Reactivation of denatured proteins by domain V of bacterial 23S rRNA
Reactivation of denatured proteins by domain V of bacterial 23S rRNADebashis Pal1, Subrata Chattopadhyay, Suparna Chandra, Dibyendu Sarkar, Abhijit Chakraborty2 and Chanchal Das Gupta*
Department of Biophysics, Molecular Biology and Genetics, University of Calcutta, 92 APC Road, Calcutta 700009, India, 1Department of Zoology, Uluberia College, Uluberia, West Bengal, India and 2Biophysics Division, Saha Institute of Nuclear Physics, Calcutta 700037, India
Received August 14, 1997;Revised and Accepted October 20, 1997
ABSTRACT
In vitro transcripts containing domain V of the 23S rRNA of Escherichia coli and Bacillus subtilis can reactivate denatured proteins almost as efficiently as the total 23S rRNA. Here we show that almost the full length of domain V is required for reactivation of denatured pig muscle lactate dehydrogenase and pig heart cytoplasmic malate dehydrogenase: the central loop of this domain alone is not enough for this purpose. The antibiotic chloramphenicol, which binds to domain V of 23S rRNA, can inhibit reactivation of these proteins completely. Activity is eliminated by EDTA at a concentration of <1 mM, even in the presence of 4 mM MgCl2, suggesting that the three-dimensional conformation of the RNA should be maintained for this activity.
Ribosomes from both prokaryotic and eukaryotic sources can reactivate several denatured proteins and, therefore, may possess a general protein folding activity (1-3). In Escherichia coli ribosomes such an activity has also been traced first to the 50S subunit and finally to its 23S rRNA component (2,4). Strong evidence has been obtained in favour of a similar role of 50S subunits/23S rRNA in vivo (manuscript submitted). Since the polypeptide chain forms on the 50S particle, one can reasonably expect it to play some role in folding of the chain (5-9). The 23S rRNA appears to play a more active role in protein synthesis than was appreciated earlier. The direct participation of 23S rRNA in the peptidyl transferase reaction during polypeptide chain elongation was demonstrated a few years ago (10). The sites of association of the ribosome with aminoacyl tRNA and antibiotics which inhibit protein synthesis have also been located on 23S rRNA (10,11). Domain V of 23S rRNA has been assigned as playing crucial roles in these processes. We have shown complete elimination of the protein folding activity of 23S rRNA by antibiotics specifically binding to domain V as well as oligodeoxynucleotides annealing with the single-stranded central loop of domain V (12). However, in all experiments reported so far on the different activities of 23S rRNA it was extracted from the 50S ribosomal subunit by the usual phenol extraction procedure (10,12). The activities have not been shown in 23S rRNA transcribed in vitro from its gene. We have transcribed only domain V of 23S rRNA and its different parts in vitro from appropriate clones and have shown that the in vitro transcript containing almost the entire domain V has protein folding activity. If the central elaborate stem-loop region is deleted from domain V then the shorter transcripts fail to reactivate denatured protein. This protein folding activity of domain V is sensitive to the 23S rRNA binding antibiotic chloramphenicol, which binds to the central loop of domain V. This activity is also sensitive to EDTA, which breaks the tertiary organization (causing hyperchromicity) of this RNA.
Lactate dehydrogenase at a concentration of 3.2 µM with respect to monomer was denatured with 1 M guanidium chloride at 20°C for 1 h (12). For refolding the enzyme was diluted 100-fold in 20 mM Tris-HCl, pH 7.5, 200 mM NaCl and 4 mM magnesium acetate and incubated at 20°C for 30 min with or without RNA unless otherwise mentioned. The enzyme concentration was 32 nM with respect to monomer during refolding. Malate dehydrogenase at a concentration of 1.15 µM with respect to monomer was denatured with 6 M guanidium chloride at 20°C for 40 min. For refolding the denatured enzyme was diluted 80-fold in 25 mM sodium phosphate, pH 7.6, 200 mM NaCl, 5 mM 2-mercaptoethanol and 4 mM magnesium acetate and incubated at 20°C for 15 min in the presence or absence of RNA unless otherwise mentioned. The enzyme concentration during refolding was 14 nM with respect to monomer. For refolding in the presence of chloramphenicol both the denatured enzymes were diluted in 50 mM Tris-HCl, pH 7.5, 200 mM NH4Cl and 10 mM magnesium acetate. The antibiotic was incubated in this buffer for 15 min at 4°C prior to addition of denatured enzyme. For refolding in the presence of lincomycin both the denatured enzymes were diluted in 60 mM Tris-HCl, pH 7.5, 400 mM KCl and 4 mM magnesium acetate. The antibiotic was incubated in this buffer for 20 min at 0°C before adding denatured enzymes. Each antibiotic needs an exclusive high salt buffer to bind 23S rRNA (36,37). The activities of the refolded enzymes were measured after dilution in the assay buffer. For lactate dehydrogenase the rate of decrease in A340 was measured at 37°C in buffer containing 100 mM Tris-HCl, pH 7.5, 5 mM sodium pyruvate and 250 µM NADH (2). For malate dehydrogenase the rate of decrease in A340 was measured at 25°C in buffer containing 150 mM sodium phosphate, pH 7.6, 2 mM 2-mercaptoethanol, 500 µM oxaloacetate and 200 µM NADH.
A part of the 23S rDNA gene covering most of domain V was PCR amplified from plasmid pKK 3535, which contains the entire 23S rDNA, using oligonucleotides (A) 5'-GAGAAGCTTGTACCCGCGGCAAGACG GAA-3' and (B) 5'-GAGGGATCCTCTCCAGCGCCCACGGCAGA-3' for the 5'- and 3'-ends of the amplified DNA. The bold sequences of the oligonucleotides constitute the two ends of the domain V segment amplified by PCR, while the rest of the nucleotides constitute restriction sites for the enzymes BamHI and HindIII. The PCR product was digested with restriction enzymes HindIII and BamHI at the 5'- and 3'-end respectively and ligated to vector pGEM4Z (38) digested with the same enzymes between the T7 and SP6 promoters in the polyclonal site. The cloned fragment contained 595 nt of 23S rDNA. This plasmid, designated pDP1, was then linearized with BamHI and transcribed with T7 RNA polymerase to give a run-off RNA containing 595 nt from 2040 to 2634 of 23S rRNA. Transcription with SP6 RNA polymerase of the plasmid linearized with HindIII gave the RNA strand complementary to 23S rRNA. The RNA strand transcribed with SP6 polymerase annealed with 23S rRNA, whereas the RNA strand transcribed with T7 polymerase failed to do so. Therefore, T7 polymerase synthesized the same strand as domain V of 23S rRNA.
The T7 and SP6 RNA polymerases, ribonucleotides and RNase inhibitors required for in vitro transcription were obtained from Ambion (Maxiscript kit no. 1322). Taq polymerase was obtained from GENEI (Bangalore, India).
The putative secondary structure (39) of the 595 nt transcript from plasmid pDP1 containing domain V of E.coli 23S rRNA is shown in Figure 1A. Domain V of 23S rRNA of Bacillus subtilis cloned in plasmid pDK105 (34) was transcribed with SP6 RNA polymerase after linearization of the plasmid with EcoRI. The secondary structure of this 660 nt RNA is shown in Figure 1B. The 337 nt RNA fragment containing the central loop of domain V of B.subtilis 23S rRNA was transcribed with SP6 RNA polymerase from plasmid pDK106 (34) linearized with EcoRI. The structure of this RNA is shown in Figure 1C. The 113 nt RNA containing the upper half of the central loop of domain V of B.subtilis 23S rRNA was transcribed by T7 RNA polymerase from plasmid pDK107 (35) linearized with EcoRI. The structure of this RNA is shown in Figure 1D. Plasmids pDK105, pDK106 and pDK107 were kind gifts from Prof. Bernard Weisblum (University of Wisconsin, Madison). The secondary structure of the four RNA species (Fig. 1A-D) are drawn according to Gutell and Fox (39). A few bases shown in the loops in Figure 1C and D were introduced by Kovalic et al. (34) to join the stems. These are not present in authentic domain V of 23S rRNA.
Figure 3 shows reactivation of denatured malate dehydrogenase and lactate dehydrogenase with E.coli 23S rRNA and domain V of 23S rRNAs from B.subtilis (660 nt) and E.coli (595 nt). The shorter fragments from domain V of B.subtilis 23S rRNA of lengths 337 and 113 nt failed to reactivate the denatured proteins. Therefore, the elaborate stem-loop regions protruding from the upper part of the central loop of domain V (Fig. 1A and B) appear to be indispensable for the protein folding reaction. The central loop alone is not enough for this activity. The extents of reactivation of the proteins were similar in the presence of complete 23S rRNA and the 660 nt RNA containing the entire domain V. The 595 nt transcript, which was missing ~40 nt from the 3'-OH end of domain V of E.coli 23S rRNA, reactivated the proteins to a little lesser extent. Like 23S rRNA the requirements of domain V RNA molecules for optimum reactivation of the enzymes were stoichiometric. The ratios of RNA to protein molecules required for optimum reactivation were identical for intact 23S rRNA and domain V RNA. Also, as observed earlier in the case of 23S rRNA (2,3), at higher concentration of domain V RNA the extent of reactivation decreased. The protein folding activity of 23S rRNA is therefore present in domain V only. Since the 595 nt RNA obtained from plasmid pDP1 is so far the smallest length of domain V having protein folding activity, we used this RNA to characterize this reaction.
The time course of reactivation of malate dehydrogenase and lactate dehydrogenase with 23S rRNA and its 595 nt fragment are shown in Figure 4A and B. The concentrations of enzymes were as described in Materials and Methods. The concentrations of both the RNAs were 14 and 28 nM respectively for malate and lactate dehydrogenase. The 595 nt RNA reactivated denatured malate dehydrogenase and lactate dehydrogenase at a rate considerably faster than 23S rRNA. With this small RNA fragment maximum reactivations were obtained at the same time as spontaneous folding for both proteins. This is in contrast to our earlier observations on 70S ribosomes, the 50S subunit and 23S rRNA-mediated protein folding, all of which took appreciably longer than spontaneous folding of the proteins. Obviously, these modulators controlled the rate limiting steps of protein folding.
We have shown that protein folding by 23S rRNA was inhibited by chloramphenicol and erythromycin (12), antibiotics which bind to this RNA. Since the antibiotics bind mainly to domain V of 23S rRNA, we checked the sensitivity of protein folding by the 595 nt RNA to such antibiotics. The antibiotics used in these experiments were chloramphenicol and lincomycin. Figure 5A shows inhibition of reactivation of denatured malate dehydrogenase and lactate dehydrogenase by chloramphenicol. At 0.2 mM chloramphenicol recovery of enzyme activity decreased to the level of spontaneous folding. Whereas lincomycin had an inhibitory effect on reactivation of denatured lactate dehydrogenase and malate dehydrogenase by full-length 23S rRNA, it did not inhibit reactivation of these enzymes by the 595 nt fragment (Fig. 5B). This could be due to the absence of nt 2032, which is implicated in lincomycin binding (11), in this RNA fragment. On the other hand, most of the nucleotides involved in chloramphenicol binding occur throughout the length of the central loop of domain V, so this antibiotic could easily bind to the RNA fragment.
We have shown that even in the presence of 4 mM MgCl2 very low concentrations of EDTA (<1 mM) could completely inhibit reactivation of proteins by 23S rRNA (3). This small amount of EDTA caused hyperchromicity of 23S rRNA (3), probably by destroying its tertiary level organization and hence protein folding activity. The 595 nt RNA-mediated reactivation of malate dehydrogenase and lactate dehydrogenase was also completely inhibited by ~0.75 mM EDTA, as shown in Figure 6.
We have shown that the protein folding activity of 23S rRNA of the bacterial ribosome (1-4,12) actually resides in domain V. The 660 and 595 nt in vitro transcripts, containing domain V or most of it of B.subtilis and E.coli 23S rRNA respectively, could reactivate denatured lactate dehydrogenase and malate dehydrogenase almost as efficiently as 23S rRNA. This observation completely eliminates the possibility of any contaminating ribosomal protein factor in our rRNA preparations which could be responsible for protein folding. Also, these transcripts take up their tertiary conformations required for protein folding spontaneously, without the assistance of any ribosomal proteins. This is a much less stringent requirement compared with the domain V conformation required for peptidyl transferase activity. For the peptidyl transferase reaction E.coli 23S rRNA was isolated by proteinase K and SDS treatment, but phenol extraction destroyed such activity (10). This reaction also involves tRNA, nucleotides, etc. Therefore, we can assume that the protein folding activity in the largest rRNA, which involves a very simple reaction without requiring any cofactors, etc. (12,4), evolved earlier than peptidyl transferase activity. These arguments are valid if we assume that the original ribosome was made up solely of RNA, as suggested by Crick (20) and asserted more strongly by Woese (21). As mentioned before, even in the 50S subunit itself a large number of nucleotides, almost exclusively in domain V of 23S rRNA, are accessible by chemical agents, amino acyl tRNA, the 30S subunit, etc. (13-19). Therefore, it is not surprising that this domain having the peptidyl transferase centre could act within or without the ribosomal protein coat to form the peptide bond and fold the newly synthesized protein. The presence of an activity in RNA in vitro which is normally seen in the riibonucleoprotein complex in vivo is not surprising: RNase P, snRNP, etc. are wonderful examples (22-30). The tertiary conformation of this RNA domain was required for this activity, as we discuss later. This points to a number of interesting possibilities. The protein folding activity in domain V of 23S rRNA could evolve before the complete ribosome structure was determined with further addition of nucleotides and proteins. In a prebiotic soup only a small RNA could assist some spontaneously synthesized polypeptides to fold to meaningful (active) conformers. The `RNA world' (31,32) envisages a versatile role of RNA in biochemical and biological evolution. In the central dogma of molecular biology the RNA then has overwhelming control over the genetic information. It transmits information from DNA to the ribosome active centre and eventually translates the information into a three-dimensional `active' protein. RNA directs the protein to attain the active three-dimensional state (2-4,10,12), a role which is perhaps the most deterministic in the central dogma of molecular biology. The requisite conformation of 23S rRNA for protein folding is destroyed by EDTA and presumably blocked by the antibiotics chloramphenicol, erythromycin, lincomycin, etc. (3,12), which bind to the 50S subunit of the E.coli ribosome.
This work was supported by grants from the Department of Biotechnology, Department of Atomic Energy and Council of Scientific and Industrial Research, Government of India. The authors gratefully acknowledge the gift of plasmids from Prof. B.Weisblum, University of Wisconsin, Madison. S.Chattopadhyay, S.Chandra and D.Sarkar are Senior Research Fellows of the University Grant Commission. We gratefully acknowledge help from the Distributive Information Centre, University of Calcutta, in the preparation of this manuscript
37Vasquez,D. (1979) Inhibitors of Protein Synthesis. Springer-Verlag, Berlin, Germany, pp. 169-175.
38Sambrook,J., Fritsch,E.F. and Maniatis,T. (1989) Molecular Cloning: A Laboratory Manual, 2nd Edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp. 1-17.
39Gutell,R.R. and Fox,G.E. (1988) Nucleic Acids Res., 16, r175-r269.MEDLINE Abstract
*To whom correspondence should be addressed. Tel: +91 33 351 0359; Fax: +91 33 554 5761; Email: ckdg@cubmb.ernet.in
D. Samanta, D. Mukhopadhyay, S. Chowdhury, J. Ghosh, S. Pal, A. Basu, A. Bhattacharya, A. Das, D. Das, and C. DasGupta Protein Folding by Domain V of Escherichia coli 23S rRNA: Specificity of RNA-Protein Interactions
J. Bacteriol.,
May 1, 2008;
190(9):
3344 - 3352.
[Abstract][Full Text][PDF]
S. C. Sanyal, S. Pal, S. Chaudhuri, and C. DasGupta 23S rRNA assisted folding of cytoplasmic malate dehydrogenase is distinctly different from its self-folding
Nucleic Acids Res.,
June 1, 2002;
30(11):
2390 - 2397.
[Abstract][Full Text][PDF]
S. Chowdhury, S. Pal, J. Ghosh, and C. DasGupta Mutations in domain V of the 23S ribosomal RNA of Bacillus subtilis that inactivate its protein folding property in vitro
Nucleic Acids Res.,
March 1, 2002;
30(5):
1278 - 1285.
[Abstract][Full Text][PDF]
S. Pal, S. Chandra, S. Chowdhury, D. Sarkar, A. N. Ghosh, and C. D. Gupta Complementary Role of Two Fragments of Domain V of 23 S Ribosomal RNA in Protein Folding
J. Biol. Chem.,
November 12, 1999;
274(46):
32771 - 32777.
[Abstract][Full Text][PDF]
CiteULike 
Connotea 
Del.icio.us What's this?
