N‐Terminomics for the Identification of In Vitro Substrates and Cleavage Site Specificity of the SARS‐CoV‐2 Main Protease

Abstract

The genome of coronaviruses, including SARS‐CoV‐2, encodes for two proteases, a papain like (PLpro) protease and the so‐called main protease (Mpro), a chymotrypsin‐like cysteine protease, also named 3CLpro or non‐structural protein 5 (nsp5). Mpro is activated by autoproteolysis and is the main protease responsible for cutting the viral polyprotein into functional units. Aside from this, it is described that Mpro proteases are also capable of processing host proteins, including those involved in the host innate immune response. To identify substrates of the three main proteases from SARS‐CoV, SARS‐CoV‐2, and hCoV‐NL63 coronviruses, an LC‐MS based N‐terminomics in vitro analysis is performed using recombinantly expressed proteases and lung epithelial and endothelial cell lysates as substrate pools. For SARS‐CoV‐2 Mpro, 445 cleavage events from more than 300 proteins are identified, while 151 and 331 Mpro derived cleavage events are identified for SARS‐CoV and hCoV‐NL63, respectively. These data enable to better understand the cleavage site specificity of the viral proteases and will help to identify novel substrates in vivo. All data are available via ProteomeXchange with identifier PXD021406.

Identification of Host Cellular Protein Substrates of SARS-COV-2 Main Protease

“…The GTRVQ*SVEQI sequence motif is fully identical in the highly homologous CTBP1 and CTBP2 human proteins ( Figure S2 ), therefore, CTBP2 is likely to be a target of SARS-CoV-2 3CL protease as well. In agreement with this, CTBP2 has been proved to be a proteolytic target of hCoV-NL63 3CLpro [ 21 ], and the highly similar cleavage site specificities imply that CTBP proteins may be potential targets of SARS-CoV and SARS-CoV-2 3CLpro enzymes, but their susceptibility for proteolytic cleavage needs to be investigated in the context of other cell types and/or species, as well.…”

“…First, we compared the autoproteolytic cleavage site sequences of SARS-CoV and SARS-CoV-2 3CLpro and found that the recognition sites closely resemble each other ( Figure 2 ). Similar to SARS-CoV [ 21 , 30 ], SARS-CoV-2 3CLpro cleavage sites also contain a conserved Gln residue in the P1 position, and there are hydrophobic (Leu, Phe, or Val) and small aliphatic residues (mainly Ser or Ala) in P2 and P1’ positions, respectively.…”

“…Additionally, we assumed that structural contexts of the putative cleavage sites need to be considered, therefore, accessibilities of target regions were also determined. A similar in silico approach has already been applied for the identification of potential cleavage sites in host selenoproteins and enzymes of glutathione synthesis [ 29 ], but neither proteomic [ 21 ] nor specifically targeted analyses proved cleavages of these targets in vitro to date. Interestingly, PAI2 was identified as a substrate of SARS-CoV and hCoV-NL63 3CLpro, as well, while cleavage of PLMN was not detected by a proteomic analysis [ 21 ].…”

“…To date, only a single in vitro study has been reported in which an LC-MS based N-terminomics approach was applied to identify host targets of SARS-CoV and CoV-2 3CLpro by incubating the recombinantly expressed enzymes with cell lysates of lung and epithelial cells. Numerous host targets have been identified, the obtained cleavage site preferences which were derived from in vitro proteomic analyses revealed a high preference for P1-Gln, P2-Leu, and P1’-Gly/Ala/Ser residues [ 21 ].…”

“…In the case of SARS-CoV-2 3CLpro, glutathione peroxidase 1, selenoprotein F, and thioredoxin reductase 1 were proposed to be host substrates by using in silico algorithms [ 29 ], but the results were not validated in vitro. These proteins were not identified in the recently reported proteomic analysis as substrates of SARS-CoV-2 3CLpro [ 21 ], although, in vitro identification of host targets in additional cell types remain to be performed. Therefore, the application of in silico methods may aid the identification of proteolytic targets, and results of in silico analyses can be correlated with those of in vitro measurements to assess the reliability of predictions, which are widely used in the computational drug design.…”