Finding Phytase

When a research team accidentally discovered plant enzymes in a batch of supposedly pure, commercially available recombinant human blood protein, it kicked off a new project into using nuclear magnetic resonance (NMR) spectroscopy as a quality control test (1).

“We were conducting binding studies between the blood plasma protein, albumin, and adenosine 5’-triphosphate (ATP) and other ligands,” explains Robert Brinson, a chemist at the Institute for Bioscience and Biotechnology Research, which is part of the US National Institute of Standards and Technology (NIST). “We purchased recombinant human serum albumin (rHSA) without consideration of its source. Commercial biotechnology-grade rHSA is typically produced in yeast. However, the new rHSA product that we purchased was produced in rice. Albumin does not have the ability to degrade ATP, but when we performed binding experiments, we found that the ATP rapidly degraded.”

The problem turned out to be contamination with phytase – a plant enzyme that hadn’t been spotted by the manufacturer in its quality control tests. Probing the problem a little further, the group conducted a systematic 31P NMR study of HSA and rHSA products, which included rice-derived products from various vendors and control products, human serum pooled HSA and yeast-derived rHSA. Varying levels of phytase contamination were found. Such residual host cell protein impurities can cause toxic or immunogenic responses in patients.

Could similar contamination occur in biotherapeutics? Brinson and the team didn’t test any drug products, but if the purification process for clinical, rice-derived rHSA is similar to biotechnology grade products, then there is a risk. Brinson says, “It is not the mission of NIST to test specific clinical products for companies. Our goal in the field of biomanufacturing is to develop measurement methods to aid industry and regulators in the characterization of biotherapeutics.” And he believes that 31P NMR could step up to the plate. 31P NMR only measures phosphorylated molecules, which means that measurements can be conducted on complex, buffered protein mixtures. In addition, 31P NMR does not require knowledge of the phosphatase – the general name for any enzyme that cleaves a phosphate from a small organic molecule, such as ATP.

The two standard methods for contamination detection are ELISA and two-dimensional liquid chromatography coupled to mass spectrometry (2D-LC-MS). “The specifics of performing these assays are markedly different, but the purpose of both techniques is to determine host cell protein and other protein contaminants. And while these are indeed very sensitive techniques, the biggest drawback is the need to have an idea of potential protein contaminants. Otherwise, the mass level of an unknown impurity may fall below the threshold of detection,” says Brinson.

In essence, a highly active enzyme that is only present in trace amounts could be missed by traditional techniques; however, the resulting cleavage reaction can be seen with 31P NMR. Brinson notes, “We see our method as complementary to the established methods. The 31P NMR method does not supplant them. NMR is not a high-throughput technique. While the 31P NMR assay is only 20 minutes, the other methods are truly high throughput and can analyze products in parallel.”