By Lynley Hargreaves 13/09/2017


Forget petri dishes: a team from the University of Auckland is using a Royal Society Te Apārangi Marsden Fund grant to organise human neural cells into grids. The group then stimulate the cells with electrodes, to better understand real communication in the brain and to mimic the effects of common neurological conditions such as epilepsy and stroke. Associate Professor Charles Unsworth tells us how it is possible to accomplish such feats.

It’s mind boggling that you can put human brain cells on to a silicon chip. How does this work and how do you keep them alive?

Associate Professor Charles Unsworth

It is just like feeding your plants at home. Cells need food and warmth. For them, their food consists of a fluid media containing nutrients which has to be maintained at an ambient temperature using an incubator. Every other day, we feed the cells by changing the old media for new. Our cells typically live for about a month and beyond. More technically, we are actually looking at an immortalised cancer stem cell line. You can take the stem cells, knock out the cancer, then differentiate the stem cells into other types of cells, in our case, brain cells to use on the chip.

We use biomaterials to coerce cells to migrate to regions on a chip in order to pattern them into grid networks. We then use lasers to perform microsurgery on the networks to prune the network to become as regular as possible. We have found that the neural cells migrate to intersections of the grids where we put small electrodes underneath to stimulate cells to fire and record the cells’ communication.

Is communication in such a network similar to a brain?

What we are developing is not a brain, by any means, but it is a deconstruction of one. That is how basic science and in vitro research works – you do a lot of deconstruction, but from it you can reveal very powerful insights into the mechanisms and processes that make things behave in the way they do, in our case these are the fundamental workings of how networks of brain cells communicate. In the brain, neural cells grow in a complex interwoven fashion. The Marsden Grant we were awarded is essentially about untangling brain cells and re-modelling them so we can understand behaviour in networks of such cells. The brain has a hundred million neurons or so but science is still in the dark about how brain cells communicate on a network level because the cells grow irregularly over the top of each other. Scientists still do not know how signals propagate from a single real neuron to the simplest of networks – world leaders have currently only managed to report the communication of up to a 16-cell organised network. Even when you grow cells in culture you still have millions of cells in a small petri dish all growing over the top of each other.

The novel thing that we are doing is the organisation of these cells in order to understand the flow of communication more easily. We have currently managed to organise about 32 neurons on a eight millimetre square chip into regular arrays so they are all connected into a grid structure, with cells at the nodes of the grids so you can record how they communicate. We aim to go higher. In addition to this we have organised 100 astrocytes into grid arrays too (to create the first astrocytic grid network). These star-shaped cells are quite an interesting cell type because they were originally thought to be just a supportive cell to neurons, but recent science has shown that actually they have much more functionality as they modulate the communication between neurons.

How do you mimic diseases such as epilepsy or stroke?

With regards to epilepsy, it is described as an abnormal synchrony of neural activity. With our technology we can use our electrodes to stimulate a neuron to fire. What is very interesting here is when you enable lots of neurons to fire you can induce the network to become synchronous, as observed in epilepsy, on that chip. Our technology enables us to specifically control where and when neurons fire in a network, and record how this communication propagates through the network allowing us to understand how epileptic circuits behave better.

In the brain, stroke occurs when there is a rapid decrease in oxygen. This can happen globally throughout the brain as a consequence of a heart attack or locally in regions of the brain due to a head injury or poor circulation in old age. It is possible to simulate a stroke event on a chip by gradually creating a more oxygen-less (hypoxic) environment causing them to die. Very little is understood about the signaling mechanisms that go on during stroke at the network level and what causes some cells to die and others to survive. Thus, our technology will enable us to better understand the kind of electrical signaling cascades which occur during such events and identify any mechanisms that may help us alleviate this.

Feature image: an astrocyte grid network.

These interviews are supported by Royal Society Te Apārangi, which supports New Zealanders to explore, discover and share knowledge.