Báo cáo khoa học: Temporal expression of heat shock genes during cold stress and recovery from chill coma in adult Drosophila melanogaster pdf

For example, heat shock proteins Hsps are considered to play crucial roles in environmental stress tolerance and in thermal adapta-tion [2–5].. The molecular basis of adaptation to nonfr

stress and recovery from chill coma in adult

Drosophila melanogaster

Herve´ Colinet1,2, Siu Fai Lee2and Ary Hoffmann2

1 Unite´ d’E´cologie et de Bioge´ographie, Biodiversity Research Centre, Universite´ catholique de Louvain, Louvain-la-Neuve, Belgium

2 Department of Genetics, Centre for Environmental Stress and Adaptation Research, Bio21 Institute, University of Melbourne, Parkville, Australia

Introduction

Temperature plays a crucial role in determining the

distribution and abundance of animals In insects and

other ectotherms, temperature simultaneously affects

physiological processes, biophysical structures, and

metabolic activities, as well as developmental rates and

growth [1] Many insect species are seasonally exposed

to suboptimal or supraoptimal temperatures, and this

has led to the evolution of protective biochemical and

physiological mechanisms For example, heat shock

proteins (Hsps) are considered to play crucial roles in

environmental stress tolerance and in thermal

adapta-tion [2–5] Hsp genes constitute a subset of a larger

group of genes coding for molecular chaperones Their functions include transport, folding, unfolding, assem-bly⁄ disassembly, and degradation of misfolded or aggregated proteins [2,5,6]

Many Hsps are upregulated in response to a diverse array of stresses [2] In arthropods, they are induced

by environmental stressors such as heat, heavy metals, ethanol, and desiccation [3,7,8] The possibility that cold stress could elicit heat stress responses has not been investigated in many biological systems [9] The molecular basis of adaptation to nonfreezing low temperatures has not received as much attention as the

Keywords

cold stress; Drosophila melanogaster; gene

expression; Hsp; recovery

Correspondence

H Colinet, Bio21 Institute, University of

Melbourne, 30 Flemington Road, Parkville,

Victoria 3010, Australia

Fax: +61 3 8344 2279

Tel: +61 3 8344 2520

E-mail: herve.colinet@uclouvain.be

(Received 9 September 2009, revised 28

October 2009, accepted 30 October 2009)

doi:10.1111/j.1742-4658.2009.07470.x

A common physiological response of organisms to environmental stresses

is the increase in expression of heat shock proteins (Hsps) In insects, this process has been widely examined for heat stress, but the response to cold stress has been far less studied In the present study, we focused on 11 Dro-sophila melanogaster Hsp genes during the stress exposure and recovery phases The temporal gene expression of adults was analyzed during 9 h of cold stress at 0C and during 8 h of recovery at 25 C Increased expres-sion of some, but not all, Hsp genes was elicited in response to cold stress The transcriptional activity of Hsp genes was not modulated during the cold stress, and peaks of expression occurred during the recovery phase

On the basis of their response, we consider that Hsp60, Hsp67Ba and Hsc70-1 are not cold-inducible, whereas Hsp22, Hsp23, Hsp26, Hsp27, Hsp40, Hsp68, Hsp70Aa and Hsp83 are induced by cold This study sug-gests the importance of the recovery phase for repairing chilling injuries, and highlights the need to further investigate the contributions of specific Hsp genes to thermal stress responses Parallels are drawn between the stress response networks resulting from heat and cold stress

Abbreviations

Cp, crossing point; Ct, cycle threshold; HSF, heat shock factor; Hsp, heat shock protein; qRT-PCR, quantitative RT-PCR; RA, recovery with agar; RF, recovery with food; RNAi, RNA interference; sHsp, small heat shock protein.

high temperature extreme [10,11] The first report of a

cold-induced heat shock response in insects was

pro-vided by Burton et al [12], who noticed the induction

of a 70 kDa protein after a cold treatment However,

the biochemical diversity of cold-induced heat shock

responses remains poorly understood, because much of

the early work [12–14] used one-dimensional gel

elec-trophoresis, which fails to discriminate different Hsps

within a family [15] Hsps are usually divided into

dif-ferent families on the basis of their sequence homology

and their molecular masses The major families include

Hsp100, Hsp90, Hsp70, Hsp60, Hsp40, and the small

Hsps (sHsps) (sizes below 30 kDa) [2,3]

The molecular mechanisms behind recovery from

cold shock are complex, and it seems that more

genes⁄ proteins are activated during the recovery phases

than during the actual period of the cold stress itself

[16] It is thus essential to differentiate between the

cold exposure and the subsequent recovery phase

[16,17], and this aspect was investigated in studies

using alternating temperatures [18–20] Although Hsps

have been implicated in cold survival, there is little

direct evidence in the literature to confirm this link

[21] In insects, there has been a long-standing focus

on Hsp70, which remains the most commonly studied

stress protein in cold-related studies [11,16,21] Even if

the level of Hsp70 expression is a good indicator of

the whole inducible stress response, studying it alone

gives an incomplete picture of the organism’s stress

response [11] Hsp70 is known to interact with a

network of other Hsps [22–24] Therefore, if Hsp70

displays mild modulation under a specific condition, it

is possible that changes in the expression of other Hsp

genes or Hsp proteins might occur and might be

over-looked [11] In the present study, we investigated the

temporal expression patterns of 11 Hsp genes in adult Drosophila melanogaster during both the cold stress exposure phase and the recovery phase

Results and Discussion

Adult flies were subjected to prolonged cold stress for

up to 9 h at 0C, and then allowed to recover at

25C for up to 8 h, with or without food (Fig 1) We investigated the temporal expression patterns of 11 Hsp genes during both the cold stress phase and the recovery phase, using quantitative RT-PCR (qRT-PCR) At 0C, adults fall directly into chill coma, because of the inability to maintain muscle resting potentials [25] In addition to this neuromuscular per-turbation, chilling injuries accumulate at low tempera-tures as a result of various physiological dysfunctions, recently reviewed by Chown and Terblanche [26] Within certain limits, these physiological injuries are reversible As previously noted in D melanogaster [27], when the cold stress period is increased, the associated chilling injuries accumulate, and the time needed for recovery increases In our experiments, recovery times (Fig 2) followed periods of stress exposure, increasing significantly with time spent at 0C (ANOVA;

F= 239.2; degrees of freedom = 3,156; P < 0.0001) Recovery time is a highly variable physiological trait

We observed that coefficients of variation ranged from 10% to 13% for stress duration of 0.25–6 h, and then increased to 28% for 9 h of cold stress As often observed in animals, the phenotypic variability of a trait increases rather than decreases when the level of stress becomes more severe [7] There was no mortality even after 9 h of cold stress, ensuring that expression changes were not confounded by mortality

Recovered with food (RF)

25 °C

0 °C

Stressed (S)

Recovered with agar (RA)

25 °C

Corresponding controls

25 °C

25 °C

Fig 1 Experimental protocol used to evaluate the cumulative effects of prolonged cold stress followed by a recovery period on expression

of Hsp genes Treated flies were stressed (S) at 0 C for periods ranging from 0.25 to 9 h, and allowed to recover at 25 C with food (RF) or with only agar (RA) for periods ranging from 0.5 to 8 h Samples for gene expression measurements were taken at several time points during the cold stress and the recovery treatments (see text for further details) Control flies were kept at 25 C and were sampled at the same time points as the treated flies.

We focused on 11 Hsp genes representing all major

Hsp families Because chilling injury is a cumulative

process, we verified whether Hsp transcripts, other

than those of Hsp70, accumulate during cold stress

All qRT-PCR assays yielded specific products (i.e

sin-gle melting peak), and qRT-PCR efficiencies were

between 80% and 101% There was no difference in

the relative expression of the housekeeping gene RpS20

between all treatment–duration combinations,

includ-ing controls (ANOVA; F = 0.656; degrees of

free-dom = 11,30; P = 0.766)

Hsp70 and Hsp68

Of the 11 Hsp genes examined, Hsp70Aa and Hsp68

were the most cold-inducible, with overall expression

being treatment-specific, and expression being

upregu-lated during recovery treatments (Table 2 and Fig 3)

Hsp70Aawas only marginally upregulated towards the

end of the cold stress, but was upregulated 68-fold

after 2 h of recovery Upon commencement of the

25C recovery, the accumulation of Hsp68 transcripts

underwent a 0.5 h delay before peaking at 2 h of

recovery with 22-fold upregulation Goto and Kimura

[28] also found that in some temperate species of

Dro-sophila, Hsp70 mRNA accumulated after the flies were

returned to 23C following cold treatment The only

exception was Drosophila watanabei, where a low level

of upregulation of Hsp70 mRNA was observed, not only during cold exposure, but also during recovery from cold Using RNA interference (RNAi), recent studies on other insect species have demonstrated that Hsp70 is critically important for cold survival [29,30] Hsp68, which belongs to the Hsp70 family, was highly expressed during recovery Hsp68 is induced by heat stress in Drosophila [23,31,32], but there is no previous report of cold induction in this gene Hsp68 is thought

to have a similar function to Hsp70 in the protection and⁄ or renaturation of proteins, but this function may

be part of a temporally different response [31]

Hsp40 and Hsp83 Hsp40 (also called DnaJ-1) had a fairly high level of basal expression in untreated flies (relative to the housekeeping gene RpS20), confirming that it is a con-stitutively expressed gene [22] Hsp40 was upregulated

Fig 2 Increasing recovery time for flies exposed to increasing

durations of cold stress Each circle represents the value of an

indi-vidual fly Means are indicated ± standard errors (n = 40).

Table 1 Primer pairs used for qRT-PCR.

Fragment length (bp)

RpS20 (reverse) TGGTGATGCGAAGGGTCTTG

Hsp22 (reverse) TCCTCGGTAGCGCCACACTC Hsp23 (forward) GGTGCCCTTCTATGAGCCCTACTAC 153 Hsp23 (reverse) CCATCCTTTCCGATTTTCGACAC

Hsp26 (reverse) TTGTAGCCATCGGGAACCTTGTAG

Hsp27 (reverse) CTCCTCGTGCTTCCCCTCTACC Hsp40 (forward) GAGATCATCAAGCCCACCACAAC 112 Hsp40 (reverse) CGGGAAACTTAATGTCGAAGGAGAC Hsp60 (forward) ACATCTCGCCGTACTTCATCAACTC 66 Hsp60 (reverse) GGAGGAGGGCATCTTGGAACTC

Hsp67Ba (forward) TGGATGAACCCACACCCAATC 89 Hsp67Ba (reverse) CGAGGCAACGGGCACTTC

Hsp68 (forward) GAAGGCACTCAAGGACGCTAAAATG 88 Hsp68 (reverse) CTGAACCTTGGGAATACGAGTG

Hsp70Aa (forward) TCGATGGTACTGACCAAGATGAAGG 98 Hsp70Aa (reverse) GAGTCGTTGAAGTAGGCTGGAACTG Hsc70-1 (forward) TGCTGGATGTCACTCCTCTGTCTC 87 Hsc70-1 (reverse) TGGGTATGGTGGTGTTCCTCTTAATC Hsp83 (forward) GGACAAGGATGCCAAGAAGAAGAAG 150 Hsp83 (reverse) CAGTCGTTGGTCAGGGATTTGTAG

Fig 3 Mean relative expression (+standard error), based on log2transformation of qRT-PCR ratios of the assayed Hsp genes relative to RpS20 White bars represent flies exposed to cold stress (S) for periods ranging from 0.25 to 9 h (S05–S9) Gray and black bars represent flies recovering from 9 h of cold stress at 25 C with food (RF) or agar (RA), respectively, for periods ranging from 0.5 to 8 h (R05F–R8F and R05A–R8A) The symbol (w) indicates mean values that are significantly (P < 0.05) different from 0 A value equal to 0 indicates no differ-ence in expression level from control flies, whereas positive and negative values indicate upregulation and downregulation, respectively.

during the recovery period (Table 2 and Fig 3),

with-out a lag period; expression was already significantly

different from that of the control after 0.5 h of

recov-ery (P < 0.05) A peak of Hsp40 expression was

observed after 2 h of recovery Hsp40 is known to

respond to heat stress in Drosophila [23,33], but this

particular Hsp gene has not previously been reported

to be cold-inducible Hsp40 is an essential cofactor

that interacts with the members of the Hsp70 family

It accelerates the dissociation of the ADP–Hsp70

com-plex, and therefore an increase in Hsp40 concentration

may cause an increase in Hsp70 activity [34] Likewise,

the expression of Hsp83 (a homolog of mammalian

Hsp90) was not modulated during cold stress, but was

upregulated during the recovery period (Table 2 and

Fig 3) After 2 h of recovery, there was significant

modulation, reaching 3.4-fold Hsp83 was highly

expressed in control flies (nearly as much as RpS20),

confirming its constitutive expression [22,35] The

Hsp90 family gene has been implicated as having a

role in the insect diapause In this context, Hsp90

mRNA levels display species-specific modulation, being

either upregulated [36,37], downregulated [29], or

unregulated [38] Whereas, in diapausing insects,

expression results are conflicting, in nondiapausing

insects upregulation of Hsp90 following cold treatment

seems to be a general rule [28,35,36], and our

Drosoph-iladata confirm this

sHsp genes

All four members of the sHsp family (Hsp22, Hsp23,

Hsp26, and Hsp27) showed similar temporal patterns

of expression (Fig 3), which differed among

treat-ments (Table 2) The mRNA levels did not show any

modulation relative to controls during the cold stress

and in the early stage of the recovery phase After this

period, gene expression was significantly upregulated

A marked peak of expression occurred after 2 h of recovery, with average fold changes relative to controls ranging from four-fold to eight-fold (Fig 3)

Only a few studies have analyzed sHsp expression in relation to cold stress in insects Yocum et al [39] found that expression of the Hsp23 transcript of the nondiapausing flesh fly was induced in response to both severe heat and cold shocks (43C and )10 C for 2 h) Sinclair et al [17] did not observe any modu-lation of Hsp23 transcription during recovery from a short cold stress (3 h at 0C) in D melanogaster Per-haps, as for Hsp70 [12], it takes several hours under mild cold stress to obtain a response in sHsp genes However, in the same species, Qin et al [40] reported the upregulation of Hsp23 and Hsp26 during a 30 min recovery phase preceded by a cold stress of only 2 h at

0C In the present study, flies were stressed for 9 h at

0C, and we observed upregulation of four sHsp genes during recovery The reason why D melanogaster has four structurally similar sHsps is still unclear [41] In addition to their molecular chaperone function, sHsps are involved in various processes [4], some of which are important for insect cold tolerance Suppressing the expression of Hsp23 by using RNAi undermines insect survival at low temperature [29] Our expression results suggest that sHsps may play an important role

in cold tolerance Hsp22 is a key player in cell protec-tion against oxidative injuries [42], a typical feature of chilling injury [43] In addition, sHsps are effective in preserving the integrity of the actin cytoskeleton and microfilaments [44] This function is particularly important, because there is increasing evidence that cytoskeletal components are involved in insect cold tolerance [19,45] Finally, sHsps prevent caspase-dependent apoptosis [46], a process that occurs during heat and cold stress in Drosophila cells [47]

Table 2 Comparison of the overall expression of Hsp genes between the three treatments: cold stress (S), RF, and RA Different letters in the same line indicate a significant pairwise difference between treatments by least significant difference tests after ANOVA (a = 0.05).

Other Hsp genes

The mRNA levels of Hsp60, Hsp67Ba and Hsc70-1

did not show any significant modulation relative to

controls during cold stress and during recovery periods

(Table 2 and Fig 3) In contrast to Hsp70, little is

known about the significance of eukaryotic Hsp60 [48]

Hsp60had a fairly high level of basal expression

(rela-tive to the housekeeping gene RpS20), confirming its

constitutive expression [49] Heat stress does not

mod-ulate the expression of Hsp60 transcripts in Drosophila

[23,32], but it increases the expression of Hsp60 in the

blowfly [48] Our data indicate that cold stress does

not upregulate transcriptional expression of Hsp60,

suggesting that this gene is not induced by thermal

stress (heat and cold), at least in D melanogaster

Hsc70, a member of the Hsp70 family, is constitutively

expressed under nonstress conditions [2] In insects,

Hsc70 displays species-specific transcriptional changes

in response to heat stress, being either induced [50,51]

or not induced [23,52] The response of Hsc70 to cold

stress is poorly documented In the flesh fly,

transcrip-tion of Hsc70 is upregulated by cold shock and not by

heat shock [52] In the rice stem borer, the level of

Hsc70 mRNA decreases slightly during cold

acclima-tion [37] In mites, the level of Hsc70 transcript is not

changed by heat or cold shock, or by recovery after

either shock [53] We found that, in D melanogaster,

cold stress does not modulate the transcriptional

expression of Hsc70-1 Finally, the multicopy gene

Hsp67 is not responsive to cold stress There are no

data on this gene in the cold stress-related literature

In Drosophila, expression of Hsp67 is upregulated by

heat stress [32,33] Therefore, it seems that, unlike the

other Hsp genes tested here, Hsp67 does not respond

similarly to heat and cold stress The absence of

tran-scriptional change in expression in these three Hsp

genes suggests that they do not contribute to the cold

repair or cold acclimation machinery However, we

cannot exclude the possibility of potential translational

or post-translational regulation

Functional significance

There are different degrees of cold inducibility in the

D melanogaster Hsp network Some genes are only

constitutively expressed, whereas others are

constitu-tively expressed and upregulated after stress or are

exclusively inducible According to their cold stress

responses, we consider that Hsp60, Hsp67Ba and

Hsc70-1are not cold-inducible, whereas Hsp22, Hsp23,

Hsp26, Hsp27, Hsp40, Hsp68, Hsp70Aa and Hsp83 are

cold-inducible Apart from Hsp67Ba, these

designa-tions match that reported in Bettencourt et al [23], in the context of heat stress induction in D melanogaster This suggests that thermal stress (heat or cold) triggers expression changes in the same set of heat shock genes However, differences exist between the two responses: the level of expression is much higher for heat stress, and the response is also direct, whereas it is delayed for cold stress Differences between heat and cold responses could arise from differential activation of the various Drosophila heat shock factor (HSF) isoforms [54] or from incomplete activation of HSFs under cold stress, as observed under mild heat stress [55]

Upregulation of Hsps may be triggered by various accumulated chilling injuries [14,56], and there is evi-dence that Hsps play a role in the repair process in insects [30] However, the possibility that the stress response observed is a result of the activation of hard-ening⁄ acclimation mechanisms should also be consid-ered, as both processes involve expression of some Hsps

in Drosophila [40,57] Finally, it has been suggested that expression of Hsps during recovery from cold might result from the thermal stress experienced during the upshift in temperature [12] However, other results have shown that cold itself acts as a cue for the induction [28] The maximum expression levels were only attained when flies were returned to an optimal temperature (peak after 2 h) The functional explanation for this delay is unknown, but it may reflect the strong repres-sion of metabolic activity at low temperature

An additional test was performed to address the possible repair⁄ protective function of Hsps during the recovery phase On the basis on recovery times (Fig 4), flies exposed to constant cold for 16 h were

Fig 4 Sigmoid models describing the cumulative proportion of flies recovering (Y) in relation to time spent after cold stress (X) Circles: 8 h treatment (i.e 8 h of cold stress) Squares: 8 + 3 + 8 h treatment (i.e two successive cold stresses of 8 h separated by

3 h of recovery) Triangles: 16 h treatment (i.e 16 h of cold stress) The equation Y = A ⁄ [1 + 10 (log B–X)C ] was used to fit the data,and

to estimate the following parameters: A, which represents the pla-teau; B, which represents the halfway point between the bottom and the top; and C represents the slope Adjusted coefficients of determination (r 2 ) are provided for each group.

the most affected (16 h treatment), followed by flies

exposed to cold for 16 h but with an intermediate

pulse of 3 h at 25C (8 + 3 + 8 h treatment), and

finally, flies exposed to cold for 8 h (8 h treatment)

were the least affected (Fig 4) With the 16 h

treat-ment, 29% of the flies did not recover at all after

90 min (but were still alive), and parameter B (i.e the

halfway point between the bottom and the plateau of

the sigmoidal curve) was 51.99 ± 0.38 min With the

8 + 3 + 8 h treatment, 13% of the flies did not

recover after 90 min (but were still alive), and

parame-ter B was 42.94 ± 0.27 min Finally all flies of the 8 h

treatment group recovered within 90 min, and

parame-ter B was 31.20 ± 0.21 min The results suggest that

the short pulse at 25C, during which some Hsp genes

are highly expressed, allows partial repair of chilling

injuries Even though flies from the 8 + 3 + 8 h

ment group suffered less than flies from the 16 h

treat-ment group, the beneficial impact of the short pulse at

25C was not sufficient to completely offset the

physi-ological cost of the first 8 h of cold stress This short

pulse did not seem to provide any protection, as flies

of the 8 + 3 + 8 h treatment group took longer to

recover than the 8 h treatment flies In summary, these

results suggest: (a) a possible role of Hsp upregulation

in repair functions, supporting the ideas of Kosˇta´l and

Tollarova´-Borovanska´ [30]; and (b) no role of Hsp

upregulation in protective functions, supporting the

ideas of Nielsen et al [57] In addition to Hsps, the

expression of other genes, proteins or metabolites

could be regulated during the recovery from cold

[16,19,20], and may be responsible for repair processes

during recovery

Because flies are immobilized at 0C (chill coma),

long-term cold stress may damage flies though a

com-bination of temperature and starvation stresses In

mites, the Hsc70 mRNA level decreases as a result of

food restriction [53] In Drosophila, Hsp26, Hsp27 and

Hsp70 are upregulated after 58 h of starvation [58]

Sinclair et al [17] analyzed the response of Hsp70 and

Hsp23 transcripts after a 5 h starvation period, and

neither showed any expression modulation In our

experimental design, flies were starved at 0C for 9 h,

and then allowed to recover with or without food for

8 h Therefore, flies recovering without food

experi-enced a total starvation period of 17 h We did not

observe any difference between these two conditions in

the expression of the 11 genes analyzed, suggesting

that starvation stress was not severe and⁄ or long

enough to cause any differential Hsp expression

The response of D melanogaster to low temperature

is complex and still not fully understood, despite the

availability of new molecular tools [59] The current

study shows that expression of some, but not all, Hsp genes is elicited in response to prolonged cold stress Some Hsp genes (Hsp22, Hsp27, Hsp40, and Hsp68) are reported as being cold-inducible for the first time Although there is a strong indication that Hsps lead to cold tolerance [29,30], the functional significance of the heat shock response for cold stress is not fully under-stood, especially as protein denaturation is unlikely to occur at 0C [14] Both Hsp70-deficient and HSF-defi-cient mutant flies maintain some degree of heat toler-ance, suggesting that compensatory modification of other Hsp genes, such as Hsp40, Hsp68, and Hsp83, may underlie the maintenance of some degree of ther-motolerance [22,60] A key feature of the heat shock response is its suppression following restoration of normal environmental conditions [6], because some Hsps can be detrimental to normal growth [5] How-ever, we show that, for cold, the stress response is essentially observed during the recovery phase This study provides an overview of the temporal expression

of D melanogaster Hsp genes in response to cold stress As the network of Hsp genes clearly shows spe-cies specificity [61], it would be interesting to compare the stress response of D melanogaster with other models Why cold stress induces the expression of eight different Hsp genes in D melanogaster and whether or not these Hsps have overlapping activities remain open questions The use of mutant (e.g deletion and extra copy numbers) and transgenic (RNAi and overexpres-sion) lines will help us to better understand the relationship between Hsps and cold tolerance

Experimental procedures

Fly culture

We conducted our experiments on a mass-bred D melano-gaster population derived from about 50 females collected

in Innisfail (Australian east coast, 1733¢S) in May 2008 Flies were maintained in 250 mL bottles for 15 generations

at 19C and 70% relative humidity under continuous light

on a medium that contained yeast (3.2% w⁄ v), agar (3.2%) and sugar (1.6%) standard fly medium [62] The fly density was kept at approximately 500 individuals per bottle Flies were transferred at 25C for three generations at the time

of experiments

Conditions for cold stress and recovery

Both sex and age can differentially affect cold resistance and Hsp expression [5] Therefore, all tests were performed using synchronized 4-day-old virgin males CO2anesthesia

is a standard technique used to sex Drosophila flies

How-ever, there is increasing evidence that anesthesia interacts

with stress recovery [63] Therefore, to avoid any potential

confusing effect on Hsp gene expression, all flies were sexed

without CO2within an 8 h window after eclosion

For measurement of gene expression during the cold stress

and during the recovery period, groups of 20 males (4 days

old, virgin) were placed in 42 mL glass vials (without food),

which were immersed in a 10% glycol solution cooled to

0C to induce chill coma Flies were sampled after 0.25, 3, 6

and 9 h of cold stress, denoted as S025, S3, S6, and S9,

respectively (Fig 1) Preliminary experiments were

per-formed to determine stress durations at low temperature that

were not associated with mortality To ensure that gene

expression was not confounded by nutritional effects during

recovery (i.e flies refeeding after cold-induced starvation),

mRNA expression was compared in flies recovering with or

without food After 9 h of cold stress, flies were returned to

25C to recover, and randomly divided into two groups:

recovery with food (RF), or recovery with agar (RA)

(Fig 1) Flies of the RF group were placed in vials with

 10 mL of standard food medium Flies of the RA group

were transferred to vials containing  10 mL of 1% agar,

which provides a source of water but not nutrients For

determination of any cumulative patterns of gene expression

during the recovery period, samples were taken after 0.5, 2, 4

and 8 h during recovery (i.e R05F, R2F, R4F and R8F with

food, and R05A, R2A, R4A and R8A with agar only)

(Fig 1) For every sampling time, there was a corresponding

control, consisting of males kept at 25C for the same period

of time (Fig 1) For each treatment–duration combination,

four vials (20 males each) were used for molecular analysis

At each specific sampling time, flies were directly transferred

into 2 mL screwcap storage tubes, snap-frozen in liquid

nitrogen, and stored at)80 C until RNA extraction

In order to detect the cumulative effect of low

tempera-ture on the time required to recover, we used the method

described in Hoffmann et al [64] Briefly, for each cold

stress duration (i.e 0.25, 3, 6 and 9 h), 40 males were

allowed to recover at 25C, and the recovery time was

recorded Flies were considered to have recovered when

they stood up

RNA extraction and reverse transcription

Flies were ground to fine powder in 1.5 mL tubes placed in

liquid nitrogen Samples were mixed with 600 lL of lysis

buffer (containing 1% b-mercaptoethanol) from RNeasy

RNA extraction kits (Qiagen Pty, Doncaster, Australia) and

vortexed for 3–5 min to complete homogenization RNA

extraction and purification was performed using an RNeasy

spin column (Qiagen), following the manufacturer’s

instruc-tions Optional on-column DNase digestion was performed

to remove any potential genomic DNA contamination, using

an RNase-Free DNase Set (Qiagen) Total RNA was eluted

in 30 lL of diethylpyrocarbonate-treated water RNA was

quantified and quality-checked with a UV spectrophotometer (Gene Quant Pro, Amersham Bioscience, Analytical Instru-ments, Golden Valley, MN, USA) (criteria: A260 nm⁄ A230 nm

> 1.85; A280 nm⁄ A260 nm> 1.85) The integrity of RNA (i.e the presence of twp intense rRNA bands) was examined by running 1 lL of each RNA sample on a 1% agarose gel Only samples that satisfied both the quality and integrity require-ments were used in subsequent experirequire-ments Four high-quality RNA samples (i.e biological replicates) were obtained for each condition except for SO25, S6, R2F, R4F, R8F, and R8A, which had three biological replicates Three hundred nanograms of total RNA was used in the reverse transcription to cDNA, using the Superscript III First-Strand Synthesis System for qRT-PCR (Invitrogen Pty, Thornton, Australia), according to the manufacturer’s instructions The cDNA was diluted 10-fold in diethylpyro-carbonate-treated water, and stored at)20 C until use

Real-time PCR

The coding sequences of the Hsp genes were retrieved from Flybase (http:⁄ ⁄ flybase.org ⁄ ), and primers were designed using the primer3 module in biomanager (http:⁄ ⁄ www angis.org.au) (see Table 1 for details) For genes with multi-ple transcripts, only sequences common to all transcripts were considered We performed electronic PCR for all primer pairs on the reference D melanogaster genome to check for primer specificity (http:⁄ ⁄ www.ncbi.nlm.nih.gov ⁄ tools⁄ primer-blast ⁄ )

Real-time PCRs were performed on the LightCycler 480 system Reactions were performed in 384-well LightCycler plates, using LightCycler 480 High Resolution Melting Mas-ter Mix (Roche Diagnostics Pty, Castle Hill, Australia), and the crossing point (Cp), equivalent to the cycle threshold (Ct), estimates were obtained using the absolute quantifi-cationmodule in the software package The PCR reactions were performed in a final volume of 10 lL containing 1 lL

of cDNA sample, 0.4 lm each primer, and 5 lL of the 2· High Resolution Melting Master Mix After 10 min at

95C, the cycling conditions were as follows: 60 cycles at

95C for 10 s, 60 C for 15 s, and 72 C for 15 s To vali-date the specificity of amplification, a postamplification melt curve analysis was performed Amplicons were first dena-tured at 95C for 1 min, and then cooled to 65 C, and the temperature was then gradually raised to 95C in incre-ments of 0.02CÆs)1 Fluorescence data were recorded con-tinuously during this period, and subsequently analyzed using the Tmcalling module in the LightCycler 480 soft-ware

Ratio¼

ðEtargetÞDCðcontroltreatedÞp target =ðEreferenceÞDCðcontroltreatedÞp reference ð1Þ

Relative expression ratios (i.e fold change) were calcu-lated using the efficiency calibrated model of Pfaffl [65],

rather than the classic 2)DDCtmethod, which assumes

opti-mal and identical amplification efficiencies in target and

ref-erence genes In the Pfaffl model (Eqn 1), Cpis the crossing

point (i.e Ct) and E the efficiency of PCRs The ratio of

the target gene is expressed in treated samples versus

matched controls (calibrators), and normalized using the

housekeeping reference gene We used RpS20 as reference

[66] instead of actin, because cytoskeletal structures are

thermally labile in insects [19,45] To demonstrate stability

of the RpS20 transcripts within and between thermal

treat-ments, the expression of RpS20 was standardized relative to

another D melanogaster housekeeping gene (DmRP140)

[67], and analyzed using one-way ANOVA Because

ampli-fication efficiency may not be exactly 100%, accurate

deter-mination of ratios requires the estimation of efficiency [65],

preferably for every reaction [68] We used a

noise-resistant iterative nonlinear regression algorithm

(real-time pcr miner; http:⁄ ⁄ www.miner.ewindup.info ⁄ ) to

determine the efficiency of every individual reaction It has

been established that this method provides the best

preci-sion for real-time PCR efficiency estimation [68,69]

Additional test

To address the functional role of Hsps during the recovery

phase, we compared the time to recovery of flies exposed to

three different cold stress treatments: constant 0C for 8 h

(8 h treatment), or 0C for 8 h followed by 3 h of recovery

at 25C and then another 8 h at 0 C (8 + 3 + 8 h

treat-ment), or constant 0C for 16 h (16 h treatment) In the

8 + 3 + 8 h treatment, flies experienced a total cold stress

duration of 16 h, but the short pulse at 25C allows the

upregulation of Hsp genes (as observed in the present

study) All flies used were 4-day-old males, as described

previously The hypotheses advanced were as follow If

expression of Hsp genes during recovery has a repair

func-tion, the recovery time should be less after the

8 + 3 + 8 h treatment than after the 16 h treatment The

Hsp induction requires exposure to a rather long period of

cold stress [12] If this cold stress period has no

physiologi-cal cost, because of complete repair during the recovery,

the time to recover should be similar after the 8 h and the

8 + 3 + 8 h treatments Finally, if the induction of Hsp

genes during recovery has a protective function, the

recov-ery time after the 8 + 3 + 8 h treatment should be less

than after the 8 h treatment We used 45 flies in each

group, and recovery times were recorded at 25C over a

maximum period of 90 min (as described before) The

pro-portion of flies that had recovered was cumulated over

time, giving a sigmoid-shaped function A nonlinear

regres-sion method (i.e Eqn 2) was used to fit the relationship

between the dependent variable Y (cumulative percentage

of recovery) and the independent variable X (time after cold

stress) In Eqn (2), the parameter A estimates the plateau,

B the halfway point between the bottom and the top, and

Cthe slope:

Y¼ A=ð1 þ 10ðlog BXÞCÞ ð2Þ

Acknowledgements

We are grateful to L Rako, J Shirriffs, S de Garis,

A Rako, L Carrington, K Mitchell and M Telonis for assistance with fly work This study was supported

by ‘Fonds de la Recherche Scientifique – FNRS’, the Australian Reseach Council, and the Commonwealth Environmental Research Fund This paper is number BRC 148 of the Biodiversity Research Centre

References

1 Sinclair BJ, Vernon Ph, Klok CJ & Chown SL (2003) Insects at low temperatures: an ecological perspective Trends Ecol Evol 18, 257–262

2 Feder ME & Hofmann GE (1999) Heat-shock proteins, molecular chaperones, and the stress response: evolu-tionary and ecological physiology Annu Rev Physiol 6, 243–282

3 Hoffmann AA, Sørensen JG & Loeschcke V (2003) Adaptation of Drosophila to temperature extremes: bringing together quantitative and molecular approaches J Therm Biol 28, 175–216

4 Frydenberg J, Hoffmann AA & Loeschcke V (2003) DNA sequence variation and latitudinal associations in hsp23, hsp26 and hsp27 from natural populations of Drosophila melanogaster Mol Ecol 12, 2025–2032

5 Sørensen JG, Kristensen TN & Loeschcke V (2003) The evolutionary and ecological role of heat shock proteins Ecol Lett 6, 1025–1037

6 Parsell DA & Lindquist S (1993) The function of heat-shock proteins in stress tolerance: degradation and reactivation of damaged proteins Annu Rev Genet 27, 437–496

7 Hoffmann AA & Parsons PA (1991) Evolutionary Genetics and Environmental Stress Oxford University Press, New York

8 Tammariello SP, Rinehart JP & Denlinger DL (1999) Desiccation elicits heat shock protein transcription in the flesh fly, Sarcophaga crassipalpis, but does not enhance tolerance to high or low temperatures J Insect Physiol 45, 933–938

9 Al-Fageeh MB & Smales CM (2006) Control and regu-lation of the cellular responses to cold shock: the responses in yeast and mammalian systems Biochem J

397, 247–259

10 Norry FM, Gomez FH & Loeschcke V (2007) Knock-down resistance to heat stress and slow recovery from chill coma are genetically associated in a central region

of chromosome 2 in Drosophila melanogaster Mol Ecol

16, 3274–3284

11 Sørensen JG & Loeschcke V (2007) Studying stress

responses in the post-genomic era: its ecological and

evolutionary role J Biosci 32, 447–456

12 Burton V, Mitchell HK, Young P & Petersen NS (1988)

Heat shock protection against cold stress of

Drosoph-ila melanogaster Mol Cell Biol 8, 3550–3552

13 Joplin KH, Yocum GD & Denlinger DL (1990) Cold

shock elicits expression of heat shock proteins in the

flesh fly, Sarcophaga crassipalpis J Insect Physiol 36,

825–834

14 Petersen NS, Young P & Burton V (1990) Heat-shock

m-RNA accumulation during recovery from cold

shock in Drosophila melanogaster Insect Biochem 20,

679–684

15 Fangue NA, Hofmeister M & Schulte PM (2006)

Intra-specific variation in thermal tolerance and heat shock

protein gene expression in common killifish,

Fundu-lus heteroclitus J Exp Biol 209, 2859–2872

16 Clark MS & Worland MR (2008) How insects survive

the cold: molecular mechanisms – a review J Comp

Physiol B 178, 917–933

17 Sinclair BJ, Gibbs AG & Roberts SP (2007) Gene

tran-scription during exposure to, and recovery from, cold

and desiccation stress in Drosophila melanogaster Insect

Mol Biol 16, 435–443

18 Colinet H, Hance T, Vernon P, Bouchereau A &

Rena-ult D (2007) Does fluctuating thermal regime trigger

free amino acid production in the parasitic wasp

Aphidi-us colemani(Hymenoptera: Aphidiinae)? Comp Biochem

Physiol A 147, 484–492

19 Colinet H, Nguyen TTA, Cloutier C, Michaud D &

Hance T (2007) Proteomic profiling of a parasitic wasp

exposed to constant and fluctuating cold exposure

Insect Biochem Mol Biol 37, 1177–1188

20 Lalouette L, Kosˇta´l V, Colinet H, Gagneul D &

Rena-ult D (2007) Cold-exposure and associated metabolic

changes in adult tropical beetles exposed to fluctuating

thermal regimes FEBS J 274, 1759–1767

21 Michaud MR & Denlinger DL (2004) Molecular

modal-ities of insect cold survival: current understanding and

future trends Int Congress Ser 1275, 32–46

22 Neal SJ, Karunanithi S, Best A, So AK, Tanguay RM,

Atwood HL & Westwood JT (2006) Thermoprotection

of synaptic transmission in a Drosophila heat shock

fac-tor mutant is accompanied by increased expression of

Hsp83 and DnaJ-1 Physiol Genomics 25, 493–501

23 Bettencourt BR, Hogan CC, Nimali M & Drohan BW

(2008) Inducible and constitutive heat shock gene

expression responds to modification of Hsp70 copy

number in Drosophila melanogaster but does not

com-pensate for loss of thermotolerance in Hsp70 null flies

BMC Biol 6, 5, doi:10.1186/1741-7007-6-5

24 Duncan RF (2005) Inhibition of Hsp90 function delays and impairs recovery from heat shock FEBS J 272, 5244–5256

25 Hosler JS, Burns JE & Esch HE (2000) Flight muscle resting potential and species-specific differences in chill coma J Insect Physiol 46, 621–627

26 Chown SL & Terblanche JS (2008) Physiological diver-sity in insects: ecological and evolutionary contexts Adv Insect Physiol 33, 50–152

27 Rako L & Hoffmann AA (2006) Complexity of the cold acclimation response in Drosophila melanogaster

J Insect Physiol 52, 94–104

28 Goto SG & Kimura MT (1998) Heat- and cold-shock responses and temperature adaptations in subtropical and temperate species of Drosophila J Insect Physiol

44, 1233–1239

29 Rinehart JP, Li A, Yocum GD, Robich RM, Hayward

SA & Denlinger DL (2007) Up-regulation of heat shock proteins is essential for cold survival during insect dia-pause Proc Natl Acad Sci USA 104, 11130–11137

30 Kosˇta´l V & Tollarova´-Borovanska´ M (2009) The

70 kDa heat shock protein assists during the repair of chilling injury in the insect, Pyrrhocoris apterus PLoS ONE 4, e4546, doi:10.1371/journal.pone.0004546

31 Palter KB, Watanabe M, Stinson L, Mahowald AP & Craig EA (1986) Expression and localization of Drosophila melanogasterhsp70 cognate proteins Mol Cell Biol 6, 1187–1203

32 Sørensen JG, Nielsen MM, Kruhøffer M, Justesen J & Loeschcke V (2005) Full genome gene expression analy-sis of the heat stress response in Drosophila melanogas-ter Cell Stress Chap 10, 312–328

33 Leemans R, Egger B, Loop T, Kammermeier L, He H, Hartmann B, Certa U, Hirth F & Reichert H (2000) Quantitative transcript imaging in normal and heat-shocked Drosophila embryos by using high-density oligonucleotide arrays Proc Natl Acad Sci USA 97, 12138–12143

34 Laufen T, Mayer MP, Beisel C, Klostermeier D, Rein-stein J & Bukau B (1999) Mechanism of regulation of Hsp70 chaperones by DnaJ co-chaperones Proc Natl Acad Sci USA 96, 5452–5457

35 Xiao H & Lis T (1989) Heat shock and developmental regulation of the Drosophila melanogaster hsp83 gene Mol Cell Biol 9, 1746–1753

36 Chen B, Kayukawa T, Monteiro A & Ishikawa Y (2005) The expression of the HSP90 gene in response to winter and summer diapauses and thermal-stress in the onion maggot, Delia antiqua Insect Mol Biol 14, 697– 702

37 Sonoda S, Fukumoto K, Izumi Y, Yoshida H & Tsumuki H (2006) Cloning of heat shock protein genes (hsp90 and hsc70) and their expression during larval diapause and cold tolerance acquisition in the rice stem