Recently a cultivation method has been described which allows the preparation .... were required in large amounts to isolate ribosomal elongation complexes for ...
1996 Oxford University Press
Nucleic Acids Research, 1996, Vol. 24, No. 5
907–913
In vivo deuteration of transfer RNAs: overexpression and large-scale purification of deuterated specific tRNAs Ralf Jünemann1, Jörg Wadzack1, Francisco J. Triana-Alonso1,2, Jörg-Uwe Bittner1, Joël Caillet3, Thierry Meinnel4, Kalju Vanatalu5 and Knud H. Nierhaus1,* 1Max-Planck-Institut
für Molekulare Genetik, Ihnestraße 73, 14195 Berlin-Dahlem, Germany, 2Centro de Investigaciones Biomedicas, Universidad de Carabobo, Maracay, Venezuela, 3Laboratoire de Biochimie, Ecole Polytechnique, 91128 Palaiseau cedex, France, 4Institut de Biologie Physico-Chimique, 13 rue Pierre et Marie Curie, 75005 Paris, France and 5Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 0026 Tallin, Estonia Received November 1, 1995; Revised and Accepted January 12, 1996
ABSTRACT Structural investigations of tRNA complexes using NMR or neutron scattering often require deuterated specific tRNAs. Those tRNAs are needed in large quantities and in highly purified and biologically active form. Fully deuterated tRNAs can be prepared from cells grown in deuterated minimal medium, but tRNA content under this conditions is low, due to regulation of tRNA biosynthesis in response to the slow growth of cells. Here we describe the large-scale preparation of two deuterated tRNA species, namely DtRNAPhe and D (the method is also applicable for other tRNA Met f tRNAs). Using overexpression constructs, the yield of specific deuterated tRNAs is improved by a factor of two to ten, depending on the tRNA and growth condition tested. The tRNAs are purified using a combination of classical chromatography on an anion exchange DEAE column with reversed phase preparative HPLC. Purification yields nearly homogenous deuterated tRNAs with a chargeability of ∼1400–1500 pmol amino acid/A260 unit. The deuterated tRNAs are of excellent biological activity. INTRODUCTION For structural investigations of biological molecules which function as part of multi-component systems, electron microscopy, X-ray diffraction, neutron scattering and NMR techniques are the most important direct physical methods that allow studies of structure–function relationships (1). Information from these methods is an indispensable prerequisite for the incorporation of the details of sequence-based data into consistent models. The great advantage of neutron scattering is that it can be used for very large molecules or multi-subunit complexes which can be analysed in solution, thus retaining the functional conformation.
* To
whom correspondence should be addressed
One central molecule of the translational apparatus is tRNA, with its various complexes (2). The three-dimensional structure of tRNAs was solved for tRNAPhe more than 20 years ago (3,4) and appears to be in general the same for all tRNA species. Nevertheless, the tRNA-containing complexes formed during protein synthesis are still the subject of intense structural investigation. In the last decade the structures of some tRNA complexes with their specific aminoacyl synthetase (aaRS) have been solved at atomic resolution by X-ray diffraction (reviewed in 5). In addition, the interaction between the tRNA and aaRS can be understood in detail by dynamic studies using various NMR techniques (6). Recently the ternary complex elongation factor Tu·GTP·tRNA was crystallized successfully and the structure has been solved at atomic resolution (7). The situation for structural investigation of ribosome complexes is more difficult, because it is a multi-component ribonucleoprotein particle (57 components in Escherichia coli 70S ribosomes) with a mass of ∼2300 kDa. To investigate such a large particle most of the structural methods cannot be easily applied. Crystallization and X-ray diffraction would probably lead to a detailed structural model, but crystallization of functional complexes is quite laborious and the phase problem is difficult to solve. Therefore, structures derived from X-ray diffraction of crystals will not be available for many years. At present only neutron scattering techniques (8) are capable of yielding a medium resolution overall structure of the tRNA–ribosome complex (9) by a direct physical method. Unfortunately, NMR and neutron scattering techniques need partially or even fully deuterated compounds in large quantities. However, the production of deuterated molecules is expensive and often high biological activity cannot be achieved easily, since cells grow only slowly in deuterated medium, resulting in low yield and severely reduced activity. Due to the growth rate regulation of tRNAs (10,11) the yield of tRNA is dramatically reduced when prepared from cells grown in deuterated medium.
908
Nucleic Acids Research, 1996, Vol. 24, No. 5
Recently a cultivation method has been described which allows the preparation of E.coli cells in kilogram quantities with high biological activity and almost 100% deuteration (12). Here we combine this method with the use of overexpression systems for specific tRNAs to increase the yield of fully deuterated tRNAs by a factor of up to 10. In addition, we describe a large-scale method to purify these tRNAs to near homogeneity preserving full biological activity. MATERIALS AND METHODS Chemicals and bacterial strains Radioactively labelled amino acids were purchased from Amersham-Buchler (Braunschweig, Germany) and restriction enzymes from New England Biolabs (Beverly, MA). All other chemicals were pro analysi grade and purchased from Merck (Darmstadt, Germany). Escherichia coli MRE600rif (12) is a strain which: (i) is adaptable to growth on deuterated media (13); (ii) contains low levels of ribonuclease I activity (14); and (iii) tolerates high doses of rifampicin. This strain was used for all cultivations in deuterated media. As a reference strain HB101 (15) containing the same plasmids as the MRE600rif derivatives was grown in protonated LB medium (see below). Plasmids The plasmid pPhe was previously described as pPP15 (16). It is a pBR322 derivative containing the pheV gene, which codes for tRNAPhe, under the control of the natural P2 promoter, which is the second of a tandem promoter pair. Plasmid pMet (17) was a kind gift of U. RajBhandary. It carries the gene for tRNA Met behind the f gene cloned into the plasmid natural promoter. The E.coli tRNA Met f plppMet (previously described as pBStRNAMetfY; 18) is under the control of a synthetic lipoprotein promoter lpp (19) and has several modifications at the level of the 5′ maturation sequence allowing maturation by RNase P in vivo. The tRNA transcription region is terminated by the strong terminator of the rrnC operon. This construct allows very high levels of overexpression. Plasmids pBR322 (Boehringer, Mannheim, Germany) and pBluescript (Stratagene, La Jolla, CA) were used for cloning purposes and as control plasmids (minus tRNA gene). Cloning of tRNA genes Plasmids pPhe, pMet and plppMet were used directly for transformation of HB101 or MRE600rif respectively by electroporation using a BioRad gene pulser. Plasmid plppMetPhe was constructed by subsequent cloning of the PstI–HpaI fragment of pPhe (the fragment contains the tRNAPhe gene) and the XhoI–HindIII fragment of plppMet (containing the cassette lpp promoter–tRNA Met gene–rrnC terminator) into the respective f restriction sites within pBluescript. The successful cloning of both tRNA genes were confirmed by sequencing using an automated laser fluorescent (ALF) DNA sequencer (Pharmacia, Uppsala, Sweden). Growth of cells overexpressing specific tRNAs Cells were grown in 50 ml batch cultures in Luria–Bertani (LB) medium (1% bactotryptone, 0.5% yeast extract, 0.5% NaCl in
H2O) or M3 medium [0.15% NaCl, 0.2% (NH4)2SO4, 0.01% MgCl2, 0.65% KH2PO4, 1% K2HPO4, pH 7.2] with 0.5% each of protonated acetate and succinate as carbon sources. M3 medium was either made with H2O or D2O. Each medium was supplemented with 200 µg/ml ampicillin (Boehringer) as a selection marker for pMet, plppMet and plppMetPhe or 12 µg/ml tetracyclin (Boehringer) for pPhe respectively. Large-scale fermentation in fully deuterated medium was performed using fed batch cultivation with a computer controlled pregiven growth rate as described previously (12), except that 200 mg/l ampicillin or 12 mg/l tetracyclin was added at the beginning of cultivation to ensure maintenance of the plasmids. A mixture of 0.67 M succinic and 1.95 M acetic acids served as carbon source. Due to very poor growth the succinic acid was increased to 0.77 M while cultivating the cells carrying plppMetPhe. The critical growth rates (µcrit) corresponding to the condition of substrate limitation of recombinant strains were significantly reduced (µcrit = 0.05, 0.035 and 0.025/h for the cells carrying pMet, pPhe and plppMetPhe respectively) compared with the host cells of E.coli MRE600rif without a plasmid (µcrit = 0.08/h). The biomass yield coefficient (mol C biomass formed/mol C substrate utilized) was 0.25 for cells carrying pPhe and 0.30 for cells carrying pMet and plppMetPhe and cells without a plasmid. Determination of the cellular tRNA content For analytical assays, a method for rapid preparation of tRNAbulk was adapted from the protocol of Xue et al. (20). About 0.2 g wet wt E.coli cells were lysed in 1.5 ml 50% phenol in 10 mM Tris–HCl, pH 6.0 (0C), 10 mM MgCl2 by shaking for 45 min at 4C. The aqueous phase was extracted with phenol twice and once with chloroform/isomayl alcohol (24:1). rRNAs were precipitated on ice by adjusting the NaCl concentration to 2 M. The tRNA-containing supernatant was precipitated with ethanol and the pellet was resuspended in a suitable amount of water (∼200 µl). Aminoacylation capacity was checked in an assay system with 50 mM HEPES–KOH, pH 7.8 (0C), 10 mM MgCl2, 100 mM KCl, 4 mM β-mercaptoethanol, supplemented with an optimized amount of tRNA-free S-100 post-ribosomal supernatant (21) as a source of synthetases. An aliquot of 0.2 A260 units tRNAbulk preparation, 500 pmol [14C]Met or [14C]Phe respectively and 500 pmol [3H]Leu for normalization were incubated for 15 min at 37C and afterwards precipitated with ice-cold TCA, filtered through a glassfibre filter (no. 6; Schleicher & Schüll, Dassel, Germany) and counted according to Rheinberger et al. (21). Large-scale purification of DtRNAs Escherichia coli cells grown in fully deuterated medium (12) were taken to prepare tight coupled ribosomes according to Bommer et al. (22). DtRNAbulk was isolated from the S-100 fraction by stepwise elution from a DEAE–cellulose 52 (Whatman, Springfield Mill) column (23). DtRNAMet was separated from DtRNAPhe by chromatography of the tRNA on a large-scale DEAE–Sephadex A-50 column (40 × 1200 mm) at 4C using a linear gradient (7 l) formed from buffer A [20 mM HEPES–KOH, pH 7.5 (0C), 10 mM magnesium chloride, 4 mM β-mercaptoethanol, 250 mM sodium chloride] and buffer B [20 mM HEPES–KOH, pH 7.5 (0C), 10 mM magnesium chloride, 4 mM β-mercaptoethanol and 500 mM sodium chloride] (24). The fractions were checked by an analytical aminoacylation assay for
909 Nucleic Acids Acids Research, Research,1994, 1996,Vol. Vol.22, 24,No. No.15 Nucleic their content of either DtRNAPhe or DtRNAMet. Fractions containing one of the two tRNAs were combined and precipitated with ethanol. The two tRNAs were further purified by HPLC on a preparative Vydac Silica Gel 214 TP-510 column (250 × 10 mm; Vydac). Chromatography was performed as described (20) except that the buffer system was varied slightly [buffer A, 20 mM ammonium acetate, pH 5.0, 8 mM magnesium chloride, 1 M sodium formiate; buffer B, 20 mM ammonium acetate, pH 5.0, 10% (w/v) methanol]. The DtRNAPhe fraction was then aminoacylated and N-acetylated (21). For aminoacylation and formylation of the DtRNAMet fraction, the tRNA was incubated in 20 mM HEPES–KOH, pH 7.5 (0C), 7 mM MgCl2, 150 mM KCl, 3 mM ATP, 4 mM β-mercaptoethanol for 30 min at 37C together with a 3-fold excess of [14C]Met, an optimized amount of tRNA-free S-100 and a 700 molar excess of formyl donor (folinic acid; Serva, Heidelberg, Germany). The reaction was stopped by addition of a 1/10 volume of 3 M sodium acetate, pH 5.0. After two phenol extractions and precipitation with ethanol the remaining formyl donor was separated from the f-[14C]Met- DtRNA Met by gel f filtration using a NAP-25 disposable column (Pharmacia, Uppsala, Sweden). A second HPLC chromatography on the Vydac Silica Gel column using the same conditions as described above but on the aminoacyltRNAs (Ac-[14C]Phe-DtRNAPhe and f-[14C]Met- DtRNA Met f ) yielded essentially pure tRNAs (>1300 pmol/A260 unit). For isolation of highly purified deacylated tRNAs the N-aminoacyl-tRNAs can be deacylated using purified RNase-free peptidyltRNA hydrolase (PTH, see next section). The enzyme works quite efficiently on Ac-Phe-tRNA in 20 mM HEPES–KOH, pH 7.6 (0C), 10 mM MgCl2, 4 mM β-mercaptoethanol within 10 min at 30C, whereas the deacylation of f-Met-tRNA needs a basic milieu (pH 8.0) and a longer incubation time (20 min, 30C) to achieve quantitative deacylation (25). The optimal ratio of enzyme to tRNA depends on its activity and has to be optimized for each preparation (it is in the region of 150 ng enzyme/pmol tRNA). Purification of peptidyl-tRNA hydrolase Peptidyl-tRNA hydrolase (PTH, EC 3.1.1.29) was purified from E.coli. A sample of 100 ml S-100 fraction from E.coli was fractionated by stepwise precipitation with 32 and 72.5% (w/v) (NH4)2SO4. The second pellet (72.5%) was dissolved in 10 mM KH2PO4, pH 6.4, and dialysed three times each for 5 h against 100 volumes of the same buffer. The protein solution (∼150 ml) was then applied to a CM–cellulose column (60 ml) equilibrated with the same buffer at a flow rate of 0.5 ml/min. The unbound and weakly bound material was washed out with 250 ml 10 mM KH2PO4, pH 6.4, 50 mM NaCl. The enzyme activity was eluted with 500 mM NaCl, concentrated and desalted by ultrafiltration using Centricon 10 microconcentrators (Amicon) and stored in 10 mM KH2PO4, pH 6.4, with 5% (w/v) glycerol at –80C. Gel electrophoresis of tRNA Approximately 0.2 A260 units of RNA were analysed on a 15% denaturing polyacrylamide gel (acrylamide:bis-acrylamide 38:2% w/v, 7 M urea). The gel was stained with 0.1% (w/v) toluidine blue.
909
Ribosome binding assay Preparation of E.coli 70S ribosomes and site-specific binding of tRNAs to 70S particles followed the protocols described (22). The ribosomes were programmed with a 5-fold excess of a 46 nt mRNA analogue containing unique codons for Met and Phe in the middle (MF-mRNA; 26) and protonated or deuterated Ac-[14C]Phe-tRNA (up to a 5-fold molar excess over ribosomes) was added either directly for P site binding or after pre-incubation with a 2-fold molar excess of DtRNA Met for A site binding of f Ac-[14C]Phe-tRNA. The extent of binding was determined by nitrocellulose filtration and location to the A or P site was distinguished by puromycin reactivity (22). RESULTS AND DISCUSSION For the structural investigation of functional ribosomal complexes with neutron scattering methods it was necessary to develop an efficient method to obtain specific highly purified 100% deuterated tRNAs from cell extracts. To optimize the tRNA yield from deuterated cells MRE600rif was transformed with tRNA overexpression constructs and grown in batch or fed batch cultivation (12). The tRNAs from these cells were purified in a multistep protocol and the quality of the resulting specific tRNAs was determined. Here we describe the large-scale preparation for two deuterated tRNAs, tRNAPhe and tRNA Met f . Those two deuterated tRNAs were required in large amounts to isolate ribosomal elongation complexes for neutron scattering analysis. However, the protocol can be easily adapted for preparation of other tRNA species. Evaluation of different constructs for overexpression of tRNAs Three prerequisites have to be taken into account in the design of overexpression constructs for large-scale preparation of fully deuterated tRNAs. (i) Since growing cells in fully deuterated media is very expensive, it is important that ribosomes and tRNAs can be prepared from a single batch. Because in vitro ribosome activity drops when protein synthesis is hampered in vivo, the level of tRNA overexpression should not interfere with cellular protein synthesis. (ii) Proton incorporation into tRNAs by metabolism-inducing agents like IPTG should be avoided, as well as changes in growth conditions due to a temperature shift. Therefore, overexpression should be constitutive. Consequently, the use of strong induction systems (27) or runaway replication plasmids (28), in which overproduction of a single tRNA kills the cell, are not suitable for overproducing deuterated tRNAs, although the yield of specific tRNAs is very high in these systems (up to 70% of total tRNA). (iii) A strain capable of growing in deuterated medium is needed. The most suitable strain for this purpose is MRE600, which can be grown in fully deuterated medium to very high densities (12). In this study we tested four different constructs (Fig. 1) for their effect on tRNA content in deuterated cells. One construct contains the gene for tRNAPhe under the control of its natural promoter P2 (pPhe), two constructs harbour genes for tRNA Met f , either with the natural promoter (pMet) or an artificial constitutive promoter of the lipoprotein Lpp (plppMet), and one construct contains both tRNA genes (plppMetPhe). The last plasmid was constructed by cloning the tRNAPhe gene (the PstI–HpaI fragment of pPhe) into the respective restriction sites of the high
910
Nucleic Acids Research, 1996, Vol. 24, No. 5
Figure 1. Characterization of the tRNA-overexpressing plasmids used. (A) The tRNA gene-containing regions. (B) The plasmids: 1, pPhe; 2, pMet; 3, plppMet; 4, plppMetPhe. The tRNAMet gene codes for initiator tRNAfMet in all cases.
copy number plasmid pBluescript. Subsequently the expression cassette for tRNA Met from plasmid plppMet (a XhoI–HindIII f and fragment comprising the lpp promoter, the gene for tRNA Met f the rrnC terminator) was inserted, yielding plasmid plppMetPhe. The presence of both tRNA genes in the plasmid construct was shown by sequencing the region of the plasmid using an automated DNA sequencer (data not shown). The overexpression level of the different plasmid constructs was analysed for different strains (MRE600rif versus HB101) in minimal and rich medium (M3 versus LB) with or without deuteration (M3/D2O versus M3/H2O). To compare data from different tRNA preparations the cellular levels of chargeable tRNAPhe or tRNAMet were normalized with respect to chargeable tRNALeu. The level of tRNALeu is assumed to be largely independent of the overexpression constructs. Representative results for batch cultivations are shown in Figure 2. In general the overexpression level of tRNAs tested in protonated rich medium was higher in HB101 (white columns) compared with MRE600rif (hatched columns), but it was still significant in MRE600rif, the strain used for growth in deuterated minimal medium. The level of tRNAMet overproduction (Fig. 2A) from its natural promoter (pMet) was not very high (3- to 4-fold) under the different conditions tested. For this reason DtRNA Met expression f was tested with the gene under the control of the strong lpp promoter (plppMet and plppMetPhe respectively). These constructs worked quite efficiently in HB101 (∼15-fold overexpression), but were less active in MRE600rif, where overexpression was only ∼5-fold.
Figure 2. Expression of tRNAMet (A) or tRNAPhe (B) in different strains and under different growth conditions. The respective chargeable tRNA content is normalized to the chargeable cellular content of tRNALeu.
Changing to minimal medium (solid black columns) slightly increased the overexpression level for tRNA Met f , whereas tRNAPhe overexpression from pPhe or plppMetPhe was increased by a factor of two to three upon a shift to minimal medium (Fig. 2B). In contrast, the switch from H2O to D2O in minimal medium had almost no effect on overexpression (data not shown). Under the conditions of fed batch cultivation, which allows growth of cells to very high densities, overexpression was clearly reduced (Fig. 3; note the different scales on the y-axis in Figs 2 and 3). Nevertheless, tRNAPhe content of the cells was 3-fold higher in the presence of pPhe or 7-fold higher when expressed from plppMetPhe. Thus the overexpression plasmids had clear effects, since the higher ratio of tRNAPhe to other tRNAs increased the yield and facilitated purification of deuterated tRNAPhe. In the little overexpression was observed (up to case of DtRNA Met f 2-fold). Purification of deuterated tRNAs The large-scale tRNA purification method described here combines well-known classical chromatographic techniques with the powerful tool of HPLC. First, tRNAbulk has to be isolated from the S-100 fraction, which contains more or less all cellular proteins and small nucleic acid compounds. Both tRNAs and proteins bind tightly to a DEAE–cellulose 52 matrix at low salt concentration (∼150 mM
911 Nucleic Acids Acids Research, Research,1994, 1996,Vol. Vol.22, 24,No. No.15 Nucleic
911
Figure 4. Absorption profile at 260 nm of DtRNAbulk applied to a DEAE–Sephadex A50 column and distribution of tRNAMet and tRNAPhe in the linear NaCl gradient.
of tRNAPhe and
tRNAMet in MRE600rif grown in fully
Figure 3. Expression deuterated minimal medium using the fed batch cultivation technique. The tRNA content is normalized to the cellular content of tRNALeu.
NaCl). In a batch-like operation tRNAbulk separates from the cellular protein fraction. While the proteins elute at salt concentrations 90% of the input material can be obtained as ). f-Met- DtRNA Met f The charged and N-blocked tRNAs (Ac-Phe-DtRNAPhe and ) elute at a higher buffer B concentration in a f-Met- DtRNA Met f second reversed phase HPLC step, away from all contaminants (Fig. 5C and D). The quality of the purified tRNAs was satisfactory; 1400–1600 pmol/A260 unit for Ac-Phe-DtRNAPhe and 1300–1400 pmol/A260 unit for f-Met- DtRNA Met (Table 1), f corresponding to a 40- to 50-fold enrichment relative to DtRNAbulk. Quality of the purified deuterated tRNAs The deuterated tRNAs obtained using the overexpression strains and the purification protocol described here are of excellent quality, as checked by three different sets of experiments. First, homogeneity of the isolate was controlled either by PAGE or by analytical HPLC analysis, where a single peak was observed (not shown). For biological activity interaction with synthetase was tested in a charging assay. Finally, interaction with the ribosome was analysed in a site-specific binding assay.
912
Nucleic Acids Research, 1996, Vol. 24, No. 5
Figure 5. Absorption profile at 260 nm of an HPLC preparation of deacylated DtRNAPhe (A) and deacylated DtRNAMet (B) and of the tRNAs after charging and acetylation or formylation resulting in Ac[14C]Phe-DtRNAPhe (C) or formyl-Met-DtRNAfMet (D) respectively. The acceptor activity of each fraction was determined before the fractions were pooled, as indicated by the shadowed area. The gradient is indicated by a dashed line.
Table 1. Characterization of fractions in a DtRNA purification protocol with overexpressed DtRNAPhe Purification step
Fraction
DtRNAMet
DtRNAPhe
fraction
Yield
Chargeable
with Met
factor
(A260 units)
(pmol/A260 unit)
Enrichment
Yield
(relative to DtRNAbulk)
(%)
fraction
Yield
Chargeable
with Phe
factor
(A260 units)
(pmol/A260 unit)
Control
HtRNAbulk
-
83
-
-
-
39
DEAE–cellulose
DtRNAbulk
(4500)
24
1
100
(4500)
67
52 batch column DEAE–Sephadex A50 column
(35) Enriched
640
120 (207)
145
312
DtRNAbulk
Enrichment
Yield
(relative to DtRNAbulk)
(%)
1
100
(27) 5
71
1580
13
42
52
150 (94)
2.2
79
1092
16
25
23
13
fractions First reversed
Deacylated DtRNA
phase HPLC Charging and second HPLC PTH treatment/ deacylation
(427) N-blocked aminoacyl-DtRNA Deacylated DtRNA
17
1320
(564) 55
21
(1360) 1010
13
1540 (1430)
42
14
Not done
(965)
The table shows a typical example using a DtRNAbulk fraction obtained from cells with the plasmid pPhe. The starting material of 4500 A260 units DtRNAbulk was taken as 100%. At this stage DtRNAMet and the overexpressed DtRNAPhe species were not separated. (For comparison the respective charging values for a DtRNA preparation from cells without the overexpression plasmid are given in parentheses.)
(i) Deacylated deuterated tRNAs are fully active in interaction with their corresponding aminoacyl synthetases: DtRNAPhe could be charged to ∼1350 pmol radioactive [14C]Phe/A260 unit when the middle fraction of the corresponding preparative HPLC peak (see
Fig. 5A) was taken. In contrast, purification of highly purified D required charging and deacylation prodeacylated DtRNA Met f cedures, since deacylated DtRNA Met could not be separated in a f
913 Nucleic Acids Acids Research, Research,1994, 1996,Vol. Vol.22, 24,No. No.15 Nucleic
913
REFERENCES
Figure 6. Binding of protonated or deuterated Ac[14C]Phe-tRNAPhe to the ribosomal P and A sites respectively.
single HPLC step. DtRNA Met could be recharged to ∼1000 f pmol/A260 unit after deacylation using PTH treatment (see Table 1). In both cases charged tRNA could be separated from uncharged material, yielding an even better quality. (ii) A prerequisite for the structural investigation of ribosomal complexes is a purity of the DtRNAs of >1000 pmol/A260 unit chargeable with its cognate amino acid. A purity of 1000 pmol/A260 unit is sufficient for preparation of specific ribosomal complexes, since the in vitro accuracy of the decoding process ensures specificity of tRNA binding. The purity obtained for deuterated tRNAs were higher than this minimal value. Deuterated tRNAs showed normal behaviour in binding to ribosomes. In 70S ribosomes programmed with heteropolymeric MF-mRNA deacylated DtRNA Met blocked the ribosomal P site quantitatively. More f than 80% of Ac-Phe-tRNA bound in the second binding step was directed to the ribosomal A site. Deuterated Ac-[14C]Phe-tRNA was as active as its protonated counterpart (Fig. 6) in binding to the ribosomal P site (binding to ribosomes programmed with MFmRNA saturates at 0.6 molecules/ribosome, which means that 60% of the 70S ribosomes carried an Ac-[14C]Phe-tRNA in the P site). For A site binding the activity of deuterated Ac-[14C]PhetRNA was ∼90% compared with that of protonated material (binding of 0.45 molecules/ribosome in comparison with 0.5 for the protonated counterpart). ACKNOWLEDGEMENTS We thank Dr M. Springer for help and discussion. We are grateful to Dr U. L. RajBhandary for providing us with the tRNA Met plasmid f and for advice and suggestions on the manuscript. Work in his laboratory is supported by grant GM17151 from the National Institutes of Health. This work was supported by grant CIPACT93-0158(DG12HSMU) from the European Commission.
1 Moore,P.B. (1993) Curr. Opin. Struct. Biol., 3, 340–344. 2 Söll,D. and RajBhandary,U.L. (eds)(1995) tRNA. Structure, Biosynthesis and Function. ASM press, Washington, DC. 3 Kim,S.H., Suddath,F.L., Quigley,G.J., McPherson,A., Sussman,J.L., Wang,A.H.J., Seeman,N.C. and Rich,A. (1974) Science, 185, 435–440. 4 Robertus,J.D., Ladner,J.E., Rhodes,J.T.F.D., Brown,R.S., Clark,B.F.C. and Klug,A. (1974) Nature, 250, 546–551. 5 Delarue,M. (1995) Curr. Opin. Struct. Biol., 5, 48–55. 6 Niimi,T., Kawai,G., Takayanagi, M., Noguchi,T., Hayashi,N., Kohno,T., Muto,Y., Watanabe,K., Miyazawa,T. and Yokoyama,S. (1993) Biochimie, 75, 1109–1115. 7 Nissen,P., Kjeldgaard,M., Thirup,S., Polekhina,G., Reshetnikova,L., Clark,B.F.C. and Nyborg,J. (1995) Science, 270, 1464–1472 8 Knop,W., Hirai,M., Olah,G., Meerwinck,W., Schink,H.-J., Stuhrmann,H.B., Wagner,R., Wenkow-EsSouni,M., Zhao,J., Schärpf,O., Crichton,R.R., Krumpolc,M., Nierhaus,K.H., Niinikoski,T.O. and Rijllart,A. (1991) Physica B, 174, 275–290. 9 Stuhrmann,H.B., Burkhardt,N., Diedrich,G., Jünemann,R., Meerwinck,W., Schmitt,M., Wadzack,J., Willumeit,R., Zhao,J. and Nierhaus,K.H. (1995) Nucl. Instrum. Methods Phys. Res. A, 356, 124–132. 10 Jinks-Robertson,S. and Nomura,M. (1987) In Neidhardt,F.C., Ingraham,J.L., Low,K.B., Magasanik,B., Schaechter,M. and Umbarger,H.E. (eds), Escherichia coli and Salmonella typhimurium. Cellular and Molecular Biology. ASM press, Washington, DC, Vol. 2, pp. 1358–1385. 11 Emilsson,V., Nälsund,A.K. and Kurland,C.G. (1993) J. Mol. Biol., 230, 483–491. 12 Vanatalu,K., Paalme,T., Vilu,R., Burkhardt,N., Jünemann,R., May,R., Rühl,M., Wadzack,J. and Nierhaus,K.H. (1993) Eur. J. Biochem., 216, 315–321. 13 Moore,P.B. (1979) Methods Enzymol., 59, 639–655. 14 Cammack,K.A. and Wade,H.E. (1965) Biochem. J., 96, 671–680. 15 Bolivar,F. and Backman,K. (1979) Methods Enzymol., 68, 245–267. 16 Caillet,J., Plumbridge,A. and Springer,M. (1985) Nucleic Acids Res., 13, 3699–3710. 17 Seong,B.L. and RajBhandary,U.L. (1987) Proc. Natl. Acad. Sci. USA, 84, 334–338. 18 Meinnel,T. and Blanquet,S. (1995) J. Biol. Chem., 270, 15908–15914. 19 Meinnel,T., Mechulam,Y. and Fayat,G. (1988) Nucleic Acids Res., 16, 8095–8096. 20 Xue,H., Shen,W. and Wong,J.T.-F. (1993) J. Chromatogr., 613, 247–255. 21 Rheinberger,H.-J., Geigenmüller,U., Wedde,M. and Nierhaus,K.H. (1988) Methods Enzymol., 164, 658–670. 22 Bommer,U., Burkhardt,N., Jünemann,R., Spahn,C.M.T., Triana-Alonso,F.J. and Nierhaus,K.H. (1996) In Graham,J. and Rickwood,D. (eds), Subcellular Fractionation—A Practical Approach. IRL Press, Washington, DC, in press. 23 Gillam,I., Milward,S., Blew,D., Tigerstrom,M. von, Wimmer,E. and Tener,G.M. (1967) Biochemistry, 6, 3043–3056. 24 Nishimura,S. (1971) In Cantoni,G.L. and Davis,D.R. (eds.) Procedures in Nucleic Acid Research. Harper, New York, NY, Vol. 2, pp. 542–564. 25 Yot,P., Paulin,D. and Chapeville,F. (1971) Methods Enzymol., 20, 194–199. 26 Triana-Alonso,F.J., Dabrowski,M., Wadzack,J. and Nierhaus,K.H. (1995) J. Biol. Chem., 270, 6298–6307. 27 Perona,J.J., Swanson,R., Steitz,T.A. and Söll,D. (1988) J. Mol. Biol., 202, 121–126. 28 Borel,F., Härtlein,M. and Leberman,R. (1993) FEBS Lett., 324, 162–166. 29 Seno,T., Kobayashi,M. and Nishimura,S. (1968) Biochim. Biophys. Acta, 169, 80–94. 30 Roy,K.L., Bloom,A. and Söll,D. (1971) In Cantoni,G.L. and Davis,D.R. (eds), Procedures in Nucleic Acid Research. Harper, New York, NY, Vol. 2, pp. 524–541.
