Glycans and cellular functions I - protein folding, ER processing, quality...

Glycans and cellular functions I - proteinfolding, ER processing, quality control

BCMB8130: Glycobiology Feb 23, 2009

Complex Carbohydrate Research Center315 Riverbend Rd.

University of GeorgiaAthens, GA 30602

[email protected]

Kelley Moremen

http://bmbiris.bmb.uga.edu/moremen/BCMB8130Glycobiology/index.html

Asn-linked glycosylation: what does it doand why do we care???

Precursor structure conserved from yeast tomammals and plants (+ possibly archaebacteria).

• Enormous cost in energy• Amazing conservation in linkage pattern and necessary

genes/gene products (enzymes)

Standard dogma• Provides hydrophilicity (solubility) to the newly

synthesized protein• Influences protein folding• Protection from proteases• Provides a scaffold for adding targeting information

(i.e. lysosomal enzymes)• Scaffold for recognition by endogenous lectins

(carbohydrate-receptor interactions)• Scaffold for exogenous recognition events (pathogen,

toxin binding)

β4

Man

α3

α2 α2

α2 α3 α6

α6

β4

α2

Asn

α3

α3

α2

Man

Man

Man

ManMan

Man Man

Man

GlcNAc

GlcNAc

GlcGlc

Glc

BUT: Why start with this and only this structure?

Precursor synthesisThe oligosaccharide is assembledsugar by sugar onto the carrier lipiddolichol

High energy pyrophosphate bond

http://darwin.bio.uci.edu/~bardwell/231B_2005_Suetterlin_Lec2.ppt

Helenius, A. and Aebi, M. (2004) Ann. Rev Biochem 73, 1019-49

The steps in the synthesis of the lipid-linked precursor and the final structure are highly conserved.

Even the exceptions are indicative of a highly conserved process:

(From Alberts el al (1994) Molecular Biology of the Cell, 3rd ed.)

Co-translational glycosylation is the rule from yeast to plants and animals.

The diversity of dolichol-linked precursors to Asn-linked glycanslikely results from secondary loss of sets of glycosyltransferases

Samuelson et al (2005) P.N.A.S. 102, 1548-1553

• The vast majority of eukaryotes synthesize Asn-linked glycans (Alg) by means of a LL precursor(Dol-PP-GlcNAc2Man9Glc3)

• Characterized Alg glycosyltransferases and dolichol-PP-glycans of diverse protists.

Models for gain or loss of N-glycan biosynthesic enzymesSamuelson etal (2005) P.N.A.S. 102, 1548-1553

Conclusions:• Common ancestry predicts the Alg glycosyltransferase inventory of each eukaryote.

• This inventory accurately predicts the Dol-PP-glycans observed.

• Alg glycosyltransferases aremissing in sets from eachorganism.

• Dol-PP-GlcNAc2Man5 and Dol-PP-and N-linked GlcNAc2 have notbeen identified previously insome of the wild type organisms.

• Diversity of protist and fungalDol-PP-linked glycans likelyresults from secondary lossof Alg genes from a commonancestor containing the completeset of Alg glycosyltransferases.

Transfer of the glycan to nascent polypeptide chains

Higher eukaryotes: greatest efficiency for transferof Glc3Man9GlcNAc2, but smaller glycans can betransferred under stress conditions (i.e. type 1CDG patients where the LLO biosyntheticenzymes are compromised, mutant cell lines thatare defective in LLO biosynthesis)

Structure of the Sec61 pore demonstrates the dimensionsof co-translational extrusion of the polypeptide chain

through the ER membrane

van den Berg et al (2004) Nature 427, 36-44

van den Berg et al (2004) Nature 427, 36-44

Unfoldedprotein

Native structure/multimeric complex

Upon arrival in the ER


Protein folding

• Chaperones as general folding helpers: BiP prevents premature,incorrect folding of segments that arrive in the ER lumen

• Formation of Disulfide bridges: ER is an oxidative environment whichpromotes the formation of SS bridges.Protein Disulfide Isomerase (PDI) to help with the formation ofdisulfide bonds or the rearrangement of disulfide bonds

• PPI (peptide prolyl isomerase) to catalyze the isomerization ofpeptidyl-prolyl bonds, which can be rate-limiting in protein folding

• Enzymes of the quality control cycle: calnexin (soluble) andcaltreticulin (membranes associated)


Chaperones as folding helpers

BiP (Hsp70 family member in the ER)keeps the newly arrived protein in anunfolded state to prevent prematurefolding of domains that have emergedearlier from the translocon


Cleaved signalsequence

Protein disulfide isomerase(PDI)

Oxidized PDI facilitates theformation of SS bonds, acts asoxidant that is reduced.

contains 2 cysteines that are close together and that are easilyinterconverted between the reduced SH form and the oxidized S-S form.NOTE: S-S bonds are only formed in the ER.

Reduced PDI helps with theformation of the correct disulfidebonds within the substrate (seefigure). Exits unchanged fromthe reaction


Following protein folding in the ER

Molinari & Helenius (2000)Science 288: 331

Use of viruses:shut down host protein synthesisglycoproteins that transit ER

“Tags” for following folding in ER:disulfide bondsglycosylation

Semliki Forest virus infection of CHO cells E1*; p62 (cleaved to generate E2)

Ab: BiP Cnx Crt

Pulse 1’;

p62 has glycosylation site nearN-terminus; E1 does not

SH

SH

HSSH

S-S

SH

SH

S-SS-S

E1 foldingintermediates

E1red

Eox1

Eox2

http://www.biochem.wisc.edu/biochem702/ppts/lecture12.ppt

Benham & Braakman (2000) Crit Rev Biochem Mol. Biol. 35, 433-473

Peptidyl-prolyl-isomerase(PPI)

PPI catalyzes the rotation about peptide prolyl-bonds, which can berate limiting in the folding of protein domains


Classical ATP-drivenChaperones

BiPGRP170

GRP94

OST

Dol-P-OS

SRP

SRP-Rec

BiP

PDI

CNX

CRT

ERp57

Sec61

RibosomemRNA

ER exit sites Transport to theGolgi complex

• Translation is initiated in thecytoplasm

• Binding of SRP to the emergingpolypeptide.

• Ribosome-peptide-SRPcomplex binds to the SRPreceptor on the ER membrane

• Peptide is extruded through theSec61 translocon in the ERmembrane.

• Glycans are added from Dol-P-OS to the emerging peptide viaOST.

Lectin-based Chaperones

Protein Folding in the ER:

Chaperones bind to facilitate protein folding:• ATP driven chaperones bind directly to polypeptides• Work in conjunction with PDI and PPIase to facilitate folding.• Lectin chaperones, CNX and CRT, bind to the glycan and facilitate folding in conjunction with

ERp57, a PDI homolog.• Folded proteins are packaged at ER exits sites for transport to the Golgi complex

Carbohydrate structures influence glycoprotein folding in the ER

ER membrane

Cytoplasm

OST

Glc I

DolPP

Glc II

Glc II

Protein Folding(other chaperonins??)

Glc II


Glc Trans (Parodi enzyme)(only acts on unfolded glycoproteins)

Calnexin

Calreticulin

Anteriograde Transport

to GolgiGlc II

Glc II

Legendα1,2-Manα1,3-Manα1,6-Manβ1,4-Manβ1,4-GlcNAcα1,2-Glcα1,3-Glc

Endoplasmic Reticulum

-KDEL -KDEL

• Membrane-anchored CNX recognizemonoglucosylated high-mannose sugarsattached to the unfolded or incompletelyfolded glycoprotein (red line).

• Interactions with ERp57 facilitate theformation of proper disulfide bondsin the substrate glycoprotein.

• Two possible tracks (blue or green arrows)upon release from CNX or calreticulin.

Folding and transport (green track)

• Release of properly folded protein from CNXleads to deglucosylation of the terminal glucose by Glc II (ii).

• ERGIC-53 recognizes the Man9GlcNAc2 glycoprotein for export out of the ER (iii).

Terminal misfolding -> disposal (blue track)

• Terminal glucose of the glycan (pale blue) is removed by glucosidase II (Glc II) (ii).

• Improperly or incompletely folded glycoproteins are reglucosylated (iii) by UGGT or a mannoseresidue (yellow) is trimmed by ER Man I (iv).

• Additional cycles of binding by CNX or CRT follow reglucosylation by UGGT

• Mannose trimming by mannosidase marks the protein for retrotranslocation and degradation.

Events in the calnexin (CNX) cycle.Schrag et al (2003) TIBS 28, 49-57

• Lectin domain (blue)• Extended arm called the P-domain containing

four repeat modules, P1–P4 (cyan, green,yellow and red).

• Repeat modules formed by antiparallelinteractions of two different proline-richsequence motifs.

• Glucose-binding site in the lectin domain isindicated by a ball-and-stick model (green)

• Glc1Man9GlcNAc2 glycoprotein modeled bysuperimposing the terminal glucose on theobserved position of bound glucose.

• Position of the model glycoprotein(RNase B, magenta) suggeststhat interaction with the P-domainis likely.

Domain structures of calnexin and calreticulin.

Schrag et al (2003) TIBS 28, 49-57

calnexin

calreticulin

glycan-binding

protein-substratebinding ??

3-D structure of calnexin

Helenius, A. and Aebi, M. (2004) Ann. Rev Biochem 73, 1019-49

• Canine CNX and rat ERGIC-53 sequenceswere aligned according to the structuralsimilarities (P-domains were omitted).

• Pink boxes mark those residues whose Cαatoms are within 3 A° of their counterparts inthe other structure.

• Secondary structure assignments accordingto assignments in the Swiss-Pdb Viewer areshown (CNX in dark blue, ERGIC-53 ingreen).

• Sequence identities of equivalent residues(those in pink boxes) are in bold purplecharacters.

• Aligned residues that are farther apart than3 A° are red.

• Residues important in ligand binding inERGIC-53 and their counterparts in VIP36are yellow.

• The alignment of the VIP36 amino acidsequence is based solely on visualcomparison with that of ERGIC-53.

Sequence alignment of lectindomains of CNX, ERGIC-53 andVIP-36

Schrag et al (2003) TIBS 28, 49-57

Calnexin and ERp57 act cooperatively to ensure a proper folding of proteins in the endoplasmic reticulum (ER). Calnexincontains two domains: a lectin domain and an extended arm termed the P-domain. ERp57 is a proteindisulfide isomerase composed of four thioredoxin-like repeats and a short basic C-terminal tail. Here we showdirect interactions between the tip of the calnexin P-domain and the ERp57 basic C-terminus by using NMRand a novel membrane yeast two-hybrid system (MYTHS) for mapping protein interactions of ER proteins. Ourresults prove that a small peptide derived from the P-domain is active in binding ERp57, and we determinethe structure of the bound conformation of the P-domain peptide.

Specific interaction of ERp57 and calnexin determined byNMR spectroscopy and an ER two-hybrid system

Pollock etal (2004) EMBO J. 23, 1020-1029

UGGT acts as a folding sensor by specifically glucosylating N-linked glycans in misfolded glycoproteins thus retaining them in thecalnexin/calreticulin chaperone cycle. To investigate how GTsenses the folding status of glycoproteins, we generated RNase Bheterodimers consisting of a folded and a misfolded domain. Onlyglycans linked to the misfolded domain were found to beglucosylated, indicating that the enzyme recognizes foldingdefects at the level of individual domains and onlyreglucosylates glycans directly attached to a misfoldeddomain. The result was confirmed with complexes of soybeanagglutinin and misfolded thyroglobulin.

Recognition of local glycoprotein misfolding by the ER foldingsensor UDP-glucose:glycoprotein glucosyltransferase

Ritter and Helenius (2000) Nat. Structure Biol. 7, 278-280

BS: RNaseB subtilisin treatedAS: RNaseA subtilisin treatedBsc: RnaseB with scrambled disulfidesAsc: RnaseA with scrambled disulfidesSolid bar: denstured domainOpen bar: native domain

What happens if there arefolding problems?

glucosidases

glucosyltransferases

Foldingproblem

• Simply trapped in a malfoldedconformation

• mutation that leads tomisfolding

• unassembled multimer subunit

Calnexin and BiP bind irreversibly to misfolded proteins

Unfolded protein response pathway

Increased transcription ofchaperones and folding catalysts


Model for IRE1, XBP1, and ATF6 function in UPR

Yoshida (2001) Cell 107, 881

Lee et al (1983) J. Biol. Chem. 259: 4616

Induction of proteins during glucose starvation

+ glucose

-glucose

[3H] leucine labeling for 16 hours of mammalian (K12) cells; 2d gel electrophoresis; autoradiography

a) Grp94 (hsp90 of ER lumen)b) Grp78 (hsp70 of ER lumen - BiP)hs) Hsc70 of cytosolAc) actin

Response to many stimuli - anything affectsprotein folding in ER: affects glycosylation (glucose starvation, inhibitors of glycosylation such as tuni- camycin); ß-mercaptoethanol; calcium ionophores

UPR - Unfolded protein response (found in all eucaryotes)

http://www.biochem.wisc.edu/biochem702/ppts/lecture12.ppt

Failure of quality control

OST

Glc I

DolPP

Glc II

Glc II


Glc II


Glc Trans (Parodi enzyme)(only acts on unfolded glycoproteins)

Calnexin

Calreticulin

Glc II

Glc II

Legendα1,2-Manα1,3-Manα1,6-Manβ1,4-Manβ1,4-GlcNAcα1,2-Glcα1,3-Glc

ER membrane

Cytoplasm


Anteriograde Transport

to Golgi

-KDEL -KDEL

What happens when proteins can not fold correctly in the ER: failure in quality control

OST

Glc I

DolPP

Glc II



Glc Trans (Parodi enzyme)(only unfolded glycoproteins)

Calnexin

Calreticulin

Glc II

Glc II

Proteasome Degradation+ Amino acids

Translocation to cytoplasm through the Sec61 pore

complex

ER membrane

Cytoplasm


ER-associated Degradation

(ERAD)

What determines whether aprotein is terminally misfolded?

Glucose residues: determine interaction withcalnexin/calreticulin

The removal of a single mannose residue (by ERmannosidase I) while still in the cycle promotesassociation with EDEM (ER-degradation enhancing1,2 mannosidase like protein) and targets the proteinfor ER associated degradation (ERAD).


Benham & Braakman (2000) Crit Rev Biochem Mol. Biol. 35, 433-473

Multiple steps in ER-associated degradationof unfolded glycoproteins

OST

Glc I

DolPP

Glc II


Glc Trans(only unfolded glycoproteins)

Calnexin

Calreticulin

Glc II

Glc II

ER membrane

Cytoplasm


Proteosome Degradation Cytosolic N-glycanase


ER Man I

dMNJ Kif

ER Man I gene

disruption

Lectin?(EDEM?, Htm1p?)

Step 1

Step 2

Step 3

Step 4

Sec61 Translocon

OtherComponents?

Two of the steps involve Class 1 mannosidases or mannosidase-related proteins

(Jakob, C. A., Burda, P., Roth, J., and Aebi, M. (1998) J. Cell Biol. 142, 1223-1233)

N-Linked oligosaccharide processing, but not association withCNX/CRT is highly correlated with ERAD of antithrombin Glu313-deleted mutant

Tokunaga et al (2003) Arch. Biochem Biophys 411, 235-242

• Examined the combined effects of inhibitors of glycosidases, protein synthesis, proteasome, and tyrosinephosphatase on ERAD of a Glu313-deleted mutant of antithrombin.

• Kifunensine suppressed ERAD (mannose trimming critical.

• Cycloheximide and puromycin suppress ERAD, the effects cancelled by pretreatment with Cast.

• Kifunensine suppresses ERAD even in Cast-treated cells, (does not require binding to CNX and/or CRT).

• Inhibitors of ER mannosidase I and protein synthesis suppress ERAD at different stages and processingof N-linked oligosaccharides highly correlated with the efficiency of ERAD.

Overexpression of ERManI accelerates the timing clockfor disposal (even for wild type proteins!!)

Wu et al (2003) P.N.A.S. 100, 8229-8234

ERManI overexpression accelerates AAT-Z disposal (non-proteasomal->proteasomal)ERManI overexpression targets WT AAT from fully secreted to fully degraded proteinERManI acts is the rate-limiting timer for targeting proteins for disposal

Glycoprotein Biosynthesis

Legendα1,2-Manα 1,3-Manα 1,6-Manβ 1,4-Manβ 1,4-GlcNAc

α 1,2-Glcα 1,3-Glc

β 1,4-Galβ1,2-GlcNAc

α 2,6-NeuNAc

GnTIGolgi

Man III

/IIx(??

)

ER M

an II

ER membrane

Endoplasmic Reticulum Golgi Complex

OST Glc IGlc II

Glc II

UGGT

Endo a-Man

ER Man IGolgi

Man IA/IBGnTI Golgi

Man IIGnTII

SwdMNJKif

dMNJ Kif

A

B

C

DolPP

B

Failure of Quality Control

Proteosomedegradation

GlycosylasyaraginaseLysosome

AminoAcids

Lysα-Man

Lysα1,6-Man

Lysβ-Man

Chitobiase

Quality Control Glycoprotein Degradation (ERAD)

Cytosol

Cytosolic α-Man

Structure of the yeast ER α-mannosidase I

Side view of (αα)7 barrel End view of (αα)7 barrel

Howell and Herscovics labs (Vallee, F., Lipari, F., Yip, P., Sleno, B., Herscovics, A., and Howell, P. L. (2000) EMBO J. 19, 581-588)

Structure of the yeast ER α-mannosidase I

Display of two protein units in the crystal lattice

Protein unit 1N-glycan 1

Protein unit 2N-glycan2

Active site in core of barrel

Howell and Herscovics labs (Vallee, F., Lipari, F., Yip, P., Sleno, B., Herscovics, A., and Howell, P. L. (2000) EMBO J. 19, 581-588)

Domain structures of Class 1 mannosidases

. . . .10 . . . .20 . . . .30 . . . .40 . . . .50 . . . .60 . . . .70 . . . .80 . . . .90 . . . 100 . . . 110 . . . 120 . . . 1EDEM1 RGMFVFGYDNYMAHAFPQDELNPIHCRGRGPDRGDPSNLNINDVLGNYSLTLVDALDTLAIMGNSSEFQKAVKLVINTVSFDKDSTVQVFEATIRVLGSLLSAHRIITDSKQPFGDMTIKDYDNELLYEDEM2 KAMFYHAYDSYLENAFPFDELRPLTCDG.............HDTWGSFSLTLIDALDTLLILGNVSEFQRVVEVLQDSVDFDIDVNASVFETNIRVVGGLLSAHLLSKKAG..VEVEAGWPCSGPLLREDEM3 LEMFDHAYGNYMEHAYPADELMPLTCRGRVRGQ.EPSRGDVDDALGKFSLTLIDSLDTLVVLNKTKEFEDAVRKVLRDVNLDNDVVVSVFETNIRVLGGLLGGHSLAIMLK..EKGEYMQWYNDELLQERManI IDVFLHAWKGYRKFAWGHDELKPVSRSF....S..........EWFGLGLTLIDALDTMWILGLRKEFEEARKWVSKKLHFEKDVDVNLFESTIRILGGLLSAYHLSGDS..............LFLRMan1A KEMMKHAWNNYKGYAWGLNELKPISKGG.........HSSSLFG.NIKGATIVDALDTLFIMEMKHEFEEAKSWVEENLDFNVNAEISVFEVNIRFVGGLLSAYYLSGE.......EI..FR.....KMan1B KEMMKHAWDNYRTYGWGHNELRPIARKG.........HSPNIFGSSQMGATIVDALDTLYIMGLHDEFLDGQRWIEDNLDFSVNSEVSVFEVNIRFIGGLLAAYYLSGE.......EI..FK.....IMan1C KEMMQFAWQSYKRYAMGKNELRPLTKDG.........YEGNMFG.GLSGATVIDSLDTLYLMELKEEFQEAKAWVGESFHLNVSGEASLFEVNIRYIGGLLSAFYLTGE.......EV..FR.....I

30 . . . 140 . . . 150 . . . 160 . . . 170 . . . 180 . . . 190 . . . 200 . . . 210 . . . 220 . . . 230 . . . 240 . . . 250 . . .EDEM1 MAHDLAVRLLPAFENTKTGIPYPRVNLKTGVP.P....DTNNETCTAGAGSLLVEFGILSRLLGDSTFEWVARRAVKALWNLRSNDTGLLGNVVNIQTGHWVGK.QSGLGAGLDSFYEYLLKSYILFGEDEM2 MAEEAARKLLPAFQ.TPTGMPYGTVNLLHGVN.P....GETPVTCTAGIGTFIVEFATLSSLTGDPVFEDVARVALMRLWESRS.DIGLVGNHIDVLTGKWVAQ.DAGIGAGVDSYFEYLVKGAILLQEDEM3 MAKQLGYKLLPAFN.TTSGLPYPRINLKFGIRKPEARTGTETDTCTACAGTLILEFAALSRFTGATIFEEYARKALDFLWEKRQRSSNLVGVTINIHTGDWVRK.DSGVGAGIDSYYEYLLKAYVLLGERManI KAEDFGNRLMPAFR.TPSKIPYSDVNIGTGVAHP...PRWTSDSTVAEVTSIQLEFRELSRLTGDKKFQEAVEKVTQHIHGLSGKKDGLVPMFINTHSGLFTHLGVFTLGARADSYYEYLLKQWIQGGMan1A KAVELGVKLLPAFH.TPSGIPWALLNMKSGIGRNWPWASGGSS.ILAEFGTLHLEFMHLSHLSGNPIFAEKVMNIRTVLNKLEK.PQGLYPNYLNPSSGQWGQH.HVSVGGLGDSFYEYLLKAWLMSDMan1B KAVQLAEKLLPAFN.TPTGIPWAMVNLKSGVGRNWGWASAGSS.ILAEFGTLHMEFIHLSYLTGDLTYYKKVMHIRKLLQKMDR.PNGLYPNYLNPRTGRWGQY.HTSVGGLGDSFYEYLLKAWLMSDMan1C KAIRLGEKLLPAFN.TPTGIPKGVVSFKSG...NWGWATAGSSSILAEFGSLHLEFLHLTELSGNQVFAEKVRNIRKVLRKIEK.PFGLYPNFLSPVSGNWVQH.HVSVGGLGDSFYEYLIKSWLMSG

260 . . . 270 . . . 280 . . . 290 . . . 300 . . . 310 . . . 320 . . . 330 . . . 340 . . . 350 . . . 360 . . . 370 . . . 380 . .EDEM1 ..EKEDLEMFNAAYQSIQNYLRRGREACNEGEGDPPLYVNVNMFSGQLMN.TWIDSLQAFFPGLQVLIGD..............VEDAICLHAFYYAIWKRYGA..LPERYNWQL..........QAPEDEM2 ..DKKLMAMFLEYNKAIRNYTR...........FDDWYLWVQMYKGTVSM.PVFQSLEAYWPGLQSLIGD..............IDNAMRTFLNYYTVWKQFGG..LPEFYNIPQGY........TVEEDEM3 ..DDSFLERFNTHYDAIMRYIS...........QPPLLLDVHIHKPMLNARTWMDALLAFFPGLQVLKGD..............IRPAIETHEMLYQVIKKHNF..LPEAFTTDFR...........VERManI KQETQLLEDYVEAIEGVRTHLLRHS........EPSKLTFVGELAHGRFS.AKMDHLVCFLPGTLALGVYHG.......LPASHMELAQELMETCYQMNRQMETGLSPEIVHFNLYPQPGRRDVEVKPMan1A KTDLEAKKMYFDAVQAIETHLIRK.........SSSGLTYIAEWKGGLLE.HKMGHLTCFAGGMFALGADAAPEGMAQHYLELGAEIARTCHESYNRTFMKLG....PEAFRFDGGVEA....IATRQMan1B KTDHEARKMYDDAIEAIEKHLIKK.........SRGGLTFIGEWKNGHLE.KKMGHLACFAGGMFALGADGSRADKAGHYLELGAEIARTCHESYDRTALKLG....PESFKFDGAVEA....VAVRQMan1C KTDMEAKNMYYEALEAIETYLLNV.........SPGGLTYIAEWRGGILD.HKMGHLACFSGGMIALGAEDAKEEKRAHYRELAAQITKTCHESYARSDTKLG....PEAFWFNSGREA....VATQL

. 390 . . . 400 . . . 410 . . . 420 . . . 430 . . . 440 . . . 450 . . . 460 . . . 470 . . . 480 . . . 490 . . . 500 . . . . EDEM1 DVLFYPLRPELVESTYLLYQATKNPFYLHVGMDILQSLEKYTKVKCG.YATLHHVID...KSTEDRMESFFLSETCKYLYLLFDED.NPVHKSG.................TRYMFTTEGHIVSVEDEM2 KREGYPLRPELIESAMYLYRATGDPTLLELGRDAVESIEKISKVECG.FATIKDLRD...HKLDNRMESFFLAETVKYLYLLFDPT.NFIHNNGSTFDAVITPYGECILGAGGYIFNTEAHPIDPEDEM3 HWAQHPLRPEFAESTYFLYKATGDPYYLEVGKTLIENLNKYARVPCG.FAAMKDVRT...GSHEDRMDSFFLAEMFKYLYLLFADK.EDIIFDI.................EDYIFTTEAHLLPLERManI ADRHNLLRPETVESLFYLYRVTGDRKYQDWGWEILQSFSRFTRVPSGGYSSINNVQDPQKPEPRDKMESFFLGETLKYLFLLFSDDPNLLSLDA...................YVFNTEAHPLPIMan1A NEKYYILRPEVMETYMYMWRLTHDPKYRKWAWEAVEALENHCRVNGG.YSGLRDVYLLH.ESYDDVQQSFFLAETLKYLYLIFSDD.DLLPLEH...................WIFNSEAHLLPIMan1B AEKYYILRPEVIETYWYLWRFTHDPRYRQWGWEAALAIEKYCRVNGG.FSGVKDVYSST.PTHDDVQQSFFLAETLKYLYLLFSGD.DLLPLDH...................WVFNTEAHPLPVMan1C SESYYILRPEVVESYMYLWRQTHNPIYREWGWEVVLALEKYCRTEAG.FSGIQDVYSST.PNHDNKQQSFFLAETLKYLYLLFSED.DLLSLED...................WVFNTEAHPLPV

α1 α2 α3 α4

α5 α6 α6 α7 α8

α9 α10 α11

α12 α13 α14 α14

EDEM1

EDEM2

EDEM3

ER MAN I

Golgi Man IA

Protease-associated domain

KDEL

Threading of yeast HTM1 results in closefit with ER Man I

Human ER Man I + DMJ Yeast HTM1

Cα backbone overlap Overlap of side chains in catalytic core

Mast et al (2005) Glycobiology 15, 421-436

EDEM2 overexpression accelerates mutant α1antitrypsin disposal

Steve Mast

EDEM2 overexpression can accelerate the degradation ofAAT mutants PI Z or NHK

Acceleration of ERAD by EDEM2 requires a glycosylatedERAD substrateTruncation of the COOH-terminal extensioneliminates EDEM function in ERADEDEM2 in cell extracts can be found in acomplex with calnexin by co-IP (data not shown)Partially purified EDEM2 has been tested forcleavage of Man9-5GlcNAc2: no cleavagedetected with any oligosaccharide structure

Analogies for ERAD

Sifers, R.N. (2003) Science 299, 1330-1331

Table 1. Examples of ERAD substrates of medical relevanceERAD substrate Associated disease Ref.α1antitrypsin Z variant (AAT-Z) Emphysema and liver disease (30,31)ΔF508 CFTR Cystic fibrosis (42,43)Insulin receptor mutants Type A insulin resistance (44)HMG co-A reductase mutants Cancer, heart disease, cholesterolemia (45)LDL receptor mutants Hypercholesteroemia (46,47)MPO Y173C Myeloperoidase deficiency (48)Pro-PTHrP Hypercalcemia associated with malignancy (49)Tyrosinase Amelanotic melanomas (50)β-Hexosaminidase Tay-Sachs disease (51)Pro-collagen IA1 Osteogenesis imperfecta (52)α2-plasmin inhibitor Severe hemorragic disease (41)Thyroglobulin Congenital goiter with hypothyroidism (53)MHC class I HCMV (54)CD4 AIDS (55)Copper-transporting ATPase Wilson disease (56)Adapted from Brodsky and McCracken (57)

Cellular response to unfolded proteins in the ER

Only let correctly folded proteins out of the ER (Quality Control)Degrade the unfolded proteins before they accumulate in the ER(ER-associated degradation, ERAD)

Two routes for exit from the ER:gating the exit of correctly folded proteins down the secretory pathway

OST

Glc I

DolPP

Glc II

Protein Folding

Glc Trans (Parodi enzyme)(only unfolded glycoproteins)

Calnexin

Calreticulin

Glc II

Glc II

Proteasome Degradation+ Amino acids

Timing?Recognition

?

ER membrane

Cytoplasm


ER-associated degradation

(ERAD)

Glc II

-KDEL

to GolgiComplex

9

121110

87 6 5

4

32

1

• Cholera toxin is transported from the plasma membrane to the ER

• The A1 subunit (CTA1) is transferred to the cytosol.

• Export of CTA1 from the ER to the cytosol was investigated in a cell-free assayusing microsomes loaded with CTA1.

• Export of CTA1 from the microsomes was time- and adenosine triphosphate-dependent and required lumenal ER proteins.

• By co-IP CTA1 was shown to be associated during export with the Sec61pcomplex

• Export of CTA1 was inhibited when the Sec61p complexes were blocked bynascent polypeptides arrested during import, thus export of CTA1 depended ontranslocation competent Sec61p complexes.

• Export of CTA1 indicated insertion of the toxin into the Sec61p complex from thelumenal side.

• Suggests that Sec61p complex–mediated protein export from the ER is notrestricted to ER-associated protein degradation but is also use by bacterialtoxins for entry into the cytosol of the target cell.

Cholera Toxin Is Exported from Microsomes by the Sec61pComplex

Schmitz et al (2000) J. Cell Biol. 148, 1203-1212

• Efficient degradation of soluble malfolded proteins in yeast requires a fullycompetent early secretory pathway (i.e. mutant forms of sec12, sec23, sec18,ufe1, or sed5 all caused delays in degradation of CPY*).

• Mutations in proteins essential for ER-Golgi protein traffic severely inhibit ERdegradation of the model substrate CPY*.

• ER localization of CPY* was found in WT cells, but no other specific organelle forER degradation could be identified by electron microscopy studies.

• CPY* is degraded in COPI coat mutants, but only a minor fraction of CPY* orsome other proteinaceous factor that is required for degradation seems to enterthe recycling pathway between ER and Golgi.

• They propose that the disorganized structure of the ER and/or the mislocalizationof Kar2p, observed in early secretory mutants, is responsible for the reduction inCPY* degradation.

• Also, mutations in proteins directly involved in degradation of malfolded proteins(Der1p, Der3/Hrd1p, and Hrd3p) lead to morphological changes of theendoplasmic reticulum and the Golgi, escape of CPY* into the secretory pathwayand a slower maturation rate of wild-type CPY.

ER-Golgi Traffic Is a Prerequisite for Efficient ER Degradation

Taxis, Vogel, and Wolf (2002) Mol. Biol. Cell 13, 1806-1818

Quality control in theendoplasmicreticulum protein factory

Sitia and Braakman (2003)18, 891-894

Gardinier lecture “Principles in Molecular and Cell Biology http://www.uiowa.edu/~c156201/PDFLecs/PDFlecs2004/Gardinier.12.1.04.pdf

http://www.micro.biol.ethz.ch/~clajakob/research/ERAD.htm

Amino acids

Polyubiquitin-proeasome cycle

Hampton, R.Y. (2002) Curr Opin. Cell Biol. 14, 476-482

• Hrd1p/Hrd3p complex (left) and Doa10p ligase (right).• Associated ubiquitin conjugating enzymes (E2s) are also shown• Cue1p anchors Ubc7p to the lumenal face of the ER membrane.• Both multi-spanning RING-domain proteins (Hrd1p or Doa10p) are key components

of the complex.• In all likelihood, still other ubiquitin E3s are involved in ER degradation of some

substrates.

Two ER-associated ubiquitin ligase complexes.

Ahner & Brodsky (2004) Trend Cell Biol. 14, 474-478

• Black stars represent misfolded domains

• Black triangles are N-linked core glycans

• All ERAD substrates tested in vivo require theproteasome and the Cdc48–Ufd1p–Npl4pcomplex.

• Variety and distinctions among the rest of theERAD machinery.

1. Soluble substrates interact with lumenal chaperonesto remain aggregation-free (a),

2. Transmembrane proteins with prominent cytosolicdomains require cytosolic chaperones (e.g. Hsp70 orSsa1p) for degradation.

3. Misfolded ER-lumenal and transmembrane proteinscan be transported from the ER and retrieved from theGolgi before being degraded by the ERAD-lumenal(ERAD-L) pathway.

4. Degradation requires Der1p, EDEM (Mnl1p), andubiquitin-conjugating enzymes (Ubc1p and Ubc7p)and ligases (Der3p or Hrd1p).

5. Transmembrane proteins with misfolded domains inthe cytosol (b) are retained in the ER and degraded bythe ERAD-cytosolic (ERAD-C) pathway, which alsorequires a ubiquitin-conjugating enzyme (Ubc7p) andubiquitin ligase (Doa10p).

Distinct ERAD pathways lead to protein degradation.

Cytosolic complex involved in ERAD: Cdc48-Npl4-Ufd1

• Members of the AAA ATPase family are involved in extracting anddegrading membrane proteins.

• Another member of this family, Cdc48 in yeast and p97 in mammals,is required for the export of ER proteins into the cytosol.

• Cdc48/p97 was previously known to function in a complex with thecofactor p47 in membrane fusion

• Its role in ER protein export requires the interacting partners Ufd1 andNpl4.

• The AAA ATPase interacts with substrates at the ER membrane

• Is required to release them as polyubiquitinated species into thecytosol.

• Propose that the Cdc48/p97±Ufd1±Npl4 complex extracts proteinsfrom the ER membrane for cytosolic degradation.

The AAA ATPase Cdc48/p97 and its partners transport proteins from the ERinto the cytosol Ye et al (2001) Nature 414, 652-656

The conserved Npl4 protein complex mediatesproteasome-dependent membrane-bound transcriptionfactor activation Hichcock et al (2001) Mol Biol Cell 12, 3226-3241

• A membrane associated complex containing Npl4p, Ufd1p, and Cdc48p mediates proteasome-regulated cleavage of Mga2p and Spt23p, two model proteins examined in this study.

• Mutations in NPL4, UFD1, and CDC48 cause a block in Mga2p and Spt23p processing

• Our data indicate that the Npl4 complex may serve to target the proteasome to theubiquitinated ER membrane-bound proteins

• Given that NPL4 is allelic to the ERAD gene HRD4, we further propose that this NPL4 functionextends to all ER-membrane–associated targets of the proteasome.

Co-IP of Cdc48pand Ufd1p withPA-tagged Npl4p

McCracken and Brodsky (2003) BioEssays 25, 868-877

• Previous reports that SCF-Fbs1,2 ubiquitin-ligase complexescontribute to ubiquitination of ERAD substrates.

• SCF-Fbs1,2 complexes are shown to interact with unfoldedglycoproteins.

• The SCF-Fbs1 complex associated with p97 bound to integrin-β1 in a manner dependent on Cdc48/p97 ATPase activity.

• Both Fbs1 and Fbs2 proteins interacted with denaturedglycoproteins (both high-mannose and complex-type) moreefficiently than native proteins.

• Fbs proteins interact with innermost chitobiose in N-glycans

• Fbs proteins distinguish native from unfolded glycoproteins bysensing the exposed chitobiose structure.

Glycoprotein-specific ubiquitin ligases recognizeN-glycans in unfolded substrates

Yoshida et al (2005) EMBO Reports 6, 1-6

Hampton, R.Y. (2002) Curr Opin. Cell Biol. 14, 476-482

• High levels of FPP -> Hmg2p is recognized as an ERAD substrate by theHRD/DER machinery.

• Lowering FPP levels or glycerol causes rapid, reversible stabilization of Hmg2p.

• Subtle determinants may be presented when FPP is high, triggering ERAD.

• Regulated transitions to ERAD have broad potential as undiscoveredmechanisms in normal cellular regulation (and targets for pharmacologicalmodulation).

Model for regulated degradation of Hmg2p(HMG-CoA reductase).

Where do we go from here?

Many proteins (even wt proteins) have inefficient folding kinetics (i.e. slowconformational maturation of wt CFTR results in >60% disposal)

Mutant forms can delay folding kinetics but not eliminate function (i.e. ΔF508mutant of CFTR: kinetics of folding is slowed, quantitatively degraded, chloridetransporter is functional if disposal is delayed).

Many loss-of-function human genetic diseases have at least one allele withdelayed folding kinetics (but can produce functional protein), degraded beforefolding is complete.

Some protein misfolding disorders can present as either dominant or recessivegain-of-toxic-function (i.e. liver disease in AAT-Z).

Growing literature on strategies to increase the kinetics of folding of mutantproteins: employ substrate analogs (inhibitors) that nucleate protein folding(“chemical chaperones”)

Selective chemical chaperones may be effective therapeutics for genetic diseaseif they can rescue folding and function prior to disposal.

Goal: selectively accelerate protein folding

Kim and Arvan (1998) Endocrine Rev. 19, 173-202

Pharmacologic rescue of conformationally-defectiveproteins: implications for the treatment of human disease

Pharmacologic chaperones (or ‘‘pharmacoperones’’): chemical mimics of in vivo ligands,arrest or reverse loss-of-function diseases by inducing mutant proteins to adopt native type-like conformations leading to normal pattern of cellular localization and function (focus ongonadotropin-releasing hormone receptor).

Ulloa-Aguirre et al (2004) Traffic 5, 821-837


Pharmacologic Rescue of Conformationally-DefectiveProteins: Implications for the Treatment of Human Disease


• Figure on the right shows inositol phosphate production (IP) by COS-7 cells transiently expressing each receptor

• Hydrophobic peptidomimetics penetrate cells and interact specifically with protein targets, including antagonist of GnRH such as IN3 (132–134)

• Pharmacological chaperones tested were the indoles IN30, IN31b, and IN3 (left figure).

• All peptidomimetics studied with an IC50 value for the human GnRHRof 2.3 nM displayed a measurable efficacy in rescuing GnRHR mutants

• Results from four rat GnRHRs and C278A rat GnRHR are also shown.

Pharmacological rescue of WT and mutant human GnRH receptors

Glycans and cellular functions I - protein folding, ER processing, quality...

Documents

Transcript of Glycans and cellular functions I - protein folding, ER processing, quality...