Negative Control of the Poly(A)-binding Protein mRNA Translation Is Mediated by the Adenine-rich Region of Its 5′-Untranslated Region
Translation of the mRNA for the poly(A)-binding protein (PABP) may be autoregulated by the binding of PABP to the A-rich segment of its 5′-untranslated region (UTR). To test this hypothesis, we examined the effect of different fragments of the 5′-UTR from human PABP cDNA on the translation of the β-...
Сохранить в:
| Главные авторы: | , |
|---|---|
| Формат: | Artigo |
| Язык: | английский |
| Опубликовано: |
1998
|
| Online-ссылка: | https://doi.org/10.1074/jbc.273.51.34535 http://www.jbc.org/article/S0021925819884831/pdf |
| Метки: |
Добавить метку
Нет меток, Требуется 1-ая метка записи!
|
| Итог: | Translation of the mRNA for the poly(A)-binding protein (PABP) may be autoregulated by the binding of PABP to the A-rich segment of its 5′-untranslated region (UTR). To test this hypothesis, we examined the effect of different fragments of the 5′-UTR from human PABP cDNA on the translation of the β-galactosidase (β-Gal) reporter gene. Presence of the A-rich sequence from the 5′-UTR of PABP mRNA inhibited expression of the chimeric β-Gal gene in transfected HeLa cells. The differences in expression of β-Gal polypeptide was due to the translational repression of β-Gal mRNA containing the A-rich 5′-UTR of PABP mRNA. The A-rich region of the 5′-UTR located within nucleotides 58–146 of PABP mRNA was sufficient to mediate translational control of this mRNA expression. We also examined the effect of overexpression of PABP mRNA in HeLa cells. The ectopic PABP mRNA without the A-rich 5′-UTR region was translated efficiently, whereas the translation of the endogenous PABP mRNA was substantially inhibited in the transfected cells. In contrast, the ectopic PABP mRNA containing the A-rich 5′-UTR region did not show similar effect on the translation of the endogenous PABP mRNA in these cells. These results suggest that feedback control of mRNA translation is involved in regulating PABP expression in HeLa cells. Translation of the mRNA for the poly(A)-binding protein (PABP) may be autoregulated by the binding of PABP to the A-rich segment of its 5′-untranslated region (UTR). To test this hypothesis, we examined the effect of different fragments of the 5′-UTR from human PABP cDNA on the translation of the β-galactosidase (β-Gal) reporter gene. Presence of the A-rich sequence from the 5′-UTR of PABP mRNA inhibited expression of the chimeric β-Gal gene in transfected HeLa cells. The differences in expression of β-Gal polypeptide was due to the translational repression of β-Gal mRNA containing the A-rich 5′-UTR of PABP mRNA. The A-rich region of the 5′-UTR located within nucleotides 58–146 of PABP mRNA was sufficient to mediate translational control of this mRNA expression. We also examined the effect of overexpression of PABP mRNA in HeLa cells. The ectopic PABP mRNA without the A-rich 5′-UTR region was translated efficiently, whereas the translation of the endogenous PABP mRNA was substantially inhibited in the transfected cells. In contrast, the ectopic PABP mRNA containing the A-rich 5′-UTR region did not show similar effect on the translation of the endogenous PABP mRNA in these cells. These results suggest that feedback control of mRNA translation is involved in regulating PABP expression in HeLa cells. poly(A)-binding protein eukaryotic initiation factor untranslated region glyceraldehyde-3-phosphate dehydrogenase β-galactosidase reverse transcription-polymerase chain reaction reverse transcription auto-regulatory sequence 4-morpholinepropanesulfonic acid. Cellular mRNAs are often complexed with a group of RNA-binding proteins. One of the best studied RNA-binding proteins is the 72-kDa poly(A)-binding protein (PABP),1 that shows specific interaction with the 3′ poly(A) region of all eukaryotic mRNAs (1Sachs A.B. Bond M.W. Kornberg R.D. Cell. 1986; 45: 827-835Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 2Grange T. Martins de Sa C. Oddos J. Pictet R. Nucleic Acids Res. 1987; 15: 4771-4787Crossref PubMed Scopus (142) Google Scholar, 3Zelus B.D. Giebelhaus D.H. Eib D.W. Kenner K.A. Moon R.T. Mol. Cell. Biol. 1989; 9: 2756-2760Crossref PubMed Scopus (80) Google Scholar). PABP is ubiquitous and highly conserved in eukaryotic cells. The poly(A)-binding region contains 90 amino acid domains (1Sachs A.B. Bond M.W. Kornberg R.D. Cell. 1986; 45: 827-835Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 2Grange T. Martins de Sa C. Oddos J. Pictet R. Nucleic Acids Res. 1987; 15: 4771-4787Crossref PubMed Scopus (142) Google Scholar, 3Zelus B.D. Giebelhaus D.H. Eib D.W. Kenner K.A. Moon R.T. Mol. Cell. Biol. 1989; 9: 2756-2760Crossref PubMed Scopus (80) Google Scholar). Plants and mammals both contain several related polypeptides with specific affinity toward the poly(A) region of the mRNA. These PABP-related polypeptides show tissue-specific expression and may be regulated by growth conditions (4Belostotsky D.A. Meagher R.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6686-6690Crossref PubMed Scopus (72) Google Scholar, 5Yang H. Collins S. Duckett H.J. Lindsten T. Mol. Cell. Biol. 1995; 15: 6770-6776Crossref PubMed Scopus (67) Google Scholar). PABP is essential for viability of eukaryotic cells and is involved in regulating mRNA translation and stability. Results of both in vitro (6Bernstein P. Peltz S.W. Ross J. Mol. Cell. Biol. 1989; 9: 659-670Crossref PubMed Scopus (291) Google Scholar) and in vivo (7Sachs A.B. Davis R.W. Cell. 1989; 58: 857-867Abstract Full Text PDF PubMed Scopus (388) Google Scholar) studies suggest that the complex of PABP and poly(A) tail plays an important role in mRNA translation. It is believed that the poly(A)-PABP complex is involved in joining the 60 S subunit to the 48 S preinitiation complex (7Sachs A.B. Davis R.W. Cell. 1989; 58: 857-867Abstract Full Text PDF PubMed Scopus (388) Google Scholar). PABP also stimulates the binding of 40 S subunits to the mRNA (8Tarun Jr., S.Z. Sachs A.B. Genes Dev. 1995; 9: 2997-3007Crossref PubMed Scopus (328) Google Scholar). The ability of PABP to interact with the eIF4G and eIF4B is important for its role in the initiation of mRNA translation (9Tarun Jr., S.Z. Sachs A.B. EMBO J. 1996; 15: 7168-7177Crossref PubMed Scopus (574) Google Scholar, 10Tarun Jr., S.Z. Wells S.E. Deardoroff J.A. Sachs A.B. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9046-9051Crossref PubMed Scopus (262) Google Scholar, 11Le H. Tanguay R.L. Balasta M.L. Wei C.-C. Browning K.S. Metz A.M. Goss D.J. Gallie D.R. J. Biol. Chem. 1997; 272: 16247-16255Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar, 12Craig A.W.B. Haghighat A. Yu A.T.K. Sonenberg N. Nature. 1998; 392: 520-523Crossref PubMed Scopus (324) Google Scholar, 13Boeck R. Tarun Jr., S.Z. Rieger M. Deardorff J. Muller-Auer S. Sachs A.B. J. Biol. Chem. 1996; 271: 432-438Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar). The interactions between PABP and these initiation factors also take place in absence of poly(A) (11Le H. Tanguay R.L. Balasta M.L. Wei C.-C. Browning K.S. Metz A.M. Goss D.J. Gallie D.R. J. Biol. Chem. 1997; 272: 16247-16255Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar). Therefore, it is not clear whether the free PABP or PABP-poly(A) complex is involved in mRNA translationin vivo. In addition, PABP is also involved in poly(A) tail shortening probably through its interaction with a poly(A)-specific 3′-exonuclease (12Craig A.W.B. Haghighat A. Yu A.T.K. Sonenberg N. Nature. 1998; 392: 520-523Crossref PubMed Scopus (324) Google Scholar, 14Ko C.G. Wahle E. J. Biol. Chem. 1997; 272: 10448-10456Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar).Several studies indicated that PABP expression is regulated at the post-transcriptional level during the cell cycle (15Maundrell K. Imaizumi-Scherrer M.T. Maxwell E.S. Civeli O. Scherrer K. J. Biol. Chem. 1983; 258: 1387-1390Abstract Full Text PDF PubMed Google Scholar, 16Ullrich S.J. Appela E. Mercer W.E. Exp. Cell. Res. 1988; 178: 273-280Crossref PubMed Scopus (8) Google Scholar, 17Mercer W.E. Jaskulski D. Shields M.T. Exp. Cell. Res. 1989; 181: 531-541Crossref PubMed Scopus (9) Google Scholar, 18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar, 19Thomas G. Thomas G. J. Cell Biol. 1986; 103: 2137-2144Crossref PubMed Scopus (89) Google Scholar, 20Berger L.C. Bag J. Sells B.H. Biol. Cell Biol. 1992; 70: 770-778Google Scholar). In resting cells, the majority of cellular PABP mRNA was found in the non-translated state. Following stimulation of growth by serum, a rapid transition of the mRNA to the translated state occurred in mouse fibroblasts (19Thomas G. Thomas G. J. Cell Biol. 1986; 103: 2137-2144Crossref PubMed Scopus (89) Google Scholar, 20Berger L.C. Bag J. Sells B.H. Biol. Cell Biol. 1992; 70: 770-778Google Scholar). In contrast, during differentiation of murine myoblasts growth arrest resulted in the migration of repressed PABP mRNA to the translated state. This process was also accompanied by reduced stability of the PABP mRNA in differentiated myotubes resulting in little change in PABP synthesis (18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar).Translation of a specific mRNA is often regulated by the interaction of regulatory proteins with defined sequences of the mRNA. For instance, feedback control of ribosomal protein L10 synthesis by the level of free L10 protein is mediated by its binding to the 5′-UTR of the L10 mRNA (21Dean D. Yates J.L. Nomura M. Nature. 1981; 289: 89-91Crossref PubMed Scopus (51) Google Scholar, 22Baughman G. Nomura M. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 5389-5393Crossref PubMed Scopus (34) Google Scholar). Similarly, synthesis of ferritin is regulated in vertebrates by the iron level (23Eisenstein R.S. Munro H.N. Enzyme (Basel). 1991; 44: 42-58Crossref Scopus (37) Google Scholar). The inhibition of ferritin mRNA translation requires the binding of a 98-kDa translation repressor polypeptide to the conserved 25-nucleotide-long sequence of its 5′-UTR (24Neupert B. Thompson N.A. Meyer C. Kuhn L. Nucleic Acids Res. 1990; 18: 51-55Crossref PubMed Scopus (60) Google Scholar, 25Dandekar T. EMBO J. 1991; 10: 1903-1915Crossref PubMed Scopus (275) Google Scholar). In addition the 3′-UTR of some mRNAs is involved in regulating translation. The translations of protamine (26Braun R.E. Peschon J.J. Behringer R.R. Brinster R.L. Palmiter R.D. Genes Dev. 1989; 3: 793-802Crossref PubMed Scopus (184) Google Scholar) and myocyte enhancer factor-2a mRNAs are controlled by RNA-binding proteins that interact with sequences located within the 3′-UTR of these mRNA (27Black B.L. Lu J. Olson E.N. Mol. Cell. Biol. 1997; 17: 2756-2763Crossref PubMed Scopus (46) Google Scholar). How the 3′-end of these mRNAs influence ribosome binding at the 5′-end is unknown. It is possible that 3′-UTR binding protein(s) may interfere with the function of poly(A)-PABP complex.In contrast to the examples cited above, regulation of PABP mRNA translation appears to be more subtle. Instead of completely blocking PABP mRNA translation, the distribution of this mRNA between translationally active and inactive populations is modulated. Interestingly, analysis of PABP mRNA sequences in various species showed the presence of conserved highly A-rich sequence in its 5′-UTR. This raises the possibility of negative feedback control of PABP mRNA translation. Previous studies from this and other laboratories have shown that PABP is able to bind to the 5′-UTR A-rich regions and inhibit PABP mRNA translation in a rabbit reticulocyte lysate cell-free system (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar, 29de Melo Neto O.P. Standart N. de Sa C.M. Nucleic Acids Res. 1995; 34: 2198-2205Crossref Scopus (95) Google Scholar), and removal of this sequence from the PABP mRNA enhances its translation in vitro. We have also shown that the translation of full-length PABP mRNA, but not the truncated mRNA from which the A-rich region of the 5′-UTR was removed, can be inhibited in a cell-free system by adding purified human PABP. Furthermore, by using UV-mediated cross-linking of RNA and proteins, we have shown that PABP binds to the first 224 nucleotides of the 5′-UTR of PABP mRNA, and this may repress its own expression (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar). Therefore in the present study, we extended these investigations to determine whether similar inhibition of mRNA translation occursin vivo by the A-rich 5′-UTR of PABP mRNA. We have also examined if PABP synthesis was regulated in HeLa cells when the attempt was made to increase PABP synthesis by ectopic expression of this mRNA.DISCUSSIONThe presence of approximately 60% adenine base between nucleotides 71 and 131 of the PABP mRNA 5′-UTR offers potential PABP-binding site(s) for autoregulation of its translation. In support of this autoregulation model it was shown that removal of the first 223 nucleotides of its 5′-UTR improves PABP synthesis in vitro(28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar, 29de Melo Neto O.P. Standart N. de Sa C.M. Nucleic Acids Res. 1995; 34: 2198-2205Crossref Scopus (95) Google Scholar). In the studies reported here further examination of the influence of this oligo(A)-rich 5′-UTR sequence in cultured HeLa cells also demonstrated the inhibitory effect of PABP on mRNA translation. The presence of the 58–146 nucleotides from PABP mRNA 5′-UTR resulted in inhibition of translation of the reporter β-Gal mRNA in HeLa cells. Although the adenine-rich region was located between nucleotides 71 and 131, we used a slightly larger region (nucleotides 58–146) of the 5′-UTR in our analyses. Due to the predominance of a single base the A-rich region was not clonable without some flanking sequences. The results of our studies did not rule out the possibility of other regulatory sequences existing within the 5′-UTR of PABP mRNA. PABP mRNA translation appeared to be differently regulated in different cells (5Yang H. Collins S. Duckett H.J. Lindsten T. Mol. Cell. Biol. 1995; 15: 6770-6776Crossref PubMed Scopus (67) Google Scholar, 16Ullrich S.J. Appela E. Mercer W.E. Exp. Cell. Res. 1988; 178: 273-280Crossref PubMed Scopus (8) Google Scholar, 17Mercer W.E. Jaskulski D. Shields M.T. Exp. Cell. Res. 1989; 181: 531-541Crossref PubMed Scopus (9) Google Scholar, 18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar). It is therefore possible that its translation can be regulated by sequence(s) other than the putative PABP-binding sequence in other cell lines or in a tissue-specific manner. It is conceivable that the presence of cell-specific RNA-binding proteins may be involved in differential regulation of translation of PABP mRNA. However, our major concern was to test the possibility of autoregulation and the potential role of the oligo(A)-rich sequence in mRNA translation in vivo. To test further this model, we examined the effect of ectopic expression of this mRNA on the translation of endogenous mRNA. Our results showed that translation of the endogenous PABP mRNA was repressed when attempts were made to overexpress PABP. These results suggest that the level of PABP and its mRNA may determine how much of the cytoplasmic pool of the PABP mRNA is translated. This checkpoint in PABP expression depends on the presence of the A-rich region of its 5′-UTR.In cells transfected with the PABP expression vectors, the mRNA levels were increased by more than 2-fold. But only a modest increase (25%) in the PABP level was found (Fig. 7). In the cells expressing the unregulated ectopic mRNA, this was achieved by almost completely blocking the translation of endogenous mRNA and translating the ectopic mRNA. On the other hand when the ectopic mRNA had the auto-regulatory sequence (ARS) both endogenous and ectopic mRNAs translation were controlled. However, the translation of the ectopic PABP mRNA with a shorter 5′-UTR and the ARS positioned closer to the initiation codon than in the endogenous mRNA was repressed more than that of the endogenous mRNA. The reason for this difference is not clear but may reflect the influence of other regulatory sequences. As a result of nearly complete inhibition of the ectopic mRNA translation and a small shift of the endogenous PABP mRNA to the non-translated state, the levels of PABP in these cells were reduced by 50%. Thus, there was over-compensation of PABP expression.The precise mechanism of PABP-mediated feedback control of translation is not known. However, the results suggest that 48 S preinitiation complexes were formed but were not efficiently converted to the initiation complex. It is therefore possible that binding of PABP at the ARS may decrease the rate of the 40 S ribosomal subunit movement along the mRNA. According to this model the distance between the initiation codon and the ARS is unlikely to have a major influence on mRNA translation. The distance of the ARS from the 5′-cap region, however, might have stronger influence than that of the ARS and the AUG codon. This is conceivable since the 3′ poly(A)-PABP complex may be involved in the translation initiation through its interaction with the eIF4G. Therefore, formation of a similar complex close to the 5′-capped end may mute their functional interaction. Further studies are required to understand how ARS-PABP complex works.Feedback control of PABP mRNA translation may serve an important purpose in regulating the cellular level of free PABP. Under normal circumstances, cytoplasmic PABP is present in excess over the availability of the poly(A)-binding sites (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar). The reason for the presence of free PABP is not clear. However, some of this may result from degradation or shortening of the poly(A) tail. Presence of multiple poly(A)-binding domains in PABP (38Sachs A.B. Davis R.W. Kornberg R.D. Mol. Cell. Biol. 1987; 7: 3268-3276Crossref PubMed Scopus (319) Google Scholar) potentially contributes to the dynamic nature of its interaction with the 3′ poly(A) tail (39Greenberg J.R. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 2923-2926Crossref PubMed Scopus (32) Google Scholar). This could allow movement of PABP between the 3′ poly(A) tail and other lower affinity PABP binding sequences (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar) including the ARS at the 5′ UTR of PABP mRNA. How efficiently PABP would interact with the lower affinity binding sequences may depend on its cellular level.Presumably an optimal level of PABP is required for maintaining an equilibrium between its various functions. Our studies have shown that the presence of the putative PABP-binding site at the 5′-UTR of PABP mRNA reduces by approximately 3-fold the translation of this mRNA both in vitro (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar) and in vivo (Fig. 3). Although the presence of a certain excess amount of PABP is tolerated and may even be required for normal cellular function, its expression beyond a threshold could be controlled at the mRNA translation level. Excess PABP beyond the threshold level may detrimentally alter the pattern of protein synthesis in the cell, for example by increasing translation of inherently inefficient mRNAs. In addition stability of some unstable mRNAs may also increase if excess PABP is made. Together, these events could have a dramatic effect on the growth regulation and cellular differentiation. This possibility is supported by the previous observation that overexpression of eIF4E results in cellular transformation (41Lazaris-Karatzas A. Montine K.S. Sonenberg N. Nature. 1990; 345: 544-547Crossref PubMed Scopus (799) Google Scholar). Since the role of PABP in translation is believed to be mediated through its interaction with the cap structure and eIF4G (of which eIF4E is a component), it will be important to control PABP expression through feedback mechanisms. This is further supported by the observation that overexpression of PABP in Xenopus oocytes influenced the deadenylation and translation of maternal mRNAs (42Wormington M. Searfoss A.M. Hurney C.A. EMBO J. 1996; 15: 900-909Crossref PubMed Scopus (91) Google Scholar). Cellular mRNAs are often complexed with a group of RNA-binding proteins. One of the best studied RNA-binding proteins is the 72-kDa poly(A)-binding protein (PABP),1 that shows specific interaction with the 3′ poly(A) region of all eukaryotic mRNAs (1Sachs A.B. Bond M.W. Kornberg R.D. Cell. 1986; 45: 827-835Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 2Grange T. Martins de Sa C. Oddos J. Pictet R. Nucleic Acids Res. 1987; 15: 4771-4787Crossref PubMed Scopus (142) Google Scholar, 3Zelus B.D. Giebelhaus D.H. Eib D.W. Kenner K.A. Moon R.T. Mol. Cell. Biol. 1989; 9: 2756-2760Crossref PubMed Scopus (80) Google Scholar). PABP is ubiquitous and highly conserved in eukaryotic cells. The poly(A)-binding region contains 90 amino acid domains (1Sachs A.B. Bond M.W. Kornberg R.D. Cell. 1986; 45: 827-835Abstract Full Text PDF PubMed Scopus (270) Google Scholar, 2Grange T. Martins de Sa C. Oddos J. Pictet R. Nucleic Acids Res. 1987; 15: 4771-4787Crossref PubMed Scopus (142) Google Scholar, 3Zelus B.D. Giebelhaus D.H. Eib D.W. Kenner K.A. Moon R.T. Mol. Cell. Biol. 1989; 9: 2756-2760Crossref PubMed Scopus (80) Google Scholar). Plants and mammals both contain several related polypeptides with specific affinity toward the poly(A) region of the mRNA. These PABP-related polypeptides show tissue-specific expression and may be regulated by growth conditions (4Belostotsky D.A. Meagher R.B. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6686-6690Crossref PubMed Scopus (72) Google Scholar, 5Yang H. Collins S. Duckett H.J. Lindsten T. Mol. Cell. Biol. 1995; 15: 6770-6776Crossref PubMed Scopus (67) Google Scholar). PABP is essential for viability of eukaryotic cells and is involved in regulating mRNA translation and stability. Results of both in vitro (6Bernstein P. Peltz S.W. Ross J. Mol. Cell. Biol. 1989; 9: 659-670Crossref PubMed Scopus (291) Google Scholar) and in vivo (7Sachs A.B. Davis R.W. Cell. 1989; 58: 857-867Abstract Full Text PDF PubMed Scopus (388) Google Scholar) studies suggest that the complex of PABP and poly(A) tail plays an important role in mRNA translation. It is believed that the poly(A)-PABP complex is involved in joining the 60 S subunit to the 48 S preinitiation complex (7Sachs A.B. Davis R.W. Cell. 1989; 58: 857-867Abstract Full Text PDF PubMed Scopus (388) Google Scholar). PABP also stimulates the binding of 40 S subunits to the mRNA (8Tarun Jr., S.Z. Sachs A.B. Genes Dev. 1995; 9: 2997-3007Crossref PubMed Scopus (328) Google Scholar). The ability of PABP to interact with the eIF4G and eIF4B is important for its role in the initiation of mRNA translation (9Tarun Jr., S.Z. Sachs A.B. EMBO J. 1996; 15: 7168-7177Crossref PubMed Scopus (574) Google Scholar, 10Tarun Jr., S.Z. Wells S.E. Deardoroff J.A. Sachs A.B. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9046-9051Crossref PubMed Scopus (262) Google Scholar, 11Le H. Tanguay R.L. Balasta M.L. Wei C.-C. Browning K.S. Metz A.M. Goss D.J. Gallie D.R. J. Biol. Chem. 1997; 272: 16247-16255Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar, 12Craig A.W.B. Haghighat A. Yu A.T.K. Sonenberg N. Nature. 1998; 392: 520-523Crossref PubMed Scopus (324) Google Scholar, 13Boeck R. Tarun Jr., S.Z. Rieger M. Deardorff J. Muller-Auer S. Sachs A.B. J. Biol. Chem. 1996; 271: 432-438Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar). The interactions between PABP and these initiation factors also take place in absence of poly(A) (11Le H. Tanguay R.L. Balasta M.L. Wei C.-C. Browning K.S. Metz A.M. Goss D.J. Gallie D.R. J. Biol. Chem. 1997; 272: 16247-16255Abstract Full Text Full Text PDF PubMed Scopus (227) Google Scholar). Therefore, it is not clear whether the free PABP or PABP-poly(A) complex is involved in mRNA translationin vivo. In addition, PABP is also involved in poly(A) tail shortening probably through its interaction with a poly(A)-specific 3′-exonuclease (12Craig A.W.B. Haghighat A. Yu A.T.K. Sonenberg N. Nature. 1998; 392: 520-523Crossref PubMed Scopus (324) Google Scholar, 14Ko C.G. Wahle E. J. Biol. Chem. 1997; 272: 10448-10456Abstract Full Text Full Text PDF PubMed Scopus (198) Google Scholar). Several studies indicated that PABP expression is regulated at the post-transcriptional level during the cell cycle (15Maundrell K. Imaizumi-Scherrer M.T. Maxwell E.S. Civeli O. Scherrer K. J. Biol. Chem. 1983; 258: 1387-1390Abstract Full Text PDF PubMed Google Scholar, 16Ullrich S.J. Appela E. Mercer W.E. Exp. Cell. Res. 1988; 178: 273-280Crossref PubMed Scopus (8) Google Scholar, 17Mercer W.E. Jaskulski D. Shields M.T. Exp. Cell. Res. 1989; 181: 531-541Crossref PubMed Scopus (9) Google Scholar, 18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar, 19Thomas G. Thomas G. J. Cell Biol. 1986; 103: 2137-2144Crossref PubMed Scopus (89) Google Scholar, 20Berger L.C. Bag J. Sells B.H. Biol. Cell Biol. 1992; 70: 770-778Google Scholar). In resting cells, the majority of cellular PABP mRNA was found in the non-translated state. Following stimulation of growth by serum, a rapid transition of the mRNA to the translated state occurred in mouse fibroblasts (19Thomas G. Thomas G. J. Cell Biol. 1986; 103: 2137-2144Crossref PubMed Scopus (89) Google Scholar, 20Berger L.C. Bag J. Sells B.H. Biol. Cell Biol. 1992; 70: 770-778Google Scholar). In contrast, during differentiation of murine myoblasts growth arrest resulted in the migration of repressed PABP mRNA to the translated state. This process was also accompanied by reduced stability of the PABP mRNA in differentiated myotubes resulting in little change in PABP synthesis (18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar). Translation of a specific mRNA is often regulated by the interaction of regulatory proteins with defined sequences of the mRNA. For instance, feedback control of ribosomal protein L10 synthesis by the level of free L10 protein is mediated by its binding to the 5′-UTR of the L10 mRNA (21Dean D. Yates J.L. Nomura M. Nature. 1981; 289: 89-91Crossref PubMed Scopus (51) Google Scholar, 22Baughman G. Nomura M. Proc. Natl. Acad. Sci. U. S. A. 1984; 81: 5389-5393Crossref PubMed Scopus (34) Google Scholar). Similarly, synthesis of ferritin is regulated in vertebrates by the iron level (23Eisenstein R.S. Munro H.N. Enzyme (Basel). 1991; 44: 42-58Crossref Scopus (37) Google Scholar). The inhibition of ferritin mRNA translation requires the binding of a 98-kDa translation repressor polypeptide to the conserved 25-nucleotide-long sequence of its 5′-UTR (24Neupert B. Thompson N.A. Meyer C. Kuhn L. Nucleic Acids Res. 1990; 18: 51-55Crossref PubMed Scopus (60) Google Scholar, 25Dandekar T. EMBO J. 1991; 10: 1903-1915Crossref PubMed Scopus (275) Google Scholar). In addition the 3′-UTR of some mRNAs is involved in regulating translation. The translations of protamine (26Braun R.E. Peschon J.J. Behringer R.R. Brinster R.L. Palmiter R.D. Genes Dev. 1989; 3: 793-802Crossref PubMed Scopus (184) Google Scholar) and myocyte enhancer factor-2a mRNAs are controlled by RNA-binding proteins that interact with sequences located within the 3′-UTR of these mRNA (27Black B.L. Lu J. Olson E.N. Mol. Cell. Biol. 1997; 17: 2756-2763Crossref PubMed Scopus (46) Google Scholar). How the 3′-end of these mRNAs influence ribosome binding at the 5′-end is unknown. It is possible that 3′-UTR binding protein(s) may interfere with the function of poly(A)-PABP complex. In contrast to the examples cited above, regulation of PABP mRNA translation appears to be more subtle. Instead of completely blocking PABP mRNA translation, the distribution of this mRNA between translationally active and inactive populations is modulated. Interestingly, analysis of PABP mRNA sequences in various species showed the presence of conserved highly A-rich sequence in its 5′-UTR. This raises the possibility of negative feedback control of PABP mRNA translation. Previous studies from this and other laboratories have shown that PABP is able to bind to the 5′-UTR A-rich regions and inhibit PABP mRNA translation in a rabbit reticulocyte lysate cell-free system (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar, 29de Melo Neto O.P. Standart N. de Sa C.M. Nucleic Acids Res. 1995; 34: 2198-2205Crossref Scopus (95) Google Scholar), and removal of this sequence from the PABP mRNA enhances its translation in vitro. We have also shown that the translation of full-length PABP mRNA, but not the truncated mRNA from which the A-rich region of the 5′-UTR was removed, can be inhibited in a cell-free system by adding purified human PABP. Furthermore, by using UV-mediated cross-linking of RNA and proteins, we have shown that PABP binds to the first 224 nucleotides of the 5′-UTR of PABP mRNA, and this may repress its own expression (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar). Therefore in the present study, we extended these investigations to determine whether similar inhibition of mRNA translation occursin vivo by the A-rich 5′-UTR of PABP mRNA. We have also examined if PABP synthesis was regulated in HeLa cells when the attempt was made to increase PABP synthesis by ectopic expression of this mRNA. DISCUSSIONThe presence of approximately 60% adenine base between nucleotides 71 and 131 of the PABP mRNA 5′-UTR offers potential PABP-binding site(s) for autoregulation of its translation. In support of this autoregulation model it was shown that removal of the first 223 nucleotides of its 5′-UTR improves PABP synthesis in vitro(28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar, 29de Melo Neto O.P. Standart N. de Sa C.M. Nucleic Acids Res. 1995; 34: 2198-2205Crossref Scopus (95) Google Scholar). In the studies reported here further examination of the influence of this oligo(A)-rich 5′-UTR sequence in cultured HeLa cells also demonstrated the inhibitory effect of PABP on mRNA translation. The presence of the 58–146 nucleotides from PABP mRNA 5′-UTR resulted in inhibition of translation of the reporter β-Gal mRNA in HeLa cells. Although the adenine-rich region was located between nucleotides 71 and 131, we used a slightly larger region (nucleotides 58–146) of the 5′-UTR in our analyses. Due to the predominance of a single base the A-rich region was not clonable without some flanking sequences. The results of our studies did not rule out the possibility of other regulatory sequences existing within the 5′-UTR of PABP mRNA. PABP mRNA translation appeared to be differently regulated in different cells (5Yang H. Collins S. Duckett H.J. Lindsten T. Mol. Cell. Biol. 1995; 15: 6770-6776Crossref PubMed Scopus (67) Google Scholar, 16Ullrich S.J. Appela E. Mercer W.E. Exp. Cell. Res. 1988; 178: 273-280Crossref PubMed Scopus (8) Google Scholar, 17Mercer W.E. Jaskulski D. Shields M.T. Exp. Cell. Res. 1989; 181: 531-541Crossref PubMed Scopus (9) Google Scholar, 18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar). It is therefore possible that its translation can be regulated by sequence(s) other than the putative PABP-binding sequence in other cell lines or in a tissue-specific manner. It is conceivable that the presence of cell-specific RNA-binding proteins may be involved in differential regulation of translation of PABP mRNA. However, our major concern was to test the possibility of autoregulation and the potential role of the oligo(A)-rich sequence in mRNA translation in vivo. To test further this model, we examined the effect of ectopic expression of this mRNA on the translation of endogenous mRNA. Our results showed that translation of the endogenous PABP mRNA was repressed when attempts were made to overexpress PABP. These results suggest that the level of PABP and its mRNA may determine how much of the cytoplasmic pool of the PABP mRNA is translated. This checkpoint in PABP expression depends on the presence of the A-rich region of its 5′-UTR.In cells transfected with the PABP expression vectors, the mRNA levels were increased by more than 2-fold. But only a modest increase (25%) in the PABP level was found (Fig. 7). In the cells expressing the unregulated ectopic mRNA, this was achieved by almost completely blocking the translation of endogenous mRNA and translating the ectopic mRNA. On the other hand when the ectopic mRNA had the auto-regulatory sequence (ARS) both endogenous and ectopic mRNAs translation were controlled. However, the translation of the ectopic PABP mRNA with a shorter 5′-UTR and the ARS positioned closer to the initiation codon than in the endogenous mRNA was repressed more than that of the endogenous mRNA. The reason for this difference is not clear but may reflect the influence of other regulatory sequences. As a result of nearly complete inhibition of the ectopic mRNA translation and a small shift of the endogenous PABP mRNA to the non-translated state, the levels of PABP in these cells were reduced by 50%. Thus, there was over-compensation of PABP expression.The precise mechanism of PABP-mediated feedback control of translation is not known. However, the results suggest that 48 S preinitiation complexes were formed but were not efficiently converted to the initiation complex. It is therefore possible that binding of PABP at the ARS may decrease the rate of the 40 S ribosomal subunit movement along the mRNA. According to this model the distance between the initiation codon and the ARS is unlikely to have a major influence on mRNA translation. The distance of the ARS from the 5′-cap region, however, might have stronger influence than that of the ARS and the AUG codon. This is conceivable since the 3′ poly(A)-PABP complex may be involved in the translation initiation through its interaction with the eIF4G. Therefore, formation of a similar complex close to the 5′-capped end may mute their functional interaction. Further studies are required to understand how ARS-PABP complex works.Feedback control of PABP mRNA translation may serve an important purpose in regulating the cellular level of free PABP. Under normal circumstances, cytoplasmic PABP is present in excess over the availability of the poly(A)-binding sites (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar). The reason for the presence of free PABP is not clear. However, some of this may result from degradation or shortening of the poly(A) tail. Presence of multiple poly(A)-binding domains in PABP (38Sachs A.B. Davis R.W. Kornberg R.D. Mol. Cell. Biol. 1987; 7: 3268-3276Crossref PubMed Scopus (319) Google Scholar) potentially contributes to the dynamic nature of its interaction with the 3′ poly(A) tail (39Greenberg J.R. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 2923-2926Crossref PubMed Scopus (32) Google Scholar). This could allow movement of PABP between the 3′ poly(A) tail and other lower affinity PABP binding sequences (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar) including the ARS at the 5′ UTR of PABP mRNA. How efficiently PABP would interact with the lower affinity binding sequences may depend on its cellular level.Presumably an optimal level of PABP is required for maintaining an equilibrium between its various functions. Our studies have shown that the presence of the putative PABP-binding site at the 5′-UTR of PABP mRNA reduces by approximately 3-fold the translation of this mRNA both in vitro (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar) and in vivo (Fig. 3). Although the presence of a certain excess amount of PABP is tolerated and may even be required for normal cellular function, its expression beyond a threshold could be controlled at the mRNA translation level. Excess PABP beyond the threshold level may detrimentally alter the pattern of protein synthesis in the cell, for example by increasing translation of inherently inefficient mRNAs. In addition stability of some unstable mRNAs may also increase if excess PABP is made. Together, these events could have a dramatic effect on the growth regulation and cellular differentiation. This possibility is supported by the previous observation that overexpression of eIF4E results in cellular transformation (41Lazaris-Karatzas A. Montine K.S. Sonenberg N. Nature. 1990; 345: 544-547Crossref PubMed Scopus (799) Google Scholar). Since the role of PABP in translation is believed to be mediated through its interaction with the cap structure and eIF4G (of which eIF4E is a component), it will be important to control PABP expression through feedback mechanisms. This is further supported by the observation that overexpression of PABP in Xenopus oocytes influenced the deadenylation and translation of maternal mRNAs (42Wormington M. Searfoss A.M. Hurney C.A. EMBO J. 1996; 15: 900-909Crossref PubMed Scopus (91) Google Scholar). The presence of approximately 60% adenine base between nucleotides 71 and 131 of the PABP mRNA 5′-UTR offers potential PABP-binding site(s) for autoregulation of its translation. In support of this autoregulation model it was shown that removal of the first 223 nucleotides of its 5′-UTR improves PABP synthesis in vitro(28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar, 29de Melo Neto O.P. Standart N. de Sa C.M. Nucleic Acids Res. 1995; 34: 2198-2205Crossref Scopus (95) Google Scholar). In the studies reported here further examination of the influence of this oligo(A)-rich 5′-UTR sequence in cultured HeLa cells also demonstrated the inhibitory effect of PABP on mRNA translation. The presence of the 58–146 nucleotides from PABP mRNA 5′-UTR resulted in inhibition of translation of the reporter β-Gal mRNA in HeLa cells. Although the adenine-rich region was located between nucleotides 71 and 131, we used a slightly larger region (nucleotides 58–146) of the 5′-UTR in our analyses. Due to the predominance of a single base the A-rich region was not clonable without some flanking sequences. The results of our studies did not rule out the possibility of other regulatory sequences existing within the 5′-UTR of PABP mRNA. PABP mRNA translation appeared to be differently regulated in different cells (5Yang H. Collins S. Duckett H.J. Lindsten T. Mol. Cell. Biol. 1995; 15: 6770-6776Crossref PubMed Scopus (67) Google Scholar, 16Ullrich S.J. Appela E. Mercer W.E. Exp. Cell. Res. 1988; 178: 273-280Crossref PubMed Scopus (8) Google Scholar, 17Mercer W.E. Jaskulski D. Shields M.T. Exp. Cell. Res. 1989; 181: 531-541Crossref PubMed Scopus (9) Google Scholar, 18Adamou J. Bag J. Eur. J. Biochem. 1992; 209: 803-812Crossref PubMed Scopus (15) Google Scholar). It is therefore possible that its translation can be regulated by sequence(s) other than the putative PABP-binding sequence in other cell lines or in a tissue-specific manner. It is conceivable that the presence of cell-specific RNA-binding proteins may be involved in differential regulation of translation of PABP mRNA. However, our major concern was to test the possibility of autoregulation and the potential role of the oligo(A)-rich sequence in mRNA translation in vivo. To test further this model, we examined the effect of ectopic expression of this mRNA on the translation of endogenous mRNA. Our results showed that translation of the endogenous PABP mRNA was repressed when attempts were made to overexpress PABP. These results suggest that the level of PABP and its mRNA may determine how much of the cytoplasmic pool of the PABP mRNA is translated. This checkpoint in PABP expression depends on the presence of the A-rich region of its 5′-UTR. In cells transfected with the PABP expression vectors, the mRNA levels were increased by more than 2-fold. But only a modest increase (25%) in the PABP level was found (Fig. 7). In the cells expressing the unregulated ectopic mRNA, this was achieved by almost completely blocking the translation of endogenous mRNA and translating the ectopic mRNA. On the other hand when the ectopic mRNA had the auto-regulatory sequence (ARS) both endogenous and ectopic mRNAs translation were controlled. However, the translation of the ectopic PABP mRNA with a shorter 5′-UTR and the ARS positioned closer to the initiation codon than in the endogenous mRNA was repressed more than that of the endogenous mRNA. The reason for this difference is not clear but may reflect the influence of other regulatory sequences. As a result of nearly complete inhibition of the ectopic mRNA translation and a small shift of the endogenous PABP mRNA to the non-translated state, the levels of PABP in these cells were reduced by 50%. Thus, there was over-compensation of PABP expression. The precise mechanism of PABP-mediated feedback control of translation is not known. However, the results suggest that 48 S preinitiation complexes were formed but were not efficiently converted to the initiation complex. It is therefore possible that binding of PABP at the ARS may decrease the rate of the 40 S ribosomal subunit movement along the mRNA. According to this model the distance between the initiation codon and the ARS is unlikely to have a major influence on mRNA translation. The distance of the ARS from the 5′-cap region, however, might have stronger influence than that of the ARS and the AUG codon. This is conceivable since the 3′ poly(A)-PABP complex may be involved in the translation initiation through its interaction with the eIF4G. Therefore, formation of a similar complex close to the 5′-capped end may mute their functional interaction. Further studies are required to understand how ARS-PABP complex works. Feedback control of PABP mRNA translation may serve an important purpose in regulating the cellular level of free PABP. Under normal circumstances, cytoplasmic PABP is present in excess over the availability of the poly(A)-binding sites (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar). The reason for the presence of free PABP is not clear. However, some of this may result from degradation or shortening of the poly(A) tail. Presence of multiple poly(A)-binding domains in PABP (38Sachs A.B. Davis R.W. Kornberg R.D. Mol. Cell. Biol. 1987; 7: 3268-3276Crossref PubMed Scopus (319) Google Scholar) potentially contributes to the dynamic nature of its interaction with the 3′ poly(A) tail (39Greenberg J.R. Proc. Natl. Acad. Sci. U. S. A. 1981; 78: 2923-2926Crossref PubMed Scopus (32) Google Scholar). This could allow movement of PABP between the 3′ poly(A) tail and other lower affinity PABP binding sequences (37Gorlach M. Burd C.G. Dreyfuss G. Exp. Cell Res. 1994; 211: 400-407Crossref PubMed Scopus (213) Google Scholar) including the ARS at the 5′ UTR of PABP mRNA. How efficiently PABP would interact with the lower affinity binding sequences may depend on its cellular level. Presumably an optimal level of PABP is required for maintaining an equilibrium between its various functions. Our studies have shown that the presence of the putative PABP-binding site at the 5′-UTR of PABP mRNA reduces by approximately 3-fold the translation of this mRNA both in vitro (28Bag J. Wu J. Eur. J. Biochem. 1996; 237: 143-152Crossref PubMed Scopus (46) Google Scholar) and in vivo (Fig. 3). Although the presence of a certain excess amount of PABP is tolerated and may even be required for normal cellular function, its expression beyond a threshold could be controlled at the mRNA translation level. Excess PABP beyond the threshold level may detrimentally alter the pattern of protein synthesis in the cell, for example by increasing translation of inherently inefficient mRNAs. In addition stability of some unstable mRNAs may also increase if excess PABP is made. Together, these events could have a dramatic effect on the growth regulation and cellular differentiation. This possibility is supported by the previous observation that overexpression of eIF4E results in cellular transformation (41Lazaris-Karatzas A. Montine K.S. Sonenberg N. Nature. 1990; 345: 544-547Crossref PubMed Scopus (799) Google Scholar). Since the role of PABP in translation is believed to be mediated through its interaction with the cap structure and eIF4G (of which eIF4E is a component), it will be important to control PABP expression through feedback mechanisms. This is further supported by the observation that overexpression of PABP in Xenopus oocytes influenced the deadenylation and translation of maternal mRNAs (42Wormington M. Searfoss A.M. Hurney C.A. EMBO J. 1996; 15: 900-909Crossref PubMed Scopus (91) Google Scholar). We thank David Murray for constructing PCMVΔPABP and PCMV88ΔPABP and Neil Maharaj for measuring the ectopic PABP expression as a partial fulfillment of the requirement of their undergraduate research project program. Our sincere thanks to M. Choudhury for help in the preparation of the diagrams. |
|---|