Phospholipase A2 up-regulation during mycorrhiza formation in Tuber borchii

www.newphytologist.org

229

Research

Blackwell Publishing, Ltd.

Phospholipase A

2

up-regulation during mycorrhiza

formation in

Tuber borchii

Laura Miozzi

1

, Raffaella Balestrini

2

, Angelo Bolchi

3

, Mara Novero

1

, Simone Ottonello

3

and Paola Bonfante

1,2

1

Dipartimento di Biologia Vegetale, Università di Torino;

2

IPP-CNR, Sezione di Micologia del Terreno, Viale Mattioli 25, 10125 Torino, Italy;

3

Dipartimento

di Biochimica e Biologia Molecolare, Università di Parma, Parco Area delle Scienze 23/A, 43100 Parma, Italy

Summary

•

TbSP1 is a secreted and surface-associated phospholipase A

2

previously found tobe up-regulated in C- or N-deprived free-living mycelia from the ectomycorrhizalascomycete

Tuber borchii

. As nutrient limitation is considered an important environ-mental factor favouring the transition to symbiotic status, TbSP1 was suggested tobe involved in the formation of mycorrhizas.

•

An

in vitro

symbiosis system between

Cistus incanus

and

T. borchii

was set up:

TbSP1

mRNA levels in free-living mycelia and in mycorrhizas sampled in differentdistricts of the plant–fungus interaction were examined. In the same samples, TbSP1protein expression was analysed by immunoelectron microscopy.

•

A substantially enhanced

TbSP1

mRNA expression, compared with nutrient-limited but free-living mycelia, was detected in the presence of the plant andreached maximal levels in fully developed mycorrhizas. A similar expression trendwas revealed by immunolocalization experiments.

•

We have shown that TbSP1 appears to respond to two partially overlapping yetdistinct stimuli: nutrient starvation and mycorrhiza formation.

Key words:

ectomycorrhizal fungi, gene expression, mycorrhiza formation,phospholipase A

2

,

Tuber borchii

Vittad.

New Phytologist

(2005)

167

: 229–238

©

New Phytologist

(2005)

doi

: 10.1111/j.1469-8137.2005.01400.x

Author for correspondence:

Paola Bonfante Tel: +39 011670 5965; +39 011650 2927Fax: +39 011670 5962Email: [email protected]

Received:

13 October 2004

Accepted:

24 January 2005

Introduction

Ectomycorrhizal (ECM) fungi are soilborne, filamentousmicrobes that associate with the roots of many tree species toproduce characteristic structures: the ectomycorrhizal root tips.The resulting symbiosis is mutualistic, as both partners benefitfrom a bidirectional nutrient exchange – photosyntheticcarbon from the host plant to the fungus; and various mineralnutrients (especially inorganic nitrogen and phosphate) fromthe fungus to the phytobiont (Smith & Read, 1997).

Like all other mycorrhizal fungi, the ascomycete

Tuberborchii

goes through two crucial morphogenetic check pointsin its life cycle: the transition from the filamentous growthtypical of free-living mycelia towards mycorrhizas (symbioticphase) and towards ascomata (reproductive phase). The lattertwo phases are both characterized by pseudo-tissues with well

differentiated cytological features: packed vs loose vegetativehyphae in the peridium and gleba of ascomata, where asci andascospores represent the reproductive structures; and denselypacked and highly branched hyphae, respectively, in themantle and Hartig net districts of mycorrhizas. Despite theanatomical and functional specialization displayed by symbi-otic hyphae involved in either soil exploration (extraradicalmycelium) or root contact (Hartig net region), very fewattempts have been made so far towards the molecular dissec-tion of these distinct mycorrhizal districts (Nehls

et al

., 2001).According to Tagu

et al

. (2002), the above transitions proba-bly result from a balance between genetic determinism andenvironmental conditions that may enhance or repress sym-biosis establishment and ascoma formation. Gene-expressionprofiles during symbiosis establishment have been investigatedin two ECM systems, both of them from Basidiomycetes

New Phytologist

(2005)

167

: 229–238

www.newphytologist.org

©

New Phytologist

(2005)

Research230

(Voiblet

et al

., 2001; Johansson

et al

., 2004; Duplessis

et al

.,2005). During

Paxillus involutus–Betula pendula

mycorrhizaformation, the expression of a significant number of genes inboth the fungal and the plant partner was found to be affected( Johansson

et al

., 2004). Some of the up-regulated fungalgenes appear to be related to development (e.g. hydrophob-ins); to C metabolism (e.g. hexokinase); or to stress responses(e.g. superoxide dismutase). Differential gene expressionduring presymbiotic interaction was recently detected in

T. borchii

when cultivated without contact with its symbioticplant partner,

Tilia americana

L. (Menotta

et al

., 2004).Sequence analysis showed that most of the expressed genesmight be involved in several biochemical pathways, e.g. secre-tion and apical growth, cellular detoxification and generalmetabolism. On the other hand, environmental conditionsthat may influence the transition of free-living (saprotrophic)mycelia to either a symbiotic or a reproductive lifestyle can, tosome extent, be reproduced under laboratory conditions. Forexample, a macroarray analysis has shown that glutaminesynthetase is one of the most highly expressed genes in natural

T. borchii

ascomata (Lacourt

et al

., 2002), and the same genewas subsequently found to be up-regulated also in N-starved,

in vitro

-cultured mycelia (Montanini

et al

., 2003), leading tothe suggestion that fruitbody formation

in vivo

takes placeunder N shortage conditions.

Another gene that is strongly up-regulated by nutrientdeprivation in

T. borchii

is

TbSP1

, which codes for a calcium-dependent, secreted and surface-associated phospholipase A

2

(PLA

2

;

M

r

= 19 000) – the first enzyme of this kind (groupXIV sPLA

2

s) to be described in microorganisms, either bacte-ria or fungi (Soragni

et al

., 2001). The

TbSP1

messenger wasfound to be strongly up-regulated by growth of free-livingmycelia under either C- or N- but not phosphate-deprivationconditions. It is well known that plant roots take up variousmineral nutrients (including inorganic N) within theirabsorption area, thus creating a ‘nutrient-poor’ depletionregion (Marschner, 1995). On the other hand, root exudatesreleased by the plant enrich the rhizosphere in C compoundsof photosynthetic origin (especially carbohydrates). Thesechanges in nutrient composition are probably perceived byboth symbiotic partners, and may act as trophic signals forsoilborne fungi (Tagu

et al

., 2002). We thus hypothesizedthat nutrient depletion might represent an important envir-onmental factor capable of promoting (or at least favouring)the mycorrhizal transition in

T. borchii

.To begin to test this hypothesis, and to investigate the

expression profile of the TbSP1 PLA

2

during early plant–fungus interaction stages, we set up an

in vitro

mycorrhizationsystem between

Cistus incanus

L. and

T. borchii

. This allowedus to reproduce various precolonization growth conditions,ultimately leading to fully developed mycorrhizal tips, and toexamine TbSP1 expression (both mRNA and protein) indifferent developmental stages and in topographically distinctdistricts of the plant–fungus complex by a combination of

RT–PCR, real-time PCR and immunoelectron microscopyanalyses. An enhanced

TbSP1

expression compared with thatof free-living mycelia was detected in extraradical hyphaebefore plant colonization and in established mycorrhizas. Wefind that TbSP1 is positively modulated, not only by nutrientdeprivation, but also by mycorrhization.

Materials and Methods

Biological material

Tuber borchii

Vittad. mycelium (isolate ATCC 96540) wasgrown in the dark at 25

°

C for 30 d. The modified Melin–Norkrans medium (MMN; Mischiati & Fontana, 1993)and the minimal M medium (MN; Bécard & Fortin, 1988),commonly employed for mycorrhization experiments, wereutilized for mycelial cultures. Apart from supplementationof MMN with malt extract (3 g l

−

1

), the major differencebetween the two media was the C source: glucose (10 g l

−

1

)for MMN; sucrose (10 g l

−

1

), which is very poorly utilized by

T. borchii

(Saltarelli

et al

., 1998), for MN. When required,free-living mycelia were grown on a semipermeable cellophanemembrane (Bio-Rad, Hercules, CA, USA) to facilitate mediumreplacement.

Cistus incanus

L. seeds were abraded delicatelywith emery paper, soaked for 10 min into a diluted (1 : 10)bleach solution, and finally rinsed three times with sterilewater.

Cistus

seedlings were grown on sucrose-supplemented(3% w/v) Murashige and Skoog solid medium (MS; Murashige& Skoog, 1962).

In vitro

mycorrhization

The

in vitro

plate system developed for

Eucalyptus globulus–Pisolithus tinctorius

mycorrhizas (Martin

et al

., 1998) wasused with slight modifications. Briefly,

T. borchii

myceliumwas grown in Petri dishes (100

×

20 mm, Corning, SigmaAldrich Italia, Milano, Italy) on MN medium for 2 wk in thedark at 25

°

C. After this time, 30-d-old

Cistus

seedlings weretransferred into mycelium-containing dishes, which were thenkept in a growth chamber at 22

°

C with a 16 h light : 8 h darkphotoperiod. Similarly to what has been observed previouslywith a different

T. borchii–Tilia platyphyllos in vitro

mycorrhiza-tion system (Sisti

et al

., 1998), mycorrhizal tips usually formedafter 60–70 d of co-cultivation.

Tuber borchii

mycelia co-cultivated for only 30 d with

C. incanus

seedlings in MNmedium were used to examine plant–fungus signalling eventspreceding mycorrhiza establishment (precolonization stage).Co-cultivation was also carried out in the presence of a septum(3MM Chr, Whatman) separating the plant from the fungus.

RT–PCR analysis

Total RNA for RT–PCR was extracted from the followingsamples: (i) free-living mycelia grown for 30 or 60 d (as

©

New Phytologist

(2005)

www.newphytologist.org

New Phytologist

(2005)

167

: 229–238

Research 231

specified in the text) on MN medium (FLM-MN); (ii) free-living mycelia grown for the same length of time on richmedium (FLM-MMN); (iii) mycelia cultured in the presenceof the plant but physically separated from it by a septum(FLM-SEP); (iv) mycelium cultured in the presence of theplant and sampled at an early, precolonization stage (30 d)at the edge of the fungal colony (‘extraradical hyphae’,EXRH); (v) in the region of contact with the roots (‘earlyroot-contacting hyphae’, ERCH); (vi) fully developed mycor-rhizal tips (MYC). RNA extraction was carried out accordingto Viotti

et al

. (1982), except for RNA from mycorrhizaswhich, because of the extremely small amount of startingmaterial, was extracted with the SV total RNA IsolationSystem kit (Promega, Madison, WI, USA). All RNA sampleswere checked routinely for DNA contamination by PCRanalysis, conducted with the 18S rDNA universal primersS1A and S2B (S1A: 5

′

-GCCAGAAGGAAAGATCCG-3

′

;S2B: 5

′

-TCATGATAACTTAACGAATCGC-3

′

) without theretro-transcription step.

The One-Step RT–PCR kit (Qiagen, Hilden, Germany)was used for RT–PCR experiments. Reactions were carriedout in a final volume of 10 µl containing 2 µl 5

×

buffer,400 µ

dNTPs, 0.8 µ

of each primer, 0.4 µl One-StepRT–PCR enzyme mix, 6.66 U RNase inhibitor (AmershamBiosciences, Freiburg, Germany), and 1–2 µl total RNA. Foreach experiment, the amount of RNA in each sample was firstdetermined by RT–PCR using the rDNA primers S1A andS2B. Reverse transcription was allowed to proceed for 30 minat 50

°

C, followed by 15 min at 95

°

C and by 28 cycles of PCRamplification (30 s at 94

°

C, 45 s at 62

°

C, 45 s at 72

°

C).

TbSP1

transcript levels were determined by semiquantitativeRT–PCR, which was conducted on internally calibratedamounts of input RNA using oligonucleotide primers designedon the sequence of the

TbSP1

cDNA (accession No. AF162269):Tbsp1-plus, 5

′

-ACCCAAATTACAGAGCCAACC-3

′

; Tbsp1-minus, 5

′

-CACTATTTCCTTCCTCCATCC-3′. Amplifica-tion reactions (30 s at 94°C, 45 s at 50°C, 45 s at 72°C) wereallowed to proceed for 36 cycles. Because of the differentrelative abundance of the two transcripts, a different numberof cycles, corresponding in each case to the exponential phase ofthe reaction (as determined in preliminary set-up experiments),were utilized for rDNA and TbSP1 amplification. Afterelectrophoretic fractionation on a 1.3% agarose gel, RT–PCRproducts were stained with ethidium bromide and blottedonto a nylon membrane (Hybond-N+, Amersham Biosciences).This was followed by hybridization with a TbSP1-specificprobe (derived from PCR amplification of the TbSP1 cDNA)under high-stringency conditions. Labelling, hybridization anddetection were performed as described in the ECL DirectNucleic Acid Labelling and Detection System protocol(Amersham Biosciences). Ethidium bromide-stained bands andhybridization signals were quantified by densitometric analy-sis (-/PC software; Bio-Rad). All experimentswere performed on two independent biological samples.

Real-time RT–PCR analysis

RNA samples for real-time RT–PCR were denatured at 65°Cfor 5 min, followed by reverse transcription (37°C, 60 min)in a final volume of 20 µl containing 5 µl total RNA, 10 µrandom primers, 0.5 m dNTPs, 10 U RNase inhibitor,2 ml 10× buffer and 1 ml Sensiscript RT (Qiagen). Tominimize potential heterogeneities in RT reaction yield, eachcDNA sample was derived by pooling three separate RTreactions. Real-time PCR was carried out with an iCyclerapparatus (Bio-Rad). Individual reactions were assembledwith oligonucleotide primers (0.15 m each), 12.5 ml 2× iQSYBR Green Supermix (Bio-Rad) plus an appropriate volumeof each cDNA preparation in a final volume of 25 µl. ThePCR cycling program (30 s at 95°C followed by 30 s at 62°C)included a heating step (3.5 min at 95°C) at the beginning ofevery run. The following primers were utilized for real-timePCR analysis:5′-CAGAGCCAACGAGTTCATCA-3′ (18S rRNA for-ward; 18SLP1);5′-GTACAAAGGGCAGGGACGTA-3′ (18S rRNA reverse;18SRP1);5′-CCGCTTCACAGAGGCTAATC-3′ (TbSP1 forward;TbLP4);5′-AAGTGTTCGCGATTTTACGG-3′ (TbSP1 reverse;TbRP4).

Single amplicons (159 and 128 bp long for the 18S rDNAand TbSP1, respectively) were produced by both primer sets.A melting curve (55–95°C with a heating rate of 0.5°C per10 s and a continuous fluorescence measurement) was recordedat the end of every run to assess amplification product specif-icity (Ririe et al., 1997). All reactions were performed at leastin duplicate; data were analysed with the i software.The ‘comparative threshold cycle (Ct)’ method was used tocalculate relative TbSP1 expression levels with the 18S rRNAas a reference (Rasmussen, 2001). Real-time PCR efficiencies[E = 10(−1/–slope)] were derived from standard curves constructedon the target (TbSP1) and the reference (18S rDNA) gene.

Microscopy

Stereomicroscopy analysis was utilized to monitor mycorrhizadevelopment and to select mycorrhiza specimens forultrastructural examination. Samples were embedded in 10%agarose and cut into 50 µm sections with a Vibratome. Aftertreatment with 1.5% NaHClO in water for 10 min (threetimes), sections were stained by incubation for 1 h in 10 µg ml−1

fluorescein isothiocyanate-conjugated wheat germ agglutinin(FITC–WGA). Stained sections were examined with a NikonOptiphot-2 microscope (Tokyo, Japan) equipped with a Bio-Rad (Hemel Hempstead, UK) View Scan DVC-250 confocalsystem.

Mycelia and mycorrhizal tips were prepared for transmis-sion electron microscopy (TEM) by overnight fixation at 4°C

New Phytologist (2005) 167: 229–238 www.newphytologist.org © New Phytologist (2005)

Research232

in 2.5% (v/v) glutaraldehyde dissolved in 10 m sodiumphosphate buffer pH 7.2 (Balestrini et al., 1996). Afterwashing with the same buffer, osmium tetroxide (1% w/v)postfixation, ethanol dehydration, and ethanol/LR White(Polysciences, Warrington, PA, USA) infiltration, sampleswere embedded in LR White resin. Following immunogoldlabelling with the anti-TbSP1 antibody (Soragni et al., 2001),sections were analysed with a Philips CM10 TEM. Samplesfrom which the primary antibody was omitted were used aslabelling-specificity controls; immunogold experiments werereplicated at least six times. Immunogold labelling density wasevaluated according to Bendayan (1984) by determining thenumber of gold particles (Ni) within a chosen area (Sa),followed by calculation of the labelling density, Ns, as the Ni/Saratio, on at least five prints at the same magnification. Thesignificance of density-labelling differences in different sampleswas assessed by variance analysis (P < 0.05 significant; P < 0.01highly significant) carried out with the program two-way with Tukey’s test as a posthoc ( 10.5).

Results

Cistus incanus–Tuber borchii mycorrhizal roots

After 60 d co-cultivation of T. borchii mycelium–C. incanusseedlings in a minimal, sucrose-containing solid medium(MN), fully developed mycorrhizas usually formed in ≈50%of the plates (Fig. 1a,b). Abundant extramatrical hyphae wereobserved associated with the fungal mantle, which was welldeveloped and consisted of three or four layers, while hyphaespreading among epidermal and cortical root cells were presentin the Hartig net region. As revealed by WGA labelling (Fig. 1c),the latter hyphae displayed a characteristic labyrinthinemorphology with many branches and irregular septa. Afull set of intracellular organelles was easily detectable inthese hyphae (Fig. 1d). Mycorrhizas produced in this in-platesystem thus resemble morphologically those obtained undermore natural (yet less tractable) experimental conditionsusing T. platyphyllos as the host plant (Sisti et al., 2003).

TbSP1 expression levels in mycorrhizas vs mycelia

TbSP1 mRNA abundance in free-living mycelia andmycorrhizas was assessed initially by semiquantitativereverse transcription PCR (RT–PCR) analysis. Sequence-specific amplification primers designed on the TbSP1 cDNAsequence were first tested on total RNA extracted from theroots of C. incanus. The negative result obtained from such

control amplifications (data not shown) allowed us to excludeany cross-hybridization with the plant transcriptome. TotalRNA extracted from free-living mycelia grown for 60 d onMN minimal medium (FLM-MN) and from fully developedmycorrhizal tips (MYC) obtained after 60 d co-cultivationwas then used for RT–PCR analysis. The amount of fungalRNA contained in mycelium- and mycorrhiza-derived sampleswas first quantified by PCR amplification with fungus-specific 18S rRNA primers previously found not to produceany amplicon in reactions programmed with plant-derivedRNA (data not shown). Balanced amounts of both RNAsamples (Fig. 2a) were then amplified with TbSP1-specificoligonucleotide primers. Because of the limited amount ofmycorrhiza-derived RNA available for analysis, amplificationsignals were not readily detectable in ethidium bromide-stained agarose gels (Fig. 2b). To gain more sensitivity,amplification reaction products, previously electrophoresedon agarose gels, were transferred onto a nylon membraneand subsequently hybridized with a horseradish peroxidase-labelled TbSP1-specific cDNA probe. A hybridizationsignal corresponding to an amplified cDNA fragment of theexpected size (536 bp) was thus detected in both FLM-MNand MYC samples, indicating that the TbSP1 mRNA isexpressed under both conditions (Fig. 2c). As revealed bythis analysis, however, the amount of TbSP1 mRNA-derivedamplicons was about three-fold higher in mycorrhizal root tipsthan in free-living mycelia grown for the same length of timeon minimal medium.

TbSP1 expression levels in different districts and conditions of plant–fungal interaction

Semiquantitative RT–PCR analysis was then extended tofree-living mycelia subjected to different nutritional regimensand to topographically distinct districts of plant–fungusinteraction (see Materials and Methods). Taking advantage ofthe availability of more abundant mycelial samples (comparedwith mycorrhizas), TbSP1 mRNA-derived amplicons werevisualized directly by ethidium bromide staining. Datareported in Fig. 3 show that a TbSP1 amplicon was barelydetectable in mycelia cultured in MMN medium (FLM-MMN).An intense amplification signal was detected, instead, inparallel-grown mycelia cultured in mycorrhization medium(FLM-MN). Under the latter conditions, TbSP1 expressionlevels were comparable with those detected in mycelia grownin the presence of the plant, but physically separated from itby a septum (FLM-SEP), as well as to those of early root-contacting hyphae (ERCH). Even higher TbSP1 expression

Fig. 1 Cistus incanus–Tuber borchii ectomycorrhizas. (a) In vitro plate mycorrhization system used in the present study. Bar, 1.9 cm. (b) Magnification of Cistus roots growing in the presence of T. borchii in the plate-culture system. Bar, 0.5 cm. (c) Transverse section of a mycorrhizal tip stained with FITC–WGA, showing the loose external and the inner compact mantle (m) ensheathing the root, and the Hartig net mycelium progressing between host cells (arrowheads). Bar, 5 µm. (d) Ultrastructural view of the Hartig net. A full set of intracellular organelles is visible within fungal hyphae (h): H, host cell; n, nucleus; v, vacuole. Bar, 1.7 µm.

© New Phytologist (2005) www.newphytologist.org New Phytologist (2005) 167: 229–238

Research 233

New Phytologist (2005) 167: 229–238 www.newphytologist.org © New Phytologist (2005)

Research234

levels were detected in extraradical hyphae sampled at the edgeof the fungal colony, far away from the host plant (EXRH).

Real-time RT–PCR analysis of TbSP1 expression during mycorrhizogenesis

The above results concerning the expression of TbSP1 mRNAin free-living mycelia, fully developed mycorrhizas and inintermediate steps of mycorrhizogenesis were verified indep-endently by a real-time RT–PCR analysis. The 18S rRNAwas also used as an internal calibration standard for theseexperiments, which, in the investigated range of cDNA input,exhibited efficiency rates higher than 96% for both the 18SrRNA and TbSP1 transcripts. Under the same RT reaction

conditions employed previously for RT–PCR analysis, TbSP1transcript levels in fully developed mycorrhizal tips wereabout eight-fold higher than those of the corresponding free-living mycelium grown in MN medium. Real-time RT–PCRanalysis was next utilized to determine relative TbSP1 mRNAexpression levels in all conditions previously examined bysemiquantitative RT–PCR, using FLM-MN as a referencesample for data normalization (Table 1). As further shown inTable 1, albeit with some quantitative differences (most notablya higher expression level in fully developed mycorrhizas anda less pronounced difference between extraradical and earlyroot contacting hyphae), an expression profile comparablewith that previously obtained by semiquantitative RT–PCRwas produced by real-time RT–PCR analysis. Despite the useof an identical minimal medium, TbSP1 mRNA abundancein all conditions (except for the FLM-MMN control) washigher than that measured in FLM-MN-derived RNAsamples, and was maximal in fully developed mycorrhizaltips. This suggests that the TbSP1 PLA2 is required not onlyduring early stages of symbiosis formation, but also for themaintenance of established mycorrhizas.

Immunogold visualization of the TbSP1 protein

To verify whether the up-regulation of the TbSP1 mRNAobserved under both presymbiotic and symbiotic conditionswas accompanied by a similar increase in TbSP1 proteinlevels, a detailed TEM-immunolabelling analysis was conductedon samples representative of the above-mentioned conditions.Of critical importance for this analysis was the specificity (i.e.lack of any cross-reaction with proteins other than TbSP1, ofeither plant or fungal origin) of the anti-TbSP1 antiserumutilized for immunolocalization. What should be noted inthis regard is: (i) the lack of any detectable labelling in the

Fig. 2 RT–PCR analysis of TbSP1 mRNA levels in free-living mycelia and mycorrhizas. (a) Calibration of mycelium (FLM-MN)- and mycorrhiza (MYC)-derived total RNA samples using the 18S rRNA as an internal reference. (b) RT–PCR analysis of TbSP1 mRNA abundance in internally calibrated RNA samples derived from 60-d-old free-living mycelia grown on MN medium (FLM-MN) and on the corresponding mycorrhizal tips (MYC); a larger amount (> 100-fold) of mycelium-derived total RNA was used as RT–PCR control (positive control). (c) Hybridization analysis of a DNA blot derived from the gel shown in (b).

Fig. 3 RT–PCR analysis of TbSP1 mRNA levels in: FLM-MN, free-living mycelium grown for 30 d in mycorrhization medium (MN); FLM-MMN, free-living mycelium grown for the same length of time in rich medium (MMN); FLM-SEP, mycelium cultured in the presence of the plant but physically separated from it by a septum; EXRH, mycelium cultured in the presence of the plant and sampled at an early, precolonization stage (30 d) at the edge of the fungal colony (‘extraradical hyphae’); or ERCH, in the region of contact with the roots (‘early root-contacting hyphae’). (a) Calibration of total RNA samples using the 18S rRNA as an internal reference. (b) RT–PCR analysis of TbSP1 mRNA abundance in internally calibrated RNA samples.

Table 1 Relative expression levels of the TbSP1 mRNA

Sample

Relative expression

Real-time PCR* RT–PCR†

FLM-MN 1 1FLM-MMN 0.4 0.5FLM-SEP 1.6 1.7EXRH 2.5 2.4ERCH 2.4 1.6MYC 7.8 2.5

*Real-time RT–PCR, relative expression data obtained with the comparative (Ct) method using the 18S rRNA as a reference RNA and normalized with respect to TbSP1 mRNA levels in FLM-MN samples. Data are the average of Ct values obtained from two technical replicates; only sets of data with SD ≤ 0.2 were considered.†Comparison with the semiquantitative RT–PCR data derived from the experiments reported in Figs 2,3, also normalized with respect to TbSP1 mRNA levels in FLM-MN samples. Data are the average of two biological replicates; SD for all samples < 0.7.

© New Phytologist (2005) www.newphytologist.org New Phytologist (2005) 167: 229–238

Research 235

root compartments of mycorrhizal sections (see below); and(ii) the fact that TbSP1 is encoded by a single-copy gene in theTuber genome, and that a single immunoreactive polypeptidehas always been detected by immunoblot analyses performedon Tuber samples with the same anti-TbSP1 antibodiesutilized for the present study (Soragni et al., 2001; A.B. andS.O., unpublished observations). As reported previously bySoragni et al. (2001), the TbSP1 protein was entirely localizedat the fungal cell surface. Very few cell wall-associated goldgranules were detected in the thin walls of FLM-MMN-derived hyphae (Fig. 4a). In keeping with previous datashowing that TbSP1 (both mRNA and protein) increasesstrongly on persistent N or C deprivation (Soragni et al.,2001), a much more conspicuous labelling was detectedin hyphae sampled from mycelia grown for the same length oftime in nutrient-poor medium (FLM-MN; Fig. 4b).

TEM-immunolabelling analysis was then extended todifferent plant–fungus interaction conditions. To select hyphae

that were topographically comparable with those previouslyutilized for RT–PCR and real-time PCR analyses, mycor-rhizal sections were first screened by light stereomicroscopy.Extraradical hyphae, not in contact with the roots (EXRH),as well as hyphae belonging to either the mantle or the Hartignet (MYC), were thus picked and analysed. Regardless of thespecific location of the hyphae and of their wall ultrastructure(which changes with hyphal age), TbSP1-associated goldgranules were always found in the fungal cell wall. By com-parison, no immunogold labelling was detected in plantcell walls, nor in control experiments (not shown) in whichthe primary anti-TbSP1 antibody was omitted. Extraradicalhyphae, displaying the usual bilayered wall typical of fullydeveloped hyphae, were intensely labelled and were oftensurrounded by an abundant (yet fairly loose) electron-densematerial (Fig. 4c,d). TbSP1 labelling was consistently foundassociated with the chitin-rich, electron-transparent innercell-wall layer (Balestrini et al., 1996). In contrast, only weak

Fig. 4 Immunogold localization of TbSP1 in free-living hyphae and presymbiotic mycelia. (a) Immunogold-TEM localization of TbSP1 in FLM-MMN hyphae transversally sectioned close to the tip. Bar, 0.28 µm. (b) More abundant TbSP1 labelling in FLM-MN hyphae. Bar, 0.27 µm; arrowheads point to TbSP1-associated gold granules within the electron-transparent inner cell-wall layer. (c) Intense labelling (arrowheads) is observed within the inner cell-wall layer of extraradical hyphae (EXRH) from mycelia grown in the presence of the plant; n, nucleus. Bar, 0.59 µm. (d) Extraradical hyphae at a higher magnification. Bar, 0.32 µm. (e) A mucilaginous material (asterisk) surrounds ‘root-contacting hyphae’ (ERCH) which exhibit a significantly reduced TbSP1 labelling compared with extraradical hyphae (see Table 2). h, hypha. Bar, 1.23 µm. (f ) Higher-magnification view of ‘root-contacting hyphae’, demonstrating the low TbSP1 labelling observed in such hyphae; h, hypha; w, plant cell wall. Bar, 0.20 µm.

New Phytologist (2005) 167: 229–238 www.newphytologist.org © New Phytologist (2005)

Research236

labelling was detected in hyphae juxtaposed to the root sur-face, but not yet in intimate contact with root epidermal cells(ERCH) (Fig. 4e,f; Table 2). In this phase, which precedestrue mycorrhization (Sisti et al., 2003), hyphal walls wereusually thin and surrounded by a mucilaginous materialproduced by the plant host (Fig. 4e). At a more advanced stageof mycorrhiza formation, when the mantle was developing,TbSP1 labelling again became intense and reached levelscomparable with those of EXRH samples (Fig. 5a; Table 2).Similarly, a strong, cell wall-localized TbSP1 labelling wasdetected in Hartig net-forming hyphae (Fig. 5b,c; Table 2).Abundant gold granules were also found associated withhyphal branching zones, visualized as ‘corner’ cell walls(Fig. 5c). In keeping with recent observations showing thattruffles are capable of penetrating root epidermal and corticalcells (Montanini et al., 2003; Selosse et al., 2004), also in thisnovel experimental system, some senescent host cells werefound to be penetrated and colonized by T. borchii hyphae. Astrong TbSP1 labelling was also found to be associated withthe cell walls of such intracellular hyphae (Fig. 5d; Table 2).

A roughly similar trend of TbSP1 abundance was observed

at the RNA and protein levels. The sole exception was foundin ERCH samples which, despite a substantial mRNA up-regulation compared with FLM-MN (Table 1), displayed anextremely reduced density of gold granule labelling (Fig. 4e,fand Table 2).

Discussion

TbSP1 mRNA levels have previously been shown to bestrongly up-regulated following nutrient deprivation (Soragniet al., 2001), and to respond to both N and C starvation, twoconditions that are considered important environmentaldeterminants of symbiosis establishment (Buscot et al., 2000).The additional information provided by the present data isthat TbSP1 expression responds not only to extremeconditions of nutrient deprivation, but also to more subtlenutritional changes (FLM-MN) and, more importantly, toco-cultivation in the presence of the plant and to mycorrhizaestablishment. The latter finding suggests that additionalfactors, in addition to nutrient availability, are sensed bymycelia grown in a rhizospheric environment. Nutrient shortage

Table 2 Gold grain quantification in immunolabelled samples*

Free-living mycelium Precolonization mycelium Fully developed mycorrhizas

FLM-MN 11 237 ± 4692 (a) EXRH 12 990 ± 3064 (a) MYC mantle/Hartig net region 11 369 ± 4567 (a)FLM-MMN 1767 ± 0569 (b,c) ERCH 1922 ± 0561 (b) MYC intracellular hyphae 9254 ± 2881 (a,c)

*Values refer to the density of cell wall-associated gold granules (± SD) detected by the anti-TbSP1 antibody; values not sharing the same letter are significantly different according to ANOVA (P ≤ 0.05).

Fig. 5 TEM-immunogold analysis of TbSP1 in fully developed mycorrhizal tips. (a) TbSP1 labelling in mantle-forming hyphae (m); n, nucleus. The immunonegative, electron-opaque material interposed between such hyphae is marked with an asterisk. Bar, 0.43 µm. (b) TbSP1-associated gold granules (arrowheads) in the chitin-rich, inner cell-wall layer of hyphae progressing between root cells in the Hartig net region (Hn); no immunopositive material is present within the host cell walls (w). Bar, 1 µm. (c) A different view of Hartig net-forming hyphae at higher magnification. Bar, 0.29 µm. (d) Intense TbSP1 labelling is associated with the inner cell-wall layer of intracellular hyphae (ih). Bar, 075 mm; inset, gold granules localized in a septum (S). Bar, 0.70 µm.

© New Phytologist (2005) www.newphytologist.org New Phytologist (2005) 167: 229–238

Research 237

is probably exacerbated by the plant, which competes with thefungus for the limited amounts of nutrients available in our invitro system. A possible nutritional role of PLA2 up-regulationis suggested by recent observations showing that soil-modifying, phospholipid surfactants are released by plantroots (Read et al., 2003). An alternative, yet not mutuallyexclusive, possibility is that the host plant may induce TbSP1up-regulation directly through as yet unidentified signals thatspread from the plant–fungus contact site to external hyphae,far away from the roots. A potential role of lipid transfer fromthe plant to the fungus in the functioning of the ECMsymbiosis has been highlighted by a recent study showing thatfatty acids of plant origin are mobilized through the symbioticinterface and accumulate in the extraradical mycelium(Laczko et al., 2003). A few cases have been reported recentlyof genes thought to be activated by plant-derived factors(Martin et al., 2001; Podila et al., 2002). For example, rootexudates have been shown to affect hyphal growth andbranching in a Pisolithus–Eucalyptus mycorrhizal system(Lagrange et al., 2001), where a fungal Ras gene (PtRas) wasfound to be induced before root contact (Martin et al., 2001).In addition, some of the symbiosis-regulated proteins iden-tified in Pisolithus (Laurent et al., 1999) are among the veryfew fungal proteins containing a putative cell adhesionmotif (RGD), which is also present in TbSP1, but not in itshomologues from nonsymbiotic ascomycetes (A.R. Viscomi,A.B. and S.O., unpublished results). These RGD-containingproteins have been classified as ‘adhesins’, possibly mediatinghomologous or heterologous cell–cell interactions and/orsurface-modification events (Hostetter, 2000).

The other main outcome of this study was the finding thatfully developed mycorrhizal tips, with their surroundingextramatrical hyphae, express the highest levels of the TbSP1mRNA and of immunopositive epitopes recognized by theanti-TbSP1 antibody. In contrast, a similar increase in TbSP1mRNA levels, but a relatively low TbSP1 protein densitylabelling (nearly the same as that measured in MMN-culturedfree-living mycelia), was observed in ERCH before mycor-rhiza establishment. Whether this is caused by some kind oftranslational interference caused by the particular nutrientsupply (or local microenvironment) provided by the plantmucilage that surrounds these hyphae is presently unknown.It should be noted, however, that a similar uncouplingbetween mRNA and protein synthesis was observed previ-ously in T. borchii (Montanini et al., 2003) as well as in otherfilamentous ascomycetes (Navarro & Aguirre, 1998).

The mantle and the Hartig net are complex pseudo-tissuesin which hyphae aggregate by increasing their reciprocal adhe-sion, augment in size, and branch in an irregular fashion, witha conspicuous membrane remodelling. Although the presenceof an RGD cell-adhesion motif may explain the surface accu-mulation of TbSP1 in the pseudo-tissue structures typical offully developed mycorrhizas, the causal relationship betweenPLA2 up-regulation and symbiosis establishment is, at

present, largely hypothetical. A direct role of TbSP1 in medi-ating hyphal breaching through the cutin-like layer thatoften coats root cell walls (Kottke, 1997) has been discountedrecently by the observation that TbSP1 is completely devoidof hydrolytic activity towards cutin analogues such as 4-nitro-phenyl palmitate (A.R. Viscomi, A.B. and S.O., unpublishedresults). So, in addition to a purely nutritional role (possiblyrelated to fatty acid mobilization), more direct signalling rolescould be envisaged. These may be mediated by either theunsaturated fatty acid or lysophospholipid moieties resultingfrom PLA2 action. For example, polyhydroxylated derivativesof linoleic acid (‘oxylipins’), that are produced by fungallinoleate diol synthases and have also been detected in truffles(Gao et al., 2001; A.B. and S.O., unpublished results), haverecently been shown to serve as hormone-like signals modu-lating the timing and balance of sexual and asexual sporedevelopment in Aspergillus nidulans (Tsitsigiannis et al., 2004and references therein). TbSP1 action in mycorrhiza develop-ment may thus fit into the as yet largely unexplored field oflipid-mediated signalling in fungi.

So, although the physiological role of PLA2 up-regulationis not known, we have shown that TbSP1 can respond to twopartially overlapped, yet distinct stimuli: nutrient starvationand mycorrhiza formation. TbSP1 can thus be viewed as anadditional and potentially interesting representative of fungalgenes that are modulated in response to symbiotic (or presym-biotic) stimuli. These genes, only a few of which have beenidentified so far in Tuber (Polidori et al., 2002; Menotta et al.,2004) and in other ECM fungi (Kim et al., 1999; Sundaramet al., 2001; Voiblet et al., 2001; Podila et al., 2002; Peteret al., 2003; Johansson et al., 2004), are of great interest asmolecular markers of the many (as yet largely unknown)events that accompany plant–fungus interaction.

Acknowledgements

We thank Arturo Roberto Viscomi (Department ofBiochemistry and Molecular Biology, University of Parma)for sharing results before publication. This work wassupported by grants from the National Research Councilof Italy (Progetto Strategico ‘Biotecnologia dei funghi eduliectomicorrizici’) to P.B. and S.O.; from the Regione EmiliaRomagna to S.O.; and from the Ministry of Education,University and Research (FIRB project ‘Plant/MicrobeInteractions’) to P.B. and A.B. The CRT of Cuneo (Italy) isacknowledged for funding the PhD course of L.M.

References

Balestrini R, Hahn MG, Faccio A, Mendgen K, Bonfante P. 1996. Differential localization of carbohydrate epitopes in plant cell walls in the presence and absence of arbuscular mycorrhizal fungi. Plant Physiology 111: 203–213.

Bécard G, Fortin JA. 1988. Early events of vesicular–arbuscular mycorrhiza formation on Ri T-DNA transformed roots. New Phytologist 108: 211–218.

New Phytologist (2005) 167: 229–238 www.newphytologist.org © New Phytologist (2005)

Research238

Bendayan M. 1984. Concentration of amylase along its secretory pathway in the pancreatic acinar cell as revealed by high resolution immunocytochemistry. Histochemical Journal 16: 85–108.

Buscot F, Munch JC, Charcosset JY, Gardes M, Nehls U, Hampp R. 2000. Recent advances in exploring physiology and biodiversity of ectomycorrhizas highlight the functioning of these symbioses in ecosystems. FEMS Microbiology Reviews 24: 601–614.

Duplessis S, Courty PE, Tagu D, Martin F. 2005. Transcript patterns associated with ectomycorrhiza development in Eucalyptus globulus and Pisolithus microcarpus. New Phytologist 165: 599–611.

Gao JM, Wang CY, Zhang AL, Liu JK. 2001. A new trihydroxy fatty acid from the ascomycete Chinese truffle Tuber indicum. Lipids 36: 1365–1370.

Hostetter MK. 2000. RGD-mediated adhesion in fungal pathogens of humans, plants and insects. Current Opinions in Microbiology 3: 344–348.

Johansson T, Le Quéré A, Ahren D, Söderström B, Erlandsson R, Lundeberg J, Uhlén M, Tunlid A. 2004. Transcriptional responses of Paxillus involutus and Betula pendula during formation of ectomycorrhizal root tissue. Molecular Plant–Microbe Interactions 17: 202–215.

Kim SJ, Hiremath ST, Podila GK. 1999. Cloning and identification of symbiosis-regulated genes from the ectomycorrhizal Laccaria bicolor. Mycological Research 103: 168–172.

Kottke I. 1997. Fungal adhesion pad formation and penetration of root cuticle in early stage mycorrhizas of Picea abies and Laccaria amethystea. Protoplasma 196: 55–64.

Lacourt I, Duplessis S, Abbà S, Bonfante P, Martin F. 2002. Isolation and characterization of differentially expressed genes in the mycelium and fruit body of Tuber borchii. Applied and Environmental Microbiology 68: 4574–4582.

Laczko E, Boller T, Wiemken V. 2003. Lipids in roots of Pinus sylvestris seedlings and in mycelia of Pisolithus tinctorius during ectomycorrhiza formation: changes in fatty acid and sterol composition. Plant, Cell & Environment 27: 27–40.

Lagrange H, Jay-Allemand C, Lapeyrie F. 2001. Rutin, the phenolglycoside from Eucalyptus root exudates, stimulates Pisolithus hyphal growth at picomolar concentrations. New Phytologist 149: 349–355.

Laurent P, Voiblet C, Tagu D, De Carvalho D, Nehls U, De Bellis R, Balestrini R, Bauw G, Bonfante P, Martin F. 1999. A novel class of ectomycorrhiza-regulated cell wall polypeptides in Pisolithus tinctorius. Molecular Plant–Microbe Interactions 12: 862–871.

Marschner A. 1995. Mineral Nutrition of Higher Plants, 2nd edn. San Diego, CA, USA: Academic Press.

Martin F, Boiffin V, Pfeffer PE. 1998. Carbohydrate and amino acid metabolism in the Eucalyptus globulus–Pisolithus tinctorius ectomycorrhiza during glucose utilization. Plant Physiology 118: 627–635.

Martin F, Duplessis S, Ditengou F, Lagrange H, Voiblet C, Lapeyrie F. 2001. Developmental cross talking in the ectomycorrhizal symbiosis: signals and communication genes. New Phytologist 151: 145–154.

Menotta M, Amicucci A, Sisti D, Gioacchini AM, Stocchi V. 2004. Differential gene expression during pre-symbiotic interaction between Tuber borchii Vittad. & Tilia americana L. Current Genetics 46: 158–165.

Mischiati P, Fontana A. 1993. In vitro culture of Tuber magnatum mycelium isolated from mycorrhizas. Mycological Research 97: 40–44.

Montanini B, Betti M, Marquez AJ, Balestrini R, Bonfante P, Ottonello S. 2003. Distinctive properties and expression profiles of glutamine synthetase from a plant symbiotic fungus. Biochemical Journal 373: 357–368.

Murashige T, Skoog FB. 1962. A revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiologia Plantarum 15: 473–497.

Navarro RE, Aguirre J. 1998. Posttranscriptional control mediates cell type-specific localization of catalase A during Aspergillus nidulans development. Journal of Bacteriology 180: 5733–5738.

Nehls U, Bock A, Ecke M, Hampp R. 2001. Differential expression of the hexose-regulated fungal genes AmPAL and AmMst1 within Amanita/Populus ectomycorrhizas. New Phytologist 150: 583–589.

Peter M, Courty PE, Kohler A, Delaruelle C, Martin D, Tagu D, Frey-Klett P, Duplessis S, Chalot M, Podila GK, Martin F. 2003. Analysis of expressed sequence tags from the ectomycorrhizal basidiomycetes Laccaria bicolor and Pisolithus microcarpus. New Phytologist 159: 117–129.

Podila GK, Zheng J, Balasubramanian S, Sundaram S, Hiremath S, Brand JH, Hymes MJ. 2002. Fungal gene expression in early symbiotic interactions between Laccaria bicolor and red pine. Plant and Soil 244: 117–128.

Polidori E, Agostini D, Zeppa S, Potenza L, Palma F, Sisti D, Stocchi G. 2002. Identification of differentially expressed cDNA clones in Tilia platyphyllos–Tuber borchii ectomycorrhizae using a differential screening approach. Molecular Genetics and Genomics 266: 858–864.

Rasmussen R. 2001. Quantification on the LightCycler. In: Mener S, Wittwer C, Nakagawara K, eds. Rapid Cycle Real-Time PCR Methods and Applications. Springer, Heidelberg, Germany, 21–34.

Read DB, Bengough AG, Gregory PJ, Crawford JW, Robinson D, Scrimgeour CM, Young IM, Zhang K, Zhang X. 2003. Plant roots release phospholipid surfactants that modify the physical and chemical properties of soil. New Phytologist 157: 315–321.

Ririe KM, Rasmussen RP, Wittwer CT. 1997. Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Annals of Biochemistry 245: 154–160.

Saltarelli R, Ceccaroli P, Vallorani L, Zambonelli A, Citterio B, Malatesta M, Stocchi V. 1998. Biochemical and morphological modifications during the growth of Tuber borchii mycelium. Mycological Research 102: 403–409.

Selosse MA, Faccio A, Scappaticci G, Bonfante P. 2004. Chlorophyllous and achlorophyllous specimens of Epipactis microphylla (Neottieae, Orchidaceae) are associated with ectomycorrhizal septomycetes, including truffles. Microbial Ecology 47: 416–426.

Sisti D, Giomaro G, Rossi I, Ceccaroli P, Citterio B, Stocchi V, Zambonelli A, Benedetti PA. 1998. In vitro mycorrhizal synthesis of micropropagated Tilia platyphillos Scop. plantlets with Tuber borchii Vittad. mycelium in pure culture. Acta Horticulture (ISHS) 457: 379–388.

Sisti D, Giomaro G, Cecchini M, Faccio A, Novero M, Bonfante P. 2003. Two genetically related strains of Tuber borchii produce Tilia mycorrhizas with different morphological traits. Mycorrhiza 13: 107–115.

Smith SE, Read DJ. 1997. Mycorrhizal Symbiosis, 2nd edn. Academic Press, San Diego, CA, USA.

Soragni E, Bolchi A, Balestrini R, Gambaretto C, Percudani R, Bonfante P, Ottonello S. 2001. A nutrient-regulated, dual localization phospholipase A(2) in the symbiotic fungus Tuber borchii. EMBO Journal 20: 5079–5090.

Sundaram S, Kim SJ, Suzuki H, Mcquattie CJ, Hiremah ST, Podila GK. 2001. Isolation and characterization of a symbiosis-regulated ras from the ectomycorrhizal fungus Laccaria bicolor. Molecular Plant–Microbe Interactions 14: 618–628.

Tagu D, Lapeyrie F, Martin F. 2002. The ectomycorrhizal symbiosis: genetics and development. Plant and Soil 244: 97–105.

Tsitsigiannis DI, Zarnowski R, Keller NP. 2004. The lipid body protein, PpoA, coordinates sexual and asexual sporulation in Aspergillus nidulans. Journal of Biological Chemistry 279: 11344–11353.

Viotti A, Abildsten D, Pogna N, Sala E, Pirrotta V. 1982. Multiplicity and diversity of cloned in cDNA sequences and their chromosomal localization. EMBO Journal 1: 53–58.

Voiblet C, Duplessis S, Encelot N, Martin F. 2001. Identification of symbiosis regulated genes in Eucalyptus globulus–Pisolithus tinctorius ectomycorrhiza by differential hybridization of arrayed cDNAs. Plant Journal 25: 181–191.