Monitoring of complex industrial bioprocesses for metabolite concentrations using modern spectroscopies and machine learning: application to gibberellic acid production

Two rapid vibrational spectroscopic approaches (diffuse reflectance-absorbance Fourier transform infrared [FT-IR] and dispersive Raman spectroscopy), and one mass spectrometric method based on in vacuo Curie-point pyrolysis (PyMS), were investigated in this study. A diverse range of unprocessed, industrial fed-batch fermentation broths containing the fungus Gibberella fujikuroi producing the natural product gibberellic acid, were analyzed directly without a priori chromatographic separation. Partial least squares regression (PLSR) and artificial neural networks (ANNs) were applied to all of the information-rich spectra obtained by each of the methods to obtain quantitative information on the gibberellic acid titer. These estimates were of good precision, and the typical root-mean-square error for predictions of concentrations in an independent test set was <10% over a very wide titer range from 0 to 4925 ppm. However, although PLSR and ANNs are very powerful techniques they are often described as "black box" methods because the information they use to construct the calibration model is largely inaccessible. Therefore, a variety of novel evolutionary computation-based methods, including genetic algorithms and genetic programming, were used to produce models that allowed the determination of those input variables that contributed most to the models formed, and to observe that these models were predominantly based on the concentration of gibberellic acid itself. This is the first time that these three modern analytical spectroscopies, in combination with advanced chemometric data analysis, have been compared for their ability to analyze a real commercial bioprocess. The results demonstrate unequivocally that all methods provide very rapid and accurate estimates of the progress of industrial fermentations, and indicate that, of the three methods studied, Raman spectroscopy is the ideal bioprocess monitoring method because it can be adapted for on-line analysis.     

Assessment of near-infrared spectral information for rapid monitoring of bioprocess quality.

Access to real-time process information is desirable for consistent and efficient operation of bioprocesses. Near-infrared spectroscopy (NIRS) is known to have potential for providing real-time information on the quantitative levels of important bioprocess variables. However, given the fact that a typical NIR spectrum encompasses information regarding almost all the constituents of the sample matrix, there are few case studies that have investigated the spectral details for applications in bioprocess quality assessment or qualitative bioprocess monitoring. Such information would be invaluable in providing operator-level assistance on the progress of a bioprocess in industrial-scale productions. We investigated this aspect and report the results of our investigation. Near-infrared spectral information derived from scanning unprocessed culture fluid (broth) samples from a complex antibiotic production process was assessed for a data set that incorporated bioprocess variations. Principal component analysis was applied to the spectral data and the loadings and scores of the principal components studied. Changes in the spectral information that corresponded to variations in the bioprocess could be deciphered. Despite the complexity of the matrix, near-infrared spectra of the culture broth are shown to have valuable information that can be deconvoluted with the help of factor analysis techniques such as principal component analysis (PCA). Although complex to interpret, the loadings and score plots are shown to offer potential in process diagnosis that could be of value in the rapid assessment of process quality, and in data assessment prior to quantitative model development.     

Prospective inference of bioprocess cell viability through chemometric modeling of fluorescence multiway data

In this investigation, the fermentation step of a standard mammalian cell-based industrial bioprocess for the production of a therapeutic protein was studied, with particular emphasis on the evolution of cell viability. This parameter constitutes one of the critical variables for bioprocess monitoring since it can affect downstream operations and the quality of the final product. In addition, when the cells experiment an unpredictable drop in viability, the assessment of this variable through classic off-line methods may not provide information sufficiently in advance to take corrective actions. In this context, Process Analytical Technology (PAT) framework aims to develop novel strategies for more efficient monitoring of critical variables, in order to improve the bioprocess performance. Thus, in this work, a set of chemometric tools were integrated to establish a PAT strategy to monitor cell viability, based on fluorescence multiway data obtained from fermentation samples of a particular bioprocess, in two different scales of operation. The spectral information, together with data regarding process variables, was integrated through chemometric exploratory tools to characterize the bioprocess and stablish novel criteria for the monitoring of cell viability. These findings motivated the development of a multivariate classification model, aiming to obtain predictive tools for the monitoring of future lots of the same bioprocess. The model could be satisfactorily fitted, showing the non-error rate of prediction of 100%.     

Epstein Barr virus/complement C3d receptor is an interferon alpha receptor.

Interferon alpha contains a sequence motif similar to the complement receptor type two (CR2/CD21) binding site on complement fragment C3d. Antibodies against a peptide with the CR2 binding sequence on C3d react with a peptide carrying the IFN alpha CR2 binding motif (residues 92-99) and with recombinant IFN alpha. The IFN alpha-derived peptide, as well as recombinant IFN alpha, inhibits C3bi/C3d interaction with CR2 on the Burkitt lymphoma Raji. The direct interaction of IFN alpha and CR2 is inhibited by polyclonal anti-IFN alpha, anti-CR2 and anti-C3d peptide antibodies as well as by C3bi/C3d, EBV coat protein gp350/220 and IFN but not by IFN gamma. [125I]IFN alpha binding to Raji cells is inhibited by polyclonal anti-IFN alpha and anti-CR2 antibodies, by peptides with the CR2 binding motif and partially by C3bi/C3d. Monoclonal anti-CR2 antibody HB5, but not OKB-7, blocks IFN alpha binding to Raji cells. CR2 or CR2-like molecules may therefore be the major IFN alpha receptors on B lymphocytes.     

Rapid screening for metabolite overproduction in fermentor broths, using pyrolysis mass spectrometry with multivariate calibration and artificial neural networks

Binary mixtures of model systems consisting of the antibiotic ampicillin with either Escherichia coli or Staphylococcus auresu were subjected to pyrolysis mass spectrometry (PyMS). To deconvolute the pyrolysis mass spectra, so as to obtain quantitative information on the concentration of ampicilin in the mixtures, partial least squares regression (PLS), principal components regression (PCR), and fully interconnected feedforward artificial neural networks (ANNs) were studied. In the latter case, the weights were modified using the standard backpropagation algorithm, and the nodes used a sigmoidal squsahing funciton. It was found that each of the methods could be used to provide calibration models which gave excellent predictions for the concentrations of ampicillin in samples on which they had not been trained. Furthermore, ANNs trained to predict the amount of ampicilin in E. coli were able to generalise so as to predict the concentration of ampicillin in a S. aureus background, illustrating the robustness of ANNs to rather substantial variations in the biological background. The PyMS of the complex mixture of ampicilin in bacteria could not be expressed simply in terms of additive combinations of the spectra describing the pure components of the mixtures and their relative concentrations. Intermolecular reactions took place in the pyrolysate, leading to a lack of superposition of the spectral components and to a dependence of the normalized mass spectrum on sample size. Samples from fermentations of a single organism in a complex production medium were also analyzed quantitatively for a drug of commercial interest. The drug could also be quantified in a variety of mutant-producing strains cultivated in the same medium. The combination of PyMS and ANNs constitutes a novel, rapid, and convenient method for exploitation in strain improvement screening programs. (c) 1994 John Wiley & Sons, Inc.     

Rapid quantitative analysis of binary mixtures of Escherichia coli strains using pyrolysis mass spectrometry with multivariate calibration and artificial neural networks

Pyrolysis mass spectrometry (PyMS) and multivariate calibration were used to show the high degree of relatedness between Escherichia coli HB101 and E. coli UB5201. Next, binary mixtures of these two phenotypically closely related E. coli strains were prepared and subjected to PyMS. Fully interconnected feedforward artificial neural networks (ANNs) were used to analyse the pyrolysis mass spectra to obtain quantitative information representative of the level of E. coli UB5201 in E. coli HB101. The ANNs exploited were trained using the standard back propagation algorithm, and the nodes used sigmoidal squashing functions. Accurate quantitative information was obtained for mixtures with >3% E. coli UB5201 in E. coli HB101. To remove noise from the pyrolysis mass spectra and so lower the limit of detection, the spectra were reduced using principal components analysis (PCA) and the first 13 principal components used to train ANNs. These PCA-ANNs allowed accurate estimates at levels as low as 1% E. coli UB5201 in E. coli HB101 to be predicted. In terms of bacterial numbers, it was shown that the limit of detection for PyMS in conjunction with ANNs was 3 × 10⁴ E. coli UB5201 cells in 1·6 × 10⁷ E. coli HB101 cells. It may be concluded that PyMS with ANNs provides a powerful and rapid method for the quantification of mixtures of closely related bacterial strains.     

Rapid authentication of animal cell lines using pyrolysis mass spectrometry and auto-associative artificial neural networks

Summary Pyrolysis mass spectrometry (PyMS) was used to produce biochemical fingerprints from replicate frozen cell cultures of mouse macrophage hybridoma 2C11-12, human leukaemia K562, baby hamster kidney BHK 21/C13, and mouse tumour BW-O, and a fresh culture of Chinese hamster ovary CHO cells. The dimensionality of these data was reduced by the unsupervised feature extraction pattern recognition technique of auto-associative neural networks. The clusters observed were compared with the groups obtained from the more conventional statistical approaches of hierarchical cluster analysis. It was observed that frozen and fresh cell line cultures gave very different pyrolysis mass spectra. When only the frozen animal cells were analysed by PyMS, auto-associative artificial neural networks (ANNs) were employed to discriminate between them successfully. Furthermore, very similar classifications were observed when the same spectral data were analysed using hierarchical cluster analysis. We demonstrate that this approach can detect the contamination of cell lines with low numbers of bacteria and fungi; this approach could plausibly be extended for the rapid detection of mycoplasma infection in animal cell lines. The major advantages that PyMS offers over more conventional methods used to type cell lines and to screen for microbial infection, such as DNA fingerprinting, are its speed, sensitivity and the ability to analyse hundreds of samples per day. We conclude that the combination of PyMS and ANNs can provide a rapid and accurate discriminatory technique for the authentication of animal cell line cultures.     

Discrimination of higher plant calluses based on embryogenic capacity and taxonomic classification by pyrolysis mass spectrometry

Pyrolysis mass spectrometry (PyMS) is a rapid, simple, high-resolution analytical method based on thermal degradation of complex material in a vacuum, and has been widely applied to the discrimination of closely related microbial strains. Minimally prepared samples of embryogenic and non-embryogenic calluses derived from various higher plants (sweet potato, morning glory, Korean ginseng, Siberian ginseng, and balloon flower) were subjected to PyMS for spectral fingerprinting. A dendrogram based on the unweighted pair group method, with arithmetic mean of pyrolysis mass spectra, divided the calluses into Siberian ginseng embryogenic callus and the others, which were subsequently divided into embryogenic and non-embryogenic callus groups, regardless of plant species from which the calluses were derived. In the non-embryogenic callus group, the dendrogram was in agreement with the known taxonomy of the plants. These results indicate that PyMS analysis could be applied for discriminating plant calluses based on embryogenic capacity and taxonomic classification.     

Use of the green fluorescent protein variant (YFP) to monitor MetArg human proinsulin production in Escherichia coli.

Green fluorescent protein (GFP), a relatively new reporter gene, is making an impact on many aspects of science. The attributes of GFP could also be applied to the area of recombinant protein production. The work described here represents the first experiments using GFP as a tool to monitor recombinant protein production in real time in the fermentation process. We have constructed plasmids containing an operon fusion of the gene encoding MetArg-human proinsulin and reporter gene GFP (GFP, BFP, and YFP variants). The MetArg-proinsulin and GFP variant reporter protein were overexpressed in Escherichia coli BL21(DE3) after isopropyl beta-d-thiogalactoside induction. The MetArg-proinsulin to YFP protein ratio did not change in the cells during the bioprocess. Since there is a quantitative relationship between the level of MetArg-proinsulin concentration and YFP fluorescence, it is possible to measure only YFP fluorescence in order to monitor the production of MetArg-proinsulin during the bioprocess. The expression level of MetArg-proinsulin could reach 20-25%. Some 140 mg recombinant MetArg-human proinsulin could be obtained easily from 1 liter of fermentation medium. The MetArg-proinsulin could simply be changed into human insulin by trypsin and carboxypeptidase B treatment in later steps. These experiments provide possibilities for using the YFP reporter gene as a convenient tool to monitor protein expression in biotechnological processes. The proposed technique could reduce the time- and labor-intensive analysis of protein production and would improve the efficiency of process development.     

Taxonomic discrimination of higher plants by pyrolysis mass spectrometry

Pyrolysis mass spectrometry (PyMS) is a rapid, simple, high-resolution analytical method based on thermal degradation of complex material in a vacuum and has been widely applied to the discrimination of closely related microbial strains. Leaf samples of six species and one variety of higher plants (Rosa multiflora, R. multiflora var. platyphylla, Sedum kamtschaticum, S. takesimense, S. sarmentosum, Hepatica insularis, and H. asiatica) were subjected to PyMS for spectral fingerprinting. Principal component analysis of PyMS data was not able to discriminate these plants in discrete clusters. However, canonical variate analysis of PyMS data separated these plants from one another. A hierarchical dendrogram based on canonical variate analysis was in agreement with the known taxonomy of the plants at the variety level. These results indicate that PyMS is able to discriminate higher plants based on taxonomic classification at the family, genus, species, and variety level.Abbreviations ANNs Artificial neural networks - CVA Canonical variate analysis - GC/MS Gas chromatography/mass spectrometry - PCA Principal component analysis - PyMS Pyrolysis mass spectrometry - UPGMA Unweighted pair group method with arithmetic mean Communicated by I.S. Chung     

Extracellular vesicles concentration is a promising and important parameter for industrial bioprocess monitoring

Extracellular vesicles (EVs) are membrane vesicles that are produced by cells to be released into their microenvironment. In this study, we present the EV concentration as a new factor for optimization of industrial bioprocess control. The release of EVs depends on many cell properties, including cell activation and stress status, and cell death. Therefore, the EV concentration might provide a readout for identification of the cell state and the conditions during a bioprocess. Our data show that the EV concentration increased during the bioprocess, which indicated deteriorating conditions in the bioreactor. This increase in EV concentration in the fermentation broth was the consequence of two different processes: cell activation, and cell death. However, the release of EVs from activated living cells had a much weaker impact on EV concentration in the bioreactor than those released during cell death. EVs and cells in the bioprocess environment were quantified by flow cytometry. The most accurate data were obtained directly from unprocessed samples, making the monitoring of the EV concentration a rapid, easy, and cheap method. These EV concentrations reflect the conditions in the bioreactor and provide new information regarding the state of the bioprocess. Therefore, we suggest EV concentration as a new and important parameter for the monitoring of industrial bioprocesses.     

Rapid monitoring of recombinant protein products: a comparison of current technologies

Specific measurement of recombinant protein titer in a complex environment during industrial bioprocessing has traditionally relied on labor-intensive and time-consuming immunoassays. In recent years, however, developments in analytical technology have resulted in improved methods for protein product monitoring during bioprocessing. The choice of product-monitoring technology for a particular bioprocess will depend on a variety of assay factors and instrument-specific factors. In this article, we have compiled an overview of the advantages and disadvantages of the most commonly used technologies used: electrochemiluminescence, optical biosensors, rapid chromatography and nephelometry. The advantages of each technology for measuring both small and large recombinant therapeutic proteins are compared with a conventional enzyme-linked immunosorbent assay (ELISA) technique.     

Constitutive expression of protective antigen gene of Bacillus anthracis in Escherichia coli

The fatal bacterial infection caused by inhalation of the Bacillus anthracis spores results from the synthesis of protein toxins-protective antigen (PA), lethal factor (LF), and edema factor (EF)--by the bacterium. PA is the target-cell binding protein and is common to the two effector molecules, LF and EF, which exert their toxic effects once they are translocated to the cytosol by PA. PA is the major component of vaccines against anthrax since it confers protective immunity. The large-scale production of recombinant protein-based anthrax vaccines requires overexpression of the PA protein. We have constitutively expressed the protective antigen protein in E. coli DH5alpha strain. We have found no increase in degradation of PA when the protein is constitutively expressed and no plasmid instability was observed inside the expressing cells. We have also scaled up the expression by bioprocess optimization using batch culture technique in a fermentor. The protein was purified using metal-chelate affinity chromatography. Approximately 125 mg of recombinant protective antigen (rPA) protein was obtained per liter of batch culture. It was found to be biologically and functionally fully active in comparison to PA protein from Bacillus anthracis. This is the first report of constitutive overexpression of protective antigen gene in E. coli.     

Immunological functions of macrophages and their regulation by interferons

Important progress has been recently made on the identification, physico-chemical properties, purification and production by recombinant technology of a series of immunoregulatory mediators, including interferons (IFNs). It is now possible to appreciate which part play IFNs in the regulation and activation of monocyte-macrophage (MO) functions. The present review discusses old concepts, such as Macrophage-Activating Factor (MAF) activity, at the light of recent data on the effects of IFN alpha, beta and gamma on four main immunological properties of MO, namely intracellular killing of microorganisms, non specific lysis of tumor cells, antigen presentation and secretion of soluble immunoregulatories molecules such as interleukin 1. All three IFN species enhance such functions, whereas enhancement of membrane class II major histocompatibility complex antigens is a unique property of IFN gamma. The role of IFNs, especially IFN gamma, as major MAF in vitro and in vivo is discussed.     

The suppressive effects of recombinant human tumor necrosis factor-alpha on normal and malignant myelopoiesis: synergism with interferon-gamma

The modulation of growth of normal and leukemic myeloid progenitor cells in soft agar cultures by recombinant human tumor necrosis factor-alpha (TNF alpha) and recombinant human interferon-gamma (IFN gamma) was investigated. TNF alpha inhibited colony formation of all colony types representing different maturational stages of normal progenitor cells committed to the myeloid lineage with different orders of sensitivity. Blast-type colonies derived from patients with acute myelogenous leukemia were more sensitive to TNF alpha inhibition than progenitor cells purified from normal bone marrow or bone marrow from patients with stable-phase chronic myelogenous leukemia. The response of most colony types to IFN gamma was poor. However, when IFN gamma was administered together with TNF alpha, synergistically enhanced antiproliferative effects were detected in all colony types tested. The antiproliferative action of IFN gamma on myelopoiesis was enhanced in culture by the presence of autologous monocytes, presumedly by inducing endogenous production of TNF alpha. However, TNF alpha seemed to act directly on the progenitor cells themselves to suppress their clonal growth, rather than involving accessory marrow elements such as monocytes and/or T lymphocytes.     

Optimized bioprocess for production of fructofuranosidase by recombinant Aspergillus niger

A comprehensive approach of bioprocess design at various levels was used to optimize microbial production of extracellular fructofuranosidase, important as biocatalyst to derive fructooligosaccharides with broad application in food or pharmaceutical industry. For production, the recombinant strain Aspergillus niger SKAn1015 was used, which expresses the fructofuranosidase encoding gene suc1 under control of a strong constitutive promoter. In a first screening towards an optimized medium, glucose, nitrate, Fe²⁺, and Mn²⁺ were identified as beneficial for production. A minimal medium with optimized concentration of these key nutrients, obtained by central composite design experiments and quadratic modelling, provided a threefold increased fructofuranosidase activity in the culture supernatant (400 U/mL) as compared to the originally described medium. Utilizing the optimized medium, the process was then transferred from shake flask into a fed-batch-operated bioreactor. Hereby, the intended addition of talc microparticles allowed engineering the morphology of A. niger into a highly active mycelial form, which strongly boosted production. Fructofuranosidase production was highly specific as confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis. The secreted enzyme activity of 2,800 U/mL, corresponding to about 3 g/L of fructofuranosidase, achieved by the microparticle-enhanced fed-batch process, is tenfold higher than that of any other process reported so far, so that the presented bioprocess strategy appears as a milestone towards future industrial fructofuranosidase production.     

Mycoplasma interaction with lymphocytes and phagocytes: role of hydrogen peroxide released from M. pneumoniae

Interferon (IFN) production by human peripheral lymphocytes stimulated with M. pneumoniae was investigated. The hydrogen peroxide released from M. pneumoniae was responsible for the induction of IFN from lymphocytes, since horseradish peroxidase inhibited the IFN production and abrogated the activity of IFN production in the supernatant of M. pneumoniae. The antiserum neutralizing IFN alpha and IFN beta failed to neutralize partially interferon produced by lymphocytes. Treatment either with pH 2.0 or antiserum neutralizing human IFN gamma resulted in a partial reduction of interferon. These results indicate that interferon produced by human lymphocytes stimulated with M. pneumoniae includes both types of IFN gamma and IFN beta.