Molecular modelling of the HCMV IL-10 protein isoforms and analysis of their interaction with the human IL-10 receptor

Abstract

The human cytomegalovirus (HCMV) UL111A gene encodes several homologs of the cellular interleukin 10 (cIL-10). Alternative splicing in the UL111A region produces two relatively well-characterized transcripts designated cmvIL-10 (isoform A) and LAcmvIL-10 (isoform B). The cmvIL-10 protein is the best characterized, both structurally and functionally, and has many immunosuppressive activities similar to cIL-10, while LAcmvIL-10 has more restricted biological activities. Alternative splicing also results in five less studied UL111A transcripts encoding additional proteins homologous to cIL-10 (isoforms C to G). These transcripts were identified during productive HCMV infection of MRC-5 cells with the high passage laboratory adapted AD169 strain, and the structure and properties of the corresponding proteins are largely unknown. Moreover, it is unclear whether these protein isoforms are able to bind the cellular IL-10 receptor and induce signalling. In the present study, we investigated the expression spectrum of UL111A transcripts in fully permissive MRC-5 cells and semi permissive U251 cells infected with the low passage HCMV strain TB40E. We identified a new spliced transcript (H) expressed during productive infection. Using computational methods, we carried out molecular modelling studies on the three-dimensional structures of the HCMV IL-10 proteins encoded by the transcripts detected in our work (cmvIL-10 (A), LAcmvIL-10 (B), E, F and H) and on their interaction with the human IL-10 receptor (IL-10R1). The modelling predicts clear differences between the isoform structures. Furthermore, the in silico simulations (molecular dynamics simulation and normal-mode analyses) allowed us to evaluate regions that contain potential receptor binding sites in each isoform. The analyses demonstrate that the complexes between the isoforms and IL-10R1 present different types of molecular interactions and consequently different affinities and stabilities. The knowledge about structure and expression of specific viral IL-10 isoforms has implications for understanding of their properties and role in HCMV immune evasion and pathogenesis.

Fig 1. HCMV IL-10 transcripts detected in infected cells.

(A) Acrylamide gel electrophoresis showing the HCMV IL-10 transcripts obtained by nested PCR, in MRC-5 and U251 cells, infected for the time points indicated. The identitity of the transcripts determined by DNA sequencing is indicated by the letters (U) unspliced; (A) transcript A (cmvIL-10); (B) transcript B (LAcmvIL-10); transcript E, F and H. (B) Squematic illustration of HCMV IL-10 pre-mRNA (unspliced) and the transcripts identified. Exon (continuous lines) and intron (dotted lines) positions are shown according to the HCMV TB40e strain (gene Bank: KF297339). The start and end positions of each intron are numbered, relative to the TB40E strain.

Fig 2. Structure and amino acid sequences of cIL-10 and HCMV IL-10 isoforms.

(A) Structures of cIL-10 (green) and isoform A (cmvIL-10) (yellow), obtained from Protein Data Bank (PDB) (PDB: 1Y6K), and the modelled structures of isoform A (cmvIL-10 (red), isoform B (LAcmvIL-10) (blue), isoform E (orange), isoform F (pink) and isoform H (light green). The helices A, B, C, D, E, F and H, are highlighted in the structures. Amino acids described as important for cmvIL-10/IL-10R1 binding are shown in grey and identified with the residue name and position. (B) Amino acid alignment between the cIL-10 and HCMV isoforms. Predicted secondary structures are highlighted as red (helix) and blue (beta sheet). The nomenclature adopted for the helix structures (A, B, C, D, E, F and H) are presented.

Fig 3. Particularities of the complex formed between the Isoform F and IL-10R1.

(A) Binding sites detected in the isoforms. Site 1: Pink; site 2: Green; site 3: Lilac; site 4: Brown; site 5: Olive green. (B) Representation of the isoform F on the surface. (C) Bioactive conformation of the isoform F coupled to the IL-10R1. (D) Vectors obtained by calculating the normal modes, where we observe regions with less movement (more rigid) (blue), regions composed of residues with high movement (pink and red), and regions with intermediate movement (white).

Fig 4. Difference in root mean square fluctuation (RMSF) Presented values between isoform F complexed with the IL-10R1 and isoform F in Apo conformation, in green and blue respectively.

Fig 5. Stability of the complex formed by isoform F and IL-10R1.

(A) Stability of the complex formed by isoform F and IL-10R1, according to the root mean square fluctuation (RMSD) values. (B) The stability of the hydrogen bonds of the complex formed by isoform F and IL-10R1.