Introduction
Tuberous sclerosis complex (TSC) is a genetic disorder characterized by multisystem involvement and wide phenotypic variability. This condition results in the development of non-cancerous tumors in various organs and most frequently affects the brain, skin, kidney, lung, heart, and retina. TSC manifestations in the central nervous system include cortical tubers, subependymal nodules, subependymal giant cell astrocytomas, and scattered abnormal cells throughout the brain (1). A majority of patients with TSC reveal neurological and/or psychiatric symptoms, including epilepsy, intellectual disability, autism spectrum disorder (ASD), attention deficit, depression, and anxiety disorder, which range from mild to severe and may impair their ability to live an independent life.
Mutation of either TSC1 or TSC2 causes TSC (2,3). Protein products of TSC1 (hamartin) and TSC2 (tuberin) form a complex that inhibits the Ras homologue enriched in the brain (Rheb), a small G protein that activates mammalian target of rapamycin complex 1 (mTORC1). Defects of TSC1 or TSC2 cause excessive mTORC1 activation, which in turn provokes abnormal regulation of important cellular processes such as cellular growth and proliferation (4,5). The Knudson’s ‘two-hit’ model (6) has been the working molecular model for tumor development in TSC for several years. In fact, loss of heterozygosity (LOH) of TSC1 or TSC2 has been demonstrated in renal angiomyolipomas (7–9) and in subependymal giant cell astrocytomas (10). However, evidence for LOH in TSC cortical tubers is limited (11). On the other hand, haploinsufficiency of these genes is also speculated to be involved in TSC pathogenesis. To reveal Tsc mutation-related mechanisms of the pathogenesis, rodents harboring a defect of the Tsc1 or Tsc2 gene have been extensively investigated (12–15). For instance, Tsc1+/− and Tsc2+/− mouse models exhibit learning and memory deficits (16,17). Eker rats are heterozygous for a mutation of Tsc2 and develop hereditary kidney cancer by the age of 1 year (18–20). Although kidney cancer is rare in human patients with TSC, it is the only cancer known to occur at an increased incidence in TSC. The embryonic lethality of Tsc2−/− Eker rat embryos is characterized by disrupted neuroepithelial growth (21). Although cortical tubers are rare (22), 63% of Eker rats develop brain lesions comprising a mixture of large and elongated cells in both subependymal and subcortical regions (23,24). In contrast, among Tsc1 and Tsc2 knock-out mouse models, only conditional ablation in the brain can induce such lesions (25). Consequently, with regard to brain lesions, the Eker rat model is more similar to the human patients compared with other mouse models.
We observed that the tumorigenicity of Tsc2−/− cells derived from mice was effectively inhibited by rapamycin treatment (26). Other groups reported a similar effect when Eker rats or knock-out mice were treated with rapamycin, although some residual tumors were detected (27,28). These findings have provided the rationale for therapy with rapalogues to treat TSC lesions such as lymphangioleiomyomatosis, SEGAs, and angiomyolipomas, directed at the abnormal activation of mTORC1 (29–31). Although decreased tumor volume has been documented, complete cure was not achieved in most cases. In addition, there are several problems associated with long-term use of rapalogues, including various undesirable side-effects. Consequently additional therapeutic molecular targets are required. The pathogenesis of TSC is assumed to be related to abnormal differentiation as a result of TSC1/2 deficiency. For instance, abnormal giant cells that appear in brain lesions of patients with TSC express both neuronal and glial lineage markers (32). In recent years, a number of articles have revealed differentiation- and cell type-specific abnormalities using in vitro differentiation protocols to investigate differentiation of embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs). To evaluate roles of Tsc2 from the viewpoint of differentiation and tissue-specific pathogenesis as well as to compare and combine in vivo and in vitro models, we established ESCs from Eker rats.
In 2008, authentic rat ESCs were established for the first time (33,34), lagging behind the establishment of mouse (35,36) and human ESCs (37). Using methods described by Buehr et al (33), we generated ESCs from blastocysts of Eker rats to establish an in vivo experimental system to explore the role of Tsc2 in TSC pathogenesis. Although several reports have indicated the necessity of Tsc1 and Tsc2 regulation to maintain ESCs (38) and somatic stem cells (39) or to establish iPS cells (40), we were able to establish not only Tsc2+/+ and Tsc2+/− ESCs but also Tsc2−/− ESCs. To our knowledge, this is the first report describing the generation of Tsc2-deficient ESCs.
Materials and methods
Ethics statement
All animal experiments were conducted in strict accordance with the institutional guidelines of Juntendo University for animal experiments. The protocol was approved by the Animal Experimentation Committee of Juntendo University (Tokyo, Japan) (approval no. 250105). All surgical procedures were performed under isoflurane anesthesia, and all efforts were made to minimize animal suffering.
Animals
Genetic homogeneity of Eker rats was maintained in our laboratory by brother-sister mating. Wistar rats, Brown Norway rats, C57BL/6J mice, and nude mice were purchased from Charles River Laboratories Japan, Inc. (Kanagawa, Japan). All animals were housed under specific pathogen-free conditions.
Mouse embryonic fibroblasts (MEFs)
MEFs were derived from embryonic day 14.5 C57BL6/J mouse embryos. MEFs were cultured in Knockout DMEM supplemented with 10% fetal bovine serum, 1% L-glutamine, and penicillin streptomycin (all from Gibco Life technologies, Carlsbad, CA, USA) on gelatin-coated dishes. MEFs were treated with mitomycin C (Sigma-Aldrich, St. Louis, MO, USA) for use as feeder cells.
Culture of ESCs
We generated ESCs from Eker rats according to the method reported by Buehr et al (33). After double heterozygous mating of Eker rats, E4.5 blastocysts were gently flushed out from uteri using the N2B27 medium (StemCells, Inc., Newark, CA, USA). After removal of zonae pellucidae with acid Tyrode’s solution, whole blastocysts were plated and cultured on mitomycin C-treated MEFs in N2B27 medium supplemented with 3 μM of CHIR99021, 1 μM of PD0325901 (both from Axon Medchem BV, Groningen, The Netherlands), 1,000 U/ml rat leukemia inhibitory factor (LIF) (ESGRO®; Millipore, Bedford, MA, USA) [two inhibitors (2i) + LIF condition]. After 5–7 days, blastocyst outgrowths were cut into pieces and replated in the same 2i + LIF medium. Thereafter, emerging ESC colonies were dissociated using Accutase (Innovative Cell Technologies, Inc., San Diego, CA, USA) and passaged every 2–4 days.
Alkaline phosphatase staining
Alkaline phosphatase staining was performed with an alkaline phosphatase kit (85L3R; Sigma-Aldrich) according to the manufacturer’s instructions.
Chromosomal analysis
A standard chromosome preparation method using colchicine treatment was employed. Chromosome preparations were analyzed after Giemsa staining. At least 30 metaphase chromosome sets were analyzed for each line.
Genotyping polymerase chain reaction (PCR)
Genotyping of ESCs was conducted using PCR on ESC DNA. To discriminate Tsc2 mutant or wild-type alleles, the following primers were used: 5MFJ (5′-ACC ATC AGG ATG CTG CTG AA-3′), 3MFJ2 (5′-CTA TGG CCA CAT GTG ACC AA-3′), and TSR27 (5′-GCG CCA GAT TCA CCT CAT TA-3′) (41). PCR was used to identify the gender of ESCs by amplification of the rat Y chromosome-specific Sry gene using the primer pair Sry-F (5′-CAT CGA AGG GTT AAA GTG CCA-3′) and Sry-R (5′-ATA GTG TGT AGG TTG TTG TCC-3′) (33).
Reverse transcription (RT)-PCR
Total RNA was obtained using a NucleoSpin® RNA II kit (Macherey-Nagel GmbH & Co. KG, Düren, Germany) according to the manufacturer’s instructions. Complementary DNA was synthesized using a SuperScript III First-Strand Synthesis SuperMix kit (Invitrogen Life Technologies, Carlsbad, CA, USA) and an oligo-dT primer, according to the manufacturer’s instructions. PCR was performed in a thermal cycler (Hybaid MBS 0.2G Thermal Cycler; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The following primer pairs were used: Oct4-F (5′-GGG ATG GCA TAC TGT GGA C-3′), Oct4-R (5′-CTT CCT CCA CCC ACT TCT C-3′), Sox2-F (5′-GGC GGC AAC CAG AAG AAC AG-3′), Sox2-R (5′-GTT GCT CCA GCC GTT CAT GTG-3′), rat Nanog-F (5′-GCC CTG AGA AGA AAG AAG AG-3′), rat Nanog-R (5′-CGT ACT GCC CCA TAC TGG AA-3′) (33), rat nestin-F (5′-AGC CAT TGT GGT CTA CTG A-3′), rat nestin-R (5′-TGC AAC TCT GCC TTA TCC-3′), Sox17-F (5′-AGG AGA GGT GGT GGC GAG TAG-3′), and Sox17-R (5′-GTT GGG ATG GTC CTG CAT GTG-3′) (34).
Western blotting
Cells were harvested and lysed in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, and 10% glycerol). Proteins were separated by SDS-PAGE and transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore). The membrane was blocked with 1% skimmed milk in Tris-buffered saline containing 0.05% Tween-20 and probed with appropriate antibodies using the EnVision System (DakoCytomation, Glostrup, Denmark). Antibody signals were developed using ECL reagents and Hyperfilm ECL film (both from GE Healthcare, Little Chalfont, UK), which were then scanned using CEPROS SV (Fujifilm, Tokyo, Japan). The following primary antibodies were used: anti-Tsc2 antibody (C20; 1:200; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), anti-Tsc1 primary antibody (c-Tsc1, 1:500), anti-phospho-S6 ribosomal protein (Ser235/236) rabbit polyclonal antibody (1:1,000, no. 2211), anti-S6 ribosomal protein rabbit monoclonal antibody (1:1,000, no. 2217) (both from Cell Signaling Technology, Inc., Danvers, MA, USA), and anti-β-actin mouse monoclonal antibody (1:1,000; Sigma-Aldrich).
Embryoid body (EB) formation
ESCs were plated into low-adhesion 96-well dishes (MS-9096; Sumitomo Bakelite Co., Ltd., Tokyo, Japan). After 10 days of suspension culture, EBs were plated onto Matrigel-coated dishes in GMEM/10% fetal bovine serum medium (both from Gibco Life Technologies).
Immunocytochemistry
Cells were fixed and permeabilized with 4% paraformaldehyde and 0.25% Triton X-100 (both from Wako Pure Chemical Industries, Ltd., Osaka, Japan) in PBS for 30 min at 4°C and then washed (3×5 min) with PBS/0.1% bovine serum albumin (BSA) (Iwai Kagaku Co., Tokyo, Japan). Cells were incubated with a primary antibody in PBS with 1% BSA for 1 h at room temperature. Thereafter, cells were washed and incubated with fluorophore-conjugated secondary antibodies and 4′,6-diamidino-2-phenylindole (DAPI) for 1 h at room temperature. Immunofluorescent images were captured using a Leica TCS SP5 v2.0 system (Leica, Heidelberg, Germany). The following primary antibodies were used: anti-Oct3/4 mouse monoclonal antibody (1:50, C-10; Santa Cruz Biotechnology, Inc.), anti-Sox2 rabbit polyclonal antibody (1:100, poly6309; BioLegend, San Diego, CA, USA), anti-β-III tubulin mouse monoclonal antibody (1:500, Tuj-1; Covance Laboratories, Princeton, NJ, USA), anti-myosin heavy chain mouse monoclonal antibody (1:50, MF20; R&D Systems, Minneapolis, MN, USA), and anti-Gata4 mouse monoclonal antibody (1:50; Santa Cruz Biotechnology, Inc.). Alexa Fluor (488 or 568)-conjugated goat anti-mouse or goat anti-rabbit secondary antibodies (Invitrogen Life Technologies) were used at 1:1,000 dilutions.
Teratoma formation
Approximately 5×105 cells were injected under kidney capsules of nude mice. Tumors were dissected after 4–5 weeks and fixed in 10% buffered formalin. Tumor tissues were embedded in paraffin wax, sectioned, and examined after hematoxylin and eosin staining.
Blastocyst injection and generation of chimeric rats
Because collection of many blastocysts from Brown Norway rats is inefficient, we attempted the chimera formation assay using Brown Norway as well as Wistar rats. Rat blastocysts at E4.5 days were collected on the day of injection and cultured for 2–3 h to ensure cavitation. ESCs were disaggregated using Accutase, and 10–12 cells were injected into blastocyst cavities. Injected embryos were transferred into uteri of pseudopregnant rats.
Dead embryos were collected from uteri by cesarean section. For Tsc2 genotyping PCR, genomic DNA was obtained from several parts of each embryo or pup.
Gene expression microarray analysis
The Rat Affymetrix GeneChip Gene 1.0 ST Array (Affymetrix, Inc., Santa Clara, CA, USA) was used for microarray analysis. Amplification and labeling of probes and hybridization were performed according to the manufacturer’s instructions. Hierarchical clustering analysis was performed using GeneSpring software version 12.1 (Agilent Technologies, Inc., Santa Clara, CA, USA).
Results
Establishment of Tsc2-deficient stem cells from Eker rat embryos
After mating of double heterozygous Eker rats, a total of 34 blastocysts were collected. Zonae pellucidae were removed, and most blastocysts were successfully cultured on feeder cells, revealing outgrowths from embryonic fibroblasts (MEFs) in N2B27 medium supplemented with 2i + LIF (33). After several passages, a total of 26 cell lines were established. We routinely passaged these cells every 2–4 days by dissociating them into single cells and replating onto new feeder cells. They grew as dome-shaped or spherical colonies and were maintained for >25 passages without losing their morphology (Fig. 1A). A majority of colonies expressed alkaline phosphatase, an indicator of stem cell character (Fig. 1A). Thereafter, we checked Tsc2 genotypes of established cell lines by PCR. Surprisingly, we identified that not only Tsc2+/+ and Tsc2+/− cell lines but also Tsc2−/− cell lines had been established (Fig. 1B). Considering that previous reports had indicated that Tsc2 is necessary for the maintenance of stem cell characteristics, this result was unexpected. Both male and female cell lines were established for each genotype (Fig. 1C). Chromosome analysis revealed that most cell lines had normal ploidy (n=42, data not shown). RT-PCR analysis revealed that Tsc2−/− cells expressed the pluripotency markers Oct4, Sox2, and Nanog (Fig. 2A). Oct4 and Sox2 expressions were confirmed by immunofluorescence microscopy (Fig. 2B). These results indicate that Tsc2-deficient stem cells could be established from Eker rat embryos. We performed further experiments using two independent cell lines of each genotype.
Activation of the mTORC1 pathway in Tsc2<sup>−/−</sup> ESCs
To evaluate the mTORC1 activation status, we analyzed ESCs by western blotting. Tsc2 protein was detected in Tsc2+/+ and Tsc2+/− cells but not in Tsc2−/− cells, thereby confirming results of the genotype analysis. Tsc1 protein levels were decreased in Tsc2−/− ESCs, thereby reflecting the reciprocal stabilization between Tsc1 and Tsc2 proteins (42). As expected, an increase in S6 phosphorylation was detected in Tsc2−/− cells compared with that in Tsc2+/+ and Tsc2+/− cells, which indicates abnormal activation of the mTORC1 pathway in Tsc2−/− ESCs (Fig. 3). These results indicate that despite abnormal activation of the mTORC1 pathway, Tsc2−/− ESCs can be established.
In vitro differentiation of Tsc2<sup>−/−</sup> ESCs into three germ layers
Using the EB formation assay, we evaluated the differentiation potential of the established cell lines. We assessed the expression of differentiation markers by RT-PCR. Expression of markers for ectoderm (Nestin), endoderm (Sox17), and mesoderm (Flk1) were all observed in EBs (Fig. 4A). We plated EBs onto Matrigel-coated dishes and assessed their differentiation status by immunofluorescent staining for β-III tubulin (neuroectoderm), myosin (mesoderm), and Gata4 (endoderm) (Fig. 4B). Not only Tsc2+/+ and Tsc2+/− cells but also Tsc2−/− cells demonstrated the potential to differentiate into all three germ layers. In addition, we observed spontaneously beating areas in EBs of all Tsc2 genotypes (data not shown). These results suggest that most differentiation processes of ESCs were not blocked by Tsc2 deficiency.
Differentiation of Tsc2<sup>−/−</sup> ESCs into multiple lineages in teratomas
When Tsc2−/− ESCs were transplanted under the kidney capsule of nude mice, they differentiated into tissues derived from all three germ layers, including gut-like epithelium (endoderm), cartilage and adipocytes (mesoderm), stratified squamous epithelium, and neuroepithelium (ectoderm) (Fig. 5). These results indicate that Tsc2−/− ESCs are multipotent, although detailed characterization of each of the differentiated tissues remains to be elucidated. Interestingly, we observed that abnormal ductal structures appeared in Tsc2−/− teratomas (Kawano H, et al, unpublished data). Further characterization of these abnormal structures is described in another report (Kawano H, et al, unpublished data).
Contribution of Tsc2<sup>+/+</sup> ESCs in chimeras
Next, to determine the ability of established ESCs to form chimeras, we injected Tsc2+/+ and Tsc2−/− ESCs into blastocysts of Wistar rats or Brown Norway rats (Materials and methods). Although ratios were low, four chimeras with black coat color were born from Wistar blastocysts injected with Tsc2+/+ ESCs, indicating the contribution of ESCs from the Eker rat strain (Fig. 6A). In contrast, we were unable to obtain pups demonstrating chimeric coat color in repeated trials using two independent Tsc2−/− ESCs. However, in these trials, we detected dead embryos in the uterus of recipient mother rats at term (Fig. 6B1 and 2). The appearance of dead embryos suggested developmental retardation. In addition, one live pup was delivered by cesarean section but died shortly after birth (Fig. 6B3). This pup revealed various morphological abnormalities such as an enlarged trunk. PCR genotyping of dead embryos and the pup indicated the contribution of Tsc2−/− ESCs in their tissues (Fig. 6C). On the basis of the band pattern of two dead embryos, we concluded that they had a greater contribution of Tsc2−/− ESCs compared with the live pup. These results suggest that a greater contribution of Tsc2−/− ESCs in the chimera results in embryonic lethality. Although germline transmission has not been confirmed yet, the contribution in chimeras suggests that ESCs established in this study possessed characteristics of multipotent stem cells.
Distinct gene expression pattern in Tsc2<sup>−/−</sup> ESCs on microarray analysis
To compare gene expression profiles of established ESCs, we employed microarray analysis (Fig. 7). Similar expression levels of pluripotency-related genes were identified in all these cells. Moreover, hierarchical clustering analysis revealed that gene expression profiles of Tsc2+/+ and Tsc2+/− ESCs resembled each other, but those of Tsc2−/− ESCs revealed an apparently distinct pattern. These results suggest that the homozygous Tsc2 mutation causes extensive gene expression changes in rat ESCs.
Discussion
In this study, we successfully established Tsc2−/− ESCs from Eker rats. These cells possessed characteristic features of ESCs, including expression of pluripotency markers, long-term self-renewal, and the capacity to differentiate into derivatives of all three germ layers. Although detailed mechanisms are still not clear, there have been several reports indicating the importance of the Tsc2-mTOR pathway in stem cell maintenance and differentiation (38,39,43). Gan et al reported that Tsc1 is a critical regulator of self-renewal, mobilization, and multilineage development in hematopoietic stem cells and that it executes these phenomena via both mTORC1-dependent and -independent pathways (39). Further, it was reported that the activation of S6K by expression of the constitutively active S6K1 or siRNA-mediated knockdown of TSC2 and RICTOR induced differentiation of human ESCs (38). Recently, Betschinger et al reported that siRNA-mediated knockdown of Tsc2 or Flcn inhibits differentiation of ESCs (43). In contrast to results in these reports, we successfully established Tsc2−/− ESCs possessing multipotent differentiation capacity despite the presence of the activated mTORC1 pathway. There are several possible reasons why the derivation of Tsc2−/− ESCs was possible in this study. Previous studies utilized siRNA- or shRNA-mediated knockdown of Tsc2 in already established ESCs or conditional knockout of Tsc1 in somatic stem cells. Such ‘acute’ downregulation of Tsc1/Tsc2 may cause some aberrant gene regulation that restrains the maintenance of the multipotent nature and differentiation capacity of stem cells. Microarray analysis revealed a distinct gene expression pattern in Tsc2−/− ESCs compared with their Tsc2+/+ and Tsc2+/− counterparts. Our system enables comparison of gene expression profiles between Tsc2−/− ESCs and Tsc2+/− ESCs with Tsc2 knockdown. Such analysis is of interest to further explore Tsc2 mutation-related pathogenesis.
We were unable to obtain pups demonstrating chimeric coat color using Tsc2−/− ESCs. Results of dead embryos suggested that higher contribution of Tsc2−/− cells in chimeras induced embryonic lethality. Further, it has been reported that when human TSC2-deficient fibroblast-like cells were grafted into mice, differentiated tissues revealed features of TSC skin tumors and that TSC2-deficient cells directly or indirectly induce abnormal follicular neogenesis and epidermal proliferation (44). Because Tsc2−/− ESCs may cause abnormal differentiation of hair in chimeras, it may not be appropriate to determine the contribution of ESCs on the basis of hair color of chimeras.
He et al reported that reprogramming of somatic cells derived from Tsc2−/− mouse embryos to iPSCs was not possible (40). In this study, we provide evidence that the derivation of Tsc2−/− ESCs from Eker rat embryos is possible. In somatic cells, some epigenetic abnormalities caused by Tsc2 deficiency may not be corrected even under reprogramming conditions. Conversely, during early embryogenesis, epigenetic abnormalities in Tsc2−/− cells may be tuned to maintain the stemness. With reprogramming experiments using Eker rat-derived embryonic fibroblasts, ESCs established in this study will serve as useful tools to compare effects of Tsc2 deficiency on epigenetic status in reprogramming and ESC derivation.
In recent years, various patient-derived iPSCs have been used for in vitro differentiation experiments to mimic the pathogenesis of human diseases (45,46). Moreover, such cellular models are useful to research novel drug target molecules by high-throughput screening (47). With regard to tumorigenesis, tissue specificity and abnormal differentiation are relevant to its molecular basis. Lineage-specific in vitro differentiation of tumor suppressor-deficient ESCs will provide valuable experimental models to explore the mechanism of pathogenesis. However, in humans, establishment of tumor suppressor-deficient (i.e., homozygously inactivated) ESCs or iPSCs has been technically difficult. In rodents, homozygous mutant ESCs for tumor suppressors, including Rb, Tp53, and Apc, have been established (48–50). To date, none of those ESCs have been extensively used for in vitro differentiation experiments. For example, Apc-deficient ESCs failed to differentiate into multiple lineages in the teratoma formation assay, suggesting that the induction of various cell types was not applicable to these ESCs (50). In contrast, Tsc2−/− ESCs exhibited the potential to differentiate into all germ layers and multiple cell lineages, both in vitro and in vivo. We already observed development of abnormal ductal structures in Tsc2−/− teratomas, suggesting that cell type-specific effects of Tsc2 deficiency could be reproduced in differentiation of ESCs (Kawano H, et al, unpublished data). Combined with in vivo experiments, in vitro differentiation models using ESCs established in this study will facilitate understanding of Tsc2 mutation-related pathogenesis as well as aid in the search for therapeutic target pathways.