Capturing the wow factor with the cryo-TEM

When we mammals have a feed, our endocrine pancreas is stimulated to produce insulin. This job is done exclusively by the beta cells that reside in the islets of Langerhans.

The death or dysfunction of these beta cells means less insulin production and eventually will lead to type I or II diabetes. Dr Brad Marsh wants to understand the insulin response mechanisms of normal pancreatic beta cells from a structural perspective. More particularly, he wants to know what happens in the cell when things go wrong.

Marsh is one of NANO's core research scientists, working at the Institute for Molecular Bioscience (IMB) at the University of Queensland. His appointment as one of the founding scientific group leaders in the newly formed NANO node in Brisbane in 2004 was just like coming home for him. He was coming home, in fact, having completed his PhD at UQ in the late 1990s on how glucose is transported to the cell surface following an insulin stimulus.

Marsh says this project sparked an interest in finding better ways to understand the intracellular glucose transport compartment involved in the insulin response. To pursue this interest, he took up a postdoctoral position at the Boulder Laboratory for 3D Electron Microscopy of Cells in Colorado, an NIH-funded national research resource.

He says his move to the US aimed to "take advantage of and build upon the advances already made at that facility for overcoming the limitations of then-available cell-imaging techniques". The goal was to answer key questions about the synthesis, processing and release of insulin, and what happens in the beta cell when this process goes awry.

Conventional EM studies are limited in allowing only 2D imaging of very thin sections (~60-120 nm) of cellular material, and so do not provide any useful information of the '3D space' of a cell. The sort of questions that Marsh and others wanted to address about cell structure and function at that time required different ways of preserving and imaging mammalian cells in situ.

Electron microscopy

In Colorado, Marsh learnt everything there was to know about the relatively new EM tomography as quickly as possible, and became keenly involved in further developing these techniques, particularly in relation to the preservation and preparation of entire pancreatic islets for imaging.

As a postdoc at one of only a handful of groups in the world doing high-resolution EM tomography and 3D analysis of biological structures, Marsh rapidly advanced to become a leading researcher in his field, evidenced by his success two years later in securing a highly competitive postdoctoral fellowship from New York's Juvenile Diabetes Research Foundation International from the preliminary data and new methods he had generated since arriving there.

So, what is actually involved in producing the final 'wow' images often used to illustrate Marsh's findings? The answer is thousands of human and computer hours per one image (this is where his graduate students come in).

The basic process, developed and honed over many years, involves cutting relatively thick slices (in EM terms) of rapidly frozen and freeze-substituted cells, then collecting nearly 500 images of these slices while tilting the specimen in the microscope over a range of angles (EM tomography). The images are then combined to produce a single, high-resolution 3D reconstruction of a small area of the cell.

The freezing arrests cellular structure within milliseconds, providing a physiological 'snapshot' of the cell and minimising the artefacts of conventional EM preparation. EM tomography and image 3D reconstruction then allows all of these frozen-in-time structures to be reassembled visually on the computer in a process taking at least two months and lots of highly sophisticated and expensive equipment and computing power.

Marsh's work thus far has provided significant insight into how insulin is packaged into granules inside a beta cell for release into the bloodstream. Ultimately he wants to know how these granules are modified and recruited to the cell surface prior to the release of insulin by the pancreas, and of course what cellular defects are at play when insulin secretion is impaired.

A key goal of Marsh and the Queensland NANO node is the Visible Cell Project - a 'big-picture', multidisciplinary initiative to reconstruct an entire pancreatic islet beta cell in 3D at less than 5 nm resolution, which is 100 times larger than anything achieved so far. The project would involve four or five people working fulltime for about five years, the parallel development of complex 3D algorithms to extract meaningful data, the availability of at least some degree of automation ... and, of course, continued funding. This ambitious project will enormously advance our understanding of the production and release of insulin in healthy cells, and have important implications for the treatment of diabetes.

The final frontier

Marsh's research relies totally on NANO's FEI Technai F30 cryo-TEM - it is truly the workhorse of his lab. Commissioned in February 2004, it has already been to allow high throughput digital imaging. The wish list for the next few years includes another scope, which Marsh describes as "a genuine revolution in microscopy".

Helping Marsh's return to Australia was further support from the JDRFI in the form of a career development award, no more than seven of which are awarded worldwide. This continued investment in Marsh's research - US$625,000 over five years - together with a recent and prestigious NIH grant, more than anything illustrates the importance of his work and the respect that it draws from his scientific peers.

"An important yet unexpected" outcome of this work is his increased involvement in public education. "Public and scientific communities alike have embraced our imagery of the beta cell for a multitude of reason ... a curiosity of the cutting-edge methods ... combined with increased public awareness of diabetes, has meant that our images have graced the pages and covers of newspapers, textbooks, science journals, kids' science magazines and school material around the world," he proudly boasts.

To understand how his work captures such a wide audience, one only has to see his fantastically colourful and 'real' cell images and hear his Trekker-like promises for the future. "Only by incorporating space, the 'final dimension', can cell biologists gain a complete picture of how cells work." Beam me up.