Australasian Science: Australia's authority on science since 1938

Future Chemistry from the Distant Past

Credit: bluebay2014/Adobe

Credit: bluebay2014/Adobe

By James Behrendorff, Yosephine Gumulya & Elizabeth Gillam

Enzymes resurrected from the past can survive tough industrial conditions better than their modern-day counterparts, leading to safer drugs and better biofuels.

Enzymes are protein-based catalysts that allow the chemistry of life to occur on reasonable timescales. Without them, life as we know it would not be possible. The breadth of chemical reactions catalysed by enzymes is staggering. It’s the basis to all the biochemical diversity seen in nature, like the flavours and fragrances found in foods, the drugs isolated from natural sources, and the powerful hormones that regulate growth and development.

For much of the past two centuries, chemists have been attempting to reproduce naturally occurring chemicals, but many of these molecules are extremely difficult or practically impossible to produce by conventional synthetic chemistry. Where this is possible, it often requires extreme temperatures or the use of toxic chemicals.

Since naturally occurring enzymes catalyse reactions at mild temperatures and without harsh chemicals, it seems logical to use them to catalyse challenging chemical reactions in laboratories. Why, then, haven’t they replaced conventional synthetic approaches yet?

A big part of the problem is durability. Enzymes are not built to last forever, and they haven’t evolved to work under industrial conditions. Enzymes are structurally flexible, and they are often exposed to highly reactive chemical intermediates. With prolonged use or fluctuating operational conditions, they tend to unfold and become non-functional.

This is not a problem in their natural setting. Living cells contain mechanisms to remove and replace dead enzymes. Industrial processes, however, require enzymes that can operate for as long as possible at peak performance, and often at elevated temperatures. We need enzymes with better thermostability (a measure of how long an enzyme will remain functional at a given temperature) before we can replace synthetic chemical catalysis with enzymatic biocatalysis. But the very complexity that gives enzymes their exquisite selectivity also means they can be extremely difficult to re-engineer.

Professor Frances Arnold was awarded the 2018 Nobel Prize in Chemistry for her efforts in developing techniques to improve the properties of enzymes, but despite many technological advances only modest, step-wise improvements in thermostability have been achieved. Moreover, it is still impossible to redesign enzymes from scratch to increase their thermo­stability. We simply don’t yet understand enough about how enzymes work to make rational changes to their performance.

Ancient Enzymes Were More Durable

It now appears that going back in time might provide a solution. Biochemists and evolutionary biologists have determined that the early microbial life forms that existed more than 1.5 billion years ago, during the pre-Cambrian era, must have tolerated much higher temperatures than their present-day descendants, and therefore their enzymes would have needed to be thermo­stable. Indeed, a number of pre-Cambrian enzymes have been shown previously to be thermostable.

Surprisingly, however, by winding the clock back a mere 450 million years to the start of the vertebrates, when temperatures were similar to those today, we discovered enzymes that can withstand temperatures around 30°C higher, and last more than 100 times longer, than their modern counterparts. What’s more, they were just as active, or in some cases even more efficient, than present-day enzymes.

How Can You Produce an Extinct Enzyme?

All proteins, including enzymes, are encoded by the DNA sequences that we commonly call genes. We need these gene sequences to be able to produce enzymes in the laboratory, but we cannot easily retrieve DNA from organisms that have been extinct for 450 million years.

Instead, a computational approach known as “ancestral sequence reconstruction” (ASR) is used to infer a pre­historic ancestor’s genes. With modern techniques, a predicted gene can be synthesised in the lab and the physical and chemical properties of the protein that it encodes can be explored.

To deduce an ancestral gene sequence for an enzyme of interest, you first need to gather as many present-day gene sequences as possible. One of the fundamental concepts of evolution is that modern organisms diverged from a common ancestor by slowly accumulating mutations that result in genetic differences, a process known as genetic drift. We usually think of this concept in terms of differences that accumulate in whole genomes and eventually result in the divergence of distinct species, but many of these differences show up as mutations in individual genes that encode enzymes. Organisms that exist today, from bacteria to plants and humans, often contain enzymes that are similar in what they do but have accumulated hundreds of differences in the time that it has taken for the organisms to diverge evolutionarily.

ASR infers ancestral sequences by comparing the differences in present-day proteins and determining how they could have come from a single ancestral sequence by accumulating mutations. In most cases it is possible to calculate more than one possible solution to this problem, so ASR uses a statistical model to determine the most likely common ancestral sequence. This is based on the observation that some genetic mutations are more likely to occur than others.

Once we have determined the best prediction of the ancestral protein sequence, the corresponding gene encoding it is synthesised and placed in a bacterium or other host to produce the ancestral protein. The ancient protein then undergoes performance testing in the laboratory.

Ancestral Enzymes for Safer Drugs and Biofuels

We used these methods to improve two enzymes that were both very useful but completely different to each other in terms of their activity and distribution in nature. The first was an enzyme from vertebrate animals that is useful in drug development.

Cytochrome P450 enzymes are found in all vertebrates. In humans, one particular P450 enzyme called CYP3A4 metabolises roughly half of all pharmaceuticals to allow them to be removed from the body more easily.

New drugs that are under development can be incubated with this P450 enzyme to find out how the drugs will be metabolised once they get into the human body. In some cases, a new drug will be metabolised into a compound with unexpected or even dangerous side-effects.

Being able to produce large quantities of these drug metabolites at an early stage of development is important to accurately test the safety of new drugs. Unfortunately, the present-day P450 enzyme, CYP3A4, is notoriously unstable and tends to fall apart after a relatively short amount of time, confounding its use at an industrial scale to produce drug metabolites.

The ancestral P450 enzyme that we reconstructed can metabolise the same broad range of drugs as its modern-day descendant but it lasts more than 100 times longer and can survive temperatures up to 30°C higher. It can also withstand much higher concentrations of the chemical solvents needed to dissolve many drugs.

The second enzyme we reconstructed was a bacterial ketol-acid reducto-isomerase (KARI). This enzyme can be used to produce biobutanol, a superior biofuel to ethanol.

The KARI enzyme that we studied exists in a wide variety of organisms from bacteria to plants, where it plays a key role in amino acid biosynthesis. More recently, synthetic biologists have repurposed this enzyme to develop industrial micro­organisms that produce biobutanol from sugars.

The ancestral KARI enzyme that we reconstructed showed an improvement in its thermostability of almost 20°C. To our surprise, the ancestral KARI also functioned up to eight times faster than contemporary KARIs – a discovery that could be important to improving the commercial viability of biobutanol as a low carbon, renewable fuel for transport.

As well as enabling bio-based chemical production for a new wave of biomanufacturing industries, enzymes that work faster and last longer could be used to repair some of the environmental damage caused by traditional chemical industries. Enzymes have been discovered that can break down dangerous environmental contaminants like pesticides and explosives, but they cannot be used effectively to clean up contaminated sites because of their instability and slow reaction rates. With better enzymes, hazardous chemicals could be degraded safely and cheaply. The enzymes themselves would eventually be broken down by soil-dwelling microbes, ultimately leaving no trace.

By studying the enzymes of our prehistoric ancestors and understanding how they have evolved, we have a real opportunity to redesign natural biocatalysts that will contribute to a cleaner future. Who would have thought that the answer to challenges presented by 21st century chemistry could be found by rewinding the evolutionary tape millions of years?


James Behrendorff and Yosephine Gumulya contributed to the research discussed in this article in the laboratory of Prof Elizabeth Gillam at the University of Queensland’s School of Chemistry and Molecular Biosciences. Yosephine is currently a research scientist at the CSIRO and James is a postdoctoral researcher at The University of Copenhagen. The study described here has been published in Nature Catalysis (https://go.nature.com/2Jbu4VG).