Investigating autocatalytic gene expression systems through mechanistic modeling

A structured model of gene expression, which incorporates the stochastic behavior of cellular processes, was developed to examine the "all-or-none" phenomenon observed in autocatalytic systems (e.g. the lac operon). Autocatalytic expression systems typically have the genes encoding the inducer transport proteins controlled by internal inducer levels, so that transport of the inducer increases production of the transport protein. The model was able to predict the unique behaviors of autocatalytic expression systems that have been experimentally observed and provided valuable insight into the role of population heterogeneity in these systems. The simulations substantiate the importance of stochastic processes on induction of gene expression in autocatalytic systems. The simulation results show that the all-or-none phenomenon is governed largely by random cellular events, and that population-averaged variations in gene expression are due to changes in the frequency of full gene induction in individual cells rather than to uniform variations in gene expression across the entire population. In addition, the model shows how concentrations of inducer too low to induce expression in uninduced cells can maintain induction in pre-induced cultures. A comparison of induction behaviors from an autocatalytic system and a system having constitutive synthesis of the transport protein showed that transport protein levels must be decoupled from inducer control to achieve homogeneous expression of a gene of interest in all cells of a culture.

Gene-expression tools for the metabolic engineering of bacteria

Recent advances in metabolic engineering have led to new methods for the synthesis of novel molecules, improved production of existing compounds and improved degradation of recalcitrant environmental contaminants. Increasing the flux through an existing pathway and introducing a new pathway into a host organism demand coordinated expression of the genes that encode the enzymes, tight control over gene expression and consistent expression in all cells. Although several gene-expression tools have been developed for the overproduction of proteins, they may not be ideal for pathway redirection. Metabolic engineering requires certain characteristics of gene-expression tools, and some new tools meet these needs.

Polyphosphate kinase of Acinetobacter sp. strain ADP1: purification and characterization of the enzyme and its role during changes in extracellular phosphate levels

Polyphosphate (polyP) is a ubiquitous biopolymer whose function and metabolism are incompletely understood. The polyphosphate kinase (PPK) of Acinetobacter sp. strain ADP1, an organism that accumulates large amounts of polyP, was purified to homogeneity and characterized. This enzyme, which adds the terminal phosphate from ATP to a growing chain of polyP, is a 79-kDa monomer. PPK is sensitive to magnesium concentrations, and optimum activity occurs in the presence of 3 mM MgCl(2). The optimum pH was between pH 7 and 8, and significant reductions in activity occurred at lower pH values. The greatest activity occurred at 40 degrees C. The half-saturation ATP concentration for PPK was 1 mM, and the maximum PPK activity was 28 nmol of polyP monomers per microg of protein per min. PPK was the primary, although not the sole, enzyme responsible for the production of polyP in Acinetobacter sp. strain ADP1. Under low-phosphate (P(i)) conditions, despite strong induction of the ppk gene, there was a decline in net polyP synthesis activity and there were near-zero levels of polyP in Acinetobacter sp. strain ADP1. Once excess phosphate was added to the P(i)-starved culture, both the polyP synthesis activity and the levels of polyP rose sharply. Increases in polyP-degrading activity, which appeared to be mainly due to a polyphosphatase and not to PPK working in reverse, were detected in cultures grown under low-P(i) conditions. This activity declined when phosphate was added.

Ultrasonic flexural-plate-wave sensor for detecting the concentration of settling E. coli W3110 cells

The flexural-plate-wave (FPW) sensor, a type of ultrasonic sensor, can detect changes in E. coli W3110 concentration in solution as the cells settle onto the sensor under the influence of gravity. A model of the sensor's response to cell settling has been developed and is in good agreement with the experimental data. The FPW technique improves on conventional methods for determining cell concentrations; this technique allows for on-line data collection, is nondestructive, and requires only small sample volumes. The FPW sensor has applications as a device to measure cell concentrations and growth rates in industrial fermentors, biofilms, and wastewater treatment facilities.

Effect of polyphosphate metabolism on the Escherichia coli phosphate-starvation response

A previously developed dynamic model of the Escherichia coli Pho regulon was extended to investigate the effect of polyphosphate synthesis and degradation on this control system. Differential equations for ATP and polyphosphate were formulated, and the model was applied to the growth of cells containing the ppk and ppx genes under control of separate, inducible promoters. In agreement with recent experimental observations, the degradation of polyphosphate by PPX during a period of phosphate limitation could repress the phosphate-starvation response. This is attributed to the release of phosphate from the cell into the periplasm, where it can be detected by the external phosphate sensor. A segregated model was then developed to account for differences in K(I), the dissociation constant for the repression complex, among cells of the population. Since K(I) is the key parameter in determining whether the Pho response is induced or repressed at a particular surface phosphate concentration, this permitted the induction of some cells while others remained repressed. The induction profiles resulting from the population-averaged values more closely matched experimental results than did those with the nonsegregated model.

Reductive dechlorination of chlorinated ethene DNAPLs by a culture enriched from contaminated groundwater

A microbial culture enriched from a trichloroethene-contaminated groundwater aquifer reductively dechlorinated trichloroethene (TCE) and tetrachloroethene (PCE) to ethene. Initial PCE dechlorination rate studies indicated a first-order dependence with respect to substrate at low PCE concentrations, and a zero-order dependence at high concentrations. Studies of TCE and vinyl chloride (VC) dechlorination indicated a first-order dependence at all substrate concentrations. VC had little or no effect on the initial rate of TCE dechlorination. With subsaturating concentrations of chlorinated ethenes, nearly stoichiometric amounts of the toxic intermediate vinyl chloride accumulated prior to its dechlorination to ethene. In contrast, under saturating conditions, in which a dense, nonaqueous-phase liquid existed in equilibrium with the aqueous phase, the chlorinated ethene was dechlorinated to ethene, at a rapid rate, with the accumulation of relatively small amounts of chlorinated intermediates.

Library of synthetic 5' secondary structures to manipulate mRNA stability in Escherichia coli

A DNA cassette system has been developed to allow for the convenient introduction of synthetic DNA oligonucleotides between the transcription and translation start sites of a gene in order to examine the effect of 5' hairpin structure and strength on mRNA stabilization. Rationally designed synthetic DNA cassettes were introduced into the 5' untranslated region of a modified lacZ gene to form hairpins at the 5' end of the mRNA. These DNA inserts influenced mRNA half-lives over an order-of-magnitude range, with some groups of predicted structures having half-lives that showed a strong correlation with hairpin strength while half-lives for another group of predicted structures exhibited little or no dependence on this property. These results indicate the importance of 5' hairpin structure and strength in determining stabilization of Escherichia coli mRNA. This synthetic library, as well others generated using the DNA cassette system described here, should prove useful in understanding the mechanisms of mRNA stabilization and in designing structures for recombinant gene expression control.

Effect of Escherichia coli biomass composition on central metabolic fluxes predicted by a stoichiometric model

The amino acid composition of proteins and the fatty acid composition of the cell membranes were measured in Escherichia coli growing exponentially in batch culture on glucose, succinate, glycerol, pyruvate, and acetate, and growing under continuous culture conditions on glucose at dilutions rates equivalent to the growth rates of the batch cultures. Although the fatty acid composition of the membranes did change significantly with carbon source and dilution rate, the amino acid content of proteins did not change significantly under either condition. A previously developed stoichiometric model of metabolism was used to calculate the fluxes through the metabolic reactions and to determine their sensitivity to changes in fatty acid and amino acid composition.

Construction and characterization of F plasmid-based expression vectors

A low-copy expression vector has been constructed from a 9 Kbp region of the Escherichia coli F plasmid containing the oriV and oriS origins of replication. This plasmid carries the beta-lactamase gene (Apr) and the araBAD promoter/araC regulator for arabinose-inducible gene expression. A derivative which carries a lacZ reporter gene was found to be stably maintained for at least 150 generations. A related multi-copy plasmid was stably maintained in arabinose-free medium, but no plasmid-bearing segregants remained after 60 generations when lacZ expression was induced. Induced expression resulted in 27% (multi-copy) and 12% (low-copy) decreases in growth rate. The uninduced levels of beta-galactosidase were 200 units (multi-copy) and 15 units (low-copy).

Optimization of polyphosphate degradation and phosphate secretion using hybrid metabolic pathways and engineered host strains

Polyphosphate degradation and phosphate secretion were optimized in Escherichia coli strains overexpressing the E. coli polyphosphate kinase gene (ppk) and either the E. coli polyphosphatase gene (ppx) or the Saccharomyces cerevisiae polyphosphatase gene (scPPX1) from different inducible promoters on medium- and high-copy plasmids. The use of a host strain without functional ppk or ppx genes on the chromosome yielded the highest levels of polyphosphate, as well as the fastest degradation of polyphosphate when the gene for polyphosphatase was induced. The introduction of a hybrid metabolic pathway consisting of the E. coli ppk gene and the S. cerevisiae polyphosphatase gene resulted in lower polyphosphate concentrations than when using both the ppk and ppx genes from E. coli, and did not significantly improve the degradation rate. It was also found that the rate of polyphosphate degradation was highest when ppx was induced late in growth, most likely due to the high intracellular polyphosphate concentration. The phosphate released from polyphosphate allowed the growth of phosphate-starved cells; excess phosphate was secreted into the medium, leading to a down-regulation of the phosphate-starvation (Pho) response. The production of alkaline phosphatase, an indicator of the Pho response, can be precisely controlled by manipulating the degree of ppx induction. Copyright 1998 John Wiley & Sons, Inc.

mRNA stability and plasmid copy number effects on gene expression from an inducible promoter system

The effects of mRNA stability and plasmid copy number on gene expression in Escherichia coli were evaluated by constructing multicopy (pMB1-based) and low-copy (F-based) plasmids containing an arabinose-inducible promoter system, the lacZ reporter gene, and mRNA-stabilizing 5' hairpin structures. Product formation and cell growth were evaluated under a number of inducer concentrations. The introduction of a 5' hairpin into the untranslated region of the mRNA resulted in significantly higher gene expression from the multicopy plasmids at low inducer concentrations and increased gene expression from the low-copy plasmids across all inducer concentrations investigated. With high inducer concentrations, expression from high-copy plasmids significantly slowed cell growth, whereas expression from the low-copy plasmids had little effect on growth rate. At inducer concentrations between 1 x 10(-4) and 4 x 10(-4)%, the productivity of low-copy plasmids containing the 5'-hairpin was equal to or greater than that from multicopy plasmids. Together, these two gene expression strategies may find important use in metabolic engineering and heterologous gene expression.

Cycle-specific replication of chromosomal and F-plasmid origins

There have been various proposals for the pattern of F-plasmid replication during the division cycle. Here we show that the recent studies of Gordon et al. (Cell 90, 1113-1121, 1997) on the duplication and segregation of green fluorescent protein (GFP) labeled replication origins of the Escherichia coli chromosome and the F plasmid during the division cycle support the proposal that the F plasmid replicates with a cell-cycle-specific (artiocyclic) pattern.

Engineering polyphosphate metabolism in Escherichia coli: implications for bioremediation of inorganic contaminants

Polyphosphate metabolism plays an important role in the bioremediation of phosphate contamination in municipal wastewater, and may play a key role in heavy metal tolerance and bioremediation. However, little is known about the regulation of polyphosphate metabolism in microorganisms and its role in heavy metal toxicity. We have manipulated polyphosphate metabolism in Escherichia coli by overexpressing the genes for polyphosphate kinase (ppk) and for polyphosphatase (ppx) under control of their native promoters and inducible promoters. Overexpression of ppk results in high levels of intracellular polyphosphate, improved phosphate uptake, but no increase in tolerance to heavy metals. Overexpression of both ppk and ppx results in lower levels of intracellular polyphosphate, secretion of phosphate from the cell, and increased tolerance to heavy metals. Metabolic flux analysis indicates that the cell responds to increased flux through the PPK-PPX pathway by altering flux through the TCA cycle.

A dynamic model of the Escherichia coli phosphate-starvation response

A mathematical model of the Escherichia coli Pho regulon was developed to study the induction of the phoA gene by starvation for inorganic phosphate. The model includes phosphate transport, detection of the phosphate concentration at the cell surface, and the signal transduction cascade ultimately leading to the induction of various Pho-controlled genes. Four parameters were manipulated to match the dynamic response of a culture growing with phosphate as the growth-limiting substrate to available experimental data for alkaline phosphatase production and internal phosphate concentration. Steady-state analysis demonstrates that the cascade design of this genetic control system gives rise to a harp transition between the uninduced and induced state for a small change in the external phosphate concentration. Parameter sensitivity indicates that the dissociation constant of the repression complex (which holds PhoR in the inactive form when phosphate is in excess), the rate constants for PhoB and PhoR phosphorylation, and the rate constant for induced transcription of Pho genes have the most influence over the expression of Pho-controlled genes. Changes in the repression complex dissociation constant and the PhoB/PhoR phosphorylation rates alter the sensitivity of the phosphate-starvation response to external phosphate concentration, whereas changes in the transcription rate constant affect the gain of the system. The model also predicts that additional Pho promoter (i.e., for the production of a heterologous protein from the phoA promoter on a plasmid) titrate activator protein PhoB A, such that a lower phosphate concentration is required to initiate expression from a high-copy plasmid than from a single-copy plasmid or the chromosome.

Complete reductive dechlorination of trichloroethene by a groundwater microbial consortium

Bioremediation promises to be an important technique in the removal of trichloroethene (TCE) and tetrachloroethene (PCE) from contaminated waste sites and contaminated groundwater systems. However, the use of aerobic degradation to degrade these compounds is not always possible. Thus, anaerobic degradation is a promising alternative that may be used to remediate these sites. Recently, literature reports indicate complete anaerobic dechlorination of TCE and PCE by microorganisms enriched from wastewater treatment plants. We report here the complete dechlorination of TCE to ethene in anaerobic microcosms by microorganisms enriched from a TCE contaminated groundwater aquifer using glucose as an electron donor. Initial TCE degradation activity occurred after 10 days of incubation and TCE was no longer detected after 20 days of incubation. During the incubation period, the reductive dechlorination products associated with TCE degradation were detected. Ultimately, all of the TCE was converted to ethene. The glucose culture was further enriched and demonstrated increased rates of TCE conversion to ethene. Our results show that organisms isolated from a contaminated groundwater site are capable of completely degrading TCE to ethene at appreciable rates, and indicate the potential of using in situ anaerobic bioremediation to clean up TCE contaminated sites.

Regulation of intracellular toxic metals and other cations by hydrolysis of polyphosphate

Heavy metal tolerance in a number of microorganisms has been correlated with the presence of long-chain polymers of inorganic phosphate called polyphosphate. It has been proposed that the polyphosphate sequesters the metals, thereby reducing their effective intracellular concentration. However, recent evidence indicates that it is not only the amount of stored polyphosphate that is important for heavy metal tolerance but also the ability to degrade polyphosphate to orthophosphate. It is proposed that, in the presence of heavy metals, polyphosphate is degraded to orthophosphate by polyphosphatase and that the metal phosphates are transported out of the cell by the inorganic phosphate transport (PIT) system. Evidence supporting this hypothesis is presented.

Mechanistic modeling of prokaryotic mRNA decay

A mechanistic model of gene expression was developed to test three prevailing and sliding prokaryotic mRNA decay theories: ribosome protection of mRNA from endonucleases, 5' binding and sliding of endonucleases on mRNA, and hybrid 5' binding/ribosome protection. The discrete event simulation incorporates the molecular events that determine both cellular mRNA and protein levels. A Monte Carlo technique was used to approximate the inherent randomness of the molecular processes involved in gene expression. Each of the decay theories was tested for the ability to predict the effects of ribosome loading and translation rate on mRNA stability as well as the observed 5' to 3' directionality of mRNA decay. The modeling results show that the hybrid decay mechanism best predicts the experimentally-observed mRNA decay behaviors. The 5' binding mechanism fails to adequately predict the sensitivity of mRNA stability to changes in translation rate and ribosome loading, while the ribosome protection mechanism does not correctly predict 5' to 3' decay directionality. In addition to discriminating between the three decay theories, the simulations provide insights into hybrid decay mechanism specific details such as RNase binding and cleavage characteristics. Finally, we discuss the application of the current mechanistic model for analysing and predicting expression from more complex genetic systems.

Kinetics of BTEX degradation by a nitrate-reducing mixed culture

A culture enriched from a xylene-contamined, groundwater aquifer degraded m- and p-xylene, toluene, and ethylbenzene completedly within 5 days and o-xylene to 55% of its original concentration under nitrate-reducing conditions. The culture did not degrade benzene. The mixed culture had a maximum growth rate of 0.017 h-1 and a yield of 0.66 g dry cell weight/g toluene consumed. Substrate limitation and then inhibition were observed with increasing concentrations of toluene. The Ks and Ki for toluene were found to be 516 microM and 332 microM, respectively, when fitted to the Andrews model for substrate limitation and inhibition and 410 microM and 492 microM, respectively, when fitted to a multiplicative model for substrate limitation and inhibition. A Monod-type dependence of toluene degradation on the nitrate concentration was observed with a Ks for nitrate of 0.21 mM. Nitrate was not inhibitory to growth or to toluene degradation at concentrations up to 10 mM.

Controlling messenger RNA stability in bacteria: strategies for engineering gene expression

Recent advances in the understanding of prokaryotic gene expression have led scientists to look beyond traditional promoter control for new methods of regulating gene expression. A promising, new technique centers on controlling the stability of messenger RNA. To exploit the potential of mRNA stability for gene expression control, it is important to understand the mechanisms of prokaryotic mRNA decay as well as the cellular factors that can be used to enhance bacterial gene expression through mRNA stabilization. Factors involved in controlling prokaryotic mRNA stability such as nucleases, secondary structures, translation influences, and transcription effects are discussed and analyzed within the context of three prevailing mRNA decay theories. Several strategies for manipulating mRNA stability in genetically-engineered cells are developed from these discussions and presented as a future direction in gene expression control. In the near future, it should be possible to use these strategies to control mRNA stability in such applications as pharmaceutical protein production and metabolic pathway design.

Cadmium removal by a new strain of Pseudomonas aeruginosa in aerobic culture

A fluorescent pseudomonad (strain CW-96-1) isolated from a deep-sea vent sample grew at 30 degrees C under aerobic conditions in an artificial seawater medium and tolerated cadmium concentrations up to 5 mM. After 140 h, strain CW-96-1 removed > 99% of the cadmium from solution. Energy dispersive microanalysis revealed that the cadmium was removed by precipitation on the cell wall; sulfide production was confirmed by growth on Kligler's agar. Based on 16S ribosomal DNA sequencing and fatty acid analysis, the microorganism is closely related to Pseudomonas aeruginosa.

Manipulation of independent synthesis and degradation of polyphosphate in Escherichia coli for investigation of phosphate secretion from the cell

The genes involved in polyphosphate metabolism in Escherichia coli were cloned behind different inducible promoters on separate plasmids. The gene coding for polyphosphate kinase (PPK), the enzyme responsible for polyphosphate synthesis, was placed behind the Ptac promoter. Polyphosphatase, a polyphosphate depolymerase, was similarly expressed by using the arabinose-inducible PBAD promoter. The ability of cells containing these constructs to produce active enzymes only when induced was confirmed by polyphosphate extraction, enzyme assays, and RNA analysis. The inducer concentrations giving optimal expression of each enzyme were determined. Experiments were performed in which ppk was induced early in growth, overproducing PPK and allowing large amounts of polyphosphate to accumulate (80 mumol in phosphate monomer units per g of dry cell weight). The ppx gene was subsequently induced, and polyphosphate was degraded to inorganic phosphate. Approximately half of this polyphosphate was depleted in 210 min. The phosphate released from polyphosphate allowed the growth of phosphate-starved cells and was secreted into the medium, leading to a down-regulation of the phosphate-starvation response. In addition, the steady-state polyphosphate level was precisely controlled by manipulating the degree of ppx induction. The polyphosphate content varied from 98 to 12 mumol in phosphate monomer units per g of dry cell weight as the arabinose concentration was increased from 0 to 0.02% by weight.

Mathematical model of the lac operon: inducer exclusion, catabolite repression, and diauxic growth on glucose and lactose

A mathematical model of the lactose (lac) operon was developed to study diauxic growth on glucose and lactose. The model includes catabolite repression, inducer exclusion, lactose hydrolysis to glucose and galactose, and synthesis and degradation of allolactose. Two models for catabolite repression were tested: (i) cyclic AMP (cAMP) synthesis inversely correlated with the external glucose concentration and (ii) synthesis inversely correlated with the glucose transport rate. No significant differences in the two models were observed. In addition to synthesis, degradation and secretion of cAMP were also included in the model. Two models for the phosphorylation of the glucose produced from lactose hydrolysis were also tested: (i) phosphorylation by intracellular hexokinase and (ii) secretion of glucose and subsequent phosphorylation upon transport back into the cell. The latter model resulted in weak catabolite repression when the glucose produced from lactose was transported out of the cell, whereas the former model showed no catabolite repression during growth on lactose. Parameter sensitivity analysis indicates the importance of key parameters to lac operon expression and cell growth: the lactose and allolactose transformation rates by beta-galactosidase and the glucose concentrations that affect catabolite repression and inducer exclusion. Large values of the allolactose hydrolysis rate resulted in low concentrations of allolactose, low-level expression of the lac operon, and slow growth due to limited import and metabolism of lactose; small values resulted in a high concentration of allolactose, high-level expression of the lac operon, and slow growth due to a limiting concentration of glucose 6-phosphate formed from allolactose. Changes in the rates of all beta-galactosidase-catalyzed reactions showed similar behavior, but had more drastic effects on the growth rate. Changes in the glucose concentration that inhibited lactose transport could extend or contract the diauxic growth period during growth in the presence of glucose and lactose. Moreover, changes in the glucose concentration that affected catabolite repression affected the cAMP levels and lac operon expression, but had a lesser effect on the growth rate.

Genetic manipulation of polyphosphate metabolism affects cadmium tolerance in Escherichia coli

The polyphosphate metabolic pathways in Escherichia coli were genetically manipulated to test the effect of polyphosphate on tolerance to cadmium. A polyphosphate kinase (ppk) and polyphosphatase (ppx) mutant strain produced no polyphosphate, whereas the same strain carrying multiple copies of ppk on a high-copy plasmid produced significant quantities. The doubling times of both strains increased with increasing cadmium concentrations. In contrast, the mutant strain carrying multiple copies of ppk and ppx produced 1/20 of the polyphosphate found in the strain carrying multiple copies of ppk only and showed no significant increase in doubling time over the same cadmium concentration range.

The recA gene and cadmium toxicity in Escherichia coli K12

The influence of the recA gene on cadmium toxicity was studied in Escherichia coli K12 strains. Those strains mutant in the recA gene showed a 1,000-fold loss of viability upon exposure to cadmium, but recovered and started growing approximately 16-20 h following the initial exposure to cadmium. In contrast to previous studies with E. coli B strains, the E. coli K12 strains carrying a functional recA gene showed little or no loss of viability upon exposure to cadmium. The cells also exhibited a significant lag period in which no net growth occurred and then began growing 16-20 h after initial exposure to cadmium. These results indicate the importance of the recA gene to cell survival during exposure to metals, and also support the hypothesis that cadmium causes DNA damage.