its important role in these events. Infact, morphological analyses (Menotta et al. 2004b) haveshown that, to promote branching, T. borchii doubles theamount of TbCDC42 to ensure the formation and devel-opment of a double apex number. The distribution ofTbCDC42 in sites of septum formation is consistent withits probable role in actin ring formation, which is respon-sible for this event (Salo et al. 1989, Raudaskoski et al.1991, Merla and Johnson 2000), as supported by localisa-tion experiments with Suillus bovinus (Gorfer et al. 2001).

We also performed studies to gain insight into the func-tion of CDC42 in T. borchii. The involvement of TbCDC42in cytoskeleton assembly processes in T. borchii was shownby in vitro actin polymerisation experiments. Moreover,S. cerevisiae cells were transformed with Tbcdc42 andvery surprising results were obtained. Firstly, the correctexpression of TbCDC42 in yeast cells showed that CDC42is able to interact successfully with the signalling path-ways and cytoskeleton organisation engine in yeast cells.Very interestingly, the expression of the constitutively ac-tive TbCDC42 Q63L in yeast cells switched on a series ofevident morphological modifications. Compared to wild-type yeast cells and to cells transformed with wild-typeTbCDC42, S. cerevisiae cells that expressed the mutatedhyperactive gene acquired an elongated shape, and giantcells and cell aggregations were present. These peculiarphenotypes led us to conclude that TbCDC42 has a funda-mental role in hyphal morphology.

Since CDC42 is commonly distributed along the innerside of the membrane (Ziman et al. 1991, Richman et al.2002), the volume increase might be provoked by the cen-tripetal push of actin filaments towards the membrane. Insome cases, a localised stochastic aggregation of TbCDC42Q63L might induce actin polymerisation and the subse-quent formation of elongated cells from which a pseudo-hyphal growth can start.

It is known that pseudohyphal filamentous growth is in-duced in both haploid and diploid yeast cells in particularenvironmental conditions (Johnson 1999). In response tonitrogen starvation and fermentable carbon abundance,yeast diploid cells undergo pseudohyphal filamentousgrowth. The cells grow longer, continue to physically ad-here to each other, and invade the substrate, allowing thisnonmotile organism to access new nutritional sources inadverse conditions. An analogous mechanism of dimor-phic growth is also fundamental to the priming of viru-lence in pathogenic fungi, as reported in a study oninvasive hyphal growth of Candida albicans (Bassilanaet al. 2003).

It has been reported that CDC42 is able to primepseudohyphal growth in S. cerevisiae by activating theMAP kinase cascade responsible for the necessary geneexpression (Lengeler et al. 2000). One of these genes,Flo11, encodes the membrane protein responsible for cellflocculation (Lo and Dranginis 1998, Lengeler et al. 2000).This protein contributes to pseudohyphal growth, keepsthe cells linked after budding, and might be responsiblefor the observed aggregation.

In conclusion, these results represent a first step to-wards gaining a better understanding of the complex mo-saic of cell signal transduction events that lead to apicalgrowth in the ectomycorrhizae of T. borchii. The data sug-gest that TbCDC42 is involved in cell signalling leadingto morphogenesis of T. borchii vegetative hyphae in theearly phase of symbiosis instauration. TbCDC42 might beactivated and subsequently anchored to the membrane layer,where it can play a role in cytoskeleton assembly and acti-vate cell signal transduction pathways through a MAP ki-nase cascade (Menotta et al. 2006). This mechanism mightregulate the synthesis of the proteins involved in apicalgrowth and cell proliferation with a consequent increasein overall cell metabolism. The strong effect of root ex-udates on Tbcdc42 gene expression suggests that low-molecular-weight compounds previously detected in themedium during the presymbiosis phase (Menotta et al.2004a) might have a key role not only in hyphal growthbut also in ectomycorrhizal development, acting as mol-ecular messengers. CDC42 is only one of the numerous

M. Menotta et al.: CDC42 and apical growth in Tuber borchii 235

elements that play a role in the complex network of cellsignalling processes. Although several putative CDC42effectors have been identified in other organisms bytheir ability to interact with GTP-bound CDC42, itis not completely clear how CDC42 signals to theactin cytoskeleton. Further studies are required in orderto fully understand the signalling pathway that leadsto the particular tightly regulated morphological changesduring filamentous growth and the instauration ofsymbiosis.

Acknowledgments

This work was supported by the project “CIPE” 36/02.

References

Amicucci A, Guidi C, Menotta M, Pierleoni R, Polidori E, Zeppa S,Stocchi V (2005) Towards a better understanding of the ectomycor-rhizal fungus Tuber borchii Vittad. life cycle. Recent Res Dev Micro-biol 9: 129–158

Bassilana M, Blyth J, Arkowitz R (2003) Cdc24, the GDP-GTP ex-change factor for Cdc42, is required for invasive hyphal growth ofCandida albicans. Eukaryot Cell 2: 9–18

Boyce KJ, Hynes MJ, Andrianopoulos A (2005) The Ras and RhoGTPases genetically interact to co-ordinately regulate cell polarityduring development in Penicillium marneffei. Mol Microbiol 55:1487–1501

Buee M, Rossignol M, Jauneau A, Ranjeva R, Bécard G (2000) The pre-symbiotic growth of arbuscular mycorrhizal fungi is induced by abranching factor partially purified from plant root exudates. Mol PlantMicrobe Interact 13: 693–698

Cerione RA (2004) Cdc42: new roads to travel. Trends Cell Biol 14:127–132

Drubin DG (1991) Development of cell polarity in budding yeast. Cell65: 1093–1096

Eitzen G, Thorngren N, Wickner W (2001) Rho1p and Cdc42p act afterYpt7p to regulate vacuole docking. EMBO J 20: 5650–5656

Erland S, Henrion B, Martin F, Glover LA, Alexander IJ (1994) Identifi-cation of ectomycorrhizal basidiomycete Tylospora fibrillosa Donk byRFLPs analysis of the PCR-amplified ITS and IGS regions of riboso-mal DNA. New Phytol 126: 525–532

Etienne-Manneville S (2004) Cdc42 – the centre of polarity. J Cell Sci117: 1291–1300

Gorfer M, Tarkka MT, Hanif M, Pardo AG, Laitiainen E, Raudaskoski M(2001) Characterization of small GTPases Cdc42 and Rac and the re-lationship between Cdc42 and actin cytoskeleton in vegetative and ec-tomycorrhizal hyphae of Suillus bovinus. Mol Plant Microbe Interact14: 135–144

Gurr SJ, Unkles SE, Kinghorn JR (1987) The structure and organ-isation of nuclear genes of filamentous fungi. In: Kinghorn JR(ed) Gene structure in eukaryotic microbes. Society of General Mi-crobiology special publications, vol 22. IRL Press, Oxford, pp 93–139

Hall A (1998) Rho GTPases and the actin cytoskeleton. Science 279:509–514

Hall IR, Wang Y, Amicucci A (2003) Cultivation of edible ectomycor-rhizal mushrooms. Trends Biotechnol 21: 433–438

Harris SD, Momany M (2004) Polarity in filamentous fungi: moving be-yond the yeast paradigm. Fungal Genet Biol 41: 391–400

Hazan I, Sepulveda-Becerra M, Liu H (2002) Hyphal elongation is regu-lated independently of cell cycle in Candida albicans. Mol Biol Cell13: 939–949

Hoffman GR, Nassar N, Cerione RA (2000) Structure of the Rho familyGTP-binding protein Cdc42 in complex with the multifunctional regu-lator RhoGDI. Cell 100: 345–356

Jaffe AB, Hall A (2005) Rho GTPases: biochemistry and biology. AnnuRev Cell Dev Biol 21: 247–269

Johnson DI (1999) CDC42: an essential Rho-type GTPase controlling eu-karyotic cell polarity. Microbiol Mol Biol Rev 63: 54–105

Johnson DI, Pringle JR (1990) Molecular characterization of CDC42, aSaccharomyces cerevisiae gene involved in the development of cellpolarity. J Cell Biol 111: 143–152

Kaibuchi K, Kuroda S, Amano M (1999) Regulation of the cytoskeletonand cell adhesion by the Rho family GTPases in mammalian cells.Annu Rev Biochem 68: 459–486

Lengeler KB, Davidson RC, D’souza C, Harashima T, Shen WC, Wang P,Pan X, Waugh M, Heitman J (2000) Signal transduction cascades regu-lating fungal development and virulence. Microbiol Mol Biol Rev 64:746–785

Lo W-S, Dranginis AM (1998) The cell surface flocculin Flo11 is re-quired for pseudohyphae formation and invasion by Saccharomycescerevisiae. Mol Biol Cell 9: 161–171

Lowe M, Kreis T (1998) Regulation of membrane traffic in animal cellsby COPI. Biochim Biophys Acta 1404: 53–66

Mahlert M, Leveleki L, Hlubek A, Sandrock B, Bölker M (2006) Rac1and Cdc42 regulate hyphal growth and cytokinesis in the dimorphicfungus Ustilago maydis. Mol Microbiol 59: 567–578

Menotta M, Gioacchini AM, Buffalini M, Amicucci A, Sisti D, StocchiV (2004a) Headspace solid-phase microextraction with gas chro-matography and mass spectrometry in the investigation of volatile or-ganic compounds in an ectomycorrhizae synthesis system. RapidCommun Mass Spectrom 18: 206–210

Menotta M, Amicucci A, Gioacchini AM, Sisti D, Stocchi V (2004b)Differential gene expression during pre-symbiotic interaction betweenTuber borchii Vittad. and Tilia americana L. Curr Genet 46: 158–165

Menotta M, Pierleoni R, Amicucci A, Sisti D, Millo E, Chiarantini L,Cerasi A, Stocchi V (2006) Characterization and complementation of theTuber borchii MAPK, TBMK. Mol Plant Microbe Interact 2: 126–134

Merla A, Johnson DI (2000) The Cdc42p GTPase is targeted to the siteof cell division in the fission yeast Schizosaccharomyces pombe. Eur JCell Biol 79: 469–477

Molina R (1979) Pure culture synthesis and host specificity of red aldermycorrhizae. Can J Bot 57: 1223–1228

Mosch HU, Roberts RL, Fink GR (1996) Ras2 signals via theCdc42/Ste20/mitogen-activated protein kinase module to induce fila-mentous growth in Saccharomyces cerevisiae. Proc Natl Acad SciUSA 93: 5352–5356

Mosch HU, Kohler T, Braus GH (2001) Different domains of the essen-tial GTPase Cdc42p required for growth and development of Saccha-romyces cerevisiae. Mol Cell Biol 21: 235–248

Müller O, Johnson DI, Mayer A (2001) Cdc42p functions at the dockingstage of yeast vacuole membrane fusion. EMBO J 20: 5657–5665

Murashige T, Skoog F (1962) A revised medium for rapid growth andbioassays with tobacco tissue cultures. Physiol Plant 15: 473–497

Nickel W, Brügger B, Wieland FT (2002) Vesicular transport: the coremachinery of COPI recruitment and budding. J Cell Sci 115:3235–3240

Palecek SP, Parikh AS, Kron SJ (2002) Sensing, signalling and integrat-ing physical processes during Saccharomyces cerevisiae invasive andfilamentous growth. Microbiology 148: 893–907

Peter M, Neiman AM, Park HO, van Lohuizen M, Herskowitz I (1996)Functional analysis of the interaction between the small GTP bindingprotein Cdc42 and the Ste20 protein kinase in yeast. EMBO J 15:7046–7059

236 M. Menotta et al.: CDC42 and apical growth in Tuber borchii

Polidori E, Agostini D, Zeppa S, Potenza L, Palma F, Sisti D, Stocchi V(2002) Identification of differentially expressed cDNA clones in Tiliaplatyphyllos-Tuber borchii ectomycorrhizae using a differentialscreening approach. Mol Genet Genomics 266: 858–864

Raudaskoski M, Rupes I, Timonen S (1991) Immunofluorescence mi-croscopy of the cytoskeleton in filamentous fungi after quick-freezingand low-temperature fixation. Exp Mycol 15: 167–173

Richman TJ, Sawyer MM, Johnson DI (2002) Saccharomyces cerevisiaeCdc42p localizes to cellular membranes and clusters at sites of polar-ized growth. Eukaryot Cell 1: 458–468

Rivero F, Somesh BP (2002) Signal transduction pathways regulated byRho GTPases in Dictyostelium. J Muscle Res Cell Motil 23: 737–749

Rivero F, Illenberger D, Somesh BP, Dislich H, Adam N, Meyer A-K(2002) Defects in cytokinesis, actin reorganization, and the contractilevacuole in cells deficient in RhoGDI. EMBO J 21: 4539–4549

Salo V, Niini SS, Virtanen I, Raudaskoski M (1989) Immunocytochem-istry of the cytoskeleton in filamentous multinucleate hyphae. J CellSci 94: 11–24

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratorymanual, 2nd edn. Cold Spring Harbor Laboratory Press, New York

Schmidt A, Hall A (1998) Signaling to the actin cytoskeleton. Annu RevCell Dev Biol 14: 305–338

Sisti D, Zambonelli A, Giomaro G, Rossi I, Ceccaroli P, Citterio B,Benedetti PA, Stocchi V (1998) In vitro mycorrhizal synthesis of mi-

cropropagated Tilia platyphyllos Scop plantlets with Tuber borchiiVittad. mycelium in pure culture. Acta Hortic 457: 379–387

Weber M, Salo V, Uuskallio M, Raudaskoski M (2005) Ectopic ex-pression of constitutively active Cdc42 small GTPase alters the mor-phology of haploid and dikaryotic hyphae in the filamentoushomobasidiomycete Schizophyllum commune. Fungal Genet Biol 42:624–637

Winer J, Jung CK, Shackel I, Williams PM (1999) Development and val-idation of the real-time quantitative reverse transcriptase-polymerasechain reaction for monitoring gene expression in cardiac myocytes invitro. Anal Biochem 270: 41–49

Zambonelli A, Iotti M, Giomaro G, Hall IR, Stocchi V (2002) Tuberborchii cultivation: an interesting perspective. In: Hall IR, Wang Y,Danell E, Zambonelli A (eds) Edible mycorrhizal mushrooms andtheir cultivation: proceedings of the Second International Conferenceon Edible Mycorrhizal Mushrooms. CD-ROM. New Zealand Institutefor Crop and Food Research, Christchurch

Zhang X, Bi E, Novick P, Du L, Kozminski KG, Lipschutz JH, Guo W(2001) Cdc42 interacts with the exocyst and regulates polarized secre-tion. J Biol Chem 276: 46745–46750

Ziman M, O’Brien JM, Ouellette LA, Church WR, Johnson DI (1991)Mutational analysis of CDC42Sc, a Saccharomyces cerevisiae genethat encodes a putative GTP-binding protein involved in the control ofcell polarity. Mol Cell Biol 11: 3537–3544

M. Menotta et al.: CDC42 and apical growth in Tuber borchii 237