PYK10, a β-glucosidase located in the endoplasmatic reticulum, is crucial for the beneficial interaction between Arabidopsis thaliana and the endophytic fungus Piriformospora indica

PYK10, a b-glucosidase located in the endoplasmaticreticulum, is crucial for the beneficial interaction betweenArabidopsis thaliana and the endophytic fungusPiriformospora indica

Irena Sherameti1, Yvonne Venus1, Corinna Drzewiecki1, Swati Tripathi2, Vipin Mohan Dan2, Inke Nitz3, Ajit Varma2,

Florian M. Grundler3 and Ralf Oelmuller1,*

1Friedrich-Schiller-Universitat Jena, Institut fur Allgemeine Botanik und Pflanzenphysiologie,

Dornburger Str. 159, 07743 Jena, Germany,2Amity Institute of Herbal and Microbial Studies, Sector 125, Noida 201303, UP, India, and3Institute of Plant Protection, Department of Applied Plant Sciences and Plant Biotechnology,

BOKU – University of Natural Resources and Applied Life Sciences Vienna, Peter Jordan-Strasse 82, A-1190 Vienna, Austria

Received 13 December 2007; revised 16 January 2008; accepted 18 January 2008.*For correspondence (fax +49 3641 949232; e-mail [email protected]).

Summary

Piriformospora indica, an endophyte of the Sebacinaceae family, promotes growth and seed production of

many plant species, including Arabidopsis. Growth of a T-DNA insertion line in PYK10 is not promoted and the

plants do not produce more seeds in the presence of P. indica, although their roots are more colonized by the

fungus than wild-type roots. Overexpression of PYK10 mRNA did not affect root colonization and the response

to the fungus. PYK10 codes for a root- and hypocotyl-specific b-glucosidase/myrosinase, which is implicated to

be involved in plant defences against herbivores and pathogens. Expression of PYK10 is activated by the basic

helix-loop-helix domain containing transcription factor NAI1, and two Arabidopsis lines with mutations in the

NAI1 gene show the same response to P. indica as the PYK10 insertion line. PYK10 transcript and PYK10

protein levels are severely reduced in a NAI1 mutant, indicating that PYK10 and not the transcription factor

NAI1 is responsible for the response to the fungus. In wild-type roots, the message level for a leucine-rich

repeat protein LRR1, but not for plant defensin 1.2 (PDF1.2), is upregulated in the presence of P. indica. In

contrast, in lines with reduced PYK10 levels the PDF1.2, but not LRR1, message level is upregulated in the

presence of the fungus. We propose that PYK10 restricts root colonization by P. indica, which results in the

repression of defence responses and the upregulation of responses leading to a mutualistic interaction

between the two symbiotic partners.

Keywords: growth promotion, NAI1, Piriformospora indica, plant/microbe interaction, PYK10, Sebacinaceae.

Introduction

The endophytic fungus Piriformospora indica, a basidio-

mycete of the Sebacinaceae family, interacts with many

plant species, including Arabidopsis. Like other members of

the Sebacinaceae, P. indica colonizes the roots, grows inter-

and intracellularly, and forms pear-shaped spores that

accumulate in the roots as well as on the root surface. The

endophyte promotes nutrient uptake, allows plants to sur-

vive under water and salt stress, confers resistance to toxins,

heavy metal ions and pathogenic organisms, and stimulates

growth and seed production (cf. Oelmuller et al., 2004, 2005;

Peskan-Berghofer et al., 2004; Pham et al., 2004; Sahay and

Varma, 1999; Shahollari et al., 2005, 2007; Sherameti et al.,

2005; Varma et al., 1999, 2001; Verma et al., 1998; Waller

et al., 2005). P. indica is a cultivable fungus and can grow on

synthetic media without a host (Peskan-Berghofer et al.,

2004; Varma et al., 2001). The host range includes trees,

agricultural, horticultural and medicinal plants, monocots

and dicots, and mosses (Barazani et al., 2005; Glen et al.,

428 ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd

The Plant Journal (2008) 54, 428–439 doi: 10.1111/j.1365-313X.2008.03424.x

2002; Peskan-Berghofer et al., 2004; Shahollari et al., 2005,

2007; Sherameti et al., 2005; Varma et al., 2001; Waller et al.,

2005; Weiss et al., 2004), suggesting that the interaction is

based on general recognition and signalling processes.

Our goal was to identify plant genes that are targeted by

the fungus. Thus, we screened for Arabidopsis mutants that

do not respond to the fungus with regard to growth

promotion and enhanced seed production (Oelmuller et al.,

2004; Shahollari et al., 2007). Here we describe Arabidopsis

mutants that are impaired in the expression of PYK10, a gene

for an abundant myrosinase located in the endoplasmatic

reticulum (ER; Nitz et al., 2001; Matsushima et al., 2004).

PYK10 has recently been identified as a target of P. indica in

Arabidopsis roots (Peskan-Berghofer et al., 2004). Matsushi-

ma et al. (2003a) have shown that PYK10 is a major protein

in spindle-shaped structures of�10 lm in length and�1 lm

in width, which they named ER bodies (cf. Haseloff et al.,

1997; Hawes et al., 2001; Hayashi et al., 1999; Ridge et al.,

1999). Similar structures have been reported for more than

50 plant species (Behnke and Eschlbeck, 1978; Bones et al.,

1989; Bonnett and Newcomb, 1965; Gunning, 1998; Iversen,

1970). ER bodies are surrounded by ribosomes (Hayashi

et al., 1999) and are highly enriched in cotyledons, hypotoc-

yls and in the roots of young seedlings (Matsushima et al.,

2002). During later phases, the ER bodies decrease in the

cotyledons, whereas they remain constant in hypocotyls and

roots. ER bodies can also be induced in rosette leaves by

jasmonate (McConn et al., 1997), and the jasmonate-insen-

sitive coronatine insensitive1 (coi1; Xie et al., 1998) mutant

does not form ER bodies (Matsushima et al., 2002).

The physiological role and natural substrate(s) of PYK10

are unknown at present. b-Glucosidases and myrosinases

hydrolyze b-glucosidic bonds of aryl and alkyl b-D-gluco-

sides, as well as glucosides with carbohydrate moieties such

as cellobiose and other b-linked oligosaccharides (Esen,

1993). In particular, myrosinases hydrolyse non-toxic gluc-

osinolates to biologically active isothiocyanates, thiocya-

nates, nitriles or epithio nitriles (cf. Bones and Rossiter, 1996;

Poulton, 1990; Rask et al., 2000; Wittstock and Halkier, 2002),

and it is believed that the biological function of a myrosinase

depends upon the nature of the aglycon moieties released

from the substrates. The best-studied role of these agylcons

is their involvement in plant defence against herbivores and

microbes (Rask et al., 2000; Stotz et al., 1999, 2000; Tierens

et al., 2001), although they are also involved in the synthesis

of naturally occurring pesticides (Bones and Rossiter, 1996;

Poulton, 1990), the activation of glycosylated plant hor-

mones (Brzobohaty et al., 1993; Schliemann, 1984; Smith

and van Staden, 1978) and cell-wall catabolism (cf. Dharma-

wardhana et al., 1995; Leah et al., 1995; and references

therein). The phosphate inducibility of the PSR3.2 b-gluco-

sidase from Arabidopsis (Malboobi and Lefebvre, 1997) also

suggests an involvement of these enzymes in phosphate

metabolism. Furthermore, Zeng et al. (2003) have shown

that myrosinase activity stimulates the growth of ectomy-

corrhiza fungi. Here, we present evidence that PYK10 is

required for P. indica-mediated growth promotion and

higher seed yield in Arabidopsis because the enzyme

restricts root colonization by the fungal hyphae.

Results

Growth of mutants impaired in PYK10 accumulation

is not promoted by P. indica

In order to identify genes and proteins in Arabidopsis that

are required for growth stimulation and higher seed yield

induced by P. indica, we isolated mutants that grow in the

presence of the fungus like the uninfected wild-type plants

(cf. Oelmuller et al., 2004, 2005; Shahollari et al., 2007).

Three of these mutants are described here. One of them has

a T-DNA insertion in PYK10 (At3g09260, N871638). In the

presence of the fungus, seedlings of this line grow like wild-

type seedlings without the fungus (Figure 1), and after

transfer to soil the plants do not produce significantly more

seeds than the uncolonized wild-type controls (Table 1).

Wild-type plants produce > 20% more seeds in the presence

of the fungus (cf. Shahollari et al., 2007). Comparable results

were obtained for Sebacina, another member of the Seba-

cinaceae (data not shown). This suggests that PYK10, a

glycosyl b-D-hydroxylase/myrosinase located in the ER, is

required for beneficial effects induced by P. indica and

Sebacina in Arabidopsis.

Further analysis of the PYK10 insertion line found that root

growth of the seedlings was not stimulated by P. indica over

a period of 22 days of co-cultivation (Figure 1b). The mor-

phology of the roots as well as the ratio of lateral to main

roots was comparable with the wild type. The branching of

the roots of 22-day-old wild-type seedlings grown in the

absence of P. indica (58.3 � 4.4 lateral roots on 17.4 � 0.5-

cm-long main roots) was almost identical to that of the roots

of the PYK10 insertion line grown in the absence or presence

of P. indica (absence, 56.3 � 4.2 lateral roots on 17.0 � 0.6-

cm-long main roots; presence, 58.4 � 5.0 lateral roots on

16.9 � 0.7-cm-long main roots). When the mutant seedlings

were transferred to soil, some growth promotion was

observed; however, the response was less compared with

the wild type (Figure 2).

The importance of PYK10 for the beneficial interaction

between Arabidopsis and P. indica is further supported by

an ethylmethane–sulfonate (EMS) mutant, called P. indica-

insensitive-4 (pii-4). Similar to the PYK10 insertion line, pii-4

did not respond to P. indica at the seedlings stage (Fig-

ure 1a), and root growth was comparable with uncolonized

control seedlings (Figure 1b). Adult plants responded to the

fungus, although less than the wild-type plants (Figure 2).

Furthermore, the PYK10 mRNA level was severely reduced

in pii-4 (Figure 3), although the PYK10 gene was not affected

PYK10 in P. indica–A. thaliana interaction 429

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

by the EMS mutagenesis (data not shown). PYK10 is located

on chromosome 3, whereas the pii-4 mutation was mapped

on chromosome 2 using the pARMS set (Schaffner, 1996).

The CAPS markers PhyB (hy3) and T9D9 were used to

position the mutation in the middle of the chromosome,

within a 4.78-million-bp region, between the positions

8.146713 and 12.930159. This region contains NAI1. As the

NAI1 mRNA level was at the detection limit in roots of the pii-

4 mutant (Figure 3), we sequenced a genomic PCR product

of NAI1 from pii-4. No difference from the wild-type

sequence within the coding region and the introns could

be detected. This suggests that regulatory elements

required for the expression/transcription of NAI1 or for the

stability of the NAI1 message might be mutated in pii-4. We

sequenced approximately 900 bp both upstream and down-

stream of the ATG and stop codons, respectively, and found

an 8-bp deletion directly upstream of the ATG codon (NAI1

in pii-4, gtcaaaagagttcttgtaATG; NAI1 in wild type, gtcaaaa-

gagaaaaagagttcttgtaATG; ATG is the start codon). Whether

this deletion is responsible for the low NAI1 mRNA level in

pii-4 (Figure 3) remains to be determined.

To confirm that NAI1 is required for the response of

Arabidopsis seedlings to P. indica and the expression of

PYK10, we analysed a T-DNA insertion line (N397417). The

Table 1 Seed production (expressed in mg/plants) of wild-typeArabidopsis plants as well as of N871638 (T-DNA insertion line inPYK10), pii-4, N397417 (T-DNA insertion line in NAI1), and N341573(T-DNA insertion line in PBP1) grown in the presence and absence ofPiriformospora indica

Seeds (%)

)P. indica þP. indica

Wild type 100.00 � 2.02 121.49 � 3.32N871638 (DPYK10) 104.11 � 2.35 110.21 � 2.42pii-4 97.56 � 2.34 101.18 � 3.14N397417 (DNAI1) 96.37 � 3.13 100.56 � 3.24N341573 (DPBP1) 103.66 � 4.01 129.66 � 4.73

Seed production of wild-type plants was taken as 100%(154.9 � 3.3 mg seeds plant)1), and the other values were expressedrelative to it. In all cases, 1000 seeds weighed 18.23 � 0.03 mg.Stimulation of seed production by P. indica was significantly lower(P < 0.01) for the N871638, pii-4 and N397417 mutants compared withthe wild-type and the N341573 mutant.

(a)

(b)

Figure 1. N871638 (T-DNA insertion in PYK10), pii-4 and N397417 (T-DNA

insertion in NAI1) seedlings do not respond to Piriformospora indica.

(a) Wild-type, N871638, pii-4 and N397417 seedlings, which were grown in the

absence () P. indica) or presence (+ P. indica) of P. indica for 10 days.

(b) Analysis of the growth response of Arabidopsis roots [DPYK10 (N871638,

A), pii-4 (B) and DNAI1 (N397417, C)] cultivated on agar with a nylon net in

glass jars. Closed (open) symbols, seedlings grown in the presence (absence)

of P. indica: re, wild-type controls; h, mutants. Data are based on eight

independent experiments (number of plants per experiment was 50); bars

represent SEs.

430 Irena Sherameti et al.

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

homozygote mutant contained no NAI1, and severely

reduced PYK10 transcript levels in the roots (Figure 3). The

response to P. indica was identical to that of the PYK10

insertion line and pii-4: no growth promotion in response to

P. indica was observed at the seedlings stage (Figure 1a),

and P. indica-colonized adult plants showed little growth

response to the fungus after transfer to soil (Figure 2). In

addition, seed production was not significantly higher

compared with the uncolonized control, again comparable

with the results obtained for the PYK10 insertion line and pii-

4 (Table 1). As the PYK10 mRNA level was also reduced in

the NAI1 T-DNA insertion line (Figure 3), it is likely that

PYK10, and not NAI1, is primarily required for P. indica-

mediated growth promotion in Arabidopsis.

Analysis of an extract enriched in plasma membrane

proteins from roots by two-dimensional gel electrophoresis

uncovered several spots that were severely reduced in pii-4

compared with the wild type. Most obvious was the

reduction of an abundant spot of approximately 60 kDa

and with a pI value of 6.5 (Figure 4, marked ‘1’). Mass

spectrometrical analysis uncovered that this spot

Figure 2. Phenotypes of Arabidopsis plants co-cultivated with Piriformospora

indica [wild type as well as the DPYK10 (N871638), pii-4 and DNAI1 (N397417)

mutants] 4 weeks after transfer to soil. ) P. indica, uncolonized control plants;

+ P. indica, plants that were checked for root colonization before transfer to

soil. Transfer to soil occurred after 20 days of co-cultivation with the fungus in

glass jars; at that time point, P. indica-colonized wild-type seedlings were

bigger than the uncolonized control (cf. Fig. 1), whereas the mutant seedlings

grown in the absence or presence of P. indica were indistinguishable.

Figure 3. Real-time PCR analysis of PYK10, PBP1 and NAI1 transcript levels in

roots of 20-day-old seedlings of the wild type, the PYK10 T-DNA insertion line

N871638, pii-4 and the NAI1 T-DNA insertion line N397417. An equal quantity

of cDNA was used for real-time PCR with gene-specific primers for PYK10,

PBP1 and NAI1. The actin gene was used as a control (data not shown), and its

mRNA levels differs <5% in the individual cDNA samples. Fold induction

values of the gene were calculated with the DDCP equation of Pfaffl (2001) and

are expressed relative to the mRNA level of wild-type seedlings grown in the

absence of Piriformospora indica (set as 1.0). The data are based on four

independent experiments and the error bars represent SEs.

1

2

WT pii-4

Figure 4. Two-dimensional electrophoresis gels of plasma-membrane--

enriched protein preparations from wild-type (WT) and pii-4 roots. The WT

and pii-4 seedlings were grown on MS medium in glass jars for 20 days

before the roots were harvested. After separation on two-dimensional gels

and staining with silver, the two spots ‘1’ and ‘2’, which are present in WT and

missing in pii-4 extracts, were analysed my mass spectrometry: ‘1’, PYK10; ‘2’,

PBP1. For details, see text.

PYK10 in P. indica–A. thaliana interaction 431

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

corresponds to PYK10 [calculated molecular weight (pI) of

processed protein, 57.2 kDa (6.3); identified peptides,

IGIAHSPAWFEAHDLADSQDGASIDR; EYADFVFQEYGGK;

SGYEAYLVTHNLLISHAEAVEAYR; At3g09260). A second

spot was also reduced in pii-4 (Figure 4, marked ‘2’). The

two identified peptides (QLTAFGSDDGTVWDDGAYVGV and

STLLGFEEFVLDYSEYITAVDGTYD) correspond to a recently

identified PYK10-binding protein (PBP1, At3g16420; Nagano

et al., 2005). PBP1 has a calculated molecular weight of

approximately 32 kDa and a pI of 5.5, and is a jacalin lectin

family protein. Although PYK10 is located in the ER, PBP1 is

a cytoplasmic protein (Nagano et al., 2005). Real-time PCR

analysis for pii-4 demonstrated that not only the reduction in

PYK10, but also a reduction in PBP1 protein levels, might be

caused by fewer messages (Figure 3). The PYK10 message

level is reduced by approximately 80%, and that of PBP1 is

reduced by more than 50%. As a homozygote knock-out line

of At3g16420 (insertion in an exon; N341573), which codes

for PBP1, responded to P. indica in a similar way as the wild

type (data not shown), it appears that PYK10 rather than

PBP1 is crucial for the interaction.

P. indica promotes growth of PYK10 overexpressors

The PYK10 cDNA was overexpressed in Arabidopsis under

the control of the 35S promoter. Two of the lines that

showed the highest PYK10 mRNA levels were further anal-

ysed. The mRNA levels in the shoots were significantly

higher than in the wild type, because the low expression

level of the PYK10 gene under the control of its own

promoter was circumvented by the use of the 35S CaMV

promoter. A significant increase in the PYK10 mRNA level

was also observed in roots, although the stimulatory effect

was less because of the high activity of the endogenous

promoter in the root tissue (Figure 5a). Growth and the

phenotype of the overexpressors did not differ from the wild

type. Furthermore, promotion of root and shoot growth

by the fungus was comparable with the wild type (Fig-

ure 5b). We conclude that the higher PYK10 mRNA levels in

the overexpressor lines have no effect on P. indica-induced

growth promotion in Arabidopsis.

PYK10 restricts root colonization

To test whether modulation of the PYK10 level has an effect

on root colonization, we determined the quantity of the

fungal translation elongation factor 1 (Pitef1) mRNA relative

to the plant actin mRNA in the roots of colonized seedlings

with altered PYK10 levels. We noticed that the Pitef1/actin

mRNA ratio was higher in roots with reduced PYK10 mRNA

levels when the seedlings were co-cultivated with the fungus

for more than 5 days. At this time point, the analysed roots

were surrounded by a thin layer of fungal mycelia on the

agar plates. Quantitative data based on real-time PCR anal-

yses were obtained from roots that were co-cultivated with

the fungus for10 days. Roots with reduced PYK10 mRNA

levels contain significantly more Pitef1 mRNA when

compared with the wild type and with the overexpressor

lines ox-1 and ox-2 (Figure 6b). We propose that lower PYK10

levels result in better root colonization.

Furthermore, the message level for the leucine-rich-

repeat containing atypical receptor protein LRR1 is tran-

siently upregulated in wild-type roots in response to

P. indica and Sebacina (Shahollari et al., 2005, 2007). Real-

time PCR analysis demonstrates that this response does

not occur in the three mutant lines with reduced PYK10

mRNA levels (Figure 7). In contrast, the message level for

plant defensin 1.2 (PDF1.2), which codes for an antimicro-

bial defensin (Penninckx et al., 1996), does not respond to

P. indica in wild type, but is upregulated in these mutants

in the presence of the fungus (Figure 7). This indicates

that PYK10 restricts the expression of PDF1.2 and allows

the upregulation of the fungus-inducible LRR1 gene (see

Discussion).

PYK10 and PYK10-like proteins in the

Arabidopsis–P. indica interaction

PYK10 and PYK10-like proteins are encoded by a least 19

genes in Arabidopsis; however, transcripts for only 11 of

them can be detected in roots. Their absolute mRNA

Figure 5. Analysis of the PYK10 overexpressor line. Upper panel: RT-PCR

products of PYK10 isolated from the shoots and roots of 20-day-old PYK10-

overexpressor-1 (ox-1) and wild-type (WT) seedlings, based on equal quan-

tities of actin mRNA, which is not shown. Lower panel: the PYK10 overex-

pressor was grown in the absence () P. indica) or presence (+ P. indica) of

Piriformospora indica for 10 days. Results are representative for four

independent experiments.

432 Irena Sherameti et al.

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

levels differ substantially. Approximately 60% of all tran-

scripts for PYK10 and PYK10-like proteins in roots derive

from PYK10 (At3g09260; Figure 8). Among the 11 genes is

also PENETRATION2 (PEN2, At2g44490), which codes for

an enzyme that restricts pathogen entry into plant leaf

cells (Lipka et al., 2005; cf. Discussion); however the PEN2

mRNA level in the roots represents less than 10% of the

total PYK10 and PYK10-like mRNA level. None of the

transcript levels was significantly regulated by P. indica in

the wild type (Figure 8). The same was observed for the

PBP1 and NAI1 transcript levels (data not shown). Also,

transgenic lines expressing the uidA gene under the

control of the PYK10 (At3g09260) promoter did not show

a higher GUS activity in Arabidopsis roots co-cultivated

with the fungus (data not shown). The identified peptides

for PYK10, which are missing in pii-4, matched only with

At3g09260. Thus, it appears that PYK10 cannot be

replaced by PYK10-like proteins in the Arabidopsis–P.

indica interaction. This might be because PYK10 is the

most abundant b-glucosidase in the roots, or because the

different enzymes catalyze different reactions.

(a)

(b)

Figure 6. The transcript levels of the fungal translation elongation factor 1

(cPitef1) and genomic DNA (gPitef1) in the roots of colonized Arabidopsis

seedlings was compared with the levels of the plant actin nucleic acids.

(a) RNA and/or DNA were isolated from the roots of N871638 (DPYK10), pii-4,

N397417 (DNAI1), wild type (wt) and the overexpressor lines ox-1 and ox-2 (cf.

Experimental procedures). Co-cultivation of both organisms was performed

for 10 days. After reverse transcription, cPitef1 (38 cycles) and actin

(20 cycles) were amplified (left six lanes). For the right three lanes, genomic

DNA was amplified with the same primers. The sizes of the fragments are

given.

(b)To obtain quantitative data, real-time PCR was performed. The actin-

mRNA-normalized Pitef1 transcript levels of the different lines are expressed

relative to the level in wild-type roots, which was taken as 1.0. Data are based

on four independent experiments, and dots indicate a significant difference to

WT at P < 0.05 (•) and P < 0.001 (••).

WT N871638 pii-4 N397417

3(a)

(b)

2

1

0

Rel

ativ

e P

DF

1.2

mR

NA

leve

lsR

elat

ive

LRR

1 m

RN

A le

vels

WT N871638 pii-4 N397417

20

15

10

5

0

Figure 7. The LRR1 (At5g16590) message level is upregulated in response to

Piriformospora indica in Arabidopsis roots, but not in pii-4 and the insertion

lines N871638 (DPYK10) and N397417 (DNAI1).

The PDF1.2 mRNA level is regulated in the opposite direction. Arabidopsis

seedlings were grown in the absence (black bars) or presence (white bars) of

P. indica. After 6 days of co-cultivation, RNA was isolated from the roots and

used for cDNA synthesis. An equal quantity of cDNA was used for real-time

PCR with gene-specific primers for LRR1, PDF1.2 and the actin gene (Pfalz

et al., 2006). Fold induction values of the gene were calculated with the DDCP

equation of Pfaffl (2001), and are expressed relative to the mRNA level of wild-

type (WT) seedlings grown in the absence of P. indica (set as 1.0). The actin

values (not shown) differed less than 5% in the individual cDNA samples. The

data are based on four independent experiments and the error bars represent

SEs.

5000

4000

3000

2000

1000

0

At3g09

260

At1g66

280

At1g52

400

At2g44

480

At2g44

450

At3g03

640

PEN2

At3g60

140

At3g60

130

At2g32

860

At1g26

560

Sign

al

Figure 8. Relative expression of PYK10 and PYK10-related genes in Arabid-

opsis roots that were co-cultivated with Piriformospora indica for either 2 or

6 days.

Based on three (2 days) and two (6 days) independent microarray analyses,

the data represent average values. The absence of the response to P. indica

was also confirmed by semi-quantitative PCR analyses for the five most

abundantly expressed PYK10 and PYK10-related genes (data not shown).

Black (light grown), 2 days without (with) P. indica; white (dark grown),

6 days without (with) P. indica.

PYK10 in P. indica–A. thaliana interaction 433

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

Discussion

PYK10 is required for the beneficial interaction

between Arabidopsis and P. indica

Three mutants (the insertion lines in PYK10 and NAI1, as

well as the EMS mutant pii-4) demonstrate that PYK10 is

required for growth promotion and higher seed yield

induced by P. indica in Arabidopsis (Figures 1–2, Table 1),

whereas higher PYK10 mRNA levels did not affect the

beneficial interaction. Gene mapping data, in combination

with sequence analyses, indicate that the reduced level of

PYK10 in pii-4 is probably caused by the downregulation of

NAI1, which codes for a basic helix-loop-helix-type tran-

scription factor required for PYK10 expression (Matsushima

et al., 2004). This was confirmed by the analysis of an NAI1

insertion line, which also contained reduced levels of

PYK10 transcripts (Figure 3, as reported by Matsushima

et al., 2003a,b, 2004) and exhibited the same response to

P. indica as pii-4 (Figures 1–3, Table 1). Thus, PYK10 and

not the transcription factor, appears to be required for the

response to P. indica. Interestingly, the response to the

fungus is identical in mutants with reduced PYK10 tran-

script levels (in pii-4 and the NAI1 insertion line) and in the

mutant that lacks PYK10 transcripts completely because of

an insertion in a PYK10 exon (Figures 1–4). Also, higher

PYK10 mRNA levels had no effect on the growth response

induced by P. indica. This indicates that the beneficial ef-

fects are not linearily correlated to the levels of PYK10 in

the roots, if higher PYK10 mRNA levels lead to more PYK10,

which accumulates at the required place in the root cell.

Furthermore, we did not observe higher mRNA levels for

PYK10 in roots co-cultivated with the fungus. Also, the blue

stain in root cells expressing uidA under the control of the

PYK10 promoter was not visibly increased in those cells,

which are in contact with fungal hyphae (data not shown).

This suggests that the PYK10 level in wild-type roots is

sufficient to protect the plant against over-colonization;

however, severe reduction in the PYK10 protein level leaves

the cells unprotected against invading fungal hyphae, and

results in a less beneficial interaction (cf. below). Finally,

other PYK10-like proteins cannot replace the missing PYK10

in the roots.

PYK10 is an abundant b-glucosidase of 65 kDa in the ER

bodies with the ER-retention signal KDEL (Matsushima

et al., 2003b), and the protein is also found in plasma-

membrane-enriched protein preparations from Arabidopsis

roots (Figure 4; cf. Peskan-Berghofer et al., 2004). This

suggests that this fraction contains endomembranes and

proteins that interact with endomembrane-associated pro-

teins. Alternatively, PYK10 is released from the endosomal

system and reacts with PBP1, forming a multimeric complex

that is partially present in our plasma-membrane prepara-

tion. The substrate(s) of PYK10 might be separated from the

enzyme through membranes (cf. references in Nagano et al.,

2005). Thus, destruction of the cell and cellular compart-

ments is required to bring these components together. It is

likely that this occurs during root colonization, after the two

organisms come into contact with each other. Some

b-glucosidases form multimeric complexes of more than

1000 kDa, which led to the idea that they might also be

involved in structural organization within cells (Fieldes and

Gerhardt, 1994; Kim et al., 2000; Nisius, 1988; Selmar et al.,

1987). This might also play a role in our interaction system,

as hosting fungal hyphae requires substantial reorganiza-

tion in the root cell.

PBP1, another protein that is reduced in pii-4 (Figure 4), is

located in the cytoplasm and consists of two repeated

regions, each of which is highly homologous to the a chain

of jacalin, a carbohydrate-binding lectin of jackfruit. As PBP1

can be detected in our plasma-membrane-enriched protein

fraction, it appears to bind to other components. It is known

that PBP1 binds PYK10 in damaged Arabidopsis tissue

(Nagano et al., 2005). Other myrosinases also interact with

myrosinase-binding proteins (Falk et al., 1995; Geshi and

Brandt, 1998; Lenman et al., 1990), which contain jacalin-like

lectin domains with lectin activities (cf. Taipalensuu et al.,

1997). However, inactivation of PBP1 did not affect the

beneficial interaction between Arabidopsis and P. indica.

Proposed role of PYK10 in the beneficial interaction

between Arabidopsis and P. indica

The comparative analysis of fungal Pitef1 and root actin

mRNA levels suggests that the beneficial interaction

between Arabidopsis and P. indica is based on a highly

sophisticated balance between the two symbiotic part-

ners. It is conceivable that increasing quantities of fungal

hyphae lead to a degree of root colonization that pro-

vokes plant defense responses and represses beneficial

responses, whereas decreasing quantities of hyphae in

the root environment results in suboptimal exchanges of

information and nutrients between the two partners. This

resembles mycorrhizal symbioses, in which initially acti-

vated defense responses against the symbiont are

reduced during later phases of the interaction, or are even

actively repressed (cf. Pozo and Azcon-Aguilar, 2007). The

P. indica–Arabidopsis interaction system described here

might help to identify plant components that control root

colonization, and that determine whether a symbiosis is

mutualistic or parasitic.

Although the role of PYK10 in the interaction between

Arabidopsis and P. indica is unclear at present, the obser-

vation that Arabidopsis lines with reduced PYK10 protein

levels are more susceptible to fungal colonization/associ-

ation supports the idea that the enzyme is involved in

defending the root cells against an excess of invading

hyphae, which could result in a disturbance of the

434 Irena Sherameti et al.

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

balanced mutualistic interaction. Two lines of evidence

support this idea. (i) PYK10 exhibits striking sequence

similarities to PEN2, a glycosyl hydrolase, which restricts

pathogen entry of two ascomycete powdery mildew fungi

into Arabidopsis leaf cells (Lipka et al., 2005). Like PEN2,

PYK10 belongs to the class of glycosyl hydrolase family 1,

both proteins are located in intracellular organellar struc-

tures (PYK10 in ER bodies and PEN2 in peroxisomes), and

both proteins share a high degree of sequence similarity.

The catalytic domains of both proteins contain two

conserved nucleophilic glutamates. Lipka et al. (2005) have

shown that glutamate183 is required for PEN2 function

in vivo, which suggests that PEN2 catalytic activity is

required for restricting pathogen entry. Thus, PYK10 might

have a similar biological function in our system. (ii) The

beneficial traits in this symbiosis are highly dependent on

the density of the hyphae in and around the root.

Increasing quantities of hyphae in our co-cultivation sys-

tem resulted in a suboptimal interaction, and marker genes

for the beneficial interaction (such as LRR1) were down-

regulated and those for defence processes (such as

PDF1.2) were upregulated in the roots in a dose-dependent

manner (Oelmuller, 2008). Similar response patterns were

observed here (see Figure 7). In order to maintain a

mutualistic interaction with benefits for both partners, the

degree of root colonization might be controlled by activat-

ing PYK10-dependent defence responses, when too many

hyphae colonize the roots and the cells become damaged

or wounded by hyphal penetration. In barley, for instance,

less-defended root cells undergo cell death after coloniza-

tion with P. indica (Deshmukh et al., 2006). To further

elucidate the role of PYK10 in this interaction, Arabidopsis

lines in which better characterized defence compounds are

manipulated, can be analysed. Furthermore, because PEN2

is also expressed in roots, manipulation of the PEN2 level

might have an influence on root colonization. Finally, the

identification of PYK10 product(s) and the characterization of

its (their) role(s) in this interaction appears to be possible, for

example by comparing the composition of glucosinolates

and of other secondary metabolites in the roots of Arabid-

opsis lines with manipulated PYK10 levels growing in the

presence or absence of P. indica.

Experimental procedures

Growth conditions of plants and fungus

Wild-type Arabidopsis thaliana seeds, EMS mutant seeds (Colum-bia; Lehle, http://www.arabidopsis.com), seeds from the homozy-gote T-DNA insertion lines and lines expressing the uidA geneunder the control of the PYK10 promoter (Nitz et al., 2001), orexpressing PYK10 under the control of the 35S CaMV promoter,were surface-sterilized and placed on Petri dishes containing MSnutrient medium (Murashige and Skoog, 1962). After cold treatmentat 4�C for 48 h, plates were incubated for 7 days at 22�C under

continuous illumination (100 lmol m)2 sec)1) to allow growth ofthe seedlings without P. indica. P. indica was cultured as describedpreviously (Peskan-Berghofer et al., 2004; Verma et al., 1998) onaspergillus-minimal medium (Kaldorf et al., 2005). For solidmedium, 1% (w/v) agar was included.

To quantify root development, the seedlings were grown onsolid MS medium in sterile glass jars (ø 9 cm; height, 5 cm). Aftercounting the lateral roots, the lengths of the main root and theweight of the total root were determined (cf. Shahollari et al.,2007).

Co-cultivation experiments and estimation of plant growth

Nine days after the beginning the experiments, A. thaliana seed-lings were transferred to nylon discs (mesh-size, 70 lm) and placedon top of a modified PNM culture medium (5 mM KNO3, 2 mM

MgSO4, 2 mM Ca(NO3)2, 0.01 lM FeSO4, 70 lM H3BO3, 14 lM MnCl2,0.5 lM CuSO4, 1 lM ZnSO4, 0.2 lM Na2MoO4, 0.01 lM CoCl2,10.5 g l)1 agar, pH 5.6) in glass jars. One seedling was used per jar.After 24 h, fungal plugs of approximately 5 mm in diameter wereplaced at a distance of 1 cm from the roots. The uninfected controlplants received the same plugs without the fungus. The jars wereincubated at 22�C under continuous illumination from the side(80 lmol m)2 sec)1). Fresh weights were determined directly afterseedlings were removed from the jars.

Experiments on soil

For the experiments on soil, Arabidopsis seedlings were germi-nated on MS medium in Petri dishes without the fungus. Afterinfection with the fungus and co-cultivation for additional 20 days injars, they were transferred to sterile soil. Uninfected controls weretreated in the same way, except that the plugs introduced to the jarswere without the fungus. For experiments with the fungus, the soilwas mixed carefully with the fungus (1%, w/v). Although growthpromotion and higher seed yield also occur in uninfected soil, wenoticed that the response is more homogenous in inoculated soil.The fungal mycelium was obtained from liquid cultures after themedium was removed, and the mycelium was washed with anexcess of distilled water. Before being transferred to soil, the rootswere examined under the microscope to ensure that hyphae andspores had developed within and around the roots. Cultivationoccurred in multi-trays with Aracon tubes in a temperature-con-trolled growth chamber at 22�C under long-day conditions (16 hlight, 8 h dark; light intensity, 80 lmol m)2 sec)1). The sizes of theplants were monitored daily. For the mutant screen, the heights ofEMS mutant plants grown in the presence of P. indica were com-pared with those of control plants. Seeds were collected from theplants that were shorter than the wild type in the presence ofP. indica, but not shorter than the wild type without the fungus. Thereduced response to P. indica was confirmed in the next two gen-erations. The physiological results for pii-4 presented here wereobtained from the M3 and M4 generations. Seed production(g seeds per plant) was monitored by collecting seeds from indi-vidual plants grown under the standardized conditions describedabove. Seeds were dried for 4 weeks in paper bags before theweight was determined.

Staining fungal hyphae and spores

Small parts of the roots from seedlings that were co-cultivatedwith P. indica were transferred to 10% KOH and boiled for

PYK10 in P. indica–A. thaliana interaction 435

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

10 min. After washing with water for 1 min, the roots were putinto a 0.01% acid fuchsin-lactic acid solution and boiled againfor 10 min. Excess dye was removed with water prior tomicroscopy.

Fluorescence measurements

Autofluorescence in the developing root hairs was detected with theLSM 510 META microscope (Carl Zeiss Inc., http://www.zeiss.com).Relative values (550 nm) were obtained for the emission spectra (cf.Peskan-Berghofer et al., 2004).

Isolation of plasma-membrane-enriched protein fractions

A 20-g portion of Arabidopsis roots were used to isolate micro-somes. The material was homogenized in a buffer containing50 mM Tris/HCl, pH 7.4, 330 mM sucrose, 3 mM EDTA, 1 mM 1,4-dithiothreitol and 5% (w/v) polyvinylpolypyrrolidone. Thehomogenate was filtered through four layers of cheesecloth andcentrifuged for 20 min at 10 000 g. The supernatant was thencentrifuged at 50 000 g for 60 min to pellet the microsomes.Plasma membranes were prepared from three microsome prep-arations by two-phase partitioning with 6.4% (w/w) dextraneT-500 and 6.4% (w/w) polyethylene glycol (average molecularweight, 3350) (Briskin et al., 1987; Larsson et al., 1987; Peskanet al., 2000). The plasma membranes were resuspended in abuffer containing 50 mM Tris/HCl, pH 7.4; 3 mM EDTA and 1 mM

1,4-dithiothreitol.Preparation of protein extracts from plasma-membrane prepara-

tions, two-dimensional gel electrophoresis, staining of the gels andextraction of the protein spots was described in Sherameti et al.(2004).

Mass spectrometry

Aliquots of the eluted protein fractions were used for mass spec-trometry. Trypsin digestion of protein mixtures was performedaccording to Sherameti et al. (2004). Peptide analysis by couplingliquid chromatography with electrospray ionization mass spec-trometry (ESI-MS) and tandem mass spectrometry (MS-MS) wasdescribed previously (Shahollari et al., 2004; Sherameti et al., 2004;Stauber et al., 2003).

Protein identification

The measured MS-MS spectra were matched with the amino-acidsequences of tryptic peptides from the A. thaliana database inFASTA format. Cys modification by carbamidomethylation (+57 Da)was taken into account, and known contaminants were filtered out.Raw MS-MS data were analyzed by the Finnigan Sequest/TurboSequest software (revision 3.0; ThermoQuest, San Jose, CA, USA).The parameters for the analysis by the Sequest algorithm were setaccording to Stauber et al. (2003). The similarity between the mea-sured MS-MS spectrum and the theoretical MS-MS spectrum,reported as the cross-correlation factor (Xcorr), was equal or above1.5, 2.5 and 3.5 for singly, doubly or triply charged precursor ions,respectively. In order to identify corresponding loci, identified pro-tein sequences were subjected to BLAST searches at NCBI (http://www.ncbi.nlm.nih.gov) and FASTA searches by using the AGIprotein database at TAIR (http://www.arabidopsis.org). Conserveddomains and signal peptides were identified using SMART (Schultzet al., 1998).

RNA analysis

RNA was isolated with an RNA isolation kit (RNeasy; Qiagen, http://www.qiagen.com). For quantitative RT-PCR (cf. legend to Figure 6),RNA from Arabidopsis roots grown in the absence or presence ofP. indica was used with gene-specific and several control primerpairs (Sambrook et al., 1989). RT-PCR was performed by reversetranscription of 5 g of total RNA with gene-specific reverse primers.First-strand synthesis was performed with a kit (#K1631) from MBIFermentas (http://www.fermentas.com). After PCR, the productswere analyzed on 1.5% agarose gels and stained with ethidiumbromide, and visualized bands were quantified with the ImageMaster Video System (Amersham, GE Life Sciences, http://www.gelifesciences.com).

Real-time PCR

Real-time quantitative RT-PCR was performed using the iCycler iQreal-time PCR detection system and iCycler software version 2.2(Bio-Rad, http://www.bio-rad.com). Total RNA was isolated fromat least four independent replicates of Arabidopsis roots. For theamplification of the PCR products, iQ SYBR Supermix (Bio-Rad)was used according to the manufacturer’s instructions in a finalvolume of 20 ll. The iCycler was programmed to 95�C for 2 min,35 cycles of 95�C for 30 sec, 55�C for 40 sec, 72�C for 45 sec, and72�C for 10 min, followed by a melting-curve programme (55–95�Cin increasing steps of 0.5�C). All reactions were repeated at leasttwice. The mRNA levels for each cDNA probe were normalizedwith respect to the actin message level. Fold induction valueswere calculated with the DDCP equaltion of Pfaffl (2001) and werecompared with the mRNA level in the target genes in wild-typeroots, which were defined as 1.0. The following primer pairs wereused: LRR1-for, CGGCGAGTTTGATCTTGATGG, LRR1-rev,CTCAGGAACCACGACATCTCTC; PYK10-for, CGCATTTCCGG-TAAGCTTC, PYK10-rev, AAAGGCACCTGGTCGTTGCT; PBP1-for,GGATCCGATGAGGGTACTCA, PBP1-rev, GGCAGGAGTCAACG-GAGTTG; NAI1-for, CCGGGTTTGAGTTGCTAGC, NAI1-rev,GGAGACCCAAATGAGATCAC; PDF1.2a (At5g44420)-for, AT-GGTCAGGGGTTTGCGGAAA, PDF1.2a-rev, AT-GGTCAGGGGTTTGCGGAAA; P. indica was monitored with aprimer pair for Pitef1 (Butehorn et al., 2000), AC-CGTCTTGGGGTTGTATCC and TCGTCGCTGTCAACAAGATG. Thecolonized (and control) roots were removed from the agar plate,rinsed 12 times with an excess of sterile water (50 ml each) toremove the loosely attached fungal hyphae, and were then frozenin liquid nitrogen for RNA or DNA extraction.

Microarray analysis

Arabidopsis seedlings, grown as described above, were co-culti-vated (or mock-treated) with P. indica for either 2 or 6 days. RNAwas extracted from 70 mg of root material with the RNeasy PlantMini Kit (Qiagen), followed by an On-Column DNAse treatment(Qiagen). Microarray hybridization was performed with the Ara-bidopsis Genome Array ATH1 from Affymetrix (http://www.affymetrix.com), and the data were analysed with GCOS1.4 soft-ware (Affymetrix).

Miscellaneous

DNA extraction and sequence analysis were performed according tostandard protocols (Stockel and Oelmuller, 2004). For cloning ofPCR products, the PCR cloning kit from Quiagen was used. To

436 Irena Sherameti et al.

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

assign the mutant pii-4 locus to one of the Arabidopsis chromo-somes, a segregating F2 progeny was generated by crossing malepollen donor plants with homozygote lines of pii-4. Restrictionfragment length polymorphism analyses of the F2 plants were per-formed with the pARMS set (Schaffner, 1996). The PYK10 cDNA wascloned into a modified pMO9819 vector (Puzio, 1997) and intro-duced into Arabidopsis via Agrobacterium tumefaciens. Elevenplants with the PYK10 cDNA expressed in sense orientation wereregenerated and initially analysed. Two of them with the highestPYK10 mRNA levels were used for this study (ox-1 = 3c/38 andox-2 = 3f/42). All statistical analyses were performed by one-wayANOVAS.

Acknowledgements

We thank the Salk Institute Genomic Analysis Laboratory for pro-viding the sequence-indexed Arabidopsis T-DNA insertion mutants.Work was supported by the SFB 604, a grant from the DFG (Oe133/19-1), the BMBF (IND 03/013), the Friedrich-Schiller-University Jenaand the IMPRS Jena.

References

Barazani, O., Benderoth, M., Groten, K., Kuhlemeier, C. and

Baldwin, I.T. (2005) Piriformospora indica and Sebacinavermifera increase growth performance at the expense ofherbivore resistance in Nicotiana attenuata. Oecologia, 146,234–243.

Behnke, H.-D. and Eschlbeck, G. (1978) Dilated cisternae in Cappa-rales – an attempt towards the characterization of a specificendoplasmic reticulum. Protoplasma, 97, 351–363.

Bones, A.M. and Rossiter, J.T. (1996) The myrosinase-glucosinolatesystem, its organization and biochemistry. Physiol. Plant. 97, 194–208.

Bones, A.M., Evjen, K. and Iversen, T.-H. (1989) Characterization anddistribution of dilated cisternae of the endoplasmic reticulum inintact plants, protoplasts, and calli of Brassicaceae. Isr. J. Bot. 38,177–192.

Bonnett, H.T., Jr and Newcomb, E.H. (1965) Polyribosomes andcisternal accumulations in root cells of radish. J. Cell Biol. 27, 423–432.

Briskin, D.P., Leonard, R.T. and Hodges, T.K. (1987) Isolation ofplasma membrane: membrane markers and general principles.Methods Enzymol. 148, 542–568.

Brzobohaty, B., Moore, I., Kristoffersen, P., Bako, L., Campos, N.,

Schell, J. and Palme, K. (1993) Release of active cytokinin by ab-glucosidase localized to the maize root meristem. Science, 262,1051–1054.

Butehorn, B., Rhody, D. and Franken, P. (2000) Isolation and char-acterization of Pitef1 encoding the translation elongation factorEF-1a of the root endophyte Piriformospora indica. Plant Biol. 2,687–692.

Deshmukh, S., Huckelhoven, R., Schafer, P., Imani, J., Sharma, M.,

Weiss, M., Waller, F. and Kogel, K.H. (2006) The root endophyticfungus Piriformospora indica requires host cell death for prolif-eration during mutualistic symbiosis with barley. Proc. Natl Acad.Sci. USA, 103, 18450–18457.

Dharmawardhana, D.P., Ellis, B.E. and Carlson, J.E. (1995) A b-glu-cosidase from lodgepole pine xylem specific for the ligninprecursor coniferin. Plant Physiol. 107, 331–339.

Esen, A. (1993) b-glucosidases: overview. In B-Glucosidases, Bio-chemistry and Molecular Biology (Esen, A., ed). Washington, DC:American Chemical Society, pp. 1–14.

Falk, A., Taipalensuu, J., Ek, B., Lenman, M. and Rask, L. (1995)Characterization of rapeseed myrosinase-binding protein. Planta,195, 387–395.

Fieldes, M.A. and Gerhardt, K.E. (1994) An examination of theb-glucosidase (linamarase) banding pattern in flax seedlingsusing Ferguson plots and sodium dodecyl sulphate-polyacryl-amide gel electrophoresis. Electrophoresis, 15, 654–661.

Geshi, N. and Brandt, A. (1998) Two jasmonate-inducible myrosin-ase-binding proteins from Brassica napus L. seedlings withhomology to jacalin. Planta, 204, 295–304.

Glen, M., Tommerup, I.C., Bougher, N.L. and O’Brien, P.A. (2002)Are Sebacinaceae common and widespread ectomycorrhizalassociates of Eucalyptus species in Australian forests? Mycor-rhiza, 12, 243–247.

Gunning, B.E.S. (1998) The mystery organelles in Arabidopsisexpressing GFP. Trends Plant Sci. 3, 417.

Haseloff, J., Siemering, K.R., Prasher, D.C. and Hodge, S. (1997)Removal of a cryptic intron and subcellular localization of greenfluorescent protein are required to mark transgenic Arabidopsisplants brightly. Proc. Natl Acad. Sci. USA, 94, 2122–2127.

Hawes, C., Saint-Jore, C., Martin, B. and Zheng, H.-Q. (2001) ERconfirmed as the location of mystery organelles in Arabidopsisplants expressing GFP. Trends Plant Sci. 6, 245–246.

Hayashi, M., Toriyama, K., Kondo, M., Hara-Nishimura, I. and

Nishimura, M. (1999) Accumulation of a fusion protein containing2S albumin induces novel vesicles in vegetative cells of Arabid-opsis. Plant Cell Physiol. 40, 263–272.

Iversen, T.-H. (1970) Cytochemical localization of myrosinase(b-thioglucosidase) in root tips of Sinapis alba. Protoplasma, 71,451–466.

Kaldorf, M., Koch, B., Rexer, K.H., Kost, G. and Varma, A. (2005)Patterns of interaction between Populus Esch5 and Piriformos-pora indica: a transition from mutualism to antagonism. PlantBiol. 7, 210–218.

Kim, Y.-W., Kang, K.-S., Kim, S.-Y. and Kim, I.-S. (2000) Formation offibrillar multimers of oat b-glucosidase isoenzymes is mediatedby the As-Glu1 monomer. J. Mol. Biol. 303, 831–842.

Larsson, C., Widell, S. and Kjellbom, P. (1987) Preparation of high-purity plasma membranes. Methods Enzymol. 148, 558–568.

Leah, R., Kigel, J., Svendsen, I. and Mundy, J. (1995) Biochemicaland molecular characterization of a barley seed b-glucosidase.J. Biol. Chem. 270, 15789–15797.

Lenman, M., Rodin, J., Josefsson, L.-G. and Rask, L. (1990) Immu-nological characterization of rapeseed myrosinase. Eur. J.Biochem. 194, 747–753.

Lipka, V., Dittgen, J., Bednarek, P. et al. (2005) Pre- and postinva-sion defenses both contribute to nonhost resistance in Arabid-opsis. Science, 310, 1180–1183.

Malboobi, M.A. and Lefebvre, D.D. (1997) A phosphate-starvationinducible b-glucosidase gene (psr3.2) isolated from Arabidopsisthaliana is a member of a distinct subfamily of the BGA family.Plant Mol. Biol. 34(1), 57–68.

Matsushima, R., Hayashi, Y., Kondo, M., Shimada, T., Nishimura, M.

and Hara-Nishimura, I. (2002) An endoplasmic reticulum-derivedstructure that is induced under stress conditions in Arabidopsis.Plant Physiol. 130, 1807–1814.

Matsushima, R., Kondo, M., Nishimura, M. and Hara-Nishimura, I.

(2003a) A novel ER-derived compartment, the ER body, selec-tively accumulates a b-glucosidase with an ER retention signal inArabidopsis. Plant J. 33, 493–502.

Matsushima, R., Hayashi, Y., Yamada, K., Shimada, T., Nishimura,

M. and Hara-Nishimura, I. (2003b) The ER body, a novel endo-plasmic reticulum-derived structure in Arabidopsis. Plant CellPhysiol. 44, 661–666.

PYK10 in P. indica–A. thaliana interaction 437

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

Matsushima, R., Fukao, Y., Nishimura, M. and Hara-Nishimura, I.

(2004) NAI1 gene that encodes a basic-helix-loop-helix-typeputative transcription factor that regulates the formation of a novelER-derived structure, the ER body. Plant Cell, 16, 1536–1549.

McConn, M., Creelman, R.A., Bell, E., Mullet, J.E. and Browse, J.

(1997) Jasmonate is essential for insect defense in Arabidopsis.Proc. Natl Acad. Sci. USA, 94, 5473–5477.

Murashige, T. and Skoog, F. (1962) A revised medium for rapidgrowth and bioassays with tobacco tissue culture. Physiol. Plant,15, 473–497.

Nagano, A.J., Matsushima, R. and Hara-Nishimura, I. (2005) Acti-vation of an ER-body-localized b-glucosidase via a cytosolicbinding partner in damaged tissues of Arabidopsis thaliana. PlantCell Physiol. 46, 1140–1148.

Nisius, A. (1988) The stromacentre in Avena plastids: an aggrega-tion of b-glucosidase responsible for the activation of oat-leafsaponins. Planta, 173, 474–481.

Nitz, I., Berkefeld, H., Puzio, P.S. and Grundler, F.M.W. (2001) Pyk10,a seedling and root specific gene and promoter from Arabidopsisthaliana. Plant Sci. 161, 337–346.

Oelmuller, R. (2008) Interaction between Piriformospora indica withArabidopsis thaliana: early plant signaling events involved in abeneficial symbiosis with a growth- and seed yield-promotingendophytic fungus. In Biology of Molecular Plant–Microbe Inter-actions. Vol. 6, (Matteo, L., Woo, S. and Scala, F., eds). Springer(in press).

Oelmuller, R., Shahollari, B., Peskan-Berghofer, T., Trebicka, A.,

Giong, P.H., Sherameti, I., Oudhoff, M., Venus, Y., Altschmied, L.

and Varma, A. (2004) Molecular analyses of the interactionbetween Arabidopsis roots and the growth-promoting fungusPiriformospora indica. Endocytobiosis Cell Res. 15, 504–517.

Oelmuller, R., Peskan-Berghofer, T., Shahollari, B., Sherameti, I. and

Varma, A. (2005) MATH-domain containing proteins represent anovel gene family in Arabidopsis thaliana and are involved inplant/microbe interactions. Physiol. Plant. 124, 152–166.

Penninckx, I.A.M.A., Eggermont, K., Terras, F.R.G., Thomma,

B.P.H.J., De Samblanx, G.W., Buchala, A., Metraux, J.-P., Manners,

J.M. and Broekaert, W.F. (1996) Pathogen-induced systemicactivation of a plant defensin gene in Arabidopsis follows asalicylic acid–independent pathway. Plant Cell, 8, 2309–2323.

Peskan, T., Westermann, M. and Oelmuller, R. (2000) Identificationof low-density Triton X-100 insoluble plasma membrane micro-domains in higher plants. Eur. J. Biochem. 267, 6989–6995.

Peskan-Berghofer, T., Shahollari, B., Giong, P.H., Hehl, S., Markert,

C., Blanke, V., Varma, A.K. and Oelmuller, R. (2004) Association ofPiriformospora indica with Arabidopsis thaliana roots representsa novel system to study beneficial plant-microbe interactions andinvolves early plant protein modifications in the endoplasmaticreticulum and at the plasma membrane. Physiol. Plant, 122, 465–477.

Pfaffl, M.W. (2001) A new mathematical model for relative quanti-fication in real-time RT-PCR. Nucleic Acids Res. 29, 2002–2007.

Pfalz, J., Liere, K., Kandlbinder, A., Dietz, K.J. and Oelmuller, R.

(2006) pTAC2, -6, and -12 are components of the transcriptionallyactive plastid chromosome that are required for plastid geneexpression. Plant Cell, 18, 176–197.

Pham, G.H., Kumari, R., Singh, An. et al. (2004) Axenic cultures ofPiriformospora indica. In Plant Surface Microbiology (Varma, A.,Abbott, L., Werner, D. and Hampp, R., eds). Germany: Springer-Verlag, pp. 593–616.

Poulton, J.E. (1990) Cyanogenesis in plants. Plant Physiol. 94, 401–405.

Pozo, M.J. and Azcon-Aguilar, C. (2007) Unraveling mycorrhiza-induced resistance. Curr. Opin. Plant Biol. 10, 393–398.

Puzio, P.S. (1997) Identifizierung und Charakterisierung nemato-denresponsiver regulatorischer und kodierender DNA-Sequen-zen in der Pflanze Arabidopsis thaliana. PhD Thesis, Kiel,Germany: Christian-Alberts Universitat.

Rask, L., Andreasson, E., Ekbom, B., Eriksson, S., Pontoppidan, B.

and Meijer, J. (2000) Myrosinase: gene family evolution andherbivore defense in Brassicaceae. Plant Mol. Biol. 42, 93–113.

Ridge, R.W., Uozumi, Y., Plazinski, J., Hurley, U.A. and Williamson,

R.E. (1999) Developmental transitions and dynamics of the corti-cal ER of Arabidopsis cells seen with green fluorescent protein.Plant Cell Physiol. 40, 1253–1261.

Sahay, N.S. and Varma, A. (1999) Piriformospora indica: a newbiological hardening tool for micropropagated plants. FEMSMicrobiol. Lett. 181, 297–302.

Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Clon-ing. A Laboratory Manual, Second Edition. New York: Cold SpringHarbor Laboratory.

Schaffner, A. (1996) pARMS, multiple marker containing plasmidsfor easy RFLP analysis in Arabidopsis thaliana. Plant Mol. Biol.Rep. 14, 11–16.

Schliemann, W. (1984) Hydrolysis of conjugated gibberellins byb-glucosidases from dwarf rice (Oryza sativa L. cv. Tan-ginbozu)..J. Plant Physiol. 116, 123–132.

Schultz, J., Milpetz, F., Bork, P. and Ponting, C.P. (1998) SMART, asimple modular architecture research tool: identification ofsignaling domains. Proc. Natl Acad. Sci. USA, 95, 5857–5864.

Selmar, D.R., Lieberei, B., Biehl, B. and Voigt, J. (1987) Hevea lina-marase-A nonspecific b-glucosidase. Plant Physiol. 83, 557–563.

Shahollari, B., Peskan-Berghofer, T., Varma, A. and Oelmuller, R.

(2004) Receptor kinases with leucine-rich repeats are enriched inTriton X-100 insoluble plasma membrane microdomains fromplants. Physiol. Plant. 122, 397–403.

Shahollari, B., Varma, A. and Oelmuller, R. (2005) Expression of areceptor kinase in Arabidopsis roots is stimulated by the basid-iomycete Piriformospora indica and the protein accumulates inTriton X-100 insoluble plasma membrane microdomains. J. PlantPhysiol. 162, 945–958.

Shahollari, B., Vadassery, J., Varma, A. and Oelmuller, R. (2007) Aleucine-rich repeat protein is required for growth promotion andenhanced seed production mediated by the endophytic fungusPiriformospora indica in Arabidopsis thaliana. Plant J. 50, 1–13.

Sherameti, I., Shahollari, B., Landsberger, M., Westermann, M.,

Cherepneva, G., Kusnetsov, V. and Oelmuller, R. (2004) Cytokininstimulates polyribosome loading of nuclear-encoded mRNAs forthe plastid ATP synthase in etioplasts of Lupinus luteus: thecomplex accumulates in the inner-envelope membrane with theCF1 moiety located towards the stromal space. Plant J. 38, 578–593.

Sherameti, I., Shahollari, B., Venus, Y., Altschmied, L., Varma, A.

and Oelmuller, R. (2005) The endophytic fungus Piriformosporaindica stimulates the expression of nitrate reductase and thestarch-degrading enzyme glucan-water dikinase in tobacco andArabidopsis roots through a homeodomain transcription factorwhich binds to a conserved motif in their promoters. J. Biol.Chem. 280, 2641–2647.

Smith, A.R. and van Staden, J. (1978) Changes in endogenouscytokinin levels in kernels of Zea mays L. during imbibition andgermination. J. Exp. Bot. 29, 1067–1075.

Stauber, E.J., Fink, A., Markert, C., Kruse, O., Johanningmeier, U.

and Hippler, M. (2003) Proteomics of Chlamydomonas reinhardtiilight-harvesting proteins. Eukaryotic Cell, 2, 978–994.

Stockel, J. and Oelmuller, R. (2004) A novel protein for photosystemI biogenesis. J. Biol. Chem. 279, 10243–10251.

Stotz, H.U., Kroymann, J. and Mitchell-Olds, T. (1999) Plant–insectinteractions. Curr. Opin. Plant Biol. 2, 268–272.

438 Irena Sherameti et al.

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439

Stotz, H.U., Pittendrigh, B.R., Kroymann, J., Weniger, K., Fritsche,

J., Bauke, A. and Mitchell-Olds, T. (2000) Induced plant defenseresponses against chewing insects. Ethylene signaling reducesresistance of Arabidopsis against cotton worm but not dia-mondback moth. Plant Physiol. 124, 1007–1017.

Taipalensuu, J., Andreasson, E., Eriksson, S. and Rask, L. (1997)Regulation of the wound-induced myrosinase-associated proteintranscript in Brassica napus plants. Eur. J. Biochem. 247(3), 963–971.

Tierens, K.F., Thomma, B.P., Brouwer, M., Schmidt, J., Kistner, K.,

Porzel, A., Mauch-Mani, B., Cammue, B.P. and Broekaert, W.F.

(2001) Study of the role of antimicrobial glucosinolate-derivedisothiocyanates in resistance of Arabidopsis to microbial patho-gens. Plant Physiol. 125, 1688–1699.

Varma, A., Verma, S., Sudha, S., Sahay, N.S., Butehorn, B. and

Franken, P. (1999) Piriformospora indica, a cultivable plantgrowth promoting root endophyte. Appl. Environ. Microbiol. 65,2741–2744.

Varma, A., Singh, A., Sudha, S. et al. (2001) Piriformospora indica: acultivable mycorrhiza-like endosymbiotic fungus. In Mycota IX(Hock, B., ed). Germany: Springer Series, pp. 123–150.

Verma, S.A., Varma, A., Rexer, K.-H., Hassel, A., Kost, G., Sarbhoy,

A., Bisen, P., Butehorn, B. and Franken, P. (1998) Piriformosporaindica, gen. et sp. nov., a new root-colonizing fungus. Mycologia90, 898–905.

Waller, F., Achatz, B., Baltruschat, H. et al. (2005) The endophyticfungus Piriformospora indica reprograms barley to salt-stresstolerance, disease resistance, and higher yield. Proc. Natl Acad.Sci. USA, 102, 13386–13391.

Weiss, M., Selosse, M.A., Rexer, K.H., Urban, A. and Oberwinkler, F.

(2004) Sebacinales: a hitherto overlooked cosm of heterobasid-iomycetes with a broad mycorrhizal potential. Mycol. Res. 108,1003–1010.

Wittstock, U. and Halkier, B.A. (2002) Glucosinolate research in theArabidopsis era. Trends Plant Sci. 7, 263–270.

Xie, D.-X., Feys, B.F., James, S., Nieto-Rostro, M. and Turner, J.G.

(1998) COI1: an Arabidopsis gene required for jasmonate-regu-lated defense and fertility. Science, 280, 1091–1094.

Zeng, R.S., Mallik, A.U. and Setliff, E. (2003) Growth stimulation ofectomycorrhizal fungi by root exudates of Brassicaceae plants:role of degraded compounds of indole glucosinolates. J. Chem.Ecol. 29, 1337–1355.

PYK10 in P. indica–A. thaliana interaction 439

ª 2008 The AuthorsJournal compilation ª 2008 Blackwell Publishing Ltd, The Plant Journal, (2008), 54, 428–439