Unraveling viral interleukin-6 binding to gp130 and activation of STAT-signaling pathways independently of the interleukin-6 receptor

Abstract

Human herpesvirus 8 encodes a viral version of interleukin-6 (vIL-6) which shows 25% sequence homology with human IL-6. In contrast to human IL-6, which first binds to the IL-6 receptor (IL-6R) and only subsequently associates with the signal transducing receptor subunit gp130, vIL-6 has been shown to directly bind to gp130 without the need of IL-6R. As a functional consequence, vIL-6 can activate far more target cells in the body since all cells express gp130, but only cells such as hepatocytes and some leukocytes express IL-6R. We sought to understand which amino acid sequences within the vIL-6 protein were responsible for its ability to bind and activate gp130 independent of IL-6R. As a first approach, we constructed chimeric IL-6 proteins in which all known gp130 interacting sites (sites II and III) were sequentially transferred from vIL-6 into the human IL-6 protein. To our surprise, human IL-6 carrying all gp130 interacting sites from vIL-6 did not show IL-6R-independent gp130 activation. Even more surprisingly, the loop between helix B and C of vIL-6, clearly shown in the crystal structure not to be in contact with gp130, is indispensable for direct binding to and activation of gp130. This points to an IL-6R induced change of site III conformation in human IL-6, which is already preformed in vIL-6. These data indicate a novel activation mechanism of human IL-6 by the IL-6R that will be important for the construction of novel hyperactive cytokine variants.

FIG. 1.

Binding of vIL-6 to gp130. (A) Ribbon model of the recently solved vIL-6/gp130 X-ray structure. (B) Interaction of gp130 (only immunoglobulinlike domain D1) and vIL-6. Sites II and III are indicated. Detailed sequence information is shown in supplemental Fig. 1. (C) The BC loop of site III (site IIIc) is not in direct contact to gp130-D1. (D) Schematic drawing of the common four-helix bundle cytokine fold. The different parts of site II and site III are color coded as green (site II) or red (site III).

FIG. 2.

Characterization of the vIL-6/hIL-6 chimeras IV1-4. (A) Schematic representation of vIL-6, hIL-6, and chimeras IV1 to IV4. Sequence stretches that are part of the exchanged areas of vIL-6 and hIL-6 are indicated in gray (site II) and black (site III). Detailed sequence information is shown in supplemental Fig. 2. (B) Equal numbers of stably transfected Ba/F3-gp130 or Ba/F3-gp130/IL-6R cells were cultured for 3 days in the presence of IL-6 (hIL6; 100 ng/ml), Hyper-IL-6 (HIL6; 100 ng/ml), IV1 (100 ng/ml), IV2 (100 ng/ml), IV3 (100 ng/ml), or IV4 (1,000 ng/ml). Proliferation was measured by pulse-labeling the cells after 72 h with [³H]thymidine for 4 h. Cells were harvested, and the incorporated radioactivity was measured by scintillation counting. Bioassays were performed, with each value being determined in triplicate.

FIG. 3.

Characterization of the vIL-6/hIL-6 chimeras IV5 to IV9. (A) Schematic representation of vIL-6, hIL-6, and chimeras IV5 to IV9. Sequence stretches that are part of the exchanged areas of vIL-6 and hIL-6 are indicated in black. Detailed sequence information is shown in supplemental Fig. 3. (B) Western blot analysis of supernatants of COS-7 cells transiently transfected with hIL-6, vIL-6, and IV5 to IV9 using anti-c-myc antibodies. (C) Equal numbers of stably transfected Ba/F3-gp130 cells were cultured for 3 days with conditioned supernatant containing hIL-6, vIL-6, IV5, IV6, IV7, IV8, or IV9 in the presence or absence of sIL-6R (1 μg/ml). Proliferation was measured by using a CellTiter-Blue cell viability assay. Bioassays were performed, with each value being determined in triplicate.

FIG. 4.

IV9 was specifically inhibited by anti-IL-6 MAb and sgp130Fc. Equal numbers of stably transfected Ba/F3-gp130 cells were cultured for 3 days with conditioned supernatant containing hIL-6, vIL-6, or IV9 in the presence or absence of sIL-6R (1 μg/ml), sgp130Fc (10 μg/ml) or anti-IL-6 MAb (10 μg/ml). Proliferation was measured by using a CellTiter-Blue cell viability assay. Bioassays were performed, with each value being determined in triplicate.

FIG. 5.

IV9-induced STAT3 phosphorylation and coimmunoprecipitated of IV9 by sgp130Fc. (A) NIH 3T3 cells were stimulated with conditioned supernatant containing IV9 for 5 min in the presence or absence of sgp130Fc (10 μg/ml) or anti-IL-6 MAb (10 μg/ml). Unstimulated cells were used as controls. STAT3 phosphorylation was detected by Western blotting with anti-phospho-STAT3 specific antibodies. Western blotting against STAT3 served as a loading control. (B) Conditioned supernatant containing IV7 and IV9 was incubated with or without recombinant sgp130Fc protein in the presence or absence of sIL-6R and precipitated with protein A-Sepharose. Precipitated c-myc-tagged IV9, IV7, and IL-6 protein was detected by Western blotting with anti-c-myc specific antibodies. (C) Conditioned supernatant containing IV9 was incubated with or without recombinant sgp130 protein in the presence or absence of sIL-6R and precipitated with anti-c-myc tag antibodies. Precipitated sgp130 protein was detected by Western blotting with the anti-gp130 specific antibody BP-4. (D) Conditioned supernatant containing IV9 was incubated with or without recombinant sgp130Fc protein in the presence or absence of sIL-6R and the BC-forming peptide at a final concentration of 100 μM and precipitated with protein A-Sepharose. Precipitated c-myc-tagged IV9-protein was detected by Western blotting with anti-c-myc specific antibodies.