Amino acids are the building blocks of proteins. They are raw materials for biosynthesis and have fundamental roles in various physiological and pathophysiological processes, such as epigenetic regulation and tumor metabolism. Therefore, it is crucial to detect and identify amino acids with a high spatiotemporal resolution, especially in the field of single-molecule protein sequencing. Owing to alternative RNA splicing and PTMs, the resulting proteoforms are highly complicated and contain deeper-level information that cannot be accessed directly from the transcriptome. In addition, there is no existing method similar to DNA amplification for amplifying proteins. Consequently, it is difficult to use mass-spectrometry-based methods to identify low-abundance proteins from the proteome. To address these problems, single-molecule sequencing methods that can distinguish the 20 proteinogenic amino acids are needed.
Fluorophore-based techniques allow specific amino acids, such as cysteine and lysine, to be selectively modified by fluorescent molecules. Then, by sequentially degrading the peptide using Edman chemistry, or direct imaging using single-molecule fluorescence resonance energy transfer (FRET), the relative position of labeled amino acids can be deduced from the fluorescent signals. Additionally, fluorophore-labeled amino-terminal recognizers of amino acids have been engineered to bind specific amino acids reversibly. The repetitive signals of the same amino acid can greatly improve the accuracy of single-molecule peptide identification. Although these methods have high throughput and reliability, it is difficult for chemists to label the 20 amino acids. For label-free methods, techniques such as tunneling current measurement and molecular junctions enable rapid, precise detection of up to 12 amino acids, which is still not sufficient for protein sequencing.
Given that the nanopore technique has demonstrated its superiority in single-molecule DNA sequencing, it is considered to be an ideal candidate for amino acid detection and protein sequencing. Studies have shown that peptides with different properties, such as molecular weight, length, PTMs and single-amino acid substitutions, can be detected directly and distinguished using nanopores. For further analysis of the peptide sequence, peptide translocation must be precisely controlled to generate sequence-dependent signals. The protein unfoldase ClpX has been used to unfold proteins and drive them through a nanopore, successfully discerning different protein segments. Electro-osmotic flow can be engineered to facilitate unidirectional translocation of peptides with a heterogeneous charge distribution. Moreover, the ratcheting motion of DNA–peptide conjugation through the nanopore has been achieved using DNA helicase or polymerase, generating clear sequence-dependent signals. However, there are 20 types of amino acid, so deconvoluting the signals produced by 5–6 amino acids is more complex than analyzing the signals from the four types of nucleotide, because there are many more possible combinations of amino acids. Consequently, the analysis of individual amino acids can provide valuable information and could be an alternative to peptide sequencing. Taking advantage of the pore structure, the aerolysin nanopore can differentiate 13 out of 20 amino acids when coupled with a polyarginine carrier. Furthermore, copper-ion-modified α-hemolysin and the solid-state MoS 2 nanopore have been developed to detect underivatized amino acids. Most recently, the MspA-NTA nanopore with a Ni 2+ modification has been able to distinguish the 20 proteinogenic amino acids and their PTMs with high resolution. Meanwhile, an exopeptidase protein-sequencing method in which amino acids were coupled to the peptide probe FGGCD 8 through a chemical linker was developed using an α-hemolysin nanopore. It enables an integrated approach to peptide sequencing. However, real-time detection of cleaved amino acids during peptide hydrolysis has not yet been achieved, hampering the development of single-molecule peptide sequencing.
Here, we report the direct detection of 20 proteinogenic amino acids using a copper(II)-functionalized MspA nanopore, with the limit of detection at the nanomolar range. We introduced histidine substitutions in the constriction region of the pore lumen to construct the binding sites for copper ions. With the copper ion binding to histidine residues, the reversible coordination between amino acid and copper–nanopore complex could generate well-defined current signals, enabling the detection of all 20 proteinogenic amino acids, 2 amino acids with PTMs (O-phosphoryl-l-serine (P-S) and Nε-acetyl-l-lysine (Ac-K)) and 1 unnatural amino acid (S-carboxymethyl-l-cysteine (CMC)). Furthermore, by analyzing the composition of peptide hydrolysate using exopeptidase, we identified ten different peptides. Our method enables the real-time detection of the cleaved amino acids during peptide hydrolysis and offers the possibility of inferring peptide sequences.
The conical pore geometry of the MspA nanopore makes it an ideal choice for examining small molecules.
