Structural and functional analysis of metal regulatory elements in the promoter region of genes encoding metallothionein isoforms in the Antarctic fish Chionodracohamatus (icefish)

Gene 274 (2001) 199–208 www.elsevier.com/locate/gene

Structural and functional analysis of metal regulatory elements in the promoter region of genes encoding metallothionein isoforms in the Antarctic fish Chionodraco hamatus (icefish) Rosaria Scudiero a, Vincenzo Carginale b, Clemente Capasso b, Marilisa Riggio a, Stefania Filosa c, Elio Parisi b,* a

Dipartimento di Biologia Evolutiva e Comparata, Universita` Federico II, Napoli, Italy b Istituto di Biochimica delle Proteine ed Enzimologia del CNR, Napoli, Italy c Istituto Internazionale di Genetica e Biofisica del CNR, Napoli, Italy

Received 28 February 2001; received in revised form 2 July 2001; accepted 11 July 2001 Received by M. D’Urso

Abstract To investigate the regulation of Chionodraco hamatus metallothionein (MT) encoding genes about 1000-bp regions of both MT-I and MTII gene promoters were cloned and sequenced. Both promoters were rich in A–T content, and lacked the canonical TATA box; several putative cis-regulatory sequences were also present. In the MT-I promoter, four MREs were identified within the first 300 bp from the ATG codon. In the MT-II promoter, seven MREs were organized into two clusters, one containing three MREs located close to the ATG codon, and the other consisting of four MREs lying 500–900 bp upstream of the transcription starting point. The alignment of the MT-I and MT-II promoter regions showed 57% identity, which increased to 87% in the 300-bp region upstream of the ATG. Only the three proximal putative MREs identified were conserved both in position and sequence. Functional analysis of MT-I and MT-II promoters was performed by introducing deletion mutants of the 5 0 -flanking regions into vector pGL-3, directly upstream of the firefly luciferase reporter gene. Each construct was tested in the HepG2 cell lines in the absence or presence of zinc or cadmium ions. Maximum inducibility of the MT-II gene promoter was achieved with a construct containing both the proximal and the distal MRE clusters. The lack of the most distally located MRE dramatically affected MT-II promoter sensitivity to metals; removal of the distal cluster of MREs also reduced metal inducibility. The MT-I promoter was more compact, since maximal activity and metal inducibility depended on the presence of the proximal cluster of four MREs. This study suggests that the different organization of the MT-I and MT-II gene promoter regions might account for the observed differences in the basal and metal-induced expression of MT-I and MT-II isoforms in the C. hamatus liver. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Cadmium; Gene regulation; Luciferase reporter gene; Metal-responsive elements (MREs); Transfection; Zinc

1. Introduction Metallothioneins (MTs) are small-molecular-mass nonenzyme proteins that act as biological chelators of heavy metals through the formation of metal-thiolate bonds by Abbreviations: AP-1, activator protein 1; bp, base pairs; b-gal, b-galactosidase; cDNA, DNA complementary to RNA; dNTP, deoxyribonucleoside triphosphate; HepG2, human hepatoblastoma cells; MRE, metal responsive element(s); MT, metallothionein(s); ONPG, O-nitro-phenylgalactoside; PCR, polymerase chain reaction; pfu, plaque-forming unit(s); Sp1, transcription factor 1; u, unit(s); UTR, untranslated region(s) * Corresponding author. Istituto di Biochimica delle Proteine ed Enzimologia, via Marconi, 10, 80125 Napoli, Italy. Tel.: 139-81-725-7323; fax: 139-81-239-6525. E-mail address: [email protected] (E. Parisi).

their numerous cysteine residues (Ka¨gi and Schaffer, 1988). MTs are functionally flexible proteins with multifunctional roles including homeostasis of trace elements, detoxification of poisonous heavy metals and scavenging of superoxide radicals (Palmiter, 1998; Nordberg and Nordberg, 2000). MTs isolated from several eukaryotic species show a strong degree of evolutionary conservation (Binz and Ka¨gi, 1999; Bargelloni et al., 1999). Metallothionein genes are organized as multigene families displaying developmentally regulated expression patterns, as well as cell-type-specific regulation (Jahroudi et al., 1990; Freedman et al., 1993; Moilanen et al., 1999). Expression of MT is controlled mainly at the transcriptional level by several agents, including metals, hormones, UV irradiation and free radicals (Imbert et al., 1990; Thiele, 1992; Andrews, 2000). Metals

0378-1119/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0378-111 9(01)00609-6

200

R. Scudiero et al. / Gene 274 (2001) 199–208

are the most general and powerful of these inducers (Durnam and Palmiter, 1981). Analysis of metallothionein gene promoters has led to the identification of short cis-acting elements, termed metal responsive elements (MREs), present in the 5 0 -flanking region of MT gene(s) (Varshney et al., 1986; Searle et al., 1987; Culotta and Hamer, 1989). MREs have been shown to mediate transcriptional response of MT genes to zinc and cadmium through trans-acting factors interacting with the MREs (Seguin and Prevost, 1988; Brugnera et al., 1994; Koizumi et al., 1999). In studies of human MT genes, several additional cis-acting elements have been identified, which are involved in the regulation of basal and induced expression of MT genes (Friedman and Stark, 1985; Samson and Gedamu, 1998; Andrews, 2000). Evidence of metal-specific post-transcriptional regulation has also been reported (Sadhu and Gedamu, 1989; Vasconcelos et al., 1996; Scudiero et al., 1997). In spite of the ubiquity of MTs in the animal kingdom, there are wide differences in the number and complexity of the MT genes observed in different organisms. Human MTs are encoded by a multigene family of at least 12 members, with numerous non-functional and processed pseudogenes and a number of functional isoforms (Karin and Richards, 1984; Walker and Gedamu, 1990; Miles et al., 2000). In fish, the MT system is not so complex, consisting of one or two known genes (Zafarullah et al., 1988). The rainbow trout (Oncorhyncus mykiss) expresses two MT genes, which are differentially regulated depending on the heavy metal and the developmental stage (Olsson et al., 1990; Zafarullah et al., 1990). We have recently described the presence of two distinct MT cDNA isoforms, namely MT-I and MT-II, in several notothenioid fish species endemic to the waters surrounding Antarctica (Carginale et al., 1998; Bargelloni et al., 1999). MT-I and MT-II gene isoforms are differentially expressed in red-blooded and haemoglobinless Antarctic fish. In the Antarctic haemoglobinless fish Chionodraco hamatus (icefish), MT-I and MT-II transcripts are differentially regulated by heavy metals: whilst the MT-II transcript is expressed constitutively, the MT-I transcript preferentially accumulates in the C. hamatus liver in response to cadmium exposure (Carginale et al., 1998). In order to elucidate the regulatory mechanisms involved in MT gene expression of notothenioids, we investigated the C. hamatus MT-I and MT-II promoter regions. In the present paper, we report the results of structural and functional analyses of the 5 0 -flanking regions of C. hamatus MT gene isoforms.

2. Materials and methods

by Dr L. Carratu`. All the enzymes were from Boehringer Mannheim and New England Biolabs Inc. Antibiotics and other chemicals were from Sigma; nylon papers and deoxynucleoside [a- 32P]- and [a- 35S]-triphosphates were from Amersham Pharmacia Biotech. 2.2. Library screening The genomic library was screened according to the method described by Benton and Davis (1977). It was plated at a density of 60,000 pfu per plate on a bacterial lawn of E. coli XL1-Blue, and screened using 32P-labelled C. hamatus MT-I cDNA coding sequence (Carginale et al., 1998) as a probe under high-stringency conditions. 2.3. Restriction mapping Phage DNAs containing genomic inserts hybridizable to fish MT cDNA were mapped by using restriction endonucleases. MT cDNA-containing sequences were localized by performing Southern transfers of the restriction fragments followed by hybridization under the conditions described above. 2.4. PCR amplification, subcloning and nucleotide sequencing of the 5 0 -flanking region of C. hamatus MT genes PCR was used to obtain the 5 0 -flanking region of the MT genes from phage DNAs containing genomic inserts corresponding to the MT genes. The 27-mer ICE-N-ter (5 0 GCTAAGCTTGGGTCCATGTTCTCAGGT-3 0 ), complementary to the N-terminal sequence of C. hamatus MT(s), was used as specific primer; the 27-mer T7 (5 0 -GCTGGTACCTAATACGACTCACTATAG-3 0 ) and the 29mer T3 (5 0 -GCTGGTACCATTAACCCTCACTAAAGGGA-3 0 ), complementary to the bacteriophage T7 and T3 promoters adjacent to the right and left lDASH polycloning sites, respectively, were used as non-specific primers. PCR reaction mixture (50 ml) consisted of 1 ml phage DNA and 49 ml PCR cocktail (Expand High Fidelity PCR System, Boehringer Mannheim), containing 50 pmol of the T7 or T3 primer, and the ICE-N-ter specific primer, 0.2 mM dNTPs, and 2.5 units of Expand enzyme mix. Twenty-five cycles of the first-strand PCR amplification were performed (temperature profile, 30 s at 948C, 30 s at 508C, and 4 min at 688C) in a DNA thermal cycler (Hybaid PCR Express). The PCR amplification products were purified and directly cloned into the pCRII-TOPO vector (Invitrogen) using the TOPO TA cloning kit (Invitrogen). Double-stranded sequencing was performed using the T7 sequencing kit (Amersham Pharmacia Biotech). 2.5. Construction of MT-luciferase hybrid plasmids

2.1. Library and reagents The Chionodraco hamatus genomic library, contained in bacteriophage lDASH II (Stratagene), was kindly provided

PCR was used to construct the desired deletion mutants after sequencing the 5 0 -flanking region of the C. hamatus MT genes and identifying the putative cis-acting elements.

R. Scudiero et al. / Gene 274 (2001) 199–208

201

Fig. 1. Oligonucleotides used for creating deletion mutants of the C. hamatus MT-I (A) and MT-II (B) gene promoters. The 3 0 ICE N-ter primer contains a Hind III site; the remaining 5 0 primers contain a Kpn I site. Hind III and Kpn I sites are underlined.

The oligonucleotides used are shown in Fig. 1. Reaction mixtures and conditions for PCR were as described above. The PCR products were resolved on agarose gel, purified and directly cloned into the pCRII-TOPO vector as previously described. The nucleotide sequence of each fragment was confirmed utilizing universal reverse and forward primers. Each deletion mutant was subcloned into the pGL3-Basic vector (Promega) using Kpn I and Hind III restriction sites. Cloning in the desired orientation was confirmed by DNA sequence analysis using the GLprimer2 (Promega), which is complementary to the N-terminal sequence of the luciferase gene. 2.6. Cell culture and transfections HepG2 (ATCC HB 8065, human hepatoblastoma) cells were grown at 378C in Dulbecco’s modified Eagle’s medium with Earle salts supplemented with 10% foetal bovine serum, 2 mM l-glutamine, penicillin (100 U/ml), streptomycin (100 mg/ml), in an atmosphere of 5% CO2. Transfections were carried out by electroporation. Briefly, cells were replated at approximately 70% confluence 24 h before transfection. They were then trypsinized and resuspended in growth medium at approximately 0.5–1 £ 10 7 cell/ml density. Amounts of 60 mg hybrid pGL-3 plasmid

and 20 mg pSV-b-galactosidase plasmid (Promega) were mixed with 0.5 ml of cells in a Gene pulser cuvette (0.4 cm) and electroporated at 250 V and 960 mF using a gene Pulser unit (Bio-Rad). Cells were allowed to recover for 24 h at 378C in 100-mm plates. Fresh complete medium containing 5.0 mM Cd or 150 mM Zn was then added to the cells, and induction was allowed to proceed for 24 h. 2.7. Reporter gene assay Cells were harvested using 900 ml of reporter lysis buffer (Promega) and centrifuged for 2 min at 48C to pellet cellular debris. Protein concentration was determined by the Bradford assay (Bio-Rad). The firefly luciferase activity was quantified from photon release upon oxidation of the substrate beetle luciferin in the presence of ATP, using a liquid scintillation counter (TRI-CARB 2100TR, Packard Bioscience Company). The assay was carried out using the luciferase assay kit (Promega) according to the manufacturer’s instructions. Transfection efficiency was measured by co-transfection of the b-galactosidase control vector. The b-gal activity was measured by incubating an aliquot of the protein extract with 1.3 mg/ml O-nitrophenylgalactoside (ONPG) in a buffer containing 0.1 M NaPO4, pH 7.3, 1 mM MgCl2 and 50 mM b-mercaptoetha-

202

R. Scudiero et al. / Gene 274 (2001) 199–208

nol for 30 min at 378C. The reaction product was measured by reading the absorbance at 420 nm. The luciferase activity in the samples transfected with the pGL3-Basic vector was arbitrarily set at 1.0. The activities of the different constructs are shown as relative activity with the activity of the pGL3Basic vector subtracted. Each experiment was performed in triplicate, and the results are presented as the average of the mean ^ S.D. 2.8. Statistical analysis ANOVA on experimental data sets was performed with a program contained in the Systat package, version 5.0 (SYSTAT Intelligent Software, Evanston, IL, USA).

had a reverse orientation. MREd (TGCGCAC) was located from 2541 to 2548 bp, MREe (TGCGCAC) from 2574 to 2581 bp, MREf (TGCACAC) from 2585 to 2592 bp, and MREg (TGCACAC) from 2923 to 2930 bp. In addition to the MREs, two sequences similar to the activator protein-1 (AP-1) were identified: one (5 0 -CTATCTCA) located from 227 to 235 bp upstream of the ATG codon, and the other (5 0 -TGACTAG) located with a reverse orientation in the distal portion of the MT-II promoter sequence, from 2633 to 2640 bp (Fig. 2B). The alignment of the MT-I and MT-II promoter regions showed a 57% identity (Fig. 3), which increased to 87% in the 300-bp region upstream of the ATG codon. Only the three proximal putative MREs identified (MREa, MREb and MREc) were conserved, both in position and sequence.

3. Results 3.1. Cloning and structure of the C. hamatus MT-I and MTII gene promoters We screened a C. hamatus genomic library and obtained two different clones with inserts of about 9 kb. By using the PCR strategy described in Section 2, we obtained an about 1000 bp fragment of the 5 0 -flanking region of the C. hamatus MT genes. The identification of each MT isoform was performed on the basis of the sequences of the MT-I and MT-II 5 0 -UTRs (Bargelloni et al., 1999). The nucleotide sequences of the MT-I and MT-II promoter regions are shown in Fig. 2. Analysis of these regions revealed that both promoters were rich in A–T content (57.3% for MT-I and 58.7% for MT-II), and lacked a canonical TATA box, that in both cases was modified into a TTTA sequence. In addition, several putative cis-regulatory sequences were identified. The MT-I promoter contained four MREs, organized into a single proximal cluster located within the first 300 bp from the ATG codon. The first MRE (MREa, TGCACCC) was located from 2116 to 2123 bp upstream of the transcription starting site. MREb (TGCACGC) and MREc (TGCACAC) were located with a reverse orientation from 2140 to 2147 and from 2179 to 2186 bp, respectively. MREd (TGCACAC) was located with a forward orientation from 2256 to 2263 bp. The promoter region of the C. hamatus MT-I gene also contained a site for the transcription factor-1 (Sp-1) (5 0 -CACCGCC) located from 278 to 286 bp upstream of the transcription starting point (Fig. 2A). In the MT-II promoter, seven MREs were identified. They were organized into two clusters, one containing three MREs located close to the ATG codon, and the other consisting of four MREs lying 500–900 bp upstream of the transcription starting point. The three MREs of the proximal cluster were located from 2128 to 2135 bp (MREa, TGCACCC), from 2151 to 2158 bp (MREb, TGCACGC) and from 2190 to 2197 bp (MREc, TGCACAC). MREb and MREc had a reverse orientation in the 5 0 flanking region. All of the four MREs of the distal cluster

3.2. 5 0 -deletion analyses of the C. hamatus MT-I and MT-II promoters To define the function of the MREs identified in the C. hamatus MT promoters, segments of the 5 0 -flanking region having different lengths were inserted into a luciferasereporter plasmid. A total of seven constructs (Fig. 4) were produced for each promoter. The MT-I promoter constructs were: (i) the pMT-I vector containing the full 21010 bp region of C. hamatus MT-I 5 0 -flanking region; (ii) pMT-I6 containing all the identified cis-acting elements and lacking the most distal 383-bp fragment; (iii) pMT-I5, containing all the cis-acting elements and lacking the distal 657-bp fragment; (iv) pMT-I4 with three MREs; (v) pMT-I3, containing two MREs; (vi) pMT-I2 containing only the most proximal MRE; vii) the pMT-I1 construct containing the TATA-box and lacking all the MREs (Fig. 4A). The MT-II promoter constructs were: (i) the pMT-II vector, containing the full 2986 region of C. hamatus MT-II 5 0 -flanking region; ii) pMT-II6, containing the 669-bp fragment lacking the most distal MRE; iii) pMT-II5 only containing the proximal cluster of three MREs; (iv) pMT-II4 made of a 233-bp fragment ending immediately downstream of the third MRE; (v) pMT-II3 containing the first two MREs; (vi) pMT-II2 only containing the most proximal MRE; (vi) the pMT-II1 construct only containing the TATA-box and lacking all the MREs (Fig. 4B). The pGL-3 basic vector without any insert was used as control. These promoter constructs were transfected into HepG2 cells, since fish metallothionein promoters have been shown to be functional in heterologous systems (Zafarullah et al., 1988; Olsson et al., 1995). The pMT-I, pMT-II and pMT-II6 constructs were tested in the HepG2 cells in the absence or presence of three different zinc concentrations (50, 100, and 150 mM) and two different cadmium concentrations (2.5 and 5.0 mM). The results in Fig. 5 show that pMT-I was induced only in the presence of the highest concentrations of zinc and cadmium, whereas pMT-II could be induced at all the concentrations tested, the maximal response being elicited at 50 mM zinc and 5.0 mM cadmium. Removal of the most distal MRE in the pMT-II

R. Scudiero et al. / Gene 274 (2001) 199–208

203

Fig. 2. Nucleotide sequences of the 5 0 -flanking regions of the C. hamatus MT-I (A) (EMBL accession no. AJ308478) and MT-II (B) (EMBL accession no. AJ308479) genes. Putative metal responsive elements (MREs) and other consensus sequences of binding of known transcriptional activators (AP-1, Sp-1) are shown, with arrows indicating orientation with respect to transcription. The TATA-box (underlined) is represented by the variant TTTA.

promoter resulted in a reduction of metal inducibility in the presence of cadmium; zinc was again the most potent inducer, but maximal induction was achieved by using the highest concentration of metal ions. Transfection of the different pMT-I vectors into the HepG2 cell line revealed that the MT-I promoter is very compact (Fig. 6A): removal of the distal 657 bp fragment, lacking MREs, enhanced basal activity (P , 0.0001) and metal inducibility (P , 0.0001) of the promoter; the remaining proximal 353 bp region (pMT-I5) showed high basal activity and elicited maximal response to metals. Deletion of the fragment between 2353 and 2222 bp (pMT-I4),

containing the most distal MRE, significantly modified both basal (P , 0.0001) and metal-induced (P , 0.0001) expression with respect to pMT-I5. Deletion of the region containing MREc consensus sequence (pMT-I3) caused a decrease in both basal and metal-induced luciferase activity; by contrast, deletion of MREb (pMT-I2) had no effect on the remaining activity, which was completely lost after the deletion of all MREs (pMT-I1). Transfection of the HepG2 cells with the pMT-II vector containing all the putative MREs resulted in maximal metal inducibility (Fig. 6B). Deletion of MREg (pMT-II6), the most distal element, did not significantly affect metal-induced

204

R. Scudiero et al. / Gene 274 (2001) 199–208

Fig. 3. Alignment of the C. hamatus MT-I and MT-II promoter sequences. The putative metal responsive element (MRE) consensus sequences are boxed in grey.

R. Scudiero et al. / Gene 274 (2001) 199–208

205

Fig. 4. Schematic representation of MT-I (A) and MT-II (B) 5 0 -deletion mutants. The left panel shows single deletion mutants with the location and orientation of the metal responsive elements (MRE, dark B) ( ! , forward; ˆ , reverse), the location of the activation protein-1 (AP-1, open A) and transcription factor-1 sequences (Sp-1, hatched g) the TATA box (grey A) and the position of the luciferase gene (solid black B). The right panel shows the ratio of the metalstimulated vs. basal activity, calculated for each promoter construct using the data reported in Fig. 6.

luciferase activity, whereas removal of the distally located cluster, containing MREf, MREe, MREd, reduced inducibility by zinc (P ¼ 0.003) and, to some extent, by cadmium (P ¼ 0.012). No significant difference in basal activity was observed among constructs pMT-II, pMT-II6, pMT-II5 and pMT-II4 (Fig. 6B). Additional deletion of one or more MREs from the proximal cluster (i.e. MREc, MREb and MREa) dramatically decreased both basal and metal-induced luciferase activity (constructs pMT-II3, pMT-II2 and pMT-I1). The induction ratios of each promoter construct in the presence of zinc or cadmium are reported in Fig. 4. The pMT-I5 construct showed maximal response to metals, with a 2-fold stimulation by cadmium and 5-fold stimulation by zinc (Fig. 4A). Among the pMT-II vector series, pMT-II showed maximal response: activity was stimulated 4.5-fold by cadmium and 6.5-fold by zinc (Fig. 4B).

4. Discussion The results reported in the present paper describe the structure and the regulatory function of the 5 0 flanking

Fig. 5. Functional analysis of pMT-I, pMT-II and pMT-II6 constructs in the presence of different concentrations of zinc or cadmium. Relative luciferase activity is given as relative luminescence standardized for transfection efficiency by measurements of b-galactosidase activity. The background activity of the pGL-3 basic vector was set at 1.0, and all the other activities were adjusted accordingly. Transfection conditions are given in Section 2. Each experiment was performed in triplicate. The bar indicates the standard error of the measurements.

206

R. Scudiero et al. / Gene 274 (2001) 199–208

Fig. 6. Functional analysis of MT-I (A) and MT-II (B) 5 0 deletion mutants. The different vectors are described in Fig. 4. Relative luciferase activity is given as relative luminescence standardized for transfection efficiency by measurements of b-galactosidase activity. The background activity of the pGL-3 basic vector was set at 1.0 and all the other activities were adjusted accordingly. Transfection conditions are given in Section 2. Each experiment was performed in triplicate. The bar indicates the standard error of the measurements.

regions of metallothionein genes from Chionodraco hamatus, a member of the major taxonomic group of Antarctic fish. Antarctic fish have two distinct chromatografic MT isoforms encoded by two paralogous genes (Bargelloni et al., 1999), that in C. hamatus are expressed at different basal levels and are differentially responsive to metal ions (Carginale et al., 1998). The promoters of the C. hamatus MT genes are A 1 T rich, a feature found also in the trout MT-B 5 0 flanking region, whereas the regulatory regions of mammalian MT genes are about 75% GC-rich (Zafarullah et al., 1988). Like in other fish, the variant TTTAA substitutes for the canonical TATAA-box. Reportedly, this substitution results in a lowered metal inducibility of the gene (Olsson and Kille, 1997). The comparison of the 5 0 flanking regions of C. hamatus MT genes shows that these sequences display high similarity only at the level of the

2300-bp fragment, whereas in the distal region, ranging between 21000 and 2300 bp, the degree of similarity considerably decreases. In vertebrates, all MT promoters contain a typical MRE consensus sequence, always present in multiple copies, that mediates the response to metals (Varshney et al., 1986; Searle et al., 1987; Culotta and Hamer, 1989). Analysis of the promoter regions of trout MT genes has indicated the presence of two clusters of MREs, one located within a 250bp region upstream of the start of the transcription signal, and the other in the distal portion of the 5 0 -flanking region of the MT gene (Zafarullah et al., 1988; Olsson et al., 1995; Samson and Gedamu, 1998). Multiple distal (.500 bp) consensus MREs have been found in pike and stone loach promoters (Kille et al., 1993; Olsson and Kille, 1997) as well as in the chicken MT promoter (Fernando and Andrews, 1989). The mammalian MT promoters with numerous proximal MREs do not generally contain distal MRE sequences (Imbert et al., 1990; Samson and Gedamu, 1998). C. hamatus is an exception among fish, because the organization of MT-I and MT-II promoters markedly differs both in the number and position of the putative cis-acting elements. The MT-II promoter region contains seven MREs organized into proximal and distal clusters, like in other fish. In particular, three MREs, two of which with a reverse orientation, form the proximal cluster; three have a reverse orientation and make up the distal cluster; the last is positioned between 2930 and 2923-bp upstream of the distal cluster. Conversely, the MT-I promoter region contains only four MREs organized in a single proximal cluster. Three of these MREs are identical in sequence, position and orientation to the MREs forming the proximal cluster of the MT-II promoter. Transfection analysis of 5 0 -deletion mutants showed that the different number of MREs in MT-I and MT-II promoters brings about different activity and inducibility to metals. The MT-II promoter with seven MREs shows high susceptibility to metals, maximal activity being elicited by 50 mM Zn 21. Removal of the most distal MRE (i.e. MREg) considerably modifies sensitivity to zinc: in this case, Zn 21 concentration must be raised to 150 mM to achieve stimulation, which, however, is about 50% that obtained with the complete MT-II promoter tested with 50 mM Zn 21. Transfection of cells with the promoter region containing only the proximal cluster results in a remarkable decrease in the promoter activity. These results demonstrate that the distally located cluster of putative MREs in the MT-II promoter is required for maximal response to zinc. Apparently, the three MREs of the proximal cluster functionally cooperate, as removal of a single MRE from such a cluster results in the complete loss of both basal and metal-induced activity. For the MT-I promoter, metal inducibility is regulated by four MREs arranged in a single proximal cluster. Transfection of cells with the two longest MT-I enhancer regions, pMT-I and pMT-I6, demonstrates that the presence of the

R. Scudiero et al. / Gene 274 (2001) 199–208

distal 5 0 -flanking region reduces the transcriptional activation of the luciferase reporter gene, at both basal and metalinduced levels. This distal region contains no known transcription enhancer elements, as the two motifs at position 2562/2555 and position 2932/2925 differ from the MRE consensus sequence in a single base substitution which is known to inactivate MRE function (Samson and Gedamu, 1998). The marked difference in structure and function between the MT-I and MT-II promoter sequences is in contrast with the high level of similarity shared by the coding sequences. This demonstrates that the divergence of the MT isoforms proceeded independently of the functional modification in the promoter regions. The polypeptide sequences of MT-I and MT-II differ in a single amino acid substitution, although it is not clear whether such a modification is sufficient to produce a change in the biological function. More likely, the structural and functional differences in the promoter regions are responsible for the differential expression of the two MT isoforms. The two MT genes present in Antarctic fish are the result of a duplication event that occurred in the ancestral lineage of notothenioids independently of that which gave origin to MT isoforms in trout (Bargelloni et al., 1999). Hence, it would not be surprising if, in Antarctic fish and salmonids, the MT genes were subjected to different transcriptional regulation to heavy metals. An example is given by Notothenia coriiceps, a red-blooded Antarctic notothenioid, in which MT-I and MT-II are differentially expressed at the level of tissues like brain, liver and kidney (Scudiero et al., 2000). The higher sensitivity of the MT-II promoter to Zn 21 might result in a preferential expression of the MT-II isoform in the tissues with a low cytoplasmic concentration of Zn 21. The expression of the MT-I gene is expected to occur preferentially in tissues with a higher Zn 21 level, or in response to an increase in heavy metal concentration. Hence, the MT-I gene may be considered as a ‘metal-shock’ gene. Acknowledgements We are grateful to Luisella Carratu` for providing us with the C. hamatus genomic library and Peter Kille for his helpful suggestions during the early stages of this work and for the sequencing of the MT-I and MT-II promoters. This study was supported financially by the Italian National Programme for Antarctic Research (PNRA). References Andrews, G.K., 2000. Regulation of metallothionein gene expression by oxidative stress and metal ions. Biochem. Pharmacol. 59, 95–104. Bargelloni, L., Scudiero, R., Parisi, E., Carginale, V., Capasso, C., Patarnello, T., 1999. Metallothioneins in Antarctic fish: evidence for independent duplication and gene conversion. Mol. Biol. Evol. 16, 885–897. Benton, W.D., Davis, R.W., 1977. Screening lambda gt recombinant clones by hybridization to single plaques in situ. Science 196, 180–182.

207

Binz, P.A., Ka¨ gi, J.H.R., 1999. Metallothionein: molecular evolution and classification. In: Klaassen, C.D. (Ed.), Metallothionein IV. Birkhauser Verlag, Switzerland, pp. 7–13. Brugnera, E., Georgiev, O., Radtke, F., Heuchel, R., Baker, E., Sutherland, G.R., Schaffner, W., 1994. Cloning, chromosomal mapping and characterization of the human metal-regulatory transcription factor MTF-1. Nucleic Acids Res. 22, 3167–3173. Carginale, V., Scudiero, R., Capasso, C., Capasso, A., Kille, P., di Prisco, G., Parisi, E., 1998. Cadmium-induced differential accumulation of metallothionein isoforms in the Antarctic icefish, which exhibits no basal metallothionein protein but high endogenous mRNA levels. Biochem. J. 332, 475–481. Culotta, V.C., Hamer, D.H., 1989. Fine mapping of a mouse metallothionein gene metal response element. Mol. Cell. Biol. 9, 1376–1380. Durnam, D.M., Palmiter, R.D., 1981. Transcriptional regulation of the mouse metallothionein-I gene by heavy metals. J. Biol. Chem. 256, 5712–5716. Fernando, L.P., Andrews, G.K., 1989. Cloning and expression of an avian metallothionein-encoding gene. Gene 81, 177–183. Freedman, J.H., Slice, L.W., Dixon, D., Fire, A., Rubin, C.S., 1993. The novel metallothionein genes of Caenorhabditis elegans. Structural organization and inducible, cell-specific expression. J. Biol. Chem. 268, 2554–2564. Friedman, R.L., Stark, G.R., 1985. Alpha-Interferon-induced transcription of HLA and metallothionein genes containing homologous upstream sequences. Nature 314, 637–639. Imbert, J., Culotta, V., Furst, P., Gedamu, L., Hamer, D., 1990. Regulation of metallothionein gene transcription by metals. Adv. Inorg. Biochem. 8, 139–164. Jahroudi, N., Foster, R., Price-Haughey, ?, Beitel, G., Gedamu, L., 1990. Cell-type specific and differential regulation of the human metallothionein genes. Correlation with DNA methylation and chromatine structur. J. Biol. Chem. 265, 6506–6511. Karin, M., Richards, R.I., 1984. The human metallothioneingene family: structure and expression. Environ. Health Perspect. 54, 111–115. Kille, P., Kay, J., Sweeney, G.E., 1993. Analysis of regulatory elements flanking metallothionein genes in Cd-tolerant fish (pike and stone loach). Biochim. Biophys. Acta 1216, 55–64. Koizumi, S., Suzuki, K., Ogra, Y., Yamada, H., Otsuka, F., 1999. Transcriptional activity and regulatory protein binding of metal-responsive elements of the human metallothionein-IIA gene. Eur. J. Biochem. 259, 635–642. Ka¨ gi, J.H.R., Schaffer, A., 1988. Biochemistry of metallothionein. Biochemistry 27, 8509–8515. Miles, A.T., Hawksworth, G.M., Beattie, J.H., Rodilla, V., 2000. Induction, regulation, degradation, and biological significance of mammalian metallothioneins. Crit. Rev. Biochem. Mol. Biol. 35, 35–70. Moilanen, L.H., Fukushige, T., Freedman, J.H., 1999. Regulation of metallothionein gene transcription. J. Biol. Chem. 274, 29655–29655. Nordberg, M., Nordberg, G.F., 2000. Toxicological aspects of metallothionein. Cell. Mol. Biol. 46, 451–463. Olsson, P.E., Kille, P., 1997. Functional comparison of the metal-regulated transcriptional control regions of metallothionein genes from cadmiumsensitive and tolerant fish species. Biochim. Biophys. Acta 1350, 325– 334. Olsson, P.E., Zafarullah, M., Foster, R., Hamor, T., Gedamu, L., 1990. Developmental regulation of metallothionein mRNA, zinc and copper levels in rainbow trout, Salmo gairdneri. Eur. J. Biochem. 193, 229– 235. Olsson, P.E., Kling, P., Erkell, L.J., Kille, P., 1995. Structural and functional analysis of the rainbow trout (Oncorhyncus mykiss) metallothionein-A gene. Eur. J. Biochem. 230, 344–349. Palmiter, R.D., 1998. The elusive function of metallothioneins. Proc. Natl. Acad. Sci. USA 95, 8428–8430. Sadhu, C., Gedamu, L., 1989. Metal-specific post-transcriptional control of human genes. Mol. Cell. Biol. 9, 5738–5741.

208

R. Scudiero et al. / Gene 274 (2001) 199–208

Samson, S., Gedamu, L., 1998. Molecular analyses of metallothionein gene regulation. Prog. Nucleic Acid Res. Mol. Biol. 59, 257–288. Scudiero, R., Carginale, V., Riggio, M., Capasso, C., Capasso, A., Kille, P., di Prisco, G., Parisi, E., 1997. Difference in hepatic metallothionein content in Antarctic red-blooded and haemoglobinless fish: undetectable metallothionein levels in haemoglobinless fish is accompanied by accumulation of untranslated metallothionein mRNA. Biochem. J. 322, 207–211. Scudiero, R., Verde, C., Carginale, V., Kille, P., Capasso, C., di Prisco, G., Parisi, E., 2000. Tissue-specific regulation of metallothionein and metallothionein mRNA accumulation in the Antarctic notothenioid, Notothenia coriiceps. Polar Biol. 23, 17–23. Searle, P.F., Stuart, G.W., Palmiter, R.D., 1987. Metal regulatory elements of the mouse metallothionein-I gene. Exp. Suppl. 52, 407–414. Seguin, C., Prevost, J., 1988. Detection of a nuclear protein that interacts with a metal regulatory element of the mouse metallothionein 1 gene. Nucleic Acids Res. 16, 10547–10560.

Thiele, D.J., 1992. Metal-regulated transcription in eukaryotes. Nucleic Acids Res. 20, 1183–1191. Varshney, U., Jahroundi, N., Foster, R., Gedamu, L., 1986. Structure, organization, and regulation of human metallothionein IF gene: differential and cell-type-specific expression in response to heavy metals and glucocorticoids. Mol. Cell. Biol. 6, 26–37. Vasconcelos, M.H., Tam, S.C., Beattie, J.H., Hesketh, J.E., 1996. Evidence for differences in the post-transcriptional regulation of rat metallothionein isoforms. Biochem. J. 315, 665–671. Walker, J.M., Gedamu, L., 1990. A human metallothionein pseudogene containing AG/CT repetitive elements. J. Mol. Evol. 31, 211–220. Zafarullah, M., Bonham, K., Gedamu, L., 1988. Structure of the rainbow trout metallothionein B gene and characterization of its metal-responsive region. Mol. Cell. Biol. 8, 4469–4476. Zafarullah, M., Olsson, P.E., Gedamu, L., 1990. Differential regulation of metallothionein genes in rainbow trout fibroblasts. RTG-2. Biochim. Biophys. Acta 1049, 318–323.