Crucial Role of Bysl in Mammalian Preimplantation Development as an Integral Factor for 40S Ribosome Biogenesis

Microinjection of siRNAs into mouse embryos.siRNA duplexes of 21 nucleotides against the following target sequences were in vitro transcribed with a CUGA7 in vitro siRNA synthesis kit (Nippon Genetech, Japan): siEGFP-441, CAGCCACAACGUCUAUAUCAU (66); siOct3/4-670, AAGAGAAAGCGAACUAGCAUU (66); siBysl-773, GAGAUGACAUUGCUGAAUACA; siBysl-845, CUGGAGCCUGGUUCAAAGGAA; and siBysl-976, CUGAAGAUCGCAGAAAUGGAA. The siRNA number represents the nucleotide position within the coding region that corresponds to the 3′ end of the siRNA antisense strand. Transcribed siRNAs were purified and resuspended in RNase-free water. Microinjection was performed under an inverted microscope using a programmable pneumatic microinjector (IM-300; Narishige). At E0.5 (24 h after hCG administration), fertilized eggs from BDF1 × BDF1 mating were injected with a few picoliters of 20 μM siRNA duplexes and cultured in M16 medium (Sigma).

RT-PCR analysis.Total RNA was isolated with Sepasol RNA I Super reagent (Nacalai Tesque, Kyoto, Japan). First-strand cDNA was synthesized from total RNA with SuperScript III reverse transcriptase (Invitrogen). The following primers were used for RT-PCR: Bysl-F (AGGAGTCCGGGAGGTGTTAT), Bysl-R (GCCAGAAAGTGGAAGACCAG), Oct3/4-F (GGCGTTCTCTTTGGAAAGGT), Oct3/4-R (CTCGAACCACATCCTTCTCT), Cdx2-F (ACATCACCATCAGGAGGAAAAG), Cdx2-R (CACTGGGTGACAGTGGAGTTTA), EndoA- F (GGCAGATCCATGAAGAGGAG), EndoA-R (CTTGCGGTAGGTGGTGATCT), Nanog-F (AGGGTCTGCTACTGAGATGC), Nanog-R (CAACCACTGGTTTTTCTGCC), Rex1-F (CAATAGAGTGAGTGTGCAGTGC), Rex1-R (CCTCTGTCTTCTCTTGCTTCG), Hnf4α-F (TGCCCTCTCACCTCAGCAATG), Hnf4α-R (CCCCTCAGCACACGGTTTTG), brachyury-F (GTCTTCTGGTTCTCCGATGT), brachyury-R (CCAGGTGCTATATATTGCCT), Gapdh-F (GTGTTCCTACCCCCAATGTG), Gapdh-R (GTCATTGAGAGCAATGCCAG), βactin-F (AAGTGTGACGTTGACATCCG), and βactin-R (GATCCACATCTGCTGGAAGG) (F indicates the forward primer, and R indicates the reverse primer).

BrdU incorporation assay.Embryos were cultured in M16 medium supplemented with 10 μM 5-bromo-2′-deoxyuridine (BrdU) for 60 min to label dividing cells. The embryos were then fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS), and DNA was denatured by incubation in 2 N HCl containing 0.2% Triton X-100 for 60 min at 37°C. After neutralization with 0.1 M sodium borate (pH 8.5), the embryos were blocked with 10% normal goat serum in PBS. BrdU incorporation was detected by incubation with mouse monoclonal anti-BrdU antibody (2 μg/ml; Roche) in PBS containing 1% bovine serum albumin at 4°C overnight, followed by Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G antibody.

Immunofluorescence analysis.For immunofluorescence staining, embryos were fixed with 4% paraformaldehyde in PBS and permeabilized in 0.2% Triton X-100 in PBS. After blocking with 10% normal goat serum in PBS, the embryos were incubated overnight at 4°C with the following antibodies in PBS containing 1% bovine serum albumin: rat monoclonal anti-cytokeratin 8 or EndoA (TROMA-1; Developmental Studies Hybridoma Bank) at 1:100, rabbit anti-Oct3/4 antiserum (44) at 1:10,000, and mouse monoclonal anti-Cdx2 (CDX2-88; BioGenex, CA) at the concentration given by the manufacturer. The embryos were then incubated for 30 min with Alexa Fluor 488- or 546-conjugated goat secondary antibodies against rat, rabbit, or mouse immunoglobulin at a dilution of 1:1,000. For nuclear staining, the embryos were incubated with 10 μg/ml Hoechst 33342 (Sigma) for 5 min and then observed under a confocal laser scanning microscope (Radiance 2100; Bio-Rad). For immunofluorescence staining of fibrillarin, mouse monoclonal anti-fibrillarin (38F3; Abcam) was used at 1:500, followed by Alexa Fluor 546-conjugated secondary antibody.

Plasmid construction.For episomal short hairpin RNA (shRNA) expression vectors, a human U6 promoter was amplified by PCR from pENTR/U6 (Invitrogen) with the following primers containing BbsI and AflIII sites for cloning of shRNA template oligonucleotides into the immediate downstream of the transcriptional initiation site: TACCAAGGTCGGGCAGGAAGAGGG (forward) and ACATGTGAAGACACGGTGTTTCGTCCTTTCCACAAG (reverse) (the U6 promoter sequences are underlined). This U6 cassette was ligated to a PstI-EcoRV fragment of pHPCAG (42) harboring a polyoma origin of replication with a mutated enhancer, which enables the episomal maintenance of the plasmid, and a PGK-pac-pA cassette, generating pPPU6. For each shRNA expression construct, both top and bottom template oligonucleotides were annealed and ligated into BbsI/AflIII-digested pPPU6. The following shRNA template oligonucleotides were used: shEGFP-441 Top, caccgCAGTCACAATGTCTATGTCATgtgtgctgtccATGATATAGACGTTGTGGCTGttttt; shEGFP-441 Bottom, catgaaaaaCAGCCACAACGTCTATATCATggacagcacacATGACATAGACATTGTGACTGc; shBysl-845 Top, caccgCTGGAGCTTGGTTCGAGGGAAgtgtgctgtccTTCCTTTGAACCAGGCTCCAGttttt; shBysl-845 Bottom, catgaaaaaCTGGAGCCTGGTTCAAAGGAAggacagcacacTTCCCTCGAACCAAGCTCCAGc; shBysl-976 Top, caccgCTGAGGATCGTAGAGATGGAAgtgtgctgtccTTCCATTTCTGCGATCTTCAGttttt; and shBysl-976 Bottom, catgaaaaaCTGAAGATCGCAGAAATGGAAggacagcacacTTCCATCTCTACGATCCTCAGc. Each pair of oligonucleotides consists of the 21-nucleotide sense strand of the target (first set of capital letters) with three G-T mismatch mutations (underlined), the optimized loop sequence derived from the human microRNA (lowercase letters), the antisense strand (second set of capital letters), the RNA polymerase III transcription terminator, and the overhanging sequences (37).

For the Bysl episomal expression vectors, the full-length mouse Bysl cDNA (GenBank accession number AK143027; 1,308 bp) was amplified by RT-PCR from blastocyst RNA and cloned into pHPCAG or pPPCAG in which the PGK-hph-pA cassette in pHPCAG was replaced by the PGK-pac-pA cassette. For the Bysl-Venus fusion constructs, Venus (39) was fused to the C terminus of Bysl by the linker peptide RSESLVNSAVDGTAGPGST derived from pEGFP-N1 (Clontech).

The RNAi-resistant Bysl-Venus construct was obtained by introducing silent mutations that do not change the coded amino acid sequence into Bysl-Venus cDNA using inverse PCR. Three silent mutations were generated in each sequence complementary to shBysl-845 and shBysl-976. The wild-type coding region in the pHPCAG or pPPCAG expression vector was replaced by the mutated sequence. The deletion mutants of Bysl fused to Venus were also constructed using inverse PCR.

For the expression of Bysl-Venus in embryos, the EGFP cDNA in the pcDNA3.1EGFP-poly(A83) vector (70) was replaced by the cDNA encoding Bysl-Venus. After linearization of the expression vector, in vitro mRNA transcription was performed using the RiboMAX Large Scale RNA Production Systems-T7 kit (Promega) in the presence of m⁷G(5′)ppp(5′)G RNA capping analog (Invitrogen). A few picoliters of 100 ng/μl denatured mRNA was microinjected.

ES cell culture and supertransfection.MGZ5 ES cells (44) were cultured in the absence of feeder cells in Dulbecco's modified Eagle medium supplemented with 15% fetal calf serum, 1 mM sodium pyruvate, 0.1 mM 2-mercaptoethanol, 0.1 mM nonessential amino acids, 50 U/ml penicillin-streptomycin, 1,000 U/ml of leukemia inhibitory factor (LIF), 100 μg/ml G418, and 10 μg/ml zeocin on gelatin-coated dishes. For supertransfection with episomal vectors, MGZ5 ES cells, which had been plated the previous day at a density of 3 × 10⁴ cells/well in 24-well plates, were transfected with 1 to 4 μg/ml supercoiled plasmid vectors using Lipofectamine 2000 (Invitrogen) in 300 μl of Opti-MEM I containing 10% fetal calf serum for 3 to 4 h and then replated into six-well plates. One day after transfection, the cells were selected with 1 μg/ml puromycin or 100 μg/ml hygromycin B. Six to seven days after transfection, the cells were fixed and stained for alkaline phosphatase (AP) activity.

Pulse-chase labeling of rRNA.Metabolic labeling of rRNA was performed as previously described (59). At 48 h posttransfection, ES cells were preincubated in methionine-free medium for 15 min and then incubated in medium containing 50 μCi/ml l-[methyl-³H]methionine (Amersham Biosciences) for 30 min. The cells were subsequently chased with nonradioactive medium containing 15 μg/ml l-methionine for various time periods. Following pulse-chase labeling, RNA was isolated using Sepasol RNA I Super reagent. Label incorporation was measured by scintillation counting, and samples containing equal amounts of radioactivity were electrophoresed on a 1% agarose-formaldehyde gel and transferred to a nylon membrane (GeneScreen Plus; PerkinElmer), which was sprayed with En³Hance (PerkinElmer) and exposed to Kodak BioMax MS film at −80°C for 3 days.

Sucrose density gradient fractionation.Polysome profiles were analyzed as previously described (59). At 48 h posttransfection, ES cells were treated with 50 μg/ml cycloheximide (CHX) for 10 min and harvested by trypsinization. Equal numbers of cells from each sample were resuspended in lysis buffer (20 mM Tris-HCl [pH 7.4], 130 mM KCl, 10 mM MgCl₂, 2.5 mM dithiothreitol, 0.5% NP-40, 0.5% sodium deoxycholate, 10 μg/ml CHX, 0.2 mg/ml heparin, and 200 U/ml RNase inhibitor [TOYOBO]) and incubated on ice for 15 min. The lysates were centrifuged at 15,000 × g for 15 min and the supernatants were layered over 10 to 45% (wt/wt) sucrose density gradients in polysome buffer (10 mM Tris-HCl [pH 7.4], 60 mM KCl, 10 mM MgCl₂, 1 mM dithiothreitol, 0.1 mg/ml heparin) formed with Biocomp Gradient Master. The gradients were centrifuged at 220,000 × g for 30 min or at 160,000 × g for 100 min at 4°C in an RPS55T-2 rotor (Hitachi) and fractionated using a BIOCOMP piston gradient fractionator equipped with a Bio-Mini UV monitor for continuous measurement of the absorbance at 260 nm.

Transmission electron microscopy.Preimplantation embryos microinjected with siRNAs were exposed to acidic Tyrode's solution (Sigma) for removal of the zona pellucida and fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.0) at 4°C overnight. The embryos were washed, postfixed in 1% osmium tetroxide for 60 min on ice, dehydrated with a series of ethanol gradients, and embedded in Epon 812 resin mixture (TAAB). The resin was polymerized at 70°C for 2 days. Ultrathin sections were stained with 2% uranyl acetate in 70% ethanol and Raynold's lead citrate and examined on a Hitachi H-7500 electron microscope.

Bysl is crucial for the development of preimplantation embryos.In an attempt to elucidate the molecular mechanisms underlying developmental processes of preimplantation embryos, we identified several genes that are highly expressed in preimplantation embryos by in silico analysis. We selected 89 such genes whose functions have not been fully characterized and analyzed their expression patterns by RT-PCR. Of 77 genes detected in preimplantation embryos, we picked 32 genes that were preferentially expressed in embryos compared to in adult tissues. To investigate possible roles in preimplantation development, we screened these genes by using siRNA. A mixture of two distinct siRNA duplexes for each gene was microinjected into the cytoplasm of fertilized eggs at E0.5. In this screen, five genes were identified as factors whose depletion by siRNA inhibited blastocyst development. Among these genes, the development of embryos injected with siRNA against Bysl was most significantly impaired just prior to blastocyst formation, suggesting direct involvement of Bysl in this process.

Microarray analyses have shown that Bysl is enriched in embryonic stem cells and certain types of adult stem cells (hematopoietic, neural, and retinal stem/progenitor cells) in mice (18, 25, 26, 49, 63) and humans (1). We found that Bysl is highly expressed in preimplantation embryos as well as some adult tissues, including brain, bone marrow, and gonads (Fig. 1A). During the preimplantation stage, Bysl mRNA was detected continuously from unfertilized eggs to blastocysts (Fig. 1B), suggesting that Bysl is expressed maternally.

• Open in new tab
• Download powerpoint

FIG. 1.

Bysl is crucial for the development of preimplantation embryos. (A) Bysl expression in various mouse tissues was analyzed by RT-PCR. Oct3/4 is a marker for pluripotent and germ line cells. (B) Bysl expression during the preimplantation period. Oct3/4 expression was restricted to pluripotent stem cells, while Cdx2 and EndoA were specifically expressed in the trophoblast lineage. TS, trophoblast stem; EB, embryoid body 10 days after plating; MEF, mouse embryonic fibroblast. (C) Three distinct siRNAs against Bysl inhibited blastocyst formation. Fertilized eggs were microinjected with 20 μM siRNA solution at E0.5, and blastocyst formation was evaluated four days after injection (E0.5+4.0). Each value represents the mean ± standard deviation from three independent experiments. Total numbers of embryos injected with siEGFP-441, siBysl-773, siBysl-845, and siBysl-976 were 40, 46, 54, and 51, respectively. Significance levels were determined by Student's t test (**, P < 0.01; ***, P < 0.001). (D) The severity of the phenotype was correlated with the effective downregulation of Bysl mRNA. Bysl expression was analyzed by RT-PCR at E0.5+3.25. (E) Developmental arrest caused by Bysl siRNAs was overcome by coinjecting Bysl mRNA in a dose-dependent manner. Embryos were injected with a mixture containing 5 μM each of siBysl-845 and siBysl-976 and various concentrations of in vitro synthesized Bysl-Venus mRNA used in the experiment whose results are described in Fig. 4B. The embryos were photographed at E0.5+4.0. Scale bar, 100 μm.

Although RNAi is a process of sequence-specific gene silencing, off-target effects can be encountered. To ensure the specificity of RNAi, we performed independent experiments with three distinct siRNAs targeted to different regions of Bysl coding sequence. Each Bysl siRNA exerted similar phenotypic effects with various efficiencies (Fig. 1C); the severity of the phenotype was correlated with the observed reduction in Bysl mRNA levels (Fig. 1D). We used a mixture of highly effective siRNAs (siBysl-845 and siBysl-976) for further analyses, both of which inhibited blastocyst formation in a dose-dependent manner (see Fig. S1 in the supplemental material), thus assuring that the minimum effective amount of siRNA was used. Moreover, the siRNA-dependent developmental arrest could be overcome by coinjecting in vitro synthesized Bysl mRNA (Fig. 1E), indicating the specificity of the RNAi effect.

The development of embryos injected with Bysl siRNAs was arrested at the 16-cell stage.To determine the time point at which development was arrested, morphological alterations in the Bysl siRNA-injected embryos were examined during preimplantation development. When Bysl siRNAs were injected into fertilized eggs, compaction at the eight-cell stage occurred normally in vitro. However, blastocyst formation was completely inhibited, although intracellular fluid accumulation in some embryos was observed (Fig. 2A). These embryos failed to hatch from the zona pellucida, could not outgrow in culture even when the zona was removed, and eventually degenerated (all injected embryos). Oct3/4 siRNA, which was used as a control, also inhibited blastocyst development albeit with a lower efficiency than Bysl siRNAs (22% ± 15% developmental rate compared with 96% ± 8% for EGFP siRNA). Chronological analysis of the number of nuclei visualized with Hoechst 33342 showed that most of the embryos injected with Bysl siRNAs were arrested at the 16-cell stage (Fig. 2B). This was accompanied by reduced BrdU incorporation into newly synthesized DNA (Fig. 2C), while no increase in apoptosis, as determined by terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling assay, was detected up to this stage (data not shown). However, prolonged arrest over 24 h at the 16-cell stage (3.7 days after E0.5 [E0.5+3.7]) inevitably caused embryonic lethality. The developmental arrest at this stage was also observed when siRNAs were injected into both blastomeres of two-cell stage embryos at E1.5 (see Fig. S2 in the supplemental material), suggesting temporal specificity of the requirement for Bysl.

• Open in new tab
• Download powerpoint

FIG. 2.

The development of embryos injected with Bysl siRNAs was arrested at the 16-cell stage, resulting in a defect in differentiation toward trophectoderm. (A) Morphological alterations in siRNA-injected embryos were examined during blastocyst formation. Embryos were injected with 20 μM siEGFP-441, siOct3/4-670, or siBysl mixture containing 10 μM each of siBysl-845 and siBysl-976. Scale bar, 50 μm. (B) Cell numbers were determined periodically. Embryos were exposed to 10 μg/ml Hoechst 33342 and squashed under a coverslip, and the numbers of fluorescent nuclei were counted. Similar results were obtained in three independent experiments. **, P < 0.01; ***, P < 0.001. (C) The proliferative activity was assayed by BrdU incorporation. Untreated, without treatment with BrdU. Scale bar, 100 μm. (D) Marker gene expression was analyzed by RT-PCR at E0.5+3.25. (E) Confocal immunofluorescence imaging of cytokeratin EndoA with TROMA-1 antibody. Nuclei were stained with Hoechst 33342 (H33342). Fourfold-magnified views of the boxed areas are shown in the smaller, lower panels. Scale bar, 50 μm. (F) Immunofluorescent localization of Oct3/4 and Cdx2. Overlapping expression of Oct3/4 and Cdx2 in the nuclei appears white in the merged images. Scale bar, 50 μm. DIC, differential interference contrast.

Embryos injected with Bysl siRNAs failed to induce differentiation into trophectoderm.Blastocyst formation requires development of the TE layer with concomitant apical-basolateral membrane polarity which allows vectorial fluid secretion and a tight junctional permeability seal that facilitates fluid accumulation (57). Therefore, we determined whether differentiation into TE was induced in Bysl siRNA-injected embryos. RT-PCR analysis of the injected embryos showed that the expression of the cytokeratin EndoA gene, an early TE marker, was slightly reduced (Fig. 2D). The failure to induce TE-specific gene expression was corroborated by immunostaining. Control blastocysts showed assembled EndoA structures in the TE layer (Fig. 2E, upper panels). In contrast, no such organized structures were detected in the Bysl siRNA-injected embryos (Fig. 2E, lower panels). We also examined the expression of Oct3/4 and Cdx2, the two mutually antagonistic and lineage-specific transcription factors that are involved in segregation of the pluriblast and trophoblast lineages, respectively (41, 44, 60). Oct3/4 was detectable in both inner and outer cells of control morulae and early blastocysts (Fig. 2F, upper and middle panels, respectively). On the other hand, Cdx2 was repressed in the inner cells as early as the morula stage but was maintained in the outer cells at the morula stage through the blastocyst stage, consistent with a previous report (44). In the Bysl siRNA-injected embryos (Fig. 2F, lower panels), however, Cdx2 was barely detectable, whereas Oct3/4 was maintained at the equivalent stage. These results indicate that the nascent differentiation toward the TE lineage is abolished by the silencing of Bysl.

Episomal expression of Bysl shRNAs inhibited proliferation of embryonic stem cells.Next, we investigated the function of Bysl in ES cells, because the expression and functional requirement for Bysl in the epiblast, the embryonic counterpart of ES cells, have been shown (5). To achieve stable and homogeneous expression of siRNA in ES cells, we constructed a polyoma-based episomal RNAi vector harboring the human U6 promoter for shRNA expression. Using episomal supertransfection (19), we found that expression of both shBysl-845 and shBysl-976 markedly inhibited proliferation of ES cells (Fig. 3A). Fluorescence-activated cell sorting analysis showed no significant differences in cell cycle distribution 48 h posttransfection (see Fig. S3 in the supplemental material). Bysl shRNA-treated ES cells exhibited no obvious morphological differentiation, although they appeared to be disorganized (Fig. 3B). Expression of undifferentiated ES cell markers such as Oct3/4, Nanog, and Rex1, or differentiation markers such as Cdx2, Hnf4α, and brachyury, was not altered (Fig. 3C). Moreover, inhibition of proliferation caused by Bysl shRNAs was also observed in differentiated ES cells after LIF withdrawal (data not shown), indicating that Bysl is required not only for the proliferation of undifferentiated stem cells but also for differentiated cells. Similar results were obtained when another ES cell line, ZHBTc4, was transiently transduced with Bysl siRNAs or shRNAs (data not shown).

• Open in new tab
• Download powerpoint

FIG. 3.

Episomal expression of Bysl shRNAs inhibited proliferation of embryonic stem cells. (A) MGZ5 ES cells were transfected with 2 μg/ml of the pPPU6 episomal shRNA expression vector. One day following transfection, the cells were selected with 1 μg/ml puromycin. Seven days after transfection, undifferentiated ES cell colonies were visualized by staining for AP activity. Scale bar, 10 mm. (B) Morphological alterations in ES cells cotransfected with 1 μg/ml pHPCAG-Venus and 2 μg/ml pPPU6-shRNA. The cells were selected with 1 μg/ml puromycin and 100 μg/ml hygromycin B, and then photographed under epifluorescence at three days posttransfection. Scale bar, 200 μm. (C) Marker gene expression was analyzed by RT-PCR at 48 h posttransfection. Cdx2, Hnf4α, and brachyury are markers for trophoblast, endoderm, and mesoderm differentiation, respectively. ZHBTc4 ES cells (43) cultured in the presence of 1 μg/ml tetracycline for 24 h, MGZ5 ES cells cultured in the absence of LIF for six days, and embryoid bodies five days after plating were used as positive controls (P.C.) for Cdx2, Hnf4α, and brachyury, respectively.

Bysl localized to the nucleus and was concentrated in the nucleolus.In a previous report, immunocytochemical analysis showed that human BYSL is localized in the cytoplasm of trophoblast cells (62). To analyze the subcellular localization of mouse Bysl, we produced Bysl-Venus by fusing the yellow fluorescent protein Venus to the C terminus of Bysl. Exogenously expressed Bysl-Venus was highly enriched in the nucleolus and was also present diffusely in the nucleoplasm of COS-7 cells, whereas no significant signal was detected in the cytoplasm (Fig. 4A). We obtained a similar localization pattern using Bysl-DsRed fusion protein (see Fig. S4 in the supplemental material). The nucleolar/nucleoplasmic localization of Bysl-Venus was also observed not only in embryos (Fig. 4B) and undifferentiated ES cells (Fig. 4C, upper panels) but also in ES cells differentiated into the trophoblast lineage (Fig. 4C, lower panels).

• Open in new tab
• Download powerpoint

FIG. 4.

Bysl localized to the nucleus and was concentrated in the nucleolus. (A) A Bysl-Venus fusion protein was expressed in COS-7 cells. The cells were fixed at one day posttransfection, immunostained for the nucleolar marker fibrillarin, and observed under a confocal laser scanning microscope. Venus alone diffusely distributed throughout the cytoplasm and nucleus (upper panels). H33342, Hoechst 33342. Scale bar, 20 μm. (B) Bysl-Venus mRNA transcribed in vitro was injected into fertilized eggs and photographed under epifluorescence one day later (upper panels), or injected into both blastomeres of two-cell stage embryos and photographed one day (middle panels) or two days (lower panels) later. Scale bar, 50 μm. (C) ZHBTc4 ES cells were transfected with a Bysl-Venus construct and cultured in the absence (undifferentiated; upper panels) or presence (differentiated into trophoblast lineage; lower panels) of 1 μg/ml tetracycline (Tc) for four days. Scale bar, 20 μm. DIC, differential interference contrast.

RNAi-resistant Bysl, which localizes to the nucleolus/nucleoplasm, overcame the proliferation inhibition caused by Bysl shRNAs.We next determined whether Bysl-Venus could functionally substitute for endogenous Bysl. For this, an RNAi-resistant Bysl-Venus construct was produced by introducing silent mutations into each of the two shRNA target sequences (Fig. 5A and B). Using this construct, we tested whether the expression of RNAi-resistant Bysl-Venus could restore the proliferation of Bysl shRNA-treated ES cells. Cotransfection of Bysl shRNA and wild-type or silently mutated Bysl-Venus expression vectors demonstrated that the mutated Bysl-Venus was sufficiently resistant to Bysl shRNAs (Fig. 5C). ES cells were first transfected with wild-type or RNAi-resistant Bysl-Venus vector and subsequently transduced with Bysl shRNAs. Wild-type Bysl-Venus partially rescued colony formation in the Bysl shRNA-treated ES cells, although the number and size of the colonies were significantly reduced compared to those of the EGFP shRNA-treated colonies (Fig. 5D, middle panels). In contrast, the expression of RNAi-resistant Bysl-Venus nearly overcame the proliferation inhibition caused by depletion of endogenous Bysl (Fig. 5D, lower panels). Taken together, these findings provide evidence that the inhibitory effect is solely due to the silencing of Bysl and that Bysl-Venus, which predominantly localizes to the nucleolus/nucleoplasm, can functionally substitute for endogenous Bysl.

• Open in new tab
• Download powerpoint

FIG. 5.

RNAi-resistant Bysl, which localizes to the nucleolus/nucleoplasm, overcame the proliferation inhibition caused by Bysl shRNAs. (A) The predicted structures of two shRNAs against Bysl. The three G-U mismatch mutations in the sense strand are presented in lowercase letters. (B) Silent mutations (underlined) were generated within each of the shRNA target sequences in Bysl-Venus to confer RNAi resistance. (C) MGZ5 ES cells were cotransfected with shRNA (2 μg/ml) and wild-type or silently mutated (sm) Bysl-Venus (2 μg/ml) expression vectors. The cells were photographed under epifluorescence at two days posttransfection. Note that smBysl-Venus was resistant to Bysl shRNAs. Scale bar, 200 μm. (D) Rescue of the knockdown phenotype by expression of RNAi-resistant Bysl. MGZ5 ES cells were first transfected with 2 μg/ml pHPCAG-ByslVenus or -smByslVenus followed by selection with 100 μg/ml hygromycin B. Four days after transfection, equal numbers of cells from each sample were subsequently transfected with 4 μg/ml pPPU6-shRNA followed by selection with 100 μg/ml hygromycin B and 1 μg/ml puromycin. The cells were stained for AP activity six days after the second transfection. Scale bar, 10 mm.

Deletion mutants of Bysl, which retain nucleolar localization, exerted a dominant negative effect.Bysl is highly conserved throughout evolution but does not contain any conserved functional domains or nuclear localization motifs. To identify the regions critical for its nuclear localization and functionality, deletion mutants were generated (Fig. 6A). The deleted regions share various degrees of amino acid identity/similarity among mice, humans, and yeast. By Western blot analysis, the Bysl-Venus deletion mutants were detected at the predicted molecular sizes, except the d1 (with amino acids 1 through 133 deleted [Δ1-133]) mutant, which was slightly smaller than expected (Fig. 6B). The subcellular localizations of these mutants were altered to various extents compared to that of wild-type Bysl (Fig. 6C). Only d1 and d2 (Δ134-175) retained nucleolar localization, though both were detected in minimal amounts in the nucleoplasm, and d2 displayed a nuclear speckle-like pattern as well. The other three mutants, d3 (Δ176-236), d4 (Δ237-309), and d5 (Δ310-436), accumulated in the cytoplasm and were excluded from the nucleus.

• Open in new tab
• Download powerpoint

FIG. 6.

Deletion mutants of Bysl, which retain nucleolar localization, exerted a dominant negative effect. (A) Schematic representation of Bysl deletion mutants fused to Venus. Deletion mutants were generated based on the sequence similarity among humans, mice, and yeast (32.5% and 54.7% amino acid identity and similarity in total). (B) Western blot analysis of the mutant proteins expressed in COS-7 cells using anti-GFP fluorescent protein antibody (SC-8334; Santa Cruz Biotechnology). The deletion mutants were detected at the predicted molecular masses (MM), except the d1 mutant, which had an observed MM of 55 to 60 kDa. (C) Subcellular localizations of the mutants were determined in COS-7 cells. Note that d1 and d2 retained nucleolar localization. Scale bar, 10 μm. H33342, Hoechst 33342. (D) Functionalities of the deletion mutants were assessed by rescue of the knockdown phenotype in MGZ5 ES cells as described in the legend to Fig. 5D. (E) MGZ5 ES cells were transfected with 2 μg/ml pPPCAG constructs expressing the mutant proteins followed by selection with 1 μg/ml puromycin. The cells were stained for AP activity at six days posttransfection. Note that overexpression of d1 or d2 suppressed the proliferation of ES cells. (F) Summary of the properties of Bysl deletion mutants. The exact relationship between nucleolar localization and inhibitory effect is noted. Cy, cytoplasm; Np, nucleoplasm; No, nucleolus; NuSp, nuclear speckle-like; ND, not determined.

To determine whether the deletion mutants could functionally substitute for endogenous Bysl, ES cells were serially transfected with the Bysl-Venus mutant and Bysl shRNA expression vectors. None of the mutants were able to rescue the knockdown phenotype in ES cells (Fig. 6D), indicating that all of them are nonfunctional. In addition, d1 and d2, which retain nucleolar localization, suppressed proliferation when overexpressed (Fig. 6E), suggesting that they exerted a dominant negative effect. The close correlation between nucleolar localization of the mutant molecules and their inhibitory effects (Fig. 6F) reinforces the idea that endogenous Bysl functions not in the cytoplasm but in the nucleolus.

Bysl is required for 40S ribosome biogenesis.To investigate the molecular function of Bysl, we examined the involvement of Bysl in ribosome biogenesis because functional Bysl is localized in the nucleolus (Fig. 4 to 6) and yeast Enp1 is involved in pre-rRNA processing and 40S subunit biogenesis (11). The nucleolus is a subnuclear structure that functions primarily in processes of ribosome biogenesis, including rRNA transcription, pre-rRNA processing, and ribosome subunit assembly (45). First, pre-rRNA processing was analyzed by pulse-chase labeling after treatment with Bysl shRNAs. Pre-rRNA was pulse-labeled with [methyl-³H]methionine at 48 h posttransfection and its processing to mature rRNA was analyzed at various time intervals after chase. During rRNA processing, the primary 47S transcript is rapidly processed to the 20S and 32S intermediates, which are further processed to the mature 18S and 5.8/28S rRNAs, which are components of the 40S and 60S ribosomal subunits, respectively (9, 15, 32). In the Bysl shRNA-treated ES cells, the synthesis of the 28S rRNA was not affected. In contrast, the production of the 18S rRNA was greatly inhibited and a marked accumulation of the 20S precursor was observed (15 to 60 min after chase) (Fig. 7A).

• Open in new tab
• Download powerpoint

FIG. 7.

Bysl is required for 40S ribosome biogenesis. (A) Inhibition of rRNA processing in Bysl shRNA-treated ES cells. MGZ5 ES cells, which had been plated the previous day at a density of 6 × 10⁴ cells/well in 12-well plates, were transfected with 2 μg/ml pPPU6-shRNA followed by selection with 1 μg/ml puromycin. At 48 h posttransfection, the cells were pulse-labeled with l-[methyl-³H]methionine for 30 min and subsequently chased in nonradioactive medium for the indicated time. The expected positions of mature rRNAs and major precursors are indicated on the right. Similar results were obtained in three independent experiments. (B and C) A deficit in 40S ribosomal subunits in ES cells transfected with Bysl shRNA (B) or deletion mutant (C) expression vectors. At 48 h posttransfection, cytoplasmic extracts from 1.0 × 10⁶ (B) or 1.5 × 10⁶ (C) cells per sample were isolated and centrifuged on 10 to 45% sucrose density gradients at 160,000 × g for 100 min (B, left panel, and C) or at 220,000 × g for 30 min (B, right panel). Absorbancy at 260 nm was continuously monitored. Similar results were obtained in three independent experiments. (D) Impairment of nucleologenesis and ribosome biogenesis in Bysl siRNA-injected embryos. Embryos injected with siRNAs were fixed at the 16-cell stage (E0.5+2.75) and examined by transmission electron microscopy. The control embryos contained fully differentiated, reticulated nucleoli composed of FCs surrounded by DFCs and GCs. In contrast, the Bysl siRNA-injected embryos contained immature nucleoli in which inactive electron-dense fibrillar spheres (FSs) were apparent at the core. Moreover, in spite of the comparable amounts of preribosomal particles (arrows in left panels) at the nucleolar periphery as well as in the nucleoplasm, only few clustered ribosomes (arrowheads) and rough endoplasmic reticula (arrows in right panel) were observed in the cytoplasm, compared with the control embryos. Scale bar, 1 μm.

Next, the ribosomal profile of cytoplasmic lysates was analyzed by sucrose density gradient sedimentation. The Bysl shRNA-treated ES cells showed a dramatic decrease in the quantity of free 40S subunits, resulting in an accumulation of 60S subunits: the 40S/60S ratios of cells treated with shBysl-845 and shBysl-976 were 14% and 22% of control, respectively (Fig. 7B, left panels). Polysomal mRNAs that are actively engaged in translation were also decreased (Fig. 7B, right panels). Overexpression of deletion mutant d1 or d2 also provoked a decrease in cytoplasmic 40S ribosomal subunits (the 40S/60S ratio was 40% of control for both mutants) (Fig. 7C), which can be attributed to the dominant negative effect. Taken together, these results demonstrate that Bysl plays a crucial role in the maturation of pre-40S ribosomal particles through involvement in 18S rRNA processing.

Finally, we determined whether Bysl siRNA-injected embryos exhibit any defects in ribosome biogenesis by using transmission electron microscopy. In mice, embryonic nucleologenesis, in which an inactive nucleolus precursor body (NPB) is gradually organized into a functional nucleolus, occurs during preimplantation development (17). At the early cleavage stage, the NPB, seen as an electron-dense fibrillar sphere, is composed of a dense network of uniformly packed filaments. At the onset of rRNA transcription in two- to four-cell stage embryos, a reticulated structure, the nucleolonema, is formed at the periphery of the NPB. Extensive reticulation of the NPB periphery takes place in the four- to eight-cell stage, while the fibrillar sphere remains in the core of the nucleolus until the morula stage (21). Active transcription was detected in the peripheral regions preceded by the assembly of the nucleolar components therein (7, 13, 71), whereas the fibrillar spheres take no part in transcription (30).

In control morulae (around the 16-cell stage), fully differentiated, reticulated nucleoli were apparent, and they showed three morphologically distinct nucleolar subcompartments: fibrillar centers (FCs), spherical electron-lucent structures; dense fibrillar components (DFCs), more-electron-dense materials surrounding FCs; and granular components (GCs), in which FC-DFC complexes are embedded (Fig. 7D, upper left panel) (14). These control nucleoli exhibited a substantial number of relatively small FCs, reflecting active transcription (24). The GCs, which are distributed throughout the nucleolus, contained a large quantity of preribosomal particles. It is likely that most of such preribosomal particles in the nucleolus represent precursors of the large subunits because the small ribosomal subunits mature more rapidly than large subunits and are immediately exported to the cytoplasm (22, 48, 65). In contrast, the nucleolar structures remained rudimentary in Bysl siRNA-injected embryos. Solid fibrillar spheres were apparent in the cores of the nucleoli, while reticulated structures at the periphery contained several FCs (Fig. 7D, lower left panel). Condensed chromatin was scarcely observed in the nucleoplasm. Furthermore, in spite of the comparable quantities of preribosomal particles at the nucleolar periphery as well as in the nucleoplasm, clustered ribosomes and rough endoplasmic reticula were inconspicuous in the cytoplasm (Fig. 7D, lower right panel). In conclusion, these data demonstrate that Bysl is crucial for early embryonic development as an integral factor for ribosome biogenesis.