Continuous recombinant bacterial fermentations utilizing selective flocculation and recycle.

Selective recycle has successfully been used to maintain an unstable plasmid-bearing bacterial strain as dominant in a continuous reactor, whereas the culture reverts to 100% segregant cells when selective recycle is not used. The plasmid-bearing strain is slower growing and flocculent; however, when the cells lose their plasmid, the resulting segregant cells are nonflocculent and grow at a faster rate due to their decreased metabolic burden. Both types of cells exit a chemostat and enter an inclined settler where the flocculent plasmid-bearing cells are separated from the nonflocculent segregant cells by differential sedimentation. The underflow from the cell separator, which is enriched with plasmid-bearing cells, is recycled back to the chemostat, while the segregant cells are withdrawn off the top of the settler and discarded. The experimental results agree well with selective recycle reactor theory. On the basis of the theory, a criterion is presented that has been shown to successfully predict whether or not a selective recycle reactor can maintain a plasmid-bearing strain.     

An adjustable expression system for controlling growth rate, plasmid maintenance, and culture dynamics

Recombinant bacterial cells express various levels of model product proteins if the genes of interest are regulated by controllable promoters. The level of gene expression influences the growth-rate differential between plasmid-bearing and plasmid-free cells, and thereby affects the culture dynamics of a plasmid-containing cell population. An expression system has been designed in which host Escherichia coli cells contain the pil operon controlled by a tac promoter; these cells are transformed with plasmids that contain the repressor gene, lacl, for the tac promoter, in combination with an expression system for a model protein, chloramphenicol acetyl tranferase (CAT). Experimental and theoretical results show that plasmid-bearing cells can be maintained as dominant in continuous cultures without selective pressure when 12% or less of the cells' total protein is the model product protein, CAT. This is because the segment cells produce pili greatly in excess of normal wild-type levels, and thus have more of a metabolic burden than do the plasmid-bearng cells that overproduce CAT. However, when the level of the plasmid-directed CAT expression is increased above 12% of the cells' total protein, the growth rate of the plasmid-bearing cells decreases to a value lower than that of the segregant cells. Therefore, plasmid-containing cells lose their selective advantage at this expression level, and cannot be maintained as the dominant cell type in a continuous culture unless antibiotic or other positive selection methods are used. By controlling the growth rate differential of this bacterial host/plasmid system, a variety of interesting competitive culture dynamics is investigated. All experimental measurements for continuous cultures are in very good agreement with theory using kinetic parameters determined from independent batch experiments. (c) 1992 John Wiley & Sons, Inc.     

Production of an immunoglobulin gene product by the plasmid expression vector L factor in mouse myeloma cells.

L factor, originally discovered in a subclone of mouse L cells, is a multicopy mammalian plasmid whose structure is related to that of polyoma. When a composite DNA consisting of L factor, pBR, bacterial neo, and an immunoglobulin (kappa) gene was introduced into mouse myeloma cells, the DNA was established as plasmids in the cells without rearrangement or integration into the chromosomes. The plasmid-bearing myeloma cells produced kappa mRNA and the gene product, kappa immunoglobulin, which were apparently derived from the gene located on plasmid L factor. These results suggest that L factor can be used as a plasmid expression vector for studies on gene expression and production of biologically active substances in mammalian cells.     

Modelling the stability of the pBR322 plasmid derivative pBB210 in Escherichia coli TG1 under non-selective and selective conditions

A mathematical model has been developed to describe the stability behaviour of the pBR322 plasmid derivative pBB210 with the β-lactamase gene and the human interferon-α1 gene in Escherichia coli TG1 under non-selective, selective and modified selective conditions in a chemostat. The model was formulated on the basis of experimental investigations. It includes the interaction between β-lactam antibiotics (ampicillin and sulbactam) and cells (with and without plasmids), in particular the correlation between the growth rate of plasmid-free cells and ampicillin concentration in the medium; ampicillin transport into the periplasm of the plasmid-bearing cells; ampicillin degradation in the periplasm by by plasmid-encoded β-lactamase and the inhibition of the latter by sulbactam. The results obtained by the simulation of chemostat cultivations under various conditions and by steady state analyses are closely related to the results of experiments. Under non-selective conditions, the fraction of plasmid-bearing cells was approaching zero. Under selective and modified selective conditions, a coexistence between plasmid-free and plasmid-bearing cells was reached at steady state. Under these conditions, the steady state fraction of plasmid-bearing cells was proportional to the ampicillin concentration in the feed and inversely proportional to the cell concentration in the chemostat. During high-density cultivation, a large amount of ampicillin is necessary to suppress plasmid-free cells. Even small concentrations of the β-lactamase inhibitor sulbactam in the feed increased the steady state fraction of plasmid-bearing cells (from 17.2% to 99.6% at sulbactam-Na concentrations of 0 to 5 mg/l).     

Survival of genetically-engineered and wild-type strains of the yeast Saccharomyces cerevisiae under simulated environmental conditions: a contribution on risk assessment

H. FUJIMURA, Y. SAKUMA AND E. AMANN. 1994. A genetically-engineered strain of Saccharomyces cerevisiae employed for the industrial production of the human coagulation Factor XIIIa (rhFXIIIa) was used for a survival study under simulated environmental conditions. The homologous strain devoid of the recombinant plasmid and the homologous strain bearing the 2 μm-based vector plasmid without the rhFXIIIa-encoding DNA insert were compared. The strains were introduced into natural soil/water suspension, into soil/medium suspension and into waste water. After intervals, samples of cell suspensions were taken and viable cell numbers were determined by plating on antibiotic-containing medium. In addition, a non-radioactive technique involving enhanced chemiluminescence was employed to detect plasmid-bearing yeast cells. The rhFXIIIa expression plasmid showed a high stability during the simulated environmental condition. No differences in survival rates, however, could be detected for the plasmid-bearing and plasmid-less strains under the three conditions tested, suggesting that the presence of plasmid does not confer selective advantages on the survival of the yeast cells. It is concluded that, even after accidental release of the engineered yeast cells into the environment, elimination rates would be comparable to those for non-recombinant yeast strains.     

Synthesis of F pilin.

Transfer of the Escherichia coli fertility plasmid, F, is dependent on expression of F pili. Synthesis of F-pilin subunits is known to involve three F plasmid transfer (tra) region products: traA encodes the 13-kDa precursor protein, TraQ permits this to be processed to the 7-kDa pilin polypeptide, and TraX catalyzes acetylation of the pilin amino terminus. Using cloned tra sequences, we performed a series of pulse-chase experiments to investigate the effect of TraQ and TraX on the fate of the traA product. In TraQ- cells, the traA gene product was found to be very unstable. While traA polypeptides of various sizes were detected early in the chase period, almost all were degraded within 5 min. Rapid traA product degradation was also observed in TraX+ cells, although an increased percentage of these products persisted during the chase. In TraQ+ cells, most of the traA product was processed to the 7-kDa pilin polypeptide within the 1-min pulse period; this product [7(Q)] was not degraded but was increasingly converted to an 8-kDa form [8(Q)] as the chase continued, suggesting that host enzymes can modify the pilin polypeptide. Similar results were observed in TraQ+ TraX+ cells, but the primary 7-kDa product appeared to be N-acetylated pilin (Ac-7). An 8-kDa product (Ac-8) was also detected, but this band did not increase in intensity during the chase. We suggest a pathway in which TraQ prevents the traA product from folding to a readily degradable conformation and assists its entry into the membrane, Leader peptidase I cleaves the traA product signal sequence, and a subset of the pilin polypeptides becomes modified by host enzymes; TraX then acetylates the N terminal of both the modified and unmodified pilin polypeptides.     

Plasmid maintenance in Escherichia coli recombinant cultures is dramatically, steadily, and specifically influenced by features of the encoded proteins

A set of eight closely related plasmid constructs carrying CI857-controlled recombinant genes has been used as a model to study plasmid stability in Escherichia coli, in the absence of antibiotic selection. Plasmid loss rates and relative interdivision times of plasmid-bearing cells and plasmid-free cells have been analyzed throughout prolonged cultures. Whereas the calculated plasmid loss rates are not consistent for a given plasmid and set of conditions, the relative growth fitness of plasmid-bearing cells is highly reproducible. In the absence of gene expression, plasmid maintenance is influenced by the length of the cloned segment, the growth temperature, and the plasmid copy number, but not by the plasmid size. At high, inducing temperatures, the effects of the metabolic burden are eclipsed by the toxicity exhibited by the different proteins produced, which is determined by structural features. Despite the multifactorial nature of the negative pressures acting independently on plasmid-bearing cells, the relative cell fitness in a mixed cell population is very reproducible for a given vector, resulting in a monotonous spread of the plasmid-free cells in recombinant cultures.     

Stability of the pBR322 plasmid derivative pBB210 in Escherichia coli TG1 under non-selective and selective conditions

The stability behaviour of the pBR322 plasmid derivative pBB210 with β-lactamase gene and human interferon-α1 gene in Escherichia coli TG1 was studied in chemostat cultures under non-selective (medium without antibiotics), selective (medium with β-lactam antibiotic ampicillin) and modified selective (medium with ampicillin and the β-lactamase inhibitor sulbactam) conditions. Under non-selective conditions, a behaviour typical of unstable systems was found. Under selective conditions, the behaviour predicted by the models was obtained – the fraction of plasmid-bearing cells in the population approached a constant value which was dependent on the ampicillin concentration in the feeding and on the cell concentration in the chemostat. Under modified selective conditions, the higher the concentration of sulbactam in the medium was, the higher the fraction of plasmid-bearing cells was in steady state conditions.     

The effect of temperature and selective agents on the growth of Yersinia enterocolitica serotype O:3 in pure culture

This study examined the individual and combined effects of the selective agents normally present in Yersinia-selective agar (i.e. cefsulodin, irgasan and novobiocin) on the growth kinetics of plasmid-bearing (P+) and plasmid-cured (P-) Yersinia enterocolitica serotype O:3 at 25 and 37 degrees C. Growth studies were carried out in pure culture, and the data obtained were subjected to linear regression analysis to determine lag phase duration(s) and growth rates of the examined strains. In general, the presence of selective agents increased the duration of the lag phase at 37 degrees C, with longer lag phases noted in all cases in which two or more selective agents were present. Growth rates in CIN broth base (CIN NA) and CIN NA plus commercial supplement (SR 109) (CIN) were faster at 37 than 25 degrees C, but in some cultures of incomplete CIN NA broth with less than three supplements added, growth tended to be faster at 25 than 37 degrees C. Generally, plasmid-bearing strains grew slower than plasmid-cured strains in most media at 37 degrees C due to virulence plasmid expression retarding growth. In some instances at 37 degrees C, it was observed that the growth rates of both plasmid-bearing and plasmid-cured strains were comparable, indicating the influence of added selective agent/s negating any effects associated with virulence plasmid expression. The effects of selective agents, incubation temperature and virulence plasmid carriage on the growth kinetics of Y. enterocolitica are discussed.     

Estimating the instability parameters of plasmid-bearing cells. I. Chemostat culture

What determines the stability of plasmid-bearing cells in natural and laboratory conditions? In order to answer this question in a quantitative manner, we need tools allowing the estimation of parameters governing plasmid loss in different environments. In the present work, we have developed two methods for the estimation of the instability parameters of plasmid-bearing cells growing in chemostat. These instability parameters are: (i) selection coefficient (or cost of the plasmid)alpha and (ii) the probability of plasmid loss at cell division tau(0). We have found that generally selection coefficient alpha changes during elimination of plasmid-bearing cells due to changes in substrate concentration; hence, methods which assume constancy of alpha are intrinsically imprecise. Instead, one can estimate selection coefficient at the beginning and the end of cultivation when the substrate concentration is approximately constant. Applying developed techniques to two sets of experimental data, we have found that (i) the cost of the plasmid pBR322 depended on the dilution rate in chemostat and was higher at low dilutions; (ii) high levels of plasmid gene expression led to a high cost of the plasmid pPHL-7; (iii) the probability of plasmid loss was lower at high levels of plasmid gene expression and independent of the dilution rate. We have also discussed the application of our results to understanding the basic biology of bacterial plasmids.     

Induction of CAT gene expression on a plasmid vector (L factor) by retinoic acid in mouse embryonal carcinoma (F9) cells

We reported previously that composite DNA constructed from a mammalian plasmid (L factor) and foreign gene can be reestablished as a plasmid in mouse embryonal carcinoma (F9) cells after transfection and the plasmid-bearing F9 cells undergo normal in vitro differentiation in response to retinoic acid, an inducer for F9 cell differentiation. We constructed F9 cells bearing plasmidal L factor DNA in which a reporter (chloramphenicol acetyltransferase; CAT) gene was placed under the control of a differentiation-responsive viral (Moloney murine leukemia virus or simian virus 40) enhancer-promoter. When such plasmid-bearing cells were treated with retinoic acid, the CAT gene was inducibly expressed. These results indicate that mammalian gene expression can be studied with the plasmidal expression vector which is structurally dissociated from complex chromosomes.     

Mechanisms of plasmid stable maintenance with special focus on plasmid addiction systems

The stable inheritance of bacterial plasmids is achieved by a number of different mechanisms. Among them are resolution of plasmid oligomers into monomers, active plasmid partitioning into dividing cells and selective killing of plasmid-free segregants. A special focus is given to the last mechanism. It involves a stable toxin and an unstable antidote. The antidotes neutralize their cognate toxins or prevent their synthesis. The different decay rates of the toxins and the antidotes underlie molecular mechanisms of toxin activation in plasmid-free cells. By eliminating of plasmid-free cells from the population of plasmid-bearing ones the toxin-antidote couples therefore act as plasmid addiction systems.     

Plasmid retention and gene expression in suspended and biofilm cultures of recombinant Escherichia coli DH5alpha(pMJR1750)

Differences in plasmid retention and expression are studied in both suspended and biofilm cultures of Escherichia coli DH5alpha(PMJR1750). An alternative mathematical model is proposed which allows the determination of plasmid loss probability in both suspended batch and continuously fed biofilm cultures. In our experiments, the average probability of plasmid loss of E. coli DH5alpha(pMJR1750) is 0.0022 in batch culture in the absence of antibiotic selection pressure and inducer. Under the induction of 0.17 MM IPTG, the maximum growth rate of plasmid-bearing cells in suspended batch culture dropped from 0.45 h(-1) to 0.35 h(-1) and the beta-galactosidase concentration reached an experimental maximum of 0.32. pg/cell 4 hours after the initiation of induction. At both 0.34 and 0.51 mM IPTG, growth rates in batch cultures decreased to 0.16 h(-1), about 36% of that without IPTG, and the beta-galactosidase concentration reached an experimental maximum of 0.47 pg/cell 3 hours after induction.In biofilm cultures, both plasmid-bearing and plasmid-free cells in increase with time reaching a plateau after 96 hours n the absence of both the inducer and any antibiotic selection pressure. Average probability of plasmid loss for biofilm-bound E. coli DH5beta(pMJR1750) population was 0.017 without antibiotic selection. Once the inducer IPTG was added, the concentration of plasmid-bearing cells in biofilm dropped dramatically while plasmid-free cell numbers maintained unaffected. The beta-galactosidase concentration reached a maximum in all biofilm experiments 24 hours after induction; they were 0.08, 0.1, and 0.12 pg/cel under 0.17, 0.34, and 0.51 mM IPTG, respectively. (c) 1993 John Wiley & Sons, Inc.     

A mathematical method for analysing plasmid stability in micro-organisms

A mathematical model describing the instability of plasmids in micro-organisms has been developed. The model is based on the assumption that the overall causes of plasmid instability are described by the segregational instability of the plasmid, R (i.e. the rate at which plasmid-free cells are generated from plasmid-bearing cells), and the growth rate difference, d mu (i.e. the difference in growth rate between plasmid-free and plasmid-bearing cells). A method for determining the values of R and d mu (accompanied by 95% confidence limits) for any plasmid-bearing micro-organism is described. This method is based on the observation that, depending on the plasmid, various exponential patterns of plasmid instability are observed. The stability of Escherichia coli 1B373(pMG169), where d mu much greater than R, and E. coli RV308(pHSG415), where R much greater than d mu, are analysed in order to demonstrate the method.     

Plasmid stabilization of an Escherichia coli culture through cycling.

The problem of plasmid instability of fermentations that involve plasmid-bearing recombinant organisms is dealt with in this work. Previous theoretical work demonstrated that under certain conditions (where plasmid-bearing species are slower in responding to changes in the fermentation environment than the wild species) the washout of the plasmid-bearing species can be prevented. In the sequel, Weber and San showed that cycling the dilution rate can delay the washout of plasmid-bearing species for a plasmid-bearing Escherichia coli culture. This work shows that it is indeed possible to secure the presence of the plasmid-bearing species at all times through appropriate cycling.     

Stability in continuous cultures of recombinant bacteria: A metabolic approach

Continuous-culture population dynamics of recombinant bacteria are predicted with a structured kinetic model. The instantaneous specific growth rates of the plasmid-bearing and plasmidfree cells are explicitly calculated from their metabolic activities. The resultant growth-rate differential (between plasmid-bearing and plasmid-free cells) is dynamic and changes over the course of a fermentation. Further, the growth-rate differential is a function of dilution rate. We present the experimental determination of model constants governing plasmid replication and foreign protein expression for a host/vector systemE. coli RR1 [pBR329]. For a different experimental system, we estimate the increased polypeptide expression from a DNA insert solely from the instability population dynamics. Stability predictions agree quite well with experimental observations from the literature and our lab.     

Stability of a recombinant plasmid carrying a-amylase gene in Bacillus subtilis

Plasmid pSB6 is a streptococcal recombinant plasmid carrying the a-amylase gene of Bacillus amyloliquefaciens and the chloramphenicol resistance gene. The segregational and structural instabilities of this plasmid were examined under non-selective conditions in Bacillus subtilis. These instabilities were modelled according to a kinetic expression derived from the difference in the growth between plasmid-bearing and plasmid-free cells. This plasmid showed slight segregational instability and much higher levels of structural instability under the conditions examined.     

Stability of a recombinant plasmid carrying a-amylase gene in Bacillus subtilis

Plasmid pSB6 is a streptococcal recombinant plasmid carrying the a-amylase gene of Bacillus amyloliquefaciens and the chloramphenicol resistance gene. The segregational and structural instabilities of this plasmid were examined under non-selective conditions in Bacillus subtilis. These instabilities were modelled according to a kinetic expression derived from the difference in the growth between plasmid-bearing and plasmid-free cells. This plasmid showed slight segregational instability and much higher levels of structural instability under the conditions examined.     

Characterization of traX, the F plasmid locus required for acetylation of F-pilin subunits.

Acetylation of F-pilin subunits has previously been shown to depend upon expression of the F plasmid transfer operon gene traX. To assess the requirement for pilin acetylation in conjugative transfer of F, we constructed traX::kan insertion mutations and crossed them onto the transmissible F derivative pOX38. Under standard conditions, the function of traX seemed to be dispensable. Although pilin synthesized by mutant plasmids pOX38-traX482 and pOX38-traX483 was not acetylated, F-pilus production and F-pilus-specific phage infection appeared to be normal and transfer occurred at wild-type frequency. Analysis of labeled products showed that TraX+ plasmids expressed two approximately 24- (TraX1) and 22-kDa (TraX2) polypeptides that localized in the cytoplasmic membranes of cells. No product that was similar in size to the product predicted from the traX open reading frame (27.5 kDa) was detected. Therefore, we used site-directed mutagenesis, stop codon linker insertions, and phoA fusion analysis to investigate traX expression. Both TraX1 and TraX2 appeared to be encoded by the traX open reading frame. Insertion of a stop codon linker into the traX C-terminal coding region led to synthesis of two correspondingly truncated products, and fusions to phoA indicated that only the traX reading frame was translated. Expression was also very dependent on the traX M1 start codon; when this was altered, no protein products were observed. However, pilin acetylation activity was still detectable, indicating that some other in-frame start codon(s) can also be used. All sequences that are essential for activity are contained between traX codons 29 and 225. Sequence analysis indicated that traX mRNA is capable of forming a variety of base-paired structures. We suggest that traX expression is translationally controlled and that F-pilin acetylation activity may be regulated by physiological conditions in cells.     

A single-cell assay of beta-galactosidase in recombinant Escherichia coli using flow cytometry

A flow cytometric method was developed for the assay of beta-galactosidase in single Escherichia coli cells. A new fluorogenic substrate for beta-galactosidase, C(12)FDG, contains a lipophilic group that allows the substrate to penetrate through cell membranes under normal conditions. When the substrate is hydrolyzed by intracellular beta-galactosidase, a green fluorescent product is formed and retained inside the cell. Consequently, the stained beta-galactosidase-positive cells exhibit fluorescence, which is detected by flow cytometry. This new assay was used to analyze the segregational instability caused by a reduction in specific growth rate of the plasmid-bearing cells in the T7 expression system. Induction results in a substantial accumulation of intracellular beta-galactosidase along with a rapid increase in the fraction of plasmid-free cells. Once the cells lose the plasmid, they no longer produce beta-galactosidase, which is reduced by at least half every generation; thus, after staining, the fluorescent, plasmid-bearing cells can be distinguished from the nonfluorescent, plasmid-free cells using flow cytometry. This article describes the feasibility of the flow cytometric assay for single E. coli cells and reports the optimal assay conditions. A direct relationship between beta-galactosidase activity and green fluorescence intensity was found, and the fractions of recombinant cells in batch cultures were analyzed after various levels of induction.