Natural Products – History, Diversity and Discovery

MICROBIAL NATURAL PRODUCTS: A PAST WITH A FUTURE

Arnold L. Demain

Fermentation Microbiology Laboratory, Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA

For over 50 years, antibiotics have served us well in combating infectious bacteria and fungi. The recently increased development of resistance to older antibacterial and antifungal drugs is being challenged by (i) newly discovered antibiotics (e.g., pneumocandins), (ii) new semi-synthetic versions of old antibiotics (e.g., glycylcyclines), (iii) older underutilized antibiotics (e.g., teicoplanin), and (iv) new derivatives of previously undeveloped narrow-spectrum antibiotics (e.g., streptogramins, everninomycin). Many of these products are in late stage clinical testing at the moment. In addition, many antibiotics are used commercially, or are potentially useful, in medicine or agriculture for activities other than their antibiotic action. They are used as antitumor agents, enzyme inhibitors including powerful hypocholesterolemic agents, immunosuppressive agents, and anti-migraine agents, in medicine. Agricultural products include bioherbicides, antiparasitic agents, bioinsecticides and growth promotants for animals (especially ruminants) and plants. A number of these products were first discovered as mycotoxins, or as antibiotics which failed in their development as such. Combinatorial chemistry will accelerate the discovery of new derivatives of natural products. It will join structure-function drug design, semi-synthesis, and recombinant DNA technology as techniques complementing the screening of natural products.

1 INTRODUCTION

Natural products have been an overwhelming success in our society. The doubling of our life span in the twentieth century has been attributed to the use of plant and microbial secondary metabolites. These have reduced pain and suffering, and revolutionized medicine by allowing for the transplantation of organs. Natural products are the most important anticancer and anti-infective agents. Over 60% of approved and pre-NDA candidates are either natural products or related to them, not including biologicals such as vaccines and monoclonal antibodies. Almost half of the best selling pharmaceuticals are natural or are related to natural products. Often, the natural molecule has not been used itself but served as a lead molecule for manipulation by chemical or genetic means.

Secondary metabolism has evolved in nature in response to needs and challenges of the natural environment. Nature has continually carried out its own version of combinatorial chemistry for the period of over 3 billion years during which bacteria have inhabited the earth. During that time, there has been an evolutionary process in progress in which producers of secondary metabolites evolve according to their local environments. If the metabolites are useful to the organism, the biosynthetic genes are retained and genetic modifications further improve the process. Combinatorial chemistry practiced by nature is much more sophisticated than combinatorial chemistry in the laboratory, yielding exotic structures rich in stereochemistry, concatenated rings and reactive functional groups. As a result, an amazing variety and number of products have been found in nature. The total number of natural products produced by plants has been estimated to be over 500,000. About 100,000 secondary metabolites of molecular weight less than 2500 have been characterized, mainly produced by microbes and plants: some 50,000 are from microorganisms. The enormous diversity existing in secondary metabolism can be illustrated by the following two examples.

(i) 22,000 terpenoids have been described from living organisms. They are all produced from hydroxymethylglutaryl coenzyme A via mevalonate. Their structures contain repeats of the five-carbon isoprene unit unless subsequently modified. Their functions include intercellular communication in animals (steroid hormones) and in plants (gibberellins), aroma (volatile terpenoids), pigments (carotenoids) and sexual hormones in animals and fungi. There are 86 known gibberellins, of which 26 are produced by fungi and the rest by plants.

(ii) 10,000 polyketides are known, most of which are produced by bacteria and fungi. These include antibiotics (e.g., erythromycin, tetracyclines, rifamycins), antitumor agents (e.g., doxorubicin, daunorubicin, enediynes), immunosuppressants (e.g., FK 506, rapamycin), antiparasitic agents (e.g., avermectins), antifungals (e.g., amphotericin, griseofulvin), cardiovascular agents (e.g., lovastatin, pravastatin) and veterinary products (e.g., monensin, tylosin).

Natural product research is at its highest level now due to met needs, remarkable diversity of structures and activities, utility as biochemical probes, novel and sensitive assay methods, improvements in isolation, purification and characterization, and new production methods. Many new products have been made by genetic methods involving modification or exchange of genes between organisms to create hybrid molecules; the technique is known as combinatorial biosynthesis.

The enormous diversity of microorganisms is a factor that must be kept in mind for future drug development. Only a minor proportion of bacteria and fungi have been cultured or examined for secondary metabolite production. For example, only 5% of the total number of fungal species have been described. Of those described 69,000), only 16% (11,500) have been cultured. It has been estimated that 1 gram of soil contains 1000 to 10,000 species of undiscovered prokaryotes! The extensively used concept of isolation of microbial strains from different geographical and climatic locations around the world still gathers support.

2 MEDICAL ANTIBIOTICS

The selective action exerted on pathogenic bacteria and fungi by microbial secondary metabolites ushered in the antibiotic era and for fifty years, we have benefited from this remarkable property of these "wonder drugs". The successes were so impressive that these antibiotics were virtually the only drugs utilized for chemotherapy against pathogenic microorganisms. Antibiotics are defined as low molecular weight organic natural products (secondary metabolites; idiolites) made by microorganisms which are active at low concentration against other microorganisms. Of the 12,000 antibiotics known in [1995, 55% were produced by filamentous bacteria (=actinomycetes) of the genus Streptomyces, 11% from other actinomycetes, 12% from non-filamentous bacteria and 22% from filamentous fungi. New bioactive products from microbes continue to be discovered at an amazing pace: 200 to 300 per year in the late 1970s increasing to 500 per year by 1997. The world market for antimicrobials involves some 150-300 products, either natural, semi-synthetic or synthetic, and includes cephalosporins (45%), penicillins (15%), quinolones (11%), tetracyclines (6%) macrolides (5%), aminoglycosides, ansamycins, glycopeptides and polyenes.

About 30 years ago, the difficulty and high cost of isolating novel structures and agents with new modes of action for such uses became apparent and the field looked like it might enter a phase of decline. This is understandable because the chance of finding useful antibiotics from microbes is very low, i.e. one in 10,000 to 100,000 compounds. Similar data have been observed with plants. Indeed, the number of anti-infective investigational new drugs (INDs) declined by 50% from the 1960s to the late 1980s. By the technique of semi-synthesis, chemists had been improving antibiotics for many years but despite success, new screening techniques were sorely needed in 1970 to isolate new bioactive molecules from nature. Many felt that the golden era of antibiotic discovery was over, but this was far from the truth. Due mainly to the development of novel target-directed screening procedures, important new antibiotics appeared on the scene and became commercial successes in the 1970s and 1980s. These included cephamycins (e.g., cefoxitin), fosfomycin, carbapenems (e.g., thienamycin), monobactams (e.g., aztreonam), glycopeptides (e.g., vancomycin, teicoplanin), aminoglycosides (e.g., amikacin, sisomicin) as well as semisynthetic versions of cephalosporins and macrolides. Thienamycin is the most potent and broadest in spectrum of all antibiotics known today. Although a β-lactam, it is not a member of the penicillins or cephalosporins; rather it is a carbapenem. It is active against aerobic and anaerobic bacteria, both Gram-positive and Gram-negative including Pseudomonas. This novel structure was isolated in Spain from a new soil species which was named Streptomyces cattleya. Interestingly, this culture also produces penicillin N and cephamycin C. Despite a low degree of production and instability of the molecule, the structure of thienamycin was elucidated in 1976. The main differences between the conventional β-lactams and thienamycin are the possession of a carbon atom instead of sulfur in the ring attached to the β-lactam ring and the trans configuration of the hydrogen atoms of the β-lactam ring. In addition to its broad spectrum and high potency, the molecule possesses resistance to plasmid and chromosomal β-lactamases and activity against pathogens resistant to penicillins, cephalosporins and aminoglycosides. Its chemical instability is due to a type of self-destruction, i.e., the β-lactam ring of one molecule of thienamycin is broken via aminolysis by the amine group of a second thienamycin molecule. To eliminate this problem, chemists created the semi-synthetic carbapenem, imipenem, by adding a formididoyl group to the side chain mine. Not only was imipenem found to be more stable than thienamycin, but also twice as active! The mode of action of imipenem (and thienamycin) is inhibition of cell-wall peptidoglycan synthesis and it is bactericidal.

In humans, imipenem was found to be metabolized by an enzyme in the kidney, renal dehydropeptidase-I, which acts as a β-lactamase. Since the enzyme appears to serve no essential role in human metabolism, scientists were able to develop a synthetic competitive inhibitor, cilastatin, which they then used with imipenem to produce the combination drug, primaxin (Tienam). Primaxin was introduced into medical practice in 1985.

The unique activity of thienamycin is due to several factors: (i) it permeates the Gram-negative outer cell membrane through porin channels at 10-20 times the rate of classical β-lactams; (ii) it is not destroyed by the β-lactamases of the periplasmic space; (iii) it binds to and inhibits all penicillin-binding proteins (PBPs) but is principally active against PBPs-2 and-1b rather than PBP-3 which is attacked by conventional penicillins and cephalosporins. Sequential inhibition of PBPs-2 and-1b converts the cells of the pathogen to non-growing spheroplasts which rapidly die. On the other hand, inhibition of PBP-3 blocks septum formation resulting in long viable filaments which take longer to die; (iv) after removal of thienamycin, there is a long delay before regrowth of any unkilled spheroplasts. With conventional β-lactams, unkilled filaments (which contain 20 or more cells per unit) septate and regrow immediately.

It is quite interesting that despite the very high potency of thienamycin, it took over 25 years of worldwide screening before it was discovered. Today, we know that carbapenems are not rare. Many members of this family have been isolated in laboratories all over the world although none equals thienamycin in potency and spectrum of activity. The answer probably lies in its very low level of production by wild strains and its instability. Conventional screening procedures evidently missed this activity and only after the development of specific and sensitive modern assay procedures was it found. Thienamycin was discovered by a sensitive mode of action screen, the details of which have never been revealed.

Microbiologists knew in 1970 that technology had not yet won the war against infectious microoganisms due to resistance development in pathogenic microbes. Indeed, it may never win the war and we may have to be satisfied to merely stay one step ahead of the pathogens for a long time to come; thus, the search for new drugs must not be stopped. New antibiotics are continually needed because of (i) the development of resistant pathogens, (ii) the evolution of new diseases (e.g., AIDS, Hanta virus, Ebola virus, Cryptospiridium,] Legionnaire's disease, Lyme disease, Escherichia coli 0157:H7), (iii) the existence of naturally resistant bacteria (e.g., Pseudomonas aeruginosa causing fatal wound infections, burn infections and chronic and fatal infections of lungs in cystic fibrosis patients), Stenotrophomonas maltophilia, Enterococcus faecium, Burkholderia cepacia and Acinetobacter baumanni, and (iv) the toxicity of some of the current compound. The resistant bacteria are generally uninhibited by all commercial antibiotic. Other organisms exist which are not normally virulent but do infect immunocompromised patients. Semi-synthetic tetracyclines, e.g. glycylcyclines, are being developed for use against tetracycline-resistant bacteria. Exploitation of old but underutilized antibiotics is occurring. In recent years, there has been great concern about resistance development among Gram-positive pathogens. Clinical isolates of penicillin-resistant Streptococcus pneumoniae, the most common cause of bacterial pneumonia, increased in the US from 1987 to 1992 by 60-fold. Methicillin-resistant Staphylococcus infections increased to an alarming extent throughout the world. At present, vancomycin is the molecule of choice to treat infections caused by such organisms; however, resistance is developing to this glycopeptide antibiotic, especially in the case of nosocomial vancomycin-resistant Enterococcus infections. Fortunately, some vancomycin-resistant enterococci are treatable by the related underutilized glycopeptide antibiotic, teicoplanin. Furthermore, a number of "old" compounds are now available for development which were not previously developed because of their narrow antibacterial spectrum which was restricted to Gram-positive bacteria. At that time (in the 1970's and 1980's), breadth of spectrum was the commercial goal but today, a major aim is to inhibit resistant Gram-positive pathogens. About 90% of natural antibiotics fail to inhibit Gram-negative organisms such as Escherichia coli. The reasons include its outer membrane which contains (i) narrow porin channels which retard the entry into the cell of even small hydrophilic compounds and (ii) a lipopolysaccharide moiety which slows down the transmembrane diffusion of lipophilic antibiotic. Furthermore, Gram-negative bacteria often possess a multiple-drug efflux pump which eliminates many antibiotics from the cells.