Isolation and characterization of the gene coding for Artemia franciscana TATA-binding protein: expression in cryptobiotic and developing embryos

Biochimica et Biophysica Acta 1445 (1999) 271^282 www.elsevier.com/locate/bba

Isolation and characterization of the gene coding for Artemia franciscana TATA-binding protein: expression in cryptobiotic and developing embryos Leandro Sastre * Instituto de Investigaciones Biome¨dicas, CSIC/UAM, C. Arturo Duperier, 4, 28029 Madrid, Spain Received 1 February 1999; accepted 6 April 1999

Abstract Genomic and cDNA clones coding for the Artemia franciscana homolog of the TATA box-binding protein (TBP) were isolated. The C-terminal region of the predicted protein displays up to 92% sequence identity with the conserved C-terminal regions of TBPs from other species. The gene is divided in seven exons that expand over a region of 33 kb. The position of the four introns located in the conserved C-terminal region has been compared with those of other species. Two of these introns have been generally conserved during evolution, another is an arthropod specific intron, present in Drosophila melanogaster and A. franciscana, and the other is only conserved between vertebrates and A. franciscana. Primer extension experiments detected several transcription initiation sites. Northern blot analyses showed the presence of four mRNAs of estimated sizes of 6.8, 2.6, 1.6 and 1.1 kb. Except for the low expression of the 6.8 and 2.6 kb RNAs in encysted embryos, steady-state levels showed little variation during the activation of the encysted embryo and the first steps of embryonic and larval development. The amount of TBP protein expressed in encysted embryos and developing larvae has been analyzed by Western blot. Cryptobiotic embryos contain significant amounts of TBP although the level of expression increased almost twice during the first 20 h of development. The presence of TBP protein in cryptobiotic embryos suggests that TBP does not play, by itself, a critical role in the arrest of transcription characteristic of these resistance forms. ß 1999 Elsevier Science B.V. All rights reserved. Keywords: Development; Dormancy; Gene expression; TATA binding protein; Transcription

1. Introduction The crustacean Artemia franciscana is a typical inhabitant of highly salted aquatic environments.

Abbreviations: DTT, 1,4-dithiothreithol ; PMSF, phenylmethylsulfonyl £uoride; TBP, TATA binding protein * Fax: +34 (1) 585-4587; E-mail: [email protected]

These habitats are frequently exposed to seasonal changes that impair animal life. A. franciscana embryos survive during these periods because they are able to enter into a cryptobiotic state. Under adverse environmental conditions, embryos at the gastrula stage secrete a hard shell, arrest their metabolic activity and get dehydrated to form the cyst. When environmental conditions are adequate, cysts rehydrate and resume metabolic activity as well as their developmental program. In 15^20 h, a swimming larvae (nauplii) hatches from the cyst shell [1]. This

0167-4781 / 99 / $ ^ see front matter ß 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 7 8 1 ( 9 9 ) 0 0 0 5 2 - 4

BBAEXP 93269 27-5-99

272

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

animal system has been often used to study the mechanisms that control the entry into cryptobiosis and the later activation of the cysts (reviewed by Slegers [2]). One of the functions that need to be regulated during these processes is gene transcription. There is no transcriptional activity in the cysts, but several studies have shown the induction of gene expression a few hours after their activation [3,4]. One possible explanation for the lack of transcriptional activity in the cyst would be the absence of functional transcriptional machinery. However, cyst extracts have been shown to contain active RNA polymerases [5]. A cluster of general or basal factors are also involved in the main step responsible for transcriptional regulation, the formation of the transcription initiation complex [6,7]. The absence or inactivity of one or more general transcription factors could also explain the lack or transcriptional activity in the cyst. The more general basal transcription factor is the TATA-binding protein (TBP), that has been shown to be involved in the formation of the initiation complex by the three eukaryotic RNA polymerases [8^ 10]. In some RNA polymerase II and III promoters TBP interacts directly with the DNA through the TATA box sequence [11]. In TATA-less promoters (RNA polymerase I and some RNA polymerase II and III promoters) TBP seems to interact with other DNA binding proteins to form the initiation complex. In all these interactions, TBP is complexed with other proteins, the TBP-associated factors or TAFs. These complexes constitute the active transcription factors SL1, TFIID and TFIIIB, required for the formation of the initiation complex by RNA polymerases I, II and III, respectively [12^ 14]. TBP is also implicated in the regulation of transcription through its interaction with other general transcription factors, sequence-speci¢c transcription factors, repressors and coactivators [7,14^16]. DNA and protein interactions are established through a TBP region of 180 amino acids, the C-terminal region, whose sequence is highly conserved in all the organisms where this protein has been studied [17]. The role played by TBP in the formation of the initiation complex by the three RNA polymerases suggested that this factor could participate in processes where the transcription of all genes is regulated

coordinately, like the one previously described for the A. franciscana cyst. The study of the possible role of this protein during cryptobiosis and activation of the cysts has been initiated through the isolation of cDNA and genomic clones coding for A. franciscana TBP. The levels of expression of TBP mRNA and protein in cryptobiotic embryos and in the ¢rst developmental stages after their activation have been also studied. These experiments have shown the presence of TBP mRNA and protein in cryptobiotic cysts and a moderate increase of their expression during the ¢rst developmental stages. 2. Materials and methods 2.1. Isolation and characterization of genomic and cDNA clones A. franciscana genomic and cDNA clones were isolated from libraries constructed in the lambda phage vectors VEMBL3 and VZAPII, respectively [18]. Genomic clones (7U105 ) were initially screened using a Drosophila melanogaster TBP cDNA clone [19] as probe. Transfer of the phages to nylon ¢lters (Colony/Plaque Screen, NEN Life Science Products) and hybridization conditions were as previously described [20]. Filters were washed three times in 2USSC, 0.1% SDS at 42³C. cDNA clones were isolated after the screening of 4U105 phages of a cDNA library made from cryptobiotic embryos RNA [18] using as probe a fragment of the ¢rst genomic clone that was isolated. Hybridization conditions were the same described above but the ¢lters were washed three times in 0.1USSC, 0.1% SDS at 65³C. Other overlapping genomic clones were subsequently isolated from the same genomic library using as probes terminal fragments of genomic clones previously characterized. Filters were hybridized as previously described and washed three times in 0.1USSC, 0.1% SDS at 65³C during these screenings. Restriction maps of the genomic clones were determined by single and double digestions with the enzymes EcoRI and HindIII. The fragments originated in some EcoRI and HindIII digestions were inserted into the plasmid vector pBluescript and were used to further de¢ne the restriction maps and to determine

BBAEXP 93269 27-5-99

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

their nucleotide sequences. The Taq dye Deoxy Terminator Cycle Sequencing kit and a 373A DNA Sequencer from Applied Biosystems were used to determine nucleotide sequences. 2.2. Rapid ampli¢cation of cDNA ends (RACE) The 5P-region of TBP mRNA was ampli¢ed according to Dorit and Ohara [21]. Poly(A)‡ RNA (100 ng) from 20-h-old nauplii were used as substrate in a reverse transcriptase reaction primed with the oligonucleotide TBP-2: 5P-GGAATAATACCAGGATCAGC-3P, complementary to nucleotides 401^ 382 of the cDNA sequence shown in Fig. 2. The cDNA ¢rst strand was d(G)-tailed and used as substrate in a PCR reaction primed by the oligonucleotide TBP-2 and the d(C)-tailed oligonucleotide described by Kasturi and Bona [22]. One hundredth of the PCR reaction was used for a second PCR primed with the oligonucleotide TBP-4: 5-CCTGAGTTTGTGATATATTGG-3P, complementary to nucleotides 274^254 of the cDNA and the same d(C) tailed oligonucleotide described above. The ampli¢cation products were puri¢ed on a 2% agarose gel and cloned in the pGEM-T plasmid vector (Promega). 2.3. Primer extension Primer extension experiments were carried out according to the method described by Triezenberg [23]. Brie£y, the indicated amounts of RNA were hybridized with 5U105 cpm of the 32 P-labeled oligonucleotides TBP-10 (5P-CTCAGTAATTAAATTCTTACAAAC-3P, complementary to nucleotides 90^ 67, located in exon 1 of the TBP gene) or TBP-21 (5P-GCCAGGACTTGGCAACATATTATCCATGGTGG-3P, complementary to nucleotides 125^94 in exon 2). After hybridization the samples were incubated for 90 min at 42³C either in the presence or absence of AMV reverse transcriptase, as speci¢ed in each sample. Reaction products were analyzed on 6% polyacrylamide-7 M urea gels. Nucleotide sequence reactions primed with the same oligonucleotides utilized in primer extension were used as markers to identify the position of the transcription initiation sites. The T7 Sequencing kit from Pharmacia was used in these sequencing reactions.

273

2.4. Northern blot analyses Fifteen micrograms of RNA isolated from cryptobiotic A. franciscana embryos (San Francisco Bay Brand, Newark, CA, lot number SFB3554), either uncultured or after incubation in 0.25 M NaCl at 30³C for the times speci¢ed in each sample, were analyzed in 1.5% agarose-2.2 M formaldehyde gels, transferred to nylon membranes (ZetaProbe, BioRad) and hybridized to the 32 P-labeled probes. Hybridization and washing conditions were as described previously [4]. The hybridization signals obtained were quanti¢ed by densitometry using the program Image-1.55 and normalized to the amount of ribosomal RNA present in each sample. 2.5. Western blot analyses Cryptobiotic embryos or nauplii obtained after incubation of the embryos for 20 h were homogenized in 6 ml/g of 15 mM HEPES, 10 mM KCl, 5 mM MgCl2 , 0.1 mM EDTA, 0.5 mM EGTA, 350 mM sucrose, 1 mM DTT, 1 mM metabisul¢te, 0.2 mM PMSF, pH 7.6. Homogenates were centrifuged for 15 min at 10 000Ug, 4³C. The supernatants were saved as soluble fractions. The precipitates were resuspended in 1 ml/g of the same bu¡er, containing 100 mM KCl and 0.4 M ammonium sulfate and centrifuged again at 90 000Ug during 60 min at 4³C. The supernatant of this centrifugation was supplemented with 0.3 g/ml of ammonium sulfate. The precipitate was collected by centrifugation at 30 000UG during 20 min, 4³C, resuspended in 0.20 ml/g of 25 mM HEPES, 40 mM KCl, 12.5 mM MgCl2 , 0.1 mM EDTA, 10% glycerol, 1 mM DTT, 0.5 mM metabisul¢te, 0.1 mM PMSF, pH 7.6 and stored at 370³C as the nuclear fraction. The speci¢ed amounts of each subcellular fraction were electrophoresed on 10% polyacrylamide gels. Proteins were electrophoretically transferred to Immobilon P ¢lters (Millipore), blocked and incubated with the monoclonal antibody 58C9 (Santa Cruz Biotechnology) or normal rat serum as described. Horse radish peroxidase coupled anti-rat serum was used as secondary antibody. Antibody binding was detected by chemiluminescence (ECL, Amersham). The signals obtained were quanti¢ed by densitometry using the Image-1.55 program.

BBAEXP 93269 27-5-99

274

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

3. Results cDNA and genomic clones coding for A. franciscana TBP were isolated by hybridization to a Drosophila cDNA clone (generously gifted by R. Tjian) [19]. The ¢rst screening was made on 7.5U105 clones of an A. franciscana genomic library [18] under low stringency conditions and yielded the clone gTBP1 (Fig. 1A). The nucleotide sequence of the HindIII^ HindIII fragment containing Exon 5 (Fig. 1A) was highly similar to a fragment of the Drosophila cDNA. This fragment was used as probe in the screening of an A. franciscana cDNA library made from cysts RNA in the VZAPII vector [18]. Screening of 4U105 clones of the cDNA library allowed the isolation of three overlapping clones: cTBP4, cTBP21 and cTBP210, shown in Fig. 1B. Subsequent screenings of the genomic library with probes ob-

tained from gTBP1 or from the ends of other genomic clones allowed the isolation of the overlapping genomic clones gTBP11, 15, 18, 22 and 29 (shown in Fig. 1A). The restrictions fragments containing exons were identi¢ed and sequenced. The analyses of these fragments led to the determination of the TBP gene structure (Fig. 1A). The complete nucleotide sequence of the mRNA coding region of the gene as well as the promoter region and intron/exon borders are shown in Fig. 2. The gene is about 33 kb long, is composed of 7 exons and codes for a 1564-nt-long mRNA. The nucleotide sequence of intron/exon borders is very similar to the consensus sequences that have been described [24]. Exon 1 contains most of the 5P-untranslated region of the gene. This region was not contained in the isolated cDNA clones and was obtained by RACE, using as primers oligonucleotides

Fig. 1. Genomic and cDNA clones coding for A. franciscana TBP. (A) The six genomic clones coding for TBP that have been isolated are schematically represented as thin lines, with the name at their right (gTBP1^29). The restriction map of the DNA fragment covered by these overlapping clones is shown in the upper thick line for the enzymes EcoRI (E) and HindIII (H). The size of the clones and their restriction fragments is indicated in the lower scale bar. The position of the seven exons coding for the TBP mRNA is indicated by boxes on the restriction map, their order number is indicated underneath. (B) Schematic representation of the overlapping cDNA clones coding for A. franciscana TBP. Each clone is represented as a thin line with its name to the right (cTBP4, 21 and 210). The upper diagram indicates the position of the protein coding region as a box and untranslated regions as thick lines. The lower line indicates the size of the clones in kilobases.

BBAEXP 93269 27-5-99

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

complementary to exon 2 regions. Four di¡erent RACE clones were analyzed and extended to nucleotides 22, 37, 47 and 59 of the nucleotide sequence shown in Fig. 2. Oligonucleotides derived from the region ampli¢ed by RACE were used to map and sequence exon 1 in gTBP29. Transcription initiation sites were determined by primer extension. The results obtained using two different oligonucleotides as primers are shown in Fig. 3. Several transcription initiation sites were detected. The position of the three more prominent bands obtained with both oligonucleotides is indicated in Fig. 3 and are also labeled with asterisks in Fig. 2. These initiation sites are speci¢cally obtained by reverse transcription of A. franciscana RNA from 6- or 20-h-old animals (Fig. 3A, lane RT+, and Fig. 3B, lanes RT1 and 2, respectively), they are not obtained when RNA from Torula is used as substrate (Fig. 3B, lane RT3) or without reverse transcriptase (Fig. 3A, lane RT3). Two di¡erent hybridization conditions, 0.3 M and 0.75 M KCl, were used with one of the oligonucleotides and the same extension products were obtained (Fig. 3B, lanes RT1 and 2). These fragments do not correspond to preferential pausing sites for the transcriptase since they are not major bands observed when in vitro synthesized TBP RNA is used as substrate, in any of the hybridizations conditions tested (Fig. 3B, samples RT4 and 5). The nucleotide sequence upstream of the determined transcription initiation sites does not contain any putative TATA box, which has been often related to the existence of multiple initiation sites. Putative binding sites for transcription factors were searched for homology in the TFD and TFMatrix transcription factors binding site data bases. The more relevant sites found are indicated in Fig. 2, between them are binding sites for the cAMP response element binding protein (CREB), CAATbinding protein (CBP), Nuclear factor 1 (NF1) and factors homologous to MyoD (bHLH). In the 70-bp region upstream of the transcription start sites there are also two GC-rich hexanucleotides (named GCboxes in Fig. 2). The protein coding sequence begins in exon 2 where two very close in-frame methionines are located. The ¢rst of these methionines has been proposed as the initiation codon since it closely resembles the consensus translation initiation site [25].

275

Starting from this ¢rst methionine, the open reading frame codes for 275 amino acids. cDNA clones cTBP4 and 210 also code for a 640-nt-long 3P-untranslated region, fully contained in exon 7. Comparison of amino acids 97^275 of the sequence deduced from A. franciscana TBP clones with that of other TBP genes shows a considerable sequence identity with their conserved C-terminal domain, as shown in Fig. 4 where only disparities with the A. franciscana sequence are indicated. The percentage of amino acid sequence identity ranges from 80% (S. pombe [26]) to 92% (D. melanogaster [19] and Homo sapiens [17]). Besides, many of the amino acid di¡erences are conservative and all the amino acids that are completely conserved between species are also present in A. franciscana TBP. The non-conserved 96-amino acid-long N-terminal region of the deduced A. franciscana protein also shows some features observed in the N-terminal region of other TBPs, like the presence of polyglutamine stretches (amino acids 30^34) or the abundance of proline, serine and threonine residues. At the C-terminal end, the A. franciscana protein is 1^6 amino acids shorter than other TBPs. This small C-terminal tail, present in other TBPs, is not conserved between species. The A. franciscana TBP mRNA coding sequence is interrupted by six introns. Two of them are located in the non-conserved N-terminal coding region and the other four in the region coding for the conserved C-terminal region of the protein, as shown in Fig. 4. A. franciscana introns 3 and 5 interrupt the gene in the same positions as introns found in several other species [27]. Intron 4 is in the same position that the unique intron described in D. melanogaster [28] and intron 6 is in identical position as an human intron [29]. The expression of TBP mRNA during A. franciscana cyst development has been studied by Northern blot and the results obtained are shown in Fig. 5A. Total RNA from encysted embryos (0 h) or embryos and larvae cultured for 4, 8, 12, 16 or 20 h were hybridized to the cDNA clone cTBP210. Four hybridization signals were obtained for each sample, corresponding to RNA molecules of estimated sizes of 6.8, 2.6, 1.6 and 1.1 kb. The estimated size of the more abundant RNA (1.6 kb) coincides with the length of the cDNA clones and probably corre-

BBAEXP 93269 27-5-99

276

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

BBAEXP 93269 27-5-99

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

277

Fig. 2. Nucleotide sequence of the A. franciscana TBP gene coding region, promoter and intron/exon junctions. The nucleotide sequence was determined from the cDNA and genomic clones schematically represented in Fig. 1. RNA coding sequences are shown in capital letters and promoter and intron sequences in lowercase letters. Determined transcription start sites are indicated by asterisks. The nucleotide sequence has been numbered, on the left, from the more upstream transcription start site. Negative numbers correspond to the promoter sequence and positive numbers to the nucleotide sequence of the exons. Possible regulatory sites found for homology to transcription factors binding sites data bases are underlined and the putative binding transcription factor or the motif name indicated underneath. The position of two GC-rich hexanucleotides found in the promoter region is similarly indicated. The size estimated for each intron from the restriction maps shown in Fig. 1 is indicated between parentheses. The amino acid sequence of the encoded protein is shown over the DNA sequence and numbered from the proposed starting methionine on the right. This sequence has been deposited to the EMBL Nucleotide Sequence Data Base (accession number AJ002478).

sponds to mature mRNA. The other mRNAs could code for alternatively spliced isoforms since alternative ¢rst exons [30] and polyadenylation sites [31] have been described in other organisms. Probes from di¡erent regions of the gene were used to test this hypothesis. Oligonucleotide probes complementary to the 5P-untranslated region of exon 1 hybridized to the four mRNAs (data not shown). However, a probe containing the 3P-untranslated region present in exon 7 and in cDNA clones cTBP210 and cTBP4 hybridized to the mRNAs of 6.8, 2.6 and 1.6 kb, but not to the 1.1-kb mRNA, suggesting that this mRNA uses a di¡erent polyadenylation site. Larger RNAs (6.8 and 2.6 kb) could be precursor molecules of the 1.6-kb mRNA. The intensity of the hybridization signal at each developmental stage was quanti¢ed by densitometry and normalized to the amount of rRNA, that has

been shown to remain constant during this period of embryonic development [32]. The results obtained for the 6.8 and 2.6 RNAs showed similar levels of expression between 4 and 16 h of development and signi¢cantly lower levels in cysts and at 20 h of development. The 1.6-kb RNA was expressed at similar levels at all the developmental stages studied, with a small transient increase at 4 h of development. The 1.1-kb RNA increased its steady-state level of expression by 3^4 times in the ¢rst 4 h of development and remained constant at later developmental stages. The amount of TBP protein was determined by Western blot analyses of A. franciscana extracts with the monoclonal antibody 58C9 (Santa Cruz Biotechnology). Cysts or 20-h-old nauplii were homogenized and a nuclear fraction obtained by centrifugation. Duplicate samples of the supernatant or nuclear fractions were electrophoresed and trans-

BBAEXP 93269 27-5-99

278

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

Fig. 3. Determination of transcription start sites by primer extension. (A). Five micrograms of Poly(A)‡ RNA isolated from embryos incubated for 6 h was hybridized in duplicated sample with the oligonucleotide TBP-10: 5P-CTCAGTAATTAAATTCTTACAAAC-3P, complementary to nucleotides 90^67 of the nucleotide sequence shown in Fig 2. The hybridization products were then incubated in extension reactions in the presence (lane RT+) or absence (lane RT3) of reverse transcriptase. Reaction products were analyzed on a 6% polyacrylamide-7 M urea gel. Lanes A, C, G and T show the products of sequencing reactions of DNA from the genomic clone gTBP29, primed with the same oligonucleotide used in primer extension. Arrowheads at the right indicate the migration of the main extension products. (B) Seventy-¢ve micrograms of 20-h-old nauplii RNA (lanes RT1 and 2), 75 Wg of Torula RNA (RT3) or 10 ng of in vitro transcribed RNA containing the sense strand of the largest RACE clone and upstream vector sequences (RT4 and 5) were incubated with the oligonucleotide TBP-21: 5P-GCCAGGACTTGGCAACATATTATCCATGGTGG-3P, complementary to nucleotides 125^94 of the nucleotide sequence shown in Fig. 2. Hybridizations were made in the presence of 0.3 M KCl (lanes RT1, 3 and 4) or 0.75 M KCl (lanes RT2 and 5). After hybridization, all the samples were incubated with 25 U of AMV reverse transcriptase. Reaction products were analyzed on a 6% polyacrylamide-7 M urea gel. Lanes A, C, G and T show the sequencing reactions obtained using as substrate the RACE clone utilized to generate the TBP sense RNA and the oligonucleotide TBP-21 as primer. Arrows indicate the migration of the main extension products. (C) Nucleotide sequence of the proximal region of the A. franciscana TBP promoter. The position of the 3P-end of the extension products obtained with both oligonucleotides is indicated with asterisks over the nucleotide sequence.

C

ferred to two membranes. Equivalent samples from cysts or nauplii contained identical amounts or protein. One of the membranes was incubated with normal rat serum (data not shown) and the other with the anti-TBP monoclonal antibody. The results obtained are shown in Fig. 5B. Speci¢c binding was observed in a polypeptide of the expected electrophoretic mobility (30 kDa) in the nuclear fractions from cysts and nauplii. Densitometric quanti¢cation of the signals obtained showed that twice as much TBP protein was present in nauplii than in encysted embryos. 4. Discussion Genomic and cDNA clones coding for A. franciscana TBP have been isolated and characterized. The

very high identity (up to 92%) of the C-terminal 180 amino acids of the largest open reading frame with the conserved C-terminal region of TBPs from other species strongly indicates that the cloned gene codes for a functional TBP homolog in A. franciscana. The A. franciscana gene also codes for 96 additional Nterminal amino acids that are not conserved with respect to other TBPs, but share with them some general characteristics, such as the abundance of proline, serine and threonine residues and the presence of polyglutamine stretches. The size of the non-conserved N-terminal region has been shown to be very variable between species and is generally larger in animals than in plants and lower eukaryotes. The size or the A. franciscana N-terminal region, 96 amino acids, is smaller than those of mammals (134 residues) [17,33], reptiles (120 residues) [27] or D.

BBAEXP 93269 27-5-99

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

279

Fig. 4. Alignment of the conserved C-terminal domain of A. franciscana TBP with those of other species, with indication of intron insertion sites. The amino acid sequence of A. franciscana TBP from amino acid 97^275 (Af lane) is compared with the equivalent sequences of Drosophila melanogaster (Dm) [19], Homo sapiens (Hs) [17], Arabidopsis thaliana (At) [38], Schizosaccharomyces pombe (Sp) [26] and Acanthamoeba castellanii (Ac) [42]. Amino acids shown under the A. franciscana sequence indicate amino acids di¡erences while empty spaces indicate identical amino acids. Vertical lines indicate intron insertion sites. The position of A. franciscana introns 3^6 is indicated.

melanogaster (170 residues) [19], but signi¢cantly larger than those of the characterized lower eukaryotes and plants TBPs. TBP proteins from other species also present 1^6 non-conserved amino acids at the very end of the protein that are not encoded in the A. franciscana gene. The intron/exon structure of the A. franciscana TBP gene is very similar to those described for vertebrate TBPs. The gene is divided into seven exons while the genes from human, mouse and snakes are composed of eight exons [27,29,30,34]. The size of the A. franciscana gene, 33 kb, is larger than the vertebrate genes (about 15^20 kb) due to the larger size of its introns (over 5 kb in average). The other arthropod TBP gene that has been described to date, that of D. melanogaster, only has one intron [28]. Other A. franciscana genes that have been studied also resemble more in their structure to the homologous vertebrate genes than to those of D. melanogas-

ter [35^37]. The main di¡erence with respect to vertebrate genes is that the number of introns is generally slightly smaller and their size larger in the A. franciscana genes. The homologous D. melanogaster genes, however, present few, small introns. Two of the six A. franciscana introns are located upstream of the region coding for the conserved Cterminal part of the protein, one in the 5P-untranslated region and the other in the protein coding region. Similarly, two introns have been described in the N-terminal regions of vertebrate genes [27,29,30,34], one in the 5P-untranslated region and other in the coding region. One intron in the 5P-untranslated region has also been described in the plant A. thaliana [38]. The position of these introns cannot be compared between species because of the lack of sequence identity in this region. The position of the four introns located in the conserved C-terminal region is similar in A. franciscana and in some other

BBAEXP 93269 27-5-99

280

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

Fig. 5. TBP mRNA and protein expression during A. franciscana development. (A) Fifteen micrograms of total RNA from encysted embryos directly (lane 0) or after culture for 4, 8, 12, 16 or 20 hours (lanes 4, 8, 12, 16 and 20) were electrophoresed on a 1.5% agarose-2.2 M formaldehyde gel, transferred to a nylon ¢lter and hybridized with a cTBP210 insert probe. The migration of the ribosomal RNAs from rat (18S and 28S) and E. coli (16S and 23S) is shown on the right. (B) Fifteen micrograms of total RNA puri¢ed from 20-h-old nauplii were electrophoresed in duplicate samples on a 1.5% agarose-2.2 M formaldehyde gel and transferred to a nylon ¢lters. One of the samples (lane cDNA) was hybridized to the cTBP210 insert probe, the same used in A. The other ¢lter (lane 3PUTR) was hybridized to a probe containing the 3P-untranslated region present in cTBP210, generated by PCR. The migration of rat ribosomal RNAs (18S and 28S) is indicated on the left. (C) Encysted embryos or larvae obtained after 20 h of incubation of the cysts were homogenized and a nuclear fraction obtained by centrifugation. Twenty micrograms of the supernatant (lanes S) and 10 or 20 Wg (lanes N) of the nuclear fraction from cysts (lanes Cysts) or nauplii (lanes Nauplii) were analyzed on a 10% polyacrylamide gel. Proteins were transferred to a ¢lter and incubated with an anti-TBP monoclonal antibody. The arrow indicates the speci¢c signal detected on nuclear extracts.

organisms. The position of introns 3 and 5 is specially conserved since they are found in the same nucleotide position in most of the animal and vegetal TBP genes [27]. Intron 6 interrupts the coding sequence in the same position as mouse and human introns [29,34]. Intron 4 is located at the same position as the D. melanogaster intron [28]. Nakashima et al. [27] have compared the position of TBP introns in di¡erent species and related them with the L-sheets and K-helices found in the conserved C-terminal domain of the protein. These authors describe the existence of three introns conserved between species. The more N-terminal of these three introns interrupts the L-sheet S1 and is present in A. franciscana (intron 3). The second conserved intron is located between L-sheets S2 and S3. This intron is not present in the A. franciscana gene although it has been found in vertebrates, plants and lower eukaryotes. The third conserved intron corresponds to the A. franciscana intron 5 and interrupt Lsheet S2P. A. franciscana intron 6, shared with vertebrates, is located at the end of the L-sheet S5P. Intron 4, whose position is conserved between the arthropods D. melanogaster and A. franciscana, separates the region coding for the core domain of the protein in their two direct repeats. Taken together, these data suggest possible gain and loss of speci¢c introns which would be in support of the two main theories of intron evolution: late or ancient intron origin, respectively, For example, the existence in A. franciscana of two of the three conserved introns could represent ancient introns present in the ancestral TBP gene. The third intron conserved in other organisms should have been speci¢cally lost in A. franciscana during evolution. The data obtained also suggest the insertion of a new intron in arthropods (A. franciscana intron 4 and the D. melanogaster single intron). The identity of the position of intron 6 from A. franciscana with a vertebrate intron, but none of D. melanogaster could also be explained by the new insertion on this intron in the ancestral animal TBP gene and its posterior deletion in D. melanogaster. The analyses of the promoter region have shown that transcription is initiated from several sites. The analyses of the nucleotide sequence upstream of these sites did not allow the identi¢cation of any putative TATA box similar to those found in other A. franciscana genes: TATAAG in actin302 [36] and sarco/

BBAEXP 93269 27-5-99

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

endoplasmic reticulum Ca-ATPase [35] or TATATC in the Na‡ ,K‡ -ATPase K1 -subunit [37] genes. Both characteristics, the absence of TATA box and the use of several transcription initiation sites, have been also described for the TBP genes of D. melanogaster [28], mouse [39,40] or human [41]. The comparison of D. melanogaster TBP and TFIIB promoter region sequences identi¢ed a shared sequence motif ATTATTATT, this motif has not been found in the A. franciscana TBP promoter region. There are two GC-rich regions in similar positions in the promoters of A. franciscana and mouse TBP genes [39]. The comparison of the promoter sequence with data banks containing binding sites for transcription factors also detected several putative regulatory sites. The possible functional relevance or all these sites needs to be determined experimentally. The isolation of the A. franciscana TBP gene is the ¢rst step in the study of the possible role of this transcription factor in the resumption of gene expression after activation of the encysted embryos. As mentioned in the introduction, the encysted embryo is metabolically and transcriptionally inactive, but, when the embryo is hydrated and resumes its embryonic development, these processes are activated. The possible role of TBP during this process has been approached through the study of its mRNA and protein levels of expression in encysted and developing embryos. The data obtained showed that the encysted embryos express signi¢cant levels of TBP mRNA and protein. The presence of mRNA in the cyst could provide the necessary store to initiate the synthesis of TBP protein shortly after cyst activation. The presence of TBP protein in the cyst, in about half of the amount found after resumption of development, suggests that the transcriptional arrest is not due to the lack of TBP in encysted embryos. As mentioned in the introduction, multiple studies have shown that TBP participate in transcription initiation by the three RNA polymerases in association with other proteins, generally designed TAFs (TBPassociated factors). Several transcriptional repressors and coactivators have also been described to associate with TBP. It is possible that TBP could participate in transcriptional regulation in A. franciscana through its association with di¡erent transcriptional repressors or activators in encysted or developing embryos, respectively. The small increase observed

281

between cysts and nauplii could be physiologically relevant in these interactions. Transcriptional regulation could also be achieved through the regulation of the expression or activity of other general or sequence-speci¢c transcription factors. Finally, other mechanisms, such as changes in chromatin structure, could also participate in this regulatory process. Acknowledgements The author thanks Dr. Robert Tjian for donation of the Drosophila TBP cDNA clone, Mar|¨a del Mar Sa¨nchez de Miguel for her excellent technical assistance during the initial part of this work, Dr. Ricardo Escalante for critical reading of the manuscript and Antonio Fernandez and Ricardo Un¬a for preparation of the ¢gures. This work has been supported by Grant PB95-0096 from the Direccio¨n General de Ensen¬anza Superior.

References [1] L.E. Drinkwater, J.S. Clegg, in: R.A. Browne, P. Sorgeloos, C.N.A. Trotman (Eds.), Artemia Biology, CRC Press, Boca Raton, FL, 1991, pp. 93^117. [2] H. Slegers, in: R.A. Browne, P. Sorgeloos, C.N.A. Trotman (Eds.), Artemia Biology, CRC Press, Boca Raton, FL, 1991, pp. 37^74. [3] R. Marco, R. Garesse, J. Cruces, J. Renart, in: R.A. Browne, P. Sorgeloos, C.N.A. Trotman (Eds.), Artemia Biology, CRC Press, Boca Raton, FL, 1991, pp. 1^19. [4] R. Escalante, A. Garc|¨a-Sa¨ez, M.-A. Ortega, L. Sastre, Biochem. Cell Biol. 72 (1994) 78^83. [5] J. Renart, J. Sebastian, Cell Di¡er. 5 (1976) 97^107. [6] L. Zawel, D. Reinberg, Annu. Rev. Biochem. 64 (1995) 533^ 561. [7] J.-P. Halle, M. Meisterernst, Trends Genet. 12 (1996) 161^ 163. [8] B.P. Cormack, K. Struhl, Cell 69 (1992) 685^696. [9] P.W. Rigby, Cell 72 (1993) 7^10. [10] N. Hernandez, Genes Dev. 7 (1993) 1291^1308. [11] J.L. Kim, D.B. Nikolov, S.K. Burley, Nature 365 (1993) 520^527. [12] C.A. Radebaugh, J.L. Matthews, G.K. Geiss, F. Liu, J.-M. Wong, E. Bateman, S. Camier, A. Sentenac, M.R. Paule, Mol. Cell. Biol. 14 (1994) 597^605. [13] R.E. Meyers, P.A. Sharp, Mol. Cell. Biol. 13 (1993) 7953^ 7960. [14] S.K. Burley, R.G. Roeder, Annu. Rev. Biochem. 65 (1996) 769^799.

BBAEXP 93269 27-5-99

282

L. Sastre / Biochimica et Biophysica Acta 1445 (1999) 271^282

[15] B.F. Pugh, Curr. Opin. Cell Biol. 8 (1996) 303^311. [16] L.A. Stargell, K. Struhl, Trends Genet. 12 (1996) 311^315. [17] A. Ho¡mann, E. Sinn, T. Yamamoto, J. Wang, A. Roy, M. Horikoshi, R.G. Roeder, Nature 346 (1990) 387^390. [18] R. Escalante, L. Sastre, J. Biol. Chem. 268 (1993) 14090^ 14095. [19] T. Hoey, B.D. Dynlacht, M.G. Peterson, B.F. Pugh, R. Tjian, Cell 61 (1990) 1179^1186. [20] M.-T. Macias, I. Palmero, L. Sastre, Gene 105 (1991) 197^ 204. [21] R.L. Dorit, O. Ohara, in: F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Struhl (Eds.), Current Protocols in Molecular Biology, Current Protocols, New York, 1994, pp. 15.6.1^15.6.10. [22] K.N. Kasturi, C.A. Bona, Nucleic Acids Res. 19 (1991) 6339^6340. [23] S.J. Triezenberg, in: F.M. Ausubel, R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, K. Struhl (Eds.), Current Protocols in Molecular Biology, Current Protocols, New York, 1993, pp. 4.8.1^4.8.5. [24] M.B. Shapiro, P. Senepathy, Nucleic Acids Res. 15 (1987) 7155^7174. [25] M. Kozak, Nucleic Acids Res. 12 (1984) 857^872. [26] A. Ho¡mann, M. Horikoshi, C.K. Wang, S. Schroeder, A. Weil, R.G. Roeder, Genes Dev. 4 (1990) 1141^1148. [27] K. Nakashima, I. Nobuhisa, M. Deshimaru, T. Ogawa, Y. Shimohisaghi, M. Fukumaki, M. Hattori, Y. Sakaki, S. Hattori, M. Ohno, Gene 152 (1995) 209^213. [28] L.M. Lira-DeVito, T.W. Burke, J.T. Kadonaga, Gene 153 (1995) 203^207. [29] C. Chalut, Y. Gallois, A. Poterszman, V. Moncollin, J.-M. Egly, Gene 161 (1995) 277^282.

[30] E.E. Schmidt, U. Schibler, Dev. Biol. 184 (1997) 138^149. [31] J. Yamauchi, A. Sugita, M. Fujiwara, K. Suzuki, H. Matsumoto, T. Yamazaki, Y. Ninomiya, T. Ono, T. Hasegawa, S. Masushige, M. Muramatsu, T. Tamura, S. Kato, Biochem. Biophys. Res. Commun. 234 (1997) 406^411. [32] H.T. Koller, K.A. Frondorf, P.D. Maschner, J.C. Vaughn, Nucleic Acids Res. 15 (1987) 5391^5411. [33] T. Tamura, K. Sumita, I. Fujino, A. Aoyama, M. Horikoshi, A. Ho¡mann, R.G. Roeder, M. Muramatsu, K. Mikoshiba, Nucleic Acids Res. 19 (1991) 3861^3865. [34] K. Sumita, Y. Makino, K. Katoh, T. Kishimoto, M. Muramatsu, K. Mikoshiba, T. Tamura, Nucleic Acids Res. 21 (1993) 2769. [35] R. Escalante, L. Sastre, J. Biol. Chem. 269 (1994) 13005^ 13012. [36] M.-A. Ortega, M. D|¨az-Guerra, L. Sastre, J. Mol. Evol. 43 (1996) 224^235. [37] A. Garc|¨a-Sa¨ez, R. Perona, L. Sastre, Biochem. J. 321 (1997) 509^518. [38] A. Gasch, A. Ho¡mann, M. Horikoshi, R.G. Roeder, N.-H. Chua, Nature 346 (1990) 390^394. [39] T. Ohbayashi, E.E. Schmidt, Y. Makino, T. Kishimoto, Y. Nabeshima, M. Muramatsu, T. Tamura, Biochem. Biophys. Res. Commun. 225 (1996) 275^280. [40] E.E. Schmidt, T. Ohbayashi, Y. Makino, T. Tamura, U. Schibler, J. Biol. Chem. 272 (1997) 5326^5334. [41] C.E. Foulds, D.K. Hawley, Nucleic Acids Res. 25 (1997) 2485^2494. [42] J.M. Wong, F. Liu, E. Bateman, Nucleic Acids Res. 20 (1992) 4817^4824.

BBAEXP 93269 27-5-99