Research Overview

Genetics and Bioinformatics Group

Bioinformatics is a new discipline, combining mathematics, computer science and biology. It evolved to model the vast amounts of data generated by undertakings such as the human genome project. Accordingly, one of the major tasks of the Division is to bridge the gap between the voluminous amounts of information in the genome databases and the questions posed by scientists at the lab bench. The Genetics group examines human populations and mouse models to find genes underlying complex genetic diseases. Autoimmune disease is of particular interest to us and we have published a significant association between a particular gene, IL-12p40, and type 1 diabetes. Using mouse models, we are narrowing the search for further diabetes genes and for genes responsible for host response to parasitic disease and cancer. Our association with the Cooperative Research Centre (CRC) for the Discovery of Genes for Common Human Diseases has produced the completed genotyping - the genetic fingerprinting, in effect - of candidate regions in a large collection of individuals with multiple sclerosis. A novel approach to finding disease genes is being applied to this population and initial results are promising. We also have an initiative investigating the biology of previously-identified disease-causing genes.

Background

Dr Simon Foote, Dr Grant Morahan and their colleagues in the Genetics team are gene hunters. Amongst their quarry are genes that predispose to insulin-dependent diabetes, multiple sclerosis and glaucoma. Other targets are genes that influence susceptibility to cancer or infection by microrganisms. The researchers study DNA from families with a history of disease, using technologies spawned from the Human Genome Project. Partnered by Professor Terry Speed's Bioformatics team, they harness powerful computer programs and advanced statistical techniques. In a complementary approach, Dr Richard Simpson's group is pioneering technologies to study protein composition of normal and diseased cells.

Technical terms are linked to a Glossary. Use your browser's "Back" button to return to the main text.

Genetics and disease

Translated, this classic biological equation states that an organism's phenotype (P) is the result of the interaction between its genes (G) and the environment (E).

For about 2500 known genetic diseases, there is a clear link between cause and effect - the 'G' part of the equation dominates, and the 'E' contribution is negligible. Such inherited diseases include haemophilia, thalassaemia, cystic fibrosis, Huntington's chorea and Duchenne's muscular dystrophy.

For other common diseases, such as insulin-dependent diabetes, multiple genes are involved and the links are more obscure. The genes concerned are often variants of normal genes and, individually, they often have no untoward effect on individuals who inherit them. Only in certain 'environments', and in the presence of variants of other genes, do these 'complex traits' cause disease.

Thus, for such diseases, genes are not destiny. Individuals may inherit a genetic predisposition to a particular disease, yet remain healthy if they avoid environmental trigger(s). Typically, the environmental triggers are still a mystery, partly because they have been difficult to differentiate from the genetic causes. Elucidating the genetic causes will lay the groundwork for understanding the environmental triggers.

While biochemistry and physiology focus on interactions within or between cells, genetics offers a more global view of the organism. 'A disease can be seen to be akin to a pyramid,' says Dr Foote. 'A small number of defects at the top can escalate and amplify to produce an extremely abnormal biology at both the cellular and whole animal level. If those small number of differences at the top of the pyramid could be understood and prevented, then the cascade to the base of the pyramid may be preventable.'

Genetics also offers prognostic or predictive capability, but only in relation to the genetic contribution to disease - obviously, it cannot predict if someone will encounter the as yet unknown environmental agent(s) that will trigger the disease. "If we can identify the genes involved in an autoimmune disorder like insulin-dependent diabetes mellitus (IDDM), we should be able to predict which individuals are at risk of getting diabetes," Dr Foote said.

Genetics of diabetes

The major feature of insulin-dependent diabetes mellitus (IDDM) is the inability to produce insulin. Insulin-producing pancreatic beta cells are destroyed by the body's own immune system, just as if they were a pathogen to be rejected. We do not know what causes this autoimmune response which leads to IDDM - why the immune system turns to the dark side - but we do know that part of the risk for developing the disease is inherited (ie genetic).

IDDM is a complex genetic disease. The contribution to the risk of diabetes is equally shared between environmental and genetic factors. Dr Grant Morahan's group is trying to identify the genetic factors responsible for the disease.

A genetic approach offers a number of advantages. No prior assumptions need to be made about the nature of contributing factors and the genes concerned may also be involved in other similar diseases. Of course, once identified, the genes will help identify those individuals at most risk of developing disease.

Dr Morahan and his group believe that 20 or so genes are involved in susceptibility to IDDM - the exact combination of genes differing with the individual. The major histocompatibility complex (MHC) of genes (called HLA in humans) may account for about half of the total genetic risk of developing IDDM. However, those responsible for the remainder of the genetic risk have yet to be identified.

Hunting for human disease susceptibility genes: The affected sib-pair method

Finding diabetes susceptibility genes involves an "affected sib" approach -researchers test siblings (brothers or sisters) who have diabetes, looking for shared alleles at genetic markers dispersed along the chromosomes.

The basic logic of the method is that if two sibs share the same disease, they tend to do so because they have inherited the same susceptibility genes. Thus, they will also tend to share genetic markers linked to these disease genes. As with many things in genetics, the concept is very simple, yet provides a base for extremely sophisticated mathematical development.

Mapping a human diabetes susceptibility gene: IDDM13

Dr Morahan and his colleagues have identified several chromosomal regions likely to contain IDDM susceptibility genes. One of these, dubbed IDDM13, is located on human chromosome 2.

As previously mentioned, the HLA complex has the biggest influence on IDDM susceptibility. Do HLA genes interact with IDDM13?

To answer this question, the researchers divided the families in the study into two groups: those in which the sib pairs are HLA-identical and those who differ. A striking difference emerged. The HLA-identical group showed no linkage to IDDM13 at all, whereas the HLA-mismatched group showed increased linkage.

These results suggest that, in the presence of two copies of the powerful HLA-linked susceptibility genes, IDDM13 is not required for disease. On the other hand, those sibs who do not have the HLA-linked susceptibility genes develop IDDM through the influence of genes such as IDDM13.

Role of IDDM13 in disease progression

If IDDM develops in multiple stages, each regulated by specific gene(s), can we determine where each susceptibility gene operates?

Some of the families analysed are enrolled in the Melbourne Pre-Diabetes Family Study, (co-ordinated by Prof. L. Harrison, Walter and Eliza Hall Institute, and Dr P. Colman, Royal Melbourne Hospital). Nondiabetic relatives are tested for circulating autoantibodies to islet cell antigens. Such antibodies are a sign of an ongoing autoimmune process and are currently the best predictor for subsequent development of IDDM. However, not all individuals who display this 'preclinical IDDM' will go on to become diabetic, suggesting that they proceed only part of the way along the IDDM pathway. Prediabetic sibs showed increased linkage to IDDM13, suggesting that IDDM13 acts at an early stage, i.e. before the autoimmune response results in total beta cell destruction.

Interaction of IDDM13 and gender

Autoimmune diseases tend to affect more females than males, but in IDDM approximately equal numbers of females and males are affected. Could IDDM13 affect one gender more than another? Again, Morahan tested this possibility simply by dividing the families into two groups: those in which mostly sisters are affected and those having a bias to affected brothers. The results were quite striking: there was significantly increased linkage in the families with a bias to affected females, suggesting some interaction of IDDM13 with gender-specific factors.

The emerging picture

By looking at families according to simple - yet biologically relevant - criteria, Morahan's team has been able to learn a lot about IDDM13. It seems that this gene tends to affect females rather than males, acts early in the IDDM process, and promotes disease in individuals lacking the influence of certain HLA-linked genes. It is now clear that there are at least two categories of IDDM patients: those in whom the early phase of disease is controlled by HLA, and those in whom it is dependent on IDDM13.

Genetics of host-parasite interactions

Infectious diseases are a special case of environment and genetics working closely together. The environmental factor is the causative micro-organism. The genetics of the host is also important, as variants in some genes can protect the host from death from certain infectious organisms and other variants can increase the risk of death.

Dr Foote and his colleagues are looking for human genes that modulate the immune response to parasitic infections - specifically, malaria, which is endemic in tropical regions of the world, and Leishmania major, which is common in Africa and Israel.

Genes Influencing susceptibility to Leishmaniasis

Dr Foote and his colleagues have been working with Dr Emanuela Handman's research group in the Infection and Immunity Division to identify host genes that confer resistance against leishmaniasis.

To model leishmaniasis, researchers infect mice with Leishmania major, the same parasite that causes leishmaniasis in humans. Like humans, some mice suffer almost no symptoms, while others develop a severe infection.

A certain strain of mouse, known as C57BL/6, is resistant to L.major, while the BALB/c strain develops full-blown leishmaniasis. The mouse T-cell response seems to be crucial - the resistant C57BL/6 mouse uses T-helper cells that are of the TH1-type, while the BALB/C mouse, which develops a severe infection, produces TH2-type T-helper cells. Why the two mice strains develop different T-cell responses is unknown. If the genetic studies identify genes that control the T-helper type switch, they should shed light on why humans also differ in their T-cell responses to Leishmania.

The researchers are crossing C57BL/6 and BALB/c mice, to produce a genetically uniform F1 hybrid strain. Another round of crosses between the F1 mice yields genetically diverse F2 offspring which segregate for parental traits, including disease resistance and/or susceptibility.

At meiosis, chromosome pairs exchange large chunks of genes, thereby creating genetically unique sperm and eggs. Using DNA markers, spaced regularly across the mouse genome, the researchers can track these chunks. Each marker has two forms, or alleles, one from each parent. These polymorphic markers identify the parental origins of all chromosomal segments for each F2 mouse.

Since the grandparents of the F2 generation were inbred (i.e. each of their two chromosomes was identical), there are only three possibilities at each chromosomal location, or locus in the F2 mice. Both alleles may have been inherited from the C57BL/6 grandparent; both from the Balb/C grandparent; or one may derive from the C57BL/6 grandparent and the other from the BALB/c grandparent. [If both alleles are identical, the locus is said to be homozygous; if they differ, the locus is said to be heterozygous.]

The researchers are using computer programs to correlate the genotype of each mouse (ie the particular set of DNA markers present) with its phenotype (susceptibility vs resistance). If a particular allele (eg. from a C57BL/6 mouse) is found to occur at high frequency in resistant F2 mice, and is absent from susceptible F2 mice, it is highly likely that the marker lies close to a resistance gene.

The frequency of association between a DNA marker and the resistant phenotype provides an indirect measure of the physical distance between the marker and the unknown gene on the chromosome.

The smaller a gene's contribution to resistance (i.e. environmental influences dominates), the more mice must be screened to detect its effect. Dr Foote and his colleagues may screen up to 500 F2 mice before they can be confident that they located a candidate gene for resistance (or susceptibility) at a specific locus.

With these techniques, the researchers have identified three loci linked to leishmaniasis resistance/susceptibility. The loci lie on chromosomes 9, 17 and the X chromosome. The genes have yet to be identified, but chromosome 17 is the headquarters of the mouse immune system, the Major Histocompatibility Complex, or MHC locus.

The next stage: Proving culpability

To prove that a candidate gene is responsible for leishmaniasis resistance, researchers will introduce it into a susceptible strain. This may be done physically - for example, the gene from resistant C57BL/6 mice would be injected into embryos of the susceptible BALB/c strain. Alternatively, the resistance gene can be tested by repeated backcrosses.

The researchers can then study the now-resistant BALB/C mice or the now-susceptible C57BL/6 mice, to see what changes have occurred in their immune-system - "If there are differences in the behavior of T-cells or macrophages, it's likely the differences trace directly to the incoming "new" DNA," Dr Foote said.

After a candidate mouse locus is identified, the researchers can search for a homologous locus in humans. This work will require working with a human population and looking for segregation of the equivalent human locus with disease.

The gene causing resistance in the mouse may not always do so in humans eg if it does not differ between humans as it does in mice. Even so, the mouse gene may be a flag for a biochemical pathway important in disease resistance - identifying the pathway can be very useful, as other genes in the pathway can then be tested as potential disease resistance genes.

Tracking down genes which promote resistance to malaria

Hunting for human genes enhancing resistance

The Genetics and Bioinformatics Group is collaborating with researchers in Papua-New Guinea and at Case Western Reserve University in the U.S., in a project to determine which host genes are protective in the East Sepik River region of Papua New Guinea, where malaria is endemic.

"We're trying to find out what genes predispose children to survive infection in infancy - it's not a trivial task, and there is no easy way to go about it," Dr Foote said.

Between 30-40% of people in the Sepik region carry a recessive gene that results in a condition called Melanesian ovalocytosis, in which the red blood cells develop an oval shape. These carriers are resistant to malaria, but in the 3-5% of embryos that inherit a double dose of the ovalocytosis gene, the condition is lethal - the embryo aborts during pregnancy.

Natural selection favors the persistence of this potentially lethal gene in areas of Melanesia where malaria is endemic. Other potentially lethal genes known to be selected in malria regions all affect red blood cells - the haemoglobin disorders thalassaemia and sickle-cell anaemia, as well as Glucose-6-Phosphatase Deficiency (G6PD).

In each case, the strong selective pressures applied by the malaria parasite means the genes occur at high frequency in local human populations. The genes came to light because of their lethality, but could there be similar, non-lethal genes that contribute to malaria resistance?

"We presume there are, so we are sampling populations in malaria areas, looking for other genes with unusually skewed frequencies," Dr Foote said.

A simple guide, he says, is that children who carry anonymous resistance genes will probably be healthier than those who don't - they are less likely to be anaemic after recurring bouts of parasitaemia.

The approach being taken by the researchers is to find chromosomal regions where there is a non-random distribution of "allele frequencies" in a population under intense pressure from malaria. The study is using polymorphic DNA markers to track variants of genes that may be involved in malaria resistance - including haemoglobin genes, HLA genes, the genes for erythropoietin (Epo) and its receptor (which regulate red cell production) and those for tumor necrosis factor (TNF) and its receptors. The frequencies of the different forms of these genes are being measured in regions of PNG that are under intense malarial attack and the results are then compared to other regions of the country where there is no malaria. Any major differences may be due to underlying malarial resistance genes.

The mouse model approach

Another approach to finding genes involved in the host response to malaria is to use mouse models. Researchers challenge inbred mouse strains with Plasmodium chabaudii, a relative of the parasites that cause human malaria. Just as in humans, where some people suffer only mildly from malaria infections, and others develop fulminating, lethal malaria, mice vary enormously in their response to mouse malaria.

The C57BL/6 mouse strain suffers only mildly and recovers rapidly after a Plasmodium chabaudii infection, but the same parasite causes fulminating infection in either C3H/He or SJL mouse strains. The difference in infection severity must be entirely due to genetic differences between the strains. Just as in the approach used for leishmania resistance, a cross was established between the resistant C57BL/6 strain and two susceptible strains, C3H/He and SJL.

Both the C3H and C57BL/6 X SJL crosses have identified a locus on chromosome 9, while the C57BL/6 X C3H cross has identified another gene on chromosome 8.

Several other loci have been identified that modulate the disease course later on during the infection. It is possible that these mouse crosses have given researchers a genetic look at genes that influence the disease at different periods of infection. For example the chromosome 8 and 9 genes probably act before activation of the specific immune response and genes on other chromosomes may modulate the specific immune response.

Finding glaucoma genes using a founder effect

Glaucoma is one of the most common eye diseases - glaucoma causes pressure to build up inside the eyeball, which if untreated, causes blindness. Certain families carry a gene - or genes - which enhance the probability that this condition will develop.

Foote and Speed are trying to identify these genes, using a novel approach which takes advantage of 'the founder effect'. Glaucoma occurs in Tasmania at the same frequency as elsewhere in Australia. However, about 60 per cent of present day Tasmanians trace their ancestry - and genetic legacy - to a small group of individuals who successfully colonised the island state in the middle of last century. Thus people with glaucoma in Tasmania are more likely to have inherited the disease gene from a common ancestor - "the founder effect" - than in most other regions of Australia, a fact that greatly helps people looking for the gene.

Tasmania is still a young population in genetic terms and records exist that allow genealogists to trace the descendants of many of the early colonists. This results in large extended family pedigrees which are like gold to geneticists. In collaboration with the Menzies Centre in Tasmania, the WEHI team is trying to identify every latter-day Tasmanian with glaucoma, and retrace their pedigrees back for six or seven generations, to identify patterns of inheritance for glaucoma. Such pedigrees have huge theoretical potential but Professor Speed and his colleagues need to develop new algorithms to use such pedigrees to identify the relevant genes.

It is already clear that the most common form of glaucoma in Tasmanians is due to a dominant gene with incomplete penetrance - only a minority of those who inherit the mutant gene develop glaucoma. But by looking for specific genetic markers that have traveled from generation to generation in company with the mystery gene, the researchers should be able to pinpoint its chromosomal location, clone it, and then determine its role in glaucoma.

Protein composition of cells (offsite - JPSL)

Glossary

algorithm - a step by step method of solving a problem

allele - alternative forms of a gene located at a particular chromosomal position (locus)

antigen - any substance capable of provoking an immune response and of reacting with the products of that response

autoimmune disorder - a disorder caused by an immune reponse against constituents of the body's own tissues

autoimmune reponse - an immune response against constituents of the body's own tissues

autoantibody - an antibody directed against a person's own tissue constituents

backcrosses - a cross between an offspring and one of its parents or an individual genetically identical to one of its parents

candidate gene - a gene suspected to play a role in a disease

cytotoxic - poisonous to cells

endemic - prevalent in a population or geographical area at all times

fulminating infection - sudden, intense infection

genealogist - a person who traces family trees

genetic disease - a hereditary disease that arises from an abnormality in genetic makeup

genetic marker - a piece of DNA of known sequence and known location on the chomosome

genome - the genetic material of a cell or organism, ie, the complete set of genes

genotype - the genetic "type" or constitution of an individual at a particular locus or across the entire genome

HLA/MHC - see major histocompatibility complex

homozygote - an individual having identical alleles at a given locus

hostile immune response - a reaction by the body's natural defence system in which one's own body is attacked.

locus - a place or position occupied by a gene on a chromosome.

macrophages - large cells that can engulf other cells or other microorganisms.

major histocompatibility complex - major antigens on cells which are recognised during graft rejection, known as HLA (human) or MHC (mouse) 

mapping/gene mapping - locating the relative position of genes on chromosomes.

meiosis - cell division of the sex cells of in which the chromosome number is halved.

MHC - see major histocompatibility complex

paradigm - an example serving as a model or pattern.

parasitaemia - the presence of parasites in the blood

pathogen - a micro-organism which can cause disease.

phenotype - the observable characteristics of an organism, for example, eye colour

polymorphic - occurring in several or many forms

predisposition - a concealed susceptibility to disease which may be activated under certain conditions

prognostic - a symptom or sign on which a diagnosis may be based

recessive gene - a gene whose phenotypic effect is apparent only when homozygous

T-cell response - a form of immune response performed by specialised blood cells called T lymphocytes. There are two different types of T-cells: helper T cells and cytotoxic T cells. Helper T cells fall into two categories: TH1 and TH2

variants - one of several forms of a gene