Chaperonin Contributes to Cold Hardiness of the Onion Maggot Delia antiqua through Repression of Depolymerization of Actin at Low Temperatures

Winter-diapause and cold-acclimated non-diapause pupae of the onion maggot, Delia antiqua (Diptera: Anthomyiidae), show strong cold hardiness. To obtain insights into the mechanisms involved in the enhancement of cold hardiness, we investigated the expression patterns of genes encoding subunits of chaperonin (CCT) and the morphology of actin, a substrate of CCT, at low temperatures. Quantitative real-time PCR analyses showed the mRNA levels of CCT subunits in pupal tissues to be highly correlated with the cold hardiness of the pupae. While actin in the Malpighian tubules of non-cold-hardy pupae showed extensive depolymerization after a cold treatment, actin in the same tissue of cold-hardy pupae was not depolymerized. Damage to cell membranes became apparent after the depolymerization of actin. Moreover, administration of Latrunculin B, an inhibitor of actin polymerization, to the larvae markedly decreased the cold hardiness of the pupae obtained. These findings suggest that CCT contributes to the cold hardiness of D. antiqua through the repression of depolymerization of actin at low temperatures.


Introduction
Low temperatures endanger the life of animals and plants through chill injuries, even when well above freezing. One of the mechanisms that cause chill injuries in cells is the denaturation and subsequent aggregation of proteins [1,2]. To avoid this, a wide range of organisms have acquired heat shock proteins (HSPs), which are classified into five major families based on molecular mass (small HSP, HSP60, HSP70, HSP90, and HSP100) [2][3][4]. Although HSPs usually work as molecular chaperones that fold and assemble newly synthesized proteins, they also prevent misfolding and aggregation of existent proteins, and convert proteins denatured under stressful conditions to their native conformations [2,[5][6][7].
Chaperonin (HSP60) in eukaryotes, often referred to as CCT (chaperonin containing t-complex polypeptide-1), is a heterooligomeric protein composed of 8 to 9 different subunit species of approximately 60 kDa [8][9][10][11]. Whereas the chaperonin in prokaryotes, GroEL, takes various substrates [12][13][14][15], until recently only actin and tubulin were known as substrates of CCT in vivo. However, numerous additional non-cytoskeletal proteins are now known to be substrates of CCT [10,11,16,17]. GroEL expression is reported to be upregulated in response to various forms of stress including low temperature, and play important roles in resistance to stress [18][19][20]. In contrast, the roles of CCT under cold stress are not well understood, although production of CCT was recently shown to be induced by heat shock in human cells [21], by chemical stress (CdCl 2 ) in the ciliate Oxytricha granulifera [22], and by cold shock in the yeast Saccharomyces cerevisiae [23].
We have long been interested in the strong cold hardiness of the diapause and cold-acclimated pupae of the onion maggot, Delia antiqua, a freeze intolerant insect. The percent survival of winterdiapause (WD) pupae after a 5-day exposure to 220uC exceeds 90% ( Fig. 1) [24,25]. Meanwhile, non-diapausing pupae also show a significantly increased survival rate when they are preexposed to 5uC (cold-acclimated non-diapause pupae, CA) ( Fig. 1) [25,26]. To explore the molecular mechanisms underlying the strong hardiness in D. antiqua, we sought for differentially expressed genes in the cold-acclimated and diapause pupae. Differential display (DD) analysis revealed the expression of a gene encoding TCP-1 (tcomplex polypeptide-1), the a subunit of CCT, in D. antiqua (DaTCP-1) to be upregulated in some tissues of pupae that exhibit enhanced cold hardiness [25]. To clarify the relationship between the enhanced cold hardiness and upregulation of CCT, the present study investigated the expression patterns of genes encoding eight subunits of CCT and the changes in the higher structure of actin, the major substrate of CCT, at low temperatures.

Cloning of All CCT Subunits and RACE
Fragments of all CCT subunits except the a subunit, which had already been obtained through a differential display analysis in the previous study [25], were obtained using degenerate PCR and RACE. The length of mRNA, predicted amino acid sequences and accession numbers for the GenbankH database are shown in Table S1. The eight CCT subunits of D. antiqua showed high amino acid identity (.70%) with corresponding CCT subunits of insect species, Drosophila melanogaster, Anopheles gambiae, Tribolium castaneum, Apis mellifera, and Bombyx mori (Table 1). Moreover, each subunit showed .57% identity with the corresponding subunit in human, yeast and rice, with the exception of the h subunit, which showed somewhat lower identity ($45%). The amino acid sequences of putative ATP-binding motifs in the N terminus of the eight subunits of D. antiqua showed a relatively high degree of identity ( Fig. S1) [27], although the identity between subunits was only 23% to 44% when calculated based on the entire sequence ( Fig. S1 and Table 2).

Expression of Each CCT Subunit
Changes in the mRNA levels of CCT subunits in association with cold acclimation and winter diapause were analyzed by Q-RT-PCR (Fig. 2). The expression of CCT subunits increased 2-to 6-fold in the brain, midgut and Malpighian tubules, as the period of cold acclimation was extended ( Fig. 2A). In winter diapause pupae, the expression levels in all tissues examined were increased 2-to 9-fold (Fig. 2B). The mRNA levels of every CCT subunit in the three tissues of cold-acclimated and winter-diapause pupae increased in accordance with the increase in cold hardiness (Fig. 1,  Fig. 2).

Cell Membrane Injury and Structure of Actin Filaments at Low Temperature
Cell membrane injury and the structure of actin were observed after staining with Trypan Blue and TRITCphalloidin staining, respectively. Cell membranes in the Malpighian tubules of cold-hardy pupae (CA32 and WD50), maintained at 17uC for 1 or 24 h after exposure to 220uC for 12 days, were not disrupted ( Fig. 3A-D). In contrast, those of ND5 were damaged by the incubation at 17uC for 24 h after the cold treatment, but not by the 1-h incubation ( Fig. 3E-G). The distribution of actin filaments in the Malpighian tubules of D. antiqua was similar to that found in Drosophila melanogaster [28] Figure 1. Schematic presentations of changes in cold hardiness of Delia antiqua pupae. Cold hardiness was assessed by percent survival after a 5-day treatment at 220uC [24][25][26]. (A) Non-diapause pupae acclimated at 5uC (CA). Larvae were reared at 23uC and 16L:8D to prepare non-diapause pupae, and 5-day-old pupae were transferred to 5uC for a specified period before testing cold hardiness. (B) Winterdiapause pupae (WD). Larvae were reared at 15uC and 12L:12D to prepare diapause pupae. Pupae were maintained under the same conditions as for rearing larvae. doi:10.1371/journal.pone.0008277.g001 Table 1. Identity between the predicted amino acid sequence of each CCT subunit in D. antiqua and corresponding sequences in other insects, human, rice, and yeast a .  Table 2. Identity between the predicted amino acid sequences of CCT subunits in D. antiqua.    Effect of the Inhibition of Repolymerization of Actin on Cold Hardiness The importance of the structure of actin to the cold hardiness of the cells was examined using an inhibitor of actin polymerization, Latrunculin B. Survival rates of ND5 pupae reared at 17uC and 16L:8D after feeding on a diet containing 0, 0.15, or 1.5 mM Latrunculin B (Lat0, Lat0.15, and Lat1.5, respectively) were not significantly different, suggesting that the concentrations of Latrunculin B used in this study are not toxic to D. antiqua (Fig. 4A). ND5 pupae that eventually died after the exposure to 220uC for 5 days appeared to have died at two discrete developmental stages: after attaining the adult morphology (Type 1), and at a stage indistinguishable from ND5 in appearance (Type 2) (Fig. 4B) [29]. The difference in morphology between Type 1 and Type 2 represents the difference in the degree of cold injury: pupae exhibiting the Type-2 form suffered from severer damage than those exhibiting the Type-1 form [29]. Most (.80%) of the Lat0 (control) pupae exhibited Type-1 death, and a transition in the type of death, from Type 1 to Type 2, was observed as the concentration of Latrunculin B was increased (Fig. 4B). These results indicate that inhibition of the repolymerization of actin brought about rapid death of the pupae during the recovery from cold stress.

Discussion
Our results present evidence that CCT contributes to the enhancement of cold hardiness in D. antiqua through repression of the depolymerization of actin at low temperatures. Recently, findings that suggest the importance of actin in cold hardiness have been made in several insects. In Culex pipiens, upregulation of two actin genes (actin 1 and actin 2) and redistribution of actin were observed under cold stress and during diapause [30]. In Sarcophaga crassipalpis, the expression of TCP-1, which encodes the a subunit of CCT, was upregulated by the entering of diapause [4]. Moreover, in Aphidius colemani, the gene expression of an actindepolymerizing factor was shown to be upregulated by low temperatures, suggesting that rearrangement of actin plays an important role in the survival of insects at low temperatures [31].
The changes in actin and tubulin, the major substrates of CCT [16], at low temperature have been investigated in several species of plants and animals. For example, actin filaments and microtubules of tobacco cells are depolymerized by low temperature, and are repolymerized during recovery from cold stress [32]. In winter wheat, Triticum vulgare, the disassembly and reorganization of microtubules occur during cold acclimation, and contribute to cold hardiness [33,34]. Upadhya et al. proposed the following hypothesis on the mechanism of disassembly of actins based on experiments using a cell line of the rat liver [35,36]. Low temperatures induce a decrease in the activity of Ca 2+ ATPase, and thereby a slow increase in the intracellular Ca 2+ level, due to a unidirectional leak of Ca 2+ from endoplasmic reticula (ER) because ER have higher Ca 2+ levels than cytoplasm. The increase in Ca 2+ levels brings about increased activity of calpain, which is known as a calcium-dependent intracellular protease that disassembles cytoskeletons including actin.
In D. antiqua, actin in the Malpighian tubules of cold-hardy pupae (cold-acclimated and diapause pupae) was not depolymerized by low temperatures, whereas actin in non-cold-hardy pupae showed extensive depolymerization after incubation at 17uC for 1 h or 24 h following the cold treatment (Fig. 3). Interestingly, damage to cell membranes was not apparent in 1 h at 17uC after the cold treatment, but became apparent after 24 h (Fig. 3). These results suggest that the disruption of cell membranes occurred after the depolymerization of actin. Thus, it was suggested that actin filaments are involved in the maintenance of cell membranes. The dense distribution of actin at the cell surface is consistent with this idea. Actually, inhibition of actin polymerization by Latrunculin B markedly decreased the survival rate of the pupae after cold treatment (Fig. 4).
The CCT subunits of D. antiqua showed 23-44% identity to each other (Table 2), the level of identity being similar to CCT subunits in mice (,30%) [37]. In Drosophila melanogaster, CCT subunits are encoded by independent genes scattered on different chromosomes (FlyBase; http://flybase.bio.indiana.edu/). Interestingly, the gene expression of all CCT subunits in D. antiqua was synchronously increased in association with the increase in cold hardiness (Fig. 2). In mouse, the genes encoding CCT subunits respectively have heat shock elements (HSEs) in the 59 non-coding region or in the first intron, and the transcription of CCT subunits is activated by the binding of heat shock transcription factors (HSFs) to HSEs [38]. Therefore, it is likely that the synchronous increase in the mRNA of CCT subunits at low temperature in D. antiqua is controlled by HSF. In D. antiqua, the expression of HSP70 and HSP90, which also presumably have HSEs in their 59 non-coding region, was upregulated by thermal stress and upon the entering of diapause [39,40].
Taken altogether, we consider the function of CCT and actin in the protection of cells from cold injury to be as follows (Fig. 5). In  [29]. Type 2: the pupal morphology was not changed from that of ND5 pupae [29]. doi:10.1371/journal.pone.0008277.g004 non-cold-hardy pupae of D. antiqua, low temperatures cause depolymerization of actin and subsequent disruption of cell membranes, although the involvement of Ca 2+ and Ca 2+ ATPase in this process has not been verified in D. antiqua (Fig. 5A). In the cold-hardy pupae, the increased levels of CCT protein maintain the higher structure of actin through the suppression of depolymerization or rearrangement of actin monomers, which results in the maintenance of cell membranes and thereby the increase in cold hardiness (Fig. 5B).

Insects
Onion maggots were originally collected at Sapporo, Japan, and their offspring have been reared on an artificial diet for about 40 generations in the laboratory [41]. Adults were reared in fabric screen cages (30630630 cm) with a supply of water, sugar and yeast extract (Dried Yeast Extract D-3, Wako Pure Chemicals, Kyoto). The cages were maintained in an environmental chamber controlled at 2361.0uC, under a 16 h light: 8 h dark photoperiod (16L: 8D) and 50-70% relative humidity. A plastic dish (8 cm in diameter 65 cm) containing damp glass beads and several pieces of onion was placed in the cage as the oviposition substrate. Eggs were inoculated on a piece of artificial diet placed on moistened fine sand in a plastic container (13 cm in diameter 68 cm). To obtain pupae with varied cold hardiness, larvae were reared under the conditions described below [24]. In the present article, the age of pupae is expressed by the days after pupariation, with the day of pupariation being day 0.
1) Cold-acclimated non-diapause pupae (CA). Larvae were reared at 20.060.2uC and 16L: 8D. After pupariation, the nondiapause pupae were maintained at 17.0uC60.2uC and 16L: 8D for 5 days (ND5). ND5 pupae were then housed in a Styrofoam box and placed in an incubator adjusted to 5.060.2uC for 0, 2, 8 or 32 days (CA0, CA2, CA8 or CA32). Because the lower threshold temperature for pupal development in D. antiqua is 5.7uC [42], no development was considered to have occurred during acclimation. Hence, all cold-acclimated pupae were regarded as ND5, irrespective of the duration of acclimation.

Dissection
The thin outermost shell of the puparia (puparial case) was removed using fine forceps, and the pupa was dissected in cold phosphate-buffered saline (PBS, 137 mM NaCl, 8 mM Na 2 HPO 4 , 2.7 mM KCl, and 1.5 mM KH 2 PO 4 , pH 7.4). The brain, midgut, and Malpighian tubules were isolated from 25 to 30 animals and snap frozen in liquid nitrogen. These tissues were chosen because they could be easily isolated and occupied a large portion of the pupal body.
Degenerate PCR mRNA was extracted from the brains of WD50 pupae using a MicroPoly (A) Purist TM kit (Ambion). cDNA was then synthesized using a TaKaRa RNA PCR Kit (AMV), Ver.3.0 (TaKaRa). Degenerate PCRs for amplifying gene fragments encoding seven subunits (b,c, d, e, f, g, and h) of CCT were conducted using the primers listed in Table S2. Thermal cycling conditions were 94uC for 3 minutes, 35 cycles of 98uC for 10 s, 48uC for 30 s and 72uC for 2 min, and 72uC for 5 min. The PCR products were separated on a 2% agarose gel, and stained with SYBR TM Green I (Molecular Probes). The fragment of expected size was extracted from the agarose gel, and cloned using pGEMH-T easy vector system I (Promega). The inserts were sequenced using a BigDye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems) on an ABI PRISM 377HN Sequencer (Applied Biosystems).

Cloning of cDNA Fragments by RACE
The 59 and 39 ends of the cDNA of CCT subunits were obtained by a modified RACE method using a FirstChoiceH RLM RACE Kit (Ambion). The first-strand cDNA was synthesized from mRNA extracted from the brains of WD50 pupae. The first PCR in the 59 RACE (39 RACE) was performed using the 59 RACE (39 RACE) outer primer supplied in the kit (Ambion) and primers specific for the respective subunit (Table S2). The nested PCRs were performed using the 59 RACE (39 RACE) inner primer and primers specific for the respective subunit (Table S2). The PCR products obtained by 59 and 39 RACE were subcloned using pGEMH-T Easy Vector System I (Promega) and sequenced.

Quantitative Real-Time PCR (Q-RT-PCR)
Total RNA was extracted from the brain, midgut, and Malpighian tubules using an RNeasy mini kit and RNase-free DNase I (Qiagen). The cDNAs synthesized by using a SYBRH RT-PCR Kit (Takara) were diluted to 1/50 for the detection of CCT subunits and 1/1000 for the 18S rRNA gene (control). The primers used for Q-RT-PCR for CCT subunits are shown in Table S2, and the primers for the 18S rRNA gene were 59-TTAAGCCATGCATGTCTAAGTAC-39 and 59-TCTCAG-GCTCCCTCTCCGGAATCG-39. The transcripts of CCT subunits were quantified using an ABI PRISM 7700 thermal cycler (Applied Biosystems). The Q-RT-PCR was carried out in a 20-ml reaction volume containing 1 ml of template cDNA, SYBR Premix Ex Taq, ROX Reference Dye (Takara) and 0.2 mM each of the primers. Shuttle PCR conditions were 95uC for 10 s followed by 40 cycles of 95uC for 5 s and 60uC for 30 s. After Q-RT-PCR, the absence of by-products was confirmed by a melting curve analysis and agarose gel electrophoresis of the PCR products. The relative molar amounts of CCT subunits and 18S rRNA transcripts were calculated based on a crossing point analysis, using standard curves generated from a cDNA standard. mRNA levels of CCT subunits were normalized with those of 18S rRNA in the same samples quantified in the same manner (three replicates). The relative mRNA levels of CCT subunits in each tissue of CA0 and WD4 pupae were set as 1.

Cold Treatment of Pupae
To avoid the inoculative freezing of pupae, the surface of the pupae was dried at the rearing temperature, and moist filter paper in the Petri dish was replaced with dry filter paper before the exposure to 220uC. The non-diapause, cold-acclimated and winter-diapause pupae were exposed to 22060.2uC and 0L:24D for 12 days.

Staining of Damaged Cells and Actin
After exposure to 220uC for a defined period, the pupae were maintained at 1760.2uC and 16L:8D for 1 or 24 h. Then the pupa was dissected in PBS as described above. The Malpighian tubules were isolated within 2 min, and immersed in Trypan Blue solution (3 mg/ml PBS) for 3 min at room temperature to stain the cells with a disrupted membrane. The tissues were rinsed in PBS three times, and immediately inspected with a stereomicroscope.

Administration of Latrunculin B
Ten-day-old larvae reared under the non-diapause conditions were fed an artificial diet containing 0, 0.15, or 1.5 mM of Latrunculin B, an inhibitor of actin polymerization, until they pupariated. ND5 pupae in Petri dishes (50 pupae/dish) were exposed to 22060.2uC for 5 days, and then maintained at 1760.2uC and 16L:8D. The survival rate and the morphology of pupae were examined 25 days after the exposure to 220uC (3 replicates).