Laboratory automation has become a driving force—some say necessity—in solid-phase extraction (SPE), removing one of the bottlenecks in high-throughput analysis, according to researchers presenting at “Pittcon 2007” in Chicago.
Phenomenex (www.phenomenex.com) has developed an automated method development protocol for SPE to remove tetracycline antibiotics from biological samples, which cuts the time for sorbent optimization and extraction from more than a day to one hour, according to Krishna M. Kallury, Ph.D., senior scientist and group leader for sample purification products. The protocol, he says, is universal for acidic, basic, or neutral drugs.
Tetracyclines are broad-spectrum antibiotics that are difficult to extract because of their high polarity and complexing properties. For example, Dr. Kallury said, “they exist in multiple conformations and tautomeric structures and can interconvert from one form to another depending upon pH, making them tough candidates for SPE.”
In the normal course of drug discovery, pharma generates 2,000–10,000 test samples per day. To analyze that many, you need fast extraction. Phenomenex’ solution packs four different sorbents into a single 96-well deep well plate in 10-mg or 30-mg bed masses for optimizing the SPE parameters, said Dr. Kallury.
The Packard Multiprobe 2 liquid-handling system automates liquid introduction, transfer, and withdrawal into the 96-well plate during the load, wash, and elution steps and transfers the eluates from the plate into HPLC vials ready for injection into the LC/MS system.
“The SPE protocol is thus optimized in an hour,” according to Dr. Kallury. “The interactions of tetracyclines with the four sorbents were correlated with their conformations or tautomeric forms at the pH conditions used for the extraction.” In this case, the weak cation exchange sorbent strata-X-CW delivered the cleanest extracts and good recoveries. However, Dr. Kallury stressed that the other sorbents—a neutral styrene-divinylbenzene-based functionalized polymer strata-X, a strong cation exchanger strata-X-C, and a weak anion exchanger strata-X-AW—are important components of the system, as they allow analytes to be selectively plucked from the solution.
SPE in Zero Gravity
At the University of Virginia, a lab, run by James Landers, Ph.D., is automating SPE for biological testing during extended space missions and has developed a fully integrated microfluidic genetic analysis system for unprocessed biological samples, like whole blood. “Astronauts take medications in flight for general pain, congestion, motion sickness, etc.,” said graduate student Daniel J. Marchiarullo.
Unfortunately the medications aren’t always as effective as on earth because space flight causes physiologic changes that alter a drug’s absorption, distribution, metabolism, and elimination properties. Therefore, a small, portable device to monitor the levels of medication in the body is needed, especially for long missions.
Marchiarullo, working with Lakshmi Putcha, Ph.D., at Johnson Space Center, is in the early stages of developing a prototype device to do just that. Ultimately, the device should integrate sample processing and analysis in one unit.
They have succeeded in using SPE to separate the anti-motion sickness medication promethazine and co-extract two hydroxyl free radical formation markers from saliva at recovery rates of 90–100%, reported Marchiarullo. “Concentration enhancements as high as 80-fold have been achieved by collecting only the fraction of eluent with the most analyte. Such high concentrations mean that solvent evaporation and reconstitution aren’t required because the eluent was compatible with electrophoretic separation.”
Landers’ lab already developed a glass microchip with three functional domains for genetic analysis—two for SPE and PCR and one for microchip electrophoresis—and are working on a gating device to control liquid transport through the microchip. The resulting device includes differential channel flow resistances, elastomeric valves, laminar flow, and electrophoretic mobility along with external fluid flow control using a syringe. Using this system, solid-phase extraction, PCR, and microchip electrophoresis, amplicon separation and detection takes less than 30 minutes. “This is one of the first systems with true sample-in-answer-out capability,” Marchiarullo said. It has detected Bacillus anthracis from 750 nL of whole blood from asymptomatic mice and Bordetella pertussis from 1 µL of nasal aspirate from a human patient. “Because the microfluidic device can use nanoliters or picoliters of fluid, quantities of reagents needed are dramatically reduced, which allows the device to be about the size of a shoe box.”
Integrating the microchannel with electrophoresis is in the beginning stages of development and is based on previous work with DNA analysis. Once a working prototype is developed, Marchiarullo said, it will be tested in collaboration with NASA, first in a hypergravity or microgravity environment and perhaps eventually in the space station.
Gilson (www.gilson.com) developed an automated approach to isolate capsaicin (the substance that produces the heat in chili peppers and topical analgesic creams) from a matrix, enhancing HPLC separation methods by automating method development for sample preparation, according to Mark E. Crawford, applications specialist. This allows chemists “to focus on the science, rather than on the mundane processes of manual SPE or HPLC,” he said.
This system is designed for bench work and can run one, four, or eight extractions simultaneously. Typically, SPE runs are 15 minutes, depending on the number of washes and the amount of sample to be captured. Gilson’s method runs in less than eight minutes and makes the wash step unnecessary, explained Crawford.
“The challenge was to optimize the method for several different matrices, while maintaining the same method for all.” To ensure that all of the compound was extracted from the solution, Crawford only collected the results when the capsaicin was present. His tests used hot sauce, pepper spray, two peppers, and a lotion. “The method extracted and eluted all the target analytes well.”
Crawford used a solution of 75% acetonitrile and 25% water to dilute the samples before solid phase extraction. “Acetonitrile incorporates well with lotion, helping draw the hydrophobic capsaicins into solution,” said Crawford. “Ethyl alcohol is effective too.”
Gerstel (www.gerstel.com) has automated sampling for solid-phase microextraction to increase sample throughput, precision, and accuracy, according John R. Stuff. He and his team reconfigured the MPS 2 (MultiPurpose Sampler), developing the MPS 2 PrepStation. The PrepStation is a fully programmable, dual-rail/dual-robot instrument for sample preparation and introduction.
One rail is an automated liquid sample handler that offers dilution, derivatization, heading, rinsing, liquid-liquid extraction, standard addition, spiking, and agitation. The other can be configured to introduce standard and large-volume liquids, headspace injection, twister back-extraction, membrane-assisted solvent extraction, or to perform solid-phase microextraction.
The automation can simplify sample preparation and quantification steps, including derivatization, salt addition, equilibration, and internal standard addition, and has applications in derivatization of drugs, according to Stuff. It offers simplified programming and advanced automation features without the need to write macros, according to Gerstel.
Online Concentration Determination
Abbott Laboratories (www.abbott.com) validated an online method to determine concentrations of a drug of interest in plasma for toxicology studies. Research scientist Azza M. Wagdy, Ph.D., said the method combines SPE with HPLC analysis “to eliminate some of the problems you have offline” and to halve 96-well plate format sample-preparation time from about four hours to about two.
Dr. Wagdy’s method reduces the number of steps required and eliminates some of the common problems, like wells clogged with plasma preventing elution, and an “under-equivalent vacuum that affects the analyte recovery from particular wells,” she said.
Dr. Wagdy and her colleagues separated the compound of interest from dog plasma using a 96-well plate for protein precipitation and transferred the samples and reagents to the SPE cartridge for loading and washing. They were analyzed using the HPLC Tandem Mass Spectrometric method for determining the analyte and stable-labeled, deuterated internal standard.
“Three separate analytical pumps were used,” said Dr. Wagdy. “One delivered the conditioning and washing reagents to the SPE cartridge, another delivered the mobile phase to the analytical column before analysis, and the third delivered reconditioning reagents to the chromatographic column. Switching valves also were engaged.” Using the time program allows one sample to be assayed while another is being loaded.
Results showed the quantitative method “was linear over a concentration range of 6.04 to 3,774.98 nanograms per milliliter, with correlation coefficients greater than 0.99.” The mean percentage bias ranged from 2.7 to 4.3%, and the coefficient variable ranged from 3.5 to 4.6%, showing “the ruggedness of the assay.”
This approach was used to determine the concentration of a confidential Abbott compound in dog plasma to support the analysis of toxicological study samples. “The validated concentration range was targeted specifically to match the toxicology study samples based on the dosage level to minimize reassaying samples below or above the quantifiable limits,” she said. “Although the method was validated at the lower limits of quantitation of 6.04 nanograms per milliliter, it is possible to achieve lower limits of detection by method optimization.”
At the University of Illinois, Urbana-Champaign, Timothy A. Richmond, Ph.D., and colleagues are using individual SPE beads to selectively sample neuropeptides secreted from tissue and individual cells, collecting the neuropeptides near their release sites. “Multiple neuropeptides can be secreted from even a single neuron,” he elaborated. “These heterogeneous brain systems are organized into spatially distinct networks that require sampling approaches that can discriminate peptide release between such regions.” The technique Dr. Richmond is helping to develop collects a range of compounds using the same principles as reverse phase HPLC. “The polymer beads are only about 50 microns in diameter and they can be placed at discrete locations on or around intact cells and tissues.”
The beads act as probes, collecting and preconcentrating analytes, and then, the tissue or cell sample is stimulated “triggering the exocytosis of chemical messenger molecules,” and additional samples are taken for comparison via MS analyses
Additionally, unlike immumohistochemical-based methods, the beads may be coupled to MALDI-TOF MS detection without preselecting the analyte of interest. This method obtains multiple compounds, which are distinguished by their mass to charge ratios, and compares the samples before and during tissue stimulation, yielding chemical information about the secreted compound that includes its sequence, any post-translational modifications that occur during cellular processing, its sites, and the conditions of its release.
Consequently, “our approach allows a directed strategy for discovering novel neuropeptides,” said Dr. Richmond.
He said the method is particularly applicable in neuroscience and works with both invertebrates and vertebrates. “Once the peptides are bound to our SPE probes, the enzymes, which begin to degrade the neuropeptides once they are released from the cells, no longer affect their structures, thus preserving the peptides for the MALDI-TOF detection.”