What’s ‘Hot’ to Like About Thermophilic Proteins?

When we think of living organisms we generally imagine them in conditions we are used to, but life can be found in almost any environment on Earth. In extremely hot places (such hot springs, geysers, or underwater hydrothermal vents) you find microorganisms that thrive in temperatures from 60 to over 100 °C, known as thermophiles.

These microorganisms cannot simply shield themselves from heat, due to their size, so instead must be adapted to withstand it [1]. One way thermophiles survive at such high temperatures is through using unusually bulky and highly-branched fatty acids in their membrane, sometimes using ether rather than ester linkages, to provide structure and resist hydrolysis. Their DNA can have a high GC content to prevent thermal dissociation, or be super-coiled to increase stability, and thermophiles can rapidly generate heat-sensitive biomolecules such as ATP or glutamic acid; but, what we find most interesting, is the proteins they use.

Alterations in protein sequences can cause structural changes that result in highly stable proteins that are able to function in environments where other proteins would be as denatured as scrambled egg. Scientists routinely use such thermophilic enzymes in the lab. The most immediate example is DNA polymerase from the bacteria Thermus aquaticus Taq, or the archaea Pyrococcus furiosus (Pfu), which can survive the high temperatures needed for the denaturing DNA during PCR. In the earliest days of development, PCR used a mesophilic DNA polymerase which was thermally inactivated during denaturation, requiring addition of fresh enzyme in each and every PCR cycle [2]. Other examples include a host of thermostable proteases, xylanases, and lipases which have long-standing applications as industrial biocatalysts [3].

Of course, not every mesophilic enzyme has a convenient thermophilic counterpart so readily available. What’s remarkable is that thermophiles are able to produce their incredibly stable enzymes using the same set of 20 amino acids as mesophilic organisms. At the molecular level, increased stabilisation can take the form of increased number of hydrogen bonds, ion bridges, dense hydrophobic cores, and metal ion binding [4], [5], but the same secondary and tertiary structures are found in mesophilic proteins.

This suggests that it might be possible to engineer any given protein to act at higher temperatures – and we don’t always have to end up with something that works in superheated water to make something useful. Enzymes that work in higher temperatures can have all sorts of advantages such as working faster, lasting longer, or being able to act on substrates that need heating to dissolve into solution. Thermostable enzymes can also be more stable in other challenging conditions such as high or low pH, organic solvents, or detergent solutions [4]. a-Amylase, one of the most important industrial enzymes, provides a good example. It has a wide range of applications that include the production of bioethanol for the alternative fuel industry, baking, brewing, paper recycling and starch soil removal in laundry powders. During the starch liquefaction and saccharification process, both low pH and high temperatures are used. a-Amylase from thermophiles are used to good effect, but directed evolution has produced enzymes with even greater pH tolerance, that are active at even higher temperatures [6]–[8].

There are a few strategies you can use to make an enzyme more thermostable. Firstly, you can look at the protein structure and try to pick out specific mutations that reproduce the properties found in naturally thermostable proteins, perhaps looking to stabilise hairpin loops by adding a proline cap, introduce some ionic interactions, or increase the hydrophobicity of the protein core structure. This can require a lot of carefully thought through experimentation and, of course, relies on a structure of your protein target being available.

A less specific approach, which doesn’t require a structure, is to use directed evolution, where random mutagenesis is coupled with a screening strategy to generate a protein with almost any desired property. Generally, this would involve heat-treating a batch of mutants and then performing an activity screen to select the best candidates for subsequent rounds. In each successive round, the heat treatment can be made more stringent (either using higher temperatures or longer incubation times) until the required target is met. In this case, it’s important to screen for activity alongside thermostability or you may end up with an enzyme that can withstand being boiled but no longer works. Some strategies also include an activity assay before the heat treatment, to eliminate all mutations that disrupt activity – particularly important if a protein is needed to function across a wide range of temperatures.

Computational approaches lie somewhere between these two strategies. Generally, they require a protein structure and use various modelling approaches to predict mutations likely to have the most impact on thermostability. However, as these models do not know about the activity of the target protein, it is important to rule out mutations of catalytically important residues or within active site pockets. The key advantage here is that by effectively trialling many of the random mutants in silico, the number of muteins needed to be generated in the lab is much reduced. As gene synthesis has become increasingly cheaper over the past decade, this strategy has become more appealing as the cost of generating 10s or even 100s of mutants becomes attractive in comparison to the cost of doing rounds of mutagenesis.

Typically, thermostable proteins (whether naturally occurring or engineered) are not produced in their native host. Growing microorganisms under such extreme conditions requires specialised incubation equipment and can be very slow, often to produce minimal yields of protein. Instead, many thermostable proteins are produced in recombinant hosts, such as E. coli, where milligram quantities of protein can be produced quite routinely. For some thermostable proteins, the purification can be as simple as heating the lysate to remove all the endogenous protein [9], while others might need a more traditional purification strategy.

Our staff at Peak Proteins have considerable experience expressing and purifying a wide range of classes of protein. Should that be of interest to you and your research, please don’t hesitate to get in touch with us via our website or by email (info@peakproteins.com). We look forward to hearing from you.

References

Bibliography

[1]        K. O. Stetter, ‘Extremophiles and their adaptation to hot environments’, FEBS Lett., vol. 452, no. 1–2, pp. 22–25, Jun. 1999, doi: 10.1016/S0014-5793(99)00663-8.

[2]        Thermofisher, ‘Advances in PCR would not be possible without the evolution of DNA polymerases’. https://www.thermofisher.com/uk/en/home/brands/thermo-scientific/molecular-biology/molecular-biology-learning-center/molecular-biology-resource-library/spotlight-articles/history-pcr.html.

[3]        B. L. Zamost, H. K. Nielsen, and R. L. Starnes, ‘Thermostable enzymes for industrial applications’, J. Ind. Microbiol., vol. 8, no. 2, pp. 71–81, Sep. 1991, doi: 10.1007/BF01578757.

[4]        O. P. Ward and M. Moo-Young, ‘Thermostable enzymes’, Biotechnol. Adv., vol. 6, no. 1, pp. 39–69, Jan. 1988, doi: 10.1016/0734-9750(88)90573-3.

[5]        D. W. Sammond, N. Kastelowitz, M. E. Himmel, H. Yin, M. F. Crowley, and Y. J. Bomble, ‘Comparing Residue Clusters from Thermophilic and Mesophilic Enzymes Reveals Adaptive Mechanisms’, PLoS One, vol. 11, no. 1, p. e0145848, Jan. 2016, doi: 10.1371/journal.pone.0145848.

[6]        A. JONES et al., ‘Directed evolution of a maltogenic α-amylase from Bacillus sp. TS-25’, J. Biotechnol., vol. 134, no. 3–4, pp. 325–333, Apr. 2008, doi: 10.1016/j.jbiotec.2008.01.016.

[7]        Y.-W. Kim et al., ‘Directed Evolution of Thermus Maltogenic Amylase toward Enhanced Thermal Resistance’, Appl. Environ. Microbiol., vol. 69, no. 8, pp. 4866–4874, Aug. 2003, doi: 10.1128/AEM.69.8.4866-4874.2003.

[8]        T. H. Richardson et al., ‘A Novel, High Performance Enzyme for Starch Liquefaction’, J. Biol. Chem., vol. 277, no. 29, pp. 26501–26507, Jul. 2002, doi: 10.1074/jbc.M203183200.

[9]        P. Ferralli, J. Duick Egan, and F. L. Erickson, ‘Making Taq DNA polymerase in the undergraduate biology laboratory’, BioOne Complet., vol. 78, no. 2, pp. 69–74, 2007.

Written by Ravneet Mandair and Duncan Smith