Resilience of BST-2/Tetherin structure to single amino acid substitutions

Abstract

Human tetherin, also known as BST-2 or CD317, is a dimeric, extracellular membrane-bound protein that consists of N and C terminal membrane anchors connected by an extracellular domain. BST-2 is involved in binding enveloped viruses, such as HIV, and inhibiting viral release in addition to a role in NF-kB signaling. Viral tethering by tetherin can be disrupted by the interaction with Vpu in HIV-1 in addition to other viral proteins. The structural mechanism of tetherin function is not clear and the effects of human tetherin mutations identified by sequencing consortiums are not known. To address this gap in the knowledge, we used data from the Ensembl database to construct and model known human missense mutations within the ectodomain to investigate how the structure of the ectodomain influences function. From the data, we identified an island of sequence stability within the ectodomain, which corresponds to a functionally and structurally important region identified in previous biochemical and biophysical studies. Most of the modeled mutations had little effect on the structure of the dimer and the coiled-coil, suggesting that the coiled-coil compensates for changes in primary structure. Thus, many of the functional defects observed in previous studies may not be due to changes in tetherin structure, but rather, due to in changes in protein-protein interactions or in aspects of tetherin not currently understood. The lack of structural effects by mutations known to decrease function further illustrates the need for more study of the structure-function connection for this system. Finally, apparent flexibility in tetherin sequence may allow for greater anti-viral activities with a larger number of viruses by reducing specific interactions with anti-tetherin proteins, while maintaining virus restriction.

Keywords: BST2; Molecular dynamics; Protein Structure; SNP; Tetherin.

Figure 1. Structure of human tetherin bound to the lipid membrane. Membrane anchors are shown in red (N) or magenta (C).

The coiled-coil region within the ectodomain is shown in cyan. Image derived from the model produced by Ozcan & Berndsen, (2017). The C-terminal anchor is shown as an alpha helix, however there is evidence that in some forms of tetherin, this may be a GPI anchor (Kupzig et al., 2003; Rollason et al., 2007).

Figure 2. Human mutations within tetherin.

(A) Cartoon showing domain organization and location of synonymous (purple) and missense mutations (red) in tetherin (B) Model showing the location of the mutants on the inside (red) or outside (blue) of the tetherin ectodomain. The green box in (A) and (B) indicates the location of amino acids 93–117.

Figure 3. No large changes in structure induced by mutations as shown by the root mean square deviation plotted against amino acid number.

Red points are those positions that are found on the inside of the tetherin dimer, while blue points are those positions on the outside of the dimer. Black, purple, and magenta lines show the mean, mean + one and mean + two standard deviations of the RMSD values respectively. Mutations at positions C-terminal to amino acid 163 were not analyzed because this region appears unstructured in crystal structures (Yang et al., 2010; Hinz et al., 2010; Swiecki et al., 2011).

Figure 4

(A) Secondary structure of tetherin mutant simulations over time for Arg64Pro, Ala100Pro, Gly118Cys, Ile120Phe, Asp129Glu, and Asp103Asn. Secondary structures were assigned by YASARA based on hydrogen bonding distances and psi angles at 25 picosecond intervals. ‘H’ stands for alpha helix, ‘T’ for turn, ‘C’ for coil, and ‘G’ for 3/10 helix. (B) Plot showing the cumulative number of times that the secondary structure was perturbed at each location for all simulations. Outliers were identified as sites that had distinct secondary structure more than two standard deviations above the average secondary structure value at that site across all simulations. A value of one means that the amino acid appeared as an outlier in either molecule of the dimer in 1 simulation.

Figure 5

(A) Differences in side chain position between the wild-type Isoleucine (purple) and the mutation Phenylalanine (yellow) (B) Ridgeline plot showing the range of distances between the A and B molecules (mol) or the Calpha at the indicated positions comparing the Ile120Phe simulation (yellow) and the wild-type simulation (purple). Overlap is shown in pink. (C) Ridgeline plot showing the range of distances between the A and B molecules (mol) or the Calpha at the indicated positions comparing the Ala100Pro simulation (yellow) and the wild-type simulation (purple). Overlap between the two conditions is indicated by grey coloring.

Figure 6. Mutation effects on protein dynamics (A) correlation of k-means and PAM values (B) scatterplot of k-mean values at each amino acid position.

Magenta lines indicate the mean value (solid) and the mean plus one standard deviation (dashed) of the k-means. Mutation effects on protein dynamics (A) correlation of k-means and PAM values (B) scatterplot of k-mean values at each amino acid position.

Figure 7

(A) Position of the charge swap mutations within the tetherin coiled-coil (B)–(D) Ridgeline plots showing the range of distances between the A and B molecules (mol) or the Calpha at the indicated positions comparing the charge swap mutant simulations (yellow) and the wild-type simulation (purple). Overlap is shown in grey A blue box indicates the area where these mutants are within the sequence. (B) Arg54, Arg 58, Arg64 to aspartate. (C) Glu76, Asp83, Glu85 to arginine. (D) Both sets of mutants.

Figure 8. Ridgeline plots showing the range of distances between the Calpha at the indicated positions comparing the tetra-substituted simulation (yellow) and the wild-type simulation (purple).

Overlap is shown in grey. Blue boxes indicate the region where the mutation occurs within the protein. (A) Asp55, Gly 56, Leu57, Arg58 to alanine (B) Cys63, Asn64, Arg65, Val66 to alanine. (C) Val74, Val84, Leu137, Leu144 to serine.