A Rhizosphere Fungus Enhances Arabidopsis Thermotolerance through Production of an HSP90 Inhibitor

Abstract

The molecular chaperone HEAT SHOCK PROTEIN90 (HSP90) is essential for the maturation of key regulatory proteins in eukaryotes and for the response to temperature stress. Earlier, we have reported that fungi living in association with plants of the Sonoran desert produce small molecule inhibitors of mammalian HSP90. Here, we address whether elaboration of the HSP90 inhibitor monocillin I (MON) by the rhizosphere fungus Paraphaeosphaeria quadriseptata affects plant HSP90 and plant environmental responsiveness. We demonstrate that MON binds Arabidopsis (Arabidopsis thaliana) HSP90 and can inhibit the function of HSP90 in lysates of wheat (Triticum aestivum) germ. MON treatment of Arabidopsis seedlings induced HSP101 and HSP70, conserved components of the stress response. Application of MON, or growth in the presence of MON, allowed Arabidopsis wild type but not AtHSP101 knockout mutant seedlings to survive otherwise lethal temperature stress. Finally, cocultivation of P. quadriseptata with Arabidopsis enhanced plant heat stress tolerance. These data demonstrate that HSP90-inhibitory compounds produced by fungi can influence plant growth and responses to the environment.

Full text

Running Head: Plant Hsp90 Inhibited by a Fungal Metabolite
Corresponding Author:
A.A. Leslie Gunatilaka
Southwest Center for Natural Products Research
University of Arizona,
250 E. Valencia Road
Tucson, Arizona 85706
Tel: (520) 741-1691
Fax: (520) 741-1468
E-mail: leslieg@ag.arizona.edu.
Journal Research Area: Plants Interacting with Other Organisms
Plant Physiology Preview. Published on July 13, 2007, as DOI:10.1104/pp.107.101808
Copyright 2007 by the American Society of Plant Biologists

2
A Rhizosphere Fungus Enhances Arabidopsis
Thermotolerance Through Production of an HSP90
Inhibitor
1
Catherine A. McLellan
2,3
, Thomas J. Turbyville
2
, E. M. Kithsiri Wijeratne, Arthur
Kerschen, Elizabeth Vierling, Christine Queitsch
4
, Luke Whitesell, and
A. A. Leslie Gunatilaka
*
FAS Center for Systems Biology, Harvard University, Cambridge, MA 02138 (C.A.M.,
C.Q.); Southwest Center for Natural Products Research, Office of Arid Lands Studies,
College of Agriculture and Life Sciences, University of Arizona, 250 E. Valencia Road,
Tucson, Arizona 85706 (T.J.T., E.M.K.W., A.A.L.G.); Department of Plant Sciences,
University of Arizona, Tucson, AZ 85721 (A.K.); Department of Biochemistry and
Molecular Biophysics, University of Arizona, Tucson, Arizona 85721 (E.V.);
Whitehead
Institute, Cambridge, MA 02142 (L.W.).

3
1
This work was supported in part by the U.S. National Institutes of Health (N.I.H.)
Grants R01-CA09025 (to A.A.L.G.) and R21-CA091056 (to L.W.), and Energy
Biosciences Program Grant DE-RG03-99ER20338 of the U.S. Department of Energy (to
E.V.). Fellowship support of T.J.T. was provided by the University of Arizona Bio5
Institute.
2
These authors contributed equally to this work.
3
Current address: Whitehead Institute, Cambridge.
4
Current address: Department of Genome Sciences, University of Washington, Seattle.
*
Corresponding author; e-mail leslieg@ag.arizona.edu; fax 520-741-0113

4
The molecular chaperone Heat Shock Protein 90 (HSP90) is essential for the maturation
of key regulatory proteins in eukaryotes and for the response to temperature stress.
Earlier we have reported that fungi living in association with plants of the Sonoran desert
produce small molecule inhibitors of mammalian HSP90. Here, we address whether
elaboration of the HSP90 inhibitor monocillin I [MON] by the rhizosphere fungus
Paraphaeosphaeria quadriseptata affects plant HSP90 and plant environmental
responsiveness. We demonstrate that MON binds Arabidopsis HSP90 and can inhibit the
function of HSP90 in lysates of wheat (Triticum aestivum) germ. MON treatment of
Arabidopsis seedlings induced HSP101 and HSP70, conserved components of the stress
response. Application of MON, or growth in the presence of MON, allowed Arabidopsis
wild type, but not AtHSP101 knock-out mutant seedlings to survive otherwise lethal
temperature stress. Finally, co-cultivation of P. quadriseptata with Arabidopsis enhanced
plant heat stress tolerance. These data demonstrate that HSP90-inhibitory compounds
produced by fungi can influence plant growth and responses to the environment.

5
Environmental factors and interactions between organisms are major forces that
impact the development of organisms and have helped shape the evolutionary histories of
species. Recent studies have identified the highly conserved and environmentally
responsive molecular chaperone HSP90 as a potential molecular link between the biotic
and abiotic environment of an organism and its phenotype. HSP90 is essential for the
maturation of many key regulatory proteins in eukaryotes and for the evolutionarily
conserved response to temperature stress (Young et al., 2001; Picard, 2002; Pratt and
Toft, 2003). In plants, reduced HSP90 function dramatically alters responses to
environmental stimuli and can globally affect plant phenotype. For example, plants with
reduced HSP90 function are more sensitive to microbial pathogens and also show altered
responses to abiotic cues such as gravity and light (Sangster and Queitsch, 2005). Several
studies have demonstrated that manipulation of HSP90 function results in expression of
previously cryptic genetic and epigenetic variation, thereby dramatically altering
organism phenotype in a heritable manner (Rutherford and Lindquist, 1998; Queitsch et
al., 2002; Sollars et al., 2003; Yeyati et al., 2007). It has been proposed that the
manifestation of such variation could occur through environmental stress which might
reduce HSP90 buffering capacity (Sangster et al., 2004). Alternatively, HSP90 function
could be modulated by HSP90-specific small molecule inhibitors elaborated by several
fungi in natural environments (Turbyville et al., 2006). To date, however, no evidence
for targeting of HSP90 in the interactions between organisms with consequences for
organismic phenotypes has been reported.
In a screen of ethyl acetate extracts from over 500 Sonoran desert plant-associated
endophytic and rhizosphere fungal strains, we previously identified two highly specific
inhibitors of mammalian HSP90, monocillin I (MON; Fig. 1A, R = H) and radicicol
(RAD; Fig. 1A, R = Cl), among many other secondary metabolites (Turbyville et al.,
2006). Extracts of the rhizosphere fungal strain Paraphaeosphaeria quadriseptata, which
inhabits the Sonoran desert Christmas cactus (Opuntia leptocaulis DC.; Cactaceae),
contained MON in surprisingly high amounts – about 30% by weight of dry extract
(Wijeratne et al., 2004). Extracts of the endophytic fungal strain Chaetomium chiversii,

6
which colonizes the stem tissues of Mormon tea (Ephedra fasciculata A. Nels.;
Ephedraceae), contained up to 10% RAD (Turbyville et al., 2006). Production of MON
and RAD are not limited to these fungi; in fact, MON was first identified as a secondary
metabolite elaborated by a mycoparasite of pine trees in North America (Delmotte and
Delmotte-Plaquee, 1953; Omura et al., 1979; Ayer et al., 1980). Additional reports
document the production of HSP90 inhibitors by bacteria and fungi in diverse
ecosystems. Furthermore, P. quadriseptata is not restricted to the desert, but has also
been isolated as the major fungal inhabitant of the rhizosphere of young maize plants in
Brazil, indicating that this plant-fungus association is not limited to members of the
cactus family (Gomes et al., 2003). Therefore, the extent of plant associations with fungi
or other microorganisms which produce HSP90 inhibitors is most likely vastly
underappreciated.
While extensive studies have been directed toward understanding the mechanisms
and activities of small molecule HSP90 inhibitors in mammalian systems, especially for
possible treatment of cancers (Whitesell and Lindquist, 2005), there has been virtually no
investigation of the impact of production of HSP90 inhibitors on plant-microorganism
associations. As an essential step in beginning to define the impact of such inhibitors on
plants in the environment, we demonstrate that MON binds and inhibits plant HSP90
function and can induce components of the plant heat shock response. In addition,
application of MON or co-cultivation with P. quadriseptata enhances plant heat
tolerance. To fully understand the importance of HSP90 in plant biology, these findings
make it clear that consideration must be given to HSP90 as a prominent target in
mediating plant-microorganism interactions.

7
RESULTS
MON Binds and Inhibits Plant Hsp90
We first sought to establish that the fungus-derived inhibitors of mammalian
HSP90, MON and RAD, can bind and inhibit plant HSP90. Although HSP90 is a
conserved molecular chaperone, small molecule inhibitors can show species specificity.
For example, the prototypic HSP90 inhibitor geldanamycin (GDA) fails to bind and
inhibit Caenorhabditis elegans HSP90 (David et al., 2003), despite the fact that this
protein is 83% similar (73% identical) to its human ortholog which is readily inhibited by
the compound. Given that the protein sequences of plant Hsp90 are only 63-71%
identical to human and yeast Hsp90 (although they are 88-93% identical between species)
(Krishna and Gloor, 2001) we first tested whether fungus-derived MON and RAD could
bind plant Hsp90. Using a previously reported solid phase competition approach
(Whitesell et al., 1994), an amine derivative of GDA was immobilized on agarose beads
and then incubated with Arabidopsis seedling lysates which had been supplemented with
various concentrations of MON or RAD as soluble competitors. After washing under
moderately stringent conditions, bound proteins were eluted and analyzed by SDS-PAGE
and staining with Sypro Ruby. Several non-specific bands of comparable size were
identified in the eluates of both control beads without attached GDA and beads with
GDA. Importantly, a prominent band in the 90kD range was present only in eluate from
beads with covalently bound GDA (data not shown). To establish the identity of this
band, we immunoblotted eluates with an antibody to plant HSP90. Immunodetection of a
specific ~90kDa band confirmed that both MON and RAD efficiently competed with
immobilized GDA for binding to Arabidopsis HSP90 in a concentration-dependent
manner (Fig.1B). These data suggest that both inhibitors bind to the nucleotide-binding
site in the amino terminus of HSP90, which is known to be the binding site of GDA and
RAD in other organisms (Roe et al., 1999; Schulte et al., 1999).
Because ATP hydrolysis is required for its chaperoning activity, GDA binding to
the nucleotide binding site of Hsp90 markedly alters its function (Roe et al., 1999). To
confirm that MON and RAD could affect plant HSP90 function, we performed standard
assays for chaperone activity based on the ATP-dependent refolding of heat-denatured
firefly luciferase in wheat germ
lysate (Kolb et al., 1994). Indeed, MON and RAD

8
inhibited luciferase refolding consistent with MON- or RAD-mediated inhibition of plant
HSP90 function (Fig. 1C). Control experiments confirmed that neither MON nor RAD
had any direct effect on the enzyme activity of native luciferase (data not shown).
MON Induces Expression of Heat Stress Responsive Genes
Having established that HSP90 from wheat and Arabidopsis were bound and
inhibited by both MON and RAD, we next evaluated whether exposure to MON could
alter the environmental responsiveness of plants. We first examined the impact of MON
on the transcriptional activation of specific heat-stress-responsive genes. Under
physiological conditions, HSP90-containing chaperone complexes are thought to
sequester heat stress transcription factor (HSF) monomers in the cytoplasm. In the most
widely accepted current model, HSP90 inhibitors and stressors such as heat that lead to
accumulation of misfolded proteins are proposed to titer HSP90 and other chaperones,
away from HSF resulting in its release and thereby allowing the trimerization and
phosphorylation that are required for its activation of HSP gene transcription (Zou et al.,
1998; Guo et al., 2001; Voellmy and Boellmann, 2007).To assess potential HSF
activation, mRNA levels of AtHSP101 were measured in Arabidopsis seedlings after
MON exposure using semi-quantitative RT-PCR. Exposure of seedlings to MON induced
a rapid increase in AtHSP101 mRNA level at 90 min post MON application, as is also
observed during heat stress. AtHSP101 mRNAs then declined substantially 210 min after
MON exposure (Fig. 2A). A similar response was observed for AtHSP70, another HSP
that relies on HSF for induction (data not shown). Heat stress at 38° C for 90 min. was
used as a positive control for these experiments, and it markedly increased AtHSP101
mRNA levels as expected. To determine whether MON-induced changes in mRNA levels
would result in increased protein amounts, lysates of MON-treated seedlings were
examined for HSP101 protein levels. A concentration-dependent increase in HSP101
level was observed in response to overnight treatment with a solution of MON in DMSO
as compared to the level found in control plants which had been exposed to DMSO alone
(Fig. 2B). As a positive control, heat-stressed seedlings were analyzed, and increased
HSP101 levels were detected (Fig. 2B). From these results, we conclude that MON-
mediated modulation of the heat shock response in plants is rapid, tunable and reversible.

9
MON Enhances HSP101-Dependent Heat Tolerance
In several plant species and in Saccharomyces cerevisiae, the HSF-dependent
induction of HSP101 is crucial for acquired thermotolerance. Therefore,
we hypothesized that MON-mediated induction of HSP101 might enhance
thermotolerance in treated seedlings. To test this hypothesis, a quantitative assay of
acquired thermotolerance based on hypocotyl elongation of dark-grown seedlings was
used (Hong and Vierling, 2000; Queitsch et al., 2000). Hypocotyl elongation in the dark
is slowed or completely arrested after exposure to temperature stress depending upon its
severity. Prior induction of HSP101 through a pre-conditioning heat exposure, however,
allows continued hypocotyl elongation even after severe stress, which can be readily
measured in large numbers of replicate seedlings, permitting statistical analysis.
Conversely, AtHSP101 mutant seedlings do not acquire thermotolerance following heat
pre-treatment and hypocotyl elongation remains arrested in response to severe stress
(Hong and Vierling, 2000; Queitsch et al., 2000).
In the absence of heat stress (control conditions), topical MON application to
dark-grown seedlings on plates did not alter hypocotyl elongation (Fig. 3A). MON
application prior to heat stress at 45
o
C for 50 or 75 min, however, resulted in significant
preservation of hypocotyl elongation consistent with enhanced thermotolerance. The 10
µM treatment appeared to cause some cytotoxicity to the heat stressed plants as measured
by decreased hypocotyl elongation (Fig. 3A). This could be due to non-specific effects of
the small molecule, or the loss of other essential HSP90 functions related to growth. Most
strikingly, when the same plates were subsequently transferred to a growth chamber
under day/night light and temperature cycling it was apparent that prior MON treatment
resulted in a dramatic rescue of the seedlings from an otherwise lethal heat stress (Fig.
3B). After two weeks, the heat stressed MON-treated seedlings showed dramatic
preservation of seedling growth and greening compared to the vehicle-treated seedlings.
To address whether root exposure to MON could enhance thermotolerance,
comparable to topical treatment, seedlings were germinated on media supplemented with
varying concentrations of MON. An Arabidopsis HSP101 null mutant, hot1-3 (Hong and
Vierling, 2000), was included in these experiments to determine if differences in response

10
to MON are linked to the observed MON-induced expression of this critical chaperone.
Growth of seedlings on MON-containing plates resulted in modest stunting of overall
hypocotyl growth when compared to controls, similar to what has been reported for the
HSP90 inhibitor GDA. However, growth on MON-supplemented plates protected dark-
grown seedlings from the adverse effects of severe temperature stress as demonstrated by
their normal light-induced greening response in contrast to untreated seedlings (Fig. 4).
Most importantly, the HSP101 null mutant hot1
−
3 failed to green and develop normally,
confirming that induction of HSP101 is required for the enhanced thermotolerance
observed after MON treatment (Fig. 4).
Co-cultivation of Arabidopsis and P. quadriseptata Enhances Heat Tolerance
To test whether the presence of MON-producing fungus in the rhizosphere could
alter plant responsiveness to thermal stress, we investigated the consequences of co-
cultivation of P. quadriseptata with Arabidopsis as a model for this type of interaction.
When Arabidopsis seedlings were co-cultured with P. quadriseptata on plates, hypocotyl
elongation under non-heat shock conditions was reduced, most likely due to the presence
of MON. Importantly, despite overall reduced growth, the hypocotyls of seedlings co-
cultivated with P. quadriseptata continued to elongate after heat stress, resulting in
significantly longer hypocotyls than seen in seedlings heat-stressed in the absence of
fungus (Fig. 5A and B, p<0.0001 for the effect of fungus, effect of heat treatment and the
interaction of the two variables). Although the heat treatment appears to have decreased
fungal growth, extraction of fungus recovered from the plates and of the agar substrate
showed that the fungus produced MON under these culture conditions and that MON had
diffused into the growth medium (Fig. 6).
Co-cultivation with P. quadriseptata also improved the survival of soil-grown
plants after severe heat stress (Fig. 7). In the presence of the fungus, surviving soil-grown
plants recovered readily from heat stress, appearing indistinguishable from non-stressed
controls with no significant difference in leaf number or rosette diameter observed after
12 days of growth (data not shown).
DISCUSSION

11
We have demonstrated that the fungal secondary metabolites MON and RAD can
bind and inhibit plant HSP90. Application of MON leads to expression of major
components of the heat stress response, HSP101 and HSP70, and can promote heat
tolerance of Arabidopsis seedlings in an HSP101-dependent manner. In addition, co-
cultivation of the MON-producing fungus, P. quadriseptata, with Arabidopsis leads to
enhanced heat tolerance of Arabidopsis. Taken together, these data demonstrate that the
presence of an HSP90 inhibitor-producing fungus can dramatically alter plant
responsiveness to environmental stresses such as heat. Based on extensive evidence in
other systems that HSF activation of HSP gene expression is stimulated by inhibition of
HSP90, we conclude that the enhanced Arabidopsis heat tolerance observed in our
studies is the result of MON inhibition of HSP90. The elaboration of HSP90 inhibitors by
plant-associated microorganisms implicates HSP90 as a direct target in organismic
interactions, possibly among microbial communities competing in the plant rhizosphere.
In light of the multi-faceted role that plant HSP90 plays in response to biotic and abiotic
stresses (Sangster and Queitsch, 2005), in normal plant development (Krishna and Gloor,
2001), and in the buffering of genetic variation (Queitsch et al., 2002), HSP90-based
plant-fungus interactions could impact plant phenotypes in numerous ways.
Induced responses to environmental stimuli such as heat and drought and defense
responses against microbial pathogens are typically tightly regulated. Constitutive
activation of these normally inducible environmental response pathways can result in
slower growth and abnormal pleiotropic phenotypes (Kasuga et al., 1999; Noutoshi et al.,
2005). At this point, it is not known whether fungal production of MON or RAD might
be regulated by abiotic conditions or by plant signals, either in the Sonoran desert, where
our isolates were obtained, or in other environments where these fungi occur. In
laboratory monoculture, however, elaboration of these secondary metabolites can be
markedly altered by growth conditions (Gunatilaka, 2006). Even if such compounds are
continually produced by fungi in association with plants, it remains possible that stress
pre-conditioning as a result of plant exposure to fungal HSP90 inhibitors could be
advantageous in certain environments such as the desert, where plants must survive rapid,
drastic temperature changes. Indeed, extreme environments may favor constitutive
upregulation of HSPs as demonstrated for larvae, but not adults, of the polar insect

12
Beligca antarctica. The larvae of this flightless midge are exposed to dramatic
temperature differences, being encased in ice during the lengthy Antarctic winter and
then emerging to complete their life cycle during the austral summer (Rinehart et al.,
2006). In contrast, adults show the typical stress-inducible expression of HSPs.
Previous studies of grass-endophyte associations have also shown that continual
presence of an endophyte can improve the tolerance of infected plants to high
temperatures (Marquez et al., 2007). In these studies, however, molecular mechanisms
responsible for induced thermotolerance including the potential elaboration of heat
shock-active secondary metabolites were not investigated. Much more research on the
occurrence, concentration and environmental stability of HSP90 inhibitors, as well as the
control of their production by plant-associated microorganisms is clearly warranted.
Ecological and physiological studies will also be advanced by development of genetically
tractable microorganisms, in which production of MON can be controlled, an approach
which is not currently feasible in P. quadriseptata.
While the responses we have measured are most likely the result of the
interactions of HSP90 with plant HSFs, HSP90 inhibitors can also be expected to disrupt
processes mediated by other substrates that HSP90 chaperones, its so called client
proteins. The only endogenous HSP90 clients characterized so far in plants are R-
proteins, whose rapid degradation in HSP90-deficient mutant plants translates into
greater sensitivity to microbial pathogens (Belkhadir et al., 2004; Schulze-Lefert, 2004).
Because the entry of symbiotic organisms into the root system of host plants is
accompanied by a local defense response similar to that observed in response to
pathogens, manipulation of host HSP90 by secondary metabolites might facilitate the
establishment and maintenance of fungal symbionts. MON and other HSP90 inhibitors
could also be toxic to certain plant pathogens (Ayer et al., 1980), suggesting another
potentially beneficial aspect of MON production by fungi for their associated plant
communities. As the clients of HSP90 in plants are further elucidated, the range of
possible phenotypes impacted by HSP90-producing microorganisms will no doubt
continue to expand.
HSP90 clearly plays an important role in eukaryotic development and
environmental responsiveness. The elaboration of HSP90 inhibitors by microorganisms

13
in both natural and agricultural ecosystems and the myriad ways in which such small
molecules could affect organismic interactions and phenotypes open up many exciting
new avenues for investigation.
MATERIAL AND METHODS
Solid Phase Competition Assay
MON binding to Arabidopsis HSP90 was assessed using a solid phase
competition assay (Whitesell et al., 1994). Plant tissue was prepared by grinding aerial
tissue from 14-day old Arabidopsis in a liquid nitrogen-cooled mortar and pestle. The
tissue was extracted with assay buffer containing 20 mM HEPES pH 7.3, 50 mM KCl, 5
mM MgCl
2
, 20 mM Na
2
MoO
4
and 0.01% NP-40 by rocking for 2 h at 4ºC. The lysate
was clarified by centrifugation for 30 min at 4ºC, followed by supplementation with 1
mM fresh DTT, and frozen at -80ºC. Binding assays were performed using 0.5 ml
aliquots (1.2 mg of total protein) of plant lysate supplemented with the indicated
concentrations of soluble drug as competitor or an equal volume of DMSO vehicle. After
addition of agarose beads on which geldanamycin (GDA) had been previously
immobilized as described, the reaction tubes were incubated for 1 h with gentle agitation
at 4ºC. Beads were then washed extensively and bound proteins eluted into Laemmli
sample buffer followed by SDS-PAGE and western blotting with an Arabidopsis-specific
anti-HSP90 antibody (Santa Cruz Biotechnology, Inc.).
Luciferase Renaturation Assay
Inhibition of chaperone-mediated re-folding of heat-denatured firefly luciferase
was analyzed using an adaptation of previously published methods (Kolb et al., 1994).
Recombinant firefly luciferase [Promega, 3
g/ml in sample buffer (SB) (Thulasiraman
and Matts, 1996) was denatured at 40ºC for 5 min. Native or denatured luciferase was
then diluted 1:10 into a 50 % solution of wheat germ lysate in H
2
O (Promega) that had
been previously supplemented with the indicated compounds or an equal volume of
DMSO vehicle. After incubation at 25ºC for the indicated time intervals, 5
l aliquots of
lysate mixture were removed, added to 50
l of luciferase assay reagent (Promega) and
the light intensity measured using a microplate luminometer. Each time point was

14
assayed in triplicate and the results presented depict the mean +/- SD of pooled data from
two independent experiments.
Measurement of AtHSP101 mRNA and Protein
Surface sterilized Arabidopsis seeds (Col-O accession) were germinated on
nutrient agar plates and allowed to grow for 10 d at 24
o
C. Seedlings were then sprayed
with aqueous solutions of MON at various concentrations. RNA was extracted from
pools of 12 seedlings at 90 and 210 min post MON application while for protein lysates,
seedlings were harvested approximately 18 h after being sprayed. For semi-quantitative
measurement of AtHSP101 mRNA, 5
g of total RNA from each sample was reverse
transcribed and a 648 bp AtHSP101 fragment amplified using forward
(CTGCTCAGCTGTCTGCTCG) and reverse (GCCCTTGACCTTAGAATTGCC)
primers derived from Arabidopsis locus At1g74310. The AtHSP101 fragment was co-
amplified in the same PCR reactions with primers for a control 542 bp fragment of
glyceraldehyde-3-phosphate dehydrogenase C subunit (GAPC) cDNA (Arabidopsis locus
At3g04120) and 20% GAPC amplification inhibitors as described previously (Kerschen
et al., 2004). After agarose gel electrophoresis, relative PCR product amounts were
measured by UV image optical density (Labworks Analysis Software, UVP, Inc., Upland,
CA). The reduced intensity of GAPC control bands in heat-shocked samples is due to
competition for amplification by elevated AtHSP101 transcript levels. The ratio of target
to control band intensity in the untreated sample was assigned a value of 1.0 and the
ratios determined for all other samples were normalized based on this measurement to
yield values for relative AtHSP101 transcript levels. The relative level of HSP101 protein
in seedlings was measured by immunoblotting protein extracts (10
g/lane) using a rabbit
polyclonal primary antibody (AZ 561, 1:5,000), peroxidase-conjugated secondary
antibody and chemiluminescent detection. As a positive control, seedlings were heat
shocked at 38
o
C for 90 min. RNA and protein extracts were analyzed in the same manner
as extracts prepared from MON-exposed seedlings.
Thermotolerance Assays

15
As a quantitative indicator of thermotolerance, hypocotyl elongation following
heat shock of 2.5 d old, synchronized dark grown seedlings (between 10-15 per
condition) was measured as previously described (Hong and Vierling, 2000). The
evening prior to heat shock, plates were sprayed with aqueous solutions of MON at
various concentrations or an equivalent volume of DMSO vehicle. Plants were
maintained in the dark continuously and the extent of hypocotyl growth over a 4.5 d
following heat shock was measured. After measuring hypocotyl growth, plates were
transferred to a growth chamber with a day/night cycle and photographed 14 days later.
This same experimental design was repeated three times with similar results. Mutant hot
1-3 seedlings (expressing no functional HSP101) and wild type (Col-0) planted on the
same plates were sprayed, as above, with varying concentrations of MON or DMSO. 18 h
later the plates were heat shocked at 45
o
C for 50 min, transferred to a growth champber
with a day/night cycle and then photographed 14 d later. To directly assess the effect of
the fungus P. quadriseptata on Arabidopsis heat tolerance, seeds were co-cultured with
fungus on plates and in soil. For these experiments seeds were sterilized and imbibed in
water at 4°C for a minimum of 2 d prior to transfer to sucrose-enriched plates or
autoclaved soil. Approximately 25 colony forming units (cfu) of P. quadriseptata were
then applied on top of each seed in 10
l of water. Cfu were determined experimentally
by plating dilutions of fungal suspension on potato dextrose agarose plates followed by
incubation at 28°C for 4 d. Seeds grown on plates were heat treated at 45
o
C 3 d after
plating. Hypocotyl elongation occurring in the dark after 4 d period was measured. Soil
grown plants were cultivated in 24-well tissue culture plates with drainage holes drilled
into the bottom of each well to facilitate watering. Plates were incubated under 50%
humidity and a physiological light/dark cycle. After 7 days of growth, the plants were
photographed, counted and transferred during the light phase of their cycle to an identical
incubator maintained at 45°C. After 2 h treatment at the higher temperature, the plants
were returned to their original growth conditions for 5 d at which time they were
photographed and recounted.
Preparation of Extracts

16
Samples of nutrient agar or actual fungal mass were transferred to three 50 ml
centrifuge tubes, sonicated twice in methanol (30 ml each time for 30 min) and filtered.
Combined filtrates from each of the three tubes were evaporated separately under reduced
pressure to afford crude methanolic extracts. Each of these extracts was then partitioned
between water and ethyl acetate (3 x 10 ml). Combined ethyl acetate layers were washed
separately with water (3 x 10 ml), dried over anhydrous Na
2
SO
4
and evaporated under
reduced pressure to yield ethyl acetate extracts for LC-MS analysis.
Analysis of Extracts for the Presence of MON by LC-MS
To detect the presence of MON, ethyl acetate extracts were dissolved in methanol
to a final concentration of 1.0 mg/ml and analyzed by LC-MS using a Shimadzu LCMS-
QP800
equipped with LC-10AD Liquid Chromatograph, SCL-10A System controller,
DGU-14A Degasser and SPD-M10A Diode Array Detector. The following conditions
were used: Cromasil C−18 5
m column (250 mm x 4.6 mm), 35 min linear gradient
from 60% methanol in water (containing 0.25% HCOOH) to 100% methanol, 0.4 ml/min
flow rate. APCI (Atmospheric Pressure Chemical Ionization) positive mode was used for
data acquisition. Spectra were monitored and processed using Shimadzu LabSolutions
LC-MS software.
ACKNOWLEDGEMENTS
We thank Vicki Chandler (University of Arizona) for generously providing some
equipment and reagents, and Susan Lindquist (Whitehead Institute) for useful
suggestions. H.D. VanEtten, L.S. Pierson, S. Faeth, and E.E. Pierson assisted in the
collection and identification of microorganisms while M.T. Marron, L.A. Luevano, and
C.J. Seliga provided expert technical assistance.
LITERATURE CITED
Ayer WA, Lee SP, Tsuneda A, Hiratsuka Y (1980) The isolation, identification, and
bioassay of the antifungal metabolites produced by Monocillium nordinii. Can J
Microbiol 26: 766-773

17
Belkhadir Y, Subramaniam R, Dangl JL (2004) Plant disease resistance protein
signaling: NBS-LRR proteins and their partners. Curr Opin Plant Biol 7: 391-399
David CL, Smith HE, Raynes DA, Pulcini EJ, Whitesell L (2003) Expression of a
unique drug-resistant Hsp90 ortholog by the nematode Caenorhabditis elegans.
Cell Stress Chaperones 8: 93-104
Delmotte P, Delmotte-Plaquee J (1953) A new antifungal substance of fungal origin.
Nature 171: 344-345
Gomes NCM, Fagbola O, Costa R, Rumjanek NG, Buchner A, Mendona-Hagler L,
Smalla K (2003) Dynamics of fungal communities in bulk and maize rhizosphere
soil in the tropics. Appl Environ Microbiol 69: 3758-3766
Guo Y, Guettouche T, Fenna M, Boellmann F, Pratt WB, Toft DO, Smith DF,
Voellmy R (2001) Evidence for a mechanism of repression of heat shock factor 1
transcriptional activity by a multichaperone complex. J Biol Chem 276: 45791-
45799
Gunatilaka AAL(2006) Natural products from plant-associated microorganisms:
distribution, structural diversity, bioactivity, and implications of their occurrence.
J Nat Prod 69: 509-526
Hong SW, Vierling E (2000) Mutants of Arabidopsis thaliana defective in the
acquisition of tolerance to high temperature stress. Proc Natl Acad Sci USA 97:
4392-4397
Kasuga M, Liu Q, Miura S, Yamaguchi-Shinozaki K, Shinozaki K (1999) Improving
plant drought, salt, and freezing tolerance by gene transfer of a single stress-
inducible transcription factor. Nat Biotechnol 17: 287-291
Kerschen A, Napoli CA, Jorgensen RA, Muller AE (2004) Effectiveness of RNA
interference in transgenic plants. FEBS Letts. 566: 223-228
Kolb VA, Makeyev EV, Spirin AS (1994) Folding of firefly luciferase during
translation in a cell-free system. Embo J 13: 3631-3637
Krishna P, Gloor G (2001) The Hsp90 family of proteins in Arabidopsis thaliana. Cell
Stress Chap. 6: 238-246
Marquez LM, Redman RS, Rodriguez RJ, Roossinck MJ (2007) A virus in a fungus
in a plant: three-way symbiosis required for thermal tolerance. Science 315: 513-
515
Noutoshi Y, Ito T, Seki M, Nakashita H, Yoshida S, Marco Y, Shirasu K, Shinozaki
K (2005) A single amino acid insertion in the WRKY domain of the Arabidopsis
TIR-NBS-LRR-WRKY-type disease resistance protein SLH1 (sensitive to low
humidity 1) causes activation of defense responses and hypersensitive cell death.
Plant J 43: 873-888
Omura S, Iwai Y, Takahashi Y, Sadakane N, Nakagawa A (1979) Herbimycin, a new
antibiotic produced by a strain of Streptomyces. J. Antibiotics 32: 255-261
Picard D (2002) Heat-shock protein 90, a chaperone for folding and regulation. Cell.
Mol. Life Sci. 59: 1640-1648
Pratt WB, Toft DO (2003) Regulation of signaling protein function and trafficking by
the hsp90/hsp70-based chaperone machinery. Exp Biol Med (Maywood) 228:
111-133
Queitsch C, Hong SW, Vierling E, Lindquist S (2000) Heat shock protein 101 plays a
crucial role in thermotolerance in Arabidopsis. The Plant Cell 12: 479-492

18
Queitsch C, Sangster TA, Lindquist S (2002) Hsp90 as a capacitor of phenotypic
variation. Nature 417: 618-624
Rinehart JP, Hayward SA, Elnitsky MA, Sandro LH, Lee RE, Jr., Denlinger DL
(2006) Continuous up-regulation of heat shock proteins in larvae, but not adults,
of a polar insect. Proc Natl Acad Sci USA 103: 14223-14227
Roe SM, Prodromou C, O'Brien R, Ladbury JE, Piper PW, Pearl LH (1999)
Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor
antibiotics radicicol and geldanamycin. J. Med. Chem. 42: 260-266
Rutherford SL, Lindquist S (1998) Hsp90 as a capacitor for morphological evolution.
Nature 396: 336-342
Sangster TA, Lindquist S, Queitsch C (2004) Under cover: causes, effects and
implications of Hsp90-mediated genetic capacitance. Bioessays 26: 348-362
Sangster TA, Queitsch C (2005) The HSP90 chaperone complex, an emerging force in
plant development and phenotypic plasticity. Curr Opin Plant Biol 8: 86-92
Schulte TW, Akinaga S, Murakata T, Agatsuma T, Sugimoto S, Nakano H, Lee Y,
Simen B, Argon Y, Felts S, Toft D, Neckers LM, V. SS (1999) Interaction of
radicicol with members of the heat shock protein 90 family of molecular
chaperones. Mol Endo 13: 1435-1448
Schulze-Lefert P (2004) Plant immunity: the origami of receptor activation. Curr Biol
14: R22-24
Sollars V, Lu X, Xiao L, Wang X, Garfinkel MD, Ruden DM (2003) Evidence for an
epigenetic mechanism by which Hsp90 acts as a capacitor for morphological
evolution. Nat Genet 33: 70-74
Thulasiraman V, Matts RL (1996) Effect of geldanamycin on the kinetics of
chaperone-mediated renaturation of firefly luciferase in rabbit reticulocyte lysate.
Biochemistry 35: 13443-13450
Turbyville TJ, Wijeratne EMK, Liu MX, Burns AM, Seliga CJ, Luevano LA, David
CL, Faeth SH, Whitesell L, Gunatilaka AAL (2006) Search for Hsp90
inhibitors with potential anticancer activity: isolation and SAR studies of radicicol
and monocillin I from two plant-associated fungi of the Sonoran desert. J Nat
Prod 69: 178-184
Voellmy R, Boellmann F (2007) Chaperone regulation of the heat shock protein
response. Adv Exp Med Biol 594: 89-99
Whitesell L, Lindquist SL (2005) HSP90 and the chaperoning of cancer. Nat Rev
Cancer 5: 761-772
Whitesell L, Mimnaugh EG, De Costa B, Myers CE, Neckers LM (1994) Inhibition of
heat shock protein HSP90-pp60v-src heteroprotein complex formation by
benzoquinone ansamycins: essential role for stress proteins in oncogenic
transformation. Proc Natl Acad Sci USA 91: 8324-8328
Wijeratne EMK, Carbonezi CA, Takahashi JA, Seliga CJ, Turbyville TJ, Pierson
EE, Pierson III LS, VanEtten HD, Whitesell L, Bolzani VdaS, Gunatilaka
AAL (2004) Isolation, optimization of production and structure-activity
relationship studies of monocellin I, the cytotoxic constituent of
Parapheosphaeria quadriseptata. J Antibiot 57: 541-546
Yeyati PL, Bancewicz RM, Maule J, van Heyningen V (2007) Hsp90 Selectively
Modulates Phenotype in Vertebrate Development. PLoS Genet 3: e43

19
Young JC, Moarefi I, Hartl FU (2001) Hsp90: a specialized but essential protein-
folding tool. J Cell Biol 154: 267-273
Zou J, Guo Y, Guettouche T, Smith DF, Voellmy R (1998) Repression of heat shock
transcription factor HSF1 activation by HSP90 (HSP90 complex) that forms a
stress-sensitive complex with HSF1. Cell 94: 471-480

20
Figure Captions
Figure 1. MON binds plant HSP90 and inhibits its activity. (A) Structures of MON
(R = H) and RAD (R = Cl). (B) HSP90 binding. Aliquots of drug-supplemented lysate of
14-day old aerial tissue from Arabidopsis seedlings were incubated with bead-
immobilized GDA. The relative binding of HSP90 was assessed by immunoblot using an
Arabidopsis HSP90-specific antibody. Control indicates a sample of resin beads without
immobilized GDA. (C) Luciferase renaturation. Heat-denatured luciferase was added to
wheat germ lysate supplemented with the indicated compounds and the time-dependent
recovery of activity was measured. Each time point was assayed in triplicate. The mean
and standard deviation of data derived from two independent experiments is depicted.
Recovery of total input luciferase activity for DMSO control samples over the course of
experiments was 61 % +/- 4%.
Figure 2. MON increases HSP101 expression in Arabidopsis seedlings. (A) Semi-
quantitative measurement of AtHSP101 mRNA by RT-PCR. Seedlings were sprayed with
aqueous solutions of MON or DMSO vehicle and harvested for RNA extraction at the
times indicated. As a positive control, seedlings were heat shocked (Heat). Fragments of
the AtHSP101 gene (black arrow) and the control housekeeping gene GAPC (gray arrow)
were co-amplified from the same reverse transcribed RNA samples and analyzed by gel
electrophoresis. Relative levels of AtHSP101 normalized to GAPC signal are indicated
below the gel. (B) Measurement of relative HSP101 protein levels by immunoblotting.
Seedlings were sprayed with aqueous solutions of MON at the indicated concentrations
or DMSO vehicle, and total protein was extracted the following day. Lysate from heat-
shocked seedlings (38
o
C for 90 min) served as a positive control (Heat).
Figure 3. MON induces a thermotolerant phenotype in Arabidopsis seedlings. (A)
Measurement of hypocotyl elongation after heat treatment. MON was applied topically to
dark grown seedlings 18 h prior to heat shock under the conditions indicated. Hypocotyl
elongation occurring in the dark over a 4 d period following heat treatment was measured

21
as previously described (Hong and Vierling, 2000; Queitsch et al., 2000). The mean
increase observed for 15 plants per condition is plotted. Error bars: SD. Star indicates that
no column appears because the value was 0. (B) MON-treated seedlings continue to
develop normally. Plates were transferred to a growth chamber with a day/night cycle,
and photographed 14 d later. Results presented are representative of 3 independent
experiments.
Figure 4. HSP101 is required for MON-induced thermotolerance. A solution of MON (2
µm) or DMSO vehicle were applied to dark grown wild type (Col) or mutant (hot 1-3)
Arabidopsis seedlings (Col-O accession). hot 1-3 mutant seedlings express no functional
HSP101 (Hong and Vierling, 2000). 18 h after MON application, seedlings were heat
shocked at 45
O
C for 50 min (45 X 50). Plates were then transferred to a growth chamber
with a day/night cycle, and photographed 14 d later.
Figure 5. Co-culture with the fungus P. quadriseptata induces a thermotolerant
phenotype in Arabidopsis seedlings. (A) Measurement of hypocotyl elongation after
heat stress. Seeds were plated along with approximately 25 colony forming units (cfu) of
P. quadriseptata. Three days later, plates were heat treated at 45
o
C for the indicated time
intervals and hypocotyl elongation occurring in the dark over a 4 d period following heat
treatment was measured. The mean increase observed for 20 to 30 plants per condition is
indicated. Error bars: SEM. (B) Plants that were not heat shocked and plants heat
shocked at 45°C for 75 min were photographed at the conclusion of the experiment.
Figure 6. Detection of MON in nutrient agar recovered from plates following plant-
fungus co-culture. (A) Analysis of methanolic extract prepared from agar approximately 1
cm distant from the location of Arabidopsis − P. quadriseptata growth zone. (B) Analysis
of authentic MON as a standard. The liquid chromatograms (LC) in panels A and B
demonstrate a major UV absorbance peak eluting at approximately 17.5 min as indicated
by the arrows. UV and mass spectrometry (MS) performed on material recovered from
these peaks confirm that the peak in panel A contains MON with the expected molecular
mass [M+1]
+
of 331 Da. The chromatogram and spectra of extract prepared directly from

22
plate-grown fungus appeared essentially identical to that shown in panel A, while extract
of agar from plates containing seedlings but no fungus was found to contain no detectable
MON (data not shown).
Figure 7. Cultivation with the fungus P. quadriseptata induces a thermotolerant
phenotype in soil grown Arabidopsis. Height of each bar depicts the percent of plants
surviving 5 d after a 2 h heat shock at 45°C. Results from two independent experiments
are displayed in which each experimental group consisted of approximately 20 plants. In
both experiments, 90-100% of plants that received no heat shock survived over the 5 d
time period, irrespective of the presence or absence of fungus.