Abstract
Selecting developmentally competent oocytes and zygotes based on their morphology is more often influenced by personal judgments and lacks universal standards. Therefore, this experiment was conducted to investigate the rate of development and mRNA level of dielectrophoretically separated oocytes and zygotes to validate dielectrophoresis (DEP) as non-invasive option for selection of oocytes and zygotes. In the first experiment, metaphase II oocytes with (PB+) and without (PB−) first polar body and zygotes were subjected to DEP at 4 MHz and 450 μm electrode distance and classified into fast, very fast, slow, and very slow depending on the time elapsed to reach one of the electrodes in the electric field. Parthenogenetic activation was employed to monitor the embryonic development of dielectrophoretically classified oocytes. The result revealed that at 6 and 7 days of post-activation, the blastocyst rate of very slow dielectrophoretic PB+ and PB− oocytes was significantly (P < 0.05) lower than other groups. Similarly, in zygotes, the blastocyst rate at 7 days post-insemination was higher (P < 0.05) in the very fast dielectrophoretic categories when compared with the slow and very slow categories. In the second experiment, mRNA level was analyzed in the very fast and very slow dielectrophoretic PB+ oocytes and zygotes respectively using the bovine cDNA microarray. The result showed that 36 and 42 transcripts were differentially regulated between the very fast and very slow dielectrophoretic categories PB+ oocytes and zygotes respectively. In conclusion, dielectrophoretically separated oocytes and zygotes showed difference in the rate of blastocyst development accompanied by difference in transcriptional abundances.
Introduction
Oocytes retrieved from cattle ovaries collected from slaughterhouse are used as a source for embryo in vitro production (IVP). However, these oocytes are extremely heterogeneous in developmental competence and ultimately reduce the efficiency of embryo transfer, nuclear transfer, and the blastocyst yield.
With advancement of IVP, different methods have been employed to select oocytes that yield the highest percentages of blastocysts after fertilization or parthenogenetic activation. The majorities of these methods emphasize on categorization of the cumulus–oocyte complex into different morphological groups to correlate with their developmental competence (Blondin & Sirard 1995, Mayes & Sirard 2001). Moreover, the numbers of blastomeres, the ratio between inner cell mass and trophectoderm cell numbers, metabolic rates, and cryotolerance (Mayes 2002) have been used as indicators of developmental competence of oocytes. Furthermore, selection of oocytes relative to the reproductive status of the animal also increases the in vitro development rates (Chohan & Hunter 2003).
Similar to the oocytes, morphological criteria, namely the position of second polar body, the pronuclear morphology, and the nuclear morphology have been employed to correlate with the developmental capacity of zygotes. For instance, development of good-quality human embryos can be obtained from zygotes with centralized and juxtaposed pronuclear zygotes (Gianaroli et al. 2003).
Although morphological evaluation of oocytes or zygotes is easier to accomplish the routine IVP, selection of competent zygotes based on the morphological parameters, namely polar body enlargement, pronuclear and nuclear morphology (Tesarik & Greco 1999, Gianaroli et al. 2003) requires careful handling and observation under the microscope until both pronuclei and polar body are inspected. Moreover, zygotes with invisible pronuclear (Tesarik & Greco 1999) are excluded from evaluation. Furthermore, morphologically best looking oocytes do not necessarily have the highest developmental competence (Blondin & Sirard 1995) and those oocytes with better cumulus expansions during maturation may not necessarily show the highest blastocyst rate. In addition, morphological criteria are more often influenced by personal judgments and they lack universal standards. Therefore, investigating other non-invasive techniques is required to predict and select competent oocytes and zygotes to increase the efficiency of IVP.
Dielectrophoresis (DEP), the motion of neutral particles due to the application of an external nonuniform electric field, has been a useful non-invasive technique for the extensive manipulation of living cells and DNA (Kadaksham et al. 2004). This method does not require knowledge of cell surface antigens or the use of antibodies, or other reporter molecules to label bioparticles (Burt et al. 1996).
Living cells can be electrically polarized depending on their composition, morphology, and phenotype and frequency of the applied electrical field (Huang et al. 1996). Therefore, different cell types in different physiological states can possess distinctly different dielectric properties that can be utilized for separation (Gascoyne et al. 1992, Wang et al. 1993). DEP has been employed for the separation of dead and live bacteria (Lapizco-Encinas et al. 2004), gram-positive and gram-negative bacteria (Markx et al. 1994), differential analysis of human leukocytes (Yang et al. 2000), manipulation and characterization of red blood cells (Minerick et al. 2003), separation of viable and non-viable cells (Huang et al. 1992), and controlling sperm trajectories to characterize sperm motility and morphology before intracytoplasmic injection (Fuhr et al. 1998). However, the application of DEP in screening of developmentally competent oocytes and zygotes has not been documented. Hence, this experiment was conducted to 1) evaluate the dielectrophoretic behavior of oocytes and zygotes, 2) investigate the relationship between the dielectrophoretic mobility of metaphase II (MII) oocytes and zygotes and their developmental competence, 3) evaluate the size of the blastocyst derived from dielectrophoretically separated zygotes, and 4) investigate the dielectrophoretic behavior of MII oocytes and zygotes in relation to the mRNA expression to generate candidate genes related to the developmental competence.
Materials and Methods
Experiment 1: dielectrophoretic behavior of oocytes and zygotes
Oocyte collection and in vitro maturation
Bovine ovaries were obtained from a nearby slaughterhouse and transported to the laboratory within 2–4 h in a thermos flask containing a 0.9% saline solution supplemented with streptocombin. The cumulus–oocyte complexes (COCs) were aspirated from follicles of 2–6 mm in diameter using a sterilized 5 ml syringe fixed to 18 gauge needle. The COCs were washed thrice with modified parker medium supplemented with 15% estrus cow serum (OCS), 0.5 mM l-glutamine, 0.2 mM pyruvate, 50 mg/ml gentamicin sulfate, and 10 μl/ml follicle-stimulating hormone (Folltropin, Vetrepharm, Canada, Belleville, Canada). A group of 40–50 COCs were then cultured in 400 μl maturation medium covered with mineral oil (Sigma) in four-well dish (Nunc, Roskilde, Denmark). Maturation was performed for 24 h at 39 °C in a humidified atmosphere of 5% CO2. At the end of maturation, the cumulus cells were removed by vortexing the COCs in 200 μl Dulbecco’s PBS (Sigma) supplemented with 2 mg bovine hyalur-onidase enzyme/ml and washed thrice with TCM air (TCM 199 contains l-glutamine and 25 mM HEPES (Sigma)) supplemented with 22 mg/ml pyruvate, 350 mg/ml NaHCO3, 50 mg/ml gentamicin, and 0.1% BSA (Sigma). The oocytes were then classified into two groups based on the presence (PB+) or absence (PB−) of the first polar body for further classification according to their dielectrophoretic behavior.
Dielectrophoretic classification of oocytes and zygotes
Am/phase Lock Generator Model 7056 (Geodätisches Institute Bonn) was used for DEP. The generator was connected to the voltmeter and the dielectrophoretic chamber through coaxial cable 7056. The dielectro-phoretic chamber (Krüss, Hamburg, Germany) consists of base plate with electric connection points, an interchangeable microscope slide, two collapsible holders with two electrodes, and adjusting screw for electrodes distance. The two holders were fixed to the base plate by laminated springs. The electrodes are platinum–iridium wire with a diameter of 200 μm and the two electrodes are rod-shaped and parallel to each other. The distance between the electrodes was adjusted using adjusting screw and for this experiment, it was adjusted to 450 μm.
Prior to DEP, the oocytes and the zygotes were washed thrice in 400 μl of 0.3 M sorbitol solution (osmolarity of 300 mOsm and conductivity of 80 μs/cm and pH 7.0) to avoid any incidence of electric shock in the chamber due to contamination of maturation or culture medium. Afterwards, 400 μl of 0.3 M sorbitol solution (osmolarity of 300 mOsm and conductivity of 80 μs/cm and pH 7.0) were decanted onto the microscope slide to submerge the electrodes. Each of the oocytes or zygotes was then placed on the microscope slide midway between the two electrodes. The center precision was determined by scratching the interchangeable microscope slide at the center, which is the midway between the two electrodes. Therefore, every time, the oocytes or the zygotes were placed at the scratched position (center) so as to keep the precision. One oocyte or zygote was treated at a time. A 14 peak-to-peak volt and 4 MHz frequency electric field was applied. This combination of voltage and electric field frequency was selected based on the preliminary experimental results. Moreover, the repeatability of movement of bovine oocytes after repeated exposure in the dielectrophoretic chamber was tested in the preliminary experiment. Accordingly, measurement of the dielectrophoretic behavior of a single oocyte up to ten times exposure in the electric field showed a (mean ± s.e.m.) 5.0 ± 0.3 s for very fast oocyte, 13.7 ± 0.4 s for fast oocyte, 23.5 ± 0.5 s for slow-moving oocytes, and 38.5 ± 0.3 s for very slow oocytes group. The low standard error mean has revealed the repeatability of measurements of a single oocyte in all the four dielectrophoretic categories. Therefore, in this study, the time elapsed by each of the oocytes or zygotes to reach one of the electrodes was recorded by one-time exposure in the electric field and one oocyte or zygote was treated at a time. Oocytes and zygotes were then classified into four different dielectrophoretic groups according to their speed in the electric field (Table 1). The time point to classify PB+, PB−, and zygotes into different dielectrophoretic groups for monitoring the embryonic development was based on the preliminary experimental results (data not shown).
Parthenogenetic activation of oocytes and in vitro culture
Due to low fertilization efficiency of denuded oocytes, parthenogenetic activation was used to assess the developmental competence of dielectrophoretically classified oocytes. For this, the dielectrophoretic categories of PB+ and PB− (very fast, fast, slow, and very slow) and the control oocytes were washed thrice with TCM air and were parthenogenetically activated with 400 μl of 5 μM ionomycine and 400 μl of 2 mM 6-dimethylaminopurine. After the end of activation, parthenogenetic-activating agents were removed by washing the oocytes thrice using Charles Rosenkrans 1 (CR1) medium. Following this, the oocytes were cultured in CR1 medium supplemented with 10% OCS, 20 μl/ml Eagle’s basal medium (BME), and 10 μl/ml minimum essential medium (MEM). The culture condition was maintained at 39 °C in an incubator with a humidified atmosphere of 5% CO2.
IVP and dielectrophoretic categories of zygotes
In vitro fertilization was performed as described in Tesfaye et al.(2004). Briefly, a group of 50 matured oocytes were transferred into a four-well dish containing 400 μl fertilization medium. Frozen–thawed spermatozoa were swim-up and added to the fertilization medium at a final concentration of 1 × 106 spermatozoa/ml. Oocytes were co-incubated with spermatozoa for 20 h in a humidified atmosphere of 5% CO2 at 39 °C. Afterwards, the cumulus cells were removed by vortexing for 90 s in CR1 culture medium. Cumulus-free zygotes were washed twice in CR1 culture medium and transferred into 400 μl TCM air covered with mineral oil. Following this, the zygotes were washed thrice with the dielectrophoretic emersion medium and subjected to the dielectrophoretic procedure and classified according to their speed of movement into four groups (Table 1). Each dielectrophoretically grouped zygote was incubated in CR1 culture medium supplemented with 10% OCS, 20 μl/ml BME, and 10 μl/ml MEM at 39 °C in a humidified atmosphere of 5% CO2 to monitor the embryonic development.
Data collection
The dielectric behavior of oocytes and zygotes was evaluated by recording the time spent by oocytes and zygotes to reach the electrodes from the centre. The cleavage rate at 3 days of post-parthenogenetic activation (dpa) and the blastocyst rates at 6, 7, 8, and 9 at dpa were recorded. Similarly, for each dielectrophoretic groups of zygotes, the cleavage rate (at 2-day post-insemination (dpi)), the morula rate (at 5 dpi), and the blastocyst rates (at 6, 7, 8, and 9 dpi) were recorded. The diameters of oocytes, zygotes, and blastocysts derived from zygotes were measured using image tool Version 3.0 (University of Texas Health Science, San Antonio, TX, USA).
Experiment 2: transcriptional profiling of dielectro-phoretically separated oocytes and zygotes
In this experiment, the transcriptional abundance of the very fast and very slow dielectrophoretic categories of PB+oocytes and zygotes was analyzed using the bovine cDNA microarray.
RNA isolation and cDNA synthesis
mRNA was isolated from each three biological pools of 20–30 very fast and very slow dielectrophoretic categories of oocytes and zygotes using Dynabead Oligo (dT)25 (Dynal Biotech, Oslo, Norway) following manufacturer’s recommendation. Briefly, each pool of oocytes or zygotes was lysated with 50 μl binding buffer (20 mM Tris–HCl (pH 7.5), 1 M LiCl, 2 mM EDTA (pH 8.0)) at 70 °C for 5 min in a water bath. The cell lysate was incubated with 20 μl Dynabead Oligo (dT)25 suspension at room temperature for 30 min. The bead hybridized with mRNA was washed thrice with washing buffer (10 mM Tris–HCl with pH 7.5, 0.15 mM LiCl, 1 mM EDTA with pH 8.0) and mRNA was eluted with 11 μl RNase-free water.
First-strand cDNA was synthesized in a 20 μl reaction volume. For this, 11 μl mRNA were incubated with 1 μl oligo (dT)21 primer, TCT AGT CGA CGG CCA GTG AAT TGTAATACG ACT CAC TATAGG GCG (T)21, for 3 min at 70 °C to which 8 μl RT mix (4 μl of 5 × first-strand synthesis buffer (50 mM Tris–HCl (pH 8.3), 75 mM KCl, and 3 mM MgCl2), 2 μl of 0.1 mM dithiothreitol (DTT), 1 μl of 0.3 M dNTP, 0.1 μl of 10 U RNase inhibitor (Promega), and 0.5 μl of 200 U superscriptase II (Invitrogen)) was added. This reaction was incubated at 42 °C for 90 min followed by reaction termination at 70 °C for 15 min.
Double-strand cDNA synthesis and in vitro transcription
Double-strand cDNA was synthesized using degenerated oligonucleotide primer (DOP)-PCR master kit (Roche Diagnostics GmbH). For this, 2 μl DOP, 2 μl T7 oligo (dT)23 primer, 40 μl 2 × DOP-PCR master mix, and 16 μl RNase-free water were added to 20 μl single-strand cDNA. The mix was then incubated at 95 °C for 5 min, 94 °C for 1 min, and 30 °C for 2 min. After the end of annealing time, the temperature was jumped to 72 °C within 3 min and continued for 15 cycles at 94 °C for 1 min, 62 °C for 1 min, and 72 °C for 2 min. The reaction was terminated after final extension at 72 °C for 7 min and subjected to phenol–chloroform purification.
In vitro transcription was performed using Ampli-Scribe T7 transcription kit (Epicentre technologies, Oldendorf, Germany). Briefly, 2 μl of 10 × reaction buffer, 4 μl dNTP, 2 μl DTT, and 2 μl T7 RNA polymerase were added to the 10 μl double-strand cDNA and incubated at 42 °C for 3.5 h. After the end of incubation, 1 μl DNase was added and incubated further at 37 °C for 30 min. The amplified RNA (aRNA) was purified using RNeasy mini kit (Qiagen) according to manufacturer’s instruction.
Aminoallyl labeling and dye coupling
The amplified RNA (aRNA) samples of oocytes or zygotes representing very fast and very slow dielec-trophoretic groups were first labeled with aminoallyl dUTP (AA-dUTP) using CyScribe post-labeling kit (Amersham Biosciences). Briefly, 1.5 μl anchored oligo (dT) primer, 1.5 μl random primer, and RNase-free water were added to 2 μg mRNA template to make a 10 μl reaction volume. This reaction was incubated at 70 °C for 5 min and at 25 °C for 10 min consecutively. Following this, 10 μl mix containing 4 μl of 5 × CyScript buffer, 2 μl of 0.1 M DTT, 1.5 μl oligonucleotide mix, 1.5 AA-dUTP, and 1 μl CyScript reverse transcriptase were added to the 10 μl reaction volume and incubated at 42 °C for 90 min. At the end of this reaction, 2 μl of 2.5 M NaOH were added and incubated at 37 °C for 15 min to hydrolyze the mRNA template and then 10 μl of 2 M HEPES-free acid were added. Purification of aminoallyl-labeled cDNA was performed using the CyScribe GFX purification kit (Amersham Biosciences). Cy3 and Cy5 dye coupling and purification of the aminoallyl dUTP-labeled very fast and very slow dielectrophoretic groups of zygotes or oocytes were performed using CyScribe post-labeling kit and CyScribe GFX purification kit (Amersham Biosciences) respectively. Three biological replicates were used for each experiment during hybridization. The dye effect during hybridization was controlled by performing dye swap. The cDNA concentration, Cy3, and Cy5 dye incorporation were measured at 260, 550, and 650 nm wavelength respectively using Ultraspec 2100 pro-u.v./visible spectrophotometer (Amersham Biosciences).
Target clones
The ready-made bovine cDNA array (BlueChip, Department of Animal Science, Laval University, Quebec, Canada; Sirard et al. 2005) was used in this study. This chip consisting of more than 2000 clones (4928 spots) was separated into two subarrays. In addition to the target clones, each subarray of the BlueChip consists of nine clones to be used as a negative control, namely vide (32 spots), alien1 (8 spots), alien2 (8 spots), GFP (8 spots), GFP1 (4 spots), GFP1/2 (4 spots), GFP1/4 (4 spots), GFP1/8 (4 spots), GFP1/16 (4 spots) H2O/DMSO (50 spots), and plant (8 spots). The housekeeping genes, namely tubulin (eight spots), ubiquitin (eight spots), and actin (eight spots) were included in each array to be used as positive controls.
Probe preparation and hybridization
Before hybridizing the samples, the arrayed slides were placed in a corning GAPS II slide container containing 50 ml pre-hybridization buffer (0.5 g BSA (Roche) +0.5 ml of 10% SDS+7.5 ml of 20% SSC+42 ml sterile water) and incubated for 20 min at 55 °C. After the end of incubation, the slides were sequentially washed with boiled water, cold water, and isopropanol followed by the centrifugation at 2000 g for 2 min. Immediately before hybridization, dye-labeled probes were dissolved in 55 μl formamide-based buffer (15 μl hybridization buffer+30 μl of 100% formamide+10 μl distilled water) to which 2.5 μl yeast tRNA and 2.5 μl human cot-DNA (Invitrogen) were added to avoid non-specific hybridization. After denaturing at 95 °C for 5 min, the mix was hybridized to the array and covered with cover slips (Roth, Karlsruhe, Germany). The hybridized array slide was then placed in the hybridization chamber (GFL, Dülmen, Germany) and incubated at 42 °C for 20 h in darkness. At the end of incubation, the slides were sequentially washed for 10 min with 2 × SSC+0.1% SDS, 5 min each with 0.2 × SSC and 0.1% SSC buffers, 1 min each with water and isopropanol and centrifuged at 2000 g for 2 min.
Image capture and array data analysis
Array scanning and image analysis were performed using Axon GenePix 4000B scanner and GenePix Pro-analysis software (version 4.0; Axon Instruments, Foster City, CA, USA) respectively. Locally weighted scatter plot smoothing (LOWESS) fit normalization method was applied to normalize the channel intensity using GPROCESSOR 2.0, a free software developed at Yale University (New Haven, CT, USA). From the merged data, a mean log2-transformed value of (Cy5/Cy3) was calculated for the replicates to obtain one value per clone. This result was submitted to statistical analysis for microarray (SAM), free software developed at Stanford University (Stanford, CA, USA), to obtain the list of differential expressed genes. Heat map and average linkage hierarchical clustering were performed using the PermutMatrix software (Caraux & Pinloche 2004).
Quantitative real-time PCR
A total of seven and nine differentially regulated genes in the very fast and very slow dielectrophoretic categories of PB+ oocytes and zygotes were selected respectively to validate the result of microarray data. Sequence-specific primers (Table 2) were designed using Primer Express v. 2.0 (Applied Biosystems, Foster City, CA, USA). For each primer, a serial dilution of 101–109 copy number or molecules was prepared from the plasmid DNA. Quantification reaction was carried out using independent cDNA samples from three biological pools of 30–40 very fast and very slow dielectrophoretic groups of oocytes or zygotes. The real-time PCR was performed in 20 μl reaction volume containing 9 μl of 2.5 × RealM asterMix/20 × SYBR (Eppendorf, Hamburg, Germany), the specific forward and reverse primers, distilled water, and the serial dilution in ABI PRISM 7000 sequence detection system (Applied Biosystems). The thermal cycling parameter was 2 min at 50 °C, 10 min at 95 °C, 45 cycles of 15 s at 95 °C and 60 s at 60 °C. The transcriptional abundance of the samples was calculated by the relative standard curve method using histone (H2A) as endogenous control.
Statistical analysis
The general linear model procedure of SPSS version 12.0 (SPSS Inc., Chicago, IL, USA) was used to test the significant variation in the cleavage and blastocyst rates by considering dielectrophoretic categories as a fixed effect and the list significant difference t-test was employed to separate means between the dielectro-phoretic categories of oocytes or zygotes.
Results
Experiment 1: the dielectrophoretic behavior of oocytes and zygotes
Uniform time point was selected for comparing the dielectrophoretic behavior between PB+, PB−, and zygotes. For this, the dielectrophoretic behavior data from 948 PB+, 468 PB−, and 1006 zygotes were considered. The result indicates that 15.7% of PB+ oocytes, 13.2% of PB− oocytes, and 15.8% of zygotes reached the electrode within 1–5 s. Furthermore, 30% of PB+ oocytes, 24% of PB− oocytes, and 23% of zygotes spent more than 20 s to reach the electrode indicating that PB+ oocytes have relatively higher proportion of slow-moving population than the PB− oocytes and zygotes (Fig. 1).
Dielectrophoretic behavior in relation to the size of oocytes and zygotes
A correlation analysis was performed between the diameter of the cell and the speed of dielectrophoretic movement in order to validate whether the size of the cell is influencing the dielectrophoretic behavior of oocytes and zygotes. The result showed that the dielectrophoretic speed has no significant correlation to the diameter of both the PB+ (n = 49, P = 0.118, r = 0.23) and PB− (n = 63, P = 0.75, r = 0.63) oocytes, indicating that the movement of the oocytes in the electric field was not related to the size of the cell itself. On the other hand, the speed of zygotes in the electric field tends to show a negative correlation (n = 37, P = 0.06, r = −0.31) to the size of the cell. Nevertheless, the correlation was not statistically significant (P > 0.05).
Dielectrophoretic behavior and developmental competence of oocytes
For investigating the dielectrophoretic behavior of PB+ oocytes in relation to developmental competence, a total of 457 oocytes derived from four biological replicates were subjected to DEP, in which 152, 121, 90, and 94 were classified as very fast, fast, slow, and very slow categories respectively. Moreover, 152 PB+ oocytes were used as control (not exposed to the electric field). The embryonic developmental competence of these dielectrophoretically classified oocytes and the control group was monitored after parthenogenetic activation. Results have shown that the cleavage rate (mean ± s.e.m.) was 89.9 ± 0.3 in very fast, 91.3 ± 1.3 in fast, 88.5 ± 2.7 in slow, 91.7 ± 6.2 in very slow, and 91.5 ± 0.8 in control groups of PB+ oocytes. These differences were not significant (P > 0.05). On the other hand, the very slow moving group had significantly (P < 0.05) lower blas-tocyst rate when compared with the very fast, fast and slow, and the control groups at 6 and 7 dpa (Fig. 2). However, at 8 and 9 dpa, the blastocyst rate was not statistically different (P > 0.05) but tended to be higher in the very fast dielectrophoretic categories.
Similarly, 210 PB− oocytes obtained from a total of four biological replicates were subjected to the DEP procedure to study their dielectrophoretic behavior in relation to developmental competence, in which 54, 100, 50, and 16 were found to be very fast, fast, very slow and slow groups respectively. Moreover, 95 PB− oocytes were used as control. The in vitro embryonic development of these dielectrophoretic groups indicates that the cleavage rate (mean ± s.e.m.) was 72.3 ± 1.9 in very fast, 77.1 ± 4.3 in fast, 93.3 ± 6.7 in slow, 76.4 ± 1.6 in very slow, and 95.5 ± 2.8 in control groups showing significantly (P < 0.05) higher cleavage rate in the slow-moving category and the control groups. The blastocyst rate at 6, 7, 8, and 9 dpa was significantly higher in the very fast and fast categories compared with the slow and very slow groups (Fig. 3).
Dielectrophoretic behavior and developmental competence of zygotes
A total of 940 zygotes from seven biological replications were subjected to dielectrophoretic procedure to study the dielectrophoretic behavior in relation to developmental competence and classified into very fast (n = 329), fast (n = 329), slow (n = 97), and very slow (n = 245) groups. Moreover, 323 zygotes were used as control group. After culturing these dielectrophoretic groups, the cleavage rate of zygotes was monitored at 2 dpi. The result showed that the cleavage rate was 74.3 ± 3.1 in very fast, 75.5 ± 7.1 in fast, 71.4 ± 6.7 slow, 72.2 ± 7.6 in very slow dielectrophoretic groups, and 77.0 ± 3.6 in control groups and differences were not significant (P > 0.05), indicating that although zygotes have different dielectric properties, separating the zygotes according to their dielectric properties has no significant effect on cleavage rates. The blastocyst rate at 7 dpi was significantly (P < 0.05) higher in the very fast (16.1 ± 2.2) than in the slow (9.1 ± 2.5) and very slow (10.6 ± 1.9) groups. However, it was not significantly (P > 0.05) higher than the fast (12.2 ± 4.1) and control (12.3 ± 2.7; Fig. 4) groups. The total blastocyst rate at 9 dpi was not significantly (P > 0.05) different between the dielectrophoretic categories of zygotes (Fig. 4).
The dielectrophoretic behavior of zygotes in relation to blastocyst size
The very fast dielectrophoretic categories of zygotes yield significantly (P < 0.05) larger sized blastocysts (179.8 ± 1.9 μm, n = 74) at 7 dpi compared with fast (168.5 ± 1.8 μm, n = 68), slow (168.9 ± 1.7 μm, n = 12), very slow (168.8 ± 1.7 μm, n = 52), and control (173.7 ± 1.5 μm, n = 55) groups. Moreover, the diameter of blastocyst that appears at 8 dpi tends to be bigger in the blastocysts derived from the very fast (173.5 ± 1.9 μm, n = 40) group when compared with the fast (170.1 ± 3.7 μm, n = 26), slow (172.4 ± 3.9 μm, n = 12), and very slow (169.4 ± 2.3 μm, n = 47) groups but tends to be lower than the control (179.3 ± 4.0 μm, n = 30) group.
Experiment 2: transcriptional profiling of dielectrophoretically separated oocytes
The microarray analysis between the very fast and the very slow oocytes revealed that out of 2000 transcripts, 713 were detected in both groups. The expression intensities (heat map) of these transcripts are shown in Fig. 5A. Scatter plot diagram was plotted for the log2 value of Cy5 (very fast oocytes) total intensity against the log2 value of Cy3 (very slow oocytes). Similar scatter plot was generated for the dye swap by plotting the log2 value of Cy5 (very slow oocyte) total intensity against the log value of Cy3 (very fast oocytes). The coefficient of determination (R2) for the target and dye reversal hybridization was 0.941 and 0.933 respectively. Similar value of R2 reveals the absence of the dye effect during the analysis.
The data analysis performed by SAM revealed that out of 713 commonly detected transcripts, 36 genes including the novel transcripts were found to be significantly differentially expressed in the very fast and very slow oocytes. Of these, 31 and 5 were up- and down-regulated respectively in the very fast group when compared with the very slow dielectrophoretic categories of oocytes. The average linkage clustering analysis (Fig. 5B) reveals that there were many subgroups within the up- or down-regulated genes (or clusters) sharing similar expression patterns. Such co-expressed genes may possibly be sharing common biological function. The magnitude of differentially regulated genes revealed by SAM showed that the expression level of eight genes was 3.2–5.0 times, 11 genes was 2.0–2.8 and 12 genes was 1.5–1.9 times higher in the very fast compared with the very slow dielectrophoretic categories of oocytes. Moreover, the expression level of five genes was from 1.7 to 2 times lower in the very fast compared with the very slow dielectrophoretic categories (Fig. 5B). The functions of differentially expressed transcripts were determined using gene ontology (http://www.geneontology.org/), NCBI (http://www.ncbi.nlm.nih.gov), and gene cards (http://www.genecards.org/index.shtml). Accordingly, 63.9% of the differentially regulated genes have known function involving in rRNA binding and structural constituent of ribosome, cell cycle regulation, protein and ion binding, metabolic regulation, including amino acid metabolism and other different functions (Fig. 6). The lists and functions of all differentially regulated genes are indicated in Tables 3 and 4.
Transcriptional profiling of dielectrophoretically separated zygotes
Similar to the oocytes, six hybridizations (three biological and three technical replicates as dye swap) were carried out between the very fast and the very slow dielectro-phoretic categories of zygotes. After performing LOWESS normalization, the scatter plot diagram was plotted for the log2 value of Cy5 total intensity against the log2 value of Cy3 as described previously for the oocytes. The coefficient of determination (R2) for the target and dye reversal hybridization was 0.9137 and 0.9274 respectively, showing the absence of the dye effect during the analysis.
A total of 793 transcripts were commonly detected in both zygote groups. The expression intensities (heat map) of these transcripts are shown in Fig. 7A. The SAM analysis confirmed a total of 42 differentially expressed transcripts among all the detected transcripts. Of these, 25 and 17 genes were up- and down-regulated respectively in the very fast zygotes. The average linkage hierarchical clustering, relative fold change difference, and the q (%) value of differentially regulated genes are shown in Fig. 7B.
The function of differentially expressed genes was described using the same source described for oocytes. Accordingly, some transcripts are involved in ion binding (ZNF85, ZNF519, and NANOS1), regulation of cell cycle (NASP, DDX10, SMARCA5, and AURKA), and signal transduction (RALA). The lists and functions of all differentially expressed genes are presented in Tables 5 and 6 and their ontological classification is shown in Fig. 8.
Validation of differentially regulated genes with quantitative real-time PCR
In oocytes, the validation results confirm that seven out of seven transcripts were found to be in accordance with the microarray data indicating higher validity of the expression pattern which was observed in the microarray data. This higher confirmation rate was brought by the fact that differentially regulated genes were selected after it was confirmed by three dye-swap replications. However, the fold change difference obtained by real-time PCR is much higher than the microarray data. In microarray data, the maximum fold change difference between the very fast and the very slow oocytes was fivefold upregulation in the very fast oocytes. Nevertheless, the maximum fold change obtained in real-time PCR was 72 and the minimum was 6 (Fig. 9) and the correlation coefficient of the microarray data and the real-time PCR was 0.458.
Similarly, in zygotes, the real-time PCR results revealed that eight out of nine transcripts were in accordance with the microarray data (Fig. 10). Furthermore, the coefficient of correlation between the microarray result and the real-time PCR was 0.552.
Discussion
Dielectrophoretic behavior of oocytes and zygotes
Understanding the mechanism of dielectrophoretic mobility of oocytes and zygotes enables appropriate validation of DEP technique to be used as one of the non-invasive tools for selecting developmentally competent oocytes or zygotes. The movement of the cells toward the higher electric field was driven by the higher polarizable ability of oocytes and zygotes compared the sorbitol solution used as dielectrophoretic emersion medium (conduction medium). It is also stated that the cell or particle can be pulled toward the higher electric field or repelled away depending on the medium used during DEP (Ramírez et al. 2003). As the conductivity of the medium is lower than the cells, the cells are more polarized than the medium and therefore the cells migrate toward the electrode (higher electric fields; Green & Morgan 1997, Goater & Pethig 1998). The movement toward the higher electric field is usually known as a positive DEP (Poh et al. 1981, Ikeda et al. 2005, Lao & Hsing 2005). Therefore, in our experimental setup, the oocytes and the zygotes also exhibited a positive DEP because the medium used during DEP has a conductivity of 80 μs/cm which triggers the oocytes and the zygotes to be more polarizable than the medium.
In view of the dielectrophoretic behavior, both the oocytes and zygotes tended to exhibit a similar dielectric behavior. However, the PB+ oocytes have relatively higher proportion of slow-moving population when compared with zygotes and PB− oocytes (Fig. 1). This may be due to the lower electrical conductivity of the PB+ oocytes associated with the declining porosity in oocyte membrane during maturation by losing the K+ permeable channel (Lansman 1983). Moreover, a decrease in plasma membrane conductivity as the oocyte progress to the MII stage is reported (Tosti & Boni 2004). Hence, the oocytes that extruded the first polar body during maturation were the ones which completed the maturational process and entered the MII stages.
Examining the causes attributing for differences in dielectrophoretic mobility of oocytes and zygotes can be useful to understand the electrical and physiological properties of the cells. The difference in dielectrophore-tic mobility within bovine oocyte and/or zygotes can be driven by their intrinsic property within the cell. As reviewed by many authors, the features of a cell, including the morphology, the cytoplasm content (Arnold & Zimmerman 1988), the conductivity and permittivity of the cytoplasm (Wanichapichart et al. 2002), the membrane morphology and properties (Wang et al. 1997), the cell size (Pethig 1991, Goater & Pethig 1998), shape and composition of the plasma membrane, and the cell cytoplasm and nucleus (Archer et al. 1999) can be the major factors resulting in differences in dielectrophoretic mobility. However, in this experiment, it has been shown that the speed of oocytes and zygotes in the electric field was not related to the size of the cells as confirmed by the correlation analysis.
Difference in the cytoplasmic conductivity is the major contributor for the variation in cell conductivity at frequencies higher than 1 MHz (Docoslis & Alexandridis 2002). Hence, bovine oocytes and zygotes with different cytoplasmic contents can possess different tendencies of polarization and speed of migration in the electric field. Furthermore, the variability in the speed within the zygotes can be attributed by difference in the accumulation of ions during and after fertilization. Fertilization causes a constant increase in the ionic conductance of the plasma membrane (Jaffe et al. 1983). The concentration of ions may foster some of the cells to acquire charges and induce dipole. Based on this notation, we can explain that higher ionic concentration on the plasma membranes after fertilization in some of the zygotes may have triggered the motion of the zygotes. Similarly, a report by Arnold et al.(1989) revealed an increase in the cell membrane conductivity of rabbit oocyte when compared with the unfertilized oocyte as a result of changing the membrane structure due to fertilization. This may lead to the conclusion that zygotes which moved very fast in the electric field might be those derived from successfully fertilized oocytes.
Dielectrophoretic behavior, developmental competence, and mRNA expression of oocytes and zygotes
Unpublished data in our laboratory on bovine oocytes show that zona-free and zona-intact oocytes have the same moving time in the electric field showing the absence of the zona pellucida effect on the dielectro-phoretic mobilities of oocytes. This in turn can suggest that the cytoplasmic content may govern the dielectrophoretic mobility. Therefore, the difference in dielectrophoretic mobility of individual oocytes or zygotes can be driven by variation in the abundance of transcripts in the cytoplasm that can be associated with the modification in morphology of a cell. For instance, Cristofanilli et al.(2002) noted the difference in dielectric properties that have occurred in MCF-7 sublines because of the overexpression of HER-2/ neuoncogen. Moreover, after separating cell membrane, dielectric properties of five different cultivated cell lines and human peripheral blood mononuclear cells, it was shown that the gene expression profiles of the post-separated cells were significantly different from those of the pre-separated cell mixtures (Huang et al. 2002). Similarly, the expression analysis in our experiment showed that PTGS2, DNA (cytosine 5) methyltransferase 1 (DNMT1), and ANXA2 (Table 3) were abundant in fast-moving oocytes; and ZNF85, ZNF519, and NANOS1 (Table 5) were enriching very fast zygotes and all these genes are believed to play in ion binding. In addition, IQGAP1, which is enriched in the very fast dielectro-phoretic zygotes, is involved in cell motility. As reported by Mataraza et al.(2003), the overexpression of IQGAP1 in mammalian cells can enhance cell migration and cell motility. However, the involvement of IQGAP1 in passive migration of oocytes in the electric field needs further investigation.
In addition to investigating the electrical behavior of the bovine MII oocytes and zygotes, looking into the developmental behavior is paramount important to predict the validity of the dielectrophoretic technique as the one non-invasive option for selecting developmentally competent oocytes and zygotes. In this regard, our result revealed a significant difference in the rate of development at 6 and 7 dpa of dielectrophoretic separated PB+ oocytes and at 6, 7, 8, and 9 dpa of PB− oocytes. Furthermore, at 7 dpi, the very fast dielectro-phoretic categories of zygotes yielded significantly higher number of blastocysts compared with the very slow dielectrophoretic groups of zygote. Therefore, this variation in the rate of development of oocyte and zygote of different dielectrophoretic categories is an indicator of the tendency of dielectrophoretic procedure to discriminate oocytes and/or zygotes according to the physiological status.
The critical point remaining as a challenge is to find out the fundamental causes which trigger the very fast moving dielectrophoretic categories of PB+ oocytes to become developmentally competent at 6 and 7 dpa compared with the very slow moving counter parts, despite the total blastocyst yield at 9 dpa, which was not significantly different.
In living cells, transmembrane enzymes can likely absorb free energy from an oscillating electric field and transduce it to chemical energy (Westerhoff et al. 1986). Similarly, the application of a 1 or 10 Hz electric field to human hepatoma cells was found to induce a fourfold increase in Ca2+ within 30 min of continuous field exposure (Cho et al. 1999). However, these assertions cannot signify the differences for developmental competence of the very fast and very slow dielectrophoretic groups of oocytes and zygotes as both groups were exposed to electric field. Therefore, investigating their molecular basis was essential to evaluate the differences in the cytoplasmic content. In order to get insight of the molecular differences between these groups of oocytes and zygotes, transcriptional analysis of the very fast and very slow oocytes or zygotes was performed using the bovine cDNA microarray.
The transcriptional abundance and developmental competence of oocytes has been assessed by many authors (Lonergan et al. 2003, Lequarre et al. 2004). High rates of RNA and protein synthesis, storage of ribosome mRNAs and protein in ooplasm during the oocyte growth are needed in high developmental capacity (Eichenlaub-Ritter & Peschke 2002). Since the molecular mechanism related to the development involves mRNA and protein synthesis, degradation and modification are required to enrich the oocyte with sufficient molecular stores for triggering the development to the start of embryonic genome activation (Donnison & Pfeffe 2004). Similarly, our microarray experiments showed that 86 and 60% of the differentially regulated genes were found to be more abundant in the very fast moving oocytes and zygotes respectively revealing the differences in the accumulation of specific transcripts in oocytes and zygotes with different developmental competences.
Among the transcripts enriching the very fast moving oocytes, RPL2, RPL8, RPL35, RPLP0, LOC52280L, and LOC390110 (Table 3; Fig. 5B) are known to be involved in amino acid metabolism, biosynthesis biological process, and encode ribosomal protein.
ANXA2, a Ca2+- and phospholipid-binding protein (Dreier et al. 1998, Filipenko et al. 2004), found to be higher in competent oocytes (Costa et al. 2006), in virally transformed cell lines and in human tumors (Filipenko et al. 2004) and involved in ion channel activity (Burger et al. 2006). Similarly, ANXA2 was abundant in the very fast moving categories of oocytes which exhibited higher blastocyst rate at 6 and 7 dpa.
Others, namely CDC91L1, NUSAP1, FAM58A, and CKS1B which are known to be involved in cell division were also upregulated in the very fast dielectrophoretic categories (Table 5). For instance, cells transfected with CDC91L1 gene grew at a faster rate and had increased anchorage-independent growth capability compared with control cells (Guo et al. 2004). Furthermore, the suppression of NUSAP1 in a cell by RNA interference has shown mitotic defects, aberrant mitotic spindles, defective chromosome segregation, and cytokinesis that interfere with normal cell cycle progression (Raemaekers et al. 2003). In addition, these authors noticed the upregulated NUSAP expression in proliferating cells during G2/M-phase of the cell cycle. The higher expressional pattern of these genes in the very fast dielectrophoretic oocytes might have fostered the rate of embryonic development at days 6 and 7 dpa due to co-ordinated and proper cell programing and cell division compared with the very slow groups where these genes are downregulated.
Similarly, differentially expressed genes, including NASP, DDX10, IQGAP, SMARCA5, RGS2, and AURKA (aurora kinase A) were more abundant in the very fast zygotes that exhibited significantly higher blastocyst rate at 7 dpi compared with the very slow dielectrophoretic categories. These genes are believed to be involved in cell division and progression, transcriptional regulation, and other similar functions. For instance, the nuclear auto-antigenic sperm protein (NASP), a homolog of the N1/N2 gene expressed in oocytes of Xenopus laevis, was found in all dividing cells that is regulated by the cell cycle (Richardson et al. 2000). Moreover, the human somatic and testis NASP mRNAs are expressed in all the transformed cell lines and human tumors revealing the involvement of NASP in the cell cycle of dividing cells (Richardson et al. 2001).
The DDX10 gene belongs to the DEAD-box proteins and involves translation initiation, nuclear and mRNA splicing, ribosome and spliceosome assembly, and mRNA stability (Schmid & Linder 1992, Iost and Dreyfus 1994, Arai et al. 1997).
SMARCA5, which contains an open reading frame of 3156 nucleotides encoding a 1052 amino acid peptide (Aihara et al. 1998), also upregulated in the very fast moving zygotes and plays a central role in eukaryotic transcriptional regulation (Mohrmann et al. 2004).
IQGAP binds calmodulin, E-cadherin and involves a crucial role in transuding Cdc42 signaling to the cytoskeleton (Sokol et al. 2001, Swart-Mataraza et al. 2002). The overexpression of IQGAP1 can significantly increase the amount of active Cdc42 in embryonic cells which in turn control the organization of the cytoskeleton, cell cycle progression, and gene transcription (Sokol et al. 2001). The higher expression of IQGAP in the very fast zygote may be one of the transcripts for fast development of the very fast zygotes compared with the very slow zygotes. Moreover, RGS2, a regulator of G-protein signaling family member (Zmijewski et al. 2001), possibly involved in regulating the intracellular Ca2+ mobilization and T-cell proliferation at the maternal–fetal interface during embryo implantation (Huang et al. 2003). The higher expression level of this gene in the very fast moving zygotes exhibited higher rate of development at 7 dpi may indicate the involvement of RGS2 in growth and development of pre-implantation embryo.
The very fast zygotes are also enriched with AURKA. It is involved in cell cycle control by regulating M-phase events. In mammals, AURKA is located at the mitotic spindle to regulate the centrosome and mitotic micro-tubules (Udayakumar et al. 2006). Furthermore, it is believed to be involved in mitotic control (Prigent & Giet 2003), which means that the AURKA is required for centrosome maturation, proper chromosome segregation, and assembly and maintenance of a bipolar spindle (Satinover et al. 2006).
DNMT1 is one of the genes enriching both the very fast moving oocytes and zygotes. DNMT1 is thought to maintain genomic methylation through DNA replication (Grohmann et al. 1997). The loss of phenotype associated with reduction in DNMT1 has been reported to reveal the importance of DNMT1 for mammalian cell development and growth. For instance, the loss of DNMT1 from somatic cells causes p53-dependent apoptosis (Jackson-Grusby et al. 2001). Depletion of DNMT1 from X. laevis embryos causes embryonic lethality and inappropriate gene expression (Stancheva & Meehan 2000). Similarly, the lower level of blastocyst of the very fast oocytes at 6 and 7 dpa and at the very slow zygotes may be attributed by deficient reprograming and inefficient demethylation as a result of low level of DNMT1.
In conclusion, the difference in the rate of development between the dielectrophoretically categorized oocytes and zygotes is accompanied by the differences in transcriptional abundance. Furthermore, the results showed the potential of DEP to be applied as a non-invasive tool to screen developmentally competent oocytes and embryos.
Dielectrophoretic categories of oocytes and zygotes depending on the time elapsed to reach one of the electrodes in the electric field.
| Dielectrophoretic categories | |||||
|---|---|---|---|---|---|
| Developmental stages | Very fast (s) | Fast (s) | Slow (s) | Very slow (s) | Control |
| Control, oocytes and zygotes which were not exposed to dielectrophoresis; s, seconds. | |||||
| PB+ oocytes | 1–9 | 10–15 | 16–28 | >29 | – |
| PB− oocytes | 1–8 | 9–15 | 16–19 | >20 | – |
| Zygotes | 1–6 | 7–15 | 16–20 | >21 | – |
Sequence of primers used for quantification of differentially expressed genes.
| Gene name | Accession no. | Primer sequence (5′–3′) | Product length | Development stages |
|---|---|---|---|---|
| aGenes differentially regulated in oocyte and zygote categories. | ||||
| RPLP0 | AB098748 | For: GAATCCCATATCCTCGTCCGA Rev: TTGCTGAAAAGGTCAAGGCCT | 162 | Oocytes |
| MGC128854 | BC103473 | For: GGGAGGGCTCAACTTCAACAC Rev: CATGGCTCTCATGGCATTTG | 182 | Oocytes |
| GCNT3 | AY283766 | For: AGCGGTTGCCTTTCACTGAC Rev: CATACACAGCTCGCAGCAGC | 183 | Oocytes |
| PGHS-2 | AF031698 | For: GGAAATCCTCAGCTCAAAAC Rev: GTCTCCCTGGGAACTATTC | 221 | Oocytes |
| DNMT1a | NM_182651 | For: TCGGGAAGCGAATGGATGT Rev: TGATGGTGGTCTGCCTGGT | 223 | Oocytes and zygotes |
| KRT8a | X12877 | For: CACCAGTTCCAAGCCTGTGG Rev: TCAGGTCTCCTGTGCAGATGC | 176 | Oocytes and zygotes |
| ANXA2 | NM_174716 | For: CGTGCTCCAGCTAACAGTTCT Rev: GGAAAGCCAGGTAATGCGTA | 139 | Oocyte |
| DDX10 | NM_004398 | For: TCGGCTTCTTCAACACATGG Rev: CAAACTCAAGCGTGCAAGGTC | 198 | Zygotes |
| NASP | BT006757 | For: CCTAGAGCTTGCCTGGGATATG Rev: TCGTGGGCTTCCAGGTACTG | 198 | Zygotes |
| ARL6IP | BC010281 | For: AGGCGCTAGAATGGGAACAAG Rev: GGCTGCAGAGACTGCAAGTCT | 232 | Zygotes |
| SMARCA5 | NM_003601 | For: AGTGAACTTTCGCCCATCTTG Rev: AGGCTTGTGGATCAGAATCTG | 194 | Zygotes |
| RGS2 | NM_002923 | For: GCAGTTGTAAAGCAGCCACTTG Rev: ATATTGAATTCTGGCTGGCCTG | 193 | Zygotes |
| RPS18 | AB098899 | For: TTCTTGGACACACCCACAGTG Rev: GGACGTGAAGGACGGGAAATA | 189 | Zygotes |
| ZP3 | NM_173974 | For: CCACCTGTTGGAGGACTTGC Rev: TCTCACCGATGCCTCCTCTG | 161 | Zygotes |
| H2a | NM_178409 | For: CTCGTCACTTGCAACTTGCTATTC Rev: CCAGGCATCCTTTAGACAGTCTTC | 148 | |
List of upregulated genes in the very fast dielectrophoretic categories of PB+ oocytes and their function.
| Accession no. | Gene bank name | Function |
|---|---|---|
| BC000047 | Homo sapiens ribosomal protein L8, mRNA (cDNA clone MAGE:3504599) (RPL8) | rRNA binding, structural constituents of ribosome |
| XM_372375 | Predicted: Homo sapiens similar to 1-aminocyclopropane-1-carboxylate synthase (LOC390110) | Amino acid metabolism, biosynthesis |
| BC062798 | Homo sapiens chromosome 6 open reading frame 173 (C6orf173) | Unknown |
| XM_230305 | Rattus norvegicus similar to 2610203E10Rik protein (LOC311218) | Unknown |
| XM_601090 | Predicted: Bos taurus similar to 1-aminocyclopropane-1-carboxylate synthase (LOC522802) | Amino acid metabolism, biosynthesis |
| BC103473 | Bos taurus similar to RNA-binding motif protein 3 (MGC128854) | RNA binding |
| BC004170 | Homo sapiens polymerase (DNA directed), epsilon 3 p17 subunit (POLE) | Protein binding, epsilon DNA polymerase activity |
| NM_174716 | Bos taurus annexin A2, mRNA (ANXA2) | Calcium- and phospholipid binding |
| BC032121 | Homo sapiens family with sequence similarity 58 member A (FAM58A) | Regulation of cell cycle |
| BC011008 | Homo sapiens nucleolar and spindle-associated protein 1 (NUSAP1) | Cytokinesis, microtubule binding |
| X12877 | Bovine mRNA fragment for cytokeratin A8 (KRT8) | Protein binding |
| U38619 | Sus scrofa small intestine diadenosine tetraphosphatase (NUDT2) | Tetraphosphatase activity |
| BC007751 | Homo sapiens CDC28 protein kinase regulatory subunit 1B (CKS1B) | Cell division, protein kinase activity |
| NM_174788 | Bos taurus ribosomal protein, large (RPLP2) | rRNA binding, structural constituents of ribosome |
| BC003185 | Homo sapiens coenzyme Q7 homolog, ubiquinone (yeast) (COQ7) | Metabolic regulator |
| AY550044 | Sus scrofa 60S ribosomal protein L35(RPL35) | rRNA binding, structural constituents of ribosome |
| AY283766 | Bos taurus type 4 mucus-type core 2 β-1,6-N-acetylglucosaminyltransferase (GCNT3) | Acetylglucosaminyltransferase activity |
| AB098748 | Bos taurus mRNA similar to acidic ribosomal phosphoprotein PO (RPLP0) | rRNA binding and structural constituents of ribosome |
| BC030512 | Homo sapiens CDC91 cell division cycle 91-like 1 (S. cerevisiae) (CDC91L1) | Cell division control, GPI anchor binding |
| XM_230305 | Rattus norvegicus similar to 2610203E10Rik protein (RGD 309314_predicted) | Unknown |
| AF031698 | Bos taurus prostaglandin G/H synthase-2 (PGHS-2) or (PTGS2) | Ion binding, cell motility, and peroxidase activity |
| AB099147 | Bos taurus mRNA similar to 16S RNA clone | Unknown |
| AF091090 | Homo sapiens clone 669 unknown mRNA, complete sequence (RP11- 82K18.3) | Carboxylate synthase activity |
| NM_182651 | Bos taurus DNA (cytosine 5) methyltransferase 1 (DNMT1) | Zinc ion binding, DNA methylation |
| X54703 | Bovine mRNA for pancreatic anionic trypsinogen | Unknown |
| BC062798 | Homo sapiens cDNA clone IMAGE: 6495746 | Unknown |
| NM_031706 | Rattus norvegicus ribosomal protein S8 (Rps8) | Structural constituents of ribosome |
List of downregulated genes in the very fast dielectrophoretic categories of PB+ oocytes and their function.
| Accession no. | Gene bank name | Function |
|---|---|---|
| AC083812 | Homo sapiens 12 BAC RP11-288D9 | Amino acid dephosphorylation |
| BC039831 | Homo sapiens intraflagellar transport 52 homolog (Chlamydomonas), mRNA (cDNA clone MGC:48784 IMAGE:6095634) (IFT52) | Unknown |
| BC000852 | Homo sapiens RAN, member RAS oncogene family (RAN) | Protein binding, transcription coactivator activity |
| XM_595660 | Predicted: Bos taurus similar to PHD finger protein 22 (LOC517489) | Unknown |
| NM_130809 | Homo sapiens hypothetical protein MGC12103 (LOC133619) | Unknown |
List of upregulated genes in the very fast dielectrophoretic categories of zygotes and their function.
| Accession no. | Gene bank name | Function |
|---|---|---|
| NM_004398 | Homo sapiens DEAD (Asp–Glu–Ala–Asp) box polypeptide 10, mRNA (DDX10) | Cellular growth and division, embryogenesis, RNA helicase activity |
| BT006757 | Homo sapiens nuclear auto-antigenic sperm protein (histone binding; NASP) | Encodes a H1 histone-binding protein, cell division |
| NM_182651 | Bos taurus DNA (cytosine 5) methyltransferase 1, mRNA (DNMT1) | Metal ion binding, zinc ion binding, DNA methylation |
| BC010281 | Homo sapiens ADP-ribosylation factor-like 6-interacting protein (ARL6IP) | Protein binding |
| NM_003870 | Homo sapiens IQ motif containing GTPase-activating protein 1 (IQGAP1) | Calmodulin binding |
| NM_003601 | Homo sapiens SWI/SNF related, matrix associated, actin dependent (SMARCA5) | Replication of pericentric heterochromatin in S-phase, regulate transcription |
| BK001637 | TPA: Mus musculus regulator of sex-limitation candidate 2 (Rslcan2) | Unknown |
| BC039858 | Homo sapiens v-ral simian leukemia viral oncogene homolog A (ras) RALA | Signal transduction, GTP binding |
| BC047646 | Homo sapiens zinc finger protein 85 (HPF4, HTF1), mRNA (cDNA clone ZNF85) | Metal ion (zinc) binding, nucleic acid binding, transcription factor activity |
| AC150515 | Bos taurus BAC CH240-118E9 (Children’s Hospital Oakland Research) | Unknown |
| NM_018158 | Homo sapiens solute carrier family 4 (anion exchanger) member 1 (SLC4A1AP) | Double-stranded RNA binding |
| AF275269 | Homo sapiens NANOS1 mRNA, complete cds | Zinc ion binding, RNA binding, regulation of translation |
| XM_132350 | Mus musculus RIKEN cDNA 4930438O03 gene (4930438O03Rik), mRNA | Unknown |
| AC096629 | Bos taurus clone rp42-152a4, complete sequence | Unknown |
| BC000764 | Homo sapiens chromosome 6 open reading frame 166, mRNA (cDNA clone (C6orf166) | Unknown |
| BC024227 | Homo sapiens zinc finger protein 519, mRNA (cDNA clone MGC:27166 (ZNF519)) | Metal ion binding, DNA binding, regulation of transcription |
| BC018147 | Homo sapiens senescence downregulated leo1-like (LEO1) | Unknown |
| NM_002923 | Homo sapiens regulator of G-protein signaling 2, 24 kDa, mRNA (RGS2) | Signal transducer, GTPase activator activity, calmodulin binding |
| NM_032598 | Homo sapiens testes development-related NYD-SP20 (NYD-SP20) | Unknown |
| BC042755 | Homo sapiens regulator of G-protein signaling 2, 24 kDa, mRNA cDNA (RGS2) | Signal transducer activity, calmodulin binding, GTPase activator activity |
| AC098686 | Bos taurus clone rp42-146c17, complete sequence | Unknown |
| BC063533 | Homo sapiens apoptosis, caspase activation inhibitor, mRNA cDNA (AVEN) | Protein binding, anti-apoptosis |
| AF312827 | Homo sapiens tectorin-β (TECTB) mRNA, complete cds | Extracellular matrix structural constituent |
| BC027464 | Homo sapiens aurora kinase A (AURKA) | Cell cycle regulation |
List of downregulated genes in the very fast dielectrophoretic categories of zygotes and their function.
| Accession no. | Gene bank name | Function |
|---|---|---|
| BC000047 | Homo sapiens ribosomal protein L8, mRNA (RPL8) | rRNA binding, structural constituents of ribosome |
| AY283766 | Bos taurus type 4 mucus-type core 2 | Carbohydrate metabolism |
| NM_033301 | Homo sapiens ribosomal protein L8, transcript variant 2 (RPL8) | rRNA binding, structural constituents of ribosome |
| AB098899 | Bos taurus mRNA similar to ribosomal protein S18, partial cds (RPS18) | rRNA binding, structural constituents of ribosome |
| BC019262 | Homo sapiens Huntingtin-interacting protein K (HYPK) | Protein binding |
| AB098748 | Bos taurus mRNA similar to acidic ribosomal phosphoprotein PO (RPLP0) | rRNA binding, structural constituents of ribosome |
| BC062798 | Homo sapiens cDNA clone IMAGE:6495746, partial cds (C6orf173) | Unknown |
| NM_031706 | Rattus norvegicus ribosomal protein S8 (Rps8) | Structural constituents of ribosome |
| BC004170 | Homo sapiens polymerase (DNA directed), epsilon 3 (p17 subunit) | DNA transcription, replication, and packaging |
| AC110780 | Homo sapiens BAC clone RP11-357A22 from 2 | Amino acid dephosphorylation |
| AB099089 | Bos taurus mRNA similar to ribosomal protein S14 (RPS14) | rRNA binding, structural constituents of ribosome |
| BC003751 | Mus musculus degenerative spermatocyte homolog Drosophila (Degs1) | Zinc ion binding, oxidoreductase activity |
| AC012175 | Homo sapiens chromosome 16 clone RP11-19H6 | Unknown |
| NM_173974 | Bos taurus zona pellucida glycoprotein 3 (ZP3) | Prevents post-fertilization polyspermy |
| X06423 | Rat mRNA for ribosomal protein S8 (RPS8) | Structural constituents of ribosome |
| NM_174731 | Bos taurus Finkel-Biskis-Reilly murine sarcoma virus (FBR-MuSV) | Structural constituent of ribosome |
Proportion of oocytes and zygotes in relation to the time spent to reach one of the electrodes in the electric field.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Blastocyst rate of dielectrophoretic categories of PB+oocytes. Bars with different letters on the same stage differ from each other statistically (P < 0.05).
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Blastocyst rate of dielectrophoretic categories of PB−oocytes. Bars with different letters on the same stage differ from each other statistically (P < 0.05).
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Blastocyst rate of dielectrophoretic categories of zygotes. Bars with different letters on the same stage differ from each other statistically (P < 0.05).
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Detected and deferentially expressed transcripts in the very fast and very slow dielectrophoretic categories of PB+ oocytes. (A) Heat map of 713 normalized transcripts. D, E, and F describe the dye swap of A, B, and C. The 635/532 log2 ratio was calculated to determine the color intensity of the heat map and the hierarchical clustering. Upregulated transcripts (higher 635/532 log2 ratio > 0.5), non-differentially regulated transcripts (log ratio = −0.5 < x < 0.5), and downregulated transcripts (log ratio ≤ −0.5) correspond to red, dark, and green colors respectively. (B) Hierarchical clustering of differentially expressed transcripts. A, B, and C on the top of the dendrogram refer the biological replications during hybridization. NNCBISY, no NCBI gene symbol.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Ontological classification of differentially regulated transcripts in the very fast and very slow PB+oocytes.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Detected and deferentially expressed transcripts in the very fast and very slow dielectrophoretic categories of zygotes. (A) Heat map of 793 normalized transcripts. D, E, and F describe the dye swap of A, B, and C. The 635/532 log2 ratio was calculated to determine the color intensity of the heat map and the hierarchical clustering. Upregulated transcripts (higher 635/532 log2 ratio > 0.5), non-differentially regulated transcripts (log ratio = −0.5 < x < 0.5), and downregulated transcripts (log ratio ≤ −0.5) correspond to red, dark, and green colors respectively. (B) Hierarchical clustering of differentially expressed transcripts. A, B, and C on the top of the dendrogram refer the biological replications during hybridization. NNCBISY, no NCBI gene symbol.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Ontological classification of differentially regulated genes in the very fast and very slow zygotes.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Relative abundance of differentially regulated transcripts in the very fast versus very slow dielectrophoretic categories oocytes.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
Relative abundance of differentially regulated genes in the very fast versus very slow dielectrophoretic categories zygotes.
Citation: Reproduction 133, 5; 10.1530/REP-06-0277
The authors are grateful to Mr Horst Fassbender and Mr Plech (Geodätisches Institut University of Bonn, Germany) for providing the AM/phase lock AC generator and voltmeter reader. They are also grateful to Dr Andreas Waha (Institute of Neuropathology, University of Bonn) for his technical support during array scanning. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
References
Aihara T, Miyoshi Y, Koyama K, Suzuki M, Takahashi E, Monden M & Nakamura Y1998 Cloning and mapping of SMARCA5 encoding hSNF2H, a novel human homologue of Drosophila ISWI. Cytoge-netics and Cell Genetics 81 191–193.
Arai Y, Hosoda F, Kobayashi H, Arai K, Hayashi Y, Kamada N, Kaneko Y & Ohki M1997 The inv(11)(p15q22) chromosome translocation of de novo and therapy-related myeloid malignancies results in fusion of the nucleoporin gene, NUP98, with the putative RNA helicase gene, DDX10. Blood 89 3936–3944.
Archer S, Morgan H & Rixon FJ1999 Electrorotation studies of baby hamster kidney fibroblasts infected with herpes simplex virus type. Biophysical Journal 76 2833–2842.
Arnold WM & Zimmerman U1988 Electrorotation: development of a technique for dielectric measurements on individual cells and particles. Journal of Electrostatics 21 151–191.
Arnold WM, Schmutzler RK, Al-Hasani S, Krebs D & Zimmermann U1989 Differences in membrane properties between unfertilized and fertilized single rabbit oocytes demonstrated by electro-rotation: comparison with cells from early embryos. Biochimica et Biophysica Acta 979 142–146.
Blondin P & Sirard MA1995 Oocyte and follicular morphology as determining characteristics for developmental competence in bovine oocytes. Molecular Reproduction and Development 41 54–62.
Burger A, Berendes R, Liemann S, Benz J, Hofmann A, Gottig P, Huber R, Gerke V, Thiel C, Romisch J & Weber K2006 The crystal structure and ion channel activity of human annexin II, a peripheral membrane protein. Journal of Molecular Biology 257 839–847.
Burt JPH, Chan KL, Dawson D, Parton A & Pethig R1996 Assays for microbial contamination and DNA analysis based on electrorota-tion. Annales de Biologie Clinique 54 253–257.
Caraux G & Pinloche S2004 PermutMatrix: a graphical environment to arrange gene expression profiles in optimal linear order. Bioinfor-matics 21 1280–1281.
Cho MR, Thatte HS, Silvia MT & Golan DE1999 Transmembrane calcium influx induced by ac electric fields. Official Publication of the Federation of American Societies for Experimental Biology 13 677–683.
Chohan KR & Hunter AG2003 Effect of reproductive status on in vitro developmental competence of bovine oocytes. Journal of Veterinary Science 4 67–72.
Costa LFS, Machado MSN,Oliveira JFC, ZamberlanG & GonçalvesPBD2006 Annexin II mRNA expression in bovine oocytes during follicular development. Genetics and Molecular Biology 29 396–400.
Cristofanilli M, De Gasperis G, Zhang L, Hung MC, Gascoyne PR & Hortobagyi GN2002 Automated electrorotation to reveal dielectric variations related to Her-2/neu over expression in MCF-7 sub lines. Clinical Cancer Research 8 615–619.
Docoslis A & Alexandridis P2002 One, two, and three-dimensional organization of colloidal particles using non-uniform AC electric fields. Electrophoresis 23 2174–2183.
Donnison M & Pfeffer PL2004 Isolation of genes associated with developmentally competent bovine oocytes and quantification of their levels during development. Biology of Reproduction 71 1813–1821.
Dreier R, Schmid KW, Gerke V & Riehemann K1998 Differential expression of annexins I, II and IV in human tissues: an immunohistochemical study. Histochemistry and Cell Biology 110 137–148.
Eichenlaub-Ritter U & Peschke M2002 Expression in in vivo and in vitro growing and maturing oocytes: focus on regulation of expression at the translational level. Human Reproduction Update 8 21–41.
Filipenko NR, MacLeod TJ, Yoon CS & Waisman DM2004 Annexin A2 is a novel RNA-binding protein. Journal of Biological Chemistry 279 8723–8731.
Fuhr G, Muller T, Baukloh V & Lucas K1998 High-frequency electric field trapping of individual human spermatozoa. Human Reproduction 13 136–141.
Gascoyne PRC, Huang Y, Pethig R, Vykoukal J & Becker FF1992 Dielectrophoretic separation of mammalian cells studied by computerized image analysis. Measurement Science & Technology 3 439–445.
Gianaroli L, Magli MC, Ferraretti AP, Fortini D & Grieco N2003 Pronuclear morphology and chromosomal abnormalities as scoring criteria for embryo selection. Fertility and Sterility 80 341–349.
Goater AD & Pethig R1998 Electrorotation and dielectrophoresis. Parasitology 117 177–189.
Green NG & Morgan H1997 Dielectrophoretic investigations of submicrometre latex spheres. Journal of Physics D: Applied Physics 30 2626–2633.
Grohmann M, Spada F, Schermelleh L, Alenina N, Bader M, Cardoso MC, Green NG & Morgan H1997 Dielectrophoretic investigations of sub micrometre latex spheres. Journal of Physics D: Applied Physics 30 2626–2633.
Guo Z, Linn JF, Wu G, Anzick SL, Eisenberger CF, Halachmi S, Cohen Y, Fomenkov A, Hoque MO, Okami K, Steiner G, Engles JM, Osada M, Moon C, Ratovitski E, Trent JM, Meltzer PS, Westra WH, Kiemeney LA, Schoenberg MP, Sidransky D & Trink B2004 CDC91L1 (PIG-U) is a newly discovered oncogene in human bladder cancer. Nature Medicine 10 374–381.
Huang Y, Holzel R, Pethig R & Wang XB1992 Differences in the AC electrodynamics of viable and non viable yeast cells determined through combined dielectrophoresis and electrophoresis studies. Physics in Medicine and Biology 37 1499–1517.
Huang Y, Wang XB, Becker FF & Gascoyne PR1996 Membrane changes associated with the temperature sensitive p85gagmos-dependant transformation of rat kidney cells as determined by dielectrophoresis and electrorotation. Biochimica et Biophysica Acta 1282 76–84.
Huang Y, Joo S, Duhon M, Heller M, Wallace B & Xu X2002 Dielectrophoretic cell separation and gene expression profiling on microelectronic chip arrays. Analytical Chemistry 74 3362–3371.
Huang ZP, Ni H, Yang ZM, Wang J, Tso JK & Shen QX2003 Expression of regulator of G-protein signalling protein 2 (RGS2) in the mouse uterus at implantation sites. Reproduction 126 309–316.
Ikeda I, Monjushiro H & Watarai H2005 Measurement of dielec-trophoretic mobility of single micro-particles in a flow channel. Analyst 130 1340–1342.
Iost I & Dreyfus M1994 mRNAs can be stabilized by DEAD-box proteins. Nature 372 193–196.
Jackson-Grusby L, Bear C, Possemat R, Tudor M, Fambroug D, Csankovszki1 G, Dausman J, Lee P, Wilson C, Lander E & Jaenisch R2001 Loss of genomic methylation causes p53-dependent apoptosis and epigenetic deregulation. Nature Genetics 27 31–39.
Jaffe LA, Sharp AP & Wolf DP1983 Absence of an electrical polyspermy block in the mouse. Developmental Biology 96 317–323.
Kadaksham AT, Singh P & Aubry N2004 Dielectrophoresis of nano particles. Electrophoresis 25 3625–3632.
Lansman JB1983 Voltage-clamp study of the conductance activated at fertilization in the starfish egg. Journal of Physiology 345 353–372.
Lao AI & Hsing IM2005 Flow-based and sieving matrix-free DNA differentiation by a miniaturized field flow fractionation device. Lab on a chip 5 687–690.
Lapizco-Encinas BH, Simmons BA, Cummings EB & Fintschenko Y2004 Insulator-based dielectrophoresis for the selective concentration and separation of live bacteria in water. Electrophoresis 25 1695–1704.
Lequarre AS, Traverso JM, Marchandise J & Donnay I2004 Poly(A) RNA is reduced by half during bovine oocyte maturation but increases when meiotic arrest is maintained with CDK inhibitors. Biology of Reproduction 71 425–431.
Lonergan P, Gutierrez-Adan A, Rizos D, Pintado B, de la Fuente J & Boland MP2003 Relative messenger RNA abundance in bovine oocytes collected in vitro or in vivo before and 20 h after the preovulatory luteinizing hormone surge. Molecular Reproduction and Development 66 297–305.
Markx GH, Huang Y, Zhou XF & Pethig R1994 Dielectrophoretic characterization and separation of micro-organism. Microbiology 140 585–591.
Mataraza JM, Briggs MW, Li Z, Entwistle A, Ridley AJ & Sacks DB2003 IQGAP1 Promotes cell motility and invasion. Journal of Biological Chemistry 278 41237–41245.
Mayes M2002 The meiotic arrest of bovine oocytes. PhD. thesis, Department of Animal Sciences, Faculty of Agricultural Sciences, University of Laval, Quebec, Canada.
Mayes MA & Sirard MA2001 The influence of cumulus–oocyte complex morphology and meiotic inhibitors on the kinetics of nuclear maturation in cattle. Theriogenology 55 911–922.
Minerick AR, Zhou R, Takhistov P & Chang HC2003 Manipulation and characterization of red blood cells with alternating current fields in micro devices. Electrophoresis 24 3703–3717.
Mohrmann L, Langenberg K, Krijgsveld J, Kal AJ, Heck AJ & Verrijzer CP2004 Differential targeting of two distinct SWI/SNF-related Drosophila chromatin-remodeling complexes. Molecular and Cellular Biology 8 3077–3088.
Pethig R1991 Application of AC electrical fields to the manipulation and characterization of cells. In Automation in Biotechnology, pp 159–185. Ed. I Karube. Amsterdam: Elsevier.
Poh HA, Kaler K & Pollock K1981 The continuous positive and negative dielectrophoresis of micro organisms. Journal of Biological Physics 9 67–86.
Prigent C & Giet R2003 Aurora A and mitotic commitment. Cell 114 531–532.
Raemaekers T, Ribbeck K, Beaudouin J, Annaert W, Van Camp M, Stockmans I, Smets N, Bouillon R, Ellenberg J & Carmeliet G2003 NuSAP, a novel microtubule-associated protein involved in mitotic spindle organization. Journal of Cell Biology 162 1017–1029.
Ramírez A, Zehe A & Starostenko O2003 Dielectrophoretic field fractionation of rouleaux formed of human erythrocytes: a feasibility study. Revista Mexicana de Ingeniería Biomédica 24 14–22.
Richardson RT, Batova IN, Widgren EE, Zheng LX, Whitfield M, Marzluff WF & O’Rand MG2000 Characterization of the histone H1-binding protein, NASP, as a cell cycle-regulated somatic protein. Journal of Biological Chemistry 275 30378–30386.
Richardson RT, Bencic DC & O’Rand MG2001 Comparison of mouse and human NASP genes and expression in human transformed and tumor cell lines. Gene 274 67–75.
Satinover DL, Leach CA, Stukenberg PT & Brautigan DL2006 Activation of Aurora-A kinase by protein phosphatase inhibitor-2, a bi functional signaling protein. PNAS 23 8625–8630.
Schmid SR & Linder P1992 DEAD protein family of putative RNA helicases. Journal of Molecular Microbiology and Biotechnology 6 283–292.
Sirard MA, Dufort I, Vallee M, Massicotte L, Gravel C, Reghenas H, Watson AJ, King WA & Robert C2005 Potential and limitations of bovine-specific arrays for analysis of mRNA levels in early development: preliminary analysis using a bovine embryonic array. Reproduction, Fertility and Development 17 47–57.
Sokol SY, Li Z & Sacks DB2001 The effect of IQGAP1 on Xenopus embryonic ectoderm requires Cdc42. Biological Chemistry 276 48425–48430.
Stancheva I & Meehan RR2000 Transient depletion of xDnmt1 leads to premature gene activation in Xenopus embryos. Genes & development 14 313–327.
Swart-Mataraza JM, Li Z & Sacks DB2002 IQGAP1 is a component of Cdc42 signaling to the cytoskeleton. Journal of Biological Chemistry 277 24753–24763.
Tesarik J & Greco E1999 The probability of abnormal pre-implantation development can be predicted by a single static observation on pronuclear stage morphology. Human Reproduction 14 1318–1323.
Tesfaye D, Ponsuksili S, Wimmers K, Gilles M & Schellander K2004 A comparative expression analysis of gene transcripts in post fertilization developmental stages of bovine embryos produced in vitro or in vivo. Reproduction in Domestic Animals 39 396–404.
Tosti E & Boni R2004 Electrical events during gamete maturation and fertilization in animals and humans. Human Reproduction Update 10 53–65.
Udayakumar TS, Belakavadi M, Choi KH, Pandey PK & Fondell JD2006 Regulation of Aurora-A kinase gene expression via GABP recruitment of TRAP220/MED1. Journal of Biological Chemistry 281 14691–14699.
Wang XB, Huang Y, Burt JPH, Markx GH & Pethig R1993 Selective dielectrophoretic confinement of bioparticles in potential energy wells. Journal of Physics D: Applied Physics 26 1278–1285.
Wang XB, Huang Y, Wang X, Becker FF & Gascoyne PR1997 Dielectrophoretic manipulation of cells with spiral electrodes. Biophysical Journal 4 1887–1899.
Wanichapichart P, Bunthawin S, Kaewpaiboon A & Kanchanapoom K2002 Determination of cell dielectric properties using dielectro-phoretic technique. ScienceAsia 28 113–119.
Westerhoff HV, Tsongt TY, Chocks PB, Chen YID & Astumiant RD1986 How enzymes can capture and transmit free energy from an oscillating electric field. PNAS 83 4734–4738.
Yang J, Huang Y, Wang XB, Becker FF & Gascoyne PR2000 Differential analysis of human leukocytes by dielectrophoretic field flow-fraction. Biophysical Journal 78 2680–2689.
Zmijewski JW, Song L, Harkins L, Cobbs CS & Jope RS2001 Oxidative stress and heat shock stimulate RGS2 expression in 1321N1 astrocytoma cells. Archives of Biochemistry and Biophysics 392 192–196.
