Introduction
Although molecular studies are now considered especially important (Bhojwani et al., Reference Bhojwani, Rudolph, Kanitz, Zuehlke, Schneider and Tomek2006; Fair et al., Reference Fair, Carter, Park, Evans and Lonergan2007; Evsikov & Marin de Evsikova, Reference Evsikov and Marin de Evsikova2009; Siemer et al., Reference Siemer, Smiljakovic, Bhojwani, Leiding, Kanitz, Kubelka and Tomek2009) classical concepts concerning the maturation of mammalian oocytes (Szollosi, Reference Szollosi1975b; Thibault et al., Reference Thibault, Szollosi and Gerard1987) still focus on two aspects: nuclear maturation and cytoplasmic maturation (Sirard, Reference Sirard2001; Ferreira et al., Reference Ferreira, Vireque, Adona, Meirelles, Ferriani and Navarro2009) that are usually highly coordinated (Eppig, Reference Eppig1996). Nuclear maturation involves the transition from a germinal vesicle nucleus to a second metaphase arrangement of the chromosomes and formation of a first polar body by the time of ovulation in most species so far studied. Cytoplasmic maturation is expressed as changes in protein composition, but most conspicuously in the redistribution of organelles that are termed the cortical granules (CG) (Szollosi, Reference Szollosi1962, Reference Szollosi1967). Such granules are, in fact, small vesicles that contain enzymes. During resumption of meiosis, the CG migrate from the Golgi apparatus to close to the vitelline surface, assuming a position 0.4–0.6 μm below the plasma membrane (Ducibella & Buetow, Reference Ducibella and Buetow1994). Only when situated just beneath the plasma membrane can they undergo exocytosis by fusing with the egg membrane. This fusion enables release of the CG contents into the perivitelline space, an important step in membranous maturation and in instigating a block to polyspermy (Szollosi, Reference Szollosi1967; Hosoe & Shioya, Reference Hosoe and Shioya1997; Wang et al., Reference Wang, Hosoe, Li and Shioya1997).
There have been diverse studies that focussed on the formation and distribution of the CG in mammalian oocytes (Szollosi, Reference Szollosi1967, Reference Szollosi, Gerard, Menezo and Thibault1978; Flechon, Reference Flechon1970; Thibault et al., Reference Thibault, Szollosi and Gerard1987; Hosoe & Shioya, Reference Hosoe and Shioya1997; Wang et al., Reference Wang, Hosoe, Li and Shioya1997; Wessel et al., Reference Wessel, Brooks, Green, Haley, Voronina, Wong, Zaydfudim and Conner2001, Reference Wessel, Conner and Berg2002) but none in the present context. In this study on maturation of bovine oocytes, the influence of heat stress on both nuclear stage and distribution of cortical granules has been evaluated in each oocyte. As heat stress seems to induce premature aging of oocytes (Lawrence et al., Reference Lawrence, Payton, Godkin, Saxton, Schrick and Edwards2004; Edwards et al., Reference Edwards, Saxton, Lawrence, Payton and Dunlap2005; Schrock et al., Reference Schrock, Saxton, Schrick and Edwards2007) the effect of overmaturation on nuclear and cortical granules distribution features was also evaluated. This research has relevance to bovine reproduction as heat stress has been associated with reduced fertility in many countries (de Rensis & Scaramuzzi, Reference Rensis and Scaramuzzi2003; López-Gatius, Reference Lopez-Gatius2003) especially when it coincides with the insemination time (Putney et al., Reference Putney, Drost and Thatcher1989; García-Ispierto et al., Reference Garcia-Ispierto, Lopez-Gatius, Bech-Sabat, Santolaria, Yaniz, Nogareda, De Rensis and Lopez-Bejar2007).
Materials and Methods
Experiment design
Three groups were established: (1) control oocytes (CO; n = 75), maintained under culture conditions (38.5°C for 22 h); (2) heat stressed oocytes (HSO; n = 80) submitted to the heat treatment; and (3) overmaturated oocytes (OMO; n = 20) cultured for 28 h (Fig. 1).
A time-line figure illustrating the experimental setup for cumulus–oocyte complexes (COCs) culture and groups of study: CO (control oocytes; n = 75), HSO (heat stress oocytes; n = 80) and OMO (overmaturation oocytes; n = 20).
As a 3 h heat treatment mimics hyperthermia real conditions (Tseng et al., Reference Tseng, Chen, Chou, Yeh and Ju2004) of the 22 h of the in vitro maturation (IVM) process, the HSO group was exposed to 41.5°C during the period of 18 to 21 h of maturation.
Chemicals and reagents
All chemicals were purchased from Sigma unless otherwise indicated.
Collection of oocytes
Ovaries recovered from slaughterhoused young heifers and placed into Dulbecco's phosphate buffered saline solution (PBS; DPBS 10×; GIBCO-Invitrogen) including 1% (v/v) antibiotic antimycotic solution (AA; 10,000 units/ml penicillin, 10 mg/ml streptomycin and 25 mg/ml amphotericin B) and carried to the laboratory at room temperature. Ovaries were then washed twice in warm sterile PBS and were kept at 37.5°C until ovarian puncture that took place within 2 h from their recovery. Next, 2 to 8 mm-sized follicles were aspirated using an 18-gauge needle joined to a 5ml syringe. Cumulus–oocyte complexes (COCs) were obtained and placed into working medium (WM; Medium 199 with Earle's salts 25 mM HEPES and NaHCO3, 1% v/v AA solution). Only oocytes enclosed in three or more layers of compact cumulus cells and presenting a homogeneous and translucent ooplasm were selected.
In vitro maturation
Selected COCs were washed twice in WM and once in maturation medium (MM; Medium 199 with Earle's salts, l-glutamine and NaHCO3, supplemented with 20 μg/ml epidermal growth factor, 2 mM sodium pyruvate and 1% v/v AA solution) that had been pre-equilibrated for 3 h at 38.5°C in 5% (v/v) CO2 in humidified air.
After washing, COCs were randomly allocated to groups of 20–25 and placed into 4-well dishes (Nunc) containing 500 μl of MM. All the processes were performed in a laminar flux booth during approximately 2 h from follicular aspiration to the entrance of COCs to IVM.
COCs were cultured at different temperature conditions for 22 h (CO and HSO groups) or 28 h (OMO group) according to the experimental design in an atmosphere of 5% (v/v) CO2 in humidified air. To reduce variation of temperature during culture, incubation of control and overmaturation groups and heat stress group COCs were performed in two different CO2 incubators. While one incubator was set at 38.5°C, the other was used for the heat treatment. Both incubators were kept closed during the whole maturation period and their temperature was verified by checking of clinical thermometers that were placed inside of the incubators.
Zona pellucida digestion and oocyte fixation
After 22 h (CO and HSO groups) or 28 h (OMO group) of IVM, COCs were removed from maturation wells and denuded of cumulus cells by pipetting and washed twice in PBS with 0.05% (w/v) of bovine serum albumin (BSA; fraction V BSA). Oocytes were then immersed in PBS containing 0.4% (w/v) pronase for 3 min to dissolve zona pellucida and washed five times in PBS with 0.05% (w/v) BSA. Oocytes were fixed in a PBS solution containing 4% (w/v) of paraformaldehyde for 45 min at room temperature and washed five times in PBS with 0.05% (w/v) BSA.
Oocyte permeabilization and cortical granules staining
Oocytes were immersed in a permeabilizing solution of PBS containing 0.3% Triton X100 and 0.05% (w/v) BSA for 5 min at room temperature and washed five times in PBS with 0.05% BSA. Oocytes were then incubated in the dark for 30 min at room temperature into a staining solution of PBS containing 100 μg/ml of fluorescein isothiocyanate-labelled Lens culinaris agglutinin (FITC–LCA) and 0.05% BSA.
Nuclear staining and mounting
Stained oocytes were thoroughly washed in PBS with 0.05% (w/v) BSA to remove excess of FITC–LCA before mounting, and mounted between a pretreated with poly-l-lysine coverslip and a glass slide supported by a washer. The antifade mounting medium contained 4,6-diamidino-2-phenilidole (DAPI; Vectashield; Vector laboratories, Inc.) for counterstaining DNA. The coverslip was sealed with nail polish and preparations were kept at 4°C and protected from light until examination by fluorescence and laser confocal microscopy.
Nuclear and cytoplasmic maturation evaluation
Nuclear and cytoplasmic maturation status was evaluated for each oocyte. Nuclear content and apical and equatorial section of each oocyte were photographed under UV epifluorescent microscope (Nikon Eclipse TE 2000S) and laser confocal microscope (Leica TCS SP2) respectively.
Cultured oocytes were checked to have reached metaphase II (MII) or not. Normality in MII morphology was furthermore registered according to the modified classification of Tseng (2004) (normal MII: uniform alignment of the chromosomes on the spindle; anomalous MII: nuclear content changed into chromatin-like structure forming condensed aggregates or forming aberrantly distributed chromosomes).
Translocation of CG to the oolemma was used as an indicator of cytoplasmic maturation (Damiani et al., Reference Damiani, Fissore, Cibelli, Long, Balise, Robl and Duby1996). The distribution of GC was classified into four patterns according the classification of Hosoe & Shioya (Reference Hosoe and Shioya1997) (pattern I: GCs distributed in clusters or large aggregates; pattern II: GC individually dispersed and partially clustered or aggregated; pattern III: GC completely dispersed; pattern IV: no CG).
Statistical analysis
Only oocytes that reached MII nuclear stage were included in the statistical analysis. The following data were recorded for each oocyte: replicate (1–4), group (CO, HSO or OMO), MII morphology (normal MII versus anomalous MII) and CG distribution pattern (I, II, III or IV).
On data from each oocyte, a first logistic regression analysis was performed using MII anomalous morphology as the dependent variable (0 or 1) and replicate, group and CG distribution pattern as independent factors. A second logistic regression analysis was performed using CG distribution pattern IV as dependent variable (0 or 1) and replicate, group and MII morphology as independent factors. A further logistic regression analysis was performed using anomalous oocyte maturation, considering both MII anomalous morphology and GC distribution pattern IV as the dependent variable (0 or 1) and replicate and group as independent factors. All variables above were considered as class variables.
Logistic regressions analyses were performed using the SPSS package, version 17.0 (SPSS Inc.) according to the method of Hosmer & Lemeshow (Reference Hosmer and Lemeshow1987). Basically, this method involves five steps as follows: (1) preliminary screening of all variables for univariate associations; (2) construction of a full model using all the variables found to be significant in the univariate analysis; (3) stepwise removal of non-significant variables from the full model and comparison of the reduced model with the previous model for model fit and confounding; (4) evaluation of plausible two ways interactions among variables; and (5) assessment of model fit using Hosmer–Lemeshow statistics. Variables with univariate associations showing p-values <0.25 were included in the initial model. We continued modelling until all the main effects or interaction terms were significant according to the Wald statistic at p < 0.05.
Results
Effects of heat stress on nuclear maturation
Of the 175 cultured and evaluated oocytes, four showed immature nuclear stages after IVM and were excluded from the study.
Logistic regression analysis indicated no significant effect of replicate or CG distribution pattern on MII morphology. Table 1 shows the adjusted odds ratios of the variable finally included in the logistic model. No significant interactions were found. Based on the odds ratio, oocytes in HSO and OMO groups were, respectively, 14.5 and 5.4 times more likely to showed anomalous MII morphology after the IVM than those matured under control conditions.
Odds ratios of variables included in the final logistic regression model for metaphase II (MII) anomalous morphology.
Likelihood ratio test 48.85; 2 d.f., p < 0.0001. Nagelkerke r 2 = 0.337.
aConfidence interval for the odds ratio.
Effects of heat stress on cortical granule distribution
Figure 2 shows laser confocal microscopic images of the equatorial section of FITC–LCA-labelled bovine oocytes showing representative CG distribution patterns and the number of oocytes for each pattern and group of study.
Laser confocal microscopic images of the equatorial section (A) and its detail (B) of fluorescein isothiocyanate Lens culinaris agglutinin (FITC–LCA)-labelled bovine oocytes showing representative patterns of CG distribution for the different groups: pattern II (1A, 1B: CO = 14/73, HSO = 12/78, OMO = 1/20), pattern III (2A, 2B: CO = 54/73, HSO = 45/78, OMO = 12/20), pattern IV (3A, 3B: CO = 14/73, HSO = 21/78, OMO = 7/20). CO, control oocytes; HSO, heat stress oocytes; OMO, overmaturation oocytes.
Logistic regression analysis indicated no significant effect of replicate or MII morphology on CG distribution pattern IV. Table 2 shows the adjusted odds ratios of the variable finally included in the logistic model. No significant interactions were found. Based on the odds ratio, the likelihood for an oocyte of showing CG distribution pattern IV was 6.3 and 9.3 times higher for HSO and OMO groups, respectively, than for the CO group.
Odds ratios of variables included in the final logistic regression model for cortical granules (CG) distribution pattern IV.
Likelihood ratio test 17.88; 2 d.f., p < 0.0001. Nagelkerke r 2 = 0.154.
aConfidence interval for the odds ratio.
Effects of heat stress on oocyte maturation considering both nuclear and cytoplasmic maturation
Logistic regression analysis indicated no significant effect of replicate on anomalous maturation considering both MII anomalous morphology and CG distribution pattern IV. Table 3 shows the adjusted odds ratios of the variable finally included in the logistic model. No significant interactions were found. Based on the odds ratios, the risk of undergo anomalous oocyte maturation was 17.1 and 18 times greater in oocytes cultured in HSO and OMO groups, respectively, than those in CO group.
Odds ratios of variables included in the final logistic regression model for anomalous oocyte maturation, considering both anomalous patterns of metaphase II (MII) morphology and cortical granules (CG) distribution pattern.
Likelihood ratio test 16.45; 2 d.f., p < 0.0001. Nagelkerke r 2 = 0.178.
aConfidence interval for the odds ratio.
Discussion
Heat stress proved to be valuable in aging oocytes. For this purpose nuclear and cytoplasmic maturation for each oocyte were assessed simultaneously. Similar figures of aged oocytes (Szollosi, Reference Szollosi1971, Reference Szollosi1974, Reference Szollosi1975a) were registered for heat stressed and overmaturated oocytes. By using logistic regression procedures, the models were highly explanatory and significant. Three consecutive analyses showed how dramatically heat stress advanced age for nuclear and cytoplasmic processes. Including both nuclear and cytoplasmic maturation measurements, and based on the odds ratio, heat stressed and overmaturated oocytes were 17 and 18 times, respectively, more likely to age than controls, used as reference.
Heat stress during the last period of IVM resulted in a high percentage of oocytes showing a pattern IV of CG distribution. Pattern IV, which is characterized by the loss of the continuous layer of CG, was firstly described in bovine oocytes by Hosoe Shioya (1997) and is the most common pattern of CG distribution for oocytes that had underwent exocytosis after insemination or activation. Although heat stress advanced oocyte maturation in previous studies (Payton et al., Reference Payton, Romar, Coy, Saxton, Lawrence and Edwards2004; Edwards et al., Reference Edwards, Saxton, Lawrence, Payton and Dunlap2005), pattern IV of CG distribution related to the final maturation of oocyte was not described. We describe herein how heat treatment or overmaturation can result in a pattern IV of CG distribution in a large number of oocytes. These results suggest that heat stress applied at the end of the period of oocyte maturation can induce CG loss.
As observed in the present study, an anomalous progression to second meiotic metaphase was a feature of heat stressed oocytes. A modified behaviour of the cytoskeleton may be the underlying cause, as it has been described in previous studies (Tseng et al., Reference Tseng, Chen, Chou, Yeh and Ju2004; Ju et al., Reference Ju, Jiang, Tseng, Parks and Yang2005; Roth & Hansen, Reference Roth and Hansen2005). Nonetheless, other aspects of nuclear behaviour may be modified by temperature stress, such as the rate and normality of chromosomal condensation (Roti, Reference Roti2008). The behaviour of contractile proteins of the cytoskeleton will be perturbed by inappropriate temperature. Aspects that remain to be clarified are: (1) how much variation in temperature an oocyte can withstand; and (2) whether tolerance to shifts in temperature changes with the stage of meiotic maturation.
Overall, there is the question as to whether tolerance to temperature modifications in vitro differs from that in the living animal. There is already a body of evidence indicating that the temperature of preovulatory Graafian follicles is lower than temperature elsewhere in the ovary of rabbits (Grinsted et al., Reference Grinsted, Blendstrup, Andreasen and Byskov1980), women (Grinsted et al., Reference Grinsted, Kjer, Blendstrup and Pedersen1985), pigs (Hunter et al., Reference Hunter, Grondahl, Greve and Schmidt1997, Reference Hunter, Bogh, Einer-Jensen, Muller and Greve2000, Reference Hunter, Einer-Jensen and Greve2006) and perhaps cattle (Greve et al., Reference Greve, Grondahl and Schmidt1996). This finding suggests the existence of sensitive systems of temperature regulation in vivo that may be difficult to mimic in vitro.
A further experiment that would be relevant would be to culture individual Graafian follicles (Baker & Neal, Reference Baker, Neal, Biggers and Schuetz1972, Reference Baker and Neal1974; Picton et al., Reference Picton, Harris, Muruvi and Chambers2008) under different protocols of heat stress. Such a study might reveal greater perturbations in oocyte maturation when within a heat stressed follicle.
Acknowledgements
This study was supported by Spanish Ministry of Science and Innovation, MICINN, Ref. RTA2008–00070-C02–02. Andreu-Vázquez was supported by an FPU Grant from the Spanish Ministry of Science and Education, MCE, Ref.AP2007-01598
