DDT Vol. 6, No. 9 May 2001 updatenews

1359-6446/01/$ – see front matter ©2001 Elsevier Science Ltd. All rights reserved. PII: S1359-6446(01)01792-5 443

Several genes associated with schizo-phrenia have recently been discovered,which could speed up the search for effective drugs, diagnostics and markersof treatment safety and efficacy for thedisease. Researchers at deCODE genetics(Reykjavik, Iceland) have identified aschizophrenia-associated gene that en-codes a protein that is expressed at thesynapses of neurons in the CNS. Studiesare now under way by deCODE to further dissect the molecular pathway ofschizophrenia; indeed, the experimentalexamination of the identified diseasegene in a two-hybrid yeast system has already revealed several other genes thatare expressed at synapses, thus leadingto the elucidation of a novel signalingpathway.

A difficult diseaseSchizophrenia is a complex, debilitatingillness that affects 0.5–1.0% of theworldwide population, starting mostcommonly in early adulthood, and acuteexacerbations of the disease are typifiedby psychosis, hallucinations, thoughtdisorder, delusions and disturbed bodymovements. There is currently no effec-tive causative therapy for schizophreniabecause little is known about the molecular pathology of the disease.However, it has long been speculatedthat the disease has a genetic compo-nent, because of the substantial degreeof inheritability of the disease, as indicated by familial clustering.

The population-driven approachAs part of its population-based geneticresearch, deCODE has conductedgenome-wide scans of DNA samplesfrom large families in Iceland in whichseveral family members are afflicted withschizophrenia. Strong population-wide

linkages were observed and the diseasegene was subsequently identified by finemapping and haplotype analysis (Fig. 1).Despite studying a discrete populationfor disease genotypes, Kari Stefansson,CEO of deCODE, is optimistic that the

results will be relevant to other popu-lations: ‘Icelandic models of human disease are valid, and prevalent diseasesin Iceland are prevalent elsewhere. Theopinion that disease differs regionally isnaïve.’ However, he adds that variation

New genes linked to schizophreniaJoanna Owens, Joanna.Owens@current-trends.com

Figure 1. The deCODE research process. Using DNA samples from large Icelandicfamilies, genome-wide scans are carried out to identify significant gene linkages. Disease-associated genes are then localized by further scrutiny of a specific region on thechromosome by fine-mapping and haplotype analysis. The candidate genes and theirproduct(s) can then used as putative therapeutic targets and will help to elucidate themolecular pathogenesis of diseases such as schizophrenia.

Familial analysis

Expressed protein

Drug discovery

Significant gene location

Dissection of molecularpathway

Candidategene

Fine-mappingHaplotype analysis

Identification of additionalcandidate genes

Genome-wide scan

Drug DiscoveryToday

update news DDT Vol. 6, No. 9 May 2001

444

in the frequency or prevalence of a disease gene within a population mightbe a factor to consider in the develop-ment of diagnostics for diseases such as schizophrenia. deCODE is workingwith groups of patients in the USA and Europe in their quest to study the genetic basis of ~60 diseases. These include Alzheimer’s disease, type 1 andtype 2 diabetes mellitus, osteoarthritis,osteoporosis, stroke, manic depression,anxiety, rheumatoid arthritis, inflamma-tory diseases, hypertension and a varietyof cancers, in addition to aging.

The contribution of gene variantsAlthough several gene localizations havebeen linked to schizophrenia (reviewedin Ref. 1), these data represent the first time that a single gene has beenidentified as contributing to the disease.However, Stefansson stresses that although deCODE has demonstrated a direct link between this gene and thedisease, it does not eliminate the role ofother genetic and environmental factors:‘All of the common diseases are com-plex, and it is this complexity that makesthem common.’ He continues, ‘For example, in Iceland, no non-smokers develop emphysema; however, only15% of those who do smoke develop the disease. In these people there is a genetic component, but without the environmental factor (i.e. smoking)there would be no cases of emphysema.’Stefansson believes that locating thegenetic components of a disease couldteach us more about the environmentalfactors that are involved.

Hot on the heels of this discovery bydeCODE is the identification of anotherschizophrenia-associated gene, WKI1, byKlaus Peter Lesch and colleagues at Julius Maximilian-University of Würzberg(Germany)2. From a study of three gen-erations of an extended German family,of which eight members had schizo-phrenia, a gene on chromosome 22 wasidentified by linkage analysis. Further investigation has revealed that the gene

encodes a protein that is similar toknown ion channels and might be in-volved in the transport of nerve impulsesin the CNS. It is thought that a mutatedform of the protein, in which a leucineresidue is replaced with a methionineresidue in one of the seven transmem-brane domains, might cause abnormaltransmission of electrical impulses in thebrain.

Validated targetsA major advantage of genetic studies,such as those carried out by deCODEand the University of Würzburg, is thevalidation of targets at an early stage in drug discovery. As Stefansson high-lighted, ‘The pharmaceutical industryagrees that it is overburdened with tar-gets from genomics and can’t prioritize’;the targets are plentiful, but hypotheti-cal, as they are derived mostly from look-ing at differences in gene expression between diseased and healthy tissue.Stefansson points out the limitation ofthis: ‘If you hit a pancreas with a ham-mer, you will change the expression ofmany genes, but this won’t lead you to the hammer as the cause, that is, 99% of this change will be reactive notcausative’. Stefansson believes that drugdiscovery is better when based on de-ductive reasoning from what is alreadyknown about a disease. Therefore, evenif the disease gene associated with schiz-ophrenia proves to be inappropriate as a target for drug discovery, it will have provided essential new insights to our understanding of the complexschizophrenia pathogenesis.

Future drug discoveryThe schizophrenia-associated gene dis-covered by deCODE, and its proteinproduct, are now being studied by researchers at F. Hoffmann-La RochePharmaceuticals (Basel, Switzerland), aspart of its drug discovery collaborationwith deCODE. ‘Only a more specificunderstanding of the pathologicalprocesses on the molecular level – where

disease first happens – will ultimatelyallow continued progress in medicine,’commented Jonathan Knowles, GlobalHead of Research at F. Hoffmann-LaRoche. ‘Genetic research is one of themost powerful approaches to gather thisknowledge and has become an essentialcomponent of biomedical research,’ hecontinued. The aim is to use these puta-tive targets, and others that are revealedfrom further dissection of the schizo-phrenia pathway, in the search for con-ventional small-molecule therapeutics.‘It will still be a long way to go from tar-get to a new medicine that is approvedby the regulatory authorities and can begiven to patients suffering from schizo-phrenia’, Knowles points out, ‘but usingtargets that we know are truly involvedin the causation of the disease shouldmake such new medicines overall moresuccessful, and particularly in those zpa-tients in whom the particular mecha-nism linked to the gene targeted by thenew drug contributes importantly.’ KlausLindpaintner (Head of Roche Genetics,Europe) also commented that, ‘As wemove forward, more of the genetic factors predisposing or contributing tovarious diseases are discovered and, indue course, new drugs are developed, It will become increasingly important to understand better the presence or absence of any of these genetic factorsin a given patient, to provide the “cus-tom-tailored” medical regimen thatpromises the greatest likelihood of beingsuccessful.’

The ultimate prize for population-based genomics will be the developmentof efficient and effective diagnostics for genetically based diseases such asschizophrenia, which will enable theearly identification of individuals who arepredisposed to the disease and thus offer the opportunity for lifestyle changes, in addition to effective personalizedtreatments.

In keeping with this logic, RocheDiagnostics has recently announced its intention to enter into a 5-year

Promising results from a study with anew intelligent inhalation technology inasthmatic children has prompted the development of the technology for usein a variety of other conditions, includ-ing the administration of systemic drugs.The six-month study using the AdaptiveAerosol Delivery (AAD™) system fromProfile Therapeutics (Bognor Regis, WestSussex, UK) was shown to improve accuracy of dose delivery, decrease thequantity of drug required and improvecompliance compared with conventionalnebulizers.

One of the problems found with manynebulizer systems is that they are depen-dent on the breathing pattern of the patient, the way the nebulizer system is used (which can cause variability ofsometimes >60%)1 and the drug outputcharacteristics of the nebulizer2. Everypatient breathes differently, especially inrelation to flow, volume and frequency3.If the patients’ inhalation flow rate doesnot exceed the nebulizer flow rate, thenmore of the nebulizer output is wastedand this is particularly important in dos-ing for the paediatric population4. Usingnebulizers has been found to waste up to60–70% of the drug as only 30–40% ofthe respiratory cycle is accounted for byinspiration5,6. The quantity of drug actu-ally inhaled by the patient is especially

important for the administration of drugswith a narrow therapeutic window.

The systemThe AAD technology was designed to tryto avoid these problems by deliveringthe drug only during the optimal phaseof inspiration (i.e. the first 50% of eachbreath). Sensors in the system monitor

the flow rate and the length of the firstthree breaths of the patient before start-ing the dose. The results are then aver-aged and used to calculate the correctquantity of drug to be delivered intoeach breath. The patient’s breathing isthen monitored every 30 msec duringdosing to enable the device to adapt therelease of aerosol to changes in the

DDT Vol. 6, No. 9 May 2001 updatenews

445

Figure 1. Patient adherence with four-week domiciliary use of conventional nebulizers (a)and (b) compared with using the AAD system (c) and (d) for the treatment of COPD. Allpatents were aged over 65 years. In (a) and (b), the regimen consisted of 2.5 mgsalbutamol four times daily. In (c) and (d), the regimen consisted of 2.5 mg salbutamolplus 500 µg ipatropium bromide four times daily.

Drug DiscoveryToday

1 10 20 30 40

19:12

14:24

09:36

00.00

24.00

04:48

Tim

eT

ime

Day Day1 5 10 15 20 25 30

500

400

300

200

100

0

Inha

led

dose

(sec

)

Mean dose 166 sec (CV 65%)Adherence to regimen 69%

1 10 20 30 40

19:1214:2409:36

00.00

24.00

04:48

Day1 5 10 15 20 25 30 35

100

50

0

Pre

-set

dos

e(%

)

Day

Compliance to pre-set dose 98%Adherence to regimen 97%

(a) (b)

(c) (d)

1359-6446/01/$ – see front matter ©2001 Elsevier Science Ltd. All rights reserved. PII: S1359-6446(01)01791-3

Intelligent inhalers for systemicadministration?Rebecca N. Lawrence, News & Features Editor

collaboration with deCODE to developDNA-based diagnostics from the resultsof deCODE’s genetic studies. This alliance will complement the existingdrug discovery partnership between the two companies, and involves Roche

providing research funding, milestonepayments and product royalties todeCODE.

References1 Bray, N.J. and Owen, M.J. (2001) Searching

for schizophrenia genes. Trends Mol. Med. 7,

169–174

2 Meyer, J. et al. (2001) A missense mutation

in a novel gene encoding a putative cation

channel is associated with catatonic

schizophrenia in a large pedigree. Mol.

Psychiat. 6, 304–308