Working at the limits of protein crystallography


Share this blog

Just when you think you’ve discovered everything there is to know about a subject and there’s little left to learn or discover, something comes out of nowhere to prove you wrong! But in crystallography that’s not necessarily a bad thing…

This was certainly true for CSIRO's Collaborative Crystallization Center* (C3) in Australia when setting up seeded crystallization experiments of AtzD; an enzyme involved in the detoxification of the pesticide atrazine. The results demonstrated the first X-ray structure for this class of enzyme. But the really cool bit was finding that it performed an interesting ring opening chemistry together with a novel fold that had not previously been observed in a protein structure (termed the ‘Toblerone fold’ because of its three-fold rotational symmetry). The high-resolution structure allowed the identification of the binding pocket residues that are involved in substrate specificity.


Image 1: a ribbon representation of a tetramer of AtzD, crystals of the AtzD protein set up at C3 and lysozyme in cubic B3

An integral part of all the hard work which yielded this success was their mosquito LCP –  a fundamental addition to their existing range of cutting edge lab automation. It’s low-volume positive displacement technology gives rapid and accurate dispensing, while ensuring zero cross-contamination. 

Facility manager Dr. Janet Newman (below) also confirmed: "The disposable tips eliminate the need for time-consuming wash stages between the set up of individual screening plates, which significantly speeds up the process of setting up crystallization experiments and allows us to set up difficult samples without fear of them clogging up the machines."



Interestingly mosquito® LCP – isn’t just limited to vapor diffusion. C3 have a second mosquito® LCP located in the cold room, used for bicelle crystallization, which requires low-temperature dispensing. This second instrument is very helpful for setting up crystallization trials with light and/or temperature-sensitive samples.

The instrument’s small footprint and ability to take up multiple small volumes of highly viscous materials has significantly enhanced C3’s throughput. This has enabled the team to setup experiments which fall outside the normal mandate of a crystallization laboratory, for example, setting up lipid/solution mixtures for high-throughput lipid phase analysis.

The CSIRO* have also recently published two brand new Toblerone-fold enzymes in 
AEM Accepts: a barbituric acid hydrolase and an enzyme of no known function As well as confirming that this is a structural family of proteins, they also “uncovered some interesting features, such as an unusual pair of vicinal disulphide bonds in one of the enzymes that can be reversibly oxidized and reduced.”


Image 2: Details of two brand new Toblerone-fold enzymes and an enzyme of no known function

As Australia’s only full fee-for-service crystallization facility, these are only a couple of the many success stories gained from their academic and commercial collaborations. Another is Cinder, a crystal-image classification tool used to successfully score a wide variety of crystallization experiment images.

According to CSIRO* the ultimate goal is: “to have computer software classify images so that we no longer need to trawl through 1000s of images, and for that we need an easy way to let many people score a training set (enter Cinder). With enough development support we will release a client specific version of Cinder, an alternative to desktop applications. This will make it easier for more people to score more often, and we can capture an ever-growing data set.”

Now that sounds like an interesting take on Tinder!