Chemistry Highlights 2004

Important advances in chemistry are made each week by researchers in academia, government, and industry. At the end of each year, C&EN editors take a look back at the developments that we’ve reported on during the year and review some of the most significant achievements among them.

We especially seek to identify long-sought or surprising breakthroughs, first-of-a-kind achievements, and findings of potentially wide-ranging influence. We’ve found about 45 developments reported in 2004 that we believe meet those criteria—just shy of one for each issue of C&EN, on average.

Such a list of highlights can never be comprehensive. selections are necessarily subjective, and we hope we haven’t overlooked one or more of your favorites.

This year’s highlighted advances make for an impressive list. Taken together, these developments represent major strides toward attaining a better understanding of our chemical and biological world and point toward novel applications of that knowledge in the future.

BIOCHEMISTRY & MOLECULAR BIOLOGY. In findings with implications for tissue regeneration, Sheng Ding, Peter G. Schultz, and coworkers at Scripps Research Institute discovered reversine--a compound that converts mouse muscle-precursor cells into progenitor cells that can differentiate into a range of tissue types--in a combinatorial library of heterocyclic compounds [J. Am. Chem. Soc., 126, 410 (2004)].

A natural ribozyme that regulates gene expression when it senses the presence of a key metabolite in bacteria represents a new class of RNA-based catalyst, few types of which are known to exist in nature. Ronald R. Breaker and coworkers at Yale University discovered the ribozyme, glmS, at one end of the messenger RNA for an amidotransferase that catalyzes glucosamine-6-phosphate synthesis [Nature, 428, 281 (2004)]. When levels of this sugar get too high, the sugar binds to the ribozyme, inducing the ribozyme to catalyze its own cleavage, and this halts further GlcN6P production. "Similar RNA switches might have played an even greater role in biological sensing before the evolutionary emergence of proteins," Breaker says.

X-ray crystal structures of two different copper-containing metalloenzymes revealed copper adducts of nitric oxide (NO) and oxygen that had not previously been seen in nature. Michael E. P. Murphy, Elitza I. Tocheva, and colleagues at the University of British Columbia, Vancouver, identified a Cu-NO complex in which the NO molecule is bound "side on" to the enzyme's copper atom--a binding mode that is unprecedented in both synthetic and biological systems [Science, 304, 867 (2004)]. And L. Mario Amzel and Sean T. Prigge of Johns Hopkins University School of Medicine and coworkers observed the first Cu-O₂ complex in which an O₂ molecule is bound to copper via just one of its oxygen atoms [Science, 304, 864 (2004)]. The novel copper adducts could represent important catalytic intermediates.

A potential competitor to Monsanto's lucrative Roundup Ready crop system for endowing plants with resistance to the herbicide glyphosate was developed by Linda A. Castle of Verdia, Redwood City, Calif. (now part of Pioneer Hi-Bred, Johnston, Iowa), and coworkers [Science, 304, 1151 (2004)]. Roundup Ready crops are engineered to contain a glyphosate-immune version of an enzyme that plants require to biosynthesize aromatic amino acids. This allows farmers to spray entire fields with Roundup to kill competitive plants without adversely affecting their crops. In their alternative approach, Castle and coworkers engineered transgenic plants to make an enzyme that N-acetylates glyphosate herbicide and thus converts it into harmless N-acetylglyphosate.

Homme W. Hellinga and coworkers at Duke University Medical Center used a combination of computational design and directed evolution to change ribose-binding protein into an enzyme that mimics the natural enzyme triose phosphate isomerase--thus endowing a noncatalytic protein with catalytic activity [Science, 304, 1967 (2004)]. "This is one of the very first times that we have been able to design, essentially from first principles, an enzymatic reaction using structure-based design," Hellinga said--adding that he hoped eventually to be able to design any enzyme at will.

After a long wait, proof of the prion theory finally surfaced this year. Researchers generated artificial prions in bacteria, converted them to amyloid fibrils, demonstrated that the fibrils cause disease in mice (as natural prion fibrils are believed to do), and showed that this infectivity can be propagated from one mouse to another. The results largely substantiate the theory that prion protein alone, and not any conventional DNA- or RNA-based agent, causes transmissible spongiform encephalopathies like mad cow disease and Creutzfeldt-Jakob disease. Stanley B. Prusiner of the University of California, San Francisco, devised the protein-only theory and won the 1997 Nobel Prize in Physiology or Medicine for it, and he and his coworkers also provided the compelling new evidence for it [Science, 305, 673 (2004)].

The first X-ray crystal structure of a cell membrane gas-transport protein--a bacterial ammonia channel protein--was obtained by Robert M. Stroud, Shahram Khademi, and coworkers at the University of California, San Francisco [Science, 305, 1587 (2004)]. The structure reveals the channel's three identical membrane-spanning subunits, each containing a hydrophobic tunnel just wide enough for NH₃ molecules to pass through single file. The study could shed light on the workings of related human proteins and on ammonia toxicity.

This year's Nobel Prize in Chemistry honored Aaron Ciechanover and Avram Hershko of Technion, in Haifa, Israel, and Irwin Rose of the University of California, Irvine, for their discovery, primarily in the 1970s and '80s, of ubiquitin-mediated protein degradation, the regulated cellular process by which unwanted proteins are cleaved into peptides. Protein degradation is now known to play a key role in cell division, DNA repair, protein synthesis, and the immune system. Dysregulation of the process can cause inflammation, muscle atrophy, and cancer.

The 2004 Nobel Prize in Physiology or Medicine, awarded to Richard Axel of Columbia University and Linda B. Buck of Fred Hutchinson Cancer Research Center, Seattle, was also chemistry related. It honored "their discoveries of odorant receptors and the organization of the olfactory system."

The ability of a bifunctional molecule to inhibit a disease-related protein-protein interaction was demonstrated by Jason E. Gestwicki, Gerald R. Crabtree, and Isabella A. Graef at Stanford University Medical School [Science, 306, 865 (2004)]. They showed that a small molecule with one part that binds tightly to a helper protein and another part that interacts with -amyloid peptide prevents aggregation of the peptide into fibrils--a process associated with Alzheimer's disease. Small molecules don't normally have sufficient bulk to block such an interaction, so the team borrowed the surface area of another protein to help the small molecule do the job.

The first detailed structural and mechanistic analysis of a sialyltransferase, a membrane-bound enzyme that transfers sialic acid groups to cell-surface glycoproteins and glycolipids, was achieved by Natalie C. J. Strynadka and Stephen G. Withers of the University of British Columbia; Warren W. Wakarchuk of the Institute for Biological Sciences of the National Research Council, Ottawa; and coworkers [Nat. Struct. Mol. Biol., 11, 163 (2004)]. The study revealed the enzyme's mechanism of action, which could lead to the development of inhibitors as potential therapeutics. The researchers had also solved the structure of one of the enzymes on the sialic acid biosynthesis pathway and were working on others.

DRUG RESEARCH. The first industrial-scale total synthesis of the promising anticancer agent (+)-discodermolide, which has a mechanism of action similar to that of paclitaxel (Taxol), was developed by principal scientist Stuart J. Mickel of Novartis Pharma, Basel, Switzerland, and coworkers, based in part on two earlier total syntheses [Org. Process Res. Dev., 8, 92, 101, 107, 113, and 122 (2004)]. "It's probably the best piece of synthetic work to come out from an industrial company," commented Steven V. Ley of the University of Cambridge. The new total synthesis makes it possible to produce discodermolide in quantities sufficient for clinical trials. Formerly, it could only be obtained from a marine sponge harvested with the help of manned submersibles.

Die, fat, die! A technique in which an apoptosis (cell suicide-inducing) agent is directed specifically to the linings of blood vessels in fat tissue was shown to be effective for reversing obesity in overfed mice [Nat. Med., 10, 625 (2004)]. Mice given the agent while on a high-calorie diet were transformed from rotund to svelte in four weeks--"a consummation devoutly to be wished," as a pudgy Hamlet might have put it. Mikhail G. Kolonin, Renata Pasqualini, and Wadih Arap of the University of Texas M. D. Anderson Cancer Center, Houston, and coworkers "used a powerful tool which, in essence, identifies a zip code for the blood vessels in fat tissue with which to send targeted apoptotic instructions," a researcher explained.

A new technique called DNA display makes it possible to use short DNA sequences to direct the synthesis of large libraries of small organic molecules. The sequences serve as a sort of genetic code for organic chemistry, according to its developers, David R. Halpin, Pehr A. B. Harbury, and coworkers at Stanford University [PLoS Biol., 2, 1015, 1022, and 1031 (2004)]. In the technique, small molecules are assembled on the ends of DNA oligomers in a reaction network. The researchers showed that DNA display could be used to carry out two cycles of in vitro selection (screening for binding or activity) on a library of 1 million nonnatural peptides, yielding a high-affinity protein ligand. They hope to use it to "evolve" compounds with improved properties.

A novel cell-surface-labeling method, in which a nonnatural azido sugar is displayed on cell surfaces from mice and then selectively derivatized with a peptide-phosphine compound, was devised by Carolyn R. Bertozzi and coworkers at the University of California, Berkeley [Nature, 430, 873 (2004)]. The technique could be useful for creating and studying changes in cell surfaces in vivo, perhaps in real time.

Derivatives of the natural product artemisinin are the most potent antimalarial drugs available, but they've been difficult and expensive to synthesize and their bioavailability has been limited. This year, Jonathan L. Vennerstrom of the University of Nebraska Medical Center, Omaha, and coworkers designed and synthesized an ozonide, OZ277, that's more potent and longer lasting than the artemisinins, structurally simple, and amenable to industrial scale-up [Nature, 430, 900 (2004)]. Phase II clinical trials are scheduled to start in early 2005.

A potentially powerful approach was developed to combat a common cause of antibiotic resistance--the incorporation into bacteria of plasmids that contain genes for antibiotic-destroying enzymes [J. Am. Chem. Soc., 126, 15402 (2004)]. Paul J. Hergenrother and coworkers at the University of Illinois, Urbana-Champaign, showed that a small molecule could mimic a process called plasmid incompatibility, causing drug-resistant bacteria to eliminate their resistance-carrying plasmids and making the bacteria susceptible once again to an antibiotic. The type of small molecule they used (an aminoglycoside) tends to be toxic, but the approach could prove to be more broadly applicable.

Faster tuberculosis treatments are urgently needed, but no new antibiotic selective for tuberculosis and other mycobacteria had reached clinical trials in the past 40 years. The first such agent, the diarylquinoline R207910, has now been identified by Koen Andries of Johnson & Johnson Pharmaceutical R&D, Beerse, Belgium, and coworkers [Science, published online Dec. 9, http://dx.doi.org/10.1126/science.1106753]. When administered with other drugs to tubercular mice, it shortened treatment times by about half, and it was well tolerated in human Phase I studies. It acts by ATP synthase inhibition, a mechanism of action never before described for any antibiotic.

ORGANICS & CARBS. "An interesting combination of antiadhesive therapy and receptor-mediated drug delivery that will prompt new ways of thinking about antibiotic design"--that was one researcher's comment on a novel type of carbohydrate-protein conjugate with antibiotic activity. The conjugate, called a glycodendriprotein, was designed and synthesized by Benjamin G. Davis of the University of Oxford, England; Marjorie M. (Kelly) Cowan of Miami University of Ohio, Oxford; and coworkers [J. Am. Chem. Soc., 126, 4750 (2004)]. Its glycodendrimer component targets and binds bacterial surface receptors, and its subtilisin component catalyzes the breakdown of a key bacterial surface protein (adhesin), making the bacteria unable to bind to host cells.

David W. C. MacMillan of California Institute of Technology and coworkers developed a highly flexible and efficient two-step enzymatic technique for carbohydrate synthesis. In this approach, starting materials are derivatized prior to the synthesis, eliminating the tedious and time-consuming protection and deprotection steps typically required in sugar synthesis. The first step is a proline-catalyzed aldol addition of two -oxyaldehydes [Angew. Chem. Int. Ed., 43, 2152 (2004)]. The second step is an aldol addition and cyclization in which a third -oxyaldehyde is added and the product cyclized to form the desired sugar [Science, 305, 175 (2004)].

The first study of the bioactivity of pure, synthetic chondroitin sulfates helped to elucidate their neuronal effects [J. Am. Chem. Soc., 126, 7736 (2004)]. Linda C. Hsieh-Wilson and coworkers at Caltech synthesized two chondroitin sulfate oligomers and found that a chondroitin sulfate tetrasaccharide stimulated neuronal growth and differentiation, whereas a disaccharide did not--suggesting that a four-unit structure may be the smallest chondroitin sulfate capable of neuronal bioactivity. The work points the way toward potential clinical applications of chondroitin sulfates and could lead to a better understanding of neuronal growth and regeneration.

Thiostrepton, a highly acid- and base-sensitive bioactive natural product with 10 rings, 11 peptide bonds, extensive unsaturation, and 17 stereogenic centers, was synthesized by K. C. Nicolaou and coworkers at Scripps Research Institute and the University of California, San Diego [Angew. Chem. Int. Ed., 43, 5087 and 5092 (2004)]. A reviewer called the synthesis "a masterpiece that highlights state-of-the-art technique and opens the field to meaningful structure-activity and mode-of-action studies."

A unique degree of remote stereocontrol in organic synthesis was demonstrated this year by Jonathan Clayden and coworkers at the University of Manchester, England [Nature, 431, 966 (2004)]. They showed that the stereochemistry of a chiral center at one end of amide-substituted xanthenes can influence the stereochemistry at other chiral centers up to 22 bonds away--almost twice the distance previously achieved in a synthetic system. The phenomenon has possible information-processing and nanotechnology applications.

A copper(I)-catalyzed fusion reaction of azides and alkynes to form 1,2,3-triazoles was applied to dendrimer synthesis for the first time and was found to give dendrimer yields higher than those achieved with any other reactions. Dendrimers have traditionally been prepared by stepwise syntheses, which typically give low yields. Valery V. Fokin and K. Barry Sharpless at Scripps Research Institute had earlier applied the copper-catalyzed fusion reaction to biological systems. Fokin and Sharpless, along with Craig J. Hawker of IBM Almaden Research Center, San Jose, Calif., and others, demonstrated that the power of this reaction also extends to materials chemistry by preparing and characterizing high-purity and high-surface-diversity dendrimers in better yields than have been achieved before [Angew. Chem. Int. Ed., 43, 3928 (2004)].

Researchers found that decomposing liquid formic acid over a heated platinum or palladium catalyst makes it possible to carry out lab-scale industrial hydrogenations without having to use potentially dangerous high-pressure gases [Chem. Commun., 2004, 1482]. The hydrogen and supercritical carbon dioxide generated simultaneously in the decomposition reaction are mixed with the material to be hydrogenated. The resulting mixture is then passed over a noble metal catalyst to complete the process. The technology was devised by Martyn Poliakoff's group at the University of Nottingham, England, which has been working with HEL, in Hertfordshire, England, to develop it commercially.

ANALYSIS & SPECTROSCOPY. A new type of microfluidic device that can be used for DNA analysis of complex biological sample solutions (such as blood) was developed by Robin H. Liu, Piotr Grodzinski, and coworkers at Motorola Labs, Tempe, Ariz. [Anal. Chem., 76, 1824 (2004)]. The low-cost, automated device consists of microfluidic mixers, valves, pumps, channels, chambers, heaters, and DNA microarray sensors. It carries out four separate steps required to analyze DNA--sample preparation, polymerase chain reaction, DNA hybridization, and electrochemical detection. Potential applications include point-of-care genetic analysis, environmental testing, and biological warfare agent detection.

Peter Abbamonte at Cornell University (now at Brookhaven National Laboratory) and coworkers demonstrated that electron dynamics can be monitored by applying a new form of mathematical analysis to highly precise X-ray scattering data [Phys. Rev. Lett., 92, 237401 (2004)]. They demonstrated the approach by producing movies of electron motion in water on a 4-attosecond timescale--considerably faster than the approximately 250-attosecond time resolution achieved previously with laser techniques. The new approach could make it possible to study chemical reaction mechanisms in unprecedented detail.

Daniel Rugar and coworkers at IBM Almaden Research Center in San Jose, Calif., achieved an important advance in magnetic resonance imaging this year by showing that the 2-attonewton force associated with the magnetic moment (spin) of a single electron can be measured with a customized magnetic resonance force microscope [Nature, 430, 329 (2004)]. By suspending the magnetic tip of the instrument's specially fabricated silicon cantilever about 100 nm above a silica sample while applying a high-frequency magnetic field, the team was able to use an interferometer to measure small changes in the cantilever's vibrational frequency caused by interactions between an electron spin and the magnetic tip.

Another IBM group showed that a scanning tunneling microscope could be used to change the charge of a single atom on a surface. Jascha Repp of IBM Zurich Research Laboratory, Rüschlikon, Switzerland, and colleagues used an STM to add an electron to a gold atom sitting on a thin insulating layer (creating an ion) and then to remove an electron, returning the atom to its original neutral state [Science, 305, 493 (2004)]. The work opens numerous research avenues, from studying charge states in single atoms and groups of atoms to controlling magnetic moments or catalytic activity.

Desorption electrospray ionization (DESI), the first general approach for mass spectrometric analysis of samples in ambient air (instead of in vacuum) was developed by R. Graham Cooks of Purdue University and coworkers [Science, 306, 471 (2004)]. An ionized stream of solvent from an electrospray source is aimed at a sample surface, releasing sample ions that are then directed to a mass spectrometer for analysis. Potential applications include blood analysis, explosives detection, chemical warfare agent monitoring, drug metabolism studies, and two-dimensional surface analysis.

NANOTECHNOLOGY & MATERIALS. The first working electrical circuit that incorporates both carbon nanotubes and standard silicon technology was devised by a group led by Jeffrey Bokor and Yu-Chih Tseng at the University of California, Berkeley, and Hongjie Dai and Ali Javey at Stanford University [Nano Lett., 4, 123 (2004)]. Such hybrid technologies could ease the commercialization of carbon-nanotube-based products.

A group led by Stig Helveg of Haldor Topsøe A/S and Jens K. Nørskov of the Technical University of Denmark, both in Lyngby, Denmark, used in situ transmission electron microscopy to study gas-solid interactions on surfaces with high temporal and spatial resolution. They obtained angstrom-resolution videos of nanofiber growth and combined the images with theoretical (density functional) calculations to determine a detailed mechanism of the growth process [Nature, 427, 426 (2004)]. The method could be applicable to studies of a wide range of catalytic surface reactions and other molecular processes. Nørskov notes that the work exemplifies a general trend in many scientific studies today--"that theory and experiment go hand in hand in revealing what is going on in complex systems."

Going up--and down! A molecular complex that can function as a nanometer-scale elevator was designed and synthesized by a team including J. Fraser Stoddart of the University of California, Los Angeles, and Vincenzo Balzani and Alberto Credi of the University of Bologna, Italy [Science, 303, 1845 (2004)]. Energy supplied by an acid-base reaction is used to raise and lower a platformlike component of the complex. Next stop: use of the technique to design and construct useful molecular motors and machines.

If you've been looking for the biaxial nematic liquid-crystal phase, your search is over: The long-sought biaxial phase was found this year by Edward T. Samulski of the University of North Carolina, Chapel Hill; Satyendra Kumar at Kent State University in Ohio; and coworkers [Phys. Rev. Lett., 92, 145505 and 145506 (2004)]. Prior nematic liquid crystals had uniaxial symmetry. The biaxial phase "permits one to design a display or device where every pixel is likely to show four extreme optical states," Kumar says. "This is equivalent to having two bits of information at every site for data storage, as opposed to one bit in current technology." Biaxial nematics could thus make it possible to create liquid-crystal displays with faster refresh rates and dramatically lower power consumption.

Nanotube synthesis techniques tend to be complex and involve careful design, preparation, and removal of templates around which the tubes are grown. But this year researchers developed a technique that yields hollow nanotubes in a single step [J. Am. Chem. Soc., 126, 5376 (2004)]. The nanotubes form when two immiscible or poorly miscible liquids emerge coaxially from a pair of concentric needles to which a high voltage is applied. The technique was devised by Ignacio G. Loscertales of the University of Málaga, Spain; Gustavo Larsen of the University of Nebraska, Lincoln; and coworkers. It can be used to make nanotubes from a variety of materials, suggesting potential applications in microfluidics, sensing, field-emission devices, and other areas.

Buffer layer-assisted laser patterning, a versatile method for forming nanometer-scale patterns using combinations of materials that are otherwise difficult to manipulate and pattern, was devised by Micha Asscher and Gabriel Kerner at Hebrew University of Jerusalem [Surf. Sci., 557, 5 (2004)]. In the method, a gaseous buffer layer is used to separate a film of the material to be patterned (such as potassium or gold) from a metallic, semiconductor, or oxide substrate. Laser patterning is then used to form periodic grating-like patterns in the material. Compared to the photolithography methods that currently dominate the semiconductor industry, the new technique is faster, cleaner, far more environmentally friendly, and potentially capable of forming narrower lines at lower cost, Asscher says.

A technique for making electrical contacts to nanometer-sized semiconductor circuit elements using metallic wires of the same dimensions as the semiconductors was devised by Charles M. Lieber and coworkers at Harvard University [Nature, 430, 61 (2004)]. In the technique, segments of silicon nanowire are converted selectively to metallic nickel silicide, and the metalized segments are then attached to semiconductor nanodevices with fine spatial control. Previously, such metal contacts were often hundreds of times larger than the nanotubes or nanowires to which they were linked. The approach could aid the fabrication of a range of nanoscale electronic and optoelectronic devices.

A long-sought method to grow defect-free silicon carbide (SiC) single crystals for use as semiconductor wafers was developed by Daisuke Nakamura and Kazumasa Takatori at Toyota Central R&D Laboratories, in Aichi, Japan, and colleagues [Nature, 430, 1009 (2004)]. They solved the problem by growing SiC in multiple stages. The advance could make it possible for SiC to be used as a more durable replacement for standard silicon semiconductors in a range of electronic devices.

Using a single-walled carbon nanotube as a confining reaction vessel, researchers polymerized the fullerene epoxide C₆₀O to form a linear, unbranched polymer that had never before been observed [Chem. Commun., published online Nov. 18, http://xlink.rsc.org/?DOI=b414247k]. The product is also different from the one obtained when the same derivative is heated in bulk. The encapsulated reaction technique was developed by David A. Britz of the University of Oxford, Andrei N. Khlobystov of the University of Nottingham, and colleagues, who believe it could allow other unprecedented linear polymers to be synthesized.

SUPRAMOLECULAR CHEMISTRY. A macrocyclic molecule having the topology of Borromean rings was prepared by the template-directed self-assembly of 18 components [Science, 304, 1308 (2004)]. Borromean rings, whose use on the crest of the Italian Borromeo family can be traced back to the 15th century, comprise three mechanically interlinked rings that are inseparably united, although no two are linked. When any ring is cut, the other two separate. The work was carried out by J. Fraser Stoddart of the University of California, Los Angeles; Jerry L. Atwood of the University of Missouri, Columbia; and coworkers.

Dipeptides covalently linked to dendritic units (dendrons) were found to assemble into helical structures containing pores that mimic channels produced by pore-forming proteins. The structures provide access to functional synthetic pores that can potentially be used in artificial channels, antimicrobial agents, membranes, and sensing applications. Preparing synthetic mimics of protein pores had been a research goal for decades, but there was previously limited success in achieving it, notes Virgil Percec of the University of Pennsylvania, who led the group that carried out the study [Nature, 430, 764 (2004)].

 ORGANOMETALLIC & INORGANIC CHEMISTRY. Scientists at the Joint Institute for Nuclear Research, Dubna, Russia, and colleagues from Lawrence Livermore National Laboratory synthesized and detected elements 113 and 115 for the first time [Phys. Rev. C, 69, 021601(R) (2004)]. The discovery broadens understanding of the properties of superheavy nuclei and adds to a growing number of experimental findings that support theoretical predictions that some isotopes of elements such as 115 are unusually stable compared with those of other transactinides.

The first stable molecular compound containing a zinc-zinc bond was prepared serendipitously by Ernesto Carmona of the University of Seville, Spain, and coworkers [Science, 305, 1136 (2004)]. X-ray crystallography of the compound--decamethyldizincocene--revealed that it consists of a Zn₂²⁺ unit sandwiched between two C₅(CH₃)₅ rings.

Akira Sekiguchi and coworkers at the University of Tsukuba, Japan, reported the synthesis and characterization of the first stable disilyne--a compound with a silicon-silicon triple bond [Science, 305, 1755 (2004)]. Until this study, the silicon-silicon triple bond had been the only alkene or alkyne analog of heavier group-14 elements (from silicon to lead) that had not been synthesized. A researcher in the field called the work a milestone for both silicon chemistry and multiple-bond chemistry.