Dunstone Lab research
About Associate Professor Michelle Dunstone
X-ray vision reveals protein gateway to cell life and death.
Understanding the life and death of cells is crucial information for scientists around the world working to treat illness and disease. Through the process of X-ray crystallography, Associate Professor Michelle Dunstone has been able to make an important contribution to this field of science. She has been able to capture images that allow scientists to visualise the behaviour of "hole punching" proteins, called pore forming toxins. These toxins are used by our immune system to combat disease and also used by bacteria to cause disease.
As a molecular biologist, Michelle has been studying the structure of several toxic proteins that create holes in the cell surface, resulting in cell death. She has captured world acclaim by imaging how these proteins work through X-ray crystallography.
The success of this technique relies largely on the skill of the researcher in preparing samples - purifying and crystallising proteins. This is an important prerequisite to using advanced imaging technology to visualise the protein molecules and to identify their biological purpose.
This understanding will allow future researchers to consider how these proteins can be used or manipulated to achieve desired outcomes, such as promoting the growth of beneficial cells or killing disease.
Michelle says the proteins she studies create pores in the cell membrane and are used by organisms in all kingdoms of life, from bacteria to humans. They come in two varieties. There are life-givers in humans, allowing the immune system to kill disease, and there are life-takers used by disease-causing bacteria.
Her research targets a class of proteins that behaves like toxins, called MACPF toxin. They are used by the human immune system to kill invading bacteria, for example. But bacteria also use them as part of their infection strategy.
Michelle has been able to crystallise and image the free-floating form of several MACPF toxin specimens. It turns out these proteins used by the immune system are similar in their three-dimensional structure to those used by bacteria, which indicates they may all use the same mechanism to punch holes. By studying these structures we can come to understand how they do it.
'When the proteins come across a target cell membrane with the right composition then these toxins change shape and bind to the membrane,' Michelle says. 'As more individual proteins do this, they form a doughnut-shaped ring in the membrane, which triggers the formation of a pore that punctures the membrane.'
When she is not at the synchrotron bombarding her crystallised samples with high-energy beams, Michelle has another passion. Fascinated by the antiquity of the proteins she images, she trawls the world's genome sequence database, trying to understand the evolution of the hole-punching proteins.
'These proteins are present in everything from bacteria all the way through to humans. It seems that an ancestor may have existed in the organism that gave rise to both modern bacteria and mammals,' she says. 'That places the origin of this molecule deep in the past, before there was divergence among bacteria. So it was present in living cells very early in evolution.'
She calls this activity 'genome gazing' along the evolutionary tree of life. She keeps finding new members of the toxin family in all sorts of different organisms. In the process, she has started to see her proteins in a different light.
'It is not always obvious that the organism needs the protein for a pore-forming activity,' she says. 'That implies they may have some other role. What we would postulate is that a common activity across the family is that they can recognise lipid membranes - see them chemically - and insert into the recognised subtype. But we need to identify more family members before we can assign a universal role.'
Associate Professor Dunstone is currently an ARC Future Fellow and group leader in Monash University's Department of Biochemistry & Molecular Biology. Her passion lies in discovering the conformational change that occurs in the Membrane Attack Complex / Perforin-like superfamily of pore forming proteins.
Associate Professor Dunstone was initially trained in crystallography in the laboratory of Professor Michael Parker at St. Vincent's Institute, Australia (NHMRC Dora Lush scholarship). Her original PhD project focused on the structure of C9 from the Membrane Attack Complex. Michelle continued researching structural immunology (Prof Rossjohn lab) and the wider MAC/perforin family of pore forming proteins (NHMRC ECF fellowship) at Monash University.
In 2007 her bioinformatic and crystallography research culminated in the first representative structure of a MAC/perforin-like pore forming proteins. Her research on the fungal toxin, pleurotolysin, explores how these hole-punching proteins work by combining single particle cryo-electron microscopy, biophysics and computational biology. Her most recent research on the Membrane Attack Complex structure not only shows how the immune system uses hole punching proteins to kill bacteria but gives insight in to how these proteins evolve for different functions. (NHMRC CDA fellowship).
Michelle continues to research the real-time assembly of pore forming proteins as part of her new ARC Future Fellowship.
- Structural and immunological studies of the Membrane Attack Complex
- MPEG: Understanding how macrophage use pore forming toxins
- Developing a fungal pore forming toxins as model systems and as pest control agents
Visit Associate Professor Dunstone's Monash research profile to see a full listing of current projects.
1. The human immune pore forming proteins; the immune Membrane Attack Complex (MAC)
What is the MAC?
The Membrane Attack Complex (MAC) is an immune based hole-puncher evolved to assemble and punch into any cell membrane. Typically, this is bacteria, protozoa and helminths. This part of the immune system is located in the plasma and interstitial fluid. Some aspects of the complement pathway are also found to work in the brain.
The MAC can assemble on surfaces where there is C5 convertase. There are no identified binding domains or factors and in vitro studies show assembly without the required of specific membrane binding regions. This is a very different system to that characterised for CDCs and pleurotolysin (fungal toxin).
How does the MAC punch?
There is sequential assembly of C5, C6, C7, C8 and C9. Our recent work has shown how the domains of these proteins can overcome the requirement of dedicated membrane binding domains (Dudkina) and that there is a loop in C9 that acts as a brake to control the exact timing of C9 inserting into the membrane. Together these data show how the MAC has evolved a sequential insertion mechanism in order to be a global targeting pore forming protein.
Why doesn't the MAC kill our own cells?
One trick to controlling the MAC is that host cells can express MAC inhibitors, such as CD59, on their surface thereby stalling MAC formation and escaping cell death. People who lack CD59 may develop paroxysmal nocturnal haemoglobinuria (PNH), a disorder characterised by MAC mediated haemolysis and anemia. Some pathogens have also evolved MAC inhibitors (or scavenge these from their hosts) to avoid death by the MAC.
Future projects tackling inflammatory disease and cancer
Why do cancerous cells make more of the inhibitor, CD59? Is the MAC important for cancer eradication? What if it were possible to engineer the MAC to specifically target cancerous cells? We are now looking at how inhibitors of the MAC and if proteins can be engineered to target specific membranes. Can we develop MAC inhibitors to stop unwanted MAC formation in inflammation?
The video below shows how the immune system uses the Membrane Attack Complex (MAC) to attack a bacteria. My latest research in Nature Communications (Dudkina, Nat Comm, 2016) unravels how the MAC assembles in a specific way in order to target bacteria with variable membrane surfaces.
This scientific animation is about a "hole-punching" machine, known as the Membrane Attack Complex (MAC),
which our immune system uses to punch and destroy bacteria entering the blood vessels.
2. MacroPhage Expressed Gene-1 (MPEG1)
MPEG1 is an immune effector
Macrophage Expressed Gene-1 (MPEG1/perforin-2) is a human immune effector that belongs to the MACPF/CDC family of pore forming proteins. Being a close homolog of perforin-1 and components of the Membrane Attack Complex (C6, C7, C8 and C9) suggested that MPEG1 may have a pore forming function in immunity.
Originally MPEG1 was identified in macrophages, hence its namesake, however it is now know to be expressed in many cell types. Intriguingly, MPEG1 expression and subcellular localisation has been shown to be modulated by LPS, TNFα (molecules present during immune response) further implicating MPEG1 in immunity. Currently its structure and function are yet to be fully understood.
MPEG1 targets intracellular bacteria
MPEG1 is known to be an intra-cellular, membrane bound factor meaning that unlike most MACPF/CDC proteins, it is membrane tethered.
Most of our understanding of MPEG1 function has been generated by in vitro cell assays and in vivo in mice. MPEG1-/- cell lines show significantly reduced ability to clear intracellular bacterial infections. Furthermore, it has been demonstrated that MPEG1 localises to the late-endosome and forms pore-like structures on E. coli derived membranes.
MPEG1 deviates from the common PFP mechanism
MACPF/CDC are thought to undergo pore formation via an archetypal three-step mechanism. Initially (1) monomers recognise and bind to their target membrane and (2) subsequently assemble into large oligomeric pre-pores. The pre-pores then transition into pores, by undergoing substantial conformational rearrangement (3) forming giant amphipathic β-barrels that span the membrane. MACPFs are known to play integral roles in innate and adaptive immunity in vertebrates, as well as developmental roles. MPEG1 however appears to deviate from this mechanism.
To read more, check out our recent publication The structure of Macrophage Expressed Gene-1, a phagolysosome immune effector that is activated upon acidification in collaboration with the Whisstock lab.
3. Fungal protein: Pleurotolysin
What are Pore Forming Proteins?
MACPF proteins belong to a class of molecules known as pore forming proteins (PFPs). These proteins target membrane bilayers and upon oligomerisation perforate the membrane by inserting amphipathic regions into the bilayer (left). PFPs exist in every domain of life and hence have diverse targets and function. In humans PFPs aide in the immune response by targeting and killing various cells. PFPs are also involved in development, however this is remains to be fully understood. Conversely, PFPs can also act as toxins to help in bacterial virulence or for digestion of prey. Industrial applications of PFPs include next generation sequencing (DNA and protein), as well as genetically modified crops for pesticide-free pest resistance.
I am fascinated with how pore forming proteins recognise their target cell, come together and then punch a hole into the cell. Here is a video of how the MACPF/CDC pore forming toxins work.
Model of how the MACPF protein domain is able to change shape. The red and yellow loops represent
the region that can insert into the cell membrane thereby forming holes in the cell.
These pore forming proteins are used by the animal immune system to kill pathogenic bacteria, virally infected cells and cancerous cells (Law, Nature, 2010). And bacteria use the same tools to kill our cells (Reboul, PLoS Comp Biology, 2014). They are also used by fungi to defend themselves against bacteria and nematodes (Lukoyanova, PLoS Biology, 2015) and animal venoms (Ellisdon, PNAS, 2015). In fact there are a lot of unknown functions of pore forming proteins.
The fungal protein, Pleurotolysin
Pleurotolysin is a PFP found in oyster mushrooms. The structures of the monomeric forms of the pleurotolysin system (PlyA, PlyB) in the soluble precursor state have been solved previously. Pleurotolysin membrane recognition is achieved via the PlyA component, while a separate PlyB component binds and docks to a PlyA dimer to form a functional subunit of the final oligomer. This enables one to decouple the system and study membrane recognition separately from membrane insertion. We have also generated cysteine trapped mutants that effectively lock the protein and stall the process of pore formation at various intermediate states. Lastly the material is very well behaved: soluble at high concentrations, stable at 4-20°C for long durations, low degree of aggregation and easy to produce in high quantities.
A model system for studying the mechanism of pore formation
While several structures of MACPF proteins are available in at various states of the general pathway, we lack a complete understanding of the mechanism. Notably, no intermediates have been captured during membrane insertion and most of our understanding is based on indirect measurements. A detailed model and understanding of the mechanism and conformational changes of pore formation are of great value. This would guide development of novel PFPs with specific properties for industry applications and help to design and develop compounds to modulate PFP activity to reduce the burden of some human diseases.
My lab uses X-ray crystallography, cryo-electron microscopy, computational biology and biophysics to answer these questions from the atomic level through to the cellular level.
We collaborate with many scientists and research organisations around the world. Some of our more significant national and international collaborators are listed below. Click on the map to see the details for each of these collaborators (dive into specific publications and outputs by clicking on the dots).
Dr Peter Boag, Monash BDI
Student research projects
The Dunstone Lab offers a variety of Honours, Masters and PhD projects for students interested in joining our group. There are also a number of short term research opportunities available.
Please visit Supervisor Connect to explore the projects currently available in our Lab.