Scientists at the U.S. Food and Drug Administration (FDA) used a common laboratory technique for increasing production of therapeutic proteins, to show that a genetic mutation that is frequent in the population and has no measurable impact on the structure of an enzyme, still enhances its production and activity.

The enzyme, called ADAMTS13, acts like a biochemical safety switch that prevents blood clotting by binding to and cutting a protein that normally triggers blood clotting. If ADAMTS13 is absent or occurs in only small amounts, its protein target, von Willebrand Factor (VWF), remains active. This increases the chance that VWF will form unneeded clots, which can cause complications such as stroke or heart attacks. The increase in ADAMTS13 production and activity caused by the mutation might protect against this abnormal blood clotting by reducing VWF activity. On very rare occasions, in severe von Willebrand Disease, when VWF levels are extremely low, such overproduction of ADAMTS13 may pose a risk.

The FDA study is important because this type of genetic change, called a synonymous mutation, is often used to make therapeutic proteins by “optimizing” codons. Codons are groups of three nucleotide building blocks of DNA or its decoded form, messenger RNA (mRNA). Each codon encodes a specific amino acid building block of a particular protein, which is made at the ribosome—the protein manufacturing factory found in cells. But there is more than one codon that can represent each amino acid in the cell and some codons work better than others. That’s because some transfer RNAs (tRNAs) are more abundant than others. tRNAs are the molecules that, based on the codon that is present, deliver specific amino acids to the growing protein strand at the ribosome. So, by genetically engineering a codon that works optimally at the ribosome (optimizing the codon), scientists can boost protein production.   

Some regular (i.e., non-synonymous) mutations can disrupt protein structure and function, as occurs when a mutation in the hemoglobin gene causes sickle-cell disease. In contrast, synonymous mutations were long thought to have no effect on either the production or function of proteins. However, findings made by scientists from both FDA and other institutions have challenged this idea.

It is now known that synonymous mutations can either reduce or enhance protein synthesis. For example, a deleterious synonymous mutation could change the shape of mRNA, disrupting its normal interaction with the ribosome, decreasing or eliminating protein production. Or, it could attract a different—and less efficient—tRNA.  This could slow or disrupt the rate of protein production at the ribosome or cause production of a misshaped protein that fails to work properly.  In addition, some synonymous mutations disrupt the process that normally modifies newly-made mRNA so it can efficiently guide protein production. This can lead to deletion of a segment in the protein, which disrupts its shape and function.

A beneficial synonymous mutation, however, might enhance protein production, for example, by attracting a more efficient tRNA.

In the current study, the FDA scientists showed that a particular synonymous mutation in the ADAMTS13 gene increases production of that protein. They used cell-free assays to study how a naturally-occurring variation of a codon in the ADAMTS13 gene called c.354G>A (p.P118P), which has not been shown to cause disease, affected the activity of the ADAMTS13 enzyme.

The scientists had previously discovered that the p.P118P variant of the gene was translated into the enzyme at higher levels than the unmutated gene. The new study showed that the p118P mutation optimized the codon, causing a 1.4-fold enhancement of production of the enzyme. The synonymous mutation also made the codon work more efficiently with both the codon preceding it and the codon following it, during translation of the mRNA into the ADAMTS13 enzyme. This cooperation appeared to improve the efficiency of the ribosome in translating the mRNA into the enzyme.

Despite the increased production and efficiency of ADAMTS13, this variant had virtually the same structure as the original (wildtype) enzyme. Nevertheless, the variant enzyme was more active in cleaving VWF than the wildtype enzyme. Based on previous research by others on the role of a critical part of ADAMTS13 in binding to and cutting VWF, the results of the FDA study suggest that this mutation improves the ability of the enzyme to bind to its target. This binding occurred away from the enzyme’s “active site,” where it cleaves VWF. The additional binding appears to stabilize the enzyme on the target protein as it cuts VWF.

In summary, the new FDA finding provides evidence that certain synonymous mutations can have beneficial effects that protect against disease. In addition, the results show that knowing which synonymous mutations boost protein production and activity could increase efficiency in bioengineering therapeutic proteins.

TITLE

A single synonymous variant (c.354G>A [p.P118P]) in ADAMTS13 confers enhanced specific activity

Int J Mol Sci. 2019 Nov 15;20(22). pii: E5734. doi: 10.3390/ijms20225734 .

AUTHORS

Ryan C. Hunt^1#, Gaya K. Hettiarachchi^1#, Upendra K. Katneni^1#, Nancy E. Hernandez¹,David D. Holcomb¹, Jacob M. Kames¹, Redab Alnifaidy¹, Brian C. Lin¹, Nobuko Hamasaki-Katagiri¹, Aaron M. Wesley², Tal Kafri², Christina Morris³, Laura Bouché⁴, Maria Panico^3,4, Tal Schiller¹, Juan C. Ibla⁵, Haim Bar⁶, Amra Ismail⁷, Howard R. Morris^3,4, Anton A. Komar⁷ and Chava Kimchi-Sarfaty¹*

 ¹ Hemostasis Branch, Division of Plasma Protein Therapeutics, Office of Tissues and Advanced Therapies, Center for Biologics Evaluation & Research, US FDA, Silver Spring, MD, USA
²Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
³BioPharmaSpec Ltd, St. Saviour, Jersey, UK
⁴Department of Life Sciences, Imperial College London, South Kensington Campus, London, UK
⁵Departments of Cardiac Surgery and Anesthesiology, Perioperative and Pain Medicine, Boston Children’s Hospital and Harvard Medical School, Boston MA, USA
⁶Department of Statistics, University of Connecticut, Storrs, CT 06269, USA
⁷Center for Gene Regulation in Health and Disease, Department of Biological, Geological & Environmental Sciences, Cleveland State University, Cleveland, OH, USA

^# These authors contributed equally to this work.

 ^*Address correspondence to:

Chava Kimchi-Sarfaty, Ph.D.
[email protected]