Microwave heating has been used for years in synthetic and medicinal chemistry, but it is just beginning to be applied in combinatorial chemistry.
Catherine Shaffer
Shaffer is a freelance writer based in Ann Arbor, Mich.
Combinatorial chemistry had been hailed as chemistry’s answer to the genomic revolution. Driven by the advent of high-throughput screening, which created a need for faster chemical synthesis and larger compound libraries, combinatorial chemistry (combichem) showed the potential to create hundreds or thousands of new compounds in just a few days. A related technique, multiple parallel synthesis, uses a smaller subset of reaction components to explore the possibilities of a few desired chemical groups. The terms “combinatorial chemistry” and “multiple parallel synthesis” are often used interchangeably. However, combichem has its own share of challenges.
One of the bottlenecks in combinatorial chemistry is the timing of chemical reactions, some of which take hours or days to complete. These slow reactions are often inadequate to feed the demand of today’s high-throughput drug discovery programs. One increasingly popular solution is microwave heating, which has been used in synthetic and medicinal chemistry for years but has been somewhat slower to catch on in combinatorial chemistry. Fortunately, many researchers are finally beginning to take advantage of microwave heating. But with microwave heating as an easy, relatively inexpensive technique with many advantages and no disadvantages, why isn’t everybody using it?
Helen Blackwell, PhD, assistant professor of chemistry at the University of Wisconsin, Madison, uses microwave-assisted chemistry to accelerate the discovery of
| Microwaves and the High-Throughput Combinatorial Laboratory In a report published in the January, 2005 issue of the Journal of Combinatorial Chemistry, Richard Hoogenboom, PhD, and colleagues at the Eindhoven University of Technology in the Netherlands, described the integration of a monomode microwave synthesizer into a high-throughput experimentation workflow. Hoogenboom coupled an Emrys Liberator synthesizer manufactured by Sweden’s Biotage with an ASW2000 synthesis robot. The robot accepted the microwave racks as a custom rack type and executed a three-part workflow. First the robot prepared reaction mixtures by dispensing stock solutions. It then transported the solutions in microwave vials to the microwave synthesizer. Last, it pipetted samples for analysis using gas chromatography and gel-permeation chromatography. The purpose of the experiment was to study the microwave-assisted ring-opening polymerization of 2-nonyl-2-oxazoline. The reaction was carried out in a solvent gradually changed from acetonitrile to dichloromethane at 140 °C. The authors concluded that “the polymerizations under microwave irradiation proceeded significantly faster than conventionally heated polymerizations (at ambient pressure).” This was the first successful living polymerization of 2-oxazolines in dichloromethane under microwave irradiation. It also demonstrated that a microwave synthesizer can be readily integrated into a high-throughput synthesis workflow. |
new probe molecules that help elucidate complex biological mechanisms. She developed a new solid-phase, array platform for chemical synthesis. She describes microwave-assisted chemistry and combinatorial chemistry as “a beautiful combination. People have moved away from doing solid-phase chemistry because of the rate problems. A lot of times the reactions are extremely slow, but the beauty of solid phase is the purification. . . . If you combine microwaves with solid phase, you have the rate benefits and the purification, merging the best of both worlds.”
Combinatorial chemistry had been evolving for a few decades before researchers thought to use microwaves. Bruce Merrifield was probably the first chemist to use iterative reactions to synthesize peptides in 1962 [Journal of the American Chemical Society, vol. 85, p. 2149 (1963)], and about a decade later, Hungarian researcher Arpád Furka pioneered simultaneous synthesis of peptides. The basic idea behind combinatorial chemistry is to combine multiple compounds in a reaction, either in the same or separate vessels, resulting in a library numbering the square of the number of compounds used. For example, reacting 10 compounds would yield a library of 100 compounds. In peptide chemistry, all reaction products are linear peptides resulting from the hydrolysis of amino acids, so possibilities for a reaction of A, B, and C would be nine compounds such as AA, AB, AC, etc. Additional innovations included the “split and mix” or “split and pool” strategy, whereby separate reaction products AA, AB, and AC would be pooled and reacted with a fourth amino acid, E, yielding AAE, ABE, ACE, etc. Through the use of these iterative reactions, many different peptides could be synthesized.
In shifting from peptides to drug-like molecules, an adjustment needed to be made in conceptualizing the reaction products. No longer did a matrix of reactions produce a linear chain of amino acids. Instead, the core of the molecule, or “centroid,” served as a base on which to attach chemical groups, often referred to as adornments. Rather than assembling trains of molecules, researchers developed other strategies for finding potential drug compounds.
The set of all possible organic compounds that could be used as drugs is said to number 1060 or more, a number so vast that it would be impossible to synthesize and test them all, even with combinatorial chemistry and high-throughput screening. Thus, the synthesis of new compounds has been guided by the rule of five devised by Pfizer’s Christopher Lipinski, PhD, which is as follows: five or fewer hydrogen bond donors, 10 or fewer hydrogen bond acceptors, molecular weight less than or equal to 500, and LogP less than or equal to 5. The development of combinatorial chemistry has been optimized universally in the pharmaceutical industry to accommodate these rules, and thus minimize the number of compounds synthesized that have no biological relevance.
The microwave effect
The common, household, domestic microwave was the first laboratory microwave. Microwaves occupy the range of the electromagnetic spectrum from 1,200 to 30,000 MHz. The 2,450 MHz band is used for microwave ovens, which corresponds to a wavelength of about 12.2 cm. This means that in spite of the misleading name, microwaves are quite large compared to other types of radiation. Microwaves affect polar molecules by causing them to rapidly align and realign with the magnetic field. The intense agitation causes rapid heating. Nonpolar species remain unaffected.
There are two basic types of microwave reactors. The common kitchen microwave is a multimode reactor. Because the magnetron emits microwaves in all directions,
 |
the distribution of the electric field is therefore uneven, which can lead to hot spots. To improve temperature control, researchers began modifying domestic microwaves with temperature detectors and controllers. The other type of microwave reactor is a monomode instrument in which the magnetron is separated from the reaction vessel by a narrow “corridor” or waveguide. The monomode instrument has the advantage of having highly focused, even heating without any hot spots or variations in the electric field.
Among other things, microwave irradiation has simplified reactions in the solution phase. In a paper published early this year [Journal of Combinatorial Chemistry, vol. 7, pp. 322-330 (2005)], researcher Peter Wipf, PhD, a professor of chemistry at the University of Pittsburgh, described the development of a library of chemical analogs of C-cyclopropyl-alkylamide 1, a compound with promising anticancer properties similar to tamoxifen. Wipf expanded a collection of 20 allylic amides and C-cyclopropylalkylamides into a 100-member library by means of parallel N-acylation, N-carbamoylation, and N-sulfonation in a “libraries from libraries” strategy. He carried out the synthesis in a single pot in a monomode microwave reactor, decreasing standard reaction times of 12 hours to just minutes. His team has more projects in the pipeline.
At Cephalon Inc., Frazer, Pa., Blanca Martinez, PhD, is setting up a new hit-to-lead discovery group using the newest microwave reactors as a core technology. Martinez, impressed with microwave heating from her experiences at Adolor Corp., Exton, Pa., considers microwaves indispensable in the laboratory. “Once you’ve tried it, there’s no reason to go back. When I was at Adolor, we wouldn’t even try reactions under traditional conditions anymore. The microwave was our starting point.” Martinez prefers the term parallel synthesis to describe the work her team does in taking results from high-throughput screening and bringing them forward to lead development for medicinal chemistry. “Combinatorial chemistry got sort of a bad reputation in the past few years. The first combinatorial libraries that were made were of low quality, with higher numbers of compounds, but very little information about the purity of each compound. The field has evolved.” Even so, she says her peers have been slow to adopt microwave synthesis. “It’s taken a little bit [of time] to spread. There’s a certain kind of chemist that is a little traditional in the way they do things. If I’ve done things in a certain way and I know it works, there’s a reluctance to change.”
Microwave chemistry is also beginning to be a popular topic at conferences around the world focusing on combinatorial chemistry. At the Eurocombi-3 meeting this year in the United Kingdom many of the presentations featured microwave heating. Gerardo Byk, PhD, a professor of chemistry at Bar-Ilan University, Ramat Gan, Israel, presented a proof-of-concept paper showing that microwave heating can be used to assist the microcyclization of peptides, a reaction that Byk says is applicable to combinatorial chemistry. “Under these conditions, the reactions are very fast. That’s why it has potential for combinatorial chemistry. One can plan multiple reactions in very fast times.”
Nonthermal microwave effects
The early modified kitchen microwave instruments often produced inconsistent and poorly reproducible results, leading to speculations that there were “nonthermal
| The History of the Microwave The development of the common microwave oven began with the invention of the magnetron in 1940 by Sir John Randall and H. A. Boot. A magnetron is a device that uses a combination of electric and magnetic fields to generate a spinning cloud of electrons, which then emits microwaves. It wasn’t until 1946 that anyone made a connection between microwaves and the cooking of food. Percy Lebanon Spencer of the Raytheon Corp. invented the first microwave oven by accident when he melted a chocolate bar in his pants pocket while doing experiments on a magnetron. He repeated this phenomenon with popcorn and an egg before concluding that these shortest of radio waves were actually cooking the food. Raytheon commercialized microwave cooking in 1947 with the Raytheon Radarange. The history of organic synthesis might have been altered completely if the first commercialization of microwave heating had been scientific, rather than culinary in nature, if Spencer’s mind had leapt not to cooking, but to chemical reactions. Instead of scientists using modified kitchen ovens to perform laboratory experiments, would the average person now be using modified monomode laboratory ovens to cook food? In that case, a microwave “oven” might more closely resemble a blender or a bread maker, with a reaction vessel filled through the top and a set of buttons and monitors on the outside. |
microwave effects” at work. Since then, the advent of monomode microwaves and multimode instruments with temperature control and monitoring resulted in more consistent results, and the prevailing belief is that nonthermal effects do not exist and are simply the result of uneven microwave dielectric heating.
In any instance where microwaves are used in the laboratory, no matter the scientific discipline, the potential for nonthermal effects exists. One problem in isolating these effects scientifically is that it is not possible to induce such rapid heating without microwaves, so there isn’t an appropriate control for the experiment. The idea of nonthermal microwave effects persists in part because there is a great deal of concern in the popular media over health effects of microwaves and radio waves, particularly exposure to radiation emanated by cell phones.
This does not mean, however, that all nonthermal microwave effects are a myth and not a valid subject for scientific inquiry. “I’ll believe it when I see it. I’ve heard arguments both ways,” says Martinez. “I have not seen convincing proof about the microwave effect yet. That doesn’t mean it’s not true, but I’m a little skeptical at this point.” The issue for combinatorial chemistry is not necessarily whether nonthermal effects exist, but whether they result in uncontrollable or irreproducible experimental results. Fortunately, advances in technology have shown that these complicating factors do not occur.
Square peg, round hole
A more persistent obstacle in the widespread adoption of microwave heating in the combinatorial laboratory is a very practical one. The most popular format for combinatorial chemistry experiments is parallel solution synthesis in 96-well plates. However, multimode microwave instruments result in uneven heating between wells, and monomode instruments do not accommodate a 96-well plate in the reaction chamber. Chemists essentially have a choice between running sequential reactions in an automated single-mode microwave, or running parallel reactions in a multiwell plate in a multimode instrument. To address the issue of reproducibility between these two approaches Jesús Alcázar, PhD, of Johnson & Johnson Pharmaceutical Research and Development, Toledo, Spain, compared the preparation of a set of compounds by both methods. The multimode system included a fiberoptic temperature sensor attached beneath the reaction vessel for improved temperature control. Although there were differences in yield between wells in the center of the multimode plate and the corners, overall reproducibility was satisfactory compared to serial experiments conducted in a monomode instrument. Interestingly, the multimode instrument produced slightly higher yields than the monomode instrument. Writes Alcázar, “. . . the preparation of compounds in multiwell plates under microwave irradiation effectively combines productivity and speed.”
Milestone Inc., Bergamo, Italy, manufactures a multimode microwave with a reaction vessel designed specifically to accommodate microwell plates, as well as a choice of four other reaction vessels. The Microsynth system can accommodate reaction volumes from 250 µL to 1L, while also incorporating a suite of reaction monitoring and control tools. Roy Mirchandani, the company’s president, cites scalability as an advantage of their system. “Other cavities don’t allow for scalability or parallel chemistry. We can give companies the ability to scale existing successful reactions they’ve done.”Mirchandani views microwave-assisted combinatorial chemistry as a large, emerging market. “We’re seeing that [microwave-assisted synthesis] has been completely accepted by the medicinal chemists. In combinatorial chemistry work, using a titer plate, we have a capability not being offered by the broader market.”
Another obstacle to adoption of microwave chemistry most frequently cited by academic scientists is the expense. A typical microwave reactor can cost $15,000, compared to $100 for a hot plate. Additionally, there remain some applications and reactions for which microwave heating is still not appropriate. “We often start relatively cold, and then heat up the reaction,” says Wipf. “You mix things that might be exothermal in a flask, then heat them up. That’s a process that’s not easy in a microwave.”
Technological hurdles and concerns about controllability and reproducibility have most likely hindered the development of microwave-assisted synthesis in the combinatorial chemistry laboratory. However, new products and technologies have overcome these problems and concerns. It’s apparent that microwave heating could enhance many, if not most combinatorial laboratories, and that in the near future, a microwave reactor will be as common there as a microwave oven is in the American kitchen.
Filed Under: Drug Discovery
