BIOCHEMISTRY

Structure of the yeastoligosaccharyltransferasecomplex gives insight intoeukaryotic N-glycosylationRebekka Wild,1* Julia Kowal,1* Jillianne Eyring,2* Elsy M. Ngwa,2

Markus Aebi,2† Kaspar P. Locher1†

Oligosaccharyltransferase (OST) is an essential membrane protein complex in theendoplasmic reticulum, where it transfers an oligosaccharide from a dolichol-pyrophosphate–activated donor to glycosylation sites of secretory proteins. Here wedescribe the atomic structure of yeast OST determined by cryo–electron microscopy,revealing a conserved subunit arrangement. The active site of the catalytic STT3 subunitpoints away from the center of the complex, allowing unhindered access to substrates.The dolichol-pyrophosphate moiety binds to a lipid-exposed groove of STT3, whereas twononcatalytic subunits and an ordered N-glycan form a membrane-proximal pocket forthe oligosaccharide. The acceptor polypeptide site faces an oxidoreductase domain instand-alone OST complexes or is immediately adjacent to the translocon, suggesting howeukaryotic OSTs efficiently glycosylate a large number of polypeptides before their folding.

N-linked glycosylation is a posttranslationalmodification of asparagine residues foundin all domains of life (1). The covalentlyattached glycans are essential for correctprotein folding, sorting, and secretion, or

for modulating specific cell surface interactions(2–4). The central enzyme in the N-glycosylationpathway is the oligosaccharyltransferase (OST),which catalyzes the initial transfer of a definedglycan (Glc3Man9GlcNAc2 in higher eukaryotes)from the lipid carrier dolichol-pyrophosphate topolypeptide chains entering the secretory path-way in the endoplasmic reticulum (ER) (5, 6).OST specifically recognizes the Asn-Xaa-Ser/Thrmotif, where Xaa can be any amino acid exceptproline (7). Whereas in bacteria and some lowereukaryotes this process is carried out by a single-subunit oligosaccharyltransferase (ssOST) (8–10),mosteukaryotesencode large,membrane-embeddedOST complexes that contain multiple subunits:eight in yeast and possiblymore inmulticellularorganisms (11, 12). In Saccharomyces cerevisiae,five subunits (STT3, SWP1,WBP1,OST1, andOST2)were shown tobe essential for cell survival,whereasdeletion of the remaining three subunits wasfound to reduce complex stability and glycosyl-ation activity (11, 13).Eukaryotic OST enzymes have been visualized

by using single-particle cryo–electron microscopy(cryo-EM) or cryo–electron tomography (14–16),but the resolution of these studies, ranging from20 to 9 Å, did not allow unambiguous assignment

of OST subunits. X-ray structures of ssOSTs frombacteria and archaea (PglB and AglB proteins,homologous to the catalytic STT3 protein of mul-tisubunit OSTs) have provided insight into thesubstrate recognition and glycan transfer mech-anisms of prokaryotic enzymes (8, 9, 17, 18).X-ray structures of the isolated luminal domainsof yeast OST6 and of the human Tusc3 (alsonamed N33) have revealed the disulfide formationand cleavage mechanism of this redox chaper-one (19, 20). However, in the absence of a high-resolution structure of a eukaryotic OST complex,it is unclear how the noncatalytic subunits ex-tend the range of acceptor polypeptides over thatof ssOST enzymes. To reveal the architecture ofeukaryotic OSTs and to understand how theyrecognize and process a large number of acceptorproteins (21), we used single-particle cryo-EMto determine a high-resolution structure of theyeast OST complex.

Preparation of OSTcomplex forcryo-EM analysis

Yeast OST exists in two isoforms, containingeither OST3 or the homologous OST6 subunit.To avoid potential heterogeneity in our purifiedsamples, we generated a S. cerevisiae strain lack-ing the OST6 gene and overexpressed OST3 froma plasmid to compensate for the absence of OST6.A 1D4 tag fused to the C terminus of the OST4subunit facilitated efficient affinity purificationand sufficient yields of the OST complex despitelow endogenous expression levels. The resultingstrain (Dost6 pOST3 OST4-1D4) showed almostcomplete N-glycosylation of the OST1 and WBP1proteins, whereas a strain expressing only anOST6-containing complex (Dost3 pOST6 OST4-1D4) exhibited hypoglycosylation (Fig. 1A). Massspectrometric analyses based on SILAC (stable

isotope labeling by amino acids in cell culture)demonstrated that for six out of eight subunits,the detergent-purified, OST3-containing complexhas the same subunit composition and stoichi-ometry as OST in wild-type yeast cells (Fig. 1Band table S1). OST4-derived peptides could notbe detected, and the observed ratios for OST3 andOST6 showed that the two paralogs are mutuallyexclusive in the assembled OST complex. TheOST3-containing complex reconstituted into li-pidic nanodiscs was fully functional, as shownby an in vitro glycosylation assay developed ear-lier (10, 22), using a synthetic lipid-linked oligosac-charide (LLO) analog and a fluorescently labeledacceptor peptide (Fig. 1C). The measured glyco-sylation activity of 3.5 peptides per minute perOST matches previously reported rates of a eu-karyotic ssOST (10) (Fig. 1D).The resolution of the 3D reconstruction of

nanodisc-reconstituted OST was 3.3 Å, on thebasis of the Fourier shell correlation = 0.143criterion (figs. S1 and S2). The EM map was ofexcellent quality in the transmembrane (TM) re-gion and in most of the luminal regions. Theluminal domain of SWP1, most distant to themembrane, featured lower map quality, probablyindicating higher domain flexibility, but still dis-played secondary structure features (Fig. 2A andfigs. S1, D to F, and S3). Missing parts in themodel include the luminal domain and TM1 ofOST3 and the external loop EL5 of STT3, forwhich no density was observed. Because wevisualized OST in an apo state, these segmentsare likely mobile (see below).

Architecture and subunit structure

Themembrane-embedded part of yeast OST con-tains a total of 28 TM helices, with each subunitcontributing at least one TM segment (Fig. 2, Bto D). The membrane topologies of the subunitsagree with previous predictions (11), except forOST5, whose N and C termini are located in theER lumen. Previous in vivo experiments sug-gested that OST assembly occurs through theformation of three subcomplexes (13). Our struc-ture revealed that the spatial arrangement of theOST subunits agrees with this subdivision: Sub-complex 1 contains OST1 and OST5; subcomplex2 contains STT3, OST3, and OST4; and subcom-plex 3 contains OST2, WBP1, and SWP1 (Fig. 2D).At the membrane-embedded interfaces betweensubcomplexes, several orderedphospholipids couldbe identified, particularly in the lipid layer facingthe ER lumen (fig. S4, F to I).In subcomplex I, the N-terminal helix of OST5

is in close contact with the OST1 subunit. OST1has a single TM helix and a luminal domaincontaining two subdomains that have similarfolds and superimpose well on each other (rootmean square deviation = 2.2 Å for 153 out of 191residues) (fig. S5A). The fold features two stackedb sheets and no a helices and was previouslyobserved in other multidomain proteins—for ex-ample, in aminopeptidases and leukotriene hydro-lases (23, 24) (fig. S5B). Because it does not containthe catalytic residues in these proteins, its functionwithin OST1 cannot be deduced.

RESEARCH

Wild et al., Science 359, 545–550 (2018) 2 February 2018 1 of 5

1Institute of Molecular Biology and Biophysics, Departmentof Biology, ETH Zurich, CH-8093 Zurich, Switzerland.2Institute of Microbiology, Department of Biology, ETHZurich, CH-8093 Zurich, Switzerland.*These authors contributed equally to this work.†Corresponding author. Email: markus.aebi@micro.biol.ethz.ch(M.A.); locher@mol.biol.ethz.ch (K.P.L.)

on June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://science.sciencemag.org/

Subcomplex II includes the catalytic subunitSTT3, which is homologous to the functionallyand structurally related ssOST enzymes PglB fromCampylobacter lari and AglB from Archaeoglobusfulgidus (8, 9, 18). Our structure confirms thatyeast STT3 contains a similar TM topology with13 TM helices (Fig. 2C) (25, 26), of which TM9 ispoorly resolved in the EM map and is probablyflexible in the absence of bound substrate (fig. S6).The luminal domain of STT3 also resembles thePglB andAglB proteins, although the folds are notidentical (fig. S7). We found the external loop EL5of STT3 to be disordered, in line with previousobservations that EL5 of PglB only becomes fullyordered when substrates are bound (18). The TMhelices of STT3 are tightly interacting with OST4(which consists of a single TMhelix) andwithTM2to TM4 of OST3 (Fig. 2C). Although the presenceof full-length OST3 in the OST complex was con-firmedbySDS–polyacrylamide gel electrophoresis(SDS-PAGE) and mass spectrometric analysis(Fig. 1A and table S1), no clear density for TM1and the N-terminal luminal domain of OST3 wasvisible in the EM map. We conclude that in theabsence of a peptide substrate, the luminal do-main of OST3 is highly flexible. Both TM1 of OST3and TM9 of STT3 are in close proximity to thelikely LLO-binding site, which suggests that theirflexibility might be associated with the absenceof bound LLO.Subcomplex III contains the OST2 subunit,

whose TMhelicesmediate contacts between STT3and the TM helices of WBP1 and SWP1. OST2contains an N-terminal a helix [amino acid (aa)21 to aa 38] located at the cytosolic membraneboundary and parallel to the membrane plane,where it forms contacts with TM8 and TM9 ofSTT3. The WBP1 subunit contains two luminaldomains. The N-terminal domain shares struc-

tural homology with various proteins of distinctfunctions, including an intraflagellar transportprotein or glutamine amidotransferase (27, 28)(fig. S5C), whereas the central domain of WBP1features a fold found in proteins of the comple-ment system (29, 30) (fig. S5D). However, none ofthese structural homologs unambiguously identifya potential function of WBP1. The SWP1 subunitcontains a single luminal domain, which is mostdistant to the membrane of all OST domains andconnected to a single TM helix by a long linker.OST1, WBP1, and SWP1 were previously sug-

gested to be involved in substrate recognition andto act as chaperones, which would coordinateprotein folding and glycosylation (31–33). Ourstructural data show that none of their luminaldomains adopt a chaperone-like fold. However,these domains might serve as docking platformsfor interaction partners, including chaperonesor enzymes acting on nascent glycoproteins. Forexample, the human OST1 homolog ribophorin Iwas shown to interact with the carbohydrate-binding protein malectin (34, 35).To analyze whether eukaryotic OST complexes

share common structural features, we plottedthe degree of sequence conservation between dif-ferent species onto our yeast OST structure. Wefound that the active site groove in STT3 is highlyconserved (Fig. 3A and fig. S7D). In addition,regions of high sequence conservation werefound at the interfaces between subunits OST1and STT3 (1129 Å2 of buried surface area), as wellas between WBP1 and SWP1 (2369 Å2 of buriedsurface area) (Fig. 3B). This suggests that boththe arrangement of the luminal domains and theactive site are conserved in eukaryotic, multi-subunit OST.Previous genetic and mass spectrometric

analyses suggested that yeast OST contains

seven N-glycosylation sites (36). We observedEM density for five of these glycans (at Asn539

of STT3, Asn336 and Asn400 of OST1, and Asn60

and Asn332 of WBP1), whereas the remaining twoglycans (at Asn99 and Asn217 of OST1) are locatedin flexible loop regions and are disordered. Thebest-ordered N-glycan is attached to the strictlyconserved STT3 residue Asn539, where we ob-served density covering eight glycan moieties(Man6GlcNAc2). This N-glycan forms interac-tions with WBP1 and SWP1 (fig. S4, A to E) andis in the immediate vicinity of the proposedbinding site for the 14-unit saccharide of boundLLO (see below).

Active site and substrate-binding pockets

The STT3 subunit contains the active site andthe acceptor peptide and donor LLO bindingpockets. Although we visualized yeast OST in theapo state with no substrates bound, key mecha-nistic insight could be deduced by comparingour structure to bacterial and archaeal ssOSTs(8, 9, 37). A superposition of yeast STT3 andsubstrate-bound PglB (18) (fig. S7, A and B) re-vealed that functionally important conservedresidues andmotifs have a similar spatial arrange-ment (Fig. 4A). These include the Asp-X-Asp/Glumotif (X, any amino acid) involved in coordinat-ing the catalytic metal ion, an aspartate residue(Asp47 in yeast STT3 corresponding to Asp56 inPglB) that binds both the metal ion and thecarboxamide group of the acceptor asparagine,the WWD motif providing hydrogen bond con-tacts to the b-hydroxyl group of the +2 serine/threonine of the acceptor sequon, and the Lys586

residue of the so-called “DK motif” that contrib-utes additional contacts to the +2 serine/threonineof the acceptor peptide (Fig. 4A). Residues shownto be involved in LLO binding to PglB (18) arealso conserved. These include an essential argi-nine (Arg404 in yeast STT3) that interacts withthe pyrophosphate group of LLO and a tyro-sine residue (Tyr521 in yeast STT3) that forms ahydrogen bond with the N-acetyl group of thereducing-end GlcNAc moiety (Fig. 4B).Comparison of the electrostatic surface poten-

tial map of the peptide-binding pocket of PglBwith that of the predicted binding pocket ofyeast STT3 provides a molecular explanation forsome of the differences in acceptor peptide spe-cificities of bacterial and eukaryotic OSTs (Fig. 4C)(38, 39). In PglB, an arginine residue (Arg311) wasfound to interact with a negatively charged sidechain (Asp or Glu) at the −2 position of the sequon.This arginine is conserved in bacterial ssOST en-zymes and correlates with an extended sequon re-quirement (DxNxS/T) for bacterial N-glycosylation(8, 38) (Fig. 4C). In contrast, no such requirementis present at the −2 position of acceptor sequonsin eukaryotes, and no positively charged residueis present in yeast STT3 where the −2 side chainis expected to bind (Fig. 4C). Instead, a largercavity providing space for more voluminous sidechains at the −2 position of sequons is observed,which is in line with previous findings that hu-man OSTs display an increased glycosylation effi-ciency for substrates with aromatic residues at

Wild et al., Science 359, 545–550 (2018) 2 February 2018 2 of 5

7055

35

25

15

10

100130

OST1STT3

WBP1

OST6SWP1

OST2

OST4-1D4

OST3

OST5

purif

ied

prot

ein/

who

le c

ell e

xtra

ct [

%]

0

50

100

150

200

250

300

350

400

450

OST1 OST2 OST3 OST5 OST6 WBP1 SWP1 STT3 Reaction time [min]

Gly

cope

ptid

e[%

]

0 10 20 30 60 90 180 [min]

DANYTK

DANYTK

GlcNAc2

Fig. 1. Purification, stoichiometry, and functional characterization of yeast OST. (A) Silver-stained SDS-PAGE analysis of detergent-purified OST complexes containing either OST6 or OST3.A 1D4 affinity tag was fused to the C terminus of OST4. Multiple bands for OST1 and WBP1 indicateheterogeneous N-glycosylation. MW, molecular weight marker. (B) Mass spectrometry–basedquantification of subunit abundance in purified OST samples containing either OST3 (green bars) orOST6 (blue bars) relative to OST complexes from wild-type cell extract (n = 3 technical replicates,error bars indicate SD). (C) In vitro glycosylation of a fluorescently labeled peptide [TAMRA-DANYTK(TAMRA, tetramethylrhodamine fluorophore; D, Asp; A, Ala; N, Asn; Y, Tyr; T, Thr; K, Lys)] by ananodisc-reconstituted, OST3-containing yeast OST complex, using a synthetic C20-LLO(NerylCitronellyl-PP-GlcNAc2) as a donor substrate. Glycosylated and nonglycosylated peptideswere separated using a Tricine gel. (D) Following quantification of band intensities in (C), the ratioof glycosylated to unreacted peptide was plotted against the reaction time and fitted using aMichaelis-Menten saturation curve (n = 3 biological replicates, error bars indicate SD).

RESEARCH | RESEARCH ARTICLEon June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://science.sciencemag.org/

the −2 position (39). At the expected binding siteof the −1 residue of the sequon, we found a “knob”in the yeast STT3 structure formed by the sidechain of Glu45 (Fig. 4C). The reduced space mayresult in a more efficient binding of sequons withsmaller side chains at the −1 position.A groove between TM6 and TM11 of STT3 in-

dicateswhere the dolichol tail of the LLOwill likelybind (Fig. 3A). A similar groovewas demonstratedto represent the polyprenyl-binding site in PglB,with four prenyl units ordered in the x-ray struc-ture (18). Given their locations, it is conceivablethat in yeast OST, the partially ordered TM9 ofSTT3 and the disordered TM1 of OST3 mightinteractwith boundLLO.Our results suggest thatPglB and yeast STT3 share a common substraterecognition and glycan transfer mechanism de-spite differing in their substrate specificities.Because glycosylation is an essential process

in yeast, abolishing OST function prevents cellgrowth (40). To validate the importance of theresidues identified in the yeast OST structure, wetherefore used a tester strain that expressed thessOST enzyme LmSTT3D from Leishmania major(41) and generated six chromosomal mutationsin the STT3 locus (Asp47→Ala47, Asp166→Ala166,Glu168→Gln168, Glu350→Ala350, Arg404→Ala404,and Lys586→Ala586). These mutations were pre-dicted to affect metal binding, peptide binding,or LLO binding, but they did not affect OSTcomplex stability (fig. S7C). With the exceptionof Lys586→Ala586, which resulted in a temperature-sensitive phenotype, all generated mutations pre-vented growth in the absence of LmSTT3D (Fig. 4D).We conclude that these mutations impaired thecatalytic activity of the STT3 subunit and thusof OST function in vivo (40).

Acceptor polypeptide delivery by theredox chaperone OST3 or the translocon

Higher eukaryotes express two paralogs of thecatalytic STT3 subunit, termed STT3AandSTT3B,which are part of distinct OST complexes. STT3A-containing OST associates with the translocon,thereby bringing native peptide chains enteringthe ER into close proximity to the glycosylationmachinery (14, 15, 42–46). In contrast, STT3B-containing OST complexes (including the yeastenzyme) are stand-alone units that contain eitheran OST3 or OST6 subunit (homologous to themammalian Tusc3 andMagT1, respectively), nei-ther of which is present in translocon-associatedOST. The luminal domains of OST6 and Tusc3have oxidoreductase activity and feature a thio-redoxin fold. Both can form disulfide bonds withacceptor proteins, but OST3- or OST6-containingcomplexes process different subsets of disulfide-forming acceptor polypeptides in vivo (19, 47, 48).In our apo structure of yeast OST, the oxido-reductase domain of OST3 appears disorderedand is likely flexible, which may allow diverseand transiently bound OST substrates to be effi-ciently glycosylated.Recent studies reported amammalian ribosome-

translocon-OST supercomplex visualized by cryo–electron tomography (14, 15). We docked ourhigh-resolution OST structure into this tomog-

raphy map and found a good fit with the portionof the density previously assigned to OST (14, 15)(Fig. 5A). This provides additional evidence thatyeast and mammalian OST complexes share acommon architecture. Additional density in thetomography map near the N terminus of SWP1(Fig. 5A) probably corresponds to an N-terminalextension of ~300 residues of the mammalianSWP1 homolog ribophorin II, as compared withyeast SWP1 (11).

Our docking analysis suggests that the activesite of STT3A faces the heterotrimeric transloconcomplex (49) (Fig. 5, A and B). The contact pointbetween OST and the translocon comprises theregion corresponding to TM2 to TM4 of theOST3 subunit in yeast OST (Fig. 5B). However,OST3 is not present in the mammalian translocon-associated OST (43). Instead, recent studies suggestthat the OST-translocon interaction is mediatedby the DC2 protein (50). A sequence comparison

Wild et al., Science 359, 545–550 (2018) 2 February 2018 3 of 5

~130 Å

~13

0 Å

nanodisc

~70 Å

180°90°

OST2

OST4OST5

OST1

WBP1

SWP1

OST3STT3

1

1 2

3

1 2

11

13

12

13

98

7 5

116

10

2a

2b

34

24

1

2 3

ER lumen

CytosolOST2 OST4

OST5

OST1WBP1

SWP1

sub-complex I sub-complex II sub-complex III

ER lumen

Cytosol

OST2OST4OST5 OST1 WBP1 SWP1OST3STT3

1

(TMEM258) (Ribophorin I) (Ost4) (Stt3B) (Tusc3) (Dad1) (Ost48) (Ribophorin II)

Fig. 2. Structure of yeast octa-subunit OST. (A) EM map of the OST complex, with densitycovering individual subunits colored as in (D). (B) Cartoon representation of the yeastOST structure. Ordered glycans are shown in stick representation. (C) Cytosolic view ontothe TM region of OST. (D) Structures of the single subunits in cartoon representation. EL5 andTM9 of STT3, as well as TM1 of OST3, are shown schematically. The crystal structure of theluminal domain of the homologous OST6 subunit (PDB ID: 3G9B) was used to illustrate the OST3luminal domain (gray ribbon in dashed red box). The names of the corresponding human OSTsubunits are indicated in parentheses.

ER lumen

Cytosol

peptide

LLO

OST1 STT3

WBP1SWP1

low high

30°

sequence conservation

Fig. 3. Sequence conservation in eukaryotic OSTcomplexes. (A) Surface representation ofthe yeast OST complex, colored according to sequence conservation. The predicted peptide andLLO binding sites are indicated. (B) Conserved subunit interfaces between OST1 and STT3 (left) andbetween WBP1 and SWP1 (right) are marked by dotted circles and arrows.

RESEARCH | RESEARCH ARTICLEon June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://science.sciencemag.org/

Wild et al., Science 359, 545–550 (2018) 2 February 2018 4 of 5

Fig. 5. OST-translocon interactions.(A) Yeast OST structure docked into acryo–electron tomography map of themammalian ribosome-translocon-OSTcomplex (EMDB ID: 3068). A close-up ofthe dashed region is shown in the leftpanel. Additional EM density above SWP1and WBP1 (dashed oval) probablycorresponds to a ~300-aa N-terminalextension present in mammalian SWP1homolog ribophorin II. (B) View fromthe ER lumen onto the TM regionsof OST and of the Sec61 translocon(PDB ID: 5A6U) (14) after the dockingshown in (A). Although the OST3subunit provides all contacts to Sec61in this docking, it is replaced by theDC2 subunit in translocon-associatedOST complexes. (C) Model of translocon-associated OSTcomplex architecture andfunction. The orange, curved line depictsa nascent polypeptide entering the ERthrough the translocon and binding tothe active site of STT3. An LLO moleculewas manually placed in its likely bindingpocket, with a red line representingdolichol, circled “P” denoting phosphatemoieties, and the Glc3Man9GlcNAc2moiety depicted by blue and green symbolsaccording to standard glycan nomencla-ture.The black arrow depicts the proposeddirection of the nucleophilic attack duringglycan transfer.

ER lumen

OST

CytosolSEC61

Ribosome

90°

SEC61

SWP1WBP1

OST3/DC2

OST4

OST1

OST5OST2

STT3LLO

SWP1

WBP1OST1

OST5

STT3OST3/DC2

OST4OST2

ER lumen

Cytosol

Cytosol

ER lumen

peptide

LLO

N-ter

PP

NxS/T

membrane

Fig. 4. Active center ofSTT3. (A) Ribbon repre-sentation of peptide-boundPglB (PDB ID: 5OGL) andyeast STT3. Residuesinvolved in substrate rec-ognition or metal ionbinding are shown assticks. The numbering inthe bound peptide(DQNATF sequence) isrelative to the acceptorAsn residue. Q, Gln;F, Phe; W, Trp; I, Ile; E, Glu.(B) Superposition ofyeast STT3 (green ribbon)and LLO-bound PglB(gray ribbon). Functionallyimportant residues (pinkfor PglB residues; blue foryeast STT3 residues) areshown as sticks, withinteractions observed inPglB indicated by dashedlines. (C) Electrostaticsurface representations ofPglB and yeast STT3.The peptide bound to PglBwas modeled into the yeast STT3 structure using the WWD motif and residues D47/D56 (PglB/STT3) as anchors. EL5 (aa 294 to aa 322) of PglBwas removed for clarity. Corresponding regions revealing the structural basis of distinct peptide specificity are indicated with dashed ovals or boxes.(D) In vivo activity assay of STT3 point mutants.

STT3

wtΔstt3::STT3

Δstt3::stt3D47AΔstt3::stt3D166AΔstt3::stt3E168QΔstt3::stt3E350AΔΔ

stt3::stt3R404A

- LmSTT3D

stt3::stt3K586A

+ LmSTT3D

Δstt3::Leu

PglBD518

W517

W516

K586

R404D47

D166

E168

AcceptorAsn

D465

W464

W463

I572

R375

D56

D154

D156

Mn2+

+1

+2

+3

-1

-2

N

C

D47/D56

Y521/Y468

R404/R375LLO

Mn2+

AcceptorAsn

AcceptorAsn

-2

-1

+1

+2

+3+3

+2

E45

R331

-1

-2

+1

PglB STT3

modeledpeptide

RESEARCH | RESEARCH ARTICLEon June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://science.sciencemag.org/

reveals that the three TM helices of DC2 sharea marked similarity with TM2 to TM4 of OST3(fig. S8), lending further weight to the dockingmodel shown in Fig. 5. Multisubunit OSTs maythus be described as a modular assembly of anOST core complex formed by seven subunits (STT3,OST1, OST2, OST4, OST5, WBP1, and SWP1),which associates with either oxidoreductases(OST3/6 or Tusc3/MagT1) or the translocon.The arrangement shown in Fig. 5C allows un-

hindered access of the polypeptide substrate fromthe luminal exit of the translocon to the catalyticcenter of OST. The observed separation of ~40 Åis in agreement with earlier distance estimatesbased on the minimal polypeptide length of65 residues between the peptidyltransferasecenter of the ribosome and the first possibleglycosylation site (51). It is also in line with thepreviously reported minimal distance of 30 to40 Å between the OST active site and the ERmembrane deduced from evaluating the glyco-sylation of integral membrane proteins (52).

Conclusions: Functions of OST subunits

Organisms encoding multisubunit OSTs arefound to glycosylate a substantially expandedrange of protein substrates compared with single-subunit OST enzymes (53). This implies that theauxiliary subunits increase the efficiency of thecatalytic STT3 core by contributing to substrateacquisition or by affecting the folding of acceptorproteins. Our yeast OST structure suggests thatOST2, OST4, and OST5, which contain mostlyTM helices, have a scaffolding function and arethus important for complex stability, but withoutdirectly contacting the substrates (11, 54). Theluminal domains of OST1, SWP1, and WBP1 mayalso have structural roles by stabilizing the STT3subunit conformation. In addition, some of themmay directly interact with the substrates. Ourstructure revealed a cavity ranging from the ac-tive site of STT3 to the WBP1 and SWP1 subunits,just above the membrane boundary (Fig. 5C). Thecavity is lined by the highly ordered N-glycanattached to Asn539 of STT3 and is sufficiently largeto accommodate the glycan moiety of bound LLOsubstrate. It is conceivable that WBP1, SWP1, andpossibly even the ordered N-glycan contribute to therecognition of the lipid-linked Glc3Man9GlcNAc2moiety and thus help to define the preferencefor an LLO substrate containing terminal a–1,2-linked glucose (55). For OST1, it is worth speculat-ing that, given its proximity to the peptide-bindingpocket of STT3, it might interact with acceptor

proteins and influence their folding. The activesite of the OST complex faces the peptide-bindingOST3 (or OST6) subunit or the translocon,both of which present polypeptide substrates inan unfolded state. This arrangement favors N-glycosylation over the competing folding reactionsand thus extends the substrate range of OST.

REFERENCES AND NOTES

1. F. Schwarz, M. Aebi, Curr. Opin. Struct. Biol. 21, 576–582 (2011).2. C. G. Gahmberg, M. Tolvanen, Trends Biochem. Sci. 21,

308–311 (1996).3. A. Grigorian, H. Mkhikian, M. Demetriou, Ann. N.Y. Acad. Sci.

1253, 49–57 (2012).4. A. Helenius, M. Aebi, Annu. Rev. Biochem. 73, 1019–1049 (2004).5. M. Chavan, W. Lennarz, Trends Biochem. Sci. 31, 17–20 (2006).6. R. Kornfeld, S. Kornfeld, Annu. Rev. Biochem. 54, 631–664

(1985).7. R. D. Marshall, Annu. Rev. Biochem. 41, 673–702 (1972).8. C. Lizak, S. Gerber, S. Numao, M. Aebi, K. P. Locher,

Nature 474, 350–355 (2011).9. S. Matsumoto et al., Proc. Natl. Acad. Sci. U.S.A. 110,

17868–17873 (2013).10. A. S. Ramírez et al., Glycobiology 27, 525–535 (2017).11. D. J. Kelleher, R. Gilmore, Glycobiology 16, 47R–62R (2006).12. N. A. Cherepanova, R. Gilmore, Sci. Rep. 6, 20946 (2016).13. S. Mueller et al., Mol. Biol. Cell 26, 2596–2608 (2015).14. S. Pfeffer et al., Nat. Commun. 6, 8403 (2015).15. S. Pfeffer et al., Nat. Commun. 8, 14516 (2017).16. H. Li, M. Chavan, H. Schindelin, W. J. Lennarz, H. Li,

Structure 16, 432–440 (2008).17. M. Igura et al., EMBO J. 27, 234–243 (2008).18. M. Napiórkowska et al., Nat. Struct. Mol. Biol. 24, 1100–1106

(2017).19. B. L. Schulz et al., Proc. Natl. Acad. Sci. U.S.A. 106,

11061–11066 (2009).20. E. Mohorko et al., Structure 22, 590–601 (2014).21. D. F. Zielinska, F. Gnad, J. R. Wiśniewski, M. Mann, Cell 141,

897–907 (2010).22. S. Gerber et al., J. Biol. Chem. 288, 8849–8861 (2013).23. M. M. Thunnissen, P. Nordlund, J. Z. Haeggström,

Nat. Struct. Biol. 8, 131–135 (2001).24. A. Mpakali et al., J. Med. Chem. 60, 2963–2972 (2017).25. H. Kim, G. von Heijne, I. Nilsson, J. Biol. Chem. 280,

20261–20267 (2005).26. P. Lara et al., J. Biol. Chem. 292, 11349–11360 (2017).27. S. Korolev et al., Proteins 49, 420–422 (2002).28. M. Taschner et al., EMBO J. 35, 773–790 (2016).29. V. Krishnan et al., Structure 17, 611–619 (2009).30. N. S. Laursen et al., EMBO J. 30, 606–616 (2011).31. R. Pathak, T. L. Hendrickson, B. Imperiali, Biochemistry 34,

4179–4185 (1995).32. Q. Yan, G. D. Prestwich, W. J. Lennarz, J. Biol. Chem. 274,

5021–5025 (1999).33. C. M. Wilson et al., J. Biol. Chem. 280, 4195–4206 (2005).34. T. Schallus et al., Mol. Biol. Cell 19, 3404–3414 (2008).35. S.-Y. Qin et al., J. Biol. Chem. 287, 38080–38089 (2012).36. G. Li, Q. Yan, A. Nita-Lazar, R. S. Haltiwanger, W. J. Lennarz,

J. Biol. Chem. 280, 1864–1871 (2005).37. C. Lizak et al., Nat. Commun. 4, 2627 (2013).38. M. Kowarik et al., EMBO J. 25, 1957–1966 (2006).39. A. N. Murray et al., Chem. Biol. 22, 1052–1062 (2015).40. Q. Yan, W. J. Lennarz, J. Biol. Chem. 277, 47692–47700 (2002).41. F. P. Nasab, B. L. Schulz, F. Gamarro, A. J. Parodi, M. Aebi,

Mol. Biol. Cell 19, 3758–3768 (2008).

42. M. Chavan, A. Yan, W. J. Lennarz, J. Biol. Chem. 280,22917–22924 (2005).

43. N. A. Cherepanova, S. Shrimal, R. Gilmore, J. Cell Biol. 206,525–539 (2014).

44. Y. Harada, H. Li, H. Li, W. J. Lennarz, Proc. Natl. Acad.Sci. U.S.A. 106, 6945–6949 (2009).

45. T. Shibatani, L. L. David, A. L. McCormack, K. Frueh,W. R. Skach, Biochemistry 44, 5982–5992 (2005).

46. T. A. Rapoport, L. Li, E. Park, Annu. Rev. Cell Dev. Biol. 33,369–390 (2017).

47. L. F. Zacchi, B. L. Schulz, Mol. Cell. Proteomics 15, 2435–2447(2016).

48. K. Poljak et al., Mol. Cell. Proteomics 17, 18–30 (2017).49. B. Van den Berg et al., Nature 427, 36–44 (2004).50. S. Shrimal, N. A. Cherepanova, R. Gilmore, J. Cell Biol. 216,

3625–3638 (2017).51. M. Kowarik, S. Küng, B. Martoglio, A. Helenius, Mol. Cell 10,

769–778 (2002).52. I. M. Nilsson, G. von Heijne, J. Biol. Chem. 268, 5798–5801

(1993).53. A. Dell, A. Galadari, F. Sastre, P. Hitchen, Int. J. Microbiol. 2010,

148178 (2010).54. A. Dumax-Vorzet, P. Roboti, S. High, J. Cell Sci. 126,

2595–2606 (2013).55. P. Burda, M. Aebi, Glycobiology 8, 455–462 (1998).

ACKNOWLEDGMENTSWe thank J. Boilevin, T. Darbre, and J.-L. Reymond for providing theLLO analog and P. Tittmann for technical support. Funding: Thiswork was supported by the Swiss National Science FoundationSinergia programs TransGlyco (CRSII3_147632) and GlycoStart(CRSII5_173709) and grant 310030_162636 to M.A. R.W.acknowledges support from the ETH postdoctoral fellowshipprogram. Author contributions: K.P.L. and M.A. designed theproject. J.E. generated the yeast strain, developed initial purificationprotocols, and purified OST for mass spectrometry analyses.R.W. purified OST for structural studies, reconstituted OST innanodiscs, and performed in vitro glycosylation assays. E.M.N.carried out in vivo mutational analysis in yeast. J.K. and R.W.performed negative-stain EM experiments and prepared cryo-EMgrids. J.K. collected cryo-EM data and performed data analysis.R.W. built the OST model and performed model refinement. K.P.L.revised the model. R.W., J.K., and K.P.L. analyzed the structure.R.W. and K.P.L. wrote the manuscript with the help of J.K. and J.E.;all authors contributed to its revision. Competing interests:None declared. Data and materials availability: Cryo-EM datawere collected at the electron microscopy facility of ETH Zurich(ScopeM). Atomic coordinates of the de novo built yeast OST modelhave been deposited in the Protein Data Bank (PDB) under ID 6EZN.The three-dimensional cryo-EM density maps have been depositedin the Electron Microscopy Data Bank (EMDB) under accessionnumbers EMD-4161 and EMD-4257. All data needed to evaluatethe conclusions of this paper are provided either in the paper orin the supplementary materials.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/359/6375/545/suppl/DC1Materials and MethodsFigs. S1 to S8Tables S1 and S2References (56–84)

16 November 2017; accepted 22 December 2017Published online 4 January 201810.1126/science.aar5140

Wild et al., Science 359, 545–550 (2018) 2 February 2018 5 of 5

RESEARCH | RESEARCH ARTICLEon June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://www.sciencemag.org/content/359/6375/545/suppl/DC1http://science.sciencemag.org/

N-glycosylationStructure of the yeast oligosaccharyltransferase complex gives insight into eukaryotic

Rebekka Wild, Julia Kowal, Jillianne Eyring, Elsy M. Ngwa, Markus Aebi and Kaspar P. Locher

originally published online January 4, 2018DOI: 10.1126/science.aar5140 (6375), 545-550.359Science

, this issue p. 545Scienceglycosylation.substrates and is flanked by accessory subunits that may facilitate delivery of newly translocated proteins forOST, which includes eight separate membrane proteins. The central catalytic subunit contains binding sites for

report a cryo-electron microscopy structure of yeastet al.chain of sugars to asparagine residues of target proteins. Wild pathway begins in the endoplasmic reticulum with the enzyme oligosaccharyltransferase (OST), which attaches a long

Eukaryotes have an elaborate trafficking and quality-control system for secreted glycoproteins. The glycosylationRemember the sugar when making proteins

ARTICLE TOOLS http://science.sciencemag.org/content/359/6375/545

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2018/01/03/science.aar5140.DC1

REFERENCES

http://science.sciencemag.org/content/359/6375/545#BIBLThis article cites 83 articles, 30 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on June 14, 2021

http://science.sciencemag.org/

Dow

nloaded from

http://science.sciencemag.org/content/359/6375/545http://science.sciencemag.org/content/suppl/2018/01/03/science.aar5140.DC1http://science.sciencemag.org/content/359/6375/545#BIBLhttp://www.sciencemag.org/help/reprints-and-permissionshttp://www.sciencemag.org/about/terms-servicehttp://science.sciencemag.org/