Receptor use by the Whitewater Arroyo virus glycoprotein

Abstract

Whitewater Arroyo virus (WWAV) is a North American New World arenavirus, first isolated from rats in New Mexico in 1993, and tentatively associated with three human fatalities in California in 1999-2000. However, it remains unclear whether WWAV was the cause of these, or any other, human infections. One important characteristic of viruses that influences pathogenic potential is the choice of cellular receptor and the corresponding tropism of the virus. In the arenaviruses, these properties are determined largely by the viral glycoprotein (GP). We have previously noted for the New World clade B arenaviruses, which include four severe human pathogens, that the ability to cause human disease correlates with the ability of the GP to use the human transferrin receptor 1 (hTfR1) to enter cells. In addition, pseudotyped retroviral vectors displaying the GPs from pathogenic clade B viruses transduced a range of cell lines in vitro that were distinct from those that could be transduced by non-pathogenic clade B viruses. WWAV was initially classified as a New World clade A virus, based on sequence analysis of its nucleoprotein gene. However, more extensive analyses have revealed that WWAV and the other North American arenaviruses are probably recombinant clade A/B viruses, and that the WWAV GP is more closely related to the clade B GPs. Based on this finding, we sought to understand more about the possible pathogenic potential of WWAV by determining whether its clade B-like GP exhibited the characteristics of a pathogenic or non-pathogenic clade B virus. Our studies found that WWAV GP did not use hTfR1 for entry, and that its overall in vitro tropism was most similar to the GPs from the non-pathogenic clade B viruses. Although many viral factors in addition to GP receptor use and tropism determine whether a virus is able to cause disease in humans, our analysis of the WWAV GP does not support the idea that WWAV is a human pathogen.

Figure 1. Recombinant origin for the North American New World arenaviruses

Sequence comparison of the GP and NP genes from the New World arenaviruses revealed three distinct clade, A, B and C. In addition, the North American arenaviruses, that include WWAV, are most closely related to clade A in the NP gene, but clade B in GP. A/B Rec. refers to this putative recombinant lineage. The strains that are pathogenic for humans (*) are found throughout clade B, in each of the three distinct sub-groups, B1, B2 and B3. The pathogenicity of WWAV remains unclear. Adapted from Charrel et al., 2001, 2002)

Figure 2. Titers of GP pseudotyped retroviral vectors on various cell lines

Titers were measured on (A) human kidney endothelial cells (293A, human T lymphocytes (CEM) and murine T lymphocytes (TIB27), and (B) rodent cell lines. Values shown are mean +/− SE for 2-8 independent experiments. All vector/cell combinations that gave no titer (*) were confirmed using 10 X concentrated stocks of vectors.

Figure 3. NH₄Cl sensitivity

Effect of NH₄Cl treatment on titer of indicated vectors on (A) 293A and (B) K562 cells. In each case, the titers were made relative to the values obtained in the absence of drug. Control vectors were pseudotyped with the amphotropic murine leukemia virus (A-MLV) Env, which is pH-independent and VSV-G, which is pH-dependent.

Figure 4. Effect of DG expression in murine ES cells on vector titers

The titers of GP pseudotyped vectors were measured on DG −/− and +/− murine ES cells and the ratios determined. Values were normalized to the ratio obtained with VSV-G vectors, to control for any differences between the two cell lines in their ability to support retroviral vector transduction. LASV GP vectors were highly dependent on DG for entry, while MACV and WWAV GP vectors were not affected by the loss of DG. Values shown are mean +/− SE for 2-3 independent experiments.

Figure 5. Effect of hTfR1 on vector titers on CHO-K1 cells

(A) Flow cytometric analysis of CHO-K1 and TRVb-1 cells was performed using an anti-hTfR1 antibody. TRVb-1 cells do not express endogenous CHO-K1 TfR1 but are stably transfected with hTfR1. Narrow lines represent secondary antibody only. (B) Comparison of vector titers on CHO-K1 versus TRVb-1 cells. Values shown are mean +/− SE for of 3-8 independent experiments.

Figure 6. Role of hTfR1 in viral entry

Human 293A (A) and rodent TRVb-1 (B) cells were pretreated with anti-hTfR1 antibody at two different doses, then challenged with the indicated pseudotyped vectors. Titers were converted to percent of control (titers obtained with no antibody pretreatment). Data are reported as mean ± SE of 2 independent experiments. (C) Titers of pseudotyped retroviral vectors on mock transfected CHO-K1 cells, or cells transiently transfected with an expression plasmid for hTfR1. Values shown are mean +/− SE for 3 independent experiments. (D) Western blotting confirming expression of hTfR1.

Figure 6. Role of hTfR1 in viral entry

Human 293A (A) and rodent TRVb-1 (B) cells were pretreated with anti-hTfR1 antibody at two different doses, then challenged with the indicated pseudotyped vectors. Titers were converted to percent of control (titers obtained with no antibody pretreatment). Data are reported as mean ± SE of 2 independent experiments. (C) Titers of pseudotyped retroviral vectors on mock transfected CHO-K1 cells, or cells transiently transfected with an expression plasmid for hTfR1. Values shown are mean +/− SE for 3 independent experiments. (D) Western blotting confirming expression of hTfR1.