| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 John Curtin School of Medical Research, Division of Immunology and Genetics, Australian National University, PO Box 334, Canberra, ACT 2600, Australia
2 Department of Microbiology, Monash University, Clayton, VIC, Australia
3 Department of Cellular and Molecular Medicine, University of Bristol, Bristol BS8 1TD, UK
Correspondence
Eva Lee
Eva.Lee{at}anu.edu.au
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Published online ahead of print on 20 June 2006 as DOI 10.1099/vir.0.82164-0.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Esko & Selleck, 2002). They display an enormous structural diversity and have a critical role in a wide range of biological processes (see Bernfield et al., 1999; Esko & Selleck, 2002; Handel et al., 2005; Lindahl et al., 1998; Perrimon & Bernfield, 2000; Sasisekharan et al., 2002). Numerous viruses utilize GAG for attachment to host cells, although subsequent infection may require virus association with additional cellular receptors (reviewed by Rostand & Esko, 1997; Spillmann, 2001). Virus binding to GAG involves primarily the electrostatic interaction of clusters of basic amino acids on the virion surface with negatively charged sulfate groups on the polysaccharide. The crystal structure of a foot-and-mouth disease virus–heparin complex (Fry et al., 1999) and computer simulation of heparin docking to the adeno-associated virus capsid protein (Kern et al., 2003) suggest that GAG binds along shallow depressions on the virions in which the GAG-binding determinants are situated, whilst a cryo-electron microscopy investigation on heparin association with a GAG-binding mutant of Ross River virus showed that binding was mediated by the distal end of the E2 spike proteins (Zhang et al., 2005).
Virus affinity for GAG is a determinant of cell tropism, and the acquisition of additional positive charge at GAG-binding sites is a common adaptation of viruses grown in cell culture (reviewed by Spillmann, 2001). However, our studies on the encephalitic flaviviruses, Murray Valley encephalitis virus (MVE), Japanese encephalitis virus (JEV) and West Nile virus (WNV) (Lee et al., 2004; Lee & Lobigs, 2002), and the work of others on Foot-and-mouth disease virus (Sa-Carvalho et al., 1997), Sindbis virus (Byrnes & Griffin, 2000), Venezuelan equine encephalitis virus (Bernard et al., 2000) and Classical swine fever virus (Van Gennip et al., 2004) have demonstrated that the increased infectivity in cell culture of host cell-adapted viruses is a disadvantage for virus growth in vivo, because the high GAG-binding affinity prevents efficient virus spread through the circulation. The loss of virulence due to high GAG-binding affinity raises the question as to whether, and to what extent, this mechanism also accounts for the virulence attenuation of live vaccines or vaccine candidates. In case of a live JEV vaccine widely used in China (strain SA14-14-2), high GAG-binding affinity contributes, at least in part, to its attenuated virulence phenotype (Lee & Lobigs, 2002).
The interaction of Dengue virus (DEN) with GAG was first shown by Chen et al. (1997) who also identified two putative GAG-binding regions, one spanning domains I and III and a second on domain III on the crystal structure of the viral envelope (E) protein (Modis et al., 2003; Rey et al., 1995). DEN is a mosquito-borne flavivirus causing a high incidence of human disease, ranging from fever to life-threatening haemorrhagic fever and shock, in tropical and subtropical countries (reviewed by Burke & Monath, 2001). Others have confirmed the putative involvement of cell surface GAG in attachment to and infection of host cells by DEN for each of the four serotypes (DEN-1–DEN-4) (Germi et al., 2002; Hilgard & Stockert, 2000; Hung et al., 1999, 2004; Lee et al., 2006; Lin et al., 2002; Thullier et al., 2001) and the antiviral activity against DEN of heparan sulfate mimetics has been shown in vitro and in vivo (Lee et al., 2006; Marks et al., 2001; Ono et al., 2003; Talarico et al., 2005). The molecular GAG-binding determinants of DEN and the role of GAG-binding affinity in DEN virulence have, so far, not been defined and these topics have been investigated in this study.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2006). Working stocks of DEN-2 strains NGC, isolated in 1944, passaged 24 times in suckling mouse brain, four times in equine kidney (EK) cells and twice in C6/36 cells (Gruenberg et al., 1988), and PUO-218, isolated from Thailand in 1988 from patient sera, passaged once in Toxorhynchites splendens mosquitoes, once in monkey kidney (LLC-MK) cells and twice in C6/36 cells, were C6/36 cell supernatants.
Serial passage of DEN-2 viruses in SW13 and BHK-21 cells.
SW13 or BHK-21 cell monolayers in 60 mm dishes (1x106 and 5x105 cells, respectively) were infected with DEN-2 NGC or PUO-218 strains at an m.o.i. of 0.1 Vero p.f.u., and supernatants were collected and supplemented with HEPES (pH 8.0; 20 mM final concentration) at 5 days post-infection (p.i.). Four subsequent passages were performed for each passage series using 0.2 ml of the corresponding supernatants and harvesting culture medium following appearance of cytopathic effects (cpe) or at 5 days p.i. After five passages, plaque-purified stocks were prepared as described previously (Lee et al., 2004) by amplification in BHK-21 cells.
Sequence analysis.
Total cellular RNA from infected BHK-21 cells (for passage variants) or infected Vero cells (for parent viruses) was extracted using Trizol (GibcoBRL) as described previously (Lee et al., 2004). RT-PCR was performed as described previously (Lee & Lobigs, 2002) using random hexamers in the reverse transcription (RT) step and the DEN-2-specific primer pair, p2425: 5'-CAGCTCACAACGCAACCACT-3' and p400: 5'-GCAGGCATGATCATTATGCT-3', for PCR. PCR-amplified cDNA fragments were gel-purified and used in cycle sequencing with ABI Big-Dye v3 chemistry (Applied Biosys). Additional primers for sequencing of DEN-2 prM and E genes were P787: 5'-TGGAAACATGCCCAGAGAAT-3', P1567: P5'-CAATGGTTCCTAGACCTGCC-3' and P1669: 5'-ACAACATCCTGTTTCTTCGC-3'.
Plasmids.
Plasmid pDVWS601 contains the full-length cDNA of the DEN-2 NGC genome flanked by a T7 RNA polymerase promoter at the 5' end and an XbaI restriction site at the 3' end (Gualano et al., 1998). Viral RNA from parental virus and passaged variants of DEN-2 PUO-218 and NGC strains was reverse-transcribed as described above and cDNA encoding the E protein was amplified by PCR using primers P787 and P2863: 5'-TTCCAGCGAATTCCAAGCTCT-3'. To introduce changes at E protein residue 227 or 249, a 1 kb region flanked by SphI and NheI sites in pDVWS601 was replaced with SphI–NheI fragments derived from RT-PCR-amplified cDNA of SW13 cell-passaged or BHK-21 cell-passaged NGC, respectively. To replace the DEN-2 strain NGC prM and E protein genes in pDVWS601 with those from PUO-218, two fragments from pDVWS601, a 1.5 kb fragment flanked by NotI (upstream of the T7 RNA polymerase promoter) and HindIII (nt 1547) sites and a 1.3 kb fragment flanked by SphI (nt 1380) and PstI (nt 2731) sites were subcloned into vector plasmids pACdel, in which the BamHI–BanI fragment from pACYC174 (New England Biolabs) had been replaced with the polylinkers from pBluescriptKS+ (Stratagene) or pGEM-5Zf (Promega), respectively, to generate plasmids pACprME5' and pGEME3'. Regions flanked by BclI (nt 407) and SphI (nt 1380) or by SphI and SpeI (nt 2370) in pACprME5' and pGEME3', respectively, were replaced with the corresponding cDNA fragments derived from PUO-218 viral RNA, or those derived from SW13 or BHK-21 cell-passaged PUO-218 to introduce the PUO-218 prM and E genes with or without coding changes to the E aa 120, 124 or 202 into the DEN-2 NGC genome. Full-length DEN-2 plasmids were generated by ligating three DNA fragments: a NotI–SphI fragment from pACprME5' derivatives, an SphI–NheI fragment from pGEME3' derivatives and a NotI–NheI fragment from the pDVWS601 plasmid. All plasmid constructs were verified by sequencing the entire prM and E protein genes.
RNA transcription and electroporation of BHK-21 cells.
Plasmids containing the full-length DEN-2 cDNAs were digested with XbaI, transcribed using T7 RNA polymerase and full-length RNA transcripts electroporated into BHK-21 cells as described previously (Lee & Lobigs, 2000) with the following modifications. For determination of electroporation efficiency, 200, 20 or 2 µl of electroporated cells (25 ml at 4x105 cells ml–1) were plated onto Vero cell monolayers in six-well trays (3.5x105 cells per well; Nunc), incubated for 1 h and agar overlay medium (M199 medium plus 1 % agar and 2 % fetal calf serum) added. Neutral red (0.02 %) was added at 4 days p.i. and plaques counted after overnight incubation.
Heparin–Sepharose binding and heparin inhibition assays.
Binding of DEN-2 viruses (105–106 p.f.u.) to heparin–Sepharose beads was assayed as described previously (Lee et al., 2004). Inhibition by heparin of infectivity of DEN-2 viruses was examined by incubating virus (1–5x106 p.f.u.) with heparin (0.2, 2, 20 or 200 µg ml–1) for 15 min prior to its addition to BHK-21 cell monolayers (4x105 cells per well in six-well plates) treated with similar concentrations of heparin. Mock-treated controls were performed by preincubation of the same amount of virus in Hanks' balanced salts solution (HBSS)-BSA in the absence of heparin. Cells were harvested by trypsinization at 21 h p.i. and the percentage of infected cells determined by flow-cytometry as described previously (Lee et al., 2006).
Specific infectivity determination.
Specific infectivity of DEN-2 virus particles was determined as described previously (Lee et al., 2004) with the following modifications. Infectivity of virus, released from infected BHK-21 cells over a 2 h incubation period, on BHK-21, Vero and SW13 cells was determined by TCID50 assay in 96-well trays using NS1 protein-specific mAb 4G4 for the detection of virus-infected cell monolayers by ELISA (Licon Luna et al., 2002). Viral RNA in the infected culture supernatant (100 µl) was extracted for quantification of virion RNA content by competitive RT-PCR. The internal standard was a DEN-2 competitor RNA (DEN2-cRNA). This was generated by in vitro transcription using T7 RNA polymerase (Promega) of a recombinant pBKS+ (Invitrogen) plasmid containing the DEN-2 NGC genome from nt 400 to 2444 (Gruenberg et al., 1988), but with an insertion of a 
100 bp DNA fragment from the polylinker region of pBluescriptKS+ at the MunI site in the E protein gene (nt 2092). Plasmid DNA linearized at the ApaI site at the 3' end of the DEN-2 insert was used as template for T7 RNA polymerase and a trace amount of [3H]UTP was added for determination of the yield of DEN-2-cRNA transcript (Sambrook et al., 1989). RT of mixtures of viral RNA and serially diluted DEN-2-cRNA was performed using the antisense primer P2425, followed by PCR using primer P2425 and the sense primer P1567 as described previously (Lee et al., 2004). PCR products were electrophoresed in 1.3 % agarose gels. The concentrations of DEN2-cRNA spiked into PCR reactions in the presence of a constant amount of virion RNA at which two PCR product DNA bands of 878 and 978 bp of similar intensity were visible allowed estimation of the concentration of the virion RNA.
Mouse experiments.
Six-week-old alpha and gamma interferon receptor double knockout (IFN-
/
-R–/–) mice (van den Broek et al., 1995) were used for virulence assays. Virus inocula (100 µl), diluted in HBSS-BSA, were injected by the intraperitoneal (i.p.) or intravenous (i.v.) routes. For intracerebral (i.c.) inoculation, the virus inoculum (30 µl) was given to mice anaesthetized with ketamine/xylazol using a 26G needle. Mortality and morbidity were monitored over a period of 28 days. Clearance of virus from the circulation was assayed as described previously (Lee et al., 2004).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), and the natural isolate, PUO-218, were serially passaged five times in SW13 or BHK-21 cells. The same approach was employed previously for the selection of GAG-binding variants of JEV serotype flaviviruses and Tick-borne encephalitis virus (TBE) (Goto et al., 2003; Lee et al., 2004; Lee & Lobigs, 2002; Lobigs et al., 1990; Mandl et al., 2001). Tissue culture supernatants were harvested after each passage between 4 and 6 days p.i., or when cpe was detectable. By the fifth passage, significant cpe was observed at 3 days p.i. in both BHK-21 and SW13 cells. Sequence analysis was performed on virion RNA extracts from uncloned as well as four plaque-purified virus stocks from each passage series. There was at least 1 nt change in the E protein gene of each passage variant, which resulted in a non-conservative amino acid change (Table 1). Notably, the E protein changes in each of the passaged variants resulted in the net gain of positive charge consistent with enhanced GAG binding. The mutations appeared to be dominant in the fifth passage virus stocks, as sequence analysis of uncloned variant stocks did not show the parental sequence at the corresponding positions in the virus populations. BHK-21 cell-passaged PUO-218 comprised two genotypes: PUOK120, which represented the majority population on the basis of sequenced uncloned virus and PUOD124/K202, which showed a minor presence.
|
; Lee et al., 2004; Lee & Lobigs, 2000, 2002; Mandl et al., 2001). To investigate further the generality of altered virus–host receptor interaction(s) among the flaviviruses due to adaptation to the two cell lines, virus binding to heparin–Sepharose was measured. Heparin is a highly sulfated reference GAG. The PUO-218 strain did not show detectable binding to heparin–Sepharose beads, while SW13 and BHK-21 cell-passaged PUO-218 variants showed greater than 97 % binding (Table 1
). Interestingly, the parental DEN-2 NGC strain bound efficiently to heparin–Sepharose. Nevertheless, the binding of NGCR227 and NGCN249 was slightly increased relative to the parent (Table 1).
Heparin-binding phenotypes of DEN-2 NGC and NGC/PUO-218 prM–E chimeric infectious cDNA clone-derived viruses with or without host cell adaptive mutations
To test the impact of single amino acid changes at PUO-218 strain E protein residues 120, 124 and 202 on heparin-binding affinity, the prM and E protein coding regions of PUO-218 with or without these host cell adaptive mutations were inserted into the full-length infectious cDNA clone of NGC (Gualano et al., 1998); the NGC/PUO-218 prM–E chimeric viruses are denoted rPUO, rPUOK120, rPUOD124 and rPUOK202. Virus rPUO closely resembled the PUO-218 strain in the lack of detectable binding to heparin–Sepharose (Fig. 1). Viruses rPUOK120 and rPUOK202, which differ from rPUO by the gain of a single positive charge in the E protein, showed >99 % binding to heparin–Sepharose. In contrast, the Asn124
Asp substitution in the E protein of rPUOE124 did not increase binding to heparin–Sepharose relative to that of rPUO. This mutation was selected in addition to the Glu202Lys change during passage of PUO-218 in BHK-21 cells.
|
15 %) relative to rNGC (Fig. 1). Accordingly, Lys at E protein residue 126 is also a determinant of GAG affinity.
The already high GAG-binding affinity of NGC may obscure detection of further enhancement of GAG binding by other determinants. To corroborate that Arg at E protein residue 227, which was selected during passage of NGC in SW13 cells, is a major GAG-binding determinant, binding of rNGCE126 and rNGCE126/R227 to heparin–Sepharose was compared. The presence of a positive charge at residue 227 in variant rNGCE126/R227 significantly improved binding relative to rNGCE126 (Fig. 1).
Effect of GAG-binding determinants on susceptibility of virus infectivity to heparin
We next investigated whether the DEN-2 GAG-binding determinants influenced dependence on cell surface GAG during virus entry. For this assay, only DEN-2 NGC and NGC/PUO-218 prM–E chimeric infectious clone-derived viruses were used. Virus and BHK-21 cells were pre-incubated with heparin at given concentrations followed by infection and quantification of infected cells by flow-cytometry (Lee et al., 2006). The infectivity of rNGC was reduced by more than 50 % at heparin concentrations of 
2 µg ml–1 (Fig. 2a), consistent with the previously determined 50 % effective concentration value of heparin for DEN-2 NGC using this assay (Lee et al., 2006). The presence of an additional positive charge in the E protein of rNGCR227 gave rise to a marked increase in heparin susceptibility, while the Asp249Asn change in rNGCN249 consistently resulted in 10 % greater inhibition compared with rNGC at heparin concentrations between 2 and 200 µg ml–1 (Fig. 2a).
|
). Notably, rPUOD124 displayed reduced susceptibility to heparin inhibition relative to rPUO (Fig. 2b), consistent with its lack of binding to heparin–Sepharose. Thus, the increase in negative charge on the viral surface of this variant appears to reduce further the relatively low GAG-binding affinity of PUO-218.
To investigate the role in attachment/entry of E protein residue 126, the susceptibility of infectivity to heparin inhibition of rNGC, rNGCE126, rPUO and rPUOK126 was compared. The results show that the rNGCE126 and rPUO, and rPUOK126 and rNGC pairs are comparable in regard to their moderate and high susceptibility to heparin inhibition, respectively (Fig. 2b), and confirm the importance of residue 126 as a GAG-binding determinant.
Influence of GAG-binding determinants on specific infectivity of DEN-2 for BHK-21, SW13 and Vero cells
The impact of E protein substitutions at GAG-binding sites on specific infectivity of DEN-2 for BHK-21, SW13 and Vero cells was examined using infectious clone-derived viruses collected over a 2 h period at 28 h p.i. The virus harvests were used for infectivity titration in Vero, BHK-21 and SW13 cells by TCID50 assay and RNA quantification by RT-PCR (Lee et al., 2004), and specific infectivity was determined as genome equivalent/TCID50 ratio. Gain-of-positive-charge mutations at E protein residues 120, 202, 227 and 249 increased infectivity for BHK-21 and SW13 cells by 10- to 170-fold, whereas that for Vero cells was comparable or slightly poorer (<10-fold) relative to the parental strains (Table 2). The Asn124Asp change, which contributed little to heparin binding (see above), resulted in a moderate enhancement (<10-fold) of specific infectivity for BHK-21 and SW13 cells, without markedly influencing the specific infectivity for Vero cells (Table 2).
|
), suggesting that Lys126 was mainly responsible for the high infectivity of NGC for BHK-21 cells. Notably, a similar effect of residue 126 on the specific infectivity for SW13 cells (compare rNGC and rPUO; Table 2) was not observed.
Mouse virulence of DEN-2 NGC and host cell-adapted variants
Natural isolates of DEN do not cause mortality or overt morbidity in adult immunocompetent mice by extraneural routes of inoculation. However, mice deficient in types I and II interferon responses (IFN-/-R–/– mice) provide a suitable mouse model for assessment of virulence properties of DEN, given that they are susceptible to DEN-2 NGC infection by the i.p. route (Johnson & Roehrig, 1999), as well as to natural DEN isolates when inoculated at a high dose i.v. route (Shresta et al., 2004). To compare mouse virulence of NGC parent and passage variants, groups of 6-week-old IFN-/-R–/– mice were inoculated via the i.c. or i.p. route with 103 and 106 p.f.u., respectively. While the parental NGC virus caused 100 % mortality via both routes, NGCR227 was attenuated with no mortality observed by i.p. inoculation and significantly prolonged mean survival time (mst) relative to the parent following i.c. inoculation (Table 3a). NGCN249 was partially attenuated by the i.p. route, showing reduced mortality and prolonged mst but was equally virulent by the i.c. infection route in comparison to the parental NGC virus (Table 3a). All surviving mice were protected against i.p. challenge with 106 p.f.u. NGC given 4 weeks after primary infection, whilst challenged naïve IFN-/-R–/– mice at 10 weeks of age showed 100 % mortality (mst 18 days).
|
/-R–/– mice was further examined using NGC and NGC/PUO-218 prM–E chimeric infectious clone-derived viruses (Table 3b). Mice infected i.v. with 106 p.f.u. of rPUO showed 100 % mortality (mst=12.6 days); rPUOD124, which showed reduced susceptibility to heparin inhibition, was also uniformly lethal but slightly more virulent than rPUO according to the shorter mst. All GAG-binding variants except rPUOK202 were significantly attenuated relative to rPUO. Unexpectedly, the GAG-binding variant rPUOK202 gave 100 % mortality; however, with a significantly prolonged mst relative to rPUO. To test whether the apparent lack of virulence attenuation of rPUOK202 was the result of phenotypic instability, sequence analysis was performed on viral RNA derived from two individual rPUOK202-infected mouse brains. This showed complete reversion of Lys202Glu and implies that the second site mutation (Asp124) in the E protein of rPUOD124/K202 stabilizes the GAG-binding and virulence attenuation phenotypes produced by Lys202.
Interestingly, rNGC was less virulent than rPUO when inoculated by the i.v. route in IFN-/-R–/– mice (Table 3b). This was surprising, given that the PUO-218 parental strain was not lethal for these mice when inoculated i.p. (106 p.f.u.) and only gave 20 % mortality when the same dose was inoculated i.v. (data not shown). Thus, it appears that the prM and E proteins of PUO-218 in combination with additional regions in the NGC polyprotein and/or non-coding regions of the NGC genome enhance neuroinvasiveness relative to the PUO-218 and NGC strains. Furthermore, we showed by i.c. inoculation of 103 p.f.u. that the neurovirulence of rNGC and rPUO was comparable in IFN-/-R–/– mice: both strains gave 100 % mortality with mst of 8.0 and 8.8 days, respectively (Table 3b). The conflicting virulence profiles of rNGC and rPUO in IFN-/-R–/– mice may be attributed to their differential capacity to spread in extraneural tissues and, in turn, be a consequence of the differential GAG-binding properties of the two viruses. This hypothesis is further supported by virulence data for the rPUOK126 and rNGCE126 pair: mortality for the strongly GAG-binding virus, rPUOK126, following i.v. infection was significantly reduced relative to rPUO, whilst it was slightly increased for rNGCE126, which binds poorly to GAG, relative to rNGC. Notably, groups of mice infected with rNGCE126 showed significantly reduced mst relative to rNGC (15.3 versus 24.5 days; P=0.0004).
In addition to being a determinant of GAG-affinity and neuroinvasiveness, residue 126 also influences mouse neurovirulence (Gualano et al., 1998), since new-born BALB/c mice inoculated i.c. with the 103 p.f.u. of rNGC or rNGCE126 showed mortality of 100 and 80 % and significantly different mst of 7.4 and 10.4 days, respectively (P=0.008), while mice similarly infected with rPUO or rPUOK126 showed mortality of 0 and 100 % (mst of 8 days) (our unpublished results, which confirm those of Gualano et al., 1998). The neurovirulence phenotype mediated by the presence of a positive charge at residue 126 is unlikely to be due to increased GAG affinity because other GAG-binding variants (rPUOK120 and rPUOD124/K202) did not show greater mouse neurovirulence than rPUO when tested similarly (data not shown).
Rapid clearance of DEN-2 GAG-binding variants from the circulation
Our mouse virulence data showed a remarkably consistent correlation between GAG-binding affinity of DEN-2 viruses and virulence attenuation in IFN-/-R–/– mice following inoculation by an extraneural route. This raised the possibility that virulence attenuation resulted from the rapid clearance of GAG-binding variants from the circulation by non-productive binding to surfaces enriched in GAG, as has been demonstrated for JEV serogroup flaviviruses (Lee et al., 2004; Lee & Lobigs, 2002). To test this possibility, virus titres in serum samples collected over a 30 min period from mice inoculated, i.v., with a high dose of rPUO or rPUOK120 was determined. The rPUO virus titres did not show significant reduction over 30 min, while rPUOK120 showed 2 log reduction in serum titres over the same period (Fig. 3).
|
) representing five subtypes and isolated between 1944 and 1990, only substitutions that gave rise to an increase in positive charge were found in a small number of isolates at these residues, with most isolates displaying the low GAG-binding determinant. When mapped onto the crystal structure of DEN-2 E protein (Modis et al., 2003), residues 120, 126, 202, 227 and 249, without exception, cluster in the central region of the E protein dimer and are exposed on the upper face of domain II (Fig. 4).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Sabin, 1952, 1955; Sabin & Schlesinger, 1945; Schlesinger et al., 1956). Consistent with recent studies by us and others on the encephalitic flaviviruses MVE, JEV, WNV and TBE, GAG-binding variants selected by serial passage of DEN-2 in BHK-21 and/or SW13 cells displayed reduced mouse virulence by peripheral infection routes (Goto et al., 2003; Lee et al., 2004; Lee & Lobigs, 2000, 2002; Mandl et al., 2001). Accordingly, flaviviruses appear to universally engage with high affinity to cell surface GAG for efficient attachment/entry of some cell types, an adaptation that is detrimental for virus spread in the animal host.
The altered host cell tropism of DEN-2 GAG-binding variants was confirmed by specific infectivity determinations: gain-of-positive-charge mutations at E protein residues 120, 202, 227 and 249 significantly increased infectivity for BHK-21 and SW13 cells, whereas that for Vero cells was comparable or slightly poorer relative to the parental strains. Thus, distinct cell surface receptors appear to be involved in attachment/entry of DEN-2 during infection of Vero, BHK-21 and SW13 cells; the contribution of virus binding to GAG seems to be minimal for the efficient infection of Vero cells but is beneficial for that of BHK-21 and SW13 cells. The GAG-binding determinant at E protein residue 126 differed from the others in that a Lys at this residue markedly enhanced the specific infectivity for BHK-21 but not for SW13 cells. This may underscore the influence of the large structural diversity of cell surface GAG (Bernfield et al., 1999) on the specificity of virus–host cell interaction. Given that specific quantitative and qualitative changes in expression of cell surface GAG are a hallmark of metastatic carcinoma cells (Sasisekharan et al., 2002), it is possible that the marked change in cell tropism of DEN-2 and other flaviviruses selected for growth in SW13 cells (adrenal gland small cell carcinoma) relates to a characteristic cell surface GAG phenotype associated with malignant transformation. This raises the question whether the selective incorporation of GAG-binding determinants can improve the effectiveness of oncolytic viruses for clinical application.
The location of DEN-2 GAG-binding determinants on the E protein crystal structure was of interest, in particular in comparison to that of other flaviviruses belonging to different serocomplexes. Remarkably, residues 120, 126, 202, 227 and 249 are exclusively clustered on the exposed surface of domain II in the centre of the E protein dimer (Modis et al., 2003, 2005). This contrasts with the location of key GAG-binding determinants of the JEV serotype flaviviruses in two clusters on domain III (residues 306 and 389/390) and domain I (residues 59 and 138) (Lee et al., 2004; Lee & Lobigs, 2000, 2002), whilst adaptation of TBE to BHK-21 cells selected for GAG-binding variants altered at residues distributed throughout the upper surface across all three structural domains (Mandl et al., 2001). It was surprising that none of the DEN-2 GAG-binding determinants were localized to E protein domain III, given that this part of the protein is thought to function in receptor binding (Lobigs et al., 1990; Rey et al., 1995) and that the attachment of a soluble DEN-2 domain III to BHK-21 cells could be blocked by heparin (Hung et al., 2004). Comparative analysis of the crystal structures of DEN-2 and -3 E protein ectodomains identified two clusters of positively charged residues predicted to function in GAG binding, one in domain I and a second in domain II, but could not confirm the existence of GAG-binding residues in domain III (Modis et al., 2005). The GAG-binding determinants identified here cluster with the putative GAG-binding residues in domain II (Modis et al., 2005), suggesting that the host cell adaptive changes enhance affinity of an existing GAG-binding site in the DEN E protein. Collectively, the spatial distribution of GAG-binding residues on the E protein from the different flaviviruses implies that multiple regions on the E proteins surface can engage with host cell receptors for virus attachment/entry, which is not inconsistent with the flat, elongated architecture of the E protein dimer.
We and others have reported that among the encephalitic flaviviruses an increase in GAG-binding affinity invariably gives rise to virulence attenuation (Goto et al., 2003; Lee et al., 2004; Lee & Lobigs, 2000, 2002; Mandl et al., 2001) by a mechanism of rapid clearance of virus from the circulation (Lee et al., 2004; Lee & Lobigs, 2002). Here, we show that this principle also applies to DEN encephalitis in mice. Although human DEN infection gives rise primarily to a febrile illness with arthralgia and occasional haemorrhagic manifestation, which cannot be reproduced in laboratory animals, the IFN-/-R–/– mouse model used in this study is pertinent to the understanding of DEN virulence in primates, where a reduction in viraemia titre and duration is considered the best marker of attenuation of DEN vaccine candidates (Saluzzo, 2003). In our mouse model, mortality as the result of encephalitis following virus inoculation by an extraneural route was used as virulence read-out and largely reflects the ability of a virus strain to produce viraemia of sufficient magnitude and/or duration to cross the blood–brain barrier. A defect in the interferon responses was required for this to occur in adult mice. Therefore, the conclusions drawn here from studies in mice on the attenuation of DEN-2 GAG-binding variants most likely also apply to their pathogenicity in humans. Notably, DEN-2 vaccine candidates NGC and PR-159-S1 (Eckels et al., 1976, 1980; Meiklejohn et al., 1952; Sabin, 1955; Scott et al., 1980) produced by serial passage in mouse brain and primary green monkey kidney cells display acidic to basic amino acid substitutions at GAG-binding determinants 126 (Gualano et al., 1998) and 202 (Hahn et al., 1988), respectively. We hypothesize that the mechanism of attenuation of these candidate DEN vaccines involves a defect in the ability of the viruses to spread in the infected host via the reticuloendothelial system due to high affinity GAG interaction.
Repeated mouse brain passage of DEN and Yellow fever virus (YFV), which are not neurotropic in natural infections, yields variants that are attenuated in viscerotropism in humans and monkeys but have acquired high neurovirulence in mice (reviewed by Burke & Monath, 2001). For DEN-2 NGC, a GluLys mutation at E protein residue 126 was implicated as a major determinant of mouse neurovirulence (Bray et al., 1998; Gualano et al., 1998). Notably, a YFV–DEN-2 chimera with the prM–E of PUO-218 is far less neurovirulent and replicates poorly in the brain relative to the YFV 17D vaccine and a YFV–DEN-4 chimera with the prM–E of a neurovirulent DEN-4 strain (Chambers et al., 2003). Taken together with our investigation showing that Lys126 is a GAG-binding determinant which attenuates neuroinvasiveness, it is apparent that a single mutation can reduce disease in extraneural tissues (viscerotropism) as well as increase neurovirulence.
Given that cell-to-cell spread in neurons is a route for virus dissemination in the central nervous system, it is not unexpected that the influence of GAG affinity on virulence can differ in virus infections of the brain and in extraneural tissues. Indeed, the greater neurovirulence associated with DEN-2 NGC as a consequence of Lys126 relative to variants with low GAG-binding affinity suggests that the efficiency of interaction between the virus and GAG on the surface of neuronal cells contributes to virus pathogenicity in the brain. A correlation between high GAG-binding affinity of strains of Murine leukemia virus (Jinno-Oue et al., 2001), Theiler's murine encephalomyelitis virus (Reddi et al., 2004) and high neurovirulence has also been reported. With regard to the neurotropism of DEN-2 NGC in comparison to that of non-brain-adapted strains (e.g. PUO-218), we find that the prM and E proteins of the latter (when incorporated into the NGC infectious clone; rPUO) confer comparable neurovirulence in IFN-/-R–/– mice. However, this is not the case in i.c. infection of new-born mice where rPUO is attenuated but NGC and rPUOK126 are both virulent (Gualano et al., 1998). This raises the possibility that Lys126 in the E protein of NGC reduces the susceptibility of virus growth to interferons.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bernfield, M., Götte, M., Park, P. W., Reizes, O., Fitzgerald, M. L., Lincecum, J. & Zako, M. (1999). Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68, 729–777.[CrossRef][Medline]
Bray, M., Men, R., Tokimatsu, I. & Lai, C. J. (1998). Genetic determinants responsible for acquisition of dengue type 2 virus mouse neurovirulence. J Virol 72, 1647–1651.
Burke, D. S. & Monath, T. P. (2001). Flaviviruses. In Fields Virology, 4th edn, pp. 1043–1126. Edited by D. M. Knipe & P. M. Howley. Philadelphia, PA: Lippincott Williams & Wilkins.
Byrnes, A. P. & Griffin, D. E. (2000). Large-plaque mutants of Sindbis virus show reduced binding to heparan sulfate, heightened viremia, and slower clearance from the circulation. J Virol 74, 644–651.
Chambers, T. J., Liang, Y., Droll, D. A., Schlesinger, J. J., Davidson, A. D., Wright, P. J. & Jiang, X. (2003). Yellow fever virus/dengue-2 virus and yellow fever virus/dengue-4 virus chimeras: biological characterization, immunogenicity, and protection against dengue encephalitis in the mouse model. J Virol 77, 3655–3668.
Chen, Y., Maguire, T., Hileman, R. E., Fromm, J. R., Esko, J. D., Linhardt, R. J. & Marks, R. M. (1997). Dengue virus infectivity depends on envelope protein binding to target cell heparan sulfate. Nat Med 3, 866–871.[CrossRef][Medline]
Eckels, K. H., Brandt, W. E., Harrison, V. R., McCown, J. M. & Russell, P. K. (1976). Isolation of a temperature-sensitive dengue-2 virus under conditions suitable for vaccine development. Infect Immun 14, 1221–1227.
Eckels, K. H., Harrison, V. R., Summers, P. L. & Russell, P. K. (1980). Dengue-2 vaccine: preparation from a small-plaque virus clone. Infect Immun 27, 175–180.
Esko, J. D. & Selleck, S. B. (2002). Order out of chaos: assembly of ligand binding sites in heparan sulfate. Annu Rev Biochem 71, 435–471.[CrossRef][Medline]
Fry, E. E., Lea, S. M., Jackson, T. & 7 other authors (1999). The structure and function of a foot-and-mouth disease virus-oligosaccharide receptor complex. EMBO J 18, 543–554.[CrossRef][Medline]
Germi, R., Crance, J. M., Garin, D., Guimet, J., Lortat-Jacob, H., Ruigrok, R. W., Zarski, J. P. & Drouet, E. (2002). Heparan sulfate-mediated binding of infectious dengue virus type 2 and yellow fever virus. Virology 292, 162–168.[CrossRef][Medline]
Goto, A., Hayasaka, D., Yoshii, K., Mizutani, T., Kariwa, H. & Takashima, I. (2003). A BHK-21 cell culture-adapted tick-borne encephalitis virus mutant is attenuated for neuroinvasiveness. Vaccine 21, 4043–4051.[CrossRef][Medline]
Gruenberg, A., Woo, W. S., Biedrzycka, A. & Wright, P. J. (1988). Partial nucleotide sequence and deduced amino acid sequence of the structural proteins of dengue virus type 2, New Guinea C and PUO-218 strains. J Gen Virol 69, 1391–1398.
Gualano, R. C., Pryor, M. J., Cauchi, M. R., Wright, P. J. & Davidson, A. D. (1998). Identification of a major determinant of mouse neurovirulence of dengue virus type 2 using stably cloned genomic-length cDNA. J Gen Virol 79, 437–446.[Abstract]
Hahn, Y. S., Galler, R., Hunkapiller, T., Dalrymple, J. M., Strauss, J. H. & Strauss, E. G. (1988). Nucleotide sequence of dengue 2 RNA and comparison of the encoded proteins with those of other flaviviruses. Virology 162, 167–180.[CrossRef][Medline]
Handel, T. M., Johnson, Z., Crown, S. E., Lau, E. K. & Proudfoot, A. E. (2005). Regulation of protein function by glycosaminoglycans – as exemplified by chemokines. Annu Rev Biochem 74, 385–410.[CrossRef][Medline]
Hilgard, P. & Stockert, R. (2000). Heparan sulfate proteoglycans initiate dengue virus infection of hepatocytes. Hepatology 32, 1069–1077.[CrossRef][Medline]
Hung, S. L., Lee, P. L., Chen, H. W., Chen, L. K., Kao, C. L. & King, C. C. (1999). Analysis of the steps involved in Dengue virus entry into host cells. Virology 257, 156–167.[CrossRef][Medline]
Hung, J. J., Hsieh, M. T., Young, M. J., Kao, C. L., King, C. C. & Chang, W. (2004). An external loop region of domain III of dengue virus type 2 envelope protein is involved in serotype-specific binding to mosquito but not mammalian cells. J Virol 78, 378–388.
Jinno-Oue, A., Oue, M. & Ruscetti, S. K. (2001). A unique heparin-binding domain in the envelope protein of the neuropathogenic PVC-211 murine leukemia virus may contribute to its brain capillary endothelial cell tropism. J Virol 75, 12439–12445.
Johnson, A. J. & Roehrig, J. T. (1999). New mouse model for dengue virus vaccine testing. J Virol 73, 783–786.
Kern, A., Schmidt, K., Leder, C., Muller, O. J., Wobus, C. E., Bettinger, K., Von der Lieth, C. W., King, J. A. & Kleinschmidt, J. A. (2003). Identification of a heparin-binding motif on adeno-associated virus type 2 capsids. J Virol 77, 11072–11081.
Lee, E. & Lobigs, M. (2000). Substitutions at the putative receptor-binding site of an encephalitic flavivirus alter virulence and host cell tropism and reveal a role for glycosaminoglycans in entry. J Virol 74, 8867–8875.
Lee, E. & Lobigs, M. (2002). Mechanism of virulence attenuation of glycosaminoglycan-binding variants of Japanese encephalitis virus and Murray Valley encephalitis virus. J Virol 76, 4901–4911.
Lee, E., Hall, R. A. & Lobigs, M. (2004). Common E protein determinants for attenuation of glycosaminoglycan-binding variants of Japanese encephalitis and West Nile viruses. J Virol 78, 8271–8280.
Lee, E., Pavy, M., Young, N., Freeman, C. & Lobigs, M. (2006). Antiviral effect of the heparan sulfate mimetic, PI-88, against dengue and encephalitic flaviviruses. Antiviral Res 69, 31–38.[CrossRef][Medline]
Lewis, J. A., Chang, G. J., Lanciotti, R. S., Kinney, R. M., Mayer, L. W. & Trent, D. W. (1993). Phylogenetic relationships of dengue-2 viruses. Virology 197, 216–224.[CrossRef][Medline]
Licon Luna, R. M., Lee, E., Mullbacher, A., Blanden, R. V., Langman, R. & Lobigs, M. (2002). Lack of both Fas ligand and perforin protects from flavivirus-mediated encephalitis in mice. J Virol 76, 3202–3211.
Lin, Y. L., Lei, H. Y., Lin, Y. S., Yeh, T. M., Chen, S. H. & Liu, H. S. (2002). Heparin inhibits dengue-2 virus infection of five human liver cell lines. Antiviral Res 56, 93–96.[CrossRef][Medline]
Lindahl, U., Kusche-Gullberg, M. & Kjellen, L. (1998). Regulated diversity of heparan sulfate. J Biol Chem 273, 24979–24982.
Lobigs, M., Usha, R., Nestorowicz, A., Marshall, I. D., Weir, R. C. & Dalgarno, L. (1990). Host cell selection of Murray Valley encephalitis virus variants altered at an RGD sequence in the envelope protein and in mouse virulence. Virology 176, 587–595.[CrossRef][Medline]
Mandl, C. W., Kroschewski, H., Allison, S. L., Kofler, R., Holzmann, H., Meixner, T. & Heinz, F. X. (2001). Adaptation of tick-borne encephalitis virus to BHK-21 cells results in the formation of multiple heparan sulfate binding sites in the envelope protein and attenuation in vivo. J Virol 75, 5627–5637.
Marks, R. M., Lu, H., Sundaresan, R., Toida, T., Suzuki, A., Imanari, T., Hernaiz, M. J. & Linhardt, R. J. (2001). Probing the interaction of dengue virus envelope protein with heparin: assessment of glycosaminoglycan-derived inhibitors. J Med Chem 44, 2178–2187.[CrossRef][Medline]
Meiklejohn, G., England, B. & Lennette, E. T. (1952). Propagation of dengue virus strains in unweaned mice. Am J Trop Med Hyg 1, 51–58.
Modis, Y., Ogata, S., Clements, D. & Harrison, S. C. (2003). A ligand-binding pocket in the dengue virus envelope glycoprotein. Proc Natl Acad Sci U S A 100, 6986–6991.
Modis, Y., Ogata, S., Clements, D. & Harrison, S. C. (2005). Variable surface epitopes in the crystal structure of dengue virus type 3 envelope glycoprotein. J Virol 79, 1223–1231.
Ono, L., Wollinger, W., Rocco, I. M., Coimbra, T. L., Gorin, P. A. & Sierakowski, M. R. (2003). In vitro and in vivo antiviral properties of sulfated galactomannans against yellow fever virus (BeH111 strain) and dengue 1 virus (Hawaii strain). Antiviral Res 60, 201–208.[CrossRef][Medline]
Perrimon, N. & Bernfield, M. (2000). Specificities of heparan sulphate proteoglycans in developmental processes. Nature 404, 725–728.[CrossRef][Medline]
Reddi, H. V., Kumar, A. S., Kung, A. Y., Kallio, P. D., Schlitt, B. P. & Lipton, H. L. (2004). Heparan sulfate-independent infection attenuates high-neurovirulence GDVII virus-induced encephalitis. J Virol 78, 8909–8916.
Rey, F. A., Heinz, F. X., Mandl, C., Kunz, C. & Harrison, S. C. (1995). The envelope glycoprotein from tick-borne encephalitis virus at 2 Å resolution. Nature 375, 291–298.[CrossRef][Medline]
Rostand, K. S. & Esko, J. D. (1997). Microbial adherence to and invasion through proteoglycans. Infect Immun 65, 1–8.[Medline]
Sabin, A. B. (1952). Research on dengue during World War II. Am J Trop Med Hyg 1, 30–50.
Sabin, A. B. (1955). Recent advances in our knowledge of dengue and sandfly fever. Am J Trop Med Hyg 4, 198–207.
Sabin, A. B. & Schlesinger, R. W. (1945). Production of immunity to dengue with virus modified by propagation in mice. Science 101, 640–642.
Sa-Carvalho, D., Rieder, E., Baxt, B., Rodarte, R., Tanuri, A. & Mason, P. W. (1997). Tissue culture adaptation of foot-and-mouth disease virus selects viruses that bind to heparin and are attenuated in cattle. J Virol 71, 5115–5123.[Abstract]
Saluzzo, J. F. (2003). Empirically derived live-attenuated vaccines against dengue and Japanese encephalitis. Adv Virus Res 61, 419–443.[CrossRef][Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor: Cold Spring Harbor Laboratory.
Sasisekharan, R., Shriver, Z., Venkataraman, G. & Narayanasami, U. (2002). Roles of heparan-sulphate glycosaminoglycans in cancer. Nat Rev Cancer 2, 521–528.[CrossRef][Medline]
Schlesinger, R. W., Gordon, I., Frankel, J. W., Winter, W., Patterson, P. R. & Dorrance, W. R. (1956). Clinical and serologic response of man to immunization with attenuated dengue and yellow fever viruses. J Immunol 77, 352–364.
Scott, R. M., Nisalak, A., Eckels, K. H., Tingpalapong, M., Harrison, V. R., Gould, D. J., Chapple, F. E. & Russell, P. K. (1980). Dengue-2 vaccine: viremia and immune responses in rhesus monkeys. Infect Immun 27, 181–186.
Shresta, S., Kyle, J. L., Snider, H. M., Basavapatna, M., Beatty, P. R. & Harris, E. (2004). Interferon-dependent immunity is essential for resistance to primary dengue virus infection in mice, whereas T- and B-cell-dependent immunity are less critical. J Virol 78, 2701–2710.
Spillmann, D. (2001). Heparan sulfate: anchor for viral intruders? Biochimie 83, 811–817.[Medline]
Talarico, L. B., Pujol, C. A., Zibetti, R. G., Faria, P. C., Noseda, M. D., Duarte, M. E. & Damonte, E. B. (2005). The antiviral activity of sulfated polysaccharides against dengue virus is dependent on virus serotype and host cell. Antiviral Res 66, 103–110.[Medline]
Thullier, P., Demangel, C., Bedouelle, H., Megret, F., Jouan, A., Deubel, V., Mazie, J. C. & Lafaye, P. (2001). Mapping of a dengue virus neutralizing epitope critical for the infectivity of all serotypes: insight into the neutralization mechanism. J Gen Virol 82, 1885–1892.
van den Broek, M. F., Muller, U., Huang, S., Aguet, M. & Zinkernagel, R. M. (1995). Antiviral defense in mice lacking both alpha/beta and gamma interferon receptors. J Virol 69, 4792–4796.[Abstract]
Van Gennip, H. G., Vlot, A. C., Hulst, M. M., De Smit, A. J. & Moormann, R. J. (2004). Determinants of virulence of classical swine fever virus strain Brescia. J Virol 78, 8812–8823.
Zhang, W., Heil, M., Kuhn, R. J. & Baker, T. S. (2005). Heparin binding sites on Ross River virus revealed by electron cryo-microscopy. Virology 332, 511–518.[CrossRef][Medline]
Received 27 April 2006;
accepted 13 June 2006.
This article has been cited by other articles:
|
|
 |
|
  R. Suzuki, E. R. Winkelmann, and P. W. Mason Construction and Characterization of a Single-Cycle Chimeric Flavivirus Vaccine Candidate That Protects Mice against Lethal Challenge with Dengue Virus Type 2 J. Virol., February 15, 2009; 83(4): 1870 - 1880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Nickells, M. Cannella, D. A. Droll, Y. Liang, W. S. M. Wold, and T. J. Chambers Neuroadapted Yellow Fever Virus Strain 17D: a Charged Locus in Domain III of the E Protein Governs Heparin Binding Activity and Neuroinvasiveness in the SCID Mouse Model J. Virol., December 15, 2008; 82(24): 12510 - 12519. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. R. Prestwood, D. M. Prigozhin, K. L. Sharar, R. M. Zellweger, and S. Shresta A Mouse-Passaged Dengue Virus Strain with Reduced Affinity for Heparan Sulfate Causes Severe Disease in Mice by Establishing Increased Systemic Viral Loads J. Virol., September 1, 2008; 82(17): 8411 - 8421. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Lee and M. Lobigs E Protein Domain III Determinants of Yellow Fever Virus 17D Vaccine Strain Enhance Binding to Glycosaminoglycans, Impede Virus Spread, and Attenuate Virulence J. Virol., June 15, 2008; 82(12): 6024 - 6033. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
