Characterization of the New World monkey homologues of human poliovirus receptor CD155

Abstract

In contrast to Old World monkeys, most New World monkeys (NWMs) are not susceptible to poliovirus (PV), regardless of the route of infection. We have investigated the molecular basis of restricted PV pathogenesis of NWMs with two kidney cell lines of NWMs, TMX (tamarin) and NZP-60 (marmoset), and characterized their PV receptor homologues. TMX cells were susceptible to infection by PV1 (Mahoney) and PV3 (Leon) but not by PV2 (Lansing). Binding studies to TMX cells indicated that the formation of PV/receptor complexes increased when measured first at 4 degrees C and then at 25 degrees C, whereas PV2 did not significantly bind to TMX cells at either temperature. On the other hand, NZP-60 cells were not susceptible to infection by any of the PV serotypes. However, a low amount of PV1 bound to NZP-60 cells at 4 degrees C, but there was no increase of binding at 25 degrees C. In contrast, both NWM cell lines supported genome replication and virion formation when transfected with viral RNAs of either serotype, an observation indicating that infection was blocked in receptor-virus interaction. To overcome the receptor block, we substituted 3 amino acids in the marmoset receptor (nCD155), H80Q, N85S, and P87S, found in the human PV receptor, hCD155. Cells expressing the mutant receptor (L-nCD155mt) were now susceptible to infection with PV1, which correlated with an increase in PV1-bound receptor complexes from 4 degrees C to 25 degrees C. L-nCD155mt cells were, however, still resistant to PV2 and PV3. These data show that an increase in the formation of PV/receptor complexes, when measured at 4 degrees C and at 25 degrees C, correlates with and is an indicator of successful infection at 37 degrees C, suggesting that the complex formed at 25 degrees C may be an intermediate in PV uptake.

FIG. 1.

The predicted structure of CD155 homologues. Schematic diagram of the CD155 structures of human (hCD155α), AGM CD155 (AGMα1), tamarin (tCD155), and marmoset (nCD155). Three extracellular immunoglobulin-like domains (circles) are formed by disulfide bonds. The transmembrane domain and the C terminus are shown with binding domains for Tctex-1 (empty box) and μ1B subunit (shaded box). The predicted N-glycosylation sites are shown on immunoglobulin domains (squares), as well as the number of amino acids that compose each domain. hCD155 is produced as four different splice variants (α, β, γ, and δ) that differ in the presence of the transmembrane domain and the length of the C-terminal domain. AGM CD155 occurs in two splice variants, AGMα1 and AGMδ1, and AGMα2 is encoded by a second gene.

FIG. 2.

One-step growth curves of PV strains on various cell lines. One-step growth curves in mouse (Ltk⁻), human (HeLa), marmoset (NZP-60), and tamarin (TMX) cells were carried out as described in Materials and Methods. (A) PV1; (B) PV2; (C) PV3. Each point represents the mean virus titer from three experiments.

FIG. 3.

Translation and replication of PV RNA and of PV-Luc replicon in different cell lines. (A) Virus titers of PV1 (Mahoney), PV2 (Lansing), and PV3 (Leon) in marmoset (NZP-60), tamarin (TMX), and mouse (Ltk⁻) cells after transfection of RNA as described in Materials and Methods. (B) Firefly luciferase activity of PV-Luc replicons. Methods for the transfection of PV-Luc replicon RNAs into various cell lines, with or without 2 mM GnHCl, and the measurement of luciferase activity are described in the text.

FIG. 4.

Analysis of expression levels of CD155 on the cell surface by flow cytometry. The CD155 expression levels on the surface of human (HeLa), mouse (Ltk⁻), tamarin (TMX), and marmoset (NZP-60) cell lines were determined by flow cytometry using MAb p286 and a secondary Ab anti-mouse-FITC conjugate, as described in Materials and Methods. The data are expressed in arbitrary units.

FIG. 5.

Binding assay of PV strains to various cell lines. The binding of 10⁸ PFU [³⁵S]methionine-labeled PV1 (A), PV2 (B), and PV3 (C) to 1 × 10⁶ cells of human (HeLa), tamarin (TMX), marmoset (NZP-60), and mouse (Ltk⁻) cells was measured as described in Materials and Methods. The values are averages of three experiments.

FIG. 6.

Sucrose density gradient fractionation of extracts of [³⁵S]methionine-labeled PV1-infected cells. Purified native virions were attached to the cells at 25°C and incubated at 37°C for 45 min before lysis. The lysates were laid on a sucrose gradient, and the gradients were fractionated from the bottom. Unheated and heated labeled virions were used as controls.

FIG. 7.

Amino acid alignment of the V-domain of CD155 homologues (modified from that reported in reference 16). The figures shows the amino acid alignments of the V-domains of human (hCD155), African green monkey (AGMα1 and AGMα2), tamarin (tCD155), and marmoset (nCD155) proteins. Residues identified for binding all three PV strains are shown in bold, residues identified as binding two serotypes are shaded and bold, and residues identified as binding only one serotype are shown in gray. The arrows indicate the β-strands of the V-domain. Residues implicated in binding to PV by previous mutational analyses are marked with an X (5, 14) or a + (10, 30, 35, 40) in the lines above the residue numbers. The boxed residues donate amino acids mutated for this study. The numbering of amino acids corresponds to V-domains of hCD155.

FIG. 8.

Infection and virus binding of stably transfected Ltk⁻ cell lines. (A) Viral titers of PV1 growing in Ltk⁻ cells stably expressing human CD155 (L-hCD155), marmoset CD155 (L-nCD155), mutant nCD155 (L-nCD155mt), tamarin CD155 (L-tCD155), and mutant tCD155 (L-tCD155mt) were determined at 0 h and 24 h. (B) The binding of [³⁵S]methionine-labeled PV1 was measured as described in Materials and Methods.