Search

It is increasingly apparent that oxidants can play an important role in mediating specific cell responses and expression of genes involved in degenerative pathophysiologic states, such as inflammation. In particular, oxidant-induced activation of the multisubunit nuclear transcription factor, NFκB, has been implicated in the transcriptional upregulation of inflammatory genes like endothelial cell adhesion glycoproteins. A second emerging concept is the recognition that the oxidant effects in cellular and molecular regulation may be mediated by oxidant-induced loss of cellular oxidation-reduction (redox) balance. This review will provide an overview of our current understanding of leukocyte-endothelial cell interactions derived from in vitro studies of endothelial cell monolayers exposed to anoxia/reoxygenation, with specific emphasis on the molecular determinants mediating this inflammatory process and the contribution of reoxygenation-induced cellular redox imbalance to the activation of NFκB and the expression of endothelial surface adhesion molecules. [Copyright &y& Elsevier]

• Dynamic simulation of batch and perfusion fermentation of Chinese hamster ovary cells. • Kinetic models describe cell growth, substrate consumption and generation of products. • Sensitivity analysis performed establishes design and operating parameters for perturbation. • Economic analysis and production time evaluation is heavily dependent on media and cell usage. • Modelling problem open to dynamic optimization given detailed industrial and economic data. Chinese Hamster Ovary (CHO) cells are widely used in fermentation towards biopharmaceutical manufacturing. The present paper presents dynamic mathematical models of two different CHO culture modes: one batch mode for the production of interferon (IFN)- γ , and one perfusion mode for the production of a monoclonal antibody (mAb). The dynamic models have been used for simulating cell, substrate, by-product and product concentration trajectories, which have been compared against previously published experimental results. A sensitivity analysis of both models has been conducted, in order to analyse the relative importance of different operating parameters towards biopharmaceutical process design. An economic analysis has also been subsequently performed: production time and net present cost for given target capacities have been evaluated, using the validated dynamic models for the batch and perfusion modes. Economic trends are discussed for variable initial concentration of viable CHO cells to be used in bioreactors: the latter has been recognised as the most sensitive model parameter for both culture modes. [ABSTRACT FROM AUTHOR]

Humanization of monoclonal antibodies by complementary determinant region (CDR)-grafting has become a standard procedure to improve the clinical usage of animal antibodies. However, antibody humanization may result in loss of activity that has been attributed to structural constraints in the framework structure. In this paper, we report the complete humanization of the 6.7 anti-human CD18 monoclonal antibody in a scFv form. We used a germline-based approach to design a humanized VL gene fragment and expressed it together with a previously described humanized VH. The designed humanized VL has only 14 mutations compared to the closest human germline sequence. The resulting humanized scFv maintained the binding capacity and specificity to human CD18 expressed on the cell surface of peripheral blood mononuclear cells (PBMC), and showed the same pattern of staining T-lymphocytes sub-populations, in comparison to the original monoclonal antibody. We observed an unexpected effect of a conserved mouse–human framework position (L37) that hinders the binding of the humanized scFv to antigen. This paper reveals a new framework residue that interferes with paratope and antigen binding and also reinforces the germline approach as a successful strategy to humanize antibodies. [Copyright &y& Elsevier]

Chromatographic separation of biopharmaceuticals in general and monoclonal antibodies (mAbs) specifically is the bottleneck in terms of cost and throughput in preparative purification. Still, generalized platform processes are used, neglecting molecule specific characteristics, defining protein-resin interaction terms. Currently used in silico modeling approaches do not consider the orientation of the molecule towards the chromatographic resins as a result of the structural features on an atomic level. This paper describes a quantitative structure–activity relationship (QSAR) approach to model the orientation of mAbs on ion exchange chromatographic matrices as a function of property distribution and mobile phase characteristics. 6 mAbs were used to build a predictive QSAR model and to investigate the preferred binding orientations and resulting surface shielding on resins. Thereby different dominating orientations, caused by composition of F ab fragments of the mAbs, could be identified. The presented methodology is suitable to gain extended insight in molecule orientation on chromatographic resins and to tailor purification strategies based on molecule structure. [ABSTRACT FROM AUTHOR]

Thorough product understanding is one of the basic tenets for successful implementation of Quality by Design (QbD). Complexity encountered in analytical characterization of biotech therapeutics such as monoclonal antibodies (mAbs) requires novel, simpler, and generic approaches towards product characterization. This paper presents a methodology for implementation of QbD for analytical method development. Optimization of an analytical cation exchange high performance liquid chromatography (CEX-HPLC) method utilizing a sigmoidal gradient has been performed using a hybrid mechanistic model that is based on Design of experiment (DOE) based studies. Since sigmodal gradients are much more complex than the traditional linear gradients and have a large number of input parameters (five) for optimization, the number of DOE experiments required for a full factorial design to estimate all the main effects as well as the interactions would be too large (243). To address this problem, a mechanistic model was used to simulate the analytical separation for the DOE and then the results were used to build an empirical model. The mechanistic model used in this work is a more versatile general rate model in combination of modified Langmuir binding kinetics. The modified Langmuir model is capable of modelling the impact of nonlinear changes in the concentration of the salt modifier. Further, to get the input and output profiles of mAb and salts/buffers, the HPLC system, consisting of the mixer, detectors, and tubing was modelled as a sequence of dispersed plug flow reactors and continuous stirred tank reactors (CSTR). The experimental work was limited to calibration of the HPLC system and finding the model parameters through three linear gradients. To simplify the optimization process, only three peaks in the centre of the profile (main product and the adjacent acidic and basic variants) were chosen to determine the final operating condition. The regression model made from the DoE data yielded a R 2 > 0.97 which made it possible to predict and choose the design space where the optimal resolution between the acidic/main peaks and the basic/main peaks could be achieved (>1.2 and >2.5, respectively). The optimal operating condition was validated using experimental runs and was found to give the same resolution as what was predicted by the simulation. The proposed approach aims to significantly reduce the time required for method optimization as well as the extent of experimentation. [ABSTRACT FROM AUTHOR]

• Machine learning techniques enable real time process control and product quality prediction. • Proposed method assists process development in accordance with QbD approach. • Tree based regression techniques outperformed neural network and gradient boosting techniques. • Gradient boosting regression model performance depends on hyperparameters tuning. • Neural network requires large datasets for training and validation. As continuous manufacturing of biotherapeutics gains steam, there is an increasing interest in using machine learning (ML) techniques for real time prediction of product quality and for process control. This paper focuses on application of different ML techniques for predicting critical process attributes that are pertinent to capture and polishing chromatography. Data from pH, UV, and conductivity sensors are acquired and pre-processed. For the present case study, tree-based regression techniques (decision tree and random forest) outperformed in all cases. The final model, random forest regression model, resulted in prediction errors of <5% for all predicted attributes. The results showed that random forest models exhibit optimal performance for smaller process-scale datasets with low chances of overfitting, are computationally inexpensive and do not require a graphics processing unit. The proposed approach is well suited for implementation in a continuous mAb train for real time prediction of chromatography process attributes. [ABSTRACT FROM AUTHOR]

Abstract: Objective: Oral health conditions have a significant relationship with diabetes mellitus (DM) as well as dyslipidemia. In this study, we investigated the levels of apolipoprotein B (apoB) and oxidized low-density lipoprotein (oxLDL) in the gingival crevicular fluid (GCF) from patients with DM. Methods: GCF and blood samples from 18 DM patients and 18 healthy subjects were examined. GCF was collected with paper points without inflicting any harm. The apoB and oxLDL levels were measured by sandwich ELISA assays. Results: The number of teeth with a deep probing pocket depth and the number of teeth with bleeding on probing, two typical periodontal parameters, correlated with the DM parameters, such as hemoglobin A1c. The GCF volume and the concentrations of protein, apoB and oxLDL in GCF were significantly higher in the DM patients than in the healthy subjects. In particular, the apoB concentration in GCF was increased 6-fold in the DM patients. The GCF apoB concentration correlated well with the DM parameters in plasma. Conclusion: GCF could be a clinical source for examining not only the oral status of patients, but also certain systemic conditions. [Copyright &y& Elsevier]

Abstract: First of all, the aim of this paper is to systematically analyze the metabolism of mammalian cell cultures and reduce the complicated metabolic networks to a manageable macroscopic reaction scheme. Herein, the central carbon metabolism (glycolysis, TCA cycle and glutaminolysis) of glucose and amino acids was investigated whereas the synthetic metabolism of cell and antibody was simply represented with macroscopically balanced reactions. Without losing generality, perfusion and fed-batch cultures of two typical cell lines (hybridoma and GS-CHO) were carried out in bioreactor. With macroscopic balance analysis and metabolic flux analysis (stationary and dynamical), the metabolic networks were rationally reduced thanks to the experimental facts that most of the amino acids (except glutamine, glutamate, and alanine) contributed negligibly to the central carbon metabolism, i.e., they were allocated mainly to the syntheses. Finally, with the aid of analysis of elementary flux mode analysis and extreme pathways, nine macro reactions were derived with physiological definitions. Furthermore, the results from above analyses (MBA, MFA, EFM and EPs) support following conclusions: the respiratory quotient value in mammalian cell culture generally should be near one; several potential metabolic bottlenecks (pyruvate to acetyl-CoA, cytosol NADH translocation into mitochondrion, glutamate dehydrogenation into TCA) may control and regulate cellular metabolism. These results will benefit the future study of dynamic modeling for process monitoring and control. [Copyright &y& Elsevier]

Abstract: An increasing number of protein therapies require chronic administration at high doses (>200mg) by subcutaneous (sc) injection. Due to the injection volume limitation (<1.5mL) associated with sc administration, high protein concentration formulations at or exceeding 100mg/mL are required to achieve the dose. Development of a high concentration protein formulation can be challenging due to increased aggregation at higher concentration and/or chemical instability, which necessitates the development of lyophilized formulation for high protein concentration drug products. Unique challenges, such as long reconstitution time for a lyophilized high protein concentration drug product, can limit practical usage and commercial marketability of the product. In this paper, a systematic approach is presented to develop a lyophilized high concentration protein formulation. The focus is on achieving reasonable reconstitution times with multidisciplinary strategies. Many strategies have been shown to provide nominal improvement in reconstitution times, such as adding wetting agents in the diluents, incorporating high annealing steps in the lyophilization cycle and reconstituting under vacuum. The reconstitution strategy of reduced diluent volume, however, has enabled significant decrease in reconstitution time (4–7-fold) of lyophilized high protein concentration formulations. [Copyright &y& Elsevier]

Abstract: Since the anthrax spore bioterrorism attacks in America in 2001, the early detection of Bacillus anthracis spores and vegetative cells has gained significant interest. At present, many polyclonal antibody-based quartz crystal microbalance (QCM) sensors have been developed to detect B. anthracis simulates. To achieve a simultaneous rapid detection of B. anthracis spores and vegetative cells, this paper presents a biosensor that utilizes an anti-B. anthracis monoclonal antibody designated to 8G3 (mAb 8G3, IgG) functionalized QCM sensor. Having compared four kinds of antibody immobilizations on Au surface, an optimized mAb 8G3 was immobilized onto the Au electrode with protein A on a mixed self-assembled monolayer (SAM) of 11-mercaptoundecanoic acid (11-MUA) and 6-mercaptohexan-1-ol (6-MHO) as adhesive layer. The detection of B. anthracis was investigated under three conditions: dip-and-dry, static addition and flow through procedure. The results indicated that the sensor yielded a distinct response to B. anthracis spores or vegetative cells but had no significant response to Bacillus thuringiensis species. The functionalized sensor recognized B. anthracis spores and vegetative cells specifically from its homophylic ones, and the limit of detection (LOD) reached 103 CFU or spores/ml of B. anthracis in less than 30min. Cyclic voltammogram (CV) and scanning electronic microscopy (SEM) were performed to characterize the surface of the sensor in variable steps during the modification and after the detection. The mAb functionalized QCM biosensor will be helpful in the fabrication of a similar biosensor that may be available in anti-bioterrorism in the future. [Copyright &y& Elsevier]

Different single-cell analyses for the detection of antigen-specific T cells based on antigen-triggered induction of cytokine production (elispot, intracellular cytokine staining, cytokine secretion assay, etc.) have been analyzed. In this paper we present the data of a thorough validation of the IFNγ Secretion Assay (ISA, Miltenyi Biotec, Bergisch Gladbach, Germany). In this assay the secreted IFNγ is bound to the cell surface and is then stained as an artificial surface molecule and analyzed by flow-cytometry. The introduction of five quality criteria markedly improved the reproducibility of this assay and made it very reliable (intra-assay variability<5%; inter-assay variability<20%). Recovery experiments further demonstrated that almost 100% of IFNγ+ labeled cells could be detected by this technology. In order to analyze which cell subsets contribute to IFNγ-production, we compared the results obtained in different individuals after VZAg-stimulation. Three different IFNγ-secretion patterns could be discerned. In Pattern 1 there is a predominant and almost equal contribution of T cells and NK cells with a minor contribution of CD3+CD56+ and B cells. Pattern 2, which is most abundant, is characterized by a predominance of NK cells (60–70%). Pattern 3 differs from the previous one in its minor contribution of NK cells. Here T cells predominate the IFNγ secretion. These results clearly demonstrate that the IFNγ+ subset distribution after VZAg-stimulation is not uniform and differs individually. Furthermore, the ISA-technology proves to be very useful in vaccine research. This was demonstrated by testing the IFNγ+ secretion pattern after HBsAg-stimulation in PBMC from HBsAg-vaccinated individuals. [Copyright &y& Elsevier]

Given the key role 4-1BB plays in the stimulation of T cells, humanization of agonistic anti-human 4-1BB monoclonal antibody (mAb) may have important clinical applications. In this paper, we present the humanization of agonistic anti-human 4-1BB mAb, BBK-4, using a phage display library. We first prepared the combinatorial library by incorporating murine and human alternative at positions representing buried residues that might affect the structural integrity of the antigen binding site. Six humanized single chain Fv (scFv) fragments were selected from the combinatorial library expressing phage-displayed humanized scFv. They were found to retain the epitope specificity of the original mAb but had affinities of lower than 1/10 of the original. In spite of the lower affinity, the humanized scFv coated on the surface expanded human peripheral blood mononuclear cells (PBMCs) in MLR similarly to the original mAb in the presence of anti-CD3 mAb. These results suggest that humanized anti-human 4-1BB scFvs can be used as a valuable reagent for clinical application. [Copyright &y& Elsevier]

Recent advancements in antibody-based therapies require the development of an efficient method for generation of monoclonal antibodies (MAbs) against the native form of membrane proteins. We examined DNA immunization followed by a single boost with cells as a protein-free immunization protocol for production of MAbs. Mice immunized with plasmid cDNAs encoding human CD30 or Ret tyrosine kinase were given a single boost with cells expressing the corresponding antigen prior to cell fusion. A total of nine cell fusion experiments revealed that the cell boost is necessary for efficient generation of hybridomas and the DNA-cell boost method gave good yields of specific MAbs (5–59 MAbs from one mouse). All IgG isotypes except IgG3 were generated, although IgG2a was the dominant isotype. All the MAbs reacted with native antigens expressed on cells in a fluorescence-activated cell sorter (FACS) analysis as well as with recombinant CD30 or Ret protein genetically fused with human Fc in an enzyme-linked immunosorbent assay (ELISA). The affinities of the anti-CD30 MAbs to CD30-Fc protein ranged from 0.9 to 12.4 nM Kds, which were comparable to existing MAbs to these proteins, which range from 3.0 to 13.0 nM. Western blot analysis and topographical epitope mapping experiments based on the mutual competition of pairs of the anti-CD30 MAbs revealed that about 40% of the epitopes were linear epitopes and that each epitope was topographically classified into one of six groups. The large number of MAbs that react with high affinities to a variety of epitopes on the native form of antigens indicates that the method presented in this paper could be generally useful for generating MAbs to other membrane proteins. [Copyright &y& Elsevier]

• Measurement of Pluronic F-68 concentration in samples with high protein content. • Mixed-mode chromatography with charged aerosol detection. • Impurity clearance study for monoclonal antibody drug. Pluronic F-68 is a nonionic surfactant, which is often used in the upstream process of biopharmaceutical production. However, the number of analytical methods developed for determination of Pluronic F-68 in the in-process and drug substance samples of biological drug production process is quite low. The lack of chromophore groups on the molecule and the interference caused by the high protein content of the samples hamper analysis. In this paper the development and qualification of a mixed-mode (MM) HPLC method with charged aerosol detection is reported. The method enables the analysis of samples with up to 85 g/L protein concentration. The range of the method was set to 250−500 μg/mL, where it was found to be accurate (89–111 % recovery) and precise (0.8–3.2 % relative standard deviation). The high sensitivity of the method indicates that even lower concentration range can be feasible. The novel method successfully demonstrates Pluronic F-68 clearance during the downstream process of the monoclonal antibody production. [ABSTRACT FROM AUTHOR]