S Functions During Coronavirus Entry

To appreciate the unique characteristics of these S proteins, one must visualize their activities in the context of the infection cycle. We begin with virion binding to susceptible host cells. As mediators of virus attachment to cells, S proteins are set apart by their ability to evolve remarkably varied attachment specificities. Sialic acid, a ubiquitous component of cell-surface carbohydrate complexes, is a documented low-affinity ligand for porcine and bovine coronavirus spikes (Schultze et al., 1991). Aminopeptidase N (APN), a type II-oriented membrane glycoprotein found in abundance on respiratory epithelia, is a receptor for antigenic "group 1" coronaviruses (Delmas et al., 1992; Tresnan et al., 1996). Members of this antigenic cluster include human respiratory viruses such as human CoV 229E, as well as several devastating animal pathogens such as transmissible gastroenteritis virus of swine and infectious peritonitis virus of cats. CarcinoEmbryonic Antigen-related Cell Adhesion Molecules (CEACAMs), immunoglobulin-like type I-oriented membrane glycoproteins that are prevalent in the liver and gastrointestinal tract, serve as receptors for the prototype member of the antigenic "group 2" coronavirus mouse hepatitis virus (Dveksler et al., 1991; Godfraind et al., 1995). Receptors for group 3 coronaviruses, which include several bird viruses causing severe bronchitis in chickens and turkeys, are currently unknown.

High-resolution structures are predicted for APN (Sjostrom et al., 2000; Firla et al., 2002), and are actually known for CEACAM (Tan et al., 2002). Structural homologies between these two proteins are not readily apparent. Thus, the adaptation of coronaviruses to either receptor likely involves substantial remodeling of binding sites on S proteins. In this regard, it is important to remember that Apn or Ceacam receptor usage correlates with the antigenic and genetic relationships used to divide coronaviruses into groups (Siddell, 1995). Therefore, one can reasonably infer that S variations adapt viruses to particular receptor usage, that receptor usage dictates the ecological niche of infection, and that coronaviruses in distinct niches then evolve somewhat independently to create recognizable antigenic/ phylogenetic groups. Suggestions that the SARS-CoV constitutes the first member of a fourth coronavirus group (Marra et al., 2003; Rota et al., 2003) may imply that this pathogen has adapted some time ago to bind a novel receptor set apart from either Apn or CEACAM.

High-resolution crystallographic structures for coronavirus S proteins are not yet known. Therefore, one can only speculate about the detailed architecture of their receptor-binding sites. The S proteins are moderately amenable to protein dissection techniques in which expressed fragments are assayed for receptor-binding potential, and these studies have roughly localized the sites of receptor interaction on primary sequences ((Suzuki and Taguchi, 1996; Bonavia et al., 2003), see Figure 4.3). Current hypotheses suggest that, as coron-aviruses diverge into types with particular receptor specificities, amino acid changes are fixed into S proteins at putative receptor-binding sites (Baric et al., 1999). This may be the case; however, S protein variabilities are relatively complex, and while many strain differences cluster in amino-terminal regions where receptors are thought to bind (Matsuyama and Taguchi, 2002a; see Figure 4.3), several changes are also found outside of this area. This complex variability can be appreciated by recalling the multifunctional properties of the S proteins, which contain receptor-binding sites as well as the machinery necessary to fuse opposing membranes (Figure 4.2). For S proteins, this membrane fusion activity is not constitutive, but is (with few exceptions) manifest only after receptor binding. In part, complex variability in S proteins may reflect the fact that receptor-binding and membrane fusion processes are coupled during virus entry. Put another way, the S-receptor interaction releases energy that is then used to create the conformational changes leading to S-induced membrane fusions. This coupling of receptor binding with membrane fusion activity suggests that subtle strain-specific polymorphisms virtually anywhere in the large S proteins might affect either or both of these essential functions, and by doing so, alter the course of coronavirus entry into cells.

The mechanism by which coronavirus S proteins mediate membrane fusion, recently clarified in studies by Bosch et al. (2003), involves a process in which the proteins respond to target cell receptor binding by undergoing conformational change (Gallagher, 1997; Matsuyama and Taguchi, 2002; Zelus et al., 2003) (Figure 4.4A). Next, a currently unidentified hydrophobic portion of the protein termed the fusion peptide (FP) harpoons target cell membranes (Figure 4.4B). This is followed by irreversible conformational changes in which alpha-helical portions of the protein condense into helical bundles (Figure 4.4C), ultimately bringing opposing membranes into sufficient proximity to coalesce them together (Figure 4.4D). This is a well-documented mechanism by which several viral and cellular proteins catalyze membrane coalescence (Weissenhorn et al., 1999; Russell et al., 2001; Jahn, et al., 2003) and is now classified as a "class-1" type fusion reaction. Algorithms predicting

GROUP 1 (HCoV-229E)

APN $1:S2

1 MnI "i"

!

3=oo-

GROUP II (MHV)

X

1

1-1,350 a.a.]

NH;-| CEACAM ATTACHMENT "DPR"

SHI S 2

1 «m

5

fcoo-

GROUP III (IBV)

X

H

S1EIS2

I hS,

S

_fcoo-

[-1.200 a.a]

S1;S2

1 *» T

!

^coo-

HIV-1 env

X

[-»50 a.a.]

w

V-|_gp120

CD4 l^ini ""

!

gp41_)-C00-

Figure 4.3. Linear depictions of coronavirus spike glycoproteins. Spikes representing groups I-III and the SARS coronavirus are depicted. Scissors indicate proteolytic cleavage between S1 and S2 of spikes. #eptad repeat regions (HR) as indicated by Learn Coil-VMF and MultiCoil are indicated in S2 and are upstream of transmembrane (TM) spans and conserved cysteine-rich stretches (C). Group I spikes recognize aminopeptidase N (CD13) metalloprotease. Group II spikes bind to carcinoembryonic antigen-related cell adhesion molecule (CEACAM) receptors. Downstream of the CEACAM-binding domain for group II MHV lies a deletion prone region (DPR). Amino acid (a.a.) lengths are drawn approximately to scale and relative to the size of the HIV type I envelope (env) fusion glycoprotein. FP is the hydrophobic fusion peptide for HIV-I. Hydrophobic residues are present N-terminal to HR1 regions in coronaviruses.

Figure 4.4. Proposed mechanism for S-mediated membrane fusion based on models of class 1-driven viral fusion.

(A) Attachment of oligomeric spike S1 domains to receptor. Cylinders represent alpha-helical secondary structure.

(B) Arrows depict exposure and insertion of hydrophobic fusion peptides into target membrane, subsequent to displacement of S1. CC indicates formation of alpha-helical coiled coil. (C) Fold-back or collapse of S2 leading to membrane coalescence. S1 domains have been removed for clarity and may in fact be absent as S1 sheds from S2 during fusion activation (see text). (D) Formation of end-stage coiled-coil bundle, fusion pore, and subsequent expansion of pore.

Figure 4.4. Proposed mechanism for S-mediated membrane fusion based on models of class 1-driven viral fusion.

(A) Attachment of oligomeric spike S1 domains to receptor. Cylinders represent alpha-helical secondary structure.

(B) Arrows depict exposure and insertion of hydrophobic fusion peptides into target membrane, subsequent to displacement of S1. CC indicates formation of alpha-helical coiled coil. (C) Fold-back or collapse of S2 leading to membrane coalescence. S1 domains have been removed for clarity and may in fact be absent as S1 sheds from S2 during fusion activation (see text). (D) Formation of end-stage coiled-coil bundle, fusion pore, and subsequent expansion of pore.

secondary protein structure (Singh et al., 1999) suggest regions of alpha helicity (designated heptad repeat, or HR 1 and 2, see Figure 4.3) in all coronavirus S proteins, including SARS-CoV. Thus, it is generally agreed that the core fusion machinery for all coronaviruses is built in such a way as to catalyze a conserved class-1-type fusion reaction.

It is notable that the fusion module (FP, HR1, HR2, TM span) occupies only about 20% of the inordinately large coronavirus S proteins, and the functional relevance of all but a portion of the remainder is largely unknown. Given that receptor-binding sites are distant from membrane fusion machinery in the primary structures (Figure 4.3), a sensible speculation is that much of the S protein structure is involved in linking receptor binding to the activation of membrane fusion. This is, after all, a crucial coupling that controls the timing and location of virus entry; that is, viral S proteins undergo conformational changes and proceed irreversibly through the class-1-type membrane fusion reaction only when engaged by cellular receptors embedded into the target cell membrane. In considering the activation mechanism, presently available genetic data point toward noncovalent linkages between receptor-binding regions and the fusion machinery (Grosse and Siddell, 1994; Matsuyama and Taguchi, 2002). In the best-studied MHV system, CEACAM binding does cause N-terminal S regions to separate from C-terminal, integral-membrane fragments (Gallagher, 1997), in all likelihood revealing the fusion apparatus (Matsuyama and Taguchi, 2002). This is a process that is augmented by cellular protease(s) that cleave the MHV S proteins at a site between receptor-binding and fusion-inducing domains ((Stauber et al., 1993; Bos et al., 1995; see also Figure 4.3). Proteolytic cleavage likely increases overall S protein conformational flexibility and eases the constraints on exposure of the fusion module, allowing it to advance more readily through the "class-1" pathway (Figure 4.4). These findings are beginning to point toward therapeutic targets interfering with coronavirus entry, and further breakthroughs will likely come from detailed S protein structure determinations.

Was this article helpful?

0 0
Swine Influenza

Swine Influenza

SWINE INFLUENZA frightening you? CONCERNED about the health implications? Coughs and Sneezes Spread Diseases! Stop The Swine Flu from Spreading. Follow the advice to keep your family and friends safe from this virus and not become another victim. These simple cost free guidelines will help you to protect yourself from the swine flu.

Get My Free Ebook


Post a comment