Improving CRISPR/Cas9-mediated genome editing efficiency in Yarrowia lipolytica using direct tRNA-sgRNA fusions

Yarrowia lipolytica is an important oleaginous yeast currently used in the production of specialty chemicals and has a great potential for further applications in lipid biotechnology. Harnessing the full potential of Y. lipolytica is, however, limited by its inherent recalcitrance to genetic manipulation. In contrast to Saccharomyces cerevisiae, Y. lipolytica is poor in homology-mediated DNA repair and thus in homologous recombination, which limits site-specific gene editing in this yeast. Recently developed CRISPR/Cas9-based methods using tRNA-sgRNA fusions succeeded in editing some genomic loci in Y. lipolytica. Nonetheless, the majority of other tested loci either failed editing or editing was achieved but at very low efficiency using these methods. Using tools of secondary RNA structure prediction, we were able to improve the design of the tRNA-sgRNA fusions used for the expression of single guide RNA (sgRNA) in such methods. This resulted in high efficiency CRISPR/cas9 gene editing at chromosomal loci that failed gene editing or were edited at very low efficiencies with previous methods. In addition, we characterized the gene editing performance of our newly designed tRNA-sgRNA fusions for both chromosomal gene integration and deletion. As such, this study presents an efficient CRISPR/Cas9-mediated gene-editing tool for efficient genetic engineering of Yarrowia lipolytica.

Eliciting the impacts of cellular noise on metabolic trade-offs by quantitative mass imaging

Optimal metabolic trade-offs between growth and productivity are key constraints in strain optimization by metabolic engineering; however, how cellular noise impacts these trade-offs and drives the emergence of subpopulations with distinct resource allocation strategies, remains largely unknown. Here, we introduce a single-cell strategy for quantifying the trade-offs between triacylglycerol production and growth in the oleaginous microorganism Yarrowia lipolytica. The strategy relies on high-throughput quantitative-phase imaging and, enabled by nanoscale secondary ion mass spectrometry analyses and dedicated image processing, allows us to image how resources are partitioned between growth and productivity. Enhanced precision over population-averaging biotechnologies and conventional microscopy demonstrates how cellular noise impacts growth and productivity differently. As such, subpopulations with distinct metabolic trade-offs emerge, with notable impacts on strain performance and robustness. By quantifying the self-degradation of cytosolic macromolecules under nutrient-limiting conditions, we discover the cell-to-cell heterogeneity in protein and fatty-acid recycling, unmasking a potential bet-hedging strategy under starvation.

Origins of Cell-to-Cell Bioprocessing Diversity and Implications of the Extracellular Environment Revealed at the Single-Cell Level

Bioprocess limitations imposed by microbial cell-to-cell phenotypic diversity remain poorly understood. To address this, we investigated the origins of such culture diversity during lipid production and assessed the impact of the fermentation microenvironment. We measured the single-cell lipid production dynamics in a time-invariant microfluidic environment and discovered that production is not monotonic, but rather sporadic with time. To characterize this, we introduce bioprocessing noise and identify its epigenetic origins. We linked such intracellular production fluctuations with cell-to-cell productivity diversity in culture. This unmasked the phenotypic diversity amplification by the culture microenvironment, a critical parameter in strain engineering as well as metabolic disease treatment.

Diffusion coefficients of glucose and ethanol in cell-free and cell-occupied calcium alginate membranes

The diffusivities of glucose and ethanol in cell-free and cell-occupied membranes of calcium alginate were measured in a diffusion cell. The lag time analysis was used. Diffusivities decreased with increasing alginate concentration and were comparable with those in water for a 2% alginate membrane. Glucose and ethanol concentrations had no effect on the respective diffusion coefficients. The ratio of ethanol diffusivity to glucose diffusivity in 2 and 4% alginate agreed closely with the inverse ratio of the hydrodynamic raii for the two molecules in water, indicating that the hydrodynamic theory of diffusion in liquids may be applicable to diffusion in dilute alginate gels. Also, the presence of 20% dead yeast cells had no effect on the diffusivities. The data reported can be used to study reaction and diffusion in immobilized cell reactors and cell physiology under immobilized conditions.

Studies on on-line bioreactor identification. IV. Utilization of pH measurements for product estimation

Relationships between the total rate of biomass growth and the rate of ammonia addition to a fermentor for pH control are presented. These equations make use of the concept of reaction invariants and provide the additional information needed for bioreactor identification. They are especially useful when the RQ measurement is not sufficient for this purpose, such as when sensitivities arise with the measured values of the respiratory quotient or when fermentation products are formed. The cases of batch, fed-batch and continuous fermentations, forming products with or without acidic/basic properties are considered. The derived relationships were successfully tested with nonbiological acid-base continuous flow reaction systems and subsequently applied to the identification of the continuous yeast fermentation of glucose to ethanol. Results of these experimental studies are also presented.

Metabolic flux distributions in Corynebacterium glutamicum during growth and lysine overproduction

The two main contributions of this are the solidification of Corynebacterium glutamicum biochemistry guided by bioreaction network analysis, and the determination of bansal metabolic flux distributions during growth and lysine synthesis. Employed methodology makes use of stoichiometrically based mass balances to determine flux distributions in the C. glutamicum metabolic network. Presented are a brief description of the methodology, a through literature review of glutamic acid bacteria biochemistry, and specific results obtained through a combination of fermentation studies and analysis-directed intracellular assays. The latter include the findings of the lack of activity of glyoxylate shunt, and that phosphoenolpyruvate carboxylase (PPC) is the only anaplerotic reaction expressed in C. glutamicum cultivated on glucose minimal media. Network simplifications afforded by the above findings facilitated the determination of metabolic flux distributions under a variety of culture conditions and led to the following conclusions. Both the pentose phosphate pathway and PPC support fluxes during growth and lysine overproduction branch point does not appear to limit lysine synthesis.

Computer-aided synthesis of biochemical pathways

The synthesis of biochemical pathways satisfying stoichiometric constraints is discussed. Stoichiometric constraints arise primarily from designating compounds as required or allowed reactants, and required or allowed products of the pathways; they also arise from similar restrictions on intermediate metabolites and bioreactions participating in the pathways. An algorithm for the complete and correct solution of the problem is presented; the algorithm satisfies each constraint by recursively transforming a base-set of pathways. The algorithm is applied to the problem of lysine synthesis from glucose and ammonia. In addition to the established synthesis routes, the algorithm constructs several alternative pathways that bypass key enzymes, such as malate dehydrogenase and pyruvate dehydrogenase. Apart from the construction of pathways with desired characteristics, the systematic synthesis of pathways can also uncover fundamental constraints in a particular problem, by demonstrating that no pathways exist to meet certain sets of specifications. In the case of lysine, the algorithm shows that oxaloacetate is a necessary intermediate in all pathways leading to lysine from glucose, and that the yield of lysine over glucose cannot exceed 67% in the absence of enzymatic recovery of carbon dioxide.

A model of secretory protein trafficking in recombinant AtT-20 cells

After their synthesis, secretory proteins in animal cells undergo a series of transport and processing steps before they are secreted. The amount and quality of protein obtained in culture medium depends on the rates of these intracellular steps. We present a model of recombinant protein trafficking in mouse pituitary AtT-20 cells based on currently available biological knowledge, plausible hypotheses, and quantitative secretion results, and we use it to simulate the dynamics of basal and induced secretion and to predict the dynamics of intracellular trafficking events. Besides the endoplasmic reticulum and Golgi, the model recognizes a conversion compartment (CC) where final processing of protein occurs, a storage compartment from which protein is secreted only in the presence of secretion stimulus, and constitutive and pseudoregulated (PR) pathways of secretion. The model further assumes that the protein flux is split between CC and PR and that the storage compartment exerts a negative feedback on protein flux through CC. The model predictions are compared with experimental results on secretion of human growth hormone (hGH) and insulin related peptides and on accumulation of hGH upon removal of secretion stimulus. The model is in agreement with data when either of two hypotheses is implemented: (a) cells always exhibit a high sorting efficiency at the trans-Golgi, but CC has the capacity to process only a fraction of the protein flux leaving the Golgi compartment; (b) the processing capacity of CC never becomes saturated, but significant missorting at the trans-Golgi occurs; in the case, the protein flux toward the plasma membrane becomes split both at the trans-Golgi cisternae and between CC and PR. The usefulness of the type of model considered in providing a quantitative description of intracellular events and in designing new experiments is discussed.

Application of macroscopic balances to the identification of gross measurement errors

A systematic method is presented which is capable of both detecting the presence of grossly biased measurement errors and locating the source of these errors in a bioreactor through statistical hypothesis testing. Equality constraints derived from material and energy balances are employed for the detection of data inconsistencies and for the subsequent identification of the suspect measurements by a process of data analysis and rectification. Maximum likelihood techniques are applied to the estimation of the states and parameters of the bioreactor after the suspect measurements have been eliminated. The level of significance is specified by the experimenter while the measurments are assumed to be randomly, normally distributed with zero mean and known variances. Two different approaches of data analysis, batchwise and sequential, that lead to a consistent set of adjustments on the experimental values, are discussed. Several examples based on the fermentation data taken from literature sources are presented to demonstrate the utility of the proposed method, and one set of data is solved numerically to illustrate the computational aspect of the algorithm.

Effect of pH oscillations on a competing mixed culture

Two microorganisms, E. coli and S. cerevisiae, competing for glucose were maintained in a stable cycle of coexistence by alternating the growth advantage between the two organisms by oscillating the pH in a Chemostat. Pure culture experiments found S. cerevisiae to be insensitive to pH between 5 and 4.3 with a maximum specific growth rate (micro(max)) of 0.4/hr; while micro(max) of E. coli decreased from 0.6 h(-1) at pH 5 to 0.1 h(-1) at pH 4.3. Steady-state and cross-inoculation chemostat runs at a dilution rate of 0.17 h(-1) confirmed the expectation that the mixed culture system is unstable at constant pH with E. coli dominating at pH 5 and S. cerevisiae dominating at pH 4.3. Three pH oscillation experiments were performed at D =0.17 h(-1) with 1 g per liter glucose feed. The 16 h/16 h cycle was stable for six periods with a stable alternating cycle of E. coli and S. cerevisiae being quickly established. A 18 h pH 5/14 h pH 4.3 cycle was found to be stable with smaller yeast concentrations. A 6 h/6 h cycle was found unstable with yeast washout. Simulation results were compared with these runs and were used to predict the onset of instability. Oscillations of pH can force stable persistence of a competing mixed culture that is otherwise unstable. Thus, varying conditions are experimentally demonstrated to be one explanation for competitive coexistence.

Genes and related proteins involved in amelogenesis imperfecta

Dental enamel formation is a remarkable example of a biomineralization process. The exact mechanisms involved in this process remain partly obscure. Some of the genes encoding specific enamel proteins have been indicated as candidate genes for amelogenesis imperfecta. Mutational analyses within studied families have supported this hypothesis. Mutations in the amelogenin gene (AMELX) cause X-linked amelogenesis imperfecta, while mutations in the enamelin gene (ENAM) cause autosomal-inherited forms of amelogenesis imperfecta. Recent reports involve kallikrein-4 (KLK4), MMP-20, and DLX3 genes in the etiologies of some cases. This paper focuses mainly on the candidate genes involved in amelogenesis imperfecta and the proteins derived from them, and reviews current knowledge on their structure, localization within the tissue, and correlation with the various types of this disorder.

Identifying gene targets for the metabolic engineering of lycopene biosynthesis in Escherichia coli

The identification of genetic targets that are effective in bringing about a desired phenotype change is still an open problem. While random gene knockouts have yielded improved strains in certain cases, it is also important to seek the guidance of cell-wide stoichiometric constraints in identifying promising gene knockout targets. To investigate these issues, we undertook a genome-wide stoichiometric flux balance analysis as an aid in discovering putative genes impacting network properties and cellular phenotype. Specifically, we calculated metabolic fluxes such as to optimize growth and then scanned the genome for single and multiple gene knockouts that yield improved product yield while maintaining acceptable overall growth rate. For the particular case of lycopene biosynthesis in Escherichia coli, we identified such targets that we subsequently tested experimentally by constructing the corresponding single, double and triple gene knockouts. While such strains are suggested (by the stoichiometric calculations) to increase precursor availability, this beneficial effect may be further impacted by kinetic and regulatory effects not captured by the stoichiometric model. For the case of lycopene biosynthesis, the so identified knockout targets yielded a triple knockout construct that exhibited a nearly 40% increase over an engineered, high producing parental strain.

Metabolic flux control analysis of branch points: an improved approach to obtain flux control coefficients from large perturbation data

An overview of published approaches for the metabolic flux control analysis of branch points revealed that often not all fundamental constraints on the flux control coefficients have been taken into account. This has led to contradictory statements in literature on the minimum number of large perturbation experiments required to estimate the complete set of flux control coefficients C(J) for a metabolic branch point. An improved calculation procedure, based on approximate Lin-log reaction kinetics, is proposed, providing explicit analytical solutions of steady state fluxes and metabolite concentrations as a function of large changes in enzyme levels. The obtained solutions allow direct calculation of elasticity ratios from experimental data and subsequently all C(J)-values from the unique relation between elasticity ratio's and flux control coefficients. This procedure ensures that the obtained C(J)-values satisfy all fundamental constraints. From these it follows that for a three enzyme branch point only one characterised or two uncharacterised large flux perturbations are sufficient to obtain all C(J)- values. The improved calculation procedure is illustrated with four experimental cases.

Metabolic flux redistribution in Corynebacterium glutamicum in response to osmotic stress

Osmotic stress constitutes a major bacterial stress factor in the soil and during industrial fermentation. In this paper, we quantified the metabolic response, in terms of metabolic flux redistribution, of a lysine-overproducing strain of Corynebacterium glutamicum grown under continuous culture, to gradually increasing osmolality. Oxygen and carbon dioxide evolution rates, and the changes in concentration of extracellular, as well as intracellular, metabolites were measured throughout the osmotic gradient. The metabolic fluxes were estimated from these measurements and from the mass balance constraints at each metabolite-node of the assumed metabolic reaction network. Our results show that formation rates of compatible solutes--trehalose first and proline at a later stage of the gradient--increased with osmotic stress to equilibrate the external osmotic pressure. Estimated flux distributions indicate that the observed increase in the glucose specific uptake rate with osmotic stress is channeled through the main energy generating pathways-- glycolysis and the tricarboxylic acid cycle--while the flux through the pentose phosphate pathway remains constant throughout the gradient. This results in a significant increase in the net specific ATP production rate, which may possibly be used to support the higher energy requirements required for cellular maintenance at high osmolalities. Finally, nodal analysis confirmed that the PEP/pyruvate node is essentially rigid and that the glucose-6-phosphate, oxaloacetate and alpha-ketoglutarate nodes are flexible and therefore adaptable to changes in osmotic pressure in C. glutamicum.

Genome-wide screening for trait conferring genes using DNA microarrays

We report a DNA microarray-based method for genome-wide monitoring of competitively grown transformants to identify genes whose overexpression confers a specific cellular phenotype. Whereas transcriptional profiling identifies differentially expressed genes that are correlated with particular aspects of the cellular phenotype, this functional genomics approach determines genes that result in a specific physiology. This parallel gene-trait mapping method consists of transforming a strain with a genomic library, enriching the cell population in transformants containing the trait conferring gene(s), and finally using DNA microarrays to simultaneously isolate and identify the enriched gene inserts. Various methods of enrichment can be used; here, genes conferring low-level antibiotic resistance were identified by growth in selective media. We demonstrated the method by transforming Escherichia coli cells with a genomic E. coli library and selecting for transformants exhibiting a growth advantage in the presence of the anti-microbial agent Pine-Sol. Genes conferring Pine-Sol tolerance (19 genes) or sensitivity (27 genes) were identified by hybridizing, on DNA microarrays containing 1,160 E. coli gene probes, extra-chromosomal DNA isolated from transformed cells grown in the presence of various levels of Pine-Sol. Results were further validated by plating and sequencing of individual colonies, and also by assessing the Pine-Sol resistance of cells transformed with enriched plasmid library or individual resistance genes identified by the microarrays. Applications of this method beyond antibiotic resistance include identification of genes resulting in resistance to chemotherapeutic agents, genes yielding resistance to toxic products (recombinant proteins, chemical feedstocks) in industrial fermentations, genes providing enhanced growth in cell culture or high cell density fermentations, genes facilitating growth on unconventional substrates, and others.

Metabolic engineering of indene bioconversion in Rhodococcus sp

We have applied the methodology of metabolic engineering in the investigation of the enzymatic bioreaction network in Rhodococcus sp. that catalyzes the bioconversion of indene to (2R)-indandiol suitable for the synthesis of cis-1-amino-2-indanol, a precursor of the HIV protease inhibitor, Crixivan. A chemostat with a novel indene air delivery system was developed to facilitate the study of steady state physiology of Rhodococcus sp. 124. Prolonged cultivation of this organism in a continuous flow system led to the evolution of a mutant strain, designated KY1, with improved bioconversion properties, in particular a twofold increase in yield of (2R)-indandiol relative to 124. Induction studies with both strains indicated that KY1 lacked a toluene-inducible dioxygenase activity present in 124 and responsible for the formation of undesired byproducts. Flux analysis of indene bioconversion in KY1 performed using steady state metabolite balancing and labeling with [14C]-tracers revealed that at least 94% of the indene is oxidized by a monooxygenase to indan oxide that is subsequently hydrolyzed to trans-(1R,2R)-indandiol and cis-(1S,2R)-indandiol. This analysis identified several targets in KY1 for increasing (2R)-indandiol product yield. Most promising among them is the selective hydrolysis of indan oxide to trans-(1R,2R)-indandiol through expression of an epoxide hydrolase or modification of culture conditions.

After a decade of progress, an expanded role for metabolic engineering

Over the past decade, metabolic engineering has emerged as an active and distinct discipline characterized by its over-arching emphasis on integration. In practice, metabolic engineering is the directed improvement of cellular properties through the application of modern genetic methods. Although it was applied on an ad hoc basis for several years following the introduction of recombinant techniques [1,2], metabolic engineering was formally defined as a new field approximately a decade ago [3]. Since that time, many creative applications, directed primarily to metabolite overproduction, have been reported [4]. In parallel, recent advances in the resolution and acquisition time of biological data, especially structural and functional genomics, has amplified interest in the systemic view of biology that metabolic engineering provides. To facilitate the burgeoning scientific exchange in this area on a more regular and convenient basis, a new conference series was launched in 1996 followed by a new journal in 1999.

Metabolic engineering

Metabolic engineering is the science that combines systematic analysis of metabolic and other pathways with molecular biological techniques to improve cellular properties by designing and implementing rational genetic modifications. As such, metabolic engineering deals with the measurement of metabolic fluxes and elucidation of their control as determinants of metabolic function and cell physiology. A novel aspect of metabolic engineering is that it departs from the traditional reductionist paradigm of cellular metabolism, taking instead a holistic view. In this sense, metabolic engineering is well suited as a framework for the analysis of genome-wide differential gene expression data, in combination with data on protein content and in vivo metabolic fluxes. The insights of the integrated view of metabolism generated by metabolic engineering will have profound implications in biotechnological applications, as well as in devising rational strategies for target selection for screening candidate drugs or designing gene therapies. In this article we review basic concepts of metabolic engineering and provide examples of applications in the production of primary and secondary metabolites, improving cellular properties, and biomedical engineering.

A compendium of gene expression in normal human tissues

This study creates a compendium of gene expression in normal human tissues suitable as a reference for defining basic organ systems biology. Using oligonucleotide microarrays, we analyze 59 samples representing 19 distinct tissue types. Of approximately 7,000 genes analyzed, 451 genes are expressed in all tissue types and designated as housekeeping genes. These genes display significant variation in expression levels among tissues and are sufficient for discerning tissue-specific expression signatures, indicative of fundamental differences in biochemical processes. In addition, subsets of tissue-selective genes are identified that define key biological processes characterizing each organ. This compendium highlights similarities and differences among organ systems and different individuals and also provides a publicly available resource (Human Gene Expression Index, the HuGE Index, http://www.hugeindex.org) for future studies of pathophysiology.

Rapamycin reduces hybridoma cell death and enhances monoclonal antibody production

Rapamycin was used as a medium additive to slow the progression of CRL 1606 hybridomas through the cell cycle, under the hypothesis that such a modulation might reduce cell death. Cell cycle distributions for CRL hybridomas in the G1 phase of the cell cycle ranged from 20% to 35% during batch, fed-batch, and continuous culture experiments, independent of culture time, dilution rate, growth rates, or death rates. Rapamycin, an mTOR signaling inhibitor, immunosuppressant, and G1-phase arresting agent, was identified and tested for efficacy in restraining cell cycle progression in CRL 1606 hybridoma cultures. However, in the presence of 100 nM rapamycin, the percentage of cells in the G1 phase of the cell cycle during fed-batch cultures was only increased from 28% to 31% in control cultures to 37% to 48% for those with rapamycin. Accordingly, rapamycin only slightly reduced culture growth rate. Instead, the use of rapamycin more notably kept viability higher than that of control cultures by delaying cell death for 48 h, thereby enabling viable proliferation to higher maximum viable cell densities. Furthermore, rapamycin enhanced specific monoclonal antibody production by up to 100% during high-viability growth. Thus, over the course of 6-day fed-batch cultivations, the beneficial effects of rapamycin on viable cell density and specific productivity resulted in an increase in final monoclonal antibody titer from 0.25 to 0.56 g/L (124%). As rapamycin is reported to influence a much broader range of cellular functions than cell cycle alone, these findings are more illustrative of the influence that signal transduction pathways related to mTOR can have on overall cell physiology and culture productivity.

Application of radiolabeled tracers to biocatalytic flux analysis

Radiolabeled tracers can provide valuable information about the structure of and flux distributions in biocatalytic reaction networks. This method derives from prior studies of glucose metabolism in mammalian systems and is implemented by pulsing a culture with a radiolabeled metabolite that can be transported into the cells and subsequently measuring the radioactivity of all network metabolites following separation by liquid chromatography. Intracellular fluxes can be directly determined from the transient radioactivity count data by tracking the depletion of the radiolabeled metabolite and/or the accompanying accumulation of any products formed. This technique differs from previous methods in that it is applied within a systems approach to the problem of flux determination. It has been used for the investigation of the indene bioconversion network expressed in Rhodococcus sp. KY1. Flux estimates obtained by radioactive tracers were confirmed by macroscopic metabolite balancing and showed that indene oxidation in steady state chemostat cultures proceeds primarily through a monooxygenase activity forming (1S,2R)-indan oxide, with no dehydrogenation of trans-(1R,2R)-indandiol. These results confirmed the significance of indan oxide formation and identified the hydrolysis of indan oxide as a key step in maximizing the production of (2R)-indandiol, a chiral precursor of the HIV protease inhibitor, Crixivan.

Metabolic engineering as an integrating platform for strain development

Integration of the analytical framework and experimental tools of metabolic engineering with emerging technologies such as DNA microarrays and directed evolution stands to dramatically improve the approaches by which strain improvement and biocatalyst design are pursued in the future. Progress in genomics and applied molecular biology, together with increasing emphasis on renewable resource utilization for chemical production, has advanced metabolic engineering to the forefront of biotechnological interest.

Effect of insulin stimulation on the proliferation and death of Chinese hamster ovary cells

The effect of environmental and genetic parameters on cell death was studied in Chinese hamster ovary cell cultures. Experiments were performed using an anchorage-dependent CHO cell line expressing gamma-IFN, and a second cell line obtained by transfection of the previous one with bcl-2. In serum-free medium the two cell lines showed a considerable degree of growth control entering quiescence while maintaining high viabilities. The addition of transferrin did not have any effect but insulin addition allowed cells arrested by serum withdrawal to reenter the cell cycle. However, insulin supplementation also resulted in cell death, which was possible to avoid through bcl-2 overexpression or in the presence of serum. We propose that the expression of c-myc, shown to be induced by insulin, plays an important role in the cell death observed after insulin addition in an inappropriate environment, deficient in protective factors. This hypothesis is supported by measurements of c-myc expression and cell cycle distribution.

Targeted, PCR-based gene disruption in cyanobacteria: inactivation of the polyhydroxyalkanoic acid synthase genes in Synechocystis sp. PCC6803

A PCR-based method is described for the efficient construction of targeted gene disruptions and gene fusions in the cyanobacterium Synechocystis sp. PCC6803. In a simple two-step PCR approach, a gene conversion cassette was synthesized targeting the polyhydroxyalkanoic acid (PHA) synthase genes. Upon transformation, PHA production in Synechocystis under normal as well as high production culture conditions was undetectable. The application of this method to the genetic inactivation of the phaE-C(Syn) gene cluster demonstrates its potential for genetic engineering of cyanobacteria and the study of functional genomics in Synechocystis.

Mitochondrial membrane potential differentiates cells resistant to apoptosis in hybridoma cultures

Previous research has implicated mitochondrial physiology and, by extension, respiratory capacity in the initiation and progress of apoptosis of cells in culture and tissue environments. This hypothesis was tested by separating a hybridoma cell population into subpopulations of varying mitochondrial membrane potential (MMP) using Rhodamine 123 stain and fluorescence-activated cell sorter analysis and subjecting them to two apoptosis inducers, rotenone and staurosporin. Apoptotic death was characterized morphologically through the determination of apoptosis-related chromatin condensation and biochemically through the measurement of caspase-3 enzymatic activity. We found dramatic differences in the apoptotic death kinetics for the subpopulations, with the high MMP cells showing higher resistance to apoptotic death. After incubation with 30 microM rotenone, the low MMP cells exhibited one-third of the viability of the high MMP cells and a three-fold increase in the capsase-3 enzymatic activity. No changes were observed in the DNA content or the cell cycle distributions of the two cell subpopulations, which maintained their mean MMP difference after 20 generations. These results suggest that heterogeneity exists in mammalian cell populations with respect to mitochondrial physiology, which correlates with resistance to apoptotic death.

Identification and analysis of the polyhydroxyalkanoate-specific beta-ketothiolase and acetoacetyl coenzyme A reductase genes in the cyanobacterium Synechocystis sp. strain PCC6803

Synechocystis sp. strain PCC6803 possesses a polyhydroxyalkanoate (PHA)-specific beta-ketothiolase encoded by phaA(Syn) and an acetoacetyl-coenzyme A (CoA) reductase encoded by phaB(Syn). A similarity search of the entire Synechocystis genome sequence identified a cluster of two putative open reading frames (ORFs) for these genes, slr1993 and slr1994. Sequence analysis showed that the ORFs encode proteins having 409 and 240 amino acids, respectively. The two ORFs are colinear and most probably coexpressed, as revealed by sequence analysis of the promoter regions. Heterologous transformation of Escherichia coli with the two genes and the PHA synthase of Synechocystis resulted in accumulation of PHAs that accounted for up to 12.3% of the cell dry weight under high-glucose growth conditions. Targeted disruption of the above gene cluster in Synechocystis eliminated the accumulation of PHAs. ORFs slr1993 and slr1994 thus encode the PHA-specific beta-ketothiolase and acetoacetyl-CoA reductase of Synechocystis and, together with the recently characterized PHA synthase genes in this organism (S. Hein, H. Tran, and A. Steinbüchel, Arch. Microbiol. 170:162-170, 1998), form the first complete PHA biosynthesis pathway known in cyanobacteria. Sequence alignment of all known short-chain-length PHA-specific acetoacetyl-CoA reductases also suggests an extended signature sequence, VTGXXXGIG, for this group of proteins. Phylogenetic analysis further places the origin of phaA(Syn) and phaB(Syn) in the gamma subdivision of the division Proteobacteria.

Mining of biological data I: identifying discriminating features via mean hypothesis testing

Large volumes of data are routinely collected during bioprocess operations and, more recently, in basic biological research using genomics-based technologies. While these data often lack sufficient detail to be used for mechanism identification, it is possible that the underlying mechanisms affecting cell phenotype or process outcome are reflected as specific patterns in the overall or temporal sensor logs. This raises the possibility of identifying outcome-specific fingerprints that can be used for process or phenotype classification and the identification of discriminating characteristics, such as specific genes or process variables. The aim of this work is to provide a systematic approach to identifying and modeling patterns in historical records and using this information for process classification. This approach differs from others in that emphasis is placed on analyzing the data structure first and thereby extracting potentially relevant features prior to model creation. The initial step in this overall approach is to first identify the discriminating features of the relevant measurements and time windows, which can then be subsequently used to discriminate among different classes of process behavior. This is achieved via a mean hypothesis testing algorithm. Next, the homogeneity of the multivariate data in each class is explored via a novel cluster analysis technique called PC1 Time Series Clustering to ensure that the data subsets used accurately reflect the variability displayed in the historical records. This will be the topic of the second paper in this series. We present here the method for identifying discriminating features in data via mean hypothesis testing along with results from the analysis of case studies from industrial fermentations

Mining of biological data II: assessing data structure and class homogeneity by cluster analysis

An important step in data analysis is class assignment which is usually done on the basis of a macroscopic phenotypic or bioprocess characteristic, such as high vs low growth, healthy vs diseased state, or high vs. low productivity. Unfortunately, such an assignment may lump together samples, which when derived from a more detailed phenotypic or bioprocess description are dissimilar, giving rise to models of lower quality and predictive power. In this paper we present a clustering algorithm for data preprocessing which involves the identification of fundamentally similar lots on the basis of the extent of similarity among the system variables. The algorithm combines aspects of cluster analysis and principal component analysis by applying agglomerative clustering methods to the first principal component of the system data matrix. As part of a rational strategy for developing empirical models, this technique selects lots (samples) which are most appropriate for inclusion in a training set by analyzing multivariate data homogeneity. Samples with similar data structures are identified and grouped together into distinct clusters. This knowledge is used in the formation of potential training sets. Additionally, this technique can identify atypical lots, i.e., samples that are not simply outliers but exhibit the general properties of one class but have been given the assignment of the other. The method is presented along with examples from its application to fermentation data sets.

Metabolic flux distributions in Corynebacterium glutamicum during growth and lysine overproduction. Reprinted from Biotechnology and Bioengineering, Vol. 41, Pp 633-646 (1993)

The two main contributions of this article are the solidification of Corynebacterium glutamicum biochemistry guided by bioreaction network analysis, and the determination of basal metabolic flux distributions during growth and lysine synthesis. Employed methodology makes use of stoichiometrically based mass balances to determine flux distributions in the C. glutamicum metabolic network. Presented are a brief description of the methodology, a thorough literature review of glutamic acid bacteria biochemistry, and specific results obtained through a combination of fermentation studies and analysis-directed intracellular assays. The latter include the findings of the lack of activity of glyoxylate shunt, and that phosphoenolpyruvate carboxylase (PPC) is the only anaplerotic reaction expressed in C. glutamicum cultivated on glucose minimal media. Network simplifications afforded by the above findings facilitated the determination of metabolic flux distributions under a variety of culture conditions and led to the following conclusions. Both the pentose phosphate pathway and PPC support significant fluxes during growth and lysine overproduction, and that flux partitioning at the glucosa-6-phosphate branch point does not appear to limit lysine synthesis.

Effect of glutamine limitation on the death of attached Chinese hamster ovary cells

The effect of glutamine depletion on the death of attached Chinese hamster ovary (CHO) cells was investigated. Experiments were performed using an anchorage dependent CHO cell line expressing gamma-IFN and a second cell line obtained by transfection of that cell line with the human bcl-2 (hbcl-2). Either cell line could grow in media devoid of glutamine with minimal cell death due to endogenous glutamine synthetase activity that allowed cells to synthesize glutamine from glutamic acid in the medium. However, compared to control cultures in glutamine-containing media, the cell growth rate in glutamine-free media was slower with an increased fraction of cells distributed in the G0/G1 phase. The slower rate of cell cycling apparently protected the cells from entering apoptosis when they were stimulated to proliferate in an environment devoid of other protective factors, such as serum or over-expressed hbcl-2. The depletion of both glutamine and glutamic acid did cause cell death, which could be mitigated by hbcl-2 over-expression.

Metabolic flux analysis of hybridoma continuous culture steady state multiplicity

Physiological state multiplicity was observed in continuous cultures of the hybridoma cell line ATCC CRL-1606 cultivated in glutamine-limited steady state chemostats. At the same dilution rate (0.04 h-1), two physiologically different cultures were obtained which exhibited similar growth rates and viabilities but drastically different cell concentrations (7.36 x 10(5) and 1.36 x 10(6) cells/mL). Metabolic flux analysis conducted using metabolite and gas exchange rate measurements revealed a more efficient culture for the steady state with the higher cell concentration, as measured by the fraction of pyruvate carbon flux shuttled into the TCA cycle for energy generation. The low-efficiency steady state was achieved after innoculation by growing the cells in a nutrient rich environment, first in batch mode followed by a stepwise increase of the dilution rate to its set point at 0.04 h-1. The high-efficiency steady state was achieved by reducing the dilution rate to progressively lower values to 0.01 h-1 resulting in conditions of stricter nutrient limitation. The high energetic efficiency attained under such conditions was preserved upon increasing the chemostat dilution rate back to 0.04 h-1 with a higher nutrient consumption, resulting in approximate doubling of the steady state cell concentration. This metabolic adaptation is unlikely due to favorable genetic mutations and could be implemented for improving cell culture performance by inducing cellular metabolic shifts to more efficient flux distribution patterns.

Isolation and characterization of indene bioconversion genes from Rhodococcus strain I24

Rhodococcus strain 124 is able to convert indene into indandiol via the actions of at least two dioxygenase systems and a putative monooxygenase system. We have identified a cosmid clone from 124 genomic DNA that is able to confer the ability to convert indene to indandiol upon Rhodococcus erythropolis SQ1, a strain that normally can not convert or metabolize indene. HPLC analysis reveals that the transformed SQ1 strain produces cis-(1R,2S)-indandiol, suggesting that the cosmid clone encodes a naphthalenetype dioxygenase. DNA sequence analysis of a portion of this clone confirmed the presence of genes for the dioxygenase as well as genes encoding a dehydrogenase and putative aldolase. These genes will be useful for manipulating indene bioconversion in Rhodococcus strain 124.

Analysis of the pathway structure of metabolic networks

Metabolic networks comprise a multitude of enzymatic reactions carrying out various functions related to cell growth and product formation. Although such reactions are occasionally organized into biochemical pathways, a formal procedure is desired to identify the independent pathways in a bioreaction network and the degree of engagement of each individual reaction in these pathways. We present a procedure for the identification of the independent pathways of bioreaction networks of any size and complexity. The method makes use of the steady-state internal metabolite stoichiometry matrix and defines the independent pathways through the reaction membership of its kernel matrix. Examples from the aromatic amino acid biosynthetic pathway and central carbon metabolism of cells in culture are provided to illustrate the method. Applications to the analysis of the control structure of bioreaction networks are also discussed.

Flux through the tetrahydrodipicolinate succinylase pathway is dispensable for L-lysine production in Corynebacterium glutamicum

The N-succinyl-LL-diaminopimelate desuccinylase gene (dapE) in the four-step succinylase branch of the L-lysine biosynthetic pathway of Corynebacterium glutamicum was disrupted via marker-exchange mutagenesis to create a mutant strain that uses only the one-step meso-diaminopimelate dehydrogenase branch to overproduce lysine. This mutant strain grew and utilized glucose from minimal medium at the same rate as the parental strain. In addition, the dapE- strain produced lysine at the same rate as its parent strain. Transformation of the parental and dapE- strains with the amplified meso-diaminopimelate dehydrogenase gene (ddh) on a plasmid did not affect lysine production in either strain, despite an eightfold amplification of the activity of the enzyme. These results indicate that the four-step succinylase pathway is dispensable for lysine overproduction in shake-flask culture. In addition, the one-step meso-diaminopimelate dehydrogenase pathway does not limit lysine flux in Corynebacterium under these conditions.

Metabolite and isotopomer balancing in the analysis of metabolic cycles: II. Applications

In a previous paper (Klapa et al., 1999), we presented a model for the analysis of isotopomer distributions of the TCA cycle intermediates resulting from 13C (or 14C) labeling experiments. Results allow the rigorous determination of the degree of enrichment at specific carbon atoms of metabolites, of the molecular weight distribution of metabolite isotopomers, as well as of the fine structure of NMR spectra in terms of a small number of metabolic fluxes. In this paper we validate the model by comparing model predictions with experimental data and then apply it to the analysis of metabolic networks that have been investigated in previous studies. The results have allowed us to conclude that: (1) there is no evidence of propionyl-CoA carboxylase pathway in Escherichia coli; and (2) the possibility that acetone utilization in mammals occurs solely via the "lactate/methylglyoxal" pathway is consistent with available labeling data. The presented modeling framework provides additional constraints that must be satisfied by experimental data in a biochemical network structure and therefore enhances the power of labeling methods for resolving in vivo metabolic fluxes.

Metabolism of peptide amino acids by Chinese hamster ovary cells grown in a complex medium

Metabolic flux analysis is a useful tool for unraveling relationships between metabolism and cell function. Material balancing can be used to provide estimates of major metabolic pathway fluxes, provided all significant metabolite uptake and production rates are measured. Potential sources of metabolizable material in many serum-free media formulations are low molecular weight digests of biological material such as yeast extracts and plant or animal tissue hydrolysates. These digests typically contain large amounts of peptides, which may be utilized as amino acids. This article demonstrates the need for accounting for amino acids liberated from peptides in order to accurately estimate pathway fluxes in Chinese hamster ovary cells grown in a complex (hydrolysate containing) medium. A simplified model of central carbon metabolism provides the framework for analyzing external metabolite measurements. Redundant measurements are included to ensure the consistency of data and assumed biochemistry by comparing redundant measurements with their predicted values from a minimum data set, and by expressing the degree of agreement using a statistical "consistency index." The consistency index tests whether redundancies are satisfied within expected experimental error. For chemostat steady states of CHO cultures grown in a hydrolysate-supplemented medium, consistent data were obtained only when amino acids liberated from peptides were taken into account.

Metabolic effects on recombinant interferon-gamma glycosylation in continuous culture of Chinese hamster ovary cells

Asparagine linked (N-linked) glycosylation is an important modification of recombinant proteins, because the attached oligosaccharide chains can significantly alter protein properties. Potential glycosylation sites are not always occupied with oligosaccharide, and site occupancy can change with the culture environment. To investigate the relationship between metabolism and glycosylation site occupancy, we studied the glycosylation of recombinant human interferon-gamma (IFN-gamma) produced in continuous culture of Chinese hamster ovary cells. Intracellular nucleotide sugar levels and IFN-gamma glycosylation were measured at different steady states which were characterized by central carbon metabolic fluxes estimated by material balances and extracellular metabolite rate measurements. Although site occupancy varied over a rather narrow range, we found that differences correlated with the intracellular pool of UDP-N-acetylglucosamine + UDP-N-acetylgalactosamine (UDP-GNAc). Measured nucleotide levels and estimates of central carbon metabolic fluxes point to UTP depletion as the cause of decreased UDP-GNAc during glucose limitation. Glucose limited cells preferentially utilized available carbon for energy production, causing reduced nucleotide biosynthesis. Lower nucleoside triphosphate pools in turn led to lower nucleotide sugar pools and reduced glycosylation site occupancy. Subsequent experiments in batch and fed-batch culture have confirmed that UDP-sugar concentrations are correlated with UTP levels in the absence of glutamine limitation. Glutamine limitation appears to influence glycosylation by reducing amino sugar formation and hence UDP-GNAc concentration. The influence of nucleotide sugars on site occupancy may only be important during periods of extreme starvation, since relatively large changes in nucleotide sugar pools led to only minor changes in glycosylation.

Metabolite and isotopomer balancing in the analysis of metabolic cycles: I. Theory

Proper analysis of label distribution in metabolic pathway intermediates is critical for correct interpretation of experimental data and strategic experimental design. While, for example, 13C nuclear magnetic resonance (NMR) spectroscopy is usually limited to the measurement of degrees of 13C enrichment, more information about metabolic fluxes can be extracted from the fine structure of NMR spectra, or molecular weight distributions of isotopomers of metabolic intermediates (measured by gas chromatography-mass spectrometry). For this purpose, rigorous accounting for the contribution of all pathways to label distribution is required, especially contributions resulting from multiple turns of metabolic cycles. In this paper we present a mathematical model developed to analyze isotopomer distributions of tricarboxylic acid cycle (TCA) intermediates following the administration of 13C (or 14C) labeled substrates. The theory presented provides the basis to analyze 13C NMR spectra and molecular weight distributions of metabolites. In a companion paper (Park et al., 1999), the theory is applied to the analysis of several cases of biological significance. Copyright 1999 John Wiley & Sons, Inc.

Metabolic fluxes and metabolic engineering

Metabolic engineering is the directed improvement of cellular properties through the modification of specific biochemical reactions or the introduction of new ones, with the use of recombinant DNA technology. As such, metabolic engineering emphasizes metabolic pathway integration and relies on metabolic fluxes as determinants of cell physiology and measures of metabolic control. The combination of analytical methods to quantify fluxes and their control with molecular biological techniques to implement genetic modifications is the essence of metabolic engineering. Strategies for metabolic flux determination are reviewed in this paper and it is shown how metabolic fluxes can be used in the systematic elucidation of metabolic control in the framework of reaction grouping and top-down metabolic control analysis.

Sequence of the Corynebacterium glutamicum pyruvate carboxylase gene

Pyruvate carboxylase is an important anaplerotic enzyme replenishing oxaloacetate consumed for biosynthesis during growth, or lysine and glutamic acid production in industrial fermentations. We used regions of homology from pyruvate carboxylase sequences of 12 different species (corresponding to the ATP- and pyruvate-binding sites), to design polymerase chain reaction (PCR) primers for amplifying a fragment of the pyruvate carboxylase (pc) gene from C. glutamicum genomic DNA. This 850-base-pair fragment was used to probe a C. glutamicum cosmid library and four candidate pc cosmids were identified. The fragment was sequenced and the sequence of the complete gene was obtained by several rounds of primer synthesis, PCR on one of the positive cosmids, and sequencing. The C. glutamicum pc sequence shows 64% homology with the pc gene of Mycobacterium tuberculosis and 44% homology with the human pc gene. Regions of ATP, pyruvate and biotin binding have also been identified.

Experimental determination of group flux control coefficients in metabolic networks

Grouping of reactions around key metabolite branch points can facilitate the study of metabolic control of complex metabolic networks. This top-down Metabolic Control Analysis is exemplified through the introduction of group (flux, as well as concentration) control coefficients whose magnitudes provide a measure of the relative impact of each reaction group on the overall network flux, as well as on the overall network stability, following enzymatic amplification. In this article, we demonstrate the application of previously developed theory to the determination of group flux control coefficients. Experimental data for the changes in metabolic fluxes obtained in response to the introduction of six different environmental perturbations are used to determine the group flux control coefficients for three reaction groups formed around the phosphoenolpyruvate/pyruvate branch point. The consistency of the obtained group flux control coefficient estimates is systematically analyzed to ensure that all necessary conditions are satisfied. The magnitudes of the determined control coefficients suggest that the control of lysine production flux in Corynebacterium glutamicum cells at a growth base state resides within the lysine biosynthetic pathway that begins with the PEP/PYR carboxylation anaplorotic pathway. Copyright 1998 John Wiley & Sons, Inc.

Effect of reversible reactions on isotope label redistribution--analysis of the pentose phosphate pathway

The pentose phosphate pathway plays several key roles in metabolism including supply of biosynthetic carbon skeletons and reducing power. Previous research has focused on determining the fluxes through the reactions of this pathway using carbon-labeled substrates and models that make certain assumptions about the reversibility of the transketolase and transaldolase reactions in the nonoxidative pathway. These assumptions, however, have resulted in inconsistencies between the predicted carbon label distributions using these models and those determined experimentally. A general metabolic reaction network model developed in this paper and applied to the pentose phosphate pathway not only incorporates reaction reversibility but also accounts for the effect of individually varying extents of reaction reversibility on labeled carbon fractional enrichment values for intermediate metabolites. In addition, an algorithm is presented that can be used to calculate the three individual transaldolase and transketolase extents of reversibility. The results of this method show that varying extents of reaction reversibility have an observable effect on the metabolite carbon label distributions which can in turn affect flux calculation for other parts of the metabolic network such as the tricarboxylic acid cycle. In addition, the observability of reversibility extent and accuracy of flux calculations depend on the particular choice of metabolite carbon enrichments measured. In particular, [6-13C]hexose 6-phosphate and [4-13C]erythrose 4-phosphate carbon enrichment values resulting from [1-13C]glucose feeding contained more information as compared to those from ribose 5-phosphate. This analysis was applied to literature data of metabolite carbon labeling that resulted from supplying either 13C- or 14C-enriched substrates to several cell types growing under various conditions. The specific activities of metabolite carbon atoms taken from rat epididymal adipose tissue, goosefish islet cells, Corynebacterium glutamicum, and Escherichia coli supplied with either [2-14C]glucose or [1-13C]glucose demonstrate how reversibility is present in the pentose phosphate pathway and the extents of reversibility can be estimated from labeled carbon data sets.

Growth factor and bcl-2 mediated survival during abortive proliferation of hybridoma cell line

Cultures of the CRL-1606 hybridoma (ATCC) have been reported to undergo continuous proliferation with simultaneous death during nutrient limited fed-batch fermentations. The bcl-2 proto-oncogene has been shown to prevent cell death under a variety of otherwise death inducing conditions. We were interested in elucidating the nature of the massive death observed in cultures of CRL-1606, specifically with respect to the possible environmental causes, and the ability of overexpressed human bcl-2 (hbcl-2) to mitigate cell death. Abortive proliferation, or continuous proliferation in the presence of continuous death, could be induced in serum free cultures of CRL-1606 through the withdrawal of insulin provided the culture was competent for cell proliferation. Culture competency for proliferation was found to be solely determined by the presence of cell culture nutrients. Abortive proliferation was defective in cultures transfected with hbcl-2 and the enhanced viability observed resulted from an increased viable cell population and at the expense of the nonviable cell population normally found in untransfected cultures. Abortive proliferation was also observed in serum containing cultures upon serum shiftdowns. Like the insulin-supplemented serum free culture system, hbcl-2 transfected cultures exhibited defects in the abortive proliferation process. These results suggest that the massive death observed during nutrient-limited fed-batch fermentation originate, in part, from growth or survival factor limitations. Hence, approaches to design cell culture media that account for the cell's proliferation requirements without accounting for the cell's survival requirements may represent a cell death sentence. Given the transformed nature of the hybridomas, we conclude that the abortive proliferation of CRL-1606 is a consequence of inappropriate cell cycle entry in a survival factor limited environment.

Processing and secretion of insulin-related peptides in an insulinoma cell line

Certain classes of prohormones and other neuroendocrine or endocrine-derived secretory proteins are post-translationally modified in the secretory storage granules. If such molecules were to be biosynthesized to acceptable quantity and yield using endocrine-derived cell lines, it would be important to understand the relationship between the secretory dynamics and the conversion and release of the immature and mature forms of the molecule. We studied aspects of such a relationship using the endocrine-derived cell line betaTC-3, which synthesizes murine proinsulin, sequesters it into secretory granules, and converts it into mature insulin. In T-flask experiments with confluent cultures of betaTC-3 cells, intracellular and secreted (pro)insulin was sampled before and after episodes of stimulated exocytosis and recharging and quantified by radioimmunoassay and reversed-phase high-performance liquid chromatography (HPLC). Under conditions of steady-state secretion in glucose-rich growth medium the cells turned over their (pro)insulin inventory (90 +/- 5% mature insulin) at 2-3% per hour through secretion of (pro)insulin which was less than 70% mature. During an episode of hyperstimulated exocytosis induced by the combined secretagogues carbachol (1 microM) and isobutylmethylxanthine (1 mM), approximately 80% of the intracellular (pro)insulin stores were depleted within 2 h and 84 +/- 4% of the secreted (pro)insulin was in the mature form. Following the discharging episode, exocytosis was suppressed to 10% of its steady-state rate with a treatment which attenuated calcium influx (20 microM verapamil with reduced levels of calcium in the medium). Under this condition the secreted protein was only approximately 50% converted to mature insulin, but 85 +/- 10% of the net (pro)insulin accumulating within the intracellular stores was converted to the mature form. The inverse relationship between rate of secretion and degree of conversion of secreted (pro)insulin is consistent with a previously observed phenomenon of preferential basal secretion from immature secretory granules. This tends to enrich the secreted peptides in immature forms relative to the total intracellular pool. Preferential early secretion can best be overcome by rapid discharging of the long-term and predominantly mature stores. Thus, a cyclic controlled secretion process wherein product is collected during intermittent discharging episodes would provide a better yield of mature product than would steady-state secretion.