The use of genetic markers to predict disease susceptibility.

Can genetic markers be used to predict susceptibility to disease and disease phenotypes?
It is rightly said that people’s curiosity drives their passion for knowledge. Acquiring knowledge becomes an adventure when there are codes that can reveal secrets. Ever since we were introduced to the reality of DNA and genes, we have been trying to decode the information that holds the secrets of our existence. Our curiosity has driven our search, which has led to great rewards; the greatest reward so far has been the success of Human Genome Project. This has revealed the sequence of the entire human genome. We now know that the human genome contains 30,000 – 35,000 genes. This information has been widely used for the study of various disorders that affect the human race. An interesting phenomenon that was noticed was that of the ‘genetic markers’. These refer to the association of certain DNA sequences which have normal and disease genes. The genetic markers are usually characterised by repeats in the sequence of microsatellites, macrosatellites, SNPs and STRs. Genetic markers may be in the mutant genes or may be some base pairs apart. Some markers even flank the mutant gene. These markers are usually highly polymorphic, i.e. most of the population is heterozygous for these markers. Hence, genetic markers are interesting tools for inheritance studies.

‘Prevention is better than cure’!
Our basic medical principle that ‘prevention is better than cure’ always guides us towards predictive diagnoses. And there is no reason not to use genetic markers for predictive diagnosis. Predictive diagnosis is carried out to evaluate a person’s susceptibility to disease. Usually, people are screened for diseases with a potential genetic influence such as Huntington’s disease, phenylketonuria, cancer, hypertension, neurological disorders and diabetes. Knowing an individual’s genetic predisposition to a disorder will help to avert or lessen the phenotype of the disorder. A physician can suggest behavioural changes with respect to diet, exercise etc. Certain drugs could be administered beforehand to avert the disease (Hall et al., 2004). Screening can be carried out on a regular basis to detect the early stages of diseases like breast cancer.

An aspect of treatment that is now considered to be important is ‘patient specificity’ whereby drug response is influenced by the patient’s genes. People with similar genetic makeups have similar responses. An example that would best illustrate this genetic individuality would be a case referred to in a review article by Guttmacher et al. (2002) – a four year old boy, John, who has acute lymphoblastic leukemia, received mercaptopurine drug therapy. However, he was then diagnosed as having an homozygous mutation for gene encoding S-ethyl transferase that inactivates mercaptopurine. His treatment was altered accordingly and the test is now used to screen patients who have been prescribed the mercaptopurine drug. Such flexibility in treatment could never have been achieved by the use of conventional methods of autoantibodies and clinical chemistry.

All prenatal or newborn screening is based on genetics. The use of genetic markers for the understanding of the genetic basis of complex human diseases has opened doors to such screening methods. A new-born can be screened for a set of disease markers, like CF, phenylketonuria, galactosemia, homocystinuria etc. (Khoury M. J. et al.,2003).

How are genetic markers used for disease-specific diagnosis?
Functional disorders caused by genetic mutations can be broadly classified as monogenic or polygenic. Monogenic diseases – involving one gene, are rare, and have severe phenotypic abnormalities. They could be further categorised as autosomal recessive, autosomal dominant, and sex-linked diseases; examples are: cystic fibrosis, Tay-Sachs disease, and β- thalassemia. On the other hand, polygenic disorders involve many genes; they are common disorders and are influenced by environmental factors. Examples of these diseases are: hypertension, Type1 diabetes, rheumatoid arthritis and cancer. The use of genetic markers for predictive diagnosis is easier in the case of monogenic disorders since there is just one gene involved.

Monogenic diseases:
Cystic fibrosis (CF) is one of the most common autosomal recessive disorders and has been extensively studied. The disorder is characterised by mucus formation in the epithelial cells and results in damage to many organs. The gene called ‘cystic fibrosis transmembrane conductance regulator’ (CFTR) is mapped on chromosome 7. It encodes a transmembrane protein, present on apical epithelial cells, which have 1480 amino acid residues. This gene strongs over 250kb and is characterised by 27 exons. The transmembrane protein is involved in chloride ion transport. The functional regulation of the transmembrane involves camp. In the case of CF, there is an over-expression of the proteins on the epithelial cells of sweat glands, the pancreatic ducts, the lungs, respiratory cells, billary ducts, salivary glands, the digestive tract and, mostly, in the serous cells of submucosal glands (Davidson et al.,1998).

About 600 mutations have been associated with CF. (Davidson et al.,1998). Various markers have been identified to be associated with the normal and mutant genes. The most common markers are IVS6aGATT, IVS8CA, IVS17BCA, IVS17BTA (microsatellites) and TUB20. (Korytina et al., 2003; Morral et al., 1996).

Mutations DF508, G542X and NI303K showed maximum slipping events at these microsatellite markers, which is evidence that they are among the oldest and most common mutations (Morral et al., 1996). The map distance between the marker and the mutant gene, in this case, determines the extent of the association of the two. Nowadays, newborn screening for CF is carried out for 170+ mutations. Markers for delF508 (DF508/F∆508) are usually found in more than 68% of CF cases, suggesting a high penetration of DF508 mutation (Davidson et al., 1998). Other common mutations are G480C and R117H. The severity of mutation is based on the sufficiency of the conferring pancreatic enzyme. DF508 is severe and is characterised by meconium-ileus obstruction in the fetal intestine. Lung disorders are associated with CF independent of the mutation that causes CF. All mutations lead to a typical phenotype. The genetic markers, which are associated with the mutation, give us a predictive picture of the phenotype that we should expect and help in improving pharmacological interventions for treatment. For example, DF508 is associated with cAMP regulated chlorine ions. So, on diagnosis of a predisposition to CF caused by DF508 mutation, cAMP analogs can be administered in high doses to elicit chloride ion conductance. Many other polymorphisms/markers have been found to be associated with musculoskeletal changes in CF due to DF508 mutations (Norek et al., 2007). Hence, genetic markers also help us to predict the specific character of the broad-spectrum phenotype of a disease and help in developing precise treatment.

A different aspect of predictive diagnosis can be seen in Huntington’s disease (HD). This is an autosomal dominant disease that causes neuropsychiatric disorder. Its onset is usually at about the age of 40. Patients experience mood swings, loss of memory and involuntary movements of the limbs and trunk. In 1983, a DNA marker closely associated with Huntington’s disease was discovered by Gusella et al. (Benjamin et al., 1994).HD is the only disorder in which a direct relationship between a simple genetic marker and a disease gene has been found and this is used for screening purposes. The gene for HD is mapped on chromosome 4 and it is a trinucleotide repeat disorder. The normal gene contains a trinucleotide CAG, but in case of HD, there are several repeats of the trinucleotide that expands in the HD gene at the 5’ end (Benjamin et al., 1994). These trinucleotide repeats act as strong genetic markers. The number of repeats gives a quantitative evaluation of genetic predisposition. The length of CAG repeats is directly proportional to susceptibility. If CAG repeats are less than 29, the person is normal. If the repeats are between 29 to 35, CAG are called intermediate allele and when there are repeats of more than 36, there is strong evidence of susceptibility (Almqvist et al., 1997). When the risk of developing the disease has been detected in patients, a multi-disciplinary approach of treatment is recommended. Patients are encouraged to partake in health-support therapies and diet is supplemented with neurotonics. Both physical and mental exercise and/or speech therapy is recommended. Huntington’s disease does not skip generations, so it is recommended that all high-risk parents undergo prenatal diagnosis during for every expected birth.

Polygenic disease:
It is difficult to identify a one-to-one relationship between a genotype and phenotype in polygenic disorders, since there are many genes involved. But the main hindrance is caused by environmental influences on the phenotype expression. In the case of a monogenic disorder, the genetic make up can be qualitatively assessed because mutation in the candidate gene will give a disease phenotype. But this is not the case with polygenic disorders. A quantitative genetic analysis is required since various genes may act in combination to give the disease phenotype. The same genes may not be responsible for causing the same disease in members of a single family (Weissman 1995). Genetic markers for polygenic diseases can be used to predict certain associated aspects of the disease phenotype such as the secondary spread of cancer or the severity of the phenotype.

An example to explain the multiple-approach strategy for polygenic diseases would be breast cancer. Breast cancer is the most common cancer in women and has been studied extensively with respect to genetics. As more and more people are affected, attention has been directed towards treating high-risk patients before the cancer develops. Breast cancer risk is evaluated by the use of many approaches, the major one being by detecting the expression of the frequency of oestrogen receptors or progesterone receptors. Tamoxifem is an oestrogen receptor modulator and has proved to be very successful in preventing the development of breast cancer in susceptible individuals. Major genes involved in ER positive patients are BRCA1 and BRCA2. These genes can be screened to evaluate genetic predisposition and, accordingly, the patient can be prescribed drugs, given a good diet and screening to detect cancer at an early stage. Mastectomy or oophorectomy has been shown be successful in averting breast cancer in some high-risk patients. As cancer is usually shows secondary spread or metastasis, a predictive analysis of metastasis susceptibility will help in improving the treatment. Research in identifying genetic markers for metastasis genes are in progress. Crawford et al. (2006), using the mouse model, studied the metastasis efficiency modifier locus linked to the proximal end of chromosome 9. They identified a gene called Sipa1 (signal induced proliferation associated gene1) or Spa1 in mice, which they thought to be the candidate modifier gene. It was found that mutation in this gene was responsible for altered protein-protein interaction. It has been hypothesised that this gene is related to the human SIPA1 gene on chromosome 11q13.3 that has a positive correlation with breast cancer metastasis. Three markers SNPs are identified to be associated with SIPA1 – 313G>A in the promoter region and two other SNPs, 545C>T and 2760G>A in exon 1 and exon 12, respectively. The geontypes associated with the presence of a positive lymph node are 313G>A and 2760G>A. ER negative tumor cells are associated with 545cT>. A greater understanding of these markers will help to reduce metastasis, which is the major fatality factor of cancer.

Genetic markers have proved very useful in understanding immunological auto-immune disorders. The best example is that of rheumatoid arthritis (RA). It has a multifactorial origin, and, hence, environmental factors have the most influence. It has also been found that the concordance for RA is higher in monozygotic twins compared with dizygotic twins. Hence, the genetic factor also is key in the regulation of this disorder. Most of the auto-immune disorders are considered to be associated with the HLA genes mapped in the chromosomal region 6p21 that is the locus of the TNFA- tumor necrosis factor alpha (Miterski et al., 2004). We can consider this to be true for RA since there is an increased TNFA concentration found in the serum and synovial fluids of RA patients. Another form of rheumatoid arthritis is juvenile rheumatoid arthritis (JRA). Genetic markers have been useful in understanding the different genetic components associated with these two disorders. Even though the basic regulatory aspect is the same, different genetic combinations are associated with each disorder. Association of HLA DRB1 *04 with RA, was first discovered by Stastny et al. (1978) and is found to be valid in most of the cases (Miterski et al., 2004) and a weak association of HLA DRB1* 08 and 11 with JRA has been noticed. Polymorphisms in the TNFA genes vary for both disorders. Other genes: IKBL, CTLA4, PTPCR, MIF and IFNG also show different genetic marker distribution for the two disorders (Miterski et al., 2004). Hence, a different application of genetic markers can be seen in this case, which increases our knowledge of the immunological aspect of the diseases and which can be used for treatment purpose.

Genetic markers: Hope for future?
Genetic markers have proved to be efficient tools to evaluate the genetic aspects of diseases. But the question that needs to be asked is about how far we have come to know these tools. Can we be certain of our findings regarding the association of genes with diseases? Can predictive analysis based on genetic markers be interpreted efficiently by physicians and can physicians be sure about what treatment to prescribe? I think not yet. Yes, we can use genetic markers as tools, but we are not ready to use them as standard diagnostic tools yet. More profound knowledge is required. But, we cannot deny the fact that development in such predictive analysis will be an evolution in medicine. The success of the Human Genome Project is proof that we can, in the near future, be more confident. There is large scope for development. By combining microarray techniques, automation of such screening will be possible. Making individual genetic records for disease susceptibility, drug sensitivity, immunological status, and the like will be possible in future.