RESEARCH PAPER

https://doi.org/10.1071/FP21238

Endophytic fungi and drought tolerance: ecophysiological

adjustment in shoot and root of an annual mesophytic host grass
Milena E. Manzur A,B , Fabián A. Garello B,C, Marina Omacini C,D, Hans Schnyder E, Moira R. Sutka F and
Pablo A. García-Parisi C,G,*

ABSTRACT
For full list of author affiliations and
declarations see end of paper
Epichloid endophytic fungi, vertically transmitted symbionts of grasses, can increase plant tolerance
to biotic and abiotic stress. Our aim was to identify ecophysiological mechanisms by which the
*Correspondence to:
Pablo A. García-Parisi
endophyte Epichloë occultans confers drought tolerance to the annual grass Lolium multiflorum
IFEVA-CONICET, Facultad de Agronomía, Lam. Endophyte-associated or endophyte-free plants were either well-watered or subjected to
Universidad de Buenos Aires, Avenida San water deficit. We evaluated plant biomass, root length and nitrogen concentration, and we assessed
Martín 4453, C1417DSE Buenos Aires,
intrinsic water use efficiency (iWUE) and its components net photosynthesis and stomatal
Argentina
Email: [email protected] conductance, by carbon and oxygen isotope analysis of shoot tissues. Endophyte-free plants
produced more biomass than endophyte-associated ones at field capacity, while water deficit
Handling Editor: strongly reduced endophyte-free plants biomass. As a result, both types of plants produced
Wieland Fricke similar biomass under water restriction. Based on oxygen isotope composition of plant cellulose,
stomatal conductance decreased with water deficit in both endophyte-associated and endophyte-
free plants. Meanwhile, carbon isotope composition indicated that iWUE increased with water
deficit only in endophyte-associated plants. Thus, the isotope data indicated that net photosynthesis
decreased more strongly in endophyte-free plants under water deficit. Additionally, endophyte
presence reduced root length but increased its hydraulic conductivity. In conclusion, endophytic
fungi confer drought tolerance to the host grass by adjusting shoot and root physiology.

Keywords: ecophysiology, fungal infection, mutualism, plant stresses, root, symbiosis, water stress
physiology, water use efficiency.

Introduction

Water limitation is one of the major constraints to plant productivity in many natural
ecosystems and agroecosystems (Knapp and Smith 2001). As predicted by climate
change models, the intensity and frequency of drought may increase in some regions
(Easterling et al. 2000; Sheffield and Wood 2007). Plants that show a lesser reduction in
fitness with increasing water deficits are considered more tolerant (Vile et al. 2012).
These tolerant plants can develop different responses to an environmental gradient of
water deficit, including structural and/or functional adjustment either in shoot or roots
Received: 19 January 2021
(Sultan 2000). In this respect, different response types have been extensively described
Accepted: 20 December 2021 for drought-tolerant crops, including the isohydric and anisohydric strategies (Klein
Published: 8 February 2022 2014; Martínez-Vilalta et al. 2014; Sutka et al. 2016). The isohydric strategy implies
that plants reduce stomatal conductance, keeping water potential and relative water
Cite this: content constant (Klein 2014; Martínez-Vilalta et al. 2014); while the anisohydric
Manzur ME et al. (2022) strategy sustains transpiration rates, facilitated by decreases of the water potential and
Functional Plant Biology
relative water content (Moshelion et al. 2015). Furthermore, some plant species can
doi:10.1071/FP21238
increase drought tolerance through different symbiotic associations (Douglas 2010).
Here we explore putative ecophysiological mechanisms by which a vertically transmitted
© 2022 The Author(s) (or their
employer(s)). Published by
endosymbiont may confer drought tolerance to its host.
CSIRO Publishing. Among vertically transmitted symbionts, Epichloë fungal endophytes (Clavicipitaceae,
Hypocreales) constitute one of the most conspicuous plant endosymbionts since they
M. E. Manzur et al. Functional Plant Biology

inhabit aerial tissues of many cool-season grasses of the is that endophyte presence confers capacities to maintain
subfamily Poöideae (Poaceae), including many important water status through hydraulic adjustment in the host
cultivated and wild forage species (Leuchtmann et al. plant, by reducing water loss or maintaining water uptake.
2014). This plant–fungal interaction has been labelled as a We predict that endophyte-associated plants will show
defensive mutualism, since the fungus protects the host higher iWUE, maintaining assimilation rates, and a greater
plant from herbivory (Clay 1988; Li et al. 2014). However, root system with increased hydraulic conductance compared
several additional benefits provided by the symbiont to the to endophyte-free plants. As a study system, we used the
host grass have been reported, including pathogen tolerance, symbiosis between the endophyte Epichloë occultans
nutrient acquisition and abiotic stress tolerance (Cheplick and (C.D.Moon, B.Scott and M.J.Chr.) Schardl and the grass
Faeth 2009). In particular, tolerance to drought is commonly L. multiflorum. This mesophytic grass inhabits grasslands
mentioned as an additional benefit for host plants (e.g. Davitt and pastures of temperate subhumid ecosystems in which
et al. 2011; but see Gundel et al. 2016). While the ability the occurrence of drought events has increased due to
of these fungal endophytes to provide drought tolerance climate change. Furthermore, as an annual grass,
in host plants has been described in many studies, the propagation by seed is a crucial process of plant population
underlying mechanisms are incompletely characterised dynamics, which strengthens the relationship between host
(Dastogeer 2018). and symbiont fitness (Gundel et al. 2008).
Isohydric and anisohydric strategies provide an
appropriate framework for understanding endophyte effects
on drought tolerance of host plants. Isohydric mechanisms Materials and methods
have been observed in endophytic plants, and include
changes in stomatal closure timing and rate (Elmi and The study comprised two manipulative experiments that
West 1995; Malinowski et al. 1997) and root morphological evaluate endophyte presence effect on different plant attributes
traits (Latch et al. 1985; De Battista et al. 1990; Malinowski under different water availability conditions. In experiment 1
et al. 1997) of endophytic plants. The greatest challenge for we studied plant performance and shoot physiology: plant
plants using these mechanisms is maintaining or minimising biomass production and partitioning, iWUE, net photosyn-
losses in photosynthesis and growth rates. Stomatal closure is thesis and N concentration. In experiment 2 we studied root
the first plant response aimed at reducing water losses. physiology: root hydraulic conductivity and conductance,
However, stomatal closure can also affect plant growth root length and root diameter.
rate since it reduces carbon dioxide (CO2) uptake for The two experiments were performed with individual plants
photosynthesis. In this sense, intrinsic water use efficiency of L. multiflorum either associated or not with endophyte
(iWUE) links carbon (C) assimilation rates with stomatal E. occultans (E+ and E− respectively). We generated
conductance. Endophytes can affect iWUE components by endophyte-free and endophyte-associated L. multiflorum from
increasing or maintaining photosynthesis rates when stomatal seeds collected 1 year prior to the experiment from an old-field
conductance decreases due to water limitation (Swarthout Pampean grassland (Carlos Casares, Argentina 34°06 0 S,
et al. 2009; Xia et al. 2018). Similarly, endophyte-mediated 60°25 0 W) dominated by a L. multiflorum population with
changes in anisohydric mechanisms have also been ≈95% endophytic association (Omacini et al. 2006). Half of
reported, including the translocation of assimilates from the seeds collected were treated with the fungicide triadimenol
the leaves (Richardson et al. 1992), osmotic adjustment (0.5 g per 100 g seeds) to eliminate the endophyte. Fungicide-
(reviewed in Malinowski and Belesky 2000), and changes treated and untreated seeds were cultivated in 1 m2 plots and
in cell wall elasticity (White et al. 1992). Despite all these the seeds produced by those plants were harvested and used in
studies, it is not known whether endophytes change shoot the experiment as E− and E+ seeds, respectively. Endophyte
and root functioning simultaneously in annual mesophytic incidence of the seeds was evaluated through microscopic
grasses, which are currently being subjected to increasing observation of 30 seeds collected from each plot, stained
intensity or frequency of drought events. with bengal rose (Bacon and White 1994). All seeds from the
The aim of this research is to understand the ecophysi- E+ population showed presence of the endophyte, while no
ological mechanisms by which an epichloid endophytic endophyte presence was observed in any of the seeds from
fungus confers water deficit tolerance to the annual grass the E− population.
Lolium multiflorum Lam. Our first hypothesis is that
endophyte presence confers drought tolerance, considering
Experimental design and growth conditions
tolerance as the change in plant performance in response to
stress (modified from Simms 2000; Stowe et al. 2000). Experiment 1
Thus, we predict that the slope of the reaction norm that The experiment 1 had a factorial design with the factor
describes the change in performance in response to water endophyte level (E− and E+) and the factor water
deficit will be lower (closer to 0) in endophyte-associated availability treatment (five levels: 0%, 5%, 25%, 50% and
plants than in endophyte-free plants. Our second hypothesis 100% of field capacity). Each combination of endophyte

B
www.publish.csiro.au/fp Functional Plant Biology

level and water treatments was repeated four times (four and the shoot mass fraction by dividing shoot biomass by
blocks) resulting in 40 experimental pots. The experiment was total plant biomass.
carried out in a plant growth chamber, under controlled con- On days 1, 2, 4 and 7 after starting the water restriction
ditions with a 16/8 h light/dark cycle, 205 ± 10 μmol m−2s−1 treatments, we measured the length of one elongating
(mean ± s.d.) photosynthetic photon flux density, 60 ± 2.5% leaf per plant to verify if (and how much) plants were
(mean ± s.d.) relative humidity and 21 ± 2°C (mean ± s.d.) effectively growing (Supplementary Fig. S1). Furthermore,
air temperature. to verify that growth actually occurred during the water
Endophyte-free or endophyte-associated seeds of restriction periods, relative whole plant biomass of each
L. multiflorum were germinated in Petri dishes over 4 days. treatment was calculated as the difference between plants
Then, seedlings were transferred to 40 individual plastic with FC, 50, 25 or 5% FC and the average of the 0% FC
containers (330 mL, one seedling per pot) filled with plants of the corresponding endophytic status, relative to
sterilised sand and moistened with Hoagland’s solution, that average (Fig. S2).
arranged in four blocks in a growing chamber. Pots were
irrigated with Hoagland’s solution until the 3rd true leaf Experiment 2
was completely expanded in most of the plants (25 days). Experiment 2 had a factorial design with the factor
After that, water availability treatments were applied, and endophyte level (E− and E+) and the factor water avail-
no more Hoagland’s solution was provided. For the control ability treatment (three levels: 0%, 25% and 100% of field
treatment, pots were irrigated every 2 days with the amount capacity). Each combination of endophyte level and water
of water needed to reach field capacity (FC), judged by the availability was repeated 18 times (18 blocks) resulting
onset of drainage. Each time, the amount of water was in 108 experimental pots. The experiment was performed
measured. Then, the other treatments were irrigated with in a greenhouse, during early spring, at 25.5 ± 1.0°C
50%, 25%, 5% and 0% of that amount of water. Before (mean ± s.d.) and 14.0 ± 0.5°C (mean ± s.d.) as daily
watering, a small portion of the sandy substrate was removed maximum and minimum temperature, a daily average relative
to estimate water content of every pot by weighing the sand humidity of 80.4 ± 2.3 % (mean ± s.d.) and midday
before and after oven-drying. Water dynamics during the photosynthetic photon flux density 249 ± 38 μmol m−2s−1
15 day-long drying cycle are shown in Fig. 1a. Each treatment (mean ± s.d.).
was maintained for 15 days, until plants at 0% FC died. Then, This experiment used 220 mL pots, filled with sterilised dry
all plants were sampled. Shoot and root biomass was sand. Pots were weighed when dry and after watering to FC.
separated. Shoot tissue was separated into blade and sheath The pot dry weight and the amount of water at FC was
parts. All biomass was oven dried (60°C) during 72 h and registered for each pot (calculated as the difference between
weighed. Then, shoot biomass was ground for nitrogen (N) weight at FC and pot dry weight). Then, endophyte-free (E−)
and isotopic analyses. Whole plant biomass was obtained as or endophyte-associated (E+) seeds of L. multiflorum were
the sum of shoot and root biomass. The blade mass fraction germinated in Petri dishes. Four days after germination,
was obtained by dividing blade mass by total shoot mass, seedlings were transferred to the plastic pots (one seedling

Fig. 1. Evolution of water availability in pots during the experiments. (a) Experiment 1: watering treatments were applied
when plants reached the stage where the third leaf was completely expanded (day 0), being applied every 2 days. Pots in
the control treatment were watered with the amount of water necessary to reach field capacity, or with the 50%, 25%, 5%
and 0% of this value. (b) Experiment 2: water treatments were applied at day 5 after germination started, keeping pots at
FC, at 25%; or were never watered again (0% FC). Each value represents mean ± s.e.m.

C
M. E. Manzur et al. Functional Plant Biology

per pot), and the pots were moistened to FC with Hoagland’s For the isotopic analysis, dried material of the shoot
solution and arranged in the greenhouse. Water availability samples was ground to a fine powder in a ball mill. Tissue
treatments were applied on day 5 after germination started C and N concentration (%C and %N) and δ13Cp were
(i.e. 1 day after transplanting). After that, pots were weighed determined on aliquots of 0.7 ± 0.05 mg dry mass weighed
daily and watered with the amount of water required to reach into tin cups (IVA Analysetechnik, Meerbusch, Germany),
FC, or 25% of FC or not watered at all (0% FC). Water combusted in an elemental analyser (NA1110, Carlo
dynamics within the pots during the experiment is shown Erba Instruments, Milan, Italy) interfaced to a continuous-
in Fig. 1b. flow isotope ratio mass spectrometer (IRMS, Delta Plus,
Plants were harvested 14 days after germination. The shoot Finnigan MAT, Bremen, Germany). Samples were measured
(3 mm above the crown) was clipped. Roots were carefully against a working CO2 gas standard previously calibrated
cleaned, scanned and analysed with WinRHIZO (V5.0, Regent against a secondary isotope standard (International Atomic
Instruments Inc., Quebec, Canada) for morphological Energy Agency sucrose, IAEA-CH6, accuracy ± 0.06‰ s.d.).
analyses. Roots from six blocks were chosen to analyse root A laboratory standard (wheat flour) was run after every
hydraulic conductance (L0) and conductivity (Lp). 10 samples to estimate the precision of the isotope analyses
(±0.10‰ s.d.) and as a standard for determination of N
concentration.
Physiological variables determinations
Oxygen isotope composition was determined on cellulose
Water use efficiency and its components from aerial tissues (δ18Op). α-cellulose was extracted from
Endophyte effects on intrinsic water use efficiency (iWUE), 50 mg or 25 mg of dry sample material using the procedure
assimilation rate (A) and stomatal conductance (gH2O) of of Brendel et al. (2000) as modified by Gaudinski
plants from the first experiment were estimated using the et al. (2005). Cellulose samples were re-dried at 40°C for
so-called dual-isotope approach. The dual-isotope approach 24 h, 700 μg aliquots were packed in silver cups (size:
involves carbon (δ13CP) and oxygen (δ18OP) isotope analysis 3.3 × 5 mm, LüdiSwiss, Flawil, Switzerland) and stored
in shoot biomass or cellulose, and evaluates whether (whole above Silica Gel orange (2–5 mm, ThoMar OHG, Lütau,
shoot, lifespan-based) differences in leaf-level iWUE were Germany) in exsiccator vessels prior to analysis. For 18O
driven by changes in assimilation rate or stomatal conduc- analysis, samples were pyrolysed at 1400°C in a pyrolysis
tance (Scheidegger et al. 2000; Moreno-Gutiérrez et al. 2012; oven (HTO, HEKAtech, Wegberg, Germany), equipped with
see below). δ13CP relates to intrinsic water use efficiency a helium-flushed zero blank auto-sampler (Costech Analytical
(iWUE) via its relationship with Ci/Ca, the ratio of leaf technologies, Valencia, CA, USA) interfaced to a continuous-
internal CO2 to atmospheric CO2 concentrations, as flow isotope ratio mass spectrometer (ConFlo III, Finnigan
MAT, Bremen, Germany; Delta Plus, Finnigan MAT). Solid
iWUE = Ca ð1 − Ci =Ca Þ=1.6, (1) internal laboratory standards (SILS, cotton powder) were
run as a control after every fifth sample. All samples and
with 1.6 the ratio of stomatal conductance for water vapour SILS were measured against a laboratory working standard
and CO2, and carbon monoxide gas, which was previously calibrated
against a secondary isotope standard (IAEA-601). The long-
δ13 CP = δ13 Ca − a − ðb − aÞCi =Ca , (2) term precision for the internal laboratory standards was
better than 0.3‰ (s.d. for repeated measurements).
with δ13Ca, the δ13C of atmospheric CO2, and a (4.4‰) and b
(27‰) the 13C discriminations associated with diffusion Root hydraulic conductance (L0) and
of CO2 through the stomata and CO2 fixation, mainly by conductivity (Lp)
RuBisCO (Ribulose-1,5-bisphosphate carboxylase-oxygenase; The modified protocol described by Miyamoto et al.
Farquhar et al. 1982, 1989). Eqn 2, solved for Ci/Ca can be (2001) was used to measure root hydraulic conductance
substituted in Eqn 1 to estimate iWUE from δ13Cp. Although and conductivity in L. multiflorum seedlings of the second
the full theory relating δ13CP to iWUE is more complex (Ma experiment. Before starting the measurements, sheaths and
et al. 2021), empirical work has provided strong support for blades were removed using razors. Blades and roots were
a very close linear relationship between δ13C and iWUE washed with tap water. Each root system was immersed in
particularly for plants in the same environment (Condon a 50 mL plastic tube filled with distilled water and inserted
et al. 2004; Yang et al. 2016). into a pressure chamber (Biocontrol Model 4, Argentine).
δ18Op can be used as an indicator of stomatal conductance The flux of root exudates (Jv) induced by increasing pressure
(gH2O; with stomatal conductance negatively related with to 0.2, 0.3 and 0.4 MPa was determined following Matsuo
δ18Op) if the source of variation of δ18Op is stomatal et al. (2009) with some modifications. Exuded sap at the
conductance not evaporative demand (Farquhar et al. 2007), root-shoot junction was collected at each pressure for 3 min
a condition likely met in comparisons of δ18Op measured in on tissue paper that had been weighed previously. Then,
the same environment. the tissue paper was weighed again on an analytical balance

D
www.publish.csiro.au/fp Functional Plant Biology

13
(0.1 mg accuracy), this process was repeated three times for C isotope composition and root hydraulic conductivity
each pressure increment. Root hydraulic conductance (L0) due to non-homogeneous variances among treatments. The
was obtained as the slope of the Jv vs pressure relationship. estimates produced by linear mixed effect models (LME)
To obtain root hydraulic conductivity (Lp), L0 values were allowed modelling the variances and the correlation of
divided by the total root surface area (cm2) of the errors within each group (factors levels) for the fixed effects
corresponding plant (see below). models (Pinheiro et al. 2015). The significance of the fixed
factors of each model were assessed through likelihood
Root morphology ratio test (LRT) with the function ANOVA. Analyses were
After L0 measurements, the root system of each plant performed with LME with the package nlme (Pinheiro et al.
from the second experiment was washed and scanned 2015) using statistical software R.
using the WinRHIZO root-scanning software (V5.0, Regent
Instruments Inc., Quebec, Canada). The software enabled
estimation of root length and root surface area of the
Results
complete root system. Furthermore, the software permitted
estimation of root length for different diameter classes
(fine: 0–0.25 mm, coarse: 0.25–0.8 mm). Root diameter
Plant performance and shoot physiology (exp 1)
classes greater than 0.8 mm were not considered as such Among plants of all treatments the effect of water deficit on
measurements could have resulted from overlapping of whole plant biomass differed between endophyte-associated
several roots. Fine root fraction (i.e. the proportion of the (E+) and endophyte-free (E−) plants. Although both groups
length represented by fine roots) was obtained by dividing of plants linearly and negatively responded to a reduction
the total length of fine roots by total root length (<0.8 mm in water availability (Fig. 2), the relative reduction of
diameter). Following measurements, roots were oven dried biomass due to water limitation was greater in E− plants
at 60°C for 72 h and weighed to obtain dry weight. than in E+ plants (slope test, F2,36 = 6.85, P = 0.003).
Although biomass in E− plants was much greater than in
Statistical analyses E+ plants at 100% FC, it was not significantly different in
For Experiment 1, linear regression analyses were developed drought treatments.
with water availability (0%, 5%, 25%, 50% and FC) as the When comparing 50% and 100% FC, shoot biomass was
independent variable and whole plant biomass as the interactively affected by endophyte and water deficit since
dependent variable, separately for E− and E+ plants. Slopes water deficit only reduced biomass in E− plants (Table 1;
of both regressions were compared with F-tests (slope test). Fig. 3a). Meanwhile, the leaf blade mass fraction was
If lines were statistically different, the slope of each group reduced by endophyte presence by about 56% irrespective
was compared against zero with an F-test. Lines were of water deficit (Fig. 3b). Root biomass was reduced by both
adjusted by minimising sum-of-squares with the statistical endophyte presence and water deficit (Fig. 3a). Accordingly,
software GraphPad Prism 6. Shoot biomass, blade mass
fraction, shoot mass fraction, 13C and 18O isotope composi-
tion and N concentration in biomass were analysed with
linear mixed effect models including the level of endophyte
presence on plants and the water availability (FC and 50%
FC levels of water availability) as fixed factors, and blocks
as a random factor. Experiment 2: root hydraulic conductance
(L0) and conductivity (Lp), root length and fine root fraction
were also analysed with linear mixed effect models including
the level of endophyte presence on plants and the water
availability (0%, 25% and FC) as fixed factors, and blocks
as a random factor.
Normality and homogeneity of variances were evaluated
with the Shapiro-Wilk normality test and Levene’s test
Fig. 2. Whole plant biomass of endophyte-free (E−, open symbols)
for homogeneity of variance. These tests were conducted
or endophyte-associated (E+, filled symbols) L. multiflorum plants
using the shapiro.test function from the stats-package in R
from the first experiment grown in different water availability
(R Core Team 2018), and the leveneTest function [expressed as a percentage of field capacity (FC)]. Values are
from the car package in R (Fox and Weisberg 2019). The mean ± s.e.m. of four plants. Regression lines were significantly
varFunc = varIdent function in the nlme package in R was different between E− and E+ (F1,36 = 9.57; P = 0.003), and both
used to stratify the variances within the levels of a factor slopes were significantly different to 0 (E−: F1,18 = 26.6, P < 0.001;
(Pinheiro et al. 2015) when analysing percentage of blade, E+: F1,18 = 4.56, P = 0.04, indicated with *** and *, respectively).

E
M. E. Manzur et al. Functional Plant Biology

Table 1. Statistical analyses.

Water treatment Endophyte W×E
χ2 P χ2 P χ2 P
Experiment 1
Shoot biomass 35.1 <0.001 0.039 0.84 4.45 0.034
Blade mass fraction 0.49 0.48 4.61 0.03 0.77 0.37
Root biomass 4.58 0.03 9 0.002 1.16 0.28
Shoot mass fraction 0.01 0.98 80.5 <0.001 1.42 0.23
δ13C 0.01 0.92 2.92 0.08 7.25 0.007
δ18O 10.4 0.001 13.1 <0.001 3.16 0.075
N concentration 7.07 0.007 0.74 0.38 9.14 0.002
Experiment 2
Root length 0.12 0.94 7.02 0.008 4.26 0.118
Fine root length 0.22 0.89 2.89 0.08 1.63 0.439
Coarse root length 1.38 0.49 0.09 0.76 0.23 0.88
Fine root fraction 0.97 0.61 3.71 0.05 1.31 0.51
Root hydraulic conductance 11.4 0.003 0.042 0.837 1.63 0.44
Root hydraulic conductivity 12.48 0.001 3.4 0.06 1.38 0.501

Chi-squared (χ2) values from statistical analyses of experiment 1 response variables: shoot biomass; proportion of blade; root biomass; proportion of shoot; δ13C; δ18O;
and N concentration; and from experiment 2: root length; fine root length; coarse root length; proportion of fine roots and root hydraulic conductance (L0) and
conductivity (Lp), as affected by water treatments, endophyte presence, and interaction (W × E).
P < 0.05 are denoted in bold font.
df, degree of freedom of Chi-squared (χ2) test.

Fig. 3. (a) Shoot (blade: white bars, sheath:

grey bars) and root (black bars) biomass
(g plant−1) of endophyte-free (E−) or
endophyte-associated (E+) L. multiflorum
plants grown under field capacity or 50% field
capacity; (b) proportion of the shoot that is
blade (mg blade mg shoot−1); (c) nitrogen (N)
concentration in shoot (%; mg N mg shoot
*100−1). Significance of the factors water
availability (W) or endophyte presence (E)
and the interaction (E × W) is indicated
above or inside the panels (n.s.: P > 0.05,
*P < 0.05, **P < 0.01). When interaction was
significant, different letters above the bars
indicate differences among treatments
(P > 0.05, Tukey test) for shoot biomass or N
concentration. Values are mean ± s.e.m. of
four plants in (a) and (c), and of eight plants
in (b).

the shoot mass fraction was increased by endophyte presence affected by water deficit. Nitrogen concentration was
both in FC (from 0.58 ± 0.04 to 0.66 ± 0.04) and in 50% FC interactively affected by water treatments and endophyte
(from 0.58 ± 0.082 to 0.72 ± 0.044). It was not significantly presence. Water deficit increased N concentration (from

F
www.publish.csiro.au/fp Functional Plant Biology

1.15 to 1.6%) in E− plants but did not affected it in E+ plants. Root physiology (experiment 2)
Indeed, under field capacity (100% FC), E+ plants showed
Root system function and structure were affected by both the
considerably higher N concentration than E− plants (1.76%
presence of endophyte and water treatments. Water deficit
vs 1.15%, Fig. 3a, c).
reduced root hydraulic conductance (L0) and conductivity
When grown at field capacity, shoot tissues of E+ plants
(Lp) while endophyte presence marginally increased Lp
were depleted in 13C by 1.2‰, (Fig. 4) relative to E− plants
(Fig. 5a). Instead, total root length was reduced by
tissues, indicating a lower iWUE in E+ in comparison with
endophyte presence (Fig. 5b). The proportion of fine roots
E− plants. Simultaneously, E+ plants had consistently lower
biomass at field capacity. Conversely, a considerable enrich-
ment in 13C (by 2.2‰, Fig. 4) of the shoot tissues was induced
by water deficit (50% FC) in E+ plants. This change corre-
sponded to an increase of iWUE of about 75%, when estimated
using a CO2 concentration in air of 410 μmol mol−1 and δ13Cair
of −10.6‰ (considering a small anthropogenic contribution
to δ13Cair inside the growth chamber; see simulation
assuming different δ13Cair in Fig. S3). In E− plants, no effect
of water deficit on δ13C of the shoot tissues was detected.
Furthermore, water deficit induced an increase of δ18O (+0.8
and +1.4‰, Fig. 4) of cellulose of both E− and E+ plants,
consistent with a lower stomatal conductance (gH2O) of these
plants. According to the Scheidegger et al. (2000) model,
these changes in C and O isotope compositions indicate
that in E+ plants, the water deficit led to an increase of
iWUE due to a distinct decrease in stomatal conductance.
Conversely, in E− plants, the water deficit led to a smaller
decrease in stomatal conductance, that did, however, not
increase iWUE as net CO2 assimilation changed in parallel.
Thus, together the isotope data suggested a greater drought-
related decrease in net CO2 assimilation rate in E− relative
to E+ plants.

Fig. 5. Properties of root system: (a) root hydraulic conductivity (Lp,

Fig. 4. Carbon (δ13C) and oxygen (δ18O) isotope composition of μL s−1 MPa−1cm−2 of root); (b) root length (cm plant−1), (c) proportion
endophyte-free (E−, empty symbols) or endophyte-associated of the length represented by fine roots (length of fine roots per total
(E+, filled symbols) L. multiflorum plants grown under field capacity root length−1) of endophyte-free (E−) or endophyte-associated (E+)
(FC, triangles) or 50% of field capacity (squares). The factors water L. multiflorum plants grown under field capacity, 25% or 0% FC. No
availability (W) and endophyte presence (E) significantly affected interaction between the factors water availability (W) and endophyte
δ18O (P < 0.01 and P < 0.001, respectively), while the interaction presence (E) was significant. When the main factor was significant,
(E × W) affected δ13C (P < 0.01). Different letters beside the different letters above bars indicate differences among treatments
points indicate differences among treatments for δ13C as the interaction (P < 0.05, Tukey test). # in (a) means tendency (P = 0.06). Values
was significant (P < 0.05, Tukey test). Values are mean ± s.e.m. of are mean ± s.e.m. of 54 plants in the left panels (endophyte
four plants. presence) and of 36 plants in the right panels (water availability).

G
M. E. Manzur et al. Functional Plant Biology

(less than 0.25 mm root diameter) was reduced by endophyte free plants showed similar N concentration in whole shoot
presence but was not affected by water deficit (Fig. 5c). biomass under water limitation. Yet, the leaf blade mass
fraction was reduced by endophyte presence, and assuming
that most of the shoot N is concentrated in leaf blades
Discussion (Jaramillo and Detling 1988) it seems possible that N
concentration in leaf blades of endophyte-associated plants
Here we found that endophyte presence within an annual was actually higher. Hence, an increase in instantaneous
grass conferred tolerance to water deficit to the host plants, photosynthesis could be possible, due the concentration of
in agreement with our first hypothesis. Our results demon-
N in a reduced leaf blade surface.
strate that endophyte-free plants reduced their biomass when Water dynamics within roots were also affected by
water availability decreased, while endophyte-associated
endophyte presence, although the symbiont resides in the
plants sustained biomass accumulation in the same scenario.
aboveground plant parts. Hence, our results demonstrated
Under no water deficit, endophyte-free plants produced
that endophyte-associated plants reduced carbon allocation
more biomass than endophyte-associated plants, perhaps
to roots and changed the root-system structure, without
due to the carbon cost of the endophyte in non-stressful
reducing its water-uptake capacity. In experiment 2, no
situations. This putative penalty of endophyte presence in
interaction between endophyte and water treatments was
well-watered conditions disappeared under water deficit,
observed, suggesting that these endophyte effects may be
where endophyte-associated plants produced a similar amount
constitutive and not induced by water stress. Hydraulic
of biomass as endophyte-free plants (Fig. 2). As regards the
conductivity in endophyte-associated plants was maintained
second hypothesis, we found that drought tolerance conferred
despite the reduction in the total length of the root system
by the endophyte appeared to be related to mechanisms
and, particularly, the fine root length. Thus, endophyte-
inducing changes in the structure and functioning of both
associated plants had a more efficient root system both at field
the shoot and root. First, water deficit differently affected
capacity and under water deficit conditions, than endophyte-
biomass allocation pattern: in endophyte-free plants water
free ones, as previously described (Malinowski et al. 2000).
limitation reduced both shoot and root biomass, while in
endophyte-associated plants, the reduction of biomass was Despite this first research studying root hydraulic properties
mainly in root mass. Secondly, under water deficit, the in an endophytic grass, further experiments are necessary
maintenance of shoot biomass production was associated to elucidate the underlying mechanisms related to water
with a strong reduction of stomatal conductance (Baca homeostatic function when endophytic plants grow in
Cabrera et al. 2021) and an increase in intrinsic water use stressful environments.
efficiency (iWUE) of the endophytic plants. Finally, although The above described isohydric strategy was also associated
endophyte presence reduced root biomass, its presence also with a change of C allocation between shoot and roots. When
increased root length, the proportion of fine roots and the analysing each biomass component, the typical water deficit
efficiency of the hydraulic system (i.e. hydraulic conductivity). response (i.e. marked aerial growth arrest; Xu et al. 2016) was
The presence of the endophyte altered hydric relationships only observed in endophyte-free plants, where a reduction in
within the plant trough mechanisms related to an isohydric 50% of water availability was accompanied with a reduction
strategy. This strategy involved a regulation in the above- of 42% in shoot biomass. Conversely, endophyte-associated
ground part of the plant reducing water losses, and was plants changed their allometry since shoot biomass was
related to stronger stomatal closure (Baca Cabrera et al. unaffected when water availability was reduced by 50%.
2021), increasing iWUE, but a lesser relative change of We observed further changes in allocation independent of
net CO2 assimilation and growth rates in comparison with water limitation: thus, endophyte presence increased
endophyte-free plants when exposed to water deficit. allocation to the shoot and reduced the blade:sheath ratio.
A similar response pattern was observed in endophyte-free These responses suggest that endophyte-associated plants
tall fescue, where the lower stomatal conductance induced prioritised shoots over roots. A recent meta-analysis indicates
by water deficit was associated with a decrease of iWUE that whether root or shoot partitioning is promoted could
resulting from the greater decrease of photosynthetic depend on species life cycle length (Dastogeer 2018).
activity, in comparison with endophyte-associated plants For example, the production of root biomass in perennial
(Swarthout et al. 2009; Xia et al. 2018). In our case, the species could be maintained and prioritised in water
lesser change of net CO2 assimilation in response to water scarcity scenarios (Wang et al. 2017). Instead, allocating C
deficit in endophyte-associated relative to endophyte-free to roots as a constitutive mechanism might be a successful
plants – inferred from C and oxygen (O) isotope analysis – strategy in annual species that deal with this challenging
may be related to a greater C-sink strength induced by condition over a short period. Nevertheless, due to changes
endophyte presence (as suggested by Kaschuk et al. 2009 in plant allometry, endophyte-associated plants were able
for other symbiotic microorganisms) possibly related to N to maintain total biomass along the experimental water
availability. However, endophyte-associated and endophyte- availability gradient while endophyte-free plants were not.

H
www.publish.csiro.au/fp Functional Plant Biology

Drought tolerance conferred by the endophyte to the host In conclusion, we found that a vertically transmitted
plant can arise from phenotypic plastic traits resulting from symbiont of C3 grasses has an important role in plant
changes in plant allometry. ‘Tolerance’ can be considered in ecophysiology during water deficits, increasing efficiency in
a broad sense as the change in performance in response to the aboveground and belowground parts of the plant. First,
stress, or, as the slope of the reaction norm of fitness across we found that endophyte presence increased water use
a gradient of growth-limiting conditions (Simms 2000; efficiency, by sustaining assimilation rate when stomatal
Stowe et al. 2000). In our case, the different slopes of conductance was reduced in response to water deficit,
the relationship between biomass and water availability resulting in an inducible mechanism. Secondly, endophyte-
(Fig. 2) demonstrate that endophyte-associated plants were associated plants were able to acquire the same amount of
more tolerant than endophyte-free plants: while biomass water with a reduced root system. Third, endophyte
gain of endophyte-free plants decreased abruptly when presence changed C allocation between shoot and root, thus
water availability was diminishing, endophyte-associated affecting plant allometry. As these responses were observed
plants demonstrated a more conservative response. Although irrespective of water limitation, they can be considered as
it is generally accepted that the endophyte can confer constitutive mechanisms. All these responses suggest that
drought tolerance, this idea has been proven mainly with the endophyte affected host C economy, maintaining
tall fescue (Arechavaleta et al. 1992; Nagabhyru et al. CO2 assimilation and C allocation under water deficits.
2013). Furthermore, when using other plant species the As far as we know, this is the first comprehensive study
pattern was not consistent, showing that the response may including endophyte-conditioned changes in shoot and root
depend on plant and endophyte genotypes, and biotic and functioning in an annual mesophytic grass that is currently
abiotic context (Davitt et al. 2011; Miranda et al. 2011; being subjected to increasing intensity or frequency of
Gundel et al. 2016; Wang et al. 2017; Dastogeer 2018). drought events. Finally, our results support the idea that
Interestingly, our results also suggest that the tolerance presence of moderate water stress can increase the effec-
observed in our experiments was not triggered by a positive tiveness of the mutualism, guaranteeing the stability of
effect of endophyte presence on host growth during the the symbiosis in variable environmental situations. These
results incite us to consider the long-term effect on plant
water stress; but by its cost when no stress was present.
fitness in the presence of the endophyte when the hydraulic
Symbionts may provide host plants with the ability to cope
conditions are limiting for adequate growth and develop-
with resource limitation, abiotic stresses or antagonistic
ment in further studies. Preserving symbioses with this trait
organisms (Douglas 2010). This could imply that under no
in natural systems may represent a strategy to mitigate
resource limitation or stressful conditions, symbiotic organisms
global change impacts.
may impair plant performance due to its maintenance
costs. This interpretation matched our observation that an
endophytic symbiont impaired plant performance in the
absence of abiotic stress, thus supporting the context-
Supplementary material
dependency hypothesis. Under water limitation, endophyte-
Supplementary material is available online.
associated plants gained tolerance as evidenced by a lesser
reduction of biomass production, probably due to the
ecophysiological mechanisms describe above. The lack of References
change in the endophyte-associated plant biomass along the
Arechavaleta M, Bacon CW, Plattner RD, Hoveland CS, Radcliffe DE
stress gradient suggests that the symbiont acted as a (1992) Accumulation of ergopeptide alkaloids in symbiotic tall
‘constitutive defence’. This implies that plants pay the cost of fescue grown under deficits of soil water and nitrogen fertilizer.
harbouring the symbiont against the eventuality of water Applied and Environmental Microbiology 58, 857–861. doi:10.1128/
aem.58.3.857-861.1992
deficit stress. Epichloid fungal endophytes are vertically Baca Cabrera JC, Hirl RT, Schäufele R, Macdonald A, Schnyder H (2021)
transmitted and obligate symbionts, which links the fitness Stomatal conductance limited the CO2 response of grassland in the last
of the symbiont to host performance (Gundel et al. 2008; century. BMC Biology 19, 50. doi:10.1186/s12915-021-00988-4
Bacon CW, White JFJ (1994) ‘Biotechnology of endophytic fungi of
Kiers and West 2015). Thus, the ability to provide benefits grasses.’ (CRC Press: Boca Raton, USA)
could promote the mutualistic outcome and symbiosis Brendel O, Iannetta PPM, Stewart D (2000) A rapid and simple method
to isolate pure alpha-cellulose. Phytochemical Analysis 11, 7–10.
stability in fluctuating environments (Rodriguez et al. 2008; doi:10.1002/(SICI)1099-1565(200001/02)11:1<7::AID-PCA488>3.0.
Redman et al. 2011; Walder and van der Heijden 2015). In CO;2-U
this sense, the new scenarios induced by climate change Cheplick GP, Faeth SH (2009) ‘Ecology and evolution of the grass-
endophyte symbiosis.’ Oxford Scholarship Online Monographs.
could strengthen the mutualistic condition of this interaction. (Oxford University Press: New York)
Increasing drought events in temperate-humid environments Clay K (1988) Fungal endophytes of grasses: a defensive mutualism
could drive plants to express new strategies to cope with between plants and fungi. Ecology 69, 10–16. doi:10.2307/1943155
Condon AG, Richards RA, Rebetzke GJ, Farquhar GD (2004) Breeding
stressful situations, fixing the tolerance trait (Rodriguez et al. for high water-use efficiency. Journal of Experimental Botany 55,
2008; Redman et al. 2011; Kiers and West 2015). 2447–2460. doi:10.1093/jxb/erh277

I
M. E. Manzur et al. Functional Plant Biology

Dastogeer KMG (2018) Influence of fungal endophytes on plant Leuchtmann A, Bacon CW, Schardl CL, White JF, Tadych M (2014)
physiology is more pronounced under stress than well-watered Nomenclatural realignment of Neotyphodium species with genus
conditions: a meta-analysis. Planta 248, 1403–1416. doi:10.1007/ Epichloë. Mycologia 106, 202–215. doi:10.3852/13-251
s00425-018-2982-y Li T, Blande JD, Gundel PE, Helander M, Saikkonen K (2014) Epichloë
Davitt AJ, Chen C, Rudgers JA (2011) Understanding context-dependency endophytes alter inducible indirect defences in host grasses. PLoS
in plant–microbe symbiosis: the influence of abiotic and biotic ONE 9, e101331. doi:10.1371/journal.pone.0101331
contexts on host fitness and the rate of symbiont transmission. Ma WT, Tcherkez G, Wang XM, Schäufele R, Schnyder H, Yang Y, Gong XY
Environmental and Experimental Botany 71, 137–145. doi:10.1016/ (2021) Accounting for mesophyll conductance substantially improves
13
j.envexpbot.2010.11.004 C-based estimates of intrinsic water-use efficiency. New Phytologist
De Battista JP, Bouton JH, Bacon CW, Siegel MR (1990) Rhizome 229, 1326–1338. doi:10.1111/nph.16958
and herbage production of endophyte-removed tall fescue clones Malinowski DP, Belesky DP (2000) Adaptations of endophyte-
and populations. Agronomy Journal 82, 651–654. doi:10.2134/ infected cool-season grasses to environmental stresses: mechanisms
agronj1990.00021962008200040001x of drought and mineral stress tolerance. Crop Science 40, 923–940.
Douglas AE (2010) ‘The symbiotic habitat.’ (Princeton University Press: doi:10.2135/cropsci2000.404923x
New Jersey, USA) Malinowski D, Leuchtmann A, Schmidt D, Nösberger J (1997) Growth and
Easterling DR, Evans JL, Groisman PY, Karl TR, Kunkel KE, water status in meadow fescue is affected by Neotyphodium and
Ambenje P (2000) Observed variability and trends in extreme climate Phialophora species endophytes. Agronomy Journal 89, 673–678.
events: a brief review. Bulletin of the American Meteorological Society 81, doi:10.2134/agronj1997.00021962008900040021x
417–426. doi:10.1175/1520-0477(2000)081<0417:OVATIE>2.3.CO;2 Malinowski DP, Alloush GA, Belesky DP (2000) Leaf endophyte
Elmi AA, West CP (1995) Endophyte infection effects on stomatal Neotyphodium coenophialum modifies mineral uptake in tall fescue.
conductance, osmotic adjustment and drought recovery of tall Plant and Soil 227, 115–126. doi:10.1023/A:1026518828237
fescue. New Phytologist 131, 61–67. doi:10.1111/j.1469-8137.1995. Martínez-Vilalta J, Poyatos R, Aguadé D, Retana J, Mencuccini M (2014)
tb03055.x A new look at water transport regulation in plants. New Phytologist
Farquhar GD, Cernusak LA, Barnes B (2007) Heavy water fractionation 204, 105–115. doi:10.1111/nph.12912
during transpiration. Plant Physiology 143, 11–18. doi:10.1104/ Matsuo N, Ozawa K, Mochizuki T (2009) Genotypic differences in
pp.106.093278 root hydraulic conductance of rice (Oryza sativa L.) in response to
Farquhar GD, O’Leary MH, Berry JA (1982) On the relationship between water regimes. Plant and Soil 316, 25–34. doi:10.1007/s11104-008-
carbon isotope discrimination and the intercellular carbon dioxide 9755-5
concentration in leaves. Australian Journal of Plant Physiology 13, Miranda MI, Omacini M, Chaneton EJ (2011) Environmental context of
281–292. doi:10.1071/PP9820121 endophyte symbioses: interacting effects of water stress and insect
Farquhar GD, Ehleringer JR, Hubick KT (1989) Carbon isotope herbivory. International Journal of Plant Sciences 172, 499–508.
discrimination and photosynthesis. Annual Review of Plant Physiology doi:10.1086/658921
Miyamoto N, Steudle E, Hirasawa T, Lafitte R (2001) Hydraulic
and Plant Molecular Biology 40, 503–537. doi:10.1146/annurev.pp.40.
conductivity of rice roots. Journal of Experimental Botany 52,
060189.002443
1835–1846. doi:10.1093/jexbot/52.362.1835
Fox J, Weisberg S (2019). An R Companion to Applied Regression, Third
Moreno-Gutiérrez C, Dawson TE, Nicolás E, Querejeta JI (2012)
edition. Sage, Thousand Oaks CA. https://socialsciences.mcmaster.
Isotopes reveal contrasting water use strategies among coexisting
ca/jfox/Books/Companion/.
plant species in a Mediterranean ecosystem. New Phytologist 196,
Gaudinski JB, Dawson TE, Quideau S, Schuur EAG, Roden JS, Trumbore
489–496. doi:10.1111/j.1469-8137.2012.04276.x
SE, Sandquist DR, Oh S-W, Wasylishen RE (2005) Comparative
Moshelion M, Halperin O, Wallach R, Oren R, Way DA (2015)
analysis of cellulose preparation techniques for use with 13C, 14C,
Role of aquaporins in determining transpiration and photosynthesis
and 18O isotopic measurements. Analytical Chemistry 77, 7212–7224. in water-stressed plants: crop water-use efficiency, growth and
doi:10.1021/ac050548u yield. Plant, Cell & Environment 38, 1785–1793. doi:10.1111/pce.
Gundel PE, Batista WB, Texeira M, Martinez-Ghersa MA, Omacini M, 12410
Ghersa CM (2008) Neotyphodium endophyte infection frequency in Nagabhyru P, Dinkins RD, Wood CL, Bacon CW, Schardl CL (2013) Tall
annual grass populations: relative importance of mutualism and fescue endophyte effects on tolerance to water-deficit stress. BMC
transmission efficiency. Proceedings of the Royal Society B: Biological Plant Biology 13, 127. doi:10.1186/1471-2229-13-127
Sciences 275, 897–905. doi:10.1098/rspb.2007.1494 Omacini M, Eggers T, Bonkowski M, Gange AC, Jones TH (2006) Leaf
Gundel PE, Irisarri JGN, Fazio L, Casas C, Pérez LI (2016) Inferring endophytes affect mycorrhizal status and growth of co-infected and
field performance from drought experiments can be misleading: the neighbouring plants. Functional Ecology 20, 226–232. doi:10.1111/
case of symbiosis between grasses and Epichloë fungal endophytes. j.1365-2435.2006.01099.x
Journal of Arid Environments 132, 60–62. doi:10.1016/j.jaridenv. Pinheiro JC, Bates D, DebRoy S, Sarkar D, Team RC (2015) {nlme}: linear
2016.04.008 and nonlinear mixed effects models. http://cran.r-project.org/
Jaramillo VJ, Detling JK (1988) Grazing history, defoliation, and package=nlme
competition: effects on shortgrass production and nitrogen R Core Team (2018) ‘R: a language and environment for statistical
accumulation. Ecology 69, 1599–1608. doi:10.2307/1941657 computing.’ (R Foundation for Statistical Computing). Available at
Kaschuk G, Kuyper TW, Leffelaar PA, Hungria M, Giller KE (2009) Are the https://www.r-project.org/
rates of photosynthesis stimulated by the carbon sink strength of Redman RS, Kim YO, Woodward CJDA, Greer C, Espino L, Doty SL,
rhizobial and arbuscular mycorrhizal symbioses? Soil Biology and Rodriguez RJ (2011) Increased fitness of rice plants to abiotic stress
Biochemistry 41, 1233–1244. doi:10.1016/j.soilbio.2009.03.005 via habitat adapted symbiosis: a strategy for mitigating impacts of
Kiers ET, West SA (2015) Evolving new organisms via symbiosis. Science climate change. PLoS ONE 6, e14823. doi:10.1371/journal.pone.
348, 392–394. doi:10.1126/science.aaa9605 0014823
Klein T (2014) The variability of stomatal sensitivity to leaf water Richardson MD, Chapman Jr GW, Hoveland CS, Bacon CW (1992)
potential across tree species indicates a continuum between isohydric Sugar alcohols in endophyte-infected tall fescue under drought.
and anisohydric behaviours. Functional Ecology 28, 1313–1320. Crop Science 32, 1060–1061. doi:10.2135/cropsci1992.0011183X00
doi:10.1111/1365-2435.12289 3200040045x
Knapp AK, Smith MD (2001) Variation among biomes in temporal Rodriguez RJ, Henson J, Van Volkenburgh E, Hoy M, Wright L, Beckwith
dynamics of aboveground primary production. Science 291, 481–484. F, Kim Y-O, Redman RS (2008) Stress tolerance in plants via habitat-
doi:10.1126/science.291.5503.481 adapted symbiosis. The ISME Journal 2, 404–416. doi:10.1038/ismej.
Latch GCM, Hunt WF, Musgrave DR (1985) Endophytic fungi affect 2007.106
growth of perennial ryegrass. New Zealand Journal of Agricultural Scheidegger Y, Saurer M, Bahn M, Siegwolf R (2000) Linking
Research 28, 165–168. doi:10.1080/00288233.1985.10427011 stable oxygen and carbon isotopes with stomatal conductance and

J
www.publish.csiro.au/fp Functional Plant Biology

photosynthetic capacity: a conceptual model. Oecologia 125, 350–357. temperature and water deficit: independent or interactive effects?
doi:10.1007/s004420000466 Plant Cell & Environment 35, 702–718. doi:10.1111/j.1365-3040.
Sheffield J, Wood EF (2007) Characteristics of global and regional 2011.02445.x
drought, 1950–2000: analysis of soil moisture data from off-line Walder F, van der Heijden MGA (2015) Regulation of resource exchange
simulation of the terrestrial hydrologic cycle. Journal of Geophysical in the arbuscular mycorrhizal symbiosis. Nature Plants 1, 15159.
Research: Atmospheres 112, D17115. doi:10.1029/2006JD008288 doi:10.1038/nplants.2015.159
Simms EL (2000) Defining tolerance as a norm of reaction. Evolutionary Wang J, Zhou Y, Lin W, Li M, Wang M, Wang Z, Kuang Y, Tian P (2017)
Ecology 14, 563–570. doi:10.1023/A:1010956716539 Effect of an Epichloë endophyte on adaptability to water stress in
Stowe KA, Marquis RJ, Hochwender CG, Simms EL (2000) The Festuca sinensis. Fungal Ecology 30, 39–47. doi:10.1016/j.funeco.
evolutionary ecology of tolerance to consumer damage. Annual Review 2017.08.005
of Ecology and Systematics 31, 565–595. doi:10.1146/annurev.ecolsys. White RH, Engelke MC, Morton SJ, Johnson-Cicalese JM, Ruemmele BA
31.1.565 (1992) Acremonium endophyte effects on tall fescue drought
Sultan SE (2000) Phenotypic plasticity for plant development, function tolerance. Crop Science 32, 1392–1396. doi:10.2135/cropsci1992.
and life history. Trends in Plant Science 5, 537–542. doi:10.1016/ 0011183X003200060017x
S1360-1385(00)01797-0 Xia C, Christensen MJ, Zhang X, Nan Z (2018) Effect of Epichloë gansuensis
Sutka MR, Manzur ME, Vitali VA, Micheletto S, Amodeo G (2016) endophyte and transgenerational effects on the water use efficiency,
Evidence for the involvement of hydraulic root or shoot adjustments nutrient and biomass accumulation of Achnatherum inebrians under
as mechanisms underlying water deficit tolerance in two Sorghum soil water deficit. Plant and Soil 424, 555–571. doi:10.1007/
bicolor genotypes. Journal of Plant Physiology 192, 13–20. s11104-018-3561-5
doi:10.1016/j.jplph.2016.01.002 Xu Z, Jiang Y, Jia B, Zhou G (2016) Elevated-CO2 response of stomata and
Swarthout D, Harper E, Judd S, Gonthier D, Shyne R, Stowe T, Bultman T its dependence on environmental factors. Frontiers in Plant Science 7,
(2009) Measures of leaf-level water-use efficiency in drought stressed 657. doi:10.3389/fpls.2016.00657
endophyte infected and non-infected tall fescue grasses. Environmental Yang Z, Liu J, Tischer S V., Christmann A, Windisch W, Schnyder H, Grill E
and Experimental Botany 66, 88–93. doi:10.1016/j.envexpbot.2008. (2016) Leveraging abscisic acid receptors for efficient water use in
12.002 Arabidopsis. Proceedings of the National Academy of Sciences of the
Vile D, Pervent M, Belluau M, Vasseur F, Bresson J, Muller B, Granier C, United States of America 113, 6791–6796. doi:10.1073/pnas.
Simonneau T (2012) Arabidopsis growth under prolonged high 1601954113

Data availability. Data are not shared due to privacy issues.

Conflicts of interest. The authors declare no conflicts of interest.
Declaration of funding. This work was supported by National Agency for the Promotion of Science and Technology (Argentina, ANPCyT BID PICT 3158
2015). FAG was supported by a postdoctoral fellowship from National Council for Scientific and Technical Research (CONICET Argentina). PAGP was supported
by a short-term grant from the German Academic Exchange Service (DAAD, number 57314022).
Acknowledgements. We thank Dr Rudi Schäufele for assistance on isotope analyses and interpretation, and the editor and two anonymous reviewers whose
constructive comments on earlier versions improved the manuscript.

Author affiliations
A
IIBIO-CONICET-UNSAM, Avenida 25 de Mayo y Francia, San Martín, CPA B1650HMP Buenos Aires, Argentina.
B
Departamento de Biología Aplicada y Alimentos, Cátedra de Fisiología Vegetal, Facultad de Agronomía, Universidad de Buenos Aires, Avenida San Martín 4453,
C1417DSE Buenos Aires, Argentina.
C
IFEVA-CONICET, Facultad de Agronomía, Universidad de Buenos Aires, Avenida San Martín 4453, C1417DSE Buenos Aires, Argentina.
D
Departamento de Recursos Naturales y Ambiente, Cátedra de Ecología, Facultad de Agronomía, Universidad de Buenos Aires, Avenida San Martín 4453,
C1417DSE Buenos Aires, Argentina.
E
Lehrstuhl für Grünlandlehre, Technische Universität München, D-85354 Freising-Weihenstephan, Germany.
F
DBBE-IBBEA, CONICET, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Intendente Güiraldes 2160, Ciudad Universitaria, C1428EGA,
Buenos Aires, Argentina.
G
Departamento de Producción Animal, Cátedra de Forrajicultura, Facultad de Agronomía, Universidad de Buenos Aires, Avenida San Martín 4453, C1417DSE
Buenos Aires, Argentina.