Crystal structure of mouse coronavirus receptor-binding domain complexed with its murine receptor

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up

Crystal structure of mouse coronavirus receptor-binding domain complexed with its murine receptor

Crystal structure of mouse coronavirus receptor-binding domain complexed with its murine receptor Author(s): Guiqing Pen

270 102 1MB

Pages 7 Page size 612 x 800 pts Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Recommend Papers

File loading please wait...
Citation preview

Crystal structure of mouse coronavirus receptor-binding domain complexed with its murine receptor Author(s): Guiqing Peng, Dawei Sun, Kanagalaghatta R. Rajashankar, Zhaohui Qian, Kathryn V. Holmes and Fang Li Source: Proceedings of the National Academy of Sciences of the United States of America, Vol. 108, No. 26 (June 28, 2011), pp. 10696-10701 Published by: National Academy of Sciences Stable URL: http://www.jstor.org/stable/27978676 Accessed: 02-12-2015 02:53 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]

National Academy of Sciences is collaborating with JSTOR to digitize, preserve and extend access to Proceedings of the National Academy of Sciences of the United States of America.

http://www.jstor.org

This content downloaded from 128.143.23.241 on Wed, 02 Dec 2015 02:53:14 UTC All use subject to JSTOR Terms and Conditions

Crystal structure of mouse binding domain complexed Guiqing Penga#Dawei

Coronavirus receptor with itsmurine receptor

R. Rajashankarb, Zhaohui Qiancf Kathryn V. Holmes0, and Fang Lia1

Sun3, Kanagalaghatta

of Chemistry and Chemical Biology, Cornell aDepartment of Pharmacology, University of Minnesota Medical School, Minneapolis, MN 55455; department of Microbiology, University of Colorado University, Northeastern Collaborative Access Team, Advanced Photon Source, Argonne, IL60439; and department School of Medicine, Aurora, CO 80045 Edited by Michael Coronaviruses different

G. Rossmann, Purdue University,West evolved

have

diverse

to

mechanisms

recognize and host

transmission cross-species Coronavirus (MHV) hepatitis

for their

receptors

Mouse

range expansion. N-terminal domain

IN, and approved May

Lafayette,

uses

the

as its receptor (NTD) of its spike protein Here we the crystal structure of MHV domain. present binding NTD complexed with its receptor murine carcinoembryonic antigen 1a (mCEACAMIa). related cell adhesion molecule Unexpectedly,

19, 2011 (received for review March

17, 2011)

fusionprotein (17). During maturation, the spike protein is often

cleaved

into

a

subunit

receptor-binding S2 that associate

a membrane

and

SI

together through noncovalent are relatively well conserved interactions (Fig. L4). SI sequences dif within each Coronavirus between group, but differ markedly two independent ferent groups. SI contains domains, N-terminal fusion

subunit

domain

(NTD)

that can both

and C domain,

serve as viral receptor

itsgalectin fold,MHV NTD does not bind sugars, but insteadbinds

binding domains (RBDs) (Table SI). C domain binds toAPN or ACE2 in coronaviruses that use them as receptors (3, 18-24), whereas NTD binds toCE AC AMI inMHV or sugar inTGEV (9, 25). The sugar-bindingdomain has not yet been identifiedin the spike proteins of HCoV-OC43, BCoV, or IBV. The only atomic

ical contacts

NL63

a core structure that has the same ?-sandwich structural motifs (S-lectins) and additional galectins that bind to the N-terminal Despite lg-like domain of mCEACAMIa.

MHV

NTD

contains

fold as human

mCEACAMIa

protein-protein through exclusive at the interface have been confirmed

interactions.

Crit

by mutagenesis, of coro specificities

a structural basis for viral and host providing na vi rus/CEACAM1 reveal that interactions. assays Sugar-binding Coronavi NTDs of some coronaviruses such as human galectin-like rus OC43 bind sugars. Structural analysis and bovine Coronavirus site in Coronavirus NTDs and mutagenesis localize the sugar-binding core. We to be above that Coronavirus the ?-sandwich propose NTDs originated and retained from a host galectin sugar-binding new in some contemporary but evolved functions coronaviruses, structural

for mCEACAMIa

inMHV

features

binding.

| galectin-like N-terminal domain of Coronavirus spike 1 | carcinoembryonic antigen-related cell adhesion molecule

Coronavirus evolution proteins

| sugar binding by coronaviruses

binding by coronaviruses a

use

and core receptors use of diverse sugars. The to infect a wide range of receptors and cause mammalian and avian species respiratory, enteric, sys have evolved diseases. How coronaviruses temic, and neurological to do so has been a major in virology. To solve this puzzle, puzzle we have investigated the structural basis for the complex receptor

ceptors, Coronaviruses

including has allowed

of

variety

cellular

and proteins coronaviruses

of coronaviruses. recognition mechanisms The Coronaviridae family of large, enveloped, positive-stranded RNA viruses consists of at least three major genera or groups (Table

SI). Aminopeptidase-N (APN) is the receptor for porcine trans

missible

gastroenteritis

virus

porcine

(TGEV),

respiratory

corona

virus (PRCoV), and human Coronavirus229E (HCoV-229E) from group

1

(1-3).

Carcinoembryonic

cell

antigen-related

adhesion

molecule 1 (CEACAM1), a member of the carcinoembryonicanti gen familyin the Ig superfamily,is the receptorformouse hepatitis Coronavirus(MHV) fromgroup 2 (subgroup2a) (4, 5).Angiotensin convertingenzyme2 (ACE2) is the receptorforhuman Coronavirus NL63 (HCoV-NL63) from group 1 and human severe acute re spiratorysyndromeCoronavirus(SARS-CoV) fromgroup 2 (sub group 2b) (6, 7). Sugars serveas receptorsor coreceptorsforTGEV from group

1, bovine

Coronavirus

(BCoV)

and human

Coronavirus

OC43 (HCoV-OC43) fromgroup 2a, and avian infectiousbronchitis virus (IBV) fromgroup 3 (8-14). Receptor is unknown for some coronaviruses

such as group

2a human

Coronavirus

HKU1

(HCoV

HKU1). The diversityin receptoruse is a distinctivefeatureof the Coronaviridae

family and

a few other

viruses and paramyxoviruses(15, 16). The

characteristic

large

spikes

on

virus

families

Coronavirus

such as retro

envelopes

are

composed of trimersof the spike protein. The spike protein mediates viral entryintohost cells by functioningas a class I viral 10696-10701

| PNAS

| June 28.2011

| vol.108

| no. 26

structures

available

and

for Coronavirus each

SARS-CoV,

are C

SI

domains

of HCoV re their common

with

complexed

ceptorACE2 (23, 24). Despite marked differencesin their struc tures, the twoC domains bind to overlapping regions on ACE2 Structural

(23, 24).

virus SI NTD.

information

the prototypic

MHV,

has

been

lacking

studied

and extensively

for any corona

Coronavirus,

causes

a varietyofmurine diseases. StrainA59 (MHV-A59) is primarily whereas

hepatotropic,

Murine CEACAM1 functions

are

strain

JHM

is neurotropic.

(MHV-JHM)

(mCEACAMl), whose primaryphysiological

to mediate

cipal receptor for allMHV

cell

adhesion

and

signaling,

strains (26). mCEACAMIa

is the prin

isbroadly

in epithelial cells, and microphages, cells, endothelial expressed its expression level is low in the central nervous system and

but is re

stricted to endothelial and microgolial cells (27). mCEACAMl and -lb; is encoded by two al?eles to produce mCEACAMIa is a much more efficient mCEACAMIa MHV receptor than mCEACAMlb (28, 29). mCEACAMIa contains either two [Dl and D4] or four [D1-D4] Ig-likedomains in tandem,a resultof alter nativemRNA splicing(26). The crystalstructureofmCEACAMIa [1,4] shows thata CO loop (loop connecting?-strandsC andC) in the V-set

Ig-like

Dl

domain

encompasses

key MHV-binding

resi

dues includingIle41 (30). Curiously, althoughmammalian CEA conserved (Tables S2 and S3), only CAM1 proteinsare significantly

murine Also,

can

CEACAMla

although

group

serve as an efficient MHV

2a Coronavirus

spike proteins

receptor (31). are significantly

conserved (Tables S2 and S3), onlyMHV spike protein interacts withmCEACAMIa (31). The molecular determinantsfor theviral and

main

host

of

specificities elusive.

Despite

the

lack

of

interactions

coronavirus/CEACAMl

sequence

homology

between

re

Coronavirus

spike proteins and any known sugar-bindingproteins (lectins), sugar moieties

on host

cell membranes

such as glycoproteins,

gly

colipids, and glycosaminoglycansplay importantroles in host cell infectionsby many coronaviruses (8). HCoV-OC43 and BCoV spike

proteins

recognize

cell-surface

components

acetyl-9-O-acetylneuraminicacid (Neu5,9Ac2)

containing

N

(13, 14). These

Author contributions:G.P., K.V.H., and F.L.designed research;G.P., D.S., K.R.R.,and Z.Q. performed research;G.P., K.V.H., and F.L. analyzed data; and F.L.wrote the paper. The authors declare no conflictof interest. This article isa PNAS Direct Submission. Data deposition: Crystallography,atomic coordinates, and structurefactors reported in thispaper have been deposited inthe ProteinData Bank database (accession no. 3R4D). 1Towhom correspondence should be addressed. E-mail: [email protected] This article contains supporting information 10. online atwww.pnas.org/lookup/suppl/doi: 1073/pnas.1104306108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas. 1104306108

This content downloaded from 128.143.23.241 on Wed, 02 Dec 2015 02:53:14 UTC All use subject to JSTOR Terms and Conditions

Fig. 1. Domain

Structure of MHV-NTD/mCEACAM1a complex. (A) structure of MHV spike protein. NTD: N-terminal domain; RBD: receptor-binding domain; HR-N: heptad repeat N; HR-C: heptad-repeat C; TM: transmembrane an chor; IC: intracellular tail. The signal peptide corresponds to residues 1-14 and iscleaved during molecular maturation (45). Structures and functions of gray areas have not been clearly defined. (B) Kinetics and binding affinityof NTD and mCEACAMIa. (C) Structure of NTD/mCEACAM1 a complex. Two ?-sheets of the NTD core are in green and magenta, respectively; receptor-binding motifs (RBMs) are in red; is inyel other parts of the NTD are in cyan; mCEACAMIa low; and virus-binding motifs (VBMs) are in blue. N*: terminus; C*: C terminus. (D) Sequence and secondary structures of NTD. ?-Strands are shown as arrows, and the disordered region as a dashed line.

viruses also contain a hemagglutinin-esterase as a receptor-destroying and aids enzyme

sugars on infectedcells (32). MHV (33), and

sugars

the HE

aminic such

acid

(Neu5Gc)

sugar-binding

spike protein recognizesN-glycolylneur

loop The model

but

acid (Neu5Ac), and 7V-acetylneuraminic activities are required for the enteric tropism

spike protein,

to bind

fails

sugars,

and

hence

PRCoV

has

respiratorytropismonly (3, 10). IBV spike protein recognizes

Neu5Ac

function activities How

can Coronavirus many spike proteins (11, 12). Overall, as viral lectins, but the molecular nature of their lectin is a mystery.

did

Coronavirus

spike

proteins

and

originate

evolve,

and

how did the evolutionary changes in their spike proteins allow

to explore novel cellular their receptors and expand we have determined these questions, the ranges? To address of MHV-A59 NTD with mCEA crystal structure complexed structure has elucidated the receptor CAMla[l, 4]. The recogni coronaviruses

host

tionmechanism of MHV

and identifiedthe determinantsof the

viral and host

interactions. specificities of coronavirus/CEACAMl structure has unexpectedly the NTD revealed the Furthermore, structural basis for the lectin activities of Coronavirus spike proteins of and provided structural insights into the origin and evolution Coronavirus

spike proteins.

Results and Discussion

were

phases

subsequently

by an averaging

improved

method

(35).We refined the structureat 3.1A resolution (Table S4). The

final model

strains are present

of many MHV

(9). PRCoV spike protein, an NTD-deletion mutant of

of TGEV TGEV

and

The

spike protein does not bind

genes

not expressed (34). TGEV

that functions (HE) viral detachment from

contains

residues

15-268

from residues

receptor

a (except for disordered 1-202 of mCEACAMIa.

of NTD

residues

40-64) also contains glycans TV-linked to viral residue residues 37, 55, and 70.

Structure

of MHV-NTD/mCEACAMIa

is a

NTD

and

13-stranded

?-sandwich

to

192 and

core The Complex. with two antiparallel

of MHV ?-sheets

stackedagainsteach other throughhydrophobicinteractions(Fig. 1 C andD and Fig. 2 ). Surprisingly,this?-sandwich core ofMHV NTD has a galectin fold,which will be discussed in detail later. the core

NTD contains several peripheral structure, MHV elements loops from the (Fig. 1C and Fig. 2A). Three segment to "upper" ?-sheet of the core converge with the N-terminal these four struc form a distinct, receptor-contacting substructure; are termed receptor-binding motifs (RBMs). Three tural elements

Outside

structural

disulfide

bonds,

connecting

cysteines

and

21-158,165-246,

reinforce MHV NTD (Fig. IB). MHV NTD binds to domain Dl ofmCEACAMIa is no significant

Fig. 7A). There

structural

153-187,

(Fig. 1C and

inmCEACAMIa

change

before and afterMHV binding (Fig. S2). The fourRBMs ofMHV

NTD

two virus-binding

contact

motifs

(VBMs)

on

the CC'C"

face

ofmCEACAMIa (Fig. 1C and Fig. 24). A total of 14 residues in NTD interact with a totalof 17 residues inmCEACAMIa (Fig. 3 A-C). The bindingburies 1,500A2 at the interface(Fig. IB). This interface

is

intermediate

between

the

SARS-CoV/ACE2

in

interface (1,300 terface (1,700 A2) and theHCoV-NL63/ACE2 bind to a series all of the three viral domains we designed but ment suitable for crystallization, receptor-binding A2), on its their respective protein receptors with similar affinities (23). NTD constructs based terminal truncation of MHV-A59 structure predictions. One NTD Notably, none of the observed or predicted glycans is involved in secondary fragment containing interactions(Fig. IB), and therefore in insect cells and stable in so residues 1-296 was well expressed MHV-NTD/mCEACAMIa to forma 1:1 heterodimeric MHV-NTD/mCEACAMIa lution.Itbound tomCEACAMla[l,4] binding depends exclusively on pro interactions. tein-protein complexwithKd of 21.4 nM (Fig. IB). We crystallizedthiscomplex Structure

in space

Determination.

group

P6,22,

To

a =

an MHV-A59

prepare

76.4 A,

b =

76.4 A,

and

NTD

c =

frag of C

942.1 ?

(Table S4), with twocomplexes per asymmetricunit (Fig. SI). The

structure was

determined

by single-wavelength

anomalous

diffrac

tion (SAD) phases using selenomethionine-labeledmCEACAMIa.

Detailed

MHV

Interactions.

MHV-NTD/mCEACAM1a

NTD

interactions

and mCEACAMIa with

scattered

polar

The

interface between

is dominated by hydrophobic

interactions.

Peng et al.

This content downloaded from 128.143.23.241 on Wed, 02 Dec 2015 02:53:14 UTC All use subject to JSTOR Terms and Conditions

Two

hydrophobic

g 2

? S 1

showed that single mCEACAMIa substitution I41G and single NTD substitutionsI22A,Y162A, Y162Q, Y162H, but notY162F, viral infectivity (28, 36), underscoring the significance abrogated sub of the two hydrophobic Furthermore, patches. single NTD and N26A viral in stitutions R20A, R20K, significantly decreased fectivity, confirming

the significance

of the hydrogen

bonds

between

Arg20 and mCEACAMIa and betweenAsn26 and mCEACAMIa. A naturallyoccurringQ159L mutation inMHV NTD caused small viral plaques (Fig. 3E) (37), suggestingthatthe hydrogenbond be tweenGlnl59 and Arg20 helps position Arg20 to interactwith mCEACAMIa.

Therefore, many of optimized thus substitutions inMHV

have ular

interactions

weaken

to appears through evolution MHV-A59 its interactions with mCEACAMIa, and

NTD that disrupt these specific molec or abrogate viral infectivity. the structural basis for the viral and host

Our study provides of coronavirus/CEACAMl specificities of structural analyses and mutagenesis

the basis interactions. On we determined that

data,

MHV NTD containsArg20, Ile22,Asn26, and Tyrl62, all ofwhich form energetically favorable interactions with mCEACAM 1a, whereas at the corre NTDs contain residues group 2a Coronavirus to disrupt critical hydrophobic that are expected sponding positions

other

or polar interactions withmCEACAMIa the other

on

hand,

of structural

(Fig. 3B and Fig. S3). On analyses,

we

found

that

contains Ile41, Val49, Met54, and Phe56, all of

mCEACAMIa which

the basis

form

favorable interactions with MHV, energetically bovine CEACAMla and -lb, and human mCEACAMlb, at the corresponding CEACAM1 contain residues that positions

whereas

likelydisrupt these favorable interactionswithMHV (Fig. 3C and Fig. S3). For example, hydrophobic residues Ile41 and Phe56 in mCEACAMIa become hydrophilic residuesThr41 and Thr56 in mCEACAMlb (Fig. 3C and Fig, S3). These results reveal the interface. (A) Another Fig. 2. Structural details of MHV-NTD/mCEACAMIa view of the MHV-NTD/mCEACAM1a structure, which isderived by rotating the one in Fig. 1C90? clockwise along a vertical axis. Virus-binding motif 1 (VBM1) on MHV NTD includes strands ?C, ?C, and ?C" and loops CC, CC", and C D. VBM2 on MHV NTD corresponds to loop FG. (?) Distribution of glycosylation sites and disulfide bonds. Glycans and glycosylated asparagines are inmagenta, and cysteines are inyellow. The orientation of the structure is the same as in Fig. 1C. (C) A hydrophobic patch at the interface that is important for MHV-NTD/mCEACAMIa binding. MHV residues are inma

residues are in green. The orientation of the genta, and mCEACAMIa structure isderived by rotating the structure in Fig. 1C 180? along a vertical axis. (D) Another hydrophobic patch at the interface that is important for MHV-NTD/mCEACAMIa binding. The orientation of the structure is slightly adjusted from the one in Fig. 1C.

mechanisms

mCEACAMlb

MHV whereby or CEACAM1

and ceptor, mCEACAMIa

other whereby as a receptor.

Sugar NTD

out.

centers

One

on

Ile41

from

the CC

loop

of mCEACAMIa. Ile41 is surrounded by the hydrophobic side chains ofMHV Tyrl5, Leu89, and Leul?O and the aliphatic side chains ofMHV Gin 159 and Arg20 (Fig. 2C). Additionally,MHV forms

Arg20 carbonyl

group

receptor

Asp89,

a bifurcated

hydrogen of receptor Thr39 and

bond

with

another

the main

hydrogen

chain

bond with

MHV Glnl59. Receptor Arg96 formsa bifurcatedsalt bridgewith while

stacking

with MHV

Arg20.

The

second

criticalhydrophobicpatch involvesmultiple hydrophobicresidues fromboth MHV and mCEACAMIa that includeMHV residues Ile22 and Tyrl62 and receptor residuesVal49, Met54, and Phe56 (Fig. 2D). Additionally,MHV Asn26 formsa hydrogenbond with the main

chain

amide

In protein-protein group of receptor Thr57. to binding interactions contribute more interactions, hydrophobic to bind interactions contribute more energy, whereas hydrophilic

The above key hydrogenbonds between NTD ing specificity.

chains

and

drophobic that the

the receptor main into place. patches

hydrophobic

patches

chain

These and

help bring structural additional

the adjacent

side hy

analyses suggest interactions polar

provide significantbinding energy and specificitytoMHV-NTD/

mCEACAMIa

binding interactions. of contact residues importance

The

interface

has been

tiviruses

pseudotyped

confirmed

at the NTD/mCEACAMla data. Here we

by mutagenesis

pared the efficiencyofmCEACAMla-dependent with wild-type

or mutant

com

cell entryby len

MHV-A59

spike

protein (Fig. 3D). Our data, togetherwith published data (Fig. 3?), 10698

and not only mCEACAMIa, as its re from cattle or humans, use cannot 2a coronaviruses group

core of MHV NTDs. The Binding by Coronavirus ?-sandwich shares the same 11-stranded fold as human (S galectins

lectins) and rotavirusVP4 (viral lectin) (38, 39), augmented by

two additional in the "lower" ?-sheet ?-strands NTD and human galectin-3 MHV have a Dali an rmsd value of 2.9 A over 137 matching Ca

(Fig. 4 and Fig. S4). Z-score of 7.8 and atoms (40). Impor

cores are identical of their ?-sandwich (Fig. tantly, the topologies NTD structural homology between MHV and S4). This unexpected human galectins may function as suggests that Coronavirus NTDs constructs of viral lectins. To test this possibility, we designed NTD other

stand

patches

uses

2 coronaviruses

group

that correspond

to the crystallized

fragmentbased on the sequence alignment of these

MHV-NTD

and purified each of these corona spike proteins. We expressed virus NTDs and detected their binding interactions with mucin, a mixture of highly glycosylated all of the sugar proteins containing moieties and Neu5Ac) Neu5Gc, (Neu5,9Ac2, recognized by the Coronavirus showed that NTDs of HCoV spike proteins. Results OC43

and

BCoV

HCoV-HKUl, raminic acids

bound

NTDs of MHV-A59, sugars, whereas SARS-CoV did not (Fig. 5). Removal of neu treatment prevented from mucin by neuraminidase and

bindingofHCoV-OC43 or BCoV NTD. Why do theNTDs ofHCoV-OC43 and BCoV, but not thatof

bind sugars, MHV, Coronavirus NTDs? sugar-binding

and where

is the sugar-binding in site located that binds galactose, the galectin-3 core and is located above the ?-sandwich

In human

site (site A)

involves the 10-11 loop (loop connecting ?lO and ?ll) (Fig. 4 and E) (39). In rotavirusVP4, which binds sialic acids, siteA is

blocked

by

a two-stranded

?-sheet;

instead,

the sugar-binding

site

(site B) is located in a groove between the two ?-sheets of the is ?-sandwich core (Fig. 4 C and F) (38). InMHV NTD, site

due to the narrowed the two ?-sheets of groove between the ?-sandwich site A is open and available core, whereas (Fig. 4 A and D). However, with human MHV NTD compared galectin-3, for has a markedly shortened 10-11 loop that may be responsible

blocked

its lack of lectin activity.HCoV-OC43 and BCoV NTDs likely share the same galectin fold as MHV NTD due to theirhigh se quence similarities(Fig. SI), but theyboth contain longer 10-11

I www.pnas.org/cgi/doi/10.1073/pnas.1104306108

This content downloaded from 128.143.23.241 on Wed, 02 Dec 2015 02:53:14 UTC All use subject to JSTOR Terms and Conditions

Peng et al.

V

^

Y15

mCEACAMla 4 i

F19 R20 122 I ! I N94 Y34 G29 T39 A30 R9 6 V49 21

81

Q23 L24 1 I R47 M54 F56

N/6 R47 F5 6

F56

Q59

141 D42 Q59

141

N172 I V49 N51 S52

141

?-GVIMA -kiima . gllev ; /glle? ? gyle:

L1"M I E93

ivi

161

yighfr ciqlvnknga ltgtksvsls iep y1glff gi?tvnyngi,1 l'i'g'i 1.lg , vigylk ctsdk-imdk lkgsvllsr:. vqp

Muri e -G?AC 1 Murine-GEACAKlb Bcvi:ie-CEACAKl?-? i e -CEAG !1b Human-CEACEAMl

LI60 T57 GbS Qb9

svgcyttgq:, py7dgkpntngnkligfwht syciyticqg pytpgkpktngnrvigfv.'ht svcqynmgeypqtickpklg nhrke-lv i, syccyt:-g:eyphtickpflg nkrve-l?;h?-; taggyy:"geyphtick-skg ssrne-skhf

7\ 41 31 I I ? V?NLPI.AI.GA FAKYKGNTTA IDKE-?ARFVP FAWYKGNPVG VHNLPLALGA ?NAFIVHFV7 AHNVTXNFLG YAWYRGFRVD NSQLIASYRV YSWYRGERVD AONVTKSPI.G KTQI.TA3YPV YSWYKGEPVP GNP',IVGYAI VHMIJQGI.VG

DENYRRTQA TDF.NFRRTEA TKFGLOGEPC. hp i.v.?E if:.-::i.\UEFA

D

studies of Fig. 3. Sequence analysis and mutagenesis coronavirus/CEACAMI interactions. (A) List of contact residues at the interface, (?) Partial sequence alignment of group 2a Coronavirus NTDs. Contact residues are in red, important noncontact residues are in blue, and loop 10

GEACA;-g1 a-dependent oil entry by tfH7-A59

O 100 ? a> 80

11 is ingreen. Asterisks indicate positions that have fully conserved residues. Colons indicate positions that have

(nCEACAMla)

strongly conserved residues. Periods indicate positions that have weakly conserved residues. (C) Partial sequence alignment of mammalian CEACAM1 proteins. (D) Struc ture-guided mutagenesis data on MHV NTD. Measured was mCEACAMIa-dependent cell entry by lentiviruses pseudotyped with wild-type or mutant MHV-A59 spike proteins. SEs are shown. (E) Published mutagenesis data

Q159L (MHV)

y162q (MHV)

J ^

Yl62H (MHV)

> &^

y16/f (mhv)

MHV

on MHV NTD (28, 36, 37).

loops thanMHV NTD (Fig. 3B and Fig. S3) and thusmay use site we modified the 10-11 A forsugarbinding.To test thishypothesis, loops inbothBCoV andHCoV-OC43 NTDs, usingMHV NTD as a reference

and HCoV (Fig. 3B and Fig. S3). For both BCoV OC43 the mutant and wild-type NTDs, proteins were equally well and stable in solution, but the mutant proteins expressed (OC43*

and BCoV*) lacked sugar-binding activities obser (Fig. 5). These vations confirm that the 10-11 loops are criticai for sugar binding in both BCoV and HCoV-OC43 A more refined description NTDs. of the sugar-binding site in BCoV future structural and biochemical

Coronavirus structures

and HCoV-OC43

NTDs

awaits

studies.

Use and Evolution. To date, Receptor are available of Coronavirus for RBDs

HCoV-NL63 C domain (23) (Fig. 6). Because of the significant sequence similaritiesof theSI subunitsof the spike proteinswithin each

Coronavirus

of the HCoV-NL63 naviruses SARS-CoV

core structure group, the six-stranded ?-sandwich C domain likely exists in other group 1 coro core structure of the and the 5-stranded ?-sheet

(23), C domain

likely exists

in other

group

(24). Similarly,thegalectin-likeNTD ofMHV

2 coronaviruses

likelyexists inother

1 and group 3 corona The folds of group group 2 coronaviruses. are less clear. However, virus NTDs NTD because and both TGEV core structure of IBV SI have lectin activities, the galectin-fold

NTDs may also be found in both group 1 and 3 coronaviruses in similar or variant forms. The present our understanding of the structures and functions study advances of Coronavirus spike proteins and the complex receptor-recognition of coronaviruses. mechanisms group

2a Coronavirus

group

three crystal SI: group 2a

MHV NTD, group 2b SARS-CoV C domain (24), and group 1

Fig. 4. Structural comparisons of MHV NTD, human galectins, and rotavirus VP4. (A) MHV NTD. The orientation of the structure is the same as in Fig. 1C (B) Human galectin-3 [Protein Data Bank (PDB) 1A3K]. The ?-sandwich core is la beled and colored the same as inMHV NTD. Bound galactose

is inyellow. (C) Rotavirus VP4 (PDB 1KQR). Bound sialic acid is in yellow. (D) Another view of MHV S1 NTD, which isderived by rotating the structure inA counterclockwise along a verti cal axis. Arrow indicates loop 10-11. (E) Another view of hu man galectin-3. Site A indicates itsgalactose-binding site. (F) Another view of rotavirus VP4. Site indicates its sialic-acid binding site.

Peng et al.

PNAS

I June 28, 2011

This content downloaded from 128.143.23.241 on Wed, 02 Dec 2015 02:53:14 UTC All use subject to JSTOR Terms and Conditions

| vol.108

| no. 26

j 10699

s o o ce u

Nase I MHV | OC43

| OC43* | BCoV

C-domain ?-sandwichfold

| BCoV* 1HKU1

Ancestral Coronavirus S1 C-domain

Nase
o_ O Oce u e