Документ взят из кэша поисковой машины. Адрес оригинального документа : http://rnp-group.genebee.msu.su/pdfs/febsz.pdf Дата изменения: Mon Oct 21 17:39:11 2002 Дата индексирования: Mon Oct 1 19:39:20 2012 Кодировка:

An extremely high conservation of RNA-protein S7 interactions during prokaryotic ribosomal biogenesis. Vera A. Spiridonova1, Andrey V. Golovin, Denis Yu. Drygin, Alexei M. Kopylov* Chemistry Department and
1

Institute of Physico Chemical Biology of Moscow State

University, 119899 Moscow, Russian Federation. Tel.: (095) 939-3149, FAX: (095) 939-3181, E-mail: kopylov@rnp.genebee.msu.su

Abstract. Direct determination of RNA-protein complex structures is often facilitated by the use of thermophilic proteins; however E.coli is the most investigated system so far. A hybrid approach is to form heterologous complexes of E.coli RNA with thermophilic proteins. The rationale for this approach to RNA-protein interactions in ribosomes is based on the ability of the thermophilic protein S7 to replace a homologous counterpart in vivo. In vitro, the protein S7 of Thermus thermophilus is able to form complexes with both the minimal 16S rRNA fragment and the intercistronic region of the str operon mRNA from E.coli (Kd = 1.4x10-7 M and 1.1x10-7 M respectively). The interaction of Thermus S7 with the E. coli intercistronic mRNA is surprising, because this region does not exist in the thermophilic str operon. It suggests a high degree of conservation of an RNA-binding site on S7.

Key words: RNA-protein complexes; ribosomal protein S7 of Thermus thermophilus; the 3'domain of the 16S rRNA fragment; str operon; Escherichia coli *Corresponding author.


1. Introduction

The ribosome is a paradigm of RNA-protein interactions. Not only do ribosomal proteins interact with rRNA during ribosomal assembly, but some of them also interact with their own message and regulate the translation of their operon. This translational feedback control is a characteristic of both the small and the large ribosomal subunit biogenesis [1 ].

The two main proteins that initiate folding of 16S RNA in the assembly of the small ribosomal subunit are S4 and S7 [2], and both are involved in translational feedback by binding to mRNA from their respective operons. During ribosome assembly, S7 binds to the 3'-terminal domain of the 16S rRNA and stimulates the binding of the other ribosomal proteins to this domain of the small subunit [2, 3]. Brakier-Gingras and coworkers have shown that just a small fragment of the 3'-end domain of the 16S rRNA is necessary for S7 binding [4, 5]. The gene for S7 (rpsG) is a part of the str operon, which contains in addition, the genes for ribosomal protein S12 as well as the elongation factors EF-G and EF-Tu [6]. When S7 is present in excess over ribosomal RNA, it interacts with the str mRNA and shuts off its own synthesis [7]. Nomura and Saito [8] have shown that during feedback inhibition, it protects from chemical modification about 60 nucleotides of the str mRNA within the S12-S7 intercistronic region, and it inhibits the coupled translation of the S7 cistron. A residual translation of the S7, initiating from its cistron, is not suppressed.

These results show that regulation of ribosome synthesis in the cell is based mainly on two types RNA-protein interactions: the protein S7-16S rRNA and the protein S7- str mRNA. A high resolution structural study will provide an understanding of the mechanism of ribosomal biogenesis as well as shed light on the poorly understood nature of RNA-protein interactions in general. For structural research, thermophilic proteins have an advantage compared to


mesophilic ones [9-11]. This was the reason we selected Thermus thermophilus protein S7 for structural work. A heterologous approach that combined the use of Thermus S7 with E. coli RNA was chosen to study RNA-protein interactions. The rationale for using this approach is the interchangability of mesophilic and thermophilic protein during ribosomal assembly in vivo [12]. This paper describes the first stage of this research, and will show that thermophilic S7 has a highly conserved RNA-binding site. It is able to interact with high affinity with fragments of both 16S rRNA and intercistronic str mRNA (Kd = 1.4x10-7 and 1.1x10-7 M respectively). The latter case is surprising, as thermophilic str operon does not have an extended intercistronic region [13].

2. Materials and methods

2.1. Construction of the RNA expression plasmids. The plasmid pFD3LH-1-2-3a, coding for the E.coli minimal 16S rRNA binding site for the protein S7 under control of the T7 promoter [4], was kindly provided by Prof. Brakier-Gingras (Montreal University, Canada).

The S12-S7 intercistronic region of the E.coli str operon was cloned first as a Bam H1 fragment of the plasmid pNO 3070 (kindly provided by Prof. Nomura, University of

California, Irvine, USA) into the expression vector pGEM-3Z (Promega, USA) under control of the T7 promoter. The insert of the resulted plasmid, pDD 590, was truncated with Sau 1 and Eco R1 and self-ligated (pDD 157). The structure was verified by dideoxy sequencing.

2.2. The filter binding assay. Thermus thermophilus S7 was expressed as a recombinant protein in E. coli and was isolated as described [12] (prepared by Karginov).


RNA was made by in vitro transcription with T7 RNA polymerase (MBI, Fermentas, Lithuania) as described [14]. Concentrations of UTP and 32P-UTP (Physico Energy Institute, Obninsk, Russia) were adjusted to have specific radioactivity about 2,000 cpm/pmole. The RNA was purified by PAG electrophoresis. After extraction, both radioactivity and absorption A260 were measured and specific radioactivity was calculated. 22 A260 OU was equal to 1 mg of RNA.

Complex formation was done according to Brakier-Gingras [4] using NH4Cl instead of KCl in 100 µl of the buffer with a fixed RNA concentration of about 70-100 nM and increasing concentration of the protein (30 nM stock in 1M NH4Cl) up to a ratio of 1:30 of RNA:protein. Filters of type GS, 0.22 µm (Millipore) were used, and rinsed after binding twice with 0.5 ml of the binding buffer. The filters were counted by Cherenkov. As a control, the RNA without protein was used, as well as the mixture of the protein with viral RNA fragment (kindly provided by Dr. Shatsky). Controls, usually 3-5%, have been substracted from the values.

3.Results and discussion

To study the RNA-binding properties of the thermophilic protein S7, we have chosen a phylogenetic approach [15], based on the idea that a similar function has to be performed by a similar structure. In our case the rationale of this approach is the ability of thermophilic protein S7 to replace the mesophilic counterpart during in vivo assembly of E.coli ribosomes [12]. At the least, this means that the protein is able to interact with the 16S rRNA binding site in ribosome assembly. We verified this fact by direct in vitro binding using the minimal 16S rRNA fragment of Brakier-Gingras (pFD3LH-1-2-3a) [4]. It turned out that the

thermophilic protein S7 binds to the E.coli 16S rRNA fragment more efficiently than the


homologous protein (Kd = 1.4x10-7 M versus 1.6x10-6 M respectively), though with a lower level of saturation (Fig. 1, [4]). The apparent binding constants are different by about an order of magnitude. It is worth noting that a comparison of the RNA primary structures revealed only 5 nucleotides difference [16]. At present it is difficult to state the exact nature of the difference in binding affinity. Either it is because of the thermophilic nature of the protein, or because of the presence an additional N-terminal tail [12] which might stabilize the complex. A construct of the protein that lacks this additional tail (unpublished data) will help to distinguish between these two possibilities. It cannot be excluded that our protocol of isolation of the thermophilic protein under native conditions gives the protein a higher binding affinity than the one Brakier-Gingras used [4]. Some examples of successful heterologous complex formation for other ribosomal proteins have also been reported [17-21].

Our success in forming a stable heterologous complex with E.coli 16S rRNA fragment allowed us to proceed with the next step: to check the ability of thermophilic protein S7 to bind the E.coli S12-S7 intercistronic region, though we were aware that thermophilic str operon does not have this extended intercistronic region [13]. Nomura and Saito have shown that S7 protects from chemical modification a relatively small region of about 60 nucleotides that is part of a more extended transcript [8]. We have used a convenient location of restriction sites to clone an intercistronic region that is 157 nucleotides long ( -106 +42 of str mRNA, the first A in S7 AUG is +1; the rest of the nucleotides are from the vector polylinker) under control of T7 promoter. It has all the regulatory elements: the termination triplet of S12, an apparent binding site of S7, and a site of S7 translational initiation. Preliminary experiments with E.coli protein S7 (kindly provided by Prof. Nierhaus, Max-Planck-Institut fur Molekulare Genetik, Berlin, Germany) demonstrated that this RNA fragment indeed binds E.coli protein S7 (data not shown).


The most striking result of our study is that thermophilic protein S7 binds strongly to the E.coli S12-S7 intercistronic region of str mRNA in vitro, inspite of the lack of the functionally analogous thermophilic extended region. It turned out that the dissociation constant is about the same as for the thermophilic protein-E.coli 16S rRNA fragment (Fig. 1), that is much higher than for E.coli homologous system. The saturation level is close to that for the E.coli homologous system. This binding is specific and the most stable of all of the known complexes of RNA with the S7 [4, 5].

The existence of this binding reveals a high conservation of a putative RNA-binding site of S7. If the protein S7 had had two separate RNA-binding sites for both rRNA and its own mRNA, the intercistronic mRNA-binding site of thermophilus S7 should have been absent. Most probably, the fact that the rRNA and mRNA E.coli S7 target sites do not resemble each other implies specific recognition by S7 of a common RNA structure ( for example, see [22]).

On the contrary, it may be suggested that inspite of the lack of the extended S12-S7 intercistronic region in the thermophilic str operon, a structure similar to the intercistronic region of E.coli might be present in the thermophilic S12 and/or S7 coding regions close to the translational initiation site of S7. Since the mode of regulation of the str operon of Thermus thermophilus has not been clarified, it cannot be excluded that this operon is regulated at the level of S12 translation, while other cistrons are regulated due to the coupled translation as a single polycistronic mRNA. In this case the intercistronic region of E.coli str operon is a new feature acquired during evolution, which took an additional regulatory function. A comparative analysis of the fine structure of str operons for different species will make this clear [6].


In conclusion, this work shows that a highly stable, heterologous complex can be formed from thermophilic ribosomal protein S7 with the E.coli S12-S7 intercistronic region of str mRNA, despite the absence of this region in the Thermus operon. This complex can be the basis for the study of protein-RNA interactions and the mechanism of translational feedback in ribosome biogenesis.

Acknowledgments: The authors are grateful to Prof. M.Nomura for providing us with the plasmid pNO 3070, to Prof. L.Brakier-Gingras for providing us with the plasmid pFD3LH-12-3a, to Prof. K.Nierhaus for providing us with E.coli protein S7, to A.Karginov for isolation of T. thermophilus protein S7, to V.Sergeev for help in sequencing, to E.Ekimova for help, to Prof. V.Ramakrishnan for discussion and reading the manuscript, and to Prof. A.Bogdanov for help and support. This work was supported by Russian Foundation for Basic Investigation (95-04-12823a), the program Universities of Russia (uni-009-95).


References

[1] Nomura, M., Gours, R. and Baughman, G. (1984) Annu. Rev. Biochem., 53, 75-117. [2] Nowotny, V. and Nierhaus, K.H. (1988) Biochemistry, 27, 7051-7055. [3] Samaha, R.R., O'Brien, B., O'Brien, T.W. and Noller, H.F. (1994) Proc. Natl. Acad. Sci. USA, 91, 7884-7888. [4] Dragon, F. and Brakier-Gingra, L. (1993) Nucleic Acids Res., 21, 1199-1203. [5] Dragon, F., Payant, C. and Brakie-Gingras, L. (1994) J. Mol. Biol., 244, 74-85. [6] Zengel, J.M. and Lindahl L. (1994) Prog. in Nucl. Acid Res. and Mol. Biol., 47, 331-370. [7] Saito, K., Mattheakis, L.C. and Nomura, M. (1994) 235, 111-124. [8] Saito, K. and Nomura, M. (1994) J. Mol. Biol., 235, 125-139. [9] Ramakrishnan, V., Davies, C., Gerchman, S.E., Golden, B.L., Hoffmann, D.W., Jaishree, T.N., Kycia, J.H., Porter, S. And White S.W. (1995) Biochem. Cell. Biol., 73, 979-986. [10] Liljas, A. And Garber, M. (1995) Curr. Opin. Struct. Biol., 5, 721-727. [11] Markus, M.A., Hink, A.P., Huang, S., Draper, D.E. and Torchia, D.A. (1997) Nature Struct. Biol., 4, 70-77. [12] Karginov, A.V., Karginova, O.A., Spiridonova, V.A. and Kopylov A.M. (1995) FEBS Lett., 369, 158-160. [13] Yakhunin, A.V., Vorozheykina, D.P. and Matvienko, N.I. (1990) Nucleic Acids Res.,18, 3659. [14] Gurevich, V.V., Pokrovskaya, I.D., Obukhova, T.A. and Zozulya, S.A. (1991) Biochem., 195, 207-213. [15] Draper, D.E. (1995) Annu. Rev. Biochem., 64, 593-620. Anal.


[16] Maidak, B.L., Olsen, G.J., Larsen, N., Overbeek, R., McCaughey, M.J. and Woese, C.R. (1996) Nucleic Acids Res., 24, 82-85. [17] Thurlow, D.L. and Zimmermann, R.A. (1978) Proc. Natl. Acad. Sci. USA, 75, 2859-2863. [18] Gongadze, G.M., Tishchenko, S.V., Sedelnikova, S.E. and Garber, M.B. (1993) FEBS Lett., 330, 46-48. [19] Vysotskaya, V., Tischenko, S., Garber, M., Kern, D., Mougel, M., Ehresmann, C. and Ehresmann, B. (1994) Eur. J. Biochem., 223, 437-445. [20] Evers, U., Franceschi, F., Boddeker, N. and Yonath, A. (1994) Biophis. Chem., 50, 3-16. [21] Xing, Y. and Draper, D.E. (1995) J. Mol. Biol., 249, 319-331. [22] Berglund, H., Rak, A., Serganov, A., Garber, M. and Hard, T. (1997) Nature Struct. Biol., 4, 20-23.


The figure 1 legend.

Fig. 1. Nitrocellulose filter retention assays. The fraction of

32

P-labeled RNA retained on

filters is plotted as a function of the protein S7 concentration. The curves are best fited to the indicated sets of data with the computer program Microcol Origin 4.00 (USA). Binding of T. thermophilus S7 to E.coli [!] S12-S7 intercistronic str mRNA fragment; [t] 16S rRNA fragment. The apparent dissociation constant (Kd) is estimated as the concentration of the protein required to give 50% saturation.


40 35 30 25 20 15 10 5 0 0,0 0,2 0,4 0,6 0,8 1,0 1,2

% of RNA retention

Protein concentration (µ M)