Search

A new modular gene-expression system for use in studies of translational control in Saccharomyces cerevisiae was constructed. A GAL::PGK fusion promoter (GPF) directed the inducible synthesis of mRNAs initiated at a single major site. A series of leader sequences were tested in combination with each of two reporter genes (encoding chloramphenicol acetyl transferase (cat) and luciferase (luc)). Stem-loop structures of three different sizes and predicted stabilities were inserted into each of two different unique restriction sites in the leader. After correction for relative mRNA abundance, a stem-loop of predicted stability equivalent to approximately -18 kcal mol-1 inhibited translation by up to 89%. The degree of inhibition exerted by the other stem-loops correlated positively with their predicted stabilities. Combinations of two stem-loops at different sites yielded an inhibitory effect greater than that of either individual stem-loop alone. Similar inhibitory effects were observed with both reporter genes. However, inhibition of translation, particularly of the cat gene, was more effective when the stem-loop was positioned close to the start codon rather than at the 5' end of the leader. The observed results reflect an important form of post-transcriptional control that is expected to act on a large number of genes in yeast.

IL-1 receptor heterogeneity in murine lymphocytes was investigated by cross-linking to [125I]IL-1 alpha and competition with IL-1 receptor antagonist, and the molecular identity of the IL-1 receptors was identified with PCR using primers specific for type I and type II IL-1 receptors. The thymoma cell line E14 6.1 exhibits exclusively the 80 kDa receptors which proved to be the type I receptor according to PCR analysis. In the pre-B cell line 70Z/3, predominantly a 60 kDa type II receptor but also a trace of type I receptor can be identified by PCR. The Th2 cell line D10N expresses both types of IL-1 receptors in equivalent amounts according to cross-linking experiments and PCR. The proliferative response of D10N cells to IL-1 is inhibited by IL-1 ra which according to cross-linking affects the binding to the type I receptor only. It is concluded that coexpression of both types of IL-1 receptors might be a characteristic of murine Th2 cells and that their growth-dependence on IL-1 is mediated by the type I receptor.

Postsplenectomy septic sequelae may be fatal, but the mechanisms are unclear. We hypothesized that peritoneal macrophage (PM phi) antimicrobial function is abnormal following splenectomy and that this may predispose to increased mortality from the fungal pathogen Candida albicans. Study 1 (in vivo): female CD-1 mice were randomized into control (C), laparotomy (L), or laparotomy+splenectomy (L + S) and inoculated with C. albicans (10(7) organisms, ip) and were studied for mortality. Study 2 (in vitro): mice were randomized to C, L, or L + S groups. Twenty-four hours later, PM phi were harvested and studied for their antifungal activity, including percentage PM phi ingestion of C. albicans and vacuolar sealing of C. albicans within PM phi, percentage C. albicans killing, and superoxide anion (O2-) generation, the mechanism by which candida are killed. Results showed decreased phagocytosis and killing of C. albicans in the L + S group (P < 0.05 vs C) and reduced vacuolar sealing (P < 0.05 vs C) but significantly higher O2- release compared to that in other groups (P < 0.05). Mortality in the L + S group from C. albicans sepsis was significantly higher than that in the other groups (60% compared to 20% in the L group and 13% in C, P < 0.02). This may have resulted from L + S-induced defective phagocytosis of C. albicans and depressed C. albicans killing but increased O2- release in response to candida. This discrepancy between decreased killing and increased O2- may result from increased leakage of O2- from more unsealed vacuoles in the L + S group. Thus, L + S may predispose to candida-induced mortality through defective PM phi intracellular candida killing while enhancing the release of O2- extracellularly from unsealed vacuoles, causing tissue injury.

A new modular in vivo/in vitro expression system was constructed which facilitates studies of the control and regulation of gene expression in the yeast Saccharomyces cerevisiae. We studied the influence of stem-loop structures inserted into the non-translated leader region upon the steady-state levels and translation of mRNAs bearing the cat gene from the bacterial transposon Tn9. mRNA abundance changed relatively little in response to alterations in the leader sequence and structure, whereas stem-loop structures clearly inhibited translation to a degree that was dependent upon the predicted stability as well as the position of the inserted secondary structure. A stem-loop structure with a predicted stability greater than -28 kcal mol-1 and with a stem comprising at least 15 (mainly G/C) base pairs inhibited translation in vivo by at least 98%. A stem-loop structure with a predicted stability of approximately -14 kcal mol-1, whose stem comprised at least six G/C base pairs, inhibited translation in vivo by at least 66%. The hairpins were more inhibitory when placed close to the start codon than when positioned near the 5' end of the leader. An mRNA showing extensive complementarity between the leader and trailer regions was not only poorly translated but also had a steady-state level at least three times higher than the average for all the cat constructs examined. Translation of the various mRNAs in a yeast cell-free system followed qualitatively the same pattern as the results obtained in vivo. The stem-loop structures were far less inhibitory in a reticulocyte lysate system. Overall, the data are likely to reflect the full spectrum of translational activities of yeast mRNAs in vivo determined by secondary structure and emphasize the importance of translation as a control step in gene expression.

A new modular expression system was developed to direct the in vitro synthesis of defined transcripts that were used as templates for translation in yeast cell-free extracts. The system was used to examine the influence of 5'-capping, 3'-polyadenylation and leader sequence upon the translation and stability of the synthetic Tn9 cat (chloramphenicol acetyl transferase), yeast PGK (phosphoglycerate kinase) and yeast HSP26 (heat-shock protein 26) mRNAs. The addition of a methylated cap (m7Gppp) or of a poly(A) tail enhanced translation and stabilized the mRNA. The dependence of translation upon capping was reduced in the presence of the HSP26 leader sequence. This may indicate the existence of a translational mechanism that enhances cap-independent translation. The enhancement of the translation and stability of mRNA was relatively insensitive to changes in the position of the poly(A) tail relative to the reading frame.

The 38-kDa protein (Ag38) of the Gram+ bacterium, Mycobacterium tuberculosis H37Rv, is an immunodominant antigen of potential utility for diagnosis and vaccine development. Assessment of this potential requires large amounts of the purified protein that would be difficult, if not impossible, to obtain from M. tuberculosis itself. The gene coding for Ag38 had been previously cloned and in the present study was expressed as an unfused protein in Escherichia coli under the control of strong transcriptional (bacteriophage lambda pLpR) and translational (atpE) signals. Fermentation of the recombinant E. coli K-12 strain CAG629[pMS9-2], which is deficient in Lon protease and the heat-shock response, produced recombinant Ag38 (reAg38) at high levels (about 10% of total cellular protein). The reAg38, which accumulated as inclusion bodies, was completely solubilized in 6 M guanidine.HCl, refolded and purified to apparent homogeneity. The product showed the expected amino acid composition and M(r), and had similar reactivities as the native protein with three different mAb. Polyclonal antibodies raised against reAg38 reacted strongly with the native antigen in enzyme-linked immunosorbent assay. These results demonstrate that reAg38, which cannot be distinguished antigenically from the native protein of M. tuberculosis, can be prepared in quantity from E. coli.

The physical and functional stabilities of genes in the atp operon fall into two classes. The first two genes, atpI and atpB, are rapidly inactivated and degraded at the mRNA level. The remaining seven genes are more stable. In order to investigate how these stabilities are determined, DNA sequences encoding mRNA structures that influence degradative events in other systems, including RNAse III sites and REP sequences, were subcloned or synthesized and inserted into non-coding regions of the operon. The effects of insertion of an RNAse III site depended on whether cleavage left an unstable 3' end or a stabilizing stem-loop upstream of the cutting point. Generation of an unstable 3' end destabilized the neighbouring upstream atp gene, thus modifying the course and rate control of degradation. Removal of the atp transcriptional terminator attenuated expression of the last gene of the operon, atpC. This effect was reversed by substitution of an alternative stem-loop for the terminator. REP sequences inserted into intercistronic regions apparently could not influence rate-controlling steps. The reported data shed light on the factors controlling the inactivation and degradation of genes in the polycistronic atp mRNA, and are discussed in relation to the general role of degradation processes in the control of gene expression.

The uncI gene is the most poorly expressed gene of the unc operon of Escherichia coli. We examined how the structural features of the translational initiation region (TIR) of this gene relate to the efficiency of its expression. A set of various TIR sequences was created for the uncI gene by coupling synthetic oligodeoxynucleotides to the 5' part of the plasmid-borne gene. Analyses of transcription and translation of the resulting constructs revealed that weak translational initiation efficiency of the uncI TIR sequence is a rate-controlling factor for expression. Moreover, the identification of an endonucleotytic cleavage site within the 5' part of the uncI mRNA by Northern blotting lends further support to the notion that instability of the uncI cistron within the polycistronic unc mRNA represents a further mode of control. The analysis of the roles of the different Shine-Dalgarno regions and putative start codons within the translational initiation region of the uncI gene revealed that the Shine-Dalgarno region III and the GUG codon (codon 4 of the uncI coding sequence described by Walker, J. E., Saraste, M., and Gay, N. J. (1984) Biochim. Biophys. Acta 768, 164-200) define the main site of initiation. However, manipulations of the translational initiation region led to selection of an initiation codon further upstream. The relationship between mRNA sequence and higher order structure on the one hand, and initiation site selection and initiation efficiency on the other, was analyzed.

The atp operon of Escherichia coli directs synthesis rates of protein subunits that are well matched to the requirements of assembly of the membrane-bound H(+)-ATPase (alpha 3 beta 3 gamma 1 delta 1 epsilon 1a1b2c10-15). Segmental differences in mRNA stability are shown to contribute to the differential control of atp gene expression. The first two genes of the operon, atpl and atpB, are rapidly inactivated at the mRNA level. The remaining seven genes are more stable. It has previously been established that the translational efficiencies of the atp genes vary greatly. Thus differential expression from this operon is achieved via post-transcriptional control exerted at two levels. Neither enhancement of translational efficiency nor insertion of repetitive extragenic palindromic (REP) sequences into the atplB intercistronic region stabilized atpl. We discuss the implications of these results in terms of the pathway of mRNA degradation and of the role of mRNA stability in the control of gene expression.

The new expression vector system CYTEXP is designed to facilitate the optimization of both transcription and translation in Escherichia coli, while at the same time allowing the exchange of its major components using unique restriction sites. In vitro mutagenesis can be performed in situ using single-stranded DNA generated from the bacteriophage f1 ORI sequence. The basic vector pCYTEXP1 bears a synthetic copy of the intercistronic sequence that enhances the translation of the E. coli atpE gene. Reading frames can be inserted directly downstream of this sequence. The bacteriophage lambda promoters, the atpE sequence, the bacteriophage fd transcriptional terminator, the f1 ORI, and the amp antibiotic resistance gene are all borne on exchangeable "modules." Thus, both the efficiency and the conditions of expression of cloned genes can be readily optimized.

A possible approach for altering alkaloid biosynthesis in plants is the expression of genes encoding key enzymes of a pathway such as lysine decarboxylase (ldc) in transgenic plants. Two strategies were followed here: one focused on expression of the gene in the cytoplasm, the other on subsequent targeting of the protein to the chloroplasts. The ldcgene from Hafnia alvei was therefore (a) placed under the control of the 1' promoter of the bidirectional Tr promoter from Agrobacterium tumefaciens Ti-plasmid, and (b) cloned behind the rbcS promoter from potato fused to the coding region of the rbcS transit peptide. Both ldc constructs, introduced into Nicotiana tabacum with the aid of A. tumefaciens, were integrated into the plant genome and transcribed as shown by Southern and northern hybridization. However, LDC activity was only detectable in plants expressing mRNA under the control of the rbcS promoter directing the LDC fusion protein into chloroplasts with the aid of the transit peptide domain. In plants expressing the processed bacterial enzyme cadaverine levels increased from nearly zero to 0.3-1% of dry mass.

A series of atp::lacZ fusions has been constructed for use in a study of translational coupling in the central region of the Escherichia coli atp operon. Five genes, atpE, atpF, atpH, atpA and atpG, were shown to be translationally coupled to various degrees of tightness. A new lac promoter vector, compatible with the atp::lacZ fusion vectors, was used to express individual atp genes in the same hosts as the fusion genes. The H(+)-ATPase subunits thus synthesized exercised no significant trans-regulation on the expression of the atp::lacZ fusions, indicating that the coupling is primarily cis. The mechanism of this coupling was investigated using in vitro mutagenesis. At least in the case of the pair atpHA, coupling seems to involve facilitated binding of fresh ribosomes to the atpA translational initiation regions.

Post-transcriptional control mechanisms assume special significance in polycistronic operons. Differential gene expression in the atp operon of Escherichia coli is primarily attributable to translational control and, to a lesser extent, to control of mRNA stability. At the same time, the polycistronic environment influences, to varying degrees, the relative importance of the different types of post-transcriptional control. The present article briefly reviews more recent results obtained through studies of the atp operon. Investigations of the pathway and kinetics of mRNA decay have yielded new information about the role of degradative mechanisms in the overall scheme of control. Moreover, translational coupling has been shown to feature as a major form of interaction between the atp genes. The relevance of these and other data is discussed in the wider context of the post-transcriptional control mechanisms generally available to E. coli.

Awareness of the importance of post-transcriptional control of gene expression in prokaryotes has grown enormously over the past ten years. In particular, translation features as a step where both control over constitutive rates of gene expression, as well as cis and trans regulation are exercised. Recent research has provided us with new insights into the molecular basis of these phenomena.

Transcription and translation of the atp genes encoding the subunits b, delta, alpha, gamma and epsilon of the Escherichia coli H+-ATPase were studied. The nature and quantities of the respective transcripts initiated from different promoters were compared with overall expression rates thus yielding accurate information about relative translational efficiency and its coupling to mRNA levels. Part of the highly efficient subunit c gene translational initiation region (TIR) was used as a tool in manipulating the TIRs of the other genes. Rate control of atp cistron translation occurs at the initiation level and is determined locally by each gene's TIR. In this way, individual subunit synthesis rates are set to match the requirements for H+-ATPase assembly. There is no (or very restricted) translational coupling between the cistrons. Translational initiation rates of the normally weakly expressed atp genes could be increased by up to a factor of 27 by manipulating the sequences upstream of the start codons, despite biased codon usages. In the presence of an improved upstream sequence, the N-terminal sequence of the subunit gamma gene exerted a limiting effect. This could be relieved by altering the sequence of the first seven codons. The levels of subunit gamma mRNA were more sensitive to changes in translational efficiency than the concentrations of the other atp mRNAs. The relationships between initiation efficiency and primary and secondary structure in the natural and manipulated atp TIRs are discussed in detail.

The cDNA sequences encoding mature human interleukin 2 (IL2) and beta-interferon (INF beta), respectively, were fused with various translational initiation regions and inserted into two different types of expression vector. The relative levels of expression of the two genes and the functional stability of their respective mRNAs were examined in vivo in Escherichia coli hosts. The addition of the 30-bp sequence, found immediately upstream of the E. coli atpE gene Shine-Dalgarno (SD) sequence, to the translational initiation regions of IL2 and INF beta increased the expression of both these genes by a factor of 6-10. Thus this sequence, which naturally acts within the E. coli atp operon to enhance the translational initiation frequency of the atpE gene, can increase the expression of other genes in E. coli. It may exemplify a specific type of recognition signal for the E. coli translational apparatus.

The gene for the FeS protein of the Rhodopseudomonas sphaeroides b/c1 complex was identified by means of cross-hybridization with a segment of the gene encoding the corresponding FeS protein of Neurospora crassa. Plasmids (pRSF1-14) containing the cross-hybridizing region, covering in total 13.5 kb of chromosomal DNA, were expressed in vitro in a homologous system. One RSF plasmid directed the synthesis of all three main polypeptides of the R. sphaeroides b/c1 complex: the FeS protein, cytochrome b and cytochrome c1. The FeS protein and cytochrome c1 were apparently synthesized as precursor forms. None of the pRSF plasmids directed the synthesis of the 10-kd polypeptide found in b/c1 complex preparations. Partial sequencing of the cloned region was performed. Several sites of strong homology between R. sphaeroides and eukaryotic polypeptides of the b/c1 complex were identified. The genes encode the three b/c1 polypeptides in the order: (5') FeS protein, cytochrome b, cytochrome c1. The three genes are transcribed to give a polycistronic mRNA of 2.9 kb. This transcriptional unit has been designated the fbc operon; its coding capacity corresponds to the size of the polycistronic mRNA assuming that only the genes for the FeS protein (fbcF), cytochrome b (fbcB) and cytochrome c1 (fbcC) are present. This could indicate that these three subunits constitute the minimal catalytic unit of the b/c1 complex from photosynthetic membranes.

1. Reduction in the magnitude of the respiration-dependent protonmotive force (proton electrochemical gradient in mV) of vesicles from Paracoccus denitrificans, and of submitochondrial particles, has been found to be paralleled small increases in S50% values for both ADP and Pi. For example, reduction of the protonmotive force of P. denitrificans vesicles from 145 mV to 110 mV was accompanied by an increase of S50% (ADP) from 8 microM to 18 microM, and an increase of S50% (Pi) from 0.33 mM to 1.4 mM. This result was obtained with partial uncoupling quantities of both carbonyl-cyanide p-trifluoromethoxyphenylhydrazone and of the synergistic combination of nigericin plus valinomycin in the presence of K+. In view of the similar effects of these two different methods of uncoupling it is concluded that the changes in S50% were a consequence of the diminished protonmotive force acting on the ATP synthase rather than of a secondary, direct interaction of the uncouplers with the enzyme. Changes in S50% rather than Km are described because under several sets of conditions double-reciprocal plots were nonlinear. 2. For equivalent attenuations in the rate of ATP synthesis by submitochondrial particles, 2,4-dinitrophenol caused much larger increases in S50% (ATP) than did carbonylcyanide p-trifluoromethoxyphenylhydrazone. Therefore it is concluded that the effect of 2,4-dinitrophenol was primarily a consequence of its previously recognized direct interaction with the F1 segment of the mitochondrial ATPase. The concentration range of 2,4-dinitrophenol that raised S50% (ADP) is similar to that which weakens the binding of ADP to a particular type of site on the purified F1 sector of ATP synthase. This correlation is consistent with such a site having a catalytic role during ATP synthesis. 3. A titration of the rate of ATP synthesis by vesicles of P. denitrificans with increasing quantities of carbonylcyanide p-trifluoromethoxyphenylhydrazone showed that the initial titres of the uncoupler caused large decreases in the rate of ATP synthesis for relatively small attenuations in the protonmotive force. Thus the initial 20 mV drop in the protonmotive force was accompanied by a reduction of more than 65% in the rate of ATP synthesis. Over the lowest range of values of protonmotive force that drove detectable rates of ATP synthesis however, the dependence of the rate was a less steep function of the protonmotive force. A plot of the logarithm of the rate of ATP synthesis against protonmotive force reveals a biphasic relationship. There does not appear to be a 'threshold' value of the protonmotive force below which ATP synthesis is blocked by kinetic factors. 4. The relationships of the protonmotive force with S50% values and with the rate of ATP synthesis (at near saturating concentrations of ADP and Pi) are discussed in relation to possible mechanisms for the coupling of proton translocation to ATP synthesis.

An attractive subject for investigations of post-transcriptional control is the atp operon, whose nine genes are differentially expressed. The primary mode of control of atp gene expression is exercised at the translational level. It has been clearly demonstrated for almost all of the atp genes that the primary and secondary structures of their respective translational initiation regions direct translational initiation rates that correspond well to the requirements for these subunits in the cell. The relationship between the structure of the translational initiation region, including bases upstream from the Shine-Dalgarno region and downstream from the start codon, and the rates of initiation that it determines, has been investigated in more detail using various polycistronic and monocistronic systems. No evidence could be found for a role of codon usage bias in controlling overall translation rates. The functional half-lives of atpE and of the other six cistrons downstream from it are similar. The chemical stabilities of the first two cistrons of the polycistronic atp mRNA may, however, be lower, and we are investigating the possibility that there may also be control of atp gene expression exercised at the level of mRNA stability. The effects of manipulations of the intercistronic regions of at least the plasmid borne atp operon are consistent with a model of mRNA decay in which rate control is associated with endonucleolytic cleavages within individual cistrons. The experimental data are discussed in relation to the possible ways in which primary and secondary structures of the mRNA might control translational efficiency and stability.

The c, b and delta subunit genes of the Escherichia coli atp operon were cloned individually in an expression vector between the tac fusion promoter and the galK gene. The relative rates of subunit synthesis directed by the cloned genes were similar in vitro and in vivo and compared favourably with the subunit stoichiometry of the assembled proton-translocating ATP synthase of E. coli in vivo. The rate of synthesis of subunit c was at least six times that of subunit b and 18 times that of subunit delta. Progressive shortening of the long intercistronic sequence lying upstream of the subunit c gene showed that maximal expression of this gene is dependent upon the presence of a sequence stretching greater than 20 bp upstream of the Shine-Dalgarno site. This sequence thus acts to enhance the rate of translational initiation. The possibility that similar sequences might perform the same function in other operons of E. coli and bacteriophage lambda is also discussed. Translation of the subunit b cistron is partially coupled to translation of the preceding subunit c cistron. In conclusion, the expression of all the atp operon genes could be adjusted to accommodate the subunit requirements of ATP synthase assembly primarily by means of mechanisms which control the efficiency of translational initiation and re-initiation at the respective cistron start codons.

In this study the relationship between birth weight and passive exposure to tobacco smoke is assessed for the first time by serum cotinine measurements. Among 1231 nonsmoking white women whose blood was sampled during the second trimester of pregnancy, 31.4% had serum cotinine levels between 1.0 and 9.9 ng/ml and were therefore considered to be passively exposed to tobacco smoke. The crude mean birth weight of infants of the women passively exposed to smoke was 107 gm lower than that of infants of unexposed women, and that difference remained after the application of a multivariate analysis that included the major known birth weight-associated covariates. These findings are consistent with a causal relationship between passive exposure to tobacco smoke and birth weight and suggest that the dose-response relationship may not be linear.

The order-disorder transitions of the double helices formed by the ribo-oligoadenylic acids rA8 and rA10 at pH 4.2 have been investigated in a series of organic/aqueous mixed solvents. Melting temperature data, Tm, derived from the uv melting curves were used to define the stability of the double helices in the different mixed solvent systems. It was found that the extent of helix destabilization depended in a predictable fashion on both the quantity and the nature of the added organic solvents. For the C1 through C4 aliphatic alcohols, the longer, less branched alcohols proved to be more effective destabilizers of the helical structure. Significantly, the amides proved to be more powerful destabilizers than the alcohols. Analysis of the melting curves provided the Van't Hoff enthalpy change for each transition. The data are interpreted in terms of the role of solvent in the stabilization of ribonucleic acid structure.

A range of translational initiation regions (TIR) was created by combining synthetic DNA fragments derived from the atpB-atpE intercistronic sequence of Escherichia coli with the cDNA sequence encoding mature human interleukin 2 (IL-2), the E. coli fnr gene, or an fnr::lacZ gene fusion. Both the overall rates of gene expression and the relative concentrations and stabilities of the corresponding mRNA species were estimated in strains bearing the constructs on plasmids. These measurements served as the basis for analyses of the relationship between the structure of the TIR and the true rates of translation that it promotes. The constructs involving the IL-2 cDNA were predicted to allow much less stable secondary structure within the TIR than those involving the N-terminal region of the fnr gene. Thus by combining one set of upstream sequences with two different types of N-terminal coding sequence, it was possible to distinguish between the respective influences of primary and secondary structure upon initiation. The data indicate that in the presence of a given Shine-Dalgarno (SD)/start codon combination, the decisive factor for translational initiation efficiency is the stability of base pairing involving, or in the vicinity of, this region. The sequences contributing to this secondary structure can be many bases upstream of the SD region and/or downstream of the start codon. There was no indication that the specific base sequence upstream of the SD region could, other than to the extent that it contributed to the local secondary structure, significantly influence the efficiency of translational initiation.

The translational initiation rates directed by the translational initiation regions (TIRs) of the atpB, atpH, atpA and atpG genes of Escherichia coli were investigated using lacZ fusions present on plasmids as well as integrated into the chromosome. This was the first investigation of the translational efficiency of the atpB gene, whose unfused product (subunit a) can be toxic to the cell. The specific mRNA levels, rates of in vivo protein synthesis and beta-galactosidase activities encoded by the atp::lacZ fusions were compared in order to obtain valid estimates of relative translation rates. The results indicate that in the E. coli atp operon, translation directed by the atpB, atpH and atpG TIRs is less efficient than that directed by the atpA TIR, and are thus consistent with earlier measurements of direct atp gene expression. Initiation is, however, to differing extents, controlled by coupling to the translation of upstream neighbours. There is particularly tight coupling between atpH and atpA. Increasing the distance between these two genes whilst maintaining the original atpA TIR structure decreased the degree of coupling. The influence of manipulations of the atpG TIR structure upon translational efficiency was quantitatively more pronounced when the atpG fusions were present as a single copy per chromosome. This is likely to be related to the mRNA binding characteristics of 30S ribosomal subunits and/or to the influence of other (trans-acting) factors. The control of independent and coupled initiation at the atp TIRs is discussed in relation to mRNA structure and possible cis and trans regulatory phenomena.

The atp operon of Escherichia coli comprises nine genes that are differentially expressed. The control of the atp genes' expression rates has been shown to be exercised primarily at the level of translational initiation, but how is this achieved in molecular terms? In order to study the interactions of 30 S ribosomal subunits with specifically the translational initiation regions (TIRs) of atpB, atpE and atpG, restriction fragments bearing these TIRs were excised from the atp operon and cloned into an SP6 promoter transcription vector. mRNA transcripts were made in vitro and used in primer extension inhibition studies and equilibrium mRNA-30 S ribosomal subunit binding measurements. The binding of 30 S ribosomal subunits blocked primer extension 14 to 15 bases downstream from the respective translational start codons. The affinities of binding of 30 S ribosomal subunits showed the relationship atpE greater than atpB greater than atpG. This was also the order of the efficiency of translation promoted by the respective TIRs, both in vivo and on the in vitro synthesized mRNA fragments. Thus, the affinity of 30 S ribosomal subunits is at least to some extent correlated with the rate of translational initiation.

Synthetic genes encoding bovine and human basic fibroblast growth factors (bFGFs) were assembled and cloned using established Escherichia coli expression plasmids. Transformed E. coli cells were able to synthesize either a fusion protein, comprising the first seven amino acids of beta-galactosidase, a linker fragment and bovine FGF, or genomic human bFGF. The two growth factors were purified from E. coli lysates by cation exchange and heparin-Sepharose affinity chromatography. The purified recombinant proteins were biologically active as monitored by their mitogenic activity for bovine aortic endothelial cells and their angiogenic capacity in the rabbit cornea.

New expression vectors were constructed for use in strains of Escherichia coli. Their most important feature is a polylinker system that facilitates the insertion of a gene in an optimal relationship to the highly efficient E. coli atpE translational initiation region (from nucleotide -50 to the start codon). Three ATG-containing restriction endonuclease sites can be used for the insertion of the 5' end of a gene at, or near to, its translational initiation codon. These sites may alternatively be used for the creation of a suitable translational start codon. Transcription is started by the bacteriophage lambda major promoters pR and pL in tandem and terminated by the bacteriophage fd terminator. Transcriptional initiation is very effectively repressed at 28-30 degrees C by the product of the bacteriophage lambda cIts857 gene, which is also present on the vectors. Full induction is achieved by shifting the incubation temperature to 42 degrees C. The combination of highly efficient transcriptional and translational signals on these vectors allowed high-level expression of sequences encoding human interferon beta and interleukin 2 and of the E. coli atpA, sucC and sucD genes.

The initial stages of our research on the effects of the overassessment or underassessment of the consequences of nonroutine events (a potentially important factor in decision-making by pilots) is described. Critical difficulties in the measurement of life stress are discussed and a partial solution is suggested in the form of the Life Events Questionnaire (LEQ), an instrument on which individuals report the nature of stressful events, when they happened, and how much they concern them at the time of testing. The results show that the tendency to overassess or underassess the consequences of nonroutine events is consistent within individuals, but is unaffected by life stress levels on the artificial air traffic control simulation used. Analysis of LEQ results showed that high stress subjects had much higher chronic than acute stress scores (p less than 0.001), although the chronic and acute scores were virtually identical for low stress subjects.

1. Aerobic respiration by cells of Paracoccus dentrificans drives the uptake of the lipophilic cation butyltriphenylphosphonium. Anaerobiosis or addition of an uncoupler of oxidative phosphorylation (carbonyl cyanide p-trifluoromethoxyphenylhydrazone) results in efflux of the cation. Changes in the concentration of butyltriphenylphosphonium in the suspension medium were measured by using an ion-selective electrode, the construction of which is described. 2. If the uptake of butyltriphenylphosphonium is used as an indicator of membrane potential, then at pH 7.3 an estimate of about 160 mV is obtained for cells of P. dentrificans respiring aerobically in 100 mM-Hepes [4-(2-hydroxyethyl)-1-piperazine-ethanesulphonic acid/NaOH or 100mM-NaH2PO4/NaOH. This potential, however, is decreased by more than 20 mV in reaction media containing a high concentration of phosphate (100 mM) together with at least 1 mM-K+. 3. Anaerobic electron transport with NO3-, NO2- or N2O as terminal electron acceptor generates a membrane potential of about 150mV in described suspension media. The presence of these species under aerobic conditions, moreover, has negligible effect upon the extent of uptake of butyltriphenylphosphonium normally driven by aerobic respiration. These data indicate that none of these molecules exert a significant uncoupling effect on the protonmotive force. 4. No 204Tl+ uptake into respiring cells was detected. This adds to the evidence that 204Tl+ is not a freely permeable cation in bacterial cells and therefore not an indicator of membrane potential as has been proposed. The absence of respiration-driven 204Tl+ uptake indicates that P. denitrificans cells grown under the conditions specified in the present work do not possess K+-transport systems of either the Kdp or TrkA types that have been described in Escherichia coli.

The unc (or atp) operon of Escherichia coli comprises eight genes encoding the known subunits of the proton-translocating ATP synthase (H+-ATPase) plus a ninth gene (uncI) of unknown function. The subunit stoichiometry of the H+-ATPase (alpha 3 beta 3 gamma 1 delta 1 epsilon 1 a1b2c10-15) requires that the respective unc genes be expressed at different rates. This review discusses the experimental methods applied to determining how differential synthesis is achieved, and evaluates the results obtained. It has been found that the primary level of control is translational initiation. The translational efficiencies of the unc genes are determined by primary and secondary mRNA structures within their respective translational initiation regions. The respective rates of translation are matched to the subunit requirements of H+-ATPase assembly. Finally, points of uncertainty remain and experimental strategies which will be important in future work are discussed.

1. Vesicles from Paracoccus denitrificans were prepared by applying an osmotic shock to spheroplasts derived from cells that had been grown anaerobically with succinate as carbon source and nitrate as electron acceptor. In the presence of either phenazinemethosulphate or N,N,N' N',-tetramethyl-p-phenylenediamine, the oxidation of isoascorbate supported the uptake of both S14CN- and 86Rb+ (in the presence of valinomycin), whereas NADH and succinate oxidation resulted only in S14CN- uptake. These observations show that the preparations contain both right-side-out and inside-out vesicles, and are related to the earlier proposal that the stimulation of an NADH-2,6-dichloroindophenol reductase activity by bee venom is an indicator of the proportion of right-side-out vesicles present. The implications impinge on previous conclusions [Burnell, J. N., John. P. and Whatley, F. R. (1975) Biochem. J. 150, 527-536 and FEBS Lett. 58, 215-218] about the mechanisms of sulphate and phosphate transport in P. denitrificans. 2. The relationship between the protonmotive force (delta p; transmembrane proton electrochemical gradient expressed in mV) and the phosphorylation potential (delta Gp) generated by vesicles from P. denitrificans has been studied as a function of the concentration of an uncoupler of oxidative phosphorylation. With either NADH or succinate as substrate, the uncoupler had a more pronounced effect on delta p than on delta Gp, so that the ratio delta Gp/F x delta p increased within a limited range of values of delta p close to the maximum. delta Gp/F x delta p was, however, approximately constant over the remaining range of delta p that was titrated. A fraction of 'highly coupled' vesicles, separated from the initial preparation by centrifugation through a Ficoll pad, showed similar titration behaviour. This demonstrated that heterogeneity within a vesicle preparation was not responsible for significant distortion of the true relationship between delta p and delta Gp. Values of delta p and delta Gp/F x delta p (H+/ATP) from 143-108 mV and 3.9-4.4, respectively, were determined when NADH was substrate, whereas with succinate, delta p ranged from 123-88 mV and delta Gp/F x delta p (H+/ATP) from 4.5-5.6. The variation in the value of delta Gp/F x delta p, which can be equated with a minimum value for the H+/ATP of the ATP synthase enzyme, is similar to, but less pronounced than, some of the data previously reported for mitochondria. Thus the observations with these bacterial vesicles, which represent an experimentally simpler system than mitochondria, might be taken as further evidence that measurements of the bulk phase delta p might not truly reflect the driving force for ATP synthesis sensed by the ATP synthase enzyme. However, other explanations that would make the data consistent with a chemiosmotic mechanism cannot be eliminated...

Transcriptional fusions between the phage lambda promotor pR and ATP synthase genes, atp, on plasmid pBR322 were constructed in order to study the effects upon growth and physiology of Escherichia coli of induced overproduction of H+-ATPase subunits. Constitutive overproduction of the complete enzyme had earlier been found to result in decreased growth rate and cytological defects. When a 15-fold overproduction of subunit a alone, or together with subunit c, or with all other ATP synthase subunits was suddenly induced, the following effects were observed. Inhibition of growth and protein synthesis within 10 min of induction, which effect was suppressed by N,N'-dicyclohexylcarbodiimide, also when the chromosomal atp genes coding for the Fo subunits a, b and c were deleted. Partial collapse of the membrane potential delta psi at 4-6 min after induction paralleled by inhibition of thiomethylgalactoside and guanosine transport. Respiration and alpha-methylglucoside transport was not affected. The partial collapse of delta psi, and the specific inhibition of proton-driven transport systems is taken to show that the subunit a has--when suddenly overproduced and inserted into the membrane--a protonophoric activity. It is suggested that this protonophoric activity of subunit a is related to the function of this subunit in the Fo sector in H+-ATPases.