Transmembrane signalling from cell surface receptors occurs by two broad mechanisms: (i) the rapid (ms) direct opening of an ion channel, where the ion channel is a component of the receptor complex (e.g. the nicotinic acetylcholine receptor); and (ii) the more slow (s) modulation of a membrane enzyme or more distant ion channel. Most of the examples of this second mechanism involve a GTP-binding protein or so–called G-protein, and the production of a second messenger. The production of nitric oxide is a special case in that it is eventually produced as a result of the activity of the second messenger ïnositol trisphosphate. The nitric oxide then diffuses into a second cell to give rise to the production of an additional ‘second’ messenger, cyclic GMP.

All of the surface receptors themselves exist as a number of subtypes. Additionally, most of the components of the second messenger systems – G-proteins, adenylyl cyclase, guanylyl cyclase, phosphoinositidase, C, inositol trisphosphate receptors, protein kinase A, protein kinase G, protein kinase C, cyclic nucleotide phosphodiesterases, and the enzymes involved in phosphatidylinositol resynthesis – occur in a number of isoforms. Furthermore, all the enzymes are controlled in their activity by a number of co-factors and other modulators. This diversity provides the potential for selective drug action, a potential which is already being exploited and which will be increasingly so in the near future.

Over 50 different neurotransmitter and hormone receptors have been cloned, transfected and characterised. Within the pharmaceutical industry, the thrust of this research is now moving from receptor cloning to focussed application of these new tools to the task of drug development. With human receptor clones now available, it is possible to target drug design to single human proteins from the inception of a drug-design project. A critical question is whether human genes will express a human pharmacology when expressed in non-human, non-neuronal cell lines. Relevant results from studies with the human serotonin 5-HT2 and 5-HT1D receptors are presented, including the cloning of a rat 5-HT1B receptor which is homologous to a human 5-HT1D receptor gene. Another issue is the relationship between agonist and antagonist binding sites. The ability to study individual human receptors in selected cell lines which are free from interfering receptors has helped to advance our understanding of this relationship. Studies comparing agonist and antagonist binding sites of the human 5-HT2 receptor are described. Finally, the fact that so many neurotransmitter receptors belong to the same gene superfamily, the G-protein-coupled or 7TM (7 transmembrane domain) receptor superfamily, allows us to form new insights about receptor structure and function. Examples where comparative studies of gene sequences are helping to predict pharmacological properties of drugs are described.

Guanine-nucleotide regulatory binding proteins (G-proteins) serve to transduce information from agonist-bound receptor complexes to either effector enzymes or ion-channels. Drugs that perturb the function of G-proteins may do so by one of four mechanisms, (i) They may exert negative intrinsic activity toward the G-protein. For instance, we have shown that incubation of isolated plasma-membranes with the β-adrenoceptor blocking drug sotalol blocked both GTP-stimulated and isoprenaline-stimulated adenylyl cyclase. This suggests that the empty β-adrenoceptor is capable of tonically stimulating Gsα, and therefore adenylyl cyclase; that is, empty β-adrenoceptors promote GDP-GTP exchange, (ii) They may perturb the GDP-GTP exchange reaction. For instance, certain PDE inhibitors, including SKF 94836 and rolipram, stimulate a marked increase in the pertussis toxin-catalysed NAD+-dependent ADP-ribosylation of Giα. This effect is similar to that of GDP, which promotes stabilisation of the αβγ holomer of Gi. The effect of these PDE inhibitors is completely reversed by GppNHp, which triggers afly dissociation by binding to the guanine-nucleotide binding domain of the G-protein. PDE inhibitors may serve as a class of drugs which perturb GDP-GTP exchange, (iii) They may trigger uncoupling of receptor-G-protein complexes. For instance, the polycationic drug mastoparan binds to the C-terminal end of the G-protein and mimics the effect of receptor activation by promoting GTP-γ-S binding, a reduction in pertussis toxin-catalysed ADP-ribosylation, and inhibition of adenylyl cyclase activity. Other agents, such as polyanionic drugs, bind to the receptor to promote uncoupling of receptor-mediated activation of certain G-proteins. (iv) They may alter the cross-talk mechanisms that operate between different receptor signalling systems. For instance, protein kinase C promotes the phosphorylation and inactivation of Gi. This leads to an unopposed stimulation of adenylyl cyclase via Gs and, therefore, enhanced sensitivity to agents such as glucagon. Protein kinase C inhibitors may be usefully exploited to modulate these processes which appear to be abberant in certain disease states.

Biotechnology had its initial impact on the pharmaceutical industry well before the perceived time. The use of fermentation technology to produce antibiotics was a cornerstone for the development of the industry. This event was both before cloning (BC) and before DNA (rather than after DNA – AD). Even now the antibiotic market, which is worth over 10 billion U.S. dollars a year, is the most valuable segment of the total market, (c.200 billion dollars per year). Nevertheless the impact of biotechnology in drug discovery was until recently perceived solely to be the use of recombinant DNA techniques to produce therapeutic proteins and modified versions of them by protein engineering.

There are several other places where genetic engineering is influencing drug discovery. The expression of recombinant proteins in surrogate systems (e.g. in E. coli, yeast or via baculovirus infection or in mammalian cells) provides materials for structure determination (e.g. HIV protease) and structure/function studies (e.g. various receptors). Recombinant DNA techniques are influencing assay technology by allowing access to proteins in sufficient quantity for high throughput screening.

In addition, screening organisms can be constructed where a particular protein function can be measured in a microorganism by complementation or via reporter gene expression.

Transgenic animals also illustrate the power of the technology for drug discovery. Not only will transgenic rats and mice be used as models of disease but also for efficacy and toxicological profiling. What is learned in transgenic rodents may well set the scene for somatic cell gene therapy in humans.

Oncogenes can be introduced into cells which have a limited lifespan, and immortalised cell lines isolated as a result. These cell lines often retain the differentiated, tissue-specific characteristics found in the original cells and, as such, provide an important model for pharmacological studies. The techniques used to develop cell lines from rabbit kidney, mouse macrophages and rat liver are described. A preliminary characterisation of all three types of cells has been carried out, and in each case the immortalised lines described have many of the properties of the original tissue type.

Venoms are complex mixtures of many different components, whereas a toxin is a single pure compound whose activity can be specified. Toxins often have novel, highly specific activities. Studies on toxins open the way to design therapeutically useful molecules based on the structural information obtained from the toxins.

An example of successful drug development from a component of a snake venom is that of the inhibitors of angiotensin-converting enzyme (ACE). The first ACE inhibitors were developed from work on peptides isolated from the Brazilian arrowhead viper Bothrops jaracusa that prolonged the action of bradykinin. The venom components blocked the enzyme that inactivated bradykinin; the same enzyme activated the precursor of the vasoconstrictor hormone angiotensin. Hence, a valuable new type of therapy was established by developing the leads provided by the natural toxin.

Dendrotoxins are small proteins from mamba venoms that block rapidly activating K+ channels in neurones. There are possibilities that dendrotoxin could be the basis for drug design. A small analogue that mimicked the activity of dendrotoxin in the central nervous system might boost the activity of certain neuronal pathways. Potentially, such a compound might be able to restore some of the function of systems damaged by progressive neurodegenerative diseases. Although dendrotoxin-like compounds would not be a cure for these diseases, they may provide temporary relief from symptoms.

Conversely, compounds with the specificity of dendrotoxin but with ability to activate K+ channels, rather than block them, may be a novel way of reducing abnormal electrical activity in the brain. Hence, they may form the basis of a new type of anticonvulsant drug for epilepsies.

Recent work at Strathclyde has concentrated on defining the structure-activity relationships of the naturally occurring dendrotoxins and on synthesising small analogues of the postulated ‘active site’. Preliminary results are encouraging, and it is hoped to go from green mamba venom to dendrotoxin to therapeutically useful drugs.

The polyamine amides comprise a newly-discovered class of compounds which exhibits considerable potential for the development of selective pharmacological tools and pharmaceutical agents. In mammals and other vertebrates, they are selective, non-competitive antagonists of ionotropic glutamate receptors, but they also interact with other ionotropic receptors (e.g. nicotinic acetylcholine receptors). Thus, they are channel blockers which are selective for cation channels. We report on synthetic studies undertaken to produce hybrid analogues of these toxins based upon the argiotoxins of spider venoms and the philanthotoxins of parasitic, predatory wasp venom. The synthesis and characterisation of a mono-acylated spermine is also described. In addition, an account of current views on the many possible sites and modes of actions of the polyamine amides is presented and their potential for therapeutic neurochemistry, e.g. for the possible treatment of ischaemic damage to the nervous system, is highlighted.

The diversity of biological functions that are exerted by toxins from snake and scorpion venoms is associated with a limited number of structural frameworks. At present, one predominant basic fold has been observed among scorpion toxins whereas six folds have been found among snake toxins. Most toxin folds have the capacity to accept multiple insertions, deletions and mutations and to exert various recognition functions. We suggest that such folds may serve as guides to engineer new protein functions.

The role of computers in drug discovery depends on just how much is known about the target macromolecule. If atomic detail of the receptor is known, binding free energy differences between drug variants may be computed. Major effort is being expended in extending the area of applicability of such studies by predicting protein structure based on homologies with known protein crystal data. Where no target structure is available, computational methods can provide leads by defining transition state structures and then using the approach of molecular similarity to define stable mimics to act as blockers.

With the recent advances in gene technology, a substantial premium has been placed on the ability to predict protein structure and mechanism from sequence data alone. This is because the direct experimental approaches are more time and labour intensive, and progress is slower. One area of theoretical investigation which has a clear potential to assist in predictive exercises is the natural evolution of protein structure and sequence.

Some aspects of protein evolution, such as the fact that homology of sequence often denotes common ancestry, are routinely considered in predicting three-dimensional structure and mechanism. However, there are other aspects to this natural process which are either not realised or are not appreciated as potential analytical tools.

This article introduces the different ways protein evolution can be investigated and indicates developing techniques as well as those already well established.

In order to investigate and demonstrate objective modelling of proteins as a basis for drug design, we have sought to model several proteins in particularly persuasive circumstances. This is either (a) by filing the results of the model with an independent institution prior to X-ray determination of their structure, or (b) by using wholly automatic, general and reproducible methods, or (c) most often by both. Results suggest the ability to predict the core of the protein to an accuracy of about 1 Å rms deviation between predicted and experimental all-atom coordinates, and of surface loops in the range 1-4 Å rms deviation. Although the upper end of the latter scale seems disturbing, it turns out that many of the surface loops show such large variations for the same protein as studied by different crystallographic groups, particularly when no common protein is used as a starting point for refinement in both cases. Recognising the dynamic nature of some loops on enzymes, and including in the calculation the ability to handle dynamics over long timescales, allows analysis and refinement of enzyme inhibitors as pharmaceuticals. Here we analyse these aspects, particularly by reference to X-ray crystallography data.

Iron is a critically important metal for a wide variety of cellular events. The element holds this central position by virtue of its facile redox chemistry and the high affinity of both redox states (iron II and iron III) for oxygen. These same properties also render iron toxic when levels exceed the normal binding capacity of the cell. As a result of this potential toxicity, selective iron chelators are finding an important role in the treatment of iron overload associated with many forms of thalassaemia. In addition, they appear to have potential in treating situations where a local increase in iron concentration causes an unfavourable pathology, for instance, in reperfused tissue (heart disease and stroke) and in Parkinsonian brain. There is also evidence that iron chelators may minimise the toxicity of paraquat and the side effects of bleomycin and doxorubicin.

Non-haem iron enzymes can also be inhibited by iron chelators and consequently such enzymes as ribonucleotide reductase and lipoxygenase can be selectively inhibited. Such inhibitory action is being investigated for the treatment of malaria, neoplastic disease, psoriasis and asthma.

Recent developments in these areas are discussed in the present overview.

A wide variety of normal and malignant mammalian cells, as well as inflammatory cells, release extracellular active oxygen species such as superoxide and hydrogen peroxide. Both these species at low levels can stimulate growth and growth responses in a range of normal and malignant cell types. In order to assess the significance of such findings, superoxide dismutase and catalase were added exogenously to the culture medium of hamster (BHK-21) and rat (208F) fibroblasts (non-transformed or oncogene transformed). This resulted in a reduction of cell proliferation and increased cellular uptake of trypan blue stain, suggesting that superoxide and/or hydrogen peroxide may have important roles as extracellular ‘messengers’ promoting cell proliferation and maintaining cell viability. Superoxide dismutase mimics, like superoxide dismutase itself, are also cytostatic towards these cells and provide a basis for the development of tumour-specific antiproliferative drugs.