Cloning and heterologous expression of SS10, a subtilisin-like protease displaying antifungal activity from Trichoderma harzianum

R E S E A R C H L E T T E R

Cloningand heterologousexpressionofSS10, a subtilisin-likeprotease displayingantifungal activity fromTrichodermaharzianumLiu Yan & Yang Qian

Department of Life Science and Engineering, Harbin Institute of Technology, Harbin, China

Correspondence: Yang Qian, Department of

Life Science and Engineering, Harbin Institute

of Technology, Room 92, Xidazhi Street,

Nangang District, Harbin City, Heilongjiang

Province 150001, China. Tel.: 186 0451

86402652; fax: 186 0451 86412952;

e-mail: [email protected]

Received 29 May 2008; accepted 8 October

2008.

First published online 19 November 2008.

DOI:10.1111/j.1574-6968.2008.01403.x

Editor: Bernard Paul

Keywords

Trichoderma harzianum ; subtilisin-like

protease; biocontrol; antagonistic activity.

Abstract

Trichoderma harzianum parasitizes a large variety of phytopathogenic fungi.

Trichoderma harzianum mycoparasitic activity depends on the secretion of

complex mixtures of hydrolytic enzymes able to degrade the host cell wall. A gene

(SS10) encoding a subtilisin-like protease was cloned from T. harzianum T88, a

biocontrol agent effective against soil-borne fungal pathogens. The full-length

cDNA was isolated by 50 and 30 rapid amplification of the cDNA ends. The coding

region of the gene is 1302 bp long, encoding 433 amino acids of a predicted protein

with a molecular mass of 45 kDa and a pI of 6.1. Analysis of the deduced amino

acid sequence revealed that this protein had homology to the serine proteases of

the subtilisin-like superfamily (subtilases) (EC 3.4.21.) and had a predicted active

site made up of the catalytic residues Asp 187, His 218 and Ser 376. Northern

experiments demonstrated that SS10 was induced in response to different

fungal cell walls. Subtilisin-like protease gene SS10 was expressed in Saccharomyces

cerevisiae under control of the GAL1 promoter. The enzyme activity culmi-

nates (17.8 U mL�1) 60 h after induction with galactose. The optimal enzyme

reaction temperature was 50 1C and the optimal pH was 8. The subtilisin-like

protease exerted broad-spectrum antifungal activity against Alternaria alternata,

Fusarium oxysporum, Rhizoctonia solani, Sclerotinia sclerotiorum and Cytospora

chrysosperma.

Introduction

Among the mycoparasitic fungi, Trichoderma harzianum are

considered highly effective biocontrol agents. Their myco-

parasitic activity is facilitated by antifungal products or

secondary metabolites, including peptide and nonpeptide

toxins, and a battery of lytic enzymes, mainly chitinases,

glucanases and proteases, released in the presence of a

suitable host (Chet & Chernin, 2002).

In mycoparasitism, fungal proteases may play a signifi-

cant role in cell wall lysis, because fungal cell walls contain

chitin and/or b-glucan fibrils embedded in a protein matrix

(Wessels, 1986). Cell wall-degrading enzyme preparations

are a mixture of several enzymes, but virtually all of them

contain some proteases. This activity is necessary for the

lysis of whole fungal cells (Scott & Schekman, 1980;

Andrews & Asenjo, 1987). Proteolytic enzymes produced

and secreted by Trichoderma biocontrol strains have been

suggested to play significant roles in their antagonistic

abilities (Williams et al., 2003; Szekeres et al., 2004). Fungal

proteases may be significantly involved in antagonistic

activity, not only in the breakdown of the host cell wall

(composed of chitin and glucan polymers embedded in, and

covalently linked to, a protein matrix) (Kapteyn et al., 1996),

but also by acting as proteolytic inactivators of pathogen

enzymes involved in the plant infection process (Elad &

Kapat, 1999; Suarez et al., 2004). By understanding the

biocontrol mechanisms of action and regulation of serine

proteases, the development of approaches for detecting and

increasing the biocontrol activity of beneficial fungi may be

achieved.

Despite the potential relevance of the proteolytic activity

for Trichoderma biocontrol properties, the number of pro-

tease genes cloned to date is relatively low compared with

those of other traditionally biocontrol-associated enzymatic

systems. The gene prb1 encoding a subtilisin-like protease

FEMS Microbiol Lett 290 (2009) 54–61c� 2008 Harbin Institute of TechnologyJournal compilation c� 2008 Federation of European Microbiological SocietiesPublished by Blackwell Publishing Ltd.

involved in mycoparasitism was initially isolated and char-

acterized from Trichoderma atroviride IMI 206040 (Geremia

et al., 1993). A homologous gene, tvsp1, has also been

studied in Trichoderma virens Gv29.8 (Pozo et al., 2004). In

addition, papA and a vacuolar aspartic protease encoding

gene (papB) have been isolated from Trichoderma asperellum

T-203 and related to mycoparasitic and plant root coloniza-

tion activities (Viterbo et al., 2004). One gene (p6281)

encoding a fungal cell wall-induced aspartic protease has

also been described in T. harzianum CECT2413 (Suarez

et al., 2005).

In the present paper, we reported the cloning, sequence

analyses, and analysis of expression of SS10, a gene coding

for a subtilisin-like protease of T. harzianum in Saccharo-

myces cerevisiae. The temperature stability and pH depen-

dence for activity are among the features explored in

our study of this protease. The expression pattern of SS10

in T. harzianum was analyzed in the presence of different

fungal cell walls. The antifungal activity of this protease was

assessed in vitro against five phytopathogenic fungi. To our

knowledge, no gene encoding a subtilisin-like protease from

a biocontrol fungus has yet been expressed in a heterologous

host.

Materials and methods

Fungal strains, culture conditions and plasmid

Trichoderma harzianum strain T88 was kindly provided by

the Agricultural University of Hebei. Alternaria alternata,

Fusarium oxysporum, Rhizoctonia solani and Sclerotinia

sclerotiorum were kindly provided by Heilongjiang Agricul-

tural University. Cytospora chrysosperma was obtained from

Northeast Forestry University. Phytopathogenic fungal

strains were routinely maintained on potato dextrose agar

(PDA; Difco, Detroit, MI) at 28 1C. The T. harzianum cDNA

library of Liu & Yang (2005) obtained from mycelium

cultured in mineral medium (MM) with 2% chitin (Penttila

et al., 1987) was used in this work to clone full-length cDNA

of serine protease. An area of 2–3 cm2 of aerial mycelium of

T. harzianum (age of 5 days) was scraped from the PDA plate

and used to inoculate 50 mL of MM containing 2% chitin in

a 150-mL Erlenmeyer flask. The mycelia were then grown at

28 1C on a shaker at 250 r.p.m. for 36 h. For total RNA

isolation, mycelia were collected by filtration, washed thor-

oughly with sterile water, lyophilized, and kept at � 80 1C

until RNA extraction. Saccharomyces cerevisiae H158 was

used as the host for heterologous expression and was grown

in YPD broth medium (1%, w/v, yeast extract; 2%, w/v,

peptone; 2%, w/v, glucose). For expression studies, it was

grown on SC-U (Adams et al., 1998). The pYES2 vector

(Novagen) was used for the expression of protease in

S. cerevisiae.

Data analysis

The unidirectional cDNA library from T. harzianum myce-

lium and 3298 expressed sequence tag (EST) were acquired

after sequencing (Liu & Yang, 2005). Sequences were aligned

using the basic local alignment search tool (BLAST) and the final

sequences were searched against the GenBank database using

the BLAST program (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Sequences were compared with the nonredundant protein

database using BLASTX and default parameters. Sequences

with no significant similarity to sequences in the protein

database were compared with the nucleotide database using

BLASTN. The N-terminal signal sequence was analyzed by

SIGNALP v3.0. O-glycosylation sites and N-glycosylation sites

were identified by NETOGLYC v3.1 and NETNGLYC v1.0, respec-

tively. The sequence domains were analyzed using the MOTIFS

program. Alignments were performed by CLUSTAL X method

using the MEGALIGN program of the informatic package

DNASTAR (Lasergene).

Isolation of full-length SS10 cDNA using 50 and 30

rapid amplification of cDNA ends (RACE)

Total RNA was isolated from mycelium of T. harzianum

using a Yeast RNA mini kit (Watson Biotechnologies,

China). 50 and 30 RACE were performed using the cDNA

library of T. harzianum to obtain the sequence of the full-

length SS10 cDNA (Liu & Yang, 2005). The 50 and 30 ends of

the transcripts were amplified by the BD SMARTTM RACE

cDNA Amplification Kit (Clontech Laboratories, TaKaRa).

The design of the gene-specific primers for 50 RACE and 30

RACE was based on the sequence of serine protease EST. The

30 gene-specific primer for 50 RACE was a 27-mer with the

base composition 50-TGGCGGTAGCAAACTGAGCTTG C

TTAG-30. The 50 gene-specific primer for 30 RACE was a 25-

mer with the sequence 50-TTCGGCGCTGGTGTTGATATC

TACG-30. Following 50 RACE and 30 RACE, two overlapping

PCR products that represented the complete 50 end and the

complete 30 end of the SS10 cDNA were generated. Nucleo-

tide sequencing and sequence assembly of these products

were performed.

Gene expression profiles

To study the expression of SS10 in submerged cultures, MM

with 1% glucose and 0.5% ammonium sulfate was inocu-

lated with 1� 106 conidia mL�1 and incubated in a rotary

shaker at 200 r.p.m. for 48 h at 28 1C. Mycelium was col-

lected, washed with distilled water and 2% MgCl2, and

transferred to fresh MM and a variable carbon source:

1% glucose, 1% chitin, 1% A. alternata cell walls, 1%

F. oxysporum cell walls, 1% R. solani cell walls, 1%

S. sclerotiorum cell walls, or 1% C. chrysosperma cell walls.

Nitrogen starvation condition was 10% of the nitrogen

FEMS Microbiol Lett 290 (2009) 54–61 c� 2008 Harbin Institute of TechnologyJournal compilation c� 2008 Federation of European Microbiological Societies

Published by Blackwell Publishing Ltd.

55A subtilisin-like protease related to biocontrol

concentration in MM and 1% glucose as carbon source.

Fungal cell walls used as a carbon source for protease-

inducing conditions were prepared according to Fleet &

Phaff (1974).

Expression of the T. harzianum SS10 gene inS. cerevisiae

The coding region of SS10 was ligated into the pYES2 vector

bearing the GAL1 promoter. The pYES2/SS10 plasmid was

then transformed into S. cerevisiae H158 by the lithium

acetate method (Krautwurst et al., 1998). In S. cerevisiae, the

expression of pYES2/SS10 was induced by the addition of 2%

galactose and repressed by glucose. Isolated colonies were

used to inoculate 200 mL minimal medium plus 10 mg L�1

adenine and 2% raffinose and were grown for 24 h at 30 1C.

These cells were then used to inoculate 50 mL SC-U medium

containing 2% galactose and were grown for 108 h at 30 1C.

The yeast culture supernatant collected every 12 h by

centrifugation was used for identifying enzyme activity.

Northern blot analysis

Total RNA was extracted from mycelia of T. harzianum T88

cultured in MM with different carbon sources or starvation

conditions. Mycelia were harvested at 4, 12, and 24 h. The

total RNA (20mg) was separated on a 1.2% agarose gel

containing 1.5% formaldehyde and blotted onto a nylon

membrane. DIG High Prime DNA Labeling and Detection

Starter Kit II (Roche Molecular Biochemicals, Germany)

were used for the preparation of the probe and detection of

the transcripts of the SS10 gene. Probes for hybridization

were prepared by the random primer extension method,

according to the manufacturer’s instruction.

Sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE)

For protein expression, the S. cerevisiae transformants were

cultured in SC-U medium containing 2% galactose for 60 h

at 30 1C. The cell culture was centrifuged at 3100 g and 4 1C

for 10 min. A 1-mL yeast culture supernatant was concen-

trated to 20 mL by a centrifugal filter unit (Millipore). The

concentrated protein samples were subjected to electrophor-

esis following standard techniques (Laemmli, 1970) in

5% and 12% stacking and separating acrylamide gels,

respectively. Proteins were stained with Coomassie Brilliant

Blue R-250.

Measurement of enzyme activity

The S. cerevisiae cell culture was centrifuged at 3100 g and

4 1C for 10 min. The culture supernatant (1.0 mL) and 1%

casein solution (1.0 mL) in NaH2PO4–Na2HPO4 buffer (pH

6.0–8.0) and Na2B4O7–NaOH buffer (pH 8.5–11.0) were

preincubated at 40 1C for 5 min, respectively, and then

mixed. The mixture was incubated at 40 1C for 10 min and

2 mL of 0.4 mol L�1 trichloroacetic acid solution was added

to the mixture immediately to stop the reaction. The

reaction mixture was centrifuged at 9500 g and 4 1C for

10 min. The culture supernatant (1.0 mL) was mixed with

5 mL of 0.4 mol L�1 sodium carbonate and 1 mL Folin-

phenol reagent. The mixture was incubated at different

temperatures (25–60 1C) for 20 min. The tyrosine content

in the culture supernatant was determined colorimetrically

at 650 nm using Folin-phenol reagent (Lowry et al., 1951).

Empty pYES2 and the transformant cultured in repression

medium (containing 2% glucose) served as the control,

respectively. One unit of protease is defined as the amount

of enzyme that catalyzes the release of 1 mg of L-tyrosine

min�1 under the above assay conditions.

Antifungal assays

The cylinder plate method (Johnson & Curl, 1972) was used

to make wells on medium. The mycelium of the test fungi

was inoculated in the middle of the Petri plates containing

PDA. Three days after inoculation, when the colony dia-

meter was 3–4 cm, wells were filled with a concentrated

culture supernatant from yeast cultures expressing the SS10

subtilisin-like protease [the culture supernatant was con-

centrated 50-fold and 100-fold a by centrifugal filter unit

(Millipore)]. The culture supernatant obtained from the

control yeasts (empty vector) served as the control. The

plates were further incubated at 28 1C until the mycelial

growth had enveloped the peripheral well containing the

control and had produced crescents of inhibition around the

wells loaded with antifungal protein.

Results

Amplification of the 50 and 30 ends

By screening c. 3300 clones of a cDNA library, an EST (hzm-

001510), encoding a protein as a subtilisin-like protease, was

identified by BLASTX analysis of ESTs from the

T. harzianum mycelium cDNA library against the GenBank

nonredundant protein database. The EST was an internal

segment of SS10 from which the 50 methionine start site, the

30 stop codon, and the poly A tail were missing. A homology

study using the BLAST program database search showed that

the sequence of the cDNA fragment has a high homology of

71%, 70%, and 68% with cDNA coding to Gibberella zeae,

Neurospora crassa, and Verticillium dahliae subtilisin-like

protease, respectively. Based on this cDNA sequence, a

primer was designed to perform RACE-PCR of the 50 and

30 ends to amplify a full-length cDNA. From 50 RACE, a

0.9-kb PCR product was created. A putative ATG initiation

triplet codon was found, beginning at bp 293. From


56 L. Yan & Y. Qian

30 RACE, a 0.5-kb PCR product was produced, which

included a 24-bp poly A tail. Sequencing of these PCR

products resulted in a final continuous cDNA sequence that

was 2510 bp in length.

Description of the SS10

By joining the 50 and 30 RACE products with the known

SS10 EST, a full-length cDNA sequence was acquired. The

ORF of the SS10 gene consisted of 1302 bp, encoding a

polypeptide of 433 amino acids with a predicted molecular

mass of 45 kDa and a pI of 6.1. The sequence was submitted

to GenBank under the accession no. EF063644.

Analysis of the amino acid sequence identified a cleavage

signal sequence site between positions A21 and M22 (Fig. 1).

The putative signal peptide corresponding to the first 21

amino acids shows typical features of signal peptides, such as

a highly hydrophobic region and alanine residues at the � 3

and � 1 positions (relative to the cleavage site) (Nielsen

et al., 1997). No O-glycosylation sites were found and two

potential N-glycosylation sites (Asn41–Met–Thr and As-

n211–Asp–Thr) were identified in SS10. Further analysis of

the deduced protein sequence revealed that SS10 contained

Fig. 1. Comparison of the predicted amino acid sequence for the subtilisin-like protease identified from Trichoderma harzianum T88 with subtilisin-like

proteases from Cg, Chaetomium globosum (GenBank accession no. XP_001226134); Pa, Podospora anserine (GenBank accession no. CAD60582); Nc,

Neurospora crassa (GenBank accession no. XP_959818); and Bg, Blumeria graminis (GenBank accession no. AAK84436). Identical amino acids in all

proteins are shaded in black. Those that are present in nearly all the proteins are shaded in gray. The putative signal peptide is underlined. The catalytic

residues are indicated by arrows. The three subtilase-defining regions are marked by asterisks. N-glycosylation sites are marked by black dots.




three sequence domains that were indicative of the serine

proteases of the subtilisin-like superfamily, otherwise known

as the subtilases. These sequence domains were identified in

SS10 as amino acids 183–194 (AYVVDTGIRVTH), amino

acids 218–228 (HGSHVAGTIGG), and amino acids 374–384

(GTSMATPHVAG) (Devereux et al., 1984; Bucher et al.,

1996). Each of these three consensus regions contributes to

the active site of the protease and contains a catalytic

residue. For SS10, comparison with the three consensus

domains of the subtilases suggested that Asp 187, His 218,

and Ser 376 were the catalytic residues.

Expression pattern of SS10

The effect of diverse carbon sources or different starvation

conditions on the expression levels of SS10 in T. harzianum

was examined by Northern blot analysis. No transcript was

detected for any time point from mycelium cultivated with

glucose (1%) and ammonium (5 g L�1) as carbon and

nitrogen sources, respectively (Fig. 2a, Glc). Carbon starva-

tion resulted in a relatively weak signal at 4 h, after which the

signal disappeared (Fig. 2b, MM-C), whereas nitrogen

starvation did not appear to have any detectable effect on

SS10 expression (Fig. 2c, MM-N).

Furthermore, expression of the SS10 gene was determined

under simulated mycoparasitic conditions, where growth

takes place in minimal medium with the cell walls of

phytopathogenic fungi as the only carbon source. SS10

mRNA accumulation was strong in the presence of all five

cell walls, indicating that cell walls or a derived compound

induce SS10 expression and that mRNA accumulation is not

simply due to the lack of glucose as a carbon source. As

shown in Fig. 2 (b, CC, RS and c, FO), in the presence of

fungal cell walls (C. chrysosperma, R. solani or F. oxysporum),

SS10 mRNA reached the highest levels detected at 4 h,

with a strong decay of the signal after 12 h. On using

S. sclerotiorum cell walls, a weak SS10 signal was observed at

4 h, with the maximum expression reached at 12 h, after

which the transcript levels rapidly decreased (Fig. 2a, SS).

When A. alternata cell wall were used as the only carbon

source, the maximal accumulation of mRNA was observed

at 4 h (with a strong decay after this time) (Fig. 2c, AA).

From mycelium cultivated with chitin, SS10 was transcribed

continuously at all of the times considered (Fig. 2a, Chi).

Activity of SS10

Expression plasmid pYES2/SS10, together with empty

pYES2 as a control, was transformed into S. cerevisiae. The

transformed yeast cells were cultured in medium with 2%

galactose as a carbon source, where induction of the gene

took place. SDS-PAGE analysis revealed that a specific band

of yeast culture supernatant samples appeared only in the

induced culture (Fig. 3, lane 3). The uninduced or the empty

Fig. 2. Northern analysis of SS10 expression.

Listed at the top are hours after transfer to the

various media. Total RNA (20 mg) was extracted

from mycelia of Trichoderma harzianum cultured

in MM with different carbon sources or starvation

conditions. Glc, 1% glucose; Chi, 1% chitin; SS,

1% Sclerotinia sclerotiorum cell walls; MM-C,

absence of a carbon source; CC, 1% Cytospora

chrysosperma cell walls; RS, 1% Rhizoctonia

solani cell walls; MM-N, 1/10 of a nitrogen source

with 1% glucose as a carbon source; AA, 1%

Alternaria alternata cell walls; FO, 1% Fusarium

oxysporum cell walls. Mycelia were harvested at

4, 12, and 24 h. The hybridizations were carried

out with the SS10 and 28S rRNA gene probes.

Fig. 3. SDS-PAGE analysis of the proteins present in the culture super-

natant from the SS10 yeast transformant. M, molecular mass standards;

lane 1, culture supernatant from Saccharomyces cerevisiae harboring

empty plasmid pYES2; lane 2, culture supernatant from the transformant

grown in repressed medium for 60 h; lane 3, culture supernatant from

the transformant grown in induced medium for 60 h.


58 L. Yan & Y. Qian

vector control did not express the protease (Fig. 3, lanes 1

and 2). The result showed that the heterologous protein was

secreted successfully by S. cerevisiae. Under inducing condi-

tions, the apparent molecular weight (MW) of the protein is

slightly higher than the theoretical MW estimated to be

45 kDa.

From the data of the enzyme activity assay (Fig. 4), it is

apparent that subtilisin-like protease activity was present in

cells grown in galactose, and there was a substantial increase

after 12-h growth using galactose as a carbon source. At 60 h,

the enzyme activity increased to optimum. No enzyme

activity was detected in the control. The optimal enzyme

reaction temperature was 50 1C and the optimal pH was 8

(data not shown). The maximum enzyme activity was

17.8 U mL�1 under the optimum conditions.

Antifungal assays

The inhibition of mycelial growth of the culture supernatant

from yeast cultures expressing the SS10 subtilisin-like pro-

tease against five pathogenic fungi was tested in vitro to

evaluate the antagonistic activity of subtilisin-like protease.

This protease showed a broad-spectrum antifungal activity

at 50-and 100-fold concentrated culture against a wide range

of test fungi such as F. oxysporum, S. sclerotiorum, R. solani,

C. chrysosperma and A. alternata (Fig. 5). The mycelial

growth of phytopathogenic fungi was inhibited by the

culture supernatant from yeast cultures expressing the SS10

subtilisin-like protease. The phytopathogenic fungi could

not grow in the periphery of the inhibition zone produced

by the concentrated culture supernatant. There were poor

mycelial developments, deformation and lysis of the fungal

mycelium, and inhibition of mycelial branching, whereas

the mycelium from the control was normally well developed.

Discussion

In this paper, we have cloned and described the subtilisin-

like protease gene, SS10, of T. harzianum T88. The deduced

amino acid sequences of SS10 shared identities of the

subtilisin-like protease gene with N. crassa G. zeae, and

V. dahliae were 70%, 67% and 65%, respectively. Analysis of

the SS10 protein sequence showed that it had a high

probability of being a serine protease from the subtilisin-

like superfamily (or subtilase) (Siezen & Leunissen, 1991;

Rawlings & Barrett, 1995). Using the MOTIFS program, the

three sequence domains that corresponded to the subtilase

consensus sequences were identified in SS10. A protein is

defined as a subtilase as long as two of these consensus

regions are present (Bucher et al., 1996). Each of these three

core regions contributes to the active site of the protease and

contains a catalytic residue. Sequence comparisons using the

programs BLASTP and FASTA supported the fact that SS10 was a

subtilase and showed that SS10 was most homologous to

subtilisin-like proteases of this superfamily. Recently, Siezen

& Leunissen (1991) further subdivided the subtilases into six

separate families (subtilisin, thermitase, proteinase K, lanti-

biotic peptidase, kexin, and pyrolysin families) based on

their sequence homology and their secondary structure.

SS10 was not a member of the proteinase K, lantibiotic

0

5

10

15

20

0 12 24 36 48 60 72 84 96 108 120Time (h)

Enz

yme

activ

ity (

U m

L–1)

Fig. 4. Effect of culture time on pYES2-SS10 DNA transformant’s enzy-

matic activity.

Fig. 5. Inhibitory activity of concentrated culture

supernatant from yeast cultures expressing the

SS10 subtilisin-like protease to (a) Fusarium

oxysporum, (b) Sclerotinia sclerotiorum,

(c) Rhizoctonia solani, (d) Cytospora

chrysosperma, and (e) Alternaria alternata. CK,

100� concentrated culture supernatant of

empty vector; 1, 50� concentrated culture

supernatant from yeast cultures expressing the

SS10 subtilisin-like protease, and 2,

100� concentrated culture supernatant from

yeast cultures expressing the SS10 subtilisin-like

protease.




peptidase, and the kexin families because it lacked certain

signature sequences. However, association with the subtili-

sin, the thermitase, or the pyrolysin family was not ruled

out.

The expression analysis showed that the gene encoding

subtilisin-like protease SS10 was strongly induced by the

presence of fungal cell walls and chitin. This suggests that

this subtilisin-like protease is produced to digest fungal cell

wall structural proteins and that it could therefore partici-

pate in the Trichoderma mycoparasitic process. A high

expression of SS10 seems to correspond to the inducer

stimulus, and not to starvation conditions, as incubation in

a medium lacking carbon or nitrogen did not result in a

significant increase in transcription. In T. harzianum, induc-

tion of the subtilisin-like protease gene (SS10) occurs in a

relatively short time (4 h) after being transferred to the

medium with phytopathogen cell walls, suggesting that it

might participate in the early stages of the mycoparasitic

process. In terms of the prey/predator relationship, this

represents an advantage for Trichoderma because it can very

rapidly stop the growth of its prey.

In order to obtain enough subtilisin-like protease to

analyze its function, S. cerevisiae was used as a powerful

and versatile heterologous expression system. In this work,

the expression of SS10 genes encoding a subtilisin-like

protease was investigated. SDS-PAGE analysis revealed that

a specific band was visualized under inducing conditions,

slightly higher than the theoretical MW. We speculated that

this may be due to the glycosylation of SS10. Amino acid

sequence analysis of SS10 showed that two potential N-

glycosylation sites located in Asn41, Asn211. The theoretical

MW was added by glycosylation. The enzyme activity assay

showed that subtilisin-like protease activity occurred after

12-h growth in induced medium, and approached a peak at

60 h. The results were in agreement with the Northern

analysis (data not shown). The SS10 transcript levels were

detected at 24 h under induction, with the maximum

expression occurring at 60 h. The characteristics of the

enzyme are similar to previous explanations (Mignon et al.,

1998).

The serine proteases related to biocontrol processes in

Trichoderma spp. have been receiving increased attention.

Different serine protease have been detected and/or purified

from several Trichoderma species (Haab et al., 1990; Geremia

et al., 1993; De Marco & Felix, 2002; Suarez et al., 2004), and

some of the corresponding genes have been cloned. The gene

prbl encoding a serine protease involved in mycoparasitism

has been isolated and characterized from T. atroviride

(Olmedo-Monfil et al., 2002). A homologous gene has also

been described in Trichoderma hamatum LU593 (Steyaert

et al., 2004). Furthermore, the capacity to control disease

(caused by R. solani in cotton plants) of transgenic Tricho-

derma lines carrying multiple copies of prbl has been studied

(Flores et al., 1997). However, none of these studies have

been carried out on heterologous expression and the anti-

fungal activity in vitro of serine protease from Trichoderma

spp.

In this paper, the subtilisin-like protease SS10 from

T. harzianum was functionally expressed in S. cerevisiae. In

order to evaluate the antagonistic activity of subtilisin-like

protease in vitro against pathogenic fungi, the growth

inhibition of subtilisin-like protease against five pathogenic

fungi has been studied. During the in vitro experiments, the

subtilisin-like protease showed a broad-spectrum antifungal

activity toward devastating fungal pathogens that attack

plants.

This is the first example of the successful expression of a

functional subtilisin-like protease from a biocontrol fungus

in a heterologous host. This protease was demonstrated to

have effective biological control competence against phyto-

pathogens. The expression of SS10 from T. harzianum

provides an important tool for further studies of serine

protease involving biocontrol. Our report on SS10 describes

the protease gene product from T. harzianum showing direct

activity by itself against phytopathogens. This will make it

possible to apply protease in vitro to control plant phyto-

pathogens. Protease may be useful in its own right as an

attractive alternative for control of fungi that attack plants,

avoiding chemical fungicide applications.

Acknowledgements

This research was supported by the Chinese National

Programs for High Technology Research and Development

(2003AA241140).

Reference

Adams A, Gottschling DE, Kaiser CA & Stearns T (1998) Methods

in Yeast Genetics: A Cold Spring Harbor Course Manual. Cold

Spring Harbor Laboratory, Cold Spring Harbor, NY.

Andrews BA & Asenjo JA (1987) Continuous-culture studies of

synthesis and regulation of extracellular b (1–3) glucanase and

protease enzymes from Oerskovia xanthineolytica. Biotechnol

Bioeng 30: 628–637.

Bucher P, Karplus K, Mooeri M & Hofmann K (1996) A flexible

motif search technique based on generalized profiles. Comput

Chem 20: 3–23.

Chet I & Chernin L (2002) Biocontrol, microbial agents in soil.

Encyclopedia of Environmental Microbiology, Vol. 1 (Bitton G,

ed), pp 450–465. Wiley, New York.

De Marco JL & Felix CR (2002) Characterization of a protease

produced by a Trichoderma harziauum isolate which controls

cocoa plant witches’ broom disease. BMC Biochem 3: 3–9.

Devereux JR, Haeberli R & Smithies O (1984) A comprehensive

set of sequence analysis programs for the Vax. Nucleic Acids Res

12: 387–395.


60 L. Yan & Y. Qian

Elad Y & Kapat A (1999) The role of Trichoderma harzianum

protease in the biocontrol of Botrytis cinerea. Eur J Plant Pathol

105: 177–189.

Fleet G & Phaff HJ (1974) Glucanases in Schizosaccharomyces.

Isolation and properties of the cell-wall associated b(1–3)glucanases. J Biol Chem 249: 1717–1728.

Flores A, Chet I & Herrera-Estrella A (1997) Improved biocontrol

activity of Trichoderma harzianum by over-expression of the

proteinase-encoding gene prb1. Curr Genet 31: 30–37.

Geremia RA, Goldman GH, Jacobs D, Ardiles W, Vila SB, Van

Montagu M & Herrera-Estrella A (1993) Molecular

characterization of the proteinase-encoding gene, prb1, related

to mycoparasitism by Trichoderma harzianum. Mol Microbiol

8: 603–613.

Haab D, Hagspicl K, Szakmary K & Kubicck CP (1990)

Formation of the extracellular protease from Trichoderma

reesei QM9414 involved in cellulose degradation. J Biotechnol

16: 187–198.

Johnson LF & Curl EA (1972) Methods for Research on the

Ecology of Soil-Borne Plant Pathogens. Burgess Publishing

Company, MN.

Kapteyn JC, Montijn RC, Cruz J, Llobell A, Douwes JE, Shimoi H,

Lipke PN & Klis FM (1996) Retention of Saccharomyces

cerevisiae cell wall proteins through a phosphodiester-linked

b-1, 3/b-1,6-glucan heteropolymer. Glycobiology 3: 337–345.

Krautwurst H, Bazaes S, Gonzalez FD, Jabalquinto AM, Frey PA &

Cardemil E (1998) The strongly conserved lysine 256 of

Saccharomyces cerevisiae phosphoenolpyruvate carboxykinase

is essential for phosphoryl transfer. Biochemistry 37:

6295–6302.

Leammli UK (1970) Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227:

680–685.

Liu PG & Yang Q (2005) Identification of genes with a biocontrol

function in Trichoderma harzianum mycelium using the

expressed sequence tag approach. Res Microbiol 156: 416–423.

Lowry OH, Rosebrough NJ, Farr AL & Randall RJ (1951) Protein

measurement with the Folin-phenol reagent. J Biol Chem 193:

265–275.

Mignon B, Swinnen M, Bouchara J, Hofinger M, Nikkels A,

Pierard G, Gerday C & Losson B (1998) Purification and

characterization of a 31.5 kDa keratinolytic subtilisin-like

serine protease from Microsporum canis and evidence of its

secretion in naturally infected cats. Med Mycol 36: 395–404.

Nielsen TH, Deiting U & Sitt M (1997) A b-amylase in potato is

induced by storage at low temperature. Plant Physiol 113:

503–510.

Olmedo-Monfil V, Mendoza-Mendoza A, Gomez I, Cortes C &

Herrera-Estrella A (2002) Multiple environmental signals

determine the transcriptional activation of the mycoparasitism

related gene prb1 in Trichoderma atrovirde. Mol Genet

Genomics 267: 703–712.

Penttila M, Nevalainen H, Ratto M, Salminen E & Knowles J

(1987) A versatile transformation system for the filamentous

fungus Trichoderma reesei. Gene 61: 155–164.

Pozo MJ, Baek JM, Garcia JM & Kenerley CM (2004) Functional

analysis of tvsp1, a serine protease-encoding gene in the

biocontrol agent Trichoderma virens. Fungal Genet Biol 41:

336–348.

Rawlings NG & Barrett AJ (1995) Families of aspartic peptidases,

and those of unknown catalytic mechanism. Method Enzymol

248: 105–146.

Scott JH & Schekman R (1980) Lyticase: endoglucanase and

protease activities that act together in yeast cell lysis. J Bacteriol

142: 414–423.

Siezen RJ & Leunissen JAM (1991) Subtilases: the superfamily of

subtilisin-like serine proteases. Protein Sci 6: 501–523.

Steyaert JM, Stewart A, Jaspers MV, Carpenter M & Ridgway HJ

(2004) Co-expression of two genes, a chitinase (chit42) and

proteinase (prbl) implicated in mycoparasitism by

Trichoderma hamatum. Mycologia 96: 1245–1252.

Suarez B, Rey M, Castillo P, Monte E & Llobell A (2004) Isolation

and characterization of PRA1, a trypsin-like protease from the

biocontrol agent Trichoderma harzianum CECT 2413

displaying nematicidal activity. Appl Microbiol Biot 65: 46–55.

Suarez B, Sanz L, Chamorro I, Rey M, Gonzalez FJ, Llobell A &

Monte E (2005) Proteomic analysis of secreted proteins from

Trichoderma harzianum Identification of a fungal cell wall-

induced aspartic protease. Fungal Genet Biol 42: 924–934.

Szekeres A, Kredics L, Antal Z, Kevei F & Manczinger L (2004)

Isolation and characterization of protease overproducing

mutants of Trichoderma harzianum. FEMS Microbiol Lett 233:

215–222.

Viterbo A, Harel M & Chet I (2004) Isolation of two aspartyl

proteases from Trichoderma asperellum expressed during

colonization of cucumber roots. FEMS Microbiology Lett 238:

151–158.

Wessels JGH (1986) Cell wall synthesis in apical hyphal growth.

Int Rev Cytol 104: 37–79.

Williams J, Clarkson JM, Mills PR & Cooper RM (2003)

Saprotrophic and mycoparasitic components of aggressiveness

of Trichoderma harzianum groups toward the commercial

mushroom Agaricus bisporus. Appl Environ Microb 69:

4192–4199.




Documents

Cloning and heterologous expression of SS10, a subtilisin-like protease displaying antifungal activity from Trichoderma harzianum