Applications and limitations of regulatory RNA elements in synthetic biology and biotechnology.

Synthetic biology requires the design and implementation of novel enzymes, genetic circuits or even entire cells, which can be controlled by the user. RNA-based regulatory elements have many important functional properties in this regard, such as their modular nature and their ability to respond to specific external stimuli. These properties have led to the widespread exploration of their use as gene regulation devices in synthetic biology. In this review, we focus on two major types of RNA elements: riboswitches and RNA thermometers (RNATs). We describe their general structure and function, before discussing their potential uses in synthetic biology (e.g. in the production of biofuels and biodegradable plastics). We also discuss their limitations, and novel strategies to implement RNA-based regulatory devices in biotechnological applications. We close with a description of some common model organisms used in synthetic biology, with a focus on the current applications and limitations of RNA-based regulation.

Conformation and interactions of the packaged double-stranded DNA genome of bacteriophage T7.

The structure of the packaged double-stranded DNA genome of bacteriophage T7 was compared to that of unpackaged T7 DNA using digital difference Raman spectroscopy. Spectral data were obtained at 25 degrees C from native T7 virus (100 mg/mL), empty T7 capsids (50 mg/mL), and purified T7 DNA (40 mg/mL) in buffer containing 200 mM NaCl, 10 mM MgCl2, and 10 mM Tris at pH 7.5. At these conditions, the local conformation of T7 DNA was not affected by packaging. Specifically, the local B-form secondary structure of unpackaged T7 DNA, including furanose C2'-endo pucker, anti glycosyl torsion, Watson-Crick base pairing, and base stacking, were essentially fully (> 98%) retained when the genome was condensed within the viral capsid. However, the average electrostatic environment of T7 DNA phosphates was altered dramatically by packaging as revealed by large perturbations in the Raman bands associated with localized vibrations of the DNA phosphate groups. The change in the phosphate environment was attributed to Mg2+ ions that were packaged with the genomic DNA, and the observed Raman perturbations of genomic DNA were equivalent to those generated by a 50-100-fold increase in Mg2+ concentration in aqueous phosphodiester model compounds. The T7 data were qualitatively and quantitatively similar to those observed previously for packaged DNA of bacteriophage P22 and imply that genomic DNAs of T7 and P22 are both organized in a similar fashion within their respective capsids. The results show that the condensed genome does not contain kinks or folds that would disrupt the local B conformation by more than 2%. The present findings are discussed in relation to previously proposed models for condensation and organization of double-stranded and single-stranded viral DNA.

Secondary structure and interaction of phage D108 Ner repressor with a 61-base-pair operator: evidence for altered protein and DNA structures in the complex.

Ner repressors of the transposable phages Mu and D108 play a central role in regulating the expression of the early (transposase) operon and in ensuring that phage growth proceeds along a lytic pathway. The latter function is analogous to that performed by the Cro protein of phage lambda. Unlike lambda Cro, however, the structural basis of operator recognition is not known for the Ner repressors. In order to elucidate the structural features underlying operator recognition by Ner repressors, we have employed Raman spectroscopy as a probe of the solution secondary structures of both D108 Ner and Mu Ner. Additionally, we have obtained Raman spectra of the D108 Ner repressor when bound to a 61-base-pair oligodeoxynucleotide containing the 55-base-pair D108 ner binding site. Conformation-sensitive Raman bands show that both D108 and Mu Ner contain similar, highly alpha-helical (approximately 45%) secondary structures. The Raman markers also show that the substantial nonhelical secondary structure of both D108 Ner and Mu Ner is largely beta-stranded. The protein-free 61-bp D108 ner operator exhibits Raman marker bands diagnostic of an uninterrupted B DNA duplex. In the D108 Ner:DNA complex, we find the following: (i) B DNA stereochemistry is fully conserved, although with significant perturbations to the B form backbone geometry, particularly in AT-rich regions of the bound operator. (ii) The specific interactions that occur between Ner repressor and operator involve B DNA major groove sites. (iii) A small (8 +/- 3%) increase in alpha-helix content of the Ner repressor is detected upon operator binding. (iv) Finally, the local environments of many aromatic amino acids are substantially altered in the D108 Ner:DNA complex. We propose a molecular model for binding of D108 Ner to its operator that is consistent with both the present spectroscopic findings and the results of recent biochemical studies. Essential features of this model are bending of the DNA double helix and contact of operator sites with repressor domains bearing sequence homologies with the helix-turn-helix (HTH) motifs of other DNA-binding proteins. The Raman fingerprint of the Ner:DNA complex is shown to be clearly distinguishable from that of the lambda cI:DNA complex, even though both gene regulatory complexes are presumed to employ HTH recognition motifs. The unique Raman signatures observed for these repressor complexes suggest that the Raman methodology may be useful in discriminating different modes of operator recognition by the HTH motifs of regulatory proteins.

Novel tyrosine markers in Raman spectra of wild-type and mutant (Y21M and Y24M) Ff virions indicate unusual environments for coat protein phenoxyls.

The tyrosine side chain generates a pair of distinctive Raman bands--a Fermi doublet near 850 and 830 cm-1--with relative intensities diagnostic of hydrogen bonding states of the phenolic acceptor and donor atoms [Siamwiza et al. (1975) Biochemistry 14, 4870-4876]. This structural correlation has been tested extensively and is used widely as an indicator of tyrosine interactions in globular proteins and their assemblies. However, in Ff filamentous viruses (fd, f1, M13) the apparent Fermi doublet intensity ratio (I853/I826 approximately 4.0) is much greater than the maximum predicted or observed in other proteins. To understand this anomaly, we have reevaluated the basis for the Fermi doublet assignment in Ff. We report Raman spectra of site-specific mutants of Ff in which either one (Y21M and Y24M) or both (Y21F/Y24S) tyrosines of the coat protein subunit (pVIII) have been mutated. These Raman data, together with those obtained from Ff virions carrying residue-specific tyrosyl (Y-d4) and phenylalanyl (F-d5) deuterations in pVIII, demonstrate conclusively that the 853 and 826 cm-1 bands of Ff do not constitute a typical tyrosine Fermi doublet: The observed 826 cm-1 Raman band of Ff is due not to tyrosine but to phenylalanine residues of pVIII. The 853 cm-1 Raman band thus constitutes the first known example of a "tyrosine singlet" in the Raman spectrum of a protein. The implications of this finding for Ff virion structure and its relevance to tyrosine markers in other proteins are discussed.

Subunit conformational changes accompanying bacteriophage P22 capsid maturation.

In double-stranded DNA bacteriophages, packaging of dsDNA requires the transformation of a precursor procapsid into a mature viral capsid. Lattice expansion and release of scaffolding subunits accompanying DNA packaging. Three-dimensional structures of procapsid and mature phage lattices demonstrate that the capsid transformation involves substantial changes in subunit environment. Since this transformation occurs without subunit dissociation, it represents a transition between at least two stable subunit conformations. Using Raman spectroscopy, we have identified changes in coat protein secondary structure and side-chain environments which accompany the capsid transformation. The subunits of procapsid shells contain only 2.0 +/- 0.4% more alpha-helix and less beta-sheet than those of mature capsids; however, numerous side chains are substantially altered by the transformation, including tyrosines, tryptophans, phenylalanines, and aliphatics, which are widely distributed through the subunit sequence. We propose, therefore, that procapsid expansion is accomplished through the relative motion of coat subunit domains with little change in secondary structure. Such hinge-bending conformational transitions may couple ATP-dependent dsDNA condensation with shell expansion.

Secondary structure and interactions of the packaged dsDNA genome of bacteriophage P22 investigated by Raman difference spectroscopy.

Vibrational spectra of the double-stranded DNA genome of bacteriophage P22 in packaged and unpackaged states are compared by digital difference Raman spectroscopy. The difference Raman spectrum, which is sensitive to structural changes at the level of < 2% of a given nucleotide type, reveals the effects of packaging upon sugar pucker, glycosyl orientation, phosphodiester geometry, base pairing, base stacking, and the electrostatic environment of DNA phosphate groups. For both packaged and unpackaged states, the experiments were performed on aqueous solutions at 25 degrees C containing effective P22 DNA concentrations of 30-50 mg/mL in 200 mM NaCl + 10 mM MgCl2 + 10 mM Tris at pH 7.5. At the experimental conditions employed, the B-form secondary structure of unpackaged P22 DNA is minimally perturbed by packaging the viral genome in the virion capsid. However, the electrostatic environment of DNA phosphates is dramatically altered with packaging. Specifically, we find the following: (1) C2'-endo sugar pucker and anti glycosyl orientations are conserved for all nucleosides. (2) Watson-Crick base pairing is essentially completely retained. (3) Alternative secondary structures, whether right- (A or C form) or left-handed (Z form), are not evident in either the packaged or unpackaged viral genome. (4) Small Raman hyperchromic effects (< 10%) observed for certain marker bands of dG, dA, and dT in the packaged state of P22 DNA suggest slightly reduced base-stacking interactions with packaging. These are consistent with previously reported UV hyperchromic effects, but the Raman spectrum shows that they are not associated with either base unpairing or strand separation.(ABSTRACT TRUNCATED AT 250 WORDS)

Raman spectroscopy of filamentous bacteriophage Ff (fd, M13, f1) incorporating specifically-deuterated alanine and tryptophan side chains. Assignments and structural interpretation.

Structural interpretation of the Raman spectra of filamentous bacteriophages is dependent upon reliable assignments for the numerous Raman vibrational bands contributed from coat protein and packaged DNA of the virion. To establish unambiguous assignments and facilitate structural conclusions derived from them, we have initiated a systematic study of filamentous bacteriophage Ff (fd, f1, M13) incorporating protein subunits with specifically deuterated amino-acid side chains. Here, we report and interpret the Raman spectra of fd virions which incorporate: (a) a single deuterio-tryptophan residue per coat protomer [fd(Wd5)], (b) ten deuterio-alanines per protomer [fd(10Ad3)], and (c) both deuterio-tryptophan and deuterio-alanine [fd(Wd5 + 10Ad3)]. The unambiguous assignment of coat protein Raman bands in normal and deuterated isotopomers of fd establishes the validity of earlier empirical assignments of many key Raman markers, including those of packaged ssDNA (Thomas et al., 1988). Present results confirm that deoxyguanosine residues of the packaged ssDNA molecule depart from the usual C2'-endo/anti conformation characteristic of protein-free DNA in aqueous solution, although C2'-endo/anti conformers of thymidine are not excluded by the data. The combined results obtained here on normal fd, and on fd incorporating deuterio-tryptophan [fd(Wd5) and fd(Wd5 + 10Ad3)], show also that the microenvironment of the single tryptophan residue per coat protomer (W26) can be clearly deduced as follows: (a) The indole 1-NH donor group of each protomer in fd forms a moderately strong hydrogen bond, most likely to a hydroxyl oxygen acceptor. (b) The planar indole ring exists in a hydrophilic environment. (c) The torsion angle describing the orientation of the indole ring (C3-C2 linkage) with respect to the side-chain (C alpha-C beta bond) is unusually large, i.e., magnitude of X2,1 approximately 120 degrees. With respect to alanine isotopomers, the present results show that alanine residues, and possibly other methyl-containing side chains, are significant contributors to the fd Raman spectrum. The present study provides new information on protomer side chains of fd and demonstrates a Raman methodology which should be generally useful for investigating single-site interactions and macromolecular conformations in other nucleoprotein assemblies.