|Home | About | Journals | Submit | Contact Us | Français|
Inscuteable (Insc) regulates cell fate decisions in several types of stem cells. Although it is recognized that the expression levels of mouse INSC govern the balance between symmetric and asymmetric stem cell division, regulation of mouse Insc gene expression remains poorly understood. Here, we showed that mouse Insc expression transiently increases at an early stage of differentiation, when mouse embryonic stem (mES) cells differentiate into bipotent mesendoderm capable of producing both endoderm and mesoderm in defined culture conditions. We identified the minimum transcriptional regulatory element (354 bases) that drives mouse Insc transcription in mES cells within a region >5 kb upstream of the mouse Insc transcription start site. We found that the transcription factor reticuloendotheliosis oncogene (c-Rel) bound to the minimum element and promoted mouse Insc expression in mES cells. In addition, short interfering RNA-mediated knockdown of either mouse INSC or c-Rel protein decreased mesodermal cell populations without affecting differentiation into the mesendoderm or endoderm. Furthermore, overexpression of mouse INSC rescued the mesoderm-reduced phenotype induced by knockdown of c-Rel. We propose that regulation of mouse Insc expression by c-Rel modulates cell fate decisions during mES cell differentiation.
Insc was first identified as a novel neural precursor gene in Drosophila (1). Insc protein expression has been detected in embryonic areas where cell shape changes or movement occurs (i.e. neuroectoderm, midgut primordium, and muscle precursors) (1). More precise roles have emerged for Insc protein activity based on studies using neuroblasts, stem cells found in the central nervous system of Drosophila, which undergo asymmetric cell division (2,–5). In neuroblasts, Insc localizes to the apical cell cortex by directly associating with bazooka-Par6-aPKC cell polarity protein complexes, whereas cell fate determinants, such as miranda (Mira), prospero (Pros), brain tumor (Brat), and Numb, localize to the basal cortex (6,–20). Alignment of the mitotic spindle along the apical-basal polarity axis drives asymmetric cell division to produce one self-renewing neuroblast and one ganglion mother cell fated for neural differentiation by asymmetrical inheritance of cell fate determinants (6, 10, 11, 13, 21, 22). Insc plays a critical role in apical-basal spindle positioning by connecting spindle capturing machineries, consisting of partner of inscuteable (Pins) and mushroom body defect (Mud), with apical bazooka-Par6-aPKC cell polarity complexes (23,–26).
Similar molecular machineries are conserved in neural progenitors (27, 28) and skin basal cells of mice (29,–31), whereby mouse INSC (the mouse homologue of mammalian inscuteable) regulates spindle orientation and cell fate determination together with Par-3 (vertebrate homolog of Bazooka) and LGN (vertebrate counterpart of Pins) (27,–31). Previous reports show that ectopic expression of mouse INSC promotes apical-basal spindle positioning in neuronal progenitors and keratinocytes, whereas loss of mouse INSC randomizes or promotes a planar spindle orientation (27,–31). Importantly, loss of mouse INSC in radial glia cells decreases neurogenesis, leading to defects in cortical organization, whereas mouse INSC overexpression expands the neuronal cell pool. Therefore, expression levels of mouse INSC appear to be critical for cell fate decision and generation of the correct number of differentiated cells. However, regulation of mouse Insc gene expression remains poorly understood, with little information on mouse Insc promoters. One reason for this gap in knowledge is the lack of established approaches to investigate regulation of mouse Insc gene expression during mammalian cell differentiation.
Embryonic stem (ES)2 cells are pluripotent and can be differentiated into all cell types found throughout the body (32,–35). Here, we demonstrate that expression of mouse INSC transiently increases during mouse ES (mES) cell differentiation into bipotent mesendoderm cells capable of giving rise to both endoderm and mesoderm lineages in defined culture conditions (36, 37). In this system, we identified DNA regulatory elements involved in mouse Insc gene expression, which are located more than 5 kb upstream of the mouse Insc transcription start site (TSS). We specified the minimum transcription-promoting sequences and identified c-Rel as a key transcription factor that drives mouse Insc expression in mES cells. Knockdown of mouse INSC or c-Rel protein leads to a decrease in the proportion of mesoderm cells without alterations in mesendoderm and endoderm cells, indicating a requirement for mouse INSC in the mesoderm cell fate decision. Our results provide further supporting evidence for how c-Rel regulates mesoderm differentiation by promoting mouse Insc expression. This study demonstrates for the first time that the c-Rel/mouse INSC axis regulates mesoderm cell fate decision during mES cell differentiation.
All cell culture products, unless noted otherwise, were Gibco brand purchased from Life Technologies. Goosecoid (Gsc)gfp/+ ES cells were maintained on gelatin-coated dishes in Glasgow minimum essential medium supplemented with 1% fetal calf serum (FCS), 10% KnockOutTM serum replacement, 0.1 mm nonessential amino acids, 1 mm sodium pyruvate, 0.1 mm 2-mercaptoethanol, and 1 μl/ml leukemia inhibitory factor (Wako Chemicals). Gscgfp/+ ES/mouse INSC-mCherry and Gscgfp/+ ES/mCherry cells were maintained on gelatin-coated dishes in Glasgow minimum essential medium supplemented with 1% FCS, 10% KnockOutTM serum replacement, 0.1 mm nonessential amino acids, 1 mm sodium pyruvate, 0.1 mm 2-mercaptoethanol, 1 μl/ml leukemia inhibitory factor, and 100 μg/ml Geneticin (Nakarai). For mesendoderm induction, ES cells were seeded onto type IV collagen-coated dishes at a density of 1 × 104 cells/ml in SF-O3 medium (Sanko Junyaku) containing 0.1% bovine serum albumin (BSA; Sigma-Aldrich), 50 μm 2-mercaptoethanol, and 10 ng/ml activin A (R&D Systems). HEK293T cells were cultured in Dulbecco's modified Eagle's medium with 10% FCS.
Cells were lysed in lysis buffer (50 mm Tris-HCl, pH 8.0, 150 mm NaCl, 1% Nonidet P-40, 2 mm EGTA, 2 mm MgCl2, 2 mm dithiothreitol (DTT), 1 mm phenylmethylsulfonyl fluoride, 1 mm Na3VO4, and 20 μg/ml aprotinin) and centrifuged at 13,000 rpm at 4 °C for 15 min. Supernatants were subjected to Western blotting. Primary antibodies were mouse monoclonal anti-FLAG (F3165, Sigma-Aldrich), rabbit polyclonal anti-Eomes (ab23345, Abcam), goat polyclonal anti-Foxa-2 (sc-9187, Santa Cruz Biotechnology), rabbit polyclonal anti-T-bra (sc-20109, Santa Cruz Biotechnology), mouse polyclonal anti-Par-3 (07-330, Millipore), rabbit anti-LGN (a gift from Dr. Matsuzaki (Riken CDB), rabbit monoclonal anti-Elk1 (E277, Abcam), rabbit monoclonal anti-Ets1 (14069, CST), rabbit polyclonal anti-cRel (sc-71, Santa Cruz Biotechnology), rabbit polyclonal anti-DsRed (632496, Clontech), and mouse monoclonal anti-α-tubulin (T6199, Sigma-Aldrich). An anti-mouse INSC antibody was prepared as described previously (38). Primary antibodies were detected with horseradish peroxidase-conjugated secondary antibodies (GE Healthcare) using Western Lightning® ECL reagents (PerkinElmer Life Sciences) according to the manufacturer's instructions. For immunoprecipitation of mouse INSC-mCherry, cells were lysed in lysis buffer B (50 mm Hepes, pH 7.5, 2 mm EGTA, 2 mm MgCl2, 12.5 mm β-glycerophosphate, 50 mm NaCl, 10% glycerol, 0.5% Nonidet P-40, 10 mm NaF, 1 mm DTT, 1 mm phenylmethylsulfonyl fluoride, 1 mm Na3VO4, 2 μg/ml aprotinin, and 1 μg/ml leupeptin) and centrifuged at 15,000 rpm for 15 min at 4 °C. The mouse anti-mCherry antibody (4 μg) was added to the supernatant, followed by incubation at 4 °C for 1 h. Protein A- or G-Sepharose beads (GE Healthcare), preincubated with 3% BSA-PBS, were added to the mixtures, followed by incubation at 4 °C for 2–3 h, and then washed three times with wash buffer (50 mm Hepes, pH 7.5, 2 mm EGTA, 2 mm MgCl2, 12.5 mm β-glycerophosphate, 150 mm NaCl, 0.5% Nonidet P-40, 10 mm NaF, 1 mm DTT, 1 mm phenylmethylsulfonyl fluoride, 1 mm Na3VO4, 2 μg/ml aprotinin, and 1 μg/ml leupeptin) and subjected to Western blotting.
Total DNA was isolated using phenol/chloroform and ethanol and then subjected to genomic PCR. Total RNA was isolated using an RNeasy® minikit (Qiagen) according to the manufacturer's protocol. First-strand cDNA was synthesized using Moloney murine leukemia virus reverse transcriptase (Life Technologies, Inc.) and random primers (Takara) for quantitative RT-PCR and RT-PCR. Quantitative RT-PCR was performed with KAPATM SYBR® Fast qPCR Master Mix (KAPA Biosystems) on an ABI7500 cycler (Applied Biosystems). G3pdh was used as an internal control. Genomic PCR and RT-PCR were performed with Taq DNA polymerase (Life Technologies) on a Veriti® 96-well thermal cycler (Applied Biosystems). Primers used for analyses are shown in Table 1.
siRNA sequences against Oct3/4, mouse INSC, Par-3, LGN, and luciferase are shown in Table 1. siRNAs targeting mouse Elk1, c-Ets-1, and c-Rel were purchased from Qiagen. Cells were transfected with siRNA using Lipofectamine® 2000 (Life Technologies). Expression levels were analyzed by quantitative RT-PCR.
A series of genomic regions upstream of the mouse Insc locus TSS (−5,382, −4,382, −3,382, −2,382, −1,382, −382, and +1 to + 200) were amplified by PCR and cloned into pGL4.10 (Promega). ES cells were transfected with plasmids using Lipofectamine® 2000. Prior to the luciferase assay, cells were treated for 24 h with IKK inhibitor II, wedelolactone (Calbiochem). The luciferase assay was performed using the Dual-Luciferase® reporter assay system (Promega). Firefly luciferase activity was normalized to control Renilla luciferase activity.
Cells were incubated in cell dissociation buffer (Gibco) and then stained with allophycocyanin-conjugated Avas12 (anti-VEGFR-2) (Biolegend). Stained cells were analyzed using a FACSCantoIITM (BD Biosciences). Dead cells stained with propidium iodide (BD Biosciences) were excluded from analysis.
Cells were fixed with 4% paraformaldehyde at 37 °C for 10 min, followed by incubation with 0.5% Triton X-100 at room temperature for 10 min. Cells were blocked in 3% BSA (Sigma-Aldrich) at 37 °C for 30 min and then incubated with goat anti-Sox17 (R&D Systems), anti-VEGFR-2, anti-GFP (MBL International), anti-Oct4 (Cell Signaling Technology), and anti-Nanog (BD Bioscience) antibodies at 4 °C overnight. The cells were then washed and incubated for 1 h with appropriate secondary antibodies, including Alexa Fluor® 488-goat anti-rabbit IgG, 546-goat anti-rat IgG, or 647-donkey anti-goat IgG (A11008, A11081, and A21447, respectively; Life Technologies).
Lysates prepared from ES cells transfected with the pGL4.10 reporter plasmid encoding the −5,382 to +200 base genomic region of the mouse Insc TSS were cross-linked with 1% formaldehyde at room temperature for 10 min. Glycine was added at a final concentration of 0.125 m, followed by incubation for 10 min at room temperature. The sample was washed once with wash buffer (2% BSA and 0.05% NaN3 in PBS), the supernatant was removed, and then the pellet was lysed with SDS lysis buffer (50 mm Tris, 10 mm EDTA, 1% SDS, 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml leupeptin) on ice at 10 min. Samples were sonicated on ice, centrifuged at 15,000 rpm at 4 °C for 10 min, and then diluted with ChIP dilution buffer (50 mm Tris, 167 mm NaCl, 1.1% Triton X-100, 0.11% sodium deoxycholate, 1 mm phenylmethylsulfonyl fluoride, 1 μg/ml aprotinin, and 1 μg/ml leupeptin). Rabbit anti-Elk1, c-Ets-1, or cRel antibodies (4 μg) were added to the samples, followed by incubation at 4 °C for 1 h. Protein A-Sepharose/salmon sperm DNA mixtures (50% slurry; Millipore) were preincubated with 3% BSA-PBS at 4 °C for 2–3 h and then added to the samples, followed by incubation at 4 °C overnight. The immunoprecipitates were washed sequentially with wash buffer A (50 mm Tris, 150 mm NaCl, 1 mm EDTA, 1% Triton X-100, 0.1% SDS, and 0.1% sodium deoxycholate), wash buffer B (50 mm Tris, 500 mm NaCl, 1 mm EDTA, 1% Triton X-100, 0.1% SDS, and 0.1% sodium deoxycholate), and wash buffer C (10 mm Tris, 0.25 m LiCl, 1 mm EDTA, 0.5% Nonidet P-40, and 0.5% sodium deoxycholate). The beads were washed twice with Tris-EDTA buffer, and bound DNA was eluted with ChIP elution buffer (10 mm Tris, 300 mm NaCl, 5 mm EDTA, and 0.5% SDS) at 65 °C overnight. Eluted DNA was purified with phenol/chloroform and ethanol and then subjected to qPCR analysis. qPCR was performed with KAPATM SYBR® Fast qPCR Master Mix (KAPA Biosystems) on an ABI7500 Cycler (Applied Biosystems) using the following primers: GGGGTACCTTCCTAAGTGTAATTC and CTAGCTAGCGAGCTTACAGCATGG.
Previous reports have demonstrated that mES cells differentiate into bipotent mesendoderm cells capable of further differentiation into mesoderm and endoderm lineages under defined culture conditions (Fig. 1A) (36, 37). Accordingly, we cultured ES cells harboring a Gsc-gfp transgene (Gscgfp/+ ES cells) to monitor mesendoderm differentiation on collagen IV-coated dishes in SF-O3 serum-free medium containing 10 ng/ml activin A. Consistent with previous studies, after 4 days of differentiation, about 50% of the ES cells became positive for Gsc-GFP (Fig. 1B). Furthermore, they expressed either vascular endothelial growth factor receptor-2 (VEGFR-2), a mesoderm marker (39), or SRY-box 17 (Sox17), an endoderm marker (40, 41) (Fig. 1C). Next, we examined the expression profiles of differentiation marker genes every 24 h. Preceding expression of Gsc, eomesodermin (Eomes), a mesendoderm marker (36, 37, 42, 43), was expressed at differentiation day 3 (Fig. 1G). Expression of forkhead box protein A2 (FoxA2) and T-brachyury (T-bra), mesendoderm genes downstream of Gsc and/or Eomes (42, 44, 45), and Sox17 and VEGFR-2, respective markers of definitive endoderm and mesoderm (39,–41), was increased drastically at differentiation day 4 (Fig. 1G). The expression of pluripotency gene Nanog (46, 47) decreased rapidly at differentiation day 2, whereas another pluripotency gene, POU domain class 5 transcription factor 1 (Oct3/4) (46, 47), decreased gradually until day 4 (Fig. 1G), which was consistent with previous reports indicating sustained expression of Oct3/4 during mesoderm/endoderm differentiation (47). The expression profiles of each protein were confirmed by immunostaining of Gsc-gfp, VEGFR-2, Sox17, Nanog, and Oct4 or immunoblotting of Eomes, Foxa2, and T-bra (Fig. 1, C–F and H). Under this condition, we next examined the expression profiles of Par-3, LGN, and mouse Insc. In ES cells, we detected 3 variants of Par-3 protein by immunoblotting, all of which were down-regulated by RNA interference (RNAi)-mediated knockdown of Par-3 (see Fig. 5H, right). During ES cell differentiation, the expression profiles of these Par-3 proteins varied (Fig. 1J), whereas expression of Par-3 mRNA, which was detected using primers in the sequence conserved among the Par-3 transcription variants, increased gradually over time (Fig. 1I). Expression levels of LGN mRNA and protein were slightly decreased at differentiation day 2 and then increased again to the levels in undifferentiated ES cells (Fig. 1, I and J). Interestingly, mouse Insc mRNA expression was rapidly up-regulated at differentiation day 2 before returning to basal levels at day 3 (Fig. 1I), followed by an increase in the mouse INSC protein level at days 2 and 3, and then returning to basal levels at day 4 (Fig. 1J). We detected up-regulation of mouse Insc mRNA expression even at differentiation day 1 (data not shown). These results prompted us to investigate the transcriptional regulation of the mouse Insc gene during ES cell differentiation into mesoderm and endoderm.
We attempted to identify DNA regulatory elements capable of promoting mouse Insc gene transcription. A luciferase assay vector was constructed, in which a region located −5,382 bases upstream of the mouse Insc gene locus was fused to the luciferase-coding sequence. We successfully detected transcription-promoting activity in ES cells (Fig. 2A). It is worth noting that this activity was significantly higher after 2 days of differentiation compared with that in undifferentiated ES cells. Furthermore, the transcriptional activity was decreased at differentiation day 3 (Fig. 2C, top, a), reflecting endogenous expression patterns of the mouse Insc gene (see Fig. 1I). PCR analysis confirmed the presence of each transfected plasmid, even at day 3, excluding the possibility that the decline in luciferase activity at day 3 was due to the loss of the transfected plasmids (Fig. 2C, bottom, a). Similar results were obtained when a region at −6,382 bases upstream of the mouse Insc locus was fused to the luciferase-coding sequence (data not shown). Deletion from −631 to −1 bases upstream slightly decreased the transcription-promoting activity (Fig. 2A). Interestingly, deletion from −4046 to −1 bases upstream resulted in even stronger transcription-promoting activity (Fig. 2A). In addition, a series of 5′-deletion constructs indicated that deletion of bases −5382 to −4382 diminished almost all transcriptional activity (Fig. 2A). These data indicate that transcription-promoting elements are located between bases −5382 and −4046 upstream of the mouse Insc gene locus. Furthermore, a series of additional deletion constructs revealed a minimum transcription-promoting element between −5066 and −4712 bases upstream (Fig. 2B). Promoter activity of this minimum element increased significantly at day 2 but decreased on differentiation day 3 without loss of the transfected plasmids (Fig. 2C, top and bottom, b). It is worth noting that the region between −5382 and −4712 bases upstream of the mouse Insc gene locus displayed higher transcription-promoting activity than the minimum element (Fig. 2B), suggesting the existence of co-regulatory elements that enhance the transcription-promoting activity of the minimum element within the region between −5382 and −5066 bases upstream.
Within the minimum transcription-promoting element (−5066 to −4712), we used the Match program of TRANSFAC® version 6.0 to identify putative binding sites for five transcription factors, including POU domain class 2 transcription factor 1 (Oct1), ELK1 member of the ETS oncogene family (Elk1), E26 avian leukemia oncogene 1, 5′ domain (c-Ets-1), c-Rel, and E74-like factor 1 (Elf-1) (Fig. 3A, left). Next, we inserted a mutation into each transcription factor-binding site and measured the promoter activity. Disruption of Oct1-, Elk1/c-Ets-1-, or c-Rel-biding sites, but not the Elf-1-binding site, resulted in loss of promoter activity (Fig. 3A, right). This result suggests the possibility that mouse Insc gene expression is up-regulated by Oct1, Elk1, c-Ets-1, or c-Rel. Expression levels of Oct1, Elk1, and c-Rel were decreased slightly at differentiation day 2 and then increased again to the levels in undifferentiated ES cells at day 4, whereas the expression level of c-Ets-1 increased gradually over time (Fig. 3B). Thus, we next examined knockdown of each transcription factor by RNAi. Although we were unable to successfully down-regulate expression of Oct1, expression levels of Elk-1, c-Ets-1, and c-Rel were decreased significantly by siRNA in ES cells (Fig. 3C). siRNA-mediated knockdown of each protein was confirmed by immunoblotting (Fig. 3D). We found that siRNA-mediated knockdown of c-Rel, but not Elk1 or c-Ets-1, significantly decreased promoter activity of both the minimum DNA regulatory element (Fig. 3E) and the base −5382 upstream region of the mouse Insc locus (Fig. 3F). Disruption of the c-Rel-binding site in the base −5382 upstream region of the mouse Insc locus significantly decreased its promoter activity (Fig. 3G). Moreover, ChIP-qPCR analysis showed binding of endogenous c-Rel protein to the minimum DNA regulatory element (Fig. 3H, left). We also detected binding of Elk1, but not c-Ets1, to the minimum DNA regulatory element (Fig. 3H, right). c-Rel belongs to the NF-κB family of transcription factors. Our data showed that treatment of the cells with IKK inhibitor II, wedelolactone, to suppress NF-κB-mediated gene transcription (48) significantly decreased promoter activity of the minimum DNA regulatory element in a dose-dependent manner (Fig. 3I) with only limited effects on cell survival and proliferation (data not shown), suggesting that NF-κB signaling is involved in the mouse Insc expression. Furthermore, siRNA-mediated knockdown of c-Rel suppressed expression levels of endogenous mouse Insc mRNA and protein (Fig. 3, J and D). These results demonstrate that mouse Insc transcription is regulated by c-Rel through binding to the 5-kb upstream region of the mouse Insc locus during mES cell differentiation into mesoderm and endoderm.
Although our used Match program of TRANSFAC® version 6.0 did not identify Oct4 as a candidate transcription factor that binds to the minimum transcription-promoting element (−5066 to −4712), the consensus binding sequence of Oct4 is the same as that of Oct1 (49). Oct4 protein is enriched in ES cells, raising the possibility that Oct4 regulates mouse Insc expression. Then we depleted Oct4 by siRNA to assess its contribution to mouse Insc expression during differentiation. Down-regulation of Oct4 mRNA and protein by the used siRNA was confirmed by qPCR, immunoblotting, and immunostaining (Fig. 3K). Interestingly, Oct4 knockdown significantly decreased promoter activity of both the minimum DNA regulatory element and the base −5382 upstream region of the mouse Insc locus (Fig. 3L). In addition, the expression level of endogenous mouse Insc mRNA was decreased in the Oct4-depleted cells (Fig. 3M). These results support the idea that, in addition to c-Rel, Oct4 regulates mouse Insc expression during ES cell differentiation into mesoderm and endoderm. However, because Oct4 is necessary for ES cell pluripotency (50), depletion of Oct4 could alter cell properties of ES cells. Consistently, we observed decreased expression of pluripotent gene Nanog mRNA and protein in the Oct4-depleted cells at differentiation day 0 (Fig. 3, K and M). Therefore, we cannot exclude the possibility that the down-regulation of the promoter activity and the expression level of mouse Insc in the Oct4-depleted cells are indirect effects of the changes in cell property.
Next, we attempted to investigate whether mouse INSC plays a role in cell fate determination of ES cells into mesendoderm, mesoderm, or endoderm. To examine this possibility, we knocked down mouse INSC protein by RNAi using two independent siRNAs. These siRNAs effectively decreased the protein levels of ectopically expressed FLAG-tagged mouse INSC in HEK293T cells (Fig. 4A) as well as the levels of endogenous mouse Insc mRNA and protein in ES cells differentiating into mesendoderm (Fig. 4, B and C). Knockdown of mouse Insc significantly suppressed expression of VEGFR-2, whereas expression levels of Gsc, Eomes, FoxA2, T-bra, and Sox17 were similar to those in control luciferase siRNA-transfected cells at differentiation day 6 (Fig. 4D). Additionally, flow cytometric analysis indicated a decrease in the population of VEGFR-2-positive cells at differentiation day 6, when the cells were depleted of mouse INSC, whereas the population of Gsc-GFP-positive cells remained constant (Fig. 4E). These results indicate that mouse INSC regulates mesoderm differentiation of ES cells.
Similar to mouse INSC, we found that siRNA-mediated knockdown of c-Rel significantly suppressed the expression of VEGFR-2, whereas the expression level of Sox17 was only increased slightly (Fig. 5A), indicating the requirement of c-Rel in mesoderm differentiation. Our data showed that expression of mouse Insc was dependent on c-Rel (Fig. 3, D and I). Then we examined whether c-Rel regulates mesoderm differentiation of ES cells through mouse Insc. To address this issue, we established ES cell lines with constituently expressed mouse INSC-mCherry or control mCherry driven by the EF1α promoter (mouse INSC-mCherry and mCherry Tg ES cells, respectively) (Fig. 5, B and C). Notably, mouse INSC-mCherry Tg ES cells displayed a decrease in the expression of VEGFR-2 compared with mCherry Tg ES cells at differentiation day 6, whereas expression of Sox17 was comparable (Fig. 5D). Similar to knockdown of mouse INSC, these results indicated that overexpression of mouse INSC suppresses mesoderm differentiation of ES cells. As expected, siRNA-mediated knockdown of c-Rel in mCherry Tg ES cells significantly suppressed expression of VEGFR-2 at day 6 with no detectable change in the expression level of Sox17 in mCherry Tg ES cells (Fig. 5D). However, knockdown of c-Rel did not suppress expression of VEGFR-2 in mouse INSC-mCherry Tg ES cells (Fig. 5D). Instead, the expression level of VEGFR-2 was significantly higher in mouse INSC-mCherry Tg ES cells than in mCherry Tg ES cells when the cells were depleted of c-Rel (Fig. 5D), indicating that overexpression of mouse INSC rescued the mesoderm-suppressed phenotype induced by knockdown of c-Rel. Taken together, our data support a model in which c-Rel regulates mesoderm differentiation of ES cells by promoting the expression of mouse INSC.
Insc is a well known regulator of asymmetric cell division, raising the possibility that asymmetric cell division occurs during mES cell differentiation into mesoderm and endoderm, in which mouse INSC might play a vital role. However, in differentiating mES cells, our immunohistochemical analysis showed mCherry-mouse INSC and endogenous mouse INSC protein in the cytoplasm and no evidence of asymmetric staining patterns (Fig. 5, E and F), although we detected polarized apical localization of Par-3 in interphase cells as well as polarized cortical localization of LGN in mitotic cells (Fig. 5, E and F). We found an association of mouse INSC with LGN, but not Par-3, in immunoprecipitation analyses (Fig. 5G). Furthermore, differentiation of ES cells into mesoderm and endoderm was only partially affected by siRNA-mediated knockdown of LGN or Par-3 (Fig. 5, H and I). These observations favor a model in which mouse INSC regulates mesoderm differentiation of ES cells by a mechanism that is independent of asymmetric cell division.
The regulatory mechanisms underlying mouse Insc expression have been largely unknown despite its critical role in cell fate determination of various cell types in mammals. Here, we identified a minimum promoter element that drives mouse Insc expression in differentiating mES cells. In general, TATA elements (located ~25–30 bases upstream of the TSS) and an initiator motif (a pyrimidine-rich sequence encompassing the TSS) are important for eukaryotic gene expression (51,–53). Inconsistently, however, our data showed that deletion of the region between bases −631 and −1 upstream of the mouse Insc TSS only partially decreased mouse Insc expression. In addition, no promoter activity was detected in a series of constructs in the region between −4,382 and −382. Therefore, the 4,382-base upstream region of the TSS is expendable for mouse Insc expression. Because deletion of the region between −4,046 and −1 drastically increased promoter activity in both undifferentiated and differentiated cells, we speculate that this region harbors suppressor or structural elements that suppress mouse Insc expression.
We found a minimum promoter element within a region between −5,066 and −4,712 bases upstream of the mouse Insc gene locus and identified a key transcription factor, c-Rel, which binds to the minimum promoter and drives mouse Insc expression in ES cells. Involvement of nuclear factor-κB (NF-κB) signaling in mES cell differentiation (54, 55) is consistent with these results. The relationship between c-Rel and mouse INSC is supported by previous reports that revealed functions of c-Rel in cell proliferation and differentiation of neural progenitors and epidermal cells (56, 57), whereby mouse INSC regulates their cell fate (28,–31). This observation suggests that c-Rel regulates cell fate determination by promoting mouse Insc expression. Our finding that the reduction of mesodermal cells induced by loss of c-Rel is rescued by mouse INSC overexpression further supports this model. However, it is unlikely that c-Rel is the sole regulator for mouse Insc expression. Our data showing that the region between −5,382 and −4,712 bases upstream has higher promoter activity than the minimum promoter implicates a co-regulatory element that drives mouse Insc expression within the region between −5,382 and −5,066, although this region alone is not sufficient for promoter activity. While we were preparing this manuscript, Ballard et al. (58) reported that SNAL1 enhances mouse Insc expression in mammary stem cells. They used a 224-bp DNA fragment containing the upstream region of human Inscuteable for luciferase assays. This sequence is not conserved in mice (data not shown). However, we found a SNAIL1 consensus-binding sequence, TCACA, within the co-regulatory region at 5,195–5,191 bases upstream of the mouse Insc gene locus. Therefore, SNAIL1 might function as a co-regulator that enhances c-Rel-mediated mouse Insc transcription in mES cells. In addition, our used Match program of TRANSFAC® version 6.0 also identified the Evi1 consensus-binding sequence, AGAGA, at bases 5,235 to 5,231 upstream of the mouse Insc gene locus. Moreover, our finding that mutation of the binding sites for Oct1, Elk1, or c-Ets-1 results in loss of promoter activity further supports the concept of a co-regulatory mechanism for mouse Insc transcription. It appears likely that these co-regulators function in a complementary manner, because knockdown of Elk1 alone did not suppress mouse Insc expression, despite its association with the minimum promoter. We speculate that the regulatory mechanism for mouse Insc transcription varies, depending on the cell type or differentiation conditions. The underlying mechanisms should be elucidated in future studies.
The function of Insc in asymmetric cell division is well established, and its overexpression alters the balance between symmetric/asymmetric cell division. Although asymmetric cell division during ES cell differentiation has received little attention, recent studies have demonstrated that Wnt protein-immobilized beads attached to the cell surface induce asymmetric division of mES cells with mitotic spindles aligned along a Wnt bead-induced polarity axis (59). Additionally, in defined culture conditions, mES cells have been differentiated into bipotent mesendoderm cells capable of giving rise to both endoderm and mesoderm lineages (36, 37). These data raise the possibility of Insc-dependent asymmetric cell division occurring during mES cell differentiation into mesoderm and endoderm. However, our results provided no evidence for asymmetric staining patterns of mouse INSC-mCherry protein or involvement of asymmetric cell division in the mesoderm cell fate decision of mES cells. These observations are consistent with a previous study showing that Staufen, an Insc-binding protein that functions in mRNA processing and basal localization of Pros mRNA in Drosophila neuroblasts, regulates mES cell differentiation without displaying an asymmetric distribution (60). Thus, one potential model is that mouse INSC regulates mesoderm cell fate through Staufen. However, we cannot exclude the possibility that asymmetric cell division occurs only in a very small population of differentiating ES cells. Further studies are necessary to elucidate how mouse INSC specifies the mesoderm cell lineage of differentiating ES cells.
c-Rel, a member of the NF-κB transcription factor family (61,–63), regulates gene expression of various cytokines and transcription factors important for proper functioning of B cells, T cells, macrophages, and dendritic cells (62,–65). Importantly, c-Rel also regulates development of T regulatory cells, Th1 cells, and Th17 cells (64,–67). Because our results show that mouse Insc gene expression is regulated by c-Rel, it would be interesting to investigate a potential c-Rel/mouse INSC axis capable of directing T cell development.
R. I., H. S., and F. T. designed the research; R. I., and S. Kozuki, and S. Kamakura performed the experiments; R. I. and F. T. analyzed data; and R. I and F. T. wrote the paper.
We thank T. Era (Kumamoto University) for Gscgfp/+ ES cells.
*This work was supported by grants from the Funding Program for Next Generation World-Leading Researchers (Grant LS069 to F. T.), the Naito Foundation (to F. T.), and the Platform Project for Dynamic Approaches to Living System from the Japan Agency for Medical Research and Development (to A. M. E. D.). The authors declare that they have no conflicts of interest with the contents of this article.
2The abbreviations used are: