Incomplete factorial and response surface methods in experimental design: yield optimization of tRNA(Trp) from in vitro T7 RNA polymerase transcription

Quality by design approach to improve quality and decrease cost of in vitro transcription of mRNA using design of experiments

In vitro transcription (IVT) reaction is an RNA polymerase-catalyzed production of messenger RNA (mRNA) from DNA template, and the unit operation with highest cost of goods in the mRNA drug substance production process. To decrease the cost of mRNA production, reagents should be optimally utilized. Due to the catalytic, multicomponent nature of the IVT reaction, optimization is a multi-factorial problem, ideally suited to design-of-experiment approach for optimization and identification of design space. We derived a data-driven model of the IVT reaction and explored factors that drive process yield (in g/L), including impact of nucleoside triphosphate (NTP) concentration and Mg:NTP ratio on reaction yield and how to optimize the main cost drivers RNA polymerase and DNA template, while minimizing dsRNA formation, a critical quality attribute in mRNA products. We report a methodological approach to derive an optimum reaction design, with which cost efficiency of the reaction was improved by 44%. We demonstrate the validity of the model on mRNA construct of different lengths. Finally, we maximized the yield of the IVT reaction to 24.9 ± 1.5 g/L in batch, thus doubling the highest ever reported IVT yield.

Enzymatic synthesis of RNA standards for mapping and quantifying RNA modifications in sequencing analysis

Synthesis of RNA standards that contain an internal site-specific modification is important for mapping and quantification of the modified nucleotide in sequencing analysis. While RNA containing a site-specific modification can be readily synthesized by solid-state coupling for less than 100-mer nucleotides, longer RNA must be synthesized by enzymatic ligation in the presence of a DNA splint. However, long RNAs have structural heterogeneity, and those generated by in vitro transcription have 3'-end sequence heterogeneity, which together substantially reduce the yield of ligation. Here we describe a method of 3-part splint ligation that joins an in vitro transcribed left-arm RNA, an in vitro transcribed right-arm RNA, and a chemically synthesized modification-containing middle RNA, with an efficiency higher than previously reported. We report that the improved efficiency is largely attributed to the inclusion of a pair of DNA disruptors proximal to the ligation sites, and to a lesser extent to the homogeneous processing of the 3'-end of the left-arm RNA. The yields of the ligated long RNA are sufficiently high to afford purification to homogeneity for practical RNA research. We also verify the sequence accuracy at each ligation junction by nanopore sequencing.

Optimizing Chitin Depolymerization by Lysozyme to Long-Chain Oligosaccharides

Chitin oligosaccharides (COs) hold high promise as organic fertilizers in the ongoing agro-ecological transition. Short- and long-chain COs can contribute to the establishment of symbiotic associations between plants and microorganisms, facilitating the uptake of soil nutrients by host plants. Long-chain COs trigger plant innate immunity. A fine investigation of these different signaling pathways requires improving the access to high-purity COs. Here, we used the response surface methodology to optimize the production of COs by enzymatic hydrolysis of water-soluble chitin (WSC) with hen egg-white lysozyme. The influence of WSC concentration, its acetylation degree, and the reaction time course were modelled using a Box–Behnken design. Under optimized conditions, water-soluble COs up to the nonasaccharide were formed in 51% yield and purified to homogeneity. This straightforward approach opens new avenues to determine the complex roles of COs in plants.

Investigation of Combined Cyclodextrin and Hydrogel Formulation for Ocular Delivery of Dexamethasone Acetate by Means of Experimental Designs

Dexamethasone acetate (DXMa) has proven its efficiency to treat corneal inflammation, without a great propensity to increase intraocular pressure. Unfortunately, its poor aqueous solubility, associated with a rapid precorneal elimination, results in a low drug bioavailability and a low penetration after topical ocular administration. The main objective of this study was to improve the apparent aqueous solubility of DXMa using cyclodextrins. First, hydroxypropyl-β-CD (HPβCD) and hydroxypropyl-γ-CD (HPγCD) were used to enhance DXMa concentration in aqueous solution. The β and γ HPCD derivatives allowed the increase of the DXMa amount in solution at 25 °C by a factor of 500 and 1500, respectively. Second, with the aim of improving the persistence of the complex solution after instillation in the eye, the formulations of DXMa-based CD solutions with marketed ophthalmic gels (CELLUVISC^®, GEL-LARMES^®, and VISMED^®) were investigated and optimized by means of special cubic mixture designs, allowing the defining of mixed gels loaded with 0.7% (HPβCD) and 2% (HPγCD) DXMa with osmolality within acceptable physiological range. Finally, in vitro drug release assays from the mixed gels were performed and compared with reference eye drops. Similarly to MAXIDEX^® and DEXAFREE^®, in the case of mixed gel containing HPβCD, more than 90% of the drug was released within 2 h, while in mixed gel containing HPγCD, the release of DXMa was partial, reaching ≈60% in 2 h. This difference will have to be further addressed with ex vivo and in vivo ocular delivery experiments.

Statistical approaches to maximize recombinant protein expression in Escherichia coli: a general review

The supply of many valuable proteins that have potential clinical or industrial use is often limited by their low natural availability. With the modern advances in genomics, proteomics and bioinformatics, the number of proteins being produced using recombinant techniques is exponentially increasing and seems to guarantee an unlimited supply of recombinant proteins. The demand of recombinant proteins has increased as more applications in several fields become a commercial reality. Escherichia coli (E. coli) is the most widely used expression system for the production of recombinant proteins for structural and functional studies. However, producing soluble proteins in E. coli is still a major bottleneck for structural biology projects. One of the most challenging steps in any structural biology project is predicting which protein or protein fragment will express solubly and purify for crystallographic studies. The production of soluble and active proteins is influenced by several factors including expression host, fusion tag, induction temperature and time. Statistical designed experiments are gaining success in the production of recombinant protein because they provide information on variable interactions that escape the "one-factor-at-a-time" method. Here, we review the most important factors affecting the production of recombinant proteins in a soluble form. Moreover, we provide information about how the statistical design experiments can increase protein yield and purity as well as find conditions for crystal growth.

Crystallization of Pseudomonas aeruginosa AmrZ protein: development of a comprehensive method for obtaining and optimization of protein-DNA crystals

The AmrZ protein from the pathogenic bacterium Pseudomonas aeruginosa is a transcription factor that activates and represses the genes for several potent virulence factors, which gives the bacteria a selective advantage in infection. AmrZ was crystallized in complex with DNA containing the amrZ1 repressor binding site. Obtaining crystals of the complex required the integration of a number of well known techniques along with the development of new methods. Here, these processes are organized and combined into a comprehensive method which yielded diffraction-quality crystals. Part of this method included thorough data mining of the crystallization conditions of protein-DNA complexes to create a new directed crystallization screen. An optimized technique for the verification of protein-DNA complexes in crystals is also presented. Taken together, the methods described in this article attempt to streamline the difficult process of obtaining diffraction-quality crystals of protein-DNA complexes through the organization of older methods combined with the introduction of new techniques.

Modeling and optimization of a batch process for in vitro RNA production

RNA molecules are commonly produced in vitro by transcription, utilizing a DNA template, an RNA polymerase enzyme, and nucleoside triphosphate substrates (NTPs). In addition to the full-length RNA molecule coded for by the DNA template, significant amounts of shorter RNA molecules are produced. A simplified model of this complex transcription process is presented, with the shorter RNA molecules lumped into a single pool. The rate equations do not depend on the stoichiometry of the RNA molecule of interest, which facilitates application of the model to other RNA molecules. Optimal initial conditions for batch in vitro RNA transcription to produce a dodecamer RNA containing three different nucleotides have been predicted using the model. The predicted optimal values for equimolar NTPs are 10 to 15 mM initial concentration for each NTP and 50 to 60 mM for magnesium acetate, yielding a maximum final dodecamer concentration of 0.8 +/- 0.1 mM at the 90% confidence interval. Experimental data agree well with the model results. (c) 1997 John Wiley & Sons, Inc. Biotechnol Bioeng 56: 210-220, 1997.

The use of machine learning and nonlinear statistical tools for ADME prediction

Absorption, distribution, metabolism and excretion (ADME)-related failure of drug candidates is a major issue for the pharmaceutical industry today. Prediction of ADME by in silico tools has now become an inevitable paradigm to reduce cost and enhance efficiency in pharmaceutical research. Recently, machine learning as well as nonlinear statistical tools has been widely applied to predict routine ADME end points. To achieve accurate and reliable predictions, it would be a prerequisite to understand the concepts, mechanisms and limitations of these tools. Here, we have devised a small synthetic nonlinear data set to help understand the mechanism of machine learning by 2D-visualisation. We applied six new machine learning methods to four different data sets. The methods include Naive Bayes classifier, classification and regression tree, random forest, Gaussian process, support vector machine and k nearest neighbour. The results demonstrated that ensemble learning and kernel machine displayed greater accuracy of prediction than classical methods irrespective of the data set size. The importance of interaction with the engineering field is also addressed. The results described here provide insights into the mechanism of machine learning, which will enable appropriate usage in the future.

Quantifying epistatic interactions among the components constituting the protein translation system

In principle, the accumulation of knowledge regarding the molecular basis of biological systems should allow the development of large-scale kinetic models of their functions. However, the development of such models requires vast numbers of parameters, which are difficult to obtain in practice. Here, we used an in vitro translation system, consisting of 69 defined components, to quantify the epistatic interactions among changes in component concentrations through Bahadur expansion, thereby obtaining a coarse-grained model of protein synthesis activity. Analyses of the data measured using various combinations of component concentrations indicated that the contributions of larger than 2-body inter-component epistatic interactions are negligible, despite the presence of larger than 2-body physical interactions. These findings allowed the prediction of protein synthesis activity at various combinations of component concentrations from a small number of samples, the principle of which is applicable to analysis and optimization of other biological systems. Moreover, the average ratio of 2- to 1-body terms was estimated to be as small as 0.1, implying high adaptability and evolvability of the protein translation system.

Response surface methodology-based optimisation of agarose gel electrophoresis for screening and electropherotyping of rotavirus

Management of rotavirus diarrhoea cases and prevention of nosocomial infection require rapid diagnostic method at the patient care level. Diagnostic tests currently available are not routinely used due to economic or sensitivity/specificity constraints. Agarose-based sieving media and running conditions were modulated by using central composite design and response surface methodology for screening and electropherotyping of rotaviruses. The electrophoretic resolution of rotavirus genome was calculated from input parameters characterising the gel matrix structure and running conditions. Resolution of rotavirus genome was calculated by densitometric analysis of the gel. The parameters at critical values were able to resolve 11 segmented rotavirus genome. Better resolution and electropherotypic variation in 11 segmented double-stranded RNA genome of rotavirus was detected at 1.96% (w/v) agarose concentration, 0.073 mol l(-1) ionic strength of Tris base-boric acid-ethylenediamine tetraacetic acid buffer (1.4x) and 4.31 h of electrophoresis at 4.6 V cm(-1) electric field strength. Modified agarose gel electrophoresis can replace other methods as a simplified alternative for routine detection of rotavirus where it is not in practice.

Methods for kinetic and thermodynamic analysis of aminoacyl-tRNA synthetases

The accuracy of protein synthesis relies on the ability of aminoacyl-tRNA synthetases (aaRSs) to discriminate among true and near cognate substrates. To date, analysis of aaRSs function, including identification of residues of aaRS participating in amino acid and tRNA discrimination, has largely relied on the steady state kinetic pyrophosphate exchange and aminoacylation assays. Pre-steady state kinetic studies investigating a more limited set of aaRS systems have also been undertaken to assess the energetic contributions of individual enzyme-substrate interactions, particularly in the adenylation half reaction. More recently, a renewed interest in the use of rapid kinetics approaches for aaRSs has led to their application to several new aaRS systems, resulting in the identification of mechanistic differences that distinguish the two structurally distinct aaRS classes. Here, we review the techniques for thermodynamic and kinetic analysis of aaRS function. Following a brief survey of methods for the preparation of materials and for steady state kinetic analysis, this review will describe pre-steady state kinetic methods employing rapid quench and stopped-flow fluorescence for analysis of the activation and aminoacyl transfer reactions. Application of these methods to any aaRS system allows the investigator to derive detailed kinetic mechanisms for the activation and aminoacyl transfer reactions, permitting issues of substrate specificity, stereochemical mechanism, and inhibitor interaction to be addressed in a rigorous and quantitative fashion.

RNA structure determination by NMR

This chapter reviews the methodologies for RNA structure determination by liquid-state nuclear magnetic resonance (NMR). The routine production of milligram quantities of isotopically labeled RNA remains critical to the success of NMR-based structure studies. The standard method for the preparation of isotopically labeled RNA for structural studies in solution is in vitro transcription from DNA oligonucleotide templates using T7 RNA polymerase and unlabeled or isotopically labeled nucleotide triphosphates (NTPs). The purification of the desired RNA can be performed by either denaturing polyacrylamide gel electrophoresis (PAGE) or anion-exchange chromatography. Our basic strategy for studying RNA in solution by NMR is outlined. The topics covered include RNA resonance assignment, restraint collection, and the structure calculation process. Selected examples of NMR spectra are given for a correctly folded 30 nucleotide-containing RNA.

A minimal TrpRS catalytic domain supports sense/antisense ancestry of class I and II aminoacyl-tRNA synthetases

The emergence of polypeptide catalysts for amino acid activation, the slowest step in protein synthesis, poses a significant puzzle associated with the origin of biology. This problem is compounded as the 20 contemporary aminoacyl-tRNA synthetases belong to two quite distinct families. We describe here the use of protein design to show experimentally that a minimal class I aminoacyl-tRNA synthetase active site might have functioned in the distant past. We deleted the anticodon binding domain from tryptophanyl-tRNA synthetase and fused the discontinuous segments comprising its active site. The resulting 130 residue minimal catalytic domain activates tryptophan. This residual catalytic activity constitutes the first experimental evidence that the conserved class I signature sequences, HIGH and KMSKS, might have arisen in-frame, opposite motifs 2 and 1 from class II, as complementary sense and antisense strands of the same ancestral gene.

Improving marginal crystals

The physical chemistry of crystal growth can help to identify directions in which to look for improved crystal properties. In this chapter, we summarize how crystal growth depends on parameters that can be controlled experimentally, and relate them to the tools available for optimizing a particular crystal form for crystal shape, volume, and diffraction quality. Our purpose is to sketch the conceptual basis of optimization and to provide sample protocols derived from those foundations. We hope to assist even those who chose not to use systematic methods by enabling them to carry out rudimentary optimization searches armed with a better understanding of how the underlying physical chemistry operates.

Precipitation diagram and optimization of crystallization conditions at low ionic strength for deglycosylated dye-decolorizing peroxidase from a basidiomycete

The growth of suitably sized protein crystals is essential for protein structure determination by X-ray crystallography. In general, crystals are grown using a trial-and-error method. However, these methods have been modified with the advent of microlitre dispensing-robot technology and of protocols that rapidly screen for crystal nucleation conditions. The use of one such automatic dispenser for mixing protein drops (1.3-2.0 microl in volume) of known concentration and pH with precipitating solutions (ejecting 2.0 microl droplets) containing salt is described here. The results of the experiments are relevant to a crystallization approach based on a two-step procedure: screening for the crystal nucleation step employing robotics followed by optimization of the crystallization conditions using incomplete factorial experimental design. Large crystals have successfully been obtained using quantities as small as 3.52 mg protein.

Optimization of apolipoprotein B mRNA editing by APOBEC1 apoenzyme and the role of its auxiliary factor, ACF

Expression and purification to homogeneity of the apolipoprotein B mRNA editing subunit, APOBEC1, has allowed the demonstration that this apoenzyme has considerable residual enzymatic activity on a minimal apoB mRNA substrate, even in the absence of any auxiliary factors. Assay of this activity as a function of various experimental conditions has led to substantial optimization of assay conditions through the use of incomplete factorial and response surface experiments. Surprisingly, the apoenzyme is thermostable, and has a temperature optimum near 45 degrees C. We have used these optimized conditions, to assess steady-state kinetic parameters for APOBEC1 mRNA editing activity with and without the auxiliary factor, ACF. An important effect of the auxiliary factor is to broaden the temperature range of APOBEC1 activity, lowering the optimal temperature and enabling it to function optimally at lower temperatures. A model consistent with this observation is that at lower temperatures ACF promotes a conformational transition in the RNA substrate that occurs spontaneously at higher temperature. Notably, the substantial RNA editing activity of APOBEC1 alone may be responsible for the "hyperediting" observed upon overexpression of APOBEC1 in transgenic mice.

Preparation of active tRNA gene transcripts devoid of 3'-extended products and dimers

Significant amounts (10-30%) of 3'-extended products with one or two extra nucleotides are synthesized in the course of run-off tRNA gene transcription with T7 RNA polymerase. Denaturing polyacrylamide gel electrophoresis appeared to be insufficient to provide preparative amounts of pure correct-size transcripts. Formation of dimers by tRNA gene transcripts as side products in the course of their activation is also another obstacle in preparation of biologically active transcripts. Here, we have shown that EF-Tu affinity chromatography and/or non-denaturing electrophoresis are simple and efficient tools for isolation of highly active correct-size transcripts. Conditions for transcript activation in vitro should be carefully controlled to prevent dimer formation and obtain reliable data on tRNA transcript structure and function.

Investigation of refolding condition for Pseudomonas fluorescens lipase by response surface methodology

Response Surface Methodology (RSM) was used to investigate optimal refolding conditions of Pseudomonas fluorescens lipase which was expressed as inclusion body in E. coli. Three interacting factors, protein concentration, pH and guanidine hydrochloride (GdnHCl) concentration were selected as the variables of RSM. The protein concentration did not affect the refolding yield within the selected range (50-340 micrograms ml-1) at low temperatures of 4 and -15 degrees C, but it was a critical factor at 25 degrees C and refolding yield significantly decreased with increasing protein concentration. The pH and GdnHCl were significant factors in all experimental conditions. But there was no trends of optimal pH depending on temperature and additives, just showing the optimum at neutral pH. Glycerol shifted optimal GdnHCl concentration to higher side, while arginine to lower side. Therefore, it was concluded that glycerol and arginine deprives and supplements GdnHCl, respectively. In this study, RSM made it possible to investigate successfully the optimal conditions of in vitro refolding and to elucidate interactions between refolding factors with a minimum number of experiments. Under the optimal condition determined by RSM, approximately 90% refolding yield was obtained and it was a 30% increase over the conventional method.