Microscale process evaluation of recombinant biocatalyst libraries: application to Baeyer–Villiger monooxygenase catalysed lactone synthesis

Microscale processing techniques are rapidly emerging as a cost- effective means for parallel experimentation and hence the evaluation of large libraries of recombinant biocatalysts. In this work, the potential of an automated microscale process is demonstrated in a linked sequence of operations comprising fermentation, enzyme induction and bioconversion using three whole-cell biocatalysts each expressing cyclohexanone monoxygenase (CHMO). The biocatalysts, Escherichia coli TOP 10 [pQR239], E. coli JM107 and Acinetobacter calcoaceticus NCIMB 9871, were first produced in 96-deep square well fermentations at various carbon source concentrations (10 and 20 g L⁻¹ glycerol). Following induction of CHMO activity biomass concentrations of up to 6 g_DCW L⁻¹ were obtained. Cells from each fermentation were subsequently used for the Baeyer–Villiger oxidation of bicyclo[3.2.0]hept-2-en-6-one, cyclohexanone and cyclopentanone. Each bioconversion was performed at two initial substrate concentrations (0.5 and 1.0 g L⁻¹) in order to simultaneously explore both substrate specificity and inhibition. The microscale process sequences yielded quantitative and reproducible data for each biocatalyst on maximum growth rate, biomass yield, initial rate of lactone formation, specific biocatalyst activity and bioconversion yield. E. coli TOP 10 [pQR239] was demonstrated to be an efficient biocatalyst showing substrate specificities and substrate inhibition effects in line with previous studies. Finally, in order to show that the data obtained with E. coli TOP 10 [pQR239] at microwell scale (1,000 μL) could be related to larger scales of operation, the process was performed in a 2-L stirred-tank bioreactor. Using conditions designed to enable microwell kinetic measurements under none oxygen-limited conditions, the fermentation and bioconversion data obtained at the two scales showed good quantitative agreement. This study therefore confirms the potential of automated microscale experimentation for the whole-process evaluation of recombinant biocatalyst libraries and the specification of pilot and process scale operating conditions.     

The use of microscale processing technologies for quantification of biocatalytic Baeyer-Villiger oxidation kinetics

Microscale processing techniques would be a useful tool for the rapid and efficient collection of biotransformation kinetic data as a basis for bioprocess design. Automated liquid handling systems can reduce labor intensity while the small scale reduces the demand for scarce materials such as substrate, product, and biocatalyst. Here we illustrate this concept by establishing the use of several microwell formats (96-round, 96-deep square and 24-round well microtiter plates) for quantification of the kinetics of the E. coli TOP10 [pQR239] resting cell catalyzed Baeyer-Villiger oxidation of bicyclo[3.2.0]hept-2en-6-one using glycerol as a source of reducing power. By increasing the biocatalyst concentration until the biotransformation rate was oxygen mass-transfer limited we can ensure that kinetic data collected are in the region away from oxygen limitation. Using a 96-round well plate the effect of substrate (bicyclo[3.2.0]hept-2en-6-one) concentration on the volumetric CHMO activity was examined and compared to data collected from 1.5-L stirred-tank experiments. The phenomenon and magnitude of substrate inhibition, observed at the larger scale, was accurately reproduced in the microwell format. We have used this as an illustrative example to demonstrate that under adequately defined conditions, automated microscale processing technologies can be used for the collection of quantitative kinetic data. Additionally, by using the experimentally determined stoichiometry for product formation and glycerol oxidation, we have estimated the maximum oxygen transfer rates as a function of well geometry and agitation rate. Oxygen-transfer rates with an upper limit of between 33 mmol. L(-1). h(-1) (based solely on product formation) and 390 mmol. L(-1). h(-1) (based on product formation and glycerol oxidation) were achieved using a 96-square well format plate shaken at 1300 rpm operated with a static surface area to volume ratio of 320 m(2). m(-3).     

Large scale production of cyclohexanone monooxygenase from Escherichia coli TOP10 pQR239

The cyclohexanone monooxygenase (CHMO) from Acinetobacter calcoaceticus NCIMB 9871 has been cloned into Escherichia coli in an L-arabinose inducible vector. The recombinant E. coli containing the L-arabinose inducible CHMO was grown at 1.5 litres under controlled conditions to determine the parameters for growth and induction. It was found that induction with 0.1% (w/v) L-arabinose at late logarithmic phase of growth and growth for a further 2.5 to 3 h gave the optimal CHMO titre ( approximately 3500 U.l(-1,) 630 U. g dry cell weight(-1)). High dissolved oxygen concentrations were shown to be deleterious to the CHMO titre. This influenced the strategy for growth and induction, and was optimal when the oxygen uptake rate was maximized but the dissolved oxygen concentration was zero. Finally, a 300 litre scale fermentation was carried out giving a total CHMO titre of >8 x 10(5) U.     

Mechanistic studies of cyclohexanone monooxygenase: chemical properties of intermediates involved in catalysis

Cyclohexanone monooxygenase (CHMO), a bacterial flavoenzyme, carries out an oxygen insertion reaction on cyclohexanone to form a seven-membered cyclic product, epsilon-caprolactone. The reaction catalyzed involves the four-electron reduction of O2 at the expense of a two-electron oxidation of NADPH and a two-electron oxidation of cyclohexanone to form epsilon-caprolactone. Previous studies suggested the participation of either a flavin C4a-hydroperoxide or a flavin C4a-peroxide intermediate during the enzymatic catalysis [Ryerson, C. C., Ballou, D. P., and Walsh, C. (1982) Biochemistry 21, 2644-2655]. However, there was no kinetic or spectral evidence to distinguish between these two possibilities. In the present work we used double-mixing stopped-flow techniques to show that the C4a-flavin-oxygen adduct, which is formed rapidly from the reaction of oxygen with reduced enzyme in the presence of NADP, can exist in two states. When the reaction is carried out at pH 7.2, the first intermediate is a flavin C4a-peroxide with maximum absorbance at 366 nm; this intermediate becomes protonated at about 3 s(-1) to form what is believed to be the flavin C4a-hydroperoxide with maximum absorbance at 383 nm. These two intermediates can be interconverted by altering the pH, with a pK(a) of 8.4. Thus, at pH 9.0 the flavin C4a-peroxide persists mainly in the deprotonated form. Further kinetic studies also demonstrated that only the flavin C4a-peroxide intermediate could oxygenate the substrate, cyclohexanone. The requirement in catalysis of the deprotonated flavin C4a-peroxide, a nucleophile, is consistent with a Baeyer-Villiger rearrangement mechanism for the enzymatic oxygenation of cyclohexanone. In the course of these studies, the Kd for cyclohexanone to the C4a-peroxyflavin form of CHMO was determined to be approximately 1 microM. The rate-determining step in catalysis was shown to be the release of NADP from the oxidized enzyme.     

Microbiological transformations 52.: Biocatalysed Baeyer–Villiger oxidation of 1-indanone derivatives

The microbiological Baeyer–Villiger oxidation of various substituted 1-indanones is described. Three bacterial strains have been explored: an E. coli TOP10 [pQR 239] constructed to overexpress the cyclohexanone monoxygenase (CHMO) of Acinetobacter calcoaceticus NCIMB 9871, an E. coli TOP10 [hapE] strain recently constructed to overexpress 4-hydroxyacetophenone monoxygenase (HAPMO) of Pseudomonas fluorescens ACB and the wild type Pseudomonas sp. NCIMB 9872 strain known to metabolise cyclopentanone. This last strain oxidised some of the proposed substrates, leading to the corresponding lactones with good to excellent yields depending on the aromatic ring substituents.     

Comparison of cyclohexanone monooxygenase as an isolated enzyme and whole cell biocatalyst for the enantioselective oxidation of 1,3-dithiane

Both whole cells of recombinant Escherichia coli TOP10, overexpressing cyclohexanone monooxygenase (CHMO) and isolated cyclohexanone monooxygenase, were used to carry out the enantioselective oxidation of 1,3-dithiane (1) to (R)-1,3-dithiane-1-oxide (2). The two biocatalysts were evaluated under various experimental conditions (e.g., shaken flask or bioreactor; non-bound or resin-adsorbed substrate; different substrate concentrations) in terms of volumetric productivity and enantioselectivity. While productivity was similar in the two cases (up to 0.58 g L⁻¹ h⁻¹), the optical purity of the product was much higher with the isolated enzyme (up to 98% e.e.) than with the whole cell biocatalyst (up to 85% e.e.).     

Cloning and characterization of a cyclohexanone monooxygenase gene from Arthrobacter sp. L661

Cyclohexanone monooxygenase (CHMO), a type of Baeyer-Villiger oxidation, catalyzes the oxidation of cyclohexanone into ɛ-caprolactone, which has been utilized as a building block in organic synthesis. A bacterium that is capable of growth on cyclohexanone as a sole carbon source was recently isolated and was identified as Arthrobacter sp. L661. The strain is believed to harbor a CHMO gene (chnB), considering the high degradablity of cyclohexanone. In order to characterize the CHMO, a chnB gene was cloned from Arthrobacter sp. L661. The deduced amino acids of the chnB gene evidenced the highest degree of homology (90% identity) with the CHMO of Arthrobacter sp. BP2 (accession no. AY123972). The CHMO of L661 was shown to be functionally expressed in Escherichia coli cells, purified via affinity chromatography, and characterized. The specific activity of the purified enzyme was 24.75 μmol/min/mg protein. The optimum pH was 7.0 and the enzyme maintained over 70% of its activity for up to 24 h in a pH range of 6.0 to 8.0 at 4°C. The CHMO of L661 readily oxidized cyclobutanone and cyclopentanone whereas less activity was detected with those of Arthrobacter sp. BP2, Rhodococcus sp. Phi1, and Rhodococcus sp. Phi2, thereby suggesting that the CHMO of L661 evidenced the different substrate specificities compared with other CHMOs. These results can provide us with useful information for the development of biocatalysts applicable to commercial organic syntheses, especially because only a few CHMO genes have been identified thus far.     

Preparative scale Baeyer-Villiger biooxidation at high concentration using recombinant Escherichia coli and in situ substrate feeding and product removal process

An efficient biocatalytic process based on the use of adsorbent resin (in situ substrate feeding and product removal) makes experiments at high substrate concentration possible by overcoming limitations due to substrate and product inhibition. This process was successfully applied to the preparative scale Baeyer-Villiger biooxidation of (-)-(1S,5R)-bicyclo[3.2.0]hept-2-en-6-one (25 g). Whole cells of recombinant E. coli (1 liter) overexpressing cyclohexanone monooxygenase were used as a biocatalyst and the substrate was preloaded onto the adsorbent resin. The corresponding lactone was obtained in 75-80% yield. Time for cell growth and biotransformation is about 24 h each and oxygen supply can be improved by using a tailor-made bubble column.     

Characterization of a recombinant Escherichia coli TOP10 [pQR239] whole-cell biocatalyst for stereoselective Baeyer–Villiger oxidations

This paper describes the kinetic characterization of a recombinant whole-cell biocatalyst for the stereoselective Baeyer–Villiger type oxidation of bicyclo[3.2.0]hept-2-en-6-one to its corresponding regio-isomeric lactones (−)-(1S,5R)-2-oxabicyclo[3.3.0]oct-6-en-3-one and (−)-(1R,5S)-3-oxabicyclo[3.3.0]oct-6-en-2-one. Escherichia coli TOP10 [pQR239], expressing cyclohexanone monooxygenase (CHMO) from Acinetobacter calcoaceticus (NCIMB 9871), was shown to be suitable for this biotransformation since it expressed CHMO at a high level, was simple to produce, contained no contaminating lactone hydrolase activity and allowed the intracellular recycle of NAD(P)H necessary for the biotransformation. A small-scale biotransformation reactor (20 ml) was developed to allow rapid collection of intrinsic kinetic data. In this system, the optimized whole-cell biocatalyst exhibited a significantly lower specific lactone production activity (55–60 μmol min⁻¹ g⁻¹ dry weight) than that of sonicated cells (500 μmol min⁻¹ g⁻¹ dry weight). It was shown that this shortfall was comprised of a difference in the pH optima of the two biocatalyst forms and mass transfer limitations of the reactant and/or product across the cell barrier. Both reactant and product inhibition were evident. The optimum ketone concentration was between 0.2 and 0.4 g l⁻¹ and at product concentrations above 4.5–5 g l⁻¹ the specific activity of the whole cells was zero. These results suggest that a reactant feeding strategy and in situ product removal should be considered in subsequent process design.     

Evaluating the impact of substrate and product concentration on a whole-cell biocatalyst during a Baeyer-Villiger reaction

The presence of high concentrations of substrate or product may impede the optimal functioning of a biocatalyst, more so in the case of whole cell biocatalysts where the metabolic status of the cells may be compromised. In this article we investigate these effects using as an example the Baeyer–Villiger oxidation of racemic bicyclo[3.2.0]hept-2-en-6-one to yield (?)-1(S),5(R)-2-oxabicyclo[3.3.0]oct-6-en-3-one and (?)-1(R),5(S)-3-oxabicyclo[3.3.0]oct-6-en-2-one by CHMO expressed in Escherichia coli TOP10. Multi parameter flow cytometry was used to illustrate that substrate (racemic bicyclo[3.2.0]hept-2-en-6-one) associated cell damage was concentration dependent. One of the two regio-isomeric products [(-)-1(S),5(R)-2-oxabicyclo[3.3.0]oct-6-en-3-one] was also used to identify that product associated cell damage was time dependent. In addition, both substrate and product concentrations affected the observed reaction rate.     

On oxygen limitation in a whole cell biocatalytic Baeyer-Villiger oxidation process

In this article, a recombinant cyclohexanone monooxygenase (CHMO), overexpressed in Escherichia coli has been used to study the oxidation of bicyclo[3.2.0]hept-2-en-6-one to its two corresponding lactones at very high enantiomeric excess. The reaction is a useful model for the study of biocatalytic oxidations to create optically pure molecules. The major limitations to a highly productive biocatalytic oxidation in this case are oxygen supply, product inhibition, and biocatalyst stability. In this article, we investigate the effects of whole cell biocatalyst concentration on the rate of reaction at a range of scales from shake flasks to 75 L bioreactors. At low cell concentrations (<2 g(dcw)/L) the maximum specific rate (0.65 g/g(dcw).h) is observed. However, at higher cell concentrations (> 2 g(dcw)/L), the reaction becomes oxygen limited and both the specific rate and absolute rate decrease with further increases in cell concentration. The role of oxygen limitation in reducing the rate of reaction with scale was investigated by increasing the maximum oxygen transfer rate in the reactor at a high cell concentration and observing the increase in product formation rate. We propose a qualitative model demonstrating the relationship between oxygen limitation, biocatalyst concentration, and the rate of reaction. This conceptual model will be a useful guide in the industrial scale-up of whole cell mediated Baeyer-Villiger biocatalysis.     

A generic hierarchical screening method for the analysis of microscale refolds using an automated robotic platform

The refolding of protein derived from inclusion bodies is often characterized by low yields of active protein. The optimization of the refolding step is achieved empirically and consequently is time-consuming slowing process development. An automated robotic platform has been used to develop a dilution refold process-screening platform upon which a hierarchical set of assays rapidly determine optimal refolding conditions at the microscale. This hierarchy allows the simplest, cheapest, and most generic high-throughput assays to first screen for a smaller subset of potentially high-yielding conditions to take forward for analysis by slower, more expensive, or protein specific assays, thus saving resources whilst maximizing information output. An absorbance assay was used to initially screen out aggregating conditions, followed by an intrinsic fluorescence assay of the soluble protein to identify the presence of native-like tertiary structure, which was then confirmed by an activity assay. Results show that fluorescence can be used in conjunction with absorbance to eliminate low-yielding conditions, leaving a significantly reduced set of conditions from which the highest yielding ones can then be identified with slower and often more costly activity or RP-HPLC assays, thus reducing bottlenecks in high-throughput analysis. The microwell-based automated process sequence with generic hierarchical assays was also used to study and minimize the effect on redox potential or misfolding, of oxygenation due to agitation, before demonstrating that the platform can be used to rapidly collect data and evaluate different refolding conditions to speed up the acquisition of process development data in a resource efficient manner.     

Baeyer-Villiger oxidations of representative heterocyclic ketones by whole cells of engineered Escherichia coli expressing cyclohexanone monooxygenase

Whole cells of an Escherichia coli strain overexpressing Acinetobacter sp. NCIB 9871 cyclohexanone monooxygenase (CHMO; E.C. 1.14.13.22) have been used for the Baeyer-Villiger oxidation of representative heterocyclic six-membered ketones to probe the potential impact of nitrogen, sulfur and oxygen on the chemoselectivity of these reactions. The fact that all of these heterocyclic systems were accepted as substrates by the enzyme and gave normal Baeyer-Villiger products broadens the synthetic utility of the engineered E. coli strain and emphasizes the chemoselectivity achievable with enzymatic oxidation catalysts.     

Evaluation of cell disruption effects on primary recovery of antibody fragments using microscale bioprocessing techniques

Intracellular antibody Fab' fragments periplasmically expressed in Escherichia coli require the release of Fab' from the cells before initial product recovery. This work demonstrates the utility of microscale bioprocessing techniques to evaluate the influence of different cell disruption operations on subsequent solid–liquid separation and product recovery. Initially, the industrial method of Fab' release by thermochemical extraction was established experimentally at the microwell scale and was observed to yield Fab' release consistent with the larger scale process. The influence of two further cell disruption operations, homogenization and sonication, on subsequent Fab' recovery by microfiltration was also examined. The results showed that the heat‐extracted cells give better dead‐end microfiltration performance in terms of permeate flux and specific cake resistance. In contrast, the cell suspensions prepared by homogenization and sonication showed more efficient product release but with lower product purity and poorer microfiltration performance. Having established the various microscale methods the linked sequence was automated on the deck of a laboratory robotic platform and used to show how different conditions during thermochemical extraction impacted on the optimal performance of the linked unit operations. The results illustrate the power of microscale techniques to evaluate crucial unit operation interactions in a bioprocess sequence using only microliter volumes of feed. © 2010 American Institute of Chemical Engineers Biotechnol. Prog., 2010     

The role of Tyr605 and Ala607 of thimet oligopeptidase and Tyr606 and Gly608 of neurolysin in substrate hydrolysis and inhibitor binding

The physicochemical properties of TOP (thimet oligopeptidase) and NEL (neurolysin) and their hydrolytic activities towards the FRET (fluorescence resonance energy transfer) peptide series Abz-GFSXFRQ-EDDnp [where Abz is o-aminobenzoyl; X=Ala, Ile, Leu, Phe, Tyr, Trp, Ser, Gln, Glu, His, Arg or Pro; and EDDnp is N-(2,4-dinitrophenyl)-ethylenediamine] were compared with those of site-mutated analogues. Mutations at Tyr605 and Ala607 in TOP and at Tyr606 and Gly608 in NEL did not affect the overall folding of the two peptidases, as indicated by their thermal stability, CD analysis and the pH-dependence of the intrinsic fluorescence of the protein. The kinetic parameters for the hydrolysis of substrates with systematic variations at position P1 showed that Tyr605 and Tyr606 of TOP and NEL respectively, played a role in subsite S1. Ala607 of TOP and Gly608 of NEL contributed to the flexibility of the loops formed by residues 600-612 (GHLAGGYDGQYYG; one-letter amino acid codes used) in NEL and 599-611 (GHLAGGYDAQYYG; one-letter amino acid codes used) in TOP contributing to the distinct substrate specificities, particularly with an isoleucine residue at P1. TOP Y605A was inhibited less efficiently by JA-2 {N-[1-(R,S)-carboxy-3-phenylpropyl]Ala-Aib-Tyr-p-aminobenzoate}, which suggested that the aromatic ring of Tyr605 was an important anchor for its interaction with wild-type TOP. The hydroxy groups of Tyr605 and Tyr606 did not contribute to the pH-activity profiles, since the pKs obtained in the assays of mutants TOP Y605F and NEL Y606F were similar to those of wild-type peptidases. However, the pH-kcat/Km dependence curve of TOP Y605A differed from that of wild-type TOP and from TOP Y606F. These results provide insights into the residues involved in the substrate specificities of TOP and NEL and how they select cytosolic peptides for hydrolysis.     

Consortium of fold-catalyzing proteins increases soluble expression of cyclohexanone monooxygenase in recombinant<Emphasis Type="Italic"> Escherichia coli</Emphasis>

The cyclohexanone monooxygenase (CHMO) gene of Acinetobacter sp. NCIMB 9871 was simultaneously expressed with the genes encoding molecular chaperones and foldases in Escherichia coli. While the expression of the CHMO gene alone resulted in the formation of inclusion bodies, coexpression of the chaperone or foldase genes remarkably increased the production of soluble CHMO enzyme in recombinant E. coli. Furthermore, it was found that molecular chaperones were more beneficial than foldases for enhancing active CHMO enzyme production. The recombinant E. coli strain simultaneously expressing the genes for CHMO, GroEL/GroES and DnaK/DnaJ/GrpE showed a specific CHMO activity of 111 units g^–1 cell protein, corresponding to a 38-fold enhancement in CHMO activity compared with the control E. coli strain expressing the CHMO gene alone.     

Enzyme cascade converting cyclohexanol into ε-caprolactone coupled with NADPH recycling using surface displayed alcohol dehydrogenase and cyclohexanone monooxygenase on E. coli

The application of enzymes as biocatalysts in industrial processes has great potential due to their outstanding stereo-, regio- and chemoselectivity. Using autodisplay, enzymes can be immobilized on the cell surface of Gram-negative bacteria such as Escherichia coli. In the present study, the surface display of an alcohol dehydrogenase (ADH) and a cyclohexanone monooxygenase (CHMO) on E. coli was investigated. Displaying these enzymes on the surface of E. coli resulted in whole-cell biocatalysts accessible for substrates without further purification. An apparent maximal reaction velocity V_MAX(app) for the oxidation of cyclohexanol with the ADH whole-cell biocatalysts was determined as 59.9 mU ml⁻¹. For the oxidation of cyclohexanone with the CHMO whole-cell biocatalysts a V_MAX(app) of 491 mU ml⁻¹ was obtained. A direct conversion of cyclohexanol to ε-caprolactone, which is a known building block for the valuable biodegradable polymer polycaprolactone, was possible by combining the two whole-cell biocatalysts. Gas chromatography was applied to quantify the yield of ε-caprolactone. 1.12 mM ε-caprolactone was produced using ADH and CHMO displaying whole-cell biocatalysts in a ratio of 1:5 after 4 h in a cell suspension of OD_578nm 10. Furthermore, the reaction cascade as applied provided a self-sufficient regeneration of NADPH for CHMO by the ADH whole-cell biocatalyst.     

Development, parallelization, and automation of a gas-inducing milliliter-scale bioreactor for high-throughput bioprocess design (HTBD)

A novel milliliter-scale bioreactor equipped with a gas-inducing impeller was developed with oxygen transfer coefficients as high as in laboratory and industrial stirred-tank bioreactors. The bioreactor reaches oxygen transfer coefficients of >0.4 s(-1). Oxygen transfer coefficients of >0.2 s(-1) can be maintained over a range of 8- to 12-mL reaction volume. A reaction block with integrated heat exchangers was developed for 48-mL-scale bioreactors. The block can be closed with a single gas cover spreading sterile process gas from a central inlet into the headspace of all bioreactors. The gas cover simultaneously acts as a sterile barrier, making the reaction block a stand-alone device that represents an alternative to 48 parallel-operated shake flasks on a much smaller footprint. Process control software was developed to control a liquid-handling system for automated sampling, titration of pH, substrate feeding, and a microtiter plate reader for automated atline pH and atline optical density analytics. The liquid-handling parameters for titration agent, feeding solution, and cell samples were optimized to increase data quality. A simple proportional pH-control algorithm and intermittent titration of pH enabled Escherichia coli growth to a dry cell weight of 20.5 g L(-1) in fed-batch cultivation with air aeration. Growth of E. coli at the milliliter scale (10 mL) was shown to be equivalent to laboratory scale (3 L) with regard to growth rate, mu, and biomass yield, Y(XS).     

Automated assay optimization with integrated statistics and smart robotics

The transition from manual to robotic high throughput screening (HTS) in the last few years has made it feasible to screen hundreds of thousands of chemical entities against a biological target in less than a month. This rate of HTS has increased the visibility of bottlenecks, one of which is assay optimization. In many organizations, experimental methods are generated by therapeutic teams associated with specific targets and passed on to the HTS group. The resulting assays frequently need to be further optimized to withstand the rigors and time frames inherent in robotic handling. Issues such as protein aggregation, ligand instability, and cellular viability are common variables in the optimization process. The availability of robotics capable of performing rapid random access tasks has made it possible to design optimization experiments that would be either very difficult or impossible for a person to carry out. Our approach to reducing the assay optimization bottleneck has been to unify the highly specific fields of statistics, biochemistry, and robotics. The product of these endeavors is a process we have named automated assay optimization (AAO). This has enabled us to determine final optimized assay conditions, which are often a composite of variables that we would not have arrived at by examining each variable independently. We have applied this approach to both radioligand binding and enzymatic assays and have realized benefits in both time and performance that we would not have predicted a priori. The fully developed AAO process encompasses the ability to download information to a robot and have liquid handling methods automatically created. This evolution in smart robotics has proven to be an invaluable tool for maintaining high-quality data in the context of increasing HTS demands.