Genetic resources of extremotolerant fungi: A method for identification of genes conferring stress tolerance

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 1

Bioresource Technology

Genetic resources of extremotolerant fungi: a method for

identification of genes conferring stress tolerance

Cene Gostinčara,b, Nina Gunde-Cimermana,b, Martina Turka*

a University of Ljubljana, Biotechnical Faculty, Department of Biology, Večna pot 111, SI-

1000 Ljubljana, Slovenia b Centre of Excellence for Integrated Approaches in Chemistry and Biology of Proteins

(CIPKeBiP), Jamova 39, SI-1000 Ljubljana, Slovenia

*Corresponding author: Martina Turk Department of Biology Biotechnical Faculty University of Ljubljana Večna pot 111 SI-1000 Ljubljana, Slovenia Tel: +386-1-3203392; Fax: +386-1-2573390 Email: [email protected]

Cene Gostinčar email: [email protected], [email protected] Nina Gunde-Cimerman email: [email protected]

Please note that this is the “revised personal version of the text of the final journal article”,

published according to the copyright policy of the publisher. If you would like to receive a

reprint of the final article (the content of which is identical to this file, with the exception of

the formatting), please contact the corresponding author or try to access the final article on

the official website of the publisher.

Cite this article as:

Gostinčar, C., Gunde-Cimerman, N., Turk, M., 2012. Genetic resources of extremotolerant fungi: A method for identification of genes conferring stress tolerance. Bioresource Technology 111, 360–367.

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 2

Abstract

Fungal species from extreme environments represent an underexploited source of stress-

resistance genes. These genes have the potential to improve stress tolerance of

economically important microorganisms and crops. An efficient high-throughput method for

the identification of biotechnologically interesting genes of extremotolerant fungi was

developed by constructing a cDNA expression library in Saccharomyces cerevisiae and

screening for gain-of-function transformants under stress conditions. The advantages and

possible modifications of this method are discussed, and its efficiency is demonstrated using

the stress-tolerant basidiomycetous yeast Rhodotorula mucilaginosa. Twelve R.

mucilaginosa genes are described that increase halotolerance in S. cerevisiae. These

include genes encoding a phosphoglucomutase and a phosphomannomutase. All twelve

investigated genes might be useful for the improvement of halotolerance in genetically

modified crops or industrial microorganisms.

Keywords: functional screening; gain-of-function method; stress-tolerance genes; stress-tolerant

fungi; halotolerance

Abbreviations: YNB – Yeast Nitrogen Base medium, CSM – Complete amino acid Supplement

Mixture, CSM-Ura – Complete amino acid Supplement Mixture without uracil, YPD – Yeast Peptone

Dextrose medium, LB – Luria - Bertani medium, SOC - Super Optimal broth with Catabolite repression

medium, MIC – Minimum Inhibitory Concentration, CDS – Coding Sequence

1. Introduction

Dehydration is a major stress factor for organisms living at subzero temperatures as

extracellular freezing leads to cell dehydration and reduced water absorption. When the ice

melts, the organisms are suddenly exposed to an abundance of water, a change that

provides an advantage to species that can adapt to a range of water activities. Similar stress

is encountered in hypersaline environments, where water activity is low, but can increase

quickly during rainfall. It is therefore not surprising that cellular adaptations to low

temperature and high salinity are often alike. Similar selection pressures in these

environments result in a significant overlap of fungal diversity (Gunde-Cimerman et al.,

2003), characterised by well-adapted species with specialised stress-tolerance mechanisms

(Gostinčar et al., 2011). One of the prominent representatives of these stress-tolerant

communities is the basidiomycetous yeast, Rhodotorula mucilaginosa. This species is

commonly found in cold and hypersaline environments, as well as in the deep sea and in

other unusual habitats (reviewed in Gostinčar et al., 2011). Due to its remarkable ability to

withstand a wide variety of stress conditions, R. mucilaginosa is a promising source of genes

that can confer stress tolerance.

Stress tolerance is a highly desirable trait in economically important organisms. Abiotic stress

can cause changes in morphology, physiology and biochemistry, affects growth and

productivity, and decreases yields. Even traditional processes, such as dough fermentation

and the production of Saccharomyes cerevisiae biomass, can expose yeast cells to freeze-

thaw, high sugar concentrations, air-drying, and oxidative stress (Shima and Takagi, 2009).

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 3

Similarly, crops are often exposed to drought, temperature extremes, and saline soils

(Bhatnagar-Mathur et al., 2008).

The production of ethanol as a biofuel and other industrial processes have increased the

need for improvements to multi-stress tolerance of S. cerevisiae and other microorganisms

(Zheng et al., 2010). Similarly, the need for increased food production calls for further

optimisation of crop yields. Salinisation of soil and lack of freshwater due to climate change

can increase drought frequency and the need for irrigation (Rozema and Flowers, 2008). The

connection between several abiotic stress factors and water deficit or osmotic stress in cells

(Holmberg and Bulow, 1998) further underlines the importance of osmotolerance.

Conventional breeding methods have so far failed to provide high levels of salt and drought

tolerance in crops (Rai et al., 2010). Therefore, genetic modifications remain the most

promising option (Bhatnagar-Mathur et al., 2008). While some moderate achievements have

already been made towards improvements in stress tolerance of some organisms, further

efforts are needed (Munns, 2002). These will have to involve elucidation of stress-tolerance

mechanisms and identification of genes that confer stress tolerance.

Several high-throughput approaches have been used to identify genes that are involved in

stress resistance, from microarrays, expressed sequence tags and transcriptome

sequencing, to two-dimensional protein electrophoresis, and others. These methods are

extremely important in basic research of stress responses. In biotechnology, however,

functional gene screening (which is also referred to as expression cloning) offers an

important advantage since it is targeted directly towards useful genes that can increase

stress tolerance of acceptor organisms. Functional gene screening does not require any prior

knowledge about the genome of the donor, and it is useful in cases of unsequenced

genomes, non-axenic cultures, and even whole communities.

There have been several studies that have used a variety of donor organisms and screening

systems. These have included: a plant cDNA library (Pisum sativum) in a bacterial screening

system (Escherichia coli; Joshi et al., 2009); a plant cDNA library (Thellungiella halophila) in

a fungal screening system (Schizosaccharomyces pombe; Chen et al., 2007); a plant cDNA

library (T. halophila) in a plant screening system (Arabidopsis thaliana; Du et al., 2008); a

fungal genomic library (Debaryomyces hansenii) in a yeast screening system (S. cerevisiae;

Prista et al., 2002); and even a metagenomic DNA library in a bacterial screening system (E.

coli; Kapardar et al., 2010). Neither these nor others have – to our knowledge – screened a

cDNA library of a halotolerant fungus in a fungal screening system.

Novel sources of stress-tolerance genes need to be considered to facilitate the engineering

of enhanced stress tolerance in various organisms (Somvanshi, 2009). Fungi have long been

accepted as a good source of extracellular enzyme encoding genes (Dalboge, 1997),

although in other fields their resources have been less well exploited. Patents concerning

fungal drought tolerance are few, and focused on S. cerevisiae (Somvanshi, 2009). However,

some other fungal species are much more tolerant to a variety of extreme stress conditions

(Gunde-Cimerman et al., 2003). We believe that the genetic resources of these fungi need to

be investigated for their potential to be used to improve the stress tolerance of economically

important fungi and plants.

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 4

Patents concerning fungal drought tolerance are few, and focused on S. cerevisiae

(Somvanshi, 2009) although some other fungal species are much more tolerant to a variety

of extreme stress conditions (Gunde-Cimerman et al., 2003). Therefore, in the current paper,

a method for the identification of genes from stress-tolerant fungi that can increase the stress

tolerance of recipient organisms is described. This convenient high-throughput functional

screening methodology enables the investigation of whole transcriptomes in S. cerevisiae in

short periods of time. The method facilitates identification of biotechnologically promising

genes, in our case, genes for improving salt tolerance. The effectiveness of this approach

was demonstrated by identifying 12 genes from the halotolerant basidiomycetous yeast, R.

mucilaginosa, that improved salt tolerance when expressed in S. cerevisiae. These include

the genes encoding a phosphoglucomutase and phosphomannomutase. The activities of the

corresponding proteins were investigated.

2. Materials and methods

2.1 Media, strains and growth conditions

The halotolerant basidiomycetous yeast, R. mucilaginosa (strain EXF-1630), was isolated

from an Arctic glacier in Kongsfjorden, Spitsbergen, Svalbard (Norway) and maintained in the

Ex-Culture Collection of the Department of Biology, Biotechnical Faculty, University of

Ljubljana (Slovenia). It was cultivated using the chemically defined medium Yeast Nitrogen

Base (YNB, Qbiogene), with 0.5% ammonium sulphate (w/v) and 2% glucose (w/v). Liquid

cultures were grown at 28 °C on a rotary shaker at 180 rpm.

S. cerevisiae W303a was cultivated using YNB medium (Qbiogene), with 0.5% ammonium

sulphate (w/v), 0.8% Complete Amino Acid Supplement Mixture (CSM) with/without uracil

(w/v; Qbiogene), and with 2% glucose (w/v). In the induction medium, glucose was

substituted with filter-sterilised 2% galactose (w/v) and 1% raffinose (w/v). The screening

media were prepared by adding the minimum inhibitory concentrations (MICs) of various

osmolytes to the induction medium: 1.37 M or 1.71 M NaCl, 2.7 M sorbitol, 3.4 M glycerol, or

0.4 M LiCl. The selection medium for curing of the plasmids with the URA3 selection marker

was prepared by autoclaving 20 g agar (Difco) in 750 ml distilled water, and mixing with 250

ml of a filter-sterilised solution of 5 g ammonium sulphate, 1.7 g YNB (Qbiogene), 20 g

glucose, 0.8 g CSM (Qbiogene), 40 mg/l final concentration of Uracil, and 1 g 5-fluoro-orotic

acid. The pH of the medium was not adjusted. Yeast Peptone Dextrose (YPD) medium was

prepared with 1% yeast extract, 2% peptone, 2% glucose, and 2% agar (all w/v) (Difco; in the

case of solid medium). The pH was adjusted to 6.5 with 1 M NaOH, and the medium was

sterilised by autoclaving.

E. coli (ElectroMAXTM DH10BTM T1 Phage Resistant cells; Invitrogen) was grown at 37 °C

on Luria-Bertani medium (LB, Merck) with the addition of the appropriate antibiotics. Liquid

cultures were grown at 37 °C on a rotary shaker at 200 rpm.

2.2 RNA isolation

For RNA isolation, R. mucilaginosa was grown in YNB liquid medium with 10% NaCl (w/v),

and harvested by centrifugation at mid-exponential growth phase (OD600 = 0.8-1.0). The

pellet was frozen in liquid nitrogen and homogenised using a pestle and mortar. Total RNA

was isolated from 500 mg of homogenised biomass using the TRI REAGENTTM (Sigma),

according to the manufacturer instructions. Poly(A)RNA was isolated using the FastTrack 2.0

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 5

mRNA Isolation kits (Invitrogen). The integrity and purity of the total RNA and the isolated

mRNA were evaluated with an RNA/DNA Bioanalyser (Agilent).

2.3 Library construction in E. coli

The cDNA library was constructed using the CloneMinerTM cDNA Library Construction kit

(Invitrogen), using the non-radiolabelling method and according to the manufacturer

instructions, with minor modifications. Briefly, cDNA was synthesised from 5 g poly(A)RNA

using the Biotin-attB2-Oligo(dT) Primer and SuperScript II RT. The products of the reaction

were used for second-strand cDNA synthesis using E. coli DNA polymerase I, E. coli DNA

ligase, and E. coli RNase H, followed by ligation of the attB1 adapter at the 5'-end. The cDNA

was size fractionated by column chromatography, to remove excess primers, adapters and

small cDNAs. The concentration and yield of the cDNA fractions were estimated using a

NanoDrop 2000 instrument (Thermo Scientific). Then, 150 ng of the cDNA was pooled,

precipitated, and used for the BP recombination reaction between the attB-flanked cDNA and

the entry vector pDONRTM222. The products of the reaction were electroporated into

ElectroMAXTM DH10BTM T1 Phage Resistant cells (Invitrogen), at 1,700 V with an

Electroporator 2510 (Eppendorf). The library was mixed with the freezing medium (40%

glycerol, 60% Super Optimal broth with catabolite repression (SOC) medium; Invitrogen),

and frozen. The titre of the library was determined by serial dilutions in SOC medium (10-2 to

10-4), plating on prewarmed LB plates containing kanamycin (50 mg/l), and counting of the

colonies after overnight incubation at 37 °C. The integrity of the inserts was estimated by

sequencing of 20 randomly picked clones.

2.4 Library construction in S. cerevisiae

An aliquot of the library in the entry vector pDONRTM222 that contained approximately 6.5 ×

106 transformants was inoculated into 50 ml liquid LB medium with 50 mg/l kanamycin, and

grown on a rotary shaker at 200 rpm and 37 °C. When the culture reached an OD600 of 1.0,

total plasmid DNA was isolated with the GenElute High Performance Plasmid MidiPrep Kit

(Sigma). The DNA was precipitated with the addition of 1 μl glycogen (20 mg/ml, Fermentas),

0.1 vol. 7.5 M ammonium acetate and 2.5 vol. ice-cold absolute ethanol to 1 ml of the

sample, and then centrifuged at 20,000 × g for 25 min at 4 °C. The pellet was washed twice

with 150 ml ice-cold 70% ethanol and centrifuged at 20,000 × g for 2 min at 4 °C. The

pelleted DNA was air-dried (10 min at room temperature) and dissolved in 30 μl TE buffer (10

mM Tris-HCl, pH 8, 1 mM EDTA). Then 50 ng plasmid DNA diluted to 7 μl with TE buffer was

used for the LR recombination reaction, with 450 ng pYES-DEST52 GatewayTM vector

(Invitrogen), 4 μl 5 × LR Clonase Reaction Buffer, and 6 μl LR ClonaseTM II (Invitrogen). The

reaction was performed at 25 °C for 1 h, and terminated with 1 μl Proteinase K, and an

incubation at 30 °C for 15 min, followed by an incubation at 75 °C for 10 min. The DNA was

precipitated and washed as described above, and resuspended in 9 μl TE buffer. The

products of the reaction were electroporated into ElectroMAXTM DH10BTM T1 Phage

Resistant cells (Invitrogen) in six aliquots (1.5 μl DNA suspension and 50 μl cell suspension),

as described above; the library was mixed with freezing medium, and frozen. The titre of the

library was determined by serial dilution of the sample aliquots with SOC medium (10-2 to 10-

4), plating on prewarmed LB plates containing 100 mg/l ampicillin. The colonies were counted

after an overnight incubation at 37 °C.

An aliquot of the library in pYES-DEST52 GatewayTM vector containing approximately 1 × 106

transformants was inoculated into 50 ml liquid LB medium with 100 mg/l ampicillin and grown

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 6

on a rotary shaker at 200 rpm and 37 °C. When the culture reached an OD600 of 1.0, total

plasmid DNA was isolated with the GeneJETTM Plasmid Miniprep Kit (Fermentas). The

plasmid DNA was transformed into S. cerevisiae W303a by the Library Screen

Transformation Protocol, as described by Gietz and Woods (2006), with minor modifications.

Briefly, 5 × 107 yeast cells from an overnight culture were added to 50 ml prewarmed 2 ×

YPD and incubated on a rotary shaker at 180 rpm and 28 °C until the cell density reached 2

× 107 cells/ml (determined by counting on a haemocytometer). The cells were harvested by

centrifugation (3,000 × g for 5 min), washed twice with 25 ml sterile water, and pelleted

again. The cells were mixed with the transformation mixture, prepared from 2.4 ml PEG 3500

(50%), 360 μl Li-acetate (1.0 M), 500 μl carrier single stranded salmon DNA (2 mg/ml,

denatured in boiling water for 5 min), and plasmid DNA (10 μg, diluted with sterile water to

340 μl). The mixture was vortexed and incubated at 42 °C for 1 h with inversion mixing every

5 min. Following this heat shock, the cells were pelleted by centrifugation (3,000 × g, 5 min).

The pellet was resuspended in 35 ml liquid YNB+CSM-Ura medium with 2% glucose and

0.5% ammonium sulphate (w/v). The titre of the library was determined by 10 × dilution of the

mixture prior to incubation with sterile water and plating 100 μl and 10 μl aliquots on

prewarmed plates of the same medium with 2% agar (w/v). The colonies were counted after

five days of incubation at 30 °C. The rest of the mixture was incubated overnight on a rotary

shaker at 180 rpm and 28 °C. After the incubation, the culture was mixed with 10.5 ml sterile

glycerol, aliquoted, and frozen at -80 °C.

2.5 Screening for osmotolerant transformants

To screen for transformants with increased stress tolerance, 4 ml of the frozen S. cerevisiae

library was thawed and centrifuged at 3,000× g for 5 min. The pelleted cells were washed in

4 ml fresh YNB+CSM-Ura medium with 0.5% ammonium sulphate, 2% galactose, and 1%

raffinose (w/v), re-pelleted, and resuspended again in 1 ml of the same medium. Then 100 μl

aliquots were streaked onto selection media with the MICs of the various osmolytes. The

MICs were determined by spotting S. cerevisiae W303a suspensions on YNB+CSM media

with 0.5% ammonium sulphate, 2% galactose and 1% raffinose (w/v) and with a range of

concentrations of a given osmolyte. The screening plates were incubated at 30 °C and

checked at regular intervals over a period of five weeks. All of the colonies that appeared

were transferred to fresh plates of the same medium, frozen in 50% glycerol at -80 °C, and

stored until further analysis.

2.6 Identification of the inserted genes in the osmotolerant transformants

Total DNA was extracted from the transformants with increased osmotolerance by

resuspending a medium-sized colony in 50 μl PrepMan Ultra Sample Preparation Reagent

(Applied Biosystems), and incubating the suspension in boiling water for 10 min and

centrifuging at 20,000 × g for 2 min. The supernatant was used for amplification of the insert

in the pYES-DEST52 vector by the polymerase chain reaction (PCR) in a 35-μl reaction

volume, with 10 pmol of vector -specific primers (5’-AATGCAAAAACTGCATAACCAC-3' and

5’-GGGATAGGCTTACCTTCGAAG-3’), 2.5 nmol of each dNTP (Applied Biosystems), 15

nmol MgCl2, and 0.625 U Taq polymerase (Fermentas). The thermal profile of the reaction

was as follows: 5 min denaturation at 94 °C, followed by 33 cycles of 30 s at 94 °C, 30 s at

the annealing temperature, and 30 s at 72 °C, with the final elongation step of 5 min at 72 °C.

The annealing temperature was 60 °C, and was decreased by 1 °C/cycle for the first 8

cycles, and then kept at 55 °C for the remaining 25 cycles. The products were checked by

agarose electrophoresis and sent for sequencing by Macrogen Inc. (Korea). The nucleotide

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 7

sequences were analysed and their putative functions identified by comparison with the

GenBank database entries, using BLASTX. The complete sequences of the selected genes

were obtained by primer walking.

Figure 1. A schematic overview of the

functional screening of the transcriptome

from the stress-tolerant fungus.

mRNA from a donor fungus is used for the

construction of a cDNA library, using the

CloneMiner Kit (Invitrogen). The library is

amplified in E. coli and transferred to a S.

cerevisiae expression vector by site-specific

recombination. After another round of

amplification, this library is transformed into S.

cerevisiae to create the yeast expression library.

This is then used for screening under the stress

conditions that completely inhibit the growth of

the original S. cerevisiae strain.

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 8

2.7 Curing of plasmids

The selected transformants were cured of their plasmids to demonstrate the origin of the

observed stress-tolerant phenotypes. The transformants were grown overnight in liquid YPD

medium, the culture was streak-plated on solid YPD plates, and incubated at 30 °C.

Individual colonies were transferred to solid selection medium with 5-fluoro-orotic acid and

incubated at 30 °C. The loss of the plasmid in the formed colonies was confirmed through

their inability to grow on YNB+CSM-Ura medium with 0.5% ammonium sulphate, 2%

galactose, and 1% raffinose (w/v), and with PCR using vector-specific primers.

2.8 Stress-tolerance assays

S. cerevisiae cells (strain W303a) transformants with increased stress tolerance (carrying the

vector pYES-DEST52 with a gene from R. mucilaginosa) were compared to the same

tranformants cured of their plasmids. S. cerevisiae cells (strain W303a) with an empty

plasmid pYES-DEST52 was used as control. They were grown overnight in YNB+CSM-Ura

medium with 0.5% ammonium sulphate, 2% galactose, and 1% raffinose (w/v) or YNB+CSM

medium with 0.5% ammonium sulphate and 2% glucose (w/v) to mid-exponential phase,

adjusted to an OD600 of 0.5, 10-fold serially diluted (1–104 dilutions) with fresh medium, and

spotted in 3 μl aliquots onto YNB+CSM-Ura medium with 0.5% ammonium sulphate, 2%

galactose, 1% raffinose, and 2% agar (w/v), both without added osmolytes (control) and with

the various osmolytes (0.86 M or 1.37 M NaCl, 2.7 M sorbitol, 3.4 M glycerol or 0.4 M LiCl).

The plates were incubated at 30 °C (an additional plate without the added osmolytes was

incubated at 15 °C), and photographed on days 4 and 7.

2.9 Enzyme activity assays

Cell-free extracts of soluble proteins were prepared from S. cerevisiae (expressing RmPGM2

or RmSEC53, or containing the empty pYES-DEST52 plasmid) and R. mucilaginosa grown

to mid-exponential phase in liquid YNB+CSM-Ura medium with 0.5% ammonium sulphate,

2% galactose, and 1% raffinose (w/v). The cells were harvested by centrifugation (3,000 × g

for 5 min), washed twice with 50 mM Tris-HCl (pH 7.5 at 4 °C), and resuspended in an equal

volume of lysis buffer (50 mM Tris-HCl, pH 7.5, 2 mM EDTA, 0.3 M sorbitol, 1 mM ditiotreitol,

protease inhibitor cocktail 50 μl/g cells; Sigma), and kept on ice. An equal volume of acid-

washed glass beads (425-600 m; Sigma) was added to the cells in the lysis buffer. The

cells were disrupted using a Mixer Mill MM 400 (Retsch). The homogenate was centrifuged

(20,000 × g, 15 min, 4 C) and the supernatant that contained the protein was frozen at -80

°C until use.

Phosphoglucomutase activity was measured by incubating the soluble protein extracts (0.1

μg for S. cerevisiae expressing RmPGM2, 3 μg for the other samples) in 100 μl of the

following reaction mixture: 50 mM Tris-HCl (pH 7.5), 0.5 mM NAD+, 2 U/ml glucose-6-

phosphate dehydrogenase, and 1.5 mM MgCl2. The reactions were mixed in a microtitre

plate, started by the addition of 4 mM glucose-1-phosphate, and followed at 30 °C in a

Multiskan Spectrum plate reader, by measuring the production of NAD+ at 340 nm.

Phosphomannomutase activity was measured by incubation of the soluble protein extracts

(0.5 μg for S. cerevisiae expressing RmSEC53

following reaction mixture: 50 mM Tris-HCl (pH 7.5), 0.5 mM NAD+, 2 U/ml glucose-6-

phosphate dehydrogenase, 3.5 μg/ml phosphomannose isomerase, 0.2 U/ml

phosphoglucose isomerase, 0.025 mM glucose-1,6-diphosphate, and 5 mM MgCl. The

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 9

reactions were started by the addition of 0.1 mM mannose-1-phosphate. The other

conditions were the same as for the phosphoglucomutase assays.

The enzyme assays were conducted in three biological replicates, for each of which the

enzyme activities were calculated as means of two measurements. The statistical

significances of the differences between the activities of the different species/ transformants

or different LiCl concentrations were tested using the T-test in the open-source PSPP

software.

3 Results

3.1 Library construction in E. coli, and transfer to S. cerevisiae

The total size of the library in E. coli was approximately 5 × 107 CFU. Sequencing of 20

randomly picked clones and subsequent BLASTX analysis showed that all of the sequenced

vectors contained inserts, of which 12 encoded complete coding sequences (CDS) with a

homologue in the GenBank database (e-value cut-off, 1.00 e-10), four encoded a CDS with a

homologue in the GenBank database but were truncated at the 5’-end, and four were not

identified (data not shown). The library in S. cerevisiae contained approximately 2.5 × 105

CFU.

3.2 Screening for osmotolerant transformants

The screening of the cDNA library yielded several S. cerevisiae transformants that grew on

low water activity media that prevented the growth of S. cerevisiae W303a. Fifty-one colonies

appeared after one to five weeks on a medium with 1.37 M NaCl, 47 colonies on a medium

with 1.71 M NaCl, two with 2.7 M sorbitol, 35 with 3.4 M glycerol, and 20 with 0.4 M LiCl.

3.3 R. mucilaginosa genes in osmotolerant transformants

R. mucilaginosa cDNA sequences were obtained from 92 transformants that showed

increased osmotolerance. A total of 71 unique sequences were found (data not shown).

Putative functions were identified for 44 genes by comparison with GenBank database

entries. Eight sequences were similar only to hypothetical proteins, while for 19 sequences,

no similar proteins were found. Twelve transformants were selected for further analyses

(Table 1).

Redundancy-check analysis revealed that the five most-abundant expressed sequence tags

accounted for 21.7% of the sequences (Table 2). However, the 5' and 3' cDNA ends of the

redundant clones differed in length by at least a few nucleotides, which indicated their

independent integration into the library, rather than a clonal origin.

3.4 Expression of selected genes from R. mucilaginosa in S. cerevisiae increases salt

tolerance of the transformants

Spotting assays of selected transformants on different media (Fig. 2A) did not show any

differences in growth in the absence of osmolytes and at 0.86 M NaCl. In contrast, all of the

transformants grew at 1.37 M NaCl, and three of them (carrying the RmSEC53, RmFACB

and RmPGM2 genes) also grew at 0.4 M LiCl; the control strain carrying an empty plasmid

did not grow under these conditions. Growth on galactose was not a prerequisite for incrased

salt tolerance of RmSEC52 and RmPGM2 transfromants. Although less pronounced, when

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 10

they were constitutively expressed by the TPI promotor in the plasmid pYX142, these genes

also increased halotolerance of S. cerevisiae during growth on glucose (data not shown).

Growth at 15 °C showed no substantial differences between the transformants and none

grew on 2.7 M sorbitol or 3.4 M glycerol (with the exception of RmPgm2, which grew on

glycerol, data not shown).

Growth of transformants cured of their plasmids (Fig. 2B) was comparable to growth of the

original S. cerevisiae strain without the plasmid. The exceptions were the cured

transformants for RmFACB and RmPGM2, which grew better on 0.4 M LiCl and at 15 °C,

respectively, possibly due to unknown mutations in the genome.

Figure 2. Increased salt tolerance of selected S. cerevisiae transformants by genes from R.

mucilaginosa.

A: Ten-fold serially diluted cultures of transformants carrying the indicated inserts and a transformant

carrying an empty vector (pYES-DEST52) as a control. The dilutions were plated on YNB media

supplemented with all amino acids except uracil and with 2% galactose and 1% raffinose, without

additional osmolytes (control) or with added osmolytes (as indicated). The plates were incubated at 30

°C (except one plate that was incubated at 15 °C) and photographed after 4 or 7 days (as indicated).

B: Ten-fold serially diluted cultures of the same clones as above, cured of their plasmids by treatment

with 5-fluoro-orotic acid. S. cerevisiae W303a without a plasmid was used as the control. The media

were supplemented with all amino acids, and the other growth conditions were the same as above.

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 11

3.5 Phosphoglucomutase, but not phosphomannomutase, from R. mucilaginosa is sensitive

to LiCl

The phosphoglucomutase and phosphomannomutase activities were more than 10-fold

higher in the S. cerevisiae transformants containing the RmPGM2 and RmSEC53 genes,

respectively, compared to S. cerevisiae with an empty plasmid (Fig. 3). Phosphoglucomutase

activity was also significantly (T-test, p <0.05) higher in R. mucilaginosa, compared to S.

cerevisiae.

The phosphoglucomutase activity in whole cell lysates was extremely sensitive to the

presence of LiCl and was significantly (T-test, p <0.05) lowered by 1 mM LiCl, and even

further inhibited by 5 mM LiCl, in all of the samples. In contrast, phosphomannomutase

activity was not significantly lowered by any of the LiCl concentrations tested in any of the

samples, even at 50 mM LiCl. Surprisingly, the phosphoglucomutase activity of the

transformant with the overexpressed phosphomannomutase enzyme (RmSEC53) was less

sensitive to LiCl compared to that of the other samples (Fig. 3A).

Figure 3. Effect of lithium on phosphoglucomutase and phosphomannomutase activities.

Enzyme activities were measured in cell-free extracts of soluble proteins from S. cerevisiae W303a

expressing the genes encoding phosphoglucomutase (RmPGM2) (A) or phosphomannomutase

(RmSEC53) (B), or containing the corresponding empty pYES-DEST52 plasmid (Sc) and R.

mucilaginosa EXF-1630 (Rm). The reactions were followed spectrophotometrically by measuring the

production of NAD+ at 340 nm. The activities are expressed as millimoles NAD

+ produced per min per

μg total protein. The reactions for both of the enzymes were performed in the absence of LiCl and

additionally at 0.1 mM, 1 mM and 5 mM LiCl for phosphoglucomutase (A), and at 5 mM, 20 mM and

50 mM LiCl for phosphomannomutase (B).

4. Discussion

To facilitate the identification of fungal genes involved in stress tolerance that are of potential

biotechnological interest, a high-throughput screening system was established in this study

(Fig. 1). This approach has several advantages: (a) It uses naturally stress-tolerant

organisms as gene donors; (b) Fungi are more promising gene donors for improving stress

tolerance of plants and industrially important fungi than structurally different and

phylogenetically distant prokaryotes; (c) Similarly, S. cerevisiae is a more appropriate

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 12

screening system than prokaryotes, and it is easier to work with in the laboratory, compared

to screening in plant systems; (d) As eukaryotic mRNAs are monocistronic, the use of cDNA

libraries eliminates the need for determining which of the cloned genes is responsible for

changes in the transformant. The absence of introns further reduces the required

bioinformatic work; (e) The use of commercially available chemicals and kits and the

Gateway cloning technology (Invitrogen) simplifies the procedures and increases the

flexibility of the method, by offering the ability to transfer the complete cDNA library to a

vector of choice.

In addition, several modifications of the method can be imagined. The use of a vector with a

dominant marker would allow for screening in prototrophic strains of S. cerevisiae (e.g. in

industrial strains). Screening on other carbon sources would be possible with the use of a

constitutive promoter. Changes in the cultivation conditions for the mRNA for library

construction, and changes in the screening conditions, can be used for the identification of

genes involved in various other stress responses (e.g. low or high pH, oxygen stress).

Finally, the use of environmental metatranscriptomes would further expand the pool of

genetic resources to unknown and non-culturable organisms.

We have shown the efficiency of this method for the identification of halotolerance-conferring

genes by screening the transcriptome of the stress-tolerant yeast R. mucilaginosa. All of the

12 selected S. cerevisiae transformants showed increased tolerance to NaCl, three

transformants also showed increased tolerance to LiCl, while none except RmPgm2 grew on

media containing sorbitol or glycerol at concentrations inhibitory to S. cerevisiae W303a. This

indicates that the proteins identified specifically counteract toxicity of the anorganic cations

Na+ or Li+, and not osmotic stress. These genes could therefore be used to provide increased

salt tolerance, rather than for increasing drought tolerance. Genes identified by screening on

non-ionic osmolytes (e.g. sorbitol) might prove more suitable for drought tolerance.

Although adaptations to high salinity and low temperatures sometimes overlap, substantial

differences in growth of the selected transformants at 15 °C were not observed. Screening at

low temperatures would be better suited for the identification of genes that are involved in

growth and survival at low temperatures.

The genes that were shown to increase halotolerance of S. cerevisiae fall into different

functional categories. The majority of these can be associated to responses to salt stress or

other stress factors.

The two most redundant genes identified in the screening encode two different

phosphomutases. Their redundancy might result from a combination of growth on galactose

and LiCl. One of these, RmPGM2, encodes a putative phosphoglucomutase, an enzyme that

catalyses the interconversion of glucose-6-phosphate and glucose-1-phosphate (the latter

being the product of galactose catabolism). Interestingly, when expressed with a constitutive

promoter, both RmPgm2 and RmSec53 increased halotolerance even when galactose in the

medium was substituted with glucose (data not shown). This enzyme is involved in

glycolysis, the pentose-phosphate shunt, and the metabolism of glycogen, trehalose, and

galactose (Boles et al., 1994). The expression of PGM2 in S. cerevisiae increases in

response to various abiotic stresses, including salt stress and lithium stress (Masuda et al.,

2001). Pgm2 is inhibited with high affinity by lithium, which acts as a competitive inhibitor of

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 13

yeast phosphoglucomutase activity by competing with magnesium, a cofactor of Pgm2

(Masuda et al., 2001). The phosphoglucomutase activity of S. cerevisiae cells expressing

RmPgm2 did not exhibit greater tolerance to lithium, and it was significantly inhibited even at

1 mM LiCl. However, the loss of activity was probably compensated for by the higher initial

activity due to the overexpression of the RmPgm2 enzyme, which results in increased lithium

tolerance of the transformants. The higher phosphoglucomutase activity seen for R.

mucilaginosa in comparison to S. cerevisiae even in the absence of salt, might contribute to

the greater lithium tolerance of R. mucilaginosa.

RmSEC53 is a homologue of genes that encode phosphomannomutases. These are

enzymes that catalyse an early step in the pathway of yeast O-linked and N-linked

mannosylation, the interconversion of mannose-6-phosphate and mannose-1-phosphate on

the cytosolic surface of the endoplasmic reticulum (Kepes and Schekman, 1988). This

process is important for protein glycosylation, protein sorting and secretion, and maintenance

of a functional endomembrane system in eukaryotic cells (Herscovics and Orlean, 1993). In

our analysis, phosphomannomutases proved to be relatively resistant to lithium even at 50

mM LiCl. As phosphomannomutases are known to also catalyse the interconversion of

glucose-1 phosphate and glucose-6 phosphate (Boles et al., 1994), the RmSec53

transformant was tested for this activity as well. Interestingly, while the expression of

RmSec53 the S. cerevisiae did not cause a detectable increase in phosphoglucomutase

activity, the activity appeared to be less sensitive to LiCl than the activity of the

phosphoglucomutase enzymes. Overexpression of RmSec53 could therefore increase the

lithium tolerance of S. cerevisiae in at least two ways: by altering its protein mannosylation;

or by rescuing its lithium-inhibited phosphoglucomutase activity.

In addition, four other genes identified by the functional screening are associated with

carbohydrate metabolism and energy production: RmGPA2, RmACK1, RmMCP1 and

RmPET9.

RmGPA2 is similar to the S. cerevisiae gene for the nucleotide binding alpha subunit of the

heterotrimeric G protein, which is required for the detection of extracellular glucose through

the activation of cAMP synthesis (Colombo et al., 1998). A GPA2 null-mutant of S. cerevisiae

has been shown to have decreased resistance to hyperosmotic stress (Yoshikawa et al.,

2009).

No homologue was found for RmACK1 in S. cerevisiae. However, the deduced protein

shares strong similarity with acetate kinases from other fungi. These enzymes are required

for the conversion of acetate to acetyl-CoA, and they are thus important in energy

metabolism. In Listeria monocytogenes, a decrease in the levels of this enzyme after salt

shock was related to a decrease in the intracellular acetyl-CoA pool (Duche et al., 2002).

RmMCP1 and RmPET9 encode putative mitochondrial transporters: the first has no known

function, while the second is an ADP/ATP carrier of the mitochondrial inner membrane

(Lawson and Douglas, 1988).

A further gene identified by the functional screening, RmARO4, codes for a putative 3-deoxy-

D-arabino-heptulosonate-7-phosphate (DAHP) synthase, an enzyme that catalyses the first

step in the biosynthesis of aromatic amino acids (Braus, 1991). A homologue of this gene

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 14

from the halotolerant yeast D. hansenii shows increased expression under high salinity

conditions (Calderon-Torres et al., 2006). It has also been proposed that amino-acid

availability is physiologically interconnected with high salinity (Pascual-Ahuir et al., 2001).

RmRIB4 is a gene for a putative lumazine synthase (6,7-dimethyl-8-ribityllumazine

synthase), an enzyme that catalyses the penultimate step in the riboflavin biosynthesis

pathway (Mortl et al., 1996).

Finally, four of the selected genes are involved in global regulation of the cell machinery.

RmFACB does not show significant similarity to any of the genes in S. cerevisiae, but its

protein sequence is similar to the sequence of C6 transcription factor from Aspergillus

fumigatus, and its homologues. The cloned gene is possibly 5’-truncated; however, it has a

start codon at the 5’-end.

RmANB1 is similar to a gene for translation elongation factor eIF-5A. It has been shown in

animal cells that it has a modest role in protein synthesis under normal conditions; however,

it becomes much more important under stress conditions, and it is possibly involved in the

reprogramming of protein synthesis in cells under stress (Li et al., 2010).

RmATG18 encodes a putative phosphoinositide binding protein that is involved in autophagy

vesicle formation. It is responsible for vacuolar fission in response to osmotic stress, and it

interacts with the transcriptional activator Rtg3. In turn, Rtg3 regulates the expression of

specific genes upon osmostress, in a Hog1-dependent manner (Georgakopoulos et al.,

2001; Noriega Esteban, 2009). In Arabidopsis thaliana it was suggested that the AtATG18

genes are required under multiple environmental stress conditions (Xiong et al., 2005).

RmCPR encodes a protein that is similar to cytoplasmic peptidyl-prolyl cis-trans isomerase,

which catalyses peptidyl-prolyl cis-trans isomerisation, the rate-limiting step in protein folding

(Wang and Heitman, 2005). Expression of the stress-inducible homologue CcCYP from the

pigeon-pea plants in A. thaliana increased tolerance to drought, salinity and extreme

temperatures (Sekhar et al., 2010). Similarly, the homologue from Thellungiella halophila

confers salt tolerance in S. pombe and Nicotiana tabacum (Chen et al., 2007).

Since S. cerevisiae transformants were constructed from a haploid auxotrophic laboratory

strain W303a, they could not be used for industrial application. For this purpose the genes

would have to be cloned and expressed in an industrial yeast strain, preferably by a stable

integration into the genome using a dominant selection marker. Since the impact of the

genes identified in this study on plant halotolerance has not been tested yet, several of the

described genes should be expressed in model plant organisms to select the most efficient

targets for the improvement of crops.

5 Conclusions

A high-throughput method for identification of stress-tolerance-conferring genes in fungal

transcriptomes was described, and the efficiency of this method by screening the

basidiomycetous yeast R. mucilaginosa was demonstrated. Twelve selected genes

conveyed increased NaCl and/or LiCl tolerance to S. cerevisiae. The present study

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 15

demonstrates that methods for high-throughput mining of underexploited genetic resources

of stress-tolerant fungi could facilitate the improvement of organisms used in biofuel and food

production.

Acknowledgments

The authors wish to thank Dr. Tjaša Danevčič for help with measurements of enzymatic

activities and Prof. Børge Diderichsen for critical reading of the manuscript. The scientific

work was partly financed via operation “Centre of excellence for integrated approaches in

chemistry and biology of proteins” number OP13.1.1.2.02.0005, financed by European

regional development fund (85% share of financing) and by Slovenian Ministry of higher

education, science and technology (15% share of financing). The authors also acknowledge

the financial support from the state budget by the Slovenian Research Agency

(Infrastructural Centre Mycosmo, and grant no. J4-2022).

References

Bhatnagar-Mathur, P., Vadez, V., Sharma, K.K., 2008.

Transgenic approaches for abiotic stress tolerance in

plants: retrospect and prospects. Plant Cell Rep. 27, 411-

424.

Boles, E., Liebetrau, W., Hofmann, M., Zimmermann, F.K.,

1994. A family of hexosephosphate mutases in

Saccharomyces cerevisiae. Eur. J. Biochem. 220, 83-96.

Braus, G.H., 1991. Aromatic Amino-Acid Biosynthesis in the

Yeast Saccharomyces cerevisiae - a Model System for

the Regulation of a Eukaryotic Biosynthetic-Pathway.

Microbiol. Rev. 55, 349-370.

Calderon-Torres, M., Pena, A., Thome, P.E., 2006.

DhARO4, an amino acid biosynthetic gene, is stimulated

by high salinity in Debaryomyces hansenii. Yeast 23, 725-

734.

Chen, A.P., Wang, G.L., Qu, Z.L., Lu, C.X., Liu, N., Wang,

F., Xia, G.X., 2007. Ectopic expression of ThCYP1, a

stress-responsive cyclophilin gene from Thellungiella

halophila, confers salt tolerance in fission yeast and

tobacco cells. Plant Cell Rep. 26, 237-245.

Colombo, S., Ma, P.S., Cauwenberg, L., Winderickx, J.,

Crauwels, M., Teunissen, A., Nauwelaers, D., de Winde,

J.H., Gorwa, M.F., Colavizza, D., Thevelein, J.M., 1998.

Involvement of distinct G-proteins, Gpa2 and Ras, in

glucose- and intracellular acidification-induced cAMP

signalling in the yeast Saccharomyces cerevisiae. EMBO

J. 17, 3326-3341.

Du, J., Huang, Y.P., Xi, J., Cao, M.J., Ni, W.S., Chen, X.,

Zhu, J.K., Oliver, D.J., Xiang, C.B., 2008. Functional

gene-mining for salt-tolerance genes with the power of

Arabidopsis. Plant J. 56, 653-664.

Duche, O., Tremoulet, F., Namane, A., Labadie, J.,

Consortiu, E.L.G., 2002. A proteomic analysis of the salt

stress response of Listeria monocytogenes. FEMS

Microbiol. Lett. 215, 183-188.

Georgakopoulos, T., Koutroubas, G., Vakonakis, I.,

Tzermia, M., Prokova, V., Voutsina, A., Alexandraki, D.,

2001. Functional analysis of the Saccharomyces

cerevisiae YFR021w/YGR223c/YPL100w ORF family

suggests relations to mitochondrial/peroxisomal functions

and amino acid signalling pathways. Yeast 18, 1155-

1171.

Gietz, R.D., Woods, R.A., 2006. Yeast transformation by

the LiAc/SS Carrier DNA/PEG method. Second Edition

ed. in: Yeast Protocol, Vol. 313, Humana Press, pp. 107-

20.

Gostinčar, C., Grube, M., Gunde-Cimerman, N., 2011.

Evolution of Fungal Pathogens in Domestic

Environments? Fungal. Biol. 115, 1008-1018.

Gunde-Cimerman, N., Sonjak, S., Zalar, P., Frisvad, J.C.,

Diderichsen, B., Plemenitaš, A., 2003. Extremophilic fungi

in Arctic ice: a relationship between adaptation to low

temperature and water activity Phys. Chem. Earth 28,

1273-1278.

Herscovics, A., Orlean, P., 1993. Glycoprotein-biosynthesis

in yeast. FASEB J. 7, 540-550.

Holmberg, N., Bulow, L., 1998. Improving stress tolerance

in plants by gene transfer. Trends Plant Sci. 3, 61-66.

Joshi, A., Dang, H.Q., Vaid, N., Tuteja, N., 2009. Isolation

of high salinity stress tolerant genes from Pisum sativum

by random overexpression in Escherichia coli and their

functional validation. Plant Signal. Behav. 4, 400-12.

Kapardar, R.K., Ranjan, R., Grover, A., Puri, M., Sharma,

R., 2010. Identification and characterization of genes

conferring salt tolerance to Escherichia coli from pond

water metagenome. Bioresour. Technol. 101, 3917-24.

Identification of genes conferring stress tolerance Page numbers not for citation purposes. 16

Kepes, F., Schekman, R., 1988. The yeast Sec53 gene

encodes phosphomannomutase. J. Biol. Chem. 263,

9155-9161.

Lawson, J.E., Douglas, M.G., 1988. Separate genes

encode functionally equivalent ADP/ATP carrier proteins

in Saccharomyces cerevisiae - Isolation and analysis of

Aac2. J. Biol. Chem. 263, 14812-14818.

Li, C.H., Ohn, T., Ivanov, P., Tisdale, S., Anderson, P.,

2010. eIF5A Promotes Translation Elongation, Polysome

Disassembly and Stress Granule Assembly. Plos One 5,

e9942.

Masuda, C.A., Xavier, M.A., Mattos, K.A., Galina, A.,

Montero-Lomeli, M., 2001. Phosphoglucomutase is an in

vivo lithium target in yeast. J. Biol. Chem. 276, 37794-

37801.

Mortl, S., Fischer, M., Richter, G., Tack, J., Weinkauf, S.,

Bacher, A., 1996. Biosynthesis of riboflavin - lumazine

synthase of Escherichia coli. J. Biol. Chem. 271, 33201-

33207.

Munns, R., 2002. Comparative physiology of salt and water

stress. Plant Cell Environ. 25, 239-250.

Noriega Esteban, N. 2009. The Rtg1 AND Rtg3 proteins are

novel transcription factors regulated by the yeast Hog1

MAPK upon stress. in: Unitat de Senyalització Cellular,

Departament de Ciencies Experimentals i de la Salut

(CEXS), Universitat Pompeu Fabra, pp. 153.

Pascual-Ahuir, M., Serrano, R., Proft, M., 2001. The Sko1p

repressor and Gcn4p activator antagonistically modulate

stress-regulated transcription in Saccharomyces

cerevisiae. Molecular and Cellular Biology 21, 16-25.

Prista, C., Soeiro, A., Vesely, P., Almagro, A., Ramos, J.,

Loureiro-Dias, M.C., 2002. Genes from Debaryomyces

hansenii increase salt tolerance in Saccharomyces

cerevisiae W303. FEMS Yeast Res. 2, 151-7.

Rai, M.K., Kalia, R.K., Singh, R., Gangola, M.P., Dhawan,

A.K., 2010. Developing stress tolerant plants through in

vitro selection - an overview of the recent progress.

Environ. Exp. Bot. 71, 89-98.

Rozema, J., Flowers, T., 2008. Crops for a salinized world.

Science 322, 1478-1480.

Sekhar, K., Priyanka, B., Reddy, V.D., Rao, K.V., 2010.

Isolation and characterization of a pigeonpea cyclophilin

(CcCYP) gene, and its over-expression in Arabidopsis

confers multiple abiotic stress tolerance. Plant Cell

Environ. 33, 1324-1338.

Shima, J., Takagi, H., 2009. Stress-tolerance of baker's-

yeast (Saccharomyces cerevisiae) cells: stress-protective

molecules and genes involved in stress tolerance.

Biotechnol. Appl. Biochem. 53, 155-64.

Somvanshi, V.S., 2009. Patenting drought tolerance in

organisms. Recent Pat. DNA Gene Seq. 3(1), 16-25.

Wang, P., Heitman, J., 2005. The cyclophilins. Genome

Biol. 6, 226.

Xiong, Y., Contento, A.L., Bassham, D.C., 2005. AtATG18a

is required for the formation of autophagosomes during

nutrient stress and senescence in Arabidopsis thaliana.

Plant J. 42, 535-546.

Yoshikawa, K., Tanaka, T., Furusawa, C., Nagahisa, K.,

Hirasawa, T., Shimizu, H., 2009. Comprehensive

phenotypic analysis for identification of genes affecting

growth under ethanol stress in Saccharomyces

cerevisiae. FEMS Yeast Res. 9, 32-44.

Zheng, D.Q., Wu, X.C., Tao, X.L., Wang, P.M., Li, P., Chi,

X.Q., Li, Y.D., Yan, Q.F., Zhao, Y.H., 2010. Screening

and construction of Saccharomyces cerevisiae strains

with improved multi-tolerance and bioethanol fermentation

performance. Bioresour. Technol. 102, 3020-3027.