Ectomycorrhizin Synthesis and Polypeptide Changes during - NCBI

Plant Physiol. (1991) 97, 977-984

Received for publication March 29, 1991 Accepted June 26, 1991

0032-0889/91/97/0977/08/$01 .00/0

Ectomycorrhizin Synthesis and Polypeptide Changes during the Early Stage of Eucalypt Mycorrhiza Development1 Jean-Louis Hilbert2, Guy Costa3, and Francis Martin* Laboratoire de Microbiologie Forestiere, Centre de Recherches Forestieres de Nancy, Institut National de la Recherche Agronomique, Champenoux 54280 Seichamps, France ABSTRACT

opment, including the completion of the sexual cycle, and it contributes the enzymes that assimilate the imported nutrients. Mycorrhiza organogenesis seems to be influenced and controlled by environmental as well as genetic factors, as indicated by physiological, cytological, morphological, and genetic experiments (for reviews, see refs. 4, 10, 17). Little is known about the molecular mechanisms by which environmental and genetic factors exert their influence on, or direct, ectomycorrhiza development and differentiation. However, changes in the symbiont protein synthesis have been detected in ectomycorrhiza by examining polypeptide patterns by two-dimensional PAGE. In functioning associations, biochemical alterations are accompanied by a differential accumulation of polypeptides. These changes fell into three distinct categories: (a) decreased amounts of a large number of polypeptides, (b) increased accumulation of a few polypeptides, and (c) synthesis of SR-proteins,4 referred to as ectomycorrhizins (5). The steps involved in the symbiosis formation show that major decisions determining the development of an ectomycorrhiza are made at stages preceding the differentiation of typical anatomical features of ectomycorrhizas (i.e. the ectomycorrhizal sheath and the Hartig net). Certainly, the process of mycorrhiza initiation and early development requires changes in the expression of additional symbiont genes, but these have not been characterized. For elucidating the possible functions of SR-proteins in the developmental processes in the symbionts, it is essential to know in which developmental stage ectomycorrhizin genes are expressed. In this paper, we report on changes in polypeptide patterns and in vivo protein synthesis during the early stages of mycorrhiza organogenesis of the Eucalyptus globulus-Pisolithus tinctorius association and demonstrate that these molecular events precede any morphological changes and are not induced by an incompatible isolate.

In functioning eucalypt ectomycorrhizas, biochemical alterations are accompanied by a differential accumulation of polypeptides including the synthesis of symbiosis-related proteins (JL Hilbert, Martin FM [1988] New Phytol 110: 339-346). In the present study, protein biosynthesis in the early stages of ectomycorrhiza formation on Eucalyptus globulus subsp. bicostata Kirkp. was examined using compatible and incompatible isolates of the basidiomycete Pisolithus tinctorius (Coker & Couch). Changes in polypeptide composition were observed within hours following contact of the compatible mycelium with the roots, well before the differentiation of typical symbiotic tissues. At this stage, at least seven symbiosis-related proteins (ectomycorrhizins) accumulated in root tissues. In vivo incorporation of [35S]methionine by ectomycorrhizas followed by electrophoresis of the labeled proteins revealed that most of these differences in polypeptide concentrations, including the ectomycorrhizin accumulation, are the result of differential protein biosynthesis rather than posttranslational modifications of the polypeptides. The initial development of eucalypt ectomycorrhizas, therefore, coincides with the synthesis of symbiosis-related proteins and the data presented here provide essential evidence to ascribe a functional developmental role to these proteins.

Induction of ectomycorrhizal symbiosis on tree roots by soil fungi in the classes ascomycetes and basidiomycetes has been shown to be a highly evolved and complex process, requiring a fine-tuned interaction between compatible mycorrhizal fungi and their host plant (4, 10, 17). The fungus encodes the basic enzymatic machinery for absorbing, transporting, and assimilating major mineral ions (e.g. phosphate and inorganic nitrogen). The plant maintains a unique ecological niche that is necessary for fungal growth and devel-

MATERIALS AND METHODS Biological Materials Eucalyptus globulus subsp. bicostata (Maid et al.) Kirkp. was used as the host plant. Seeds (seedlot No. 16100) were provided by the Division of Forest Research (Commonwealth Scientific and Industrial Research Organization, Australia) from 10 parent trees at Mount Nullo (New South Wales, Australia). The fungus (Pisolithus tinctorius Coker & Couch

' This paper is dedicated to the memory of Prof. J. Harley and is the second paper of a series. This work was supported by a grant from the Institut National de la Recherche Agronomique (AIP "R&gulation du Mtabolisme des Associations Mycorhiziennes" grant No 88/4630) awarded to F.M. and by a Doctoral Fellowship from the Minist&re de la Recherche et de la Technologie to J.-L.H. 2 Present address: Physiologie de la Ditferenciation et Biotechnologies Vegetales, Universite des Sciences et Techniques de LilleFlandres-Artois, UER de Biologie, Bat SN2, 59655 Villeneuve d'Asc Cedex, France. 'Present address: Laboratoire de Phytotechnie, INRA-ENSAIA, 2 Avenue de la Foret de Haye, 54500 Vandoeuvre, France.

'Abbreviations: SR-proteins, symbiosis-related proteins; 2-ME, 2mercaptoethanol. 977

978

HILBERT ET AL.

[isolates 441 and 270]) was maintained in the collection of ectomycorrhizal fungi at the Laboratoire de Microbiologie Forestiere (Institut National de la Recherche Agronomique, Nancy Forestry Research Center, Champenoux). The isolate of P. tinctorius 270 (provided by Dr. D. H. Marx) was isolated from a sporocarp collected under Pinus taeda L. in Florida, whereas isolate 441 (provided by Dr. G. A. Chilvers) was isolated from a sporocarp collected under Eucalyptus citriodora in Brasil. The cultures were subcultured monthly on 20 mL of Pachlewski's medium in 2.0% (w/v) agar (9). Tree Seedlings

Eucalypt seeds were surface-sterilized in 20% (w/v) sodium hypochlorite for 20 min and then rinsed in several changes of sterile distilled water. Germination was carried out on 2.0% distilled water agar at 25°C in the dark for 3 d. Aseptic seedlings were then transferred on 3.0% agar containing halfstrength Pachlewski's medium. Petri dishes holding seedlings were edge-sealed with Paraflim, and seedlings were grown for an additional 4 d in a controlled environment chamber. The light/dark period was 16/8 h, and the temperature (high/low) was 22/16°C. Photosynthetic photon flux density was 90 ,tmol m-2 s-'. Seedlings with tap roots 1 to 1.5 cm in length were used for the aseptic synthesis of ectomycorrhizas according to Malajczuk et al. (8). Aseptic Synthesis of Ectomycorrhizas Large round dishes (150 x 20 mm) containing 50 mL of half-strength Pachlewski's medium in 2.0% agar were covered with washed, autoclaved cellophane discs and were inoculated with four agar plugs of P. tinctorius 441 or 270 (7-mm diameter plugs cut from the edge of 2-week-old colonies). Fungal growth took place at 25°C in the dark for 3 weeks (35mm diameter fungal colonies). Then, six 7-d-old aseptic seedlings of E. globulus were laid onto the outer portion of each fungal colony so that the root tips were in contact with the youngest part of the mycelium. Dishes holding fungal colonies together with inoculated seedlings (24/plate) were edge-sealed with Parafilm and were returned to the controlled environment chamber. To detect any potential stress-induced changes resulting from the transfer, control seedlings were manipulated and grown as the inoculated seedlings except that they were transferred on agar plates with cellophane without fungal colonies.

Sampling Twelve-, 24-, 36-, 48-, and 96-h-old inoculated roots from 24 plants, free-living mycelia, and roots from 24 noninfected seedlings were sampled, immediately frozen in liquid N2, and stored at -70°C. At all stages of the P. tinctorius 270-E. globulus interaction and at 12 and 24 h after inoculation of seedlings with isolate 441, we collected that part of the tap root in contact with the fungal mycelium where mycorrhizas normally appear, and at 36 and 48 h we harvested sheathed roots. Filee-living mycelium was sampled from the edge of the colony. Excess extramatrical mycelium of 48-h-old ectomycorrhizas was excised prior to freezing.

Plant Physiol. Vol. 97, 1991

Estimation of the Fungal Biomass Fungal biomass of inoculated roots was estimated from the ergosterol content according to Martin et al. (9). Briefly, mycelium or inoculated roots were extracted with 0.5 mL of cold (4°C) absolute ethanol for 2 to 3 min with a mortar and pestle. The extracted mixture, plus an additional 0.5-mL portion of absolute ethanol used to rinse the mortar, was poured into a 1.5-mL microfuge tube and centrifuged 5 min at 14,000g in a bench centrifuge (Heraeus Biofuge) in a cold room. The supernatant was decanted, and the pellet was resuspended in 0.5 mL of ethanol, shaken for 30 s, and centrifuged as above. The two supernatants were pooled, adjusted to 1.5 mL, and filtered through a 0.45-Am filter (Sartorius, Palaiseau, France), and the filtrate was analyzed for free ergosterol content by reverse-phase HPLC. To determine the proportion offungal biomass in the inoculated roots, a conversion factor was calculated from the ergosterol/fresh weight ratio of the mycelium sampled in the immediate vicinity (within ± 5 mm) of the roots. In Vivo Labeling of Proteins

Free-living mycelium and roots of ectomycorrhizal and noninoculated seedlings were labeled in vivo by immersion in 1 mL of sterile, half-strength Pachlewski solution containing [35S]methionine (Amersham, 1.5 MBq mL-'; 37 TBq/mol). Incubation was continued for 4 h in the growth chamber under light conditions at 22°C. After the labeling period, samples were washed in unlabeled medium containing 5 mm cold methionine, immediately frozen in liquid N2, and stored at -70C until extraction. We obtained proteins of sufficient specific radioactivity to be visualized on two-dimensional fluorograms after 3 to 7 d exposure from 8 to 10 seedlings labeled by these methods. Protein Extraction Approximately 10 mycorrhizas (or parts of the roots in contact with the mycelial mat) from 1O inoculated plants, 1O noninoculated control roots, and 10 to 15 mg free-living mycelium were used for every protein extraction. Samples were powdered in liquid N2 with a mortar and pestle. The powder was homogenized in cold acetone (-20°C) (10 mL g-' fresh weight) containing TCA (10%, w/v) and 2-ME (0.07%, v/v), and proteins were precipitated at -20C for 45 min (18). After centrifugation at l0,OOOg for 30 min, supernatants were discarded, and pellets were washed with 1 mL of cold acetone (-20°C) containing 2-ME (0.07%, v/v). After a second centrifugation (l0,OOOg for 30 min), the solution was discarded, and pellets were dried overnight under vacuum. Pellets were resuspended in O'Farrell (14) lysis buffer (30 ,uL mg-' dry weight pellet) to solubilize the proteins and then centrifuged at l0,OOOg for 3 min. Supernatants were stored at -70°C for further analysis. Protein contents of the supernatants were estimated (3). For quantitative determinations of the incorporated radioactivity, 2-AL samples were spotted on Whatman GC-A filters, which were then treated with 5 mL of ice-cold 25% TCA, washed with 2 x 5 mL of ice-cold 10% TCA, and 4 x 5 mL of 100% ethanol. The amount of

PROTEIN BIOSYNTHESIS DURING MYCORRHIZA FORMATION

radioactivity incorporated into TCA-insoluble polypeptide was measured by scintillation counting. PAGE

Two-dimensional PAGE of unlabeled and labeled proteins was performed as described by O'Farrell (14) as modified by Hilbert and Martin (5). Ampholytes were added to a final concentration of 4% and consisted of 25% ampholytes (pH 3.5 to 10; LKB) and 75% ampholytes (pH 5 to 7; Pharmacia). Samples (30 ,L) containing 150 to 300,ug of protein (100,000 to 300,000 cpm in the TCA-precipitable material) were loaded on the basic end of the tube gels. The gel was run at 1200 V for 17.5 h followed by 1500 V for 0.5 h, extruded from the glass tube, equilibrated, and loaded onto the second dimension as described by O'Farrell (14), except that 2-ME was omitted (15). Electrophoresis in the second dimension was carried out as described by Blangarin and Madjar (1) using a Hoefer Electrophoresis Multi-Cell (Bioblock, Strasbourg, France). Proteins were silver-stained as in Blum et al. (2). After electrophoresis, gels were dried in a Bio-Rad model 543 slab gel drier (Bio-Rad, Ivry-sur-Seine, France). Gels containing labeled proteins were impregnated with Amplify (Amersham, Les Ulis, France). Dried gels were autofluorographed with Hyperfilm-MP (Amersham, Les Ulis, France) and Cronex enhancing screens at -70°C. The apparent mol wt and isoelectric point of polypeptides were estimated from their migration in the gel in relation to that of standard proteins with known mol wt (Electrophoresis Calibration Kit, Pharmacia AB, Uppsala, Sweden) and iso-

electric point (Isoelectric Point Calibration Kit, BDH, Poole, UK). Evaluation of Electrophoregrams and Autoradiograms

Changes in the relative accumulation of specific polypeptides were identified by careful examination of silver-stained gels and autoradiograms. Comparison among polypeptide patterns (nonmycorrhizal roots, mycorrhizal roots, and freeliving fungus) and among gels (0, 12, 24, and 48 h after contact) were performed. The concentration and the relative rate of synthesis of individual polypeptides were evaluated on each gel and autoradiogram in relation to total protein or radioactivity in the sample, by comparison with a number of reference proteins, and to the complete protein pattern. The temporal pattern of synthesis of polypeptides was evaluated by pairwise comparisons of gels and autoradiograms derived from three to six replicate experiments with separate lots of samples. Differences between gels in the intensity of spots that were larger than twofold were considered to be of biological significance. Approximately 250 polypeptides, of the 600 to 800 that were resolved by the method, were subjected to this detailed analysis. Results of visual analysis were confirmed by quantitative densitometric analysis. Areas of the gels containing selected major spots were scanned in 256-gray scale mode using a scanner connected to an Apple Macintosh II computer and then saved as TIFF files. The TIFF files were then analyzed with the freeware program Image v. 1.33 provided by Dr. W. Rasband (National Institute of Health, Bethesda, MD). The program is available for downloading from Bitnet.

979

RESULTS In this report, we describe an analysis at the level of protein synthesis during the first phase in the development ofeucalypt ectomycorrhizas. We wanted to investigate whether the alteration in root development that results from mycosymbiont colonization is preceded by or associated with an altered pattern of protein synthesis in symbionts. We monitored the changes in polypeptide concentration and synthesis from the time of contact until the Hartig net started to form. The aim was to identify polypeptides that are differentially synthesized during the early stages of mycorrhiza formation.

Time Sequence of the Infection Process

The interaction of E. globulus and P. tinctorius 441, under our experimental conditions, results in the induction of mycorrhizal tissues in the most susceptible region of the tap root, that area which is near the root cap at the time of inoculation (6). This localization causes mycorrhiza to develop as dense sheath of about 1 cm length in this region. To assess which cell types or tissues are involved in the early stages of mycorrhiza development, root segments were observed at 12, 24, 36, 48, and 96 h after inoculation. The sequences of events in the early infection processes that were observed with the compatible isolate 441 were identical to those published recently (6-8, 11, 12). However, to serve as a frame of reference for the analysis to follow the mycorrhiza polypeptides, a brief description of mycorrhiza formation is included. Twelve hours after transferring seedlings onto the fungal mat, a few contacts were observed between the fungal hyphae and root hairs. Fungal attachment to the host epidermis was evident within 24 h postinoculation. Two days after inoculation, formation of the initial layers of the fungal sheath was evident around some part of the tap root, and hyphal penetration between epidermal cells had commenced (Hartig net initia-

0 0

20

I.. w la I"

16

0

c) U,

0

12

go

0

8

uB

4

..

._

ro 'U

0

0

20

40

60 80 Hours

100 120 140

Figure 1. Proportion of fungal biomass during the colonization of E. globulus roots by P. tinctorius. Fungal infection induced by the compatible isolate 441 (-) and the incompatible isolate 270 (0) of P. tinctorius was estimated from the ergosterol content of infected roots (9). Each value represents the average of results from four replicates

(± SD).


HILBERT ET AL.

980

4l.

V3

.i

94|

a. .s.I

A

C

X

.\..E.,

$

43A

46! 4

I

:' b ':

~ ~ 4i~~

~ 43 Pop~~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~~ -215

~

~ ~~

4

~

l~~~~~~~~l~~. W

~

~

~

~

~

2

3

042

Figure 2. Silver-stained two-dimensional gels of noninoculated roots of E. globulus (A), and ectomycorrhizas of P. tinctorius 441 -E. globulus at 24 (B) and 48 h (C) after inoculation. Two hundred micrograms of proteins of noninfected roots and slightly higher amounts (7-9% according to the fungal biomass proportion) of mycorrhiza proteins have been loaded onto the gels. At these early stages of colonization, the fungal proteins do not significantly contribute to the two-dimensional patterns. Major symbiosis-regulated polypeptides are indicated by numbers. Mb39 (El) is a microsomal membrane polypeptide that is not affected during early mycorrhiza development and, therefore, was used as a reference protein. (Y), decreased polypeptides; (A), enhanced polypeptides. Open circles in the gel of noninoculated roots (A) indicate the locations where ectomycorrhizins accumulated in inoculated roots.


981

The protein concentration of infected root tissues was not altered through the early organogenesis of mycorrhiza (data

Figure 3. Changes in major root-specific polypeptides during ectomycorrhiza development. Regions of silver-stained gels of uninfected roots (A), and inoculated roots at 12 (B), 24 (C), 36 (D), and 48 h (E) postinoculation are shown. Mb39 (F) is a microsomal membrane polypeptide that is not affected during early mycorrhiza development and, therefore, was used as a reference protein.

tion). Hyphae accumulated rapidly between 24 and 48 h. By 4 d after inoculation, the ectomycorrhizal sheath was well developed and tightly appressed to epidermal cells, and development of the Hartig net between the epidermal cells was evident. During the same period, isolate 270 showed little if any contact with the root surface (7, 8). We also used the content of the fungal sterol, ergosterol, in inoculated roots to assess the colonization of roots at various times (9). Increase in ergosterol content of the eucalypt roots was used to illustrate the fungal colonization of the host plant (Fig. 1). A lag period of at least 12 h ("preinfection stage") was required for the detection of the first quantifiable fungal material. During the "mycorrhiza-formation stage" (24-96 h), fungal biomass increased linearly, reaching a maximum (16% of infected root) at the time of dense fungal sheath development (96 h). Ergosterol was not detected in roots of seedlings growing on fungal mats of the incompatible isolate 270. Changes in Polypeptide Patterns

To study the variations in protein synthesis in the preinfection and early infection developmental stages of E. globulus ectomycorrhizas, we analyzed the protein concentration (data not shown) and protein composition by two dimensional PAGE (Fig. 2) of inoculated roots at the following four stages: preattachment (12 h), formation of light weft of hyphae (24 h), formation of the initial layers of the mycorrhizal sheath (36 and 48 h), and end of the mycorrhiza formation (96 h).

not shown). The protein composition of the roots was conserved during the preinfection stage (12 h) (data not shown), but alterations in the protein patterns of the roots occurred over the entire period of colonization (24-48 h) (Fig. 2B and C). Three sets of proteins that showed different temporal patterns of accumulation in relation to mycorrhiza development can be distinguished. The first set comprises root-specific polypeptides that were abundant in the noninfected roots but accumulated at a lower level in P. tinctorius 441-infected roots. In noninoculated roots, a cluster of abundant root-specific polypeptides that comprised a large fraction of the total protein was evident (polypeptides Mb39, 12, 13, 14, 15). Dominant species are shown in Figure 3A and have apparent molecular masses of 41 kD (polypeptide 15), 40 kD (polypeptide 14), 39.5 kD (polypeptide 12), 39 kD (Mb39), and 38 kD (polypeptide 13). Polypeptides 12, 13, 14, and 15 are soluble proteins, whereas Mb39 is a microsomal polypeptide (B. Henrion and F.M. Martin, unpublished data). The quantitatively major change in protein metabolism over the period 12 to 48 h was the fast and complete degradation of polypeptides 12, 13, 14, and 15. They decreased in amount during the formation of initial hyphal wefts (24 to 36 h) (Fig. 3C-D) and fell below detectable levels when the first layers of the ectomycorrhizal sheath were

being formed (48 h) (Fig. 3E). A second set of polypeptides (polypeptides 1, 22, and 23) (indicated in Fig. 2) accumulated at higher relative levels during the early infection period and during the first 1 to 2 d of development. In addition, seven unique polypeptides appeared during the early infection stages (polypeptides 16, 17, 18, 19, 20, 21, and 24) (Fig. 2B and C) and were observed consistently on two-dimensional gels of ectomycorrhizas as early as 24 h postinoculation (approximately 12 h after fungal attachment). They represent SR-proteins, referred to as ectomycorrhizins, which may be related to those previously characterized in fully developed mycorrhizas (5). Ectomycorrhizin accumulation and changes in polypeptide patterns were not observed over the course of the experiment in the eucalypt roots treated with isolate 270 (data not shown). Two-dimensional PAGE patterns remained identical to the root polypeptide pattern (Fig. 2A). In Vivo Protein Synthesis

To examine further these variations in polypeptide composition, in vivo protein synthesis was studied. The results showed that 150 to 200 polypeptides could be resolved by two-dimensional PAGE of radiolabeled proteins. By superimposing the fluorograms of the two-dimensional pattern of labeled, inoculated roots (Fig. 4) on the corresponding silverstained gels (Fig. 2), it appeared that many of the highly labeled polypeptides on fluorograms coincided with stained spots on silver-stained patterns even if their relative abundance may differ. Comparison of the fluorograms of inoculated roots at 48 h (middle of the mycorrhiza formation stage) and 96 h (end of the mycorrhiza formation stage) with the noninfected root two-dimensional patterns (Fig. 4) indicated that two major modifications of in vivo protein synthesis

HILBERT ET AL.

982

Plant

Physiol. Vol. 97,

1991

67@ COL0 f

43Fof104

443

4304

20) aela 1 to aid i orl

432C

1~~~~~~~~~~~~~~~~

_1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _014 hw fto-iesoalgl finVv 3S hagsi roenboythssdrngetmcrriadvlpmn.Furorpsae rm oiouae ot o .gouu A, ncltdrosa 4 B n 6h()potiouain n mehoielbldprtisioae frelvng ..lum() imlraout fraiatviyhv be ode noth es Telctin fsyboi-rgltd oyetie AFar laeld oai ointtin

Fiue4

.


983

kD

...

430

I4.

30

.E!z|IM

430

14> AS

a,

I

H

G

414

E13

At

..lb.

1 -

i

:.,, ..-

.::

s

"

Figure 5. Time-course changes in ectomycorrhizin biosynthesis during ectomycorrhiza development. Fluorographs are shown of two-dimensional gel areas showing in vivo [35S]methionine-labeled ectomycorrhizins. Noninoculated roots of E. globulus (A, G), inoculated roots at 12 (B), 24 (C), 36 (D), and 48 h (E, H) postinoculation, and free-living mycelia of P. tinctorius (F, I). Similar amounts of radioactivity have been loaded onto the gels. The locations of some nonaffected polypeptides (-*) are labeled to aid in orientation. Ectomycorrhizins are indicated by their molecular masses in kD and by thick black arrows.

occurred in differentiating mycorrhizas. First, several ectomycorrhizins were synthesized at very high rates in symbiotic tissues, whereas they were not detected in the free-living partners. These proteins were clustered in two major groups, with molecular masses in the regions of 32 and 12 kD. This latter group contained three polypeptides, E13, Ella, and El lb (Fig. 5H), whereas the 32-kD subset was composed of two polypeptides, E32a and E32b (Fig. 4B and C, Fig. 5B-E). These groups of in vivo-synthesized proteins can be very easily correlated with similar sets of spots detected in the silverstained gels (Fig. 2B and C). E32a, E32b, and E69 (polypeptides 16, 17, and 24) were already synthesized at a time (48 h) when the fungal polypeptides (Fig. 4D) were barely detected on the fluorogram, strongly suggesting that these SR-proteins accumulated in the root cells. E32a and E32b were synthesized at a low rate during the preinfection and early infection stages (Fig. 5B and C). The relative rate of synthesis was higher at 36 and 48 h (Fig. 4C; Fig. 5D and E). These symbiosis-specific polypeptides are very acidic, having an isoelectric point of approximately 3 (Fig. 4B and C). Second, there was a substantial increase in the synthesis of several polypeptides including polypeptide 1. Finally, there was a reduction in several

root (polypeptide 25) and fungal polypeptides (polypeptides b, c, d, and e) in differentiating mycorrhizas. Therefore, results of the in vivo labeling study are in agreement with the data obtained by the analysis of the protein composition of eucalypt ectomycorrhizas. Moreover, in vivo labeling data confirm that major variations in protein biosynthesis occur during early mycorrhiza development.

DISCUSSION The changes in root morphology and metabolism that result from ectomycorrhizal fungus infection are associated with extensive alterations in the rates of accumulation of a large number of proteins (5). The present results show that variations in protein synthesis are not restricted to the functioning mycorrhiza but also occur at the initiation of ectomycorrhizal development, well before the differentiation of typical symbiotic tissues. Twelve hours after fungal attachment, when there were only a few hyphae near the root cap, there was already evidence of molecular interaction between the fungus and the host root in the form of an accumulation of SRproteins, the ectomycorrhizins. Alteration of protein contents

HILBERT ET AL.

984

requires that changes in protein biosynthesis take place even earlier. This is the first indication of changes in host gene expression that occur before the formation of ectomycorrhiza anatomical features. This rapid regulation suggests that these genes may respond to some early rhizospheric signals exchanged between the symbionts similar to those recently purified in endomycorrhizal associations (13). Ectomycorrhizin accumulation and changes in polypeptide patterns were not observed when eucalypt roots were growing on the incompatible fungal isolate 270, demonstrating the selective character of these apparent alterations in protein synthesis. Two days after inoculation, after the fungus was attached to the root and the ectomycorrhizal sheath was being formed, changes in polypeptide concentrations were more pronounced. The synchronous decrease of root protein synthesis and the accumulation of ectomycorrhizins at these stages of the mycorrhizal interaction suggest that a major reprograming of protein synthesis takes place early during the initial infection process. However, the primary response of eucalypt roots to the early fungal colonization does not appear to be associated with any rapid, large-scale alteration in gene expression. This is in contrast to the extensive changes in polypeptide levels observed in eucalypt ectomycorrhizas exhibiting all the basic features of a typical mycorrhiza (i.e. ectomycorrhizal sheath and Hartig net) (5). This suggests that the formation of an ectomycorrhiza initiates a sequence of developmental events leading to the final symbiotic phenotype. The role of the ectomycorrhizins in the symbiosis differentiation has yet to be ascertained, but their synthesis is directly related to initial colonization of the root. Based on gel migration patterns, it is likely that some of the ectomycorrhizins characterized in the present study (Ella, El lb, E13) and the low mol wt ectomycorrhizins that accumulated in functioning mycorrhizas (5) are identical. Ectomycorrhizins may also be related to specific proteins synthesized in other plant-microbe interactions (e.g. pathogenesis related proteins and nodulins). Recent studies of the endomycorrhizal Glomus-soybean symbiosis (16) demonstrating an immunochemical cross-reactivity between endomycorrhizins and peribacteroid membrane nodulins of nodulated soybeans support this contention. However, the precise characterization of this homology must await isolation of the SR-proteins and further immunological studies. In conclusion, our results show that the changes in protein synthesis in ectomycorrhizal symbionts take place within a few hours after contact. The data presented here provide essential evidence to ascribe a functional developmental role to the SR-proteins we have identified. Further efforts to characterize the function of these proteins will provide materials for the dissection of the ectomycorrhiza differentiation. The mechanism for induction of SR-proteins remains to be explored, and studies including comparative characterization of mRNA levels of noninoculated and mycorrhizal roots will be necessary to examine the symbiosis-regulated gene expression. Efforts are now underway to address this issue. ACKNOWLEDGMENTS

We thank Dr. Bernard Dell (Murdoch University, Perth, Australia), Dr. Frederic Lapeyrie (I.N.R.A., Nancy, France) and Dr. Raymond


Pacovsky (MSU, East Lansing, MI) for helpful conversations and critical reading of the manuscript, and Dominique Vairelles for photographs. We are particularly indebted to Dr. Nicolas Malajczuk (CSIRO, Perth, Australia) for having introduced us to the aseptic synthesis of eucalypt ectomycorrhizas. 1.

2.

3. 4.

5.

6. 7.

8.

9. 10. 11.

12.

13.

14. 15.

16.

17.

18.

LITERATURE CITED Blangarin P, Madjar JJ (1986) Simplification et optimisation de certains parametres intervenant au cours de l'electrophorese bidimensionnelle en gel de polyacrylamide. Applications aux proteines seriques. In MM Galteau, G Siest, eds, Recent Progresses in Two-Dimensional Electrophoresis, Presses Universitaires de Nancy, Nancy, pp 25-36 Blum H, Beier H, Gross HJ (1987) Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels. Electrophoresis 8: 93-99 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 Gianinazzi-Pearson V, Gianinazzi S (1989) Cellular and genetical aspects of interactions between hosts and fungal symbionts in mycorrhizae. Genome 31: 336-341 Hilbert JL, Martin FM (1988) Regulation of gene expression in ectomycorrhizas. I. Protein changes and the presence of ectomycorrhiza-specific polypeptides in the Pisolithus-Eucalyptus symbiosis. New Phytol 110: 339-346 Horan DP, Chilvers GA, Lapeyrie F (1988) Time-sequence of the infection process in eucalypt ectomycorrhizas. New Phytol 109: 451-458 Lei J, Lapeyrie F, Malajuczuk N, Dexheimer J (1990) Infectivity of pine and eucalypt isolates of Pisolithus tinctorius on roots of Eucalyptus urophylla in vitro. II. Ultrastructural and biochemical changes at the early stage of mycorrhiza formation. New Phytol 116: 115-122 Malajczuk N, Garbaye J, Lapeyrie F (1990) Infectivity of pine and eucalypt isolates of Pisolithus tinctorius on roots of Eucalyptus urophylla in vitro. 1. Mycorrhiza formation in model systems. New Phytol 114: 627-631 Martin FM, Delaruelle C, Hilbert JL (1990) An improved ergosterol assay to estimate fungal biomass in ectomycorrhizas. Mycol Res 94: 1059-1064 Martin FM, Hilbert JL (1991) Morphological, biochemical and molecular changes during ectomycorrhiza development. Experientia 47: 321-331 Massicotte HB, Peterson RL, Ashford AE (1987) Ontogeny of Eucalyptus pilularis-Pisolithus tinctorius ectomycorrhizae. I. Light microscopy and scanning electron microscopy. Can J Bot 65: 1927-1939 Massicotte HB, Peterson RL, Ackerley CA, Ashford AE (1987) Ontogeny of Eucalyptus pilularis-Pisolithus tinctorius ectomycorrhizae. II. Transmission electron microscopy. Can J Bot 65: 1940-1947 Nair MG, Safir GR, Siqueira JO (1991) Isolation and identification of vesicular-arbuscular mycorrhiza-stimulatory compounds from clover (Trifolium repens) roots. Appl Environ Microbiol 57: 434-439 O'Farrell PH (1975) High resolution two dimensional electrophoresis of proteins. J Biol Chem 250: 4007-4021 Tasheva B, Dessev G (1983) Artifacts in sodium dodecyl sulfate polyacrylamide gel electrophoresis due to 2-mercaptoethanol. Anal Biochem 129: 98-102 Wyss P, Mellor RB, Wiemken A (1990) Vesicular-arbuscular mycorrhizas of wild-type soybean and non-nodulating mutants with Glomus mosseae contain symbiosis-specific polypeptides (mycorrhizins), immunologically cross-reactive with nodulins. Planta 182: 22-26 Wong KKY, Fortin JA (1990) Root colonization and intraspecific mycobiont variation in ectomycorrhiza. Symbiosis 8: 210-223 Zivy M (1986) Influence des ampholytes sur la revelation des proteines au nitrate d'argent. In MM Galteau, G Siest, eds, Recent Progresses in Two-Dimensional Electrophoresis, Presses Universitaires de Nancy, Nancy, pp 69-72