| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Medicine and The James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research (Drs. Tebbutt and Paré), Department of Medicine, St. Paul’s Hospital, University of British Columbia, Vancouver, BC; and School of Medicine and Pharmacology (Dr. James), University of Western Australia and West Australian Sleep Disorders Research Institute, Sir Charles Gairdner Hospital, Perth, Australia.
Correspondence to: Scott Tebbutt, PhD, James Hogg iCAPTURE Centre for Cardiovascular and Pulmonary Research, St. Paul’s Hospital, University of British Columbia, 1081 Burrard St, Vancouver, BC, Canada, V6Z 1Y6; e-mail: stebbutt{at}mrl.ubc.ca
Abstract
Human genetic variation has enormous implications for individual susceptibility to lung disease, as well as for differences in prognosis and response to therapeutic interventions. Single-nucleotide polymorphisms (SNPs) are the most common type of DNA sequence variation. An SNP is the substitution of a single base in the sequence for one that is different from that present in the majority of the population. In this review, we describe in more detail what SNPs are, how they are discovered, and their potential to elucidate the genetic basis of lung disease. We illustrate several examples of how SNPs are being used—or are poised for use—in diagnostic and therapeutic applications. We conclude with a brief discussion of the future of medicine and how genetic knowledge and application can play an ever-increasing and important role in more effective diagnosis and treatment at a more personalized level.
Key Words: genetics • genotyping • lung disease • pharmacogenetics • single-nucleotide polymorphism
We are all the same! We are all different! Any two humans are approximately 99.9% identical at the DNA sequence level, yet substantial, often medically relevant phenotypic differences exist between individuals. A significant proportion of these phenotypic differences are caused by this relatively small amount of genetic variation interacting with environmental factors. A clinically important element of phenotypic variation relates to susceptibility to disease and response to therapy. In this review, we examine the potential for using the assessment of genetic variation, by genotyping, for risk prediction and individualized therapy in pulmonary disorders. To date, the revolution in molecular genetics has had a relatively minor influence on the practice of respiratory medicine. Examples where it has influenced practice include the following: (1) genotyping for a single-nucleotide polymorphism (SNP) in the gene encoding Factor V Leiden that increases risk for serious thromboembolic disease1; (2) genotyping for mutations associated with cystic fibrosis (CF) for diagnosis and genetic counseling2; and (3) genotyping non-small cell carcinomas to predict response to chemotherapy.3 Genotyping has had a greater impact on the identification of respiratory pathogens, offering exquisite sensitivity and the potential for rapid diagnosis and determination of virulence and drug resistance4; however, this subject is beyond the scope of the present review.
The variations in the DNA sequence that cause or contribute to disease are called either mutations or polymorphisms, based solely on their frequency in the population. By convention, DNA sequence variants that occur in > 1% of the population are termed polymorphisms, and those that occur in less than one percent of individuals are called mutations. Mutations are responsible for the relatively rare single-gene Mendelian disorders (Table 1 ), while polymorphisms are associated with the more common complex genetic disorders (Table 2 ). Mutations in DNA arise naturally or unnaturally (environmental exposure). They are not always disease causing (they are far more likely to occur in noncoding DNA than coding DNA because of the far greater number of base-pairs of noncoding DNA in the human genome). Variations in inherited DNA sequence between individuals can be due to the deletion or addition of bases, or to variable lengths of repetitive sequences within or between genes. However, the most common type of DNA sequence variants are SNPs in which a single base in the sequence is replaced by a different nucleotide. There are SNPs approximately every 200 to 300 base-pairs in the human genome. Since the genome contains approximately 3 billion base-pairs, this means that there are between 10 to 15 million sites at which > 1% of the population differ from the majority. Although this seems like a large potential for diversity, simple arithmetic shows that even the most genetically diverse people are still at least 99.9% identical. If the density of SNPs was evenly spaced over the entire genome, this would mean that there are approximately 300,000 to 600,000 SNPs within the estimated 30,000 human genes. Many of these SNPs cause functional changes by affecting transcription factor binding sites, influencing splicing or stability of messenger RNA, or altering the amino acid sequence of the protein (Fig 1 ). It is this variation that, in combination with environmental factors and epigenetic modification of DNA (epigenetic changes include methylation and demethylation of regulatory sequences and/or chemical modification on the histones that influence gene expression) accounts for all of human phenotypic diversity, including disease susceptibility.
|
|
|
An important aspect of the Human Genome Project was the massive governmental and industry-sponsored effort to develop a dense set of SNP markers throughout the human genome.11 This effort was spurred on by the realization that a dense set of SNP markers could yield critical information to determine specific functional SNPs and combinations of SNPs that form the genetic basis of complex diseases.12 The SNP Consortium (http://snp.cshl.org/) and the International HapMap Project13 (http://www.hapmap.org), as well as research conducted by individual laboratories throughout the world (eg, Seattle SNPs: http://pga.mbt.washington.edu/), have generated enormous SNP-based resources to allow biologists to better investigate complex genetic diseases.
SNP Genotyping
Determination of the base sequence of DNA at a specific SNP site is called genotyping. For research discovery purposes, there are a number of high through-put technologies available to optimize the genotyping of large numbers of individuals for one SNP at a time.14 Genotyping by microarray allows the opposite approach—the simultaneous determination of multiple SNPs from an individual—and it is this strategy that promises to influence the practice of medicine. Microarrays allow the fixation of hundreds or thousands of specific oligonucleotide probes in a precise configuration or array onto a small-format solid support, such as a microscope slide,15 where they can be identified.
New technologies have recently been described that will allow the complete sequencing of an individual’s DNA. Ultimately, such an approach would eliminate the need for genotyping of hundreds or thousands of SNPs across the genome; however, the cost-effectiveness and bioinformatic challenge of this approach to clinical genomics is still unclear.
General Approaches to Gene Mapping
Two major strategies have been employed to identify the genes and the mutations/polymorphisms that contribute to the development of pulmonary diseases: linkage analysis and candidate gene association studies. Linkage analysis requires recruitment of affected families, whereas candidate genes are tested by association studies of unrelated subjects.
Linkage Analysis
Linkage analysis (sometimes referred to as positional cloning or genomic scanning) is the classical method for randomly searching the entire human genome for disease-causing genes. It usually requires affected families of at least two generations, although single-generation sibling-pairs can also be used. Each family member is genotyped for DNA markers (SNPs) that are scattered throughout the genome. Linkage analysis determines whether any of the markers are inherited with the disease more often than would be predicted by chance. The genes are identified solely on the basis of their position in the genome (thus "positional cloning"). The CF transmembrane conductance regulator gene and the mutations within this gene that are the cause of CF were the first severe disease-causing gene and mutations to be identified using positional cloning.16 An advantage of this approach is that completely novel genes can be implicated in disease pathogenesis; one is not limited to a search for disease-causing polymorphisms in candidate genes that are known to be, or suspected to be, involved in the disorder. However, once an approximate position in the genome is identified, a major challenge of this approach is the painstaking research necessary to identify the functional mutations responsible for the phenotype. This work has been made much easier by the open-source publication of the human genome sequence.
Genetic Association Studies
The second major gene hunting strategy is the candidate gene-association approach in which polymorphisms in individual genes thought to be important in disease pathogenesis are tested for their involvement in a disease. One first identifies candidate genes that are hypothesized or known to be important in the pathogenesis of a condition. Such genes might be suggested by studying the biology of the disease and/or by comparing gene expression in normal and diseased tissues (for example, by using messenger RNA microarrays). The next step is to identify polymorphisms within the gene that could affect its regulation or function. Finally, one examines whether the specific polymorphisms occur more frequently in individuals who have a disease than in an appropriate control population, or if they predict the development of disease in a cohort study. In an attempt to increase the "hit" rate for candidates, a "positional candidate" approach is an option; biologically plausible candidates that are located in regions previously implicated by linkage analysis are given precedence.
Publication and on-line access to comprehensive "directories" of SNPs in different ethnic groups as part of the HapMap (http://www.hapmap.org/) and other projects (http://pga.mbt.washington.edu/) have greatly facilitated association studies. Rather than testing all of the SNPs within a gene for association, one strategy is to select "Tag" SNPs. Since some SNPs are not independent of each other, and display an inter-SNP correlation that is called linkage disequilibrium, typing of Tag SNPs provides a reliable interrogation of additional SNPs.
One of the major advantages of association studies is that one uses knowledge of biologically plausible pathogenic mechanisms to focus the search for genes on relatively few candidates, although obviously only genes of known function can be examined. Another advantage is that the study subjects are usually unrelated individuals, so that genotypic and phenotypic data from multiple generations are not required. This is especially important in diseases such as COPD, in which the late age of onset makes it very difficult to obtain DNA and phenotypic data from parents of affected individuals. It should be pointed out that a major limitation of association studies is that a positive association may not always be due to a causative role for the polymorphism in disease pathogenesis. For example, false-positive associations can occur if a different ethnic group (with different SNP frequencies) is overrepresented in the case or control groups. Such population admixture is just one of the reasons for some apparently contradictory results of association studies.17
The results of association studies may differ between different populations due to a number of factors: variations in the frequency of SNPs in different populations; the modulating effects of other SNPs or mutations within individuals; and variation in the penetrance of the effects of an SNP due to environmental factors such as age and exposures. Testing of gene/environment interaction is critical for interpretation of genetic testing in disease. A striking example is the effect of pollutant exposure on children with asthma. Children with polymorphisms in specific genes involved in the metabolism of oxidant pollutants (glutathione transferases) are selectively affected by environmental tobacco smoke and pollution.18 Many studies of complex genetic disease have been plagued by failure to replicate reported associations. Although there are a variety of possible reasons for nonreplication, many false-positive associations are due to population heterogeneity and small sample size. Increasingly, very large sample sizes and metaanalytic approaches are being employed.19
Despite these caveats, consistent patterns are starting to emerge as more and more genetic linkage and association studies are undertaken. For example, in asthma, the most thoroughly studied, complex pulmonary disease, genome-wide linkage screens have been performed in 11 different populations and have identified 18 genomic regions that contain asthma/atopy genes, with consistently replicated regions on chromosomes 5q, 2q, 13q, 6q, and 12q.8 In studies8 of unrelated individuals, > 100 genes have been associated with allergy/asthma, and 79 of these associations have been replicated in a second study. Among these candidates, six are completely novel genes that were identified by positional cloning.
Therapeutic Implications
Despite these encouraging results, the predictive value of these SNPs for the development of asthma is yet to be tested prospectively in a general population sample. However, the exponential expansion of our information and knowledge of SNPs and their effects, coupled with advances in microarray technology, has positioned us on the brink of a very different approach to clinical medicine: the routine assessment of an individual’s SNP profile in clinical decision making. Although much work remains, the prospect of real-time clinical genotyping for selected conditions is on us!
The potential clinical benefits of genotyping are several fold: early detection of disease; predicting prognosis; selecting the most appropriate therapy; estimating risk to allow more appropriate environmental modification; predicting adverse events; and discovering novel biological mechanisms. The challenges are also numerous, and include the following: costs; issues of fatalism/invincibility; protection of privacy; education of the public and their health-care providers; and the biological uncertainty associated with the modest risks imparted by a particular genotype.
Imagine the following: a 42-year-old woman who has smoked for 23 years and is currently a 1.5 pack-per-day smoker presents to her family physician complaining of increasing cough and mild shortness of breath. She has tried to quit smoking on numerous occasions, but despite repeated, constructive counseling from her physician and trials of nicotine patches, bupropion, and acupuncture, has been unsuccessful. Lung function tests are at the lower limit of normal. An uncle died of "emphysema," and a cousin who is 10 years older had received a diagnosis of "COPD." She wants to know her risk for emphysema/COPD and whether there is any therapy to prevent it. A DNA sample is sent for the COPD susceptibility screen, and 48 h later results are available. The polymorphism profile indicates that she is indeed susceptible to rapid decline in lung function that will be exaggerated by cigarette smoking (and by environmental pollution). Her susceptibility genes suggest that her predominant risk is for emphysema rather than airway fibrosis and narrowing. The patient is informed that she is susceptible to COPD and that the test results indicate the potential for accelerated decline in lung function. This information is a powerful aid to her in achieving smoking cessation. She is prescribed a newly developed therapeutic agent that specifically inhibits matrix metalloproteinases implicated in the development of emphysema. In addition, certain other SNP genotypes indicate that another new inhaled drug for COPD, which inhibits the synthesis of matrix proteins by airway myofibroblasts, is actually contraindicated in this particular patient because her airway obstruction is related to emphysema rather than airway fibrosis.
Although this example is still futuristic, it is not fanciful. There are several practical examples in which genotyping currently aid clinical practice, and these will soon become commonplace.
We are already familiar with the thrombophilic screen undertaken for patients presenting with thromboembolic disease. This includes a search for SNPs, of which the most common (3 to 5% of the white populations) causes a substitution of a glutamine for arginine at position 506 of the Factor V gene in the clotting cascade. This genotype is referred to as Factor V Leiden, and it confers an increased risk of severe thromboembolism. Symptomatic carriers of this genotype may benefit from life-long anticoagulation. For non-small cell lung cancer, both response to therapy20 and survival following therapy3 with the epidermal growth factor receptor antagonist gefitinib, have been related to point mutations of the epidermal growth factor receptor gene. Conversely, studies21 in Japanese populations have shown that resistance to gefitinib can be predicted by genotyping. The use of these and similar genotypic markers of response to cancer therapy will increasingly impact on the selection of treatments for patients or, more correctly, patients for treatment.22
The metabolism of the antituberculous drug isoniazid, and the frequency of isoniazid-induced hepatotoxicity, are influenced by SNPs in the N-acetyltransferase 2 gene.23 Although routine genotyping for N-acetyltransferase 2 is not yet performed in hospital laboratories, the use of microarray technology will soon make it the standard of care and prevent the devastating consequences of severe isoniazid-induced hepatotoxicity. Azathioprine, an immunosuppressive drug used in the treatment of some pulmonary diseases such as pulmonary fibrosis, causes cytopenia in 10 to 15% of individuals. SNPs in the thiopurine S-methyltransferase gene, which catalyzes the inactivation of azathioprine, can predict the risk for this toxicity, and it is anticipated that introduction of this test will provide a cost-effective method of avoiding toxicity.24
Pharmacogenetics in Asthma and COPD
While there are sufficient data to recommend testing for the genotypes mentioned above, additional discovery research and validation are needed for some genetic tests before they are introduced clinically. Particularly promising are pharmacogenetic studies25 of asthma medications including the ß2-adrenergic receptor agonists, glucocorticosteroids, and leukotriene receptor antagonists. Pharmacogenetics is the study of how gene variation influences an individual’s response to drugs. Genes involved in the absorption or metabolism of a drug, or that influence its receptors or signaling pathways, can increase or decrease the effectiveness of a therapeutic agent in an individual patient. Genetics may also influence the immune response to a drug.26 Studies have shown that the effect of regular treatment with ß-agonists on lung function27 and the frequency of exacerbations28 are related to polymorphisms in the ß2-adrenergic receptor gene (Fig 1). The response to treatment with montelukast, a leukotriene receptor antagonist, has been related to polymorphisms in genes for enzymes in the arachadonic metabolism pathway.29 Similarly, a polymorphism of the CYP1A2 gene that alters metabolism of theophylline was associated with decreased clearance of theophylline in a group of Japanese patients with asthma.30 Genes that alter the effects of the corticosteroids have also been studied in relation to treatment response in asthma. Polymorphisms of the gene for corticotrophin-releasing factor receptor type 1 were associated with enhanced response to glucocorticoids in children.31
Despite these studies showing associations between therapeutic responses in asthma and genetic polymorphisms, the results are currently of limited clinical value. This is because the responses measured are variable between different studies and populations, and the effects are usually small or involve only subsets of individuals. However, the evidence is compelling enough that stratification by genotype is recommended in designing and assessing the results of clinical trials of these therapeutic agents. Further refinement of the predictive value of the specific genotypes may justify their inclusion in clinical evaluation of asthmatic patients.
Pharmacogenetics and the Genetics of Behavior
In the clinical scenario described above, we describe how SNP genotyping may provide evidence of increased risk of disease related to cigarette smoking. Although smokers continue to smoke despite overwhelming evidence of the harmful effects of cigarettes, personalizing the risk may be more effective in assisting smokers to quit. Another approach that is currently being taken by a number of investigators is to examine the genetic variation that determines the risk of addiction as well as the pharmacologic and psychological effects of smoking. This knowledge may provide targets for interrupting dependency. Examples include genes for central and peripheral receptors that modify the response to nicotine and other constituents of cigarette smoke.32
Other Disorders
There are numerous additional diseases in which the discovery of SNPs will ultimately impact on how medicine is practiced. For example, susceptibility to tuberculosis is influenced by SNPs in the Toll-like receptor 2 gene33; the severity of organ dysfunction in sepsis is related to polymorphisms in the interleukin- (IL)-6 gene34; and the risk for narcolepsy is strongly associated with a specific human leukocyte antigen (HLA) subtype.35
Conclusions
The study of the genetics of single-gene pulmonary diseases is well advanced. Although specific, highly effective therapies based on this knowledge have yet to be developed, research has shed considerable light on disease pathogenesis and is likely to substantially alter diagnosis and management in the near future. For the much more common complex genetic diseases of the lung and airways, genetic studies are at an earlier stage, but the explosion in technologic and analytic capacity that has accompanied the Human Genome Project has allowed impressive progress, and it is likely that a combination of linkage, association, and gene expression studies will completely transform our approach to the diagnosis and eventual management of these conditions over the next decade.
Rapid genotyping at the point of care, with graphical visualization-based bioinformatics tools (programs to translate the mountains of data and the myriad of interactions between the variations that make up a person’s genotype), will enable researchers and clinicians to record and demonstrate the implications to patients of individual SNP patterns with respect to protein structure, pathways, gene regulation, drug choice/side effects/efficacy, environment, and lifestyle choice (Fig 2 ).
|
Acknowledgements
We thank Jennifer Myers for her help with the figure graphics in this review.
Footnotes
Abbreviations: CF = cystic fibrosis; HLA = human leukocyte antigen; IL = interleukin; SNP = single-nucleotide polymorphism
This work was supported in part by funding from the AllerGen Networks of Centres of Excellence Canada, and from the National Sanitarium Association.
The authors have no conflicts of interest to disclose.
Received for publication September 11, 2006. Accepted for publication December 11, 2006.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
