COMMENTARY

http://biotech.nature.com • APRIL 2002 • VOLUME 20 • nature biotechnology

A major change has occurred in the pharma-ceutical industry in the past several years. Thishas been caused in part by the financial com-munity’s expectation that a certain number ofblockbuster drugs per year are necessary forcompany profitability. As a result, we have wit-nessed an extensive series of mergers amongthe giants of the industry as well as theiracquisitions of smaller biopharmaceuticalcompanies. This has unfortunately not solvedthe financial problem: the commercial outputof the combined companies has not evenequaled the total number of products com-mercialized by the organizations existingbefore merger mania took over.

There are several reasons for this inefficien-cy. First, the number of competing discoverygroups, each with its own concepts, method-ologies, and ingenuity, has decreased. Thenewly merged companies tend to be less inno-vative than the original organizations.

Second, there is a perceived need to feed thediscovery machine with more and more can-didate compounds. High-throughput screen-ing, which began in the early 1990s, can assayas many as 100,000 entities in one day. It hasaccomplished miniaturization and speed, buthas neither provided the numbers of high-quality leads anticipated nor even increasedthe rate at which such leads are found.

This worship of the numbers game has vir-tually eliminated the most unique source ofchemical diversity, natural products, from theplaying field in favor of combinatorial chem-istry. It is not that Nature does not providemillions of new compounds, but that it ismore difficult and takes more brainpower toharness and analyze this great feast of undis-covered chemistry by high-throughputscreening. Combinatorial chemistry yieldsmainly minor modifications of present-daydrugs, and without new scaffolds on which tobuild, it is not up to the task at hand.

A third reason relates to the promise ofgenomics. Although comparative genomicscan disclose new targets for antimicrobialsand other types of drugs, the number of suchpotential targets is so large that it requirestremendous investments of time and moneyto set up all the screens necessary to exploitthis resource. This can only be handled by a

high-throughput screening methodology,which demands libraries of millions of chemi-cal entities. Although such targets would beexcellent for screening natural products, theindustry has failed to exploit this opportunityand has opted to save funds by eliminatingnatural product departments or decreasingtheir relevance in the hunt for new drugs.

Finally, an ever-increasing number of syn-thetic drugs have been recalled by the US Foodand Drug Administration (Rockville, MD)because of toxicities and deaths occurring inpatients after initial approval.

During the 3 billion years that bacteria haveinhabited the earth, Nature has conducted itsown type of combinatorial chemistry in amuch more exotic way than chemists, leadingto the creation of hundreds of thousands ofnovel secondary metabolites. These naturalproducts, with structures more spatially com-plex than those of synthetic chemicals, havebeen overwhelmingly useful to our society.

Microbial and plant secondary metabolitesdoubled our life span during the 20th century.They have reduced pain and suffering, andrevolutionized medicine. Over half ofapproved drugs are either natural products orrelated to them, even excluding biologicals,such as vaccines and monoclonal antibodies.It is most unfortunate that the pharmaceuticalindustry has downgraded natural productsjust at the time that new assays are availableand major improvements have been made indetection, characterization, and purificationof small molecules. With the advent of combi-natorial biosynthesis, thousands of new deriv-atives can now be made by a biological technique complementary to combinatorialchemistry. Furthermore, only a minor propor-tion (∼ 1–10%) of bacteria and fungi have thusfar been examined for secondary metaboliteproduction. New methods are being devel-oped to cultivate the so-called “unculturable”microbes from the soil and sea.

The selective action exerted on pathogenicbacteria and fungi by microbial secondarymetabolites ushered in the antibiotic era, andfor over 50 years, we have benefited from thisremarkable property of “wonder drugs”. Themarket for such compounds is over $30 bil-lion per year. Microbiologists know thatantimicrobial technology alone will not per-manently win the war against infectiousmicroorganisms because of the developmentof resistance in pathogenic microbes. We willhave to be satisfied merely to stay one stepahead of the pathogens for a long time to

come; thus, the search for new antibioticsmust not be stopped.

New entities are continually needed as aresult of the development of resistantpathogens, the existence of naturally resistantbacteria (e.g., Pseudomonas aeruginosa,Stenotrophomonas maltophilia, many entero-cocci, Burkholderia cepacia and Acinetobacterbaumanni), and the toxicity of some of thecurrent compounds. Another problem is theincreased incidence of infection by organismsthat are not normally virulent but do infectimmunocompromised patients. Fungal infec-tions doubled from the 1980s to the 1990s.Candidiasis, cryptococcosis, and aspergillosisare on the rise, and the rate of treatment fail-ure in aspergillosis exceeds 60%. Fungal infec-tions, usually by Candida and Aspergillusspecies, often occur after lung, kidney, heart,and liver transplant operations. Pulmonaryaspergillosis is the main factor in the death ofrecipients of bone marrow transplants, andPneumocystis carinii is the primary cause ofdeath in patients with AIDS from Europe andNorth America. Current treatments includethe synthetic azoles (such as fluconazole andflucytosine) or the natural polyene ampho-tericin B. However, their use is limited by thedevelopment of resistance to the azoles and bythe toxicity of amphotericin B.

In the past 20 years, there has been a majorchange in the discovery and application ofsecondary metabolites, characterized by abroadening of the scope of the search. Nolonger are microbial sources looked uponsolely as providing potential solutions formicrobial diseases. This change in screeningphilosophy has been followed by ingeniousapplications of molecular biology to detectactivities of compounds from microbes andplants for non-antibiotic applications. Theseinclude the cholesterol-lowering statins(e.g., lovastatin and pravastatin), anticanceragents (e.g., the microtubule stabilizer pacli-taxel), immunosuppressants (e.g.,cyclosporin A, tacrolimus, sirolimus, andmycophenolic acid), antiparasitic agents(e.g., avermectins and the polyethers), bio-herbicides (e.g, bialaphos), and bioinsecti-cides (e.g., the spinosins).

It is clear that the future success of the drugindustry depends on the combination of com-plementary technologies, such as naturalproduct discovery, high-throughput screen-ing, genomics, proteomics, and combinatorialchemistry. Elimination of natural productsfrom the mix is a formula for failure.

331

Arnold L. Demain is fellow of the Charles A.Dana Research Institute for ScientistsEmeriti, 330 Hall of Science, Drew University,Madison, NJ 07940 (ademain@drew.edu).

Prescription for an ailing pharmaceutical industryArnold L. Demain

©20

02 N

atu

re P

ub

lish

ing

Gro

up

h

ttp

://b

iote

ch.n

atu

re.c

om