100477

164

G. Yim ct ul. Inlcmutionul Joumul of Mcdicul Microbiology 296 (2006) 163 170

his opinion (Waksman. 1961). Allhough Ihc conccpt of antibiotics has evolved ovcr time to includc plant and animal products as well as synthetic and scmisynthetic compounds uscd in thcrapy, thc word 'antibiotic' has becomc entrcnchcd as a descriptor for any molecule produccd in thc laboratory with thc ability (cidal or static) to inhibit thc growth of microbes. Many text-books dolino thc function of antibiotics as agents produccd for intcr-bactcrial warfarc in thcir natural cnvironmcnts, disrcgarding thc fact that studics arc donc undcr complctcly arii lic ial laboratory conditions. The only rcquircmcnt is that thc combination of media and fermentation conditions make sufficicnt of the bioactive to suppress the growth of tester strains. The cullurc media requircd to provokc antibiotic formation arc unusual to say the least; mixtures as complex and conccntratcd as tryptic soy broth containing substrates such as blood meal, cottonseed flour, pork liver or herring meal are often used. These are unlikely nutrients for soil microbes (Dcmain and Solomon. 1986)!

Noncthclcss, the antibiosis notion has guidcd anti-infcctive discovery for almost 100 years with enormous succcss. rcvolutionizing the treatment of infcctious discasc and providing many important thcrapcutics (sec Tablo I); antiviral and antitumour thcrapy also beneliled. It rcsullcd in thc building of the fermentation-based pharmaccutical ind ustry in North America and clscwhcrc. in cfTcct using biotechnology in a way that went bcyond thc production of food, drink, and Chemicals.

The antibiotic conccpt menlioned above implies that smali moleeules arc produccd in naturę by bacteria and fungi in sufficicnt quantitics to suppress or kill other microorganisms. In rcality thc laboratory dcmonstra-tions of this cffcct may havc bccn fortuitous in that thc producing strains. when taken front thcir sturvation diet

l abie I. Principal antibiotics in use (2005)

Class	Datę discovered
Sulfonamidem	1937
Pcnicillins	1940
Polymixin	1947
Chloramphenicol	1949
Tetracydines	1953
Cephalosporins (4 genemtions)	1953
Aminoglycosidcs	1957
Vancomydn	1958
Clindamycin	1966
Rifamyein	1971
TMP SMX	1973
Carbapcncms	1976
Monobaetams	1982
Linczolid	1987
Daptomycin	1987
Synercid	1992

in thc cnvironmcnt, wcrc ‘shockcd* into producing larger amounts of compounds upon exposure to unnatural. rich growth conditions. It is common knowl-edge that the production of appropriatc amounts of compound for preliminary Chemical and clinical studies often rcquircs hcroic cITorts in microbiology, chcmistry, and the isolation of high-producing mutants. In all probability many smali moleeules produccd in Iow conccnlrations arc ncvcr dctcctcd by these methods.

One of the morę striking discoverics from genome scqucncing cITorts is the demonstration that. bascd on currcnt knowlcdgc of biosynthetie pathways and thcir organization, most strcptomycctcs (and other soil bacteria) possess the genetic capacity to produce upwards of 25 diITcrent smali moleeules (Bcntlcy ct al„ 2002; Ikeda et al.. 2003). Most of these compounds are not dctcctcd in traditional antimicrobial screens. sug-gesting that they are not produccd undcr laboratory growth conditions (at least not in detcctablc conccntra-tions) or that they do not havc any antibiotic propcrtics.

Antibiotic discovery

We can safely assume that all smali moleeules madę by microbes havc biological functions (although it has becn suggested that antibiotics may be simply waste products of ccllular metabolism. i.c.. microbial faccal materiał!) (Chadwick and Whclan, 1992). Suffice it to say that their environmcntal activitics are not known; smali moleeules may provc to havc many rolcs in microbial communitics, and rational scarchcs for ncw bioactivc moleeules must employ the idcntilicalion of any and all biological activitics. In contrast to traditional scrccning, a morę comprchcnsivc approach will likely pro\ide a rich surfeit of potential antibiotic moleeules: too many hits. Howcvcr. this will permit the idcniification of a larger number of uscful bioactives; it is often elaimed that currcnt scrccning methods havc harvcstcd all thc 'Iow hanging fruit*. It would seem that therc are many plums still to be picked!

It is difhcult to estimate cithcr the numbers of compounds or thc Chemical space represented by microbial smali moleeules; the word that comes to mind is 'infinilc' (Dobson. 2004). For cvolutionary rcasons. thc rangę of bioactivitics will be correspondingly largo. It has bcen known for some time that antibiotics have propcrtics in addition to thosc related to thcir ther-apeutic use; all produce side effects during thcrapy and provokc a varicty of responses in thc host (Fig. 1). Rcccntly it has bccn shown that many microbial products act as bacterial pheromones and play important rolcs in the regulation of microbial gene c.xprcssion. Surprisingly, the divcrsity of effects of antibiotics has ncvcr bccn considered to imply that the main function of these moleeules in naturę is not. in fact. antibiosis!