| Nucleic Acids Research | Pages |
Molecular cloning and expression of the mouse translation initiation factor eIF-1A
Introduction
Materials And Methods
Cloning of mouse eIF-1A cDNA by 5[prime] and 3[prime] rapid amplification of cDNA ends (RACE)
Isolation of genomic clones containing the eIF-1A promoter
Southern blot analysis
Northern blot analysis
Mapping of transcription start sites
Electrophoretic mobility shift assays
Specific amplification of cDNA ends (SPACE) to detect eIF-1A transcripts in mouse
Results
Molecular cloning of mouse eIF-1A cDNA
Isolation of genomic clones containing putative eIF-1A promoter
Northern blot analysis of eIF-1A
Mapping transcription start sites
Binding of TBP to the putative TATA element in the eIF-1A proximal promoter
Specific amplification of cDNA ends (SPACE) to detect five eIF-1A transcripts in 2-cell mouse embryos
Discussion
Acknowledgements
References
Molecular cloning and expression of the mouse translation initiation factor eIF-1A
DDBJ/EMBL/GenBank accession no. AF026481
ABSTRACT
INTRODUCTION
Following fertilization of mouse eggs, the embryonic genome is clearly activated by the mid 2-cell stage (1). This activation involves expression of transcripts that are common to both the egg and embryo, as well as expression of transcripts that are unique to the embryo and presumably essential for further development. In addition, there is a class of genes, as identified by two-dimensional gel electrophoresis, whose expression transiently increases during the 2-cell stage (2). While the identity of a few of these polypeptides has been determined, e.g. transcripts encoding the splicing factor U2afbp-rs (3), hsp70 (4) and the translation initiation factor eIF-1A (formerly called eIF-4C) (5), the paucity of biological material has seriously hampered the identification of zygotically activated genes and, in particular, the class that are transiently expressed during the 2-cell stage. Analysis of the regulation of expression of these genes whose expression transiently increases during the 2-cell stage should provide insight into the mechanism of zygotic gene activation.
Previously, we employed mRNA differential display (6), as adapted for pre-implantation mouse embryos (7), to identify genes whose expression transiently increases during the 2-cell stage. This screen resulted in the isolation of a 350 bp cDNA fragment that exhibits a transient increase in expression during the 2-cell stage (5). Sequence analysis of the fragment revealed it to be the mouse homolog of the human protein translation initiation factor eIF-1A (GenBank accession no. U28419). Eukaryotic initiation factor eIF-1A is a small protein of Mr = 16 400 that promotes dissociation of 80S ribosomes into subunits (8), stabilizes methionyl-tRNA binding to 40S ribosomal subunits (9) and promotes binding of mRNA to the 40S pre-initiation complex (10). Unlike several other translation initiation factors, it does not appear to be post-translationally modified by phosphorylation (11). In yeast, the protein is essential for cell growth and viability (12). A recent report challenges the view that eIF-1A is a pleiotropic factor in initiation of translation and suggests that it is essential to mediate the nearly quantitative transfer of the met-tRNAf@eIF2@GTP ternary complex to 40S ribosomal subunits (13).
In order to explore further the regulation of eIF-1A expression during zygotic gene activation, which is essential for further embryo development, it was necessary to clone the gene and its promoter. We report here molecular cloning of the mouse translation initiation factor eIF-1A gene and describe several aspects of its organization. In addition, we show that the individual eIF-1A transcripts can be selectively amplified; this will facilitate elucidation of the requirements for control of their expression during early embryogenesis.
MATERIALS AND METHODS
Cloning of mouse eIF-1A cDNA by 5[prime] and 3[prime] rapid amplification of cDNA ends (RACE)
5[prime] RACE. RACE was performed as described (14). Briefly, total RNA was extracted from the liver of female CF-1 mice (Harlan) and reverse transcription performed on the poly(A)+ fraction using antisense primer A1 and 200 U Superscript II reverse transcriptase (Life Technologies Inc.). The sequence of this primer and others used in this study are presented in Table 11. RNase H was added to remove the mRNA strand and the 3[prime]-end of the first strand DNA was extended with dCTP and terminal deoxynucleotidyl transferase. The first round of PCR was performed on the tailed cDNA template using sense primer S1 and antisense primer A2. A second round of PCR was performed on a 1:100 dilution of the primary amplification product using sense primer S2 and antisense primer A2. The resulting 754 bp PCR product was cloned and sequenced.
Table 1. Primers used for RT-PCR
Isolation of genomic clones containing the eIF-1A promoter
The mouse PromoterFinder (Clontech) libraries consist of five individual pools of genomic DNA fragments generated by digestion with EcoRV, ScaI, DraI, PvuII or SspI. For the initial genomic walk the primary PCR was performed according to the manufacturer's instructions using adapter primer 1 (S7) and antisense gene-specific primer 1 (A5). For the second round of amplification, reaction conditions were as described above except that 1:50 of the primary reaction was used as template and nested adapter primer 2 (S8) and antisense gene-specific primer 2 (A6) were used. An aliquot of the reactions was run on a 1.5% agarose gel and selected PCR products were cloned and sequenced. For the second genomic walk, the first round of PCR was performed as described above, except that the primary PCR used sense primer S7 and antisense gene-specific primer A7 and the second round of amplification used nested adapter primer S8 and antisense gene-specific primer A8. An aliquot of the reactions was run on a 1.5% agarose gel and selected PCR products cloned and sequenced.
Southern blot analysis
A 329 bp probe for eIF-1A was generated by PCR using a mouse genomic DNA template and sense primer S6 and antisense primer A4. The 329 bp PCR product was used as template for probe synthesis using a single gene-specific antisense primer, A4, and [[alpha]-32P]dCTP. Approximately 5 × 106 c.p.m. probe was used for Southern hybridization experiments. Mouse genomic DNA (8-10 µg) was digested with either HindIII, EcoRI, BglII or XbaI. Southern transfer and hybridization was performed according to standard procedures (15). Autoradiography was performed and results were viewed on a Molecular Dynamics PhosphorImager.
Northern blot analysis
Total mouse liver RNA (30 µg) was subjected to electrophoresis in a 1.4% agarose gel containing 37% formaldehyde. Northern transfer and hybridization were conducted according to standard conditions (15). The radiolabeled single-stranded antisense probe used in these studies (2.5 × 106 c.p.m./ml) was generated from a 465 bp cDNA template spanning the coding region of the gene. The DNA template for the probe was generated by PCR, using sense primer S4 and antisense primer A2 from a cDNA reverse transcribed template. Radiolabeling of this probe was performed as described above (Southern blot analysis). For analysis of eIF-1A transcripts in various mouse tissues, the Mouse Multiple Tissue Northern Blot (Clontech) was utilized. This blot consists of ~2 µg poly(A)+ RNA from the following mouse tissues: heart, brain, spleen, lung, liver, skeletal muscle, kidney and testis. Autoradiography was performed and results were viewed on a Molecular Dynamics PhosphorImager.
Mapping of transcription start sites
5[prime] RACE was used to map the transcription start sites (16) of the 2.8, 2.2 and 1.9 kb transcripts that were detected by northern blot analysis (see Results) and was performed similarly to the cloning of eIF-1A cDNA with the following exceptions. To map the start site of the 1.9 kb transcript, the first round of reverse transcription was performed with total liver RNA, using antisense primer A12. Following RNase H treatment of the first-strand cDNA, purification and dC tailing with terminal deoxynucleotidyl transferase, the first round of PCR amplification was performed on the cDNA template with the sense anchor primer S1 and the antisense primer A5. A second round of PCR amplification was then performed on a 1:100 dilution of the primary amplification product with sense primer S2 and antisense primer A6. The two resulting PCR products of 235 and 205 bp were cloned and sequenced.
To map the start site of the 2.8 kb transcript, reverse transcription was performed with total liver RNA, using antisense primer A7. Preparation of the cDNA template and the first round of PCR were as described above, using sense anchor primer S1 and antisense gene-specific primer A8. The second round of PCR was performed as above, using sense primer S2 and antisense gene-specific primer A11. The two resulting PCR products of 232 and 205 bp were cloned and sequenced. The transcription start site of the 2.2 kb transcript was mapped in a similar fashion, except that in the second round of PCR antisense gene-specific primer A10 was used. The resulting 126 bp PCR product was cloned and sequenced.
Electrophoretic mobility shift assays
Gel mobility shift assays were performed as previously described (17). The double-stranded oligonucleotide probes pTATA (proximal TATA element, 5[prime]-GCATTCCTTATACGCTTTCCCACGT-3[prime]), dTATA (distal TATA element, 5[prime]-ATGGCGTTTCTTACAGGGCATTCCT-3[prime]) and mTATA (mutant TATA element, 5[prime]-GCATTCCTGCTACGCTTTCCCACGT-3[prime]) were generated by end-labeling the 5[prime] or sense oligonucleotides to a specific activity of 1.7-2.9 × 109 c.p.m./µg and hybridization to an excess of unlabeled complementary oligonucleotides. Generation of the competitor pTATA and mTATA double-stranded oligonucleotides were similarly produced. In 10 µl binding reactions, 1.5 ng radiolabeled probe was incubated with 30 ng TBP protein (Santa Cruz Biotechnology) in 20 mM Tris-HCl (pH 7.9), 25 mM KCl, 2 mM spermidine, 0.1 mM EDTA, 0.025% NP-40, 10% glycerol, 0.5 mM DTT, 100 µg/ml BSA, 2 mM MgCl2 for 40 min at room temperature. In the competition experiments, a 200-fold molar excess of the pTATA or mTATA oligonucleotides was initially incubated with the protein for 20 min prior to addition of the pTATA radiolabeled probe and incubation was continued as above. For the supershift experiments, 2 µl polyclonal antibody specific for TBP were initially incubated with the protein for 20 min prior to addition of radiolabeled probe and incubation was continued as above (the antibody was a generous gift of Dr Frank Pugh). The binding reactions were separated in 4% polyacrylamide-0.5× TBE gel containing 2.5 mM MgCl2 and 0.05% NP-40. The 0.5× TBE running buffer contained 2 mM MgCl2 and 0.05% NP-40. After electrophoresis, the gels were dried onto Whatman 3MM paper and the complexes detected by autoradiography using a Molecular Dynamics PhosphorImager.
Specific amplification of cDNA ends (SPACE) to detect eIF-1A transcripts in mouse
Specific amplification of cDNA ends for eIF-1A transcripts was performed as described (18; Results gives a more general description of the method). Total RNA was isolated from 350 mid-2-cell embryos as previously described (19) and the pellet resuspended in 18 µl DEPC-treated water. Random hexamer oligonucleotides (200 pmol) were added and the reaction heated to 70°C for 10 min and transferred to ice for 2 min. The contents were collected by brief centrifugation and incubated at 23°C for an additional 10 min. Reverse transcription was performed for 90 min at 42°C in a 40 µl reaction containing 50 mM Tris-HCl, pH 8.3, 75 mM KCl, 3 mM MgCl2, 10 mM DTT, 1 mM dNTPs, 4 U RNasin (Promega) and 400 U Superscript RT II (Life Technologies). The amplification was conducted in two steps: a 10 min extension of the first strand cDNA to incorporate the dissimilar antisequence of the 5[prime]-flanking region of the extension oligonucleotide, followed by extension with the amplification primer that base pairs to the newly extended cDNA to generate a double-stranded template suitable for conventional PCR with the 5[prime] or sense amplification primer and a gene-specific 3[prime] or antisense primer. The same extension primer (S12) was used to extend the 2.8 and 1.9 kb transcripts initiated from the A transcription start site. The same extension primer (S13) was used to extend the 2.8 and 1.9 kb transcripts initiated from the A[prime] transcription start site, 30 bp downstream of the A transcription start site. The extension oligonucleotide S14 was used to extend the 2.2 kb transcript initiated from the B transcription start site. The same 5[prime] or sense amplification primer (S15) was used for amplification of the 2.8 and 1.9 kb-specific transcripts initiated from the A transcription start site. The same 5[prime] or sense amplification primer (S16) was used for amplification of the 2.8 and 1.9 kb-specific transcripts initiated from the A[prime] transcription start site. The 5[prime] or sense amplification primer S17 was used for amplification of the 2.2 kb-specific transcript initiated from the B transcription start site. The same 3[prime] or antisense gene-specific primer (A13) was used for amplification of the 2.8 kb-specific transcripts initiated from the A and A[prime] transcription start sites, respectively. The same 3[prime] or antisense gene-specific primer (A6) was used for amplification of the 1.9 kb-specific transcripts initiated from the A and A[prime] transcription start sites, respectively. Although these primers could also conceiveably amplify the 2.8 kb-specific transcript, since antisense primer A6 is located near the coding region of the gene, PCR extension times are kept short to selectively amplify the 1.9 kb-specific transcript. The 3[prime] or antisense gene-specific primer A14 was used for amplification of the 2.2 kb-specific transcript initiated from the B transcription start site. For the first step of amplification for the five individual reactions, 1 µl (8 embryo equivalents) of the cDNA from the reverse transcription reaction was used in a 50 µl reaction containing 200 µM dNTPs, 1× Pwo buffer + Mg2+ (BM), 5 pmol appropriate extension oligonucleotide and 1.75 U Pwo polymerase (BM). PCR extension was performed in a Perkin-Elmer 9600 thermocycler using the following program: 94°C for 2 min, 70°C for 30 s, 72°C for 10 min. The reactions were then heated to 90°C for 2 min and 50 pmol appropriate 5[prime] or sense amplification primer and 3[prime] or antisense gene-specific primer were added to the individual reactions. Cycle parameters for the second step of amplification were: seven cycles of 94°C for 30 s, 62°C for 30 s, 72°C for 30 s, followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 30 s. The products of PCR amplification were analyzed by electrophoresis in a 2% agarose gel.
RESULTS
Molecular cloning of mouse eIF-1A cDNA
We previously obtained DNA sequence for the mouse eIF-1A gene that includes both coding sequence and 3[prime]-untranslated region (UTR) sequence (5). In order to obtain a full-length mouse eIF-1A cDNA, we employed 5[prime] and 3[prime] RACE. Using 5[prime] RACE, a 754 bp amplification product was isolated, cloned and sequenced. This product contained the entire open reading frame (ORF) of the gene; the ATG start codon (bp +1173) is preceded by the consensus translational start site sequence CCA/GCCATG (20). The ORF extends 432 bp prior to the translational TAG stop codon (Fig.
Figure 1. Genomic organization of the cDNA encoding mouse eIF-1A. The nucleotide sequence of cDNA encoding mouse eIF-1A was determined as described in Materials and Methods. Transcription initiation site A (+1) is denoted by an arrow. The second and third transcription initiation sites, A[prime] (+ 31) and B (+628), are also denoted by arrows. The 840 bp region representing the alternatively spliced form of eIF-1A is indicated from position 76 to 916. A TATA-like sequence beginning at position 609 is demarcated with an asterisk. The open, crosshatched and shaded boxes indicate exons that are alternatively spliced to generate multiple 5[prime]-UTR regions. At position 976, the beginning of an ~8.5 kb intron is shown. The open reading frame extends from position 1173 to 1607 (black box). A polyadenylation signal is indicated at position 2631-2636. The remainder of the 3[prime]-end of the gene was cloned by 3[prime] RACE. Three rounds of amplification were performed using nested gene-specific primers to ensure specificity for the eIF-1A cDNA. Following the final nested PCR reaction, an amplification product of 1210 bp was then cloned and sequenced. The sequence of the 3[prime] RACE product was identical over 200 bp of sequence (1474-1684) previously defined (5) and extended to a putative polyadenylation signal, AATAAA (+2631) (21), that was followed by a poly(A) stretch corresponding to the beginning of the poly(A) tail of the mRNA. This approach, therefore, resulted in 1739 bp of sequence of the eIF-1A cDNA (GenBank accession no. AF026481). The sequence of the open reading frame of the mouse homolog of eIF-1A is >87% identical to that of the human at the nucleic acid level. In contrast, both the 5[prime]- and 3[prime]-UTR nucleotide sequences of the mouse homolog diverge significantly from that of the human sequence (22; GenBank accession no. L18960). At the protein level, the sequence is >98% identical over the 144 amino acids of the open reading frame, the mouse homolog differing from the human protein by two conservative substitutions.
Isolation of genomic clones containing putative eIF-1A promoter
In order to study the mechanism of activation of the eIF-1A gene at the time of zygotic gene activation during the 2-cell stage in the mouse embryo, we used the PromoterFinder (Clontech) method to isolate a single 2 kb product from the DraI library that contained putative cis-regulatory elements; this product was cloned and sequenced. In comparing the sequence upstream of the first coding exon generated by 5[prime] RACE and the same region produced from the Promoterfinder 2 kb genomic clone, the genomic sequence diverged from the cDNA ~200 bp upstream (+975; Figs
Figure 2. Schematic diagram depicting experiments to determine the size of two introns in the 5[prime]-UTR of the eIF-1A gene. (A) For the 8.5 kb intron, the genomic sequence across the splice donor and splice acceptor boundaries is shown, as well as the processed cDNA sequence following splicing. The arrows mark the approximate positions of primers used on genomic DNA and cDNA templates to determine the size of the intron. (B) For the 840 bp intron, the sequence of the 2.8 kb transcript across the splice donor and splice acceptor boundaries is shown, as well as the alternatively spliced cDNA sequence of the 1.9 kb transcript. The arrows mark the approximate positions of primers used on genomic DNA and cDNA templates to verify the alternatively spliced eIF-1A mRNAs. The presence and approximate size of such an intron was documented as follows (Fig. Two sense primers were tested. The first corresponded to the known sequence present in the cDNA within 50 bp of the putative splice donor site (S3). The second primer, S9, was 72 bp upstream of the putative splice donor site. (The sequence present for this primer was obtained from a second PromoterFinder walk that used antisense primers that corresponded to DNA sequence upstream of the putative splice donor site and present in the cDNA. This PromoterFinder walk generated a 1 kb amplification product that was cloned and sequenced.) Antisense primer A7 was chosen from sequence present in the cDNA downstream of the putative splice acceptor site. When PCR amplification was performed on the above genomic DNA templates, products of ~8.6 and ~8.5 kb, respectively, were produced. When PCR analysis was performed on the cDNA template, products of the expected sizes, 179 and 107 bp, respectively, were produced (data not shown). The latter products were sequenced to validate their specificity for the mouse eIF-1A gene. Finally, to confirm the presence of splice donor consensus signals, the ~8.5 kb PCR fragment containing the putative intron was purified and its 5[prime]-end sequenced. The sequence at the 5[prime] donor boundary, AG/GTGAGT (+975), closely corresponds to the consensus splice donor sequence, AG/GTAAGT (23). These findings suggest that an intron of ~8.5 kb is present in the 5[prime]-UTR of the eIF-1A gene. To confirm the PromoterFinder results, as well as to obtain additional genomic DNA that would include the putative promoter for eIF-1A, we screened a mouse BAC genomic library, whose average insert size is 130 kb (Research Genetics), with a probe that corresponds to bp 56-355 of eIF-1A. Two independent BAC clones (B 414 and B 435) were isolated that comprised the entire gene. A series of PCR experiments were then conducted to characterize the BAC clones. These experiments confirmed the identity of the BAC clones as specific for eIF-1A and also the presence of the large 8.5 kb intron in the 5[prime]-UTR of the gene (data not shown). Further, using primers (S11 and A2) that flanked the coding region of the gene, we isolated a PCR product that following sequencing revealed that the coding region of the gene is uninterrupted by introns (data not shown). Southern blot analysis using a probe specific to the 5[prime]-UTR of the gene (nt 56-355) was conducted to verify that a single copy of the gene containing this putative regulatory sequence was present. Southern blots of HindIII, EcoRI, BglII and XbaI digests of mouse genomic DNA with this probe revealed a single signal in each of the four restriction digests (Fig. Figure 3. Southern blot analysis of the mouse eIF-1A gene using a 5[prime]-UTR-specific probe. Genomic DNA was restricted with HindIII (lane 1), EcoRI (lane 2), BglII (lane 3) or XbaI (lane 4). The experimental procedures are described in Materials and Methods and molecular weight markers are indicated on the left.
Northern blot analysis of eIF-1A
In order to determine the number and size of eIF-1A transcripts, we performed northern blot analysis on mouse liver RNA using a radiolabeled probe that spanned the coding region of the gene. Autoradiography revealed three transcripts that corresponded to 2.8, 2.2 and 1.9 kb, respectively (Fig.
Figure 4. Northern blot analysis of mouse eIF-1A mRNA. (A) Northern blot analysis of total mouse liver RNA subjected to electrophoresis in a 1.4% formaldehyde-agarose gel. The molecular weights of the bands are indicated on the left and the positions of the 18S and 28S rRNAs are indicated on the right. (B) Multiple tissue northern blot. The mouse multiple tissue northern blot (Clontech) consists of ~2 µg poly(A)+ RNA from brain (BR), heart (HT), spleen (SP), lung (LG), liver (LI), skeletal muscle (SM), kidney (KI) and testis (TS). The RNA was subjected to electrophoresis in a 1.2% formaldehyde-agarose gel. Molecular weight markers are indicated on the left. The arrows on the right point to the positions of the 2.8 and 2.2/1.9 kb transcripts.
Mapping transcription start sites
The presence of three transcripts suggested that multiple transcription start sites may be used and/or alternative splicing occurs, although differential polyadenylation could also be involved. 5[prime] RACE was undertaken to map precisely the start sites of transcription to nucleotide resolution (18). Results of the experiments described below suggest that an alternative splicing event occurs in the 5[prime]-UTR, such that from transcription initiated at the upstream start site both the 2.8 and 1.9 kb transcripts are generated. A second transcription initiation site, which is located 628 bp downstream of the first site, generates the 2.2 kb transcript. The results of these experiments are summarized in Figure
Figure 5. Schematic diagram depicting experiments to map eIF-1A mRNA transcription start sites by RACE. Genomic organization of the eIF-1A gene is shown above the arrow and each of the transcripts derived from the gene is shown below the arrow. The exons are depicted as boxes and the transcription start sites A, A[prime] and B are indicated by rightward arrows. The leftward arrows mark the approximate positions of the 3[prime] primers used to map the transcription start sites for the 1.9, 2.8 and 2.2 kb mRNAs. The 5[prime] primers S1 and S2 were used in the first and second rounds of PCR, respectively. (A-C) Generation of the 1.9, 2.8 and 2.2 kb transcripts, respectively. The transcription start site of the 1.9 kb transcript was mapped by reverse transcription using antisense primer A12, located in the middle of the coding region of the gene. Following the first and second rounds of PCR, using upstream primers A5 and A6 (located several hundred base pairs upstream of the reverse transcription primer in the 5[prime]-UTR), two PCR products (a major product of 235 and a minor product of 205 bp) were generated and subsequently cloned and sequenced. The first major start site was mapped to a T residue (+1) that, when extended to the beginning of the poly(A) tail, comprises 1801 bp of cDNA. A minor start site was mapped to a T residue (+30), i.e. located 30 bp downstream of the major start site. Polyadenylation likely accounts for the 1.9 kb size of this transcript. When the resulting RACE cDNA sequence of the 1.9 kb transcript was compared with genomic DNA obtained from the PromoterFinder 1 kb genomic clone, the genomic sequence diverged from the cDNA (Fig. The transcription start site of the 2.8 kb transcript was mapped using antisense primer A7 for reverse transcription (Fig. Figure 6. Electrophoretic mobility shift assays of TBP protein binding to eIF-1A putative TATA elements demonstrating preferential binding of TBP to the proximal TATA element. (A) Lanes 1 and 2, 1.5 ng 5[prime]-end-labeled 25 bp pTATA box probe; lane 3, 1.5 ng 5[prime]-end-labeled 25 bp dTATA box probe. Lane 1, no protein; lanes 2 and 3, 30 ng TBP protein. (B) Lanes 1, 3, 4 and 5, 1.5 ng 5[prime]-end-labeled pTATA box probe; lane 2, 1.5 ng 5[prime]-end-labeled mTATA box probe; lanes 1-5, 30 ng TBP protein; lane 3, 200-fold molar excess pTATA competitor oligonucleotide; lane 4, 200-fold molar excess of mTATA competitor oligonucleotide; lane 5, supershifted protein-oligonucleotide complex with a TBP-specific antibody. The arrows point to the position of the specific complex. The transcription start site of the 2.2 kb transcript was mapped using antisense primer A7 for reverse transcription (Fig. 
Binding of TBP to the putative TATA element in the eIF-1A proximal promoter
The preceding experiments suggest that transcription of the eIF-1A gene is controlled by two putative promoters, separated by >600 bp. The proximal putative promoter, responsible for transcription of the 2.2 kb mRNA, contains a potential TATA element, located -20 bp from transcription start site B. To determine if this TATA element could be functional as determined by its ability to bind TBP, we performed an electrophoretic mobility shift assay (EMSA). In addition, we also characterized binding of a second putative TATA element located -34 bp upstream of the transcription start site.
The proximal TATA element (pTATA), located -20 bp upstream of transcription start site B, did in fact bind TBP by EMSA but the putative distal TATA element (dTATA), -34 bp upstream of the transcription start site, only very weakly bound the protein compared with the pTATA element (Fig.
A
 B  |
Figure 7. (A) Diagram of the general SPACE methodology to depict detection of mRNAs A and B, derived from two transcription initiation sites. Following reverse transcription of RNA, an extension oligonucleotide is annealed to the 3[prime]-end of the first strand cDNA. Amplification occurs in two steps. The first step is an extension of first strand cDNA to incorporate the dissimilar sequence of the 5[prime]-flanking region of the extension oligonucleotide. The extension oligonucleotide does not extend from its 3[prime]-end because it does not base pair match the cDNA. The second step is extension with a 5[prime] amplification primer that base pairs to the newly extended region of the cDNA to generate a double-stranded template suitable for conventional PCR with the 5[prime] amplification primer and a 3[prime] gene-specific primer. Although the cDNA produced from a co-linear mRNA that initiated transcription upstream of the first start site (B cDNA) would also anneal to the extension oligonucleotide, the complex would not serve as a polymerase template because neither the 5[prime]- nor 3[prime]-end of the extension oligonucleotide can base pair with the cDNA. (B) Schematic diagram depicting use of SPACE to detect the 2.8 and 1.9 kb eIF-1A transcripts derived from the distal promoter and the 2.2 kb eiF-1A transcript derived from the proximal promoter. For simplicity, the transcripts derived from the alternative A[prime] start site are not shown. In step 2, the extension oligo B does not bind to the 1.9 kb transcript and does not extend the 2.8 kb transcript. In step 4, to distinguish the 2.8 kb transcript from the 2.2 and 1.9 kb transcripts, the A13 primer sequence is within the intron and is 5[prime] of the B transcription start site. In step 7, the A6 primer is near the coding region and use of PCR conditions distinguishes the 2.8 from the 1.9 kb transcript, i.e. extension conditions do not permit detection of the 2.8 kb transcript.
Specific amplification of cDNA ends (SPACE) to detect five eIF-1A transcripts in 2-cell mouse embryos
Having mapped the transcription start sites of the five eIF-1A mRNAs, we wanted to confirm in vivo expression of the five independently initiated transcripts generated from the two promoters. For these experiments, we performed specific amplification of cDNA ends (SPACE), which allows detection of individual mRNAs initiated from a defined promoter and transcription start site, using 2-cell embryo RNA (Fig.
Using RNA recovered from 2-cell embryos, we successfully amplified PCR products corresponding to the five eIF-1A mRNAs (Fig.
Although the experiment was not quantitative, the B/2.2 kb-specific 230 bp fragment that was transcribed from the TATA-containing promoter appeared more abundant than those (2.8 and 1.9 kb-specific) from the TATA-less promoter. Further, although this experiment was done in 2-cell embryos, the relative amounts of the specific amplification products approximated the relative amounts of the mRNAs seen in northern blots of liver tissue.
DISCUSSION
In this report we extend our previous characterization of a partial cDNA for the mouse eIF-1A gene by obtaining sequence for a full-length cDNA, characterizing the genomic organization of the gene, as well as describing how several transcripts are generated. Southern blot analysis using either a 5[prime]-UTR- or 3[prime]-UTR-specific probe is consistent with a single copy of these putative regulatory sequences physically linked to the coding region of the gene. The open reading frame of the eIF-1A mRNA is 432 bp long (ATG start nt 1173) and generates virtually the same protein of 144 amino acids as the human homolog; it differs by only two conservative amino acid changes. This sequence conservation extends to plants, in which wheat germ eIF-1A is 68% identical and 76% similar to the mammalian protein (22). This high degree of homology between eIF-1A is reflected in conservation of function, since the mammalian protein functions in wheat germ translation systems and the wheat germ protein functions in mammalian translation systems (24). The 5[prime]- and 3[prime]-UTR sequences of the mouse gene, however, diverge significantly from those present in the human gene.
Figure 8. SPACE analysis of transcripts derived from eIF-1A in 2-cell pre-implantation mouse embryo. Lane 1, 100 bp molecular weight marker; lane 2, 246 bp product corresponding to the 2.8 kb transcript initiated from the A transcription start site; lane 3, 216 bp product corresponding to the 2.8 kb transcript initiated from the A[prime] transcription start site; lane 4, 211 bp product corresponding to the 1.9 kb transcript initiated from the A transcription start site; lane 5, 181 bp product corresponding to the 1.9 kb transcript initiated from the A[prime] transcription start site; lane 6, 230 bp product corresponding to the 2.2 kb transcript initiated from the B transcription start site. An 8.5 kb intron is present in the 5[prime]-UTR, 197 bp upstream of the methionine initiation codon. In a survey of 699 vertebrate mRNA sequences, nearly 25% possessed an intron between the promoter and the start of the major ORF (25). This 5[prime] intron in the mouse eIF-1A gene may serve a regulatory function in transcription, since a growing list of genes with 5[prime] introns reveals the presence of regulatory elements within these introns (26,27). In contrast to the 5[prime]-UTR, the coding region of the gene is uninterrupted by introns. Northern blot analysis indicates the presence of three major transcripts of 2.8, 2.2 and 1.9 kb in length. 5[prime] RACE mapping reveals that an alternative splicing event occurs in the 5[prime]-UTR resulting in generation of both the largest, 2.8 kb, and smallest, 1.9 kb, transcripts. The major start site is shared by both transcripts and was mapped to a T residue; the minor start site is mapped to a T residue 30 bp downstream. The genomic sequence in the putative promoter region upstream of the major start site has several GC-rich clusters (i.e. potential Sp1 binding sites) and lacks a TATA box or CCAAT box element that specify the start site and determine the efficiency of transcription (28). The 2.2 kb transcript is generated from a separate transcription start site that maps to a C residue (+628), and a TATA-like consensus sequence is located upstream of this start site (+609). This TATA element binds the TBP protein in electrophoretic mobility shift assays. The 2.2 kb mRNA is the most abundant of the transcripts present in the liver and its putative TATA-like box may confer specificity to the single site of initiation of transcription that is observed. Although the biological significance of the TATA-containing and TATA-less promoters present in the eIF-1A gene and the alternative splicing that occurs is not known, the ability to generate five transcripts from two promoters may provide greater flexibility in the regulation of eIF-1A gene expression. Differences in the level of transcription between alternative promoters, translational efficiency of mRNA isoforms, tissue specificity and responsiveness of the promoters and mRNA turnover may all be a consequence of transcription derived from multiple promoters and alternative splicing to generate different 5[prime]-ends (29). Although the 5[prime]-UTR of eIF-1A mRNA is not particularly GC-rich (<50%) and therefore would not be expected to generate stable secondary structures that may retard translation, there are multiple upstream AUG codons present that give rise to multiple small ORFs that may reduce the efficiency of translation of some of the eIF-1A mRNAs (30). For example, the 2.8 kb transcript contains multiple AUG codons that are not present in the spliced 1.9 kb variant and the 2.2 kb transcript contains fewer such codons than the 2.8 kb transcript. Results of the multiple tissue northern blot of eIF-1A mRNA show a significant variation in the abundance of the mRNA isoforms among the different tissues examined. These differences in expression may be due to differences in promoter utilization in response to cellular and metabolic conditions in a particular tissue type and result in mRNAs that may differ in translational capacity and/or stability. Our finding that eIF-1A expression is driven by both a TATA-containing and a TATA-less promoter affords us the opportunity to test the hypothesis that changes in TATA-box utilization could, in part, account for the spatially restricted patterns of gene expression that are present in the blastocyst. This hypothesis is based on the observation that in undifferentiated cells such as 2-8-cell blastomeres and embryonic stem (ES) cells, expression of a luciferase reporter gene driven by the tk promoter does not require a TATA box for enhancer stimulation of expression (31); in these undifferentiated cells enhancer stimulation of the TATA-less promoter is mediated by an Sp1 DNA-binding site. In contrast, differentiated cells, such as the oocyte, require a TATA box for enhancer-driven expression. Our ability to detect by SPACE each of the transcripts that are generated by the TATA-containing and TATA-less promoters in the pre-implantation mouse embryo allows us to test this attractive proposal.
ACKNOWLEDGEMENTS
W.D. would like to thank Maya Bucan and Barry Hough for assistance with the BAC libraries and Yu-ping Lu for assistance with the RACE experiments. This research was supported by a grant from the NIH (HD 22681) to R.M.S. W.D. was supported by a training grant from the NIH (5 T32 GM-07229) and portions of this work are being submitted by W.D. in partial fulfillment of the PhD requirements at the University of Pennsylvania.
REFERENCES
This article has been cited by other articles:
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: 30 Sep 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
M. Lynch, L. Chen, M. J. Ravitz, S. Mehtani, K. Korenblat, M. J. Pazin, and E. V. Schmidt
hnRNP K Binds a Core Polypyrimidine Element in the Eukaryotic Translation Initiation Factor 4E (eIF4E) Promoter, and Its Regulation of eIF4E Contributes to Neoplastic Transformation
Mol. Cell. Biol.,
August 1, 2005;
25(15):
6436 - 6453.
[Abstract]
[Full Text]
[PDF]
M. T. Fiorenza, A. Bevilacqua, S. Canterini, S. Torcia, M. Pontecorvi, and F. Mangia
Early Transcriptional Activation of the Hsp70.1 Gene by Osmotic Stress in One-Cell Embryos of the Mouse
Biol Reprod,
June 1, 2004;
70(6):
1606 - 1613.
[Abstract]
[Full Text]
[PDF]
K.-B. Choo, H.-H. Chen, T. Y.-C. Liu, and C.-P. Chang
Different modes of regulation of transcription and pre-mRNA processing of the structurally juxtaposed homologs, Rnf33 and Rnf35, in eggs and in pre-implantation embryos
Nucleic Acids Res.,
November 15, 2002;
30(22):
4836 - 4844.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (227K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (8)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Davis, W.
Articles by Schultz, R. M.
Search for Related Content
PubMed
PubMed Citation
Articles by Davis, W., Jr
Articles by Schultz, R. M.
Social Bookmarking
What's this?