Mol. Cells, Vol. 13, No. 2, pp. 272-280
M olecules and Cells KSMCB 2002
Amelioration of Mitochondrial Dysfunction and Apoptosis of Two-Cell Mouse Embryos after Freezing and Thawing by the High Frequency Liquid Nitrogen Infusion In Pyo Sohn1, Hak Joon Ahn2, Dong Wook Park2, Myung Chan Gye3, Do Hyun Jo2, Suk Young Kim4, Churl Ki Min5, and Hyuck Chan Kwon* Department of Obstetrics and Gynecology, Eulji Medical Center, Eulji University School of Medicine, Seoul 139-711, Korea; 1 Department of Biology, Kyonggi University, Suwon 442-760, Korea; 2 Department of Molecular Sciences and Technology, Ajou University, Suwon 442-749, Korea; 3 Department of Life Science, College of Natural Sciences, Hanyang University, Seoul 133-791, Korea; 4 Department of Obstetrics and Gynecology, Ghil Medical Center, Gachon Medical School, Inchon 406-203, Korea; 5 Department of Biological Sciences, Ajou University, Suwon 442-749, Korea. (Received October 16, 2001; Accepted December 7, 2001)
Liquid nitrogen (LN2) infusions are currently used in a slow controlled-rate freezing during cryopreservation. The effects of two different LN2 infusion frequencies (conventional, slow 50 infusions/min and high 120 infusions/min) were studied with frozen-thawed two-cell mouse embryos and their subsequent development to blastocysts. The embryos that were subjected to the high frequency LN2 infusion (HFLI) showed a significantly higher survival rate over the low frequency LN2 infusion (LFLI) (50.7 vs. 34.6%, P < 0.05). The blastocyst formation was also higher in HFLI (76.7%) than LFLI (44.0%, P < 0.05) with respective to the number of cells in a blastocyst of 71.6 ± 8.0 (n = 20) and 62.5 ± 4.7 (n = 20) (P < 0.05). The relative amount of H2O2 in an embryo that was assessed by a fluorescence intensity of 2',7'-dichlorofluorecein (DCF) showed a difference between the procedures with 16.6 ± 1.6 (n = 21) and 23.4 ± 1.8 (n = 24) for HFLI and LFLI, respectively (P < 0.05). Mitochondrial staining by Rhodamine 123 showed that the number and distribution of viable mitochondria were similar in both procedures, but fewer mitochondria were observed with a marked aggregation in the arrested embryos, indicating a mitochondrial disintegration. The mitochondrial membrane potential was visualized by a membrane potential-sensitive fluorescent probe, 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbocyanine iodide (JC-1). There was a decrease in
Cryopreservation of oocytes and embryos has been widely used in reproductive medicine and animal breeding (Al-Hasani et al., 1987; Bernard et al., 1992; Chen, 1986; Fuku et al., 1992). The embryo cryopreservation is largely achieved by a slow freezing to −30 to −70°C in
* To whom correspondence should be addressed. Tel: 82-2-970-8717; Fax: 82-2-970-8230 E-mail:
[email protected]
Abbreviations: HFLI, high frequency LN2 infusion; IE, intact embryo; LFLI, low frequency LN2 infusion; LN2, liquid nitrogen.
Demo
the number of mitochondria that had a high membrane potential, and they showed a peripheral redistribution along the cell membrane in LFLI. A fluorescent staining of the actin filaments revealed a discontinuity that was noticeably at the peripheral “actin band” in LFLI. The DNA fragmentation was assessed by the dUTP nick end-labeling (TUNEL). The results showed a higher DNA fragmentation of blastocyst nuclei in LFLI compared to HFLI (65.6 vs. 36.0%, P < 0.05). Based on these observations, it was concluded that HFLI was better than LFLI in the case of freezing the mouse 2-cell embryos for preserving cytoskeletons and mitochondrial integrities. This could subsequently lead to a higher survival and developmental rate of the cryopreserved mouse embryos. Keywords: Actin Filament; Cryopreservation; DNA Fragmentation; Liquid Nitrogen Infusion Frequency; Mitochondrial Membrane Potential.
Introduction
In Pyo Sohn et al.
the presence of appropriate cryoprotectants before plunging into liquid nitrogen (LN2) (−196°C). Subsequent thawing has also been carefully examined to maximize the post-thaw survival of embryos. Despite a great deal of effort, freezing and thawing still causes drastic morphological and biochemical alterations. These include polyploid formations (Balakier et al., 2000), ionic disturbances (Lane et al., 2000), damages to membranes (James et al., 1999), enzyme inactivations (Peluso et al., 2001), cellular organelle dysfunctions (Saunders and Parks, 1999; Stojkovic et al., 2001), and cytoplasm disorganizations (Fuku et al., 1995). Such detrimental effects may lead to the death of embryos, which, in turn, may decrease the implantation and development capabilities of embryos (Chedid et al., 1992; Karlsson et al., 1996; Uechi et al., 1999). A number of different cryopreservation protocols have been utilized in order to maximize the post-thaw survival in various cells and tissues (Van den Abbeel et al., 1997). The most commonly employed protocol involves an interrupted slow-freezing procedure (Sommerfeld and Niemann, 1999). In this procedure, a programmable, controlled-rate freezer is used; its cooling rate is controlled by the frequency and duration of the LN2 infusion. During the slow, interrupted cooling period, ice crystals are formed as a result of the liquid/solid phase transition in the extracellular milieu. This inevitably results in a thermal fluctuation that is due to supercooling and subsequent exotherm, which is a result of the dissipation of the latent heat of fusion as ice is formed (Woelders et al., 1997). After the supercooling, the temperature will increase to a new equilibrium between the ice and solution at a specified solute concentration before the next cooling resumes. This will provide an opportunity for ice crystal growth until the next stabilizing temperature is reached with a concomitant diffusion of water molecules from smaller ice crystals to larger ones, known as migratory recrystallization (MacFarlane and Forsyth, 1987). The migratory recrystallization increases the probability of strain in the cells due to the presence of ice. In this study, the high frequency LN2 infusion (HFLI, 120 infusions/min) was applied. The results of freezing and thawing were compared with that of conventional low frequency LN2 infusion (LFLI, 50 infusions/min). We hypothesized that migratory recrystallization could be minimized, resulting in a reduction of the temperature fluctuation when the infusion frequency increased. To ascertain this hypothesis, the cytoskeletal preservation and mitochondrial integrity were evaluated by monitoring the following: (1) the distribution of actin filament and active mitochondria, (2) H2O2 production, and (3) mitochondrial membrane potential after the freezing and thawing of the mouse 2-cell embryos. As an additional test of the validity of different infusion frequencies, development of the embryos, and degree of apoptosis were compared.
273
Materials and Methods Embryo isolation The 6−8-week old ICR female mice were superovulated by standard hormonal treatments using an intraperitoneal injection of 7.5 IU of pregnant mare serum gonadotrophin (PMSG; Sigma, USA). This was followed by an intraperitoneal injection of 7.5 IU of human chorionic gonadotrophin (hCG; IBSA, Switzerland) 48 h apart. After mating overnight with 8−10-week-old males of the same strain, the females were inspected for vaginal plugs, and killed by cervical dislocation 48 h after hCG treatment. Embryos at the late two-cell stage were recovered by flushing the oviducts with a mHTF culture medium that consisted of in mM; 2.04 CaCl2, 0.1 EDTA, 97.6 NaCl, 4.7 KCl, 0.2 MgSO4, 25 NaHCO3, 0.33 Na-pyruvate, 1.0 Glutamine, and 10 Na-lactate that was supplemented with 0.4% bovine serum albumin (BSA; Sigma, USA). Two-cell embryos that were considered normal under a microscopic observation were chosen for freezing. The healthy two-cell embryos were randomly assigned to three experimental aliquots. Two aliquots (n = 402−494 each) were used for cryopreservation by two different freezing protocols. The third aliquot (n = 240) was unfrozen and employed as a control. Embryo cryopreservation A slow-rate freezing protocol was employed, according to Lassalle et al. (1985), using two different programmable freezers (Kryo-10, Planer, USA for LFLI; CryoMagic, Booil Industry, Korea for HFLI) with a minor modification. Phosphate-buffered saline (PBS) that was supplemented with 1.5 M 1,2-propanediol (Sigma, USA) and 20% bovine serum albumin (BSA) was used as a freezing solution. Embryos were equilibrated in the freezing solution for 15 min at room temperature. Ten embryos that were placed in 0.25 ml straw were frozen at −2°C/min from room temperature to −7°C before ice seeding was formed with super-cold forceps. After 10 min of equilibration, the ice-seeded embryos were further frozen to −30°C at −0.3°C/min. However, one had 50 LN2 infusions/min (LFLI) and the other had 120 LN2 infusions/min (HFLI) before being stored at −196°C in liquid nitrogen for 7−60 d. A rapid thawing to room temperature (500°C/min) was employed, and the cryoprotectants were eliminated by a serial dilution to a mHTF medium that contained 0.4% BSA. Embryo culture and cell counting After thawing, the two-cell embryos were recovered and examined under a microscope in a cryoprotectant-free medium. Under a microscope, an embryo that had two viable blastomeres was counted as an intact embryo (IE). One that had only one viable blastomere was counted as a partial embryo (PE). One with no viable blastomere was counted as a damaged embryo (DE). Only IEs were cultured in the 0.4% BSA-containing mHTF medium in an incubator (Cellstar 2710; Queue Systems, USA) that was supplied with 5% CO2 and 95% air at 37°C. The osmolarity was maintained at 280−285 mOsm/kg. In addition to the microscopic observation, a staining with propidium iodide (PI) (BD Pharmingen, USA) (Cossarizza et al., 1994) was carried out in order to differentiate
274
Mitochondrial Dysfunction and Apoptosis after Cryopreservation
the viable embryos from the dead embryos. Every 24 h, the developing embryos were evaluated under an inverted microscope (Diaphot 300; Nikon, Japan). After 72 h in culture, the embryos that had reached the blastocyst stage were resuspended in 0.8% sodium citrate, fixed with Carnoy’s solution (acetic acid:ethanol = 1:3) on a slide glass for 24 h, and subjected to staining with 1.0 µg/ml Hoeschst 33258 (Sigma, USA) for 5 min in the dark. The number of cells in a blastocyst was counted under a fluorescence microscope (Epifluorescence BX50; Olympus, Japan). Staining of actin filaments Zona pellucidas were physically removed by fine forceps. Zona pellucida-free cells were washed twice with PBS, and fixed in 3.7% paraformaldehyde in PBS for 30 min. After washing twice with PBS, the cells were incubated in 0.25% Triton X-100 for 10 min and subsequently for 10 min in 0.25% NH4Cl in PBS. The cells were then incubated in FITCphalloidin (Molecular Probes, USA) in PBS at a final concentration of 10 µg/ml for 45 min in the dark. After washing with PBS, the cells were observed in the Bio-Rad MR-AG/2 laser-scanning confocal imaging system (Bio-Rad, USA) that was equipped with a Krypton-Argon laser and interfaced with a Nikon inverted microscope (Diaphot 300; Nikon, Japan). The wavelengths of excitation and emission were 488 nm and 510 nm, respectively. Measurement of mitochondrial viability and membrane potential The presence of viable mitochondria was identified by Rhodamine 123 staining (Garner et al., 1997). The cells were incubated in a Rhodamine 123 solution (Molecular Probes, USA) at the concentration of 10 µg/ml in a mHTF medium at 37°C for 15 min in the dark. After washing with a mHTF medium, the cells were excited at 488 nm and observed at 510 nm in the confocal imaging system. The mitochondrial membrane potential was visualized using 5,5′,6,6′-tetrachloro-1,1′,3,3′tetraethylbenzimidazolyl carbocyanide iodide (JC-1, Molecular Probes, USA) staining (Cossarizza et al., 1994). The cells were rinsed with a mHTF medium and incubated in a mHTF medium that contained 5.0 µg/ml JC-1 for 15 min at 37°C in the dark. After washing with a mHTF medium, the cells were excited at 488 nm and observed at either 510 nm (green mitochondria) or 590 nm (red to orange mitochondria) in the confocal imaging system. Measurement of H2O2 content in an embryo Measurement of H2O2 in an embryo was performed according to Nasr-Esfahani and Johnson (1991) with a minor modification. The H2O2 concentration within the embryos was measured by 2′,7′-dichlorodihydrofluorescein diacetate (DCHFDA; Molecular Probes Inc, Pitchford Ave, Eugene, USA). The principle underlying this procedure is briefly described as follows. Non-ionized DCHFDA is membrane permeant; therefore, it is able to diffuse readily into cells. Within the cell, the acetate groups are hydrolyzed by intracellular esterase activity forming 2′,7′-dichlorodihydrofluorescein (DCHF), which is polar. It is, therefore, trapped within the cell. DCHF fluoresces when it is oxidized by H2O2 or
lipid peroxides to yield 2′,7′-dichlorofluorescein (DCF). The level of DCF that is produced within the cells is related linearly to that of peroxides that are present; therefore, its fluorescent emission at 510 nm after excitation at 480 nm provides a measure of the peroxide levels. Two-cell embryos were incubated in a mHTF medium that was supplemented with 10 µM DCHFDA for 15 min at room temperature. The embryos were then washed with mHTF in order to remove traces of the dye. The fluorescence emission at 510 nm was viewed/quantified in the confocal microscopy. TUNEL measurement of the DNA fragmentation Detection of embryonic apoptosis was performed using terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-digoxigenin nick end labeling (TUNEL) with an in situ cell death detection kit (ApopTag Kit, Oncor Inc., USA) according to the manufacturer’s recommendation. Briefly, the cells in a blastocyst were fixed at room temperature for 10 min in a Carnoy’s solution. They were subsequently incubated in the equilibrium buffer for 5 min before being exposed to terminal deoxynucleotidyl transferase (TdT) and dUTP-digoxigenin for 24 h at 37°C in a humidified chamber. After washing, detection was performed using FITC-conjugated anti-digoxigenin. The cells were then viewed/counted in the confocal microscopy. Statistical analysis All of the experiments were repeated at least five times. A representative fluorescent microscopy was presented where appropriate for qualitative analysis. The quantitative results that were obtained from the cell counts, H2O2 content, mitochondrial membrane potentials, or TUNEL represented the mean ± SEM of the combined data from the replicate experiments. Statistical differences between the mean values were analyzed by using a Chi-square, or Scheffe tests. The level of significance was set at 5%.
Results Cryopreservation was a negative factor for cell viability, regardless of the employed freezing procedures. Of the 402 recovered embryos, the numbers of IE, PE, and DE were 204, 100, and 98, respectively, in HFLI. Of the 494 recovered embryos, the respective numbers of IE, PE, and DE were 171, 154, and 169 in LFLI. The survival rates of the frozen embryos were 50.7% (204/402) for HFLI and 34.6% (171/494) for LFLI (if the complete IEs were taken into account). By comparison, the survival rate of the fresh embryos (the control) was 100% (240/240). A statistical analysis using the Chi-square test showed that the overall embryo survival rate (IE + PE) significantly decreased by freezing the embryos, but independent of the procedures. However, the survival rate (based on IE only) was different between the procedures (P < 0.05) with HFLI showing a higher survival rate (Table 1). The viability of embryos was further confirmed by the PI stain-
In Pyo Sohn et al.
275
Table 1. Survival rates of two-cell mouse embryos after freezing and thawing No. of embryos (%) IE
PE
DE
Total No. of embryos
240 (100) 204 (50.7)* 171 (34.6)
0 100 (24.9) 154 (31.2)
0 98 (24.4) 169 (34.2)
240 402 494
Group Control HFLI LFLI
NA, not applicable; HFLI, high frequency LN2 infusion (120 infusions/min); LFLI, low frequency LN2 infusion (50 infusions/min). IE (intact embryo), an embryo that had all viable blastomeres; PE (partial embryo), an embryo that had one viable and one damaged blastomere; DE (damaged embryo), one without any viable blastomere. * P < 0.0000 vs. LFLI by Chi-square test.
ing in a parallel experiment. There was a good correlation between the microscopic evaluation and PI staining (data not shown). The recovered IEs were cultured in vitro for 72 h to the blastocyst formation. The percentages of the blastocyst formation were 86.7% (150/173) in the control, 76.7% (105/137) in HFLI, and 44.0% (44/100) in LFLI. The respective number of cells in a blastocyst were 79.5 ± 12.9 (n = 21), 71.6 ± 8.0 (n = 20), and 62.5 ± 4.7 (n = 20). The statistical analysis by the Scheffe test showed that the blastocyst formation and the number of cells in a blastocyst were affected by the freezing procedures. When using LFLI, both the percentage of the blastocyst formation and the number of cells were significantly decreased (P < 0.05) (Fig. 1). The relative amount of H2O2 in the individual embryos was measured by a staining with DCHFDA. The fluorescence level of DCF that was produced from DCHFDA by H2O2 oxidation within the cells is related linearly to the H2O2 contents (Nasr-Esfahanai and Johnson, 1991). The relative amount of H2O2 that was assessed by DCF was 15.3 ± 3.0 (n = 20), 16.6 ± 1.6 (n = 21), and 23.4 ± 1.8 (n = 24) in the control, HFLI, and LFLI, respectively (Fig. 2). The difference between HFLI and lFLI was statistically significant by the Scheffe test (P < 0.05). LFLI increased the amount of H2O2 in the frozen-thawed embryo over the control level, whereas HFLI did not. Viable mitochondria in the frozen-thawed two-cell embryos were visualized by Rhodamine 123 staining. The staining results showed that mitochondria were as evenly distributed throughout the cytoplasm as in the control, independent of the freezing protocols (Figs. 3A to 3C). However, in the arrested embryos that showed a concomitant DNA fragmentation, viable mitochondria appeared to be aggregated mainly along the cell membrane (Fig. 3D). The mitochondrial membrane potential was visualized by JC-1 staining. A typical result is shown in Fig. 4. The red to orange color represents a high mitochondrial membrane
Fig. 1. Development of the two-cell mouse embryos after 72 h in culture. The two-cell mouse embryos were cultured for 72 h to blastocyst formation. Nuclei were labeled with Hoeschst 33258, described in Materials and Methods. Blastocysts that were formed from the fresh embryo (control), the frozen-thawed embryo in high frequency LN2 infusion (HFLI), and the frozenthawed embryo in low frequency LN2 infusion (LFLI), as well as the cells in a blastocyst were counted microscopically. The percentages of the blastocyst formation ( ) were: 86.7% in the control, 76.7% in HFLI, and 44.0% in LFLI. The number of cells in a blastocyst (T) were: 79.5 ± 12.9 (n = 21) in the control, 71.6 ± 8.0 (n = 20) in HFLI, and 62.5 ± 4.7 (n = 20) in LFLI. * P
