Nucleotide sugars, the glycosyl donors in the biosynthesis of carbohydrates, are critical ingredients in the growth and development of all living organisms. A variety of nucleotide sugars simultaneously exist in biological samples. They, however, have only minor structural differences, which make them extremely difficult to discriminate. In this work, a phenylboronic acid (PBA)-modified Mycobacterium smegmatis porin A (MspA) hetero-octamer was applied to sense nucleotide sugars. Five representative nucleotide sugars, including guanosine diphosphate mannose (GDP-Man), adenosine diphosphate glucose (ADP-Glc), uridine diphosphate N-acetylglucosamine (UDP-GlcNAc), uridine diphosphate glucose (UDP-Glc), and uridine diphosphate glucoronic acid (UDP-GlcA), were successfully distinguished. A custom machine learning algorithm was also employed to automatically identify events, reporting a general accuracy of 99.4%. This sensing strategy provides a rapid, direct, and accurate method for identifying different nucleotide sugars. However, single-molecule identification of nucleotide sugars has never been previously reported, to the best of our knowledge.
Keywords: nanopore; nucleotide sugars; phenylboronic acid.