Reply to Behnam and Klein: Potential role of the His-tag in C-terminal His-tagged SARS-CoV-2 main protease

Behnam and Klein (1) carried out in vitro assays on two severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease (M^pro) inhibitors, including atazanavir reported by us (2), using an M^pro protein with C-terminal His-tag under various assay conditions, and find that both the Michaelis−Menten constant (K_m) and inhibitory activity (IC₅₀) are sensitive to the assay conditions. It is interesting to examine whether the assay conditions significantly affect the activities of an enzyme and its inhibitors. On the other hand, Behnam and Klein (1) do not account for possible effects of the C-terminal His-tag on the K_m and IC₅₀, despite the fact that the tag-free M^pro (without any N- or C-terminal tag) was used in our reported assays (2), whereas the C-terminal His-tagged M^pro was used in their assays (1).

As discussed recently (3), both the N and C termini of M^pro are close to the active-site cavity, according to available X-ray crystal structures. Thus, a His-tag on the N or C terminus could interfere in M^pro binding with a ligand (substrate or inhibitor). Depicted in Fig. 1 is a modeled structure of the C-terminal His-tagged M^pro protein. The modeling started from an available X-ray crystal structure (Protein Data Bank ID code 6M2N) (4). Notably, the His-tag was outside of the M^pro active site in the initial structure (Fig. 1A) built by using the Prime module (5, 6) of the Schrodinger 2013 software. After 50-ns molecular dynamics (MD) simulations, the His-tag moved to the M^pro active site (Fig. 1B), suggesting that the His-tag added to the C terminus may compete with the substrate/inhibitor for binding. For this reason, the C-terminal His-tag could lower the binding affinity of a given ligand (substrate or inhibitor) with M^pro. In fact, it has been reported that K_m = 1.41 μM for the tag-free M^pro (3) and K_m = 28.2 μM for the C-terminal His-tagged M^pro (7), suggesting that the C-terminal His-tag may lower the binding affinity with the substrate by ∼20-fold. The remarkable difference explains why the IC₅₀ values determined by Behnam and Klein (1) and Ma and Wang (8) using C-terminal His-tagged M^pro are significantly larger than the corresponding IC₅₀ values determined by us using the tag-free M^pro (2). It is also possible that the interaction between the His-tag and the active site is affected by the assay conditions. Hence, the results described by Behnam and Klein (1) are not surprising based on the structural information about the His-tagged M^pro protein discussed above.

Fig. 1.

Structure of the M^pro protein with His-tag (LEHHHHHH) added to the C terminus. (A) Initial structure of the C-terminal His-tagged M^pro protein with a ligand bound in the active site. (B) The last snapshot of the MD-simulated C-terminal His-tagged M^pro protein structure (in which the ligand in the active site was removed before the MD simulation) in a TIP3P water box. The system, after energy minimization, was first heated to 300 K in a 50-ps MD simulation (with NVT) followed by 50-ps MD simulation [with NPT, i.e., the constant amount of substance (N), pressure (P), and temperature (T)] for equilibration. Finally, a 50-ns MD simulation (with NPT) was performed without any restraints for the system.

Change History

September 8, 2021: The licensing has been updated.