| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Monash Immunology and Stem Cell Laboratories (MISCL), Monash University, Clayton, Victoria 3800, Australia
ABSTRACT
Mitochondrial dysfunction may be acquired or inherited by oocytes without detectable morphological abnormalities. This pathology may account for some examples of unexplained pregnancy loss in women following transfer of morphologically normal in vitro fertilization (IVF) embryos. The present study was intended to determine whether sublethal mitochondrial injury in mouse oocytes before IVF negatively affects pre- and postimplantation development, and to further define the latency of developmental compromise in relation to aberrant mitochondrial metabolism. Mature mouse oocytes were loaded with the mitochondrial fluorophore rhodamine-123 and photosensitized for 20 sec, a duration previously found to permit preimplantation embryo development to the blastocyst stage and so deemed "sublethal." This treatment resulted in some aberrations in cytoplasmic patterning of organelles, but did not inhibit zygote mitochondrial metabolism. Blastocyst development following IVF was not signficantly inhibited following sublethal oocyte photosensitization; however, a decrease in trophectoderm cell numbers was observed relative to untreated controls. Following intrauterine transfer, blastocysts derived from sublethally photosensitized oocytes implanted but later aborted at a higher rate, formed fetuses with lower average weights, and, in rare cases, formed abnormal fetuses relative to controls. Photosensitization for more prolonged durations resulted in failed fertilization (2 min) and rapid oocyte degeneration (10 min). Therefore, photosensitization duration and the consequent degree of mitochondrial dysfunction are negatively related to the onset of developmental compromise. Acquired low-level mitochondrial injury is heritable by the resultant embryos and can cause postimplantation developmental compromise that may be relevant to some clinically observed outcomes following human assisted reproduction strategies, including reduced birth weights for gestational age. Future strategies for the detection and prevention of mitochondrial dysfunction may assist in improving outcomes for some clinically infertile women.
early development, embryo, implantation, in vitro fertilization, oocyte development
INTRODUCTION
Oocytes contain a pool of nonreplicating yet functional mitochondria from which all cells of the resultant embryos directly inherit their own complements. It has recently been confirmed that ooplasmic mitochondrial metabolism is necessary for initiation [1] and progression [2] of preimplantation embryo development in the mouse. Morphological changes have been reported in human oocytes with respect to advancing female age [3], which may correlate with the progressive age-related decline in mitochondrial charge that has been reported relative to embryo cell number [4]. Furthermore, levels of oocyte ATP have been shown to correlate with postimplantation outcomes [5]. It has been postulated also that disordered spatial patterning of these organelles in the human oocyte could be inherited by daughter cells, resulting in local metabolic heterogeneities [6]. Mitochondrial genomes (mtDNAs) that are mutated can also occupy an oocyte together with nonmutated transcripts, resulting in heterogeneic mitochondrial pools that can be transmitted to live-born offspring, resulting in a range of pathological metabolic disorders [7]. The severity and onset of these conditions is directly related to the relative "mutant load" within oocytes and early embryos.
It has yet to be proven whether these sorts of genetic or physiological defects in mitochondria are directly responsible for unexplained etiologies of female infertility. Despite having multiple biological causes, clinical phenotypes including recurrent fertilization failure and retarded preimplantation embryo development in vitro are highly suggestive of inherent or acquired aberrations in nutrient metabolism or energy production. It is possible that even minor defects in mitochondrial energy metabolism, such as those occurring with female age, do not necessarily impinge on ongoing preimplantation development but may have adverse effects at later stages of development. This concept is raised by the Barker hypothesis, which predicts that the quality of embryonic and fetal nutrition in utero may have downstream heritable influences on adult health [8]. The same concept extends also to suboptimal nutrition during the periconception stages of development, as shown in rodent and livestock species for which suboptimal diets administered for short durations during this period caused preimplantation developmental retardation as well as fetal health problems prenatally [9, 10].
Given that mitochondria are 1) maternally inherited, 2) principal sites of oxidative damage, 3) required for cytoplasmic energy production in all embryonic cells, and 4) important regulators of preimplantation embryo development, it is feasible that inherited or acquired mitochondrial dysfunction in the oocyte can be detrimental to postimplantation development. Based on previous observations that blastocyst development in vitro is relatively resistant to low levels of mitochondrial injury induced by fluorophore photosensitization in the mouse oocyte [11], the present study was intended to identify any delayed developmental effects of this treatment after implantation. Blastocysts generated in vitro from treated oocytes were assessed for pre- and postimplantation developmental competence. The effects of extended oocyte photosensitization durations were also tested to define the tolerance of ooplasmic mitochondria to injury and define the latency of manifested developmental damage.
MATERIALS AND METHODS
Animals
In all experiments, C57BL6J x CBA/CaH (F1) hybrid mice were used. Mice were housed in an environmentally controlled room at 22–24°C with a 12L:12D photoperiod, with food and water available ad libitum. For oocyte harvest, 4–6 wk (pubertal) female mice were superovulated by i.p. injection of 5 IU eCG (Folligon; Intervet) followed approximately 48 h later with an i.p. injection of 5 IU hCG (Chorulon; Intervet) [12]. For in vitro insemination of oocytes, mature sperm were isolated from whole epididymides of sexually mature (10–12 wk) male mice. Pseudopregnant recipient female mice used for intrauterine blastocyst transfer were prepared by mating mature (>12 wk) female mice with mature (10–12 wk) vasectomized males. Both female and male mice were killed by cervical dislocation before retrieval of gametes.
Ethics of Experimentation
The present study was approved by Monash Medical Centre Animal Ethics Committee A, Monash Medical Centre, Clayton, Victoria, Australia, approval number A2003/40. Experiments were conducted in accordance with the 1997 NH & MRC Australian Code of Practice for the Care and Use of Animals for Scientific Purposes and the Victorian Prevention of Cruelty to Animals Act and Regulations, 1986.
Oocyte Isolation, In Vitro Fertilization, Blastocyst Culture, and Assessment of Blastocyst Development
Oocyte collection, sperm preparation, and in vitro fertilization were performed as described previously [13]. Normally fertilized zygotes were then cultured for 96 h to the blastocyst stage in a humidified atmosphere of 5% CO2 at 37°C in a thermostat-controlled incubator. Three replicate groups of 40 zygotes were cultured in groups of 10 in 20-µl droplets of bicarbonate-buffered KSOMaa medium [14] supplemented with 1 mg/ml BSA (Invitrogen) in sterile plastic Petri dishes overlaid with mineral oil (Sigma Aldrich). The number of blastocysts that formed in these groups was expressed as a percentage of normally fertilized oocytes (n = 120). Three independent replicate groups of 20 zygotes were grown to blastocyst stage and assessed for mitotic development by microscopic determination of inner cell mass and trophectoderm cell numbers according to a previously described method [15].
Photosensitization of Oocytes Using Epifluorescence Microscopy
Isolated and denuded oocytes were photosensitized using the mitochondrion-specific fluorophore rhodamine-123 (R123; 50 µg/ml; Molecular Probes) as described previously [11]. After loading, oocytes were sublethally photosensitized by irradiation for 20 sec on the stage of an epifluorescence microscope before insemination. For comparisons of blastocyst formation rates, oocytes from the same batch of mice were either incubated in R123 but not irradiated ("R123 only"), or not loaded or irradiated ("not treated") (n = 120). Control groups for the R123 vehicle alone or for light exposure alone were not included in this study because they have been shown previously to have no significant effects on preimplantation embryo development [11, 16]. Extended photosensitization treatment was performed at two additional time periods of 2 and 10 min, which were both chosen arbitrarily and were each applied to a set of three replicate groups of at least 30 oocytes. The 2-min exposure was based on the presumption that developmental compromise would occur earlier than that observed for the 60-sec (lethal) exposures found previously to result in early cleavage embryo arrest [11]. The 10-min exposure was chosen on the basis of preliminary observations of oocyte-specific arrest (data not shown).
Assessment of Mitochondrial Morphology and Metabolism in Zygotes after Sublethal Oocyte Photosensitization
An independent group of 30 oocytes (three replicate groups of 10 per group) were photosensitized sublethally (20 sec) and after normal fertilization, zygotes were microscopically assessed either for mitochondrial membrane potential (
) using laser-scanning confocal microscopy, or for autofluorescence (a marker of NADH/NADPH content) or ATP (a measure of mitochondrial energy production) using epifluorescence microscopy, as described previously [11]. The same assessments were performed on zygotes derived from similar-sized control groups of oocytes that were lethally photosensitized (60 sec) or loaded but not irradiated (0 sec). Values of NADH/NADPH content and were expressed as arbitrary units (a.u.) of fluorescence intensity relative to background levels. Fluorescence microscopy and laser-scanning confocal microscopy were also used to capture images of some zygotes derived from loaded (irradiated or nonirradiated) oocytes to visualize cytoplasmic patterning of mitochondria, as outlined previously [11].
Assessment of Postimplantation Development of Blastocysts after Sublethal Oocyte Photosensitization
Blastocysts derived from oocytes that were sublethally photosensitized were surgically transferred to the uteri of pseudopregnant female recipient mice as described previously [12]. Blastocysts derived from oocytes that were either loaded and not irradiated or untreated were transferred to separate females as controls. On Day 16.5 postcoitum, conceptuses were isolated from recipient females and fetuses were assessed for morphological normality and wet weight after manual detachment from the placentas and attached membranes.
Statistical Analysis
Means of blastocyst formation rates, blastocyst cell numbers, metabolic endpoint measurements, rates of zygote arrest, and fetal wet weights for treatment groups were exponentially transformed and compared to control values using ANOVA with Tukey's posttest. Proportions of blastocysts that underwent fetus formation, resorption, or loss after transfer following oocyte photosensitization were compared to untreated controls using the chi-squared test. Differences between means were considered biologically significant at a P value of less than 0.05.
RESULTS
Blastocyst Development Following Sublethal Oocyte Photosensitization
Sublethally photosensitized oocytes (20 sec) formed blastocysts after fertilization (62.9%) at similar rates to those seen for loaded, nonirradiated oocytes (0 sec, 72.5%) and untreated control oocytes (72.3%), confirming previously reported findings [11, 16] (Fig. 1). Cell numbers of inner cell mass subpopulations were similar in blastocysts derived from sublethally photosensitized oocytes (11 ± 1) compared to those derived from loaded nonirradiated (0 sec) oocytes (13 ± 1) or untreated oocytes (12 ± 1). Cell numbers of trophectoderm subpopulations were, however, lower in blastocysts derived from sublethally photosensitized oocytes (25 ± 2) as well as those derived from loaded, nonirradiated (0 sec) control oocytes (28 ± 2) in comparison to untreated oocytes (38 ± 2, P < 0.01). These decreased trophectoderm numbers were reflected in the decreased total cell numbers of blastocysts originating from sublethally photosensitized oocytes (36 ± 2) and from loaded nonirradiated control oocytes (41 ± 2) compared to blastocysts originating from untreated oocytes (49 ± 3) (Fig. 2).
|
|
Patterning and Metabolism of Mitochondria in Zygotes after Sublethal Oocyte Photosensitization
Compared to zygotes derived from loaded nonirradiated oocytes (Fig. 3A), those derived from sublethally (20 sec) photosensitized oocytes displayed a range of potentially anomalous cytoplasmic patterning, including irregular peri-pronuclear aggregations, focal organelle swelling, and asymmetric distributions (Fig. 3, B–E) as observed using epifluorescence microscopy. These anomalies were less widespread than the dense aggregation patterns that were typical of zygotes undergoing arrest and cytoplasmic degeneration (Fig. 3F) following lethal (60 sec) oocyte photosensitization. Laser-scanning confocal microscopy further revealed an absence of peri-pronuclear mitochondrial aggregations in one zygote derived from a sublethally photosensitized oocyte, despite an intense red fluorescence relative after staining with JC-1, which indicates an intact mitochondrial membrane potential (Fig. 3, G and H).
|
In terms of metabolic function, zygotes derived from sublethally photosensitized oocytes (20 sec) were observed with NADH/NADPH levels (105.9 ± 3 a.u.) that were similar to those for zygotes derived from loaded nonirradiated oocytes (104.3 ± 3 a.u.), lower than those for zygotes derived from untreated control oocytes (114.0 ± 5, P < 0.05), and higher than those for zygotes derived from lethally (60 sec) photosensitized oocytes (83.8 ± 3 a.u., P < 0.001). The of zygotes derived from sublethally photosensitized oocytes (1.14 ± 0.04 a.u.) was similar to that of loaded nonirradiated oocytes (1.25 ± 0.02 a.u.) and untreated oocytes (1.31 ± 0.08 a.u.), but was significantly higher than those for zygotes derived from lethally (60 sec) photosensitized oocytes (1.08 ± 0.04 a.u., P < 0.05). Observations of zygote energy production followed a similar trend in that ATP levels in zygotes derived from sublethally photosensitized oocytes (0.82 ± 0.03 pmol) were similar to those of zygotes derived from loaded, nonirradiated oocytes (0.90 ± 0.05 pmol) and untreated oocytes (0.96 ± 0.010), but significantly higher than for zygotes derived from lethally photosensitized oocytes (0.45 ± 0.10 pmol, P < 0.01) (Table 1).
|
Effects of Extended Photosensitization on Oocyte Developmental Competence
Zygotes derived from oocytes photosensitized for 2 min and assessed 6–8 h after insemination at the time of expected pronucleus formation were found to have arrested and degenerated at significantly higher rates (81%, P < 0.01) than control groups of loaded, nonirradiated (0 sec) oocytes (12%) or untreated (15%) oocytes (Fig. 4). Zygote arrest following photosensitization of oocytes for 2 min was marked at the time of pronucleus formation by a dark and shrunken cytoplasmic appearance and in some cases was associated with dense accumulations of zona-bound spermatozoa. Zygotes that remained intact appeared to be undergoing degeneration and exhibited no pronuclei and a translucent cytoplasm, similar to zygotes described with apoptosis and associated mitochondrial dysfunction [17, 18]. When photosensitization was extended to 10 min, all exposed oocytes underwent a progressive swelling that resulted in plasma membrane rupture and lysis approximately 2 h after treatment. In comparison, loaded, nonirradiated oocytes and unfertilized oocytes incubated in culture medium alone remained morphologically intact for up to 2 days, as described previously for unfertilized mouse oocytes [19].
|
Postimplantation Effects of Sublethal Oocyte Photosensitization
Blastocysts derived from sublethally photosensitized oocytes implanted and formed fetuses at a similar rate compared to untreated control blastocysts (16.3% vs. 17.7%). However, more transferred blastocysts from sublethally photosensitized oocytes implanted and then aborted and resorbed compared to untreated controls (28.1% vs. 15.6%, P < 0.05) with no difference in proportions of blastocysts that failed to implant (55.5% vs. 66.6%, respectively). Unexpectedly, blastocysts derived from oocytes that were loaded but not photosensitized implanted and formed significantly lower numbers of viable fetuses than untreated controls (6.47% vs. 17.7%, P < 0.01). This decrease in fetal formation was reflected in a significantly higher proportion of blastocysts that failed to implant compared to untreated controls (79.1% vs. 66.6%), not as an increase in fetal abortion after implantation (14.4% vs. 17.7%; Fig. 5). When expressed as proportions of implantations that formed a fetus, blastocysts originating from photosensitized and nonphotosensitized oocytes (31.0% and 36.6%) were similar to each other, but both were significantly lower than blastocysts derived from untreated oocytes (53.2%, P < 0.05).
|
Fetuses that developed to Day 16.5 postcoitum from blastocysts derived from sublethally photosensitized oocytes were of lower wet weight (0.52 mg, P < 0.05) than fetuses derived from loaded nonphotosensitized (0.59 mg) or untreated (0.60 g) oocytes. In addition, two of the 20 fetuses derived from sublethally photosensitized oocytes derived from an individual foster female were exencephalic (Fig. 6) with lower than average weight (0.37 and 0.35 g).
|
The actual time of onset of increased fetal resorptions in the sublethal photosensitization group could not be determined retrospectively, so this event was represented for convenience as the midpoint (Day 10) between the day of transfer (Day 2.5) and the day of fetus retrieval (Day 16.5). This time of onset after sublethal oocyte photosensitization, together with times of onset of preimplantation developmental arrest resulting from more prolonged treatment (expressed in minutes after treatment, or y) was found to correlate significantly (r2 = 0.96) with the actual photosensitization duration used (expressed in minutes of irradiation, or x). This correlation followed the mathematical relationship y = 2498.4x–1.437 (Fig. 7).
|
DISCUSSION
The method of ooplasmic mitochondrial photosensitization in the oocyte using rhodamine-123 was reemployed in the present study because of its usefulness in inducing mitochondrial damage in a controlled, exposure-dependent manner with defined phenotypic effects. Low-level ooplasmic mitochondrial photosensitization of rhodamine-123 resulted in aberrant cytoplasmic patterning of mitochondria without a detectable alteration in zygote metabolism or blastocyst development, although trophectoderm formation in vitro and fetal formation in utero were both compromised. Mitochondrial photosensitization for periods greater than those shown to be inhibitory to blastocyst development induced preimplantation developmental anomalies sooner.
A 60-sec exposure of oocytes loaded with rhodamine-123 in a previous study did not inhibit formation of pronuclear zygotes after fertilization, but did significantly inhibit blastocyst formation [11]. In contrast, a 2-min irradiation of loaded oocytes in the present study significantly inhibited zygote formation after fertilization. Arguably the most likely scenario to explain this more immediate developmental arrest is a failure of oocyte activation due to an acute inability of ooplasmic mitochondria to support calcium sequestration [1]. In a similar manner to somatic cells (reviewed in [20]), mitochondrial metabolism in the mouse oocytes is closely synchronized with endoplasmic reticulum ATP-dependent transport channels that regulate the uptake and release of cytoplasmic calcium during fertilization [1]. Because a photosensitization period of 1 min resulted in significantly decreased mitochondrial metabolism and subsequent ATP deficiency in zygotes, it is likely that a 2-min duration could cause more spatially widespread metabolic dysfunction, especially in critical ooplasmic regions where mitochondria aggregate with endoplasmic reticulum vesicles for the maintenance of calcium oscillations [21, 22]. In this case, inadequate ATP-dependent calcium sequestration may have promoted ooplasmic calcium overload and subsequent developmental demise by the induction of apoptotic protein cascades instead of oocyte activation cascades [18]. In addition, more widespread metabolic dysfunction after 2-min photosensitization may have further inhibited ATP-dependent and calcium-dependent cell cycle factors in the ooplasm that signal meiotic completion after oocyte activation (reviewed in [23]).
Upon increase of photosensitization to 10 min, oocytes themselves underwent more immediate cell death. The onset of this death at approximately 2 h after oocyte retrieval, as opposed to an immediate demise, may reflect the quiescent state of mitochondrial metabolism in the oocyte [24]. Morphological swelling and eventual lysis of the treated oocytes following 10-min photosensitization could be interpreted as a hallmark of oncotic rather than apoptotic cell death [25]. Oncosis is marked by global organelle damage that operates independently of apoptotic pathways and is more rapid in onset [26]. Both forms of cell death, however, are associated with mitochondrial defects, including mitochondrial swelling and metabolic decline [26, 27] as well as formation of mitochondrial permeability transition pores [28]. Because 60-sec photosensitization durations are sufficient to induce mitochondrial swelling, increased membrane permeability, and decreased metabolism [11], a 10-min duration may have exacerbated this effect and even resulted in complete or irreversible organelle rupture. This would have further promoted leakage of not only rhodamine-123 [11] but other factors such as reactive oxygen species and pro-oxidants from the matrix and intramembrane space into the surrounding cytoplasm, with a consequent increase in nonspecific oxidative damage to other cytoplasmic processes. Gross ooplasmic damage, mitochondrial swelling, and developmental arrest have similarly been reported as a result of oxidative damage in immature mouse oocytes following photosensitization of the mitochondrial fluorophore chloromethoxy-rosamine for durations of up to 30 sec [29].
In contrast to the high oocyte photosensitization doses, a sublethal dose of 20 sec had more subtle and delayed inhibitory effects on preimplantation embryo development. A range of morphological aberrations including altered cytoplasmic patterning, axis asymmetry, focal mitochondrial aggregations, and swelling were evident in normally fertilized zygotes derived from sublethally photosensitized oocytes. Such anomalies have been described previously in association with arrested or delayed development of rodent preimplantation embryos [30, 31]. In spite of the anomalies, zygote mitochondrial metabolism was largely uninhibited, suggesting that the photo-damaged organelles were outnumbered by functionally intact organelles [32, 33]. It was not until the blastocyst stage that developmental retardation following sublethal treatment occurred, manifest as a decrease in trophectoderm cell populations relative to those of untreated controls. Blastocoel formation, however, was not impeded, as indicated by a similarity in blastocyst formation rates between test and control groups. With a larger proportion of blastomeres than the inner cell mass, the tropectoderm is perhaps more likely to incur developmental retardation, perhaps via the random inheritance of uneven or abnormally localized densities of organelles in individual blastomeres, as postulated previously [21, 34, 35]. The axis asymmetries in mitochondrial density observed at the zygote stage may have contributed to alteration of the downstream proliferative capacity and structural symmetry of outer (trophectoderm progenitor) cells of the blastulating morula, promoting some of these blastomeres to instead relocate as inner (inner cell mass progenitor) cells, as described previously [36]. Trophectoderm cells may also have stopped dividing and proceeded to apoptosis, leading to a reduction in cell population sizes [37]. This apoptosis may have been influenced further by the increased rates of naturally occurring apoptosis in this epithelium in the regulation of cell population sizes, under the influence of the inner cell mass [38].
The decrease in trophectoderm cell numbers observed in blastocysts derived from loaded, nonirradiated oocytes compared to untreated controls was unexpected and may reflect a low-level inhibitory effect of rhodamine-123 on ATP synthesis, as reported in studies using somatic cells lines [39]. This inhibitory effect may be related to the higher levels of autofluorescence observed in zygotes derived from untreated control oocytes relative to photosensitized as well as to loaded, nonirradiated oocytes. In agreement with previously reported values [40, 41], levels of ATP and mitochondrial membrane potential in the untreated zygotes were, however, not different, suggesting that any ATP deficiencies caused by rhodamine-123 loading are likely to be acquired during the early cleavage stages. Whether via rhodamine-123 chemical toxicity, or in combination with phototoxicity, such an inhibitory effect may be more pronounced in trophectoderm mitochondria that are morphologically and metabolically more similar to those of somatic cells than those of inner cell mass cells [42].
The more delayed effect of photosensitization for 20-sec durations was also reflected in aberrant postimplantation development of conceptuses. The mechanism and tissue-specificity of the mitochondrial dysfunction responsible for increased abortion and resorption rates of conceptuses is difficult to explain. The decrease in trophectoderm cell numbers resulting from sublethal mitochondrial dysfunction and subsequent apoptosis may be an early marker of downstream compromises in placental development that may adversely influence fetal survival during gestation [43], perhaps indicating a reduced capacity for mitochondrial replication at later stages. Another candidate form of benign mitochondrial dysfunction is the degradation or mutation of mtDNA resulting in a transmissible heteroplasmic state [7, 44]. In preimplantation embryos, mtDNAs are largely inactive until the mitochondria become morphologically mature during embryonic tissue differentiation [45, 46] although total template numbers in bovine oocytes have been correlated with developmental competence [47]. Naturally occurring mtDNA mutations have been identified in humans that result in an array of systemic and tissue-specific pathologies (reviewed in [7, 48]). Importantly, the severity and tissue specificity of these pathologies is reflected in the variability in the proportion of maternally-inherited mutant transcripts that is regulated during oocyte maturation (reviewed in [49, 50]). In somatic cells, mtDNA has been estimated to be 10–20 times more susceptible to mutation than nuclear DNA [51]. Because mtDNA is capable of undergoing large-scale deletions following reactive oxygen species generation by photosensitization [52], it is plausible that the same sort of dysfunction occurred in sublethally photosensitized oocytes. Fetal compromise may therefore have been influenced by metabolic or apoptotic anomalies induced by de novo replication of mtDNAs from damaged templates, especially in developing vital organs [53]. The incidence of exencephaly in selected fetuses derived from sublethally photosensitized oocytes is therefore strongly suggestive of such a tissue-specific effect in this vital, early-differentiating, and metabolically-active organ system. Examples of this tissue-specific effect may include areas of fusing neurepithelium, which have been reported to normally undergo increased apoptosis under regulation by mitogen activated protein kinase kinase kinase (Map3k4), but which fail to fuse and result in a nonpatent neural tube and exencephaly in fetuses harboring a defective Map3k4 gene [54].
Blastocysts derived from oocytes that were loaded with rhodamine-123 without photo-irradiation underwent failure of conception sooner than sublethally photosensitized blastocysts, via failure of implantation. Of relevance to this observation is the direct teratogenic effect of rhodamine-123 when administered transplacentally to gestating female mice [55, 56], with direct inhibitory effects on ATP synthesis in mitochondria extracted from both fetal and maternal tissues in direct relation to fluorophore dose. In relation to oocytes that were loaded with rhodamine-123 and sublethally irradiated, a possible transient upregulation of stress response factors such as antioxidant proteins induced by reactive oxygen species generation after this treatment may have contributed to an absence of immediate implantation failure. Sublethal photosensitization of the fluorophore hypericin in cancer lines has been observed to result in upregulation of heat shock protein 70 (HSPA1B, formerly Hsp 70) and an apparent resistance to further phototoxicity [57]. Similarly, cancer cell lines that are inherently resistant to photosensitization with photofrin, a membrane-specific fluorophore with a high affinity for mitochondria, are also able to significantly upregulate mitochondrial heat shock protein 60 (HSPD1, formerly Hsp60) compared to more sensitive lines [58]. In this way, a protective effect of sublethal photosensitization after rhodamine-123 loading may have resulted in a relative delay in developmental demise of blastocysts derived from oocytes that were sublethally photosensitized, although this effect was transient, as indicated by lower fetal weights only in this group. Despite subtle differences in the potential mechanism of sublethal mitochondrial dysfunction, oocytes loaded with rhodamine-123 with or without photosensitization formed zygotes with similar mitochondrial metabolism, formed blastocysts with similarly decreased trophectoderm numbers, and implanted to form fetuses at similar rates. The fact that both groups were significantly different to untreated controls as far as blastocyst trophectoderm proliferation and fetus formation suggests than any form of low-level biochemical interference with mitochondrial function in the oocyte has negative developmental consequences, especially in conjunction with differentiation events.
The observed failures of conception following sublethal mitochondrial injury in the oocyte resulted in postimplantation pathologies similar to those reported for clinically subfertile women, including recurrent implantation failure or miscarriage [59] and decreased live birth weight [60]. The possible conclusion that mitochondrial dysfunction represents a potential epigenetic factor in the etiology of clinical subfertility in women is, however, made with caution in the absence of direct evidence in human oocytes. Even the most compelling evidence of correlations between mitochondrial function and human oocyte developmental [4, 35] and implantation [5] competence to date remains speculative in regard to direct proof of etiology. Nevertheless, the present study reinforces the value of mitochondrial function as a potentially valuable indicator of oocyte development competence for several reasons. First, degrees of ooplasmic mitochondrial dysfunction can influence developmental competence in a heritable manner with variable temporal effects, as predicted by the Barker hypothesis [8]. Second, mitochondrial function represents an integral part of the multifactorial, maternally inherited, preestablished biochemical state of mature oocytes that defines their developmental competence. Third, mitochondrial function is sensitive to environmental stressors, including those induced by in vitro manipulations and conditions [61], resulting in further depression of oocyte developmental competence. Development of a noninvasive assay of mitochondrial function would therefore be a challenging future task for assisted reproductive technologists.
FOOTNOTES
1 Supported by grant funding from the National Institutes of Child Health and Human Development, external scholarship funding from the Helen Macpherson-Smith Trust, and internal scholarship funding from Monash Immunology and Stem Cell Laboratories. 
2 Correspondence: George A. Thouas, Monash Immunology and Stem Cell Laboratories (MISCL), STRIP–Building 75, Monash University, Wellington Road, Clayton, Victoria 3800, Australia. FAX: 61 3 9905 0680; george.thouas{at}med.monash.edu.au
Received: 13 October 2005.
First decision: 15 November 2005.
Accepted: 1 February 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
