- Email: [email protected]
Identification of a developmentally regulated translation elongation factor 2 in Tetrahymena thermophila
Gene 326 (2004) 97 – 105 www.elsevier.com/locate/gene
Identification of a developmentally regulated translation elongation factor 2 in Tetrahymena thermophila Tania M. Malave´ 1, James D. Forney * Department of Biochemistry, Purdue University, West Lafayette, IN 47907 USA Received 28 April 2003; received in revised form 6 August 2003; accepted 15 October 2003 Received by A. Roger
Abstract Protein synthesis elongation factor 2 (eEF2) catalyzes the translocation of the peptidyl-tRNA from the A site to the P site of the ribosome. Most organisms encode a single EF2 protein and its activity is regulated by phosphorylation. We have identified a family of genes in Tetrahymena thermophila that encode proteins homologous to eEF2, yet are expressed only during sexual reproduction. These genes have been designated EFR for Elongation Factor 2 Related. EFR transcripts were not detected in vegetative cell cultures but rapidly increased about 6 h after the start of conjugation (mating). For comparison, we cloned, sequenced and analyzed the expression of the standard eEF2 gene from T. thermophila. Unlike EFR, transcripts from eEF2 were detected in vegetative cells but were present at lower concentrations during conjugation. Despite the high sequence identity between EFR and eEF2 from other organisms (about 42% at the amino acid level), key regulatory sequences that are involved in the regulation of eEF2 are altered in EFR. The sequence and expression data suggest that EFR is an eEF2 variant involved in a major translation regulatory mechanism that occurs during the formation of the macronuclear genome in conjugating cells. D 2003 Elsevier B.V. All rights reserved. Keywords: EF2; Protein synthesis; Ciliate; Protozoa
1. Introduction Protein synthesis is essential for the growth and development of all organisms and the components of the translational machinery have maintained a high degree of conservation throughout evolution. The synthesis of polypeptides consists of three major steps: (1) initiation in which mRNA comes into contact with the ribosome, the start codon is recognized and protein synthesis begins; (2) elongation, during which the polypeptide chain grows and the ribosome moves along the mRNA; and (3) termination,
Abbreviations: eEF2, eukaryotic translation elongation factor 2; eEF1A, eukaryotic translation elongation factor 1A; EFR, elongation factor 2 related; TBE, Tris borate EDTA; MOPS, 3(N-morpholino)propane-sulfonic acid; SDS, sodium dodecyl sulfate. * Corresponding author. Tel.: +1-765-494-1632; fax: +1-765-4947897. E-mail address: [email protected] (J.D. Forney). 1 Current address: Department of Molecular and Human Genetics, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA. 0378-1119/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2003.10.016
when a stop codon is recognized and the translation complex is disassembled. Protein synthesis is influenced by conditions that affect growth, such as availability of nutrients, stress, development and differentiation (reviewed in Mathews et al., 2000). Thus far, most examples of translational regulation involve initiation and fewer instances of control at the level of elongation are known. Translation elongation factor 2, eEF2, is a GTPase involved in the translocation of the peptidyl-tRNA from the A site to the P site on the ribosome. The 95-kDa protein is highly conserved, with 60% amino acid sequence identity between the human and yeast proteins. In Saccharomyces cerevisiae, there are two genes for eEF2, which encode identical proteins. Gene knockout experiments reveal that either gene can be eliminated individually without affecting cell viability, but double gene knockouts are lethal (Perentesis et al., 1992). Two major mechanisms are known to regulate protein elongation and both involve eEF2. First, eEF2 can be modulated by reversible phosphorylation (Nairn and Palfrey, 1987, Redpath et al., 1993). Increased levels of phosphorylated eEF2 reduce elongation rates
98
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
presumably because phosphorylated eEF2 fails to bind the ribosomes (Carlberg et al., 1990). Treatment of mammalian cells with agents that raise the cytoplasmic Ca2 + and cAMP levels reduce elongation rates by activating the kinase responsible for phosphorylating eEF2 (Mackie et al., 1989). In contrast, treatment of cells with insulin increases elongation rates by promoting eEF2 dephosphorylation (Redpath et al., 1996, Campbell et al., 1999). Second, the protein can be post-translationally modified by ADP-ribosylation. This reaction is performed by various bacterial toxins after modification of a specific histidine residue to diphthamide, but there is evidence for an endogenous ADP ribosylase activity (Fendrick et al., 1992). Similar to the bacterial toxins, it is presumed that modification by the endogenous enzyme also inhibits eEF2 activity. Our laboratory has been interested in the developmental events during the formation of the macronuclear genome in ciliated protozoa. Ciliates are unique in that they posses two distinct types of nuclei within a single cell: a somatic macronucleus and a transcriptionally silent micronucleus. During sexual conjugation, the existing macronucleus in the cell is destroyed and a new one is formed from the micronuclear genome (reviewed in Jahn and Klobutcher, 2002). The process of macronuclear maturation involves extensive genome remodeling. Thus far, the only developmentally regulated proteins that have been reported are associated with chromatin and DNA elimination (Nikiforov et al., 2000 and references therein). We have used differential display to identify transcripts unique to conjugation that could code for proteins involved in macronuclear development. We have identified a family of genes in Tetrahymena thermophila that encode for proteins homologous to eEF2 yet are expressed only during sexual reproduction. These genes have been designated EFR for Elongation Factor 2 Related. EFR transcripts appeared during conjugation (mating) but were not detected in vegetative cell cultures. For comparison, we cloned, sequenced and analyzed the expression of the standard eEF2 in T. thermophila. Unlike EFR, transcripts from eEF2 were detected in vegetative cells but decreased in conjugating cells. Despite the high sequence identity (roughly 50% at the amino acid level), there are key differences between the two proteins in conserved regulatory regions. The observations suggest that EFR is involved in a major translation regulatory mechanism that occurs during the formation of the macronuclear genome in conjugating cells.
2. Materials and methods 2.1. Tetrahymena growth and mating Strains CU428.1 and B2086 were obtained from Dr. David Asai (Biological Sciences Department, Purdue University). Cells were grown to a density of 2 105 cells/ml in NEFFs (0.5% dextrose, 0.25% yeast extract, 0.25% proteose
peptone, 3.3 mM FeCl3) media at 30 jC with 85 rpm of shaking. Mating reactive cells were prepared by spinning down 100 ml of a log phase culture (2 – 5 105 cells/ml), resuspending the cell pellet in 100 ml of 10 mM Tris –HCl, pH 7.5, and incubating at 30 jC with shaking (85 rpm) for 12– 24 h. After starving, the cells were counted and equal numbers of cells (at a density of 2 105 cells/ml) were added to a 2-l flask, adjusting the final volume to 100 ml. This flask was incubated at 30 jC, without shaking to allow pairs to form. Cell mixing was defined as time 0 and the cells were checked for mating efficiency at 3 – 4 h. Mating efficiency was calculated by counting the number of cells in pairs and dividing by the total number of cells. Only cultures with over 70% of the cells in pairs were used for further experiments. 2.2. RNA isolation Cells were disrupted using the Qiashredder (Qiagen) and RNA isolated with the Qiagen RNeasy Mini kit (Qiagen) Approximately 10 ml of cells at a density of 2 –5 105 cells/ml were used per column. For the time course experiments, cells were pelleted at each time point and the RNA immediately extracted. The RNA used for the Northern blots was treated with the Qiagen RNase free DNase (Qiagen). Poly(A) RNA was purified using the Promega Poly(A) system 1000 magnetic beads (Promega) using the protocol provided by the manufacturer. 2.3. Differential display Differential display was carried out using the Gene Hunter RNA Image kits (GenHunter). RNA was purified from contaminating DNA using the Message Clean DNase kit (GenHunter). Reverse transcription was performed with 0.1 Ag of total RNA and one of three different oligo dT primers provided in the kit. PCR of the resulting cDNA was then performed using the same oligo dT primer, a specific oligodecamer provided in the GenHunter kit and Qiagen Taq DNA polymerase in the presence of a-33P-dATP. Reactions were mixed with loading dye and electrophoresed on a 6% polyacrylamide gel in 1xTBE. The gel was dried, exposed to film overnight, and bands were excised. DNA was purified by placing the gel fragment in a 1.5-ml tube with water, sealing the tube with paraffin and boiling for 10 min. DNA was then amplified by PCR using the same conditions as in the original differential display but without the radioactive nucleotide. The PCR product was used as a probe for Northern blots and sub-cloned into the pGEM-Teasy vector (Promega). 2.4. Northern blots Five micrograms of RNA were mixed with 5 Al of sample buffer (3.7% formaldehyde and 4 MOPS) and 10 Al of
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
formamide and heated at 65 jC for 5 min. The samples were quick cooled on ice, 5 Al of loading buffer was added and they were electrophoresed on a 1% agarose, 2.5% formaldehyde gel in 1 MOPS (20 mM MOPS, 5 mM NaOAc, 1 mM EDTA, pH 7.0), at 90 V for 2 h. Gels were soaked in 20 SSC for 30 min, and RNA transferred to Nylon membranes, then fixed with UV cross-linking and hybridized. DNA probes were labeled with a-32P-dATP using the RediPrime kit (Amersham), and membranes hybridized overnight in 1 Denhart’s, 0.2 M phosphate buffer, 5 SSC, 0.25% SDS. After hybridization, membranes were washed twice for 15 min at room temperature in 1 SSC, 0.1% SDS and once for 30 min at 65 jC in 0.25 SSC, 0.1% SDS. 2.5. Southern blots Ten micrograms of DNA were digested with restriction enzymes overnight. DNA was purified by phenol/chloroform extraction, ethanol precipitated and electrophoresed on a 1% agarose gel. The gel was stained with ethidium bromide and DNA transferred to nylon membrane. Hybridization and wash conditions were the same as for Northern blots. 2.6. Size selected library Genomic Southern blots were probed with cDNA fragments of eEF2 or EFR to identify an appropriate restriction fragment for cloning. Size selected libraries of EcoRI fragments were constructed from T. thermophila strain B2086 genomic DNA. The appropriate size class of DNA was purified from an agarose gel using the Ultra Clean 15 DNA purification kit (MoBio Laboratories). DNA was then ligated into the pZERO vector (Invitrogen) for 30 min and transformed into TOP10 chemically competent cells. Colonies were transferred onto nitrocellulose membranes and identified by colony hybridization (Sambrook et al., 1989). Positive colonies were confirmed by Southern analysis and DNA sequencing.
99
play patterns between mating, starved and vegetatively growing cell cultures. A 10-h time point after cell mixing was chosen to isolate RNA from mating cell cultures because it corresponds to a period of major DNA rearrangements in the developing macronucleus. Differentially expressed DNA bands were purified from the gel, reamplified and used directly to probe Northern blots. These Northern blots contained poly(A) RNA as well as total RNA. One of the first products tested, A46-1 hybridized to a single band in the RNA sample from mating cells (see Fig. 1). Since the differential display technique uses a poly dT primer it was assumed that most display products would be from the polyadenylated fraction. Contrary to expectations, hybridization was much greater to the total RNA fraction than to the poly(A) fraction (compare Fig. 1, lanes 2 and 5). This result was verified with independent RNA isolations and Northern hybridizations using probes against polyadenylated transcripts (data not shown). The data suggest that EFR transcripts lack poly(A) tails. The 735-bp A46-1 differential display product was cloned into the Promega T-Easy vector and sequenced. Analysis of the sequence revealed an open reading frame of approximately 200 amino acids terminated by a UGA stop. Comparison to sequence databases revealed that the A46-1 clone had high identity to the eEF2 protein from other organisms including S. cerevisiae and human. This identity was approximately 45% and corresponded to the
2.7. DNA sequencing DNA sequencing was performed by the Purdue University Genomics Center as well as the Iowa State Sequencing Facility. Sequences were deposited in Genbank under accession numbers AF534908, EF2; AF534909, EFR1; AF534910, EFR2.
3. Results 3.1. Identification of the EFR genes To identify mRNA transcripts specific to the development of the macronucleus we compared differential dis-
Fig. 1. Differential expression of EFR transcripts. The A46-1 fragment isolated from the differential display was analyzed by Northern hybridization of RNA from vegetative, starved and mated cells. Five micrograms of poly(A) (lanes 1 – 3) or total (lanes 4 – 6) were electrophoresed on a 1% agarose/formaldehyde gel. RNA was transferred to nitrocellulose and probed with the A46-1 fragment. Lanes 1 and 4 contain RNA extracted from vegetative cell cultures, 2 and 5 from mating cell cultures at 10 h after mixing, and 3 and 6 from starved cell cultures. V, vegetative; M, mated; N, starved but not mated.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
100
Table 1 Sequence identity comparison between Tetrahymena eEF2 and EFR1, with eEF2 proteins from other organisms eEF2
T. thermophila eEF2
T. thermophila EFR1
T. pyriformis D. melanogaster C. elegans S. cerevisiae H. sapiens
92 66 63 59 65
47 41 42 42 42
Protein sequence identity from several organisms is compared to T. thermophila eEF2 and EFR1. Identity was determined using BLAST searches using the eEF2 nucleotide sequences in translated searches using the ciliate nuclear genetic code.
C-terminal portion of the eEF2 proteins (see Table 1 for comparisons of the full sequence). Because the mating specific expression suggested that A46-1was not a standard eEF2, it was designated EFR for Elongation Factor 2 Related. 3.2. EFR is a multigene family Several observations indicate that multiple transcriptionally active EFR genes exist. First, genomic Southern hybridizations probed with the cloned A46-1 fragment detected multiple bands in restriction digests (Fig. 2, lane 1). The A46-1 fragment did not contain any EcoR1 restriction sites therefore the presence of multiple genes seemed likely. Stronger evidence came from cloning and
Fig. 2. Genomic Southern blot probed with the A46-1 fragment. A total of 10 Ag of genomic DNA were digested with EcoRI (lane 1) or HindIII (lane 2), purified by phenol chloroform extraction and precipitated and electrophoresed on a 1% agarose gel. After transfer to a nylon membrane, samples were probed with the cloned A46-1 fragment, which contains one HindIII site and no EcoRI sites.
sequencing multiple genomic clones. A genomic fragment selected from a recombinant phage lambda library was only 90% identical to the A46-1 clone (data not shown). This clone, designated EFR1, contained a partial gene sequence with 1.2 kb of coding region and 1.8 kb of 3V downstream region. A size selected genomic library containing 3.6-kb EcoR1 fragments was screened to clone the 5V end of the gene and PCR experiments completed the sequence of EFR1. A second library constructed from 2-kb EcoR1 fragments was used to clone a homologous gene and it was designated EFR2. PCR was then used to amplify the entire coding region of the EFR1 and EFR2 genes from genomic DNA. Both genes were sequenced and consist of single open reading frames of 2526 (EFR1) and 2541 (EFR2) nucleotides with no introns (GenBank accession numbers AF534908, EF2; AF534909, EFR1; AF534910, EFR2). Finally, to determine whether multiple EFR genes were transcribed, we performed RT-PCR experiments using primers in regions that were conserved between EFR1 and EFR2. Eleven clones were sequenced and they fell into three different classes of transcripts that include sequences corresponding to EFR1 and EFR2 (data not shown). 3.3. EFR transcripts are developmentally regulated Initial Northern blot analysis showed that EFR genes are developmentally regulated since no transcripts were detected in vegetatively growing cells. To investigate the precise pattern of EFR transcription, we isolated total RNA from vegetative cell cultures as well as mating cell cultures at 0, 4, 6, 8, 10, 12, 14 and 24 h after mixing. Fig. 3A shows the Northern blot when a fragment of EFR1 is used as a probe under the same conditions as the genomic
Fig. 3. Time course analysis of EFR and eEF2 expression. Expression of the EFR and eEF2 genes was analyzed at various times after mating. Total RNA was extracted from vegetative cell cultures and mating cultures at 4, 6, 8, 10, 12, 14 and 24 h after mixing cells. Five micrograms of RNA were electrophoresed on a 1% agarose/formaldehyde gel, and RNA transferred to a nylon membrane. (A) Northern hybridization using the insert from pLEFR1 as a probe (EFR1). (B) The same RNA samples were probed with the pTTEF2A fragment (eEF2). (C) The samples were probed with ribosomal RNA as a loading control. Lanes 1 and 2 were relocated from more distant lanes on the original gel.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Southern blot shown in Fig. 2 (conditions that detect multiple EFR genes). EFR transcripts were not detected in vegetatively growing cells or in mating cells before 4 h. There was a sharp increase in message at 6 h with diminishing levels over the next 18 h. The decrease in EFR transcripts at 8 h (Fig. 3A, lane 5) was reproducible when cells were fed at 7 h after mixing to prevent additional pair formation. If no nutrient broth is added, then the concentration of EFR transcripts decreased gradually over the entire 24-h period (data not shown). The results suggest that EFR mRNA levels are sensitive to the availability of nutrients during conjugation. Failure to add nutrient media after mating results in some EFR transcripts detected after 24 h, presumably this was due to the formation of late mating pairs (data not shown). Interestingly, EFR transcription corresponds to the time when the anlagen or new macronucleus can first be seen (6 h) and precedes the time when most DNA rearrangements occur (12 – 14 h). It is important to note that EFR transcripts were not detected in starved cells that were not mated (Fig. 3A, lane 2). These results are consistent with a role for EFR in macronuclear development. 3.4. Cloning of T. thermophila eEF2 Analysis of the EFR1 and EFR2 sequences revealed high identity to eEF2 from human, yeast and other
101
eukaryotic organisms (Table 1). Since EFR transcripts were not detected in vegetatively growing cells, it seemed unlikely that the EFR genes code for the only elongation factor 2 proteins in T. thermophila. The vegetatively expressed eEF2 gene was cloned and sequenced in order to compare the EFR gene products to vegetative eEF2. A partial eEF2 gene sequence from Tetrahymena pyriformis, along with a fragment of the eEF2 gene from Paramecium tetraurelia (J. Forney, unpublished data) were used to select degenerate primers for PCR amplification of the T. thermophila eEF2. A 300-bp product was amplified, cloned and sequenced. The nucleotide sequence was 92% identical to the T. pyriformis eEF2 gene. This fragment was then used in a genomic Southern blot to determine an appropriate restriction digest for construction of a library. Each restriction digest resulted in hybridization to a single band and a 3.5-kb EcoRI fragment was chosen for isolation (data not shown). A 3.5-kb size selected library was constructed and clones containing the eEF2 gene were isolated. The resulting plasmid (pTTEF2) contains the entire eEF2 open reading frame of 2514 nucleotides plus an intron of 92 bp between nucleotides 2289 and 2290 (from the start of translation). The deduced amino acid sequence is approximately 60% identical to eEF2 proteins from other organisms (Table 1). Comparison of the amino acid sequence from EFR1 with eEF2 reveals 47% identity (Table 1).
Fig. 4. Protein sequence alignment of EFR1, EFR2 and eEF2. Deduced protein sequences from EFR1, EFR2 and eEF2 are compared using ClustalW alignments.
102
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
To compare the expression of eEF2 relative to EFR, the steady state mRNA levels of eEF2 were analyzed during conjugation from the same RNA samples used to analyze EFR. The Northern blot in Fig. 3B shows that eEF2 transcripts are abundant in vegetative cells but substantially decreased during the first few hours of mating (Fig. 3B, compare lanes 1 and 3). From 10 to 24 h, eEF2 transcripts increase in abundance, while EFR transcripts decrease. The data demonstrate an inverse correlation between the abundance of EFR and eEF2 transcripts. 3.5. Sequence analysis reveals differences in regulatory regions of EF2 and EFR Analysis of the EFR sequence is consistent with its function as a translation elongation factor. The deduced amino acid sequences show an overall identity of 47% between EFR1 and T. thermophila eEF2. Fig. 4 shows that the amino acid identity between T. thermophila eEF2, EFR1 and EFR2 is spread throughout the entire sequence
and not limited to a single functional domain. In addition, the two genes have open reading frames of almost identical length (838 and 842 amino acids for eEF2 and EFR1, respectively) and both contain conserved GTP binding motifs (approximately residues 20 – 40, see Fig. 4). In contrast to these similarities, key differences between the sequences suggest altered modes of regulation by posttranslational modification. As described previously, reversible phosphorylation and ADP-ribosylation regulate the activity of eEF2 protein. Phosphorylation of a conserved Thr residue in mammals and yeast (Thr-56 in mammals and Thr-57 in yeast) leads to reduced elongation rates and this event is induced by amino acid starvation (Wang et al., 1998). As can be seen in Fig. 5A, Thr-57 is conserved in T. thermophila eEF2, but EFR1 and EFR2 both contain a methionine at the corresponding position. Although Thr57 is highly conserved in multicellular animals and many protozoa (e.g. Plasmodium and Giardia), Fig. 5A shows that some additional protists (plus Candida albicans) contain a Met in this position (asterisk in
Fig. 5. Comparison of functional regions between eEF2 and EFR. The important functional regions of eEF2 are compared with EFR1 and EFR2. (A) GTPase effector domains are compared. The conserved threonine residue (57) that is phosphorylated in yeast and mammals is shown with an asterisk. (B) The ADPribosylation domain is shown. The conserved histidine residue that is converted to diphthamide in yeast is bold and the conserved glycine residue required for ADP-ribosylation is underlined. Homo, Homo sapiens, Z11692; Araba, Arabadopsis thaliana, 6056373; Caenorab, Caenorhabditis elegans, Z81068; Xenopus, Xenopus laevis, AAH44327; Drosoph, Drosophila melanogaster, X15805; Blasto, Blastocystis hominis, D79219, Bonnem, Bonnemaisonia hamifera, AAG40109; Chlorel, Chlorella kessleri, M68064; Chondrus, Chondrus crispus, AAF71704; Cyanodi, Cyanidioschyzon merolae, BAC67668; Entamoeb, Entamoeba histolytica, L02417; Crypto, Cryptospridium parvum, U21667; Euglena, E. gracilis, AAF71706; Giardia, Giardia intestinalis, D29835; Neurosp, Neurospora crassa, EAA33050.1; Plasmod, Plasmodium falciparum, BAA97565; Sacchar, Saccharomyces cerevisiae, NP014776; Trichom, Trichomonas tenax, D78480; Stylon, Stylonychia mytilus, AAF71704; Tetra, T. thermophila, AF534908; Candida, C. albicans, Y09664; Dicty, Dictyostelium discoideum, M26017; Tetrap, T. pyriformis, AAF71708; Leish, Leishmania major, LmjF36.0190; Tryp, T. cruzi, D50806; EFR1, AF534909; EFR2, AF534910.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Fig. 5A). The significance of this observation is not entirely clear, but it is possible that the Met57 group does not regulate EF2 by the same mechanism as humans and yeast. ADP-ribosylation may also have differential effects on eEF2 and EFR. Although best known as a modification by exogenous toxins (i.e. diphtheria toxin), endogenous ADPribosylation activity has been reported in mammalian cells and could be used to inactivate eEF2 as a regulatory mechanism (Fendrick et al., 1992). The conversion of a conserved histidine residue to diphthamide is required prior to ADP-ribosylation. Both eEF2 and EFR contain the conserved His, but experiments in mammalian cells and yeast have revealed that substitution of arginine for a conserved glycine located two amino acids C-terminal to the diphthamide inhibits ADP-ribosylation (Kohno and Uchida, 1987; Kimata et al., 1993). Interestingly, as can be seen in Fig. 5B, both eEF2 and EFR contain the conserved His, but EFR1 and EFR2 have an Arg residue in place of the conserved Gly. The unusual sequence at this location sets EFR apart from other eukaryotic EF2 molecules, even proteins that contain the Met57 motif in panel A (e.g. Dictyostelium and Trypanosoma cruzi) have the conserved glycine at the ADP-ribosylation site (underlined in Fig. 5B). These observations suggest that EFR is not subject to inhibitory modifications (phosphorylation or ADP-ribosylation) that normally regulate eEF2.
4. Discussion We have identified a developmentally regulated gene family in T. thermophila with high identity to translation elongation factor eEF2. This family was named EFR for Elongation Factor-2 Related because unlike the standard eEF2 gene, transcripts from EFR genes were detected only during conjugation (cell mating). At least three EFR genes are transcriptionally active based on the identification of different classes of RT-PCR products. Genomic Southern blots suggest that no more than five unique genes exist and recent experiments using PCR primers in conserved regions of EFR have successfully amplified three different genes. Analysis of DNA sequence from EFR1 and EFR2 plus RT-PCR products shows greater than 90% identity between the deduced amino acid sequences. This identity is spread relatively evenly across the protein rather than being separated into strongly conserved and more divergent regions that might suggest different functional classes of EFR. At this point, we cannot speculate on the need for multiple EFR genes especially in light of the evidence supporting a single eEF2 gene per haploid genome. The events of conjugation in Tetrahymena and other ciliated protozoa have generated considerable interest because of the extensive genome remodeling that occurs during the formation of the macronuclear genome (Jahn and Klobutcher, 2002). Differentiation of the developing
103
macronucleus begins approximately 6– 7 h after the start of conjugation and DNA rearrangements begin at approximately 12 h post-conjugation. Many of the cis-acting regulatory elements involved in specific DNA rearrangements have been investigated, but only a limited number of proteins involved in macronuclear development have been identified (Nikiforov et al., 2000 and references therein). Although it is unlikely that EFR is directly involved in genome remodeling, the rapid increase in EFR transcript levels 6– 8 h after the initiation of conjugation suggests a role in macronuclear development since it corresponds to the start of macronuclear differentiation. The potential involvement of EFR in a translation regulatory mechanism during sexual reproduction remains an interesting possibility and we present several models for EFR function at the end of this discussion. Regulation of eEF2 is known to involve phosphorylation and ADP-ribosylation, but developmentally regulated eEF2 isoforms have not been reported. S. cerevisiae contains two genes for eEF2, but they encode identical proteins (Perentesis et al., 1992). Similarly, Drosophila contains two nearly identical eEF2 genes and no differences in expression have been observed (Lasko, 2000). Nevertheless, there are examples in multicellular eukaryotes of differential expression of isoforms of translation elongation factor 1A (eEF1A). In Drosophila and Xenopus, there is evidence that one EF1A gene is continuously expressed and one or two others are developmentally regulated (Hovemann et al., 1988, Dje´ et al., 1990). In mammals, the eEF1A2 gene is expressed in heart and muscle tissue and eventually becomes the only isoform expressed in these tissues (Knudsen et al., 1993). Disruption of the eEF1A2 gene in mouse results in deficiencies in muscle and neuronal function within a few weeks after birth (Chambers et al., 1998). The biochemical rationale for different elongation factors has not been established, but Tetrahymena provides a system with strong biochemical and genetic tools (homologous gene replacement) to analyze the function of EFR. The sequence data obtained for EFR is consistent with its role as a translation elongation factor. However, the transcription pattern of the EFR genes, along with the differences in regulatory regions of the protein points to a specialized role in translation. Although sequence comparisons show that EFR is not unique in the substitution of Met57 for the conserved Thr57, a relatively small subset of protozoa uses this motif. A previous study constructed a maximum likelihood tree based on EF2 sequences and the results do not support grouping of the Met57 and Thr57 sequences based on phylogenetics (Moreira et al., 2000). For example, the tree groups together Euglena gracilis and T. cruzi, yet the two organisms differ at position 57. Surprisingly, T. pyriformis has an EFR motif at position 57 yet has greater overall similarity to T. thermophila EF2. It is difficult to interpret this result because the T. pyriformis gene was PCR amplified from genomic DNA and there is no expression data for the clone. Second, T. pyriformis is a
104
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
nonmating species therefore it does not have the same biological context as T. thermophila. Regardless of the sequence details, it is clear that EFR expression is segregated to conjugation, key regulatory sequences are divergent and the functional significance of the protein deserves additional study. In order to place the significance of EFR in its biological context, we present two hypotheses for EFR function that will be evaluated as part of future research efforts. The first model is based on the premise that EFR proteins function as normal eEF2, but they remain active during conjugation under conditions that would inhibit eEF2. It has been established that eEF2 is regulated in response to nutrient availability. Mammalian cells respond to amino acid withdrawal with increased phosphorylation of EF2 and inhibition of elongation (Wang et al., 1998). Since Tetrahymena cells mate in response to decreased nutrient levels, it is possible that vegetative eEF2 protein is phosphorylated and inactivated during conjugation. We observed that the steadystate level of eEF2 mRNA decreases during the initial stages of conjugation (Fig. 3); therefore, it is possible that a combination of lower transcript levels and phosphorylation of EF2 inhibit standard EF2 during conjugation. Yet protein synthesis continues during conjugation as demonstrated with in vivo labeling studies of nuclear proteins (Nikiforov et al., 2000 and references therein). The regulatory model predicts that the EFR proteins are not inhibited by phosphorylation (they do not contain the conserved theronine present in eEF2) and allow protein synthesis to continue during the early phase of sexual reproduction. Although this model is consistent with many of our observations, it is important to note that low levels of eEF2 transcripts are detected in both mating and starved cells up to 36 h after starvation is initiated (T. Malave, unpublished data). Therefore, it is unlikely that all of the eEF2 protein is inhibited during starvation and mating. The proposed role of EFR is to overcome the starvation induced eEF2 inhibition during a specific period of conjugation, then eEF2 takes over protein synthesis as EFR levels begin to diminish. This model has further implications since it is known that amino acid deficiency (starvation) of mammalian cells alters the phosphorylation state of many translation factors including initiation factors (Wang et al., 1998). If this model is correct, then other conjugation specific translation factors may be expected. Our second, more speculative model proposes that EFR proteins are components of special translation machinery that results in enhanced translation of specific transcripts. For example, EFR proteins might increase the frequency of frame shifting to allow synthesis of specific proteins that require programmed frame shifts. Although this might seem a rather unlikely hypothesis, the ciliates have an unusual life style that could benefit from such a scheme. During macronuclear development, extensive genome reorganization takes place which includes site specific DNA splicing, chromosome fragmentation and DNA amplification. These
are potentially dangerous enzymatic activities in a vegetative cell that could threaten the integrity of the germline DNA, or the vegetative macronuclear DNA. It is reasonable to assume that the proteins involved in macronuclear development are tightly regulated and this could include controls at the level of translation as well as transcription. Programmed ribosomal frameshifting is used predominantly by RNA viruses to induce reading frame shifts in response to cis-acting mRNA sequences (reviewed in Gesteland and Atkin, 1996). Nevertheless, several examples of proteins requiring ribosomal frameshifts have been identified in eukaryotes. These include genes encoding a protein kinase and a telomerase component in the ciliate Euplotes (Aigner et al., 2000) as well as a protein component of yeast telomerase (Morris and Lundblad, 1997). Although no frameshifts have been identified in Tetrahymena genes, a complete genome is not available and few genes have been sequenced. The specialized translation hypothesis also requires that elongation factors can alter the frequency of reading frame shifts and indirect evidence supports this notion. A search for dominant suppressors of a + 1 insertion in the yeast MET2 gene revealed a series of mutations in eEF1A (Sandbaken and Culbertson, 1988). Further analysis showed that different mutant alleles have different effects on the frequency of + 1 and 1 frame shifts (Dinmann and Goss Kizny, 1997). A recent report also demonstrates that mutant alleles of eEF2 from yeast selected for resistance to sordarin impair ribosomal frameshifting (Harger et al., 2001). These observations reveal the importance of the elongation process in frameshifting. Whether this is an indirect effect of pausing or a direct result of altered translocation remains to be determined. In the frameshift hypothesis, EFR protein would allow the synthesis of proteins that require a programmed frameshift. for synthesis. During vegetative growth even if the transcripts were present, the conjugation specific proteins would not be synthesized in the absence of the EFR proteins. The regulatory system would provide an extra measure of protection for the genome during vegetative growth. Although we consider it unlikely, it is possible that the EFR proteins have a function unrelated to translation. There is precedence for this hypothesis from studies of eEF1A. Multiple reports indicate that the alpha subunit of EF1 associates with calmodulin and regulates activities that include bundling microtubules (Durso and Cyr, 1994), bundling F-actin (Kurasawa et al., 1996) and ubiquitin-dependent degradation of N-acetylated proteins (Gonen et al., 1994). Although there are no reports of EF2 acting as a multifunctional protein, it remains possible that EFR evolved a different function from eEF2, while maintaining significant sequence similarity. Experiments using gene replacement and disruption techniques combined with biochemical analysis of EFR proteins will allow us to distinguish between the models described above.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Acknowledgements This research was supported by NSF award 0112260MCB to JDF and a Ford Foundation Predoctoral Fellowship for Minorities to TMM. This is paper number 16962 from the Purdue Agricultural Experiment Station. The nucleotide sequence(s) reported in this paper has been submitted to GenBankTM with accession numbers AF534908, AF537909, AF534910.
References Aigner, S., Lingner, J., Goodrich, K.J., Grosshans, C.A., Shevchenko, A., Mann, M., Cech, T.R., 2000. Euplotes telomerase contains an La motif protein produced by apparent translation frameshifting. EMBO J. 19, 6230 – 6239. Campbell, L.E., Wang, X., Proud, C.G., 1999. Nutrients differentially regulate multiple translation factors and their control by insulin. Biochem. J. 344, 433 – 441. Carlberg, U., Nilsson, A., Nygard, O., 1990. Functional properties of phosphorylated elongation factor 2. Eur. J. Biochem. 191, 639 – 645. Chambers, D.M., Peters, J., Abbott, C.M., 1998. The lethal mutation of the mouse wasted (wst) is a deletion that abolishes expression of a tissuespecific isoform of translation elongation factor 1alpha, encoded by the EEf1a2 gene. Proc. Natl. Acad. Sci. U. S. A. 95, 4463 – 4468. Dinmann, J.D., Goss Kizny, T., 1997. Translational misreading: mutations in translation elongation factor 1a differentially affect programmed ribosomal frameshifting and drug sensitivity. RNA 3, 870 – 881. Dje´, M.K., Mazabraud, A., Viel, A., le Marie, M., Denis, H., Crawford, E., Brown, D.D., 1990. Three genes under different developmental control encode elongation factor 1-a in Xenopus laevis. Nucleic Acids Res. 18, 3489 – 3493. Durso, N.A., Cyr, R.J., 1994. A calmodulin-sensitive interaction between microtubules and a higher plant homolog of elongation factor-1 alpha. Plant Cell 9, 893 – 905. Fendrick, J.L., Iglewski, W.J., Moehring, J.M., Moehring, T.J., 1992. Characterization of the endogeneous ADP-ribosylation of mutant and wild-type elongation factor 2 in eukaryotic cells. Eur. J. Biochem. 205, 25 – 31. Gesteland, R.F., Atkins, J.F., 1996. Recoding: dynamic reprogramming of translation. Ann. Rev. Biochem. 65, 741 – 768. Gonen, H., Smith, C.E., Siegel, N.R., Kahana, C., Merrick, W.C., Chakraburtty, K., Schwartz, A.L., Ciechanover, A., 1994. Protein synthesis elongation factor EF-1 alpha is essential for ubiquitin-dependent degradation of certain N alpha-acetylated proteins and may be substituted for by the bacterial elongation factor EF-Tu. Proc. Natl. Acad. Sci. 91, 7648 – 7652. Hovemman, B., Richter, S., Waldorf, U., Cziepluch, C., 1988. Two genes encode related cytoplasmic elongation factors 1a (EF-1a) in Drosophila melanogaster with continuous and stage specific expression. Nucleic Acids Res. 16, 3175 – 3194. Harger, J.W., Meskauskas, A., Nielsen, J., Justice, M.C., Dinman, J.D., 2001. Ty1 retrotransposition and programmed + 1 ribosomal frameshifting require integrity of the protein synthetic translocation step. Virology 286, 216 – 224.
105
Jahn, C.L., Klobutcher, L.A., 2002. Genome remodeling in ciliated protozoa. Annu. Rev. Microbiol. 56, 489 – 520. Kimata, Y., Harashima, S., Kohno, K., 1993. Expression of non-ADP-ribosylatable, diphtheria toxin-resistant elongation factor 2 in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 191, 1145 – 1151. Knudsen, S.M., Frydenberg, J., Clark, B.F., Leffers, H., 1993. Tissue-dependent variation in the expression of elongation factor-1 alpha isoforms: isolation and characterisation of a cDNA encoding a novel variant of human elongation-factor 1 alpha. Eur. J. Biochem. 215, 549 – 554. Kohno, K., Uchida, T., 1987. Highly frequent single amino acid substitution in mammalian elongation factor 2 (EF-2) results in expression of resistance to EF-2-ADP-ribosylating toxins. J. Biol. Chem. 262, 12298 – 12305. Kurasawa, Y., Hanyu, K., Watanabe, Y., Numata, O., 1996. F-Actin bundling activity of Tetrahymena elongation factor 1a is regulated by Ca++/calmodulin. J. Biochem. 119, 791 – 798. Lasko, P., 2000. The Drosophila melanogaster genome: translation factors and RNA binding proteins. J. Cell Biol. 150, 51 – 56. Mackie, K.P., Nairn, A.C., Hampel, G., Lam, G., Jaffe, E.A., 1989. Thrombin and histamine stimulate phosphorylation of elongation factor 2 in human umbilical vein endothelial cells. J. Biol. Chem. 264, 1748 – 1753. Mathews, M.B., Sonenberg, N, Hershey, J.W.B., 2000. Origins and principles of translational control. In: Sonenberg, N., Hershey, J.W.B., Mathews, M.B. (Eds.), Translational Control of Gene Expression. Cold Spring Harbor Laboratory Press, New York, pp. 1 – 32. Moreira, D., LeGuyader, H., Phillippe, H., 2000. The origin of red algae and the evolution of chloroplasts. Nature 405, 69 – 72. Morris, D.K., Lundblad, V., 1997. Programmed translational frameshifting in a gene required for yeast telomere replication. Curr. Biol. 7, 969 – 976. Nairn, A.C., Palfrey, H.C., 1987. Identification of the major Mr 1,000,000 substrate for calmodulin dependent protein kinase III in mammalian cells as elongation factor 2. J. Biol. Chem. 262, 17299 – 17303. Nikiforov, M.A., Gorovsky, M.A., Allis, C.D., 2000. A novel chromodomail protein, Pddp3p, associates with internal eliminated sequences during macronuclear development in Tetrahymena thermophila. Mol. Cell. Biol. 20, 4128 – 4134. Perentesis, J.P., Phan, L.D., Gleason, W.B., LaPorte, D.C., Livingston, D.M., Bodley, J.W., 1992. Saccharomyces cerevisiae elongation factor 2. J. Biol. Chem. 267, 1190 – 1197. Redpath, N.C., Price, N.T., Severinov, K.V., Proud, C.G., 1993. Regulation of elongation factor-2 by multisite phosphorylation. Eur. J. Biochem. 213, 689 – 699. Redpath, N.C., Foulstone, E.J., Proud, C.G., 1996. Regulation of translation elongation factor-2 by insulin via a rapamycin sensitive pathway. EMBO J. 15, 2291 – 2297. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, New York. Sandbaken, M.G., Culbertson, M.R., 1988. Mutations in elongation factor EF-1alpha affect the frequency of frameshifting and amino acid misincorporation in Saccharomyces cerevisiae. Genetics 120, 923 – 934. Wang, X., Campbell, L., Miller, C., Proud, C., 1998. Amino acid availability regulates p70 S6 kinase and multiple translation factors. Biochem. J. 334, 261 – 267.
Identification of a developmentally regulated translation elongation factor 2 in Tetrahymena thermophila Tania M. Malave´ 1, James D. Forney * Department of Biochemistry, Purdue University, West Lafayette, IN 47907 USA Received 28 April 2003; received in revised form 6 August 2003; accepted 15 October 2003 Received by A. Roger
Abstract Protein synthesis elongation factor 2 (eEF2) catalyzes the translocation of the peptidyl-tRNA from the A site to the P site of the ribosome. Most organisms encode a single EF2 protein and its activity is regulated by phosphorylation. We have identified a family of genes in Tetrahymena thermophila that encode proteins homologous to eEF2, yet are expressed only during sexual reproduction. These genes have been designated EFR for Elongation Factor 2 Related. EFR transcripts were not detected in vegetative cell cultures but rapidly increased about 6 h after the start of conjugation (mating). For comparison, we cloned, sequenced and analyzed the expression of the standard eEF2 gene from T. thermophila. Unlike EFR, transcripts from eEF2 were detected in vegetative cells but were present at lower concentrations during conjugation. Despite the high sequence identity between EFR and eEF2 from other organisms (about 42% at the amino acid level), key regulatory sequences that are involved in the regulation of eEF2 are altered in EFR. The sequence and expression data suggest that EFR is an eEF2 variant involved in a major translation regulatory mechanism that occurs during the formation of the macronuclear genome in conjugating cells. D 2003 Elsevier B.V. All rights reserved. Keywords: EF2; Protein synthesis; Ciliate; Protozoa
1. Introduction Protein synthesis is essential for the growth and development of all organisms and the components of the translational machinery have maintained a high degree of conservation throughout evolution. The synthesis of polypeptides consists of three major steps: (1) initiation in which mRNA comes into contact with the ribosome, the start codon is recognized and protein synthesis begins; (2) elongation, during which the polypeptide chain grows and the ribosome moves along the mRNA; and (3) termination,
Abbreviations: eEF2, eukaryotic translation elongation factor 2; eEF1A, eukaryotic translation elongation factor 1A; EFR, elongation factor 2 related; TBE, Tris borate EDTA; MOPS, 3(N-morpholino)propane-sulfonic acid; SDS, sodium dodecyl sulfate. * Corresponding author. Tel.: +1-765-494-1632; fax: +1-765-4947897. E-mail address: [email protected] (J.D. Forney). 1 Current address: Department of Molecular and Human Genetics, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA. 0378-1119/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2003.10.016
when a stop codon is recognized and the translation complex is disassembled. Protein synthesis is influenced by conditions that affect growth, such as availability of nutrients, stress, development and differentiation (reviewed in Mathews et al., 2000). Thus far, most examples of translational regulation involve initiation and fewer instances of control at the level of elongation are known. Translation elongation factor 2, eEF2, is a GTPase involved in the translocation of the peptidyl-tRNA from the A site to the P site on the ribosome. The 95-kDa protein is highly conserved, with 60% amino acid sequence identity between the human and yeast proteins. In Saccharomyces cerevisiae, there are two genes for eEF2, which encode identical proteins. Gene knockout experiments reveal that either gene can be eliminated individually without affecting cell viability, but double gene knockouts are lethal (Perentesis et al., 1992). Two major mechanisms are known to regulate protein elongation and both involve eEF2. First, eEF2 can be modulated by reversible phosphorylation (Nairn and Palfrey, 1987, Redpath et al., 1993). Increased levels of phosphorylated eEF2 reduce elongation rates
98
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
presumably because phosphorylated eEF2 fails to bind the ribosomes (Carlberg et al., 1990). Treatment of mammalian cells with agents that raise the cytoplasmic Ca2 + and cAMP levels reduce elongation rates by activating the kinase responsible for phosphorylating eEF2 (Mackie et al., 1989). In contrast, treatment of cells with insulin increases elongation rates by promoting eEF2 dephosphorylation (Redpath et al., 1996, Campbell et al., 1999). Second, the protein can be post-translationally modified by ADP-ribosylation. This reaction is performed by various bacterial toxins after modification of a specific histidine residue to diphthamide, but there is evidence for an endogenous ADP ribosylase activity (Fendrick et al., 1992). Similar to the bacterial toxins, it is presumed that modification by the endogenous enzyme also inhibits eEF2 activity. Our laboratory has been interested in the developmental events during the formation of the macronuclear genome in ciliated protozoa. Ciliates are unique in that they posses two distinct types of nuclei within a single cell: a somatic macronucleus and a transcriptionally silent micronucleus. During sexual conjugation, the existing macronucleus in the cell is destroyed and a new one is formed from the micronuclear genome (reviewed in Jahn and Klobutcher, 2002). The process of macronuclear maturation involves extensive genome remodeling. Thus far, the only developmentally regulated proteins that have been reported are associated with chromatin and DNA elimination (Nikiforov et al., 2000 and references therein). We have used differential display to identify transcripts unique to conjugation that could code for proteins involved in macronuclear development. We have identified a family of genes in Tetrahymena thermophila that encode for proteins homologous to eEF2 yet are expressed only during sexual reproduction. These genes have been designated EFR for Elongation Factor 2 Related. EFR transcripts appeared during conjugation (mating) but were not detected in vegetative cell cultures. For comparison, we cloned, sequenced and analyzed the expression of the standard eEF2 in T. thermophila. Unlike EFR, transcripts from eEF2 were detected in vegetative cells but decreased in conjugating cells. Despite the high sequence identity (roughly 50% at the amino acid level), there are key differences between the two proteins in conserved regulatory regions. The observations suggest that EFR is involved in a major translation regulatory mechanism that occurs during the formation of the macronuclear genome in conjugating cells.
2. Materials and methods 2.1. Tetrahymena growth and mating Strains CU428.1 and B2086 were obtained from Dr. David Asai (Biological Sciences Department, Purdue University). Cells were grown to a density of 2 105 cells/ml in NEFFs (0.5% dextrose, 0.25% yeast extract, 0.25% proteose
peptone, 3.3 mM FeCl3) media at 30 jC with 85 rpm of shaking. Mating reactive cells were prepared by spinning down 100 ml of a log phase culture (2 – 5 105 cells/ml), resuspending the cell pellet in 100 ml of 10 mM Tris –HCl, pH 7.5, and incubating at 30 jC with shaking (85 rpm) for 12– 24 h. After starving, the cells were counted and equal numbers of cells (at a density of 2 105 cells/ml) were added to a 2-l flask, adjusting the final volume to 100 ml. This flask was incubated at 30 jC, without shaking to allow pairs to form. Cell mixing was defined as time 0 and the cells were checked for mating efficiency at 3 – 4 h. Mating efficiency was calculated by counting the number of cells in pairs and dividing by the total number of cells. Only cultures with over 70% of the cells in pairs were used for further experiments. 2.2. RNA isolation Cells were disrupted using the Qiashredder (Qiagen) and RNA isolated with the Qiagen RNeasy Mini kit (Qiagen) Approximately 10 ml of cells at a density of 2 –5 105 cells/ml were used per column. For the time course experiments, cells were pelleted at each time point and the RNA immediately extracted. The RNA used for the Northern blots was treated with the Qiagen RNase free DNase (Qiagen). Poly(A) RNA was purified using the Promega Poly(A) system 1000 magnetic beads (Promega) using the protocol provided by the manufacturer. 2.3. Differential display Differential display was carried out using the Gene Hunter RNA Image kits (GenHunter). RNA was purified from contaminating DNA using the Message Clean DNase kit (GenHunter). Reverse transcription was performed with 0.1 Ag of total RNA and one of three different oligo dT primers provided in the kit. PCR of the resulting cDNA was then performed using the same oligo dT primer, a specific oligodecamer provided in the GenHunter kit and Qiagen Taq DNA polymerase in the presence of a-33P-dATP. Reactions were mixed with loading dye and electrophoresed on a 6% polyacrylamide gel in 1xTBE. The gel was dried, exposed to film overnight, and bands were excised. DNA was purified by placing the gel fragment in a 1.5-ml tube with water, sealing the tube with paraffin and boiling for 10 min. DNA was then amplified by PCR using the same conditions as in the original differential display but without the radioactive nucleotide. The PCR product was used as a probe for Northern blots and sub-cloned into the pGEM-Teasy vector (Promega). 2.4. Northern blots Five micrograms of RNA were mixed with 5 Al of sample buffer (3.7% formaldehyde and 4 MOPS) and 10 Al of
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
formamide and heated at 65 jC for 5 min. The samples were quick cooled on ice, 5 Al of loading buffer was added and they were electrophoresed on a 1% agarose, 2.5% formaldehyde gel in 1 MOPS (20 mM MOPS, 5 mM NaOAc, 1 mM EDTA, pH 7.0), at 90 V for 2 h. Gels were soaked in 20 SSC for 30 min, and RNA transferred to Nylon membranes, then fixed with UV cross-linking and hybridized. DNA probes were labeled with a-32P-dATP using the RediPrime kit (Amersham), and membranes hybridized overnight in 1 Denhart’s, 0.2 M phosphate buffer, 5 SSC, 0.25% SDS. After hybridization, membranes were washed twice for 15 min at room temperature in 1 SSC, 0.1% SDS and once for 30 min at 65 jC in 0.25 SSC, 0.1% SDS. 2.5. Southern blots Ten micrograms of DNA were digested with restriction enzymes overnight. DNA was purified by phenol/chloroform extraction, ethanol precipitated and electrophoresed on a 1% agarose gel. The gel was stained with ethidium bromide and DNA transferred to nylon membrane. Hybridization and wash conditions were the same as for Northern blots. 2.6. Size selected library Genomic Southern blots were probed with cDNA fragments of eEF2 or EFR to identify an appropriate restriction fragment for cloning. Size selected libraries of EcoRI fragments were constructed from T. thermophila strain B2086 genomic DNA. The appropriate size class of DNA was purified from an agarose gel using the Ultra Clean 15 DNA purification kit (MoBio Laboratories). DNA was then ligated into the pZERO vector (Invitrogen) for 30 min and transformed into TOP10 chemically competent cells. Colonies were transferred onto nitrocellulose membranes and identified by colony hybridization (Sambrook et al., 1989). Positive colonies were confirmed by Southern analysis and DNA sequencing.
99
play patterns between mating, starved and vegetatively growing cell cultures. A 10-h time point after cell mixing was chosen to isolate RNA from mating cell cultures because it corresponds to a period of major DNA rearrangements in the developing macronucleus. Differentially expressed DNA bands were purified from the gel, reamplified and used directly to probe Northern blots. These Northern blots contained poly(A) RNA as well as total RNA. One of the first products tested, A46-1 hybridized to a single band in the RNA sample from mating cells (see Fig. 1). Since the differential display technique uses a poly dT primer it was assumed that most display products would be from the polyadenylated fraction. Contrary to expectations, hybridization was much greater to the total RNA fraction than to the poly(A) fraction (compare Fig. 1, lanes 2 and 5). This result was verified with independent RNA isolations and Northern hybridizations using probes against polyadenylated transcripts (data not shown). The data suggest that EFR transcripts lack poly(A) tails. The 735-bp A46-1 differential display product was cloned into the Promega T-Easy vector and sequenced. Analysis of the sequence revealed an open reading frame of approximately 200 amino acids terminated by a UGA stop. Comparison to sequence databases revealed that the A46-1 clone had high identity to the eEF2 protein from other organisms including S. cerevisiae and human. This identity was approximately 45% and corresponded to the
2.7. DNA sequencing DNA sequencing was performed by the Purdue University Genomics Center as well as the Iowa State Sequencing Facility. Sequences were deposited in Genbank under accession numbers AF534908, EF2; AF534909, EFR1; AF534910, EFR2.
3. Results 3.1. Identification of the EFR genes To identify mRNA transcripts specific to the development of the macronucleus we compared differential dis-
Fig. 1. Differential expression of EFR transcripts. The A46-1 fragment isolated from the differential display was analyzed by Northern hybridization of RNA from vegetative, starved and mated cells. Five micrograms of poly(A) (lanes 1 – 3) or total (lanes 4 – 6) were electrophoresed on a 1% agarose/formaldehyde gel. RNA was transferred to nitrocellulose and probed with the A46-1 fragment. Lanes 1 and 4 contain RNA extracted from vegetative cell cultures, 2 and 5 from mating cell cultures at 10 h after mixing, and 3 and 6 from starved cell cultures. V, vegetative; M, mated; N, starved but not mated.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
100
Table 1 Sequence identity comparison between Tetrahymena eEF2 and EFR1, with eEF2 proteins from other organisms eEF2
T. thermophila eEF2
T. thermophila EFR1
T. pyriformis D. melanogaster C. elegans S. cerevisiae H. sapiens
92 66 63 59 65
47 41 42 42 42
Protein sequence identity from several organisms is compared to T. thermophila eEF2 and EFR1. Identity was determined using BLAST searches using the eEF2 nucleotide sequences in translated searches using the ciliate nuclear genetic code.
C-terminal portion of the eEF2 proteins (see Table 1 for comparisons of the full sequence). Because the mating specific expression suggested that A46-1was not a standard eEF2, it was designated EFR for Elongation Factor 2 Related. 3.2. EFR is a multigene family Several observations indicate that multiple transcriptionally active EFR genes exist. First, genomic Southern hybridizations probed with the cloned A46-1 fragment detected multiple bands in restriction digests (Fig. 2, lane 1). The A46-1 fragment did not contain any EcoR1 restriction sites therefore the presence of multiple genes seemed likely. Stronger evidence came from cloning and
Fig. 2. Genomic Southern blot probed with the A46-1 fragment. A total of 10 Ag of genomic DNA were digested with EcoRI (lane 1) or HindIII (lane 2), purified by phenol chloroform extraction and precipitated and electrophoresed on a 1% agarose gel. After transfer to a nylon membrane, samples were probed with the cloned A46-1 fragment, which contains one HindIII site and no EcoRI sites.
sequencing multiple genomic clones. A genomic fragment selected from a recombinant phage lambda library was only 90% identical to the A46-1 clone (data not shown). This clone, designated EFR1, contained a partial gene sequence with 1.2 kb of coding region and 1.8 kb of 3V downstream region. A size selected genomic library containing 3.6-kb EcoR1 fragments was screened to clone the 5V end of the gene and PCR experiments completed the sequence of EFR1. A second library constructed from 2-kb EcoR1 fragments was used to clone a homologous gene and it was designated EFR2. PCR was then used to amplify the entire coding region of the EFR1 and EFR2 genes from genomic DNA. Both genes were sequenced and consist of single open reading frames of 2526 (EFR1) and 2541 (EFR2) nucleotides with no introns (GenBank accession numbers AF534908, EF2; AF534909, EFR1; AF534910, EFR2). Finally, to determine whether multiple EFR genes were transcribed, we performed RT-PCR experiments using primers in regions that were conserved between EFR1 and EFR2. Eleven clones were sequenced and they fell into three different classes of transcripts that include sequences corresponding to EFR1 and EFR2 (data not shown). 3.3. EFR transcripts are developmentally regulated Initial Northern blot analysis showed that EFR genes are developmentally regulated since no transcripts were detected in vegetatively growing cells. To investigate the precise pattern of EFR transcription, we isolated total RNA from vegetative cell cultures as well as mating cell cultures at 0, 4, 6, 8, 10, 12, 14 and 24 h after mixing. Fig. 3A shows the Northern blot when a fragment of EFR1 is used as a probe under the same conditions as the genomic
Fig. 3. Time course analysis of EFR and eEF2 expression. Expression of the EFR and eEF2 genes was analyzed at various times after mating. Total RNA was extracted from vegetative cell cultures and mating cultures at 4, 6, 8, 10, 12, 14 and 24 h after mixing cells. Five micrograms of RNA were electrophoresed on a 1% agarose/formaldehyde gel, and RNA transferred to a nylon membrane. (A) Northern hybridization using the insert from pLEFR1 as a probe (EFR1). (B) The same RNA samples were probed with the pTTEF2A fragment (eEF2). (C) The samples were probed with ribosomal RNA as a loading control. Lanes 1 and 2 were relocated from more distant lanes on the original gel.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Southern blot shown in Fig. 2 (conditions that detect multiple EFR genes). EFR transcripts were not detected in vegetatively growing cells or in mating cells before 4 h. There was a sharp increase in message at 6 h with diminishing levels over the next 18 h. The decrease in EFR transcripts at 8 h (Fig. 3A, lane 5) was reproducible when cells were fed at 7 h after mixing to prevent additional pair formation. If no nutrient broth is added, then the concentration of EFR transcripts decreased gradually over the entire 24-h period (data not shown). The results suggest that EFR mRNA levels are sensitive to the availability of nutrients during conjugation. Failure to add nutrient media after mating results in some EFR transcripts detected after 24 h, presumably this was due to the formation of late mating pairs (data not shown). Interestingly, EFR transcription corresponds to the time when the anlagen or new macronucleus can first be seen (6 h) and precedes the time when most DNA rearrangements occur (12 – 14 h). It is important to note that EFR transcripts were not detected in starved cells that were not mated (Fig. 3A, lane 2). These results are consistent with a role for EFR in macronuclear development. 3.4. Cloning of T. thermophila eEF2 Analysis of the EFR1 and EFR2 sequences revealed high identity to eEF2 from human, yeast and other
101
eukaryotic organisms (Table 1). Since EFR transcripts were not detected in vegetatively growing cells, it seemed unlikely that the EFR genes code for the only elongation factor 2 proteins in T. thermophila. The vegetatively expressed eEF2 gene was cloned and sequenced in order to compare the EFR gene products to vegetative eEF2. A partial eEF2 gene sequence from Tetrahymena pyriformis, along with a fragment of the eEF2 gene from Paramecium tetraurelia (J. Forney, unpublished data) were used to select degenerate primers for PCR amplification of the T. thermophila eEF2. A 300-bp product was amplified, cloned and sequenced. The nucleotide sequence was 92% identical to the T. pyriformis eEF2 gene. This fragment was then used in a genomic Southern blot to determine an appropriate restriction digest for construction of a library. Each restriction digest resulted in hybridization to a single band and a 3.5-kb EcoRI fragment was chosen for isolation (data not shown). A 3.5-kb size selected library was constructed and clones containing the eEF2 gene were isolated. The resulting plasmid (pTTEF2) contains the entire eEF2 open reading frame of 2514 nucleotides plus an intron of 92 bp between nucleotides 2289 and 2290 (from the start of translation). The deduced amino acid sequence is approximately 60% identical to eEF2 proteins from other organisms (Table 1). Comparison of the amino acid sequence from EFR1 with eEF2 reveals 47% identity (Table 1).
Fig. 4. Protein sequence alignment of EFR1, EFR2 and eEF2. Deduced protein sequences from EFR1, EFR2 and eEF2 are compared using ClustalW alignments.
102
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
To compare the expression of eEF2 relative to EFR, the steady state mRNA levels of eEF2 were analyzed during conjugation from the same RNA samples used to analyze EFR. The Northern blot in Fig. 3B shows that eEF2 transcripts are abundant in vegetative cells but substantially decreased during the first few hours of mating (Fig. 3B, compare lanes 1 and 3). From 10 to 24 h, eEF2 transcripts increase in abundance, while EFR transcripts decrease. The data demonstrate an inverse correlation between the abundance of EFR and eEF2 transcripts. 3.5. Sequence analysis reveals differences in regulatory regions of EF2 and EFR Analysis of the EFR sequence is consistent with its function as a translation elongation factor. The deduced amino acid sequences show an overall identity of 47% between EFR1 and T. thermophila eEF2. Fig. 4 shows that the amino acid identity between T. thermophila eEF2, EFR1 and EFR2 is spread throughout the entire sequence
and not limited to a single functional domain. In addition, the two genes have open reading frames of almost identical length (838 and 842 amino acids for eEF2 and EFR1, respectively) and both contain conserved GTP binding motifs (approximately residues 20 – 40, see Fig. 4). In contrast to these similarities, key differences between the sequences suggest altered modes of regulation by posttranslational modification. As described previously, reversible phosphorylation and ADP-ribosylation regulate the activity of eEF2 protein. Phosphorylation of a conserved Thr residue in mammals and yeast (Thr-56 in mammals and Thr-57 in yeast) leads to reduced elongation rates and this event is induced by amino acid starvation (Wang et al., 1998). As can be seen in Fig. 5A, Thr-57 is conserved in T. thermophila eEF2, but EFR1 and EFR2 both contain a methionine at the corresponding position. Although Thr57 is highly conserved in multicellular animals and many protozoa (e.g. Plasmodium and Giardia), Fig. 5A shows that some additional protists (plus Candida albicans) contain a Met in this position (asterisk in
Fig. 5. Comparison of functional regions between eEF2 and EFR. The important functional regions of eEF2 are compared with EFR1 and EFR2. (A) GTPase effector domains are compared. The conserved threonine residue (57) that is phosphorylated in yeast and mammals is shown with an asterisk. (B) The ADPribosylation domain is shown. The conserved histidine residue that is converted to diphthamide in yeast is bold and the conserved glycine residue required for ADP-ribosylation is underlined. Homo, Homo sapiens, Z11692; Araba, Arabadopsis thaliana, 6056373; Caenorab, Caenorhabditis elegans, Z81068; Xenopus, Xenopus laevis, AAH44327; Drosoph, Drosophila melanogaster, X15805; Blasto, Blastocystis hominis, D79219, Bonnem, Bonnemaisonia hamifera, AAG40109; Chlorel, Chlorella kessleri, M68064; Chondrus, Chondrus crispus, AAF71704; Cyanodi, Cyanidioschyzon merolae, BAC67668; Entamoeb, Entamoeba histolytica, L02417; Crypto, Cryptospridium parvum, U21667; Euglena, E. gracilis, AAF71706; Giardia, Giardia intestinalis, D29835; Neurosp, Neurospora crassa, EAA33050.1; Plasmod, Plasmodium falciparum, BAA97565; Sacchar, Saccharomyces cerevisiae, NP014776; Trichom, Trichomonas tenax, D78480; Stylon, Stylonychia mytilus, AAF71704; Tetra, T. thermophila, AF534908; Candida, C. albicans, Y09664; Dicty, Dictyostelium discoideum, M26017; Tetrap, T. pyriformis, AAF71708; Leish, Leishmania major, LmjF36.0190; Tryp, T. cruzi, D50806; EFR1, AF534909; EFR2, AF534910.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Fig. 5A). The significance of this observation is not entirely clear, but it is possible that the Met57 group does not regulate EF2 by the same mechanism as humans and yeast. ADP-ribosylation may also have differential effects on eEF2 and EFR. Although best known as a modification by exogenous toxins (i.e. diphtheria toxin), endogenous ADPribosylation activity has been reported in mammalian cells and could be used to inactivate eEF2 as a regulatory mechanism (Fendrick et al., 1992). The conversion of a conserved histidine residue to diphthamide is required prior to ADP-ribosylation. Both eEF2 and EFR contain the conserved His, but experiments in mammalian cells and yeast have revealed that substitution of arginine for a conserved glycine located two amino acids C-terminal to the diphthamide inhibits ADP-ribosylation (Kohno and Uchida, 1987; Kimata et al., 1993). Interestingly, as can be seen in Fig. 5B, both eEF2 and EFR contain the conserved His, but EFR1 and EFR2 have an Arg residue in place of the conserved Gly. The unusual sequence at this location sets EFR apart from other eukaryotic EF2 molecules, even proteins that contain the Met57 motif in panel A (e.g. Dictyostelium and Trypanosoma cruzi) have the conserved glycine at the ADP-ribosylation site (underlined in Fig. 5B). These observations suggest that EFR is not subject to inhibitory modifications (phosphorylation or ADP-ribosylation) that normally regulate eEF2.
4. Discussion We have identified a developmentally regulated gene family in T. thermophila with high identity to translation elongation factor eEF2. This family was named EFR for Elongation Factor-2 Related because unlike the standard eEF2 gene, transcripts from EFR genes were detected only during conjugation (cell mating). At least three EFR genes are transcriptionally active based on the identification of different classes of RT-PCR products. Genomic Southern blots suggest that no more than five unique genes exist and recent experiments using PCR primers in conserved regions of EFR have successfully amplified three different genes. Analysis of DNA sequence from EFR1 and EFR2 plus RT-PCR products shows greater than 90% identity between the deduced amino acid sequences. This identity is spread relatively evenly across the protein rather than being separated into strongly conserved and more divergent regions that might suggest different functional classes of EFR. At this point, we cannot speculate on the need for multiple EFR genes especially in light of the evidence supporting a single eEF2 gene per haploid genome. The events of conjugation in Tetrahymena and other ciliated protozoa have generated considerable interest because of the extensive genome remodeling that occurs during the formation of the macronuclear genome (Jahn and Klobutcher, 2002). Differentiation of the developing
103
macronucleus begins approximately 6– 7 h after the start of conjugation and DNA rearrangements begin at approximately 12 h post-conjugation. Many of the cis-acting regulatory elements involved in specific DNA rearrangements have been investigated, but only a limited number of proteins involved in macronuclear development have been identified (Nikiforov et al., 2000 and references therein). Although it is unlikely that EFR is directly involved in genome remodeling, the rapid increase in EFR transcript levels 6– 8 h after the initiation of conjugation suggests a role in macronuclear development since it corresponds to the start of macronuclear differentiation. The potential involvement of EFR in a translation regulatory mechanism during sexual reproduction remains an interesting possibility and we present several models for EFR function at the end of this discussion. Regulation of eEF2 is known to involve phosphorylation and ADP-ribosylation, but developmentally regulated eEF2 isoforms have not been reported. S. cerevisiae contains two genes for eEF2, but they encode identical proteins (Perentesis et al., 1992). Similarly, Drosophila contains two nearly identical eEF2 genes and no differences in expression have been observed (Lasko, 2000). Nevertheless, there are examples in multicellular eukaryotes of differential expression of isoforms of translation elongation factor 1A (eEF1A). In Drosophila and Xenopus, there is evidence that one EF1A gene is continuously expressed and one or two others are developmentally regulated (Hovemann et al., 1988, Dje´ et al., 1990). In mammals, the eEF1A2 gene is expressed in heart and muscle tissue and eventually becomes the only isoform expressed in these tissues (Knudsen et al., 1993). Disruption of the eEF1A2 gene in mouse results in deficiencies in muscle and neuronal function within a few weeks after birth (Chambers et al., 1998). The biochemical rationale for different elongation factors has not been established, but Tetrahymena provides a system with strong biochemical and genetic tools (homologous gene replacement) to analyze the function of EFR. The sequence data obtained for EFR is consistent with its role as a translation elongation factor. However, the transcription pattern of the EFR genes, along with the differences in regulatory regions of the protein points to a specialized role in translation. Although sequence comparisons show that EFR is not unique in the substitution of Met57 for the conserved Thr57, a relatively small subset of protozoa uses this motif. A previous study constructed a maximum likelihood tree based on EF2 sequences and the results do not support grouping of the Met57 and Thr57 sequences based on phylogenetics (Moreira et al., 2000). For example, the tree groups together Euglena gracilis and T. cruzi, yet the two organisms differ at position 57. Surprisingly, T. pyriformis has an EFR motif at position 57 yet has greater overall similarity to T. thermophila EF2. It is difficult to interpret this result because the T. pyriformis gene was PCR amplified from genomic DNA and there is no expression data for the clone. Second, T. pyriformis is a
104
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
nonmating species therefore it does not have the same biological context as T. thermophila. Regardless of the sequence details, it is clear that EFR expression is segregated to conjugation, key regulatory sequences are divergent and the functional significance of the protein deserves additional study. In order to place the significance of EFR in its biological context, we present two hypotheses for EFR function that will be evaluated as part of future research efforts. The first model is based on the premise that EFR proteins function as normal eEF2, but they remain active during conjugation under conditions that would inhibit eEF2. It has been established that eEF2 is regulated in response to nutrient availability. Mammalian cells respond to amino acid withdrawal with increased phosphorylation of EF2 and inhibition of elongation (Wang et al., 1998). Since Tetrahymena cells mate in response to decreased nutrient levels, it is possible that vegetative eEF2 protein is phosphorylated and inactivated during conjugation. We observed that the steadystate level of eEF2 mRNA decreases during the initial stages of conjugation (Fig. 3); therefore, it is possible that a combination of lower transcript levels and phosphorylation of EF2 inhibit standard EF2 during conjugation. Yet protein synthesis continues during conjugation as demonstrated with in vivo labeling studies of nuclear proteins (Nikiforov et al., 2000 and references therein). The regulatory model predicts that the EFR proteins are not inhibited by phosphorylation (they do not contain the conserved theronine present in eEF2) and allow protein synthesis to continue during the early phase of sexual reproduction. Although this model is consistent with many of our observations, it is important to note that low levels of eEF2 transcripts are detected in both mating and starved cells up to 36 h after starvation is initiated (T. Malave, unpublished data). Therefore, it is unlikely that all of the eEF2 protein is inhibited during starvation and mating. The proposed role of EFR is to overcome the starvation induced eEF2 inhibition during a specific period of conjugation, then eEF2 takes over protein synthesis as EFR levels begin to diminish. This model has further implications since it is known that amino acid deficiency (starvation) of mammalian cells alters the phosphorylation state of many translation factors including initiation factors (Wang et al., 1998). If this model is correct, then other conjugation specific translation factors may be expected. Our second, more speculative model proposes that EFR proteins are components of special translation machinery that results in enhanced translation of specific transcripts. For example, EFR proteins might increase the frequency of frame shifting to allow synthesis of specific proteins that require programmed frame shifts. Although this might seem a rather unlikely hypothesis, the ciliates have an unusual life style that could benefit from such a scheme. During macronuclear development, extensive genome reorganization takes place which includes site specific DNA splicing, chromosome fragmentation and DNA amplification. These
are potentially dangerous enzymatic activities in a vegetative cell that could threaten the integrity of the germline DNA, or the vegetative macronuclear DNA. It is reasonable to assume that the proteins involved in macronuclear development are tightly regulated and this could include controls at the level of translation as well as transcription. Programmed ribosomal frameshifting is used predominantly by RNA viruses to induce reading frame shifts in response to cis-acting mRNA sequences (reviewed in Gesteland and Atkin, 1996). Nevertheless, several examples of proteins requiring ribosomal frameshifts have been identified in eukaryotes. These include genes encoding a protein kinase and a telomerase component in the ciliate Euplotes (Aigner et al., 2000) as well as a protein component of yeast telomerase (Morris and Lundblad, 1997). Although no frameshifts have been identified in Tetrahymena genes, a complete genome is not available and few genes have been sequenced. The specialized translation hypothesis also requires that elongation factors can alter the frequency of reading frame shifts and indirect evidence supports this notion. A search for dominant suppressors of a + 1 insertion in the yeast MET2 gene revealed a series of mutations in eEF1A (Sandbaken and Culbertson, 1988). Further analysis showed that different mutant alleles have different effects on the frequency of + 1 and 1 frame shifts (Dinmann and Goss Kizny, 1997). A recent report also demonstrates that mutant alleles of eEF2 from yeast selected for resistance to sordarin impair ribosomal frameshifting (Harger et al., 2001). These observations reveal the importance of the elongation process in frameshifting. Whether this is an indirect effect of pausing or a direct result of altered translocation remains to be determined. In the frameshift hypothesis, EFR protein would allow the synthesis of proteins that require a programmed frameshift. for synthesis. During vegetative growth even if the transcripts were present, the conjugation specific proteins would not be synthesized in the absence of the EFR proteins. The regulatory system would provide an extra measure of protection for the genome during vegetative growth. Although we consider it unlikely, it is possible that the EFR proteins have a function unrelated to translation. There is precedence for this hypothesis from studies of eEF1A. Multiple reports indicate that the alpha subunit of EF1 associates with calmodulin and regulates activities that include bundling microtubules (Durso and Cyr, 1994), bundling F-actin (Kurasawa et al., 1996) and ubiquitin-dependent degradation of N-acetylated proteins (Gonen et al., 1994). Although there are no reports of EF2 acting as a multifunctional protein, it remains possible that EFR evolved a different function from eEF2, while maintaining significant sequence similarity. Experiments using gene replacement and disruption techniques combined with biochemical analysis of EFR proteins will allow us to distinguish between the models described above.
T.M. Malave´, J.D. Forney / Gene 326 (2004) 97–105
Acknowledgements This research was supported by NSF award 0112260MCB to JDF and a Ford Foundation Predoctoral Fellowship for Minorities to TMM. This is paper number 16962 from the Purdue Agricultural Experiment Station. The nucleotide sequence(s) reported in this paper has been submitted to GenBankTM with accession numbers AF534908, AF537909, AF534910.
References Aigner, S., Lingner, J., Goodrich, K.J., Grosshans, C.A., Shevchenko, A., Mann, M., Cech, T.R., 2000. Euplotes telomerase contains an La motif protein produced by apparent translation frameshifting. EMBO J. 19, 6230 – 6239. Campbell, L.E., Wang, X., Proud, C.G., 1999. Nutrients differentially regulate multiple translation factors and their control by insulin. Biochem. J. 344, 433 – 441. Carlberg, U., Nilsson, A., Nygard, O., 1990. Functional properties of phosphorylated elongation factor 2. Eur. J. Biochem. 191, 639 – 645. Chambers, D.M., Peters, J., Abbott, C.M., 1998. The lethal mutation of the mouse wasted (wst) is a deletion that abolishes expression of a tissuespecific isoform of translation elongation factor 1alpha, encoded by the EEf1a2 gene. Proc. Natl. Acad. Sci. U. S. A. 95, 4463 – 4468. Dinmann, J.D., Goss Kizny, T., 1997. Translational misreading: mutations in translation elongation factor 1a differentially affect programmed ribosomal frameshifting and drug sensitivity. RNA 3, 870 – 881. Dje´, M.K., Mazabraud, A., Viel, A., le Marie, M., Denis, H., Crawford, E., Brown, D.D., 1990. Three genes under different developmental control encode elongation factor 1-a in Xenopus laevis. Nucleic Acids Res. 18, 3489 – 3493. Durso, N.A., Cyr, R.J., 1994. A calmodulin-sensitive interaction between microtubules and a higher plant homolog of elongation factor-1 alpha. Plant Cell 9, 893 – 905. Fendrick, J.L., Iglewski, W.J., Moehring, J.M., Moehring, T.J., 1992. Characterization of the endogeneous ADP-ribosylation of mutant and wild-type elongation factor 2 in eukaryotic cells. Eur. J. Biochem. 205, 25 – 31. Gesteland, R.F., Atkins, J.F., 1996. Recoding: dynamic reprogramming of translation. Ann. Rev. Biochem. 65, 741 – 768. Gonen, H., Smith, C.E., Siegel, N.R., Kahana, C., Merrick, W.C., Chakraburtty, K., Schwartz, A.L., Ciechanover, A., 1994. Protein synthesis elongation factor EF-1 alpha is essential for ubiquitin-dependent degradation of certain N alpha-acetylated proteins and may be substituted for by the bacterial elongation factor EF-Tu. Proc. Natl. Acad. Sci. 91, 7648 – 7652. Hovemman, B., Richter, S., Waldorf, U., Cziepluch, C., 1988. Two genes encode related cytoplasmic elongation factors 1a (EF-1a) in Drosophila melanogaster with continuous and stage specific expression. Nucleic Acids Res. 16, 3175 – 3194. Harger, J.W., Meskauskas, A., Nielsen, J., Justice, M.C., Dinman, J.D., 2001. Ty1 retrotransposition and programmed + 1 ribosomal frameshifting require integrity of the protein synthetic translocation step. Virology 286, 216 – 224.
105
Jahn, C.L., Klobutcher, L.A., 2002. Genome remodeling in ciliated protozoa. Annu. Rev. Microbiol. 56, 489 – 520. Kimata, Y., Harashima, S., Kohno, K., 1993. Expression of non-ADP-ribosylatable, diphtheria toxin-resistant elongation factor 2 in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 191, 1145 – 1151. Knudsen, S.M., Frydenberg, J., Clark, B.F., Leffers, H., 1993. Tissue-dependent variation in the expression of elongation factor-1 alpha isoforms: isolation and characterisation of a cDNA encoding a novel variant of human elongation-factor 1 alpha. Eur. J. Biochem. 215, 549 – 554. Kohno, K., Uchida, T., 1987. Highly frequent single amino acid substitution in mammalian elongation factor 2 (EF-2) results in expression of resistance to EF-2-ADP-ribosylating toxins. J. Biol. Chem. 262, 12298 – 12305. Kurasawa, Y., Hanyu, K., Watanabe, Y., Numata, O., 1996. F-Actin bundling activity of Tetrahymena elongation factor 1a is regulated by Ca++/calmodulin. J. Biochem. 119, 791 – 798. Lasko, P., 2000. The Drosophila melanogaster genome: translation factors and RNA binding proteins. J. Cell Biol. 150, 51 – 56. Mackie, K.P., Nairn, A.C., Hampel, G., Lam, G., Jaffe, E.A., 1989. Thrombin and histamine stimulate phosphorylation of elongation factor 2 in human umbilical vein endothelial cells. J. Biol. Chem. 264, 1748 – 1753. Mathews, M.B., Sonenberg, N, Hershey, J.W.B., 2000. Origins and principles of translational control. In: Sonenberg, N., Hershey, J.W.B., Mathews, M.B. (Eds.), Translational Control of Gene Expression. Cold Spring Harbor Laboratory Press, New York, pp. 1 – 32. Moreira, D., LeGuyader, H., Phillippe, H., 2000. The origin of red algae and the evolution of chloroplasts. Nature 405, 69 – 72. Morris, D.K., Lundblad, V., 1997. Programmed translational frameshifting in a gene required for yeast telomere replication. Curr. Biol. 7, 969 – 976. Nairn, A.C., Palfrey, H.C., 1987. Identification of the major Mr 1,000,000 substrate for calmodulin dependent protein kinase III in mammalian cells as elongation factor 2. J. Biol. Chem. 262, 17299 – 17303. Nikiforov, M.A., Gorovsky, M.A., Allis, C.D., 2000. A novel chromodomail protein, Pddp3p, associates with internal eliminated sequences during macronuclear development in Tetrahymena thermophila. Mol. Cell. Biol. 20, 4128 – 4134. Perentesis, J.P., Phan, L.D., Gleason, W.B., LaPorte, D.C., Livingston, D.M., Bodley, J.W., 1992. Saccharomyces cerevisiae elongation factor 2. J. Biol. Chem. 267, 1190 – 1197. Redpath, N.C., Price, N.T., Severinov, K.V., Proud, C.G., 1993. Regulation of elongation factor-2 by multisite phosphorylation. Eur. J. Biochem. 213, 689 – 699. Redpath, N.C., Foulstone, E.J., Proud, C.G., 1996. Regulation of translation elongation factor-2 by insulin via a rapamycin sensitive pathway. EMBO J. 15, 2291 – 2297. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, New York. Sandbaken, M.G., Culbertson, M.R., 1988. Mutations in elongation factor EF-1alpha affect the frequency of frameshifting and amino acid misincorporation in Saccharomyces cerevisiae. Genetics 120, 923 – 934. Wang, X., Campbell, L., Miller, C., Proud, C., 1998. Amino acid availability regulates p70 S6 kinase and multiple translation factors. Biochem. J. 334, 261 – 267.