Expression analysis of cloned chromosomal segments of Escherichia coli

The Escherichia coli proteome: past, present, and future prospects

Proteomics has emerged as an indispensable methodology for large-scale protein analysis in functional genomics. The Escherichia coli proteome has been extensively studied and is well defined in terms of biochemical, biological, and biotechnological data. Even before the entire E. coli proteome was fully elucidated, the largest available data set had been integrated to decipher regulatory circuits and metabolic pathways, providing valuable insights into global cellular physiology and the development of metabolic and cellular engineering strategies. With the recent advent of advanced proteomic technologies, the E. coli proteome has been used for the validation of new technologies and methodologies such as sample prefractionation, protein enrichment, two-dimensional gel electrophoresis, protein detection, mass spectrometry (MS), combinatorial assays with n-dimensional chromatographies and MS, and image analysis software. These important technologies will not only provide a great amount of additional information on the E. coli proteome but also synergistically contribute to other proteomic studies. Here, we review the past development and current status of E. coli proteome research in terms of its biological, biotechnological, and methodological significance and suggest future prospects.

Proteomic approaches to Salmonella Pathogenicity Island 2 encoded proteins and the SsrAB regulon

Type III protein secretion is a common virulence determinant in Gram-negative bacteria and the genetic information is often clustered in pathogenicity islands or on virulence plasmids. We have analyzed the type III secretion system encoded by Salmonella Pathogenicity Island 2 (SPI2) that is indispensable for systemic disease of Salmonella enterica serotype Typhimurium (S. Typhimurium) in mice. Since the low abundance of this secretion system restricted direct analysis by proteomic approaches, several putative proteins were expressed as recombinant products and analyzed by two-dimensional electrophoresis. The map obtained for SPI2 encoded proteins was correlated to the expression pattern of S. Typhimurium. The latter was compared to the proteins induced by SsrAB, the two-component system regulating SPI2 gene expression. Our results exemplify that recombinant expression is a complementary tool for analysis of low abundant proteins or membrane proteins. This approach contributes to the characterization of these proteins by subcellular fractionation. Furthermore, we show that pulse labeling was necessary to analyze growth phase regulated SPI2 proteins that might not be otherwise detectable.

Functional dissection of trigger factor and DnaK: interactions with nascent polypeptides and thermally denatured proteins

In Escherichia coli, the ribosome-associated Trigger Factor (TF) cooperates with the DnaK system in the folding of newly synthesized cytosolic polypeptides. Here we investigated the functional relationship of TF and DnaK by comparing various functional properties of both chaperones. First, we analyzed the ability of TF and DnaK to associate with nascent polypeptides and full-length proteins released from the ribosome. Toward this end, we established an E. coli based transcription/translation system containing physiological ratios of TF, DnaK and ribosomes. In this system, TF can be crosslinked to nascent polypeptides of sigma32. No TF crosslink was found to full-length sigma32, which is known to be a DnaK substrate. In contrast, DnaK crosslinked to both nascent and full-length sigma32. DnaK crosslinks critically depended on the type of chemical crosslinker. Crosslinks represent specific substrate-chaperone interactions since they relied on the association of the nascent polypeptides with the substrate binding pocket of DnaK. While DnaK is known to be the major chaperone to prevent protein aggregation under heat shock conditions, we found that TF did not prevent aggregation of thermally unfolded proteins in vitro and was not able to complement the heat-sensitive phenotype of a deltadnaK52 mutant in vivo. These data indicate that TF and DnaK show strong differences in their ability to prevent aggregation of denatured proteins and to associate with native like substrates, but share the ability to associate with nascent polypeptides.

Overproduction of acetyl-CoA carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia coli

Acetyl-CoA carboxylase (ACC) catalyzes the first committed step of the fatty acid synthetic pathway. Although ACC has often been proposed to be a major rate-controlling enzyme of this pathway, no direct tests of this proposal in vivo have been reported. We have tested this proposal in Escherichia coli. The genes encoding the four subunits of E. coli ACC were cloned in a single plasmid under the control of a bacteriophage T7 promoter. Upon induction of gene expression, the four ACC subunits were overproduced in equimolar amounts. Overproduction of the proteins resulted in greatly increased ACC activity with a concomitant increase in the intracellular level of malonyl-CoA. The effects of ACC overexpression on the rate of fatty acid synthesis were examined in the presence of a thioesterase, which provided a metabolic sink for fatty acid overproduction. Under these conditions ACC overproduction resulted in a 6-fold increase in the rate of fatty acid synthesis.

Evidence for an active role of the DnaK chaperone system in the degradation of sigma(32)

Under non-stressed conditions in Escherichia coli, the heat shock transcription factor sigma(32) is rapidly degraded by the AAA protease FtsH. The DnaK chaperone system is also required for the rapid turnover of sigma(32) in the cell. It has been hypothesized that the DnaK chaperone system facilitates the degradation of sigma(32) by sequestering it from RNA polymerase core. This hypothesis predicts that mutant sigma(32) proteins, which are deficient in binding to RNA polymerase core, will be degraded independently of the DnaK chaperone system. We examined the in vivo stability of such mutant sigma(32) proteins. Results indicated that the mutant sigma(32) proteins as similar as authentic sigma(32) were stabilized in DeltadnaK and DeltadnaJ/DeltacbpA cells. The interaction between sigma(32) and DnaK/DnaJ/GrpE was not affected by these mutations. These results strongly suggest that the degradation of sigma(32) requires an unidentified active role of the DnaK chaperone system.

Global analysis of gene expression in cells of the immune system I. Analytical limitations in obtaining sequence information on polypeptides in two-dimensional gel spots

We have outlined various aspects and limitations of the collective analysis of protein species of a cell (lymphocyte). We have indicated research directions that, in to our opinion, deserve more attention. We have evaluated mainly the approach used in our laboratory and we recognize that a bulk of important research on the interface of proteomics and genomics remains to be dealt with. It is of great value that we can proceed in our quest by trial and error. But as much as the human genome initiative was not implemented by trial and error, but by formulating new technological approaches, we hope that our approach can be incorporated in the mainstream of proteomics. We need several integrating research directions, some of which are outlined in this communication, namely the use of ordered cDNA libraries, cell-free expression systems, high density filter hybridization, identification of two-dimensional (2-D) gel spots in terms of their amino acid composition through biosynthetic labeling and identification of restriction sites in the corresponding coding sequences. In the accompanying paper the cDNA ordered library approach will be described in some detail.

Construction and use of low-copy number T7 expression vectors for purification of problem proteins: purification of mycobacterium tuberculosis RmlD and pseudomonas aeruginosa LasI and RhlI proteins, and functional analysis of purified RhlI

Purification of proteins from Escherichia coli under native conditions is often hampered by inclusion-body formation after overexpression from T7 promoter-based expression vectors. This is probably due to the relatively high copy number of the ColE1-based expression vectors. To circumvent these problems, the low-copy-number pViet and pNam expression vectors were constructed. These vectors contain the pSC101 origin of replication and allow the expression of oligohistidine and intein chitin-binding domain fusion proteins, respectively. Since pViet and pNam do not replicate in E. coli B strains, an E. coli K-12 host strain [SA1503(DE3)] was constructed. This strain is defective in the Lon and OmpT proteases and allows IPTG-inducible expression of recombinant proteins from the T7 promoter. The new vectors were successfully tested by purification of three very insoluble proteins (RmlD, LasI and RhlI) under non-denaturing conditions, and all three proteins retained enzymatic activity. The purified hexahistidine (His6)-tagged Pseudomonas aeruginosa RhlI protein was subjected to more detailed analyses, which indicated that (1) only butyryl-acyl carrier protein (ACP) and S-adenosylmethionine (SAM) were required for synthesis of N-butyryl-L-homoserine lactone; (2) when present at physiological concentrations, butyryl-coenzyme A and NADPH were not substrates for RhlI; (3) RhlI was able to synthesize N-hexanoyl-L-homoserine lactone from hexanoyl-ACP and SAM; (4) RhlI was able to direct synthesis of N-butyryl-L-homoserine lactone from crotonyl-ACP in a reaction coupled to purified P. aeruginosa FabI (enoyl-ACP reductase).

Identification and characterization of an Escherichia coli invasion gene locus, ibeB, required for penetration of brain microvascular endothelial cells

Escherichia coli K1 is the most common gram-negative organism causing neonatal meningitis, but the mechanism by which E. coli K1 crosses the blood-brain barrier is incompletely understood. We have previously described the cloning and molecular characterization of a determinant, ibeA (also called ibe10), from the chromosome of an invasive cerebrospinal fluid isolate of E. coli K1 strain RS218 (O18:K1:H7). Here we report the identification of another chromosomal locus, ibeB, which allows RS218 to invade brain microvascular endothelial cells (BMEC). The noninvasive TnphoA mutant 7A-33 exhibited <1% the invasive ability of the parent strain in vitro in BMEC and was significantly less invasive in the central nervous system in the newborn rat model of hematogenous E. coli meningitis than the parent strain. The TnphoA insert with flanking sequences was cloned and sequenced. A 1,383-nucleotide open reading frame (ORF) encoding a 50-kDa protein was identified and termed ibeB. This ORF was found to be 97% identical to a gene encoding a 50-kDa hypothetical protein (p77211) and located in the 13-min region of the E. coli K-12 genome. However, no homology was observed between ibeB and other known invasion genes when DNA and protein databases in GenBank were searched. Like the TnphoA insertion mutant 7A-33, an isogenic ibeB deletion mutant (IB7D5) was unable to invade BMEC. A 7. 0-kb locus containing ibeB was isolated from a LambdaGEM-12 genomic library of E. coli RS218 and subcloned into a pBluescript KS vector (pKS7-7B). pKS7-7B was capable of completely restoring the BMEC invasion of the noninvasive TnphoA mutant 7A-33 and the ibeB deletion mutant IB7D5 to the level of the parent strain. More importantly, the ibeB deletion mutant IB7D5 was fully complemented by pFN476 carrying the ibeB ORF (pFN7C), indicating that ibeB is required for E. coli K1 invasion of BMEC. Taken together, these findings indicate that several E. coli determinants, including ibeA and ibeB, contribute to crossing of the blood-brain barrier.

Application of two-dimensional protein gels in biotechnology

The optimal use of biological systems for technologically developed products will not be achieved until biological systems are completely defined in biochemical terms. Two-dimensional polyacrylamide gel electrophoresis, 2-D gels, are contributing to this goal. These gels separate complex mixtures of proteins into individual polypeptide species. The ultimate use of 2-D gels is the construction of cellular 2-D gel databases which identify the proteins on the gels and catalog their responses to different environmental conditions. In addition to these global analyses, many applications for 2-D gels in basic, applied and clinical research have been shown.

The arbZ gene from Lactobacillus delbrueckii subsp. lactis confers to Escherichia coli the ability to utilize the beta-glucoside arbutin

From a genomic library of the industrially used strain Lactobacillus delbrueckii subsp. lactis DSM7290, a gene designated arbZ (869bp; encoding a 33.5kDa protein) was isolated by screening E. coli transformants for the ability to utilize the beta-glucoside arbutin. Out of 9000 transformants nine were able to ferment arbutin, whereas no utilization of the beta-glucosides salicin, esculin or cellobiose could be detected. Overexpression of arbZ using the T7-polymerase-T7-promoter-system resulted in the formation of insoluble, catalytically inactive protein aggregates (inclusion bodies). Accordingly, overexpression was not accompanied by an increase in ArbZ activity. Induction of arbZ controlled by the lac promoter under conditions that reduce protein aggregation resulted in a 12-fold increase in arbutin hydrolyzing activity of intact cells and a 13-fold increase in phospho-beta-glycosidase activity in cell-free extracts of the respective transformants. Nucleotide sequence analysis revealed a second gene upstream of arbZ that was designated arbX (830bp). ArbX (32.6kDa) shared similarity with several glycosyltransferases involved in the biosynthesis of lipopolysaccharides in Gram-negative bacteria. In Lb. delbrueckii subsp. lactis DSM7290 two transcripts, one covering arbX together with arbZ and one covering arbZ alone were detected by Northern blot analysis.

Ordered cloned DNA map of the genome of Vibrio cholerae 569B and localization of genetic markers

By using a low-resolution macrorestriction map as the foundation (R. Majumder et al., J. Bacteriol. 176:1105-1112, 1996), an ordered cloned DNA map of the 3.2-Mb chromosome of the hypertoxinogenic strain 569B of Vibrio cholerae has been constructed. A cosmid library the size of about 4,000 clones containing more than 120 Mb of V. cholerae genomic DNA (40-genome equivalent) was generated. By combining landmark analysis and chromosome walking, the cosmid clones were assembled into 13 contigs covering about 90% of the V. cholerae genome. A total of 92 cosmid clones were assigned to the genome and to regions defined by NotI, SfiI, and CeuI macrorestriction maps. Twenty-seven cloned genes, 9 rrn operons, and 10 copies of a repetitive DNA sequence (IS1004) have been positioned on the ordered cloned DNA map.

Escherichia coli proteome analysis using the gene-protein database

The gene-protein database of Escherichia coli is a collection of data, largely generated from the separation of complex mixtures of cellular proteins on two-dimensional (2-D) polyacrylamide gel electrophoresis. The database currently contains about 1600 protein spots. The data are comprised of both identification information for many of these proteins and data on how the level or synthesis rates of proteins vary under different growth conditions. Three projects are underway to further elucidate the E. coli proteome including a project to localize on 2-D gels all of the open reading framed encoded by the E. coli chromosome, a project to determine the condition(s) under which each open reading frame is expressed and a project to determine the abundance and location of each protein in the cell. Applications for proteome databases for cell modeling are discussed and examples of applications in therapeutic drug discovery are given.

Lactobacillus delbrueckii subsp. lactis DSM7290 pepG gene encodes a novel cysteine aminopeptidase

A number of Escherichia coli clones were isolated from a Lactobacillus delbrueckii subsp. lactis gene library capable of hydrolysing the chromogenic substrate Gly-Ala-beta-naphthylamide (Gly-Ala-beta NA). Some of the recombinant plasmids carried by these clones have been shown to encode the cysteine aminopeptidase gene pepC. Nucleotide sequence analyses of the plasmid inserts of the remaining clones resulted in the identification of two adjacent ORFs encoding proteins exhibiting a high degree of similarity between themselves (72.6%) and with PepC. One gene, designated pepG, was overexpressed in E. coli and the crude extracts obtained were shown to be peptidolytically active both against chromogenic substrates and peptides, and in a Salmonella typhimurium growth test. PepC and PepG activities were compared using chromogenic beta NA and p-nitroanilide substrates and leucine or proline-containing peptides were applied in growth experiments of recombinant Sal. typhimurium. The results indicate that the enzymes, although structurally related, have different substrate preferences. No enzyme activity could be ascribed to the second ORF (orfW), despite the production of a visible protein using a T7 RNA polymerase system. Primer extension analysis, using mRNA isolated from Lb. delbrueckii subsp. lactis DSM7290 did establish that orfW was transcribed.

Global analysis of proteins synthesized during phosphorus restriction in Escherichia coli

The pattern of proteins synthesized in Escherichia coli during steady-state growth in media with ample inorganic phosphate (Pi), upon limitation for Pi (without an alternative phosphorous compound), and during steady-state growth in media containing phosphonate (PHN) as the sole P source was examined by two-dimensional gel electrophoresis. Of 816 proteins monitored in these experiments, all those with differential synthesis rates greater than 2.0 or less than 0.5 upon phosphate limitation (P limitation) or during growth on PHN compared with their rates in the cultures with Pi were classified as belonging to the PL or PHN stimulon, respectively. The PL stimulon included 413 proteins, 208 showing induced synthesis and 205 showing repressed synthesis. The PHN stimulon was smaller: it included 257 proteins; 227 showed induced synthesis and 30 showed repressed synthesis. The overlap of the two stimulons included 137 proteins: most (118) were ones showing induced synthesis. The promoter regions of genes for several of the proteins with induced or repressed synthesis contained sequences which resembled the consensus sequence for PhoB binding. The aggregate mass of proteins responding to P limitation or growth on PHN was 30 to 40% of the cells' total mass. By comparing the proteins responding to P limitation with those responding to growth on PHN, one can speculate which proteins are likely involved in adapting cells to new P sources or in preparing cells to survive stationary phase.

Sequencing and analysis of bacterial genomes

The complete sequences of two small bacterial genomes have recently become available, and those of several more species should follow within the next two years. Sequence comparisons show that the most bacterial proteins are highly conserved in evolution, allowing predictions to be made about the functions of most products of an uncharacterized genome. Bacterial genomes differ vastly in their gene repertoires. Although genes for components of the translation and transcription machinery, and for molecular chaperones, are typically maintained, many regulatory and metabolic systems are absent in bacteria with small genomes. Mycoplasma genitalium, with the smallest known genome of any cellular life form, lacks virtually all known regulatory genes, and its gene expression may be regulated differently than in other bacteria. Genome organization is evolutionarily labile: extensive gene shuffling leaves only very few conserved gene arrays in distantly related bacteria.

CCAAT/enhancer-binding protein beta (C/EBP beta) binds and activates while hepatocyte nuclear factor-4 (HNF-4) does not bind but represses the liver-type arginase promoter

In an attempt to elucidate the mechanism governing liver-specific transcription of the arginase gene, we previously detected two protein-binding sites designated footprint areas A and B at positions around--90 and --55 bp, respectively, relative to the transcription start site of the rat arginase gene. Based on the finding that area A was bound by a liver-selective factor(s) related to CCAAT/enhancer-binding protein (C/EBP), we performed cotransfection assay and showed that C/EBP family members and a related factor, albumin D-element-binding protein (DBP) stimulate transcription from the arginase promoter. In addition to area A, a recombinant C/EBP beta protein bound to area B, which appeared to be primarily responsible for activation by C/EBPs. We unexpectedly found that the arginase promoter activity stimulated by C/EBPs and DBP was repressed by another liver-enriched transcription factor, hepatocyte nuclear factor-4 (HNF-4). Analysis of chimeras formed between the arginase promoter and the herpes simplex virus thymidine kinase promoter allowed us to delimit the negative HNF-4-responsive element into the region overlapping with footprint area B. However, no apparent binding of HNF-4 was observed in this negative element. We speculate that HNF-4 is involved in fine regulation of the arginase gene in the liver or shutdown of the gene in nonhepatic tissues without direct binding to the promoter region.

Cloning and characterization of brnQ, a gene encoding a low-affinity, branched-chain amino acid carrier in Lactobacillus delbrückii subsp. lactis DSM7290

A gene (brnQ), encoding a carrier for branched-chain amino acids in Lactobacillus delbrückii subsp. lactis DSM7290 was cloned in the low-copy-number vector pLG339 by complementation of a transport-deficient Escherichia coli strain. The plasmid carrying the cloned gene restored growth of an E. coli strain mutated in 4 different branched-chain amino acid transport genes at low concentrations of isoleucine, and increased its sensitivity to valine. Transport assays showed that leucine, isoleucine and valine are transported by this carrier and that transport is driven by the proton motive force. Nucleotide sequence analysis revealed an open reading frame of 1338 bp encoding a hydrophobic protein of 446 amino acids with a calculated molecular mass of 47864 Daltons. The start site of brnQ transcription was determined by primer extension analysis using mRNA from Lactobacillus delbrückii subsp. lactis DSM7290. The hydropathy profile suggests the existence of at least 12 hydrophobic domains that probably form membrane-associated alpha-helices. Comparisons of the nucleotide sequence of brnQ from Lactobacillus delbrückii subsp. lactis DSM7290, the amino acid sequence of its product and the topology of the hydrophobic domains with those of the respective carrier genes and proteins of Salmonella typhimurium and Pseudomonas aeruginosa revealed extensive homology.

Low-copy-number T7 vectors for selective gene expression and efficient protein overproduction in Escherichia coli

A set of low-copy-number vectors (pPD) has been constructed that permit selective gene expression and high-level protein overproduction in Escherichia coli, based on the bacteriophage T7 RNA polymerase/T7 promoter system. These plasmids carry a chloramphenicol resistance gene (cat) as a selective marker and an extended multiple cloning site for convenient gene cloning. Their replication is mediated by ori sequences derived from the low-copy-number vector pSC101. The efficient T7 gene 10 promoter present on these vectors allows selective and high-level transcription of cloned genes carrying their own translational initiation signals. In addition, low-copy-number T7 vectors were constructed that permit expression of genes lacking their own transcription and translation initiation elements by providing a ribosome binding site, an ATG start codon and a multiple cloning site devised for the cloning in all three reading frames. The pPD expression vectors were used to achieve high-level overproduction of the E. coli integral outer membrane protein Tsx, and the cytoplasmic enzymes beta-galactosidase (beta Gal) and UTP:alpha-D-glucose-1-phosphate uridylyltransferase (GalU). The characteristics of these low-copy-number T7 expression vectors should prove very useful for the cloning and high-level overexpression of genes whose gene products are deleterious to the E. coli host.