Spike Protein Fusion Peptide and Feline Coronavirus Virulence

Hui-Wen Chang, Herman F. Egberink, Rebecca Halpin, David J. Spiro, and Peter J.M. Rottier

Author affiliations: Utrecht University, Utrecht, the Netherlands (H.-W. Chang, H.F. Egberink, P.J.M. Rottier); and J. Craig Venter Institute, Rockville, Maryland, USA (R. Halpin, D.J. Spiro)

Abstract

Coronaviruses are well known for their potential to change their host or tissue tropism, resulting in unpredictable new diseases and changes in pathogenicity; severe acute respiratory syndrome and feline coronaviruses, respectively, are the most recognized examples. Feline coronaviruses occur as 2 pathotypes: nonvirulent feline enteric coronaviruses (FECVs), which replicate in intestinal epithelium cells, and lethal feline infectious peritonitis viruses (FIPVs), which replicate in macrophages. Evidence indicates that FIPV originates from FECV by mutation, but consistent distinguishing differences have not been established. We sequenced the full genome of 11 viruses of each pathotype and then focused on the single most distinctive site by additionally sequencing hundreds of viruses in that region. As a result, we identified 2 alternative amino acid differences in the putative fusion peptide of the spike protein that together distinguish FIPV from FECV in >95% of cases. By these and perhaps other mutations, the virus apparently acquires its macrophage tropism and spreads systemically.

Coronaviruses (subfamily Coronavirinae, order Nidovirales) are enveloped, plus-strand RNA viruses that infect mammals and birds. They are quite common and cause infections in humans and a wide variety of animals; infection typically results in respiratory or enteric disease. Severe acute respiratory syndrome coronavirus (SARS-CoV), which emerged suddenly in 2002 and caused severe acute respiratory disease in humans, is the most notorious coronavirus. SARS-CoV spread rapidly around the globe, infecting thousands and killing ≈800 persons. The virus presumably originated from bats and was transmitted to humans either directly or by using civets or raccoon dogs as intermediate hosts (1,2).
SARS-CoV best illustrates the remarkable potential for CoVs to change their tropism. Tropism switching has been implicated in the zoonotic emergence of human coronavirus OC43 from a bovine coronavirus and in turning transmissible gastroenteritis virus, an enteric porcine coronavirus, into porcine respiratory coronavirus, a respiratory pathogen (3,4). Such changes can be accompanied, although not necessarily, by cross-species transmissions; thus, the erratic occurrence and unpredictable new disease manifestations of tropism switching are a matter of public health concern.
The feline coronaviruses (FCoVs) present an example of pathogenetic change apparently associated with tropism switching. These viruses occur as 2 pathotypes with an enigmatic, even controversial, relationship: the low-virulence or nonvirulent feline enteric coronavirus (FECV) and the highly lethal feline infectious peritonitis virus (FIPV). FECV and FIPV are considered independently circulating viruses by some investigators (5,6). However, accumulating evidence supports the mutation hypothesis, which proposes that FIPV evolves from FECV by mutation in individually infected cats (7–12). A responsible mutation(s) has not been identified to back this hypothesis.
FECV is ubiquitous and spreads efficiently by the fecal-oral route; hence, seropositivity among cat populations can reach 90%, depending on the field conditions (13). The infection is restricted to the enteric tract, where the virus replicates in epithelial cells lining the gut mucosa. FECV infection is mild, causing transient enteritis that often passes unnoticed. The infection cannot be cleared efficiently by the immune system and thus persists, often for weeks or months, sometimes even longer (14–18). In contrast, FIPV is rare, but the consequences of infection are devastating. FIPV infection causes a progressive systemic disease called feline infectious peritonitis. The disease affects many organs, usually inducing fatal immunopathologic disease characterized by disseminated pyogranulomas and severe inflammatory damage to serosal membranes.
In the past, sequence differences in several virus genes, including those encoding membrane and spike (S) structural proteins and the so-called group-specific proteins 3c and 7b, have been implicated in the FCoV virulence shift (5–7,10,11,19–22). However, none of these differences appeared to consistently correlate with disease phenotype. To establish a consistent cause for a virulence shift in FCoV, specifically the predominant serotype I FCoV, we sequenced the entire genome of several FECV and FIPV specimens and then concentrated on the most conspicuous region of consistent difference by collecting and sequencing additional FECV and FIPV samples.