Communication Vol. 258, No. 3, Imue of February 10, pp. 1387-1390, 1983 THE JOURNAL OF BIOLOGICAL CHEMISTRY

Prrnted in U.S. A.

Messenger RNA for the 73,000- dalton Poly(A)-binding Protein Occurs as Translationally Repressed mRNP in Duck Reticulocytes*

(Received for publication, September 17, 1982) Kinsey MaundrellS, Maria Tereza Imaizumi- Scherrer, E. Stuart Maxwell§, Olivier CivelliT, and Klaus Scherrer From the Znstitut de Recherches en Biologie Moleculaire, Tour 43, 2 Place Jussieu, 75221 Paris Cedex 05 France

Poly(A)-containing mRNA has been prepared from the polyribosomes and post-polyribosomal mRNP frac- tion of duck reticulocytes. The coding capacity of the respective mRNA populations has been examined by translation in vitro followed by two-dimensional elec- trophoresis of the 35S-labeled polypeptides. A detailed analysis of these results is given elsewhere (Imaizumi- Scherrer, M.-T., Maundrell, K., Civelli, O., and Scherrer, K. (1982) Dev. Biol. 93,126-138). Here, we focus on one of these translation products which migrates as a slightly basic protein of 73,000 molecular weight. By two-dimensional electrophoretic analysis and partial peptide mapping, we show that this protein is indistin- guishable from the poly(A)-binding protein. We con- clude that the majority of the coding sequences for this protein are translationally repressed in the reticulocyte cytoplasm.

Two classes of mRNA have been identified in the cytoplasm of eukaryotic cells, one class associated with polyribosomes and actively engaged in protein synthesis (1, 2), the other, translationally repressed and isolated as more slowly sedi- menting RNP’ complexes from the post-polyribosomal super- natant (3, 4). Investigation of the respective mRNA popula- tions by in vitro translation and kinetic hybridization has shown that most, and probably all, coding sequences present in polyribosomes have a counterpart in the nontranslated fraction; however, the distribution between the two pools

* This research was supported by Centre Nationale de la gecherche Scientifique Grant ATP 13 557, Delegation Generale a la Recherche Scientifique et Technique Grant 7 670 710, the Fondation pour la Recherche Medicale Francaise, and Institut Nationale de la Sante et de la Recherche Medicale Grant 764 1191. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Recipient of a European Molecular Biology Organization Fellow- ship. Present address is Biochemistry Laboratory, Biology Building, University of Sussex, Falmer, Brighton BN19QG, England. To whom correspondence should be addressed.

5 Fellow of the Damon Runyon Cancer Fund. Present address is Department of Biology, University of Chicago, 1103 East 57th Street, Chicago, IL 60637.

7 Boursier de la Fondation pour la Recherche Medicale Francaise. Present address is Department of Chemistry, College of Liberal Arts, University of Oregon, Eugene, OR 97403.

’ The abbreviations used are: RNP, ribonucleoprotein; SDS, so- dium dodecyl sulfate; kDa, kilodaltons.

varies widely for individual mRNA species, and appears to be independently regulated in each case (5-7). In a recent report, we have made a detailed examination of mRNA distribution in the duck reticulocyte cytoplasm (7). In keeping with the massive synthesis of globin which occurs at this stage of red blood cell maturation, we find that 92% of the globin coding sequences are polyribosome-associated while the remainder are found in the postpolyribosomal supernatant. In the pres- ent report, we examine another mRNA of the duck reticulo- cyte cytoplasm, in this case, one present predominantly in the translationally nonexpressed fraction. The translation product of this mRNA is the 73-kDa poly(A)-binding protein found associated with the 3’ end of polyribosomal mRNA (8) and under other experimental conditions with the residual protein skeleton of erythroblast nuclei (9). The intracellular distribu- tion of this mRNA is considered in relation to the general shutdown of mRNA synthesis previously shown to accompany red blood cell maturation (10).

MATERIALS AND METHODS

Preparation of mRNA from the red blood cells of anaemic ducks and mRNA translation in vitro were described recently (7). Duck blood was collected directly onto emetine (200pg/ml, final concentra- tion) to prevent ribosome run off, and lysed by hypotonic shock. The lysate was fractionated immediately on a Spinco Ti15 zonal rotor at 32,000 rpm for 2 h and the gradient was pooled into polyribosomes and postpolyribosomal mRNP particles on the basis of AZw profies and protein composition of selected fractions. Poly(A)-containing mRNA was prepared from each fraction by oligo(dT)-cellulose chro- matography and translated in a micrococcal nuclease-treated rabbit reticulocyte lysate. One-dimensional and two-dimensional analyses of proteins were performed according to published procedures in Refs. 11 and 12, respectively; solubilization of protein samples in each case is detailed in the appropriate figure legends. Affinity chromatography on poly(A)-sepharose was based on the procedure of Fukami and Itano (13). Peptide mapping in SDS-polyacrylamide gels was as described by Cleveland et al. (14).

RESULTS AND DISCUSSION

Preparations of deproteinized mRNA isolated from the polyribosomal and postpolyribosomal subfractions of duck reticulocyte cytoplasm show comparable template activity when translated in vitro in a nuclease-treated rabbit reticu- locyte lysate. The ”S-labeled translation products of the respective mRNA populations separated by two-dimensional electrophoresis are shown in Fig. 1. As discussed in more detail elsewhere (7), numerous, essentially quantitative differ- ences can be seen in this comparison. The protein we have focused on in the present study migrates as slightly basic polypeptide with a molecular weight of 73,000 (arrows in Fig. 1). This protein is considerably more abundant among the translation products of postpolyribosomal mRNA.

Conclusions regarding mRNA distribution based on in vitro translation must be made cautiously since mRNA isolated from different cytoplasmic fractions may not be translated with equal efficiency. However, under the conditions used here, equal amounts of mRNA template from polyribosomes and postpolyribosomal RNP produced virtually identical in- corporation of [35S]methionine. Thus, the relative incorpora- tion into the 73-kDa protein should give an indication of the respective amounts of mRNA template present in each frac- tion. By excision and counting of appropriate regions of the two gels shown, we estimate that about 90% of the coding sequences for the 73-kDa protein are in the translationally

1387

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

1388 mRNA for the Poly(A)-binding Protein in Duck Reticulocytes

0 0

FIG. 1. Two-dimensional electrophoretic comparison of the proteins encoded by poly(A') RNA from polyribosomes and free RNP. 0.5 pg of poly(A') mKNA from polyribosomes ( a ) and free mKNP ( 6 ) was translated in a nuclease-treated rabbit reticulo- cyte lysate as described elsewhere (7). 5 pl of each reaction mixture containing approximately 6 X 10" cpm was solubilized by addition of 20 pI of 9.5 M urea, 2% Nonidet 1'-40,2% ampholines (pH range 3-10), 5% 2-mercaptoethanol, and incubated for 30 min at 37 "C. Two- dimensional electrophoresis using isoelectric focusing in the first dimension and SDS electrophoresis in the second dimension was run according to OFarrell et al. (12). "'S-labeled translation products were detected by fluorography (18). Molecular weight markers in- cluded at the edge of the second dimension gels were phosphorylase a (90,000). bovine serum albumin (68,OOO). ovalbumin (43,000). DNase I (31,OOO), a-chymotrypsinogen (25,000). and myoglobin (17,000).

repressed fraction. In overall agreement with these results, recent experiments on the reticulocyte proteins labeled in vivo showed no detectable incorporation of radioactive precursors into the 73-kDa protein.2 These data taken together argue that the mRNA for the 73-kDa protein is predominantly, perhaps in some cells totally, sequestered in the repressed mRNP fraction.

Because the two-dimensional electrophoretic mobility of this 73-kDa protein is similar to that found for the poly(A)- binding protein, we have undertaken a more detailed exami- nation of its possible identity. A bulk preparation on nonra- dioactive poly(A)-binding protein was obtained from the sol- uble fraction of duck reticulocyte cytoplasm by affinity chro- matography on poly(A)-sepharose. Proteins were applied in low ionic strength buffer as described in the legend to Fig. 2, and bound proteins were eluted sequentially with 0.5 M KCl, and 2 M LiCl, 1 M urea according to their increasing affinities for poly(A) (13). The most strongly bound fraction shown in Fig. 2, lune e, represents 0.01% of the Anxn absorbing material applied to the column, and comprises relatively few polypep- tides of which by far the most abundant has a molecular weight of 73,000. The 73-kDa poly(A)-binding protein from the soluble protein fraction of the reticulocyte lysate is indis- tinguishable from the mRNA-bound poly(A)-binding protein in two-dimensional mobility:' and amino acid composition (15). In addition, the corresponding proteins from mouse L cells have the same peptide maps when digested with protease V8 (19). Thus, we assume the poly(A)-binding proteins from the cytosol and polyribosomes to be the same molecular species.

In the experiment reported in Fig. 3, 5 pg of the 73-kDa

0. Akkayat, unpublished observations. E. S. Maxwell and A. Vincent, unpublished observations.

protein purified electrophoretically from the LiCl/urea eluate was added to a rabbit reticulocyte lysate which had previously been used to translate duck reticulocyte postpolyribosomal mRNA (compare Fig. lb). The combined proteins were sep- arated by two-dimensional electrophoresis. Comparison of the stained polypeptides (Fig. 3u) with the autoradiograph of the same gel (Fig. 3h), reveals that the added poly(A)-binding protein and the 73-kDa translation product of postpolyribo- soma1 mRNAs have identical migration in two dimensions.

To investigate further the identity between these two pro- teins, we have carried out partial peptide mapping experi- ments illustrated in Fig. 4. In this case, 40 pg of the poly(A)- binding preparation shown in Fig. 2, lune e, was added to the in vitro translation products of nonpolyribosomal mRNA, and again separated by two-dimensional electrophoresis as in Fig. 3. It should be noted here that the endogenous 73-kDa protein of the rabbit reticulocyte lysate co-migrates with its added duck counterpart (20); however, the trace amounts present do not interfere with the experiments reported here. After elec- trophoresis, the gel was stained briefly in ethidium bromide without fixation (16) and the region of the gel containing the 73-kDa protein was visualized by UV illumination and excised. This gel fragment containing both the exogenous 73-kDa protein and the "'S-labeled in vitro translation product was transferred to the sample well of a "third dimension" SDS- polyacrylamide gel and overlaid with a-chymotrypsin. Migra- tion was interrupted in the stacking gel to allow proteolysis to occur (14), and the cleavage products were resolved by elec- trophoresis into the lower gel. The resulting peptide map visible after Coomassie blue staining (Fig. 4, lune c) was compared to the fluorograph of the same slot (Fig. 4, lune c'). The similarity between the two partial digests is striking. All peptides present in the stained preparation are present in the fluorograph; moreover, the slightly aberrant migration is du-

a b c d e

90-1

""1 431

311

FIG. 2. SDS-polyacrylamide gel electrophoresis of duck re- ticulocyte lysate fractionated by affinity chromatography on poly(A)-sepharose. Preparation of anemic duck blood and cell fractionation following hypotonic shock has been described elsewhere (7). 1500 AZn, units of the soluble protein fraction recovered from the 4 S peak after zonal centrifugation was dialyzed against 15 mM KCI, 10 mM Tris/HCI, pH 7.4, 1 mM EDTA, and applied at 22 "C to a poly(A)-sepharose column at a flow rate of 16 ml/h. The column was washed overnight at a flow rate of 24 ml/h with 0.1 M NaCI, 10 mM Tris, pH 7.4, when the Ann, reading of the effluent fell to zero. Proteins were eluted using the procedure of Fukami and Itano (13). with 0.5 M KCI, 25 mM Tris, pH 7.4, until no further A.w, absorbing material was released and then with 2 M LiCI, 1 M urea, 50 mM Tris, pH 7.4. Eluted proteins were precipitated with 9 volumes of acetone and the pellets dissolved in SDS sample buffer (11). Lane a, marker proteins as in Fig. 1; lane b, soluble proteins applied to poly(A)-sepharose; lane c, proteins in the unbound fraction; lane d, proteins eluted in 0.5 M KCI; and lane e, proteins eluted in 2 M LiCI, 1 M urea.

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

mRNA for the Poly(A)-binding Protein in Duck Reticulocytes 1389

plicated in both patterns. The peptide map thus establishes the identity of the 73-kDa product of in vitro translation as the poly(A)-binding protein. The presence of additional minor bands in the fluorograph is unexplained at present. The most likely explanation is that they originate from a contaminating polypeptide also present in the excised gel fragment. Finally, in passing, it is interesting to note that the mRNA for the poly(A)-binding protein is itself polyadenylated since mRNAs for in vitro translation were purified by oligo(dT)-chromatog- raphy. This, therefore, is a specific example of a protein which can bind its own mRNA.

In the reticulocyte cytoplasm, 88% of the total poly(A)- containing mRNA is present in polyribosomes and 12% is in nontranslated RNP complexes. As we show elsewhere (7), both mRNA fractions consist of three frequency classes of which the most abundant in each case is made up of globin mRNA. From the data in Fig. 1, it can be seen that the mRNA for the poly(A)-binding protein is in the rare class of polyri-

"1 a 43

31.

25

b 90- 68-

43-

31 -

25-

*

8

FIG. 3. Two-dimensional comparison between the 73-kDa poly(A)-binding protein and the in uitm translation products of postpolyribosomal poly(A+) mRNAs. 5 p g of electrophoreti- cally purified poly(A)-binding protein was added to 5 pl of a rabbit reticulocyte lysate containing the in vitro translation products of nonpolyribosomal mRNAs. 20 pI of 9.5 M urea, 2% Nonidet P-40, 2% ampholines (pH range 3-10), 5% 2-mercaptoethanol was added and the mixture was incubated a t 37 "C for 30 min. Analysis of proteins by two-dimensional electrophoresis using nonequilibrium isotacho- phoresis in the first dimension and SDS-polyacrylamide electropho- resis in the second dimension is described by OFarrell et al. (12). Molecular weight marker proteins (see Fig. 1) were run at the edge of the second dimension gel. After staining in Coomassie blue, the gel was treated for fluorography as described (18) and exposed to x-ray film a t -80 "C. The arrow in Q shows the added 73-kDa poly(A)- binding protein which conceals a trace amount of endogenous protein seen when the lysate is analyzed alone. The arrow in b indicates the product of in vitro translation which co-migrates exactly with the poly(A)-binding protein.

a b c C'

31 - - 25- t c c

17- -

FIG. 4. Analysis of peptides from the partial a-chymotryptic digest of the 73.000 molecular weight protein. Approximately 40 pg of LiCl/urea eluate from poly(A)-sepharose was precipitated with acetone and resuspended in 25 p1 of 9.5 M urea, 2% Nonidet P-40,2% ampholines (pH range 3-10), 5 mM 2mercaptoethanol a t 37 "C for 1 h. 5 pl of rabbit reticulocyte lysate containing the in vitro translation products of postpolyribosomal mRNAs was added and incubation continued for a further 30 min a t 37 "C. The solubilized proteins were separated by two-dimensional electrophoresis as described in the legend to Fig. 3. After electrophoresis, the gel was washed with water for 15 min, then transferred to 22 pg/ml of ethidium bromide for 20 min and illuminated with UV light as described (16). The 73-kDa protein was clearly visible. This region of the gel was excised and equilibrated with 0.1% SDS, 0.125 M Tris, pH 6.8, for 30 min at room temperature. Meanwhile, a "third dimension" gel of 16% acrylamide containing 1 mM EDTA throughout was prepared (14). The stacking gel was longer than normal to accommodate deeper slots. The equil- ibrated gel fragment was chopped into small pieces and transferred to one of the slots (lane c). It was overlaid with 5 pg of freshly prepared a-chymotrypsin (Worthington). Adjacent wells contained the same quantity of enzyme alone (lane b ) and marker proteins as detailed in Fig. 1 (lane a ) . Electrophoresis was interrupted for 1 h during stacking to allow proteolysis to occur, and peptides of the partial digest were analyzed in the lower gel. Digestion of both samples together in this way ensured that the enzyme/substrate ratio for both proteins was the same. Following Coomassie blue staining, the gel was treated for fluorography and exposed to x-ray film for 4 days a t -80 "C to reveal the pattern of labeled peptides (lane c'). Asterisks indicate the labeled bands not represented in the stained pattern.

bosomal mRNAs and in the intermediate frequency class of postpolyribosomal mRNAs. This distribution, in which the equilibrium is towards the translationally repressed fraction, is consistent with the reduced synthesis of poly(A)-containing mRNA, and thus a reduced requirement for the poly(A)- binding protein characteristic of later stages of red blood cell maturation. It is markedly different from the distribution of globin mRNA which, as shown elsewhere, is predominantly polyribosome-associated in reticulocytes (7). This demonstra- tion of differential translation seen in the case of two specific coding sequences implies a means of discriminating between different mRNAs and we are currently investigating this ques- tion in terms of the structure of cytoplasmic mRNP com- plexes. Nonrandom selection of mRNAs for translation is one of the final levels in the cascade of regulatory steps we envisage to be involved in the control of gene expression (17).

Acknowledgments-We thank Alain Vincent, Sylvie Marionnet, and Samuel Goldenberg for help and advice in carrying out these experiments.

REFERENCES 1. Perry, R. P., and Kelley, D. E. (1968) J. Mol. Biol. 35.37-59 2. Morel, C., Gander, E., Henberg, M., Dubochet, d., and Scherrer,

K. (1973) Eur. J. Biochem. 36.455-4M

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

1390 mRNA for the Poly(A)-binding Protein in Duck Reticulocytes

3. Spirin, A. S. (1966) Curr. Top. Deu. Bwl. 1, 1-38 12. O'Farrell, P. Z., Goodman, H., and OFarrell, P. H. (1977) Cell 12, 4. Vincent, A., Civelli, O., Maundrell, K., and Scherrer, K. (1980) 1133-1142

Eur. J. Biochem. 112,617-633 13. Fukami, H., and Itano, H. A. (1976) Biochemistry 15,3529-3535 5. Walters, R. A., Yandell, P. M., and Enger, M. D. (1979) Biochem- 14. Cleveland, D. W., Fischer, S. G., Kirschner, M. W., and Laemmli,

istry 18,4254-4261 U. K. (1977) J. Biol. Chem. 252, 1102-1106 6. Infante, A. A, and Heilmann, L. J. (1981) Biochemistry 20, 1-8 15. Mazur, G., and Schweiger, A. (1978) Biochem. Biophys. Res. 7. Imaizumi-Schemer, M.-T., Maundrell, K., Civelli, O., and Scher- Commun. SO, 39-45

8. Vincent, A., Goldenberg, S., and Schemer, K. (1981) Eur. J. 17. Schemer, K. (1981) in Eukaryotic Gene Regulation (Kolodney,

9. Maundrell, K., Maxwell, E. S., Puvion, E., and Schemer, K. (1981) 18. Laskey, R. A., and Mills, A. D. (1975) Eur. J. Biochem. 56,335-

rer, K. (1982) Deu. Biol. 93, 126-138 16. Vincent, A., and Scherrer, K. (1979) Mol. Biol. Rep. 5, 209-214

Biochem. 114, 179-193 G., ed) C. R. C. Press, Boca Raton, Florida

Exp. Cell Res. 136,435-445 34 1

(1966) Proc. Natl. Acad. Sci. U. S. A. 56, 1511-1518 20. Standart, N., Vincent, A., and Scherrer, K. (1981) FEBS Lett. 10. Schemer, K., Marcaud, L., Zajdela, F., London, I., and Gros, F. 19. Setyono, B., and Greenberg, J. R. (1981) Cell 24, 775-783

11. Laemmli, U. K. (1970) Nature (Lond.) 227,680-685 135,56-60

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from

K Maundrell, M T Imaizumi-Scherrer, E S Maxwell, O Civelli and K Scherrertranslationally repressed mRNP in duck reticulocytes.

Messenger RNA for the 73,000-dalton poly(A)-binding protein occurs as

1983, 258:1387-1390.J. Biol. Chem. 

  http://www.jbc.org/content/258/3/1387Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/258/3/1387.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on January 11, 2021http://w

ww

.jbc.org/D

ownloaded from