Germ-line mutations of the human TSC2 tumor suppressor gene cause tuberous sclerosis (TSC), a disease characterized by the development of hamartomas in various organs. In the Eker rat, however, a germ-line Tsc2 mutation gives rise to renal cell carcinomas with a complete penetrance. The molecular mechanism for this phenotypic difference between man and rat is currently unknown, and the physiological function of the TSC2/Tsc2 product (tuberin) is not fully understood. To investigate these unsolved problems, we have generated a Tsc2 mutant mouse. Tsc2 heterozygous mutant (Tsc2+/−) mice developed renal carcinomas with a complete penetrance, as seen in the Eker rat, but not the angiomyolipomas characteristic of human TSC, confirming the existence of a species-specific mechanism of tumorigenesis caused by tuberin deficiency. Unexpectedly, ∼80% of Tsc2+/− mice also developed hepatic hemangiomas that are not observed in either TSC or the Eker rat. Tsc2 homozygous (Tsc2−/) mutants died around embryonic day 10.5, indicating an essential function for tuberin in mouse embryonic development. Some Tsc2−/− embryos exhibited an unclosed neural tube and/or thickened myocardium. The latter is associated with increased cell density that may be a reflection of loss of a growth-suppressive function of tuberin. The mouse strain described here should provide a valuable experimental model to analyze the function of tuberin and its association with tumorigenesis.

Human TSC,3 an autosomal dominant multiorgan disorder caused by a mutation in either the TSC1 (human chromosome 9q34) or TSC2 (human chromosome 16p13.3) gene, is characterized by phacomatosis with manifestations that include mental retardation and seizures (1, 2, 3). It has been reported that hamartomas that develop in TSC patients have LOH at one of two TSC loci (4), suggesting both to be tumor suppressor genes fitting Knudson’s two-hit hypothesis (5).

In the Eker rat, a germ-line retrotransposon insertion (intron 30) in a homologue of the human TSC gene 2 (Tsc2) causes RCs with a complete penetrance in all heterozygotes (6, 7, 8, 9). At the histological level, multiple bilateral RCs develop through multiple stages from early preneoplastic lesions (e.g., phenotypically altered tubules) to adenomas in virtually all heterozygotes by the age of 1 year (6). LOH and intragenic somatic mutations of wild-type Tsc2 are characteristics of Eker rat RCs, indicating that a second hit of the Tsc2 gene is a cause of RC development (10, 11, 12). We have generated transgenic Eker rats carrying wild-type Tsc2 as a transgene and demonstrated that phenotypes of the Eker rat are suppressed by extra copies of the wild-type Tsc2 gene (13). This demonstration finally confirmed that tumors in the Eker rat are caused by the germ-line mutation of Tsc2. Phenotypes of the Eker rat differ from those of human TSC, including the incidence of RCs (in man, angiomyolipomas are more common renal lesions), although subependymal and subcortical hamartomas resembling those in TSC were reported quite recently in the Eker rat (1, 14). Thus, a mutation of the same gene causes diverse phenotypes between species.

TSC2/Tsc2 encodes tuberin, which is highly conserved in vertebrates (15, 16). It contains Rap1-GAP homology region near its COOH terminus (3). Recently, Rap1-GAP activity (17) and Rab5-GAP activity along with rabaptin 5 binding (18) of tuberin have been demonstrated. Transcriptional activator activities of tuberin have also been reported (19, 20). Because TSC patients with alterations of either of the two predisposing genes (TSC1 and TSC2) exhibit identical symptoms, the products of these two genes (hamartin and tuberin) are thought to be involved in a common biological pathway (2). Indeed, direct interactions between tuberin and hamartin have been reported previously (21). However, the precise functions of tuberin and the molecular mechanisms of tumor development associated with Tsc2 mutations have not been fully elucidated.

In this study, we have generated a Tsc2 knockout mouse to characterize the function of tuberin in vivo and to elicit insights into tumorigenesis caused by Tsc2 gene inactivation. A comparison of the phenotypes of Tsc2 mutant mice with those of human TSC and the Eker rat will provide some insights into species-specific differences in tumorigenesis caused by tuberin deficiency. Moreover, in comparison with the rat system, one of the advantages of the mouse system is the availability of various genetically modified lines for genetic cross experiments (22). The pathway in which tuberin is involved might be elucidated through such genetic analysis.

Construction of a Targeting Vector.

Overlapping phage clones covering the mouse Tsc2 gene were isolated from a 129/Sv mouse genomic library using rat Tsc2 cDNA as a probe (15). A Tsc2-targeting vector, pKO-LT, was constructed to delete a ∼3-kb fragment of the mouse Tsc2 gene that ranged from a HindIII site in exon 2 to a PstI site in exon 5 (Fig. 1 A). An open reading frame of β-gal (lacZ) with a NLS was fused to the Tsc2 gene in frame at the 3′ end of a ∼7-kb 5′ homologous region. For positive selection, a PGK-neo cassette flanked by a pair of loxP sequences was introduced between NLS-lacZ and a 1.5-kb 3′ homologous region (23). For negative selection, a DTA-expressing cassette without a polyadenylic acid addition signal was fused to the 3′ end of the 3′ homologous region (24). Transcriptional orientations of PGK-neo and DTA cassettes were the opposite and the same, respectively, with respect to that of the Tsc2 gene.

Gene Targeting in ES Cells and Generation of Tsc2 Mutant Mice.

J1 ES cells were transfected with pKO-LT, linearized at a NotI site by electroporation, and selected for resistance to G418 (175 μg/ml) as described previously (25). Genomic DNAs of surviving colonies were isolated as described previously (25) and screened for homologous recombinants by Southern blot analysis after digestion with BamHI using a 0.7-kb PstI-XbaI fragment as a probe (Fig. 1, probe A). One homologous recombinant (clone 44) was identified and used for blastocyst injection, which was performed as described previously (25). Male chimeras were bred to C57BL/6J females to obtain F1 mice. F2 offspring and embryos were obtained by F1 heterozygous intercrosses.

Genotyping.

For genotyping by Southern blot analysis, 10 μg of DNA samples prepared from mouse tail tips were digested with BamHI and separated by 1% agarose gel electrophoresis. Filter preparation, hybridization, and washing were carried out as described previously (6). A ∼0.7-kb PstI fragment containing exons 6 and 7 and a ∼ 1.3-kb NcoI-BamHI fragment containing exon 1a of the Tsc2 gene were used as the 3′outside and 5′ inside probes, respectively (Fig. 1 A). To genotype F2 embryos by PCR, DNA was prepared from yolk sacs or embryonic tails by proteinase K digestion. Two primer sets, LZF2 and LZR2 for mutated allele-specific amplification (224 bp) and MTSC3 and MTSC9 for wild-type allele-specific amplification (160 bp), were used. PCR (35 cycles) was carried out as described previously, except for the use of Taq DNA polymerase (Toyobo, Tokyo, Japan; Ref. 15), and the products were analyzed by a 3% Nusieve agarose gel (FMC BioProducts, Rockland, ME) electrophoresis. Sequences and locations of the primers were as follows: LZF2, 5′-GGTAAACTGGCTCGGATTAG-3′ in the lacZ coding region; LZR2, 5′-CCATCAGTTGCTGTTGACTG-3′ in the lacZ coding region; MTSC3, 5′-AATGCCCTAAGTGCAACCTG-3′ in intron 2; and MTSC9, 5′-CTACATGGTAAGTCCCTGTC-3′ in intron 2.

Analyses of RNA and Protein.

For RT-PCR, first-strand cDNA was synthesized from 5 μg of total RNA from whole embryos at E 9.5 (noon of the day on which a vaginal plug was detected was defined as E 0.5). PCR (35 cycles) was carried out as described for genotyping (15). Primers used for the amplification of wild-type Tsc2 cDNA were MTSC10 (sense; 5′-TGTGCAGAAGGCAAACAGAC-3′ in exon 1) and MTSC11 (antisense; 5′-CATGTTGTAGATGGCGCTGT-3′ in exon 8). Primers used for the amplification of mutant cDNA were MTSC10 (sense) and LZRT (antisense, 5′-GACGTTGTAAAACGACGGGA-3′ in lacZ). Amplified products were analyzed by 2% Nusieve agarose gel electrophoresis. For the detection of tuberin, whole E9.5 embryos of comparable body size were directly homogenized in SDS-sample buffer [2.3% SDS, 0.0625 m Tris-HCl (pH 6.8), 5% β-mercaptoethanol, and 10% glycerol]. Comparable amounts of protein from each were separated on 5% SDS-PAGE, transferred onto nylon membrane, and probed with anti-human tuberin NH2-terminal and COOH-terminal peptide antibodies (Santa Cruz Biotechnology, Santa Cruz, CA) known to cross-react with mouse tuberin.

Carcinogen Treatment.

Female C57BL/6J mice were bred with Tsc2 heterozygous mutant (Tsc2+/−) F1 males. At E 14 or E 15, ENU solution (10 mg/ml in saline) was i.p. injected into bred females at a final dose of 50 mg/kg body weight (26). Offspring were sacrificed at the ages of 11 weeks or 6 months and examined for tumor formation.

Histological Analyses.

Whole mount X-gal staining of embryos was carried out using a standard protocol (27) after fixation in 4% paraformaldehyde/PBS (4°C; 20 min). Immunohistochemical analysis of embryos was performed using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA) after fixation in Bouin’s solution (4°C; overnight) and standard sectioning in paraffin. For orientation, sections were stained with H&E. The primary antibodies used were a mouse monoclonal anti-rat PCNA antibody (Novocastra Laboratories, Newcastle, United Kingdom), a mouse monoclonal anti-rat class III β-tubulin antibody (Babco, Richmond, CA), a rabbit polyclonal anti-nestin antibody (28), and mouse monoclonal anti-MAP2 antibodies (clone AP-20 and HM-2; Sigma, St. Louis, MO). Tissues from adult mice were examined by H&E staining as described previously (10).

Analysis of LOH.

For detection of LOH by PCR, tissues were fixed in 10% buffered formalin and sectioned (10 μm thick). Tumor samples and comparable amounts of normal kidney cortex areas were dissected separately from sections with needles under a stereomicroscope. Samples were dissolved in 25 μl of proteinase K solution [1 mg/ml proteinase K, 50 mm Tris-HCl (pH 8.5), 1 mm EDTA, and 0.5% Tween 20], digested overnight at 55°C, and heated at 95°C for 20 min. Aliquots (5 μl) of each mixture were subjected to PCR (35 cycles) in a 25-μl reaction volume containing two primer sets, MTSC3 and MTSC9 for the wild-type allele and LZF2 and LZR2 for the mutated allele. Amplified products were separated by 3% Nusieve agarose gel electrophoresis, transferred onto Nylon membranes, and hybridized with 32P-labeled internal oligonucleotides [MTSC14 (5′-CCACCAAGGTAAGTCTCTCA-3′) for the wild-type allele and NZF4 (5′-TGAATTATGGCCCACACCAG-3′) for the mutated allele]. For the detection of LOH by Southern blot analysis, membrane filters of tissue DNAs were prepared after BamHI digestion and hybridized as described for genotyping. For both PCR and Southern blot analysis, radioactivity was measured with a BAS2000 Bio-Image Analyzer (Fuji Film, Tokyo, Japan).

Generation of Tsc2 Mutant Mice.

To generate a line of Tsc2 mutant mice, we constructed a targeting vector, pKO-LT, in which a fragment from codon 74 in exon 2 to codon 164 in exon 5 of the mouse Tsc2 gene was deleted (Fig. 1,A). A NLS-LacZ coding fragment was introduced in frame at exon 2 to analyze the expression profile of the Tsc2 gene (Fig. 1,A). By gene targeting, one ES clone (clone 44) was identified as a homologous recombinant that gave rise to germ-line transmission of the mutated Tsc2 allele after blastocyst injection and subsequent mating of chimeras with C57BL/6J females (Fig. 1, A and B). PCR genotyping analysis of embryos at E 9.5 from F1 heterozygous intercrosses revealed the existence of homozygous mutant embryos (Tsc2−/−; Fig. 1,C). These embryos were subjected to Western blot analysis using anti-tuberin NH2 terminus or COOH terminus antibodies. The ∼180-kDa band of normal tuberin was absent in Tsc2−/− embryos in both cases, confirming the inactivation of the Tsc2 gene (Fig. 1 D).

The expression of the mutated Tsc2 allele was examined using RT-PCR and whole mount X-gal staining of Tsc2+/−embryos (Fig. 1, E and F). When the primer set encompassing exons 1–8 of the Tsc2 gene was used, the predicted 741-bp product derived from wild-type Tsc2 mRNA was amplified in wild-type (Tsc2+/+) as well as in Tsc2+/− embryos (Fig. 1,E). A 196-bp product was amplified in a Tsc2+/− embryo-specific manner using a lacZ coding region-specific primer, suggesting that splicing occurred between exon 1 and the modified exon 2 of the mutated Tsc2 allele (Fig. 1,E). Sequence analysis of this 196-bp product confirmed in frame fusion of the Tsc2 transcript with the NLS-lacZ at the 3′ end of exon 2 (data not shown). These results indicated that the mutated Tsc2 allele expressed Tsc2-lacZ chimeric mRNA, possibly resulting in the expression of a fusion protein consisting of the NH2-terminal 73 amino acid residues of tuberin and NLS-β-gal. Tsc2+/− embryo-specific expression of β-gal was detected by whole mount staining of E 11.5 embryos with X-gal (Fig. 1 F). Although a detailed histological examination at the single cell level has not been fully accomplished, organs including the heart and central nervous system were stained as reported in an earlier work by in situ hybridization analysis of Tsc2 gene expression in mouse embryos (29).

Development of RCs and Hepatic Hemangiomas in Tsc2 Heterozygous Mutant Mice.

Although Tsc2+/− mice were born without any apparent phenotype, they developed spontaneous bilateral and multiple RCs with age (Fig. 2,A). Macroscopically visible RCs developed in most Tsc2+/− mice (20 of 21 cases) by the age of 6 months, irrespective of the sex. The earliest occurrence of a macroscopically visible RC was seen in a female mouse at the age of 14 weeks (data not shown). By 10 months of age, virtually all Tsc2+/− mice developed RCs. None of the Tsc2+/+ mice developed RCs, even after the age of 1 year (n = 43). Macroscopically, the RCs were mostly cystic, with a diameter of 1–2 mm (Fig. 2,A). Histological examination revealed RCs to be located exclusively in the cortex region, and the majority showed a papillar growth within the cysts (Fig. 2,B). Some tubular-type RCs were also observed (Fig. 2,C). As seen in the Eker rat (30), the development of RCs in Tsc2+/− mice was accelerated by transplacental ENU-treatment of embryos at E 14.0 or E 15.0 (n = 7; Fig. 2 D). All ENU-treated Tsc2+/− offspring developed macroscopically visible RCs by the age of ∼11 weeks. Thus, the results indicate that a Tsc2 germ-line mutation in the mouse can cause hereditary RCs, as seen in the rat.

We searched for loss of the wild-type Tsc2 allele in spontaneous RCs in Tsc2+/− mice by PCR and Southern blot analysis. In all three cases of smaller RCs examined by PCR, less amplification of the wild-type Tsc2 allele-specific fragment was observed in the RC region than in the normal cortex region (Fig. 2,E; data not shown). Although substantial amounts of the PCR product from the wild-type Tsc2 allele were amplified from RCs, these were thought to be derived from contaminating normal tissues, which could not be completely eliminated during sample preparation. On Southern blot analysis of larger RCs, 4 of 11 cases showed loss of the wild-type Tsc2 allele (Fig. 2 F; data not shown). Thus, we infer that the wild-type Tsc2 allele may be lost during the development of RCs in Tsc2+/− mice.

Examination of aged Tsc2+/− mice (age > 1.5 years) revealed hepatic hemangiomas in 18 of 23 cases (∼80%; Fig. 2 G). These hemangiomas developed in both sexes and in all hepatic lobes. Histologically, they resembled cavernous hemangiomas consisting of a thin layer of endothelium and large cavities filled with blood. Thus far, our analysis has not detected loss of the wild-type Tsc2 allele in these hemangiomas (data not shown). The development of macroscopically visible hemangiomas in Tsc2+/− mice was not obvious at 10 months of age. In contrast to the Tsc2+/− mice, none of the Tsc2+/+ mice over 1.5 years of age (n = 29) developed any hepatic hemangiomas. These results indicate that a Tsc2 germ-line mutation causes hepatic hemangiomas in the mouse.

Embryonic Lethality of Tsc2 Homozygous Mutants.

Although intercrossing of F1Tsc2+/− mice was carried out extensively (111 offspring from 17 intercrosses), we could not obtain any Tsc2−/−offspring (43 Tsc2+/+ mice and 68 Tsc2+/− mice), indicating the embryonic lethality of Tsc2 homozygous mutants. Therefore, we analyzed the genotype of embryos from F1 heterozygous intercrosses to determine the lethal stage (Table 1). At E 10.0–10.5, 16% of all surviving embryos (7 of 44) were homozygous mutants. On the other hand, 5 of 6 Tsc2−/− embryos were dead at E 11.0–11.5. Therefore, we concluded that most Tsc2−/− embryos died at around E 10.5. However, dead or resorbed Tsc2−/− embryos were already present at E 9.0–9.5, and some Tsc2−/− embryos survived to E 11.5–12.5, suggesting that there might be considerable variation in the lethal stage.

Macroscopically, live Tsc2−/− embryos exhibited various outlooks, sometimes showing a smaller body size compared with control Tsc2+/+ or Tsc2+/− embryos (Fig. 3). The most prominent feature found in a significant proportion (∼50%) but not all of nonresorbed Tsc2−/− embryos at E 9.0–11.5 was the nonclosure of the neural tube in the head region (Figs. 3 and 4).

Histological Analysis of Tsc2 Homozygous Mutants.

To search for developmental abnormalities that may cause embryonic lethality, we examined Tsc2−/− embryos histologically at E 9.0–12.5. As noted macroscopically, nonclosure of the neural tube in Tsc2−/− embryos was apparent in the head region (Fig. 4,A). Fig. 4,A shows the most severe consequence of nonclosure found in an E 11.5 Tsc2−/− embryo by transverse sectioning. In the head region, the subventricular layer of the neural tube was completely exposed to body surface in this embryo. However, even in this case, the neural tubes in homozygous Tsc2 mutants formed the neuroepithelial architecture. To assess the early neuronal differentiation status in Tsc2−/− embryos, we performed immunohistochemical analysis for several neuronal markers (class III β-tubulin, nestin, and MAP2; Refs. 31, 32, 33) and PCNA. In the nonclosed neural tube of Tsc2−/− embryos (E 9.5∼), nestin-positive cells showing a process-like staining pattern were prominent throughout the mantle layer as seen in Tsc2+/+ or Tsc2+/− embryos (Fig. 4,C). In E 11.5 and E 12.5 embryos, class III β-tubulin- and MAP2-positive cells also accumulated in the marginal layer, regardless of the Tsc2 genotype (Fig. 4,B; data not shown). In contrast to these neuronal markers, PCNA-positive cells were localized mainly in the subventricular layer in Tsc2+/+, Tsc2+/−, and Tsc2−/− embryos (Fig. 4 C). These results indicated that cell division in the ventricular layer, cell migration to the marginal layer, and cell differentiation to a neural lineage still occur in the neural tube of Tsc2−/− embryos.

At the histological level, hearts of E 11.5 (two cases) and E 12.5 (one case) Tsc2−/− embryos exhibited abnormal myocardia (Fig. 5; data not shown). Although these two E 11.5 embryos had a smaller body size, the size of their hearts was comparable to the size of those of control embryos (Fig. 5,B, top panels; data not shown). Myocardia of these E 11.5 and E 12.5 Tsc2−/− embryos were thickened and associated with increased cell density (Fig. 5,A, bottom panels, Fig. 5,B, top panels). Consequently, ventricular cavities were narrowed (Fig. 5,A, top panels). Endocardia of those hearts became less recognizable than normal controls because of their tight association with thickened myocardia (Fig. 5,A, bottom panels, Fig. 5,B, top panels). Such myocardial abnormalities might have resulted from altered cell proliferation activity of cardiomyocytes in the outer mantle layer or trabeculae. However, no particular differences were found on PCNA staining between Tsc2−/− and Tsc2+/+ or Tsc2+/−hearts (Fig. 5 B). Differences between the developing hearts of E 9.0–10.5 Tsc2−/− embryos and control Tsc2+/+ or Tsc2+/− embryos were not obvious on histological examination (data not shown). Thus, the myocardial anomaly became apparent at later stages (E 11.5–12.5) in Tsc2−/− embryos.

In the present study, we generated and initially characterized a Tsc2 mutant mouse strain. The most notable phenotype of the Tsc2+/− mice was the development of RCs histologically resembling those in the Eker rat (6). Consistent with the tumor suppressor characteristics of the TSC2/Tsc2 gene, our LOH analysis detected a second hit in the wild-type Tsc2 allele in those RCs, as seen in hamartomas from TSC patients and RCs from the Eker rat (4, 5, 10, 11). As in the Eker rat, the development of hamartomas characteristic of human TSC was not observed in Tsc2+/− mice. The similar RC development in both mouse and rat Tsc2 mutants suggests that some species-specific mechanism may participate in the occurrence of phenotypic differences caused by tuberin deficiency.

However, there were also several differences in the incidences of extrarenal lesions between our Tsc2+/− mice and Eker rats (34). For example, a high incidence of hepatic hemangiomas was seen, although the underlying mechanism remains to be determined. Such hepatic hemangiomas have not been detected in the Eker rat. In contrast, in Tsc2+/− mice, we have not observed the pituitary adenomas or subependymal nodules that often develop in the Eker rat (14, 34). Although we have identified a few cases of Tsc2+/− mice exhibiting tumors in the spleen, uterus, and foot pad, 4 the incidences of these tumors were not as high as those of tumors in the spleen and uterus in the Eker rat (34). Thus, differences in the cell-type specificity of tumorigenesis caused by tuberin deficiency might exist not only between man and rodents but also between rat and mouse.

Because all of the Tsc2−/− embryos died in utero, an essential role for tuberin function in mouse development was revealed (6). Although we have not detected the common lethal defect, the phenotypes of the Tsc2−/− embryos did provide some insights into the physiological functions of tuberin. The most characteristic histological feature seen in Tsc2−/− embryos in later stages was anomalies of the myocardium. The increased cell density associated with cardiac anomalies might have been caused by loss of the growth-suppressive function of tuberin. Because myocardia in Tsc2−/− embryos at earlier stages (E 9.0–10.5) exhibited a normal architecture with substantial trabeculation and the hearts of Tsc2−/− embryos beat, early differentiation of cardiomyocytes into functional heart muscle occurs normally. However, late-stage differentiation or maturation processes in cardiac development may be disturbed by tuberin deficiency. Because human neonates with TSC often develop cardiac rhabdomyomas that originate primarily in the ventricle (1), we infer that the cardiac anomaly in Tsc2−/− embryos is directly linked to the tuberin deficiency and that tuberin function is important for the growth and differentiation of cardiomyocytes. Recently, Pajak et al.(35) reported that the DNA synthetic activity of Tsc2−/− rat cardiomyocytes in the embryonic primary culture is sustained, suggesting the involvement of tuberin in cardiac development.

Nonclosure of the neural tube in the brain region was another prominent feature found in Tsc2−/− embryos. Considering its occurrence in E 10–11.5 Tsc2−/− embryos and the developmental progress of other tissues in each embryo, this neural tube abnormality may not be simply a consequence of whole embryonic growth retardation. Further investigation of the specific role of Tsc2 in neural tube development and the presence of modifier gene(s) is needed, because nonclosure was seen in only ∼50% of Tsc2−/− embryos. In contrast to the heart anomaly in Tsc2−/− embryos, which might result in the failure of heart activity, the neural tube nonclosure itself might not give rise to embryonic lethality. Indeed, some homozygous mutants were found to have a closed neural tube. In addition, mouse embryos can be born without a forebrain-midbrain region, as seen in Lim1-deficient mice (36). However, a contribution of tuberin function to the development of the nervous system might be expected because TSC patients frequently develop lesions such as cortical tubers and subependymal nodules in the brain (1). Histological studies including immunostaining for various neuronal and glial markers suggested that the failure of cell differentiation and/or migration defect may be associated with the development of those lesions (37, 38). However, the unclosed neural tubes in Tsc2−/− embryos demonstrated substantially organized neuroepithelial structures. Moreover, our immunohistochemical analysis for several neuronal differentiation markers and PCNA suggests that, at least in the mouse, the potential to differentiate into early neuronal lineage was retained normally by neuronal stem cells in Tsc2−/− embryos. Because of the embryonic lethality, the effects of tuberin deficiency in mice in later embryonic stages or in adulthood could not be analyzed. To circumvent embryonic lethality, conditional gene targeting of Tsc2 is necessary for a more complete functional analysis of tuberin (22, 39).

In summary, we have established a mouse Tsc2 mutant line and demonstrated spontaneous renal carcinogenesis and hepatic hemangiomatosis in heterozygotes as well as embryonic lethality in homozygotes. Extensive studies of this mouse system, using the genetic advantage of this species, should facilitate the elucidation of the physiological function of tuberin in cell growth and differentiation and the molecular mechanisms of tumorigenesis associated with tuberin deficiency.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

      
1

Supported in part by Grants-in Aid for Cancer Research from the Ministry of Education, Science, Sports and Culture and the Ministry of Health and Welfare of Japan and the Organization for Pharmaceutical Safety and Research.

            
3

The abbreviations used are: TSC, tuberous sclerosis; RC, renal carcinoma; E, embryonic day; LOH, loss of heterozygosity; GAP, GTPase-activating protein; NLS, nuclear localization signal; DTA, diphtheria toxin A fragment; RT-PCR, reverse transcription-PCR; ENU, N-ethyl-N-nitrosourea; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactopy-ranoside; PCNA, proliferating cell nuclear antigen; MAP2, microtubule-associated protein 2; β-gal, β-galactosidase; ES, embryonic stem.

      
4

Unpublished observation.

Fig. 1.

Targeted disruption of the mouse Tsc2 gene. A, structures of the targeting vector and the wild-type and mutated Tsc2 allele. Exons, ▪ with numbers. The coding region for β-gal with a NLS (NLS-lacZ) and expression cassettes for neomycin-resistant (neo) and DTA genes are shown by □. and , loxP sequences flanking the neo cassette and the position of a 0.7-kb fragment used as a screening probe (probe A), respectively. The lengths of BamHI (B) fragments from wild-type (∼13 kb) and mutated (∼7.5 kb) alleles detected by probe A are shown below each allele with arrows. Approximate positions of primers used for PCR genotyping are shown by triangles (▴, the wild-type allele; ▵, the mutated allele) below each allele. B, Southern blot analysis of tail DNAs from F1 littermates after digestion by BamHI. Right, positions of fragments from wild-type (wt; ∼13 kb) and mutated (mut; ∼7.5 kb) alleles are indicated. Wild-type mice, +/+; heterozygous mutant mice, +/−. C, PCR genotyping of E 9.5 embryos from a F1 heterozygous breeding. Right, positions of bands amplified from wild-type (wt; 160 bp) and mutated (mut; 224 bp) alleles are indicated. Wild-type embryo, +/+; heterozygous embryo, +/−; homozygous embryo, −/−. D, Western blot analysis of E 9.5 embryos from a F1 heterozygous breeding. After separation on 5% SDS-PAGE, samples were probed with anti-tuberin NH2-terminal (anti-N) or COOH-terminal (anti-C) antibodies. The position of the ∼180-kDa band for wild-type tuberin (arrow) and the genotypes are indicated. E, RT-PCR analysis of F1 embryos at E 11.5. RNA samples from whole embryos were subjected to RT-PCR analysis using a primer set on exons 1 and 8 (MTSC10 × MTSC11, top panel) or a mutated allele-specific primer set on exon 1 and lacZ (MTSC10 × LZRT, bottom panel). Right, sizes of products are shown. Wild-type, +/+; heterozygous mutant, +/−. F, whole mount X-gal staining of F1 embryos at E 11.5. Genotypes are indicated. Note Tsc2 ± embryo-specific staining, especially in the brain, spinal cord, and heart (arrows).

Fig. 1.

Targeted disruption of the mouse Tsc2 gene. A, structures of the targeting vector and the wild-type and mutated Tsc2 allele. Exons, ▪ with numbers. The coding region for β-gal with a NLS (NLS-lacZ) and expression cassettes for neomycin-resistant (neo) and DTA genes are shown by □. and , loxP sequences flanking the neo cassette and the position of a 0.7-kb fragment used as a screening probe (probe A), respectively. The lengths of BamHI (B) fragments from wild-type (∼13 kb) and mutated (∼7.5 kb) alleles detected by probe A are shown below each allele with arrows. Approximate positions of primers used for PCR genotyping are shown by triangles (▴, the wild-type allele; ▵, the mutated allele) below each allele. B, Southern blot analysis of tail DNAs from F1 littermates after digestion by BamHI. Right, positions of fragments from wild-type (wt; ∼13 kb) and mutated (mut; ∼7.5 kb) alleles are indicated. Wild-type mice, +/+; heterozygous mutant mice, +/−. C, PCR genotyping of E 9.5 embryos from a F1 heterozygous breeding. Right, positions of bands amplified from wild-type (wt; 160 bp) and mutated (mut; 224 bp) alleles are indicated. Wild-type embryo, +/+; heterozygous embryo, +/−; homozygous embryo, −/−. D, Western blot analysis of E 9.5 embryos from a F1 heterozygous breeding. After separation on 5% SDS-PAGE, samples were probed with anti-tuberin NH2-terminal (anti-N) or COOH-terminal (anti-C) antibodies. The position of the ∼180-kDa band for wild-type tuberin (arrow) and the genotypes are indicated. E, RT-PCR analysis of F1 embryos at E 11.5. RNA samples from whole embryos were subjected to RT-PCR analysis using a primer set on exons 1 and 8 (MTSC10 × MTSC11, top panel) or a mutated allele-specific primer set on exon 1 and lacZ (MTSC10 × LZRT, bottom panel). Right, sizes of products are shown. Wild-type, +/+; heterozygous mutant, +/−. F, whole mount X-gal staining of F1 embryos at E 11.5. Genotypes are indicated. Note Tsc2 ± embryo-specific staining, especially in the brain, spinal cord, and heart (arrows).

Close modal
Fig. 2.

Tumor development in Tsc2 heterozygous mutant mice. A, RCs in a Tsc2 heterozygous mutant mouse. Kidneys from a 7-month-old female mutant are shown. Arrows indicate spontaneous RCs developing bilaterally. B and C, H&E staining of kidney sections containing RC with cystic (B) or tubular-type features (C). The asterisk in B indicates the lumen of a cyst. Bar, 30 μm. D, kidneys from an 11-week-old female mutant treated with ENU at E 15.0. Note the accelerated development of RCs compared with the kidneys in A. E, detection of LOH by PCR in spontaneous RCs. A representative case of LOH is shown. N, normal tissue; T, tumor. Right, positions and sizes (in parentheses) of amplified products from the wild-type (wt) and mutated (mut) alleles are indicated. F, detection of LOH by Southern blot analysis in spontaneous RCs. Two representative cases of LOH are shown. N, normal tissue; T, tumor. Right, positions and sizes (in parentheses) of fragments from wild-type (wt) and mutated (mut) alleles are indicated. G, H&E staining of a hepatic hemangioma from a Tsc2 heterozygous mutant mouse. Bar, 30 μm.

Fig. 2.

Tumor development in Tsc2 heterozygous mutant mice. A, RCs in a Tsc2 heterozygous mutant mouse. Kidneys from a 7-month-old female mutant are shown. Arrows indicate spontaneous RCs developing bilaterally. B and C, H&E staining of kidney sections containing RC with cystic (B) or tubular-type features (C). The asterisk in B indicates the lumen of a cyst. Bar, 30 μm. D, kidneys from an 11-week-old female mutant treated with ENU at E 15.0. Note the accelerated development of RCs compared with the kidneys in A. E, detection of LOH by PCR in spontaneous RCs. A representative case of LOH is shown. N, normal tissue; T, tumor. Right, positions and sizes (in parentheses) of amplified products from the wild-type (wt) and mutated (mut) alleles are indicated. F, detection of LOH by Southern blot analysis in spontaneous RCs. Two representative cases of LOH are shown. N, normal tissue; T, tumor. Right, positions and sizes (in parentheses) of fragments from wild-type (wt) and mutated (mut) alleles are indicated. G, H&E staining of a hepatic hemangioma from a Tsc2 heterozygous mutant mouse. Bar, 30 μm.

Close modal
Fig. 3.

Typical abnormal appearance of Tsc2 homozygous mutant embryos. Embryos at E 10.0 (A) and E 11.5 (B) from F1 heterozygous breedings are shown (A, dark field; B, light field). Tsc2 heterozygous mutant, +/−; Tsc2 homozygous mutant, −/−. Note the smaller body size of the Tsc2−/− mutants. Arrows indicate the region of failure of neural tube closure. Bar, 750 (A) and 1000 μm (B).

Fig. 3.

Typical abnormal appearance of Tsc2 homozygous mutant embryos. Embryos at E 10.0 (A) and E 11.5 (B) from F1 heterozygous breedings are shown (A, dark field; B, light field). Tsc2 heterozygous mutant, +/−; Tsc2 homozygous mutant, −/−. Note the smaller body size of the Tsc2−/− mutants. Arrows indicate the region of failure of neural tube closure. Bar, 750 (A) and 1000 μm (B).

Close modal
Fig. 4.

Histological analysis of neural tubes. A, H&E staining of transverse sections of the head region of Tsc2 homozygous mutant (−/−, bottom panel) and control heterozygous mutant (+/−, top panel) embryos at E 11.5. The former is the most severe example of the consequences of nonclosure of the neural tube found in an E 11.5 Tsc2−/− embryo. The subventricular layer of neural tube is completely exposed to body surface. Bar, 200 μm. B, anti-βIII-tubulin staining of neural tubes of Tsc2 homozygous mutant (−/−) and control wild-type (+/+) embryos at E 11.5. Transverse sections were stained with anti-βIII-tubulin and counterstained with hematoxylin. In each case, the bottom panel is the enlarged view of the region enclosed by a rectangle in the top panel. Note the prominent staining in the marginal layer. VZ, ventricular zone; MZ, marginal zone. Bar, 200 (top panels) and 40 μm (bottom panels). C, anti-nestin and anti-PCNA staining of Tsc2 homozygous mutant (−/−) and control heterozygous (+/−) embryos at E 11.5. As in anti-βIII-tubulin staining (B), neural tubes were stained with anti-nestin followed by a light counterstain with hematoxylin (top panels) or with anti-PCNA followed by a light counterstain with eosin (bottom panels) using serial sections. VZ, ventricular zone; MZ, marginal zone. Bar, 40 μm.

Fig. 4.

Histological analysis of neural tubes. A, H&E staining of transverse sections of the head region of Tsc2 homozygous mutant (−/−, bottom panel) and control heterozygous mutant (+/−, top panel) embryos at E 11.5. The former is the most severe example of the consequences of nonclosure of the neural tube found in an E 11.5 Tsc2−/− embryo. The subventricular layer of neural tube is completely exposed to body surface. Bar, 200 μm. B, anti-βIII-tubulin staining of neural tubes of Tsc2 homozygous mutant (−/−) and control wild-type (+/+) embryos at E 11.5. Transverse sections were stained with anti-βIII-tubulin and counterstained with hematoxylin. In each case, the bottom panel is the enlarged view of the region enclosed by a rectangle in the top panel. Note the prominent staining in the marginal layer. VZ, ventricular zone; MZ, marginal zone. Bar, 200 (top panels) and 40 μm (bottom panels). C, anti-nestin and anti-PCNA staining of Tsc2 homozygous mutant (−/−) and control heterozygous (+/−) embryos at E 11.5. As in anti-βIII-tubulin staining (B), neural tubes were stained with anti-nestin followed by a light counterstain with hematoxylin (top panels) or with anti-PCNA followed by a light counterstain with eosin (bottom panels) using serial sections. VZ, ventricular zone; MZ, marginal zone. Bar, 40 μm.

Close modal
Fig. 5.

Histological analysis of embryonic hearts. A, cardiac anomalies in the Tsc2 homozygous mutant. Transverse sections of embryonic hearts from Tsc2 homozygous mutant (−/−, right panels) and control wild-type (+/+, left panels) embryos at E 12.5 were stained with H&E. RA, right atrium; LA, left atrium; RV, right ventricle; LV, left ventricle. Bottom panels are enlarged views of myocardia enclosed in rectangles in the top panels. Bar, 150 (top panels) and 30 μm (bottom panels). B, PCNA staining of embryonic hearts. Ventricular parts of E 11.5 Tsc2 homozygous mutant (−/−, right panels) and control wild-type (+/+, left panels) embryos were stained with H&E (top panels) or with anti-PCNA antibody followed by light counterstain of eosin (bottom panels) using serial sections. LV, left ventricle; CZ, compact zone. Positively stained nuclei were observed throughout both the compact zone and the trabeculated area in both embryos. Bar, 30 μm.

Fig. 5.

Histological analysis of embryonic hearts. A, cardiac anomalies in the Tsc2 homozygous mutant. Transverse sections of embryonic hearts from Tsc2 homozygous mutant (−/−, right panels) and control wild-type (+/+, left panels) embryos at E 12.5 were stained with H&E. RA, right atrium; LA, left atrium; RV, right ventricle; LV, left ventricle. Bottom panels are enlarged views of myocardia enclosed in rectangles in the top panels. Bar, 150 (top panels) and 30 μm (bottom panels). B, PCNA staining of embryonic hearts. Ventricular parts of E 11.5 Tsc2 homozygous mutant (−/−, right panels) and control wild-type (+/+, left panels) embryos were stained with H&E (top panels) or with anti-PCNA antibody followed by light counterstain of eosin (bottom panels) using serial sections. LV, left ventricle; CZ, compact zone. Positively stained nuclei were observed throughout both the compact zone and the trabeculated area in both embryos. Bar, 30 μm.

Close modal
Table 1

Genotype analysis of embryos from Tsc2 heterozygous intercrosses

No. of embryos with the indicated Tsc2 genotypes
EaLitters+/++/−−/−No. of total genotyped embryosNo. of resorbed embryos (not genotyped)
≤8.5 12 
9.0–9.5 18 45 19 (7)b 82 
10.0–10.5 18 28 10 (5) 56 
11.0–11.5 17 38 17 (15) 72 
12.0–12.5 26 6 (5) 41 
13.0–13.5 10 19 1 (1) 30 14 
≥14.0 
No. of embryos with the indicated Tsc2 genotypes
EaLitters+/++/−−/−No. of total genotyped embryosNo. of resorbed embryos (not genotyped)
≤8.5 12 
9.0–9.5 18 45 19 (7)b 82 
10.0–10.5 18 28 10 (5) 56 
11.0–11.5 17 38 17 (15) 72 
12.0–12.5 26 6 (5) 41 
13.0–13.5 10 19 1 (1) 30 14 
≥14.0 
a

Noon of the day on which a vaginal plug was checked was defined as embryonic day (E) 0.5.

b

The number in parentheses indicates the number of dead (resorbed or with no heart beat) embryos among total genotyped −/− embryos.

We thank I. Komuro and M. Miyagawa for histological analysis; Y. Tomooka for anti-nestin antibody; H. Yamanaka, Y. Sugitani, T. Takahara, Y. Hirayama, and E. Kobayashi for excellent technical assistance, and other members of the Hino and Noda laboratories for stimulating discussions and invaluable advice. We also thank Drs. H. Sugano, T. Kitagawa, and A. G. Knudson for encouragement throughout this work.

1
Gomez M. R. Tuberous Sclerosis 2nd ed. Raven Press New York  
1988
.
2
van Slegtenhorst M., de Hoogt R., Hermans C., Nellist M., Janssen B., Verhoef S., Lindhout D., van den Ouweland A., Hally D., Young J., Burley M., Jeremiah S., Woodward K., Nahmias J., Fox M., Ekong R., Osborne J., Wolfe J., Povey S., Snell R. G., Cheadle J. P., Jones A. C., Tachataki M., Ravine D., Sampson J. R., Reeve M. P., Richardson P., Wilmer F., Munro C., Hawkins T. L., Sepp T., Ali J. B. M., Ward S., Green A. J., Yates J. R. W., Kwiatkowska J., Henske E. P., Short M. P., Haines J. H., Jozwiak S., Kwiatkowski D. J. Identification of the tuberous sclerosis gene TSC1 on chromosome 9q34.
Science (Washington DC)
,
277
:
805
-808,  
1997
.
3
The European Chromosome 16 Tuberous Sclerosis Consortium. Identification and characterization of the tuberous sclerosis gene on chromosome 16.
Cell
,
75
:
1305
-1315,  
1193
.
4
Sepp T., Yates J. R. W., Green A. J. Loss of heterozygosity in tuberous sclerosis hamartomas.
J. Med. Genet.
,
33
:
962
-964,  
1996
.
5
Knudson A. G. Mutation and cancer: statistical study of retinoblastoma.
Proc. Natl. Acad. Sci. USA
,
68
:
820
-823,  
1971
.
6
Hino O., Klein-Szanto A. J. P., Freed J. J., Testa J. R., Brown D. Q., Vilensky M., Yeung R. S., Tartof K. D., Knudson A. G. Spontaneous and radiation-induced renal tumors in the Eker rat model of dominantly inherited cancer.
Proc. Natl. Acad. Sci. USA
,
90
:
327
-331,  
1993
.
7
Hino O., Kobayashi T., Tsuchiya H., Kikuchi Y., Kobayashi E., Mitani H., Hirayama Y. The predisposing gene of the Eker rat inherited cancer syndrome is tightly linked to the tuberous sclerosis (Tsc2) gene.
Biochem. Biophys. Res. Commun.
,
203
:
1302
-1308,  
1994
.
8
Kobayashi T., Hirayama Y., Kobayashi E., Kubo Y., Hino O. A germline insertion in the tuberous sclerosis (Tsc2) gene gives rise to the Eker rat model of dominantly inherited cancer.
Nat. Genet.
,
9
:
70
-74,  
1995
.
9
Yeung R. S., Xiao G-H., Jin F., Lee W-C., Testa J. R., Knudson A. G. Predisposition to renal carcinoma in the Eker rat is determined by germ-line mutation of the tuberous sclerosis 2 (TSC2) gene.
Proc. Natl. Acad. Sci. USA
,
91
:
11413
-11416,  
1994
.
10
Kubo Y., Mitani H., Hino O. Allelic loss at predisposing gene locus in spontaneous and chemically induced renal cell carcinomas in the Eker rat.
Cancer Res.
,
54
:
2633
-2635,  
1994
.
11
Yeung R. S., Xiao G-H., Everitt J. I., Jin F., Walker C. L. Allelic loss at the tuberous sclerosis 2 locus in spontaneous tumors in the Eker rat.
Mol. Carcinog.
,
14
:
28
-36,  
1995
.
12
Kobayashi T., Urakami S., Hirayama Y., Yamamoto T., Nishizawa M., Takahara T., Kubo Y., Hino O. Intragenic Tsc2 somatic mutations as Knudson’s second hit in spontaneous and chemically induced renal carcinomas in the Eker rat model.
Jpn. J. Cancer Res.
,
88
:
254
-261,  
1997
.
13
Kobayashi T., Mitani H., Takahashi R., Hirabayashi M., Ueda M., Tamura H., Hino O. Transgenic rescue from embryonic lethality, and renal carcinogenesis in the Eker rat model by introduction of a wild-type Tsc2 gene.
Proc. Natl. Acad. Sci. USA
,
94
:
3990
-3993,  
1997
.
14
Yeung R. S., Katsetos C. D., Klein-Szanto A. Subependymal astrocytic hamartomas in the Eker rat model of tuberous sclerosis.
Am. J. Pathol.
,
151
:
1477
-1486,  
1997
.
15
Kobayashi T., Nishizawa M., Hirayama Y., Kobayashi E., Hino O. cDNA structure, alternative splicing and exon-intron organization of the predisposing tuberous sclerosis (Tsc2) gene of the Eker rat model.
Nucleic Acids Res.
,
23
:
2608
-2613,  
1995
.
16
Maheshwar M. M., Sandford R., Nellist M., Cheadle J. P., Sgotto B., Vaudin M., Sampson J. R. Comparative analysis and genomic structure of the tuberous sclerosis 2 (TSC2) gene in human and pufferfish.
Hum. Mol. Genet.
,
5
:
131
-137,  
1996
.
17
Wienecke R., Konig A., DeClue J. E. Identification of tuberin, the tuberous sclerosis-2 product.
J. Biol. Chem.
,
270
:
16409
-16414,  
1995
.
18
Xiao G-H., Shoarinejad F., Jin F., Golemis E. A., Yeung R. S. The tuberous sclerosis 2 gene product, tuberin, functions as a rab5 GTPase activating protein (GAP) in modulating endocytosis.
J. Biol. Chem.
,
272
:
6097
-6100,  
1997
.
19
Tsuchiya H., Orimoto K., Kobayashi T., Hino O. Presence of potent transcriptional activation domains in the predisposing tuberous sclerosis (Tsc2) gene product of the Eker rat model.
Cancer Res.
,
56
:
429
-433,  
1996
.
20
Henry K. W., Yuan X., Koszewski N. J., Onda H., Kwiatkowski D. J., Noonan D. J. Tuberous sclerosis gene 2 product modulates transcription mediated by steroid hormone receptor family members.
J. Biol. Chem.
,
273
:
20535
-20539,  
1998
.
21
van Slegtenhorst M., Nellist M., Nagelkerken B., Cheadle J. P., Snell R., van den Ouweland A., Reuser A., Sampson J., Halley D., van der Sluijs P. Interaction between hamartin and tuberin, the TSC1 and TSC2 gene products.
Hum. Mol. Genet.
,
7
:
1053
-1057,  
1998
.
22
McClatchey A. I., Jacks T. Tumor suppressor mutations in mice: the next generation.
Curr. Opin. Genet. Dev.
,
8
:
304
-310,  
1998
.
23
Rudnicki M. A., Braun T., Hinuma S., Jaenisch R. Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development.
Cell
,
71
:
383
-390,  
1992
.
24
Yagi T., Ikawa Y., Yoshida K., Shigetani Y., Takeda N., Mabuchi I., Yamamoto T., Aizawa S. Homologous recombination at c-fyn locus of mouse embryonic stem cells with use of diphtheria toxin A-fragment gene in negative selection.
Proc. Natl. Acad. Sci. USA
,
87
:
9918
-9922,  
1990
.
25
Nakai S., Kawano H., Yudate T., Nishi M., Kuno J., Nagata A., Jishage K., Hamada H., Fujii H., Kawamura K., Shiba K., Noda T. The POU domain transcription factor Brn-2 is required for the determination of specific neuronal lineages in the hypothalamus of the mouse.
Genes Dev.
,
9
:
3109
-3121,  
1995
.
26
Shibuya T., Murota T., Horiya N., Matsuda H., Hara T. The induction of recessive mutations in mouse primordial germ cells with N-ethyl-N-nitrosourea.
Mutat. Res.
,
290
:
273
-280,  
1993
.
27
Bonnerot C., Nicolas J-F. Application of LacZ gene fusions to postimplantation development Wassarman P. M. DePamphilis M. L. eds. .
Guide to Techniques in Mouse Development
,
:
451
-469, Academic Press London  
1993
.
28
Tomooka Y., Kitani H., Jing N., Matsushima M., Sakakura T. Reconstruction of neural tube-like structures in vitro from primary neural precursor cells.
Proc. Natl. Acad. Sci. USA
,
90
:
9683
-9687,  
1993
.
29
Olsson P. G., Schofield J. N., Edwards Y. H., Frischauf A. M. Expression and differential splicing of the mouse homolog TSC2 homolog.
Mamm. Genome
,
7
:
212
-215,  
1996
.
30
Hino O., Mitani H., Knudson A. G. Genetic predisposition to transplacentally induced renal cell carcinomas in the Eker rat.
Cancer Res.
,
53
:
5856
-5858,  
1993
.
31
Caccamo D. V., Katsetos C. D., Frankfurter A., Collins V. P., Vandenburg S. R., Herman M. M. An immunohistochemical characterization of the primitive and maturing neuroepithelial components in the OTT-6050 transplantable mouse teratoma.
Neuropathol. Appl. Neurobiol.
,
15
:
389
-405,  
1989
.
32
Ferreira A., Busciglio J., Caceres A. An immunocytochemical analysis of the ontogeny of the microtuble-associated proteins MAP-2 and Tau in the nervous system of the rat.
Brain Res.
,
431
:
9
-31,  
1987
.
33
Lendahl U., Zimmerman L. B., Mackay R. D. G. CNS stem cells express a new class of intermediate filament protein.
Cell
,
60
:
585
-595,  
1990
.
34
Hino O., Mitani H., Katsuyama H., Kubo Y. A novel cancer predisposition syndrome in the Eker rat model.
Cancer Lett.
,
83
:
117
-121,  
1994
.
35
Pajak L., Jin F., Xiao G-H., Soonpaa M. H., Field L. J., Yeung R. S. Sustained cardiomyocyte DNA synthesis in whole embryo cultures lacking the TSC2 gene product.
Am. J. Physiol.
,
273
:
H1619
-H1627,  
1997
.
36
Shawlot W., Behringer R. R. Requirement for Lim1 in head-organizer function.
Nature (Lond.)
,
374
:
425
-430,  
1995
.
37
Crino P. B., Trojanowski J. Q., Dichter M. A., Eberwine J. Embryonic neuronal markers in tuberous sclerosis: single-cell molecular pathology.
Proc. Natl. Acad. Sci. USA
,
93
:
14152
-14157,  
1996
.
38
Yamanouchi H., Jay V., Rutka J. T., Takashima S., Becker L. E. Evidence of abnormal differentiation in giant cells of tuberous sclerosis.
Pediat. Neurol.
,
17
:
49
-53,  
1997
.
39
Shibata H., Toyama K., Shioya H., Ito M., Hirota M., Hasegawa S., Matsumoto H., Takano H., Akiyama T., Toyoshima K., Kanamaru R., Kanegae Y., Saito I., Nakamura Y., Shiba K., Node T. Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene.
Science (Washington DC)
,
278
:
120
-123,  
1997
.