Clare Sansom

Tandem mass spectrometry: the tool of choice for diagnosing inborn errors of metabolism

Phenylketonuria (PKU), which affects about one in 10 000 ethnic Northern Europeans, is caused by mutations in phenylalanine hydroxylase (PAH). This enzyme converts phenylalanine (Phe) to tyrosine (Tyr); if it is defective, blood Phe builds up to toxic levels leading, in ‘classic PKU’, to severe mental retardation if left untreated. Infants with PKU will develop normally if their diet is restricted to exclude Phe-rich foods, starting in the first few months of life.

FISHing for aneuploidies nets accurate results

Conventional karyotyping, which is still the routine method used for identifying chromosomal abnormalities in the fetus, is expensive and time consuming. Results are not available until 1–3 weeks after amniocentesis, causing a distressing wait for affected women and their families. Now, a comprehensive clinical study by Bernd Eiben (Institut fur Klinische Genetik und Frauenklinik am Evangelischen Krankenhaus Oberhausen, Oberhausen, Germany) and co-workers1xRapid prenatal diagnosis of aneuploidies in uncultured amniocytes by fluorescence in situ hybridisation. Eiben, B. et al. Fetal Diagn. Ther. 1999; 14: 193–197Crossref | PubMed | Scopus (70)See all References has shown that a cheaper DNA-based assay can diagnose a range of chromosomal aberrations much faster than, and as accurately as, karyotyping. The commercially available AneuVysion™ assay, produced by Vysis (Downers Grove, IL, USA), uses the technique of fluorescence in situ hybridization (FISH) to detect additional copies of chromosomes 13, 18, and 21 and numerical gonosomal aberrations (aberrations of the sex chromosomes). These account for almost all clinically important congenital abnormalities of chromosomal origin (Table 1Table 1).Table 1Frequencies of the most common chromosomal abnormalitiesChromosomal abnormalityKaryotypeAssociated syndromeFrequency/live birthsaTrisomy 2147, +21Down1 in 700Trisomy 1347, +13Patau1 in 25 000Trisomy 1847, +18Edwards1 in 8000Gonosomal47, XXYKlinefelter1 in 1000Gonosomal45, XOTurner1 in 5000 View Table in HTML aData (apart from that for Down’s syndrome) from: Hawley, R.S. and Mori, C.A. (1999) The Human Genome: A User’s Guide, Academic PressIn Western countries, it is accepted practice to offer amniocentesis to women who have been predicted, after screening, to be at high risk of chromosomal aberrations. Risks are calculated using a combination of maternal age, serum markers and markers from ultrasound scans. Certain anatomical features that can be detected with ultrasound in the first trimester, particularly nuchal oedema, have been associated with Down’s syndrome. Over 80% of all Down’s syndrome pregnancies can be identified in the first trimester after screening2xAppropriate biochemical parameters in first trimester screening for Down’s syndrome. Cuckle, H.S. and van Lith, J.M.M. Prenatal Diagn. 1999; 19: 505–512Crossref | PubMed | Scopus (138)See all References.The FISH test involves attaching fluorescent markers to DNA probes that are specific for particular chromosomes or chromosomal segments. A fluorescent signal is produced when a probe hybridizes to its target within an amniocyte. Each type of probe produces a different colour signal, so the number of copies of each chromosome in a nucleus can be detected easily under a fluorescence microscope. There is no need to culture the amniocytes, and the test can be performed accurately with only 30–50 nuclei.Eiben and colleagues compared FISH results with standard karyotyping in 3280 women at high risk of chromosomal aberrations. The FISH test was successfully performed in 3150 cases, and there were no false positive or false negative diagnoses for any of the disorders tested. In over 94% of cases, a normal FISH result was confirmed by karyotyping. As it is fairly common to see a few nuclei with aberrant FISH signals in cytogenetically normal cases, and vice versa, intermediate cases were flagged as ‘suspicious’ and detailed karyotyping was performed. Several of these led to the diagnosis of a mosaic trisomy (in which an extra chromosome is only present in some cells).In 28 cases, an abnormal karyotype was detected, although the FISH result was normal. This underlines the main disadvantage of the FISH test: that, as it can only identify these particular aneuploidies, it will not detect every birth defect associated with chromosomal abnormalities. In contrast, the standard karyotype analysis will pick up a wide range of rare abnormalities, including translocations. Most infants affected by these abnormalities are phenotypically normal, but some have major or minor birth defects. Thus, there is a slight possibility that a normal FISH result will mask a serious problem. Steven Seelig (Vysis) explains one reason why this is rare: ‘Women are usually offered amniocentesis if they are at high risk of Down’s syndrome. A woman in this position is not necessarily at unusually high risk of any of these other, very rare conditions.’ Furthermore, as, in many cases, the outcome of a particular chromosomal abnormality is unknown, it is difficult to know how to counsel the women concerned.Currently, the FISH test is almost always performed in tandem with standard karyotyping. It is not yet clear to what extent this cheaper, simpler test will replace karyotyping in the future. Clearly, cost will be an important factor in both state-funded and insurance-based health systems. In the UK, amniocentesis is currently offered to all women thought to have at least a 1:250 risk of carrying a Down’s syndrome fetus. Howard Cuckle (Reproductive Epidemiology, School of Medicine, Leeds University, Leeds, UK) says: ‘If the average cost of testing for aneuploidies were to decrease, it should be possible to offer the tests to women at slightly lower risk, which would increase the overall detection rate.’ There is certain to be pressure to replace karyotyping with the cheaper tests, although this could only be at the cost of mis-diagnosing some rare chromosomal abnormalities. Eiben knows that similar pressures exist in Germany: ‘Both tests must be paid for out of the insurance system. It may be that our politicians will decide that standard karyotyping is too expensive, now that a cheaper test is available.’Cuckle is in no doubt that DNA-based tests, including FISH, will be increasingly important in the diagnosis of aneuploidies: ‘I am sure that we won’t be doing karyotyping for routine cases in ten years’ time.’ It should also be possible to develop molecular genetic methods of detecting non-chromosomal abnormalities. These will probably include most common mutations in the cystic fibrosis transmembrane conductance regulator, and the genetic defect responsible for Fragile X syndrome (the most common inherited cause of learning disability).

A role for Borna disease virus in schizophrenia

Borna disease virus (BDV) is known to cause neurological disease in many species of animals. Antibodies to BDV have often been found in human serum, more commonly in that of patients with certain psychiatric disorders. Now, for the first time, a Japanese group claims to have isolated the virus directly from the brain of a patient with schizophrenia1xIsolation of Borna disease virus from human brain tissue. Nakamura, Y. et al. J. Virol. 2000; 74: 4601–4611Crossref | PubMed | Scopus (91)See all References.Borna disease in horses was first described in the early nineteenth century, although its causative agent – a nonsegmented, negative-strand RNA virus from the order Mononegavirales – has only recently been characterized. Evidence indicates that the host range, geographic distribution and prevalence of BDV are broader than was previously thought. It is now known to infect sheep, llamas, cattle, cats and, possibly, humans. The neurological symptoms of BDV-infected animals mimic some of the symptoms of human affective disorders. Many studies have reported the presence of antibodies to BDV in human serum, and viral RNA in human cells. In most, but not all, of these studies, traces of the virus were found in a significantly higher proportion of patients with affective disorders than in healthy controls2xBorna disease virus and the brain. Gonzales-Dunia, D. et al. Brain Res. Bull. 1997; 44: 647–664Crossref | PubMed | Scopus (74)See all References.Kazuyoshi Ikuta and colleagues (Institute of Immunological Sciences, Hokkaido University, and Research Institute for Microbial Diseases, Osaka University, Japan) examined autopsy brain tissue from four schizophrenic patients and two controls for markers of BDV (Ref. 1xIsolation of Borna disease virus from human brain tissue. Nakamura, Y. et al. J. Virol. 2000; 74: 4601–4611Crossref | PubMed | Scopus (91)See all ReferencesRef. 1). One patient (P2), who was found to be seropositive for BDV, also had viral RNA in several brain regions, including the hippocampus. Neural damage, similar to that observed in BDV-infected animals, was observed in the hippocampus of this patient but of no other subject in this study.The virus itself was successfully isolated from the brain of patient P2 using two different experimental approaches. In one, the group injected BDV-positive brain tissue from P2 into the brains of newborn Mongolian gerbils. Subsequently, the virus was isolated by inoculating oligodendroglia (OL) cells with brain homogenates from these gerbils. The second method involved a collaboration with Juan Carlos de la Torre (Scripps Institute, La Jolla, CA, USA). De la Torre isolated the virus by transfecting Vero cells with ribonucleoprotein complexes derived from BDV-positive human and gerbil brain tissues.Patient P2 was 26 years old at death and had been diagnosed with undifferentiated schizophrenia only two years previously. All previous unsuccessful attempts to isolate BDV from human brain tissue had involved elderly patients. Ikuta believes that this might be an important factor: ‘It is conceivable that infection with BDV quite early in life could influence the onset or the course of neuropsychological disorders.’Schizophrenia is a multifactorial disease with a complex etiology; almost certainly, both genetic and environmental factors are involved in its development. It is possible that infection with BDV could be one of the environmental factors that affect the course of disease in some patients. However, there are other explanations for these results. De la Torre explains one possibility: ‘Maybe individuals with schizophrenia could become more susceptible to infection with certain viruses.’The controversy over human Borna disease virus extends far beyond speculation over its precise link with schizophrenia. Some experts do not even agree that this virus can infect humans. In a recent report3xSequence similarities between human bornavirus isolates and laboratory strains question human origin. Schwemmle, M. et al. Lancet. 1999; 354: 1973–1974Abstract | Full Text | Full Text PDF | PubMed | Scopus (53)See all References, Martin Schwemmle and colleagues (Department of Virology, University of Freiburg, Germany) conclude from sequence analysis that most reports of human BDV infection are likely to be contaminants from laboratory strains. Peter Staehli from this group is not convinced by Ikuta’s work: ‘These findings fit nicely into the picture [described in Schwemmle’s paper] that suggested that most (if not all) human BDV isolates known to date might represent contamination artefacts’. The virus is known to mutate slowly, and the evidence from sequence analysis is still inconclusive. Staehli suggests that multicentre studies should be organized in which multiple samples from the various specimens are stored for independent testing. De la Torre agrees: ‘We need a more comprehensive approach, with many more patients and collaboration between labs.’ If independent multicentre studies definitely conclude either that Borna disease virus can infect the human brain, and is associated with schizophrenia or other affective disorders, or disprove this theory entirely, it would be an important step forward in elucidating the molecular basis of complex diseases.

A breath of fresh air for patients with lung disease

Inspire Pharmaceuticals, Inc. (Research Triangle Park, NC, USA) has announced promising results from its first Phase I clinical trial of a novel P2Y2 receptor agonist, INS365, as a potential treatment for respiratory diseases including cystic fibrosis (CF). Speaking at the North American Cystic Fibrosis Conference held in October 1997 in Montreal, Canada, Christy Shaffer, Inspires Vice President (Development) and Chief Operating Officer, said: ‘INS365 represents a new therapeutic approach to the treatment of CF. We expect it to promote airway hydration and enhance the natural process of mucociliary clearance.’Cystic fibrosis is the most common fatal autosomal-recessive disease in Caucasians, affecting an estimated 30 000 North Americans and 20 000 Europeans. Loss-of-function mutations in the gene for the cystic fibrosis transmembrane conductance regulator (CFTR), which block its Cl− channel function, lead to the secretion of abnormal mucus that is difficult to clear. Although the prognosis for CF patients is much improved, most still die in early adulthood from chronic lung disease.INS365 is a specific agonist of the P2Y2 subclass of G-protein-coupled receptors (Fig. 1Fig. 1), which are found in high density on airway epithelial cells and coordinate the mucociliary clearance process. Agonists of this receptor subclass stimulate this natural process by increasing cilia beat frequency, stimulating chloride efflux—via a non-CFTR Cl−channel that bypasses the defect in CF patients—and modulating mucin release from goblet cells.Figure 1INS365s mechanism of action at the plasma membrane of respiratory epithelial cells. The drug binds to the P2Y2 receptor, activating phospholipase Cβ through a G-protein-coupled mechanism. The resultant increase in inositol (1,4,5)-trisphosphate (IP3) levels causes release of CA2+ from the endoplasmic reticulum. This stimulates cellular functions such as increased cilia beat frequency and Clbtl efflux, which improve mucociliary clearance. Kindly provided by Inspire Pharmaceuticals, Inc., Research Triangle Park, NC, USA (http://www.inspirepharm.com/).View Large Image | Download PowerPoint SlideINS365 was tested pre-clinically using standard FDA acute toxicity assays in dogs and rats. ‘It showed a very clean toxicological profile, with no pulmonary or systemic toxicity, even at high doses. We were able to take it from a pre-clinical lead into the clinic in less than a year’, explains Shaffer. The Phase I clinical trial was initiated in the UK, at the beginning of 1998. This was a placebo-controlled, double-blind study, giving single escalating doses by nebuliser to human volunteers, including smokers. Regular smokers are often used in clinical trials as a model for sufferers from chronic lung disease, as their bronchi are easily irritated. ‘A difference between the smokers and non-smokers was first observed at the second dose level. The smokers began to show increased mucus clearance’, reports Sylvio Tamin [Pharmakopius, Goring, UK (now Valorum)], chief investigator in the clinical trial. The drug was well tolerated and the most frequent side effect, cough, had been expected owing to its action in mobilizing secretions.Inspire is now recruiting patients into a Phase I study in the USA, to evaluate the effect of INS365 on mucociliary clearance in smokers and non-smokers. An additional trial is being started in CF patients; this will eventually involve 48 adult and 48 paediatric subjects with mild to moderate CF lung disease. It is being organized through the Cystic Fibrosis Foundation (http://www.cff.org/), which helps pharmaceutical companies set up clinical trials by coordinating with the Therapeutic Development Network, comprising seven of the top clinical CF research centres in the USA.In September 1997, Inspire announced a partnership with the leading Japanese pharmaceutical company Kissei Pharmaceutical Co. Ltd. Kissei has acquired the rights to market INS365 in Japan. As few ethnic Japanese suffer from CF, the drug will be marketed there for chronic obstructive pulmonary disease (COPD), including bronchitis. ‘INS365 provides an excellent opportunity to augment our respiratory product line … we believe that it will become an extremely important treatment for COPD’, says Mutsuo Kanzawa, President and CEO of Kissei. Inspire is also in discussion with global pharmaceutical companies for the development of INS365 for both CF and chronic bronchitis outside Japan. Another ongoing project is the development of an aerosol delivery system for the drug, which should improve patient compliance.

AGGREGATION INHIBITORS : HOPE FOR HUNTINGTON'S

Huntington’s disease (HD) is an autosomal dominant, progressive neurodegenerative disease that affects 3–4 in every 10 000 Caucasians. Its symptoms include the involuntary movements that gave rise to its old name of Huntington’s chorea, and it leads to increasing disability and death within 10–15 years. The genetic defect that causes this devastating disease is an expanded CAG repeat in a gene encoding a protein of unknown function known as huntingtin. Unaffected individuals have fewer than 35 CAG repeats in exon 1 of this gene; individuals affected by HD have between 37 and 120 such repeats. As this mutation is in a coding region, the affected protein contains an expanded polyglutamine (polyQ) tract. The number of glutamines is one of several factors that determines the age of onset of the disease. In an introductory lecture at the FEBS meeting, Jean-Louise Mandel (Universite Louis Pasteur, Illkirch, France) explained: ‘Every extra glutamine takes 18 months off the average age of onset of Huntington’s symptoms’. Several other neurological conditions are known to be associated with expanded polyQ repeats in proteins. The disease-causing threshold of 35–40 glutamines seen in HD is typical of this type of disease.

Genotyping HIV isolates paves the way to effective treatment regimes

The success of the anti-HIV drugs commonly prescribed throughout the West is seriously limited by the development of resistant strains of the virus. In most Western countries approximately 10–13 different anti-HIV drugs are available on prescription. Patients generally take a cocktail of these, which often includes both reverse transcriptase inhibitors and protease inhibitors. Under the current optimum standard of care (SOC) regimes, up to two drugs are changed when increased viral load indicates that drug treatment is failing, without the clinician knowing which drugs have failed.Pierre Dellamonica, Jacques Durant, Philippe Clevenbergh and colleagues (Archet Hospital, Nice, France) have shown, in a recent clinical trial, that using HIV genotyping to select drug treatments can yield significant clinical benefit. This trial, known as VIRADAPT, tested a genotyping technique in which parts of the HIV genome were sequenced using an HIV sequencing kit (TruGene™) developed by Visible Genetics, Inc. (Toronto, ON, Canada). Currently, the regions of the genome encoding the protease and part of the reverse transcriptase proteins are routinely sequenced. It should be relatively straightforward to extend this sequencing if anti-HIV drugs with other targets are introduced. Those mutations that are associated with resistance to particular drugs can be identified from publicly available resistance tables. The treatment for an individual patient is selected from those drugs not ruled out by resistance. According to Clevenbergh ‘Drug selection can only be as accurate as the drug resistance tables available. We will need to refine the resistance tables continually, as new treatments are introduced and new mutations emerge.’VIRADAPT was a controlled, randomized clinical trial involving 108 HIV-1 infected patients. At its onset, all selected patients had viral loads above 10,000 copies ml−1, and were taking at least one protease inhibitor. Those patients assigned to the control arm were treated using SOC. Results from the first six months of the trial, recently published in The Lancet1xDrug-resistance genotyping in HIV-1 therapy: the VIRADAPT randomised controlled trial. Durant, J. et al. The Lancet. 1999; 353: 2195–2199Abstract | Full Text | Full Text PDF | PubMed | Scopus (643)See all References1, showed a reduction in viral load to below detectable levels in 32% of the patients on the genotyping arm versus 15% of those on the control arm. As this difference was statistically significant, patients in the control group were then switched to genotyping. Results from the full twelve-month study were presented in June at the third international workshop on Drug Resistance and Treatment Outcomes (San Diego, CA, USA). The mean viral loads in the patients originally assigned to the genotyping arm were similar to those recorded at six months. Mean viral loads in the patients on the control arm also decreased significantly after six months of genotyping. John Stevens (President and CEO of Visible Genetics) was encouraged: ‘These are the first pilot results to indicate that genotyping may have a long-term clinical value.’ The French Health Ministry will begin reimbursing the cost of HIV genotyping this autumn, and Visible Genetics’ HIV genotyping assay and system are currently in clinical trials for FDA approval.The higher the viral load, the faster mutations develop. It has been estimated that, with a viral load of 1000 copies ml−1, approximately 56 000 random mutations will emerge each day. If viral load is reduced, the frequency of mutations drops in direct proportion. Therefore, if drug-resistant mutations can be detected at low viral loads, claims Rob Lloyd, Jr. (Scientific Director of Applied Sciences, Inc., a subsidiary of Visible Genetics) ‘it should be possible to stay one step ahead of the virus and change drug regimes before they fail’. Visible Genetics has developed a test that can determine genotypes from extremely low viral loads and this should increase the efficacy of this technique. Using a new kit for extracting RNA (TruPrep™), it is now possible to obtain genotypes at viral loads as low as 60 copies ml−1. This is a more than tenfold reduction compared with current techniques. The test is an adaptation of a laboratory kit for RNA extraction produced by Qiagen, Inc. (Venlo, The Netherlands). The volume of plasma used is larger than in previous methodologies, and viral RNA is concentrated using ultracentrifugation. The PCR and DNA sequencing steps are performed using TruGene™ technology. The test is currently available for research use only. Visible Genetics is collaborating with the AIDS Research Consortium of Atlanta, Inc. (ARCA) to establish a clinical trial of ultra-low genotyping. If this is successful, it may become a part of the standard treatment for AIDS. Lloyd believes that ‘genotyping of HIV for resistance will become the next surrogate marker for monitoring AIDS infection, alongside CD4 cell counts and viral load’.This technique may also be applicable to other viral diseases. Visible Genetics is developing a kit for use in hepatitis B. Genotyping the hepatitis B virus could be used to predict both the viral resistance to lamivudine (Epivir-HBV) and the likely response to immunotherapy.

Making sense of the human proteome

Now that the task of sequencing the genome is well under way (the first draft of the human genome sequence will be made available in the first few months of next year), the focus of research will switch to the protein level, relating sequences and, where possible, structures to protein function. In June, Amos Bairoch (Swiss Institute of Bioinformatics, Switzerland), the curator of the SwissProt protein sequence database (http://www.expasy.ch/sprot)1xThe SWISS-PROT protein sequence data bank and its supplement TrEMBL in 1999. Bairoch, A. and Apweiler, R. Nucleic Acids Res. 1999; 27: 49–54Crossref | PubMed | Scopus (399)See all References, announced a project to ensure that high-quality human protein sequence data are released into the public domain as soon as practicable – the Human Proteomics Initiative (http://www.expasy.ch/sprot/hpi).SwissProt is the most widely used protein sequence database. Its value lies at least as much in the annotations and cross-links of each entry as in the sequence information itself. A typical entry for a human gene contains detailed descriptions of its known polymorphisms and any associated diseases, and cross-references to databases including Medline, Online Mendelian Inheritance in Man (OMIM) and the gene sequence databases. Bairoch described the ‘golden goals’ for SwissProt: that it should be fully annotated, complete, non-redundant, highly cross-referenced, and available through a wide range of servers and analysis tools. He acknowledged, however, that ‘there is a dichotomy between the first two aims, as annotation is time consuming and expensive’. New investment in SwissProt, resulting from the fees now charged to industrial users of the database, is now being devoted to the Human Proteomics Initiative. Its aim is nothing less than to catch up with the gene sequencers, by releasing a fully annotated set of human protein sequences and their variants into the public domain.The information in the human proteome goes far beyond coding region sequences. Bairoch and his colleagues will collect and annotate information on mammalian orthologues of human genes, post-translational modifications and polymorphisms. Over 150 different post-translational modifications of human proteins (such as cleavage sites, crosslinks and bonds to non-protein molecules) have been described in the literature. Knowledge of the complete set of human polymorphisms is necessary for investigating variations in predisposition to common diseases.The Initiative will have two phases. The aim of the first phase, which should last until Spring 2000, is to annotate the protein products of all known human genes. The second is a long-term commitment to continue releasing well-annotated sequences of protein variants for as long as researchers produce new data. The initiative is bound to make an important contribution to our understanding of the genetic basis of disease, and it cannot succeed without collaboration. Bairoch appealed for help to the whole medical research community: ‘We need you to submit all types of information to speed up the comprehensive annotating of the human proteome.’

Carbohydrate-based diagnostics: one lump or two?

Many disease states are characterized by changes in the expression patterns of oligosaccharides and glycoproteins. Research presented at the tenth European Carbohydrate Symposium, Eurocarb X (held in Galway, Ireland in July 1999) is likely to lead to improvements in carbohydrate-based diagnostic techniques for a range of diseases, from prion disease (Elizabeth Hounsell, University College, London, UK) to schistosomiasis (Henricus J. Vermeer, Bijvoet Centre, Utrecht, and Laboratory for Parasitology, Leiden, The Netherlands). But perhaps the most immediately applicable advance was in the area of breast cancer prognosis.Most deaths from breast cancer occur as a result of secondary tumours, so predicting metastasis is an important factor in predicting cancer outcome. Differences in the glycosylation patterns of glycoproteins in breast tumours can be correlated with the likelihood of the tumours metastising. Miriam Dwek (University College, London, UK) described a method of purifying oligosaccharides from the glycoproteins in breast tumour tissue and characterizing them using high-performance liquid chromatography (HPLC). Between 30 and 70 different oligosaccharides could be extracted from each tumour, and analysis of breast tumours derived from 76 patients with known outcome revealed that the tumours from patients who were free of disease five years later were different from those with more aggressive tumours. High levels of two particular oligosaccharides were often found in aggressive tumours, and these are now being sequenced. ‘Oligosaccharide fingerprints may help to predict which patients will develop metastases after breast cancer,’ explains Dwek; ‘this is particularly important for patients with small tumours detected by mammography. Once we have characterized the oligosaccharides, we plan to target them in a new therapeutic approach for the treatment of metastasis.’ The group at UCL is also collaborating with programmers at Brunel University (Brunel, UK) to analyse these complex chromatograms automatically.