The signs are unmistakable: The pharmaceutical industry hasflipped from its 1990s invincibility to uncertainty, with no easy fix in sight. That spells trouble for the fine chemicals industry, which supplies drugmakers with complex intermediates and active ingredients.

Being able to provide single enantiomers of chiral intermediates and active ingredients quickly and cost-effectively is a potent competitive advantage for fine chemicals companies. Since the 1990s, they have been honing, acquiring, developing, and expanding chiral technologies to respond to the rising demand for chiral compounds, not only from drug companies but also from the agrochemical, food and beverage, and diagnostic and research industries.

According to the market research firm Freedonia Group, demand for chiral raw materials, intermediates, and active ingredients will grow by 9.4% annually between 2000 and 2005. By 2005, it says, the market will be $15.1 billion, with $11.5 billion (76%) going to drug manufacture.

Sandra E. Erb and Jane Zhou, manager of research and senior research associate, respectively, for chiral and fine chemicals consulting at Technology Catalysts International measure the impact of chiral compounds in terms of chiral drugs. Their analysis shows that of the $410 billion in worldwide sales of formulated pharmaceutical products in 2001, $147 billion (36%) was due to single-enantiomer drugs. That number continues the rising sales and market share of single-enantiomer drugs, which were $133 billion (34%) in 2000 and $115 billion (32%) in 1999. Erb and Zhou expect this trend to continue.

RESPIRATORY AND central nervous system (CNS) disorders are among the therapeutic areas in which Erb and Zhou see a major growth in single-enantiomer drugs. For example, the fastest growing new treatment for asthma is the single-enantiomer drug Singulair (montelukast sodium) from Merck, Erb says. Sales of the product in 2001 were $1.38 billion, up 60% from 2000 and 174% from 1999.

Among CNS disorders, the biggest market is depression, Erb says. Until 2000, the leading product, with sales that year of almost $2.6 billion, was Eli Lilly's Prozac (fluoxetine), a racemate. The entry of generics last August eroded sales to about $2.0 billion in 2001. Analysts project that Prozac sales will plunge to $525 million by 2004.

Other antidepressants, however, such as GlaxoSmithKline's Paxil (paroxetine hydrochloride) and Pfizer's Zoloft (sertraline), are single enantiomers. Sales of Paxil in 2001 were almost $2.7 billion, up 20% from 2000, and those of Zoloft were $2.4 billion, up 11% from 1999. Sales are forecast to reach $3.4 billion for Paxil and $2.8 billion for Zoloft by 2004.

The pharmaceutical industry's problem is that new billion-dollar drugs are not obviously in sight. Productivity has been low, despite the almost tripling of R&D spending worldwide from $8.42 billion in 1990 to $30.3 billion in 2001 and the flowering of genomics and combinatorial chemistry.

According to the Pharmaceutical Research & Manufacturers of America (PhRMA), the total number of new drugs approved by the Food & Drug Administration each year has been declining since 1996, and total drug development time has almost doubled from 8.1 years in the 1960s to 14.2 years in the 1990s. Observers say these trends reflect the increasing complexity of the diseases that pharmaceutical companies are now targeting.

Meanwhile, the time it takes for FDA to approve a new drug, after decreasing from about 30 months in the early 1990s to 11.7 months in 1998, has been rising again. In 2001, the average was 16.4 months.

The drain on resources is exacerbated by patent expirations. With drug companies in the doldrums, some fine chemicals companies are wondering if there will be enough products to work on, Erb says.

One strategy to improve the bottom lines of drug companies is to extend the profitable life of products by redeveloping single-enantiomer forms of drugs that had been approved as racemates. The strategy has become a key defensive tool of pharmaceutical companies against generic drugmakers, Zhou points out.

A recent example is the effort of Forest Laboratories to bring out a more potent version of its antidepression drug, Celexa (citalopram), which is a racemate. The R isomer is inactive, and early last year, Forest applied for approval of escitalopram, the S-enantiomer-only drug.

Whether it's switching from a racemate to a single enantiomer, synthesizing new chiral drugs, or improving routes to approved single-enantiomer drugs, suppliers with versatile toolboxes of chiral technologies have the edge. As Erb puts it: "There are so many different ways to get the end results. The issue is what is the most economic way to do it in the fastest time."

In surveying trends in chiral technology, Zhou has seen a definite move toward biocatalysis. That's because it enables transformations in fewer steps--with fewer by-products and lower solvent use--than does traditional chemical synthesis, she says. That the metals in catalysts are potential sources of contamination and negative environmental impact is also a plus for biocatalysis.

Among fine chemicals companies that are heavily invested in biocatalysis is Degussa Fine Chemicals. This year, it achieved a breakthrough in using hydantoinase technology to produce L-amino acids.

"We became interested in hydantoinase technology because it starts from D,L-hydantoins, which are produced by easy, inexpensive, standard chemistry," says Michael Schwarm, head of R&D for pharmaceutical intermediates and exclusive synthesis at Degussa. Hydantoinases convert hydantoins to carbamoylamino acids, which yield amino acids when acted upon by carbamoylases.

	FULL SIZE - CLICK IMAGE

WHAT'S NICE about hydantoins is that they racemize readily, Schwarm says. "When one enantiomer is consumed, the remaining one racemizes and you end up, theoretically, with 100% of one enantiomer. That makes it a very attractive route to L - or D-amino acids," he says. The route to D-amino acids is well established and widely used; Degussa had been running it at ton scale, Schwarm says. But until now, a complementary path to L-amino acids had not been available.

Collaborating with Frances H. Arnold, a chemistry professor at California Institute of Technology, Degussa has developed a fairly selective L-hydantoinase by directed evolution of a wild-type hydantoinase that prefers D-hydantoins. "To our knowledge, this was the first time that the enantioselectivity of an enzyme had been changed by directed evolution," Schwarm says.

Degussa uses the enzyme in conjunction with a racemase, which catalyzes hydantoin racemization, and an extremely selective L-carbamoylase, all housed in modified Escherichia coli cells. "Provided you don't have problems with starting material going into the cells and product leaving the cells, whole-cell biocatalysts are easier to use than pure enzymes," Schwarm points out.

Amino acids that have been produced at a scale of several hundreds of kilograms include L-methionine, L-norleucine, L-2-aminobutyric acid, and L-3-(3'-pyridyl)alanine. "We have many more examples, but they are related to customer projects," Schwarm says. "The whole-cell biocatalyst digests a wide variety of raw materials to make a range of products."

Meanwhile, in Tokyo, Daicel Chemical Industries has expanded its biocatalytic technology for producing chiral alcohols. A recently discovered R-specific secondary alcohol dehydrogenase from Pichia finlandica now complements the S-specific enzyme from Candida parapsilosis, discovered in 1995. With these two enzymes, Daicel now can supply both enantiomers of a chiral secondary alcohol from the same ketone substrate.

In South Africa, researchers at CSIR Bio/Chemtek have developed a process to produce l-menthol from the readily available raw material m-cresol.

Alkylation of m-cresol generates thymol. Hydrogenation of thymol yields four pairs of diastereomers: -menthol, -isomenthol, -neomenthol, and -neoisomenthol. Acylation of this mixture using a stereoselective lipase yields l-menthyl acetate in at least 96% enantiomeric excess (ee). l-Menthyl acetate is separated from the unreacted isomers by distillation. Hydrolysis yields l-menthol.

The enzymatic resolution has been demonstrated in a continuous process at 1 kg per hour. Enzyme activity is retained even after 2,000 hours of operation. Furthermore, isomerization/racemization of the unreacted isomers regenerates the original mixture of diastereomers, which is routed again to enzyme resolution. Over several cycles, thymol is almost fully converted to l-menthol.

	FULL SIZE - CLICK IMAGE

At DSM Fine Chemicals, a stereoselective lipase also plays a key role in multiton-per-year production of an enantiopure secondary alcohol that's initially formed as a racemate. A ruthenium catalyst with proprietary ligands racemizes both enantiomers. But in the same pot, a stereoselective lipase converts only one enantiomer to an ester in high yield and greater than 99% ee. The ester is inert to the metal complex and does not racemize, according to Rinus Broxterman, competence manager for chiral technologies at DSM Fine Chemicals. Hydrolysis to the enantiopure alcohol is nearly quantitative.

Chemist Gerard Verzijl developed the chemistry at DSM research facilities in Geleen, the Netherlands. Now, the process is being scaled up at facilities in Linz, Austria. Full-scale production, for a pharmaceutical customer, is expected to begin this year, Broxterman says. "We think this is the first example of an enantiopure secondary alcohol being produced at industrial scale by dynamic kinetic resolution."

Even as biocatalysis gains ground in chiral chemicals production, research in asymmetric chemical synthesis continues unabated. One active area is immobilization of asymmetric homogeneous catalysts.

Asymmetric homogeneous catalysts are difficult to use in large-scale runs. They are not reusable, and they can contaminate the desired products. Immobilization could solve these problems, as well as open up fine chemicals production to continuous processing. "Productivity with continuous processes is much better than with batch runs, and immobilization is key to continuous processing," says David J. Moody, Avecia's director of new technology ventures.

 AMONG COMPANIES developing and commercializing catalyst-immobilizing technologies is Synetix Chiral Technologies. Methods to anchor catalysts and prevent them from becoming impurities while improving their intrinsic activity are creating valuable products for fine chemicals companies, says Fred Hancock, product manager at Synetix.

	FULL SIZE - CLICK IMAGE

Synetix' technology is based on rigid porous solids formed by controlled hydrolysis of tetraethylorthosilicate in the presence of triethoxysilane or a triethoxyaluminum salt, which provides linking groups. Further chemistry on the resulting powder anchors the catalyst metal or ligands through electrostatic or covalent interactions with the linking groups. The anchored catalyst can be added directly to a reaction mixture or packed in a fixed bed through which substrate and reagents pass. Strong binding of the catalyst to the support prevents metal leaching into product. It also improves turnover number.

Many homogeneous catalysts are deactivated by dimerization of the metal through hydride bridges, causing agglomeration and formation of metallic particles. "Suddenly, a ruthenium hydrogenation catalyst is now a ruthenium mirror in your reactor," Hancock says. "Immobilization anchors that ruthenium away from others so it can't form a dimer. That extends the life of the catalyst, which equates to its being much more recyclable in effect."

Synetix has reached agreements with Rhodia ChiRex and Avecia to apply the technology to these companies' proprietary catalysts. For Rhodia ChiRex, Synetix has successfully immobilized the Co(salen)cobalt catalyst for hydrolytic kinetic resolution, an invention of Harvard University chemistry professor Eric Jacobsen that had been licensed to Rhodia ChiRex. Further work is being done to determine the stability, activity, and recyclability of the immobilized catalyst.

For Avecia, Synetix will be immobilizing catalytic asymmetric cyanohydrin (CACHy) and transfer hydrogenation (CATHy) catalysts. According to Moody, Synetix' technology complements what Avecia already has. He refers to catalyst encapsulation (EnCat) in a polyurea framework, which Avecia uses to immobilize palladium for carbon-carbon cross-coupling reactions. With EnCat, ligands for the metal aren't needed because they are supplied by the polymer itself.

"It is unlikely that a single technology will meet all needs," Moody explains. "With CATHy and CACHy, we're moving in parallel tracks, using Synetix' technology and our own. This approach makes it extremely likely that we will come up with cost-effective manufacturing solutions."

Johnson Matthey is another company that's commercializing catalyst immobilization. Recently, it expanded its portfolio by licensing an invention of Robert L. Augustine, an emeritus professor of chemistry and biochemistry and executive director of the Center for Applied Catalysis at Seton Hall University.

Augustine anchors preformed asymmetric homogeneous catalysts--the metal and its coordinated ligands--onto various supports (such as alumina, silica, or clay) by using heteropolyacids as anchoring agents. Anchoring is achieved when the metal forms a bond with a heteropolyacid such as phosphotungstic acid.

Asymmetric hydrogenation catalysts immobilized this way are at least as active and selective as the homogeneous versions, Augustine says. Some are reusable for up to 15 times. Catalyst leaching is not observed.

THE TECHNOLOGY complements Johnson Matthey's FibreCat technology, based on anchoring catalysts to a polymer fiber backbone. Four series of fiber-anchored catalysts are already commercially available: palladium catalysts for carbon-carbon cross-coupling, rhodium catalysts for hydrogenation, osmium catalysts for cis-hydroxylations, and ruthenium catalysts for selective oxidations.

The technology has been best studied for palladium-catalyzed cross-coupling reactions, says Thomas J. Colacot, a senior development associate at Johnson Matthey. Using model systems, he and coworkers have found that fiber-anchored catalysts are more selective than the homogeneous versions and can be reused up to five times. However, reusability depends on substrate and reaction conditions.

Although the main reason for developing FibreCat was to bind expensive ligands and metals and to recover them after the chemistry is complete, other advantages soon became clear, says Richard A. Teichman, manager of chemical development at Johnson Matthey. When anchored, pyrophoric ligands, such as tert-butyl phosphines, become stable; osmium tetroxide, which is volatile and highly toxic, can be handled like a nontoxic material, he says.

 

FULL SIZE - CLICK IMAGE

SAMPLES New compounds from Synthon Chiragenics feature a range of pharmacophores at different levels of complexity.

NO WORK HAS been done yet with asymmetric reactions. Although Johnson Matthey has shown that FibreCat osmium catalysts convert octene to dihydroxyoctane, chiral modifiers have not yet been used. "Our gut feeling is that if we add the chiral modifiers we should get results comparable with those from homogeneous cis-hydroxylations," Colacot says. The same would be true for FibreCat rhodium hydrogenation catalysts, he adds.

Meanwhile, research in chiral ligands continues to be very productive.

(R)-DTBM SegPhos, developed by Takao Saito and coworkers at Takasago International Corp., is a new addition to the company's portfolio of SegPhos ligands [(4,4'-bi-1,3-benzodioxole)-5,5'-diyl-bis(diarylphosphine)s]. Under dynamic kinetic resolution conditions, ruthenium-(R)-DTBM SegPhos reduces the carbonyl group of racemic -benzamido--ketoesters to form only one of four possible isomers in greater than 98% diastereomeric excess and greater than 99% ee.

Also from Takasago are nine new chiral diphosphine ligands for rhodium-catalyzed asymmetric hydrogenation of olefins based on three structure types: SegPhos, BeePhos [1,2-bis(2-alkyl-2,3-dihydro-1H-phosphindol-1-yl)benzenes], and UCAP [1-(2,5-dialkylphospholano)-2-(diarylphosphino)-benzenes or 1-(dialkylphosphino)-2-(2,5-dialkylphospholano)benzenes].

Yoshinori Kawai, associate director for fine chemicals at Takasago's Rockleigh, N.J., office, says these catalysts are being applied to olefins that are difficult to reduce enantioselectively. Reactions are carried out at 30 °C rather than at cryogenic temperatures. Pressures are moderate, sometimes as low as 13.6 lb per sq in.

In other work on asymmetric hydrogenations, Xumu Zhang, an associate professor of chemistry at Pennsylvania State University and the chief technology officer of Chiral Quest, State College, Pa., has prepared ortho-substituted BINAPO ligands [1,1'-bi-2-naphthylbis(diphenylphosphinite)s]. He says ortho-substitution restricts the orientation of aryl groups joined to phosphorus atoms and makes the ligands more effective than the unsubstituted versions. The new ligands have been used in ruthenium-catalyzed asymmetric hydrogenation of -aryl-substituted -(acylamino)acrylates to -aryl-substituted -(acylamino)esters at up to 99% ee.

For rhodium-catalyzed hydrogenations of -(acylamino)acrylic acid derivatives and -arylenamides, Zhang offers TangPhos, a 1,2-bisphospholane named after graduate student Wenjun Tang. The ligand, which has chiral phosphorus atoms, was designed with conformational rigidity in mind as well. Enantioselectivities of up to 99% and turnovers of up to 10,000 have been achieved with the ligand. Both ligand types have been licensed by Penn State to Chiral Quest.

In Germany, meanwhile, Bayer AG's fine chemicals business group has developed a new synthesis for its proprietary Cl-MeO-BIPHEP ligands [(5,5'-dichloro-6,6'-dimethoxybiphenyl-2,2'-diyl)-bis(diphenylphosphine)s]. These ligands deliver greater than 98.7% enantioselectivity in asymmetric hydrogenations of carbonyl groups and carbon-carbon double bonds. The method enables a wide spectrum of alkyl groups to be introduced, allowing "us to fine-tune the catalyst beyond what was possible before," says Rudolf Hanko, head of Bayer's fine chemicals business unit.

In other developments, Sumitomo Chemical is now making chiral cyclopropane carboxylic acids based on addition of a diazoacetate to a terminal alkene catalyzed by dimeric rhodium triphenylacetate. Yields of up to 90% are achieved with highly functionalized substrates. The reaction produces a racemate, but it is practical because of Sumitomo's library of phenethylamine resolving agents. Enantiopurities of at least 98% are achieved.

AT SNPE, THE inversion of configuration that occurs in bimolecular nucleophilic substitutions is being used to prepare chiral 2-chloropropionates. When methyl (S)-(–)-2-(chlorocarbonyloxy)propionate--made by phosgenation of methyl (S)-(–)-lactate--decomposes in the presence of hexabutylguanidinium chloride hydrochloride, methyl (R)-(+)- 2-chloropropionate is formed in up to 90% yield and up to 98% ee. Continuous attack by chloride ion on either side of the substitution site can occur, resulting in a racemate. But continuous removal of the inversion product prevents that from happening. The chemistry is being practiced now at the 100-kg scale.

For drug discovery, Synthon Chiragenics began offering early this year complex chiral compounds that have never before been readily available. Grouped by complexity into diamond, platinum, gold, and silver collections, the compounds will enable discovery of drugs based on structures that previously were hard to get to, says the company's founder, Rawle I. Hollingsworth, chemistry professor at Michigan State University, East Lansing.

Compounds in the diamond collection are the most complex, often with two or three chiral centers, and they can be used in drug screening with minimum transformation. They include highly functionalized oxazolidinones, -amino acids, morpholines, piperazines, and pyrrolidinols.

The platinum collection is composed of functionally rich and structurally complex amino alcohols, - and -amino acids, and oxazolidinones. They are suitable as cores or scaffolds for developing lead candidates, says Padmakumar R. Kaimal, director of drug discovery chemistry at Synthon Chiragenics.

The gold and silver collections, which became available in early June, include novel compounds with one chiral center. Kaimal says they can be used to generate core molecules for drug discovery or as chiral ligands.

These few examples hardly convey the breadth of recent advances in chiral technologies. Many fine chemicals companies are ready for the challenges of chiral manufacture, however poor the outlook is for their customers at the moment.

MORE ONLINE

For a calendar of events featuring chiral chemistry through March 2003, visit http://pubs.acs.org/cen/coverstory/8023/8023calendar.html.

MORE TO COME

In C&EN's June 17 issue, Senior Editor A. Maureen Rouhi will describe nonlinear effects in asymmetric reactions.

Chemical & Engineering News
Copyright © 2002 American Chemical Society