| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Department of Physiology and Institute for Biomedical Research,3 School of Biomedical Sciences, University of Sydney, New South Wales 2006, Australia
Department of Anatomy,4 University of Cambridge, Cambridge CB2 3DY, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
conceptus, early development, gene regulation, signal transduction, trophoblast
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Voltage-gated K+ channels have been divided into subfamilies on the basis of their genetic sequence [10, 11]. The electrical properties of the early embryogenic channel resemble those of channels encoded by the ether-à-go-go gene family, which consists of eag, erg (ether-à-go-go-related), and elk (ether-à-go-go-like) members [10]. In the present study, we examined expression of the erg subfamily during preimplantation embryogenesis during the same period that the oscillatory K+-channel activity is detected. Two mouse isoforms of the erg gene product are known: MERG1A and MERG1B [12, 13]. MERG1A transcripts are expressed in many different tissues, including heart, brain, and testes, whereas the alternately spliced form, MERG1B, has been detected only in cardiac tissues [12, 13]. MERG1B can be distinguished from other family members by its distinct N-terminal cytoplasmic domain. In the human, mutations in human erg on chromosome 7 cause inherited long-QT syndrome, resulting in the development of ventricular arrhythmia and sudden death [14, 15]. Cardiac arrhythmia and long-QT symptoms also can be induced by drugs that affect ether-à-go-go-related gene (ERG)-channel function, such as cisapride [16] and E-4031 [17]. Here, we report on the expression profile of MERG during preimplantation development.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Outbred Quackenbush Swiss mice (Laboratory Animal Services, University of Sydney) were superovulated by intraperitoneal injections of 10 IU of eCG (Intervet, Boxmeer, The Netherlands) and hCG (Intervet) 48 h apart. To obtain concepti, females were paired overnight with males. Unfertilized oocytes; 1-cell, 2-cell, 4-cell, and 8-cell, and morula stages; and blastocysts were obtained from female mice at 12.5–13, 18–24, 30–50, 52–64, 64–76, 76–100, and 112–124 h post-hCG, respectively. Unfertilized oocytes and 1-cell zygotes were teased from the oviducts into M2 medium containing 4 mg/ml of bovine serum albumin (M2+BSA). The cumulus cells were removed with 0.2 mg/ml of hyaluronidase (Type II; Sigma, St. Louis, MO) in M2+BSA. The 2-cell to blastocyst stages were either flushed from the oviducts or uteri at the stage to be examined or derived by culture in vitro from earlier stages in T6 medium supplemented with 4 mg/ml of BSA under mineral oil (T6+BSA; Sigma) in 5% CO2 at 37°C. Studies were performed in accordance with the National Health and Medical Research Council of Australia Guidelines on Ethics in Animal Experimentation and were approved by the University of Sydney, Animal Ethics Committee.
Inner cell masses (ICMs) were isolated from blastocyst stages at 112–124 h post-hCG by immunosurgery according to the method of Solter and Knowles [18] using a rabbit antiserum raised against mouse spleen (gift of P. Kaye, University of Queensland, Brisbane, Australia), or a goat anti-mouse antibody (Santa Cruz Biotechnologies, Santa Cruz, CA) when immunofluorescence studies were to be performed, to avoid a nonspecific signal occurring with the anti-rabbit fluorescein isothiocyanate (FITC) secondary antibody. Trophoblast cells were obtained for reverse transcription-polymerase chain reaction (RT-PCR) analysis by bisecting expanded blastocysts with a fine glass needle and retaining the segment without the ICM.
Drug Treatment of Concepti
Groups of early 8-cell stages were cultured in T6+BSA under oil containing either 1 µg/ml of cytochalasin D (CCD; stock solution of 1 mg/ml in dimethyl sulfoxide [DMSO]; Sigma) for 12 h or 10 µM puromycin (10 mM stock solution in water; Sigma) for 24 h and then fixed for immunofluorescent analysis. Concepti undergoing blastocoele formation were exposed to puromycin for 12 h before immunofluorescence analysis. Concepti treated with E-4031 (4'-[[1-[2-(6-methyl-2-pyridyl) ethyl]-4-piperidinyl] carbonyl] methanesulfon-anilide·2HCl·H2O; 100 mM stock solution in water; Biomol, Plymouth Meeting, PA) or cisapride (5 mM stock solution in DMSO; gift from K. Wyse, Garvin Institute, Sydney, Australia) were cultured continuously in the presence of the drug from the 2-cell stage until blastocyst formation. Other groups were exposed to E-4031 or cisapride for 24 h from the late 2-cell to the compact 8-cell stage, when the cellular localization of the protein was assessed. All experiments were repeated at least three times.
Isolation of Total RNA from Oocytes, Blastocysts, Trophoblast Stem Cells, and Heart Tissue
The RNA was isolated from oocytes, blastocysts, and a trophoblast stem (TS) cell line (derived from 3.5-day postcoitum mouse blastocysts; gift of S. Tanaka, University of Tokyo, Tokyo, Japan) [19] using TriReagent (Sigma) according to the manufacturer's protocol. Pools of 200 oocytes or blastocysts were placed in 500 µl of TriReagent containing 5 µg of glycogen (Roche, Castle Hill, NSW, Australia) as a carrier, and the RNA was extracted by addition of 100 µl of chloroform and centrifuged at 12 000 x g at 4°C for 15 min. The upper phase was removed, and the RNA was precipitated by addition of half the volume of isopropanol and pelleted by centrifuging at 12 000 x g for 15 min. The supernatant was discarded, the pellet washed with 75% ethanol and air-dried, and the RNA resuspended in 10 µl of water. Total RNA was purified from 2.5 x107 TS cells using TriReagent as described above.
The RNA was purified from 40 mg of mouse heart using the GenElute Mammalian Total RNA kit (Sigma). The tissue was first ground under liquid nitrogen and the powder transferred into 100 µl of RNA extraction buffer. The tube was shaken vigorously to mix the powder and then spun at 13 000 rpm for 1 min to sediment any large debris. The supernatant was then processed according to the manufacturer's protocol. The concentration and purity of each RNA was assessed spectrophotometrically, and aliquots of the RNA were then stored at -80°C until required for use.
RNA Detection Using RT-PCR
Reverse transcription was carried out either directly on oocytes and concepti or on total RNA isolated as described above. When RT was performed without previous isolation of RNA, the oocytes or concepti were placed in groups of 50 into 200-µl, thin-walled PCR tubes in a minimal volume of M2+BSA and stored at -80°C until required for RT-PCR [20]. The RNA was reverse transcribed by addition of 8.9 µl of the reaction mixture, which contained 10 mM dithiothreitol, 1 mM of each dNTP, 0.2 µg of random hexanucleotides (Invitrogen), 24 U of RNAguard, and 0.1 mg/ml of BSA (Amersham Biosciences, Uppsala, Sweden) in Superscript II buffer (Invitrogen). The reaction mix was heated to 65°C for 1 min and then cooled to room temperature before adding 200 U of Superscript II reverse transcriptase (Invitrogen). Reverse transcription was performed at 37°C for 1 h. Reverse transcription of total RNA isolated from oocytes, blastocysts, TS cells, and heart tissue was performed using the above procedure following the addition of 17.5 ng of RNA (equivalent to 50 oocytes or 10 blastocysts) to the reaction mix.
The PCR amplification was performed in a final volume of 50 µl containing 10 µl of cDNA, 2 mmol/L of MgCl2, 2.5 U of recombinant Taq polymerase (Roche), and 0.5 µmol/L of 5' and 3' gene-specific primers. Thirty cycles of amplification were performed (94°C for 30 sec, 60°C for 30 sec, 72°C for 2 min). Three sets of primers were used to study MERG1A and MERG1B RNA expression (Fig. 1). Primers specific to MERG1A (merg1a-F, 5'-GACCCTTTCCTGGCTTCACC-3'; merg1a-R, 5'-GACCAGCAGCAGGATGAGC-3'; Sigma-Genosys, Castle Hill, NSW, Australia) generated a 219-base pair (bp) diagnostic fragment. The presence of MERG1B was examined using a second set of primers (merg1b-F, 5'-GGGAAGGAGAGCAGGACAG-3'; merg1b-R, 5'-GGTTGGGAATGGTGAAAG-3') that produced a 1047-bp fragment. Expression of MERG1A at different stages of development was assessed using a third set of primers (merg-F, 5'-TGAGTCCATGTGAGGCTGTTCC-3'; merg-R, 5'-ACTGGCTCATTCTGCTGCTGGTC-3') that spanned an intron and thus allowed amplification of cDNA (514-bp amplicon) to be unambiguously distinguished from amplification of genomic DNA (976-bp amplicon). The genomic signal was used as a measure of the number of cells in the samples analyzed. Rapid amplification of cDNA ends (RACE) was performed to confirm the full-length sequence of MERG1A in mouse oocytes [21]. The PCR products were ligated into pGEM-Teasy (Promega, Madison, WI) and independent clones were sequenced (AGRF, Brisbane, QLD, Australia). Amplification of ß-actin (forward primer, 5'-AAACGCAGCTCAGTAACAG; reverse primer, 3'-CATGTACGTAGCCATCCAG; 762-bp amplicon) was used as a positive control for cDNA integrity.
|
Detection of MERG1A by Immunoblotting
Groups of oocytes and blastocysts were collected in a mixture of M2 plus polyvinylpyrrolidine (PVP) and lysis buffer (50 mM Tris [pH 7.4], 250 mM NaCl, 0.1% Igepal CA 630, 50 mM sodium fluoride) containing protease inhibitors (cocktail set I; Calbiochem, Alexandria, NSW, Australia) and frozen immediately at -20°C. Heart and skeletal muscle protein lysates were prepared by freezing and then grinding the tissue under liquid N2 before the addition of lysis buffer. Total protein was extracted from 2.5 x 107 TS cells by lysis in PBS containing 1% Triton X-100 and protease inhibitors. These lysates were centrifuged at 13 000 rpm and 4°C for 15 min to remove cell debris, and the supernatant was stored at -80°C. An equal volume of 2x sample buffer [22] was added as the samples thawed on ice. The proteins were separated by SDS-PAGE (7% gels) and transferred electrically to nitrocellulose membranes (Hybond-C; Amersham Biosciences). Following transfer and blocking for 2 h in 3% normal goat serum in PBS plus 0.2% Triton X-100, membranes were incubated overnight at 4°C with a rabbit polyclonal anti-human ERG (HERG) antibody (gift of A. Pond, Purdue University, West Lafayette, IN) [23] diluted 1:1000 in blocking solution. This antibody is specific to the carboxyl terminal of the HERG protein (residues 1145–1159). Despite this region showing sequence identity with MERG1A and MERG1B, Pond et al. [23] only detected the expression of proteins in heart that reflect differences in N-linked glycosylation of full-length MERG1A rather than the coexpression of MERG1B. The membranes were then incubated in a goat anti-rabbit immunoglobulin G secondary antibody (Santa Cruz Biotechnology) conjugated to horseradish peroxidase diluted 1:2000 in blocking solution for 2 h at room temperature. After three rinses in PBS plus 0.2% Triton-X 100, the MERG1A protein signal was revealed using the enhanced chemiluminescence plus Western blot analysis system (Amersham Biosciences).
Immunofluorescent Staining
Oocytes and concepti were fixed in 4% paraformaldehyde in PBS containing 1 mg/ml of polyvinyl alcohol (PBS+PVA) for 30 min and then, in the same solution, with the addition of 0.3% Triton X-100 for a further 30 min. The samples were rinsed in PBS+PVA containing 0.7% BSA and 0.1% Tween 20 for 30 min to 2 h of incubation with anti-HERG antiserum diluted 1:500 in rinsing solution or anti-GLUT3 antiserum diluted 1:200 (gift of M. Pantaleon and P. Kaye, University of Queensland, Australia) [24]. The samples were then rinsed again for 30 min followed by 1 h of incubation in FITC-conjugated anti-rabbit secondary antibody (Roche) diluted to 10 µg/ml in rinsing solution. After a 30-min rinse step, the DNA was stained with propidium iodide (5 µg/ml; Sigma) for 10 min, rinsed, and mounted on glass slides in Vectashield mounting medium (Vector Laboratories, Inc., Burlingame, CA) to prevent bleaching. All steps of the procedure were carried out at room temperature.
The cellular localization of the MERG1A protein was assessed by confocal laser-scanning microscopy using a Nikon Eclipse E800 microscope (Nikon, Tokyo, Japan) equipped with a Bio-Rad Radiance Plus laser system (Bio-Rad, Regents Park, NSW, Australia) and Lasersharp 2000 software (Bio-Rad). Images were collected using a Nikon Plan Apo 60x/1.40 oil immersion lens. Fluorescein excitation was obtained using an argon/krypton laser operating at 488 nm and propidium iodide excitation by a helium/neon laser operating at 543 nm. Simultaneous optical sections (thickness, 1 µm) were collected, the number of which varied depending on the depth of each sample. Line-averaging mode was used to integrate the signal collected over eight lines to reduce noise with a raster size of 512 x 512 pixels.
The NIH Image (v1.60) software (Research Services Branch, N1H, Bethesda, MD) was used to quantify the amount of MERG1A protein in the apical and basolateral membrane regions. A rectangle was placed over individual blastomeres in an embryo, and the intensity profile was plotted and the values exported to Excel (Microsoft, Redmond, WA). The average apical and basolateral intensities were calculated from at least three blastomeres per embryo. Only blastomeres on the edge of an embryo were measured. A measure of the total MERG1A protein expression also was obtained from confocal images using NIH Image. To standardize the method of analysis, the fluorescence intensity was quantified from the section taken through the center of each conceptus.
Nuclear Counts
The number of nuclei in blastocysts cultured in the presence of E-4031 from the 2-cell stage was assessed in concepti fixed as described above for immunofluorescent staining and mounted in Vectashield mounting medium containing 1.5 µg/ml of 4',6-diamidino-2-phenylindole (DAPI). Nuclei were counted from sections obtained by confocal microscopy.
Statistical Analysis
Results are presented as the mean ± SEM, with the number of experiments (n) in parentheses. Data were compared by ANOVA, and differences between means were determined using the Student t-test and Bonferroni method for multiple comparisons. Differences were considered to be significant when P values were less than 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The RT-PCR analysis was carried out using primers designed to distinguish the long and short forms of mouse ERG (MERG1A and MERG1B) [12, 13]. Primers specific to MERG1A yielded a positive signal of 219 bp in samples of oocytes, blastocysts (Fig. 2A), and TS cells (Fig. 2C). No signal was obtained from oocytes, blastocysts (Fig. 2B), or TS cells (Fig. 2C) with primers specific to the shorter MERG1B form of the gene, although amplification of the expected-size fragment from mouse heart mRNA was achieved (Fig. 2B, Track HT). The 5'-RACE was used to obtain the full-length sequence of the merg1a gene in oocytes, and sequence analysis showed that it was identical to the published merg1a gene (GenBank accession no. NM_013569).
|
Developmental Expression Pattern of MERG1A mRNA
The RT-PCR products from samples containing equal numbers of oocytes or concepti at different developmental stages were compared directly for the presence of MERG1A mRNA. High levels of MERG1A mRNA were detected in the oocyte, but levels declined during the 2-cell stage to become barely detectable (Fig. 3A, lanes 0, 2E, and 2L). When early 2-cell stages were cultured to the equivalent of the late 4-cell stage in the presence of the transcriptional inhibitor 
-amanitin, no transcripts were observed, suggesting loss of maternal transcripts and prevention of zygotic transcription (Fig. 3B). In control concepti, restoration of MERG1A mRNA levels did not occur until the 8-cell stage and increased in morulae and blastocysts (Fig. 3A, lanes 4 to B). The primers used spanned a merg1a exon, which allowed cDNA (514-bp amplicon) to be distinguished unambiguously from a genomic DNA signal (976-bp amplicon). The latter was only detectable in those stages with a cell number greater than eight. In the absence of reverse transcriptase, only a product from genomic DNA was amplified from blastocysts (Fig. 3A, lane RT-). When the constituent tissue subpopulations of the blastocyst were examined by RT-PCR, MERG1A mRNA was detected primarily in the ICM and only relatively weakly in the trophoblast (Fig. 3C), despite this latter tissue having more cells, as indicated by the stronger genomic DNA signal.
|
Identification of the MERG1A Protein in Mouse Oocytes and Early Concepti by Immunoblotting
Immunoblotting was used to determine whether the MERG1A protein was detectable in a mouse TS cell line and in the mouse oocyte and early conceptus. As expected, immunoblots of protein from mouse heart cell lysates revealed a prominent band at 165 kDa as well as a weaker band at 205 kDa (Fig. 4A), representing different N-linked glycosylation forms, as demonstrated previously by Pond et al. [23]. Immunoblots of a TS cell lysate, unfertilized oocytes, and blastocysts probed with anti-HERG revealed bands at approximately 127 kDa (the nonglycosylated form) and two presumptive glycosylated forms at 145 and 175 kDa (Fig. 4B). All these bands in heart and TS cell lysates were eliminated by preincubation of the antibody with the immunogenic peptide (data not shown).
|
Immunofluorescence Distribution of MERG1A Protein in Mouse Oocytes and Early Concepti
The MERG1A protein was detected at all stages from mature oocytes through to the blastocyst by immunofluorescent staining (Fig. 5). Preabsorption of the anti-HERG antibody with the immunogenic peptide resulted in reduced signal (compare Fig. 5, I and K). In unfertilized oocytes, MERG1A was concentrated in the cytoplasm around the spindle, especially at its equator (Fig. 5A, arrows). Early after fertilization (18 h post-hCG), MERG1A was located centrally in the cytoplasm and excluded from the pronuclei (Fig. 5B), but later, in S phase of the first cell cycle, some MERG1A protein was detected in the expanded pronuclei as a ring around the nucleoli (Fig. 5C). This region surrounding the nucleoli also fluoresced intensely in samples stained with the DNA-specific fluorophore DAPI (data not shown), indicating it to be chromatin rich. As development progressed further, the MERG1A protein was increasingly located in the nucleus and the nuclear and plasma membranes, and less was cytoplasmic. At the 2-cell stage, MERG1A protein was detected at reduced levels in S phase, being dispersed and mainly cytoplasmic (Fig. 5D), but by the G2 phase, it was localized to the nuclear membranes and around the nucleoli and, for the first time, was clearly localized to the plasma membrane (Fig. 5E). The 4-cell stages exhibited MERG1A staining in the plasma membrane, with a suggestion of its accumulation in the regions of cell contact (Fig. 5F). This selective basolateral localization of MERG1A became quite distinct at the 8-cell stage, particularly following the cell flattening at compaction (Fig. 5G). Moreover, the level of MERG1A staining increased with compaction, becoming even more intense thereafter (Fig. 5, H and I). By the morula and blastocyst stages, the number of cells (n 
32) made it difficult to determine precisely the boundary of individual cells, but there remained a clear absence of MERG1A staining on the external, apical surfaces of polarized external cells (Fig. 5H, arrow). In contrast, most exposed surfaces of nonpolar cells in recently isolated ICMs stained despite their close contact with adjacent cells (Fig. 5J). The distinct staining pattern noted around the chromatin in oocytes arrested in metaphase of meiosis II (Fig. 6A) was observed in mitotic cells in all embryogenic stages (Fig. 6, B–D, arrows).
|
|
Many molecular redistributions at compaction involve cytoskeletal activity [25]. The basolateral localization of MERG1A from compaction onward was studied by culture of early 8-cell stages in the presence of the actin-depolymerizing drug CCD. This drug prevents intercellular flattening during the 8-cell stage, with the blastomeres remaining spherical, and although the nuclear cell cycle is completed, cytokinesis is blocked. The CCD-treated early 8-cell stages, unlike the controls, did not develop restricted basolateral staining of MERG1A, which was detected around the entire cell membrane (Fig. 7B). The overall increase in MERG1A staining normally observed after the 8-cell stage still occurred in the presence of CCD (compare Fig. 7, A and B with D), although at a slightly reduced level (Fig. 7D).
|
The increase in MERG1A staining after compaction was reduced by culture of concepti in the protein synthesis-inhibitor puromycin for 24 h (compare Fig. 7, B' with A' and C'; see also Fig. 7D'). The MERG1A remaining was concentrated almost exclusively in the membrane, with cytoplasmic staining being very weak. These puromycin-treated concepti arrested in G2 phase of the cell cycle but remained otherwise intact and undamaged [8, 26]. Treatment of blastocysts with puromycin from the onset of cavitation also showed a significantly reduced level of MERG1A staining compared to both start and end-point controls (Fig. 7, A''–D'').
Inhibitors of ERG-Channel Function Affect Both Localization of MERG1A Protein and Transition to the Blastocyst Stage
The E-4031 is a methanesulfonanilide, antiarrhythmic drug that selectively binds to ERG channels and thereby blocks the K+ current (7–10 nM) [15, 17]. At higher concentrations of E-4031 (1–5 µM), the drug interferes with ERG-protein trafficking to the plasma membrane [27, 28]. Concepti were incubated from the 2-cell stage of development in the continuous presence of various concentrations of E-4031 to determine whether ERG-channel function is required for early mouse embryogenic development. Only concentrations of 100 µM E-4031 had a significant effect on development to the blastocyst stage (Fig. 8A). This was not caused by cytotoxicity of the drug, because development from the 2-cell stage through to the early 8-cell stage was not affected by E-4031 (1–100 µM). Nuclear count analysis revealed a significant reduction in the number of cells in concepti treated with between 1 and 100 µM E-4031 (Fig. 8B). Re-expansion of blastocysts that were collapsed mechanically, by passing them gently in and out of the tip of a micropipette, was not affected by the presence of 100 µM E-4031 (data not shown).
|
Treatment of embryos from the 2-cell stage with 1–100 µM E-4031 also affected the redistribution of MERG1A protein at the late 8-cell stage (Fig. 8, E–G). This effect of E-4031 was specific for MERG1A, because the same drug regime had no obvious effect on the polarized distribution of the normal apical localization of the glucose transporter GLUT3 in trophoblast cells (data not shown) [24]. Despite altering the membrane distribution of MERG1A, E-4031 (33–100 µM) had no effect on the total amount of MERG1A protein expressed in the late 8-cell stage (Fig. 8H). This was shown by the lack of any significant effect on total MERG1A fluorescence in late 8-cell concepti following treatment with E-4031 from the late 2-cell stage.
The effect on mouse embryogenic development of a second drug, cisapride, which blocks HERG channels with high affinity (5–10 nM) [16, 29] and interferes with ERG trafficking at a higher concentration (1–5 µM) [27] also was examined. Cisapride (100 µM) affected significantly the morula-to-blastocyst transition in the majority of concepti when treated from the 2-cell stage (Fig. 9A). As with E-4031, this was not caused by cytotoxicity of the drug, because development from the 2-cell stage through to the 8-cell stage was not affected by any concentration of cisapride used. Nuclear count analysis revealed a significant reduction in the number of cells in 10–100 µM cisapride-treated concepti (Fig. 9B). Immunofluorescent and quantitative analyses both revealed a reduced basolateral localization of MERG1A after 24 h of cisapride exposure (Fig. 9, E–G), although total MERG1A expression in the late 8-cell stage remained comparable with that observed in untreated controls (Fig. 9H). Exposure of concepti to the cisapride diluent (DMSO) had no effect on the development or the distribution and levels of MERG1A protein (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The predicted molecular mass of MERG1A is 127 kDa, but a variety of higher-molecular-weight forms have been reported because of differential N-glycosylation [15, 23]. The N-glycosylation states of ERG channels are associated with its trafficking to the plasma membrane [15, 28], and the core-glycosylated, lower-molecular-weight form of the ERG protein is located in the endoplasmic reticulum [31]. Posttranslational modification leads to the formation of a mature, functional, cell surface-expressed ERG protein. Our analysis revealed that both cytoplasmic and membrane-bound forms of the protein are present during preimplantation development, but that MERG1A maturation (and, hence, localization) is regulated developmentally. Thus, at the 8-cell stage, polarization and intercellular flattening was associated with a redistribution of MERG1A from the cytoplasm to the blastomere membrane, where it was localized to the internal (basolateral) regions, adjacent to regions of cell contact. This concentration of MERG1A basolaterally seems to depend on maintained intercellular contact, because as blastomeres entered into the M phase and rounded up to reduce cell contact, basolateral localization ceased. Additionally, no basolateral accumulation of the protein was seen in the nonpolar cells of isolated ICMs or in concepti, where intercellular flattening was disturbed by treatment with the actin-depolymerizing drug CCD.
The mature, membrane-bound form of MERG1A appears to be more stable than the cytoplasmic form, because inhibition of protein synthesis at any stage examined resulted in a highly significant reduction in the intensity of staining in the cytoplasm even though cortical staining remained. The presence of protein synthesis inhibitors does not prevent the polarization and intercellular flattening at the 8-cell stage but, instead, causes them to occur prematurely [26, 32]. Two other proteins implicated in polarization and flattening are ezrin and E-cadherin. These proteins undergo posttranslational modification and increased turnover during the 8-cell stage, resulting in selective stabilization of E-cadherin to the basolateral region and of ezrin to the apical region of blastomeres [33–37]. The MERG1A appears to behave differently, and its selective localization to basolateral regions likely is secondary to cell polarization and probably is dependent on targeting of newly produced protein to the basolateral region rather than to its relative stabilization there, as occurs with E-cadherin [34].
An interesting feature was the association between MERG1A localization and the chromatin in M-phase cells and around the nucleoli in interphase nuclei. The reason for this specific association of MERG1A with chromatin is not clear, but it might be another example of a structural role for this protein, supporting the cellular architecture at different stages of the cell-division cycle. A structural role for channel proteins has been reported recently for the sodium exchanger NHE1 and the potassium channel Kv4.2 [1, 2]. In the case of Kv4.2, its association with the actin cytoskeleton is essential for the generation of appropriate Kv4.2 current densities. However, the structural role of NHE1 in regulating the cortical cytoskeleton is independent of its function as an ion exchanger [1, 2].
The descriptive studies of MERG1A do not elucidate whether it has a critical role in development. The pattern of membrane distribution suggests that it is unlikely to mediate the oscillatory-channel activity that we reported previously as occurring over the early cell cycles, when MERG1A is mainly cytoplasmic [7, 8]. Only at high concentrations of MERG1A-channel inhibitors was the number of embryos forming blastocysts reduced significantly. However, at lower concentrations of these inhibitors, those concepti that did form blastocysts had a significantly lower number of nuclei. Even so, concepti exposed to MERG1A inhibitors after completion of compaction formed blastocysts normally. Thus, any role for MERG1A in blastocyst formation appears to occur at or before compaction. Given that the drugs interfered with the relocalization of MERG1A at compaction without affecting intercellular flattening itself, it seems likely that MERG1A localization at this time is critical, and that the correct targeting of the protein to the basolateral region may be required for blastomere cleavage. Once this relocalization has been achieved, the process of blastocyst expansion occurs irrespective of MERG1A inhibitors. Zhou et al. [28] reported that the presence of either drug, at concentrations similar to those found to affect MERG1A localization in the current studies, induced the expression of the mature glycosylated form of ERG in HEK293 cells. Thus, expression of the channel protein over the entire cell surface might represent the consequence of increased glycosylation and protein trafficking to the membrane in response to feedback resulting from inhibition of channel function. The mechanisms by which these drugs have an effect on the cells and concepti require further investigation.
In conclusion, we have demonstrated that the erg gene is expressed in mouse oocytes and early concepti, that its expression and location are developmentally regulated, and that its inhibition leads to stage-specific developmental defects. It is not clear whether its electrical activity is required for its developmental effects or whether, like other K+ channels, it has a structural role in development.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Margot Day, Department of Physiology (F13), University of Sydney, NSW, 2006, Australia. FAX: 612 9351 2058; margotd{at}physiol.usyd.edu.au
Received: 1 July 2003.
First decision: 25 July 2003.
Accepted: 1 December 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Y. Ku, L. Babich, R. A. Word, M. Zhong, A. Ulloa, M. Monga, and B. M. Sanborn Expression of Transient Receptor Channel Proteins in Human Fundal Myometrium in Pregnancy Reproductive Sciences, April 1, 2006; 13(3): 217 - 225. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
