【病毒外文文献】2008 Identification of residues in the receptor-binding domain (RBD) of the spike protein of human coronavirus NL63 that

Identification of residues in the receptor binding domain RBD of the spike protein of human coronavirus NL63 that are critical for the RBD ACE2 receptor interaction Han Xin Lin 1 Yan Feng 1 Gillian Wong 1 Liping Wang 2 Bei Li 1 Xuesen Zhao 1 Yan Li 3 Fiona Smaill 1 and Chengsheng Zhang 1 Correspondence Chengsheng Zhang zhangch mcmaster ca 1 Department of Pathology and Molecular Medicine McMaster University and Department of Microbiology St Joseph s Healthcare Hamilton ON L8N 4A6 Canada 2 Department of Cancer Immunology and AIDS Dana Farber Cancer Institute Boston MA 02115 USA 3 National Microbiology Laboratory Canadian Science Center for Human and Animal Health Winnipeg MB R3E 3R2 Canada Received 22 July 2007 Accepted 17 December 2007 Human coronavirus NL63 NL63 a member of the group I coronaviruses may cause acute respiratory diseases in young children and immunocompromised adults Like severe acute respiratory syndrome coronavirus SARS CoV NL63 also employs the human angiotensin converting enzyme 2 hACE2 receptor for cellular entry To identify residues in the spike protein of NL63 that are important for hACE2 binding this study first generated a series of S1 truncated variants examined their associations with the hACE2 receptor and subsequently mapped a minimal receptor binding domain RBD that consisted of 141 residues aa 476 616 towards the C terminus of the S1 domain The data also demonstrated that the NL63 RBD bound to hACE2 more efficiently than its full length counterpart and had a binding efficiency comparable to the S1 or RBD of SARS CoV A further series of RBD variants was generated using site directed mutagenesis and random mutant library screening assays and identified 15 residues C497 Y498 V499 C500 K501 R518 R530 V531 G534 G537 D538 S540 E582 W585 and T591 that appeared to be critical for the RBD hACE2 association These critical residues clustered in three separate regions designated RI RII and RIII inside the RBD which may represent three receptor binding sites These results may help to delineate the molecular interactions between the S protein of NL63 and the hACE2 receptor and may also enhance our understanding of the pathogenesis of NL63 and SARS CoV INTRODUCTION Coronaviruses consist of a large and diverse family of enveloped positive sense single stranded RNA viruses They can infect a broad range of mammalian and avian species causing a variety of diseases in the respiratory gastrointestinal hepatic and central nervous systems Holmes Peiris et al 2003 Since then two additional human coronaviruses HCoV NL63 NL63 and HCoV HKU1 have been identified van der Hoek et al 2004 Woo et al 2005 NL63 has been found to be associated with diseases in the upper and lower respiratory tracts such as bronchiolitis conjunctivitis croup and pneumonia in young children and immunocompromised adults Arden et al 2005 Bastien et al 2005 Ebihara et al 2005 Gerna et al 2006 Kaiser et al 2005 van der Hoek et al 2005 2006 So far NL63 infections have been reported in 12 countries across Europe Asia and North America indicating that it is circulating among the human population worldwide Arden et al 2005 Bastien et al 2005 Han et al 2007 Koetz et al 2006 Lau et al 2006 Suzuki et al 2005 Vabret et al 2005 NL63 is a member of the group I coronaviruses which also includes HCoV 229E 229E feline infectious peritonitis virus 79 1146 feline enteric coronavirus 79 1683 canine coronavirus and porcine transmissible gastroenteritis virus TGEV van der Hoek et al 2004 2006 The genome of NL63 contains 27553 nt and encodes the polyproteins 1a Journal of General Virology 2008 89 1015 1024 DOI 10 1099 vir 0 83331 0 0008 3331 G 2008 SGM Printed in Great Britain 1015 and 1b spike S open reading frame 3 ORF3 envelope E membrane M and nucleocapsid N proteins van der Hoek et al 2004 Fig 1 The S protein of NL63 is a 1356 aa type I membrane glycoprotein which contains a signal peptide 1 20 aa an ectodomain 1 1302 aa a transmembrane domain 1303 1325 aa and a cytoplasmic tail 1326 1356 aa The S protein can also be divided into S1 21 717 aa and S2 718 1356 aa domains based on its similarity to S proteins of other coronaviruses Jackwood et al 2001 Sturman Gallagher Yeager et al 1992 As all previously known group I coronaviruses use the APN of their own host species as functional receptors Tresnan Tresnan et al 1996 and viral sequence analysis indicates that NL63 is most closely related to 229E van der Hoek et al 2004 it has been proposed that NL63 may also employ hAPN as its receptor However hACE2 is a functional receptor for NL63 Hofmann et al 2005 This finding was quite surprising given that the S protein of NL63 shows only limited sequence similarities to that of SARS CoV Previous studies have demonstrated that the S1 domains of coronaviruses may contain so called receptor binding domains RBDs that could directly bind to the cellular receptors The RBD of the S protein of SARS CoV has been mapped to a fragment consisting of aa 318 510 in the S1 domain Babcock et al 2004 Wong et al 2004 Whilst the first 330 residues at the N terminus of the mouse hepatitis virus MHV S1 subunit are able to bind efficiently to carcinoembryonic antigen related cell adhe sion molecule 1 CEACAM1 the RBDs of 229E and TGEV are all located at the C terminus of their S1 domains Ballesteros et al 1997 Bonavia et al 2003 Godet et al 1994 Laude et al 1995 A recent study suggested that the central and C terminal domains of the NL63 S protein are required for hACE2 association Hofmann et al 2006 However the location of the NL63 RBD remains to be defined and the critical residues for receptor binding have not been characterized In this study we generated a series of S1 truncated variants tested their association with the hACE2 receptor and subsequently mapped an RBD that consists of 141 aa aa 476 616 towards the C terminus of the NL63 S1 domain In addition we identified 15 residues in the RBD that appear to be critical for receptor binding Interestingly these critical residues seemed to cluster into three separate regions designated RI RII and RIII inside the RBD suggesting that these sites might be associated with receptor binding METHODS Construction of expression plasmids A plasmid encoding hACE2 with a c myc tag at the N terminus and a C9 tag at the C terminus of hACE2 has been described previously Li et al 2005b In this study the domain structures of the NL63 S protein were defined using computer programs from the websites of SOSUI http bp nuap nagoya u ac jp sosui sosui submit html and Pfam http www sanger ac uk Software Pfam To construct the plasmids expressing S1 and its variants we first synthesized a full length codon optimized NL63 S1 gene encoding aa 21 717 according to the method described by Babcock et al 2004 All of the NL63 S1 truncated variants used in this study were generated based on this template The RBD variants were generated by site directed mutagenesis and or random mutant library screening assays using a GeneMorph Random Mutagenesis kit Stratagene and overlapping PCR strategy The cDNA inserts encoding the S1 RBD or a null aa 12 327 fragment of SARS CoV were amplified based on a codon optimized S protein sequence of SARS CoV described previously Li et al 2003 For expression and production of soluble S1 or RBD variants the cDNA inserts were cloned into a mammalian soluble protein expression vector pSecTag Hygro Invitrogen respectively thereby producing c myc and His tagged fusion proteins In addition some constructs were cloned into a modified pSecTag Hygro Ig vector that contains the mouse Igk chain leader sequence and human IgG Fc fragment producing hACE2 Ig S1 Ig RBD Ig or other Ig fusion proteins All expression plasmids used in this study were confirmed by DNA sequencing Expressionandpurificationofproteins TomakesolublehACE2 S1 or RBD proteins 293T cells were transfected with the expression plasmids described above using a standard calcium phosphate transfection method At 16 20 h after transfection the cells were washed with PBS and cultured in serum free medium 293 SFM II medium Invitrogen The protein containing supernatants were collected on days 2 3 and 4 post transfection and pooled An aliquot of the pooled protein supernatants was analysed by Western blotting WB to confirm the target protein expression The supernatants were then passed through a 0 45 mm filter Nalgene and dialysed with binding buffer prior to affinity purification The affinity column packed with charged nickel beads Qiagen was used to purify the 66His tagged proteins To purify the Ig fusion proteins hACE2 Ig S1 Ig or RBD Ig the supernatants were incubated with protein A G Sepharose beads at 4 uC for 16 h and then washed with 0 5 M NaCl in PBS The samples were subsequently eluted with 50 mM sodium citrate 50 mM glycine pH 2 0 neutralized with NaOH and dialysed against PBS as described previously Sui et al 2004 The purified proteins were concentrated with Centricon filters Amicon and dialysed against PBS Protein concentration was measured with a protein assay kit Bio Rad and aliquots of the proteins were stored at 80 uC WB To detect proteins by WB purified proteins cell culture supernatants or cell lysates were resolved by 10 SDS PAGE and Fig 1 Schematic diagram of the genome organization of NL63 and the deduced domains of its S protein H X Lin and others 1016 Journal of General Virology 89 transferred to a nitrocellulose membrane Amersham Biosciences The membrane was blocked for 1 h at room temperature with blocking buffer 5 skimmed milk 0 05 Tween 20 in PBS and incubated with primary antibody anti c myc antibody at a 1 2000 dilution or NL63 positive human serum at a 1 500 dilution overnight at 4 uC followed by three washes with washing buffer 0 05 Tween 20 in PBS The blot was then incubated with secondary antibody HRP conjugated anti mouse IgG or HRP conjugated anti human IgG at a 1 4000 dilution Amersham Bioscience for 1 h at room temperature and washed three times with washing buffer and once with PBS The protein bands were detected using ECL reagents Amersham Biosciences The density of the protein band was quantified using Scion Image National Institutes of Health Bethesda MD USA Immunoprecipitation IP assay Cell lysate was prepared from hACE2 transfected 293T cells with lysis buffer consisting of 100 mM NH 4 2 SO 4 20 mM Tris HCl pH 7 5 10 glycerol 1 Cymal 5 and protease inhibitor cocktail To detect an S ACE2 association by IP purified S1 or RBD proteins 1 mM were mixed with 200 mg hACE2 C9 cell lysate and 2 5 ml anti His antibody Amersham Biosciences The mixture was incubated at 4 uC overnight followed by the addition of 30 ml protein A G beads and incubation was continued at 4 uC for 2 h The precipitate was washed three times with Triton X 100 in PBS and resuspended in 40 ml protein loading buffer for detection of proteins by WB Additionally we performed IP using soluble hACE2 Ig proteins that were incubated with supernat ant containing S1 c myc His or RBD c myc His protein and protein A G beads followed by the steps described above Flow cytometry To detect receptor expression and or the binding interaction between S1 or RBD and the receptors on the cell surface flow cytometry or FACS was performed as described previously Zhang et al 1996 In brief the target cells were detached with 5 mM EDTA in PBS and washed twice with immunofluorescence IF buffer 1 fetal bovine serum and 0 02 NaN 3 in PBS The cells were incubated with anti c myc antibody for the detection of receptor expression or with S1 Ig or RBD Ig protein for the detection of binding between S1 or RBD and the receptor on ice for 1 h followed by three washes with IF buffer They were further incubated with fluorescein isothiocyanate FITC conjugated anti mouse IgG or FITC conjugated anti human IgG on ice for 30 min The cells were washed three times with IF buffer fixed with 1 formaldehyde and analysed using a Beckman Coulter FC5000 cytofluorometer RESULTS The receptor binding site is located in the C terminal region of the S1 domain To localize the region in the S1 domain of NL63 that is responsible for hACE2 binding we first constructed a panel of expression plasmids that encoded full length S1 S 21 717 and six C terminally truncated S1 variants S 21 616 S 21 516 S 21 460 S 21 410 S 21 360 and S 21 260 Fig 2a We then expressed and purified these S1 variants from human 293T cells We demonstrated by WB that both anti c myc antibody Fig 2b top panel and NL63 positive human serum Fig 2b middle panel recognized these proteins suggesting that they folded properly and retained their antigenicity We also expressed and purified three S1 variants from SARS CoV full length S1 S 12 670 RBD S 318 510 and a null fragment S 12 327 which were detected by anti c myc antibody but not the NL63 positive human serum Fig 2b top and bottom panels To examine S1 hACE2 associations we performed an IP assay using anti His antibody followed by WB using anti C9 Fig 2 Localization of the receptor binding region of NL63 S1 a Diagram of the codon optimized full length S1 and six C terminally truncated S1 variants constructed for this study b Detection of purified S1 variant proteins by WB and S1 hACE2 associations by IP All of the purified c myc and His tagged S1 variants 500 ng each from both SARS CoV and NL63 were detected by WB using anti c myc antibody top panel whereas only the S1 variants of NL63 were recognized by NL63 positive human serum bottom panel For the IP assay middle panel S1 proteins 1000 nM each were incubated with 100 mg C9 tagged hACE2 protein lysate anti His antibody and protein A G beads followed by WB detection of the precipitated hACE2 receptor using anti C9 antibody The data showed that hACE2 could bind to full length S1 and to S 21 616 but not to other C terminally truncated S1 variants of NL63 RBD of HCoV NL63 http vir sgmjournals org 1017 antibody to detect precipitated ACE2 receptors Fig 2b middle panel shows that S1 and S 21 616 of NL63 were able to bind to hACE2 although their binding efficiencies appeared to be much lower than the S1 and RBD of SARS CoV In contrast we did not detect an association between hACE2 and the other S1 variants S 21 516 S 21 460 S 21 410 S 21 360 and S 21 260 suggesting that the region spanning aa 516 616 is required for receptor binding whereas the region at the very end of the C terminus aa 617 717 is dispensable for receptor binding These results are consistent with previous findings reported by others Hofmann et al 2006 Mapping a minimal RBD To map further the RBD we constructed another panel of S1 variants with deletions in the N and or C terminal regions of the S1 domain Fig 3a Protein expression of three S1 variants S 311 616 S 361 616 and S 411 616 was not detected in the supernatants of transfected 293T cells data not shown suggesting that the deleted regions could be important for stability folding and or secretion of the proteins In contrast the other five S1 variants were expressed at high levels in the supernatants of transfected 293T cells We then purified these S1 variants that were detected by WB using anti c myc antibody Fig 3b top panel and performed IP to detect an S1 hACE2 association As expected S1 and RBD of SARS CoV bound efficiently to hACE2 whilst the null fragment S 12 327 did not Fig 3b bottom panel As for the NL63 S1 variants we could detect an association between hACE2 and full length S1 S 21 717 S 210 717 S 311 717 and S 461 616 respect ively but not between hACE2 and S 311 566 or S 517 717 In particular we demonstrated that the S 461 616 variant was able to bind to hACE2 much more efficiently than full length S1 and had a binding efficiency comparable to that of S1 or RBD of SARS CoV Fig 3b bottom panel To determine whether S 461 616 is the minimal fragment that can bind to hACE2 we generated a further four truncated variants S 476 616 S 491 616 S 461 601 and S 461 586 Whilstthe expression levels of S 476 616 and S 461 601 were comparable to that of S 461 616 the other two variants S 491 616 and S 461 586 did not express detectable proteins in the supernatants of transfected 293T cells data not shown We then purified S 476 616 and S 461 601 and used equal amounts of the proteins Fig 3c toppanel forreciprocalIPassaysinwhicheitherthe S proteins Fig 3c middle panel or the hACE2 receptors Fig 3c bottom panel were precipitated and detected In both IP assays S 476 616 bound to hACE2 as efficiently as S 461 616 whilst S 461 601 did not bind We also measured the S1 hACE2 association by flow cytometry using similar amounts of S1 Ig proteins of three NL63 S variants S1 the minimal RBD S 476 616 and the null fragment S 21 460 and three SARS CoV S constructs S1 RBD and null Fig 3d and e The quantitative data presented in Fig 3 e showed dramatic differences in binding efficiency between NL63 RBD hACE2 and S1 hACE2 However we did not observe such differences among NL63 RBD hACE2 SARS CoV RBD hACE2 and SARS CoV S1 hACE2 Therefore the flow cytometry data were consistent with the IP data Collectively our study defined the fragment S 476 616 as the minimal RBD of the S protein of NL63 Identifying critical residues for RBD hACE2 binding A previous study on the RBD of SARS CoV suggested that cysteine and some negatively charged residues appear to be important for receptor association Wong et al 2004 To identify the critical residues in NL63 RBD for hACE2 binding we first generated 20 RBD variants Table 1 in which a single cysteine C or a negatively or positively charged residue was replaced by alanine A These mutants comprised six cysteine C497A C500A C516A C550A C567A and C577A two aspartic acid D484A and D538A three arginine R518A R525A and R530A three glutamic acid E572A E582A and E602A and six lysine K501A K532A K546A K556A K562A and K564A point muta tions The expression plasmids encoding these RBD variants were transiently transfected into 293T cells and protein expression levels in the cell culture supernatants were detected by WB Six of the mutants C516A C550A C567A C577A R525A and K556A either did not express or had very low expression levels in the supernatants of transfected 293T cells whereas the remaining 14 RBD mutants exhibited relatively high expression levels Table 1 and Fig 4a top panel Subsequently similar amounts of protein were used for IP to detect RBD hACE2 associations As illustrated in the bottom panel of Fig 4 a alanine substitution of seven residues C497 C500 D538 R518 R530 E582 and K501 respectively could either abolish or dramatically decrease the receptor binding efficiency To identify additional residues that are critical for RBD hACE2 binding we constructed a random mutant library of NL63 RBD and screened protein expression of 143 mutants by WB Surprisingly most of the mutants 118 143 83 had no or very low levels of protein expression in the supernatants of transfected 293T cells suggesting that protein folding and or stability of the RBD is vulnerable to mutagenic changes data not shown However consid erable levels of protein expression were observed for the remaining25mutants Fig 4b toppanels Similaramounts of protein were used for the detection of RBD hACE2 association by IP assay Fig 4b bottom panels Eleven RBD mutants 1 5 1 22 1 31 1 36 1 74 2 10 3 2 3 7 3 42 3 46 and 3 55 were found to have undetectable or dramatically decreased binding activities to hACE2 receptor whilst the other 14 mutants 1 10 1 11 1 19 1 23 1 30 1 38 1 66 1 72 1 77 1 80 2 1 2 20 3 15 and 3 40 could bind efficiently to hACE2 Sequence analysis revealed that three mutants carried a single amino acid substitution 1 22 with V531E 3 42 with V499I and 3 46 with G534A whereas the other eight mutants harboured two or more amino acid substitutions in RBD Table 2 To pinpoint H X Lin and others 1018 Journal of General Virology 89 Fig 3 Mapping of the minimal RBD a Diagram of NL63 full length S1 and eight S1 variants constructed for this study b Detection ofprotein expression by WBand S1 hACE