Echoviruses and Coxsackie B Viruses That Use Human Decay-Accelerating Factor (DAF) as a Receptor Do Not Bind the Rodent Analogues of DAF

Journal of Infectious Diseases, Jan 2000

Many serotypes of echovirus (EV) and Coxsackie B virus (CBV) bind human decay-accelerating factor (DAF) and use it as a receptor for infection. Analogues for DAF have been isolated from mice and rats and characterized; these analogues have amino acid identities to human DAF of ∼60%. EV serotypes 3, 6′, 7, 11–13, and 29 and CBV serotypes 1, 3, and 5 caused hemagglutination of human erythrocytes but not rat or mouse erythrocytes, suggesting failure to bind rodent DAF. To confirm this evidence, radiolabeled viruses were incubated with transfected Chinese hamster ovary (CHO) cells that were abundantly expressing each type of DAF. Only cells that expressed human DAF bound virus. Although binding of EV and CBV was specific for human DAF, complement inhibition by DAF expressed in CHO cells was similar for each analogue.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://jid.oxfordjournals.org/content/181/1/340.full.pdf

Echoviruses and Coxsackie B Viruses That Use Human Decay-Accelerating Factor (DAF) as a Receptor Do Not Bind the Rodent Analogues of DAF

O. Brad Spiller 1 3 4 Ian G. Goodfellow 1 2 3 David J. Evans 1 2 3 Jeffrey W. Almond 0 1 3 5 B. Paul Morgan 1 3 4 0 Virology Group, School of Animal and Microbial Sciences, University of Reading , Reading , United Kingdom 1 Virus strains. Defined EV and CBV of different serotypes were the kind gift of Brian Megson, Public Health Laboratory Service , Colindale , UK. The viruses were propagated in human embryonal rhabdomyosarcoma (RD) cells (European Collection of Animal Cell Cultures [ECACC] , Porton Down , UK) . RD cells were prop- agated in Dulbecco's modified essential medium containing glu- tamine and penicillin/streptomycin and supplemented with 10% fetal calf serum (FCS) (both purchased from GibcoBRL , Paisley , UK) . Virus infections and propagation were done in the absence of FCS. The influenza strain , PR8, was a gift from Dr. W. S. Barclay , School of Animal and Microbial Sciences, University of Reading. Antibodies. Mouse monoclonal antibody (mAb) recognizing human DAF (MBC1) , mouse mAb anti-rat DAF (RDIII7) 2 Institute of Virology, University of Glasgow , Glasgow 3 Received 18 June 1999; revised 30 August 1999; electronically published 23 December 1999. Informed consent was obtained from serum donors, and no live animals were used in these experiments. Financial support: The Wellcome Trust (098700/z/98/z and 043400/z/95 /z) and the Medical Research Council (G9006199) 4 Complement Biology Group, Department of Medical Biochemistry, University of Wales College of Medicine , Cardiff 5 Present affiliation: Pasteur Me rieux Connaught, Marcy-L'E toile, France. Group, Dept. of Medical Biochemistry, University of Wales College of Medicine , Heath Park, Cardiff, CF4 4XX , United Kingdom ( SpillerB @cardiff.ac.uk) Many serotypes of echovirus (EV) and Coxsackie B virus (CBV) bind human decay-accelerating factor (DAF) and use it as a receptor for infection. Analogues for DAF have been isolated from mice and rats and characterized; these analogues have amino acid identities to human DAF of 60%. EV serotypes 3, 60, 7, 11-13, and 29 and CBV serotypes 1, 3, and 5 caused hemagglutination of human erythrocytes but not rat or mouse erythrocytes, suggesting failure to bind rodent DAF. To confirm this evidence, radiolabeled viruses were incubated with transfected Chinese hamster ovary (CHO) cells that were abundantly expressing each type of DAF. Only cells that expressed human DAF bound virus. Although binding of EV and CBV was specific for human DAF, complement inhibition by DAF expressed in CHO cells was similar for each analogue. - Viruses initiate infection by attaching to cell-surface receptor molecules; thus, expression of specific receptors is an important determinant of viral host range and tissue tropism. Echoviruses (EVs [enteric cytopathic human orphan viruses]) and Coxsackie B viruses (CBVs) are members of the Enterovirus genus, family Picornaviridae. The wide range of clinical sequelae associated with EV and CBV infection includes rashes, diarrhea, aseptic meningitis, respiratory disease, myocarditis, and nonspecific febrile illness. The range of clinical manifestations probably reflects virus-tissue tropisms that are mediated, at least in part, by differential use of receptors. Those EV serotypes that hemagglutinate human erythrocytes, including serotypes 3, 6, 60, 7, 1113, 19, 21, 24, 25, 29, 30, and 33, bind decay-accelerating factor (DAF; CD55) on the erythrocyte [1, 2]. DAF comprises four 60amino acid short consensus repeats (SCRs), a heavily O-glycosylated region, and a glycolipid anchor (reviewed in [3]). DAF plays a central role in regulating the complement system by dissociating the convertase enzymes responsible for cleaving C3. Other members of the Enterovirus genus also use DAF as receptor; Enterovirus 70, Coxsackie A virus serotype 21, and CBV serotypes 1, 3, and 5 all have DAF-binding capacity [4]. CBVs also use the Coxsackie adenovirus receptor (CAR), which can mediate permissiveness for all serotypes [5]. Isolated reports describe infection of rodent cells in vivo and in vitro with various EV serotypes, but the receptors on rodent cells are unknown [69]. DAF analogues have recently been identified in mice and rats [10, 11]. We here examine whether these rodent DAF analogues bind EV and other viruses that use human DAF as a receptor. We characterize species selectivity of the DAF-EV and DAF-CBV interactions to elucidate whether enterovirus infection of rodent cells might involve binding endogenous DAF. Materials and Methods rat monoclonal anti-mouse DAF (3D5) were all generated in house. Phycoerythrin (PE)conjugated goat anti-mouse immunoglobulin secondary antibody was purchased from DAKO Ltd. (Ely, UK), and fluorescein isothiocyanateconjugated rabbit anti-rat immunoglobulin secondary antibody was purchased from Sigma (Poole, UK). Hemagglutination studies. Human erythrocytes were collected into EDTA (final concentration, 15 mM) and washed 3 times with 50 vol of PBS. Rat and mouse erythrocytes in 15 mM EDTA were supplied by Biomedical Services at the University of Wales College of Medicine, Cardiff. A 0.5 % suspension of washed erythrocytes was made in PBS for hemagglutination assays. Virus dilutions were made in point-bottomed 96-well plates (Dynatech, Wirral, UK) by using 1 mL of 109 TCID50U/mL of sucrose-gradientpurified EV or CBV serotypes added to 100 mL of PBS. Equal volumes of serially diluted virus and erythrocytes were mixed and allowed to settle for 4 h on ice. Positive hemagglutination was apparent as a lack of a pellet in the bottom of the well, because agglutinated cells spread over the well base. EV serotypes 3, 60, 7, 9, 1113, and 29 and CBV serotypes 1, 3, and 5 were tested. Sequence data. All amino acid sequences used for alignments were identical to those available at the NIH National Center for Biotechnology Information web site (http://www.ncbi.nlm.nih .gov/). Accession numbers for each protein are as follows: rat DAF, AAC77438; glycosylphosphatidylinositol (GPI)anchored mouse DAF, Q61475; transmembrane mouse DAF, Q61476; rat Crry, AAA91821; and human DAF, P08174. (Crry, a rodent-specific complement regulator that is structurally related to DAF, was used as a control.) The boundaries for each SCR were defined in the sequence for human DAF. Transfected cells. Chinese hamster ovary (CHO) cells obtained from ECACC were transfected with the empty eukaryotic transfection vector pDR2EF1a with or without specific inserts, as described elsewhere [12]. Rat Crry cDNA (a gift from Dr. R. Quigg, University of Chicago) was cloned into pDR2EF1a by restriction digestion. Stable clones were generated by selection and propagation in the presence of 100 mg/mL hygromycin B. Cell-surface expression for CHO-transfected cells and erythrocytes was measured by flow cytometry (FACScalibre; BectonDickinson, Oxford, UK). Virus-binding studies. EV and CBV were metabolically labeled by propagating in RD cells maintained overnight in cysteine/methionine-free medium containing 0.5 MBq of [35S]-cysteine/methionine (Amersham, Little Chalfont, UK). Cell debris was removed by low-speed centrifugation (5 min at 1300 g), and the virus was separated from unincorporated radiolabel by centrifugation through a 30% sucrose cushion in 1 M NaCl and 20 mM Trisacetate (pH 7.5) with 0.1% bovine serum albumin (125,000 g, SW28 rotor, overnight, at 47C). The pelleted virus was resuspended in serum-free medium, particulate material was removed by centrifugation (1300 g, 10 min), and virus concentration was determined by TCID50 on RD cells. The virus was stored in aliquots at 2207C. Fifty microliters of each radiolabeled virus was counted to determine input radioactivity for EV7 (11,000 cpm), EV11 (24,000 cpm), EV12 (10,000 cpm), CB1 (54,000 cpm), CBV3 (39,000 cpm), and CBV5 (64,000 cpm). TCID50 values for all viruses were 108/ mL. The binding of each virus was tested for each transfected cell line by pelleting 107 EDTA-disaggregated cells (1000 g, 5 min) in ice-cold, FCS-free cell medium, then resuspending the cell pellet in 50 mL of radiolabeled virus. After a 2-h incubation on ice, unbound virus was removed by 3 3 1 mL washes in ice-cold, serum-free cell medium, and cell-bound radioactivity was measured by scintillation counting. Each condition was done in triplicate and repeated at least twice. Calcein release assay. CHO cells transfected with the GPIanchored forms of human, rat, or mouse DAF or empty vectors were subcultured into 24-well plates at a density of 105 cells per well. Cells were loaded with calcein (Molecular Probes, Portland, OR) as described elsewhere [11] and were sensitized with a heatinactivated rabbit polyclonal antiCHO antiserum. Calcein release was measured in a Denley wellfluor fluorimeter (Life Sciences International, Basingstoke, UK) after a 1-h incubation with 20% human serum diluted in complement fixation diluent (Oxoid, Basingstoke). All assays were done in triplicate. Results Hemagglutination of human, rat, and mouse erythrocytes by EV and CBV. Erythrocytes from humans, rats, and mice were incubated with purified EV serotypes 3, 60, 7, 9, 1113, and 29 and CBV1, 3, and 5. EV9 does not bind to DAF and is the only virus of those listed that did not hemagglutinate the human erythrocytes (data not presented). None of the viruses tested hemagglutinated mouse or rat erythrocytes (data not presented). Influenza virus, which binds to sialic acid residues, was used as a positive control and caused hemagglutination of human, mouse, and rat erythrocytes, demonstrating that hemagglutination of the rodent erythrocytes was possible. Binding of radiolabeled EV and CBV. We considered the possibility that EV and CBV might bind rodent DAF analogues with a low affinity, insufficient to cause hemagglutination of erythrocytes. To address this possibility, we transfected CHO cells with human, rat, or mouse DAF or rat Crry cDNAs by using a eukaryotic expression vector that drives expression at very high levels. Expression levels were compared for erythrocytes and transfected CHO cells, and a representative histogram is shown in figure 1A. The transfected cells expressed between 45- (human DAF) and 150-fold (GPI-mouse DAF) more antigen than the levels observed naturally on erythrocytes. Radiolabeled (35S) EV serotypes 7, 11, and 12 and CBV serotypes 1, 3, and 5 were incubated with the transfected cells on ice, and the amount of labeled virus that remained bound to the cells after washing was expressed as a percentage of input virus (table 1). EV 7, 11, and 12 and CBV 1, 3, and 5 all bound strongly to human DAF, but no specific virus binding to the cells transfected with rat or mouse DAF or rat Crry was detected. Complement-regulating activity of human, rat, and mouse DAF. Preliminary experiments performed with complement regulationblocking anti-human DAF antibodies to inhibit hemagglutination by EV suggested a close association between the region in human DAF mediating EV binding and complement regulation [13, 14]. We therefore examined whether the lack of an EV-binding site in rat and mouse DAF was associated with reduced capacity to inhibit human complement. CHO cells transfected with human, rat, or mouse DAF (GPI) were loaded with calcein, antibody-sensitized, and exposed to human serum. DAF from each of the 3 species tested protected CHO cells from lysis by human serum to similar degrees, indicating that each could regulate human complement (figure 1B). Discussion Investigations of the pathogenicity of EV serotypes 6 and 9 have used newborn mice as a model, indicating that these serotypes can infect mouse cells in vivo [6, 7]. In addition, clinical isolates of EV serotypes 11 and 25 have caused, respectively, endomyocarditis- and poliomyelitis-like symptoms in newborn mice [8]. No systematic studies of susceptibility to EV infections in rats have been done. One report found infectious EV30 in feces of rats inoculated with virus, suggesting that productive infection had occurred [9]. For CBV, a murine analogue for human CAR has been found that is capable of mediating permissive infections in transfected CHO cells [15]. Although some reports of in vitro rat myocyte infections are available, CBV infection of rats has not been thoroughly studied. Because of the importance of human DAF as a receptor for enteroviruses, we have investigated the ability of several serotypes of EV and CBV to bind rodent analogues of DAF. Mouse DAF (both transmembrane and GPI forms) and rat DAF share 54%58% amino acid identity with human DAF through the Table 1. Binding of radiolabeled Echovirus (EV) and Coxsackie B virus (CBV) to transfected Chinese hamster ovary (CHO) cells, expressed as percentages. Control or transfection NOTE. DAF, decay-accelerating factor; GPI, glycosylphosphatidylinositolanchored form of mouse DAF; TM, transmembrane form of mouse DAF. 107 cells were incubated for 2 h on ice with radiolabeled EV serotypes 7, 11, and 12 and CBV serotypes 1, 3, and 5. These virus stocks were approximately the same MOI; therefore, different input cpm values represented different efficiencies of label incorporation. Values are given as percentage of bound input virus and represent triplicate measurements made for two separate experiments. Control cells are CHO cells transfected with empty vector. Other than human DAFtransfected cells, no significant binding of radiolabeled virus was observed for mouse DAF, rat DAF, or rat Crrytransfected cells (P!.001). SCRs, although no obvious clustering of conserved amino acid sequence is apparent. Rodents also have an additional C3 regulator, Crry, not found in humans, that has DAF-like activity and is more widely expressed than DAF in both rats and mice. However, rat Crry has only 26.6% amino acid identity with human DAF and is therefore an unlikely candidate to serve as an EV or CBV receptor. Other investigators have shown that the DAF SCRs required for EV binding are the same as those identified as essential for complement regulation on a macroscopic level [13, 14]. While testing a panel of anti-human DAF mouse mAbs for ability to block the hemagglutinating activity of EV and the complementregulating activity of DAF, we found that mAbs that blocked EV hemagglutination also blocked complement-regulating activity (data not presented). Taken together, these data suggest that sites essential for virus binding and complement regulation in human DAF are closely related within SCR 3. Because neither mouse nor rat DAF bound EV, it was of interest to examine their capacity to regulate human complement. Transfection of CHO cells with mouse, rat, or human DAF resulted in similar levels of protection from complement, making it unlikely that the regions in DAF required for complement regulation are identical to the sites bound by EV or CBV. It will be interesting to examine in domain-swapping studies whether human SCR 3 is sufficient to confer the capacity to bind enteroviruses on rodent DAF analogues. Acknowledgments We thank Drs. Claire Harris and Steward Hinchliffe for provision of the cDNA for mouse and rat DAF, respectively. Members of the Complement Biology Group are thanked for provision of antibodies.


This is a preview of a remote PDF: https://jid.oxfordjournals.org/content/181/1/340.full.pdf

O. Brad Spiller, Ian G. Goodfellow, David J. Evans, Jeffrey W. Almond, B. Paul Morgan. Echoviruses and Coxsackie B Viruses That Use Human Decay-Accelerating Factor (DAF) as a Receptor Do Not Bind the Rodent Analogues of DAF, Journal of Infectious Diseases, 2000, 340-343, DOI: 10.1086/315210