| Human Molecular Genetics | Pages |
A tetranucleotide polymorphic microsatellite, located in the first intron of the tyrosine hydroxylase gene, acts as a transcription regulatory element in vitro
Introduction
Results
Effects of T10i and T10p on transcriptional activity
Tetrarepeated sequence binding activity
TRE competition experiments
Discussion
Materials And Methods
Plasmid construction
Cell cultures and DNA transfection
Luciferase and chloramphenicol acetyltransferase assays
Preparation of nuclear extracts and EMSA
Abbreviations
Acknowledgements
References
A tetranucleotide polymorphic microsatellite, located in the first intron of the tyrosine hydroxylase gene, acts as a transcription regulatory element in vitro
The polymorphic HUMTH01 microsatellite, located in the first intron of the tyrosine hydroxylase gene is characterized by a tetranucleotide core motif. The 10 repeat allele of this microsatellite exhibits two sequence variants: an imperfect repeat and a perfect repeat. Here we present evidence that this tetrarepeat is endowed with regulatory properties. Constructions were made linking the 10 repetition alleles to the luciferase reporter gene under the control of a thymidine kinase minimal promoter. In transient transfection experiments in HeLa, PC12 and SK-NSH cell lines these repeated sequences increased the basal transcription up to 9-fold. This effect was independent of the sequence orientation, a feature characteristic of an enhancer element. In electrophoretic mobility shift assays these tetrameric repeated sequences form specific complexes with HeLa cell nuclear extracts. Competition experiments with heterologous sequences suggest that proteins of the Fos-Jun family may be involved in the formation of these complexes, although other unidentified transacting factors bind to these sequences. These results thus implicate the HUMTH01 microsatellite in the regulation of tyrosine hydroxylase gene expression. Tetrarepeated sequences of this type may constitute a new class of regulatory elements.
INTRODUCTION
Microsatellites (1) are highly polymorphic tandemly repeated sequences, usually constituted by di-, tri- and tetranucleotide motifs. They are distributed throughout the genome in eukaryotes (2). Their considerable polymorphism is due not only to differences in the number of repeat units but also in some instances to variations in the sequence of the core motif (3), characteristics that make them useful markers for genetic analysis (4,5). Microsatellites, according to the `neutral hypothesis' of their origin, do not have a generalised function (5), although it appears that in many cases they are under evolutionary control (6) and may exert some genetic effect (7). Dinucleotide and trinucleotide repeats have been shown to regulate transcription in vitro (8,9). The best existing evidence for a functional role of microsatellites comes from genetic studies implicating unstable trinucleotide repeats as causative elements of several neurological diseases (10). In some instances the expansion of these repeats acts through the inhibition of the transcription of the gene involved in the underlying pathology, such as the FRAXA gene or the DMAHP gene, implicated in fragile X syndrome and myotonic dystrophy, respectively (11-13). Minisatellites, the other main class of repeated sequences used as genetic markers (14), may also be endowed with regulatory functions. For example, the minisatellite located 1 kb downstream from the HRAS1 proto-oncogene and the minisatellite 5[prime] to the insulin gene [associated with increased risk of cancer (15) and insulin dependent diabetes mellitus (IDDM) (16), respectively] display transcriptional regulatory activity in vitro (17-19). Thus, the capacity to behave as transcription regulatory elements appears to be a general feature of all classes of repeated sequences (7). However, no role has yet been demonstrated for microsatellites with tetranucleotide motifs.
We investigated whether a tetrarepeat microsatellite motif has any functional role. The study is based on the HUMTH01 microsatellite which is a (TCAT)n motif localised in the first intron of the gene encoding tyrosine hydroxylase (TH) (20). TH, the rate limiting enzyme in the synthesis of catecholamines, is pivotal in neurotransmission and is thus a candidate gene for neuropsychiatric diseases. TH activity is regulated by a wide array of mechanisms involving short-term regulation acting through activation or inhibition of the enzyme as well as long-term regulation of TH gene expression that could be determinant for normal and pathological states (21).
The HUMTH01 microsatellite alleles are generated by 5-10 repetitions of the core motif TCAT. The alleles containing 5-9 repeats are iterations of the core motif TCAT without variation. Interestingly, the 10 repeat allele exhibits two sequence variants: a (TCAT)4CAT(TCAT)5 imperfect repeat (allele T10i) and a perfect repeat (TCAT)10 (allele T10p). T10i is the most frequent allele in the Caucasian general population whereas T10p is very rare (>30% and ~1%, respectively) (22,23). In a study in two independent populations, we have previously shown an association of the T10p allele with genetic predisposition to schizophrenia (24). This initial observation led us to focus our study on the 10 repeat in its perfect and imperfect forms. We used transient transfection and electrophoretic mobility shift assay (EMSA) to show that these tetranucleotide repeated sequences exhibit the properties of a regulatory motif and may thus contribute to TH gene expression. By extension, our findings suggest that microsatellites with tetranucleotide repeated sequences of this type may constitute a new class of transcriptional regulatory elements.
RESULTS
Effects of T10i and T10p on transcriptional activity
We first tested the ability of the T10i and T10p sequences to modulate the transcriptional activity of a heterologous promoter. These two alleles were placed in both orientations upstream from the herpes simplex virus thymidine kinase promoter using the firefly luciferase as the reporter gene. Expression of these constructs was analysed by transient transfection experiments in HeLa cells and in the neural cell lines PC12 and SK-NSH. PC12 cells are derived from rat pheochromocytomaand the SK-NSH cells from human neuroblastoma tumours.
The presence of the T10p sequence increased expression from the promoter by 9-fold in HeLa cells as shown (Table 1). A similar effect (an 8-fold increase) was observed with the T10i sequence. We tested whether these increases were orientation dependent. Significant increases, by ~5-6-fold the activity of the promoter, were obtained for both perfect and imperfect tetrarepeat sequences in their antisense orientation (Table 1). Similar effects on basal transcription were also obtained in the two neural cell lines (Table 1). This first series of experiments indicate that the T10i and T10p tetrarepeat sequences act as enhancer elements.
Figure
We then tested whether nuclear factors bound to T10i and T10p in 40mer oligonucleotide sequence. These experiments were performed by EMSA using HeLa cell nuclear protein extracts (Fig. 1). The two 40mer probes each gave two major discrete bands (denoted A and B) suggesting that at least two complexes were formed. The specificity of these complexes was ascertained by competition with unlabelled homologous and heterologous sequences. These experiments were performed by cross competition using molar excesses of imperfect and perfect 40mer oligonucleotides against the perfect and imperfect probe, respectively. Molar excesses of homologous competitor displaced both complexes (Fig. 1), whereas the same molar excesses of heterologous competitor in the form of a tetrarepeat (CTGC)10 of equal length was unable to displace the complexes (Fig. 2). These findings demonstrate that both complexes result from the specific binding of nuclear factors (Fig. 1). The perfect and imperfect sequences gave similar results in these competition experiments, without obvious differences.
Table 1.
Figure
To further characterise the complexes formed between the T10i and T10p sequences and nuclear proteins, a second set of experiments was conducted with 5-unit tetrarepeat sequences. These probes were 20mer oligonucleotides with the sequence (TCAT)2CAT(TCAT)2T (the T5i probe), and (TCAT)5 (the T5p probe). The T5p probe corresponds to the shortest HUMTH01 allele described and may thus represent the minimal number of repeats needed for protein binding. The T5i sequence was used to test for discrimination in the binding between a perfect and imperfect tetrarepeat sequence. The 20mer probes, both perfect and imperfect, generate a single band shift signal. This signal was specifically displaced by molar excesses of unlabelled competitor (Fig. 3). As shown in Figure 3 the cross competition of these complexes reveals a difference of efficacy between the perfect and imperfect oligonucleotides in the binding of nuclear factors. Densitometric analysis of the gel presented in Figure 3 confirms the presence of this difference. The T5p competitor more efficiently displaced both the T5i or T5p probe than the T5i competitor (Fig. 4). This reveals differences between the binding of the two probes and suggests that the T5p repeat has a higher binding affinity than the T5i to HeLa cell nuclear extracts.
Figure
Figure
Tetrarepeated sequence binding activity
Cell lines
TK-Luc activity
Induction
RLU/mU CAT
T10i
T10p
sense
antisense
sense
antisense
HeLa
482 ± 40
7.6 ± 1.4
4.7 ± 0.9
9.0 ± 1.7
6.1 ± 0.7
PC12
41844 ± 1701
4.6 ± 0.6
1.0 ± 4.0
4.0 ± 0.1
2.3 ± 0.2
SK-NSH
1121 ± 309
4.8 ± 1.0
1.5 ± 0.4
4.5 ± 0.3
2.0 ± 0.2
TRE competition experiments
The composition of the complexes seen with EMSA was further investigated by competition with the TPA responsive element (TRE) sequence. This sequence was selected because the core motif of the HUMTH01 microsatellite (TCAT)n is analogous to the TGACTCA motif that characterises the TRE recognised by the Fos-Jun complex (25,26). Displacement by the TRE consensus sequence as well as the TRE sequence (TGATTCA) present in rat (27) and human TH (TRE-TH) proximal promoter region was tested. The perfect and imperfect 20- and 40mer probes were competed with molar excesses of a 25mer oligonucleotide containing the TRE-TH motif and its flanking sequence in the human TH gene promoter. The TRE-TH sequence was unable to compete for the formation of the complex of nuclear proteins with either the perfect or imperfect 20mer probes (Fig. 5). Identical results were obtained with the consensus TRE (data not shown). Interestingly, when the same competitor was tested against the 40mer perfect and imperfect probes, which yield two major complexes (A and B), it fully abolished the signal associated with the A but not the B complex (Fig. 6). These results suggest that the AP1 complex could interact with the tetrameric repeated sequence, whereas proteins other than Fos and Jun bind to the HUMTH01 alleles.
Figure
Figure
DISCUSSION
This study establishes that stretches of a tetranucleotide microsatellite (TCAT)n exhibit the characteristic features of a transcriptional enhancer element. Tetranucleotide repeats with this type of sequence, like other repeated sequences, may therefore be endowed with regulatory functions in gene expression throughout the genome.
We studied the sequence of the HUMTH01 microsatellite marker, found in the first intron of the TH gene. Using transient transfection studies in three different cells lines we investigated a perfect and imperfect variant of the 10 repetition allele of the HUMTH01 microsatellite. These elements, in all situations, led to a substantial increase of basal transcription. Moreover, as expected for a regulatory element, this effect was not dependent on the orientation of the sequences.
In a second set of experiments, we showed that stretches of this tetrarepeat bind in a sequence specific manner to nuclear proteins. Two major complexes were detected. The HUMTH01 microsatellite core motif (TCAT)n is very similar to the TRE canonical consensus sequence (TGACTCA) (25,26) and differs by only one nucleotide from the consensus TRE sequence (TGATTCA) present in the rat and human TH gene (27). We therefore tested competition between the 10 repetition perfect and imperfect alleles and the TRE consensus sequences. One of the complexes was abolished in a dose dependent fashion by competition with a canonical TRE consensus sequence as well as the TRE consensus sequence present in the rat and human TH, suggesting that this complex is formed by members of the Fos-Jun family. Supporting this view, preliminary experiments with antibodies directed against the Jun protein indicate that Jun is involved in the complex (not shown). The TRE sequence, TGATTCA, within the proximal region of the promoter of the human and rat TH genes has been shown to be relevant for the expression of the rat TH (27), which does not, however, contain a tetrarepeat sequence. It is not known whether in humans both sites are functional and whether there is any interaction between the AP1 site in the proximal promoter region and the tetrarepeat sequence in the first intron of the TH gene. The TRE consensus sequence did not compete with the formation of the second complex. This suggests that factors other than those of the Fos-Jun family are able to bind the HUMTH01 microsatellite motif. The proteins responsible for the formation of this second complex may represent a novel class of unidentified regulatory factors. Although it remains to be established whether the HUMTH01 microsatellite exhibits transcription enhancer activity in vivo, our results suggest that this tetranucleotide repeat is a novel regulatory sequence whose action may be relevant to gene expression. Thus, it would be of interest to further study the role of the HUMTH01 microsatellite at its orthologous position in the context of the TH gene. Several different types of constructs containing the TH gene sequence extending from the 5[prime] regulatory region and encompassing the first intron with the repeated sequence are needed to characterise the effect of the different HUMTH01 microsatellite alleles.
Because of its localisation within the TH gene, a candidate gene for neuropsychiatric diseases, the HUMTH01 microsatellite has been widely used in genetic studies of bipolar illness several of which have yielded positive results (28-31). In a recent study, we found that the perfect rare allele T10 of the HUMTH01 microsatellite is significantly associated with schizophrenia in two different populations (24). However, it remains to be ascertained whether the alleles of this microsatellite or another polymorphism in linkage disequilibrium with it determines the genetic predisposition to these diseases. Most interestingly, a follow up study of a subgroup of the schizophrenic patients revealed that the plasma concentrations of catecholamine metabolites were significantly lower in the group bearing the perfect allele than in schizophrenic patients carrying the other alleles (32). In the present study we detected no significant differences between the enhancing effects of the perfect and imperfect repeated sequences. However, gel migration retardation experiments showed differences in binding between the perfect and imperfect repeat sequences, supporting the notion that allelic variations based on a single nucleotide mutation may have different effects in vivo. These results further implicate the polymorphic tetrarepeat sequence in the regulation of the TH gene.
The TH gene is closely linked to the insulin gene (33,34). Studies on the genetics of insulin dependent diabetes mellitus (IDDM) show the association of a minisatellite (Ins-VNTR) located 5[prime] to the insulin gene to the disease (16). The polymorphism causing IDDM has not been definitively identified, although several studies show an association between the HUMTH01 microsatellite and diabetes (4) (J.Hors, personal communication) or between haplotypes extending to the TH locus and diabetes (35). The Ins-VNTR can regulate transcription in vitro and thus it has been suggested that it can act on the expression of the insulin gene (18,19) as well as on the expression of other closely linked genes, such as the TH gene (18). Conversely, our results indicating that the TH tetrarepeat acts as a transcription regulator raise the possibility that the HUMTH01 microsatellite could exert some activity on the expression of nearby genes such as the insulin gene.
Comparative analyses suggest that in some instances microsatellites are under evolutionary control and this is consistent with their participation in gene regulation. Several lines of evidence show that they are preserved at orthologous positions in the genome of different species (36-39) as assessed, for example, comparing the human genome to the genome of rat (40) or non-human primates (41). Moreover, the genetic distance between humans and chimpanzees calculated using dinucleotide microsatellites is 9-fold smaller to that expected assuming that there is no selection for these sequences (41). Similarly up to a maximum of eight repeats of the core motif of the HUMTH01 microsatellite and its flanking region are highly conserved in the first intron of the TH gene of several non-human primate genera (42), suggesting that evolutionary constraints may act upon this sequence. Thus, our results imply that the presence in the human population of a perfect and imperfect variant of a 10 tetranucleotide repeated sequence showing transcriptional regulatory activity is not due to genetic drift, but may be relevant to the expression of normal and/or pathological genetic traits.
MATERIALS AND METHODS
Plasmid construction
TK-Luc (gift from Dr Hugues de Th) contains a thymidine kinase promoter fragment (-109 to +51) fused to the firefly luciferase reporter gene. The perfect (T10P) and imperfect repetition (T10I) were inserted into its SalI site. The oligonucleotide sequences used were as follows : T10P 5[prime]-(TCAT)10-3[prime], T10I 5[prime]-(TCAT)4 CAT(TCAT)5T-3[prime]. The constructs were sequenced on both strands by the dideoxy method of Sanger in the presence of a [35S]dATP following the guidelines for the Sequenase 2.0 kit (US Biochemical Corp.). The plasmids were prepared by a double caesium chloride gradient centrifugation.
Cell cultures and DNA transfection
HeLa cells were grown in Dulbecco's modified eagle medium supplemented with 10% foetal calf serum, 5% antibiotics and 100 mM sodium pyruvate. For transient assay experiments, 106 cells were re-suspended in 0.15 ml of serum free DMEM and transfected by electroporation with 4 µg of the test DNA, 10 µg of carrier DNA and 1 µg of SV40-CAT plasmid (pCAT3 control vector; Promega) to assess transfection efficiency in a total volume of 160 µl. Electroporations were performed using a Biorad gene pulser at 190 V, 960 µF for 50 ms. After electroporation cells were placed in serum-containing medium.
Luciferase and chloramphenicol acetyltransferase assays
HeLa cells were harvested in 2 ml of phosphate buffered saline and resuspended in 200 µl of lysate buffer (25 mM Tris-phosphate pH 7.8, 8 mM MgCl2, 1 mM dithiotreitol, 1 mM EDTA, 1% Triton, 15% glycerol, 1% bovine serum albumin). Cell debris was removed by centrifugation. Luciferase assays, in 150 µl of reaction mixture (0.08 mM luciferin, 0.1 mM ATP, 25 mM Tris-phosphate pH 7.8, 8 mM MgCl2, 1 mM dithiotreitol, 1 mM EDTA, 1% Triton, 15% glycerol), were carried out using a Lumat LB9501 (Berthold) luminometer. Preliminary studies indicated that the luciferase activity of cell lysates on transfection was in the linear range of the assay. The luciferase activity corresponding to each construction was normalised to that of the co-transfected SV-CAT vector. The amount of the cell lysate used was always in the linear range of the CAT assay as assessed using a reference CAT preparation. CAT assays were performed using the liquid scintillation counting method.
Preparation of nuclear extracts and EMSA
Nuclear extracts were prepared according to the procedure of Dignam (43). Protein concentrations in these extracts were determined using the Bradford method (44) and were between 1 and 2 mg/ml. The antisense oligonucleotides T10p and T10i were labelled at the 5[prime]-end by incubation with [[gamma]-32P]ATP and T4 polynucleotide kinase. Each of these labelled oligonucleotides was annealed with its cold complementary oligonucleotide. The binding reactions were performed at 20°C for 15 min with 1-2 fmol of labelled DNA in 0.21 mM MgCl2, 25% glycerol, 11.4 mM Hepes pH 7.8, 75.7 mM KCl, 0.86 mM DTT, 2.28 mM Tris, 2.28 mM Triton X100 and 0.74 mM EDTA. Unlabelled double-stranded oligonucleotides at 10-, 50- or 200-fold molar excess were used for competition experiments. DNA/protein complexes were separated by electrophoresis on a 6% polyacrylamide gel in 0.5× Tris-borate/EDTA at 10 V/cm. After drying, the gels were autoradiographed overnight at -80°C with MP film.
ABBREVIATIONS
AP1, activating protein 1; EMSA, electrophoretic mobility shift assay; Ins, insulin; IDDM, insulin dependent diabetes mellitus; TH, tyrosine hydroxylase; TPA, 12-O-tetradecanoylphorbol-13-acetate; TRE, TPA responsive element; VNTR, variable number of tandem repeats.
ACKNOWLEDGEMENTS
We thank Drs Nicole Faucon-Biguet, Leila Houhou, Marika Nosten-Bertrand and Guilane Vodjdani for critical reading of the manuscript and Dr Françoise Treilhou-Lahille for providing advice and support for cells culture. This work was supported by the Centre National pour la Recherche Scientifique, Institut National de la Santé et de la Recherche Medicale, Association Française contre les Myopathies and Rhône-Poulenc Rorer. V.A. and P.R. are supported by the Ministère de la Recherche Scientifique and Institut de Recherche sur la Moelle Epinière, respectively.
REFERENCES
*To whom correspondence should be addressed. Tel: +33 1 42 17 75 30; Fax: +33 1 42 17 75 33; Email: mallet@infobiogen.fr
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 14 Feb 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Kvetnansky, E. L. Sabban, and M. Palkovits Catecholaminergic Systems in Stress: Structural and Molecular Genetic Approaches Physiol Rev, April 1, 2009; 89(2): 535 - 606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Hunt Tyrosine Hydroxylase: Another Piece of the Genetics of Hypertension Puzzle Circulation, August 28, 2007; 116(9): 970 - 972. [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. Mudunuri and H. A. Nagarajaram IMEx: Imperfect Microsatellite Extractor Bioinformatics, May 15, 2007; 23(10): 1181 - 1187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Boeva, M. Regnier, D. Papatsenko, and V. Makeev Short fuzzy tandem repeats in genomic sequences, identification, and possible role in regulation of gene expression Bioinformatics, March 15, 2006; 22(6): 676 - 684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kiefer, F. Chatail-Hermitte, P. Ravassard, E. Bayard, I. Brunet, and J. Mallet ZENON, a Novel POZ Kruppel-Like DNA Binding Protein Associated with Differentiation and/or Survival of Late Postmitotic Neurons Mol. Cell. Biol., March 1, 2005; 25(5): 1713 - 1729. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhang, F. Rao, J. Wessel, B. P. Kennedy, B. K. Rana, L. Taupenot, E. O. Lillie, M. Cockburn, N. J. Schork, M. G. Ziegler, et al. Functional allelic heterogeneity and pleiotropy of a repeat polymorphism in tyrosine hydroxylase: prediction of catecholamines and response to stress in twins Physiol Genomics, November 17, 2004; 19(3): 277 - 291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-C. Li, A. B. Korol, T. Fahima, and E. Nevo Microsatellites Within Genes: Structure, Function, and Evolution Mol. Biol. Evol., June 1, 2004; 21(6): 991 - 1007. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Albanese, N. F. Biguet, H. Kiefer, E. Bayard, J. Mallet, and R. Meloni Quantitative effects on gene silencing by allelic variation at a tetranucleotide microsatellite Hum. Mol. Genet., August 1, 2001; 10(17): 1785 - 1792. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Croager, A. M. Gout, and L. J. Abraham Involvement of Sp1 and Microsatellite Repressor Sequences in the Transcriptional Control of the Human CD30 Gene Am. J. Pathol., May 1, 2000; 156(5): 1723 - 1731. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Godbout and E. A. Monckton Differential Regulation of the Aldehyde Dehydrogenase 1 Gene in Embryonic Chick Retina and Liver J. Biol. Chem., August 24, 2001; 276(35): 32896 - 32904. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||