screens serves as an excellent illustration of the development of more precise, targeted systems.
Antibiotics were initially detected by growing ihe potential producer on an agar plate in the presence of an organism (or organisms) against which antimicrobial action was required. Production of the antibiotic was detected by inhibition of the test organism(s). Alternatively, the microbial isolate could be grown in liquid culture and the cell-free broth tested for activity. This approach was extended by using a range of organisms to detect antibiotics with a defined antibacterial spectrum. For example, Zahner (1978) discussed the use of test organisms to detect the production of antibiotics with confined action spectra. The use of Bacillus sub-tilis and Streptomyces viridochromogenes or Clostridiun. pasteurianum allows the identification of antibiotics with a low activity against B. subtilis and a high activity against the other test organisms. Such antibiotics may be new because they would not be isolated by the more common tests using B. subtilis alone. The kirromycin group of antibiotics were discovered using methods of this type.
In the 15 years prior to 1971, no novel naturally occurring ^-lactams were discovered (Nisbet and Porter, 1989). However, the advances made in the
Table 3.4 Guidelines for 'overproduction media' (Nisbet, 1982)
1. Prepare a range of media in which different types of nutrients become growth-limiting e.g. C, N, P, O
2. For each type of nutrient depletion use different forms of the growth-sufficient nutrient
3. Use a polymeric or complexed form of the growth-limiting nutrient
4. Avoid the use of readily assimilated forms of carbon (glucose) or nitrogen (NHJ) that may cause catabolite repression
5. Ensure that known cofactors are present (Co3+, Mg2+, Mn ,
6. Buffer to minimize pH changes.
.. ,t • ■||-wall biosynthesis and the mode of undcriianJinP J ^ a|k)Wcd lhc development of mode action o) 11 ^^ [he ltJ70s which resulted in a very 0f action soiCL ^ ^ djscoveries of new ¡3-lactam signitK-'»11 (] in et rt/-s (1Q7|) discovery of the antibiotic ^ ^ on thc detection of compounds ^Ílduced' inoiphological changes in susceptible vhl,h that lhc mode of action of penicillin'"
^niiis' wa1- 'd'e 'inhibition of the transpeptidase en-
,. ...„sslmkiiisi n'.ucopeptide molecules led to the Svlopmcnt o, en//me inhibitor assays which were -vriicuhrh .iiii.iciivc because they could be auto-i' 1 leinni" «; (W82) described the devlopment T uV automated screen for the detection of car-boupeptid..^ inhibitors which has led to the detection „1 social novel ^phamycin and carbapenem com-
^''lkMiicKMsins: licquency of penicillin and cephalos-poiin lesM.mcc' amongst clinical bacteria led to the development ol iiKx Iianisrn based screens for the isola-uon ol nioic elkclive antibacterials. The logic of the Beecli.uiis I'li.u in.ii c uticals group of Brown et al. (1976) was lo mmhIi loi a compound which would inhibit /Macianusc .iml could be incorporated with ampicillin as a combination therapeutic agent. Samples were tested for their ability to increase the inhibitory effect ol .unpicillin on ii ^-lactamase producing Klebsiella uemgencs and tins strategy resulted in the discovery of cl.iuil.mic acid I mi liter examples of targeted antibiotic screens are given in the excellent review by Nisbet and I'm lei ll'ivn.
I lie coiKcpi ol lining the inhibition of enzymes as a screening mechanism was pioneered by Umezawa (1972) in his search for microbial products inhibiting key enzymes of human metabolism. His approach was based on the logic that, if a compound inhibits a key hum.in cn/vmc w . :rro, it may have a pharmacological action in vivo. Such screens have been applied to a wide range of pharmacological targets and have resulted in the isolation of several important drugs. From Aspergillus, Alberts et al. (1980) isolated mevinolin which is an inhibitor of hydroxy-methyl-glutaryl reductase, the rate limiting step in the biosynthesis of sterols. It is now marketed as an agent to lower high cholesterol levels. Bull (1992) list the following clinical situations in which microbial products have been shown to inhibit key enzymes: hypercholesterolemia, hypertension, gastric inflammation, muscular dystrophy, benign prostate hyperplasia and systemic lupus erythemosus. A further specific example is provided by the work of Hashimoto ct al. (1990). The activity of carbapenem antibiotics is lost in therapy due to renal dehydropepti-
dase activity. These workers isolated microbial products capable of inhibiting the enzyme which could then be administered along with the antibiotic to maintain its clinical activity.
The detection of pharmacological agents by receptor ligand binding assays has been developed rapidly by pharmaceutical companies (Bull, 1992). These are extensions of the enzyme inhibitor approach but agents which block receptor sites are likely to be more effective at very low concentrations. The gastrointestinal hormone cholecystokinin (CCK) controls a range of digestive activities such as pancreatic secretion and gall bladder contraction. Receptor screening identified a fungal metabolite, aperlicin (from Aspergillus alliaceus) that had a very high affinity for CCK receptors. Although the fungal product did not prove to be a suitable drug it was used as a model for the design of analogues which were receptor binders and pharmacologically acceptable.
The progress in molecular biology, genetics and immunology has also contributed extensively to the development of innovative screens, by enabling the construction of specific detector strains, increasing the availability of enzymes and receptors and constructing extremely sensitive assays. Bull (1992) summarized the major contributions as follows:
(i) The provision of test organisms that have increased sensitivities, or resistances, to known agents. For example, the use of super-sensitive strains for the detection of /8-lactam antibiotics.
(ii) The cloning of genes coding for enzymes or receptors that may be used in inhibitor or binding screens makes such materials more accessible and available in much larger amounts.
(iii) The development of reporter gene assays. A reporter gene is one which codes for an easily assayable product so that it can be used to detect the activation of a control sequence to which it is fused. Such systems have been used in the search for metabolites that disrupt viral replication.
(iv) Molecular probes for particular gene sequences may enable the detection of organisms capable of producing certain product groups. This information may be used to focus the search on these organisms in an attempt to find novel representatives of an already known commercially attractive chemical family.
(v) The development of immunologically based assays such as ELISA.
It is important to appreciate at this stage that the advances in the biological sciences which have enabled the design of sophisticated screening tests have been paralleled by the development of robotic automation systems. This means that the engineering now exists to automate such tests, resulting in an enormous throughput. Indeed, Nisbet and Porter (1989) claimed that "the modern microbial discovery programme must achieve rates of 105 tests per year, for as many as 20 different assay systems, to compete in the race for novel therapeutics."
Was this article helpful?