Search

Riboswitches are natural RNA-based genetic switches that sense small-molecule metabolites and regulate in response the expression of the corresponding metabolic genes. Within the last years, several engineered riboswitches have been developed that act on various stages of gene expression. These switches can be engineered to respond to any ligand of choice and are therefore of great interest for synthetic biology. In this review, we present an overview of engineered riboswitches and discuss their application in conditional gene expression systems. We will provide structural and mechanistic insights and point out problems and recent trends in the development of engineered riboswitches.

Results of a first successful application of a direct photo-induced affinity modification of Tet repressor (TetR(D)) protein with tetracycline within a complex of known three-dimensional structure are described. The conditions of the modification have provided suitable yields of the modified complex and allowed characterization of the modified segments of the protein. The potential of tetracycline as a fine modifying reagent was established. In the complex of TetR(D) protein with tetracycline, the antibiotic modifies at least two segments, Ile59–Glu73 and Ala173–Glu183, which form a binding tunnel for the drug according to the X-ray analysis. These data open possibilities for the use of different tetracycline targets for structural studies in solution.

Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) has been used for C-terminal amino acid sequence determination of peptides and proteins. The usefulness of MALDI-MS was demonstrated by analyzing peptide mixtures (C-terminal peptide ladder) which were generated by enzymatic digestion of substance P, glucagon, angiotensinogen, insulin B chain and myoglobin with the exopeptidases carboxypeptidase Y and P. The results clearly show that up to 11 amino acid residues can be determined in the pmol range by analyzing the molecular masses of the truncated peptides. For proteins it is possible to investigate enzymatic or chemical digests in the same manner.

The complete amino acid sequence of protein S14 from the small subunit of Bacillus stearothermophilus was determined by N-terminal sequence analysis and by sequencing of overlapping peptides obtained from enzymatic digestions. Protein S14 consists of 60 amino acid residues with a molecular mass of 7148 Da. It has a high content of basic amino acids and a predicted isoelectric point of 11.46. Protein S14 contains two pairs of cysteines in the carboxyl-terminal region, presumably linked by two sulphur bridges. A comparison between protein S14 of B. stearothermophilus and homologous proteins from other organisms revealed highly conserved carboxyl-termini for this protein in eubacteria, archaebacteria and eukaryotes.

Biosynthesis of gramicidinS in Bacillus brevis is catalysed by a multienzyme system consisting of two multifunctional proteins, gramicidinS synthetase 1 and 2 codified by the grsA and grsB genes, respectively. GramicidinS synthetase 2 shows a modular architecture of four amino acid-activating domains each containing a thioester binding motif LGG H/D S L/I highly conserved in its C-terminal region, as demonstrated by sequence analysis of the grsB gene [W. Schlumbohm et al. (1991) J. Biol. Chem. 266, 23135-23141]. This multienzyme was specifically labeled at the thioester binding site of l-valine with [3H]N-ethylmaleimide using a substrate protection technique. After enzymatic digestion a labeled active site peptide was isolated in pure form by multistep methodology. This fragment was identified by gas-phase sequencing as the active site peptide of the thiotemplate site for l-Val by comparison with the grsB gene sequence. By mass spectrometry in combination with amino acid analysis it was demonstrated that a 4'-phosphopantetheine carrier was attached to the active serine in this motif. Our results give evidence that multiple peripheral 4'-phosphopantetheine carriers are involved in the formation of gramicidinS in contrast to a central carrier arm as assumed in the original version of the thiotemplate mechanism. A ‘Multiple Carrier Model’ of nonribosomal peptide biosynthesis is proposed.

Results of a first successful application of a direct photo-induced affinity modification of Tet repressor (TetR(D)) protein with tetracycline within a complex of known three-dimensional structure are described. The conditions of the modification have provided suitable yields of the modified complex and allowed characterization of the modified segments of the protein. The potential of tetracycline as a fine modifying reagent was established. In the complex of TetR(D) protein with tetracycline, the antibiotic modifies at least two segments, Ile59-Glu73 and Ala173-Glu183, which form a binding tunnel for the drug according to the X-ray analysis. These data open possibilities for the use of different tetracycline targets for structural studies in solution.

Ribosomes and their subunits have been isolated from Halobacterium marismortui, an extremely halophilic bacterium from the Dead Sea. The stability and functional activity of the subunits were tested under a wide range of salt conditions. Three-dimensional microcrystals of the large ribosomal subunits have been obtained. Electron microscopy of positively stained thin sections of these crystals showed that the particles are closely packed with approximate cell constants of 310 × 350 Å.

The complete amino acid sequence of protein B (= CAMP factor) of Streptococcus agalactiae has been determined. The sequence data were obtained mainly by manual sequencing of peptides derived from digestion with lysyl-peptidase, clostripain ans Staphylococcus aureus protease and by solid phase sequencing of cyanogen bromide fragments. The protein contains 226 amino acids and has an M r of 25 263. The sequence was compared with sequences of other Fc-binding proteins and partial sequence homology was found between protein B and the Fc-binding region of protein A.

For the crystallization of proteins under microgravity conditions, a Chinese re-entry system was used, in which 101 experiments of 25 different biological macromolecules were accommodated. From the results obtained we conclude that under microgravity conditions crystal growth can only be expected under those crystallization conditions which also permit crystal growthon earth. A number of space-grown crystals were larger in size and of a better quality in their ability to diffract X-rays than the corresponding ground control crystals grown at the Chinese launch site. However, the space-grown crystals have not reached the X-ray diffraction quality of the crystals obtained under optimal conditions in the home laboratories.

A new form of three-dimensional crystals of the 50 S ribosomal subunits from Halobacterium marismortui has been obtained at 19°C, using polyethylene glycol in the presence of 1.2–1.7 M KCl in the crystallization mixture. The crystals diffract X-rays to 13 A and are stable in the synchrotron radiation beam for 4–8 h. Being aggregates of thin plates, the dimensions of only two unit cell edges, 147 × 181 A, with an angle of 95°, could so far be determined by both X-ray crystallography and electron microscopy. Attempts to produce thicker crystals by sophisticated seeding are in progress.

The complete sequence of protein L17 which is a component of the large subunit of the E. coli ribosome has been determined. Peptides deriving from enzymatic hydrolysis with trypsin, thermolysin, chymotrypsin and S. aureus and A. mellea protease were isolated and sequenced by the DABITC/PITC double coupling method. Some overlapping peptides were obtained after mild acid cleavage of the protein. According to the amino acid sequence protein L17 contains 127 residues and has a molecular mass of 14 365. The primary structure of protein L17 agrees well with the amino acid analysis of the intact protein and its N-terminal sequence as derived from automatic sequencing in an improved Beckman sequencer. Secondary predictions and a search for homologous sequence stretches to other ribosomal proteins were made.

A new form of well-ordered three-dimensional crystals of intact 50 S ribosomal subunits from Bacillus stearothermophilus have been obtained. Electron micrographs of positively stained sections of these crystals revealed that the ribosomal particles are packed closely. The cell parameters have been determined. Representative electron micrographs and their computed contoured filtered images are shown.

The described isocratic separation for phenylthiohydantoin (Pth) amino acid derivatives offers advantages over gradient systems: the baseline is more stable at high sensitivity, the resolution and reproducibility is excellent and the cost and consumption of solvent is very low. The separation of low picomolar quantities, on 2 different types of reverse phase columns, are presented. The method is used ‘off line’, i.e. as a separate analytical unit, or in an ‘on line’ mode, where it is directly connected to a protein sequencer for automatic identification of the released amino add derivatives.

Sequence comparison of protein B (CAMP-factor) with human apolipoprotein A-IV (apo A-IV) revealed 32% similarity between the N-terminal part of protein B and a part of the putative lipid-binding domain of apo A-IV. The significance of this similarity is discussed with respect to the structure/function relationship of protein B.

Well-ordered three-dimensional crystals of the large subunit of Bacillus stearothermophilus have been obtained. Electron micrographs of positively stained sections of these crystals revealed that the ribosomal particles are packed in several modes. Cell dimensions have been determined for 4 crystal forms. Representative electron micrographs, their optical diffraction patterns and their two-dimensional images are shown.

The chemical synthesis of 4-(N-tertbutyloxycarbonylaminomethyl)-phenylisothiocyanate starting from 4-nitrobenzylamine is described. This derivative represents an Edman-type reagent with a masked amino group which renders the thiohydantoin upon deblocking susceptible to fluorogenic detection. The coupling efficiency is determined in comparison to degradations with PITC, DABITC and FITC. The detection sensitivity on thin layer chromatograms is compared to the thiohydantoins derived from DABITC.

An open reading frame upstream of the Methanococcus vannielii L12 gene has been detected. The beginning of this open reading frame agrees with the N-terminal region of a protein (MvaL10) which has been isolated from the 50 S ribosomal subunit of M. vannielii and sequenced. The length of this gene is 1008 nucleotides, coding for 336 amino acids. Excellent sequence similarities were found to the L10-like ribosomal proteins from Halobacterium halobium and man. The N-terminal part of the MvaL10 protein shows significant sequence similarities to the E. coli L10 protein. MvaL10 is more than twice as long as E. coli L10 but is of length similar to those of the homologous halobacterial and human proteins. Interestingly, the C-terminal region of MvaL10 shows exceptionally high similarity to the C-terminal sequence of the MvaL12 protein. This is not the case for the E. coli proteins but was also observed for the human, Halobacterium and Sulfolobus proteins.

Ribosomal proteins were extracted from 30 S subunits of Halobacterium marismortui under native conditions. Their separation was based on gel filtration and hydrophobic chromatography, performed at a concentration of 3.2 M KCl to avoid denaturation. A total of nine proteins were isolated, purified and identified by partial amino-terminal sequences and two-dimensional gel electrophoresis. There is a high degree of sequence homology with 30 S proteins from H. cutirubrum, and also some with 30 S proteins of eubacteria. Proton NMR data indicate unfolding of the proteins in low salt. One of the proteins, however, retains its secondary structure at a salt concentration as low as 0.1 M NaCl, and even in 8 M urea. One reason for this outstanding stability could be the high proportion (50%) of β-structure in this protein as determined from circular dichroism measurements. In general, there is a higher β-sheet content than for 30 S proteins from Escherichia coli. Measurements of Stokes radii indicate several of the proteins to have a rather elongated shape. One of these is a complex consisting of L3/L4 and L20, similar to the L8-complex from E. coli. The presence of this 50 S complex in the preparation of the small subunit suggests a location on the interface between the subunits.

The complete primary structure of protein L2 which is the largest protein component of the E. coli 50 S subunit, has been established. A combination of enzymatic and chemical cleavages has been employed to isolate peptides, which were sequenced by the micro-DABITC/PITC double-coupling method [FEBS Lett. (1978) 93, 205–214]. The sequence determined shows protein L2 to consist of 272 amino acid residues with Mr = 29730. Secondary structure predictions were made based on the primary structure. Further, sequence regions homologous to other ribosomal proteins are presented. These results suggest protein L2, which binds specifically to the 23 S RNA, to show homologous sequence stretches to the other RNA-binding proteins.

E. coli ribosomes contain a large number of ribosomal proteins whose structural homology has not been established. Information about such homology can be obtained by comparison of the peptide patterns from isolated ribosomal proteins. The present paper describes the tryptic digestion of 43 pure ribosomal proteins isolated from 3@ S and 50 S subunits of E. coli and the separation of the tryptic peptides by column chromatography. An abstract of this work has appeared elsewhere [ l]

The complete sequence determination of protein S1, the largest protein from the Escherichia coli ribosome, revealed that it is composed of repeated internal duplications, mainly at the central region of the molecule which contains the mRNA-binding domain [Eur. J. Biochem. (1982) 123, 37–53]. With the aid of computer programs the statistical significance of the internal repeats in S1 was proven. Auto-comparison of the S1-sequence showed that it is composed of 87-residue strings with 44-residue subunits: 3 strings (residues 189–447) are highly related; 3 strings (residues 13–188 and 448–533) are less but significantly related. Statistical analysis revealed a more distant relatedness for the 44-residue subunits than for the 87-residue strings. Protein S1 was compared to all other E. coli ribosomal proteins and to the 1100 primary structures listed in the last. Atlas of Protein Sequence and Structure (1978) showing parts of S1 distantly related with parts of several ribosomal proteins. However, distinct homologies between protein S1 and the other ribosomal proteins can be ruled out. The strongest repeats within the S1 sequence were mainly found corresponding to the mRNA-binding domain. Distantly related partial sequences were also found with ribosome-associated and nucleotide-binding proteins, with some enzymes, with several peptide hormones and with contractile proteins.

The isocratic separation by reversed-phase HPLC of all DABTH-amino acid derivatives which are released by degradation of polypeptides with 4-N,N-dimethylaminobenzene-4'-isothiocyanate is described. The method allows an easy and quick identification and the quantitative determination of all derivatives in 30 min and includes the separation of isoleucine and leucine. Detection limit is 5 pmol at 436 nm. The solvent mixture is continuously recycled through the column, thereby reducing the solvent consumption considerably.

Quail cystatin, a new cysteine proteinase inhibitor protein of the cystatin superfamily, was purified from egg albumen of Japanese quail Coturnix coturnix japonica. Amino acid sequencing and mass spectrometry revealed the complete 116 amino acid residue primary structure of a phosphorylated form (13,173 Da). The inhibitor has a 90% sequence identity with chicken cystatin. Its interaction with papain is rapid and tight (Ki = 4.4 pM; k(on) = 1.8x10(7) M(-1) s(-1); k(off) = 0.8x10(-4) s(-1)) and very similar to that of chicken cystatin. Surprisingly, however, cathepsin B was inhibited 15-fold more strongly by quail cystatin (Ki = 47 pM; k(on) = 19x10(7) M(-1) s(-1); k(off) = 9x10(-4) s(-1)) than by chicken cystatin (Ki = 784 pM; k(on) = 2.9x10(7) M(-1) s(-1); k(off) = 24x10(-4) s(-1)). Intuitive comparative conformational inspection of related inhibitors and of cognate enzymes suggest that: (i) the 3D structure of quail cystatin is nearly identical to that of chicken cystatin, (ii) quail cystatin can interact with cathepsin B analogous to the stefin B-papain interaction, if the 'occluding loop' of cathepsin B possesses an 'open' conformation, (iii) the greater inhibition of cathepsin B by quail cystatin compared to chicken cystatins probably arises from two additional ionic interactions between residues Arg15 and Lys112 of the inhibitor and Glu194 and Asp124 of the enzyme, respectively. The two potential salt bridges are located outside of the known contact regions between cystatins and peptidases of the papain family.

In the figure, 30 S ribosomal proteins have been arranged according to their functional role: Protein S1 is required for mRNA binding. Proteins S3, S4, S5, S11 and S12 are involved in cistron and/or codon—anticodon recognition. They must be close to the decoding sites on the 30 S subunit. Furthermore proteins S2, S3, S10, S14, S19 and S21 function in f-Met-tRNA binding. Proteins S1, S2, S3, S10, S14, S19, S20 and S21 are important for the function of both decoding sites, whereas proteins S9, S11 and S18 are only needed for EF—Tu-dependent aminoacyl-tRNA binding. Proteins S2, S5, S9 and S11 would be close to the GTPase center of the 50 S subunit, since they are important for this activity. The present available data concerning the 50 S subunit allow the following picture to be drawn: Protein L16 is involved in binding the 3′-terminus of aminoacyl-tRNA in the A-site. Next to it in the A-site, there is protein L6. The P-site is located adjacent to the A-site of the peptidyltransferase center. Accordingly, protein L2 is near protein L6 and is located in the P-site as well as proteins L27 and L4. Protein L11, which is intimately involved in peptide bond formation, would have to border parts of both A- and P-sites. Proteins L6 and L2 stimulate binding of 5 S RNA—protein complexes to 23 S RNA. The 5 S RNA—protein complex has GTPase and ATPase activities. The proteins in this complex (L5, L18, L20, L25 and L30) seem to be located close to the A-site of the peptidyltransferase center. These proteins together with protein L11 are involved in GDP binding. Proteins L10 and L6 are implicated in reconstitution of protein L7 and L12 mediated EF—G-dependent ribosomal GTP hydrolysis. This observation is supported by the fact that the aminoacyl-tRNA binding site, e.g. proteins L16 and L6, is connected with EF—G and EF—Tu binding site, e.g. proteins L7 and L12, as well as the GTPase center. Furthermore, if one of the functional roles of 5 S RNA is to bind aminoacyl-tRNA via T—Ψ—C, then those ribosomal proteins which bind to 5 S RNA (or are close to it) would be located near or at the A-site. The model of active sites in E. coli ribosome illustrated in the figure is based on the presently available experimental results. It is far from being complete and should not be overinterpreted as an accurate topographical model. More data on the functional role of ribosomal components and on the topography of the subunits can be expected in the near future and will add to the knowledge on the active sites in ribosomes.

The N-terminal sequence of HPLC-purified protein L23 from the Methanococcus vannielii ribosome has been determined by automated liquid-phase Edman degradation. Using the N-terminal amino acid sequence, an oligonucleotide probe complementary to the 5′-end of the gene was synthesized. The 26-mer oligonucleotide, containing two inosines, was used for hybridization with digested M. vannielii chromosomal DNA. The hybridizing band from HpaII-digested genomic DNA was ligated into pUC18 to yield plasmid pMvaZl containing the entire gene of protein L23. The nucleotide sequence complemented the partial amino acid sequence, and the gene codes for a protein of 9824 Da. The amino acid sequence of protein L23 form M. vannielii was compared to that of ribosomal proteins from other archaebacteria as well as from eubacteria and eukaryotes. The number of identical amino acids is highest when the M. vannielii protein is compared to the homologous protein from yeast and lowest vs that from tobacco chloroplasts. Interestingly, the secondary structures of the proteins as predicted by computer programs are more conserved than the primary structures.