Introduction
Heat stress, which is caused by a rise in body temperature above its fixed set point, is an important factor that disrupts reproductive performance (Badinga et al., Reference Badinga, Collier, Thatcher and Wilcox1985; Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b), compromising the functioning of germ cells and the viability of early stage embryos (Hansen, Reference Hansen2009). HS has deleterious effects on the internal organelles, on oocyte maturation rate and fertilization capability, and eventually on embryonic development (Payton et al., Reference Payton, Romar, Coy, Saxton, Lawrence and Edwards2004; Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b). Our previous research (Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b) demonstrated that both short-term and long-term exposure of oocytes to heat stress led to aberrant nuclear maturation, a process dependent upon extensive rearranging of the cytoskeleton (Gallicano, Reference Gallicano2001; Shin & Kim, Reference Shin and Kim2003), which is composed of microtubules and microfilament, that were altered in mature bovine oocytes and 2-cell embryos exposed to temperatures of 41–41.5°C (Tseng et al., Reference Tseng, Chen, Chou, Yeh and Ju2004; Rivera et al., Reference Rivera, Kelley, Erdos and Hansen2004). Stressed cells release heat shock proteins (HSPs), which facilitate correct folding of newly synthesized proteins and help proteins to cross cellular membranes (Kang et al., Reference Kang, Ostermann, Shilling, Neupert, Craig and Pfanner1990). The effects of heat stress are mostly detrimental to oocytes during the first 10–12 h of maturation (Putney et al., Reference Putney, Mullins, Thatcher, Drost and Gross1989) at the time when cumulus cells are intimately associated with the oocyte. Gap junction intercellular communication (GJIC) or connexin 43 (Cx43) between cells can regulate cell death, proliferation and differentiation (Russell & Robker, Reference Russell and Robker2007). Furthermore, oocyte and embryo development is associated with DNA methylation, an epigenetic mechanism for differentially marking the parental alleles of imprinted genes (Denomme & Mann, Reference Denomme and Mann2012). It occurs at cytosine residues, mainly within CpG dinucleotides, and is catalyzed by a family of DNA methyltransferases (DNMT). DNMT1 also plays a crucial role in mammalian preimplantation embryo development (Golding & Westhusin, Reference Golding and Westhusin2003).
In previous research, gene expression analysis of embryos developed in the summer as opposed to in the winter developed aberrant expression patterns in DNMT1, Cx43, GDF9, POU5F1 and HSPA14 genes (Gendelman et al., Reference Gendelman, Aroyo, Yavin and Roth2010; Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016). A similar pattern was also observed in the embryos that were developed from heat stressed oocytes, providing evidence that the development and growth of embryos depend on maternal mRNA and proteins stored in the oocytes. The current experiment consolidates our previous study (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016), in which gene expression of the targeted gene was investigated to evaluate the kinetics of heat-shocked oocytes and of oocytes matured in the summer or in the winter.
Materials and Methods
Experimental design
For the first experiment, 25 cumulus–oocyte complexes (COCs) per replication, totalling 100 COCs/season, were collected in the summer (June, July, August, and September) and in the winter (December, January, February and March) and were matured and stored for later genetic analysis. Mean Temperature Humidity Index (THI) in Terceira Island, Azores, Portugal (38°43′N, 27°12′W) was 71.7 ± 0.7 in the summer and 62.8 ± 0.2 in the winter months, respectively. For the second experiment, 500 COCs collected during the winter months were divided into five groups of 100 COCs (25 COCs per replication): a control group which COCs matured in vitro (IVM) at 38.5°C for 24 h and four HS groups COCs in which IVM was performed at 39.5°C for 6, 12, 18 or 24 h, respectively, and placed back into control temperature (38.5°C) for completing the 24 h maturation period (except for the 24 h HS group). COCs matured at different heat exposure times were washed in DEPC water and stored at −80°C for further analysis.
Oocytes collection and in vitro maturation
All the oocyte collection and IVM procedures were performed as previously described by Pavani et al. (Reference Pavani, Baron, Faheem, Chaveiro and Moreira da Silva2015b). Briefly, COCs were collected by aspiration from antral follicles and the ones covered by at least four layers of compacted cumulus cells and evenly granulated ooplasm were washed twice in Tissue Culture Medium 199 (TCM-199), supplemented with 2% fetal bovine serum (FBS), 0.3 mg/ml glutamine and 50 μg/ml gentamycin. The COCs collected in the summer and the winter months of the first experiment were stored, and further subjected to genetic analysis. Similarly, COCs collected during the winter months and matured at HS (39.5°C) at the different exposure times were also stored for gene expression analysis.
Total RNA extraction
All matured oocytes were divided according to their heat shock treatment and seasonal differences, were washed three times with diethylpyrocarbonate (DEPC) treated water and stored as pools of 25 COCs at −80˚C. Total RNA was extracted as previously described by Pavani et al. (Reference Pavani, Baron, Faheem, Chaveiro and Moreira da Silva2015a).
Single-strand cDNA synthesis, specific primer design and standard curves
Single-strand cDNA synthesis, primer design and standard curves were performed accordingly to Pavani et al. (Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016). Specific primers for four chosen genes and two reference genes are shown in the Table 1.
Oligonucleotide primers used for quantitative real-time polymerase chain reactions (qPCR)
Quantitative real-time polymerase chain reaction (qPCR)
Real-time PCRs were performed as previously described for embryos (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016). Briefly, the template cDNA corresponded to oocytes samples from summer and winter months, control and the four HS treatment groups. A non-template control was also included. The amplification program included a pre-incubation step at the holding stage (50°C for 2 min and 95°C for 10 min), followed by 40 three-step amplification cycles consisting of a 15 s denaturation at 95°C, 1 min at 54°C and 30 s at 95°C. A melting curve for the verification of amplification product specificity was recorded at the final dissociation stage to confirm that there was no contamination from primer dimmers. The quantification was carried out using the comparative cycle threshold (Ct) method, with the results expressed in relation to endogenous reference genes and a control group. In this experiment, the reference genes were GAPDH and SDHA, and the targeted gene were Cx43, HSPA14, CDH1, and DNMT1.
Data analysis
Data analysis was performed as described by Faheem et al. (Reference Faheem, Baron, Carvalhais, Chaveiro, Pavani and Moreira da Silva2014) and Pavani et al. (Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016) for embryos. For comparisons between genes, data from GV stage COC's collected during winter were used as a calibrator group for relative gene expression analysis. The mRNAs levels of the Cx43, CDH1, DNMT1 and HSPA14 genes were compared between the summer and winter matured oocytes and exposure time (control and 6 h, 12 h, 18 h, 24 h HS samples) using GAPDH and SDHA as reference genes for normalization.
Results
Assay 1: In vitro maturation effect on gene expression during summer and winter months
The analysis of gene quantification of matured COCs during summer as compared with winter showed a significant (P < 0.05) down-regulation in HSPA14 with respect to reference genes GAPDH and SDHA (Fig. 1). As for the CDH1, DNMT1, and Cx43 genes, no difference was observed when compared with data collected for the winter months (P > 0.05), which indicates no seasonal effect on gene expression.
Gene expression transcripts of in vitro bovine oocytes in warm months compared with cold months. Columns with (*) represent significant differences (P < 0.05) within the gene. Bars indicate standard error of the mean. Mean THI (Temperature Humidity Index) in warmer months (June, July, August and September) was 71.7 ± 0.7 and in cold months (December, January, February and March) was 62.8 ± 0.2.
Assay 2: Effect of heat shock on gene expression of in vitro matured oocytes
After examining gene expression in samples exposed to different heat shock times (Fig. 2), it was found that prolonged heat stress altered gene regulation. No difference in gene expression was seen (P > 0.05) in heat shock oocytes at 6 h with respect to the control group (Fig. 2 a). Samples exposed to 12 h to 24 h exhibited a constant up-regulation of CDH1 genes (P < 0.05) (Fig. 2b–d). An up-regulation (P < 0.05) of the DNMT1 gene was identified in samples exposed to 24 h of heat stress (Fig. 2 d). No altered expression was observed in HSPA14 and Cx43 genes for any of the HS treatments.
Gene expression transcripts of in vitro bovine oocytes matured under different kinetic heat stress (39.5°C) exposure times with reference to the oocytes matured at control (38.5°C): (a) Control versus 6 h. (b) Control versus 12 h. (c) Control versus 18 h. (d) Control versus 24 h. Columns with (*) represent significant differences (P < 0.05) within the gene. Bars indicate standard error of the mean.
Discussion
Environmental stress induces an alteration in the ovarian pool of follicles and their oocytes, which seems to be a cause of reduced fertility in dairy cows during the warmest months of the year (Roth et al., Reference Roth, Inbar and Arav2008). A short-term, mild heat shock enhances the thermotolerance of many organisms (Craig & Gross, Reference Craig and Gross1991; Georgopoulos & Welch, Reference Georgopoulos and Welch1993) including mammalian embryos (Ryan et al., Reference Ryan, Blakewood, Lynn, Munyakazi and Godke1992; Arechiga et al., Reference Arechiga, Ealy and Hansen1995). However, prolonged heat exposure may be detrimental to oocyte and embryo viability. In previous experiments (Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b, Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016) we found that prolonged heat stress did affect oocyte maturation and blastocyst development. The present experiment confirms that embryos developed from oocytes under maternal heat stress (obtained in summer months) had alterations in some of the targeted developmental genes, in agreement with previous observations that prolonged heat shock of bovine oocytes at 39.5°C and 40.5°C led to slow embryo development and gene alteration (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016).
No difference on gene transcription was seen between summer and winter for CX43, CDH1, DNMT1, but HSPA14 had a significantly lower expression in summer. This could explain the lower oocyte maturation observed in the summer (44.3%) compared with winter (78.4%) (Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b), considering that high expression of HSPA14 results in augmented proteins synthesis that may aid oocyte survival (Zhang et al., Reference Zhang, Peñagaricano, Driver, Chen and Khatib2011).
We had previously demonstrated that oocyte maturation rate decreased significantly due to high temperature and exposure time (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016). The present study evaluated gene regulation of kinetic heat stressed oocytes. As no gene expression differences were observed between control and the 6 h HS samples, it can be assumed that oocytes are resistant to a 1°C rise from the control temperature for a 6 h period. Edwards & Hansen (Reference Edwards and Hansen1997) showed that oocytes matured at 41°C for the first 12 h had a reduction in blastocyst rate of 30–65%. Our results support these findings, as after heat exposure at 39.5°C, oocytes exhibited up-regulation of CDH1 across the 12 h, 18 h, and 24 h exposure groups accompanied by low maturation rates (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016). CDH1 is an important tumour suppressor gene that plays a role in suppressing migration and invasion of key transmembrane proteins essential for the maintenance of intercellular junctions in epithelial tissues (Pfeiffer et al., Reference Pfeiffer, Slatkowski, Paudel, Balbach, Baeumer, Crosetto, Drexler, Fuellen and Boiani2011). It also functions as a regulator of cell proliferation by participating in CDK inhibitor-dependent G1 arrest and DNA damage-induced G2 arrest (Sudo et al., Reference Sudo, Ota, Kotani, Nakao, Takami, Takeda and Saya2001). This function can explain the constant up-regulation of CDH1 as an attempt to protect oocytes from apoptosis, because CDH1 is up-regulated when there are signs of DNA damage due to heat stress (Eguren et al., Reference Eguren, Porlan, Manchado, García-Higuera, Cañamero, Fariñas and Malumbres2013) and is associated with the protection of cells under different stress conditions. HSPA14 remained unexpressed possibly due to its minor effect on heat stressed bovine oocytes. Similarly, no gene regulation of Cx43 was observed in response to heat stress in all heat exposure time groups.
Up-regulation in DNMT1 was only observed after 24 h of heat shock exposure, which can be explained by the fact that its up-regulation is associated with the chromatin of metaphase II stage (Cirio et al., Reference Cirio, Ratnam and Ding2008; Kurihara et al., 2008) which appears only towards the end of the maturation period (24 h). Furthermore, DNMT1 alone is sufficient to maintain the methylation of the imprinted genes (Hirasawa et al., Reference Hirasawa, Chiba and Kaneda2008), and prolonged heat stress could cause DNA damage, which is typically associated with CDH1 and DNMT1. The higher expression of CDH1 along with DNMT1 is also an indicator of apoptosis (Palii et al., Reference Palii, Van Emburgh, Sankpal, Brown and Robertson2008).
Conclusion
The combination of both experiments provided a better understanding of the effects of heat stress on oocyte maturation and gene expression. The maturation rates obtained from our previous work (Pavani et al., Reference Pavani, Carvalhais, Faheem, Chaveiro, Reis and Moreira da Silva2015b) were 78.4 ± 8.0% versus 44.3 ± 8.1% (P < 0.001) for colder and warmer months, respectively. Comparing those maturation rates with the gene expression analysis carried out, a parallelism is evident between the lower maturation rates obtained during summer months and the altered expression of HSPA14. The gene quantification of in vitro matured oocytes subjected to HS at different temperatures provided additional evidence that the low maturation rate starting at 12 h of heat shock previously detected (Pavani et al., Reference Pavani, Baron, Correia, Lourenço, Bettencourt, Sousa and Moreira da Silva2016), may be due to the constant up-regulation of CDH1 starting at 12 h and an up-regulation of DNMT1 in samples exposed to 24 h of HS. Overall, the gene regulation analysis of CDH1 and DNMT1 supports the view that after prolonged heat shock, in vitro matured oocytes could experience apoptosis. It is therefore logical to assume that such heat stress conditions could have deleterious effects on oocyte growth, protein synthesis, or the formation of transcripts, all of which are required for subsequent embryo development.
Acknowledgements
Ms Meredith Cannella is fully acknowledged for her English correction and documentation of the manuscript.
Financial support
Project supported by the Azorean Agency for Science and Technology, Grant BD M3.1.2/F/044/2011. CITA-A is also fully acknowledged.
Conflict of interest
None of the authors has any conflict of interest to declare.
