Efficient anchoring of RNA polymerase in Escherichia coli during coupled transcription-translation of genes encoding integral inner membrane polypeptides

While it has been known that supercoiling of the DNA template can be induced by transcription, the mechanism and the efficiency of this process in vivo is not fully understood. We report here that transcription of genes encoding 16 S rRNA, a stable RNA species, or cytoplasmic polypeptides leads to very little or no detectable DNA supercoiling even under the optimum conditions in Escherichia coli. This indicates that hydrodynamic drag on the transcription complex (including RNA polymerase, nascent RNA, ribosomes, and nascent polypeptides) is not sufficient to anchor RNA polymerase during coupled transcription-translation. On the other hand, transcription of membrane-associated genes encoding integral inner membrane or exported periplasmic polypeptides leads to apparent DNA supercoiling. Transcription of genes encoding integral inner membrane polypeptides leads to significantly greater anchoring of RNA polymerase than does transcription of genes encoding periplasmic polypeptides. This may reflect differences in the coupling of transcription-translation with membrane association during expression of these two classes of polypeptides. Evidence is further presented to suggest that the anchoring of RNA polymerase is probably achieved through the interaction of nascent polypeptides with the cytoplasmic surface of the inner membrane during coupled transcription-translation. Moreover, transcriptions of a membrane-associated gene can, under certain circumstances, induce topological anchoring of an RNA polymerase transcribing a neighboring gene that ordinarily is not membrane-associated. Finally, the potential biological consequences of our findings are discussed.

Molecular cloning and characterization of acrA and acrE genes of Escherichia coli

The DNA fragment containing the acrA locus of the Escherichia coli chromosome has been cloned by using a complementation test. The nucleotide sequence indicates the presence of two open reading frames (ORFs). Sequence analysis suggests that the first ORF encodes a 397-residue lipoprotein with a 24-amino-acid signal peptide at its N terminus. One inactive allele of acrA from strain N43 was shown to contain an IS2 element inserted into this ORF. Therefore, this ORF was designated acrA. The second downstream ORF is predicted to encode a transmembrane protein of 1,049 amino acids and is named acrE. Genes acrA and acrE are probably located on the same operon, and both of their products are likely to affect drug susceptibilities observed in wild-type cells. The cellular localizations of these polypeptides have been analyzed by making acrA::TnphoA and acrE::TnphoA fusion proteins. Interestingly, AcrA and AcrE share 65 and 77% amino acid identity with two other E. coli polypeptides, EnvC and EnvD, respectively. Drug susceptibilities in one acrA mutant (N43) and one envCD mutant (PM61) have been determined and compared. Finally, the possible functions of these proteins are discussed.

Polymer models for interphase chromosomes

The overall geometry of chromosomes in mammalian cells during interphase is analyzed. On scales larger than approximately 10(5) bp, a chromosome is modeled as a Gaussian polymer subjected to additional forces that confine it to a subvolume of the cell nucleus. An appropriate partial differential equation for the polymer Green's function leads to predictions for the average geometric length between two points on the chromosome. The model reproduces several of the experimental observations: (i) a square root dependence of average geometric distance between two marked chromosome locations on their genomic separation over genomic length scales from approximately 10(5) to approximately 10(6) bp; (ii) an approach of the geometric distance to a maximum value for still larger genomic separations of the two points; (iii) overall chromosome localization in subdomains of the cell nucleus, as suggested by fluorescent labeling of whole chromosomes and by radiobiological evidence. The model is also consistent with known properties of the 30-nm chromatin fiber. It makes a testable prediction: that for two markers a given number of base pairs apart on a given chromosome, the average geometric separation is larger if the configuration is near one end of the chromosome than if it is near the center of the chromosome.

Early evolution of photosynthesis: clues from nitrogenase and chlorophyll iron proteins

Chlorophyll (Chl) is often viewed as having preceded bacteriochlorophyll (BChl) as the primary photoreceptor pigment in early photosynthetic systems because synthesis of Chl requires one fewer enzymatic reduction than does synthesis of BChl. We have conducted statistical DNA sequence analyses of the two reductases involved in Chl and BChl synthesis, protochlorophyllide reductase and chlorin reductase. Both are three-subunit enzymes in which each subunit from one reductase shares significant amino acid identity with a subunit of the other, indicating that the two enzymes are derived from a common three-subunit ancestral reductase. The "chlorophyll iron protein" subunits, encoded by the bchL and bchX genes in the purple bacterium Rhodobacter capsulatus, also share amino acid sequence identity with the nitrogenase iron protein, encoded by nifH. When nitrogenase iron proteins are used as outgroups, the chlorophyll iron protein tree is rooted on the chlorine reductase lineage. This rooting suggests that the last common ancestor of all extant photosynthetic eubacteria contained BChl, not Chl, in its reaction center, and implies that Chl-containing reaction centers were a late invention unique to the cyanobacteria/chloroplast lineage.

Laser-induced protein-DNA cross-links via psoralen furanside monoadducts

We have developed a technique for cross-linking DNA binding proteins to DNA using psoralen furanside monoadducts as photoaffinity probes and a continuous-wave argon ion laser (366 nm) as a light source. Several DNA binding proteins (T7 RNA polymerase, UvrB, single-stranded DNA binding protein of Escherichia coli, T4 gp32, and RecA of E. coli) are shown to cross-link to single-stranded psoralen monoadducted DNA oligos differing in length and sequence. Increasing fluences of laser light on a fixed ratio of DNA/protein resulted in an increase in the yield of cross-links. Titration experiments were carried out to measure the apparent cross-linking constant (KappXL) for T7 RNA polymerase or UvrB to a monoadducted 24 mer DNA. The estimated values for the apparent cross-linking constant were in the range of (2-3) x 10(-7) M for both T7 RNA polymerase and UvrB. The efficiency of cross-linking was investigated as a function of the length of adducted DNA and also as a fraction of the total noncovalent binding of proteins of psoralenated DNAs. The results showed that in the cases of T7 RNA polymerase and UvrB cross-linking was more efficient with short oligos (8 and 19 mers) as compared to longer oligos (50 mer). A tryptic peptide of T7 RNA polymerase that was conjugated to a psoralen furanside monoadducted 12 mer DNA was isolated by high-performance liquid chromatography. Mass spectrometry and amino acid composition of this peptide revealed that it originated from a region between residues 558 and 608 of the primary structure of T7 RNA polymerase. Two other peptides cross-linked to oligos were also purified. Repeated attempts to perform Edman sequencing of the peptide-DNA conjugates failed. Overall evidence indicates that photo-cross-linking of furanside monoadducts occurred at multiple sites on the proteins. We have shown that T7 RNA polymerase molecules in a ternary complex arrested at the furanside monoadduct can be cross-linked to the DNA templates with laser light. Evidence suggests that the arrested polymerase molecules existed in multiple conformations on the DNA template. This method of transcriptional cross-linking offers a new method for preparing highly stable elongation complexes for further studies.

Regulation of carotenoid and bacteriochlorophyll biosynthesis genes and identification of an evolutionarily conserved gene required for bacteriochlorophyll accumulation

The temporal expression of ten clustered genes required for carotenoid (crt) and bacteriochlorophyll (bch) biosynthesis was examined during the transition from aerobic respiration to anaerobiosis requisite for the development of the photosynthetic membrane in the bacterium Rhodobacter capsulatus. Accumulation of crtA, crtC, crtD, crtE, crtF, crtK, bchC and bchD mRNAs increased transiently and coordinately, up to 12-fold following removal of oxygen from the growth medium, paralleling increases in mRNAs encoding pigment-binding polypeptides of the photosynthetic apparatus. The crtB and crtI genes, in contrast, were expressed similarly in the presence or absence of oxygen. The regulation patterns of promoters for the crtA and crtI genes and the bchCXYZ operon were characterized using lacZ transcriptional fusion and qualitatively reflected the corresponding mRNA accumulation patterns. We also report that the bchI gene product, encoded by a DNA sequence previously considered to be a portion of crtA, shares 49% sequence identity with the nuclear-encoded Arabidopsis thaliana Cs chloroplast protein required for normal pigmentation in plants.

The Rhodobacter capsulatus chlorin reductase-encoding locus, bchA, consists of three genes, bchX, bchY, and bchZ

The bchA locus of Rhodobacter capsulatus codes for the chlorin reductase enzyme in the bacteriochlorophyll synthesis pathway. Previous work has suggested that this locus might encompass a single gene. We have sequenced the bchA locus and found it to contain three coding segments, which we designate bchX, bchY, and bchZ. Each coding segment contains its own translational initiation sequence and follows codon utilization patterns consistent with those of previously published R. capsulatus genes. When various regions of the bchA locus and flanking sequences were subcloned into an expression vector and expressed in Escherichia coli, the three coding segments were all expressed as separate peptides. Finally, conservation of amino acid sequences between bchX and a subunit of the protochlorophyllide reductase (bchL, 34% identity) and the nitrogenase Fe protein (nifH, 30 to 37% identity) suggests structural and mechanistic commonalities among all three proteins.

bchFNBH bacteriochlorophyll synthesis genes of Rhodobacter capsulatus and identification of the third subunit of light-independent protochlorophyllide reductase in bacteria and plants

We present the nucleotide and deduced amino acid sequences of four contiguous bacteriochlorophyll synthesis genes from Rhodobacter capsulatus. Three of these genes code for enzymes which catalyze reactions common to the chlorophyll synthesis pathway and therefore are likely to be found in plants and cyanobacteria as well. The pigments accumulated in strains with physically mapped transposon insertion mutations are analyzed by absorbance and fluorescence spectroscopy, allowing us to assign the genes as bchF, bchN, bchB, and bchH, in that order. bchF encodes a bacteriochlorophyll alpha-specific enzyme that adds water across the 2-vinyl group. The other three genes are required for portions of the pathway that are shared with chlorophyll synthesis, and they were expected to be common to both pathways. bchN and bchB are required for protochlorophyllide reduction in the dark (along with bchL), a reaction that has been observed in all major groups of photosynthetic organisms except angiosperms, where only the light-dependent reaction has been clearly established. The purple bacterial and plant enzymes show 35% identity between the amino acids coded by bchN and chlN (gidA) and 49% identity between the amino acids coded by bchL and chlL (frxC). Furthermore, bchB is 33% identical to ORF513 from the Marchantia polymorpha chloroplast. We present arguments in favor of the probable role of ORF513 (chlB) in protochlorophyllide reduction in the dark. The further similarities of all three subunits of protochlorophyllide reductase and the three subunits of chlorin reductase in bacteriochlorophyll synthesis suggest that the two reductase systems are derived from a common ancestor.

Analysis of the promoter and regulatory sequences of an oxygen-regulated bch operon in Rhodobacter capsulatus by site-directed mutagenesis

The biosynthesis of pigments (carotenoids and bacteriochlorophylls) in the photosynthetic bacterium Rhodobacter capsulatus is regulated by the oxygen concentration in the environment. However, the mechanism of this regulation has remained obscure. In this study, transcriptional fusions of the bchCXYZ promoter region to lacZ were used to identify the promoter and regulatory sequences governing transcription of these bacteriochlorophyll biosynthesis genes. The promoter region was identified in vivo by making deletions and site-directed mutations. The 50 bp upstream of the promoter region was shown to be required for the oxygen-dependent transcriptional regulation of bchCXYZ. A previously described palindrome sequence is also likely involved in the regulation. A gel mobility shift assay further defined the interaction of transcription regulators with these DNA sequence elements in vitro and demonstrated that a DNA-protein complex is formed at this promoter region. Since the suggested promoter sequence and the palindrome sequence are found upstream of several other bch and crt operons, these sequences may be responsible for regulating oxygen-dependent pigment biosynthesis at the level of transcription in R. capsulatus. In addition, these cis-acting DNA elements are not found upstream of puh and puf operons, which encode the structural polypeptides of the reaction center and light-harvesting I complexes. This observation supports the model of different regulatory mechanism for the pigment biosynthesis enzymes and structural polypeptides required for the production of the photosynthetic apparatus.

Molecular matchmakers

Molecular matchmakers are a class of proteins that use the energy released from the hydrolysis of adenosine triphosphate to cause a conformational change in one or both components of a DNA binding protein pair to promote formation of a metastable DNA-protein complex. After matchmaking the matchmaker dissociates from the complex, permitting the matched protein to engage in other protein-protein interactions to bring about the effector function. Matchmaking is most commonly used under circumstances that require targeted, high-avidity DNA binding without relying solely on sequence specificity. Molecular matchmaking is an extensively used mechanism in repair, replication, and transcription and most likely in recombination and transposition reactions, too.

DNA repair by eukaryotic nucleotide excision nuclease. Removal of thymine dimer and psoralen monoadduct by HeLa cell-free extract and of thymine dimer by Xenopus laevis oocytes

Using a human cell-free extract, we have recently shown that thymine dimers are removed from DNA in oligonucleotides 27-29 nucleotides in length (Huang, J. C., Svoboda, D. L., Reardon, J. T., and Sancar, A. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 3664-3668). In this study we find that the excision reaction is dependent on ATP, the excised fragments range in length from 27-32 nucleotides, and have 5'-P and 3'-OH termini. We also found that a thymine-psoralen furan side monoadduct is excised from the DNA with a similar incision pattern, indicating that in humans bulky adducts are removed from DNA by the same enzyme system which hydrolyzes mainly the 22-24th and the 5th phosphodiester bonds 5' and 3', respectively, to the lesion. This incision pattern might be common to eukaryotic excision nucleases as thymine dimers were removed from DNA by the same dual incision pattern by Xenopus laevis oocytes.

In vitro expression and activity of lycopene cyclase and beta-carotene hydroxylase from Erwinia herbicola

The cyclisation of lycopene to beta-carotene and the hydroxylation of beta-carotene to zeaxanthin are common enzymatic steps in the biosynthesis of carotenoids in a wide range of bacteria, fungi, and plants. We have individually expressed in E. coli the two genes coding for these enzymatic steps in Erwinia herbicola. The cyclase and hydroxylase enzymes have apparent molecular weights of 43 kDa and 22 kDa, respectively, as determined by SDS-PAGE. Hydroxylase in vitro activity was obtained only in the cytoplasmic fraction. Cyclase also demonstrated enzyme activity in a crude cell-free lysate, although to a lesser extent.

Dynamics of DNA supercoiling by transcription in Escherichia coli

The relative rotation between RNA polymerase and DNA during transcription elongation can lead to supercoiling of the DNA template. However, the variables that influence the efficiency of supercoiling by RNA polymerase in vivo are poorly understood, despite the importance of supercoiling for DNA metabolism. We describe a model system to measure the rate of supercoiling by transcription and to estimate the rates of topoisomerase turnover in Escherichia coli. Transcription in a strain lacking topoisomerase I can lead to optimal supercoiling, wherein nearly one positive and one negative superturn are produced for each 10.4 base pairs transcribed. This rapid efficient supercoiling is observed during transcription of membrane-associated gene products, encoded by tet (the gene for tetracycline resistance) and phoA (the gene for E. coli alkaline phosphatase), when the genes are oppositely oriented. Replacement of tet by cat, the gene from Tn9 encoding resistance to chloramphenicol, whose gene product is soluble in the cytosol, reduces the efficiency of supercoiling by RNA polymerase. In a wild-type topoisomerase background, both gyrase and topoisomerase I are kinetically competent to relieve superturns produced by transcription. These results suggest that the level of DNA supercoiling in vivo is probably determined by topoisomerase activity, not by transcription.

Functional expression of zeaxanthin glucosyltransferase from Erwinia herbicola and a proposed uridine diphosphate binding site

Erwinia herbicola, a nonphotosynthetic bacterium, is yellow colored due to the accumulation of unusually polar carotenoids, primarily mono- and diglucosides of zeaxanthin. We have cloned and expressed the gene for the enzyme that catalyzes the glucosylation of zeaxanthin. The enzyme has an apparent molecular mass of 45 kDa on an SDS/polyacrylamide gel, which is consistent with its calculated molecular mass. In vitro enzymatic activity was demonstrated using UDP-[14C]glucose and zeaxanthin as substrates. The product zeaxanthin diglucoside and its intermediate monoglucoside were identified by thin layer chromatography. The optimum pH and temperature ranges of the enzyme are 7.0-7.5 and 32-37 degrees C, respectively. A hydropathy plot indicates no apparent membrane-spanning regions, and biochemical experiments suggest that the enzyme is weakly membrane-associated. The amino acid sequence derived from the zeaxanthin glucosyltransferase gene shows a small region of high similarity with other glucuronosyl- and glucosyltransferases that use either UDP-activated glucuronic acid or a sugar as one of their substrates. Based on these similarities, we propose that this conserved sequence is part of the UDP binding site.

Active site of (A)BC excinuclease. II. Binding, bending, and catalysis mutants of UvrB reveal a direct role in 3' and an indirect role in 5' incision

UvrB plays a central role in (A)BC excinuclease. To study its role in the incision reactions, conserved His and Asp residues in this subunit were mutagenized. All His and the majority of Asp mutants behaved like wild-type protein in vivo and in vitro. However, three mutants, D337A, D478A, and D510A, either completely or partially abolished UvrB activity. All three mutant proteins associate with UvrA normally but D337A and D510A were unable to bind to DNA specifically. The UvrB-D478A mutant bound to DNA specifically but failed to denature and kink the DNA. However, UvrB-D478A was efficiently loaded onto DNA preincised at the 3' site and promoted near-normal incision by UvrC at the 5' site. We propose that D478 is involved in bending DNA and catalysis of the 3' incision and that the 3' incision precedes the 5' incision. UvrB which is missing the carboxyl-terminal 43 amino acids binds to, and kinks DNA but is unable to make the 3' incision suggesting that it is missing a residue involved in catalysis. This residue was identified to be E639 by site-specific mutagenesis.

The crtE gene in Erwinia herbicola encodes geranylgeranyl diphosphate synthase

A cluster of genes essential for the biosynthesis of carotenoids in Erwinia herbicola has been isolated and characterized [Armstrong, G.A., Alberti, M. & Hearst, J. E. (1990) Proc. Natl. Acad. Sci. USA 87, 9975-9979]. Related gene clusters are found in other carotenoid-producing bacteria. Two of these genes, crtB and crtE, have been assigned to enzymes responsible for conversion of geranylgeranyl diphosphate (GGPP) to prephytoene diphosphate and prephytoene diphosphate to phytoene, respectively. We isolated crtE from the Er. herbicola cluster by PCR amplification and cloned the coding region into the Escherichia coli expression vector pARC306N. Es. coli JM101 was transformed with the expression plasmid, and transformants were assayed for GGPP synthase and phytoene synthase activity. Extracts from JM101/pSM145 accumulated [14C]GGPP when incubated with [14C]isopentenyl diphosphate and farnesyl diphosphate, whereas similar incubations with [3H]GGPP did not yield prephytoene diphosphate or phytoene. Thus, crtE encodes GGPP synthase.

Recent advances in the synthesis and structure determination of site specifically psoralen-modified DNA oligonucleotides

We have developed novel methods for the preparation of multimicromole quantities of extremely pure, uniquely photoadducted psoralen-DNA cross-links, furan-side monoadducted DNA and pyrone-side monoadducts. Psoralen cross-linked and furan-side monoadducted DNA were produced by employing high intensity argon ion and krypton ion lasers as light sources. Pyrone-side monoadducts were prepared by base-catalyzed photoreversal of psoralen cross-links. The various psoralen-adducted DNA oligomers were efficiently purified by high performance liquid chromatography. These methods have permitted us to synthesize 4 mumol each of a self-complementary 8-mer d(GCGTACGC) 4'-(hydroxymethyl)-4,5',8-trimethylpsoralen (HMT) furan-side monoadduct and HMT cross-link. Preliminary nuclear magnetic resonance (NMR) data on the HMT cross-linked 8-mer d(GCGTACGC) have been obtained which confirmed the presence of the diadducted psoralen at the unique 5'TpA3' site. NMR data obtained from the 8-mer furan-side monoadduct revealed that the psoralen molecule is intercalated into the DNA double helix. Preliminary crystals of 8-mer cross-linked DNA molecule have been grown. Conditions for the growth of X-ray diffraction-quality crystals and the further analysis of these crystals are now in progress.

Methods for the large-scale synthesis of psoralen furan-side monoadducts and diadducts

We report methods for the preparation of multimicromole quantities of extremely pure uniquely photo-adducted psoralen-DNA furan-side monoadducts and diadducts (cross-links). The methods use high-intensity krypton and argon ion lasers in the photoreactions and HPLC methods to purify the required oligonucleotides containing the photoadducts. With these methods we have synthesized 2-3 mumol of 8-mer psoralen furan-side monoadduct and diadduct. These methods allow one to generate large amounts of psoralenated DNA oligonucleotides and facilitate their study by NMR and x-ray crystallography.

Post-incision steps of nucleotide excision repair in Escherichia coli. Disassembly of the UvrBC-DNA complex by helicase II and DNA polymerase I

UvrA, UvrB, and UvrC initiate nucleotide excision repair by incising a damaged DNA strand on each side of the damaged nucleotide. This incision reaction is substoichiometric with regard to UvrB and UvrC, suggesting that both proteins remain bound following incision and do not "turn over." The addition of only helicase II to such reaction mixtures turns over UvrC; UvrB turnover requires the addition of helicase II, DNA polymerase I, and deoxynucleoside triphosphates. Column chromatography and psoralen photocross-linking experiments show that following incision, the damaged oligomer remains associated with the undamaged strand, UvrB, and UvrC in a post-incision complex. Helicase II releases the damaged oligomer and UvrC from this complex, making repair synthesis possible; DNase I footprinting experiments show that UvrB remains bound to the resulting gapped DNA until displaced by DNA polymerase I. The specific binding of UvrB to a psoralen adduct in DNA inhibits psoralen-mediated DNA-DNA cross-linking, yet promotes the formation of UrvB-psoralen-DNA cross-links. The discovery of psoralen-UvrB photocross-linking offers the potential of active-site labeling.

Torsional rigidity of positively and negatively supercoiled DNA

Time-correlated single-photon counting of intercalated ethidium bromide was used to measure the torsion constants of positively supercoiled, relaxed, and negatively supercoiled pBR322 DNA, which range in superhelix density from +0.042 to -0.123. DNA behaves as coupled, nonlinear torsional pendulums under superhelical stress, and the anharmonic term in the Hamiltonian is approximately 15 percent for root-mean-square fluctuations in twist at room temperature. At the level of secondary structure, positively supercoiled DNA is significantly more flexible than negatively supercoiled DNA. These results exclude certain models that account for differential binding affinity of proteins to positively and negatively supercoiled DNA.