Page 119 - Read Online

Page 8 of 16 Varming et al. Microbiome Res Rep 2024;3:15 https://dx.doi.org/10.20517/mrr.2023.50

RESULTS
A new crystal structure of TP901-1 CI-NTD
As introduced earlier, the TP901-1 lysogeny switch has been characterized extensively at the structural level,
and indeed, four crystal structures have already been deposited in the PDB containing the CI-NTD. Table 1
summarizes the characteristics of these structures and corresponding sequences as well as the new one
presented here. However, none of the uncomplexed structures of CI-NTD corresponded to the identical
sequence of the construct used for the interaction studies with Mor, which motivated a new crystallization
effort. The new X-ray structure, which has by far the highest resolution for a TP901-1 CI-NTD structure
obtained so far, resembles the overall previous structures of TP901-1 CI-NTD with the typical HTH motif.
Generally, the structure is of high quality. It was determined at a maximum resolution of 1.29 Å with a final
R factor of 14.97% and R free of 18.29%. 98.7% of residues are in the favored regions of the Ramachandran
plot, no residues lie in outlier regions, and stereochemical parameters are good. However, several exposed
side chains have alternate conformations (Met1, Ser6, Gln11, Ile12, Met13, Glu15, Ile23, Ser41, Ser44, Ile58,
Glu69, Met73, Met79, and Val80) or are not fully visible in the electron density maps. The termini, in
particular, were extremely difficult to place. The C-terminus beyond residue 80 is particularly problematic,
as could be expected in view of previous results indicating the start of the linker to the CTD in this region.
However, there was definite disordered electron density beyond this point, which, after many attempts, was
modeled as res 81 connected to the rest of the chain and two additional residues (83-84) after a one-residue
gap. As modeled, the positions of res 81 and the res 83-84 segment are not compatible with each other, but
each represents one of presumably multiple conformations assumed by the chain in this area. Atomic details
in this region cannot be relied upon and some electron density remains unmodeled.

Comparison of TP901-1 CI-NTD crystal structures in the interface area
We have previously analyzed in detail the dynamics of TP901-1 Mor based on NMR data and the
CI-NTD:Mor interface as revealed by the crystal structure, and discovered that rearrangement and
dynamics of aromatic residues at the Mor interface with CI-NTD are important for the formation of the
complex . The N-terminus of Mor is involved in the interface with CI, including aromatic residues Tyr3
[24]
and Tyr5. Furthermore, in the NMR structure of Mor alone, the conformation of Trp43 is incompatible
with the crystal complex with CI-NTD. Thus, molecular recognition dynamics involving repacking of two
aromatic rings (Tyr5 and Trp43) seems to be required for interaction with CI-NTD.

Here, we focus on the reorientation of residues that might occur in the TP901-1 CI-NTD when interacting
with Mor, based on a comparison of the various crystal structures in Table 1. Interface residues have
previously been identified in the N-terminal region (2-QTDT-5), C-terminal region
(68-SEAWLMGFDVP-78), and the first half of α-helix 4 (52-SPDQNRIY-59) . The N-terminus, which
[24]
differs considerably in different structures (also as the native Q marked in bold above is sometimes a K due
to cloning artifacts), has only a minor involvement in complex formation, so it will not be discussed in
detail. Gln55, Glu69, and Met73 (underlined in the sequences above) stand out as interface residues taking
on different conformations in different crystal structures. CI Gln55 [Figure 2A] in the complex has to fit
snugly into a cavity formed by aromatic residues Tyr3, Tyr5, Trp43, and Phe67 of Mor, where it comes into
hydrogen bonding distance of Tyr5. Gln55, which is overall well defined in the electron densities of the
respective structures, needs to generally make a minor rearrangement to fit in the complex, compared to the
larger rearrangements seen by some of the aromatic side chains forming the pocket in Mor. The position of
CI Glu69 in all crystal structures except for the CI:Mor complex [Figure 2B] is always pointing in the same
direction and is reasonably well-defined in the respective electron densities, which could be due to the
residue being stabilized in this position through interaction with side/main chain amines of crystal
neighbors. In the most recent high-resolution structure, there are no such interactions and two
conformations were necessary for adequate modeling of the side chain. An omit map where Glu69 was