A model for the early interactions of rotaviruses with MA104 cells

As a summary of the data presented here, we propose the following working model

(Fig. 5), which takes into account the currently available information:

(a) Wild type RRV interacts primarily with a SA-containing cell receptor through the VP8 domain of VP4. The identity of the SA-containing molecule has not been determined, although good candidates are ganglioside GM3 (Guo et al 1999, Rolsma et al 1998), or the SA present in the integrin molecules (see below). The SA-binding domain of VP8 is located between amino acids 93 to 208, with residues 155, and 188 to 190, having an important role in this function (Fig. 1) (Fiore et al 1991, Fuentes Panana et al 1995, Isa et al 1997).

(b) Subsequent to the initial interaction with SA, RRV interacts with a second cell receptor, most probably a2^1 integrin, through the DGE integrin-binding motif located in the VP5 subunit of VP4 (Zarate et al 2000b). The ability of

Rotavirus Vp4 Vp8 Vp5
SA-containing receptor C ^ 2D9-Ag

a2p1 integrin aVp3 integrin

Wa receptor lipid raft in the cell membrane the NA-resistant variant nar3 to interact directly with this integrin is likely to be the result of a slight different conformation of its VP4 protein, compared to that of the wtRRV protein (Mendez et al 1993, 1996). Although the present data clearly indicate the existence of two different interactions between wildtype RRV and the cell surface, it has not yet been established whether two cell molecules, or two sites in the same molecule (e.g. a2^1), interact with VP8 and VP5. The fact that in infection competition assays the wild-type and variant viruses compete with each other reciprocally, suggests that if it is not the same cellular entity, the two cell molecules must be in close proximity.

(c) Integrins a4^1 and aX^2 have been implicated in rotavirus cell infection (Coulson et al 1997, Guerrero et al 2000b, Hewish et al 2000). The role of these integrins has not been determined yet, however, given that no additivity was observed when mixes of antibodies to these and other integrins were tested (Guerrero et al 2000b), they may represent alternative interaction sites for rotaviruses.

(d) The results obtained in the infection competition assays described above suggest that HRV Wa initially attaches to a cell surface molecule that is used by RRV and nar3, in a subsequent step after their interaction with a2^1 integrin. It can not be ruled out, however, that the attachment receptor for Wa is not actually used by RRV and nar3, but that HRV Wa interferes with the infectivity of these viruses by binding to a molecule that might be located in close proximity to either a2^1, or to co-receptor aV^3. The cellular molecule used by HRV Wa to bind to cells has not yet been characterized, although ganglioside GM1 seems to be a good candidate (Guo et al 1999).

FIG. 5. A model for the early interactions of rotaviruses with MA104 cells. Wild type RRV interacts primarily with a SA-containing cell receptor through the VP8 domain of VP4. Subsequent to this initial interaction, which might induce a conformational change in VP4, the virus interacts with a second, NA-resistant cell receptor, here proposed to be a2;81 integrin. This interaction is through the DGE binding motif of VP5, present at aa 308—310. This second virus—cell interaction might facilitate a third interaction of the virus with P3 integrin. The SA-independent variant nar3 is proposed to interact directly, through VP5, with the a2;81 integrin. For the sake of clarity, the SA-containing and NA-resistant cellular receptors are depicted here as two separate entities (the first possibly being ganglioside GM3), however, they could be two domains of the same receptor molecule (see text). The nature of the attachment site for the HRV strain Wa has not been determined, however, we propose that it binds to a molecule that is in close proximity to a2;81, probably GM1 (see text). The antigen recognized by MAb 2D9 (2D9-Ag) has not been identified, but we assume that it should also be close to the a2^1 integrin (see text). After their initial contact with the cell, all three rotavirus strains are proposed to interact with P3 integrin, this interaction might mediate the penetration of the viruses into the cell's interior. In this model most, if not all, of the molecules involved in rotavirus binding and entry are proposed to form a complex, probably embedded in glycosphingolipid-enriched lipid microdomains on the cell surface.

Also, the viral protein domain responsible for this interaction has not been determined.

(e) We have found that integrin aV^3 plays an important role for infection of all three rotavirus strains at a post-attachment step, most likely penetration; however, the precise function of this protein has yet to be characterized.

(f) The essential components of the glycosphingolipid-enriched membrane domains, termed rafts, are glycoproteins, glycosphingolipids and cholesterol. Since these three components have been found to be important for the initial steps of rotavirus infection (Guerrero et al 2000a), and aV^3 integrin has been observed to be present in rafts (C. A. Guerrero, S. Lopez & C. F. Arias, unpublished data, Green et al 1999), we propose that some or all of the various virus-cell interactions described above might take place in these lipid microdomains.

The data presented here are consistent with the existence of several rotavirus receptors which might be tightly organized, maybe forming a complex in glycosphingolipid-enriched rafts. The requirement for several cell molecules to be present and organized in a precise fashion, might explain the exquisite cell and tissue tropism of these viruses. It remains to be established which, if any, of the receptor molecules described so far is indeed non-replaceable, and if in fact there exists a unique pathway of infectivity for rotaviruses, with distinct entry points for different strains.

In conclusion, much remains to be learned about the process of binding and penetration of rotaviruses. The characterization of the nature of the interactions that occur between the cellular and viral partners, and the signal transduction pathways potentially triggered by the early virus—cell contacts, should give insight into the elaborated mechanism used by these viruses to enter cells.

A cknowledgements

This work was partially supported by grants 75197-527106 and 75301-527108 from the Howard Hughes Medical Institute, G0012-N9607 from the National Council for Science and Technology-Mexico, and IN201399 from DGAPA-UNAM.

0 0

Post a comment