PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Send a file File manager PDF Toolbox Search Help Contact



2018 Colinet & Renault Exp Gerontol .pdf



Original filename: 2018 Colinet & Renault Exp Gerontol.pdf
Title: Similar post-stress metabolic trajectories in young and old flies
Author: Hervé Colinet

This PDF 1.7 document has been generated by Elsevier / , and has been sent on pdf-archive.com on 09/01/2018 at 16:29, from IP address 129.20.x.x. The current document download page has been viewed 237 times.
File size: 790 KB (8 pages).
Privacy: public file




Download original PDF file









Document preview


Experimental Gerontology 102 (2018) 43–50

Contents lists available at ScienceDirect

Experimental Gerontology
journal homepage: www.elsevier.com/locate/expgero

Similar post-stress metabolic trajectories in young and old flies


T

Hervé Colinet , David Renault
UMR CNRS 6553 EcoBio, Université de Rennes 1, 263 Avenue du General Leclerc, CS 74205, 35042 Rennes Cedex, France

A R T I C L E I N F O

A B S T R A C T

Section Editor: T.E. Johnson

Homeostenosis (i.e. decline in stress resistance and resilience with age) is a fundamental notion of the biogerontology and physiology of aging. Stressful situations typically challenge metabolic homeostasis and the capacity
to recover from a stress-induced metabolic disorder might be particularly compromised in senescent individuals.
In the present work, we report the effects of aging on low temperature stress tolerance and metabolic profiles in
Drosophila melanogaster females of different ages. Adult flies aged 4, 16, 30 and 44 days were subjected to acute
and chronic cold stress, and data confirmed a strong decline in cold tolerance and resilience of old flies compared
to young counterparts. Using quantitative target GC–MS analysis, we found distinct metabolic phenotypes between young (4 day-old) and old (44 day-old) flies, with glycolytic pathways being differentially affected between the two age groups. We also compared the robustness of metabolic homeostasis in young vs. old flies when
exposed to cold stress using time-series metabolic analysis. In both age groups, we found evidence of strong
alteration of metabolic profiles when flies were exposed to low temperature stress. Interestingly, the temporal
metabolic trajectories during the recovery period were similar in young and old flies, despite strong differences
in thermotolerance. In conclusion, metabolic signatures markedly changed with age and homeostenosis was
observed in the phenotypic response to cold stress. However, these changes did not reflect in different temporal
homeostatic response at metabolic level.

Keywords:
Functional senescence
Cold stress
Metabolic trajectories
Fruit fly

1. Introduction
Functional senescence describes the failure of biological systems
and functions, progressively arising near the end of life (Grotewiel
et al., 2005). It involves a large range of cellular and molecular events,
and thus, it occurs at all levels of biological organisation (Chen et al.,
2007; De Loof, 2011). In the chain of biomolecules from the genes to
the phenotypes, metabolites are the quantifiable molecules with the
closest link to phenotypes. Maintaining life is viewed as the ability to
maintain the level of metabolites from intermediate metabolism within
discrete and functional ranges (Mishur and Rea, 2012). For reasons not
yet fully understood, this range of allowed equilibria narrows with
aging, a phenomenon referred to as homeostenosis (Troncale, 1996).
Metabolic signatures of long-lived mutants Caenorhabditis elegans
are distinct from those of wild counterparts (Butler et al., 2010; Fuchs
et al., 2010). In Drosophila melanogaster, metabolic profiles of flies selected for longevity differ from those of unselected individuals (Sarup
et al., 2012). Young and old flies also exhibit divergent metabolic
profiles and pathways (Copes et al., 2015; Sarup et al., 2012; Hoffman

et al., 2014; Avanesov et al., 2014). These data support the notion that
metabolic pathways available to an organism change with longevity
and age. Analysing metabolome and metabolomic responses can thus be
useful for understanding aging-related disjunctions (Mishur and Rea,
2012; Soltow et al., 2010).
Homeostenosis (i.e. the decline in stress resistance and resilience
with age) is a fundamental notion of the biogerontology and physiology
of aging (Troncale, 1996). In brief, as organisms age, increasing physiologic reserves are used to maintain homeostasis, while, at the same
time, physiological limits decline. Consequently, the net physiological
capacity (or reserves) declines with age (Troncale, 1996; Taffett, 2003).
Thus, much less perturbation - or stress – is required to push an older
individual over the edge of its physiological limits. In early life, individuals should better cope with stress exposure, and should recover
“more easily.” In contrast, later in life, the same stress could be lethal,
or considerably more effort might be required to return the perturbed
system to homeostasis (Fig. 1). An illustration of this process can be
delineated from the work of Coquin et al. (2008), who used NMR to
study the age-related decline of hypoxia tolerance in D. melanogaster.

Abbreviations: GC–MS, Gas chromatography–mass spectrometry; ANOVA, Analysis of variance; PCA, Principal component analysis; PC1 & PC2, First & second principal components;
ASCA, ANOVA-simultaneous component analysis; Ala, Alanine; Val, Valine; Ser, Serine; Leu, Leucine; Thr, Threonine; Pro, Proline; Phe, Phenylalanine; Orn, Ornithine; Gly, Glycine; Ile,
Isoleucine; Glu, Glutamate; Fru, Fructose; Glc, Glucose; Tre, Trehalose; Rib, Ribose; Suc, Sucrose; Mal, Maltose; G6P, Glucose-6-phosphate; F6P, Fructose-6-phosphate; G3P, Glycerol-3phosphate; ETA, Ethanolamine; PO4, Free phosphate; GDL, Glucono delta-lactone

Corresponding author at: UMR CNRS 6553 Bât 14A, Université de Rennes1, 263 Avenue du Général Leclerc, CS 74205, 35042 Rennes, France.
E-mail address: herve.colinet@univ-rennes1.fr (H. Colinet).
https://doi.org/10.1016/j.exger.2017.08.021
Received 5 July 2017; Accepted 15 August 2017
Available online 16 August 2017
0531-5565/ © 2017 Elsevier Inc. All rights reserved.

Experimental Gerontology 102 (2018) 43–50

H. Colinet, D. Renault

trajectories would be observed between young and old flies during the
period of recovery, indicating that young flies have a more efficient/fast
homeostatic response. If young flies have a more efficient response than
old flies, distinct temporal metabolic trajectories should be observed
(i.e. significant interaction effects).

Failure

Stress

Capacity

Stress
Efficient
recovery

Challenging
recovery

2. Experimental procedures
2.1. Flies stock and rearing
A laboratory population of D. melanogaster was used for the experiments. The population was founded from a large number of individuals collected in October 2010 in Brittany, France. The flies were
maintained in the laboratory in 100-mL bottles at 25 ± 1 °C (L/D: 12/
12 h) on standard fly medium consisting of brewer yeast (80 g/L), sucrose (50 g/L), agar (15 g/L) and Nipagin® (8 mL/L). To generate flies
for the experiments, groups of 15 mated females were allowed to lay
eggs in 100 mL rearing bottles during a restricted period of, at most, 6 h
under laboratory conditions. This controlled procedure allowed larvae
to develop under uncrowded conditions. To avoid gender-induced
variability, we only tested females in all experiments. At emergence,
adult virgin flies were sexed by visual inspection (with an aspirator).
We did not use CO2 anaesthesia because it could cause confusing metabolic effects (Colinet and Renault, 2012). Virgin females were maintained at 25 ± 1 °C (L/D: 12/12 h) until they were 4- 16-, 30- and 44days-old. Food was changed every two days. Cold tolerance was assessed in these four experimental age groups. A second cohort was then
created, following the same procedure. In this second cohort, two age
groups were compared: 4-day-old flies (subsequently referred to as
young; YN) vs. 44-day-old flies (subsequently referred to as old; OL).
This latter cohort was used to confirm that old flies exhibit less effective
cold tolerance compared to young flies. Individuals from this cohort
were also used for the comparative time-series metabolomics.

Used reserves

Age
Fig. 1. Schematic of the homeostenosis concept. The figure shows a progressive constriction of homeostatic capacity with aging, due to the increase in reserves used to
maintain homeostasis in parallel to a decrease in physiological limits. Thus, in early life,
the metabolic system should deviate when exposed to stress (symbolised by green circles)
and return to homeostasis easily. In contrast, the metabolic system of old individuals later
in life submitted to the same stress (symbolised by red circles) should require considerably more effort to return to homeostasis. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)

The authors found that older flies took much longer to recover from
hypoxic stress than their younger counterparts. The heart rates of older
flies took longer to recover, as did the levels of cellular ATP. Fluxbalance analysis from the metabolic profiles also detected major differences in metabolic functions between young and old flies upon reoxygenation (Coquin et al., 2008), indicating different and/or altered
metabolic trajectories. Additional data are required to determine
whether decline in stress resistance and resilience with age applies
universally to all types of environmental stress.
Noxious environmental conditions typically perturb metabolic
homeostasis (Parsons, 1991), causing cells to respond by activating a
cascade of actions that progressively restore homeostasis (i.e. stress
response and degradation of harmful compounds) (Kültz, 2005). Because of homeostenosis, the capacity to cope with and recover from a
stress-induced metabolic disorder might be particularly compromised
in senescent individuals. To confirm this hypothesis, time-series metabolomics can be particularly informative, as it allows the homeostatic
status of an organism to be monitored via the inspection of metabolic
trajectories during stress or recovery (Nicholson et al., 2002;
Malmendal et al., 2006; Colinet et al., 2012, 2016; Williams et al.,
2014).
In the present study, we examined the effects of aging on the stress
tolerance and resilience of D. melanogaster using low temperature as the
stressor. Cold tolerance varies considerably with age in D. melanogaster
(Minois and Le Bourg, 1999; Le Bourg, 2007; Colinet et al., 2013a,
2015). There is evidence of strong deviation in the metabolic profiles of
Drosophila and other insects when exposed to low temperatures (Teets
et al., 2012; Colinet et al., 2012, 2016; Williams et al., 2014; Koštál
et al., 2016a). Interestingly, cold-hardy flies (either cold-adapted or
cold-acclimated) have the capacity to counteract this cold-induced
homeostatic deviation (Teets et al., 2012; Colinet et al., 2012, 2016;
Williams et al., 2014), indicating that the robustness of metabolic
networks influences cold tolerance. Thus, it can be hypothesised that
metabolic robustness under stress situation, such as cold stress, can be
challenged by aging, as suggested by previous data (Coquin et al.,
2008). In the present study, we used targeted metabolomics to track the
dynamics of the homeostatic response during cold stress and recovery
in young versus old flies. Our expectations were that: (i) young flies
would cope better with cold stress than old flies, (ii) young and old flies
would display different metabolic signatures, (iii) post-stress metabolic
robustness would be higher for young flies compared to old flies (i.e.
less perturbation in young flies), and (iv) different metabolic

2.2. Cold tolerance assessment
To assess cold tolerance, we used both acute and chronic cold stress.
For each age group (i.e. 4- 16-, 30- and 44-days-old), five replicates of
25 females were placed in 42 mL glass vials immersed for 90 min in a
circulating bath (Haake F3 Electron, Karlsruhe, Germany) filled with
ethylene glycol and set at − 3 °C (i.e. acute stress). Then, the flies were
returned to optimal conditions 25 ± 1 °C (L/D: 12/12 h), and survival
was scored after 24 h. Previous studies showed that most D. melanogaster mortality occurs within 24 h of exposure to cold stress (Rako and
Hoffmann, 2006).
To assess tolerance to chronic cold stress, seven vials of 25 flies from
each age group (i.e. 4- 16-, 30- and 44-days-old) were exposed to 0 °C
for 10 h by placing vials in a cold incubator (Model MIR-154, SANYO
Electric Co. Ltd., Munich, Germany). After this prolonged period of cold
stress, 50 females from each experimental age group were randomly
selected and returned to 25 ± 1 °C. The chill coma recovery (CCR) (i.e.
time needed for each fly to stand on legs) was recorded for each fly
individually, as described in (Colinet et al., 2013a). In addition, mortality following chronic cold stress was scored (using five groups of 25
flies) 24 h after exposure to stress.
2.3. Metabolomics
Young (4-day-old) and old (44-days-old) females were used to
compare the dynamics of metabolic trajectories during the stress and
recovery periods (five different time points). For both age groups, females were cold-stressed at 0 °C for 12 h (in a cold incubator), and were
then returned to the optimal temperature (25 °C) for recovery in agar
vials. Unstressed control flies were sampled just before the stress
(codes: YN and CO, for young and old females, respectively). Then, flies
were stressed and sampled at 0, 2, 4 and 8 h of recovery at 25 °C (codes
44

Experimental Gerontology 102 (2018) 43–50

04 16 30 44 -

80
60

a

a
b
c
c

40
20

100

a

b

a
b

80

Survival (%)

100

Survival (%)

Cumulated percent of
recovering flies

H. Colinet, D. Renault

b

60

100

a

80

c
a

60

b

40

b

20

40
0

0

4

0

20

40

60

80

30

4

44

16

30

44

Age (days)

Age (days)

Time (hours)
100

d

4 days
44 days

80

100

Survival (%)

Cumulated percent of
recovering flies

16

60
40
20

e

a

80

b
60
40

0
0

20

40

60

80

4

44

Age (days)

Time (hours)

Fig. 2. Cold tolerance assessment. (a) Comparison of chill coma recovery dynamics of females aged 4, 16, 30 and 44 days. Time to recover from chill coma was monitored individually in
flies (n = 50) that were returned at 25 °C conditions after 10 h at 0 °C. Survival rates ( ± S.D.) of females aged 4, 16, 30 and 44 days and recovering for 24 h at 25 °C after (b) chronic cold
stress (0 °C for 10 h) (n = 125) and (c) acute cold stress (− 3 °C for 1.5 h) (n = 125). Different letters indicate a significant difference (p < 0.05) among age groups. (d) Chill coma
recovery dynamics from a second cohort of flies aged 4 days (young) vs. 44 days (old), and (e) survival rate at 24 h after exposure to a chronic cold stress (0 °C for 10 h).

2.4. Statistical analyses

0-YN, 2-YN, 4-YN, 8-YN and 0-OL, 2-OL, 4-OL, 8-OL, for young and old
flies, respectively). The flies from treatments 0-YN and 0-OL correspond
to samples assessed immediately after the stress period (just before
recovery). For both age groups and for each time point, eight replicates
(each consisting of 15 pooled females) were immediately snap-frozen in
liquid nitrogen, and were then stored at −80 °C until metabolites were
extracted.
The fresh mass of each sample was measured before metabolite
extraction with a microbalance (Mettler Toledo UMX2, accurate to
1 μg). We used the sample preparation and derivatization process as
described in Colinet et al., 2013b, with minor modifications. In brief,
samples were homogenised in ice-cold methanol-chloroform (2:1, v:v,
600 μL). Then, a 400 μL volume of ultrapure water was added to each
sample. After centrifugation at 4000g for 10 min, a 100 μL aliquot of the
upper phase was transferred to a clean glass vial. Samples were vacuum-dried, and the dry residue was re-suspended in 30 μL of
20 mg mL− 1 methoxyamine hydrochloride in pyridine. After incubation at 40 °C for 60 min under orbital shaking, a 30 μL volume of
MSTFA was added to each sample. Derivatization was conducted at
40 °C for 60 min under orbital shaking. The timing of sample preparation was standardized by using a CTC CombiPal autosampler (PAL
System, CTC Analytics AG, Zwingen, Switzerland). A GC/MS platform,
consisting of a Trace GC Ultra chromatograph and a Trace DSQII
quadrupole mass spectrometer (Thermo Fischer Scientific Inc., Waltham, MA, USA), was used to analyse the samples. We used the analytical method previously described by (Colinet et al., 2013b). We
completely randomised the injection order of the samples, which were
run under the SIM mode rather than the full-scan mode; thus, we only
screened for the 63 pure reference compounds included in our custom
spectral database. Calibration curves for 63 pure reference compounds
at concentrations of 1, 2, 5, 10, 20, 50, 100, 200, 500, 750, 1000, and
1500 μM were established. Chromatograms were deconvoluted using
XCalibur 2.0.7, and metabolite concentrations were quantified based on
the calibration curves of each reference compound. Arabinose was used
as the internal standard, and quality controls at concentrations of
200 μM were run every 15 samples.

Chill coma recovery (CCR) data were used to generate temporal
CCR curves, which were compared among ages with Log-rank (MantelCox) tests. Post-stress survival was analysed as function of age by specifying a generalized linear model (GLM) with logistic link function for
binary outcome (i.e. dead/alive). Multiple comparisons were then
performed with Tukey tests. For each metabolite, the temporal changes
in the concentrations for the two age groups were analysed by analysis
of variance (ANOVA2), using age, time and interaction as explanatory
variables. Effect plots were also generated for each metabolite. These
plots show the conditional coefficients (“marginal effects”) for all
variables and interaction terms. Metabotypes (i.e. profiles resulting
from all quantified metabolites) were first analysed using a betweenclass principal component analysis (PCA) (Dolédec and Chessel, 1991)
to test the clustering effect according to the experimental modalities
(i.e. age and time). A Monte-Carlo test (number of iterations = 1000)
was further used to determine whether the samples were randomly
distributed in variable space according to their experimental modality.
All analyses were performed using the statistical software R 3.0.3 (R
Development Core Team, Vienna, Austria). Dimensionality reduction
techniques, such as PCA, are suitable for explorative analyses. However, these approaches do not consider the underlying experimental
design. Therefore, the different sources of variation, such as time and
age in the present case, might be confounded in the model. A typical
analysis that considers the design is the univariate ANOVA, which focuses on how the different sources of variation are separated. By
adapting simultaneous component analysis (SCA), Smilde and coworkers (Smilde et al., 2005) developed a multivariate method that
incorporates information on the experimental design. The so-called
ASCA (ANOVA–SCA) applies PCA to the estimated parameters for each
source of variation of an ANOVA model. This methodology is ideal for
the analysis of time-series metabolomic studies (Nueda et al., 2007;
Smilde et al., 2005). Both ANOVAs and ASCA were performed using the
temporal analysis module of MetaboAnalyst 3.0 (Xia et al., 2011). This
pipeline tests the overall and the interaction effects via ASCA
45

Experimental Gerontology 102 (2018) 43–50

Concentration (nmol/mg)

H. Colinet, D. Renault

Leu

Val

0.9

Gly

30

0.6
0.7

3.0
0.4

0.5

20
0.4

0.5

2.5

0.3
2.0

0.3

10
0.2

0.3

1.5

0.2
-1

0

2

4

8

-1

0

2

Ser
Concentration (nmol/mg)

Pro

Ile
0.5

4

-1

8

0

2

Thr

4

-1

8

0

2

Ala

4

8

-1

0

2

Glu

3.0

4

8

4

8

4

8

4

8

Phe

10
1.00

12

0.4

2.5
8
0.75

9

2.0

0.3
6

0.50

1.5

6
0.25

1.0
-1

0

2

4

8

0.2

4
-1

0

2

Orn

4

8

-1

0

2

Rib

4

8

-1

0

2

Fru

4

8

-1

0

2

Glc

Suc

Concentration (nmol/mg)

4
0.040

0.30

20

6

3
0.035
0.25

15

4

2

0.030
10

0.20

2

0.025

1
5

0.020

0.15

0

0
-1

0

2

4

8

-1

0

Tre

2

4

8

-1

0

2

Mal

4

8

-1

0

2

GDL

4

8

-1

0

2

F6P

G6P

Concentration (nmol/mg)

1.1
30

3

2.0

4

1.5

2

3

1.0
15

0.5

1

2

0.5

10

1

0.3
0

2

4

8

-1

Succinate
Concentration (nmol/mg)

0.7

20

-1

0

2

4

8

-1

0

2

Fumarate

4

8

-1

0

Malate

2

4

8

-1

0

2

Citrate

Glycerol

1.8
3.5

1.0

0.175

4

1.5
3.0

0.150
3

0.8
1.2

2.5
0.6

0.125
2

2.0

0.9

0.100
1.5

0.4
-1

0

2

4

8

-1

G3R
Concentration (nmol/mg)

5
0.9

25

0

2

4

8

1
-1

0

Sorbitol

2

4

8

-1

0

Inositol

2

4

8

0.75

14

2

4

8

4

8

4

8

0.07

0.30

0.100

0.25

0.075

0.20

0.050

0.06
0.05

0.50

10

0

Galacticol

0.35
18

-1

Erythritol

0.25

0.15

0.04

0.025

0.03

0.10

Concentration (nmol/mg)

-1

0

2

4

8

-1

0

Mannitol

0.8

2

4

8

-1

Xylitol

0

2

4

8

-1

0

Citrulline

2

4

8

0.08

9

0.4

0.06

7

0.2

0.04

5

0.0

0.02

0

2

ETA
0.4

6
0.6

-1

PO4

5
0.3
4

-1

0

2

Time (h)

4

8

0.2
3

3
-1

0

2

Time (h)

4

8

-1

0

2

Time (h)

4

8

-1

0

2

Time (h)

4

8

-1

0

2

Time (h)

Fig. 3. Variation in individual metabolites in young (green) and old (red) flies according to time. Both age groups were sampled at various time points: before cold stress (−1),
immediately after cold stress (0), and after 2, 4 and 8 h of recovery post cold stress (2, 4, 8). Refer to Table S1 for metabolites abbreviation's (indicated on top of each plot). Means are
indicated within each boxplot by a horizontal black line and are based on eight replicates. (For interpretation of the references to colour in this figure legend, the reader is referred to the
web version of this article.)

46

Experimental Gerontology 102 (2018) 43–50

H. Colinet, D. Renault

explaining the variability in the data and the clustering of the different
classes. The correlations of each metabolite to PC1 and PC2 are shown
in Fig. 5a and b, respectively. Among the metabolites contributing the
most to PC1, we found various compounds negatively correlated, such
as sorbitol and many amino acids (Ile, Leu, Val, Ser, Thr, Phe, Glu, Pro).
The concentrations of these compounds increased during stress and/or
recovery period (i.e. along PC1) (Figs. 3 and 5a). The metabolites that
contributed the most to PC2 were G6P, F6P, Tre and Glc, which were
positively correlated (i.e. more abundant in young flies), and inositol,
Fru, GDL, mannitol, Ala and GDL, which were negatively correlated
(i.e. more abundant in old flies) (Figs. 3 and 5b).
In addition to the classical PCA, an ASCA was performed to further
analyse our multivariate data and test the main effects of the model (i.e.
age and time) as well as the interaction. If young flies have a distinct
and more efficient homeostatic response than old flies, then different
temporal metabolic trajectories should be observed during recovery
between both groups, and this should result in a significant interaction
effect. The PC1 scores of the factor “age” of the ASCA are presented in
supplementary Fig. S2a. An increasing trend was detected for YN flies
compared to OL flies, supporting the PCA results shown in Fig. 4c. The
score plot based on the PC1 for the factor “time” is supplementary Fig.
S2c. A marked decrease from initial state (time −1) was detected towards the state corresponding to end of the stress period (time 0), with
a further decrease at 2 h post recovery (time 2), when the score became
negative, after which the scores remained negative at 4 and 8 h of recovery (times 4 and 8). This result was consistent with PCA results
shown in Fig. 4b, in which the metabotypes from both ages shifted in
the same direction over the time-course experiment. The observed
statistics with permutation tests were significant for both the overall
effect of age and time (observed p < 0.05), suggesting clear differentiation between young and old flies, as well as among the different
time points (supplementary Fig. S2b,d). The score plot based on the PC1
of the interaction factor “time × age” is shown in supplementary Fig.
S2e. The scores of this submodel were interpreted as the deviation of
each age group from the “time” factor. The permutation test showed
that the interaction model was not significant (observed p = 0.25),
indicating no significant differentiation in temporal patterns between
young and old flies (supplementary Fig. S2f).

permutation tests. If the observed effect differs to the no-effect distribution, then the effect is considered significant (Vis et al., 2007). All
data were scaled and mean-centred before PCA and ASCA.
3. Results
3.1. Cold tolerance
CCR curves differed with the age of flies (χ2 = 145.7; df = 3;
p < 0.001; Fig. 2a). Specifically, older flies (30- and 44-days-old) took
longer to recover from chronic cold stress than younger flies (16- and 4days-old). Survival following chronic cold stress (Fig. 2b) was affected
by age (χ2 = 29.08; df = 3; p < 0.001). Specifically, the survival of
older flies (30- and 44-days-old) was lower compared to that of younger
flies (16- and 4-days-old). The same pattern was observed for survival
following acute cold stress (Fig. 2c) (χ2 = 107.35; df = 3; p < 0.001),
with old flies being more susceptible to cold than their younger counterparts. These results were confirmed in the replicate experiment,
which confirmed that younger flies (4-days-old) recovered faster
(χ2 = 85.25; df = 1; p < 0.001; Fig. 2d) and survived from cold stress
better compared to older flies (44-days-old) (χ2 = 11.86; df = 1;
p = 0.006; Fig. 2e).
3.2. Metabolic profiles and trajectories
Among the 63 metabolites included in our spectral library, 35 were
detected in our fly samples, and were quantified (absolute quantification) based on the calibration curves of pure standards. A list of detected metabolites with their abbreviations is provided in
Supplementary Table S1. Among these 35 compounds, we found eleven
amino acids, six sugars, seven polyols, three organophosphates, four
intermediate metabolites, one polyamine, and three diverse metabolites. Glc, Pro and Tre were the most abundant metabolites detected.
Changes in the level of each individual metabolite in relation to age
and time are illustrated in panel Fig. 3. The effects identified by twoways ANOVAs are summarized in Supplementary Fig. S1 (a Venn diagram summarizing all significant effects). Tabular results of the twoways ANOVAs are also provided in Supplementary Table S2. The
abundance of 22, 24 and 11 metabolites was significantly affected by
age, time and their interaction, respectively (see Supplementary Fig.
S1). The abundance of just eight metabolic compounds was simultaneously affected by age, time and their interaction (Supplementary Fig.
S1 and Table S2).
Global metabolic changes in all experimental treatments were first
characterised with PCA. This PCA reduced the dimensionality of the
data into a few components that were much easier to analyse. The ordination of classes within the first plane is presented in Fig. 4a. PC1
(36.97%) and PC2 (27.02%), represented 64% of total inertia. The
Monte-Carlo randomisations confirmed the significance of the differences among classes (observed p < 0.001). The PC1 and PC2 scores
(i.e. projection of centroids) of each treatment group are shown in
Fig. 4b and c, respectively. The PCA showed that PC1 mainly represented a temporal pattern, with unstressed treatments opposed to
recovering treatments, and the stressed metabotypes being intermediate. Of note, temporal patterns were similar in both age groups.
The PC1 scores were high and positive before cold stress (OL & YN),
then they markedly diminished during stress (0-OL & 0-YN), reached a
minimum at 2 h post recovery (2-OL & 2-YN), and then slightly increased at 4 and 8 h post recovery (4-OL, 8-OL, 4-YN, 8-YN) (Fig. 4a,b).
Of note, the trajectories did not return to the initial state, suggesting
persistent cold-induced deviation of metabolic homeostasis in both age
groups (Fig. 4b). PC2 described a consistent opposition between young
and old metabotypes, and thus it mainly represented an age pattern,
with higher scores being observed in young (YN) flies, whatever the
sampling time (Fig. 4a,c). We looked at variables that were the most
correlated with PC1 and PC2 as these are the most important in

4. Discussion
Resistance to environmental stressors declines with age, partly because macromolecules are subjected to more intense stochastic damage
(Partridge and Gems, 2006; Grotewiel et al., 2005). Several studies
support an age-dependent decline in thermal tolerance in insects. For
instance, senescent adult fruit flies are considerably less cold tolerant
than young adults (Minois and Le Bourg, 1999; Burger and Promislow,
2006; Le Bourg, 2007), with the same pattern being described in other
insect species (Bowler and Terblanche, 2008). Newly eclosed adults of
D. melanogaster are able to withstand lethal high or low temperatures,
with this ability noticeably declining in first days of adult life (Sørensen
et al., 2002; Colinet et al., 2013a). This phenomenon might benefit
natural populations, with high resistance to environmental stressors by
young adults allowing them to reproduce at least once (Bowler and
Terblanche, 2008). The present study showed that old flies were substantially less cold tolerant than their younger counterparts. Flies became particularly susceptible to cold stress when they were 30-days-old
or more, an age that largely exceeds the expected life span of flies in
natural populations (1.3 to 6.2 days) (Rosewell and Shorrocks, 1987).
Thus, the higher cold-susceptibility of old flies is consistent with our
initial expectations and with the published literature.
We observed distinct metabotypes between young and old flies. This
clear differentiation supports the notion that metabolic signatures
change with age (Mishur and Rea, 2012; Soltow et al., 2010; Sarup
et al., 2012; Hoffman et al., 2014; Avanesov et al., 2014; Copes et al.,
2015). In the present study, old flies were characterised by a decrease in
47

Experimental Gerontology 102 (2018) 43–50

H. Colinet, D. Renault

0-YN

2

YN

Old
Young

4
2
0

-2

4-YN

-4
BEFORE

0H

2H

4H

8H

0

Time
0-OL

8-YN

OL

c

6

PC2 scores

-2

2-OL

Old
Young

4
2
0

-2
-4

4-OL

-4

PC2 - 27.02 % of total inertia

2-YN

b

6

PC1 Scores

4

a

-6

8-OL

BEFORE

0H

2H

4H

8H

Time

-6

-4

-2

0

2

4

6

PC1 - 36.97 % of total inertia
Fig. 4. Metabolic profiling based on quantitative time-series GC/MS analyses from young (4-day-old; green) vs. old (44-day-old; red) adults. (a) The plot of PC1 and PC2 (cumulating
63.99% of inertia), which was derived from principal component analysis (PCA), shows the clustering of treatments according to the global composition of metabolites. Both age groups
were sampled at various time points: before cold stress (0 °C for 12 h) (YN for young and OL for old), immediately after cold stress (0-YN and 0-OL), and then, during the subsequent
recovery period (2, 4 and 8 h post cold stress: 2-YN, 4-YN, 8-YN and 2-OL, 4-OL, 8-OL, respectively). Eight replicates of each treatment were used and are linked to a common centroid by
line segments; the ovals represent 95% confidence intervals around the centroid. The right panel shows the mean scores ( ± SE) according to sampling times, representing the projections
of the centroids on PC1 (b) and PC2 (c). PC1 explains the opposing temporal patterns in unstressed and recovering flies, while PC2 explains the opposing trends between young and old
flies. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Correlation values of the different concentrations of
metabolites in relation to the principal components PC1 (a)
and PC2 (b) in the principal component analysis.
Correlations are ranked on the Y-axis according to their
values. Refer to Table S1 for the abbreviations of metabolites.

48

Experimental Gerontology 102 (2018) 43–50

H. Colinet, D. Renault

have protective functions for cold/freeze tolerance (Kostal et al., 2011a;
Kostál et al., 2012; Koštál et al., 2016b); however, their accumulation
may likely depict a cold-induced degenerative syndrome resulting from
proteolysis or disruption of metabolic pathways (Lalouette et al., 2007;
Colinet et al., 2007; Kostál et al., 2011b; Koštál et al., 2016a).
Aging is viewed as a narrowing of the homeostenosis over a lifetime. Consequently, the ability to maintain the level of small metabolites (derived from intermediate metabolism) within their functional
ranges is expected to decrease with age (Mishur and Rea, 2012). Because of homeostenosis, we speculated that the homeostasis restoration
of the cold-perturbed metabolic system would be more challenging in
old flies than in young flies. Hence, we expected distinct metabolic
trajectories along time between old and young flies (i.e. significant age
x time interaction). We found that old flies required much longer to
recover from chill coma compared to young flies, confirming that
homeostenosis exists on this phenotypic trait. Yet, the ASCA model did
not detect an overall effect of the age x time interaction on metabolic
profiles. Some individual metabolites displayed interaction effects (Fig.
S1), reflecting distinct temporal changes between the two ages groups,
but when assessing this phenomenon at the metabotype level (i.e. integration all metabolite concentrations), we found no clear indication
that temporal metabolic trajectories were differentially affected by age.
Thus, age-related difference in cold tolerance was not mirrored by
differential homeostatic responses after stress. Cold hardy flies (coldacclimated or cold-adapted) generally return the perturbed metabolic
system to homeostasis within short time (< 4 h) after cold stress
(Colinet et al., 2012; Williams et al., 2014). Likewise, heat-acclimated
flies fully recover from heat-induced metabolic alteration after only 4 h,
while untreated flies show persistent metabolic disturbance
(Malmendal et al., 2006). In the present study, neither the young nor
the old flies were cold-hardy. This condition might explain why metabolic changes persisted during the entire recovery period in both
groups. Because metabolic fluxes occur rapidly (Nicholson et al., 2002),
we expected to detect divergence between the two age groups within a
short period of time after stress, especially since flies are able to quickly
restore metabolic homeostasis when favourable temperatures are restored (Colinet et al., 2016). In our study, neither age group returned to
homeostasis within 8 h, and the temporal dynamics was similar in both
groups. Divergence in metabolic trajectories between young and old
flies, if any, might require longer to be detected. In the old fly group, a
high proportion of individuals died within 24 h post cold stress, complicating the long-term assessment of biological functions. Evaluation of
metabolic trajectories of young vs. old acclimated flies might be an
alternative avenue for exploration. Despite a marked difference in cold
tolerance between young and old flies, our results show that aging does
not affect the homeostatic response of the two groups differentially, at
least in the short term following cold stress.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.exger.2017.08.021.

key compounds associated with glycolysis (e.g. G6P and F6P), as well as
sugar molecules from starch and sucrose metabolism (e.g. Tre, Glc and
Mal) that directly feed glycolysis. The relative concentrations of Glc and
Mal also varied in flies selected for longevity (Sarup et al., 2012),
suggesting a role in aging and longevity. Recent data from untargeted
high-resolution metabolomics also showed a clear decline in the level of
G6P and F6P with aging in Drosophila, corroborating our observations
(Hoffman et al., 2014). These similar patterns, from different experimental designs and methods, reinforce the notion that glycolysis is altered with aging. Energy metabolism changes with age and, more
specifically, the activity of glycolytic enzymes varies in aging flies
(Wilps et al., 1983). Aging organisms, from C. elegans to mammals,
display decreased mitochondrial functioning and increased glycolysis
(Feng et al., 2016). With aging, there is a shift in energy metabolism
that increases the demand of Glc as energy source (Feng et al., 2016).
Clearly, energy metabolism is reprogrammed with aging, which might
explain the altered concentrations of glycolytic and carbohydrate metabolites observed in the present study. Metabolomics provides a
snapshot of metabolism; however, dynamic information on how fluxes
occur through metabolic networks is not known, even in quantitative
time-series experiments. Two explanations might exist for altered metabolite levels within a pathway, with production increasing (pathway
is up-regulated) or consumption decreasing (pathway is down-regulated). Our data indicate that the glycolytic pathway is differentially
affected in young vs. old flies; however, we could not infer about the
direction of the fluxes through this pathway.
Inositol, mannitol, Fru, Ala and GDL were consistently more abundant in old flies. Ala readily interconverts with pyruvate; thus, the increase in this compound might be connected with changes in the metabolites involved in glycolysis (e.g. G6P, F6P). The higher
concentrations of mannitol and Fru in old flies are of interest. The interconversion of these metabolites allows NADP/NADPH to be generated. NADPH is being increasingly recognised as a key molecule in
aging, as it is the ultimate donor of reductive power for most ROSdetoxifying enzymes (Fernandez-Marcos and Nóbrega-Pereira, 2016).
For instance, an increase in the levels of NADPH via the transgenic
overexpression of glucose-6-phosphate dehydrogenase (G6PD) lowers
ROS-derived damage and promotes longevity in both mice and flies
(Legan et al., 2008; Nóbrega-Pereira et al., 2016). Thus, NADPH-producing pathways, mainly the pentose phosphate pathway (PPP), are
probably related to aging (Fernandez-Marcos and Nóbrega-Pereira,
2016). Interestingly, higher amounts of gluconolactone (GDL) were also
detected in old flies. GDL chelates metals and functions by scavenging
free radicals (Bernstein et al., 2004). In addition, GDL is the precursor
of gluconolactone-6-phosphate in the PPP, and its synthesis from Glc
generates NADPH.
All stresses typically challenge metabolic homeostasis (Parsons,
1991), and because metabolic fluxes occur rapidly, interpreting the
metabolite content at a single time point during stress response can be
misleading. Tracking dynamic metabolic changes in response to stress
requires the use of several measures to establish time-dependent trajectories (Nicholson et al., 2002). Thermal stress (heat and cold) provokes strong deviations in metabolic trajectories during both stress and
recovery periods (Malmendal et al., 2006; Colinet et al., 2012, 2016;
Williams et al., 2014). Our multivariate approach (ASCA) identified a
strong temporal effect in both age groups, with metabolic profiles deviating markedly from the initial state during stress and the two first
hours of recovery. Of note, the trajectories did not return to the initial
state during the recovery period in either age group. These results
suggest persistent cold-induced effects on metabolic homeostasis. A
recovery period longer than 8 h was likely necessary for the system to
return to the initial metabolic state. We observed that many free amino
acids (FAA; such as Ile, Leu, Val, Ser, Thr, Phe, Glu and Pro) showed
temporal accumulation in cold-stressed flies. The level of FAA typically
rises in cold-exposed insects (Colinet et al., 2007, 2012; Lalouette et al.,
2007; Singh et al., 2010; Teets et al., 2012). Increased FAA levels might

Acknowledgements
We are grateful to Richard Brieuc for technical help on fly
maintenance.Author contributions
H.C. designed and conducted all the experiments; H.C. and D.R.
conducted the GC–MS analysis; H.C. analysed all the data. H.C. drafted
the manuscript, and D.R revised the manuscript.Conflict of interest
None.
References
Avanesov, A.S., Ma, S., Pierce, K.A., SH ee, Yim, BC heon, Lee, Clish, C.B., Gladyshev,
V.N., 2014. Age- and diet-associated metabolome remodeling characterizes the aging
process driven by damage accumulation. elife 3, e02077.
Bernstein, E.F., Brown, D.B., Schwartz, M.D., Kaidbey, K., Ksenzenko, S.M., 2004. The
polyhydroxy acid gluconolactone protects against ultraviolet radiation in an in vitro
model of cutaneous photoaging. Dermatol. Surg. 30, 189–195.

49

Experimental Gerontology 102 (2018) 43–50

H. Colinet, D. Renault

Rev. Physiol. 67, 225–257.
Lalouette, L., Koštál, V., Colinet, H., Gagneul, D., Renault, D., 2007. Cold exposure and
associated metabolic changes in adult tropical beetles exposed to fluctuating thermal
regimes. FEBS J. 274, 1759–1767.
Le Bourg, E., 2007. Hormetic effects of repeated exposures to cold at young age on
longevity, aging and resistance to heat or cold shocks in Drosophila melanogaster.
Biogerontology 8, 431–444.
Legan, S.K., Rebrin, I., Mockett, R.J., Radyuk, S.N., Klichko, V.I., Sohal, R.S., Orr, W.C.,
2008. Overexpression of glucose-6-phosphate dehydrogenase extends the life span of
Drosophila melanogaster. J. Biol. Chem. 283, 32492–32499.
Malmendal, A., Overgaard, J., Bundy, J.G., Sørensen, J.G., Nielsen, N.C., Loeschcke, V.,
Holmstrup, M., 2006. Metabolomic profiling of heat stress: hardening and recovery of
homeostasis in Drosophila. Am. J. Physiol. Regul. Integr. Comp. Physiol. 291,
R205–12.
Minois, N., Le Bourg, E., 1999. Resistance to stress as a function of age in Drosophila
melanogaster living in hypergravity. Mech. Ageing Dev. 109, 53–64.
Mishur, R.J., Rea, S.L., 2012. Applications of mass spectrometry to metabolomics and
metabonomics: detection of biomarkers of aging and of age-related diseases. Mass
Spectrom. Rev. 31, 70–95.
Nicholson, J.K., Connelly, J., Lindon, J.C., Holmes, E., 2002. Metabonomics: a platform
for studying drug toxicity and gene function. Nat. Rev. Drug Discov. 1, 153–161.
Nóbrega-Pereira, S., Fernandez-Marcos, P.J., Brioche, T., Gomez-Cabrera, M.C., SalvadorPascual, A., Flores, J.M., Viña, J., Serrano, M., 2016. G6PD protects from oxidative
damage and improves healthspan in mice. Nat. Commun. 7, 10894.
Nueda, M.J., Conesa, A., Westerhuis, J.A., Hoefsloot, H.C.J., Smilde, A.K., Talón, M.,
Ferrer, A., 2007. Discovering gene expression patterns in time course microarray
experiments by ANOVA-SCA. Bioinformatics 23, 1792–1800.
Parsons, P.A., 1991. Evolutionary rates: stress and species boundaries. Annu. Rev. Ecol.
Syst. 22, 1–18.
Partridge, L., Gems, D., 2006. Beyond the evolutionary theory of ageing, from functional
genomics to evo-gero. Trends Ecol. Evol. 21, 334–340.
Rako, L., Hoffmann, A.A., 2006. Complexity of the cold acclimation response in
Drosophila melanogaster. J. Insect Physiol. 52, 94–104.
Rosewell, J., Shorrocks, B., 1987. The implication of survival rates in natural populations
of Drosophila: capture-recapture experiments on domestic species. Biol. J. Linn. Soc.
32, 373–384.
Sarup, P., Pedersen, S.M.M., Nielsen, N.C., Malmendal, A., Loeschcke, V., 2012. The
metabolic profile of long-lived Drosophila melanogaster. PLoS One 7, e47461.
Singh, A., Sharma, S.K., Sharma, B., 2010. Low temperature induced alterations in certain
biochemical constituents of 5th instar larvae of Philosamia ricini (Lepidoptera:
Satunidae). Open Access Insect physiol. 2, 11–16.
Smilde, A.K., Jansen, J.J., Hoefsloot, H.C.J., Lamers, R.-J.A.N., van der Greef, J.,
Timmerman, M.E., 2005. ANOVA-simultaneous component analysis (ASCA): a new
tool for analyzing designed metabolomics data. Bioinformatics 21, 3043–3048.
Soltow, Q.A., Jones, D.P., Promislow, D.E.L., 2010. A network perspective on metabolism
and aging. In: Integrative and Comparative Biology. Oxford University Press, pp.
844–854.
Sørensen, J.G., Loeschcke, V., Sorensen, J.G., 2002. Decreased heat-shock resistance and
down-regulation of Hsp70 expression with increasing age in adult Drosophila melanogaster. Funct. Ecol. 16, 379–384.
Taffett, G.E., 2003. Physiology of Aging. In: Geriatric Medicine. Springer-Verlag, New
York, pp. 27–35.
Teets, N.M.N.M., Peyton, J.T.J.T., Ragland, G.J.G.J., Colinet, H., Renault, D., D a. DA,
Hahn, DLDL, Denlinger, 2012. Combined transcriptomic and metabolomic approach
uncovers molecular mechanisms of cold tolerance in a temperate flesh fly. Physiol.
Genomics 44, 764–777.
Troncale, J.A., 1996. The aging process. Physiologic changes and pharmacologic implications. Postgrad. Med. 99 (111–4), 120–122.
Vis, D.J., Westerhuis, J.A., Smilde, A.K., van der Greef, J., 2007. Statistical validation of
megavariate effects in ASCA. BMC Bioinf. 8, 322.
Williams, C.M., Watanabe, M., Guarracino, M.R., Ferraro, M.B., Edison, A.S., Morgan,
T.J., Boroujerdi, A.F.B., Hahn, D. a, 2014. Cold adaptation shapes the robustness of
metabolic networks in Drosophila melanogaster. Evolution 68, 3505–3523.
Wilps, H., Collatz, K.G., Mehler, L., 1983. Energy metabolism and ageing of Phormia
terraenovae—I. Programmed changes of enzyme activities and substrate concentrations in carbohydrate metabolism. Comp. Biochem. Physiol. B 74, 331–335.
Xia, J., Sinelnikov, I.V., Wishart, D.S., 2011. MetATT: a web-based metabolomics tool for
analyzing time-series and two-factor datasets. Bioinformatics 27, 2455–2456.

Bowler, K., Terblanche, J.S., 2008. Insect thermal tolerance: what is the role of ontogeny,
ageing and senescence? Biol. Rev. 83, 339–355.
Burger, J.M.S., Promislow, D.E.L., 2006. Are functional and demographic senescence
genetically independent? Exp. Gerontol. 41, 1108–1116.
Butler, J.A., Ventura, N., Johnson, T.E., Rea, S.L., 2010. Long-lived mitochondrial (Mit)
mutants of Caenorhabditis elegans utilize a novel metabolism. FASEB J. 24,
4977–4988.
Chen, J.H., Hales, C.N., Ozanne, S.E., 2007. DNA damage, cellular senescence and organismal ageing: causal or correlative? Nucleic Acids Res. 35, 7417–7428.
Colinet, H., Renault, D., 2012. Metabolic effects of CO2 anaesthesia in Drosophila melanogaster. Biol. Lett. 8, 1050–1054.
Colinet, H., Hance, T., Vernon, P., Bouchereau, A., Renault, D., 2007. Does fluctuating
thermal regime trigger free amino acid production in the parasitic wasp Aphidius
colemani (Hymenoptera: Aphidiinae)? Comp. Biochem. Physiol. A Mol. Integr.
Physiol. 147, 484–492.
Colinet, H., Larvor, V., Laparie, M., Renault, D., 2012. Exploring the plastic response to
cold acclimation through metabolomics. Funct. Ecol. 26, 711–722.
Colinet, H., Siaussat, D., Bozzolan, F., Bowler, K., 2013a. Rapid decline of cold tolerance
at young age is associated with expression of stress genes in Drosophila melanogaster.
J. Exp. Biol. 216, 253–259.
Colinet, H., Larvor, V., Bical, R., Renault, D., 2013b. Dietary sugars affect cold tolerance
of Drosophila melanogaster. Metabolomics 9, 608–622.
Colinet, H., Chertemps, T., Boulogne, I., Siaussat, D., 2015. Age-related decline of abiotic
stress tolerance in young Drosophila melanogaster adults. J. Gerontol. A Biol. Sci. Med.
Sci. 71, 1574–1580.
Colinet, H., Renault, D., Javal, M., Berková, P., Šimek, P., Koštál, V., 2016. Uncovering
the benefits of fluctuating thermal regimes on cold tolerance of drosophila flies by
combined metabolomic and lipidomic approach. Biochim. Biophys. Acta Mol. Cell
Biol. Lipids 1861, 1736–1745.
Copes, N., Edwards, C., Chaput, D., Saifee, M., Barjuca, I., Nelson, D., Paraggio, A., Saad,
P., Lipps, D., Stevens, S.M., Bradshaw, P.C., 2015. Metabolome and proteome
changes with aging in Caenorhabditis elegans. Exp. Gerontol. 72, 67–84.
Coquin, L., Feala, J.D., McCulloch, A.D., Paternostro, G., 2008. Metabolomic and fluxbalance analysis of age-related decline of hypoxia tolerance in Drosophila muscle
tissue. Mol. Syst. Biol. 4, 233.
De Loof, A., 2011. Longevity and aging in insects: is reproduction costly; cheap; beneficial
or irrelevant? A critical evaluation of the “trade-off” concept. J. Insect Physiol. 57,
1–11.
Dolédec, S., Chessel, D., 1991. Recent developments in linear ordination methods for
environmental sciences. Adv. Ecol. 1, 133–155.
Feng, Z., Hanson, R.W., Berger, N.A., Trubitsyn, A., 2016. Reprogramming of energy
metabolism as a driver of aging. Oncotarget 5, 15410–15420.
Fernandez-Marcos, P.J., Nóbrega-Pereira, S., 2016. NADPH: new oxygen for the ROS
theory of aging. Oncotarget 7, 50814–50815.
Fuchs, S., Bundy, J.J.G., Davies, S.K., Viney, J.M., Swire, J.S., Leroi, A.A.M., 2010. A
metabolic signature of long life in Caenorhabditis elegans. BMC Biol. 8, 14.
Grotewiel, M.S., Martin, I., Bhandari, P., Cook-wiens, E., 2005. Functional senescence in
Drosophila melanogaster. Ageing Res. Rev. 4, 372–397.
Hoffman, J.M., Soltow, Q.A., Li, S., Sidik, A., Jones, D.P., Promislow, D.E.L., 2014. Effects
of age, sex, and genotype on high-sensitivity metabolomic profiles in the fruit fly,
Drosophila melanogaster. Aging Cell 13, 596–604.
Kostal, V., Zahradnickova, H., Simek, P., 2011. Hyperprolinemic larvae of the drosophilid
fly, Chymomyza costata, survive cryopreservation in liquid nitrogen. Proc. Natl. Acad.
Sci. U. S. A. 108, 13041–13046.
Kostál, V., Renault, D., Rozsypal, J., 2011. Seasonal changes of free amino acids and
thermal hysteresis in overwintering heteropteran insect, Pyrrhocoris apterus. Comp.
Biochem. Physiol. A Mol. Integr. Physiol. 160, 245–251.
Kostál, V., Šimek, P., Zahradníčková, H., Cimlová, J., Štětina, T., 2012. Conversion of the
chill susceptible fruit fly larva (Drosophila melanogaster) to a freeze tolerant organism.
Proc. Natl. Acad. Sci. U. S. A. 109, 3270–3274.
Koštál, V., Korbelová, J., Štětina, T., Poupardin, R., Colinet, H., Zahradníčková, H.,
Opekarová, I., Moos, M., Šimek, P., 2016a. Physiological basis for low-temperature
survival and storage of quiescent larvae of the fruit fly Drosophila melanogaster. Sci.
Rep. 6, 32346.
Koštál, V., Korbelová, J., Poupardin, R., Moos, M., Šimek, P., 2016b. Arginine and proline
applied as food additives stimulate high freeze tolerance in larvae of Drosophila
melanogaster. J. Exp. Biol. 219, 2358–2367.
Kültz, D., 2005. Molecular and evolutionary basis of the cellular stress response. Annu.

50


Related documents


PDF Document 2018 colinet renault exp gerontol
PDF Document 2018 henry and colinet 2018 sci nat
PDF Document 2012 colinet et al funct ecol
PDF Document 2015 colinet et al j gerontol a biol sci med sci
PDF Document 2013 colinet et al jeb
PDF Document 2010 colinet et al jeb


Related keywords