February 15, 2007 (Vol. 27, No. 4)
Determining the Right Pressure, Particle Size, Column, and Solvent for Better Results
High-pressure liquid chromatography (HPLC) remains an important workhorse in drug discovery. Continuing upgrades and enhancements to the technology are paving the way to save time and money, improve speed and sensitivity, and to more accurately detect metabolites implicated in toxicity.
A number of presentations at this year’s “Pittcon Conference,” which will be held in Chicago at the end of this month, will focus on HPLC’s role in drug discovery and development activities.
In today’s cost-sensitive pharmaceutical market, researchers are continually seeking ways to streamline analyses. One way to stretch the HPLC dollar would be to replace an expensive, commonly used solvent, acetonitrile, with the less-costly methanol for analytical and preparative chromatographic analyses.
Peter A. Tate, Ph.D., a Wyeth Research (www.wyeth.com) senior research scientist who will be displaying a poster on HPLC at Pittcon, reports, “We estimated that our group alone could save about $50,000 per year by making this change in our reverse-phase HPLC systems. But one has to look at several parameters before deciding if this strategy can save you money or not. Since methanol is weaker as a solvent, a higher concentration may be needed to separate compounds on a column. Additionally, since methanol has a lower loading capacity, you may need to do more preparative HPLC runs per sample.”
Other considerations that need to be weighed are solute pKa, hydrophobicity, and solubility. According to Dr. Tate, “If you have a very hydrophobic compound that needs a high concentration of acetonitrile, then methanol might not work for you. On the other hand, we can blend small amounts of other strong solvents, such as tetrahydrofuran or isopropyl alcohol, with methanol to boost the organic modifier strength and allow elution of sample in an optimized time frame that would still produce cost savings.”
Dr. Tate also suggests looking at whether your laboratory is a stand-alone or an open access facility. “For separations groups with trained chromatographers, this would be a noninvasive change, but for an open-access lab, we need to be sure this change is seamless and that all our chemists can get their preparative separations done without further complicating the chromatography.”
Dr. Tate’s take-home message is that “methanol should be considered as a viable alternative to acetonitrile, but you must carefully examine cost versus savings by comparing all of these parameters.”
Enhancing Speed and Sensitivity
A new version of HPLC, ultra-high-pressure liquid chromatography or U-HPLC, is sparking explosive interest in the scientific community, especially those in drug discovery, according to Ed Long, LC and LC/MS marketing manager, Thermo Fisher Scientific (www.thermo.com). His colleagues will put up a poster on the novel technique at Pittcon.
“Many scientists are finding compelling benefits to this technology, almost as if it’s a magic elixir, because it can be not only faster, but better. Conceptually, both HPLC and U-HPLC are based on similar principles of instrumentation and chromatographic separation science,” explains Long. What has changed is the use of small-particle columns (silica-based packing with particle dimensions of less than two microns in diameter).
“These require higher-pressure systems to suitably handle the higher backpressures. With Fast LC or U-HPLC, however, chromatography separations can now be reduced to run times in seconds instead of minutes or hours. This dramatically increases the number of samples processed per unit time, increasing chromatographic throughput.”
Long notes that the use of smaller particles and/or shorter columns can lead to increased sensitivity and better resolution of components. “The smaller particles provide increased theoretical plate generation, which results in higher efficiencies over a broader range of mobile phase linear velocities. The high efficiency results in more peak capacity, or peaks per unit time, with sharper peaks that typically generate higher signal-to-noise values. Many pharmaceutical companies have been the early adopters of the U-HPLC or Fast LC technology. By exploiting improved sensitivity and resolution, these systems have been applied to difficult separations and small molecule analysis, including impurity profiling, metabolite screening and identification, and pharmaceutical development.”
When laboratories transfer their methods to these newer and faster systems, Long suggests that care must be taken to ensure that operating flow rates, gradient profiles, and injection volumes are appropriately scaled to obtain an equivalent or superior separation. “Almost everyone will have to make some changes, but this impact depends on where you are in the process. For some, this can be direct and simple, but others, such as in QA/QC, will need to prepare thorough documentation. However, at the end of the day, the benefits for doing this far outweigh the time spent performing method transfer.”
In addition to being a complete supplier of these technologies, Thermo Fisher now offers Turboflow™ technology. Rohan Thakur, Ph.D., strategic marketing manager, says, “Turboflow is a technology that provides a significant benefit for atmospheric-pressure-ionization-based bioanalysis or LC-MS/MS. TurboFlow columns thoroughly clean up protein debris and other macromolecules responsible for causing ion suppression from biological samples. So, it’s a valuable and synergistic technology to minimize sample preparation, enhance sensitivity, and reduce matrix effects.”
For scientists preferring to work within the HPLC environment, upgrades and enhancements are available. Curtis R. Campbell, Ph.D., product manager for Shimadzu Scientific Instruments (www.shimadzu.com), says, “Scientists can benefit from new and improved HPLC systems and specialty software programs. These programs are wrappers for traditional chromatography data systems and allow high-quality data to be collected by less technically savvy users.
“Shimadzu offers one example in Prominence MD, a method-development program. This system lends itself to drug discovery because it allows the user to submit many screening runs with just a few mouse clicks—easily selecting from up to ten different column and mobile-phase combinations for these runs. The system automatically switches between columns, mobile phases, etc. and monitors the system until such time that it is re-equilibrated before moving to the next run. Once the run is complete, it is automatically emailed to the user for evaluation.”
Dr. Campbell notes that scientists can also improve separation and analysis through the choice of columns. “For example, HPLC performance can be pushed even further using the Shimadzu XR 2.2-µm particle-size columns.
“The choice of this particle size, along with stringent QC on the silica support for optimal packing efficiency, gives performance equal to sub-2-µm particle columns at system pressures of conventional HPLC equipment, expanding the capabilities of existing lab equipment. The Prominence Series HPLC is ideal for this kind of application.”
Overcoming Idiosyncratic Toxicity
Few drug-development surprises can be as devastating as idiosyncratic toxicity. Variations in the roles of human-drug-metabolizing enzymes may be subtle. This can mask evidence of toxicity problems during preclinical safety studies or even in large clinical trials, but if the side effects are serious, it can result in product withdrawal.
New and improved approaches are needed for discovering such problems at an early stage in order to guide medicinal chemistry to better optimize lead compounds, suggests Zhengyin Yan, Ph.D., principal scientist in the drug discovery division, Johnson & Johnson Pharmaceutical Research & Development (www.jnjpharmarnd.com), which will be making a presentation at a Pittcon session.
Dr. Yan and colleagues are working toward that end. Key players in this arena are the Cytochrome P450s (CYPs), a family of phase I liver enzymes that catalyze the primary metabolism of most drugs.
“As a general approach to detecting reactive metabolites generated by CYPs, drug candidates are usually incubated with liver microsomes in the presence of glutathione, and the reactive metabolites that are formed are trapped to form stable GSH adducts that are subsequently analyzed by liquid chromatography-tandem mass spectrometry,” explains Dr. Yan.
“To make the analysis of MS faster, collision-induced, neutral-loss MS scanning (CID-NL-MS) has been utilized to detect whether GSH adducts form in microsomal incubations, since all GSH adducts undergo a common neutral loss of 129 Da (the g-glutamyl moiety) in the collision-induced dissociation (CID). Because the neutral loss of 129 Da is not exclusive for GSH adducts, this approach suffers from low sensitivity and unreliability. Therefore, a subsequent CID MS/MS is always necessary for further verification.”
High-content screening represents an improved approach that would allow analysis of hundreds of compounds per day. Dr. Yan notes, “This level of throughput is a not big deal for a particular biological assay, but it is a great challenge in LC-MS analysis. In order to accomplish high-throughputs, we use a technique called stable isotope trapping. In the current method, a mixture of natural and stable-isotope-labeled glutathione is used to trap reactive metabolites generated in incubations, and the formed GSH adducts are analyzed by MS/MS neutral-loss scanning.
“Because both natural and stable-isotope-labeled GSH adducts undergo a common neutral loss of the g-glutamyl moiety and thus exhibit a discernible isotope doublet in NL-MS/MS spectra, the isotopic pattern is highly unique. Therefore, it can be easily used as a MS signature to identify reactive metabolites in the form of GSH adducts. The isotopic MS signature is less dependent on the signal-to-noise ratio and can be applied for the detection of reactive metabolites at low levels.
More importantly, the isotopic MS signature is consistent, and spectral analyses can be performed automatically using computer-assisted MS pattern recognition. Therefore, fully automated identification of reactive metabolites can be accomplished in a truly high-throughput fashion.”
Dr. Yan’s take-home message is that “more basic research is required in this area in order to develop better assays to evaluate reactive metabolites and understand the relationship between reactive metabolites and idiosyncratic toxicity. The advancement of this research will allow the earlier detection of toxicities, which would reduce later-stage clinical failures and decrease the risk of seeing an adverse event post-launch.”
Metabolite Analysis with Flow Cells
Identification and characterization of metabolites derived during drug discovery can present challenges due to the limited quantities usually available for analysis and the difficult chemical properties they may possess.
Darwin Asa, Ph.D., marketing director at ESA Biosciences (www.esainc.com) and a poster presenter at Pittcon, says that ESA is tackling this bottleneck with their new instrumentation. “We’ve developed a family of electrochemical (EC) flow cells that can be coupled to HPLC and MS and used to generate microgram quantities of artificial metabolites. Our EC cells are especially useful for scientists studying drug metabolism.
“Normally, the human enzyme Cytochrome P-450 drives the enzymatic oxidation of compounds in the liver. We have demonstrated that electrochemically-derived products often correspond to biological metabolites and chemical degradants. We utilize EC flow cells that are linked with HPLC and MS. Using controlled-potential electrolysis in flowing solutions provides a means for small-scale synthesis of metabolites that can assist analysis in drug discovery, as well as development.”
The company’s experimental setup consists of an HPLC system and in-line EC flow cells specifically engineered with a large surface area. Detection is made via UV and MS. Dr. Asa explains, “We begin with microgram quantities of materials and then can identify the formation of oxidative products. Our new instruments differ from earlier versions in their ability to generate larger quantities of materials that are sufficient to allow even structural identification.”
Dr. Asa notes that the company’s cells are unique because of their porous graphite composition and connections to HPLC instrumentation. “The large surface area essentially acts like a sponge. Controlling the applied voltage to the cells changes the electric potential needed to generate metabolites. This gives scientists in analytical chemistry groups ample material for further analyses.”