PDF Archive

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

Share a file Manage my documents Convert Recover PDF Search Help Contact



2013 Colinet et al. IBMB .pdf



Original filename: 2013 Colinet et al. IBMB.pdf

This PDF 1.4 document has been generated by LaTeX with hyperref package / StampPDF Batch 3.0 Windows SPDF_1085 Oct 13 2003, and has been sent on pdf-archive.com on 07/07/2015 at 22:33, from IP address 93.182.x.x. The current document download page has been viewed 723 times.
File size: 1.5 MB (14 pages).
Privacy: public file




Download original PDF file









Document preview


Author's personal copy

Insect Biochemistry and Molecular Biology 43 (2013) 352e365

Contents lists available at SciVerse ScienceDirect

Insect Biochemistry and Molecular Biology
journal homepage: www.elsevier.com/locate/ibmb

Proteomic profiling of thermal acclimation in Drosophila melanogaster
Hervé Colinet a, b, *, Johannes Overgaard c, Emmanuelle Com d, Jesper Givskov Sørensen e
a

Earth and Life Institute ELI, Biodiversity Research Centre BDIV, Catholic University of Louvain, Croix du Sud 4-5, B-1348 Louvain-la-Neuve, Belgium
Université de Rennes 1, UMR CNRS 6553 Ecobio, 263 Avenue du Général Leclerc, CS 74205, 35042 Rennes Cedex, France
c
Zoophysiology, Department of Bioscience, Aarhus University, C.F. Møllers Alle 3, Building 1131, DK-8000 Aarhus, Denmark
d
Proteomics Core Facility Biogenouest, INSERM U1085 IRSET, Campus de Beaulieu, Université de Rennes 1, 263 Avenue du Général Leclerc, CS 2407, 35042 Rennes Cedex, France
e
Department of Bioscience, Aarhus University, Vejlsøvej 25, DK-8600 Silkeborg, Denmark
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 26 November 2012
Received in revised form
11 January 2013
Accepted 31 January 2013

Thermal acclimation drastically alters thermotolerance of ectotherms, but the mechanisms determining
this plastic response are not fully understood. The present study investigates the proteomic response
(2D-DIGE) of adult Drosophila melanogaster acclimated at 11, 25 or 31 C. As expected 11 C-acclimation
improved cold tolerance and 31 C-acclimation improved heat tolerance. We hypothesized that the
marked organismal responses to acclimation could be detected at the proteomic level assuming that
changes in the abundance of specific proteins are linked to the physiological changes underlying the
phenotypic response. The 31 C-acclimated flies displayed a particular divergent proteomic profile where
molecular chaperones made up a large number of the proteins that were modulated during heat acclimation. Many other proteins showed significant modulation during acclimation including proteins
involved in iron ion and cell redox homeostasis, carbohydrate and energy metabolism, chromatin
remodeling and translation, and contractile machinery. Interestingly the changes in protein abundance
were often unrelated to transcriptional activity of the genes coding for the proteins, except for the most
strongly expressed proteins (e.g. Hsp70). The 11 C-acclimation evoked weak proteomic response despite
the marked effect on the organismal phenotype. Thus the acquired cold tolerance observed here may
involve regulatory process such as posttranslational regulation rather than de novo protein synthesis.
Ó 2013 Elsevier Ltd. All rights reserved.

Keywords:
Proteomics
Acclimation
Heat
Cold
Thermotolerance
Fruit fly

1. Introduction
Temperature is a major factor shaping the distribution of ectothermic organisms (Overgaard et al., 2010; Hoffmann et al., in
press; Kellermann et al., 2012). Ectothermic animals possess
diverse responses for dealing with stressful temperatures and
exposure to different thermal environments has led to the evolution of a wide range of biochemical and physiological protective
mechanisms (Hoffmann and Sgro, 2011; Overgaard et al., 2010). It is
well known that many ectothermic animals are capable of modifying their thermal tolerance by physiological adjustments, a phenomenon referred to as thermal acclimation (Hoffmann et al.,
2003; Angilletta, 2009; Colinet and Hoffmann, 2012). Different
forms of acclimatory responses exist and physiological ecologists
commonly distinguish between plastic responses involving short-

* Corresponding author. Present address: UMR CNRS 6553 Bât 14A, Université de
Rennes 1, 263 Avenue du Général Leclerc, CS 74205, 35042 Rennes Cedex, France.
Tel.: þ33 (0) 2 23 23 66 27; fax: þ33 (0) 2 23 23 50 26.
E-mail addresses: herve.colinet@univ-rennes1.fr, colinet.herve@gmail.com
(H. Colinet).
0965-1748/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ibmb.2013.01.006

term vs. more gradual pre-exposure, referred to as rapid hardening vs. gradual acclimation respectively (Hoffmann et al., 2003;
Loeschcke and Sørensen, 2005)
The ecological aspect of thermal acclimation has been extensively studied in insects. Since cold and heat acclimation increases
cold- and heat tolerance respectively (Hoffmann et al., 2003),
acclimation has often been considered as an adaptive response
(Whitman, 2009). However, thermal acclimation is likely to be
beneficial only when the acclimation regime correctly predicts the
future regime, which is not necessarily the case under natural
fluctuating conditions. Indeed, field experiments have revealed that
the advantage of acclimation comes at a huge cost when temperature changes (Kristensen et al., 2008). Like many species, the fruit
fly Drosophila melanogaster has the capacity to enhance thermotolerance in response to a pre-exposure to sub-lethal temperature
(Hoffmann et al., 2003; Loeschcke and Sørensen, 2005; Colinet
et al., 2012a). All acclimatory responses (short or gradual) share
some common characteristics: the detection of environmental
cues, the transduction of signals into a cellular response, and the
activation of certain genes, proteins and metabolites that cause
physiological adjustments promoting thermotolerance (Angilletta,

Author's personal copy

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

2009). Acclimation in D. melanogaster causes thermal compensation on various metabolic enzymes (Burnell et al., 1991) and affects
metabolic rates (Berrigan, 1997). It also induces adaptive changes in
(i) membrane lipids composition (Overgaard et al., 2008; Kostál
et al., 2011), (ii) metabolite composition (Malmendal et al., 2006;
Overgaard et al., 2007; Kostál et al., 2011; Colinet et al., 2012a) and
(iii) the expression of various genes such as Drosophila cold acclimation (dca) (Goto, 2000) and heat shock proteins (hsps) (Colinet
and Hoffmann, 2012). Thus, the variety of biological processes
leading to thermal acclimation is expected to generate detectable
changes across several levels of biological organization. However, in
spite of the findings outlined above and the extensive knowledge of
the genetics of D. melanogaster, the understanding of acclimationrelated physiological adjustments remains poorly understood
(Doucet et al., 2009).
Transcriptomics is a limited proxy for monitoring environmental physiology as it targets genes expression which is situated
upstream of the functional molecules. Moreover, it does not account for downstream regulations, and genes and proteins
expression might operate with considerable time lag (Bahrndorff
et al., 2009). A long-standing question is how much protein
abundance is controlled at the transcriptional, post-transcriptional,
translational and post-translational levels (Schwanhäusser et al.,
2011; Suarez and Moyes, 2012; Vogel and Marcotte, 2012).
Although proteomics does not directly quantify the activity of different proteins, this “omic” approach alleviates the problem of
post-transcriptional modulation as it directly measures the abundance of the functional active molecules. Nonetheless, there is
a surprisingly paucity of proteomic studies that have focused on
thermal- and acclimatory responses in insects (Michaud and
Denlinger, 2010; Storey and Storey, 2012).
The study presented here aimed at providing a better and more
detailed understanding of the thermal acclimatory responses of
insects across several levels of biological organization. We used
a complementary approach based on proteomics and transcriptomics to decipher the underpinnings of acclimatory responses. By measuring mRNA expression of the candidate proteins,
it is possible to investigate how protein abundance is matched by
gene expression. To our knowledge, no study has characterized the
proteome of insects that were gradually acclimated at both high
and low temperature. Our overall goal of this study was thus to
identify candidate mechanisms that are responsible for the capacity
of acclimated insects to maintain metabolic homeostasis and survive thermal stress, thereby elucidating molecular pathways
responsible for acclimatory changes in D. melanogaster. While most
of studies generally focus on one thermal extreme, our comparative
approach was to contrast proteotypes (i.e. proteomic phenotypes)
that were either cold- or heat-acclimated (CA and HA respectively).
2. Material and methods
2.1. Origin and maintenance of experimental flies
Females from a laboratory population of D. melanogaster were
used for this experiment. The population was founded from a large
number of isofemales line (>500) collected in October 2010 in
Denmark (flies were kindly shared by Mads Fristrup Schou and
Volker Loeschcke). The population was maintained on a standard
oat mealesugareyeasteagar Drosophila food at low to moderately
high larval density conditions at 25 C, relative humidity (RH) of
50% and 12 h light/12 h dark cycles.
To generate experimental flies, individuals from the mass bred
population were transferred to bottles with yeasted medium to
stimulate egg production (w500 flies on 35 ml food). Flies were
subsequently placed on spoons with medium for egg laying (w10

353

pairs per spoon). Batches of 40 eggs were collected from the spoons
14e20 h later and transferred to fresh vials with 7 ml fly food. The
larvae developed at constant 25 C and upon emergence flies were
collected and transferred to fresh bottles at 25 C until sexual
maturation.
Two days after emergence flies were sexed under light CO2
anesthesia and saved in fresh food vials with a density of 25 flies
per vial that were returned to constant 25 C for another two days
to recover. Flies were placed on fresh medium every second day
during the subsequent experiment. After the recovery flies were
transferred to one of the three acclimation treatments (constant
temperature at 11, 25 or 31 C) where they were maintained for five
days before they were tested for thermotolerance or snap-frozen in
liquid nitrogen and kept at 80 C until protein and mRNA extractions. Thus, flies were tested 7 days after CO2 anesthesia; a recovery period long enough to avoid any physiological effect of
anaesthesia (Nilson et al., 2006; Colinet and Renault, 2012). Flies
coming from three different acclimation conditions were compared: cold-acclimated at 11 C (CA), benign control regime at 25 C
(C), and heat-acclimated at 31 C (HA).
2.2. Cold tolerance
Cold tolerance of flies from the three acclimation treatments
was investigated using three different assays. i) Cold survival following 1 h acute exposure to 6 C. Ten vials with ten female flies
from each acclimation treatment were placed in empty glass vials
with a moist stopper to ensure high humidity. The flies were the
transferred to a pre-cooled water bath and after the 1 h cold shock
the flies were returned to fresh food vials at 25 C to recover for
20 h before survival was evaluated from the ability to move. ii) Chill
coma recovery was tested in 25 flies from each acclimation treatment. Flies were placed individually in 5 ml glass vials at 0 C for
11 h. After the cold treatment the vials were quickly returned to
room temperature and recovery time was measured as the time
taken for flies to spontaneously stand up. The assay was stopped
after 4 h, where almost all flies had recovered and non-recovered
flies were assigned with a recovery time of 4 h (only four flies
from the 31 C did not recover during this period). iii) Critical
minimum temperature (CTmin) was tested in 25 flies from each
acclimation treatments. Flies were individually transferred to small
air-tight glass vials (5 ml) and randomly placed in a rack fitted in
a glass water tank (25 flies per acclimation group). Flies were
ramped down from 25 C at 0.1 C min 1 and temperature of knock
down (cessation of movement) was recorded for each fly
individually.
2.3. Heat tolerance
Heat tolerance was investigated using two different assays. i)
Survival following 1 h acute heat exposure to 38.5 C was determined using ten vials with ten female flies from each acclimation
treatment. The flies were transferred to empty glass vials with
a moist stopper to ensure high humidity and acutely placed in preheated water baths. Survival was scored as for cold survival. ii)
Critical maximum temperature (CTmax) was tested using 25 flies
from each acclimation treatment. This assay was performed similarly to CTmin described above with the exception that the temperature was ramped up from 25 C at 0.1 C min 1.
2.4. Analyses of phenotypic data
One-way ANOVAs were used to test for differences with respect
to CTmin, CTmax, chill coma recovery and survival following exposure to extreme high and low temperature. Survival proportions

Author's personal copy

354

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

were Arcsin-square-root transformed to achieve homogeneity of
variances. In cases where normality and homoscedasticity could
not be verified we employed non-parametric KruskaleWallis tests.
All statistical tests were performed using Sigmaplot 12.0 software
(Systat Software Inc.). Post hoc comparisons were performed using
all pairwise multiple comparisons Bonferroni t-test when assumptions were met and Dunn’s method for non-parametric data.
Data are presented as mean þ/ SEM and differences are considered significant at the P < 0.05 level.
2.5. 2D-DIGE experiment
Four biological replicates (each consisting of 25 female flies)
were used from each of the three acclimation treatments. The
samples were ground to fine powder in liquid nitrogen and then
proteins were extracted following exactly the same procedure as
previously described (Colinet et al., 2012b). Total protein concentration was determined using the Bradford Protein Assay Kit (Biorad, Marnes-la-Coquette, France) according to the manufacturer’s
instructions. 2D-DIGE experiment was performed as previously
described (Colinet et al., 2012b). Briefly, 50 mg of protein extracts
from individual biological replicates were labeled with 400 pmol of
either cyanine dyes Cy3 or Cy5 (GE Healthcare, Orsay, France), in
a reciprocal manner (i.e. dye swapping) according to a standardized
protocol (Com et al., 2011). Dye swap consists of reverse labeling of
same treatment. So, each treatment was labeled with Cy3 (two
replicates) and with Cy5 (the two other replicates). Fifty micrograms of combined protein extracts derived from a mix of all
samples were labeled with 400 pmol of Cy2 and used as internal
standard for the normalization of spot abundances. Individual Cy2,
Cy3 and Cy5 labeling reactions were mixed and incubated in
a solubilisation buffer (DeStreakTMRehydration solution; GE
Healthcare) containing 0.5% Pharmalytes pH 3e10 in a 450 ml final
volume, for 1 h at room temperature. The isoelectric focusing (IEF)
was performed with pH 3e10 NL 24 cm IPG strips as described
previously (Colinet et al., 2012b). Equilibrated IPG strips were
transferred onto a 26 20 cm 12.5% gel casted onto nonfluorescent gel support (Serva Electrophoresis). The protein separation was carried out in 1 anodal and 1 cathodal buffers (Serva
Electrophoresis) at 1 W/gel during 1 h and 2.5 W/gel overnight.
After electrophoresis, gels were scanned at a resolution of 200 mm
(pixel size) using a TyphoonÔ 9400 imager (GE Healthcare) in
fluorescence mode as previously described (Colinet et al., 2012b).
The global fluorescence intensities of the scanned images were
normalized by adjusting the exposure times to the average pixel
values acquired. The gel image analysis was performed using the
DeCyder software (version 5.01). For each protein, the mean normalized abundance (n ¼ 4) was used to calculate fold changes
among treatments. Cutoff values of >1.3-fold in absolute value
together with P-value 0.05 (one-way ANOVA with Bonferroni
adjustment) were used for the selection of differentially modulated
spots (Com et al., 2011).

was exported to Screen Picker (Proteomics Consult, Kampenhout,
Belgium) for spot picking. In gel digestion was performed as previously described (Colinet et al., 2012b). Tryptic peptides were
analyzed by nano LCeMS/MS using nano-LC system Ultimate
3000Ô (DIONEX e LC Packings, Amsterdam, The Netherlands)
coupled on-line to a linear ion trap HCT Ultra PÔ Discovery system
mass spectrometer (BrukerDaltoniK, GmBh, Germany) with the
same conditions previously reported (Colinet et al., 2012b). MS/MS
data files were processed (Auto MSn find and deconvolution) using
the Data Analysis (version 3.4, BrukerDaltoniK, GmBh, Germany)
software. For each acquisition, a maximum of 2000 compounds
were detected with an intensity threshold of 200,000 and the
charge state of precursor ions was automatically determined by
resolved isotope deconvolution. The proteinScape 2.1 software
(BrukerDaltonik GmbH) was used to submit MS/MS data to the
following database: NCBI restricted to Drosophila (June 2011,
223,543 sequences) using the Mascot search engine (Mascot server
v2.2, http://www.matrixscience.com). Parameters were set as follows: trypsin as enzyme with one allowed miscleavage, carbamidomethylation of cysteins as fixed modification and methionine
oxidation as variable modifications. The mass tolerance for parent
and fragment ions was set to 0.5 Da. Peptide identifications were
accepted if the individual ion Mascot scores were above the identity
threshold (the ion score is 10*log(P), where P is the probability
that the observed match is a random event, P < 0.05). In case of
ambiguous assignments (one compound fit to more than one
peptide), peptide were accepted based on the peptide score,
meaning that the peptide sequence with the highest score is
accepted. The compilation of identified peptides to proteins was
performed with the ProteinExtractor algorithm (Thiele et al., 2008,
2010), so that every protein reported was identified by at least one
peptide with significant ion Mascot score (above the identity
threshold). Figures were designed using Prism V 5.01 (GraphPad
software, Inc. 2007).
2.7. Multivariate analysis of the proteomic changes
A principal component analysis (PCA) was used to initially
assess the relative importance of specific proteins in differentiating
the proteotypes from the different acclimation treatments. This
analysis was performed using the ‘ade4’ library in the statistical
software ‘R 2.13.0’ (http://cran.r-project.org). The PCA was based on
the relative abundance of the proteins whose expression profiles
changed significantly among acclimation treatments (P < 0.05, oneway ANOVA with Bonferroni adjustment). We also employed hierarchical clustering analysis (HCA) to associate the proteins with
similar expression patterns across samples such that proteotypes
with similar protein expression profiles could be grouped. The HCA
was performed with Ward’s method and Euclidean distances in
form of a heat map using Permutmatrix V.1.9.3. (Caraux and
Pinloche, 2005). Mean-centered data were used in all multivariate analyses.

2.6. Preparative gel and protein identifications

2.8. Gene expression

Four hundred micrograms of a mix of protein extracts from all
analyzed samples (i.e. internal standard) were loaded on a preparative gel and were run together with the analytical gels, using in the
same experimental conditions (see above). This preparative gel was
stained with LavaPurple and scanned using a TyphoonÔ 9400
imager (GE Healthcare) in fluorescence mode as previously
described (Colinet et al., 2012b). The gel images were analyzed
using Decyder software and matched against the spots referenced
in the picking list created after the detection of the significantly upor down-regulated protein signals in analytical gels. The picking list

Gene expression was investigated for 7 genes corresponding to
proteins identified in the proteomic analysis (Table 1) and two
additional genes (zw and GlyP) related to carbohydrate and energy
metabolism. For each treatment, gene expression was assayed in
four replicate samples of 15 flies each. Total RNA was extracted
using NucleoSpin RNA II (MachereyeNagel, Düren, Germany) and
the resulting RNA concentration was determined using an
ImplenNanoPhotometer (AH Diagnostics, Aarhus, Denmark). cDNA
was synthesized from 600 ng total RNA using Omniscript Reverse
Transcriptase kit (Qiagen, Copenhagen, Denmark) and an Anchored

Author's personal copy

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

355

Table 1
List of significantly modulated proteins identified in acclimated D. melanogaster females by nano-LCeMS/MS.
Master
num.

Absolute
fold change

Accession

Protein name

MW [kDa]

282
297
340
345
369
400

1.34
1.37
1.35
1.31
1.91
1.30

gij24665669
gij21357739
gij24585709
gij24585709
gij17647529
gij19920676

Neural conserved at 73EF, isoform A
Glycoprotein 93
Elongation factor 2b, isoform A
Elongation factor 2b, isoform A
Heat shock protein 83
Trehalose-6-phosphate synthase 1

112.5
90.2
94.4
94.4
81.8
91.1

484

1.66

gij40215525

RE72980p

499
500
513

1.32
1.37
1.33

gij157658
gij157658
gij17737967

514

1.51

gij17737967

516

1.34

gij17737967

517
523

1.70
1.61

gij157667
gij17737967

525

1.32

gij21357965

528
530

1.82
1.85

gij157667
gij24643010

Heat shock protein cognate
Heat shock protein cognate
Heat shock protein cognate
isoform A
Heat shock protein cognate
isoform A
Heat shock protein cognate
isoform A
Heat shock protein cognate
Heat shock protein cognate
isoform A
Glycyl-tRNA synthetase,
isoform A
Heat shock protein cognate
Transferrin 1

Mascot
score

Nb. peptides

Biological process

Gene name

6.4
4.8
6.2
6.2
4.8
6.4

1570.4
594.3
179.8
1609.1
2562.2
1379.5

23
8
3
23
37
20

Nc73EF
Gp93
Ef2b
Ef2b
Hsp83
Tps1

66.9

4.7

902.9

14

72
72
4,

72.2
72.2
71.1

5.1
5.1
5.2

1975.2
1657.1
2891.7

28
22
41

TCA cycle
Response to stress
Protein biosynthesis
Protein biosynthesis
Response to stress
Trehalose biosynthetic
process
Carbohydrate metabolic
process
Response to stress
Response to stress
Response to stress

Hsc70-3
Hsc70-4
Hsc70-4

4,

71.1

5.2

1713.7

25

Response to stress

Hsc70-4

4,

71.1

5.2

2387.7

33

Response to stress

Hsc70-4

71
4,

74.2
71.1

5.7
5.2

1215.5
1745

18
25

Response to stress
Response to stress

Hsc70-5
Hsc70-4

75.8

6.0

324.7

5

Aats-gly

74.2
71.8

5.7
6.8

1210.8
454.2

17
7

Transferrin 1

71.8

6.8

1431.5

22

gij994753
gij24643010

Pro-phenol oxidase A1
Transferrin 1

79.0
71.8

6.1
6.8

1298.7
1339

20
21

4.99
1.48
1.49
1.73
1.58

gij57165022
gij24641191
gij24641191
gij24641191
gij17530879

Heat shock protein
Heat shock protein
Heat shock protein
Heat shock protein
Yolk protein 1

67.3
60.8
60.8
60.8
48.7

5.7
5.2
5.2
5.2
7.8

77.8
1851.9
2271.6
2121.8
1100.3

1
26
29
29
17

956

1.74

gij17530879

Yolk protein 1

48.7

7.8

1287

19

963

1.39

gij17530879

Yolk protein 1

48.7

7.8

1374.5

20

974
1009

1.32
1.30

gij111145227
gij161077703

Enolase
Yolk protein 2

46.6
49.6

6.3
8.6

782.8
1578

9
21

1209

1.30

gij221468704

32.6

5.3

472

5

1212

1.35

gij221468704

32.6

5.3

676.4

9

Chromatin remodeling

Nurf-38

1365

1.56

gij18860103

Nucleosome remodeling
factor - 38kD, isoform B
Nucleosome remodeling
factor - 38kD, isoform B
Regucalcin, isoform C

Glycyl-tRNA
aminoacylation
Response to stress
Cellular iron ion
homeostasis
Cellular iron ion
homeostasis
Metabolic process
Cellular iron ion
homeostasis
Response to stress
Response to stress
Response to stress
Response to stress
Lipid metabolic process,
vitellogenesis
Lipid metabolic process,
vitellogenesis
Lipid metabolic process,
vitellogenesis
Glycolysis
Lipid metabolic process,
vitellogenesis
Chromatin remodeling

540

1.93

gij24643010

549
550

1.48
1.63

604
675
680
683
951

33.6

6.0

1150.1

15

regucalcin

1386
1392
1394
1406
1472
1563

1.41
1.42
2.62
1.59
1.48
1.54

gij17933672
gij17737667
gij17737667
gij17737667
gij45552007
gij78706922

23.7
20.6
20.6
20.6
21.5
27.7

4.5
5.2
5.2
5.2
4.5
7.7

1032
430
406.9
246
78.3
1005.2

14
6
5
3
1
12

Mlc2
fln
fln
fln
Mlc2
Adh

1569
1578

1.88
1.52

gij156800
gij78706922

27.7
27.7

7.7
7.7

495.4
954.6

7
12

Alcohol catabolic process
Alcohol catabolic process

Adh
Adh

1580

1.38

gij17157991

21.7

5.4

602.9

10

1.46
1.34

gij78706922
gij17157991

27.7
21.7

7.7
5.4

349.2
514.8

5
7

1596
1605
1609
1613

1.52
1.52
1.78
1.56

gij156800
gij78706622
gij6451549
gij17933722

27.7
19.8
21.8
23.1

7.7
5.1
5.3
5.5

584.1
347.5
405.6
583.4

9
5
6
7

Cell redox homeostasis,
response to DNA damage
Alcohol catabolic process
Cell redox homeostasis,
response to DNA damage
Alcohol catabolic process
Response to stress
Nucleotide phosphorylation
Cellular iron ion homeostasis

Jafrac1

1582
1592

1677

2.00

gij24649019

19.6

5.7

343.8

6

1960

2.33

gij8067

Myosin light chain 2, isoform A
Flightin
Flightin
Flightin
Myosin light chain 2, isoform B
Alcohol dehydrogenase,
isoform C
Alcohol dehydrogenase
Alcohol dehydrogenase,
isoform C
Thioredoxin peroxidase 1,
isoform A
Alcohol dehydrogenase, isoform C
Thioredoxin peroxidase 1,
isoform A
Alcohol dehydrogenase
Heat shock protein 22, isoform B
Dak1
Ferritin 1 heavy chain
homologue, isoform A
Phosphatidylethanolamine-binding
protein 1
H2A histone

Multicellular organism
reproduction
Muscle system process
Muscle contraction
Muscle contraction
Muscle contraction
Muscle system process
Alcohol catabolic process

6.7

11.5

47.8

1

2082

1.36

gij21358539

Neuropeptide-like precursor 2

9.4

5.2

72.4

1

71

70Bb
60, isoform A
60, isoform A
60, isoform A

pI

Regulation of antimicrobial
humoral response
Chromatin condensation,
centrosome duplication
Humoral immune response

Mal-A1

Hsc70-5
Tsf1
Tsf1
proPO-A1
Tsf1
Hsp70Bb
Hsp60
Hsp60
Hsp60
Yp1
Yp1
Yp1
Eno
Yp2
Nurf-38

Adh
Jafrac1
Adh
Hsp22
Dak1
Fer1HCH
Pebp1
His2A
Nplp2

Author's personal copy

356

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

Oligo(dT)20 primer (Invitrogen A/S, Naerum, Denmark) in a total
volume of 20 ml. The resulting cDNA was subsequently diluted to
a concentration of 4 ng total RNA ml 1, and stored at 20 C. Primer
sequences (Table S1) were obtained from Flybase (http://flybase.
org/) and designed using Primer3 (Rozen and Skaletsky, 2000) or
obtained from published literature (Bettencourt et al., 2008).
Primers were synthesized by SigmaeAldrich (Brøndby, Denmark).
qPCR for the target genes was run on a Stratagene MX3005P (AH
Diagnostics, Aarhus, Denmark) using Stratagene Brilliant II SYBR
Green qPCRMastermix (AH Diagnostics, Aarhus, Denmark). Each
sample was measured in duplicate and contained 5 ml of cDNA
template (equivalent to 20 ng total RNA) along with 900 nM primer
in a final volume of 15 ml. DNA amplification was initiated with
95 C for 10 min to activate the DNA polymerase, followed by 40
cycles of 95 C for 10 s and 60 C for 60 s. Finally, a single cycle with
gradually increasing temperature was performed to measure the
melting points of the products. Melting curves were visually
inspected to verify a single amplification product with no primerdimers. Raw qPCR data was analyzed with a programmed Excel
workbook, DART-PCR which also identifies and omits outlier samples (Data Analysis for Real-Time PCR) (Peirson et al., 2003). Gene
expression data was normalized using the algorithm NORMA-gene,
which calculates a normalization factor without the use of reference gene (Heckmann et al., 2011). Comparisons of acclimation
effects on gene expression among treatments were performed using one way ANOVAs.

3. Results
3.1. Thermotolerance
As expected acclimation had a profound effect on thermal tolerance of D. melanogaster female flies. Survival following acute cold
exposure was significantly higher in cold acclimated flies (CA e
11 C) compared to flies acclimated to 25 C and 31 C (Fig. 1A; One-

way ANOVA; P < 0.001). CTmin and chill coma recovery time also
differed significantly among acclimation treatments such that cold
tolerance was inversely related to acclimation temperature (Fig. 1B
and C; KruskaleWallis tests; P < 0.001). Survival following exposure to acute high temperature revealed that heat acclimated flies
(HA e 31 C) had significantly higher survival rates that both control (C e 25 C) and CA flies (Fig. 1D; One-way ANOVA; P < 0.001). In
accordance with this result, CTmax was markedly different among
the three acclimation treatments demonstrating a strong positive
relationship between acclimation temperature and heat tolerance
(Fig. 1E; KruskaleWallis test; P < 0.001).
3.2. Proteomic profiling of thermal acclimation
The 2D-DIGE map revealed about 2186 (unmatched) spots from
the D. melanogaster proteome with molecular masses ranging from
10 to 250 kDa and isoelectric point between 3 and 10. Among these,
1120 spots could be repeatedly found and matched in the four
replicated gels within each treatment. These 1120 spots were used
for further analyses of proteomic differentiation. A representation
of the protein map of the preparative gel is shown in Fig. 2.
Although 207 spots were differentially expressed in abundance
among treatments (P < 0.05), only 86 spots reached the cutoff value
of 1.3-fold. All matched proteins were ranked in a plot according to
their P-value and their absolute fold change (see Fig. 3A) and none
of the 86 proteins were found to be exclusively present in one
treatment.
A PCA was performed to evaluate how the 207 spots contributed
to the separation among acclimation groups. The first principal
component (PC1) accounted for 56.8% of the total variance and this
PC clearly separated the HA proteotype from the C and CA proteotypes, suggesting that the HA treatment had the most divergent
proteomic profile (Fig. 3B). Thus, spots that were positively or
negatively correlated to PC1 were more or less abundant in HA
phenotype respectively (i.e. heat specific). The second principal

Fig. 1. Effects on thermotolerance. Phenotypic effects of thermal acclimation at 11, 25 and 31 C for five days in adult D. melanogaster females. In all cases different letters indicate
statistically differences between treatment groups of post hoc comparisons. (A) Cold shock survival to 6 C for 1 h; (B) Critical thermal minimum (CTmin), i.e. the temperature of
knock down at a temperature ramping assay at 0.1 C/min; (C) Chill coma recovery time after exposure to 0 C for 11 h; (D) Heat shock survival to 38.5 C for 1 h; (E) Critical
thermal maximum (CTmax), i.e. the temperature of knock down at a temperature ramping assay at þ0.1 C/min.

Author's personal copy

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

357

Fig. 2. 2D protein map. Representative image of the separation of D. melanogaster proteins on the preparative gel stained with fluorescent LavaPurple. Identified proteins showing
differential expression are annotated on the gel with their respective master spot number. Summarized properties of identified proteins are given in Table 1 and comprehensive
information is provided in Table S1.

Fig. 3. Analysis on proteomic data. (A) Graphical representation of quantitative proteomics data. Proteins were ranked according to their statistical P-value (y-axis, log10 base) and
their absolute fold change ratio (log2 base). Off-centered spots are those that varied the most among acclimation groups. Cutoff values of >1.3-fold change in absolute value (vertical
broken line), together with P-value 0.05 (horizontal broken line) were used to find spots of interest. All matched spots are represented (empty symbol), together with the 50 spots
of interest that were successfully identified (full symbol) with mass spectrometry. Master spot numbers of the 50 identified proteins are indicated. (B) Multivariate analysis (PCA) on
proteomic data illustrating the plotting of PC1 against PC2. The unit “d” (top right of the plot) represents the side-length of a square in the grid. A clear separation was observed
among the three acclimation proteotypes (11, 25 and 31 C). (C). Heat map and hierarchical clustering of the 207 significant proteins. Samples were grouped based on Ward’s
method and Euclidean distances clustering (mean-centered data). Rows indicate the replicated samples for each treatment (n ¼ 4).

Author's personal copy

358

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

component (PC2) accounted for 13.4% of the variance and this PC
separated the CA and C proteotypes (Fig. 3B). Spots that were the
positively or negatively correlated to PC2 were more or less abundant in the CA phenotype, respectively (Fig. S1). Correlation values
to PC1 and PC2 are shown in Fig. S1. The remaining PC’s explained
only a small fraction of total variance (e.g. PC3: 6.5%), and were not
considered as they did not significantly separate treatment groups.
To extend our analysis, we also clustered the set of differential
proteins (207 spots) by expression profile using hierarchical cluster
analysis (HCA) (Fig. 3C). The HCA further supported a similarity in
expression between the CA and C proteotypes, compared to the HA
proteotype which appears clearly distinct (Fig. 3C).
Fifty spots were selected for identification based on the magnitude of the response (>1.3 fold), their statistical significance
(P < 0.05), their high correlation to either PC1 or PC2 (see Fig. S1), as
well as their resolution and size on the preparative gel. Some spots
of interest had to be discarded because they were fused or placed at
the border of the gel, or because they appeared too faint in the
LavaPurple-stained preparative gel (compared to Cy-dye images).
Summarized information regarding protein identity of the 50
identified proteins is provided in Table 1 and the profiles of differential abundance for these 50 identified proteins are shown in
Fig. 4. The identified proteins were involved in various biological
processes such as response to stress, iron ion and cell redox homeostasis, carbohydrate and energy metabolism, chromatin
remodeling and translation, contractile machinery, and other
metabolic processes. Among these, molecular chaperones made up
a large number of proteins that were modulated during heat
acclimation. Additional and comprehensive information regarding
the identified proteins is provided in supplementary Table S2. All
proteomics raw data have been deposited in ProteomeXchange
(http://www.proteomexchange.org/; dataset ID: PXD000043).

3.3. Gene expression
To assess the correspondence between protein abundance and
transcriptional activity a subset of genes were selected for qPCR
analysis (Fig. 5). Acclimation effect on gene expression was generally small such that only three out of nine genes showed statistical significance (hsp70, zw and fln). Results are in agreement with
what registered at protein level (Fig. 4) for hsp70 and fln. In some
cases both cold and heat acclimation showed a tendency for
increased expression (most clearly for hsp70 and hsp22, but similar
tendency for tsf1, fln and GlyP). The gene zw had increased
expression at the low acclimation temperature, and tps1expression
showed a similar tendency to be negatively related to acclimation
temperature (Fig. 5).

4. Discussion
4.1. Effects on thermotolerance
The present study investigated how five days acclimation to low
(CA e 11 C), benign (C e 25 C) and warm (HA e 31 C) temperature affected thermotolerance of D. melanogaster. Several metrics
were used to assess the thermotolerance including survival, recovery time, and critical temperatures. All metrics confirmed
a clear pattern of beneficial acclimation such that cold- or heat
acclimation improved cold- or heat tolerance, respectively. Importantly, all these indices have been linked to adaptive patterns that
match expectations based on patterns of species/population distribution and climatic conditions (Hoffmann et al., 2002, in press;
Kellermann et al., 2012).

4.2. General proteomic and transcriptomic patterns
We hypothesized that the changes in thermal tolerance afforded
by a five days acclimation will be associated with significant proteomic changes and that these changes could be traced back to
biological processes of potential importance for thermal acclimation. 2D-DIGE proteomics clearly confirmed that the proteome of
flies acclimated to the three temperatures were dissimilar and that
the HA proteotype particularly displayed a profile that differed
from the C and CA proteotypes. Considering the marked effect of
both heat and cold acclimation on organismal performance, this
suggests that heat acclimation is associated with considerable
modification in the transcriptional and translational processes
while cold acclimation might be related to regulatory process such
as posttranslational modifications of various physiological responses. Transcription and translation would presumably occur
more slowly at low than at high temperature, which might account
for difference between CA and HA profiles. However, thermotolerance data support that the acclimation was effective, even at low
temperature, suggesting that physiological adjustments had
already occurred. Still, it is possible that physiological process at
low temperature might need more time reached full potential.
Nonetheless, the CA and C proteotypes differed significantly as
attested by the variance associated with the second axis of PCA and
also by the HCA (Fig. 3B and C).
Some of the identified spots that were positioned on the gel as
a horizontal train represented the same protein identity. These
horizontally adjacent spots often exist as charge variants representing posttranslational modifications where proteins have similar molecular weights but different isoelectric points (pI) (Görg
et al., 2004). This phenomenon has been observed in various proteins such as cytoskeletal proteins (Fields et al., 2012) or Hsps
(Tomanek and Zuzow, 2010). In nearly all cases, we found similar
changes in abundance of these isoforms with a treatment, apart
from some isoforms of flightin which showed opposite changes
within the same treatment. Coordinated decrease or increase in
several isoforms of a charge train suggests a down- or upregulation respectively, while changes in opposite directions
might result from post-translational modifications, such as phosphorylation events (Coling et al., 2007). We identified a number of
promising candidates linked to acclimation and below we discuss
these proteins and their putative roles in the thermal biology of
insects.
4.3. Miscellaneous proteins
Two isoforms of yolk proteins (Yp1 and Yp2), also called vitellogenin, were identified (four spots). Yp1 was up-regulated in HA,
while Yp2 was down-regulated in CA phenotype. Vitellogenin is the
major yolk protein of eggs (Bownes, 1994). Thermal acclimation
also affects several Yps in Daphnia pulex (Schwerin et al., 2009). The
observed variance in yolk proteins could be linked to changes in the
rate of egg production which is well known to be affected by
temperature (Huey et al., 1995; Dillon et al., 2007). Thus, variation
of yolk proteins may not be directly related to acclimation ability
but rather be a consequence of thermal changes. Two spots, namely
phosphatidylethanolamine-binding protein 1 (Pebp1) and
neuropeptide-like precursor 2 (Nplp2), were down-regulated in HA
phenotype. The reason for these modulations is unclear. The Pebp
protein family is a conserved group of proteins but little is known
about their role(s) (Simister et al., 2002). Likewise, the reason for
modulation of Nplp2, a neuropeptide precursor, is difficult to
decipher as neuropeptides regulate most, if not all, physiological
processes (Liu et al., 2006). The up-regulation of a regucalcin in HA
phenotype is of particular interest for acclimation response.

Author's personal copy

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365
359

Fig. 4. Differential expression profiles for the 50 identified proteins. The plots show the mean standardized log abundance derived from the four replicated gels (n ¼ 4) among 11, 25 and 31 C acclimation treatments. Different letters
indicate when acclimation treatments were significantly different. Master spot number and abbreviation of the proteins’ names are indicated on top of each graph.

Author's personal copy

360

H. Colinet et al. / Insect Biochemistry and Molecular Biology 43 (2013) 352e365

Fig. 5. Gene expression patterns. Gene expression of candidate genes in flies acclimated at 11, 25 or 31 C, respectively. Expression levels are given relative to expression at 25 C.
Significant differences are indicated by different letters.

Regucalcin is distinct from Drosophila cold acclimation (Dca) gene,
the latter being also referred to as smp-30, although these two
genes code for proteins with 72% amino acid identities (Reis et al.,
2011). Dca most likely arose from a duplication event in the
ancestral regucalcin-like gene (Arboleda-Bustos and Segarra, 2011).
Dca is well known to be involved in the adaptive response to low
temperature and cold acclimation (Goto, 2000; Clowers et al.,
2010). By contrast, little is known about the function of regucalcin protein. Regucalcin gene does not seem implicated in cold
tolerance in Drosophila species (Reis et al., 2011), but it appears to

be linked with diapause in D. Montana (Vesala et al., 2012). Modulation of regucalcin protein suggests a putative role in heat
acclimation.
4.4. Molecular chaperones
Molecular chaperones made up a large number of proteins that
were modulated during heat acclimation (15 out of 50 identified
proteins). These changes included up-regulation of Glycoprotein 93
(Gp93), an ortholog of mammalian heat shock protein HSP90b1


Related documents


2017 colinet et al scientific reports
2019 enriquez  colinet bmc genomics 1
2012 colinet et al funct ecol
2013 colinet et al ibmb
2010 colinet et al febs
2012 colinet hoffmann funct ecol


Related keywords