Chemzymes, Challenging Nature: How Artificial Enzymes Are Becoming Nature’s Counterparts

Schematic drawing of an artificial enzyme

Enzymes were discovered as early as the nineteenth century, [15] when Eduard Buchner extracted the enzyme responsible for the fermentation of sugars from yeast into alcohol. Besides earning him the Nobel Prize, Buchner’s work began the process of elucidating the diverse functions of enzymes, thousands of which have since been identified [12]. They allow us to do everything from digestion to synthesizing new drugs. Enzymes are testaments to evolution, some having taken millions of years to perfect. The magnitude of the task of replicating their creation is daunting, to say the least, but a select few researchers are taking on this challenge in hopes of expanding the application of enzymes to both biological and laboratory chemistry.

Until very recently, the only way to make new enzymes was to take natural enzymes and force them to evolve through elaborate systems of mutation and identification until the resultant “tailor-made” enzyme performed novel reactions. Unfortunately, it was not as effortless as simply taking themes and motifs from nature in order to ‘mix and match,’ even with the development of sophisticated computer analyzers. Dr. Mikael Bols and his colleagues from the University of Copenhagen, however, are providing new hope with their creation of a truly artificial enzyme analogue: the “chemzyme.” Even though the idea was not previously unheard of, the chemzyme developed by the Copenhagen research group is the only artificial enzyme to match natural enzymes’ rate of catalysis and specificity. What is even more interesting is that no natural enzyme can perform the reaction that the chemzyme Dr. Bols’ team created is designed for. Herein, the methods of generating tailor-made enzymes will be contrasted with the methods used to generate chemzymes. The promise these new artificial enzymes hold for chemistry and medicine will also be discussed.

One primary point to be addressed is how enzymes are produced naturally. Briefly, enzymes comprise a very significant portion of the protein produced by biological organisms. An enzyme, more specifically, is a protein or piece of RNA that catalyzes a reaction. The biological functions of enzymes are as diverse as the number of known enzymes is large. For example, the biological metabolism of glucose to pyruvate (a process called glycolysis) involves ten different enzymes, and pyruvate digestion requires even more [14]. Enzymes catalyze the myriad chemical reactions necessary for life, and can be found in both the simplest living organisms and the most complex. Typically, one enzyme catalyzes one reaction and one reaction only [14], recognizing no more than the one or very few molecules, called substrates, whose reaction the enzyme takes part in. As a continued example, triose phosphate isomerase, involved in glycolysis, is said to have reached evolutionary perfection. It catalyzes reactions at a rate that approaches theoretical maximum, and only recognizes three-carbon-molecules bearing three oxygen groups [11].

The physical basis for this catalysis and specificity is derived from two key elements: the active site, and the protein scaffold upon which the active site is located [1]. The active site is minuscule, averaging three to four amino acids versus up to several thousand for the whole enzyme [1, 5]. Metal cores or specific functional groups (i.e. portions of molecules that have particular properties) actively bind the substrate, stabilizing it during the reaction [8]. The protein scaffold is the amino acid polymer upon which the active site is located, that does not directly participate in catalysis, and whose three-dimensional structure dictates the enzyme’s specificity and the environment in which it can operate [9].

To clarify, assume that two molecules have similar structures, but one is much larger than the other. If the protein scaffold restricts the binding site such that only the smaller molecule can fit, specificity is observed. The combination of active site and protein scaffold yields a highly selective, highly efficient solution to very difficult chemical syntheses under conditions that the body can handle (for example, reactions conducted in water) [2]. They are so efficient that enzymes can speed up chemical reactions to be 1019 times faster than they would be otherwise and simultaneously produce no byproducts [2]. The digestion of glucose occurs at the order of seconds in the body, but could take five million years for the same reactions to occur without catalysis [11][A].

Reaching natural enzymatic efficiency in laboratory and industrial chemistry is highly uncommon despite significant advances in catalyst development[B]. To circumvent this problem, many industries have opted to simply include certain naturally occurring enzymes into their products. For example, many cleaning materials contain proteases, enzymes that degrade proteins to make them easier to remove [10]. Although useful, these applications for enzymes only push the envelope so far. What many scientists are now doing is allowing evolution and computers to create new enzymes for them, generating novel catalysts to suit difficult chemical syntheses. This approach has proven to be quite useful and involves cutting-edge technology that places the field at the forefront of biochemistry. Commonly used are directed evolution techniques together with computational design and rational protein design. Rational protein design relies upon a detailed understanding of the three-dimensional structure of the protein in the hopes of making changes more predictable and logical [3].

Directed evolution exploits Darwinian theory to produce enzymes that have different or ‘unnatural’ specificity, solubility, or efficiency. Using well-known and relatively simple techniques such as site-directed mutagenesis and plasmid vector transformation, the gene for the enzyme of interest is induced to mutate hundreds, if not thousands, of times. Once a copy of each mutated gene is made, researchers can use bacteria (which reproduce quickly and are very inexpensive) to generate the new enzymes. If any of these new enzymes proves to be different in a beneficial way from the original, the mutant gene is induced to mutate many more times. Ultimately, many successive repetitions of the above procedure can generate an enzyme that is vastly different from the original [6]. The use of computers in such protein engineering has been pivotal, as intricate algorithms can help predict the changes that will result from mutations before they happen. This allows scientists to reorganize, mixing and matching in a sense [9], protein scaffolds with active sites so that synthetic genes can be created that produce novel enzymes [6]. Together, computational design and directed-evolution have provided many innovative enzymes capable of reactions that were previously impossible or very difficult [6].

Structure of a cyclodextrin

This active reorganization, directed evolution, and computational modeling are now a common practice in many chemical and biological laboratories, but one might be wondering why these do not qualify as chemzymes. The answer is that a chemzyme is not a protein at all, but rather a small, entirely man-made molecule capable of catalysis at rates equivalent to its natural counterparts and under very mild conditions [4][C]. Rather than using a protein scaffold, Dr. Bols and his team are using variable size, open conical structures called cyclodextrins derived from the enzymatic degradation of starches. The cyclodextrin is composed of multiple glucose molecules with a very specific arrangement that makes them soluble in water, while keeping the inside of the open cone water-free. Using some newly developed methods, functional groups like those found on enzymes can be systematically implanted onto the cyclodextrins. These functional groups are not chosen at random, either. Rather, the research team in Copenhagen makes no secret of the fact that they rely heavily on the themes and motifs of active sites found on natural enzymes. The main difference is that when using directed-evolution, besides time, very specific conditions are needed to ensure that the enzymes do not degrade. During chemzyme synthesis, if previous information about an enzyme of interest is available, the chemzyme can be made under industrial conditions, with no need to alter or grow bacteria. The project of synthesizing new chemzymes that have different reactivities and substrate specificities, and the subsequent search for their applications, will take up a great deal of the coming research in the field.

What can society expect then from chemzymes, and more importantly, when? Fortunately, the number of ways these new enzymes can expand today’s chemical repertoire is vast. The materials needed are already available, inexpensive, can be made quickly, and the active site can be altered almost endlessly to produce novel reactivities. “Tailor-made” enzymes, or the ones produced by directed-evolution, suffer serious time delays and possible dead ends in their applications because the proteins are too fragile [7-8]. As Dr. Mikael Bols puts it, “Enzyme catalysis is [already] important since it is mild, green, and sustainable, but the scope is limited by Nature… chemzymes are the only way to expand the field.”

The reason they can expand the field to a much higher degree is because of their versatility and much higher resilience to temperature fluctuations than natural enzymes. Dr. Jeannette Bjerre, one of Dr. Bols’ colleagues, discovered the chemzyme that this paper focuses on during her doctoral research. The chemzyme shows highly selective and efficient degradation of the toxin found in horse chestnuts. Her success has encouraged the possibility of “selective enzymatic remedies for poisoning,” and other medicinal applications. Unfortunately, although seemingly straightforward, the synthesis of certain new chemzymes is still complicated, and as for any product aimed for the medical industry, many tests remain. Chemzymes rely heavily on natural enzyme motifs and even today the nature of protein structure and functions remain elusive. For these reasons, Dr. Bols anticipates ten to twenty years before chemzymes are implemented in enzyme catalysis on a regular basis. However, due to the possibility of mass production, functionality under physiological conditions, and their selectivity, it would be interesting to see if chemzymes can be developed the way hormone therapies have been. Hormones like insulin, for example, are available at industrial scales. If chemzymes could be developed to treat metabolic disorders like phenylkaptonuria, which is caused by the absence of an enzyme needed to degrade phenylalanine, they could serve as a treatment option. If these enzymes could be replaced with chemzyme analogues, those suffering from phenylkaptonuria would experience a less restricted lifestyle, and thus decrease their daily stress.


[A] Breaking of the glycosidic bond without catalysis would take this long, more specifically.

[B] Nitrogen fixation is the most energy consuming process in industrial chemistry, and transition metal catalyzed reactions can still require several days to run.

[C] Mild conditions refers to a neutral pH environment, at room or physiological temperatures, and standard barometric pressure.


  1. Bartlett GJ, Porter CT, Brkakoti N, Thornton JM. Analysis of Catalytic Residues in Enzyme Active Sites. Journal of Molecular Biology. 2002 Nov 15;324(1):105-121.
  2. Bjerre J, Rousseau C, Bols M. Artificial Enzymes, “Chemzymes:” Current State and Perspectives. Applied Microbiology and Biotechnology, Copenhagen, Denmark. 2008;81:1-11
  3. Boas FE. Physics-Based Design of Protein-Ligand Binding. PhD dissertation, Department of Biochemistry, Stanford University. 2008.
  4. Bols, M. E-mail to Joseph Bartolacci ( 2011 May 19 [cited 2011 Aug 6].
  5. Brocchieri L, Karlin S. Protein length in eukaryotic and prokaryotic proteomes. Nucleic Acids Research. 2005;33(10):3390–3400.
  6. Johannes TW, Zhao H. Directed Evolution of Enzymes and Biosynthetic Pathways. Current Opinion in Microbiology. 2006;9:261–267.
  7. Khoury GA, Fazelinia H, Chin JW, Pantazes RJ, Cirino PC, Maranas CD. Computational Design of Candida Boidinii Xylose Reductase for Altered Cofactor Specificity. Protein Science. 2009;18(10):2125–2138.
  8. Kung Y, Doukov T, Seravalli J, Ragsdale S, and Drennan C. Crystallographic snapshots of cyanide- and water-bound C-clusters from bifunctional carbon monoxide dehydrogenase/acetyl-CoA synthase. Biochemistry. 2009;48(31):7432-7440.
  9. Malisi C, Kohlbacher O, Höcker B. Automated Scaffold Selection for Enzyme Design. Max Planck Institute for Developmental Biology, Germany. 2009. 74-83.
  10. Maurer, KH. Detergent Proteases. Current Opinion in Biotechnology. 2004;14(4):330-334.
  11. Nelson DL, Cox MM. Lehninger Principles of Biochemistry, Fourth Edition. 4th ed.
  12. Nobel Lectures, Chemistry. Amsterdam: Elsevier Publishing Company; 1966.
  13. Pelsajovich SG, Tawfik DS. Protein Engineers Turned Evolutionists. 4th ed.: Nature; 2007.
  14. Solomons TW, Fryhle C. Organic Chemistry Ninth Ed.: Wiley.
  15. Williams HS. A History of Science: in Five Volumes. Volume IV: Modern Development of the Chemical and Biological Sciences. New York: Harper and Brothers. 1904.
  16. Artificial Enzyme. (Wikimedia Commons). 2007 Sep 11 [cited 2011 Aug 13]. Available from:
  17. AlphaCyclodextrin structure. (Wikimedia Commons). 2004 Dec 28 [cited 2011 Aug 13]. Available from:

Joseph Bartolacci is a third-year student at the University of Chicago majoring in biological sciences, chemistry, and biological chemistry. Join The Triple Helix Online on Facebook and follow us on Twitter.