Quality control in combinatorial chemistry: determination of the quantity,purity, and quantitative purity of compounds in combinatorial libraries

The quality of combinatorial libraries determines the success of biological screening in drug discovery programs. In this paper, we evaluate and compare various methods for measuring identity, purity, and quantity (yield) of combinatorial libraries. Determination of quantitative purity reveals the true library quality and often indicates potential quality problems before full-scale library production. The relative purity can be determined for every member in a large library in a high-throughput mode, but must be cautiously interpreted. In particular, many impurities are not observable by relative purity measurements using detectors such as UV(214), UV(254), and evaporative light-scattering detection. These "invisible" impurities may constitute a significant portion of the sample weight. We found that TFA, plastic extracts, inorganic compounds, and resin washout are among these impurities. With compelling evidence, we reach a conclusion that purification is the only way to remove "invisible" impurities and improve the quantitative purity of any compound even though some compounds may have a high relative purity before purification.     

Measurement of Interferents from Dosages in Metabolite Analyses

Abstract

Polar metabolites of and impurities from dosage solutions may elute similarly by HPLC and thereby yield inaccurate quantitative results. Therefore, impurities collected with eluate fractions containing metabolites from exposed animals may interfere in metabolite analyses. Marine organisms (shrimp) were exposed to ¹⁴C-labeled naphthalene and the resulting metabolites compared to dosage solution impurities using a ³H-labeled naphthalene internal standard obtained from exposed rats. HPLC purification is shown to remove impurities from dosage solutions prior to evaluation of metabolism. Procedures which test both for impurities and for effects of other compounds on metabolite studies are recommended, including criteria for evaluating when metabolism experiments should be conducted with purified solutions to avoid interferences due to impurities.

Determining metabolites of dosage substances is important in toxicity studies¹. For such bioanalytical measurements, trace impurities in exposure solutions may interfere with accurate analyses and lead to erroneous evaluations of metabolism'. If dosage materials are not pure, it is incumbent upon the investigator to devise purification methods appropriate to the task at hand. These needs seem to be well understood in mammalian toxicology¹. However, the special pharmacokinetics shown by aquatic organisms may require special tests for purity and extensive dosage purifications for metabolism studies using aquatic species. Reviewing the aquatic toxicology literature shows that such extensive evaluations and purifications are done only occasionally.

In aquatic organisms moderately polar aromatic metabolites often tend to accumulate while their parent aromatic compounds are eliminated rapidly^2–4. Measurements of metabolic analytes in systems displaying these toxicokinetics are therefore vulnerable to interferences by polar impurities froin dosage compounds. These impurities may mimic analytes, especially if the polar impurities and the metabolites of interest exhibit similar physiochemical properties.

As an example consider a case where 0.5% of a radiolabeled intraperitoneal dosage is polar impurities which are eliminated from an experimental animal with a half-life of 1 week, while the parent compound is eliminated with a half-life of 12 hours. Residual radioactive compounds isolated from an exposed organism after a short depuration, e.g. 60 hours, may have polar physiochemical properties and therefore be measured as metabolites of the dosage parent compound. However, these compounds may include impurities introduced by the dosage solution which were selectively retained because of their polarity. Impurities in this example may comprise only 0.5% of the original radioactivity, but are bioconcentrated to be 10% of the residual radioactivity. Thus, an apparently minor impurity in the dosage may prove to be a significant interferent, and thereby cause errors in analyses.

For toxicologic experiments, reagent evaluation or purification may use a variety of procedures which separate impurities from the radiolabeled compounds to be employed in biological experiments. Often the supplier performs purifications and typically guarantees that reagent purity exceeds 98% or 99%. However, we have found that radiolabeled compounds purchased for aromatic hydrocarbon metabolism studies, even those which contain low concentrations of polar impurities, may not be sufficiently pure for aquatic organism studies⁵. Thus, dosage evaluations must precede toxicologic experiments which are sensitive to small amounts of impurities, e.g., metabolite determinations.

We have observed and measured interferences in chromato-graphic analyses for metabolites⁵ due to small amounts of impurities in experiments, and herein illustrate effective procedures using HPLC which may be used to avoid errors in analyses. Small amounts of dosage solutions may be used for purity assessment. Up to 75 mg of exposure substance may be purified using commercially available analytical HPLC columns. Other advantages of these purifications include: (a) excellent separation of similar compounds due to the high resolution of HPLC, (b) similarity between the purification method used to test and prepare the dosage solution and the HPLC methods used to separate and detect metabolites, and (c) lower cost than performing purifications by other techniques.     

阿维菌素B1a纯物质的制备纯化研究

本研究致力于制备纯化出高纯阿维菌素B1a,为核磁定量及质量平衡法提供计量溯源纯物质,从而给阿维菌素标准物质(研制中)纯度定值时提供纯度参考标准。利用制备型液相色谱对阿维菌素原料(B1a含量为95.71%)进行纯化,除去痕量杂质,真空干燥及冷冻干燥后得到阿维菌素B1a高纯物质。建立了基于制备液相色谱-真空干燥的阿维菌素B1a高纯物质的制备纯化工艺:采用Agilent Prep HT XDB-C18型制备柱,流动相为水/甲醇(15∶85),进样量500μL,流速20.0m L/min。经超高效液相色谱(UPLC)检测,产品纯度达到99.74%,可以满足核磁定量及质量平衡法的要求,并利用液相色谱串联质谱以及核磁共振法对产品进行定性分析。     

A systematic investigation of recovery in preparative reverse phase high performance liquid chromatography/mass spectrometry

In this paper we report a systematic recovery study based on reversed phase high performance liquid chromatography (RP-HPLC) separation and mass spectrometric (MS) based fractionation. Factors including a compound's physicochemical properties, column mass loading and presence of impurities were investigated through commercially available compounds. Results suggest that the delay time between MS peak detection and fraction collection, fraction detector's signal-to-noise ratio and compound's base peak width in the chromatogram have the biggest impacts on purification recovery. In an effort to assess sample recovery within our high throughput purification process, re-purification was performed on four compound libraries that were synthesized in-house. Reproducible recoveries (>80%) were achieved in all tests.     

Developing strategies for isolation of minor impurities with mass spectrometry-directed fractionation

Efficient and automated purification of new chemical entities/potential drug substances and isolation of minor impurities are important aspects of early drug discovery and development strategies, especially when combinatorial synthesis is applied. LC–MS controlled preparative LC and automated fraction collection have been developed for this purpose. The success of such an approach is greatly determined by the quality of the software controlling the application, the coordination between software and hardware, and the reliability of the hardware. The performance of a commercially-available LC–MS controlled autopurification system was evaluated by fractionating four impurities of buspirone as a model compound, eluting closely to the major component under both acidic and basic mobile-phase conditions. A purification strategy for these four components is proposed.     

Evaluation of Multiplexed CE with UV Detection for Rapid pK a Estimation of Active Pharmaceutical Ingredients

The acid dissociation constant (pK _a) is a key physicochemical parameter for characterizing active pharmaceutical ingredients (APIs). Early determination of pK _a values is highly desirable in drug discovery, pharmaceutical process research and formulation design. To overcome the challenges of limited sample availability and potential low purity of API samples at early stages of drug development, as well as to increase sample analysis throughput, a multiplexed 96-channel capillary electrophoresis with UV detection was evaluated as a practical approach for high throughput pK _a estimation of proprietary APIs in support of pharmaceutical research. Proprietary APIs with diverse structures were examined using the approach. The pK _a values were successfully determined with good accuracy and precision. System robustness was demonstrated and analysis of at least eight samples can be completed within 1 h. A rapid pK _a estimation procedure for marginally soluble APIs was proposed by performing single-point multiplexed CE–UV measurement without extrapolation using 10 or 20% methanol as co-solvent. Direct pK _a estimation of APIs using DMSO solution samples and crude reaction samples containing a large amount of solvents and reagents and high level of impurities was also demonstrated using the multiplexed CE–UV approach.     

Integration of supercritical fluid chromatography into drug discovery as a routine support tool. II. investigation and evaluation of supercritical fluid chromatography for achiral batch purification

Supercritical fluid chromatography (SFC) has recently been implemented within our analytical technologies department as a purity assessment and purification tool to complement HPLC for isomer and chiral separations. This report extends the previous work to achiral analysis and purification. This internal evaluation explores the potential impact SFC can have on high throughput, batch purification. Achiral methods have been optimised and batches of compounds purified using a retention time mapping strategy. Here the preparative retention time is predicted from a standard calibration curve and fraction windows set to ensure the peak of interest is collected in one of the four available fraction positions. In this contribution, a completely indirect scale up strategy is applied using totally independent analytical and preparative methods. This novel approach allows for fast analytical purity analysis without compromising the ability to scale up to the preparative system. The benefits and limitations of SFC for batch purification are described in comparison to HPLC across a set of standard compounds and a set of 90 research compounds.     

Residue Analysis of Fluroxypyr-meptyl in Wheat and Soil by GC�CECD

C-Phycocyanin is a natural blue pigment with many commercial applications in foods, cosmetics, and medicines. In this paper we describe the extraction and purification of C-phycocyanin from the cyanobacterium Spirulina platensis. The procedure is based on adsorption of impurities with chitosan and activated charcoal then one-step ion-exchange chromatography. The dry algal powder was soaked in potassium phosphate buffer for 2?h to furnish crude phycocyanin extract of purity 0.93. The crude extract was then treated with chitosan and activated charcoal for 5?min, which increased the purity to 2.78. After chromatography on DEAE Sephadex A-25, the purity of phycocyanin was improved to 4.3. The identity of the purified phycocyanin was confirmed by sodium dodecyl sulfate?Cpolyacrylamide gel electrophoretic analysis and by spectroscopic measurements (UV?Cvisible spectrophotometry and spectrofluorimetry). Compared with conventional methods, this method was simple, inexpensive, and time-saving.     

UV-triggered main-component fraction collection method and its application for high-throughput chromatographic purification of combinatorial libraries

A maximum-seeking, algorithm-driven fraction collection method was developed to support high-throughput chromatographic purification, which provides new possibilities for off-line high-performance liquid chromatography mass spectroscopy (HPLC/MS) quality control experiments. The method is based on manipulation of a six-port valve that is installed downstream from the UV detector and equipped with a fraction collector loop. The detector signal is monitored by a programmable microcontroller that controls the state of the fraction collector valve. After detecting a chromatographic peak, the appropriate fraction is stored in the collector loop. The height of the next peak is compared to the previous one (using a maximum-seeking algorithm) and, depending on the result, the collected fraction is or is not exchanged with the new one. At the end of the run, the stored UV main component is pumped into the external fraction vial. This configuration was used for chromatographic purification of large compound libraries (the results of the purification of 5324 compounds are reported here), as well as for high-throughput off-line HPLC quality control experiments, where the collected main component fractions of an analytical-scale separation were subjected to further mass spectrometric molecular weight verification.     

Optimization strategies for the analysis and purification of drug discovery compounds by reversed-phase high-performance liquid chromatography with high-pH mobile phases

Careful selection of both high-pH mobile phase as well as organic modifier, was performed in order to develop and optimize HPLC conditions for the separation of drug discovery compounds. High-pH mobile phases provide excellent chromatographic resolution and increased mass loading of basic compounds. The analytical methods so defined have been successfully transferred to preparative automated UV-directed purification, an important fact due to the increasing number of samples requiring purification. It should be noted that, the single prerequisite for this approach is an analytical LC-UV-MS run, therefore the system has the ability to collect only fractions likely to contain the target product. A cost-effective strategy for maximizing the purification of drug discovery compounds is proposed.     

High-performance purification of quaternary alkaloids from Corydalis yanhusuo W. T. Wang using a new polar-copolymerized stationary phase

An efficient method for purification of alkaloids from Corydalis yanhusuo W. T. Wang using HPLC was developed, featuring a polar-copolymerized stationary phase named C18HCE. As ionizable solutes, the crude alkaloid sample often suffered from serious peak tailing problem on conventional RP-LC columns, and the separation would rapidly became destroyed with the increasing of load amount. However, on the new stationary phase, good peak shapes (asymmetry factor <1.5) as well as good loadability were easily obtained in a commonly used acidic mobile phase condition. The loading amount could reach 10 mg per injection on an analytical C18HCE column for laboratory-scale purification. About 6.8 mg of palmatine (HPLC purity >98%) and 44.4 mg of dehydrocorydaline (HPLC purity >98%) were rapidly derived from 200 mg of the crude alkaloid sample, and the recoveries of these two compounds were 76.5 and 81.7%, respectively. The purified alkaloids were characterized by comparing retention times with standard compounds as well as (1)H-NMR data. The new method is simple and high yielding, and it may provide a promising tool for purification of alkaloids as well as other alkaline compounds.     

Automated post-collection concentration for purified preparative fractions via solid phase extraction

Recent advancements in preparative HPLC have improved and streamlined compound purification. However, fraction evaporation remains a bottleneck within the process. An alternative to fraction evaporation is to remove the water and reduce the overall volume of the collection by trapping the fraction onto a solid phase extraction (SPE) cartridge. This method (as opposed to analytical applications involving SPE) works by collecting and then diluting the fraction(s), passing the fraction(s) through a SPE, drying the SPE with nitrogen and ultimately eluting the concentrated fraction(s) in a small amount of 100% organic solvent. An appreciable breakthrough is not observed using this method. In addition, recovery from the SPE for the tested compounds rosmarinic acid and carvacrol, two naturally occurring antioxidants in oregano, was found to be 95-98% for a 100mg injection via preparative HPLC purification at 50mg/compound.     

Purification of folic acid candidate reference material by preparative high-performance liquid chromatography

An effective process for the purification of folic acid candidate reference material with preparative high-performance liquid chromatography (Pre-HPLC) was developed in this study. During the process of experimental operation, parameters including the influences of mobile phase, flow rate, and injection volume on the purity and yield were investigated, and the optimized conditions were as follows: the mobile phase was acetonitrile and water in a gradient mode with flow rate of 16?mL/min and injection volume of 2.0?mL at concentration of 10?mg/mL. Under the conditions, the purity and yield of folic acid product were up to 99.4% and 21.0%, respectively, whereas the purity of folic acid raw material was 95.2%. The purified folic acid product was characterized by LC–MS, HPLC, Karl Fischer coulometer, and quantitative nuclear magnetic resonance (qNMR). Results proved that the main component of the product was folic acid and the purities determined by HPLC and qNMR were consistent. Two impurities including N-(4-aminobenzoyl)-L-glutamic acid and pteroic acid were further quantified by LC–MS. Compared with the recrystallization approach, the purity of folic acid obtained by Pre-HPLC increased from 98.2% to 99.4%.     

气相色谱法检测工业用乙二醇纯度及杂质

以Rtx-624色谱柱（30 m×0.32 mm×1.8 μm）为分析柱进行分析，采用校正面积归一化法，建立了检测工业用乙二醇纯度及其中有机杂质的气相色谱分析法。该法可检测传统乙烯法制得的乙二醇中固有杂质二乙二醇、三乙二醇和1，3-二氧杂烷-2-甲醇，同时也适用于检测草酸酯加氢法制得的乙二醇中的新杂质（1，2-丁二醇、1，4-丁二醇、1，2-己二醇、碳酸乙烯酯等）。结果表明，该法具有良好的重复性和较高的检测灵敏度，检出限最低可达0.0002%（质量分数），回收率在91.2%~105.4%之间。该法在乙二醇生产控制、产品检测、市场贸易等过程中具有良好的应用前景。     

Production of milligram quantities of affinity tagged-proteins using automated multistep chromatographic purification

A new chromatography system, AKTAxpress (GE Healthcare, Amersham Biosciences, Uppsala, Sweden) has been designed to meet the demand for high-throughput purification of proteins in structural genomics and drug discovery. The system offers a number of automated multistep purification protocols for affinity-tagged proteins. All protocols start with affinity chromatography followed by combinations of desalting, ion exchange chromatography and gel filtration. As an option, tag removal can be included in the purification protocols. Up to 16 proteins can be purified per day and the yield can be as high as 50 mg of each protein at > 90% purity. To highlight the versatility of the system, this paper presents several case studies; purifications of hexahistidine- and glutathione S-transferase-tagged proteins using different protocols, automated on-column tag cleavage and optimization studies for a hexahistidine-tagged kinase.     

Two‐dimensional hydrophilic interaction chromatography × reversed‐phase liquid chromatography for the preparative isolation of potential anti‐hepatitis phenylpropanoids from Salvia prattii

Traditional Tibetan medicine is important for discovery of drug precursors. However, knowledge of the chemical composition of traditional Tibetan medicines is very limited due to the lack of appropriate chromatographic purification methods. In the present work, Salvia prattii was taken as an example, and an off‐line hydrophilic interaction liquid chromatography/reversed‐phase liquid chromatography preparative method was developed for the purification of phenylpropanoids with high purity from a crude sample of Salvia prattii. Based on the separation results of four different chromatographic stationary phases, the first‐dimensional preparation was performed on an XAmide preparative column with the crude sample concentration of 62.0 mg/mL, and five main fractions were obtained from the 12.4 g crude sample with a recovery of 54.8%. An XCharge C₁₈ preparative column was applied in the second‐dimensional preparation to further isolate the phenylpropanoids from the redissolved first‐dimensional fractions with concentration of approximately 50.0 mg/mL. The purities of the phenylpropanoids isolated from the crude sample of Salvia prattii were higher than 98%, indicating that the method was efficient for the purification of phenylpropanoids with high purity from Salvia prattii. Additionally, this method showed great potential in the preparation of phenylpropanoids and can serve as a good example for the purification of phenylpropanoids from other plant materials.     

Identification of impurities in artemisinin, their behavior in high performance liquid chromatography and implications for the quality of derived anti-malarial drugs

Previous work [1] on the HPLC analysis of artemisinin tentatively identified the two impurities present above trace levels. This identification was based on LC-MS results and NMR of impurities isolated from artemisinin. In this work the impurities have been synthesized allowing verification of their identity by LC-MS. It is found that the previously suggested elution order is incorrect. A determination of relative response factors strongly impacts suggested limits on impurity levels and explains the erroneous peak assignment. The fates of the identified impurities are explored in the transformation of artemisinin to its derivative active pharmaceutical ingredients. A survey of a wide variety of artemisinin samples isolated from different geographical regions, different growing seasons, different plant backgrounds and using different extraction and purification approaches showed that artemisinin has sufficient purity for its intended use as a raw material for anti-malarial drug products.     

New “hyphenated” CPC-HPLC-DAD-MS strategy for simultaneous isolation, analysis and identification of phytochemicals: application to xanthones from Garcinia mangostana

Centrifugal partition chromatography (CPC) coupled online with high-performance liquid chromatography (HPLC) with diode-array detection (DAD) and mass spectrometry (MS) is presented in this work. This strategy offers the possibility to obtain simultaneously CPC fractionation of natural extracts, the HPLC fingerprint of separated fractions and structural information on molecules contained in each fraction. This new approach was applied to the fractionation and purification of xanthones from Garcinia mangostana (Clusiaceae) pericarp. A biphasic solvent system of heptane/ethyl acetate/methanol/water (2:1:2:1, v/v) was used for the CPC separation of 175?mg crude ethanolic extract. The HPLC analysis was conducted with a reversed-phase monolithic column allowing fast and repeatable separation. This combined CPC-HPLC-DAD-MS method led to isolation of 33?mg α-mangostin and 6?mg γ-mangostin at 98?% and 98.5?% purity, respectively, in 140?min. Furthermore, in the same time a total of 16 other xanthones were detected in the extract, and ten of them were identified on the basis of their UV and MS spectra.     

Use of expanded bed adsorption to purify flavonoids from Ginkgo biloba L.

Three techniques (liquid–liquid extraction, packed bed adsorption and expanded bed adsorption) have been compared for the purification of flavonoids from the leaves of Ginkgo biloba L. A crude Ginkgo extract was obtained by refluxing with ethanol for 3 h. The yield of flavonoids achieved by this crude extraction was about 19% (w/w) and the purity of flavonoids in the concentrated extract was between 1.9 and 2.3% (w/w). The crude extract was then dissolved in deionized water and centrifuged where necessary to prepare clarified feedstock for further purification. For the method using liquid–liquid extraction with ethyl acetate, the purity, concentration ratio and yield of flavonoids were 25.4–31.0%, 16–18 and >98%, respectively. For the method using packed bed adsorption, Amberlite XAD7HP was selected as the adsorbent and clarified extract was used as the feedstock. The dynamic adsorption breakthrough curves and elution profiles were measured. For a feedstock containing flavonoids at a concentration of 0.25 mg/mL, the appropriate loading volume to reach a 5% breakthrough point during the adsorption stage was estimated to be 550–600 mL for a packed bed of volume 53 mL and a flow rate of 183 cm/h. The results from the elution stage indicated that the majority of impurities were eluted by ethanol concentrations of 40% (v/v) or below and efficient separation of flavonoids from the impurities could be achieved by elution of the flavonoids with 50–80% ethanol reaching an average purity of ∼25%. The recovery yield of flavonoids using the packed bed purification method was about 60% of the flavonoids present in the clarified feedstock (corresponding to around 30% for the total flavonoids in the unclarified crude extract). For the method using expanded bed adsorption also conducted with Amberlite XAD7HP as the adsorbent, the optimal operation conditions scouted during the packed bed experiments were used but unclarified crude extract could be loaded directly into the column. For an expanded bed with a settled bed height of 30 cm, the loss of flavonoids in the column flow-through was about 30%. The two-step elution protocol again proved to be effective in separating the adsorbed impurities and flavonoids. More than 96% of the bound impurities were completely removed by 40% ethanol in the first elution stage and less than 4% remained in the final product eluted by 90% ethanol in the second elution stage. Also, ∼74% of the adsorbed flavonoids on column (corresponding to 51% of the total flavonoids in the unclarified feedstock) were recovered in the product. In addition to higher recovery yield, the average process time to obtain the same amount of product was decreased in the expanded bed adsorption (EBA) process. The results suggest that the adoption of EBA procedures can greatly simplify the process flow sheet and in addition reduce the cost and time to purify flavonoids from Ginkgo biloba. These results clearly demonstrate the potential for the use of EBA to purify pharmaceuticals from plant sources.