Old mother Hubbard’s cupboard

Critical assessment of congruence of biological systems with human disease mechanisms is one of the prime pre-requisites of good drug discovery research, says Dr V Sudarsanam

Drug discovery (DD) is all about translating fundamental research to clinical advances, about focusing science to convent exciting discoveries to beneficial cures. A good and well-executed drug discovery programme can almost be a parable for teamwork, a parable of innovation, intuition, insight, imagination, experimentation, creativity, brain power and hardwork all leading to the 3Cs of "collaborate, contribute and cure" for the common good, the public good.

DD covers everything from basic disease mechanism, the validation of targets, the chemistry and the biological screening to indentifying ‘actives’,‘hits,’ optimisation via lead compounds, downstream preclinical studies including ADME-TOX, clinical studies to marketing of a new medical entity (NME) in an appropriate convenient dosage form.

The current paradigm of drug discovery emphasises the exploitation of genomic knowledge with bioinformatics playing a key role in unlocking the information contained in the DNA sequence (genome), the mRNAs (transcriptome) and proteins (proteome) of cells to:

a) identifying putative drug targets

b) aid the identification of sequence variants of genes that genetically predispose individuals to developing disease and

c) help to establish a firm association between putative targets and the disease of interest (target validation).

All these have created a great sense of exploration, a moon walk kind of excitement in modern day drug discovery.

However, all these have to be tempered with a sense of realism. Humans have hundreds of functionally specialised cell type that interact differently with environmental factors to influence disease development and to modulate the effects of drugs. The relationships between endogenous metabolic processes (coded in the genome and intrinsic to the cellular function) and xenobiotic metabolism are poorly understood, especially with respect to environmental factors, further compounded by the difficulty in the understanding of the gene-disease relationship and the vexed issue of idiosyncratic toxicity.

Further, in view of the polygenic nature of many of the diseases, there is a need to visualise and define biological systems not as one gene, or one protein, or even as one pathway, but rather as the ability to understand the total complexity of the biological system in multiple dimensions.

Even though the 3-dimensional structures of some target proteins gained either experimentally (structural genomics) or computationally (homology modelling) are available for the rational design of drugs, it needs to be remembered that many of the protein targets like GPCRs (G-protein coupled receptors) are membrane bound or incorporated into membranes, with many of them being solvated or glycoslyated to a quaternary configuration about which very little information is available.

Further, post translational modifications such as phosphorylation protein cleavage and the creation of protein complexes can greatly increase the number of functionally distinct protein entities produced from one gene further making difficult our understanding of disease process. As conformational changes bring about the transformation of a receptor from a resting or inactive state to an active one, there is a lot more to understand about the processes of "conformational selection" and/or conformational induction" producing a texture to drug effects [TIPS vol 24, (7) 346, July 2003] [Nature Biotech vol 19, 1099 Dec 2001].

In the light of such formidable challenges (to mention a few) peppering the process of drug discovery, there is an absolute need to put screening methodology on totally sound footing so that medicinal researchers can recognise patterns in biological data, relate these trends to molecular properties and ultimately use them to guide, further synthesis iterations (Drugs Disc and Dev p35, Nov 2003). Dr David Horrobin of UK highlights the importance of establishing the congruence between in-vitro and animal models of disease and the corresponding human condition -- a fundamental assumption of much biomedical research -- and points out that this is very rarely critically analysed. He also adds that in the absence of such critical assessment, the assumption of congruence may be invalid for most models.

This is all the more necessary in in-vivo models as these can predict the integrated response of a drug in which a combination of genetic biochemical, pathological and environmental influences work in concert. It can hardly be overemphasised that the response of an organic system to a drug cannot be fully predicted from the behaviour of an isolated cell, chiefly because the properties of organic systems are emergent multifactorial responses to both internal and external environmental elements.

There are other factors which point to the limitations that researchers face. Human research has been tied by law to prior research on non-human animals. This, if viewed dispassionately, appears odd as the biochemical mechanisms of different species differ enough that wholesale transfer interventions from an animal model to humans or a human model to animals can be chancy and cannot be made without some risk. However, as animal anatomy and physiology bears reasonable similarity to the human counterparts, because the latter have evolved from other species through processes of elaboration and modification of the originating species’ genome, fundamental mechanisms and processes discoverable by studying one or more animal species can be a staging ground for development of therapies. It is for this reason models that mimic relevant human processes and conditions need to be established for proper drug research work (congruence).

At this juncture it may be germane to look at a few examples. One of the approaches for the design of drugs for the treatment of obesity and diabetes has been the concept of Beta 3 - adrenoceptor agonism. In spite of major efforts spread over more than three decades, no drug has emerged in the market place based on this approach. Stimulation of this receptor results in signals through cyclic AMP - activating protein kinase A which stimulates upregulation of PGC1 messenger RNA in target tissues.

PGC1 is a transcriptional co activator that co ordinates a thermogenic programme that includes the unregulation of uncoupling proteins, genes of the mitochondrial electron transfer pathway mediated through factors, peroxisome proliferator activated receptor, thyroid hormone receptor Beta and retinoic acid receptor. Mitochondrial biogenesis is also unregulated, mediated through nuclear receptor factors resulting in the generation of heat (and ATP).

The models used for the evaluation of new chemical entities are genetically deficient, obese, insulin-resistant mice wherein the target tissue of thermogenesis is BAT (brown adipose tissue) and protein involved being UCP 1 (uncoupling protein 1). It is to be remembered that BAT is present in humans only during neonatal period and after that its presence is extremely limited. Further, in humans the main organ of thermogenesis is the skeletal muscle and the proteins involved are UCP 2 and UCP 3 (see Nature 404, 652, 2000).

What is paradoxical has been identification of new candidates based on the activity in mice involving BAT and UCP 1, when the target tissue in humans is muscle and proteins involved are UCP 2 and UCP 3, making one wonder whether the lack of success in this approach even after three and half decades is due to lack of congruence in models used for testing.

As another instance we can look at another therapeutic area as to whether we as scientists have always been wise and far sighted enough or have been carried away by immediate kudos.

Bacterial organisms consist of two populations that exist simultaneously - multiplying and non multiplying [slowly multiplying, latent, dormant (which require resuscitation from a culture negative state)]. However, all our antibacterials have been developed by targeting with methodologies oriented towards fast multiplying bacteria. Non-multiplying states are resistant to a wide range of environmental stresses such as heat and to antimicrobial drugs. Hence non-multiplying organisms can mutate merrily in the presence of drugs, become genetically resistant and when situations are favourable start to multiply.

What is not very obvious is while tubercular organisms has been targeted at both multiplying and dormant states (rifampicin), while parasitic diseases like amoebiasis have been targeted at multiplying trophozoite (metronidazole) and dormant cyst (furamide) states, how is it that the same has not been done with bacterial organisms and in the process, have we encouraged bacterial resistance.

Many other such examples can be cited to highlight the importance of congruence. It is all the more pertinent because, in spite of the great advance in new science, new knowledge and new technologies, the productivity in terms of new drug introductions in USA has drastically fallen from over 50 in early nineties to 17 in 2002 (and the 17 includes some biotechnology derived products).

In today’s changed scenario (unlike earlier years when corporate drug research was ruled by knowledgeable scientists (eg. Dr Roy Vagelos at Merck), today’s scientists report to business and are under corporate heads obsessed with only speed (rational or otherwise) and business plans. To quote O Morgan ‘‘Pharmaceuticals is a long-term game and much of the far-off stuff can only be understood by people with a strong scientific background. The corporate management on the other hand thinks ‘after lunch’ is a long time away’’ (Observer (Lond) Jan 23, 2000). In such a surcharged scenario, is it just possible that modern day scientists are forced to relax the punctilious attention to meticulous detail and overlook the mantras of good drug discovery endeavours postulated in science: [Algorithm for drug discovery vol 292, 13, 6th April 2001]

• Slow down and explore
• Pursue quality for its own sake
• Look at raw data as unusual observations can lead to breakthrough
• Cultivate smart friends (only good teams can deliver great dreams) and also nurture intellectual freedom to develop original (and sometimes heretical) concepts.

If good science, good sense and good attitudes are not nurtured, it is more than likely that the good blockbuster drugs with tangible benefits emerging from the pipeline may become as bare as old mother hubbard’s cupboard and organisations can only recite what most of us learnt as nursery rhymes:

Old mother Hubbard, went to her cupboard To fetch her poor dog a bone When she got there The cupboard was bare and the poor dog had none.

The writer is a Mumbai based research consultant