| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Department of Experimental Pathology, Cancer Institute, Tokyo 170-8455 [N. S., T. K., E. K., O. H.], and Second Department of Pathology, School of Medicine, The University of Tokushima, Tokushima 770-8503 [N. S., K. I.], Japan
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
86% amino acid sequence identity. Using PCR-single-strand conformational polymorphism analysis, we identified two splicing donor site mutations in one chemically induced rat RC (1 of 15). This suggests that alterations of the Tsc1 gene may be involved in the development of a subset of rat RCs. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tsc2 encodes tuberin, a 200-kDa protein that contains a Rap1-GAP homology region near its COOH terminus (12)
. Recently, Rap1-GAP, Rab5-GAP, and transcriptional activator activities, as well as other possible functions of tuberin, were reported (13, 14, 15, 16)
. However, the precise in vivo function of tuberin and the molecular mechanisms of tumor development associated with Tsc2 mutations have not been fully elucidated.
TSC is an autosomal dominantly inherited disease characterized by hamartomatous benign tumors in multiple organs such as the brain, kidney, and heart (17)
. Approximately half of TSC patients are sporadic cases without any familial history (18)
. In addition to the TSC2 gene localized on chromosome 16p13.3, another TSC-predisposing gene (TSC1) was recently identified on chromosome 9q34 (19
, 20) . TSC1 encodes hamartin, which contains a predicted coiled-coil region and one potential transmembrane region, with a calculated molecular mass of 130 kDa (20)
. The symptomatic similarity among human TSC patients associated with either TSC1 or TSC2 mutation suggests that hamartin and tuberin may be involved in a common biochemical pathway in vivo (19
, 21)
. Indeed, direct interactions between tuberin and hamartin have been reported (21)
.
Because of the highly conserved nature of the Tsc2 gene structure in vertebrates, the function and possible interaction of tuberin with hamartin may be conserved. Thus, we anticipated that studies of rat hamartin might provide some clues for elucidation of the molecular mechanisms of renal carcinogenesis in the Eker rat model. Therefore, we cloned and structurally characterized the rat Tsc1 gene. Moreover, we identified somatic Tsc1 mutations in one chemically induced rat RC.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DNA and RNA Isolation and Southern Blot and Northern Blot Analyses.
DNAs were isolated from tumor samples by proteinase K digestion as described previously (22)
. Total RNAs were prepared with Isogen reagent (Nippon gene) according to the manufacturer’s instructions. Southern and Northern blot analyses were performed as described previously (3)
. The membrane filters for hybrid cell panel analysis were kindly provided by Dr. G. Levan (University of Goteborg, Goteborg, Sweden; Refs. 23, 24, 25
).
RT-PCR.
For the initial isolation of partial rat Tsc1 cDNA, first-strand cDNA was synthesized from 5 µg of rat (Long Evans strain) testis total RNA by RT using Superscript II (Life Technologies) and random primers in 20 µl of reaction mixture. A 1-µl aliquot of RT solution was used for PCR in 25 µl of reaction mixture containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 0.1% Triton X-100, 1 mM MgCl2, 200 mM deoxynucleotide triphosphates, 2 units of Taq DNA polymerase (Toyobo), and 100 pmol each of forward primer HTSF2 (5'-CTGGATCCCACAGAAGCCT-3') and reverse primer HTSR2 (5'-CAGTCGACAGACTTGCTGG-3'). Temperature conditions were 92°C for 3 min for initial denaturation; 92°C for 1 min, 55°C for 1 min, and 72°C for 1.5 min for the amplification (35 cycles); and 72°C for 3 min for the final extension (protocol I). Similarly, the 5' part of the human TSC1 cDNA was obtained by RT-PCR using forward primer HTSF1 (5'-GACCATGGCCCAACAAGCAA-3'), reverse primer HTR1 (5'-GGTGAGACACAGAATAGCCA-3'), and total RNA from the G401 cell line. To obtain the full coding region of rat Tsc1 cDNA, first-strand cDNA was synthesized from rat (Long Evans) kidney total RNA using random primers, and PCR was performed using an Expand High Fidelity PCR kit (Boehringer Mannheim) according to manufacturer’s instructions. Temperature conditions for PCR were 92°C for 3 min, 55°C for 1 min, and 72°C for 1.5 min for the initial polymerization; 92°C for 1 min, 62°C for 1 min, and 72°C for 1.5 min for the amplification (35 cycles); and 72°C for 3 min for the final extension (protocol II). The primers used were as follows: primers TSS1 (forward primer; 5'-ATCTAAAGAGCTTTCTGAGATCACCACT-3') and TSA1 (reverse primer; 5'-GAGGTACCTCAGCTGTGTTCGTGATGAG-3') were used for the 3' part of the coding region; and primers TSS2 (forward primer; 5'-TGGAATTCACCATGGCCCAGCTAGCCAACA-3') and TSA2 (reverse primer; 5'-AATGTCTTCCACCTTCGAGGGTCCAGTTCA-3') were used for the 3' part of the coding region. In each case, amplified products were subcloned into pBluescript II SK(+) (Stratagene) for sequencing and further analysis.
Screening of Cosmid and Phage Libraries.
Rat (Wistar strain) cosmid and phage (Stratagene) genomic DNA libraries and a rat (Long Evans strain) kidney cDNA library (12)
were screened with 32P-labeled rat Tsc1 and human TSC1 cDNA fragments obtained by RT-PCR. Positive clones were analyzed by restriction enzyme digestion and Southern blot analysis using cDNA probes [RT-PCR fragments and human TSC1 cDNA clone HA4782 (kindly provided by Dr. T. Nagase, Kazusa DNA Research Institute, Kisarazu, Japan; Ref. 26
)]. For sequence analysis, various restriction fragments from cosmid and phage clones were subcloned into pBluescript II SK(+).
The 5'-RACE.
The 5'-RACE based on inverse PCR was performed according to the method described by Maruyama et al. (27)
with some modifications. Rat brain total RNA (3 µg) was used as a template for the RT reaction using the antisense primer TSRT6 (5'-AGGTGCTTATCATGTGGCTC-3'). After the hydrolysis of RNA with 0.5 N NaOH, cDNAs were precipitated and circularized in 40 µl of reaction mixture containing 50 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 10 mM DTT, 1 mM ATP, 20% polyethylene glycol 6000, and 50 units of T4 RNA ligase (Takara) at 16°C. Ligation mixture (0.3 µl) was subjected to PCR using primers TSRT1 (reverse primer; 5'-GGGACTCCTTGAAGATGGTT-3') and TSRT3 (forward primer; 5'-GTGGGCCTATGCTTGTAAAC-3'). Temperature conditions for PCR were the same as those for protocol I. A 1-µl aliquot of the first PCR solution was used for the second PCR using primers TSRT4 (reverse primer; 5'-GCGAATTCGAGAGTAGCTCCCCAATGTT-3') and TSRT5 (forward primer; 5'-ATCTCGAGGTTGGTGGATTATTACCTGGA-3'). Amplified products were subcloned into pBluescript II SK(+) after digestion with EcoRI and XhoI for sequence analysis.
Genomic PCR Analysis for the Determination of Exon-Intron Structures.
Intron lengths were partially determined by PCR analysis using genomic DNA or cosmid DNA as templates and the Expand Long Template PCR System (Boehringer Mannheim) according to the manufacturer’s instructions. For the amplification of introns between coding exons, primers for SSCP analysis were used. For introns between 5' leader exons and exon 3, the following primers were used: (a) TS5U1, 5'-CAATTATGCTGTGTGGGCTC-3' (reverse primer, on exon 3); (b) TS5U2, 5'-TGACCATGGAAGACACAAGG-3' (forward primer, on exon 2); (c) TS5U3, 5'-GACGGTACATCAGTTTCCAG-3' (reverse primer, on exon 2); and (d) TS5U, 5'-AGGGACTGTGAGGTAAACAG-3' (forward primer, on exon 1). All amplified products were sequenced with the same primers for PCR, and exon-intron boundaries were confirmed.
PCR-SSCP Analysis.
Details of the oligonucleotide primer sets, sizes of the PCR products, MgCl2 concentrations, and PCR annealing temperatures for the Tsc1 gene are summarized in Table 1
. The primer sets for the Tsc2 and Vhl genes were described previously (28
, 29)
. PCR was carried out in a 10-µl volume including 50 ng of template genomic DNA, 1 mM each of the primers, 100 mM deoxynucleotide triphosphate, 0.5–1.25 mM MgCl2, 2.5 µCi of [
-32P]dCTP, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 0.1% Triton X-100, and 2 units of Taq DNA polymerase (Toyobo). Temperature conditions for PCR were the same as those for protocol I (RT-PCR analysis). After PCR, samples were diluted with the loading buffer (95% formamide, 20 mM EDTA, 0.05% xylene cyanol, and 0.05% bromphenol blue) and boiled for 5 min. A 1.5-µl aliquot of each diluted sample was run on a 6% polyacrylamide/10% glycerol gel at 4°C in 0.5x Tris-borate-EDTA buffer at 1400 V. Then gels were dried and autoradiographed with X-ray film. For sequencing, bands were dissected from the gels and extracted in 100 µl of distilled water, and fragments were reamplified by PCR with the same primers used for PCR-SSCP.
|
was used as a marker. PCR was performed for the PCR-SSCP analysis using a primer set for the latter half of exon 9. After 25, 30, or 35 amplification cycles, the PCR products were analyzed by 3% Agarose-1000 (Life Technologies) gel electrophoresis.
|
-32P]dCTP. |
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.2-kb fragment was successfully amplified with a primer set (HTSF2 and HTSR2) from rat testis mRNA (data not shown). Sequence analysis of the amplified portion of this fragment revealed a derivation from rat Tsc1 mRNA, showing homology with nt 1030–2229 of the human TSC1 cDNA sequence (data not shown). Subsequently, cosmid and phage clones containing rat Tsc1 genomic DNA regions were isolated by library screening, and their exonic sequences were determined (Fig. 1A)
. Also, a partial rat Tsc1 cDNA fragment corresponding to nt 739-1722 of human TSC1 cDNA was isolated from a kidney cDNA library. After the identification of sequences corresponding to exons 3 and 23 of the human TSC1 gene in a cosmid clone, the full coding region of rat Tsc1 cDNA was obtained by RT-PCR (Figs. 1A
and 2A
; GenBank/EMBL/DDBJ accession number AB011821). The 5'-RACE was also performed to analyze the 5'-untranslated region of Tsc1 mRNA (Figs. 1A
and 2B
; GenBank accession number AB016165).
|
|
). The rat exon 11 sequence (115 nt) was found to be 3 nt longer than that of the human sequence at its 3' end. Complimentarily, the rat exon 12 sequence (119 nt) was 3 nt shorter than that of the human sequence at its 5' end. The GT/AG rule was kept at all intronic splicing donor and acceptor sites (Fig. 1B)
. The deduced aa sequence of the rat Tsc1 product consisted of 1163 residues and showed 86% identity with that of human hamartin (Fig. 2A)
. A putative transmembrane region (aa 127–144) and a potential coiled-coil region (aa 719–998) were conserved between rat and human (Fig. 2A)
. In the COOH-terminal region encoded by exon 23, a 3-aa deletion and two single aa insertions were found in rat hamartin as compared with human hamartin (Fig. 2A)
. During RT-PCR analysis, we identified a cDNA fragment lacking nt 2042–2056 (15 bp) at the exon 16-exon 17 junction (data not shown). This 15-bp sequence was pyrimidine rich and ended with an AG dinucleotide (Fig. 1B)
. These results indicate that the second splicing acceptor site exists in the 5' part of exon 17. The use of this second acceptor site causes a 5-aa (Gly-Ser-Pro-Pro-Ser) deletion (Fig. 2A)
.
In man, TSC1 mRNA has a long 3'-untranslated region (4.5 kb; Ref. 19
). Although we have not yet fully accomplished sequence analysis of the 3'-untranslated region of rat Tsc1 cDNA, substantial homology of the sequenced region of exon 23 with the 3'-untranslated region of human TSC1 cDNA was identified (data not shown). In addition, the mRNA size (8 kb) of rat Tsc1 was similar to that of human TSC1 mRNA detected by Northern blot analysis (see below). These results indicate that rat Tsc1 mRNA also has a long 3'-untranslated region. The longest 5'-untranslated region, which was obtained by 5'-RACE, was 184 bp in length and showed an 84.5% sequence identity with that of human TSC1 cDNA (Fig. 2B)
. Thus, the 5'-untranslated region of Tsc1/TSC1 mRNA was also conserved.
Chromosomal Localization and Expression of the Rat Tsc1 Gene.
The chromosomal localization of the rat Tsc1 gene was determined by Southern blot analysis of a human/rat somatic cell hybrid panel. The appearance of rat-specific bands coincided with the presence of rat chromosome 3 in each cell clone (data not shown).
The expression of Tsc1 was examined in several rat tissues by Northern blot analysis. A 8-kb band similar in size to that of human TSC1 mRNA was detected in the brain, liver, spleen, kidney, and heart (Fig. 3)
.
|
With regard to the Tsc1 gene, we identified two mutations in 1 EHEN-induced RC (sample 9) as well as nine polymorphisms in 15 RCs (Fig. 4
; Table 2
). The two mutations were in the splicing donor site of intron 16, giving rise to a GT to GA transition and a GT to GC transversion. In exon 16, a polymorphic sequence at nt 2020 (G versus C) was found (Table 2)
, and the rat in which sample 9 RC developed was a heterozygote for this polymorphism. Autoradiography of SSCP of exon 16 showed several discrete shifted bands in sample 9 that were not seen in the normal tissue sample from the same rat or other RC samples (Fig. 3)
. When these shifted bands were sequenced, both mutations were found with C at nt 2020, whereas the normal band contains either C or G. Thus, the two mutations were generated on the same allele. In sample 9, no mutations were detected in either the Vhl or Tsc2 gene (see below). In intron 9, a polymorphic 20-bp deletion was identified in the LEC allele used here (Table 2)
. We used this polymorphism as a marker for LOH detection in applicable cases of RC from LEC/F344 F1 and F2 and LEC/WKAH F2 progeny. However, LOH was not detected in any of the RCs (data not shown).
|
. These results regarding the absence of mutations in Vhl and the presence of mutations in the Tsc2 gene in chemically induced RCs were in line with those from our previous report (11)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The high degree of structural similarity observed suggests functional conservation of hamartin in these species. The predicted coiled-coil region, in particular, appears to be highly conserved and thus may be functionally important, possibly for multiple protein-protein binding. The tuberin-binding site in human hamartin reported by Slegtenhorst et al. (20) contains 70 amino-terminal residues of this coiled-coil region, and the authors also mentioned homophilic interactions through this coiled-coil region. We have in fact identified a rat tuberin-binding site in a fragment containing the amino-terminal half of the coiled-coil region of rat hamartin,4 and the available information therefore indicates that it is a key element in the in vivo function of the two proteins.
In addition to higher homology of the coding region, rat Tsc1 also demonstrated other characteristics in common with its similar human counterpart. The 5'- and 3'-untranslated regions of mRNA are conserved and may be functionally important in the regulation of gene expression at the posttranscriptional level. Our Northern blot analysis showed 8-kb Tsc1 mRNA to be highly expressed in the brain as well as in the heart and kidney. The lower expression of Tsc1 found in the liver is similar to that in the human case (20)
. Analysis of the expression profile of Tsc1 mRNA and protein at the histological level should clarify the existence of posttranscriptional regulation of gene expression.
The localization of rat Tsc1 on chromosome 3 is reasonable because other homologues of genes on human chromosome 9q32–q34, such as the c-Abl proto-oncogene and the gene for adenylate kinase 1 (AK1), have also been mapped to rat chromosome 3 (30) .
As seen in human TSC, Tsc1 and Tsc2 deficiency might cause a similar phenotype in the rat and thus facilitate RC development. As one approach to investigate this possibility, we searched for Tsc1 mutations or deletions in chemically induced RCs; some of these RCs contained somatic Tsc2 mutations that we have identified previously (11) . The two Tsc1 splice donor site mutations found in sample 9 appeared to be on the same allele, as revealed by the presence of the same polymorphic sequence in their neighborhood. These splice donor site mutations may affect splicing. Tsc2 mutations were not detected in sample 9. Several possibilities could explain the occurrence of these double mutations. One possibility is that two (or more) independently arising RCs were included in sample 9, and Tsc1 mutations were formed separately in different RCs during the early stages of renal carcinogenesis. Advanced carcinomas are known to be monoclonal, but Novelli et al. (31) reported mixed karyotype adenomas in familial adenomatous polyposis and suggested cell to cell cooperation in adenoma formation before monoclonality due to a dominantly growing clone. Merritt et al. (32) also observed that intestinal adenomas in the Min mouse, an experimental model for familial adenomatous polyposis, have a polyclonal structure. The other possibility is that the Tsc1 mutations occurred separately within the same tumor after the initiation of RC development. In either case, the finding of Tsc1 mutations suggests the involvement of the Tsc1 alteration in RC development in the rat.
The mutations analyzed in this study and reported previously are summarized in Table 3
. The 15 chemically induced RCs showed a chromophilic type histology with a tubular or papillary pattern, not a clear cell type histology. They showed the same histopathological features, comparing the one with Tsc1 mutation versus others with Tsc2 mutations.
The frequency of Tsc2 mutations (8 of 15) was higher than that of Tsc1 mutations (1 of 15) in chemically induced RCs (P < 0.05). We are now trying to ascertain whether chemical renal carcinogenesis is possible in transgenic rats constructed with extra copies of the wild-type Tsc2 gene. This experiment may also provide evidence of the involvement of Tsc1 alterations in RC development in the rat.
In man, loss of the TSC1 locus has been detected in hamartomas from TSC patients, suggesting that two hits of TSC1 may cause their development (33) . A similar case is applicable for TSC2/Tsc2 in human TSC hamartomas and RCs in Eker rats or Tsc2 knockout mice.5 Thus far, however, we have not found a loss of the Tsc1 locus in sample 9 or in other RCs. We know that our SSCP analysis may have limitations. One possible reason, other than an insufficient sensitivity of SSCP analysis or a dominant negative effect on Tsc1 function, is that these tumors could contain mutations in other extra-exonic regions that are critical for Tsc1 gene expression or could have deletions in primer annealing sites. Another possible reason is a reduction of expression of the wild-type Tsc1 allele by some epigenetic mechanism, such as methylation of the promoter region. Detection of two hits of Tsc1 in RCs will strengthen the case for the involvement of Tsc1 alterations in renal carcinogenesis. Phenotypic analysis of Tsc1 knockout mice should also clarify this possibility and the in vivo interactions of hamartin and tuberin.
In summary, we have isolated and characterized the rat homologue of the human TSC1 gene. Structural analysis of rat Tsc1 rendered mutational analysis of RC samples possible, and we identified two mutations in one EHEN-induced RC by PCR-SSCP analysis. Biochemical studies using the Tsc1 cDNA cloned here as well as further mutational analysis of rat RCs should provide clues to the functions of hamartin and tuberin and the molecular mechanisms of tumor development associated with TSC1/Tsc1 or TSC2/Tsc2 mutations.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by Grants-in-Aid for Cancer Research from the Ministry of Education, Science, Sports and Culture of Japan and the Program for Promotion of Fundamental Studies in Health Sciences of the Organization for Pharmaceutical Safety and Research. 
2 To whom requests for reprints should be addressed, at the Department of Experimental Pathology, Cancer Institute, 1-37-1 Kami-ikebukuro, Toshima-ku, Tokyo 170-8455, Japan. Phone: 81-3-5394-3815; Fax: 81-3-5394-3815; E-mail: ohino{at}ims.u-tokyo.ac.jp
3 The abbreviations used are: RC, renal cell carcinoma; EHEN, N-ethyl-N-hydroxyethyl-nitrosamine; DEN, diethylnitrosamine; TSC, tuberous sclerosis; SSCP, single-strand conformational polymorphism; LOH, loss of heterozygosity; GAP, GTPase-activating protein; RT, reverse transcription; 5'-RACE, rapid amplification of the 5' ends of cDNA; nt, nucleotide(s); aa, amino acid(s).
4 S. Hasegawa and O. Hino, unpublished observations.
5 T. Kobayashi, O. Minowa, J. Kuno, H. Mitani, O. Hino, and T. Noda, Renal Carcinogenesis, hepatic hemangiomatosis and embryonic lethality caused by a germline Tsc2 mutation in mice, submitted for publication.
Received 9/21/98. Accepted 12/18/98.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-fetoprotein genes and assignment to chromosome 14. Cytogenet. Cell Genet., 38: 142-149, 1984.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. Okimoto, J. Sakurai, T. Kobayashi, H. Mitani, Y. Hirayama, M. L. Nickerson, M. B. Warren, B. Zbar, L. S. Schmidt, and O. Hino A germ-line insertion in the Birt-Hogg-Dube (BHD) gene gives rise to the Nihon rat model of inherited renal cancer PNAS, February 17, 2004; 101(7): 2023 - 2027. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Hodges, S. Li, J. Maynard, L. Parry, R. Braverman, J. P. Cheadle, J. E. DeClue, and J. R. Sampson Pathological mutations in TSC1 and TSC2 disrupt the interaction between hamartin and tuberin Hum. Mol. Genet., December 1, 2001; 10(25): 2899 - 2905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kobayashi, O. Minowa, Y. Sugitani, S. Takai, H. Mitani, E. Kobayashi, T. Noda, and O. Hino A germ-line Tsc1 mutation causes tumor development and embryonic lethality that are similar, but not identical to, those caused by Tsc2 mutation in mice PNAS, June 28, 2001; (2001) 151033798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Murthy, L. A. Haddad, N. Smith, D. Pinney, R. Tyszkowski, D. Brown, and V. Ramesh Similarities and differences in the subcellular localization of hamartin and tuberin in the kidney Am J Physiol Renal Physiol, May 1, 2000; 278(5): F737 - F746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kobayashi, O. Minowa, Y. Sugitani, S. Takai, H. Mitani, E. Kobayashi, T. Noda, and O. Hino A germ-line Tsc1 mutation causes tumor development and embryonic lethality that are similar, but not identical to, those caused by Tsc2 mutation in mice PNAS, July 17, 2001; 98(15): 8762 - 8767. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
1a1) are indicated. Both ends of 
