The avian-specific small heat shock protein HSP25 is a constitutive protector against environmental stresses during blastoderm dormancy

Small heat shock proteins (sHSPs) range in size from 12 to 42 kDa and contain an α-crystalline domain. They have been proposed to play roles in the first line of defence against various stresses in an ATP-independent manner. In birds, a newly oviposited blastoderm can survive several weeks in a dormant state in low-temperature storage suggesting that blastoderm cells are basically tolerant of environmental stress. However, sHSPs in the stress-tolerant blastoderm have yet to be investigated. Thus, we characterised the expression and function of sHSPs in the chicken blastoderm. We found that chicken HSP25 was expressed especially in the blastoderm and was highly upregulated during low-temperature storage. Multiple alignments, phylogenetic trees, and expression in the blastoderms of Japanese quail and zebra finch showed homologues of HSP25 were conserved in other avian species. After knockdown of chicken HSP25, the expression of pluripotency marker genes decreased significantly. Furthermore, loss of function studies demonstrated that chicken HSP25 is associated with anti-apoptotic, anti-oxidant, and pro-autophagic effects in chicken blastoderm cells. Collectively, these results suggest avian HSP25 could play an important role in association with the first line of cellular defences against environmental stress and the protection of future embryonic cells in the avian blastoderm.

. Expression profiling of sHSPs. Expression analysis of sHSPs in chicken embryonic fibroblasts (CEFs) and chicken stage X blastoderm (Stage X) by RT-PCR (a) and quantitative real-time PCR (b,c). HSP25 and HSP30CL were detected specifically in stage X blastoderm (b). Real-time PCR was conducted in triplicate and normalised to expression of GAPDH. Significant differences between groups are indicated as ***P < 0.001. Error bars indicate the SE of triplicate analyses.
Scientific RepoRts | 6:36704 | DOI: 10.1038/srep36704 Expression of chicken stage X blastoderm-specific HSP30 subfamily members during intrauterine development, egg incubation, and egg storage. As shown above, chicken HSP25 and HSP30CL were expressed specifically in stage X blastoderms. Next, we investigated the expression patterns during intrauterine and early development and egg storage. First, we examined the expression patterns of the HSP30 subfamily during chicken intrauterine development. Expression of both genes was detected at low levels until EGK. III at which point zygotic gene activation (ZGA) occurred. In the case of HSP25, after EGK.VIII, expression increased sharply only shortly before oviposition, whereas, HSP30CL expression increased after EGK.III and EGK.VIII (Fig. 2a). Furthermore, after a 4 h incubation of eggs at 37.5 °C to promote embryonic development to the pre-streak stage 25 , expression of HSP25 decreased significantly. However, HSP30CL expression in the 4-h-incubated embryo was significantly higher than in stage X blastoderm (Fig. 2a). To examine HSP30 subfamily expression dynamics due to increased stress in the stage X blastoderm, we also quantified transcripts during egg storage at 16 °C. Expression of both HSP25 and HSP30CL was upregulated gradually during egg storage. However, HSP25 mRNA expression increased significantly after 5 days of storage versus the unstored blastoderm and this was maintained until 14 days, whereas upregulation of HSP30CL mRNA expression was not significant (Fig. 2b).
Multiple sequence alignment, phylogenetic analysis, and expression in other avian blastoderms of sHSPs containing the α-crystallin domain HSPB9-like. mRNA and protein sequences of chicken HSP25 were obtained from the NCBI Gallus gallus genome database. The HSP25 mRNA sequence contains an open reading frame of 582 base pairs with no intron sequence and encodes 194 amino acids, including the α -crystallin domain HSPB9-like. Next, we aligned the protein sequences containing the α -crystallin domain HSPB9-like from chicken HSP25, HSP30C-like, Japanese quail HSP25, HSP30C-like, collared flycatcher HSP30C-like, Tibetan ground-tit HSP30C-like, zebra finch HSP25-like, HSP30C-like, rock pigeon HSPB11-like, HSP30C, African clawed frog HSP30C, HSP30D Atlantic salmon HSP30, human HSPB9, and mouse HSPB9 (Fig. 3a). Pair-wise comparisons of sHSPs showed high degrees of homology among avian species including the α -crystallin domain, but they differed from other vertebrate sequences. Also, in the phylogenetic analysis, and the degree of similarity to chicken HSP25, they separated into two subfamilies, forming a branch separated from other vertebrates, one including chicken HSP25, Japanese quail HSP25, collared flycatcher HSP30C-like, Tibetan ground-tit HSP30C-like, zebra finch HSP25-like, and rock pigeon HSPB11-like, and another including chicken HSP30C-like, Japanese quail HSP30C-like, zebra finch HSP30C-like, and pigeon HSP30C due to N-terminal variations between the subfamilies (Fig. 3b,c). To identify the expression patterns of HSP25 and HSP30CL in other  avian species, we compared QEFs, ZEFs, and blastoderms at oviposition by RT-PCR. Similar to chicken, expression of the Japanese quail and zebra finch HSP25 and HSP30CL genes was detected specifically in blastoderms (Fig. 3d). Furthermore, qRT-PCR analysis showed that HSP25 mRNA expression increased significantly after 7 days of storage versus the unstored blastoderm in quail and zebra finch (Fig. 3e), similar to the case of chicken embryo.

Effect of HSP25 knockdown in chicken blastoderm cells.
To investigate chicken HSP25 function, we designed two candidate siRNAs, siRNA-296 and siRNA-497, and transfected them into blastoderm cells in vitro. At 48 h after transfection, the HSP25 transcript level was analysed using real-time PCR. As shown in Fig. 4a, siRNA-296 was the most efficient knockdown probe for HSP25, showing 68% suppression versus the control (P < 0.01). This siRNA was selected for further knockdown experiments of chicken HSP25. First of all, we examined the role of chicken HSP25 in stem cell property, like as mouse ESCs 5,9 . After knockdown of HSP25 in chicken blastoderm cells, we examined the expression of pluripotency markers, including nanog homeobox (NANOG), POU domain class 5 transcription factor 3 (POUV), SRY (sex determining region Y)-box 2 (SOX2), and cripto, FRL-1, cryptic family 1B (CRIPTO). Compared with the control, knockdown of HSP25 in blastoderm cells decreased the expression of all pluripotency genes significantly (P < 0.05; Fig. 4b). Furthermore, we performed a TUNEL assay to examine the effects of HSP25 knockdown on apoptosis in chicken blastoderm cells. Apoptotic signals were not detected in the control samples, but were strongly induced in most blastoderm cells after HSP25 knockdown (Fig. 4c).
Anti-apoptotic function of HSP25 in chicken blastoderm cells. Next, we investigated chicken HSP25 function in blastoderm cell survival under various stress conditions. First, to study the anti-apoptotic function of chicken HSP25, we assessed apoptotic cell death in blastoderm cells after HSP25 knockdown followed by mitomycin C treatment with annexin V/PI and flow cytometry analysis (Fig. 5a). After mitomycin C treatment on blastoderm cells, the proportion of live cells was decreased (annexin V − and PI − , Fig. 5b) and apoptotic cells (annexin V + only + annexin V + and PI + , Fig. 5c) was increased significantly. Then, compared with the control,  Fig. 5c shows that the repression of HSP25 increased the proportion of apoptotic cells significantly (30 μ M, P < 0.05 and 60 μ M, P < 0.01). Correspondingly, expression of anti-apoptotic genes, such as BCL2 and BCL2L1, was decreased significantly by HSP25 knockdown followed by 60 μ M mitomycin C treatment in chicken blastoderm cells (Fig. 5d).

Anti-oxidant and anti-necrotic effects of HSP25 in chicken blastoderm cells. Chicken blasto-
derm cells were treated with H 2 O 2 to create oxidative stress after HSP25 knockdown, then a DCFDA assay was performed with flow cytometry to measure ROS in blastoderm cells (Fig. 6a). The DCFDA analysis showed that ROS production was significantly higher after HSP25 knockdown than the control with or without H 2 O 2 treatment (P < 0.05; Fig. 6b). Furthermore, we conducted a gene expression analysis of anti-oxidant genes by real-time PCR. Figure 6c shows that two anti-oxidant genes, glutathione peroxidase 3 (GPX3) and glutathione peroxidase 4 (GPX4), were downregulated after HSP25 knockdown followed by 200 μ M H 2 O 2 treatment.
Generally, oxidative stress due to H 2 O 2 promotes cell death in various cell types. Thus, we also used annexin V/PI and flow cytometry to investigate effects of HSP25 on cell death with oxidative stress (Fig. 7a). After knockdown of HSP25, double negative annexin V/PI cells (live cells; Fig. 7b) were decreased significantly with H 2 O 2 treatment (100 μ M, P < 0.01 and 200 μ M, P < 0.05). Notably, although the proportion of apoptotic cells (annexin V + only + annexin V + and PI + ) was slightly, but not significantly increased (Fig. 7b), that of necrotic cells (PI + only) was increased significantly after HSP25 knockdown followed by H 2 O 2 treatment (100 μ M, P < 0.001 and 200 μ M, P < 0.01; Fig. 7c).

Pro-autophagic function of HSP25 in chicken blastoderm cells.
We examined the pro-autophagic function of chicken HSP25 in blastoderm cells by siRNA knockdown and autophagy stimulation experiments. After 48 h of knockdown of chicken HSP25, we treated cells with MG132 at 5 and 10 μ M to induce autophagy, then analysed the cells using flow cytometry for LC3 antibody-Alexa488 fluorescence (Fig. 8a). After MG132 treatment on blastoderm cells, the stimulation of autophagy was increased significantly (Fig. 8b). Then, quantification of LC3-positive cells showed that HSP25 knockdown decreased autophagy activation significantly under MG132 induction versus the control. However, there was no difference between knockdown and the control without MG132 induction (Fig. 8b). Real-time PCR analysis of pro-autophagic genes was also performed. Figure 8c shows that expression levels of several pro-autophagic genes, such as phosphatase and tensin homolog (PTEN), beclin 1, autophagy related (BECN1), UV radiation resistance associated (UVRAG), autophagy related 12 (ATG12), and autophagy related 5 (ATG5), were downregulated after HSP25 knockdown followed by 10 μ M MG132-induced autophagy. Discussion sHSPs are ATP-independent, contain an α -crystallin domain, and prevent the irreversible denaturation of other proteins. sHSPs are known to be responsible for the transfer of other proteins to the ATP-dependent chaperones or to the protein degradation machinery, such as proteasomes or autophagosomes 3 . In many organisms, sHSPs are accumulated for stress tolerance during diapause and other dormant states in which the ATP concentration is low or limited 27,28 . Among avian species, newly oviposited EGK stage X chicken blastoderms can endure in a dormant state known as cold torpor for 3-4 weeks 26,29 . Thus, we hypothesised that chicken blastoderm may show specific sHSP protein expression and function.
First, we examined the expression of 11 sHSPs, including the HSP30 subfamily, in chicken EGK stage X blastoderm cells. Among them, HSP25 and HSP30CL of the HSP30 subfamily were expressed specifically in the chicken blastoderm and HSP25 showed higher expression than HSP30CL. The HSP30 subfamily is absent from mammals and may be restricted to oviparous animals, such as frogs, fish, and avians 19,30,31 . Thus, we identified two sHSPs which may be candidates for regulating dormancy in the chicken blastoderm.
Next, we analysed the expression patterns of HSP25 and HSP30CL during intrauterine and early development. HSP25 mRNA was upregulated exclusively after EGK.VIII, shortly before oviposition, whereas HSP30CL mRNA was induced after EGK.III, subsequently decreased at EGK.VI, and upregulated again after EGK.VIII. These findings indicated that HSP25 and HSP30CL transcripts were not maternally inherited for embryonic development and stress tolerance. Additionally, developmentally constitutive HSP30 homologues in Xenopus were first detected in early and mid-tailbud embryos, after gastrulation, but not soon after ZGA, which in Xenopus occurs at the 128-to 256-cell stage 7 . It was recently discovered that ZGA starts between EGK stages II and III in chicken 32,33 . HSP25 and HSP30CL transcripts were also not induced immediately after ZGA, at EGK III, but earlier than the homologues in Xenopus. Accordingly, HSP25 expression may be related to gaining embryonic tolerance against environmental stress; however, the expression pattern of HSP30CL may indicate effects at both the cleavage stage, between EGK.III and VI, and later. Additionally, HSP25 mRNA was decreased significantly, whereas HSP30CL mRNA was upregulated significantly when embryonic development was reinitiated in the 4-h-incubated embryo versus the stage X blastoderm. Species-specific sHSP gene expression was increased and expressed exclusively in diapause and dormancy states, such as the dauer stage of C. elegans and cysts of A. franciscana, but was not detected when development continued 10,11 . Thus, HSP25 gene expression was reduced in developmental initiation, but the HSP30CL gene was induced, more similar to the gastrulation stages in Xenopus 7 .
Next, we examined HSP25 and HSP30CL expression during egg storage, a dormant state in which energy production was limited. Even in this dormant state under cold storage of eggs for 14 days, HSP25 expression was upregulated significantly by approximately four-fold after 5 days of storage versus unstored blastoderm, and this level was maintained through 14 days of storage. However, the level of HSP30CL expression after storage was only about two-fold higher (not significant). As shown in Fig. 1a, HSP25 transcript levels were the highest of the sHSPs in stage X blastoderms. Collectively, these results indicated that, during early development in chicken, HSP25 was most abundant, and may be specific for embryonic programming of self-defence against future stresses after oviposition.
To identify HSP25 homologues in vertebrates, we generated multiple alignments and performed phylogenetic tree analysis of sHSPs containing α -crystallin domains HSPB9-like in chicken, Japanese quail, collared flycatcher, Tibetan ground-tit, zebra finch, rock pigeon, frog, salmon, human, and mouse. According to the multiple alignment and phylogenetic trees, the α -crystallin domain in sHSPs was highly conserved, but there was also distinct variations in the N-terminal domain among species. In particular, the N-termini of avian sHSPs differ significantly from those of human, mouse, salmon, and frog sHSPs. Thus, HSP25 and HSP30CL in avian species are quite different to those of other vertebrates.
Although the structure and organisation of the N-terminus in sHSPs is not yet well defined, it is known that the N-terminal domains determine phosphorylation sites, and dimer and oligomer organisation 2,3 . Thus, the N-terminal similarity of avian sHSPs could indicate analogous function and substrates. Moreover, based on phylogenetic trees, they can be divided into two groups: one represented by chicken HSP25 and the other by chicken HSP30CL. The similarity between HSP25s versus other sHSPs also showed the same aspect of branching. Thus, avian HSP25 and HSP30CL are distinct sHSPs and may function differently. Furthermore, because of the specific expression of quail and zebra finch HSP25 and HSP30CL in blastoderms at oviposition, and the increased expression of HSP25 after 7 days of cold storage in both species, which are similar to that in chicken, HSP25 and HSP30CL homologues may be expressed and act in the same way throughout avian species.
Chicken HSP25, induced by various stresses, has been studied in somatic fibroblast cells 19 and adult tissues 20-23 . Chicken HSP25 was first isolated by Kawazoe et al. Subsequently, Katoh et al. observed the accumulation of HSP25 in the aggresomes of somatic fibroblasts. In terms of the genomic structure of chicken HSP25, the lack of introns may facilitate rapid expression without disturbance by stressors that could interfere with RNA splicing 34 . Also, HSP25 was the most significant sHSP expressed in testes, brain, liver, and egg muscle of egg-laying and broiler adult chickens in response to acute heat stress exposure [21][22][23] . Thus, among the sHSPs in chicken, HSP25 could be the first line of the cellular defence against environmental stresses.
Although chicken HSP25 was discovered through accumulation and inclusion formation in chicken somatic cells 19 and adult tissues [20][21][22][23] , to date, loss of function studies have not been reported. Thus, we next investigated the effects of siRNA-mediated HSP25 knockdown in chicken blastoderm cells in vitro using siRNA-296. Previous studies demonstrated stemness was composed of stress defence, as well as pluripotency in mouse embryonic stem cells. Hspb1, a unique signature for stress tolerance, was specific to mouse embryonic stem cells and was reduced along with pluripotency markers by loss of stemness during embryoid bodies formation 5,9 . Therefore, we examined pluripotency-related genes, including NANOG, POUV, SOX2, and CRIPTO 35 , in chicken blastoderm cells harbouring stemness after HSP25 knockdown. In vitro knockdown of HSP25 in blastoderm cells caused decreased expression of all pluripotency-related genes, suggesting that HSP25 seems to have a positive correlation with stem cell property in chickens, like Hspb1 in mouse embryonic stem cells.
sHSPs are known to be involved in cell survival mechanisms, such as anti-apoptosis [36][37][38][39][40][41][42] , anti-oxidative stress [43][44][45][46] , and autophagy [47][48][49] . Cellular homeostasis and integrity should be maintained appropriately for future embryonic development. Nevertheless, the role of chicken HSP25 and dormancy-specific sHSPs related to biological processes in cell protection have remained largely unknown. We investigated blastoderm cells after HSP25 knockdown followed by mitomycin C treatment to induce apoptosis. Annexin V/PI analysis showed that suppression of HSP25 decreased live cells significantly and increased apoptotic cells versus the control. Also, based on qRT-PCR analysis, anti-apoptotic genes were downregulated in knockdown blastoderm cells. These results suggest that chicken HSP25 in blastoderm cells is important for regulating apoptosis against stress. Additionally, we discovered an anti-oxidant effect of HSP25. DCFDA and qRT-PCR analysis indicated that knockdown of HSP25 followed by H 2 O 2 treatment increased ROS production significantly and downregulated some anti-oxidant genes in blastoderm cells versus the control. Furthermore, we measured live and dead cells using annexin V/PI flow cytometry. Due to increased ROS production after HSP25 knockdown, live cells were decreased significantly, but not apoptotic cells. Rather, necrotic cells were increased, consistent with a previous study indicating ROS-induce necrosis 50 . Finally, to assess the pro-autophagic function of chicken HSP25, we examined autophagy stimulation with the autophagy marker LC3 after inhibition of HSP25. Quantification using flow cytometry indicated that autophagic activation was suppressed significantly with HSP25 knockdown versus the control. Additionally, the relative expression of pro-autophagic genes was downregulated by HSP25 knockdown. Consequently, chicken HSP25 may be involved in the regulation of processes including apoptosis, anti-oxidative stress, and autophagy for cellular integrity in blastoderm cells.
In conclusion, we found that HSP25 may be associated with stress-tolerant characteristics in blastoderm cells in chicken. Also, homologues of chicken HSP25 were present in and conserved among avian species. Furthermore, we performed chicken HSP25 knockdown experiments in blastoderm cells to examine functions associated with apoptosis, anti-oxidative stress, and autophagy. Finally, we demonstrated avian HSP25 is the first line of cellular defence against environmental stresses and is important in protecting future embryonic cells in the avian blastoderm in cold torpor, as a dormancy state.

Methods
Experimental animals and animal care. The care and experimental use of animals was approved by the Institute of Laboratory Animal Resources, Seoul National University (SNU-150827-1). Animals were maintained according to a standard management program at the University Animal Farm, Seoul National University, Korea. The procedures for animal management, reproduction, and embryo manipulation adhered to the standard operating protocols of our laboratory.
Sample preparation. Blastoderms were obtained within 6 h after oviposition from White Leghorn chickens (WL), Japanese quail (JQ), and zebra finch (ZF). WL, JQ, and ZF eggs were incubated with intermittent rocking at 37.5 °C under 60-70% relative humidity. Chicken blastoderm cells were collected by gentle dissociation of EGK stage X blastoderms from WL eggs 25 . WL eggs were stored at 16 °C under 70-80% relative humidity for 2 weeks. The egg-laying times of the WL hens were recorded and intrauterine eggs from EGK stages III-VIII were harvested using an abdominal massage technique 51 . To collect oocytes and zygotes, WL hens were sacrificed and the follicles were collected. Chicken embryonic fibroblasts (CEFs), quail embryonic fibroblasts (QEFs), and zebra finch embryonic fibroblasts (ZEFs) were collected by dissociating the embryonic body of Hamburger and Hamilton (HH) stage 28 in 0.05% trypsin-EDTA (GIBCO Invitrogen, Grand Island, NY, USA) at 37 °C for 10 min 52 . Cells were then cultured in Dulbecco's modified Eagle's medium (DMEM; Thermo Fisher Scientific, Inc., Waltham, MA, USA) containing 10% FBS and 1% antibiotic-antimycotic (Invitrogen) in a 5% CO 2 atmosphere at 37 °C.
RT-PCR and quantitative real-time PCR analysis. Total RNA was isolated using the Trizol Reagent (Invitrogen, Carlsbad, CA), according to the manufacturer's protocol. For RT-PCR and quantitative real-time PCR analysis of mRNAs, total RNA (1 μ g) was used as template for cDNA synthesis using the SuperScript III First-Strand Synthesis System (Invitrogen). The cDNA was serially diluted five-fold and was equalised quantitatively for PCR amplification. Primers for real-time PCR of each gene transcript were designed using the program Primer3 (Supplementary Table S1; http://frodo.wi.mit.edu/). RT-PCR was performed with an initial incubation at 95 °C for 5 min, followed by 35  Chicken blastoderm cell culture. Stage X blastoderm cells were prepared as described previously 54 .
Transfection and chemical treatment of chicken blastoderm cells. HSP25-specific siRNAs were designed and purchased from Bioneer Corporation (Supplementary Table S2; Daejeon, Korea). For the transfection of siRNAs into the cultured blastoderm cells, Lipofectamine RNAiMAX (Invitrogen) was used according to the manufacturer's protocol for 48 h. For the in vitro induction of apoptosis, reactive oxygen species (ROS) production, and autophagy, cells were treated with mitomycin C (Sigma-Aldrich, St. Louis, MO, USA), H 2 O 2 (Sigma), and MG132 (Sigma). Specifically, 0, 30, and 60 μ M mitomycin C was added for 4 h; 0, 100, and 200 μ M H 2 O 2 was added for 4 h; and 0, 5, and 10 μ M MG132 was added for 24 h. After each treatment, cells were harvested and analysed. TUNEL assay. Cells were washed and concentrated on glass slides. After fixation in 2% paraformaldehyde for 15 min, the cells were incubated in a permeabilisation solution (0.1% Triton X-100 in PBS) for 10 min. Apoptotic cells were identified using an in situ Cell Death Detection Kit and TMR red (Roche Applied Science, Basel, Switzerland) that stains apoptotic cells red. Cells were counterstained with DAPI, mounted, and analysed under a fluorescence microscope (TU-80, Nikon).

Flow cytometry.
To examine apoptotic cell death, annexin V/propidium iodide (PI) (Thermo Fisher Scientific) staining was performed according to manufacturer's protocol and cells were analysed by flow cytometry (FACSCalibur, BD Biosciences, San Jose, CA, USA). Annexin V and PI double-negative cells were considered viable, annexin V-positive and PI-negative (early apoptotic) cells and annexin V-and PI-positive (late apoptotic) cells were considered apoptotic, and PI only-positive cells were considered necrotic. To assess cellular ROS, a 2′ ,7′ -dichlorofluorescein diacetate (DCFDA) assay (Abcam, Cambridge, UK) was performed according to the manufacturer's protocol and analysed with the FACSCalibur. To assay autophagy activity, cells were labelled with anti-light chain 3 (LC3) (Abcam) followed by staining with Alexa488-conjugated goat anti-mouse IgG antibodies. Next, cells were analysed with FACSCalibur and data were analysed with the FlowJo software (ver. 7.6.5; Tree Star, Ashland, OR, USA).
Statistical analyses. Significant differences between groups were examined statistically using Student's t-test and one-way ANOVA. A P value < 0.05 was considered to indicate statistical significance (***P < 0.001, **P < 0.01, and *P < 0.05).