Alphavirus entry: NRAMP leads the way

Abstract

The identity of the receptors that mediate alphavirus entry into host cells has been elusive. In this issue, Rose et al. (2011) use a Drosophila RNAi screen to identify NRAMP, an iron transporter with 12 transmembrane domains, as a receptor for Sindbis virus in both insect and mammalian cells.

Enveloped viruses infect cells by a membrane fusion reaction that delivers the viral genome into the cytoplasm. A critical initial step in infection is the binding of the virion to a receptor on the surface of the host cell. This virus-receptor interaction is a key determinant of viral host range, tissue tropism, and pathogenesis, and influences whether membrane fusion occurs at the plasma membrane or after uptake by cellular endocytic mechanisms. The enveloped alphaviruses include emerging viral pathogens such as chikungunya virus, which has caused recent outbreaks in India, and well-characterized viruses such as Sindbis virus and Semliki Forest virus (Kuhn, 2007). Alphaviruses are disseminated by mosquito vectors and replicate efficiently in a wide variety of insect and vertebrate cells. They contain a nucleocapsid core with the positive-sense RNA genome, surrounded by a lipid membrane with a lattice work of heterodimers of the transmembrane E1 and E2 proteins. Alphaviruses enter cells by clathrin-mediated endocytosis (Kielian et al., 2010). The low pH of the endosome triggers conformational changes in the E1 protein that drive the fusion of the viral and endosome membranes.

Alphavirus-receptor binding is mediated by the E2 protein (Kielian et al., 2010; Kuhn, 2007). Early studies showed that virus binding to the cell surface is saturable and sensitive to protease digestion of the target cells. The ability of the virus to infect both insect and vertebrate cells suggests either the existence of multiple receptors or of a broadly distributed, conserved molecule. Proteins such as the high-affinity laminin receptor can promote infection, but it is not clear whether they mediate endocytic uptake or serve primarily to increase initial virus attachment. Alphavirus infection can be increased by heparan sulfate binding but this effect is due to viral adaption to cell culture and is not observed for circulating wild type viruses. Virus produced in insect cells also attaches to DC-SIGN and L-SIGN, which may aid in transmission between insect and mammalian hosts. Thus, the identity of the receptor(s) for alphaviruses has remained outstanding.

To search for host factors involved in Sindbis virus infection, Rose and colleagues (Rose et al., 2011) carried out a genome-wide RNAi screen in Drosophila cells. Gene expression was knocked down using a dsRNA library, cells were infected with Sindbis virus, and infection was scored by microscopy. Nine genes encoding transmembrane proteins promoting infection were identified, but only one was conserved, ubiquitously expressed, and reported to localize to the plasma membrane. This protein was Natural Resistance-Associated Macrophage Protein (NRAMP). The authors found that Drosophila (d)NRAMP strongly promoted the infection of Drosophila cells by both lab-adapted and pathogenic strains of Sindbis virus but did not affect infection by the flavivirus West Nile virus or the rhabdovirus vesicular stomatitis virus. Infection by the alphavirus Ross River virus was also independent of dNRAMP, and chimera studies demonstrated that the envelope proteins were responsible for the Sindbis NRAMP requirement. Flies homozygous for a loss-of-function dNRAMP allele showed lower Sindbis virus infection than controls. The NRAMP requirement could be bypassed by direct transfection of Sindbis virus RNA into cells, suggesting that the block to infection was at the entry step.

dNRAMP is a member of a family of proteins that are conserved from bacteria to humans (Nevo and Nelson, 2006). These proteins are predicted to have 12 transmembrane domains with the N- and C-termini in the cytoplasm. Eukaryotic NRAMPs are involved in proton-coupled transport of divalent metal ions including iron across the plasma membrane and the internal phagosome or endosome membranes. There are two NRAMP genes in mammals. NRAMP1 is expressed mainly in the phagosomal membrane of macrophages while NRAMP2 (also known as DMT1, DCT1, SLC11A2) is ubiquitously expressed at the plasma membrane and in endosomes. NRAMP2 expression is regulated by high iron concentrations, which decrease mRNA levels (Eisenstein, 2000) and also cause internalization and degradation of the cell surface protein (Foot et al., 2008). Rose et al. used this sensitivity to iron as a tool to down-regulate NRAMP expression in both mosquito and mammalian cells. Infection by several Sindbis virus isolates was attenuated in iron-treated mosquito and mammalian cells, and binding was reduced in iron-treated U2OS cells, suggesting that NRAMP is also a receptor in both mosquitoes and mammals. As a more direct test, deletion of NRAMP2 in mouse embryo fibroblasts decreased primary virus infection by up to 50-fold.

Definitive evidence that a protein acts as a virus receptor includes demonstration of direct binding between the relevant viral protein and the cellular partner, which can be a tall order for a multispan membrane protein such as NRAMP2. Here Rose et al. showed that NRAMP2 could be co-precipitated with Sindbis virus. The structure of alphavirus E2 has been recently defined and used to map the locations of several residues involved in receptor binding (Li et al., 2010; Voss et al., 2010). This information plus characterization of the extracellular regions of NRAMP2 may aid in future direct binding studies of E2 and NRAMP2.

The efficient fusion of Sindbis virus with protein-free liposomes supports a model in which a receptor mediates virus uptake and delivery to the endosome while low pH provides the critical fusion trigger. This may explain why the NRAMP2 requirement in mouse cells was not more stringent. In agreement with the known alphavirus entry pathway, NRAMP2 is internalized by clathrin- and dynamin-dependent endocytosis through a motif on the C-terminal cytoplasmic domain (Lam-Yuk-Tseung et al., 2005). The use of NRAMP trafficking mutants and virus-NRAMP colocalization studies can now provide important avenues to further characterize the role of the receptor in alphavirus entry and fusion.

The identification of NRAMP2 as a receptor for Sindbis virus suggests exciting directions for future studies. It will be very interesting to characterize the role of NRAMP in an in vivo mammalian system. This is complicated by the fact that NRAMP2 knockout mice die shortly after birth and that naturally-occurring functional mutations cause severe iron deficiency and anemia (Gunshin et al., 2005; Lam-Yuk-Tseung et al., 2005). It is possible that tissue-specific knockouts will allow studies of viral infection, tropism, and pathogenesis. Can the NRAMP1 protein serve as a receptor in some cell types? What is the importance of the observed isoforms of NRAMP2, and does iron availability play any role in Sindbis virus infection? Given the lack of alphavirus vaccines and antiviral therapies, it will be important to determine if a specific receptor-E2 interaction could be exploited for potential therapeutics. Receptors remain undefined for other important alphavirus and flavivirus pathogens. The Drosophila RNAi approach described here may prove useful for other viruses with a broad host range.