synaptotagmin-3 drives ampa receptor endocytosis ......ankit awasthi 1, binu ramachandran , saheeb...

RESEARCH ARTICLE SUMMARY

NEUROSCIENCE

Synaptotagmin-3 drives AMPAreceptor endocytosis depression ofsynapse strength and forgettingAnkit Awasthi Binu Ramachandran Saheeb Ahmed Eva Benito Yo ShinodaNoam Nitzan Alina Heukamp Sabine Rannio Henrik Martens Jonas BarthKatja Burk Yu Tian Wang Andre Fischer Camin Deandagger

INTRODUCTIONMemories are stored asmo-lecular and cellular changes in the brain Synap-ses the nodes of connection between neuronscan store memories by virtue of their ability totune the efficacy of communication betweenneurons This property of synaptic plasticitymakes it possible for the brain to store andretrieve memoriesmdashto replay patterns of elec-trical activity that occurred during an impor-tant event Forgetting leads to the inability toretrieve memories by making them latent ordecaying them below any useful quality How-ever what determines whether a memory isforgotten A mechanism is the regulation ofneurotransmitter receptor numbers on the post-

synaptic plasma membrane These receptorsmediate synaptic transmission by transduc-ing presynaptically released neurotransmittersinto an electrical signal Neuronal activitystrengthens synapses by inserting receptors orweakens synapses by removing receptorsfrom the postsynaptic membrane Receptortrafficking is controlled through calcium in-flux into the neuron however the calciumsensors mediating this control are not known

RATIONALE Synaptotagmin proteins sensecalcium to trigger membrane fusion Stimulat-ing neuronal cultures elicits a calcium-mediatedexternalization of most synaptotagmin isoforms

into plasmamembranes but synaptotagmin-3(Syt3) internalizes frompostsynapticmembranesStimulating AMPA (a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic) or NMDA (N-methyl D-aspartate) receptors induces internalization ofAMPA receptors (which mediate most of thefast synaptic transmission in the brain) andSyt3This raised the intriguing possibility that Syt3mediates activity-induced internalizationof recep-tors to weaken synapses and cause forgettingWe imaged Syt3 using an isoform-specific anti-body tested its role in receptor trafficking usingelectrophysiological methods in brain slicesand neuronal cultures and tested its role inforgetting using spatial memory tasks in mice

RESULTS Syt3 is on postsynaptic membranesat endocytic zones which are clathrin-richregions close to the postsynaptic density Syt3binds the GluA2 AMPA receptor subunit and

alsobindsAP2andBRAG2two proteins implicatedin activity-dependent in-ternalization of AMPAreceptors via clathrin-mediated endocytosisSyt3 does not affect basal

AMPA receptor trafficking However knock-ing out Syt3mdashor expressing calcium-bindingndashdeficient Syt3mdashabolishes AMPA receptorinternalization induced by AMPA NMDA orelectrophysiological stimulation of long-termdepression of synaptic strength It also blocksthe AMPA receptor internalization that nor-mally decays long-term potentiation of synap-tic strength These effects are mimicked in awild-type background through acute appli-cation of the Tat-GluA2-3Y peptide whichcompetitively inhibits binding of Syt3 to atyrosine-rich (3Y) motif on the cytoplasmictail of GluA2 In spatial memory tasks mice inwhich Syt3 was knocked out (Syt3 knockoutmice) learn escape positions normally butpersevere to previously learned positionswhich can be explained by a lack of forget-ting previously acquired memories Inject-ing the Tat-GluA2-3Y peptide in wild-typemice mimics the lack of forgetting of spatialmemories and this effect is occluded inSyt3 knockout mice

CONCLUSION The persistence or degra-dation of memories is governed by a poorlyunderstood molecular machinery We havediscovered a distinct synaptotagmin isoformthat triggers calcium-mediated internalizationof AMPA receptors resulting in a weakeningof synaptic transmission and forgetting ofspatial memories in mice

RESEARCH

Awasthi et al Science 363 44 (2019) 4 January 2019 1 of 1

The list of author affiliations is available in the full article onlineThese authors contributed equally to this workdaggerCorresponding author Email cdeaneni-gdeCite this article as A Awasthi et al Science 363 eaav1483(2019) DOI 101126scienceaav1483

Wild-type

Syt3 KO

Escape position 2 Previous escape position

Recorded field EPSP (readout of synaptic strength) Dendritic spine

AMPA receptorEscape position 1

Mouse

Syt3 knockout mice do not forget Both wild-type mice and Syt3 knockout mice can learnan escape position in the water maze in which corresponding synapses are strengthenedthrough the increase of AMPA receptors These synapses are weakened by the removalof receptors if the memory is no longer neededmdashfor example when a new escape position islearned Syt3 knockout mice cannot remove receptors and therefore cannot forget previousescape positions

ON OUR WEBSITE

Read the full articleat httpdxdoiorg101126scienceaav1483

on June 18 2020

httpsciencesciencemagorg

Dow

nloaded from

RESEARCH ARTICLE

NEUROSCIENCE

Synaptotagmin-3 drives AMPAreceptor endocytosis depression ofsynapse strength and forgettingAnkit Awasthi1 Binu Ramachandran1 Saheeb Ahmed1dagger Eva Benito23 Yo Shinoda1DaggerNoam Nitzan1 Alina Heukamp1sect Sabine Rannio1 Henrik Martens4 Jonas Barth23Katja Burk1 Yu Tian Wang5 Andre Fischer23 Camin Dean1para

Forgetting is important Without it the relative importance of acquired memories in achanging environment is lost We discovered that synaptotagmin-3 (Syt3) localizes topostsynaptic endocytic zones and removes AMPA receptors from synaptic plasmamembranes in response to stimulation AMPA receptor internalization long-termdepression (LTD) and decay of long-term potentiation (LTP) of synaptic strength requiredcalcium-sensing by Syt3 and were abolished through Syt3 knockout In spatial memorytasks mice in which Syt3 was knocked out learned normally but exhibited a lack offorgetting Disrupting Syt3GluA2 binding in a wild-type background mimicked the lack ofLTP decay and lack of forgetting and these effects were occluded in the Syt3 knockoutbackground Our findings provide evidence for a molecular mechanism in which Syt3internalizes AMPA receptors to depress synaptic strength and promote forgetting

Afundamental property of the brain is theability to learn from experience by mod-ulating the strength of synaptic connec-tions between neurons The trafficking ofAMPA receptors to and from the surface

of postsynaptic membranes is a key determinantin the regulation of synaptic strength (1ndash3) High-frequency stimulation increases surface recep-tors and promotes long-term potentiation (LTP)of synaptic strength Low-frequency stimulationremoves receptors from the postsynaptic mem-brane and causes long-term depression (LTD) ofsynaptic strength This phenomenon has beenmost extensively studied inN-methyl-D-aspartate(NMDA)ndashreceptorndashdependent plasticity of CA3-CA1 synapses in the hippocampus (4) LTP iswidely believed to underlie learning (5ndash7) Con-versely weakening of potentiated synapses andLTD can promote forgetting (5 8 9)Both LTD (10 11) and the decay of LTP depend

on activity-dependent removal of postsynaptic

GluA2-containing AMPA receptors from the plas-ma membrane (8 12 13) Ca2+ influx into den-drites is critical for virtually all types of synapticplasticity (4) including decay of LTP (14 15) andactive internalization of GluA2-containing AMPAreceptors (16 17) The same signaling entity canhave divergent consequences for the synapse Fasthigh [Ca2+] influx can promote LTP whereasgradual low [Ca2+] influx can promote LTD (4)Plasma membranendashlocalized Ca2+-binding pro-teins are likely needed to remove or add receptorsto the postsynaptic membrane through regulatedendo- or exocytosis Synaptotagmins (Syts) arecandidates for such a function (18) This familyof integral membrane proteins contain a shortluminal tail transmembrane domain and twoconserved cytoplasmic Ca2+-binding C2 domains(19) that regulate Ca2+-dependent membrane re-cycling eventsOf the 17 mammalian Syt isoforms Syt3 is the

third most abundant in the brain Unlike Syt1and Syt2 which are predominantly thought to belocalized to synaptic vesicle membranes Syt3 wasreported to be localized to the plasma membrane(20 21) and to have a 10-fold higher Ca2+ affinitythan that of Syt1 and Syt2 (22) In a pHluorin-Sytscreen of Syt isoforms in response to stimulationSyt3 exhibited distinct recycling properties It wasthe only isoform to endocytose in response tostimulation and to recycle exclusively in den-drites (23) Because the kinetics of stimulus-induced pHluorin-Syt3 endocytosis resembledthose of pHluorin-GluA2 (24 25) we hypothe-sized that Syt3 may regulate Ca2+- and activity-dependent postsynaptic receptor endocytosis toaffect synaptic plasticity

ResultsSyt3 is on postsynapticplasma membranesTo examine the location of Syt3 we developeda highly specific antibody (fig S1 A and B) thatrecognized a single band in brain homogenateswhich was absent in Syt3 knockouts an antibodydeveloped by Neuromab showed similar results(Fig 1A) Syt3 was most abundant in adiposetissue heart and brain (fig S1C) where it wasfound in the hippocampus cortex thalamus andstriatum (fig S1D) Expression in the brain beganembryonically and remained high throughoutadulthood (fig S1E) Immunostains revealed Syt3signal on neuronal cell bodies and dendrites inthe CA1 (Fig 1B) CA3 and dentate gyrus (fig S1F)in wild-type but not Syt3 knockout hippocam-pal slicesTo localize Syt3 subcellularly we expressed

cytosolic green fluorescent protein (GFP) in cul-tured hippocampal neurons and immunostainedfor Syt3 andMAP2 todistinguish dendrites (MAP2-positive) from axons (GFP-positiveMAP2-negative)Syt3 signal was predominantly detected in den-drites (902 plusmn 30 of total signal) comparedwithaxons (98plusmn51of total signal) (Plt0001) (Fig 1C)Syt3 exhibited a punctate pattern in dendritesand colocalized with the pre- and postsynapticmarkers synaptophysin and PSD95 or GluA1respectively at synapses (Fig 1D) 706 plusmn 53of Syt3 signal was at excitatory synapses (figS1G) and 294 plusmn 19 was at inhibitory synapses(P lt 0001) (fig S1H) In subcellular fractionsSyt3 was associated with synaptosomal plasmamembranes and notwith synaptic vesicles (fig S1I)which is in agreement with a previous report (20)Immuno-organelle isolation of synaptic vesicleswith antibodies to Syt1 or Syb2 further confirmedthat Syt3 isnot present on synaptic vesicles (fig S1J)We further tested whether Syt3 is specifically local-ized to postsynaptic membranes using a trypsincleavage assay of synaptosomes (26) Synapto-somes form in such a way that the presynapticterminal seals trapping synaptic vesicles andother presynaptic components inside (Fig 1E)whereas the postsynaptic membrane does not re-seal Presynaptic proteinswere therefore protectedfrom trypsin cleavage whereas postsynaptic pro-teins including Syt3 were cleaved (Fig 1F)

Stimulation induces endocytosis of Syt3

The presence of Syt3 on postsynaptic membranessuggests it may regulate a post-synaptic re-cycling event Time-lapse imaging of pHluorin-Syt3 in transfected hippocampal neurons duringdepolarization with 45 mMKCl (23) or field stim-ulation revealed a calcium-dependent fluores-cence decrease requiringNMDAAMPA receptorsand L-type calcium channels likely correspondingto endocytosis (fig S2 A to D) PHluorin-Syt3 alsoexhibited calcium-dependent endocytosis in re-sponse to AMPA (Fig 2A) and NMDA (Fig 2B)mdashstimuli that promoteAMPAreceptor internalizationmdashwith kinetics similar to those of pHluorin-taggedAMPA receptors (24 25)Because interpretation of pHluorin experiments

may be confounded by intracellular acidification

RESEARCH

Awasthi et al Science 363 eaav1483 (2019) 4 January 2019 1 of 14

1Trans-synaptic Signaling Group European NeuroscienceInstitute 37077 Goettingen Germany 2German Center forNeurodegenerative Disease 37075 Goettingen Germany3Department of Psychiatry and Psychotherapy UniversityMedical Center Goettingen 37075 Goettingen Germany4Synaptic Systems GmbH 37079 Goettingen Germany 5BrainResearch Center and Department of Medicine University ofBritish Columbia Vancouver BC V6T2B5 CanadaThese authors contributed equally to this workdaggerPresent address Department of Diagnostic and InterventionalRadiology University Medical Center Goettingen 37075 GoettingenGermanyDaggerPresent address Department of Environmental Health School ofPharmacy Tokyo University of Pharmacy and Life Sciences Tokyo192-0392 Japan sectPresent address Department of NeurobiologyWeizmann Institute of Science 7610001 Rehovot IsraelparaCorresponding author Email cdeaneni-gde

on June 18 2020


Dow

nloaded from

in response to AMPA and NMDA stimulation(27) we performed an additional assay to testinternalization of Syt3 We expressed GFP-Syt3in Syt3 knockout neurons and labeled surfaceGFP-Syt3 with an antibody to GFP stimulatedneurons with AMPA or NMDA and then labeled

remaining surface GFP-Syt3 with a secondaryantibody of one color permeabilized cells andlabeled internalized GFP-Syt3 with a secondaryantibody of another color so as to quantify Syt3internalized in response to stimulation Wefound a significant increase in internalized

GFP-Syt3 after stimulation with AMPA andNMDA compared with that in control conditions(Fig 2C)AMPA receptor endocytosis is thought to occur

at postsynaptic endocytic zonesmdashmarked by flu-orescently tagged Clathrin light chainmdashwhich isadjacent to but distinct from postsynaptic den-sities marked by fluorescently tagged Homer(28ndash31) To visualize these markers exclusively atpostsynaptic sites neuronal cultures were sparse-ly transfected so that dendrites of transfectedneurons can be identified and fluorescent markercolocalization quantified at postsynaptic sites inthese dendritesWe first verified that Clathrin andHomer can be distinguished only 44 plusmn 11 ofClathrin light chain-DsRed and Homer-mycpuncta overlapped (Fig 2D) We further foundthat only 40 plusmn 13 of GFP-Syt3 puncta co-localized with Homer-myc at postsynaptic den-sities whereas 840 plusmn 18 of GFP-Syt3 puncta indendrites colocalized with Clathrin-DsRed at en-docytic zones (Fig 2D)

Syt3 internalizes AMPA receptors

Does Syt3 endocytose receptors As a first testwe pulled down binding partners of Syt3 frombrain homogenates Recombinant Syt3 pulleddown GluA2 but not other receptor subunits andalso pulled down the endocytic proteins AP-2(32) and BRAG2 (10)mdashtwo proteins implicated inreceptor endocytosismdashbut not GRIP or PICK1(Fig 2E and fig S3) which bind distinct regionsof the GluA2 C-terminal tail (Fig 2F) (1) Syt3bound GluA2 in a calcium-dependent mannerand bound AP-2 and BRAG2 in the absence ofcalcium and at concentrations up to 10 mM inthe case of AP2 (Fig 2G) Because the ninendashamino acid 3Y tail of GluA2 is important forreceptor internalization we tested whetherSyt3 interacts with this region Incubation ofbrain lysates with 1 mM Tat-GluA2-3Y peptidemdashwhich competitively inhibits protein bindingto the GluA2 3Y region and blocks activity-dependent endocytosis of receptors (10 13)mdashindeed disrupted Syt3GluA2 binding (Fig 2H)Because Syt3 pulled down GluA2 receptors

and GluA1GluA2 heteromers comprise the ma-jority of AMPA receptors in the hippocampus(33) we examined internalization of GluA1- andGluA2-containing receptors in response to stim-ulation (17 32 34) in dissociated hippocampalneurons in which Syt3 was overexpressed orknocked down postsynaptically using sparsetransfection Surface epitopes of GluA1 or GluA2in hippocampal cultures were labeled with pri-mary antibodies followed by stimulation withAMPA or NMDA to promote receptor internal-ization Surface receptors were then labeled witha secondary antibody of one color and internal-ized receptors labeled with a second color afterpermeabilization whereby the ratio of internalsurface signal reports the extent of receptorinternalization Overexpression of GFP-Syt3 in-creased internalization of GluA1 and GluA2whereas Syt3 knockdown or expression of acalcium-bindingndashdeficient mutant of Syt3 abol-ished stimulation-induced internalization of


D

CSyt3 MAP2 GFP

Syt3 Syp PSD95 merge

Syt3 Syp GluA1 merge

BA

70 kD-55 kD-

25 kD-

Wt KO Wt KO

Syt3 NT

Tubulin

Syt3 NB

Syp

no tr

10acute

20acute

30acute

60acute

90acute

Syn

Syb2

Rab3a

SNAP25

Piccolo

Syx1A

PSD95

Homer

GluA1

GluN1

Syt3 NB

Syt3 NT

Nrx

Nlg

F

Syt1

Syp

Rab3a Syn

SNAP25

Piccolo

Syx

Nrx NlgGluN1 GluA1

PSD95Homer

Syb2

Syt3

E

protectedcleaved

Wt

KO

Syt3 MAP2 merge

no tr

10acute

20acute

30acute

60acute

90acute

Fig 1 Syt3 is postsynaptic (A) Syt3 antibodies recognize a specific band of the expectedmolecular weight in wild-type (WT) but not in Syt3 knockout (KO) mouse brain homogenates inWestern blots Syt3 NB is a monoclonal antibody from Neuromab and Syt3 NT is a polyclonalantibody developed by Synaptic Systems Tubulin is a loading control (B) Syt3 immunoreactivity isdetected on pyramidal cell bodies and dendrites in the CA1 region of hippocampal slices but notin Syt3 knockouts MAP2 marks dendrites Scale bar 50 mm (C) Syt3 is predominantly in dendritesin hippocampal neurons transfected with GFP and immunostained with MAP2 to mark dendrites(MAP2-positive) or axons (GFP-positiveMAP2-negative) (n = 20 neurons3 cultures) Scale bar20 mm (left) 5 mm (right) (D) Syt3 localizes to synapses marked by synaptophysin and PSD95 (top)or GluA1 (bottom) in hippocampal cultures Scale bar 5 mm (E) Schematic of a synaptosome Thepresynaptic side reseals whereas the postsynaptic side does not leaving postsynaptic proteinsaccessible to trypsin cleavage (F) Presynaptic proteins are protected from trypsin cleavage whereaspostsynaptic proteins (including Syt3) are cleaved

RESEARCH | RESEARCH ARTICLEon June 18 2020


Dow

nloaded from


inte

nsity

(

)

time (s)

NH4ClNMDA

inte

nsity

(

)AMPA

time (s)

NH4Cl

inte

nsity

(

) 0 Ca2+

NH4ClNMDA

2mM Ca2+

inte

nsity

(

)

time (s)

0 Ca2+ 2 mM Ca2+

NH4ClAMPA

A

00

05

10

20

inte

rnal

sur

face

Syt

3

ctrl

AMPA

NMDA

15

internalized GFP-Syt3

cont

rol

AM

PAN

MD

A

surface GFP-Syt3 mergeC

GFP-Syt3

CLC-DsRed

Homer-GFP GFP-Syt3

Homer-mycCLC-DsRed

Homer-GFP CLC-DsRed GFP-Syt3 CLC-DsRed GFP-Syt3 Homer-mycD

GluA2 C-tail

EFCYFSRAEAKRMKVAKNPQNINPSSSQNSQNFATYKEGYNVYGIESVKI

AP-2 NSF BRAG2 GRIP and PICK1

inpu

tSy

t3 +

GRIP

AP-2

GluN2A

GluA2

Syt3

ndash

GluA1

GluA3

GluN1

GABAARα1

Pick1

BRAG2

FE

G

0 s 100 s 300 s0 s 100 s 300 s

B

GluA2

BRAG2

input

neg

ctrl

EGTA0

Ca2+

10 micro

M C

a2+

100

microM C

a2+

1000

microM

Ca2+

AP2

140

120

100

80

60

40

20

0

140

120

100

80

60

40

20

0 100 200 300

0 50 100

140

120

100

80

60

40

20

00 100 200 300

140

120

100

80

60

40

20

time (s)0 50 100

834 883mdash

input

neg

ctrl

100

microM C

a2+

GluA2

Tat-GluA2-3Y ndash +

H

Fig 2 Syt3 endocytoses in response to stimulation and binds GluA2AP-2 and BRAG2 (A) pHluorin-Syt3 endocytoses in response tostimulation with AMPA or NMDA (B) in 2 mM calcium (left) but not in theabsence of calcium (right) [n ROIneuronscultures AMPA 2 mM Ca2+ 3773 (left) 0 and 2 mM Ca2+ 4263 (right) NMDA 2 mM Ca2+ 3533(left) 0 and 2 mM Ca2+ 4033 (right)] NH4Cl dequenches internalizedpHluorin-Syt3 fluorescence Scale bars 5 mm (C) GFP-Syt3 expressedin Syt3 knockout hippocampal neurons internalizes in response tostimulation with AMPA and NMDA (n = 28 25 and 26 neurons per 2cultures for control AMPA and NMDA respectively one-way ANOVATukeyrsquos test) (D) GFP-Syt3 colocalizes with CLC (clathrin light chain)ndashDsred at postsynaptic endocytic zones and not with Homer-mycat PSDs (n ROIneurons 3420 Homer-GFPCLC-DsRed 2919

GFP-Syt3CLC-Dsred 3418 GFP-Syt3Homer-myc from three cultures)Scale bars 5 mm (E) Recombinant Syt3 C2AB pulls down GluA2 AP2 andBRAG2 from brain homogenates Syt3ndash is beads only (F) Binding siteson the GluA2 C-terminal tail of proteins implicated in receptor recyclingand tested in pull downs (G) Calcium-dependence of Syt3 C2AB pulldown of GluA2 AP2 and BRAG2 from brain homogenate In 0 Ca2+

conditions brains were homogenized and solubilized in calcium-freebuffer but endogenous buffered calcium in internal stores remainsAddition of EGTA eliminates any potential effects of these additionalsources of calcium (H) Western blot of GluA2 pulled down from brainhomogenate by Syt3 C2AB in 100 mM calcium with or withoutpreincubation with 1 mM Tat-GluA2-3Y peptide Negative controls arebeads only



Dow

nloaded from

receptors [Fig 3 A to C and fig S4 A to D Syt3short hairpin RNA (shRNA) knockdown valida-tion is provided in fig S4 E and F]Manipulationof Syt3 did not affect surface versus internalreceptors in basal conditions Syt3 knockoutneurons yielded similar results GluA2 receptorswere internalized in response to AMPA andNMDA stimulation in wild-type neurons butnot in Syt3 knockout neurons or in wild-typeneurons treated with the Tat-GluA2-3Y peptide(Fig 3 D and E and fig S4G)

Syt3 does not affect basal transmissionWe found no change in miniature excitatorypostsynaptic current (mEPSC) amplitude fre-quency or decay time in Syt3 overexpressingor knockdown neurons compared with GFP-transfected controls (decay time GFP 23 plusmn 02 msGFP-Syt3 21 plusmn 02 ms Syt3 KD 19 plusmn 02 ms)(fig S5 A to D) further confirming that post-synaptic Syt3 does not affect surface receptornumber in basal conditions To exclude pre-synaptic effects we also compared mEPSCs in

wild-type and Syt3 knockout cultures and againfound no difference in mEPSC amplitude orfrequency (frequency P = 02671 t test Welchrsquoscorrection) (fig S5 E to G) After stimulationhowever mEPSC amplitude decreased in wild-type but not in Syt3 knockout neurons and thiseffect was rescued by reintroducing GFP-Syt3postsynaptically into knockout neurons (fig S5H)confirming a lack of stimulation-induced inter-nalization of synaptic AMPA receptors in Syt3knockouts


00

05

10

15

20

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

inte

rnal

su

rfac

e G

luA

1

WT WTSyt3 KO Syt3 KO WT+GluA2-3Y

control AMPAGFP-Syt3 Syt3 KD Syt3 Ca2+ mutGFP GFP-Syt3 Syt3 KD GFP

control AMPA

EG

FP

Glu

A2

surf

ace

Glu

A2

inte

rnal

Glu

A2

surf

ace

Glu

A2

inte

rnal

control AMPA NMDAcontrol AMPA NMDA

GFP

Syt3 KDGFP-Syt3

Syt3 Ca2+ mut

WTSyt3 KOWT+ 3Y

control AMPA NMDA

A

CB

D E

Syt3 Ca2+ mut

Fig 3 Syt3 internalizes AMPA receptors in response to stimulation(A) Hippocampal neurons transfected with GFP GFP-Syt3 Syt3 knockdown(Syt3 KD) or Syt3 calcium-binding deficient mutant (Syt3 Ca2+ mut)constructs in control and AMPA-stimulated conditions immunostained forsurface and internalized GluA2 receptors (B) Stimulation-induced GluA2 andGluA1 (C) internalization is increased by overexpression of GFP-Syt3 andblocked by Syt3 KD or Syt3 Ca2+ mut (n neurons for GluA2GluA1 GFP

6469 ctrl 6846 AMPA 2847 NMDA GFP-Syt3 4665 ctrl 4647 AMPA2343 NMDA Syt3 KD 5141 ctrl 6938 AMPA 2324 NMDA Syt3 Ca2+mut4237 ctrl 2836 AMPA 2731 NMDA from six cultures) (D and E)Stimulation-induced GluA2 internalization is abolished in Syt3 knockoutneurons and in WTneurons treated with the Tat-GluA2-3Ypeptide (n Syt3 KOWTneurons 3129 ctrl 2422 AMPA 1817 NMDAWT+GluA2-3Y 18 AMPAfrom four cultures) Scale bars 10 mm two-way ANOVA Dunnettrsquos test



Dow

nloaded from

Basal synaptic transmissionwas also unchangedin Syt3 knockouts Recordings of synaptic re-sponses from CA1 neurons elicited by Schaffercollateral stimulation in acute hippocampalslices displayed no change in NMDAAMPAg-aminobutyric acid (GABA)AMPA or GABANMDA receptorndashmediated response ratio in Syt3knockout compared with wild-type neurons(fig S5 I to L) There was also no change in theAMPA EPSC decay time (wild type 194 plusmn 4 msknockout 219 plusmn 46 ms) or GABA inhibitorypostsynaptic current (IPSC) decay time (wild type76 plusmn 49 ms knockout 588 plusmn 7 ms) In additionwe found no change in stimulus intensity versusfEPSP (field excitatory postsynaptic potential)slope or fiber volley amplitude (fig S5 M andN) and no change in paired pulse ratio (fig S5O)suggesting that Syt3 does not affect the numberof synaptic inputs from CA3 onto CA1 neurons orshort-term presynaptic plasticity

Syt3 promotes LTD and decay of LTP

LTDmdashwhich requires activity-dependent endo-cytosis of AMPA receptorsmdashwas abolished in Syt3knockouts (Fig 4A) which is consistent withdefective activity-dependent AMPA receptor en-docytosis in the absence of Syt3 In wild-typeslices application of the GluA2-3Y peptide whichdisrupts Syt3 binding to the GluA2 cytoplasmictail mimicked the Syt3 phenotype and abolishedLTD (Fig 4A) (8 11 35)The decay of LTP induced by means of a

1XTET stimulation (a single tetanus of 16 pulsesat 100 Hz) also depends on receptor internal-ization (13) LTP induced by means of a 1XTETstimulus decayed within 1 hour in wild-typeslices (36) but was reinforced and remained po-tentiated in Syt3 knockouts (Fig 4B) this formof LTP was NMDA receptorndashdependent in bothwild-type and Syt3 knockout slices (fig S6 Aand B) The GluA2-3Y peptide again mimickedthe Syt3 knockout phenotype and reinforceddecaying LTP (Fig 4C) This reinforced decay-ing LTP was insensitive to ZIP (PKMz inhib-itory peptide) (Fig 4C) which blocks the activityof atypical protein kinases and has the unusualproperty of reversing LTP after induction throughendocytosis of AMPA receptors (12 13 37 38)Reinforced decaying LTP in Syt3 knockouts wassimilarly ZIP-insensitive (Fig 4D) The GluA2-3Ypeptide had no effect on the reinforced decayingLTP in Syt3 knockouts confirming that Syt3decays LTP by acting on the GluA2-3Y region(Fig 4D)Nondecaying LTP induced bymeans of 3XTET

stimulation (three tetanizing trains of 100 pulsesat 100 Hz) was unchanged in Syt3 knockouthippocampal slices compared with wild-typeslices (Fig 4E) However ZIP promoted decayof 3XTET-induced LTP in wild-type slices butnot in Syt3 knockout slices (Fig 4F) furtherindicating a defect in activity-dependent recep-tor internalization in Syt3 knockouts BecauseZIP had no effect on LTP in Syt3 knockoutslices it did not appear to cause excitotoxicityor neural silencing (39 40) Analysis of stimula-tion frequencyndashpotentiation dependence revealed

increased potentiation in Syt3 knockouts com-pared with wild-type slices (Fig 4G)

To test whether calcium-sensing by post-synaptic Syt3 is important for receptor internal-ization in these plasticity paradigms we injectedwild-type or calcium-bindingndashdeficient mutantSyt3 AAV12 postsynaptically into the dorsal CA1region of the hippocampus of mice in which Syt3had been knocked out (Syt3 knockout mice)(Fig 4H) We found that LTD (Fig 4I) decaying1XTET-induced LTP (Fig 4J) and ZIP-mediateddecay of 3XTET LTP (Fig 4K) were rescued byexpression of wild-type Syt3 but not calcium-bindingndashdeficient Syt3

Syt3 knockout mice learn normally buthave impaired forgetting

Because Syt3 knockoutmice had normal LTP butlacked LTD we hypothesized that they wouldlearn normally but have deficits in forgetting Totest this we used thewatermaze spatial memorytask that involves the CA1 hippocampal sub-region in which mice learn to navigate to a hid-den platform position over sequential days oftraining using visual cues Syt3 knockout miceshowed no difference in anxiety or hyperactivity(fig S7 A to C) but had a 15 decrease in bodyweight and faster swim speed compared withthose of wild-type mice (fig S7 D and E) Wetherefore plotted proximity to the platform inaddition to the traditional escape latencyparameter to control for possible effects ofswim speed The maximum average proximitydifferencemdashbetween closest (after training of allmice to a platform position) and farthest possibleproximity (by using the same data but calculat-ing proximity to a platform in the oppositequadrant)mdashwas 125 cm Syt3 knockout and wild-type mice learned the position of the hiddenplatform equally well there was no significantdifference in proximity to the platform (Fig 5A)or escape latency (fig S7F) during training Whenthe platform was removed in probe tests aftertraining to test how well the mice have learnedthe platform position the percentage of timespent in the target quadrant was well abovechance level for both Syt3 knockout and wild-type mice (Fig 5B) and proximity to the plat-form position (Fig 5C) and platform positioncrossings (Fig 5D) were similar Syt3 knockoutmice in cohort 1 (of two cohorts tested) had sig-nificantly lower proximities and more platformcrossings in the first probe test compared withthose of wild-type mice but no difference in thesecond probe test Thus Syt3 knockout micelearned the task as well or slightly faster than didwild-type miceWe observed an indication of a lack of forget-

ting in the second probe test after learning Syt3knockout mice exhibited a lack of ldquowithin-trialrdquoextinction Wild-type mice showed maximalsearching (proximity to the platform) in the first10 to 20 s of the probe test and then graduallyshifted to a dispersed search pattern of otherregions of the pool (41) whereas Syt3 knockoutmice continued to persevere to the platformposition even near the end of the trial (Fig 5E)

When the platform was shifted to the oppositequadrant in reversal training Syt3 knockoutmice learned the new platform position as wellas didwild-typemice in the probe test (probe test3) (Fig 5 B C and D) but continued to persevereto the original platform position during reversaltraining (fig S7G) and in the probe test afterreversal (Fig 5F) exhibiting a lack of forgettingof the previous platform positionIn a fourth cohort we extended the training

days after reversal (fig S7H) Syt3 knockout micepersevered to the previous platform position sig-nificantly more than did wild-type mice evenafter 7 days of reversal training (Fig 5G) Injec-tion of Syt3 AAV12 specifically into the dorsalCA1 (postsynaptic) region of the hippocampusof Syt3 knockout mice rescued this perseverencephenotype comparedwith Syt3 knockout or wild-type mice injected with control GFP AAV12(Fig 5G)To test whether the lack of forgetting exhibited

by Syt3 knockouts is due to defective AMPA re-ceptor internalization we trained wild-type andSyt3 knockout mice to learn a platform positionin the water maze during habituation to dailysaline intraperitoneal injections (fig S7I)We theninjected the Tat-GluA2-3Y peptide (5 mmolkgintraperitoneally)mdashwhich disrupts Syt3GluA2binding and blocks AMPA receptor internal-ization (11ndash13 35)mdash1 hour before the probe testafter training to the initial platform positionand then daily 1 hour before reversal trainingto a new platform position in the opposite qua-drant In the probe test after training to theinitial platform position peptide-injected wild-type and Syt3 knockout mice showed a similarlack of within-trial extinction and continuedto persevere to the platform position whereassaline-injected wild-type mice shifted to a dis-persed search pattern near the end of the trial(Fig 5H and fig S7J) Similarly in probe testsafter reversal training to a newplatformpositionwild-type mice injected with the Tat-GluA2-3Ypeptide persevered to the original platform posi-tion significantly more than did saline-injectedwild-type mice [P = 0042 one-way analysis ofvariance (ANOVA) Bonferronirsquos correction] oruninjected wild-type mice in previous cohorts(P = 0039) and to a similar extent as did Tat-GluA2-3Y peptidendashinjected Syt3 knockout mice(P = 0096) or uninjected Syt3 knockout micein previous cohorts (P = 0105) (Fig 5I) ThusTat-GluA2-3Y peptide injection mimics the Syt3knockout phenotype There was no differencein perseverence between Tat-GluA2-3Yndashinjectedand uninjected Syt3 knockouts in previous co-horts (P = 0184) indicating that the effect ofTat-GluA2-3Y peptide injection is occluded inSyt3 knockout miceTo further test whether the lack of forgetting

phenotype of Syt3 knockouts is due to a lack ofreceptor internalization via Syt3GluA2 interac-tion we tested spatial memory in the Barnesmaze in which mice learn to navigate to 1 of 20holes around the perimeter of a circular platformleading to an escape cage using visual cues Timespent in the target hole area during training and




Dow

nloaded from

in subsequent probe tests gives a readout ofspatial memory We tested four groups simulta-neously wild-type mice and Syt3 knockout miceinjected with saline or GluA2-3Y peptide duringldquoreversalrdquo training to a new hole after initiallearning of an original hole positionWepredicted

that (i) Syt3 knockout mice would exhibit a lackof forgettingmdashpersevere more to the originalhole after reversal as compared with wild-type(ii) disruption of Syt3GluA2 interaction throughinjection of the GluA2-3Y peptide in wild-typemice would mimic the lack of forgetting pheno-

type of Syt3 knockouts and (iii) injection of theGluA2-3Y peptide in Syt3 knockout mice wouldhave no effect on their lack of forgettingIn initial training all four groups (injected

intraperitoneally daily with saline only 1 hourbefore training) learned the target hole equally


Fig 4 LTD and decay of LTP is abolished inSyt3 knockouts (A) LTD is abolished in Syt3KO and Tat-GluA2-3Y peptide treated WThippocampal slices (n slicesmice 128 Syt3KO 86 WT and 85 WT + Tat-GluA2-3Y)P lt 0001 for Syt3 KOWT P lt 005 for WT+GluA2-3YWT (B) 1XTET-induced LTP is rein-forced in Syt3 KO slices compared with WT(n = 107 Syt3 KO 1010 WT) P lt 005(C) 1XTET-induced LTP in WT slices treated withthe Tat-GluA2-3Y peptide is reinforced andZIP-insensitive (n = 77) (D) Reinforced 1XTET-induced LTP in Syt3 KOs is unchanged by theTat-GluA2-3Y peptide and is ZIP-insensitive(n = 66 Syt3 KO 77 Syt3 KO + GluA2-3Ypeptide 77 Syt3 KO + ZIP) (E) 3XTET-inducedLTP is unchanged in Syt3 KO compared withWT (n = 66) (F) 3XTET-induced LTP in Syt3KOs is ZIP-insensitive (n = 66) P lt 001(G) fEPSP changes in Syt3 KO and WT slicesinduced by different stimulation frequenciesrecorded 30 min after stimulation [Syt3 KOWTn = 912 (02 Hz) 912 (1 Hz) 127 (10 Hz)and 1414 (100 Hz)] (H) GFP fluorescence in ahippocampal slice from a mouse injected withAAV12 Syt3-P2A-GFP in the dorsal CA1 region(I) LTD in Syt3 KOs is rescued by hippocampalCA1 AAV12 injection of WT Syt3 (n = 1311)but not calcium-binding deficient Syt3 (Ca2+ mut)(n = 86) P lt 001 (J) Decay of 1XTET-inducedLTP in Syt3 KOs is rescued by WT (n = 86)but not Ca2+ mut Syt3 (n = 76 mice) Plt001(K) ZIP-mediated decay of 3XTET-induced LTPin Syt3 KOs is rescued by WT (n = 55) butnot Ca2+ mut Syt3 (n = 66) P lt 005Mann-Whitney U test or Kruskal-Wallis testwith Dunnrsquos test for multiple comparisonsn = slicesmice

1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )

Syt3 KO + WT Syt3Syt3 KO + Ca2+ mut Syt3

5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08

WT + 3YKO + salKO + 3Y

Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D

PT1 PT2 PT3 PT1 PT2 PT3 PT1 PT2 PT3

WTKO

WT + GFPSyt3 KO + GFP

Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3

WT+sal WT + 3Y KO + 3YKO+sal

PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal

Fig 5 Syt3 knockout mice learn as well as wild-type mice but haveimpaired forgetting (A) Syt3 KO and WTmice had similar proximity to theplatform during training (genotype effect P = 01052 two-way ANOVABonferronirsquos test) and in probe tests exhibited similar percent of time in(B) target quadrant (C) proximity to platform and (D) platform crossings KOsin cohort 1 had lower proximities than those of WT in probe tests (P = 0027Studentrsquos t test) (E) Syt3 KO mice lack within-trial-extinction in time-binnedaverage occupancy plots of all mice and proximity to the platform position(red in schematics) in probe test 2 (F) Syt3 KOmice persevere to the previousplatform position (orange) in the probe test after reversal training (to thered platform position) (n = 913 Syt3 KO and 1014 WTmice for cohort1cohort 2) Studentrsquos t testWelchrsquos correction in (B) (C) (E) and (F) andMann-Whitney U test in (D) (G) Expression of Syt3 in the hippocampalCA1 region rescues the perseverence phenotype of Syt3 KOs in training afterreversal two-way ANOVATukeyrsquos test Black red and blue asterisks aresignificance between WT+GFPSyt3 KO+Syt3 Syt3 KO+GFPWT+GFP andSyt3 KO+GFPSyt3 KO+Syt3 respectively (n = 10 Syt3 KO+Syt3 7 Syt3

KO+GFP and 15 WT+GFP) (H) Injection of the Tat-GluA2-3Ypeptide mimicsthe lack of within-trial-extinction in probe test 2 after initial training and(I) perseverence to the previous platform position in probe tests after reversaltraining exhibited by Syt3 knockouts (n = 8 WTsaline 7 WT+3Y and 10 Syt3KO+3Y) one-way ANOVA Bonferronirsquos correction (J) All mice learned(decreased latency to) the escape hole in the Barnes maze equally well duringtraining (two-way ANOVATukeyrsquos test) and (K) (top) in probe tests 1 and 2after training (Middle) In reversal training and (bottom) probe tests 4 to 6 afterreversal trainingWTsaline mice learned the new hole position (red) butWT+3Y KO saline and KO+3Ymice persevered to the previous hole position(orange) (L) WT+3Y KO saline and KO+3Ymice had a higher perseverenceratio [time spent exploring original hole (orange) divided by total time spentexploring original and reversal (red) holes] as compared with that of WTsaline mice (n = 10 WTsaline 11 WT+3Y 11 Syt3 KO saline 12 Syt3 KO+3YP = 0060 for WTsalineSyt3 KO+3Y in probe tests 4 to 6) one-way ANOVABonferronirsquos correction All occupancy plots show average search pathdensities across all mice in a group



Dow

nloaded from

well (Fig 5J) and spent themajority of their timeexploring the target hole region in probe tests(Fig 5K top) Beginning on the day of probe test2 and during reversal training in which thetarget hole was positioned in a different quad-rant mice were injected with saline or GluA2-3Ypeptide 1 hour before training each day Saline-injected Syt3 knockout mice and GluA2-3Y-injected wild-type and Syt3 knockout mice allshowed a similar increased perseverence to the

original hole region as compared with wild-typesaline-injectedmice throughout reversal training(Fig 5K middle) and in probe tests after reversaltraining (Fig 5 K bottom and L) which is inagreement with our prediction

Syt3 knockouts have impaired workingmemory owing to lack of forgetting

To further test deficits in forgetting exhibitedby Syt3 knockout mice we examined working

memory in the delayedmatching to place (DMP)water maze task in which the platform positionis moved to a new position each day (42) Eachday the mice have four trials to learn the newplatform position and forget the previous plat-form position We first trained the mice to avisible platform for 3 consecutive days in whichSyt3 knockout and wild-type mice performedequally well (fig S8A) Because swimming speedof Syt3 knockout mice was again faster than that


WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0

Block 1 Block 2 Block 3 Block 4inter-trial interval = 75 min

trial 2 probe test

Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8 Day 9 Day 10 Day 11 Day 12 Day 13 Day 14 Day 15 Day 16

inter-trial interval = 5 min

KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT

Fig 6 Syt3 knockout mice show deficits in the delayed matching toplace task because of impaired forgetting (A) WTmice had a closerproximity to the platform in trials 2 to 4 relative to trial 1 (higher proximitysavings) as compared with that of Syt3 KO mice indicating that Syt3 KOs haddeficits in finding the platform when it appeared in a new position each dayHidden platform positions presented each day are indicated above the graphsBlue-outlined mazes indicate counter-balancing in which half the cohort istrained to one of the positions and the other half is trained to the otherposition to avoid biased search of inner- or outer-platform positions (n = 14Syt3 KO and 20 WTmice) Studentrsquos t test (B) Occupancy plots of individual

trials averaged across all mice on training days and in the probe test aftertraining reveal impaired forgetting in Syt3 KO micemdashhigher perseverence toprevious daysrsquo platform positions as compared with that of WT In mazeschematics the current dayrsquos platform is red the previous dayrsquos is orange andthe platform position 2 days previous is yellow (C) Syt3 KOs have significantlymore perseverence trials compared with those of WT in strategy analysis (P =00383 unpaired t test Welchrsquos correction) (D) In the probe test on day 16Syt3 KO mice persevere more (have closer proximity as compared with WT) toall previous positions V1 V2 and V3 indicate visible platform training daystwo-way ANOVA for genotype effect P lt 00001 Bonferronirsquos test P lt 005



Dow

nloaded from

of wild-type mice (fig S8B) we quantified prox-imity savings in hidden platform training(Fig 6A) In the first 4 days of training micebecame accustomed to the task In the second4-day block mice had learned the task and wild-type mice performed significantly better thanSyt3 knockouts In the third block from day 10onward the intertrial interval between trial 1 andtrial 2 was increased from 5 to 75 min and wild-type mice continued to perform better than Syt3knockout mice this difference was most pro-nounced in the fourth 4-day block of training(Fig 6A) Quantitation of escape latency savings(fig S8D) and path savings (fig S8E) yieldedsimilar results To test whether the poor perform-ance of Syt3 knockouts was due to difficultyforgetting previous platform positions we exam-ined perseverance to previous positions Occu-pancy plots indicated that Syt3 knockouts indeedpersevered to previous platform positions morethan did wild-typemice throughout training andsampled most previous platform positions in theprobe test whereaswild-typemice adopted amorerandom search pattern (Fig 6B) Syt3 knockoutmice had significantly more trials classified asperseverence (43) to the previous dayrsquos platformposition as compared with wild-type mice acrossall days of training (P = 00383 unpaired t testWelchrsquos correction) (Fig 6C)Neither Syt3 knockoutnor wild-type mice exhibited chaining (43) anonspatial search strategy for platform positionsat two fixed distances from the wall (fig S8C)Syt3 knockout mice also had a significantlycloser proximity to all previous platform posi-tions in the probe test at the end of the 16-daytraining period compared with that of wild-typemice (Fig 6D)

Discussion

We found that activity-dependent AMPA recep-tor internalization LTD decay of LTP and for-getting of spatial memories requires Syt3 Inrescue experiments calcium-sensing by post-synaptic Syt3 was required for LTD and thedecay of LTP The GluA2-3Y peptide competi-tively inhibited Syt3 binding to the 3Y region ofthe GluA2 receptor tail Introduction of thispeptide in a wild-type backgroundmimicked theSyt3 knockout phenotypes of lack of activity-dependent AMPA receptor internalization LTDdecay of LTP and forgetting Effects of the pep-tide were occluded in Syt3 knockout miceimplicating Syt3 in a GluA2-3Yndashdependent mech-anism of AMPA receptor internalizationOur data give rise to amodel in which Syt3 at

postsynaptic endocytic zones is bound to AP-2and BRAG2 in the absence of calcium GluA2could then accumulate at endocytic zones bybinding Syt3 in response to increased calciumduring neuronal activity This would potentiallybring GluA2 into close proximity to BRAG2where a transient interaction could activateBRAG2 and Arf6 and promote endocytosis ofreceptors via clathrin and AP-2 (10 32) PICK1 isalso important for AMPA receptor endocytosisraising the question of the interplay of Syt3 andPICK1 Two possiblemechanisms are conceivable

PICK1 could sequester receptors internally afterendocytosis (44) and act downstream of Syt3Alternatively PICK1 could transiently bind GluA2and then AP-2 after stimulation to cluster AMPAreceptors at endocytic zones and promote theirsubsequent internalization (45) Thus it is alsopossible that PICK1 acts upstream of or in concertwith Syt3 to bring GluA2 to endocytic zonesBlockade of postsynaptic expression of Syt1

Syt7 was recently reported to abolish LTP buthave no effect on LTD (18) Thus distinct Sytsmay insert and remove receptors from the post-synaptic membrane to mediate LTP and LTDrespectively Syts may regulate both exo- andendocytosis (46) but be predisposed to one orthe other depending on their calcium affinityEndocytosis occurs on slower time scales thandoes exocytosis As calcium concentration de-clines high-affinity Syts such as Syt3 (22) mayremain active whereas low-affinity Syts suchas Syt1 inactivate Alternatively Syts predisposedto endocytosis may interact with protein com-plexes that extend association with Ca2+ or bindproteins in resting conditions that are releasedupon Ca2+ binding to cause endocytosisAlthough the ability to remember is often

regarded as the most important aspect of mem-ory forgetting is equally important Deficits inforgetting can have severe consequences and leadto posttraumatic stress disorder for example Ourdata identify Syt3 as a molecule important for aforgetting mechanism by which AMPA receptorsare internalized to promote LTD and decay of LTP

Materials and methodsAnimals

Use of mice for experimentation was approvedand performed according to the specifications ofthe Institutional Animal Care and Ethics Com-mittees of Goumlttingen University (T1031) andGerman animalwelfare laws Animal sample sizewas estimated by experimental literature andpower analysis (G Power version 31) to reduceanimal number where possible while maintainingstatistical power The Syt3 and Syt6 knock-out andSyt5 and Syt10 knock-in quadruple targeted muta-tion mice (B6129-Syt6tm1Sud Syt5tm1Sud Syt3tm1Sud

Syt10tm1SudJ stockno 008413)were obtained fromThe Jackson Laboratory (wwwjaxorgstrain008413)These mice were then crossed with Black6Jmice obtained from Charles River to isolate micewith homozygous knock-out alleles of Syt3 butWT alleles of Syt5 Syt6 and Syt10 The genotypesof all breeder pairs and mice used for behavioralexperiments were confirmed by PCR

Dissociated hippocampal culture andneuronal transfection

Rat hippocampi were isolated from E18-19Wistarrats as described previously (47) Hippocampiwere treated with trypsin (Sigma) for 20 min at37degC triturated to dissociate cells plated at40000 cellscm2 on poly-D-lysine (Sigma)ndashcoatedcoverslips (Carolina Biologicals) and cultured inNeurobasal medium supplemented with 2 B-27and 2 mM Glutamax (GibcoInvitrogen) Cul-tures were fixed at 12 to 18 DIV with 4 para-

formaldehyde and immunostained with primaryantibodies followed by secondary labeling withAlexa dyes for confocal imagingHippocampi from P0 mouse brains were dis-

sected in ice colddissectionmedium(HBSS (Gibco)20 mM HEPES (Gibco) 15 mM CaCl2 10 mMMgCl2 pH adjusted to 74 with NaOH) andthen incubated in a papain digestion solution(dissectionmedium 02mgml L-Cysteine 05mMNaEDTA 1 mM CaCl2 3 mM NaOH 1 Papainequilibriated in 37degC and 5 CO2 + 01 mgmlDNAseI) for 30 min at 37degC and 5 CO2 Papainwas inactivated with serum medium [DMEM2 mM Glutamax 5 FBS 1X Mito+ supplements(VWR) 05XMEM vitamins (Gibco)] + 25 mgmlBSA + 01 mgml DNAseI followed by 3 washeswith serum medium After trituration in serummedium the cell suspension was centrifuged for5min at 500timesgat room temperature The cell pelletwas resuspended in plating medium (Neurobasalmedium 100 Uml PenStrep 2 B27 0049 mML-aspartate 005 mM L-glutamate 2 mM Gluta-max 10 serum medium) Cells were plated at60000 cellscm2 on 12 mm diameter acid-etchedglass coverslips coated with 004 Polyethylenei-mine (PEI Sigma) Glial growth was blocked onDIV4 with 200 mM FUDR 50 conditionedmedium was replaced with feeding medium(Neurobasal 2 mM Glutamax 100 Uml PenStrep 2 B-27) on DIV7For transfection of neurons on 12 mm cover-

slips in 24-well plates the culture medium fromeach well was exchanged with 400 ml Neurobasalmedium the original culturemediumwas storedat 37degC and 5 CO2 1 mg DNA50 ml Neurobasalmedium and 1 ml Lipofectamine 2000 (Gibco)50 ml Neurobasal medium were incubated sepa-rately for 5 min mixed and incubated for 20additional min before adding the mixture toneurons in a well Following incubation for2 hours neurons were washed once with Neuro-basal and conditioned media was replaced

HEK cell culture transfectionimmunostaining and Western blots forSyt3 knockdown validation

Human embryonic kidney (HEK) 293 cells (pur-chased from American Type Culture CollectionCRL-3216 not tested for mycoplasma) culturedin DMEM supplemented with 10 FBS on 10 cmdishes were transfected at 60 to 80 confluenceusing the calcium-phosphate method 20 mg plas-mid DNA in 360 ml dH2O was mixed with 40 ml25 M CaCl2 followed by addition of an equalvolume (400 ml) of transfection buffer (280 mMNaCl 15mMNa2HPO4 50mMHEPES pH 705)This mixture was incubated at room temper-ature for 20min and 800 ml added per dish 24 to48 hours later cells were harvested for Westernblots in lysis buffer (50 mM Tris-HCl pH 75150 mM NaCl 2 mM EDTA 05 NP40 andprotease inhibitors) For immunostainingHEKcellsgrowing on 12 mm coverslips were transfectedwith Lipofectamine 2000 (Invitrogen) by incu-bating 1 mg DNA50 ml medium and 1 ml Lipo-fectamine 200050 ml medium separately for5 min mixing the two solutions together and




Dow

nloaded from

then incubating for an additional 20min beforeadding the mixture to wells of a 24-well plate

Immunohistochemistry

Acute hippocampal slices were prepared from8-week-old mice anesthetized with isofluraneand decapitated The hippocampus was removedand 200 mm thick slices were cut transverselyusing a tissue chopper (Stoelting) in ice-coldartificial cerebrospinal fluid (ACSF) containingin mM 124 NaCl 49 KCl 12 KH2PO4 2 MgSO42 CaCl2 246 NaHCO3 and 10 D-glucose (saturatedwith 95 O2 and 5 CO2 pH 74 ~305 mOsm)Slices were fixed in 4 paraformaldehyde in PBSfor 30 min and washed 3 X 20 min in PBS Afterwashing slices were incubated in antibody buffer(2 donkey serum 01 Triton X-100 and 005NaN3 in PBS) for 30 min at room temperatureThen slices were incubated with primary anti-bodies in antibody buffer overnight at 4degC Sliceswere then washed with PBS 3 X 20 min andincubated with fluorescently tagged secondaryantibodies for 2 hours at room temperature Sliceswere washed 3 X 20 min in PBS and mounted onmicroscope slides with Fluoromount-G (Sigma)and sealed with nail polish Images were collectedusing 10X air and 40X oil immersion objectiveson a Zeiss A1 laser scanning confocal microscopewith Zen software (Carl Zeiss) Digital imageswere processed using Adobe Photoshop software

Antibodies and expression constructs

Antibodies used including species and catalognumber were rabbit Syt3 105133 (Synaptic Sys-tems) andmouse Syt3N27819 (NeuroMab) chickMAP2 C-1382-50 (Biosensis) rabbit GFP ab290(Abcam)mouseSynaptophysin (Syp) 101011 rabbitSynaptobrevin2 (Syb2) 104202 mouse Syt1 105101mouse SNAP25 111011 guinea pig Piccolo 142104guinea pig Homer 160004 guinea pig Beta3-tubulin 302304 rabbit VGluT1 135303 mouseVGAT 131011 mouse Rab-GDI 130011 mouseGephyrin 147011 rabbitGluA1 182003mouseGluA2182111 rabbit GRIP 151003 (Synaptic Systems)mouse GABAARa1 75136 (NeuroMab) rabbitGluA3 17311 (Epitomics) mouse PSD95 MA1-045 (Thermo Scientific) rabbit GluA1 PC246(Calbiochem)mouseGluA2MAB397 rabbitGluN1AB9864 mouse GluN2A MAB5216 (Millipore)rabbit PICK1 PA1073 (Thermo Scientific) mouseSynapsin (Syn) mouse Rab3a mouse Syntaxin1Arabbit EEA1 (provided by Reinhard Jahn MaxPlanck Institute for Biophysical ChemistryGoettingen Germany) rabbit Neurexin (Nrx) rab-bit Neuroligin (Nlg) (provided by Nils BroseMax Planck Institute of Experimental MedicineGoettingen Germany) mouse AP-2 (provided byPietro De Camilli Yale School of Medicine NewHaven CT USA) and rabbit BRAG2 (10) (providedby Hans-Christian Kornau Chariteacute University ofMedicine Berlin Germany) Mammalian expres-sion constructs used were pHluorin-Syt3 aspreviously described (23) GFP-Syt3 and mCherry-Syt3 were subcloned by replacing the pHluorin inpHluorin-Syt3 with GFP or mCherry respectivelyHomer-GFP Homer-myc and Clathrin lightchain (CLC)-DsRed constructs were provided by

Daniel Choquet (30) (University of Bordeaux)The calcium-binding deficientmutant of Syt3 wasgenerated by Genscript by mutagenesis of D386388N and D520 522 N corresponding to thecalcium-binding sites of syt1 D230 232 N andD363 365 N (48) Syt3 shRNA knockdown con-structs transfected in equal amounts were KD1TGCTGTTGACAGTGAGCGACAAGCTCATCGGT-CAGATCAATAGTGAAGCCACAGATGTATTGAT-CTGACCGATGAGCTTGGTGCCTACTGCCTCGGAKD2 TGCTGTTGACAGTGAGCGCAGGTGTCAA-GAGTTCAACGAATAGTGAAGCCACAGATGTA-TTCGTTGAACTCTTGACACCTATGCCTACTGC-CTCGGA and KD3 TGCTGTGACAGTGAGCCAGGATTGTCAGAGAAAGAGAATAGTGAAGCCACA-GATGTATTCTCTTTCTCTGACAATCCTTTGCC-TACTGCCTCGGA in a pGIPZ vector co-expressingturboGFP (Thermo Scientific Openbiosystems)

Receptor internalization assays

Neurons were labeled with primary extracellularantibodies against GluA1 (rabbit PC246 Calbio-chem) or GluA2 (mouse MAB397 Millipore) inmedium for 15 min at 37degC and 5 CO2 washedfor 2 min in medium and then stimulated with100 mM AMPA or NMDA for 2 min followed byincubation in conditioned medium for an addi-tional 8 min at 37degC and 5 CO2 before fixingcells with 4 paraformaldehyde Surface recep-tors were labeled with Alexa 647 secondary anti-bodies then cells were permeabilized and internalreceptors labeled with Alexa 546 secondary anti-bodies The internalization indexwas calculated asthe ratio of internal to surface receptor fluores-cence Cultures in which control conditions didnot show an increase in internalization followingstimulation were excluded from analysis

pHluorin imaging

Neurons were transferred to a live imaging cham-ber (Warner Instruments) containing bath salinesolution (140 mM NaCl 5 mM KCl 2 mM CaCl22 mM MgCl2 55 mM glucose 20 mM HEPES1 mM TTX pH = 73) Transfected cells wereselected and images acquired at 1 s intervals and500 ms exposure times with 48420nm excita-tion and 51720-nm emission filters on a ZeissAxio Observer invertedmicroscope with a Photo-metric Evolve EMCCD camera and LambdaDG-4 fast-switching light source interfaced withMetamorph software A baseline of at least 30images was collected before addition of highpotassium buffer (100 mM NaCl 45 mM KCl2mMCaCl2 2mMMgCl2 55mMglucose 20mMHEPES 1 mM TTX pH = 73) 100 mM AMPA100 mMNMDA or field stimulation to depolarizeneurons For AMPA stimulation 50 mMAP5 wasincluded in the bath solution to block NMDAreceptors for NMDA stimulation 20 mM CNQXwas included to blockAMPA receptors Dendriticpuncta were selected using Metamorph software(Molecular Devices) and fluorescence intensitynormalized to baseline was plotted versus time

Subcellular fractionation from whole brain

Rat brains were homogenized in ice-cold homog-enization buffer (320mM sucrose 4mMHEPES

pH 74 with NaOH) with 10 strokes at 900 rpmSamples were then centrifuged at 1000 times g for10 min The supernatant (S1) was collected theresulting pellet (P1) contains large cell frag-ments and nuclei S1 was then centrifuged at15000 times g for 15 min The supernatant (S2) con-tains soluble proteins and the pellet (P2) containssynaptosomes The pellet was then carefullyresuspended in 1 ml homogenization buffer 9mlice-cold ddH20 was added and three strokes at2000 rpmwere performed 50 ml 1MHEPES andprotease inhibitors were added The lysate wascentrifuged at 17000 times g for 25 min to separatesynaptosomal membranes (LP2) from synapto-somal cytosol (LS2) The LP2 pellet was resus-pended in 6ml 40mM sucrose and layered over acontinuous sucrose gradient from 50 mM to800 mM The sucrose cushion was then centri-fuged at 28000 rpm for 2 hours Following centri-fugation the region between approximately200 mM and 400 mM sucrose was collectedand separated by chromatography on controlled-pore glass beads (CPG column) and run overnightThe first peak (PI) contained larger membranefragments and SVs were found in the second peak

Immuno-organelle isolation ofsynaptic vesicles

Mouse monoclonal antibodies directed againstSyb2 and Syt1 were coupled to Protein G mag-netic Dynabeads (Invitrogen) in PBS for 2 hoursat 4degC Antibody-coated beads were added towhole brain S1 fractions and incubated overnightat 4degC Magnetic beads were separated fromimmuno-depleted supernatant andwashed 3 timeswith PBS Bound vesicles were eluted in samplebuffer and analyzed by SDSndashpolyacrylamide gelelectrophoresis (SDS-PAGE) and immunoblotting

Syt3 pulldowns

Recombinant His-tagged Syt3 C2AB (providedby Edwin Chapman University of WisconsinMadison) was expressed in E coli and purifiedas previously described (49) Recombinant Syt3was incubated with solubilized rat brain homog-enate for 2 hours at 4degC After incubationNi-beadswere added and incubated for an additional2 hours at 4degC Themixture was then poured intoa MT column from Biorad and washed 3 X withwash buffer (20 mM Tris pH 74 500 mM NaCl20 mM imidazole) Proteins bound to recombi-nant Syt3 in the column were eluted with elutionbuffer (20mMTris pH 74 500mMNaCl 400mMimidazole) For experiments testing inhibition ofSyt3GluA2 binding 1 mM Tat-GluA2-3Y peptide(sequence YGRKKRRQRRR-869YKEGYNVYG877provided by Yu-Tian Wang University of BritishColumbia Vancouver Canada) was pre-incubatedwith brain homogenate overnight prior to incu-bation with recombinant Syt3 and Ni-beadsEluted proteins were loaded onto SDS-PAGE gelsand analyzed by Western blot

Synaptosome trypsin cleavage assay

Synaptosomes were prepared and treated withtrypsin as described previously (26) Purified syn-aptosomes were centrifuged for 3 min at 8700 times g




Dow

nloaded from

4degCThepelletwas resuspended in320mMsucrose5 mM HEPES pH 8 For trypsin cleavage a01 mgml trypsin stock solution was added toyield a final protein-protease ratio of 1001 Synap-tosomes were incubated for 10 20 30 60 or90 min at 30degC with gentle agitation This mix-ture was then centrifuged for 3 min at 8700 times gand the resulting pellet resuspended in sucrosebuffer containing 400 mM Pefabloc (Roche) tostop trypsin cleavage activity Samples were thenanalyzed by SDS-PAGE and immunoblotting

AAV preparation andhippocampal injections

AAVESYN-GFP-P2A-Syt3 or calcium-binding defi-cient mutant Syt3 (D386 388N and D520 522N)constructs were synthesized and subcloned byGenscript and AAV12 viral particles preparedby transfecting 10 cm dishes of 70-80 confluentHEK293 cells with 125 mg of viral construct con-taining Syt3 or calcium-binding deficientmutantSyt3 with 25 mg pFdelta6 625 mg pRV1 and625 mg pH21 helper plasmids using calciumphosphate Cell media was replaced after 6 hoursand cells incubated for 48 hours prior to virusharvesting Viral particles were released by add-ing 10 sodium deoxycholate and benzonase(Sigma-Aldrich) and purified in an Optiprep(Sigma-Aldrich) step gradient by ultracentrifugationfor 90 min The pure viral fraction was concen-trated to approximately 500 ml through a 022 mmAmicon Ultra unit (MilliporeMerck) aliquotedand stored at ndash80degC Virus was used at a titerof 107 particlesml Mice were given metamizol(3mlL) in drinking water for 2 days prior tosurgery and up to 3 days after On the day ofsurgery mice were anesthetized with a singleintraperitoneal injection of ketaminexylazine(80 and 10 mgkg respectively) and given asingle subcutaneous injection of buprenorphine(01 mgkg) Mice were fixed in a stereotaxicdevice (myNeuroLab Wetzlar Germany) an inci-sion was made to expose the bone and antero-posterior (ndash175mm) and mediolateral (plusmn10 mm)coordinates from bregma were used for drillingholes bilaterally A fine glass injection capillaryfilled with 12 ml virus was then slowly lowered tondash15 (dorsoventral) and allowed to rest for 1 min09 ml was then injected over 3 min The needlewas allowed to rest in place for an additionalminute after the injection and then slowly liftedabove bone level This procedure was repeated forboth hippocampi Mice were then removed fromthe stereotaxic device and tissue glue (HistoacrylB Braun) was used to seal the wound They werethen allowed to recover on a heat blanket andtransferred to individual cages for post-operationrecovery Mice were given buprenorphin injec-tions up to 2 days after the operation and closelymonitored for signs of distress or pain Mice wereonly used for subsequent experiment after amini-mum of 2 weeks post-operation

Electrophysiological recordings fromdissociated hippocampal neurons

Following transfection on DIV10 DIV13-20 ratneurons growing on coverslips were placed in a

custom-made recording chamber in extracellularsolution containing in mM 142 NaCl 25 KCl 10HEPES 10 D-Glucose 2 CaCl2 13 MgCl2 (295mOsm pH adjusted to 72 with NaOH) Thetemperature of the bath was maintained at 30to 32degC To record miniature excitatory post-synaptic currents (mEPSCs) 1 mM TTX (to blockaction potentials) and 50 mMpicrotoxin (to inhibitGABAA receptors) was added to the extracellularsolution An Olympus upright BX51WImicroscopeequipped with a 40X water-immersion objectivefluorescent light source (Lumen 200Pro PriorScientific) and filters forGFP fluorescence imagingwere used to visualize neurons transfected withGFP GFP-Syt3 or Turbo GFP-expressing Syt3knockdownconstructs Patch pipetteswere pulledfrom borosilicate glass (15 mmOD 086 mm ID3 to 6megohms) using a P-97micropipette puller(Sutter Instruments) The internal solution con-tained inmM 130K-gluconate 10NaCl 1 EGTA0133 CaCl2 2MgCl2 10HEPES 35Na2-ATP1Na-GTP (285 mOsm pH adjusted to 74 with KOH)Whole-cell patch-clamp recordings were obtainedusing a HEKA EPC10 USB double patch clampamplifier coupled to Patchmaster acquisitionsoftware Signals were low pass filtered using aBessel filter at 29 KHz and digitized at 5 KHzmEPSCs were recorded while holding neuronsat ndash60mV in the voltage-clamp mode with fastand slow capacitance and series resistance com-pensated The series resistance was monitoredduring recording to ensure it did not change bymore than plusmn 3 megohms and neurons wererecorded from only if uncompensated Rs lt 20megohms MEPSCs were analyzed using MiniAnalysis software v603 (Synaptosoft Inc) with anamplitude threshold of 35 X average RMS noiseDIV13-19 mouse neurons were recorded from

in the same conditions except the bathwas at roomtemperature and 100 mMpicrotoxin was addedFor AMPA stimulation coverslips were transferredto 250 ml prewarmed medium containing 100 mMS-AMPA (Abcam ab120005) and incubated at 37degCand 5 CO2 for 2 min followed by 8 min incu-bation in conditioned medium without AMPAas for receptor internalization assays Whole cellrecordings were performed on an upright micro-scope (Zeiss Examiner D1) equipped with a 40Xwater immersion objective with a fluorescentlight source (Zeiss Colibri) using an ELC-03XSpatch clampamplifier (NPI Electronics Germany)with custom written data acquisition scripts forIgorPro 612A software (Wavemetrics) fromOliver Schluumlter (EuropeanNeuroscience InstituteGoettingen) Signals were digitized at 10 KHzusing an InstruTECH ITC-18 data acquisitioninterface (HEKA) The fast capacitance was com-pensatedandslowcapacitanceandseries resistancewere not compensated The series resistance didnot change by more than 10 monitored every10 s The experimenter was blinded to file namesduring analysis

Whole-cell electrophysiology in acutehippocampal slices

Patchpipetteswith a resistance of 25 to 5megohmswere prepared from glass capillaries (Harvard

Apparatus cat no 300060 15 mmOD 086mmID) using a P-97 puller (Sutter Inst Novato CA)P12-P17 male and female mouse pups wereanesthetized with isoflurane (Abbott WiesbadenGermany) decapitated and the brain extractedand transferred to cold NMDG buffer containing(in mM) 45 NMDG 033 KCl 04 KH2PO405 MgCl2 016 CaCl2 20 choline bicarbonate1295 glucose 300 mm coronal hippocampalslices were made with a Leica VT1200 vibratome(Wetzlar Germany) and a stainless steel blade(Feather) in ice cold NMDG buffer Slices weretransferred to a preincubation chamber with amesh bottom filled with ACSF containing (in mM)124 NaCl 44 KCl 1 NaH2PO4 262 NaHCO3 13MgSO4 25 CaCl2 and 10 D-Glucose bubbled with95 O25 CO2 (carbogen) and incubated at 35degCfor 05 hours followedbyanother 05 hours at roomtemperature Hippocampi were then dissectedusing a Zeiss Stemi 2000 stereoscopic microscopeA cut perpendicular to the CA3 pyramidal celllayer was made in the CA3 region and anothercut perpendicular to the CA1 field at themedialend of the slice was made to prevent recurrentactivity Slices were submerged in a chamberperfusedwith carbogen-bubbled ACSFmaintainedat 30degC-32degC andCA1 neurons visualizedwith anupright Zeiss Examiner D1 microscope equippedwith a 40X water-immersion objective The intra-cellular pipette solution contained (in millimolar)130 CsMeSO3 267 CsCl 10 HEPES 1 EGTA 4 Mg-ATP 03 Na-GTP 3 QX-314 Cl 5 TEA-Cl 15Phosphocreatine disodium and 5Uml Creatine-phosphokinase (mOsm 303 pH 744) Whole-cellpatch-clamp recordings were obtained using a NPIELC-03XS patch clamp amplifier and digitizedusing an InstruTECH ITC-18 data acquisitioninterface (HEKA) A custom written procedure(provided by Oliver Schluumlter European Neuro-science Institute Goettingen) in Igor Pro 612Awas used to visualize and store recorded dataSignals were low pass filtered using a Bessel filterat 3 KHz and digitized at 10 KHz Schaffer col-laterals were stimulated at 01 Hz using a bipolarglass electrode filled with ACSF at the distaldendritic region of the stratum radiatum close tothe border of the lacunosum-moleculare and aminimum of 30 EPSCs were averaged The fastcapacitance was compensated and slow capaci-tance and series resistance were not compensatedSynaptic responses were first recorded at a holdingpotential of ndash56mV themeasured average reversalpotential for Clndash in wild-type slices GABAA

receptor-mediated currents were then recordedat 0mV the measured average reversal potentialof NMDA and AMPA receptors 01 mM picro-toxin was then perfusedwhile holding at ndash56mVafter which the elimination of GABA IPSCs at0 mV was confirmed The residual AMPA+NMDAEPSC at 0mVwas then subtracted from the GABAIPSC TheAMPAEPSCs subsequently recorded atndash56 mV which were free of any GABAA receptorIPSC component were used for analysis TheAMPA+NMDA compound EPSC was then re-corded at +40 mV in the presence of 01 mMpicrotoxin where the amplitude of the EPSCapproximately 60ms after the peak of the AMPA




Dow

nloaded from

EPSC is considered to be purely NMDA receptor-mediated since the measured AMPA receptorEPSC decay time constant t asymp 20 ms The inputand series resistance (Rs) were monitored every10 s Recordings where Rs was gt 25 megohms orchanged by more than plusmn 20 during recordingwere excluded from analysis Input resistancesranged from 100 to 400 megohms and did notchange by more than plusmn 20 during the course ofthe recording Matlab was used to calculate cur-rent ratios and decay times

Extracellular recordings from acutehippocampal slices

Acute hippocampal slices were prepared from8 to 12-week-old male mice anesthetized withisoflurane and decapitated The hippocampuswas removed and 400 mm thick dorsal hippo-campal slices were cut transversely using a tissuechopper (Stoelting) in ice-cold artificial cerebro-spinal fluid (ACSF) containing in mM 124 NaCl49 KCl 12 KH2PO4 2 MgSO4 2 CaCl2 246NaHCO3 and 10 D-glucose (saturated with 95O2 and 5 CO2 pH 74 ~305 mOsm) Within5 min of decapitation slices were transferredto a custom-built interface chamber and pre-incubated in ACSF for at least 3 hours Followingpre-incubation the stimulation strength was setto elicit 40 of the maximal field EPSP (fEPSP)slope The slope of the rising phase of the fEPSPwas used to determine potentiation of synapticresponses For stimulation biphasic constant cur-rent pulseswere used A baselinewas recorded forat least 30min before LTDor LTP induction Fouraveraged 02 Hz biphasic constant-current pulses(01 ms per polarity) were used to test responsespost-tetanus for up to 4 hoursLTD was induced by a single tetanus of 900

pulses at 1 Hz Nondecaying LTP was inducedby a 3XTET high frequency stimulation withthree stimulus trains of 100 pulses at 100 Hzstimulus duration 02 ms per polarity with 10 minintertrain intervals (50) Decaying LTP was in-duced with a 1XTET single tetanus of 16 pulsesat 100 Hz stimulus duration 02 ms per polaritymodified for mouse hippocampal slices from theprotocol consisting of 21 pulses used for rathippocampal slices (36) The Tat-GluA2-3Y peptide(sequence YGRKKRRQRRR-869YKEGYNVYG877provided by Yu-Tian Wang University of BritishColumbia Vancouver Canada) was used at a con-centration of 1 mM and ZIP (Tocris cat no 2549)was dissolved in water as recommended by themanufacturer and used at a concentration of1 mM (13) Control and experimental recordingswere interleaved as much as possible for com-parison of conditions For each condition thecorresponding control was repeated before andintermittently throughout experiments to ensureconsistent recording conditions

Behavioral experiments

Behavioral experiments were performed on 3- to9-month-old male mice by a male experimenterexcept for the Barnes maze where a male exper-imenter performed intraperitoneal injections anda female experimenter performed experiments

All experimenters were blinded to genotype Micewere individually housed at the European Neuro-science Institute Goumlttingen Germany animalfacility in standard environmental conditions(temperature humidity) with ad libitum accessto food andwater and a 12 hours lightdark cycleMice were allowed to habituate to different hold-ing rooms for behavioral experiments for twoweeks before testing and to experimenters for atleast three days before experiments All experi-mentswere performed during the light cycle Videotrackingwas donewith the TSEmonitoring systemVideomot2 All behavioral experiments were ap-proved under project number G151794 by theNiedersaumlchsischesLandesamt fuumlrVerbraucherschutzund Lebensmittelsicherheit (LAVES LowerSaxony Germany)

Open field elevated plus maze

In the open field test mice were introduced nearthe wall of an empty opaque square plexiglasbox and allowed to freely explore the arena for 5min Before every trial urine was removed withKimwipes and the arena was cleanedwith waterand 70 EtOH Time spent in the center (2 times 2grid in the middle of a 4 times 4 grid arena) relativeto near thewalls and the total path travelled wasrecordedFor the elevated plus maze mice were intro-

duced into the center quadrant of a 4-arm mazewith two open and two closed arms Before everytrial urine was removed with Kimwipes and themaze was cleaned with water and 70 EtOHTime spent in the open arms was analyzed

Reference memory water maze

Naumlive mice from four independent cohorts weretrained to find a 13 times 13 cm square hidden plat-form (uninjected mice) or a 10 cm diametercircular platform (injectedmice) submerged 15 cmbelow the surface in a 11 m diameter circularpool filled with white opaque water at 19 plusmn 1degCMice were trained in 4 trials per day in succes-sion where each trial lasted 1 min during whichthe mice searched for the platform A trial endedwhen the mouse spent at least 2 s on the plat-form after which they were left on the platformfor 15 s prior to the start of the next trial Fourshapes around the pool (star square trianglecircle) served as visual cues Mice that failed tofind the platform after 1 min were guided to itand left on the platform for 15 s Mice that failedto swim (floaters) were excluded from analysisMice were placed into the pool facing the wall atthe beginning of each trial and the position ofpool entry from four different directions wasrandomly shuffled daily For probe tests the plat-form was removed and mice were placed into thepool near the wall in the quadrant opposite tothat of the platform location and allowed tosearch for the platform for 1 min Two probe testswere performed to monitor learning of the firstplatform position and additional probe test(s)performed after reversal ie switching the plat-form position to the opposite quadrant Onlydata from coincident days of training from thefirst 2 cohorts (Fig 5 A to F) was pooled Video

tracking data was analyzed using Matlab toextract time-tagged xy-coordinate informationand quantify escape latency platform crossingsproximity [mean distance of all tracked pathpoints to platform center which is reported to bethe most reliable parameter to quantify spatialspecificity of water maze search patterns (41)]percent time spent in target quadrant and aver-age swimming speed Occupancy plots weregenerated by super-imposing all path pointsof all mice in a group Densities of path points(normalized to total number of path points in agroup of mice) within a grid size of 21 cm times 21 cm(43)were calculatedDensity datawas smoothenedand interpolated to plot heat maps (Scattercloudfunction Matlab File ID 6037) Occupancy plotcolor maps were linearly scaled using the globalminimumandmaximumdensities for all groupsto allow comparison between them The last 05 sof the trial whenmice were on the platform andnot searching was excluded in occupancy plotsfor Fig 5G To generate occupancy plots of probetrials from two cohorts in which the platformwas in different positions (Fig 5 E and F) thetracking data from one cohort was transposedand mirror imaged with respect to the centerpool line and superimposed on data from thesecond cohort Circular areas encompassing plat-forms of both cohorts in both original and rever-sal quadrants were defined as target areasBecause forgetting cannot be analyzed in mice

that did not learn mice in ip injected cohorts(which learnedmore slowly) were excluded fromanalysis if (i) their swim speed was less than62 cms for more than three consecutive days(ii) their proximity or escape latency failed todecrease between the first day of training andthe average of the last three days of trainingbefore platform reversal or (iii) they spent 0time in the target quadrant in the second probetest These criteria were also met by all othercohorts tested Daily ip injections (maximumvolume injected 10 mlg body weight) were per-formed 1 hour before training (13) Mice wereinjected with saline (09 NaCl) during trainingto the first platform position and then injectedwith salineorwithTat-GluA2-3Ypeptide (5mmolkgbody weight) one hour before the second probetest and daily after reversal Probe tests followingreversal training of ip injected cohorts wereperformed on three consecutive days

Delayed matching to place water maze

The delayedmatching-to-place task protocol wasperformed as previously described (42) withmodi-fications Mice were first habituated to the task bytraining to a visible platform (marked by agraduated cylinder coated with multi-coloredpaper on top of a submerged platform) placedat a unique position every day for 3 days in a 11 mdiameter circular pool filled with white opaquewater at 19 plusmn 1degC The circular platform wassubmerged 15 cm below the surface and had aradius of 5 cm Four shapes around the pool (starsquare triangle circle) served as visual cuesAfter habituation mice were trained to 16 uniquehidden platform positions at two fixed distances




Dow

nloaded from

from the pool center (11 in anouter ring and5 in aninner ring) over 16 daysMice underwent 4 trials of2min every day The platformwas shifted to a newrandom position every day in a different quadrantalternating between inner and outer rings asmuchas possible A trial ended when themouse spent atleast 2 s on the platform after which they were lefton the platform for 15 s and then returned to theirhome cage Mice were warmed with infraredlamps after every trial The drop-off points withrespect to the platform position of that day werein a random order Mice were divided into twogroups for counterbalancing in which each groupexperienced different alternations of platformpositions between inner and outer rings toprevent non-spatial chaining search strategiesin which mice search for platforms within acertain distance of the pool wall Inter-trialintervals were 5 min from day 10 onwards theinter-trial interval between trial 1 and trial 2was increased to 1 hour 15 min Trial 2 on day16 was a probe trial ie the platform was re-moved and the exploratory behavior of micerecorded for 2 min Because the platform in thistrial was missing only the search path data be-fore the first crossing of the platform positionfrom trial 1 was used to calculate savings ofproximity escape latency and path length Allmice were trained at the same time of the day asfar as possible ie some mice were always trainedin themorningwhile otherswere always trained inthe evening In all occupancy plots the last 15 s ofthe trial were excluded to enhance contrast in thepool and avoid high occupancy densities on theplatform since all mice spent the last 2 s of everytrial on the platformQuantitation of search strategy was modified

from a previously describedmethod (43) AMatlabscript extracted time-tagged xy-coordinate infor-mation from video tracking data of all trials andclassified individual trials as search strategiesThe following parameters were used for classifi-cation of the indicated search strategies Directswimming strategy total path length le14 Xdistance between drop-off point and platformcenterge80pathpointswithin a 30deg goal corridorangle Focal search strategy mean proximityto platform center le035 X pool radius meanproximity to path centroid le04 X pool radiusDirected search strategy ge70 path pointswithin a 30deg goal corridor angle Perseverancestrategy mean proximity to previous platformcenter le05 X pool radius mean proximity topath centroid allowed for perseverance le06 Xpool radius Chaining strategy ge70 path pointsinside annulus around the ring of platforms(inner or outer) containing the current daysplatform by 001 to 002 X pool radius annulusfor platforms in outer ring between 052 X poolradius and 073 X pool radius annulus forplatforms in inner ring between 022 X poolradius and 042X pool radius Scanning strategyge80 of path points inside scanning radius (acircular area enclosing all platforms 073 X poolradius) total pool area scanned ge10 andle50 Thigmotaxis ge30 of path points insidecloser wall zone starting 085 X pool radius and

ge50 of path points inside wider wall zonestarting 075 X pool radius Random Search strat-egy pool area scanned ge50

Barnes Maze

The Barnes maze consisted of a white circularplatform made of plexiglass 92cm in diameterwith 20 equally spaced holes (5 cm in diameter)along the perimeter The platform was elevated40 cm above the ground An escape chamber(155 times 9 times 6 cm3) was placed under one of theholes defined as the target hole To encouragemice to enter the escape chamber it contained aplastic ramp to enable the mice to climb into itand an odorless paper towel resembling nestingmaterial Before every trial urine was removedwith Kimwipes and the arena was cleaned withwater and 70 EtOH to eliminate olfactory cuesFor each trial a mouse was placed onto the mid-dle of themaze such that the target hole was notdistinguishable from any other hole and themouse had to rely on three visual cues (26times 26 cm2

in size) placed 5 cm from the edge of the platformto identify the location of the target hole Theplatformwas uniformly illuminated by bright LEDlight which served as a mildly aversive stimulus tomotivate the mice to search for the target holeDaily ip injections (maximum volume injected

10 mlg body weight) were performed one hourbefore training (13) Mice were injected withsaline (09 NaCl) during training to the firsthole position and then injected with saline orwith Tat-GluA2-3Y peptide (5 mmolkg bodyweight) one hour before the second probe testand daily after reversalOn the day before training mice were habi-

tuated to the maze by placing them in a clearplastic cylinder (15 cm in diameter) in themiddleof the platform for 30 s before gently guidingthem to the target hole where they were given3 min to enter the escape chamber If mice didnot enter the escape chamber they were gentlynudged with the cylinder into the chamber Allmice spent 1 min in the escape chamber beforebeing returned to their home cage On subse-quent training days mice were placed into anopaque plastic cylinder (15 cm in diameter)covered by an opaque lid for 15 s to randomizestarting orientation Each trial lasting 2 minwas initiated after lifting the cylinder If themouse climbed into the escape chamber the trialwas stopped and themouse returned to its homecage after remaining inside the escape chamberfor 1 min If the mouse did not climb into theescape chamber within 2min it was gently guidedto the target holewith the clear plastic cylinder andgiven 3 min to climb into the escape chamberwhere it remained for 1 min before being returnedto its home cageIn probe tests the escape chamber was re-

moved and mice were allowed to explore themaze for 2 min The ldquoreversalrdquo hole was locatedin an adjacent quadrant since some mice showeda bias toward exploring the opposite quadrantduring initial training Mice experienced onetrial per day and were given a one-day ldquobreakrdquoafter the first three probe tests during which they

did not perform in the Barnes maze but stillreceived saline (maximum volume injected 10 ml09NaClg body weight) or GluA2-3Y (5 mmolkgbody weight) ip injectionsThree-point (head center tail) tracking data

was exported into Matlab for further analysisOccupancy plots were generated usingMatlab asdescribed above for the water maze except thathead instead of center coordinates were usedVideo tracking errors in which head and tailcoordinates were erroneously swapped (mostlyclose to hole regions) were detected by a distancethreshold and excluded Mice that did not findthe original target hole in either of the first twoprobe tests (did not learn the target hole) wereexcluded from analysis of perseverance afterreversal Target area was defined as the area ofthe target hole and its two neighboring holesusing center tracked coordinates for calculationof the perseverance ratio (time spent in reversaltarget area divided by total time spent in originaland reversal target areas)

Statistical analysis

All statistical analysis was done using GraphPadPrism or Microsoft Excel All reported values instatistical analysis represent themean error barsindicate SEM and all Students t tests are two-tailed type 2 unless otherwise indicated For allstatistical tests data met the assumptions of thetest All n numbers listed in Figure Legends referto biological replicates

REFERENCES AND NOTES

1 J M Henley K A Wilkinson Synaptic AMPA receptorcomposition in development plasticity and disease Nat RevNeurosci 17 337ndash350 (2016) doi 101038nrn201637pmid 27080385

2 J D Shepherd R L Huganir The cell biology of synapticplasticity AMPA receptor trafficking Annu Rev Cell Dev Biol23 613ndash643 (2007) doi 101146annurevcellbio23090506123516 pmid 17506699

3 G L Collingridge J T Isaac Y T Wang Receptor traffickingand synaptic plasticity Nat Rev Neurosci 5 952ndash962 (2004)doi 101038nrn1556 pmid 15550950

4 R C Malenka M F Bear LTP and LTD An embarrassment ofriches Neuron 44 5ndash21 (2004) doi 101016jneuron200409012 pmid 15450156

5 S Nabavi et al Engineering a memory with LTD and LTPNature 511 348ndash352 (2014) doi 101038nature13294pmid 24896183

6 T J Ryan D S Roy M Pignatelli A Arons S TonegawaMemory Engram cells retain memory under retrogradeamnesia Science 348 1007ndash1013 (2015) doi 101126scienceaaa5542 pmid 26023136

7 J H Choi et al Interregional synaptic maps among engramcells underlie memory formation Science 360 430ndash435(2018) doi 101126scienceaas9204 pmid 29700265

8 P V Migues et al Blocking synaptic removal of GluA2-containing AMPA receptors prevents the natural forgettingof long-term memories J Neurosci 36 3481ndash3494(2016)doi 101523JNEUROSCI3333-152016 pmid 27013677

9 A Hayashi-Takagi et al Labelling and optical erasure ofsynaptic memory traces in the motor cortex Nature 525333ndash338 (2015) doi 101038nature15257 pmid 26352471

10 R Scholz et al AMPA receptor signaling through BRAG2 andArf6 critical for long-term synaptic depression Neuron 66768ndash780 (2010) doi 101016jneuron201005003pmid 20547133

11 G Ahmadian et al Tyrosine phosphorylation of GluR2 isrequired for insulin-stimulated AMPA receptor endocytosis andLTD EMBO J 23 1040ndash1050 (2004) doi 101038sjemboj7600126 pmid 14976558




Dow

nloaded from

12 P V Migues et al PKMzeta maintains memories by regulatingGluR2-dependent AMPA receptor trafficking Nat Neurosci 13630ndash634 (2010) doi 101038nn2531 pmid 20383136

13 Z Dong et al Long-term potentiation decay and memory lossare mediated by AMPAR endocytosis J Clin Invest 125234ndash247 (2015) doi 101172JCI77888 pmid 25437879

14 M Y Xiao Y P Niu H Wigstroumlm Activity-dependent decay ofearly LTP revealed by dual EPSP recording in hippocampalslices from young rats Eur J Neurosci 8 1916ndash1923 (1996)doi 101111j1460-95681996tb01335x pmid 8921282

15 D M Villarreal V Do E Haddad B E Derrick NMDA receptorantagonists sustain LTP and spatial memory Active processesmediate LTP decay Nat Neurosci 5 48ndash52 (2002)doi 101038nn776 pmid 11740500

16 M D Ehlers Reinsertion or degradation of AMPA receptorsdetermined by activity-dependent endocytic sorting Neuron 28511ndash525 (2000) doi 101016S0896-6273(00)00129-Xpmid 11144360

17 E C Beattie et al Regulation of AMPA receptor endocytosisby a signaling mechanism shared with LTD Nat Neurosci 31291ndash1300 (2000) doi 10103881823 pmid 11100150

18 D Wu et al Postsynaptic synaptotagmins mediate AMPAreceptor exocytosis during LTP Nature 544 316ndash321 (2017)doi 101038nature21720 pmid 28355182

19 R B Sutton B A Davletov A M Berghuis T C SuumldhofS R Sprang Structure of the first C2 domain ofsynaptotagmin I A novel Ca2+phospholipid-binding fold Cell80 929ndash938 (1995) doi 1010160092-8674(95)90296-1pmid 7697723

20 S Butz R Fernandez-Chacon F Schmitz R JahnT C Suumldhof The subcellular localizations of atypicalsynaptotagmins III and VI Synaptotagmin III is enriched insynapses and synaptic plasma membranes but not in synapticvesicles J Biol Chem 274 18290ndash18296 (1999)doi 101074jbc2742618290 pmid 10373432

21 T C Suumldhof Synaptotagmins Why so many J Biol Chem277 7629ndash7632 (2002) doi 101074jbcR100052200pmid 11739399

22 S Sugita O H Shin W Han Y Lao T C SuumldhofSynaptotagmins form a hierarchy of exocytotic Ca(2+) sensorswith distinct Ca(2+) affinities EMBO J 21 270ndash280 (2002)doi 101093emboj213270 pmid 11823420

23 C Dean et al Axonal and dendritic synaptotagmin isoformsrevealed by a pHluorin-syt functional screen Mol Biol Cell 231715ndash1727 (2012) doi 101091mbce11-08-0707 pmid 22398727

24 L W Gong P De Camilli Regulation of postsynaptic AMPAresponses by synaptojanin 1 Proc Natl Acad Sci USA 10517561ndash17566 (2008) doi 101073pnas0809221105pmid 18987319

25 D T Lin R L Huganir PICK1 and phosphorylation of theglutamate receptor 2 (GluR2) AMPA receptor subunit regulatesGluR2 recycling after NMDA receptor-induced internalizationJ Neurosci 27 13903ndash13908 (2007) doi 101523JNEUROSCI1750-072007 pmid 18077702

26 J Boyken et al Molecular profiling of synaptic vesicle dockingsites reveals novel proteins but few differences betweenglutamatergic and GABAergic synapses Neuron 78 285ndash297(2013) doi 101016jneuron201302027 pmid 23622064

27 M Rathje et al AMPA receptor pHluorin-GluA2 reports NMDAreceptor-induced intracellular acidification in hippocampalneurons Proc Natl Acad Sci USA 110 14426ndash14431 (2013)doi 101073pnas1312982110 pmid 23940334

28 T A Blanpied D B Scott M D Ehlers Dynamics andregulation of clathrin coats at specialized endocytic zones ofdendrites and spines Neuron 36 435ndash449 (2002)doi 101016S0896-6273(02)00979-0 pmid 12408846

29 J Lu et al Postsynaptic positioning of endocytic zones andAMPA receptor cycling by physical coupling of dynamin-3 toHomer Neuron 55 874ndash889 (2007) doi 101016jneuron200706041 pmid 17880892

30 E M Petrini et al Endocytic trafficking and recycling maintaina pool of mobile surface AMPA receptors required for synaptic

potentiation Neuron 63 92ndash105 (2009) doi 101016jneuron200905025 pmid 19607795

31 M Rosendale D Jullieacute D Choquet D Perrais Spatial andtemporal regulation of receptor endocytosis in neuronaldendrites revealed by imaging of single vesicle formationCell Reports 18 1840ndash1847 (2017) doi 101016jcelrep201701081 pmid 28228251

32 S H Lee L Liu Y T Wang M Sheng Clathrin adaptor AP2and NSF interact with overlapping sites of GluR2 and playdistinct roles in AMPA receptor trafficking and hippocampalLTD Neuron 36 661ndash674 (2002) doi 101016S0896-6273(02)01024-3 pmid 12441055

33 W Lu et al Subunit composition of synaptic AMPA receptorsrevealed by a single-cell genetic approach Neuron 62254ndash268 (2009) doi 101016jneuron200902027pmid 19409270

34 J W Lin et al Distinct molecular mechanisms and divergentendocytotic pathways of AMPA receptor internalizationNat Neurosci 3 1282ndash1290 (2000) doi 10103881814pmid 11100149

35 Z Dong et al Hippocampal long-term depression mediatesspatial reversal learning in the Morris water mazeNeuropharmacology 64 65ndash73 (2013) doi 101016jneuropharm201206027 pmid 22732443

36 S Sajikumar S Navakkode J U Frey Identification ofcompartment- and process-specific molecules required forldquosynaptic taggingrdquo during long-term potentiation and long-term depression in hippocampal CA1 J Neurosci 275068ndash5080 (2007) doi 101523JNEUROSCI4940-062007pmid 17494693

37 P Tsokas et al Compensation for PKMz in long-termpotentiation and spatial long-term memory in mutant miceeLife 5 e14846 (2016) doi 107554eLife14846pmid 27187150

38 Y Yao et al PKM zeta maintains late long-term potentiation byN-ethylmaleimide-sensitive factorGluR2-dependent traffickingof postsynaptic AMPA receptors J Neurosci 28 7820ndash7827(2008) doi 101523JNEUROSCI0223-082008pmid 18667614

39 N Sadeh S Verbitsky Y Dudai M Segal Zeta InhibitoryPeptide a candidate inhibitor of protein kinase Mz isexcitotoxic to cultured hippocampal neurons J Neurosci 3512404ndash12411 (2015) doi 101523JNEUROSCI0976-152015pmid 26354909

40 M J LeBlancq T L McKinney C T Dickson ZIP It Neuralsilencing is an additional effect of the PKM-z inhibitor zeta-inhibitory peptide J Neurosci 36 6193ndash6198 (2016)doi 101523JNEUROSCI4563-142016 pmid 27277798

41 H R Maei K Zaslavsky C M Teixeira P W Frankland Whatis the most sensitive measure of water maze probe testperformance Front Integr Nuerosci 3 4 (2009)doi 103389neuro070042009 pmid 19404412

42 K Nakazawa et al Hippocampal CA3 NMDA receptors arecrucial for memory acquisition of one-time experience Neuron38 305ndash315 (2003) doi 101016S0896-6273(03)00165-Xpmid 12718863

43 A Garthe J Behr G Kempermann Adult-generatedhippocampal neurons allow the flexible use of spatially preciselearning strategies PLOS ONE 4 e5464 (2009) doi 101371journalpone0005464 pmid 19421325

44 A Citri et al Calcium binding to PICK1 is essential for theintracellular retention of AMPA receptors underlying long-termdepression J Neurosci 30 16437ndash16452 (2010)doi 101523JNEUROSCI4478-102010 pmid 21147983

45 M Fiuza et al PICK1 regulates AMPA receptor endocytosis viadirect interactions with AP2 a-appendage and dynamin J CellBiol 216 3323ndash3338 (2017) doi 101083jcb201701034pmid 28855251

46 J Yao S E Kwon J D Gaffaney F M DunningE R Chapman Uncoupling the roles of synaptotagmin I duringendo- and exocytosis of synaptic vesicles Nat Neurosci 15243ndash249 (2011) doi 101038nn3013 pmid 22197832

47 G A Banker W M Cowan Rat hippocampal neurons indispersed cell culture Brain Res 126 397ndash42 (1977)doi 1010160006-8993(77)90594-7 pmid 861729

48 J Rizo T C Suumldhof C2-domains structure and function of auniversal Ca2+-binding domain J Biol Chem 27315879ndash15882 (1998) doi 101074jbc2732615879pmid 9632630

49 A Bhalla M C Chicka E R Chapman Analysis of thesynaptotagmin family during reconstituted membrane fusionUncovering a class of inhibitory isoforms J Biol Chem 28321799ndash21807 (2008) doi 101074jbcM709628200pmid 18508778

50 U Frey R G Morris Weak before strong Dissociating synaptictagging and plasticity-factor accounts of late-LTPNeuropharmacology 37 545ndash552 (1998) doi 101016S0028-3908(98)00040-9 pmid 9704995

ACKNOWLEDGMENTS

We thank R Morris for recommendations of intertrial intervals inthe DMP task H Urbanke for the Barnes maze protocol andJ Schrader for technical assistance Funding This work wassupported by a Sofja Kovalevskaja grant from the Alexander vonHumboldt Foundation European Research Council (ERC) startinggrant SytActivity FP7 260916 and DeutscheForschungsgemeinschaft (DFG) grants DE19511 and -3 and theCenter for Nanoscale Microscopy and Molecular Physiology of theBrain (CNMPB) to CD DFG research group KFO241PsyCourseFi981-4Fi981 11-1 DFG project 1791-12013 ERC consolidatorgrant DEPICODE 648898 BMBF projects ENERGI (01GQ1421A)and Intergrament 01ZX1314D and funds from the German Centerfor Neurodegenerative Diseases to AF Canadian Institute forHealth Research grant FDN-154286 to YTW and a DorotheaSchloezer fellowship to KB Author contributions AAcoordinated the project designed conducted and analyzed allbehavior experiments designed and conducted receptor and Syt3internalization assays and immunocytochemistry experiments ofSyt3endocytic zones and PSDs conducted and analyzedpHluorin-Syt3 experiments conducted and trained and supervisedAH in mEPSC recordings trained and supervised NN inwhole-cell recordings from hippocampal slices and SR in Barnesmaze behavior experiments and revised the manuscript BRdesigned conducted and analyzed all LTP LTD and fieldrecording experiments and conducted all immunohistochemistryexperiments SA designed conducted and analyzed allbiochemistry experiments EB designed behavior experimentstrained AA in behavior experiments and performed allhippocampal viral injections YS designed conducted andanalyzed immunocytochemistry experiments of Syt3 localizationand all pHluorin-Syt3 experiments NN conducted and analyzedwhole-cell recordings from hippocampal slices AH conducted andanalyzed mEPSC recordings SR conducted and analyzed Barnesmaze behavior experiments HM designed developed andvalidated the Syt3 antibody JB trained AA in and helped planbehavior experiments KB performed biochemistry experimentsYTW generated and validated the GluA2-3Y peptide and helpedplan behavior experiments AF provided funding for JB andEB and helped plan behavior experiments CD conceived andcoordinated the project provided funding for AA BR SA YSand KB conducted and analyzed receptor and Syt3 internalizationassays and immunocytochemistry of Syt3 endocytic zones andPSDs and wrote and revised the manuscript Competing interestsThe authors declare no competing interests Data and materialsavailability All data are available in the manuscript or thesupplementary materials

SUPPLEMENTARY MATERIALS

wwwsciencemagorgcontent3636422eaav1483supplDC1Figs S1 to S8

18 August 2018 accepted 1 November 2018Published online 13 December 2018101126scienceaav1483




Dow

nloaded from

forgettingSynaptotagmin-3 drives AMPA receptor endocytosis depression of synapse strength and

Henrik Martens Jonas Barth Katja Burk Yu Tian Wang Andre Fischer and Camin DeanAnkit Awasthi Binu Ramachandran Saheeb Ahmed Eva Benito Yo Shinoda Noam Nitzan Alina Heukamp Sabine Rannio

originally published online December 13 2018DOI 101126scienceaav1483 (6422) eaav1483363Science

this issue p eaav1483 see also p 31Sciencethese animals showed signs of impaired forgetting and relearning during the water maze spatial memory taskabolished and decaying long-term potentiation endured In Syt3 knockout mice spatial learning was unaltered howevershort-term plasticity were unchanged However in neurons from Syt3 knockout mice synaptic long-term depression was the Perspective by Mandelberg and Tsien) In Syt3 overexpressing or knockdown neurons synaptic transmission andpredominantly found on postsynaptic endocytic zones of neurons where it promotes AMPA receptor internalization (see

found that the integral membrane protein synaptotagmin-3 (Syt3) iset aland underlies learning and memory Awasthi The trafficking of AMPA receptors to and from the surface of postsynaptic membranes regulates synaptic strength

Forgetting and receptor removal

ARTICLE TOOLS httpsciencesciencemagorgcontent3636422eaav1483

MATERIALSSUPPLEMENTARY httpsciencesciencemagorgcontentsuppl20181212scienceaav1483DC1

CONTENTRELATED

httpstkesciencemagorgcontentsigtrans12586eaav3577fullhttpstkesciencemagorgcontentsigtrans12585eaaw2040full

REFERENCES

httpsciencesciencemagorgcontent3636422eaav1483BIBLThis article cites 50 articles 19 of which you can access for free

PERMISSIONS httpwwwsciencemagorghelpreprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAASScienceScience 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 US Government WorksCopyright copy 2019 The Authors some rights reserved exclusive licensee American Association for the Advancement of

on June 18 2020


Dow

nloaded from

363_44

363_aav1483

RESEARCH ARTICLE

NEUROSCIENCE

Synaptotagmin-3 drives AMPAreceptor endocytosis depression ofsynapse strength and forgettingAnkit Awasthi1 Binu Ramachandran1 Saheeb Ahmed1dagger Eva Benito23 Yo Shinoda1DaggerNoam Nitzan1 Alina Heukamp1sect Sabine Rannio1 Henrik Martens4 Jonas Barth23Katja Burk1 Yu Tian Wang5 Andre Fischer23 Camin Dean1para

Forgetting is important Without it the relative importance of acquired memories in achanging environment is lost We discovered that synaptotagmin-3 (Syt3) localizes topostsynaptic endocytic zones and removes AMPA receptors from synaptic plasmamembranes in response to stimulation AMPA receptor internalization long-termdepression (LTD) and decay of long-term potentiation (LTP) of synaptic strength requiredcalcium-sensing by Syt3 and were abolished through Syt3 knockout In spatial memorytasks mice in which Syt3 was knocked out learned normally but exhibited a lack offorgetting Disrupting Syt3GluA2 binding in a wild-type background mimicked the lack ofLTP decay and lack of forgetting and these effects were occluded in the Syt3 knockoutbackground Our findings provide evidence for a molecular mechanism in which Syt3internalizes AMPA receptors to depress synaptic strength and promote forgetting

Afundamental property of the brain is theability to learn from experience by mod-ulating the strength of synaptic connec-tions between neurons The trafficking ofAMPA receptors to and from the surface

of postsynaptic membranes is a key determinantin the regulation of synaptic strength (1ndash3) High-frequency stimulation increases surface recep-tors and promotes long-term potentiation (LTP)of synaptic strength Low-frequency stimulationremoves receptors from the postsynaptic mem-brane and causes long-term depression (LTD) ofsynaptic strength This phenomenon has beenmost extensively studied inN-methyl-D-aspartate(NMDA)ndashreceptorndashdependent plasticity of CA3-CA1 synapses in the hippocampus (4) LTP iswidely believed to underlie learning (5ndash7) Con-versely weakening of potentiated synapses andLTD can promote forgetting (5 8 9)Both LTD (10 11) and the decay of LTP depend

on activity-dependent removal of postsynaptic

GluA2-containing AMPA receptors from the plas-ma membrane (8 12 13) Ca2+ influx into den-drites is critical for virtually all types of synapticplasticity (4) including decay of LTP (14 15) andactive internalization of GluA2-containing AMPAreceptors (16 17) The same signaling entity canhave divergent consequences for the synapse Fasthigh [Ca2+] influx can promote LTP whereasgradual low [Ca2+] influx can promote LTD (4)Plasma membranendashlocalized Ca2+-binding pro-teins are likely needed to remove or add receptorsto the postsynaptic membrane through regulatedendo- or exocytosis Synaptotagmins (Syts) arecandidates for such a function (18) This familyof integral membrane proteins contain a shortluminal tail transmembrane domain and twoconserved cytoplasmic Ca2+-binding C2 domains(19) that regulate Ca2+-dependent membrane re-cycling eventsOf the 17 mammalian Syt isoforms Syt3 is the

third most abundant in the brain Unlike Syt1and Syt2 which are predominantly thought to belocalized to synaptic vesicle membranes Syt3 wasreported to be localized to the plasma membrane(20 21) and to have a 10-fold higher Ca2+ affinitythan that of Syt1 and Syt2 (22) In a pHluorin-Sytscreen of Syt isoforms in response to stimulationSyt3 exhibited distinct recycling properties It wasthe only isoform to endocytose in response tostimulation and to recycle exclusively in den-drites (23) Because the kinetics of stimulus-induced pHluorin-Syt3 endocytosis resembledthose of pHluorin-GluA2 (24 25) we hypothe-sized that Syt3 may regulate Ca2+- and activity-dependent postsynaptic receptor endocytosis toaffect synaptic plasticity

ResultsSyt3 is on postsynapticplasma membranesTo examine the location of Syt3 we developeda highly specific antibody (fig S1 A and B) thatrecognized a single band in brain homogenateswhich was absent in Syt3 knockouts an antibodydeveloped by Neuromab showed similar results(Fig 1A) Syt3 was most abundant in adiposetissue heart and brain (fig S1C) where it wasfound in the hippocampus cortex thalamus andstriatum (fig S1D) Expression in the brain beganembryonically and remained high throughoutadulthood (fig S1E) Immunostains revealed Syt3signal on neuronal cell bodies and dendrites inthe CA1 (Fig 1B) CA3 and dentate gyrus (fig S1F)in wild-type but not Syt3 knockout hippocam-pal slicesTo localize Syt3 subcellularly we expressed

cytosolic green fluorescent protein (GFP) in cul-tured hippocampal neurons and immunostainedfor Syt3 andMAP2 todistinguish dendrites (MAP2-positive) from axons (GFP-positiveMAP2-negative)Syt3 signal was predominantly detected in den-drites (902 plusmn 30 of total signal) comparedwithaxons (98plusmn51of total signal) (Plt0001) (Fig 1C)Syt3 exhibited a punctate pattern in dendritesand colocalized with the pre- and postsynapticmarkers synaptophysin and PSD95 or GluA1respectively at synapses (Fig 1D) 706 plusmn 53of Syt3 signal was at excitatory synapses (figS1G) and 294 plusmn 19 was at inhibitory synapses(P lt 0001) (fig S1H) In subcellular fractionsSyt3 was associated with synaptosomal plasmamembranes and notwith synaptic vesicles (fig S1I)which is in agreement with a previous report (20)Immuno-organelle isolation of synaptic vesicleswith antibodies to Syt1 or Syb2 further confirmedthat Syt3 isnot present on synaptic vesicles (fig S1J)We further tested whether Syt3 is specifically local-ized to postsynaptic membranes using a trypsincleavage assay of synaptosomes (26) Synapto-somes form in such a way that the presynapticterminal seals trapping synaptic vesicles andother presynaptic components inside (Fig 1E)whereas the postsynaptic membrane does not re-seal Presynaptic proteinswere therefore protectedfrom trypsin cleavage whereas postsynaptic pro-teins including Syt3 were cleaved (Fig 1F)

Stimulation induces endocytosis of Syt3

The presence of Syt3 on postsynaptic membranessuggests it may regulate a post-synaptic re-cycling event Time-lapse imaging of pHluorin-Syt3 in transfected hippocampal neurons duringdepolarization with 45 mMKCl (23) or field stim-ulation revealed a calcium-dependent fluores-cence decrease requiringNMDAAMPA receptorsand L-type calcium channels likely correspondingto endocytosis (fig S2 A to D) PHluorin-Syt3 alsoexhibited calcium-dependent endocytosis in re-sponse to AMPA (Fig 2A) and NMDA (Fig 2B)mdashstimuli that promoteAMPAreceptor internalizationmdashwith kinetics similar to those of pHluorin-taggedAMPA receptors (24 25)Because interpretation of pHluorin experiments

may be confounded by intracellular acidification

RESEARCH


1Trans-synaptic Signaling Group European NeuroscienceInstitute 37077 Goettingen Germany 2German Center forNeurodegenerative Disease 37075 Goettingen Germany3Department of Psychiatry and Psychotherapy UniversityMedical Center Goettingen 37075 Goettingen Germany4Synaptic Systems GmbH 37079 Goettingen Germany 5BrainResearch Center and Department of Medicine University ofBritish Columbia Vancouver BC V6T2B5 CanadaThese authors contributed equally to this workdaggerPresent address Department of Diagnostic and InterventionalRadiology University Medical Center Goettingen 37075 GoettingenGermanyDaggerPresent address Department of Environmental Health School ofPharmacy Tokyo University of Pharmacy and Life Sciences Tokyo192-0392 Japan sectPresent address Department of NeurobiologyWeizmann Institute of Science 7610001 Rehovot IsraelparaCorresponding author Email cdeaneni-gde

on June 18 2020


Dow

nloaded from









D

CSyt3 MAP2 GFP



BA

70 kD-55 kD-

25 kD-

Wt KO Wt KO

Syt3 NT

Tubulin

Syt3 NB

Syp

no tr

10acute

20acute

30acute

60acute

90acute

Syn

Syb2

Rab3a

SNAP25

Piccolo

Syx1A

PSD95

Homer

GluA1

GluN1

Syt3 NB

Syt3 NT

Nrx

Nlg

F

Syt1

Syp

Rab3a Syn

SNAP25

Piccolo

Syx

Nrx NlgGluN1 GluA1

PSD95Homer

Syb2

Syt3

E

protectedcleaved

Wt

KO

Syt3 MAP2 merge

no tr

10acute

20acute

30acute

60acute

90acute




Dow

nloaded from


inte

nsity

(

)

time (s)

NH4ClNMDA

inte

nsity

(

)AMPA

time (s)

NH4Cl

inte

nsity

(

) 0 Ca2+

NH4ClNMDA

2mM Ca2+

inte

nsity

(

)

time (s)

0 Ca2+ 2 mM Ca2+

NH4ClAMPA

A

00

05

10

20

inte

rnal

sur

face

Syt

3

ctrl

AMPA

NMDA

15


cont

rol

AM

PAN

MD

A


GFP-Syt3

CLC-DsRed

Homer-GFP GFP-Syt3

Homer-mycCLC-DsRed


GluA2 C-tail



inpu

tSy

t3 +

GRIP

AP-2

GluN2A

GluA2

Syt3

ndash

GluA1

GluA3

GluN1

GABAARα1

Pick1

BRAG2

FE

G

0 s 100 s 300 s0 s 100 s 300 s

B

GluA2

BRAG2

input

neg

ctrl

EGTA0

Ca2+

10 micro

M C

a2+

100

microM C

a2+

1000

microM

Ca2+

AP2

140

120

100

80

60

40

20

0

140

120

100

80

60

40

20

0 100 200 300

0 50 100

140

120

100

80

60

40

20

00 100 200 300

140

120

100

80

60

40

20

time (s)0 50 100

834 883mdash

input

neg

ctrl

100

microM C

a2+

GluA2


H






Dow

nloaded from





00

05

10

15

20

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

inte

rnal

su

rfac

e G

luA

1



control AMPA

EG

FP

Glu

A2

surf

ace

Glu

A2

inte

rnal

Glu

A2

surf

ace

Glu

A2

inte

rnal


GFP

Syt3 KDGFP-Syt3

Syt3 Ca2+ mut

WTSyt3 KOWT+ 3Y

control AMPA NMDA

A

CB

D E

Syt3 Ca2+ mut





Dow

nloaded from


















Dow

nloaded from







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483









D

CSyt3 MAP2 GFP



BA

70 kD-55 kD-

25 kD-

Wt KO Wt KO

Syt3 NT

Tubulin

Syt3 NB

Syp

no tr

10acute

20acute

30acute

60acute

90acute

Syn

Syb2

Rab3a

SNAP25

Piccolo

Syx1A

PSD95

Homer

GluA1

GluN1

Syt3 NB

Syt3 NT

Nrx

Nlg

F

Syt1

Syp

Rab3a Syn

SNAP25

Piccolo

Syx

Nrx NlgGluN1 GluA1

PSD95Homer

Syb2

Syt3

E

protectedcleaved

Wt

KO

Syt3 MAP2 merge

no tr

10acute

20acute

30acute

60acute

90acute




Dow

nloaded from


inte

nsity

(

)

time (s)

NH4ClNMDA

inte

nsity

(

)AMPA

time (s)

NH4Cl

inte

nsity

(

) 0 Ca2+

NH4ClNMDA

2mM Ca2+

inte

nsity

(

)

time (s)

0 Ca2+ 2 mM Ca2+

NH4ClAMPA

A

00

05

10

20

inte

rnal

sur

face

Syt

3

ctrl

AMPA

NMDA

15


cont

rol

AM

PAN

MD

A


GFP-Syt3

CLC-DsRed

Homer-GFP GFP-Syt3

Homer-mycCLC-DsRed


GluA2 C-tail



inpu

tSy

t3 +

GRIP

AP-2

GluN2A

GluA2

Syt3

ndash

GluA1

GluA3

GluN1

GABAARα1

Pick1

BRAG2

FE

G

0 s 100 s 300 s0 s 100 s 300 s

B

GluA2

BRAG2

input

neg

ctrl

EGTA0

Ca2+

10 micro

M C

a2+

100

microM C

a2+

1000

microM

Ca2+

AP2

140

120

100

80

60

40

20

0

140

120

100

80

60

40

20

0 100 200 300

0 50 100

140

120

100

80

60

40

20

00 100 200 300

140

120

100

80

60

40

20

time (s)0 50 100

834 883mdash

input

neg

ctrl

100

microM C

a2+

GluA2


H






Dow

nloaded from





00

05

10

15

20

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

inte

rnal

su

rfac

e G

luA

1



control AMPA

EG

FP

Glu

A2

surf

ace

Glu

A2

inte

rnal

Glu

A2

surf

ace

Glu

A2

inte

rnal


GFP

Syt3 KDGFP-Syt3

Syt3 Ca2+ mut

WTSyt3 KOWT+ 3Y

control AMPA NMDA

A

CB

D E

Syt3 Ca2+ mut





Dow

nloaded from


















Dow

nloaded from







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483


inte

nsity

(

)

time (s)

NH4ClNMDA

inte

nsity

(

)AMPA

time (s)

NH4Cl

inte

nsity

(

) 0 Ca2+

NH4ClNMDA

2mM Ca2+

inte

nsity

(

)

time (s)

0 Ca2+ 2 mM Ca2+

NH4ClAMPA

A

00

05

10

20

inte

rnal

sur

face

Syt

3

ctrl

AMPA

NMDA

15


cont

rol

AM

PAN

MD

A


GFP-Syt3

CLC-DsRed

Homer-GFP GFP-Syt3

Homer-mycCLC-DsRed


GluA2 C-tail



inpu

tSy

t3 +

GRIP

AP-2

GluN2A

GluA2

Syt3

ndash

GluA1

GluA3

GluN1

GABAARα1

Pick1

BRAG2

FE

G

0 s 100 s 300 s0 s 100 s 300 s

B

GluA2

BRAG2

input

neg

ctrl

EGTA0

Ca2+

10 micro

M C

a2+

100

microM C

a2+

1000

microM

Ca2+

AP2

140

120

100

80

60

40

20

0

140

120

100

80

60

40

20

0 100 200 300

0 50 100

140

120

100

80

60

40

20

00 100 200 300

140

120

100

80

60

40

20

time (s)0 50 100

834 883mdash

input

neg

ctrl

100

microM C

a2+

GluA2


H






Dow

nloaded from





00

05

10

15

20

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

inte

rnal

su

rfac

e G

luA

1



control AMPA

EG

FP

Glu

A2

surf

ace

Glu

A2

inte

rnal

Glu

A2

surf

ace

Glu

A2

inte

rnal


GFP

Syt3 KDGFP-Syt3

Syt3 Ca2+ mut

WTSyt3 KOWT+ 3Y

control AMPA NMDA

A

CB

D E

Syt3 Ca2+ mut





Dow

nloaded from


















Dow

nloaded from







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483





00

05

10

15

20

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

00

05

10

15

20

inte

rnal

su

rfac

e G

luA

2

inte

rnal

su

rfac

e G

luA

1



control AMPA

EG

FP

Glu

A2

surf

ace

Glu

A2

inte

rnal

Glu

A2

surf

ace

Glu

A2

inte

rnal


GFP

Syt3 KDGFP-Syt3

Syt3 Ca2+ mut

WTSyt3 KOWT+ 3Y

control AMPA NMDA

A

CB

D E

Syt3 Ca2+ mut





Dow

nloaded from


















Dow

nloaded from







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483


















Dow

nloaded from







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483







1 microM GluA2-3Y

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

Syt3 KO

WT

E5 mV

5 ms

0 60 120 180 24050

75

100

125

150

175

200

time (min)fE

PS

P (m

Vm

s)

1 microM ZIP

F

Syt3 KO + ZIP

WT + ZIP

5 mV5 ms

0 30 60 90 12050

75

100

125

150

-30

B

time (min)

fEP

SP

(mV

ms)

WT

Syt3 KO

5 mV5 ms

C

-30 0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms)

WT + GluA2-3Y + ZIP

WT + GluA2-3Y

1 microM ZIP

1 microM GluA2-3Y

5 mV5 ms

-30 0 60 120 180 24050

75

100

125

150

time (min)

fEP

SP

(mV

ms)

Syt3 KO

Syt3 KO + ZIP

1 microM ZIP

D

Syt3 KO + GluA2-3Y

5 mV5 ms

-20 0 20 40 60 8025

50

75

100

125

time (min)

fEP

SP

(mV

ms)

WTSyt3 KO

A

WT + GluA2-3Y

5 mV5 ms

CA3

AAV-Syt3-P2A-GFP

CA1

DG

-20 0 20 40 60 8050

75

100

125

time (min)

fEP

SP

(mV

ms )


5 mV5 ms

H

0 60 120 180 24050

75

100

125

150

175

200

time (min)

fEP

SP

(mV

ms )

1 microM ZIP


5 mV5 ms

K

-20 0 20 40 60 80 100 12050

75

100

125

150

time (min)

fEP

SP

(mV

ms)


5 mV5 ms

J

60

80

100

120

140

160

frequency (Hz)

fEP

SP

(mV

ms)

WT

Syt3 KOG

02 1 10 100

I

1 microM GluA2-3Y



Dow

nloaded from


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483


A

1 2 3 4 5 6 7 8 9 10 11 12 13 1425

30

35

40

45

50

prox

imity

to p

latfo

rm (

cm)

DaysP

T1

PT

2

PT

3

Rev

WT

Syt

3 K

O

17 18 19 20 21 22 2320

30

40

Days

prox

imity

to

orig

inal

pla

tform

(cm

)

25

35

45

30

40

50

prox

imity

toor

igin

al p

latfo

rm (

cm)

35

45

WTSyt3 KO

Day 22

WT

Syt

3 K

Ore

scue

Syt

3 K

OW

T

PT20-10s 10-20s 20-30s 30-40s 40-50s 50-60s 0-60s

20

30

40

50

10-2

0 s

50-6

0 s

prox

imity

to p

latfo

rm (

cm) WT

KO

E F

G

00

02

04

06

08

00

02

04

06

08


Day

8-1

3 pe

rsev

eren

ce

PT

4-6

pers

ever

ence

WT + salL

WT

WTKO

KO cohort 1

cohort 2

PT3

K

B

0

10

20

30

40

50

60

prox

imity

to p

latfo

rm (

cm)

02468

101214161820

plat

form

cro

ssin

gs

0102030405060708090

ti

me

targ

et q

uadr

ant

C D


WTKO


Syt3 KO + Syt3

late

ncy

to c

urre

nt h

o le

( s)

1 2 3 4 5 6 7

PT

1 8 9 10 11P

T3 12 13

PT

4P

T5

PT

6

0

25

50

75

100

WT + sal

KO + 3Y

WT + 3YKO + sal

Rev

PT

2

Days

J

Day

8-1

3


PT

1-2

PT

4-6

H

WT+

sal

W

T +

3YKO

+ 3

Y0

10

20

30

40

50

30

40

50

ti

me

orig

inal

TQ

35

45

55

25

IPT210-20s 50-60s 0-60s

WT+

sal

W

T +

3Y K

O +

3Y

10-2

0 s15

25

35

45

55

50-6

0 s

WT + 3YKO + 3Y

WT + sal

prox

imity

to p

latfo

rm (

cm)

PT5

prox

imity

to o

rigin

al p

latfo

rm (

cm)

WT + 3YKO + 3Y

WT + sal





Dow

nloaded from







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483







WT Syt3 KO

Day

4D

ay 6

Day

9D

ay 1

4D

ay 1

6

B

Pro

be te

st

prox

imity

nor

m

to W

T (

cm)

Training Day

D

Syt3 KOWT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16V3V2V1

C

Training DayV1 V2 V3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

T

rials

cla

ssifi

ed a

s pe

rsev

eran

ce to

pre

viou

s da

yrsquos

plat

form

pos

ition

0

10

20

30

40WTSyt3 KO

A

-6

-4

-2

0


trial 2 probe test



KOWT

prox

imity

sav

ings

(cm

)

4

20

-2

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

15

10

5

0

1 2 3 4Trial

121086420

1 2 3 4Trial

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

prox

imity

sav

ings

(cm

)

142020

68

10

KOWT

KOWT

KOWT





Dow

nloaded from


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483


Discussion



















Dow

nloaded from









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483









pHluorin imaging







Syt3 pulldowns







Dow

nloaded from















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483















Dow

nloaded from




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483




















Dow

nloaded from




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483




Barnes Maze

























Dow

nloaded from









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483









































ACKNOWLEDGMENTS








Dow

nloaded from









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483









CONTENTRELATED


REFERENCES






on June 18 2020


Dow

nloaded from

363_44

363_aav1483

synaptotagmin-3 drives ampa receptor endocytosis ......ankit awasthi 1*, binu ramachandran *, saheeb...

Documents

synaptotagmin-3 drives ampa receptor endocytosis ......ankit awasthi 1, binu ramachandran , saheeb...