Altering Intracellular pH Disrupts Development and Cellular Organization in Preimplantation Hamster Embryos

Abstract

In early cleavage stage hamster embryos, the inability to regulate intracellular pH (pH_i) properly is associated with reduced developmental competence in vitro. The disruption of mitochondrial organization is also correlated with reduced development in vitro. To determine the relationship between pH_i and the disruption of cytoplasmic organization, we examined the effects of altering pH_i on hamster embryo development, mitochondrial distribution, and cytoskeletal organization. The weak base trimethylamine was used to increase pH_i and was found to reduce embryo development and disrupt the perinuclear organization of mitochondria. The weak acid 5,5-dimethyl-2,4-oxazolinedione was used to decrease pH_i and was also found to reduce development and disrupt the perinuclear organization of mitochondria. With either treatment, the microfilament organization was perturbed, but the microtubule cytoskeleton was not. However, the temporal progression of the disruption of mitochondrial distribution was more rapid in alkalinized embryos than acidified embryos, as revealed by two-photon imaging of living embryos. Additionally, the disruption of the microfilament network by the two treatments was not identical. The cytoplasmic disruptions observed were not due to acute toxicity of the compounds because embryos recovered developmentally when the treatment compounds were removed. These observations link ionic homeostasis, structural integrity and developmental competence in preimplantation hamster embryos.

INTRODUCTION

Homeostasis of the intracellular ionic environment is necessary to maintain cell function and viability. This includes the control of the concentration of hydrogen ions, i.e., intracellular pH (pH_i), within a limited range. Within this range, small, acute changes in pH_i accompany, and have been implicated in, the regulation of key processes in somatic cells, including metabolism [1], proliferation [2], cell fate determination [3], the activity of regulatory enzymes, and maintenance of gap junctions and the cytoskeleton [4] as well as modulation of calcium levels [5]. Given its multiple roles in somatic cells, it seems likely that pH_i also plays an important role in the regulation of cellular and developmental processes during embryogenesis. Changes in pH_i can be involved in critical transitions in the oocyte’s status, such as during meiotic maturation in the frog [6] and limpet [7] or following oocyte activation and/or fertilization in some invertebrate species [8]. Shifts in pH_i are also implicated in a variety of sperm functions, such as motility [9–11], capacitation [12], and the acrosome reaction [13, 14]. A change of pH_i can trigger a cascade of cellular events that initiate embryonic development. In the sea urchin, a sodium-dependent alkalinization accompanies fertilization and is necessary for initiating or regulating various processes [8], including DNA and protein synthesis [15], sperm aster formation and pronuclear movements [16], and microvilli elongation [17, 18]. Although this alkalinization at fertilization is limited to a few species and may be absent in mammalian embryos [19, 20], the studies on sea urchin fertilization and activation illustrate that alterations in pH_i can profoundly affect cellular events governing early development.

In mammals, the preimplantation embryo’s ability to maintain its ionic balance is crucial to development. This is particularly important for embryos that have been removed from the oviductal environment and placed into an artificial culture milieu. Recent studies have shown that hamster ova undergoing fertilization, which are notoriously difficult to culture, lack Na⁺/H⁺ antiporter activity prior to the formation of pronuclei, suggesting an important relationship between intracellular pH balance and developmental competence [21]. Furthermore, the culture environment [22–24] or embryo manipulation, such as zona drilling [25] or cryopreservation [26] may affect embryonic pH_i. However, it is not yet clear what developmentally critical events are affected by the changes in pH_i. Therefore, elucidating how cultured embryos regulate and respond to normal changes in pH_i, as well as the cellular consequences of perturbing pH_i, is important for gaining a more complete understanding of mammalian embryo physiology in addition to improving their developmental competence in culture.

A substantial body of work is beginning to emerge on how mammalian embryos regulate their pH_i [21, 22, 24, 27–29]. However, there is limited information on how changes in pH_i affect the embryo [30]. In the study presented here, the potential role of pH_i in maintaining subcellular organization and structural integrity of the embryo was investigated. A disruption in mitochondrial organization, concomitant with developmental arrest, occurs in hamster 2-cell embryos cultured in the presence of inorganic phosphate (P_i) [31], and P_i causes an increase in pH_i in these arrested embryos [23]. Therefore we examined the developmental competence as well as the organization of mitochondria, microtubules, and microfilaments in hamster 2-cell embryos treated with compounds that alter pH_i.

MATERIALS AND METHODS

Embryo Collection and Culture

These investigations were conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and in accordance with the local guidelines of the Research Animal Resources Center at the University of Wisconsin-Madison (animal use protocol number A-07-5600-A00502-3-03-97). Two-cell embryos were collected from mated, eCG-stimulated adult golden hamster females as previously described [32]. For studies of intracellular pH, development, cytoplasmic organization, and recovery, 2-cell embryos were flushed from oviducts at 28–29 h post-egg activation (PEA [33]) with warm, equilibrated HECM-10 [34], rinsed 2× in 50-µl drops of HECM-10, then distributed equally among treatment drops [35], with 7–12 embryos per drop. All incubations of living embryos were done at 37°C in 10% CO₂, 5% O₂, and 85% N₂ in culture drops under mineral oil (Sigma Chemical Co., St. Louis, MO). For developmental studies, embryos were cultured to 82 h PEA, when the developmental stage was assessed on a dissecting microscope. Statistical analysis was performed on an arcsine transformation of the percentage of embryos reaching a given developmental stage [32]. Statistical significance was determined using the general linear models (GLM) procedure of Statistical Analysis Systems (SAS) followed by a least-squares means comparison, blocked by day [36].

Treatments

5,5-Dimethyl-2,4-oxazolinedione (DMO; Sigma) and trimethylamine (TMA; Sigma) were made as 2 M stocks in culture quality water and each used at a final concentration of 20 mM in HECM-10 (abbreviated HDMO and HTMA, respectively), unless otherwise noted. Embryos were initially placed into 15-µl drops of preequilibrated HECM-10 under mineral oil (Sigma). When all embryos were distributed, 15 µl of a 2× solution of the pH_i-altering compound, in HECM-10, was added to the culture drop. For the control embryos, an additional 15 µl of HECM-10 was added to the initial culture drop.

Measuring pH_i

Two-cell embryos were cultured in either HECM-10 or HECM-10 supplemented with increasing concentrations of either DMO or TMA. After 4 h of culture, the pH_i of 2-cell embryos was determined using methods previously described [28]. Briefly, 2-cell embryos were loaded with 0.7 µM 2′,7′-bis(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (Molecular Probes, Eugene, OR) for 20 min at 37°C in HECM-10. Embryos were washed twice in medium without the probe. Measurement of pH_i was achieved using a Nikon Diaphot inverted microscope connected by a Nikon Dual Optical Path Tube to a Photometrics PXL cooled camera (Huntington Beach, CA) for high resolution recording of epifluorescent images. Analysis of fluorescent images was performed using Metamorph/Metafluor hardware and software (Universal Imaging Corporation, West Chester, PA). Emission wavelength was set to 530 nm, and the ratio of fluorescence intensities of excitation wavelengths 500 (pH sensitive) to 450 nm (pH insensitive) was obtained for each embryo. Fluorescent ratios were calibrated in situ using a nigericin/high K⁺ method at four pH levels: 6.7, 7.0, 7.4, and 7.8 for acid-loading and 7.0, 7.3, 7.7, and 8.1 for alkaline loading. The ratio of fluorescence intensities was linearly proportional to pH (r² = 0.984).

Labeling and Specimen Preparation

The mitochondria were labeled with Mitotracker-X-Rosamine (Molecular Probes) [37] by incubating live embryos for 15 min in 330 nM Mitotracker under culture conditions. The embryos were fixed in 1% paraformaldehyde (Electron Microscopy Sciences, Fort Washington, PA) in PBS/PVA (PBS with 0.1% polyvinyl alcohol; Sigma), rinsed twice with PBS/PVA, then mounted in PBS/PVA under a coverslip with Vaseline/paraffin spacers and imaged with confocal microscopy. For all cytoskeleton staining procedures, after fixation the subsequent steps were performed in microtiter wells (Falcon #3077; Becton Dickinson, Lincoln Park, NJ). Microtubules and microfilaments were labeled based on methods described in Barnett et al. [31]. Basically, for microtubule labeling, embryos were fixed in 4% EM-grade paraformaldehyde in PBS + 0.1% Triton-X 100 (PBSTX) (Sigma) overnight at 4°C. Embryos were rinsed in PBSTX and incubated for 15 min in 150 mM glycine (Sigma). After blocking in 10% normal goat serum (Sigma) in PBSTX for 15 min, embryos were rinsed in PBSTX and incubated overnight at 4°C in a mouse anti-β-tubulin primary antibody (Tub 2.1; Sigma) [38, 39] diluted 1:100 in PBSTX. After rinsing in PBSTX, embryos were incubated for 2 h at room temperature in goat anti-mouse Alexa Fluor 488 secondary antibody (Molecular Probes), diluted 1:100 in PBSTX. The embryos were rinsed and mounted as described above. For microfilament staining, embryos were fixed as for the Mitotracker-labeled embryos, then rinsed twice in PBSTX, and incubated for 20 min in PBSTX. The embryos were placed in phalloidin-Alexa Fluor 488 (Molecular Probes) at 0.4 µg/ml in PBSTX for 1 h at 37°C. The embryos were rinsed twice in PBSTX and mounted for observation as described above.

Microscopy

Laser scanning confocal microscopy (LSCM)

Fluorescently labeled fixed embryos were imaged with a Bio-Rad MRC 600 (Hercules, CA) LSCM, using either a 40× 1.3 numerical aperture (NA) or a 60× 1.4 NA oil immersion lens on a Nikon Optiphot microscope.

Two-photon laser scanning microscopy (TPLSM)

Live embryos were imaged with 1047 nm two-photon excitation as previously described [40]. Embryos were collected, stained with Mitotracker as described above, then placed in culture drops for imaging. For each replicate, only one drop containing three embryos could be imaged chronically (with each drop containing one embryo from each of three females). However, each dish contained three drops of embryos, one drop for each of the treatments: HECM-10, HTMA, and HDMO. Thus, one drop could be imaged, and the other two drops served as controls on the microscope stage. For long-term imaging, the embryos in one treatment drop were imaged such that five optical sections, 5 µm apart were collected every 10 min for 6 h. Individual images were collected of the embryos in the control drops at three time points: immediately prior to, 3 h after, and immediately following the long-term imaging. Embryos were then returned to the incubator and development assessed at 82 h PEA.

Statistical analyses were performed to compare the pattern of mitochondria distribution (see below) between treatments using a GLM procedure followed by a least-squares means analysis (blocked by day) on the data obtained from images collected from each of the treatments. In order to assess the dynamics of the disruption of the mitochondria organization, a mixed procedure [36] of the long-term time series data, using a model with day as a random variable and the ratio measurements as repeated measures, followed by a least-squares means comparison, was performed to compare changes over time within each of the two experimental treatments.

For both LSCM and TPLSM imaging, single optical sections were collected from each embryo to include a cross section of the nuclei. Data analyses were performed on raw images. For publication, image figures were assembled using Adobe Photoshop (Adobe Systems, Inc., San Jose, CA) and any adjustments for publication were performed across the entire data set to preserve relative differences in images.

Quantitation of Images

The distribution of mitochondria was quantified using a modification of the method developed by Barnett et al. [31] and illustrated in Figure 3G. All measurements were made on the LSCM or TPLSM digital images using NIH Image software. The data analyzed were the ratios of the average pixel intensity in a 4.9-µm-diameter circle set 3.8 µm in from the cortex (intermediate region) and a similar circle adjacent to the nuclear membrane (perinuclear region). For fixed embryos, data were collected from two subregions of one of the two blastomeres because it has been shown that there is little advantage to collecting additional subsamples [31]. Data for the live embryos were collected in a similar fashion except data was collected from both blastomeres. The location for collection of the regional pixel intensity was assigned using a straight line parallel to the junction of the two blastomeres and bisecting the nucleus. The circles for pixel intensity collection were centered on this line in their respective regions. For statistical analysis, treatments were compared using the GLM procedure in SAS [36] blocking on day, with embryo as the experimental unit. Treatment differences were compared using the least-squares means analysis.

FIG. 3.

Distribution of mitochondria in 2-cell embryos with altered pH_i. These confocal micrographs of 2-cell embryos show how altering pH_i disrupts the perinuclear organization of mitochondria. The embryos were cultured in either HECM-10 (A, D), HTMA (B, E), or HDMO (C, F) for either 3 h (A–C) or 6–7 h (D–F), then labeled with Mitotracker. Bar = 50 µm. G) To quantitate the pattern of mitochondria distribution (white regions) in 2-cell embryos the average pixel intensity of the region within each of the four circles was determined. The ratio of the average pixel intensity of the intermediate region to the perinuclear region from the same side of the blastomere was determined. A ratio closer to one indicates a more homogeneous distribution of mitochondria, whereas a ratio closer to zero indicates a heterogeneous pattern, with a higher accumulation of mitochondria around the nucleus than in the intermediate region [31]. The graph (H) shows these ratios for embryos cultured for either 3 h or 6–7 h in HECM-10, HTMA, or HDMO. Numbers in parentheses are numbers of embryos; bars are standard error. *Significant difference from control at P < 0.05.

Nuclear size was determined as the ratio of the nuclear diameter to the largest blastomere diameter, to compensate for changes in size due to fixation or mounting. In living embryos both the ratio of the nuclear to blastomere diameter, as well as the actual diameters, were used for analysis. For statistical analysis of the nuclear and blastomere size data, treatments were compared using the GLM procedure in SAS [36], blocked by day and using embryo as the experimental unit. Treatment differences were compared using the least-squares means analysis.

Washout Experiments

Embryos were collected as described above and placed into one of the three treatments: HECM-10, HDMO, or HTMA. After 3, 6, 9, 12, or 24 h in these treatments, embryos were rinsed three times in HECM-10, then placed in a fresh 30-µl drop of HECM-10. For the 3-h and 6-h experiments, some embryos were placed into a fresh drop of the treatment they had been initially cultured in, i.e., HDMO or HTMA (negative controls). For every rinse time point, effects of handling were controlled by a set of embryos (positive controls) that were initially cultured in HECM-10, rinsed, and then cultured in a fresh drop of HECM-10. At least three replicates were performed for each rinse time point. Because ANOVA analysis showed that there was no significant difference resulting from the time of rinsing of either the positive control embryos or the negative control embryos the statistical analysis presented was performed on the entire data set.

To determine changes in the cytoplasmic organization, embryos were cultured for 4 h in either HECM-10, HTMA, or HDMO and stained with Mitotracker as described above, rinsed three times in HECM-10, and returned either to fresh drops of HECM-10 for recovery, or into fresh drops of the treatments in which they had been originally cultured. At this time (Time 1), each embryo was imaged with TPSLM and then returned to the incubator for an additional 3 h. Embryos were each then imaged a second time (Time 2). Immediately following the completion of this second imaging, embryos were fixed and stained for microfilaments.

RESULTS

DMO and TMA Alter Intracellular pH and Disrupt Development

Control 2-cell embryos cultured for 4 h in HECM-10 exhibited a pH_i of about 7.2. In comparison, similar embryos cultured in the presence of increasing concentrations of DMO showed an acidification of pH_i while those cultured in the presence of increasing concentrations of TMA showed an alkalinization of pH_i (Table 1). Altering the pH_i of 2-cell embryos with TMA or DMO inhibited development to the morula and blastocyst stage in a dose-dependent manner (Table 1).

TABLE 1.

The pH_i, and development of 2-embryos cultured with increasing concentrations of DMO and TMA.

Compound concentrationa	pHib	% Morula and blastocyst	% Blastocyst
TMA 0 mM	7.186 ± 0.04	97.3	73.3
TMA 5 mM	7.285 ± 0.07	82.6c	34.7d
TMA 10 mM	7.383 ± 0.04c	49.1d	6.0d
TMA 20 mM	7.421 ± 0.06c	5.17d	2.6d
TMA 40 mM	7.538 ± 0.04d	0d	0d
TMA 80 mM	7.685 ± 0.05d	0d	0d
DMO 0 mM	7.242 ± 0.04	90.58	74.7
DMO 5 mM	7.221 ± 0.07	87.1	61.1c
DMO 10 mM	7.09 ± 0.05c	46.3d	21.6d
DMO 20 mM	6.87 ± 0.05d	0d	0d

^aTMA series consisted of 115 embryos per treatment over 14 replicates. DMO series consisted of 170 embryos per treatment over 7 replicates.

^bDetermined after 4 h culture; n > 20 embryos per treatment.

^cSignificantly different from control at P < 0.05.

^dSignificantly different from control at P < 0.01.

Subsequent experiments on the cytoplasmic effects of altered pH_i were performed using a concentration of 20 mM for both DMO and TMA, the lowest concentration that caused a dramatic developmental arrest. At this concentration, a significant number of embryos arrested at the 2-cell stage (Fig. 1).

FIG. 1.

Embryonic development in either 20 mM DMO or TMA. This graph shows the percentage of embryos reaching each developmental stage when assessed at 82 h PEA. Embryos were cultured continuously from the 2-cell stage in HECM-10, HTMA, or HDMO. Different letters indicate statistical difference at P < 0.05 within developmental stage.

Embryos Can Recover from pH_i Perturbation

The capability of 2-cell embryos to recover from various lengths (3 h, 6 h, 9 h, 12 h, 24 h) of exposure to TMA or DMO was determined (Fig. 2). After returning to the control medium, embryos recovered a statistically significant amount of their developmental potential (to the morula and blastocyst stage) with up to a 12-h exposure to the pH_i altering compounds, as compared to embryos cultured for the entire period in either HTMA (T/T) or HDMO (D/D), but the degree of recovery declined after 6 h. Although the treated embryos were not always able to develop to the morula and blastocyst stages at levels equivalent to the positive control, development was significantly improved compared to the negative control in most of the treatments.

FIG. 2.

Developmental recovery after pH_i perturbation. Embryos were cultured for varying times in HECM-10 (H), HTMA (T), or HDMO (D), rinsed in HECM-10, returned to fresh medium, then cultured until 82 h PEA, at which time developmental stage was assessed. The letter in front of the slash indicates first culture treatment while the letter after the slash indicates culture treatment following rinsing. Numbers in parentheses are number of embryos. Time under columns is time of culture prior to rinsing. Data represent a minimum of three replicates per time point. Data for the positive (H/H) or negative (T/T or D/D) control were combined across time points for clarity of presentation. *Statistically significant difference from positive control (P < 0.01). ●Statistically significant difference from negative control (P < 0.01).

Embryos cultured for 9 h in HTMA showed a significant reduction in development beyond the 4-cell stage compared to controls (H/H = 100%; T/H = 75.4%; P < 0.01), whereas those cultured for the same period in HDMO did not show such a reduction (D/H = 94.7%; P = 0.24). In fact, HDMO-cultured embryos did not show a significant decrease in development beyond the 4-cell stage even with 24 h of exposure (H/H = 100%; D/H = 86.2%; P = 0.38). However, embryos cultured in HDMO for 24 h did exhibit a significant decrease in development to the 8-cell stage (H/H = 100%; D/H = 48.3%; P < 0.01).

Altering pH_i Disrupts the Organization of Mitochondria

Embryos cultured in HECM-10 for either a short (3 h) or long (6–7 h) duration exhibited a perinuclear organization of mitochondria (Fig. 3, A and D), similar to that previously described for in vivo [41] and developmentally competent, in vitro-cultured [31] 2-cell hamster embryos. However, this organization was disrupted in embryos cultured in either HTMA or HDMO at either time point (Fig. 3, B and E, or C and F, respectively), with more obvious disruption following the longer culture period. Specifically, the mitochondria no longer maintained their perinuclear organization but instead were observed in the intermediate cytoplasm. The images of Mitotracker-labeled embryos were quantified to assess the degree of pattern disruption in embryos with altered pH_i as compared to control embryos, using the method developed by Barnett et al. [31] as shown in Figure 3G and described in Materials and Methods. This quantitation showed that there was a clear treatment effect on the distribution of mitochondria (Fig. 3H).

To investigate possible temporal differences in the organization-disrupting effect of either TMA or DMO, living embryos labeled with Mitotracker were imaged using TPSLM. These live imaging studies revealed the time course of the change in mitochondria organization in embryos treated with either TMA or DMO (Fig. 4A). In both treatments there was a movement of mitochondria away from the nuclei that began within 1 h of the addition of the pH_i-altering compounds. Furthermore, when the images were quantified it was clear that there was a difference in the temporal progression of the disruption of the mitochondrial organization between embryos with increased pH_i compared to those with decreased pH_i (Fig. 4B, bar graph). At the initiation of imaging (within 15 min following the addition of pH_i-altering compounds), there were no differences in the pattern of mitochondrial distribution between the treated and untreated embryos. After 3 h there was not yet a significant difference between controls and embryos cultured in HDMO, whereas there was a significant difference in the distribution of mitochondria in embryos cultured in HTMA when compared to untreated controls. By 6 h of culture, there was a significant difference between each experimental treatment when compared to the control.

FIG. 4.

Change in the distribution of mitochondria over time in living embryos with altered pH_i. A) These single optical sections of live 2-cell embryos labeled with Mitotracker and imaged with TPLSM illustrate how mitochondria move away from their perinuclear localization shortly after the addition of the pH_i-altering compounds, while embryos cultured in HECM-10 retain the perinuclear configuration. The HECM-10 embryo shown in the top row was cultured in the same dish as the HTMA-cultured embryo shown in the time series in the second row. The HECM-10 embryo shown in the bottom row was cultured in the same dish as the HDMO-cultured embryo shown in the time series in the third row. The t = 0-h images (left column) were collected prior to the addition of the pH_i-altering compounds. Scale bar = 50 µm. The graph (B) illustrates the change of mitochondria distribution over time in living embryos. The bar graph shows data from individual images from each of the three treatments collected prior to the time-lapse imaging (0 h), as well as at 3 and 6 h following the addition of the pH_i-altering compounds. The data shown represent six replicates. *Statistical difference (P < 0.05) from control within a time point. Bars indicate standard error. The line graph shows the quantitation of the mitochondrial pattern of individual images from a time-lapse sequence on a single culture. Each line represents three replicates. Bars indicate standard error. Arrows show the first time point that is statistically different (P < 0.01) from Time 0 within a treatment.

This temporal difference was also evident when comparing the distribution of mitochondria at different time points within the same treatment (Fig. 4B, line graph). These data were collected from embryos imaged sequentially over the 6-h period. Because only one treatment can be imaged in this fashion for each replicate, it was inappropriate to compare directly between treatments. However, comparing the pattern of mitochondrial distribution within the same treatment across different time points provided insight into the dynamics of these changes. By 3 h after the addition of TMA the pattern of mitochondrial distribution in embryos cultured in HTMA was significantly different from the same embryos prior to the addition of TMA. In contrast, the mitochondria pattern of embryos cultured in HDMO was not significantly different from their pretreatment pattern until 4 h after the addition of DMO.

Altering pH_i Disrupts the Structure and Organization of Other Cellular Components

Because the nuclei in some of the alkalinized embryos appeared smaller than those in control embryos (Fig. 3E), the ratio of the nuclear diameter to the blastomere diameter in fixed embryos was determined. At both 3 h and 6 h there was a significant difference in the nuclear:blastomere diameter ratio between embryos cultured in HECM and HTMA but not between HECM and HDMO (Table 2A). The reduction in the nuclear:blastomere ratio in HTMA-cultured embryos was not due to distortion in shape caused by fixation because a similar analysis of live embryos also revealed a decrease in the nuclear:blastomere diameter ratio (Table 2A). In these living embryos, prior to the addition of TMA or DMO there was no significant difference in this ratio between embryos to be cultured in HECM-10 and HTMA or HECM-10 and HDMO (0 h). However, after 3 or 6 h of culture there was a significant reduction in the nuclear:blastomere diameter ratio in embryos cultured in HTMA compared with those in HECM-10. Interestingly, in these live embryos there was a significant increase in the nuclear:blastomere diameter ratio between embryos cultured in HECM-10 and HDMO after 3 h, but not after 6 h, of culture.

TABLE 2.

Nucleus and blastomere size in 2-cell embryos cultured in media containing pH_i-altering compounds.

	HECM-10 (n)a	HTMA (n)	HDMO (n)
A. Ratio of nuclear diameter to blastomere diameter (N/B ± SEM)
Fixed embryos
3 h	0.35 ± 0.005 (23)	0.30 ± 0.005b (21)	0.33 ± 0.007 (14)
6 h	0.31 ± 0.004 (46)	0.27 ± 0.004b (52)	0.31 ± 0.002 (40)
Live embryos
0 h	0.30 ± 0.004 (17)	0.31 ± 0.003 (18)	0.31 ± 0.004 (18)
3 h	0.32 ± 0.004 (18)	0.29 ± 0.003b (18)	0.33 ± 0.004b (18)
6 h	0.32 ± 0.003 (16)	0.29 ± 0.004b (18)	0.33 ± 0.003 (17)
B. Actual nuclear and blastomere diameters in live embryos (µm ± SEM)
Nuclear diameter
0 h	17.79 ± 0.26 (17)	17.71 ± 0.17 (18)	17.84 ± 0.22 (18)
3 h	17.98 ± 0.19 (18)	17.29 ± 0.23b (18)	18.19 ± 0.23 (18)
6 h	18.24 ± 0.20 (16)	17.22 ± 0.29b (18)	18.36 ± 0.14 (17)
Blastomere diameter
0 h	58.45 ± 0.53 (17)	57.76 ± 0.41 (18)	57.99 ± 0.45 (18)
3 h	56.85 ± 0.32 (18)	58.99 ± 0.39b (18)	55.09 ± 0.34b (18)
6 h	57.01 ± 0.37 (16)	60.03 ± 0.62b (18)	55.23 ± 0.40b (17)

^a(n) = number of embryos.

^bSignificantly different at P < 0.05 from control within time point.

To determine whether the difference in the nuclear:blastomere diameter ratio was due to a change in the nuclear size or in the blastomere size, measurements of these parameters obtained from living embryos were analyzed. As shown in Table 2B, the difference in ratio seen in the fixed or live embryos cultured in HTMA resulted from both a decrease in the diameter of the nucleus as well as an increase in the size of the blastomere. This was true at both the 3- and 6-h time points. In contrast, in the HDMO-treated embryos, the blastomere diameter was decreased at both the 3- and 6-h time points, but there was no significant difference in the diameter of the nucleus when compared to controls.

In order to determine a possible proximal target of the altered pH_i for the disruption in cytoplasmic organization, the organization of microtubules and microfilaments in DMO- and TMA-treated 2-cell embryos were examined. In the control embryos, the microtubules form an interconnected network throughout the cytoplasm (Fig. 5A). There appeared to be little, if any, change in this organization in embryos cultured in HTMA or HDMO (Fig. 5, B and C). There was, however, a dramatic change in the organization of the microfilaments in both HTMA- and HDMO-treated embryos. In control embryos the actin filaments formed a network around the nucleus and at the cortex, extending into the cortical microvilli, with a few fine filaments traversing the intermediate region (Fig. 5D), essentially mirroring the mitochondria pattern. When embryos were treated with TMA the perinuclear ring disappeared and aggregates of actin were observed throughout the cytoplasm, with some fine filaments still occupying the intermediate region (Fig. 5E). The cortical actin and the microvilli appeared unaffected. This type of disruption was seen in 38% of the embryos after 3 h of culture (n = 24) and in 74% of the embryos following 6 h of culture (n = 23). The perinuclear microfilament organization in HDMO-cultured embryos was also disrupted, but in a different manner. The microfilaments were still prevalent in the cytoplasm but had dispersed away from the nucleus to fill up the intermediate region. Again, the cortical and microvillar microfilaments appeared unchanged. At 3 h of culture 65% (n = 26) of the embryos showed some disruption in the perinuclear organization of microfilaments, but by 6 h of culture all of the embryos (n = 19) showed some disruption in the perinuclear organization of microfilaments.

FIG. 5.

Cytoskeletal organization in embryos with altered pH_i. Confocal micrographs of 2-cell embryos cultured in either HECM-10 (A, D), HTMA (B, E), or HDMO (C, F) for 6 h, showing the microtubule (A–C) and microfilament (D–F) configurations in these embryos. Scale bar = 50 µm.

Embryos Can Recover Structurally from pH_i Perturbation

To determine whether embryos that were capable of developmental recovery following a perturbation in pH_i were also able to regain at least some of their structural integrity, live 2-cell embryos, with labeled mitochondria, were imaged using TPLSM (Fig. 6, A–E) after 4 h of culture in either HECM-10, HTMA, or HDMO. The same embryos were imaged a second time after being permitted to recover for 3 h in HECM-10 (Fig. 6, F–J). Following this second imaging, the embryos were fixed and stained for microfilaments (Fig. 6, K–L). The patterns of either mitochondria or microfilaments in the embryos that were rinsed and returned to their original treatment (columns 1–3) resembled those presented earlier (Figs. 3 and 4 for mitochondria and Fig. 5 for microfilaments), whereas those treated with TMA or DMO then rinsed and returned to control medium (HECM-10) initially showed some disruption in the organization of their mitochondria (Fig. 6, E and F) but either regained a more perinuclear pattern, or at least did not become more disrupted (Fig. 6, I and J). The changes in the pattern of mitochondrial distribution were quantified, and the results are shown in Figure 6P. At the initial imaging, after 4 h of treatment with TMA or DMO, there was a disruption in the pattern of mitochondrial organization in both of these treatments. However, at the second time point, embryos that had been rinsed and returned to HECM-10 recovered some of the perinuclear pattern, as demonstrated by the fact that these embryos were either not different from controls (HDMO/HECM embryos), or they were not as perturbed as their counterparts cultured continuously in the same treatment (comparing HTMA/HECM to HTMA/HTMA). Thus, the recovery of the mitochondria pattern was not necessarily equivalent to controls.

FIG. 6.

Mitochondria distribution in embryos permitted to recover from pH_i perturbation. The first row of TPLSM images (Time 1: A–E) shows the distribution of mitochondria in embryos cultured for 4 h in either control medium (HECM-10) or either HTMA or HDMO. Embryos were rinsed, stained, placed into the second culture treatment (HECM-10, HTMA, or HDMO), then imaged. Time 2 is 3 h after Time 1, when embryos were imaged a second time (F–J). Immediately after the second imaging, the embryos were fixed, stained for actin, and imaged with LSCM (K–O). Abbreviation in front of slash mark is initial culture medium; abbreviation after slash mark is second culture medium. Scale bar = 25 µm. P) The distribution of mitochondria was quantified and the results are shown in this graph. Data include three replicates. n = 12 embryos for all treatments except HECM/HECM Time 2 where n = 10 embryos. *Significant difference (P < 0.05) from control (HECM/HECM). ●Significant difference (P < 0.05) of treated embryos rinsed and placed into HECM-10 from their respective negative control (either HTMA/HTMA or HDMO/HDMO).

This was true for the microfilament organization as well, where embryos cultured in the single treatments over time (Fig. 6, L and M) showed a disrupted microfilament organization (HTMA/HTMA = 66%, n = 9; HDMO/HDMO = 100%, n = 8) compared to controls (Fig. 6K: HECM/HECM = 13%, n = 8), whereas those embryos returned to HECM-10 after culture in the pH_i-altering compounds showed patterns of microfilament organization (Fig. 6, N and O) more similar to the control. In both HTMA/HECM and HDMO/HECM, only 22% (n = 9 for each treatment) of the embryos showed disrupted microfilament organization.

DISCUSSION

Mammalian embryo development is decreased when pH_i strays beyond a relatively small range [28, 30, 42]. The embryo’s ability to cope with an in vitro culture environment may be altered depending on how the culture medium affects pH_i [22, 23, 30]. The interrelationships between changes in culture media, pH_i, and cellular processes are most certainly complex. In this study, the relationship between pH_i and the structural organization of the embryo was investigated.

The studies presented here showed that the weak acid DMO and the weak base TMA can be used to alter the pH_i of hamster 2-cell embryos. Alteration of pH_i with weak acids and bases has been used by a number of researchers to determine the role of pH_i in a variety of cellular functions [8, 16, 30, 43–46]. In the work presented here both TMA and DMO, at 20 mM, inhibited development, with many embryos in both treatments unable to progress be1852 yond the 2-cell stage. This is consistent with the observation that embryos cultured in the presence of P_i are arrested at the 2-cell stage [32].

Although nonspecific effects of the pH_i-altering compounds cannot be completely discounted [8], it seems unlikely that the developmental and structural defects observed were due to acute toxicity—that the alterations observed are simply expressions of cell death—because embryos were able to recover from limited treatment with both TMA and DMO. Because 2-cell hamster embryos do have mechanisms for maintaining pH_i [28], the extent of these embryos’ ability to recover from the pH_i perturbations was determined. The developmental recovery study presented here showed that although embryos can recover much of their developmental competence, it was frequently not equivalent to untreated controls, particularly with increasing periods of pH_i perturbation. This observation suggests that the embryos did not have an unlimited ability to recover from a pH_i stress, and this may contribute to the suboptimal development commonly observed in cultured embryos.

Both increasing and decreasing the pH_i resulted in a disruption of mitochondria organization at the 2-cell stage. The live imaging study showed that a significant disruption in the pattern of mitochondria organization occurred at least an hour earlier in TMA-treated embryos than in DMO-treated embryos. This suggests that the TMA treatment more rapidly disrupted the mitochondrial organization than did DMO treatment, supporting the concept that the effects of these two treatments, although similar, were not identical. In this time course study the mitochondria dispersed away from the nucleus. This gradual dispersion of mitochondria away from the nucleus has also been observed in living embryos cultured in the presence of P_i (unpublished data). Alterations in pH_i disrupt organelle positioning in other cell types. Specifically, in hyphae, acidification of pH_i within the range of the studies presented here, namely 6.8– 7.2, permitted mitochondria to move toward the apex, a region from which they are normally excluded [43]. In fibroblasts, changes in pH_i altered the distribution of lysosomes [47].

The organization of the mitochondria was not the only structural parameter of the embryos disrupted by alterations in pH_i. The nuclear:blastomere diameter ratio was reduced in TMA-treated embryos. In contrast, there was an increase in the nuclear:blastomere diameter ratio of live DMO-treated embryos, suggesting opposite effects of these two treatments. Further analysis of live embryos revealed that in alkalinized embryos there was a concomitant and inverse change in both the blastomere and the nuclear diameter, with the blastomere diameter increasing and the nuclear diameter decreasing. In contrast, in acidified embryos only the blastomere diameter decreased, accounting for the increase in the nuclear:blastomere diameter ratio. It seems unlikely that the change in nuclear size was due to the effect of the altered pH_i on chromatin because studies on such effects in other systems suggest that the volume of isolated nuclei is reduced by acid pH [48–50], DNA tends to have a looser structure at high pH_i [49, 51, 52], and nuclear lamins are more soluble at higher pH [53], favoring the concept that the nucleus would be more likely to swell under alkalinizing conditions. Therefore, it seems probable that some other mechanism(s) must be involved in changing the nuclear size in these embryos. For example, the disruption of the microfilament cytoskeleton may prevent the cells from maintaining their proper structural integrity.

Altering the pH_i of 2-cell hamster embryos had no obvious effect on the microtubule cytoskeleton. This is surprising in view of various studies showing that pH can have a dramatic effect on microtubule polymerization in a number of different systems [16, 54–57]. However, in fungal hyphae when the pH_i was reduced, a number of other cytoplasmic structures were disrupted, but the microtubule network remained apparently unchanged [43, 58]. Of course, a lack of detectable change in the organization of the microtubule network does not mean that functional changes have not occurred. Although no major differences are seen in the structure of proteins associated with microtubules over a pH range of 6.3–7.8 [55], the function of these proteins could be altered [59, 60].

In contrast to their lack of an observable effect on microtubules, both TMA and DMO dramatically altered the microfilament network. Microfilament organization is also changed in fungal hyphae when pH_i was decreased [43, 58]. Although it is possible that these phenomena result from a direct and global change in microfilament assembly and disassembly, this seems unlikely in the hamster embryo because microfilaments were still present and the microvilli and cortical actin appeared unchanged following disruption of pH_i. Therefore it seems plausible that the disruption in the actin organization was via a microfilament-associated protein, possibly a bundling protein. This is supported by observations in sea urchins, where alkalinization alone causes bundling of actin [17, 18] and, along with Ca²⁺, increasing pH_i promotes the formation of masses of actin filament bundles in platelets [61]. In the present study with 2-cell hamster embryos there may have been an exaggeration of microfilament bundling controlled by pH_i such that at high pH_i the bundling was excessive and resulted in the formation of aggregates, whereas at lower pH_i the bundling no longer occurred properly, and the filamentous network remained but was no longer held in its proper configuration. Why the cortical actin remained immune to these effects is unclear. However, it does support the hypothesis that the effect of pH_i alteration was not directly on the actin but either via a secondary effect, such as the alteration of metabolism [1], or through changes in a regulatory or an associated protein whose activity may be pH_i sensitive [4].

It is possible that the alterations in developmental competence and cytoplasmic organization are not a direct result of increasing or decreasing pH_i, but are likely to be steps in a network of complex cytological changes that may include changes in Ca²⁺ levels [5], protein synthesis [62], and function [4], or alterations in metabolism [1]. It has been shown that decreasing pH_i with DMO decreases glycolytic activity in mouse zygotes [30]. Alternatively, increasing pH_i with TMA increases glycolytic activity of 2-cell hamster embryos but decreases oxidative metabolism [26]. The studies presented here demonstrate a link between alterations in pH_i and structural organization in the hamster 2-cell embryo, stimulating further investigations into the interplay between ionic homeostasis, developmental competence, and other cellular processes in mammalian embryos.