Structural biology

Microglial receptors
Genome-wide association studies of a variety of neurodegenerative diseases indicate that innate immunity in the brain plays a major role. Microglial cells, the resident immune cells of the central nervous system, act as the first and main form of active immune defence in the brain. Upon detection of pathogens or damage, microglia adopt an activated state resulting in an inflammatory response which leads to the release of a host of neuroactive signalling molecules. However, in neurodegenerative diseases these cells fail to remove the large excess of neurotoxic proteins that accumulate in the brain.

Receptors expressed on the surface of microglia play a role in sensing environmental changes and regulating their activation. We are exploring the role of various microglial receptors in normal physiology and in neurodegenerative diseases. As an example, we have been studying the microglial receptor CD33 which acts as a “handbrake” on microglial activity. We have determined its crystal structure and visualised how a carbohydrate ligand called P22 binds to it. We have shown that P22 enhances the uptake of the Alzheimer’s toxin, Abeta, into cells for its destruction and clearance from the brain. Hence inhibitors of CD33 function have potential therapeutic benefit to treat Alzheimer’s and possibly other neurodegenerative diseases.

Amyloid Precursor Protein (with Prof Colin Masters, Florey Institute)
Alzheimer's disease is the most prevalent neurodegenerative disease in humans and is the fourth leading cause of death in the developed world. The disease is characterised by the presence of amyloid plaques that principally derive from amyloid precursor protein (APP). The long-term aim of this project is to determine the complete structure of APP in order to understand its normal physiological function and as a basis for structure-based drug design of anti-Alzheimer's drugs. We are solving the structure of APP by a divide-and-conquer approach: solving the structures of small, overlapping fragments with the eventual aim of piecing the molecule together. To date we have determined the structures of three regions of APP: the growth factor domain, the copper-binding domain and the E2 stalk domain.

Antibodies and diabodies
A widely advocated clinical strategy to treat Alzheimer’s disease is to use antibodies to remove neurotoxins, Abeta and hyper-phosphorylated Tau (pTau), that are thought to contribute to the disease. Our lab has visualised, by crystallography, how three such antibodies recognise the Abeta peptide, including the clinical antibodies Bapineuzumab and Solaneuzmab. We have developed our own anti-Abeta antibody by using one of our crystal structures to engineer in desirable features for a more potent therapeutic.

In a different approach we have developed bispecific diabodies. Diabodies are a small format antibody lacking Fc domains (which stimulate microglia) and constant domains. They are comprised only of a fusion of the antigen binding domains of two different antibodies. Our diabodies diabody consist of one arm with specificity for a pathogen (Abeta or pTau in Alzheimer’s and α-synuclein in Parkinson’s) and the other arm which recognises a microglial surface protein that drags the pathogen into microglia for degradation.

The diabody approach promises significant benefits over traditional antibody formats: the absence of the Fc domains might avoid the neuroinflammatory side-effects that have plagued AD antibody clinical trials and their smaller size might make them more amenable to cross the blood-brain barrier. Therefore, diabodies offer a potentially safer, targeted route to clear components of plaques, tangles and other pathologies associated with AD, Parkinson’s and other neurodegenerative diseases.

IRAP (with Dr Siew Yeen Chai, Monash University)
Central administration of the hexapeptide angiotensin IV markedly enhances memory and learning in rodents. This effect is mediated by binding to a specific, high-affinity site in the brain which our collaborators identified to be the transmembrane enzyme, insulin-regulated aminopeptidase (IRAP). The peptide binds with high affinity to the catalytic domain of IRAP inhibiting its enzymatic activity. Using a molecular model of the catalytic domain, we have screened compound databases and have identified compounds that inhibit IRAP and reverse memory deficits in animals. To date, there is no proven effective treatment for cognitive impairment. Since the causes of cognitive impairment range from birth defects (Down's syndrome, mental retardation, cerebral palsy), recreational drug abuse, opportunistic infections to neurological conditions (stroke, brain trauma, neurodegeneration such as Alzheimer's disease), prevention therapies are not effective alternatives. The IRAP inhibitors may lead to the development of new classes of cognitive enhancers.

The work undertaken by SVI scientists has provided major contributions to identification of targets for cancer therapy, and understanding the mechanisms of cancer growth and spread. This work was recognised in 2012 by a generous grant from the Australian Cancer Research Foundation to help establish the ACRF Rational Drug Discovery Centre at the Institute. Examples of the cancer projects currently being pursued by the Centre include:

GM-CSF and IL-3 receptors (with Profs Angel Lopez and Tim Hughes, Centre for Cancer Biology, Adelaide)
Cytokine receptors are transmembrane cell surface glycoproteins that bind specifically to cytokines and transduce their signals that can direct cells to proliferate, differentiate or even die. The GM-CSF, IL-3 and IL-5 family of cytokines regulates the survival, proliferation, differentiation and functional activation of hematopoietic cells. These same cytokines have also been implicated in multiple pathologies resulting from the excessive or aberrant expression of the cytokine or their receptors, in conditions such as certain types of leukaemia. We determined the structure of a GM-CSF:receptor ternary complex, representing the first structure of an "activated" receptor of this family of cytokines. Inspection of the structure revealed exciting insights into the mechanism of receptor activation and provided a unifying molecular explanation for the diverse functional properties of related cytokine:receptor systems. To maximise the drug development opportunities of this discovery we have formed a partnership with the biopharmaceutical company CSL Limited to discover and develop therapeutic antibodies that will disrupt aberrant signalling by the receptor. We have also determined the structure of the related IL-3 receptor bound to an antibody discovered by Angel Lopez’s lab which is currently under clinical development for leukaemia treatment by CSL. The structure reveals how the antibody inhibits IL-3 signalling.

Glutathione transferases (with Prof Giorgio Ricci, University of Rome, Italy; Prof Paul Dyson, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland; Prof Philip Board, Australian National University, Canberra)
Glutathione S-transferases (GSTs) are a supergene family of enzymes that play a pivotal role in the detoxification of foreign chemicals and toxic metabolites. Paradoxically, the detoxifying activity of these enzymes is in part responsible for the development of cellular multi-drug resistance towards a number of chemotherapeutic agents. We have determined more than 50 GST crystal structures which have contributed to an understanding of the molecular basis of substrate recognition and catalysis by the enzyme superfamily. Inhibitors of GSTs have been used in clinical trials as adjuvants in cancer chemotherapy with some success. However, the use of these inhibitors has produced undesirable side effects and more suitable inhibitors are urgently being sought.

We have made major contributions to our understanding of how bacterial pore-forming toxins can pass through the walls of target cells. In recent years this work has expanded into studies of other infectious organisms such as parasites and viruses. Some examples of our work include:

Protein toxins (with Prof Rod Tweten, University of Oklahoma, USA; A/Prof Laurie Comstock, Harvard Medical School, USA)
Pore-forming toxins are promising model systems for understanding the biogenesis, structure and function of membrane channels as well as being potential targets for new antibiotics. One example of our work is perfringolysin O (PFO), a 52 kDa toxin secreted by the gas gangrene bacterium Clostridium perfringens, a member of a family of more than 20 toxins produced by Gram-positive bacteria. This family is often referred to as cholesterol-dependent cytolysins (CDCs) because of their strict requirement for membrane-bound cholesterol for activity. The presumed common mode of action of these toxins involves binding to target cell membranes via cholesterol, insertion into the lipid bilayer of target cells followed by oligomerisation and pore formation leading to cytolysis. However, the details of the molecular mechanism of membrane damage are not known. We have now determined the crystal structures of a number of CDCs. Current work revolves around understanding how CDC toxins penetrate membranes using X-ray crystallography, cryo electron microscopy and computational biology.