RESEARCH ARTICLE SUMMARY

NEUROSCIENCE

Autism-associated SHANK3haploinsufficiency causes Ihchannelopathy in human neuronsFei Yi,* Tamas Danko,* Salome Calado Botelho, Christopher Patzke, ChangHui Pak,Marius Wernig, Thomas C. Südhof†

INTRODUCTION: SHANK3 is a scaffoldingprotein that is enriched in postsynaptic den-sities of excitatory synapses but ubiquitouslyexpressed in most cells. SHANK3 gene muta-tions are significantly associated with autismspectrum disorders (ASDs), and deletion ofSHANK3 is thought to cause the major symp-toms of Phelan-McDermid syndrome. Moreover,increasing evidence links SHANK3mutations toschizophrenia. Because SHANK3 is a synapticprotein, SHANK3 mutations are thought topredispose to neuropsychiatric disorders by im-pairing synaptic function. How SHANK3 mu-tations are pathogenic, however, remains unclear.

RATIONALE:Human neurons derived fromPhelan-McDermid syndrome patients displaycomplex abnormalities, including synaptic def-

icits and altered intrinsic electrical properties.Although some of these abnormalities are re-versed by SHANK3 reexpression, the alteredelectrical properties are difficult to reconcilewith a primarily synaptic impairment. Moreover,in mice, Shank3 deletions produce behavioralchanges and synaptic transmission deficits, al-though no cellular phenotype has been identified.Here, we explored the pathogenetic mechanismof human SHANK3 mutations with a condi-tional genetic approach in human neurons andcorrelated the results with those obtained inShank3-mutant mouse neurons. We introducedconditional SHANK3 deletions into humanembryonic stem cells and examined isogeniccontrol and heterozygous and homozygousSHANK3-mutant neurons derived from theseconditionally mutant cells. In addition, we

analyzed developing mouse Shank3-mutantneurons and compared their phenotype withthat of human SHANK3-mutant neurons.

RESULTS: Heterozygous and homozygousSHANK3-mutant human neurons displayeddiverse abnormalities, ranging from a mas-sive increase in input resistance to increasedexcitability, modest impairments in dendriticarborization, and decreases in synaptic trans-mission. Because the increased input resistancesuggested an altered channel conductanceas a primary impairment, we tested variousconductances. We found that the SHANK3mutations caused a profound impairment inhyperpolarization-activated cation (Ih) currents,which are mediated by hyperpolarization-activated cyclic nucleotide-gated (HCN) chan-nels. This impairment produced the increasedinput resistance; moreover, chronic pharmaco-

logical inhibition of Ihcurrents in wild-type hu-man neurons impaireddendritic arborization andsynaptic transmission sim-ilar to the SHANK3muta-tions. Mechanistically, we

detected a direct interaction of HCN channelswith SHANK3 protein and observed a de-crease in HCN-channel proteins in SHANK3-mutant neurons. Finally, we found that developinghippocampal neurons cultured from heterozy-gous and homozygous Shank3-mutant micealso exhibited an increased input resistance,reduced Ih currents, and an increased excitabilitysimilar to SHANK3-mutant human neurons.

CONCLUSION: Using human neurons withconditional SHANK3mutations, we found thatSHANK3mutations impair Ih-channel function,thereby increasing neuronal input resistanceand enhancing neuronal excitability. This im-pairment in intrinsic electrical properties ac-counts, at least in part, for the decreaseddendritic arborization and synaptic transmis-sion of SHANK3-mutant neurons. The reducedIh-current phenotype manifests early in neu-ronal development and is similarly observedin immature Shank3-mutant mouse neurons.We propose that, in addition to having a spe-cifically postsynaptic function, SHANK3 proteinmay perform a general role during neurodevel-opment by scaffolding HCN channels that me-diate Ih currents in neurons and nonneuronalcells consistent with the ubiquitous expressionof SHANK3. Thus, we hypothesize that SHANK3mutations induce an Ih channelopathy thatcontributes to ASD pathogenesis and may beamenable to pharmacological intervention.

RESEARCH

672 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

The list of author affiliations is available in the full article online.*These authors contributed equally to this work.†Corresponding author. Email: tcs1@stanford.eduCite this article as F. Yi et al., Science 352, aaf2669 (2016).DOI: 10.1126/science.aaf2669

Conditional SHANK3 deletion in human neurons impairs Ih channel. Comparison of isogeniccontrol and SHANK3-deficient human neurons reveals that heterozygous and homozygous SHANK3mutations dramatically decrease Ih-channel function, resulting in multifarious secondary impairments,including a decrease in dendritic arborization and synaptic responses and an increase in input resistanceand neuronal excitability.

ON OUR WEBSITE

Read the full articleat http://dx.doi.org/10.1126/science.aaf2669..................................................

on February 16, 2019

http://science.sciencem

ag.org/D

ownloaded from

RESEARCH ARTICLE

NEUROSCIENCE

Autism-associated SHANK3haploinsufficiency causes Ihchannelopathy in human neuronsFei Yi,1* Tamas Danko,1,2,3* Salome Calado Botelho,1 Christopher Patzke,1

ChangHui Pak,1 Marius Wernig,2,3 Thomas C. Südhof 1,4†

Heterozygous SHANK3 mutations are associated with idiopathic autism andPhelan-McDermid syndrome. SHANK3 is a ubiquitously expressed scaffolding proteinthat is enriched in postsynaptic excitatory synapses. Here, we used engineeredconditional mutations in human neurons and found that heterozygous and homozygousSHANK3 mutations severely and specifically impaired hyperpolarization-activatedcation (Ih) channels. SHANK3 mutations caused alterations in neuronal morphologyand synaptic connectivity; chronic pharmacological blockage of Ih channels reproducedthese phenotypes, suggesting that they may be secondary to Ih-channel impairment.Moreover, mouse Shank3-deficient neurons also exhibited severe decreases in Ihcurrents. SHANK3 protein interacted with hyperpolarization-activated cyclicnucleotide-gated channel proteins (HCN proteins) that form Ih channels, indicatingthat SHANK3 functions to organize HCN channels. Our data suggest that SHANK3mutations predispose to autism, at least partially, by inducing an Ih channelopathy thatmay be amenable to pharmacological intervention.

Haploinsufficiency of SHANK3 constitutesone of the more frequent single-gene muta-tions in autism spectrum disorders (ASDs)(1–5). Moreover, SHANK3 is among severalgenes deleted in Phelan-McDermid syn-

drome (22q13.3 deletion syndrome), and SHANK3haploinsufficiency may be the most importantcontributing factor to Phelan-McDermid syndromepathology (1–5). In addition, emerging evidencelinks SHANK3 mutations to schizophrenia (6, 7),and SHANK3 overexpression has also been im-plicated in neuropsychiatric disorders (8). Anal-ysis of human neurons differentiated from inducedpluripotent stem cells (iPSC) from patients withPhelan-McDermid syndrome have revealed twoapparently disparate phenotypes, an impairmentin synaptic transmission and an increase in inputresistance (9). The synaptic impairments of Phelan-McDermid neurons are rescued with SHANK3 (9),consistent with the presence of SHANK3 in post-synaptic specializations where SHANK3 is thoughtto function as a scaffolding protein that organizesreceptor signaling (10–12). Thus, Phelan-McDermid

syndrome may result from a synaptic impairmentcaused by SHANK3 haploinsufficiency. The dra-matically increased input resistance of Phelan-McDermid neurons, however, remains an enigma(9). Because human neurons with sole SHANK3mutations have not been analyzed, the role ofSHANK3 in the phenotypes of Phelan-McDermidneurons remains incompletely understood, as doesthe functional effect of a pure SHANK3 haploin-sufficiency on human neurons. In mice, hetero-zygous and homozygous Shank3mutations causea behavioral phenotype resembling autism and/orobsessive-compulsive disorders and produce changesin synaptic transmission (13–21). However, theunderlying molecular mechanisms and possiblecontributory nonsynaptic effects are unclear, asare the cellular effects of murine Shank3mutations.To address these issues that are crucial for

progress toward understanding ASDs and Phelan-McDermid syndrome, we generated human neu-rons with conditional heterozygous and homozygousSHANK3 loss-of-functionmutations. Unexpectedly,we found that, besides causing synaptic impair-ments, loss of SHANK3 function selectively andseverely impaired hyperpolarization-activated cat-ion (Ih) currents. We observed that SHANK3 pro-tein directly bound to hyperpolarization-activatedcyclic nucleotide-gated channel (HCN) proteinsmediating Ih currents and that at least some ofthe synaptic impairments in SHANK3-mutanthuman neurons are an indirect result of the Ihchannelopathy produced by the SHANK3 muta-tions during neuronal development. Finally, weobserved a similar phenotype in Shank3-deficient

mouse neurons, suggesting a general function forSHANK3 in scaffolding HCN channels.

Generation of heterozygous SHANK3conditional knockout (cKO) mutations

To investigate the role of SHANK3 mutations inhuman neurons, we constructed conditional muta-tions in the SHANK3 gene in human H1 embryonicstem cells (ES cells) using homologous recombina-tion (Fig. 1, A and B, and fig. S1) (22). We chosethis approach because it enables generation ofmatching control and mutant neurons from thesame ES cell clones, thus eliminating subclone-to-subclone variability. The SHANK3 locus expressesmultiple transcripts that encode different SHANK3isoforms with distinct protein interaction domains(Fig. 1A and fig. S2A). To conditionally delete majorSHANK3 isoforms, we targeted exon 13 of theSHANK3 gene whose deletion causes a frame-shift in all major SHANK3 mRNAs.We converted heterozygous conditionally mutant

SHANK3+/cKO ES cells into neurons by forced ex-pression of the transcription factor Ngn2, whichgenerates a homogenous population of glutama-tergic excitatory neurons with abundant synapseformation (23). We coexpressed active (Cre) ormutant Cre-recombinase (DCre) with Ngn2 duringneuronal differentiation to produce preciselymatching wild-type [DCre (SHANK3+/+)] and het-erozygous mutant neurons [Cre (SHANK3+/−)](24, 25), using two independently targeted hetero-zygous clones of SHANK3+/cKO ES cells to controlfor clonal variation (Fig. 1C). Immunoblotting andpolymerase chain reaction (PCR) demonstratedthat the heterozygous SHANK3 KO decreasedSHANK3 expression but left SHANK1 and SHANK2expression unchanged (Fig. 1D and figs. S1, D andE, S2B, and S3). No significant changes in othersynaptic proteins were observed except for a de-crease in PSD95 that is known to bind to SHANKs(fig. S2, C and D) (10–12).

SHANK3 haploinsufficiency impairsdendritic arborization, intrinsicelectrical properties, and synaptictransmission in human neurons

Human SHANK3+/− neurons exhibited a typicalneuronal morphology with abundant synapse forma-tion (Fig. 1E). When we quantified the morphologicalfeatures of matching SHANK3+/+ and SHANK3+/−

neurons derived from independently derived ES cellclones, we observed that SHANK3 haploinsufficiencysignificantly decreased the length and branching ofneurites but had no significant effect on the densityor size of synapsin-positive synapses (Fig. 1, F to I).Next, we performed whole-cell patch-clamp re-

cordings frommatching SHANK3+/+ and SHANK3+/−

neurons. SHANK3 haploinsufficiency caused a largeincrease (~25 to 33%) in input resistance with-out a change in capacitance (Fig. 2A). Moreover,SHANK3+/− neurons exhibited a major decrease inevoked excitatory postsynaptic currents (EPSCs)(~40 to 50% decrease) and in the amplitude ofspontaneous miniature EPSCs (mEPSCs) (~25% de-crease) but not in other mEPSC parameters (Fig. 2,B and C, and fig. S2E). Overall, these results resem-ble those obtained with Phelan-McDermid neurons

RESEARCH

SCIENCE sciencemag.org 6 MAY 2016 • VOL 352 ISSUE 6286 aaf2669-1

1Department of Molecular and Cellular Physiology, StanfordUniversity School of Medicine, 265 Campus Drive, Stanford, CA94305, USA. 2Institute for Stem Cell Biology and RegenerativeMedicine, Stanford University School of Medicine, 265 CampusDrive, Stanford, CA 94305, USA. 3Department of Pathology,Stanford University School of Medicine, 265 Campus Drive,Stanford, CA 94305, USA. 4Howard Hughes Medical Institute,Stanford University School of Medicine, 265 Campus Drive,Stanford, CA 94305, USA.*These authors contributed equally to this work. †Correspondingauthor. Email: tcs1@stanford.edu

on February 16, 2019

http://science.sciencem

ag.org/D

ownloaded from

(9), supporting the notion that despite the largegenomic deletion present in Phelan-McDermidneurons, SHANK3 haploinsufficiency may accountfor most Phelan-McDermid syndrome phenotypes.

Homozygous SHANK3 deletionaggravates SHANK3haploinsufficiency phenotype

To inquire whether homozygous SHANK3 muta-tions produce phenotypes similar to those of het-

erozygous SHANK3 mutations, we generatedmutant H1 ES cells with homozygous conditionalSHANK3mutations (22) and converted them intomatching SHANK3+/+ or SHANK3−/− neurons bycoexpression of Ngn2 and inactive or activeCre-recombinase, respectively (Fig. 3A and fig.S4). Immunoblotting confirmed that the majorSHANK3 protein isoforms were no longer ex-pressed in SHANK3−/− neurons (Fig. 3B and fig.S4E). Quantitative analyses uncovered a similar,

but more severe, phenotype in SHANK3−/− neuronsthan in SHANK3+/− neurons, with a significantdecrease in dendritic arborization and in synapsedensity (Fig. 3, C to F, and fig. S5). Electrophysio-logically, SHANK3−/− neurons also exhibited a largeincrease in input resistance, a massive decrease inevoked EPSC amplitude, and a significant decreasein mEPSC amplitude similar to SHANK3+/− neu-rons (Fig. 3, G to I). In addition, SHANK3−/−

neurons displayed a decrease in capacitance and

aaf2669-2 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

Fig. 1. SHANK3 haploinsufficiencyimpairs dendritic development ofhuman neurons. (A) Diagram of theSHANK3 gene and of the three majorSHANK3 transcripts that are blocked byconditional deletion of exon 13 (yellow).(B and C) SHANK3 targeting strategy inhuman ES cells. Homologous recombi-nation wasmediated using recombinantadeno-associated virus (AAV) (B) andconfirmed by PCR (fig. S1B).The PGK-puromycin resistance cassette (brownbox) was excised by Flp-recombinase togenerate the cKO allele (SHANK3+/cKO)(C). SHANK3+/cKO ES cells wereconverted into human SHANK3+/+ orSHANK3+/− neurons by coexpression ofNgn2 with either mutant inactive Cre-recombinase (DCre) or active Cre-recombinase (Cre). (D) Reduction ofSHANK3 protein levels in humanSHANK3+/− neurons derived from twoindependent SHANK3+/cKO ES cellclones. (E) Representative images ofSHANK3+/+ and SHANK3+/− neurons(day 21) labeled by double immuno-fluorescence for MAP2 (red) and syn-apsin (green). (F) Representativedendritic arborization analyses byMetaMorph software of SHANK3+/+ andSHANK3+/− neurons (sparsely trans-fected with EGFP). (G) SHANK3 hap-loinsufficiency impairs dendriticarborization (summary graphs of indi-cated parameters measured inmatching SHANK3+/+ and SHANK3+/−

neurons derived from independentSHANK3+/cKO ES cell clones; normal-ized to SHANK3+/+ controls). (H) Rep-resentative images of SHANK3+/+ andSHANK3+/− dendrites stained for MAP2(red) and synapsin (green) for analysisof synaptic puncta by MetaMorphsoftware. (I) Summary graphs of den-dritic synaptic puncta density and size inSHANK3+/+ and SHANK3+/− neuronsderived from two independentSHANK3+/cKO ES cell clones. Data in(G) and (I) are means ± SEM. Numbersof cells/independent cultures analyzedare shown in the bars. Statistical signif-icance was evaluated by Student’s t test(*P < 0.05; **P < 0.01). For additionaldata, see figs. S1 and S2.

SHANK3WT gene

SHANK3targetingvector

Exon 13 (76 bp)

5 homology arm

1302 bp 1462 bp

Neu

rite

trac

ing

G

FP

Sta

inin

g

Arb

itrar

y un

its(n

orm

aliz

ed to

con

rol)

Total neurite length Number of primaryprocesses

Total branches Soma area

0

0.4

0.6

0.8

1.0

1.2

* **

* **

0

3

6

9

12

Syn

aptic

pun

cta

per

10 µ

m d

endr

ite

0

0.4

0.8

1.0

0.6

0.2

10 µm

MAP2 / SYNAPSIN Puncta analysis

10 µm

Puncta analysis

30 µm

0.2

55/3

56/3

54/3

42/3

18/3

18/3

18/3

18/3

Pun

cta

size

(µm

2 )

SHANK3+/+ SHANK3+/-

Clone A

SK3+/+

SK3+/-

Clone B

SHANK3cKO allele LoxP

133 homology arm

PGK-Puro

FRT LoxP

LoxP FRT LoxPFRT

Flp-recombinaseSHANK3

targetedgene

PGK-Puro

homologous recombination

∆CreCre

SHANK3+/+

SHANK3+/-13

13

13

13

SHANK3+/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone B

SHANK3+/+

SHANK3+/-

Clone A

SHANK3+/+

SHANK3+/-

Clone B

0

0.4

0.6

0.8

1.0

1.2

0.2

0

0.4

0.6

0.8

1.0

1.2

0.2

0

0.4

0.6

0.8

1.0

1.2

0.2

SHANK3+/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone BSHANK3+

/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone BSHANK3+

/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone B

SHANK3+/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone BSHANK3+

/+

SHANK3+/-

Clone ASHANK3+

/+

SHANK3+/-

Clone B

Synaptic puncta sizeSynaptic puncta density

MAP2 / SYNAPSIN

SK3+/+

SK3+/-

MA

P2

/ SY

NA

PS

IN

SHANK3+/+ SHANK3+/-

250

150

50

kDa

SHANK3

GDI

132 43 81 9 1110 12 1716 19 21 2218 207 15

ANK Proline-richSH3 SAMPDZ

SHANK3a (exon 1-22) SHANK3b (exon 4-22)

SHANK3c (exon 12-22)

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

a notable decline in mEPSC frequency (Fig. 3, Gand I).Viewed together, our results suggest that con-

ditional heterozygous and homozygous deletionsof SHANK3 produce similar phenotypes in humanneurons. Although the SHANK3-mutant neuronsclearly exhibit synaptic impairments, their increasedinput resistance and decreased dendritic arboriza-tion cannot be readily explained in terms of asynaptic change, prompting us to search for otherpathogenetic mechanisms.

Heterozygous and homozygous SHANK3mutations impair Ih currents

The input resistance of neurons depends at leastin part on their ionic conductances and ion channels.Thus, we first asked whether changes in voltage-gated Na+ channels or K+ channels (delayed recti-fier), which generate action potentials and determineneuronal excitability, could be responsible for theincreased input resistance of SHANK3-mutantneurons, prompted in part by binding of SHANK3to K+ channels (26). However, we detected nochanges in Na+ or K+currents in SHANK3-mutant

neurons (fig. S6). We next investigated whetherdecreased ionotropic glutamate receptor activa-tion caused by impaired synaptic transmissionin SHANK3-mutant neurons might be responsible.However, blocking ionotropic glutamate receptorshad no effect on input resistance in SHANK3+/+ orSHANK3+/− neurons (Fig. 4A).A third candidate for an impaired membrane

conductance as a cause of the increased inputresistance is Ih currents that are mediated byHCN channels. In mammals, HCN channels areencoded by four genes (HCN1 to HCN4) and areexpressed, like SHANKs, in neuronal and non-neuronal cells (27–30) (fig. S7). HCN channelsmediate hyperpolarization-activated Ih currentsthat depolarize membranes toward the action-potential threshold and reducemembrane resistance.Ih currents control neuronal excitability, mem-brane resting potentials, dendritic integration ofsynaptic potentials, and rhythmic oscillation ofneurons (29, 30). Given their multifaceted functions,impairments of Ih currents can have profoundconsequences for neuronal network activity [e.g.,see (31–33)].

Strikingly, we found that the Ih-current inhib-itor ZD7288 (34, 35) increased the input resistanceof wild-type neurons dramatically but had only asmall effect on SHANK3-mutant neurons, therebyabolishing the difference in input resistance be-tween wild-type and SHANK3-mutant neurons(Fig. 4B). Moreover, SHANK3-deficient neuronsdisplayed an increased resting membrane potential,and the addition of ZD7288 to wild-type neuronsincreased their resting potential to that of SHANK3-deficient neurons (Fig. 4C).Because these results suggest that the changed

electrical properties of SHANK3-mutant neuronsmay be due to an impairment of Ih currents, we nextdirectly measured Ih currents in precisely matchingSHANK3+/+ and SHANK3+/− or SHANK3−/− neu-rons using whole-cell recordings (Fig. 4, D and E,and fig. S8). Ih currents were readily activated bybrief (2 s) hyperpolarizing voltage steps, exhibited areversal potential of –32 mV, and were inhibited byextracellular Cs+ and ZD7288 (fig. S8). Comparedwith SHANK3+/+ neurons, both SHANK3+/− andSHANK3−/− neurons exhibited a severe decreasein Ih-current density (Fig. 4, D and E) and a

SCIENCE sciencemag.org 6 MAY 2016 • VOL 352 ISSUE 6286 aaf2669-3

25 ms

100

pA

10 s

30 p

A

0

100

200

300

29/3

29/3

**

Rin

(G

Ω)

0

0.25

0.50

0.75

***

57/3

57/3

Cm

(pF

)

0

20

40

60

57/3

57/3

20 ms

100

pA

10 s

30 p

A

0

0.25

0.50

0.75

0

20

40

60

46/3

47/3

***

46/3

47/3

0 24/3

25/3

**

Interevent interval (s)0 10 20 30 40 50

Cum

ulat

ive

prob

abili

ty

0.0

0.5

1.0

mEPSC amplitude (pA)0 40 80 120

Cum

ulat

ive

prob

abili

ty

0.0

0.5

1.0 ***

Cum

ulat

ive

prob

abili

ty

0.0

0.5

1.0

Interevent interval (s)0 10 20 30 40

mEPSC amplitude (pA)0 40 80 120

Cum

ulat

ive

prob

abili

ty

0.0

0.5

1.0

***E

PS

C a

mpl

itude

(pA

)

100

200

300

EP

SC

am

plitu

de (

pA)

SHANK3+/+

SHANK3+/-

Clone A

SHANK3+/+ SHANK3+/-

Clone B

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/-

Clone A Clone B

SHANK3+/+

SHANK3+/-

Rin

(G

Ω)

Cm

(pF

)

SHANK3+/+ SHANK3+/-Clone A Clone B

0.0

0.2

0.4

0.6

28/3

27/3

SHANK3+/+

SHANK3+/-

0

10

20

30

**28

/3

27/3

SHANK3+/+

SHANK3+/-

mE

PS

Cfr

eque

ncy

(Hz)

0.0

0.2

0.4

34/3

33/3

SHANK3+/+

SHANK3+/-

mE

PS

Cam

plitu

de (

pA)

0

10

20 ***

34/3

33/3

SHANK3+/+

SHANK3+/-

mE

PS

Cfr

eque

ncy

(Hz)

mE

PS

Cam

plitu

de (

pA)

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/-

Input resistance & capacitance Evoked EPSCs

Spontaneous mEPSCs

Fig. 2. SHANK3 haploinsufficiency increases electrical input resistanceof human neurons and decreases overall synaptic strength. (A) Inputresistance (Rin) but not capacitance (Cm) is increased in SHANK3+/− neurons(summarygraphs frommatching humanSHANK3+/+ andSHANK3+/− neuronsderived from two independentSHANK3+/cKOEScell clones). (B) Evoked synaptictransmission is decreased in SHANK3+/− human neurons [representative EPSCtraces (left) and EPSC amplitude summary graphs (right) for independent sets

of neurons]. (C) Amplitudes but not frequencies of spontaneous mEPSCs(monitored in 1 mM tetrodotoxin) are impaired in SHANK3+/− neurons [top,representative traces; bottom, cumulative plots and summary graphs of themEPSC frequency (left) and amplitude (right)]. Data are means ± SEM.Numbers of cells/cultures analyzed are shown in bars. Statistical significancewas evaluated by either Student’s t test (bar graphs) or Kolmogorov-Smirnovtest (cumulative probability plots) (**P < 0.01; ***P < 0.001).

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

deceleration of Ih-current activation (Fig. 4F).However, other basic properties of Ih currents,such as their half-maximal activation potential(V50), were not significantly altered (fig. S8).Furthermore, depolarizing voltage sag responses

that are evoked by brief injections of hyperpolarizingcurrents in current-clamp mode (36, 37) werealso significantly impaired in SHANK3+/− andSHANK3−/− neurons; these impairments againwere occluded by ZD7288, suggesting that these

phenotypes are also due to a specific alterationin Ih-channel function (fig. S8).Viewed together, these data suggest that SHANK3

mutations cause Ih-channel dysfunction in humanneurons. To investigate the possibility that SHANK3

aaf2669-4 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

2.5 s

40 p

A

Total neurite length Number of primaryprocesses

Total # of branches Soma size

* ***** **

*

0

3

6

9

* *

0

10 µm

10 µm

SHANK3+/+ SHANK3-/-

SHANK3+/+

SHANK3-/-

Clone C

SHANK3+/+

SHANK3-/-

Clone D

Arb

itrar

y un

its(n

orm

aliz

ed to

con

rol)

0

0.4

0.6

0.8

1.0

0.2

SHANK3+/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone D

0

0.6

0.8

1.0

0

0.6

0.8

1.0

0

0.6

0.8

1.0

40/3

40/3

46/3

45/3

0.4

0.2

0.4

0.2

0.4

0.2

MAP2 / SynapsinClone C

SK3+/+

SK3-/-

Clone DMAP2 / Synapsin

SK3+/+

SK3-/-

MA

P2

/ SY

NA

PS

IN

SHANK3+/+ SHANK3-/-

18/3

18/3

18/3

18/3

12

Pun

cta

per

10 µ

m d

endr

ite

0.4

0.6

0.2

Pun

cta

size

(µm

2 )

Puncta analysis

Puncta analysis

20 ms10

0 pA

0

0.25

0.50

0.75

0

20

40

0

0.4

0.6

22/3

22/3

***

49/3

49/3

***

49/3

49/3

**

***

0.0

0.5

1.0 ***

Cum

ulat

ive

prob

abili

ty

0.0

0.5

1.0

0 10 20 30 40 0 40 80 120 160

**

0.0

0.5

1.0

1.5

30/3

30/3

**

0

10

20

30

30/3

30/3

Rin

(G

Ω)

Cm

(pF

)

Interevent interval (s) mEPSC amplitude (pA)

SHANK3+/+

SHANK3-/-

mE

PS

Cfr

eque

ncy

(Hz)

mE

PS

Cam

plitu

de (

pA)

EP

SC

am

plitu

de (

nA)

0.2

SHANK3+/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

SHANK3-/-

SHANK3+/+

Synaptic puncta sizeSynaptic puncta density

SHANK3+/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone DSHANK3+

/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone DSHANK3+

/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone D

SHANK3+/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone DSHANK3+

/+

SHANK3-/-

Clone CSHANK3+

/+

SHANK3-/-

Clone D

HomozygousSHANK3cKO alleles

LoxP FRT LoxP13

Exon 13

13 Input resistance& capacitance

Evoked EPSCs

Spontaneous mEPSCs

30 µm

250

150

50

kDa

SHANK3

GDI

Fig. 3. Homozygous SHANK3 deletion aggravates SHANK3 haploinsufficiency pheno-type. (A) Diagram of homozygous SHANK3cKO/cKO alleles (see supplementary materialsfor details). (B)SHANK3−/−neurons lackmajor SHANK3proteins. Images depict immunoblotsofmatchingSHANK3+/+andSHANK3−/−neuronsderived fromtwo independentSHANK3cKO/cKO

ES cell clones. (C) Representative images of matching SHANK3+/+ and SHANK3−/− neu-rons (day 21) stained bydouble immunofluorescence forMAP2 (red) and synapsin (green).(D) Homozygous SHANK3 deletion severely impairs dendritic arborization [summary graphsof indicated parameter (normalized to SHANK3+/+ controls) measured in matchingSHANK3+/+ and SHANK3−/− neurons derived from two independent SHANK3cKO/cKO EScell clones]. (E) Representative images of dendrites stained for MAP2 (red) and synapsin(green) for analysis of synaptic puncta in SHANK3+/+ and SHANK3−/− neurons. (F) Homo-zygous SHANK3 deletion reduces synapse density (summary graphs of synaptic punctadensity and size on proximal dendrites of isogenic SHANK3+/+ and SHANK3−/− neuronsfrom two independentSHANK3 cKO/cKOES cell clones). (G) HomozygousSHANK3 deletionincreases neuronal Rin (top) and decreases Cm (bottom). (H) Homozygous SHANK3deletion reduces evoked EPSC amplitudes (left, representative traces; right, summarygraphs of EPSC amplitudes). (I) Homozygous SHANK3 deletion decreases the frequencyand amplitude of mEPSCs [top, representative traces; bottom, cumulative plots and summary graphs of mEPSC interevent intervals and frequency (bottom left) ormEPSCamplitudes (bottom right)]. Data in bardiagramsaremeans±SEM.Numbersof cells/cultures analyzedare shown in bars.Statistical significancewas evaluatedby Student’s t test (bar graphs) or Kolmogorov-Smirnov test (cumulative probability plots) (*P < 0.05; **P < 0.01; ***P < 0.001). For additional data, see figs. S3 to S5.

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

interacts with HCN channels, we coexpressedSHANK3 with HCN1, HCN2, or HCN3 channelsin human embryonic kidney (HEK) 293T cells.Coimmunoprecipitations revealed that SHANK3bound to all three isoforms of HCN channels inthis assay (Fig. 4G). Mapping of the interactiondomains using glutathione S-transferase (GST)pull-downs suggested that the SHANK3 ankyrinrepeats directly bind to HCN channels (fig. S9).Finally, measurements of the levels of two humanHCN isoforms to which antibodies were available,

HCN3 and HCN4, demonstrated that the SHANK3deletion significantly decreased the levels of endo-genous HCN proteins in human neurons, consistentwith a direct interaction (Fig. 4H and fig. S10).Because Ih currents are major determinants

of neuronal excitability (29, 30), we examinedthe effects of SHANK3mutations on action potentialgeneration (Fig. 5). Consistent with the increasedinput resistance, SHANK3+/− and SHANK3−/− neu-rons fired significantly more action potentials thanSHANK3+/+ neurons in response to depolarizing

current injections; this phenotype was abolishedby the addition of ZD7288 and thus was Ih-current-dependent (Fig. 5). Moreover, becausesustained rhythmic oscillations are a hallmarkof neuronal circuits in various brain regions thatoverlap with highly enriched regions of SHANK3expression (14, 29, 38), we monitored the sponta-neous spiking activity of SHANK3-mutant neurons.When neurons were held at the resting membranepotential, a large proportion of wild-type cells(~60%) fired action potentials spontaneously

SCIENCE sciencemag.org 6 MAY 2016 • VOL 352 ISSUE 6286 aaf2669-5

Fig. 4. SHANK3deletions impair Ih currentsin human neurons, and SHANK3 proteinsinteract with HCN channels. (A) Neuronalsilencing by blocking glutamate receptorswith CNQX (20 mM) and AP5 (50 mM) doesnot alter neuronal Rin. (B) Blocking Ihcurrents with ZD7288 (100 mM) increasesthe Rin of wild-type neurons, abolishing thedifference between wild-type and mutantneurons. (C) SHANK3+/− and SHANK3−/−

neurons exhibit a more negative Vrest thanSHANK3+/+ neurons; inhibition of Ih cur-rents in SHANK3+/+ neurons with ZD7288(5 mM) abolishes the difference. (D and E)SHANK3+/− neurons (D) and SHANK3−/−

neurons (E) exhibit decreased Ih-currentamplitudes compared with matchingSHANK3+/+ neurons (left, experimentalprotocol and sample traces; right, current/voltage relation of Ih currents, which arevalidated by inhibition with ZD7288).(F) Kinetics of Ih-current activation isimpaired in SHANK3+/− and SHANK3−/−

neurons [left, representative capacity- andleak current–subtracted Ih-current record-ings fitted with a double-exponentialfunction (yellow line superimposed ontraces); center and right, summary graphsof time constants (t) and amplitudes of fastand slow components of Ih-current activa-tion at a test potential of –120 mV].(G) SHANK3 protein binds to HCN channelsmediating Ih currents. Representativeimmunoblots document coimmunoprecipi-tation of Myc-tagged SHANK3 with HA-tagged HCN1, HCN2, and HCN3 proteinscoexpressed in transfected HEK293Tcells;cells expressing only one or the otherprotein serve as negative controls.(H) Levels of endogenous HCN3 and HCN4proteins are decreased in SHANK3−/− neu-rons, as determined by quantitative immu-noblotting (for representative blots, see fig.S10). Data are means ± SEM. Numbers ofcells/cultures analyzed are shown in bars orparentheses. Statistical significance wasevaluated by Student’s t test [bar graphs in(F) and (H)] or two-way analysis of variance(ANOVA) [bar graphs in (A) to (C)] and two-way repeated measure ANOVA [I-V plots in(D) and (E)] followed by Bonferroni’s post hoctest (*P < 0.05; **P < 0.01; ***P < 0.001).

020

/3

21/3

20/3

0

0.4

0.8

1.2

20/3

22/3

24/3

29/3

29/3

SHANK3+/+

SHANK3+/-

CNQX+AP5

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/-

ZD7288

SHANK3+/+

SHANK3+/-

0

0.4

0.8

23/3

23/3

34/4

33/4

SHANK3+/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

Rin

(G

Ω)

0.4

0.8

Rin

(G

Ω)

Rin

(G

Ω)

*** ****** ***

ZD7288

n.s.n.s.

Effect of ZD7288 on input resistanceNeuronal silencing

100150

25010% input IP: HA-HCN

HA-HCN

kDa

HCN2

+ + - + + - + + -

HCN3

+ +- + +- + +-

HCN1

Co-Immunoprecipitations ofSHANK3 with HCN proteins

Control Control Control

0.5 s

100

pA

50 p

A

-40 mV

-120 mV

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3-/-

0.5 s

Ih current amplitudesExperimental protocol

Representative traces

0.10

1.00

0.05

0.50

0.20

2.00

fast slow

22/3

21/3

22/3

21/3

n.s

τ (se

c)

23/3

23/3

n.s23

/3

23/3

fast slow

Am

plitu

de (

pA)

0

20

40

fast slow

Am

plitu

de (

pA)

0

20

40

60

fast slow

n.sn.s

Rel

ativ

e pr

otei

n le

vel

(nor

mal

ized

to G

DI)

SHANK3+/+

SHANK3-/-

0

0.4

0.6

0.8

1.0

1.2

0.2

SHANK3+/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

ZD7288

Vre

st (

mV

)

-40

-20

0

**** *

38/6

25/4

23/4

23/6

Control

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3-/-

Ih current activation kinetics (τ)

-40 mV-90 mV

-120 mV

0.25 s

100

pA

0.10

1.00

0.05

0.50

0.20

2.00

*** ***

Representative traces of Ih currentsfitted to a double exponential

Actin L1CAM

33 33

HCN3

***

HCN4

***

33 33

HCN protein levels inSHANK3 cKO neurons

*** ***

Resting potential

Ih current amplitudes

-40 mV

-120 mV

Experimental protocol

IB: Myc

IB: HA

IB: Myc

IB: HA

IB: Myc

IB: HA

IP: Myc-SHANK3

100

150

250

100150

250

Myc-SHANK3

SHANK3+/+ SHANK3+/- SHANK3+/+ SHANK3-/-

+100 µM ZD7288

Voltage (mV)-120 -80 -40

-1.6

-1.2

-0.8

-0.4

0.0

-0.4

0.0

-120 -80 -40

-2.0

-1.5

-1.0

-0.5

0.0

-0.5

0.0SHANK3+/+ (23/3)SHANK3+/- (23/3)

SHANK3+/+ (23/3)SHANK3-/- (23/3)

SHANK3+/+ (25/3)

SHANK3+/-(26/3)

SHANK3+/+ (24/3)

SHANK3-/-(24/3)

*** ***

+100 µM ZD7288

I h c

urre

nt d

ensi

ty (

pA/p

F)

I h c

urre

nt d

ensi

ty (

pA/p

F)

Voltage (mV)

Representative traces

τ (se

c)

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

and regularly (fig. S11). In contrast, SHANK3-mutant neurons fired fewer action potentials;again, this phenotype was reversed by additionof the Ih-current inhibitor ZD7288 (fig. S11).

Shank3 deletions impair Ih currents indeveloping mouse neurons

To determine whether Shank3-mutant mice alsoexhibit an impairment in Ih currents, we culturedhippocampal neurons from newborn littermateShank3+/+, Shank3+/−, and Shank3−/− mice. Inthe Shank3-mutant mice, exons encoding the PDZdomain of SHANK3 are deleted, similar to ourconditionally SHANK3-mutant human neurons(14). We observed an overall very similar pheno-type in developing mouse neurons as in humanneurons (Fig. 6 and figs. S12 and S13). Specifically,quantitative imaging revealed that at 8 days in vitro(DIV8), homo- but not heterozygous Shank3mutations caused a significant impairment indendritic arborization and synapse density similarto human SHANK3 deletions (Fig. 6, A and B, andfig. S12). Electrophysiological recordings showedthat although the total input resistance of mouseneurons was lower than that of human neurons,both hetero- and homozygous Shank3 mutationsproduced a large increase in input resistance(Fig. 6C). Homozygous Shank3 deletions alsosignificantly decreased cell capacitance and in-creased the resting membrane potential, similarto human mutations. We then directly measuredIh currents and detected a massive impairmentin both hetero- and homozygous Shank3-deficientmouse neurons (Fig. 6, D and E, and fig. S12). TheIh-current amplitude was decreased more thantwofold by Shank3mutations, the Ih-current voltagesag was reduced, and the Ih-current activationkinetics was significantly decelerated. In all ofthese phenotypes, the homozygous mutation wasmore deleterious than the heterozygous mutation.These changes closely resemble those observed inhuman SHANK3-deficient neurons and also ledto a markedly increased excitability in mouseShank3-deficient neurons (Fig. 6F).

Chronic Ih-current inhibition impairsneuronal development similar toSHANK3 mutations

A decrease in Ih currents by SHANK3mutationslikely accounts for the electrophysiological changesof SHANK3-mutant neurons, but can it alsoaccount, at least in part, for their morphologicaland synaptic changes? To investigate this question,we chronically inhibited Ih channels in wild-typeneurons by continuous application of low concen-trations of ZD7288 (1 and 5 mM) during neuronaldifferentiation.Chronic inhibition of Ih channels dramatically

impaired neuronal morphology in a manner in-distinguishable from that of the SHANK3 haplo-insufficiency (Fig. 7A and fig. S14). Specifically,chronic application of ZD7288 caused a dose-dependent decrease in neurite outgrowth, num-ber of primary processes, and dendritic branching.Moreover, ZD7288 decreased the synapse density,again suggesting that major phenotypes of SHANK3mutations may represent indirect effects of an

aaf2669-6 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

+30 pA

-10 pA

-70 mV0 10 20 30 40 50

0

4

8

12

20 m

V

-70 mV

+30 pA

-10 pA

Injected current (pA)0 10 20 30 40 50

Num

ber

of a

ctio

n po

tent

ials

0

4

8

12

SHANK3 +/-SHANK3 +/+

SHANK3+/+(33/4)

SHANK3+/- (34/4)

SHANK3+/+ (20/3)SHANK3+/- (21/3)

-70 mV

20 m

V

200 ms-70 mV

+40 pA

-10 pA

Injected current (pA)

Num

ber

of a

ctio

n po

tent

ials

0 10 20 30 400

4

8

+40 pA

-10 pA

0 10 20 30 40 500

4

8

12

200 ms

SHANK3 +/+ SHANK3 -/-SHANK3-/- (31/4)SHANK3+/+(32/4)

SHANK3+/+ (23/3)SHANK3-/- (23/3)

***

***

n.s.

n.s.

AP

firin

gth

resh

old

(mV

)

029

/3

29/3

22/3

24/3

SHANK3+/+

SHANK3+/-

+ZD7288SHANK3+

/+

SHANK3+/-

-15

-30

AP

ampl

itude

(m

V)

0

20

40

60

29/3

29/3

22/3

24/3

SHANK3+/+

SHANK3+/-

+ZD7288SHANK3+

/+

SHANK3+/-

AH

P (

mV

)

0

29/3

29/3

22/3

24/3

SHANK3+/+

SHANK3+/-

+ZD7288SHANK3+

/+

SHANK3+/-

10A

P fi

ring

thre

shol

d (m

V)

0

33/4

33/4

23/3

23/3

-15

-30

AP

ampl

itude

(m

V)

0

20

40

60

33/4

33/4

23/3

23/3 A

HP

(m

V)

0

33/4

33/4

23/3

23/3

10

SHANK3+/+

SHANK3-/-

+ZD7288SHANK3+

/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

+ZD7288SHANK3+

/+

SHANK3-/-

SHANK3+/+

SHANK3-/-

+ZD7288SHANK3+

/+

SHANK3-/-

Neuronal excitability

Action potential properties

Neuronal excitability

Action potential properties

+100 µM ZD7288

+100 µM ZD7288

+100 µMZD7288

+100 µMZD7288

Fig. 5. SHANK3 mutations render neurons hyperexcitable by impairing Ih currents. (A) SHANK3+/−

mutantneurons reachactionpotential (AP) firing thresholdearlier thanmatchingSHANK3+/+humanneuronsandexhibit a steeper input-output relationship, as assessed by the number of APs elicited by increasing currentinjections (from –10 to +50 pA, 1-s, 5-pA increments) during current-clamp recordings. Ih-channel inhibition withZD7288abolishes thedifference (left, experimental protocol and representative traces; right, plotsof themeanAPnumber versus injected current). (B) Summarygraphs of active electrical properties ofmatchingSHANK3+/+ andSHANK3+/− neurons (measured without or with ZD7288 application). From left to right: AP firing threshold, APamplitude, and AP after-hyperpolarization amplitude (AHP). (C) Same as (A), but for matching SHANK3−/− andSHANK3+/+neurons. (D) Sameas (B), but formatchingSHANK3−/−andSHANK3+/+neurons. Data aremeans±SEM.Numbers of cells/cultures analyzed are shown in bars or parentheses. Statistical significancewas evaluatedby two-way ANOVA (bar graphs) or two-way repeated measure ANOVA followed by Bonferroni’s post hoc test(input-output plots) (n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001).

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

SCIENCE sciencemag.org 6 MAY 2016 • VOL 352 ISSUE 6286 aaf2669-7

Fig. 6. Developing hippocampalneurons from Shank3 knockoutmice reproduce phenotype ofhuman SHANK3-mutant neu-rons. (A and B) Dendritic arbori-zation (A) and synapse formation(B) are impaired in developingShank3-deficient mouse neurons[top, representative images ofhippocampal neurons (A) anddendritic segments (B); bottom,summary graphs of the indicateddendritic, cellular, and synapticparameters]. Neurons culturedfrom littermate Shank3+/+,Shank3+/−, or Shank3−/− micewere stained at DIV8 for MAP2(red) and synapsin (green).(C) Hetero- and homozygousShank3 deletions increase neuro-nal input resistance (top),decrease cell capacitance(center), and enhance the restingmembrane potential (bottom) inhippocampal neurons culturedfrom littermate Shank3+/+,Shank3+/−, and Shank3−/−

mice and analyzed at DIV8-9.(D) Hetero- and homozygousShank3 deletions impair neuronalIh currents in hippocampal neu-rons at DIV8-9 (left, experimentalprotocol and sample traces; right,summary graph of the current/voltage relation). (E) Hetero- andhomozygous Shank3 deletionsdecelerate Ih-current activation[left top, experimental protocol;left bottom, capacitance- andleak current-subtracted repre-sentative traces fitted with thesum of two exponential functionsshown superimposed on thetraces in yellow; right, summarygraphs of activation time con-stants (t) and component ampli-tudes obtained at a test potentialof –120 mV]. (F) Hetero- andhomozygous Shank3 deletionsrender hippocampal neuronshyperexcitable (left, experimentalprotocol and representativetraces of stepwise depolarizingcurrent injections; right, sum-mary plots of the AP numberversus injected current duringcurrent-clamp recordings of neu-rons analyzed at DIV8-9). Dataare means ± SEM. Numbers ofcells/cultures analyzed are shownin bars or parentheses. Statisticalsignificance was evaluated by one-way ANOVA followed by Tukey’s post hoc test (bar graphs) or two-way repeatedmeasure ANOVA followed by Bonferroni’s posthoc test (I-V plot) (*P < 0.05; **P < 0.01; ***P < 0.001). For additional data, see figs. S12 and S13.

0

100

200

300

28/3

30/3

28/3

0

40

80

120

28/3

30/3

28/3

-20

-40

-60

28/3

30/3

28/3

-120 -100 -80 -60 -40

-2.0

-1.6

-1.2

-0.8

-0.4

0.0

Voltage (mV)

I h c

urre

nt d

ensi

ty (

pA/p

F)

Shank3+/+ (24/3)Shank3+/- (30/3)Shank3-/- (28/3)

-40 mV-90 mV

-120 mV

200 ms

100

pA

Am

plitu

de (

pA)

0

50

100

150fast slow

**

**

-40 mV

-120 mV

Exptl. protocol & representative traces

-50 mV

200

pA

250 ms

fast slow

21/3

21/3

21/3

21/3

21/3

21/3

*

n.s n.s

Hip

poca

mpa

l Neu

ron

DIV

8 M

ap2

/ Syn

apsi

n30 µm

Map

2 / S

ynap

sin

10 µm

0

0.4

0.6

0.8

0.2

Syn

apsi

n pu

ncta

per

10 µ

m d

endr

ite

Pun

cta

size

(µm

2 )

Synapsin puncta size

Synapsin puncta density

0

3

6

9

12

Totaldendrite length

75/3

18/3

18/3

18/3

75/3

75/3

0

0.4

0.6

0.8

1.0

0.2

Totalbranches

*

0

0.4

0.6

0.8

1.0

0.2

Soma area

0

0.4

0.6

0.8

1.0

0.2

*

# of primaryprocesses

0

0.4

0.6

0.8

1.0 *

0.2

**

Shank3+/+

Shank3+/-

Shank3-/-

Rin

(M

Ω)

Vre

st (

mV

)

****

*

*

Ih current activation kinetics (τ)

Ih current amplitudes

Shank3-/-

Ih current traces fitted to a double exponential

Shank3+/+ Shank3+/-

τ (se

c)

0.10

1.00

0.05

0.50

0.20

2.00

******

*****

**

*****

Input resistance,capacitance &

Vrest

Cm

(pF

)A

rbitr

ary

units

(nor

mal

ized

)

Shank3+/+

Shank3+/-

Shank3-/-

*** ******

Shank3-/-Shank3+/+ Shank3+/-

Shank3-/-

Shank3+/+

Shank3+/-

Shank3+/+

Shank3+/-

Shank3-/-

Shank3+/+

Shank3+/-

Shank3-/-

Shank3+/+

Shank3+/-

Shank3-/-

Shank3+/+

Shank3+/-

Shank3-/-

Shank3+/+

Shank3+/-

Shank3-/-

Shank3-/-Shank3+/+

Shank3+/-

Neuronal development Synapse density

Neuronal excitability

+280 pA

-40 pA

20 m

V

0.1 s

-70 mV

Shank3+/+ Shank3+/- Shank3-/-

Injected current (pA)0 60 120 180 240N

umbe

r of

act

ion

pote

ntia

ls

0

4

8

12Shank3+/+ (24/3)Shank3+/- (28/3)Shank3-/- (25/3)

****

*n.

s.

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

Ih-current impairment (Fig. 7B). In addition, chronicinhibition of Ih channels in human neurons in-creased the neuronal input resistance and impairedspontaneous synaptic transmission, measured afterwashout of the ZD7288 used to inhibit Ih cur-rents (Fig. 7, C and D). These results suggest thatlong-term inhibition of Ih currents has dramaticeffects on multiple facets of neuronal development.

Summary

Many mutations are thought to predispose toidiopathic ASDs by causing primary impairmentsin synaptic transmission (1–5). Our data showthat SHANK3 haploinsufficiency impairs synapticfunction but also demonstrate that SHANK3haploinsufficiency decreases Ih-channel functionas a primary impairment, which in turn producesmajor changes in intrinsic neuronal properties

and secondarily affects synaptic function. Indeed,the fact that several salient phenotypes of hetero-and homozygous SHANK3-mutant human neu-rons were reproduced by pharmacologic inhibitionof Ih channels suggests that Ih-channel dysfunctionis a major effect of SHANK3 haploinsufficiency.Moreover, the very similar phenotypes producedby Shank3mutations in developing mouse neurons,which also severely impaired Ih currents, indicatesa general role for SHANK3 in scaffolding HCNchannels during neuronal development at a periodcoinciding with the manifestation of ASDs. A rolefor SHANK3 as a scaffolding protein for HCNchannels is plausible given the broad expressionof SHANK3 in nonneuronal cells that also expressHCN channels, and SHANK3 may function toenrich HCN channels at postsynaptic sites togetherwith other proteins (31, 32). Thus, we would like to

suggest that an Ih-current impairment is a majorpathogenetic force of SHANK3 mutations in pre-disposing to ASDs and in Phelan-McDermid syn-drome. Changes in Ih-channel function canconceivably be influenced pharmacologically, sug-gesting that pharmacological manipulation of Ihchannels may be therapeutically beneficial.HCN-channel mutations have been linked clin-

ically with human neurological disorders, includingepilepsy, sleep disorder, and impaired learning,which are commonly associated with enhancedneuron firing and/or aberrant neuronal firing pat-terns and are also observed in mice with HCN-channel mutations (29, 30, 38–41). The symptomsresulting from impaired HCN channels agree wellwith the hypothesized involvement of SHANK3deletions in ASDs and Phelan-McDermid syn-drome that are also commonly associated with

aaf2669-8 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

Effect of chronic Ih current inhibition on dendritic arborization

10 µm0

3

6

9

12

0.4

0.6

Pun

cta

per

10 µ

m d

endr

ite

MA

P2

/ Syn

apsi

n

1 µM ZD7288

SH

AN

K3+

/+

5 µM ZD7288

***

Effect of chronic Ih current inhibition on synapse density

018/3

18/3

18/3

0.2

Pun

cta

size

(µm

2 )

MA

P2

/ Syn

apsi

n

30 µm

SHANK3+/+ Control SHANK3+/++ 1 µM ZD7288

SHANK3+/++ 5 µM ZD7288

1 5µM ZD7288

0 1 5µM ZD7288

0

42/3

42/3

42/3

42/3

*** **

Arb

itrar

y un

its(n

orm

aliz

ed to

con

rol)

Total neuritelength

# of primaryprocesses Total branches Soma area

0

0.4

0.6

0.8

1.0

0.2

1 500

****** ***

1 5µM

ZD7288: 00 1 500 1 500

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/+

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/+

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/+

SHANK3+/+

SHANK3+/-

SHANK3+/+

SHANK3+/+

*

mE

PS

Cam

plitu

de (

pA)

0

10

20

30

23/3

25/3

mE

PS

Cfr

eque

ncy

(Hz)

0.0

0.4

0.8

23/3

25/3

Interevent interval (s) 0 10 20 30 40

Cum

ulat

ive

Pro

babi

lity

0.0

0.5

1.0

mEPSC amplitude (pA)

0 20 40 60 80 100 120 1400.0

0.5

1.0

***

***

0

0.4

0.8

0

10

20

30

40

25/3

24/3

25/3

24/3

***

0 µM ZD7288 1 µM ZD7288

1 s

20 p

A

SHANK3+/+ SHANK3+/+

Rin

(G

Ω)

Cm

(pF

)

Resistance & capacitance

SHANK3+/+ SHANK3+/+

10µM ZD7288

10µM ZD7288

Spontaneous mEPSCs

10µM ZD7288

10µM ZD7288

***

***

SHANK3+/+SHANK3+/+

0 µM ZD7288

Fig. 7. Chronic inhibition of Ih currents in human neurons mimics SHANK3haploinsufficiency phenotype. (A) Chronic partial inhibition of Ih currents inwild-type SHANK3+/+ neurons causes dendritic arborization defects similar toSHANK3+/− haploinsufficiency.MatchingSHANK3+/+ andSHANK3+/− neuronswere differentiated from SHANK3+/cKO ES cells; SHANK3+/+ neurons weretreated with low-dose ZD7288 (1 mM or 5 mM) from days 3 to 21 of neuralinduction (top, representative images of neurons double-labeled for MAP2 andsynapsin; bottom, summary graphs of indicated parameters normalized to theuntreatedSHANK3+/+ control). (B)Chronic inhibition of Ih currents inSHANK3

+/+

neurons decreases synapsenumbers. Experimentswere performedas describedfor (A) (left, representative images of dendritic segments double labeled forMAP2 and synapsin; right, summary graphs of density and size of synapticpuncta). (C) Chronic inhibition of Ih currents in SHANK3+/+ neurons significantly

increases Rin (left) and decreases cell Cm (right). Experiments were performedas described for (A); recordings were performed after washout of ZD7288.(D) Chronic inhibition of Ih currents in SHANK3+/+ neurons with ZD7288 (1 mM)reduces the frequency and amplitude of spontaneousmEPSCs (top, representativetraces; bottom left, summary plots of the interevent interval and summary graph ofthemEPSC frequency; bottom right, same for themEPSC amplitude). Experimentswere performed as described for (A); recordings were performed after washout ofZD7288. Data are means ± SEM. Numbers of cells/cultures analyzed are shown inthe bars. Statistical significance was evaluated by Student’s t test [bar graphsin (C) and (D)], one-wayANOVA followed by Tukey’s post hoc test [bar graphs in(B)], or two-way ANOVA followed by Bonferroni’s post hoc test [bar graphs in(A)] or Kolmogorov-Smirnov test [cumulative probability plots in (D)] (*P <0.05; **P < 0.01; ***P < 0.001).

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

intellectual disability, impaired learning andmemory, and epilepsy (1–5). Therefore, our datacollectively suggest that impairment of HCN-channel function may contribute to the mani-festations of ASDs in patients with SHANK3mutations and Phelan-McDermid syndrome.

Methods summary

See the supplementary materials for full detailsof the materials and methods (22).

Generation and analysis of human EScells with heterozygous and homozygousSHANK3 cKO alleles

SHANK3-mutant ES cells were generated fromH1 ES cells (passage 40; WiCell, www.wicell.org)using recombinant adenoassociated virus (AAV)–mediated homologous recombination (Fig. 1A andfig. S1A) (24, 25). AAVs used for gene targetingcontained a puromycin-resistance cassette sur-rounded by FRT sites (for Flp-recombinase medi-ated deletion). The cassette was inserted into the3′ intron of exon 13 (76 base pairs) of the humanSHANK3 gene, exon 13 was flanked by LoxP sites(for Cre-recombinase mediated deletion), and 5′and 3′ DNA homology arms were added. Thesame gene-targeting template was also used inthe generation of homozygous SHANK3 cKO EScells, except that a CRISPR/Cas9 targeting methodwas used to specifically target the second wild-type allele of the SHANK3 gene and not the already-targeted first SHANK3 cKO allele. Heterozygousand homozygous SHANK3 cKO ES cell clone muta-tions were confirmed by PCR, and the puromycin-resistance cassette was removed by Flp-recombinase.To generate isogenic control and SHANK3-mutanthuman neurons, hetero- or homozygous SHANK3cKO ES cells were transdifferentiated to neuronsusing forced expression of Ngn2, as described (23).Lentiviruses encoding active (Cre) or inactive Cre-recombinase (DCre) were applied at 1 day beforeinduction of neuron differentiation. Neurons wereanalyzed at day 21 to 23 in most experiments. Forthe experiments of chronic Ih-current inhibitionwith low-dose ZD7288 (Fig. 7), wild-type humanneurons were treated with 1 mM or 5 mM ZD7288from day 3 to day 21.

Generation and analysis of Shank3-mutant hippocampal mouse neurons

Breedings of heterozygous Shank3-mutant micewith deletion of the PDZ domain (exons 13 to 16)(14) (B6.129-Shank3tm2Gfng/J; Jackson Labs stockNo. 017688) were used to produce littermate wild-type, heterozygous, and homozygous Shank3-mutant mice. Hippocampal neurons were culturedfrom newborn mice (42, 43) and analyzed as im-mature developing neurons at DIV8-9. Mouse geno-typing was performed by PCR using the JacksonLaboratory protocol.

Morphological analyses

Dendritic arborizations were analyzed in neuronsthat were sparsely transfected with an enhancedgreen fluorescent protein (EGFP) to obtain fluores-cent images of individual neurons. Confocal im-ages were analyzed unbiasedly using the “neurite

outgrowth” application on MetaMorph software.For synapse morphology analyses, fixed neuronswere stained by double immunofluorescence withantibodies to mitogen-activated protein kinasekinase (MAP2) (to stain for dendrites) and syn-apsin (to label presynaptic terminals) or HOMER1(to label postsynaptic specializations), and imageswere again quantified using MetaMorph software(Molecular Devices) (44).

Electrophysiological recordings

Whole-cell patch-clamp recordings were performedessentially as described (23, 42). EPSCs werepharmacologically isolated with 50 mM picro-toxin (PTX) and recorded at a –70 mV holdingpotential in voltage-clamp mode in response toextracellular stimulation with a concentric bipolarelectrode (43). Spontaneous mEPSCs were mon-itored in the presence of tetrodotoxin (1 mM).Recordings of the intrinsic and active membraneproperties were generally recorded in humanneurons in the presence of 50 mM PTX (unlessotherwise stated), and in hippocampal mouseneurons in the presence of CNQX (6-cyano-7-nitroquinoxaline-2,3-dione) (20 mM), AP5 (2-amino-5-phosphonopentanoic acid) (50 mM), and PTX(50 mM).Input resistance (Rin) was calculated as the

slope of linear fits of current-voltage plots gen-erated from a series of increasing current injectionsteps in current-clamp mode. Ih-channel activitywas measured in voltage-clamp mode as the am-plitude of the slowly activating inward current com-ponent elicited by 2-s voltage steps from –50to –120 mV in 10-mV increments from a holdingpotential of –40 mV with 2 mM 4-aminopyridine(4-AP) and 0.5 mM BaCl2 in the bath solution.Depolarizing voltage-sag responses were evokedby brief injections of hyperpolarizing currentsin current-clamp mode. Voltage-dependent Na+

(INa) and K+ (IKD) currents were recorded involtage-clamp mode at a holding potential of–70 mV in the presence of 2-mM 4-AP; voltagesteps ranging from –90 to +40 mV were deliveredat 10-mV increments. Intrinsic action potentialfiring properties of neurons were recorded incurrent-clamp mode. To assess neuronal excitabil-ity, first minimal currents were introduced to holdmembrane potential around −65 to −70 mV, thenincreasing amounts of depolarizing currents wereinjected for 1 s in stepwise manner. For sponta-neous AP firing, cells were held at their restingmembrane potential (Vrest); no current was injected.The Vrest obtained during spontaneous firing wasdetermined as the mean steady-state voltage re-corded during interspike intervals. All experimentswere performed at room temperature.

Protein-protein interaction analyses

Protein-protein interaction analyses were per-formed by coimmunoprecipitation of Myc-taggedfull-length mouse Shank3 protein with hemag-glutinin (HA)–tagged full-length human HCNproteins expressed in HEK293T cells. Coimmuno-precipitations of truncated Shank3 proteins withfull-length HCN proteins were also performed tomap the responsible interaction domain on Shank3.

To further demonstrate the interaction betweenShank3 and HCN proteins, a GST pull-down assaywas performed using purified Shank3 ankyrinrepeats domain and full-length HCN1 protein.

Immunoblotting

All immunoblots were visualized by fluorescentlylabeled secondary antibodies and quantified onOdyssey CLx Infrared Imager and Odyssey software(LI-COR Biosciences). Signals were normalizedto human guanine nucleotide dissociation inhib-itor (GDI) as the neuronal loading control.

REFERENCES AND NOTES

1. C. S. Leblond et al., Meta-analysis of SHANK mutations inautism spectrum disorders: A gradient of severity in cognitiveimpairments. PLOS Genet. 10, e1004580 (2014). doi: 10.1371/journal.pgen.1004580; pmid: 25188300

2. C. Betancur, J. D. Buxbaum, SHANK3 haploinsufficiency: A“common” but underdiagnosed highly penetrant monogeniccause of autism spectrum disorders. Molecular Autism 4, 17(2013). doi: 10.1186/2040-2392-4-17; pmid: 23758743

3. R. O. Rosti, A. A. Sadek, K. K. Vaux, J. G. Gleeson, The geneticlandscape of autism spectrum disorders. Dev. Med. ChildNeurol. 56, 12–18 (2014). doi: 10.1111/dmcn.12278;pmid: 24116704

4. A. M. Grabrucker, M. J. Schmeisser, M. Schoen, T. M. Boeckers,Postsynaptic ProSAP/Shank scaffolds in the cross-hair ofsynaptopathies. Trends Cell Biol. 21, 594–603 (2011).doi: 10.1016/j.tcb.2011.07.003; pmid: 21840719

5. S. Carbonetto, A blueprint for research on Shankopathies: Aview from research on autism spectrum disorder. Dev.Neurobiol. 74, 85–112 (2014). doi: 10.1002/dneu.22150;pmid: 24218108

6. J. Gauthier et al., De novo mutations in the gene encodingthe synaptic scaffolding protein SHANK3 in patientsascertained for schizophrenia. Proc. Natl. Acad. Sci. U.S.A. 107,7863–7868 (2010). doi: 10.1073/pnas.0906232107;pmid: 20385823

7. A. Guilmatre, G. Huguet, R. Delorme, T. Bourgeron, Theemerging role of SHANK genes in neuropsychiatric disorders.Dev. Neurobiol. 74, 113–122 (2014). doi: 10.1002/dneu.22128;pmid: 24124131

8. K. Han et al., SHANK3 overexpression causes manic-likebehaviour with unique pharmacogenetic properties. Nature503, 72–77 (2013). doi: 10.1038/nature12630; pmid: 24153177

9. A. Shcheglovitov et al., SHANK3 and IGF1 restore synapticdeficits in neurons from 22q13 deletion syndrome patients.Nature 503, 267–271 (2013). doi: 10.1038/nature12618;pmid: 24132240

10. S. Naisbitt et al., Shank, a novel family of postsynaptic densityproteins that binds to the NMDA receptor/PSD-95/GKAPcomplex and cortactin. Neuron 23, 569–582 (1999).doi: 10.1016/S0896-6273(00)80809-0; pmid: 10433268

11. J. C. Tu et al., Coupling of mGluR/Homer and PSD-95complexes by the Shank family of postsynaptic densityproteins. Neuron 23, 583–592 (1999). doi: 10.1016/S0896-6273(00)80810-7; pmid: 10433269

12. M. Sheng, E. Kim, The Shank family of scaffold proteins.J. Cell Sci. 113, 1851–1856 (2000). pmid: 10806096

13. O. Bozdagi et al., Haploinsufficiency of the autism-associatedShank3 gene leads to deficits in synaptic function, socialinteraction, and social communication. Molecular Autism 1, 15(2010). doi: 10.1186/2040-2392-1-15; pmid: 21167025

14. J. Peça et al., Shank3 mutant mice display autistic-likebehaviours and striatal dysfunction. Nature 472, 437–442(2011). doi: 10.1038/nature09965; pmid: 21423165

15. H. E. Speed et al., Autism-associated insertion mutation (InsG)of Shank3 exon 21 causes impaired synaptic transmission andbehavioral deficits. J. Neurosci. 35, 9648–9665 (2015).doi: 10.1523/JNEUROSCI.3125-14.2015; pmid: 26134648

16. M. Kouser et al., Loss of predominant Shank3 isoforms resultsin hippocampus-dependent impairments in behavior and synaptictransmission. J. Neurosci. 33, 18448–18468 (2013). doi: 10.1523/JNEUROSCI.3017-13.2013; pmid: 24259569

17. M. Yang et al., Reduced excitatory neurotransmission and mildautism-relevant phenotypes in adolescent Shank3 null mutantmice. J. Neurosci. 32, 6525–6541 (2012). doi: 10.1523/JNEUROSCI.6107-11.2012; pmid: 22573675

SCIENCE sciencemag.org 6 MAY 2016 • VOL 352 ISSUE 6286 aaf2669-9

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

18. E. Drapeau, N. P. Dorr, G. A. Elder, J. D. Buxbaum, Absence ofstrong strain effects in behavioral analyses of Shank3-deficientmice. Dis. Model. Mech. 7, 667–681 (2014). doi: 10.1242/dmm.013821; pmid: 24652766

19. X. Wang et al., Synaptic dysfunction and abnormal behaviors inmice lacking major isoforms of Shank3. Hum. Mol. Genet. 20,3093–3108 (2011). doi: 10.1093/hmg/ddr212; pmid: 21558424

20. J. Lee et al., Shank3-mutant mice lacking exon 9 show alteredexcitation/inhibition balance, enhanced rearing, and spatialmemory deficit. Front. Cell. Neurosci. 9, 94 (2015).doi: 10.3389/fncel.2015.00094; pmid: 25852484

21. N. Uppal et al., Ultrastructural analyses in the hippocampusCA1 field in Shank3-deficient mice. Molecular Autism 6, 41(2015). doi: 10.1186/s13229-015-0036-x; pmid: 26137200

22. For experimental procedures, see the supplementary materialsand methods on Science Online.

23. Y. Zhang et al., Rapid single-step induction of functionalneurons from human pluripotent stem cells. Neuron 78,785–798 (2013). doi: 10.1016/j.neuron.2013.05.029;pmid: 23764284

24. C. Pak et al., Human neuropsychiatric disease modeling usingconditional deletion reveals synaptic transmission defectscaused by heterozygous mutations in NRXN1. Cell Stem Cell17, 316–328 (2015). doi: 10.1016/j.stem.2015.07.017;pmid: 26279266

25. C. Patzke et al., Analysis of conditional heterozygous STXBP1mutations in human neurons. J. Clin. Invest. 125, 3560–3571(2015). doi: 10.1172/JCI78612; pmid: 26280581

26. C. Proepper, S. Putz, R. Russell, T. M. Boeckers, S. Liebau, TheKvb2 subunit of voltage-gated potassium channels isinteracting with ProSAP2/Shank3 in the PSD. Neuroscience261, 133–143 (2014). doi: 10.1016/j.neuroscience.2013.10.045;pmid: 24211303

27. A. Ludwig, X. Zong, M. Jeglitsch, F. Hofmann, M. Biel, A familyof hyperpolarization-activated mammalian cation channels.Nature 393, 587–591 (1998). doi: 10.1038/31255; pmid: 9634236

28. S. Chen, J. Wang, S. A. Siegelbaum, Properties ofhyperpolarization-activated pacemaker current defined bycoassembly of HCN1 and HCN2 subunits and basal modulationby cyclic nucleotide. J. Gen. Physiol. 117, 491–504 (2001).doi: 10.1085/jgp.117.5.491; pmid: 11331358

29. E. E. Benarroch, HCN channels: Function and clinicalimplications. Neurology 80, 304–310 (2013). doi: 10.1212/WNL.0b013e31827dec42; pmid: 23319474

30. O. Postea, M. Biel, Exploring HCN channels as novel drug targets.Nat. Rev. Drug Discov. 10, 903–914 (2011). doi: 10.1038/nrd3576;pmid: 22094868

31. M. F. Nolan et al., The hyperpolarization-activated HCN1channel is important for motor learning and neuronalintegration by cerebellar Purkinje cells. Cell 115, 551–564(2003). doi: 10.1016/S0092-8674(03)00884-5;pmid: 14651847

32. M. F. Nolan et al., A behavioral role for dendritic integration:HCN1 channels constrain spatial memory and plasticity atinputs to distal dendrites of CA1 pyramidal neurons. Cell 119,719–732 (2004). pmid: 15550252

33. J. C. DiFrancesco, D. DiFrancesco, Dysfunctional HCN ionchannels in neurological diseases. Front. Cell. Neurosci. 6, 174(2015). doi: 10.3389/fncel.2015.00071; pmid: 25805968

34. R. E. BoSmith, I. Briggs, N. C. Sturgess, Inhibitory actions ofZENECA ZD7288 on whole-cell hyperpolarization activatedinward current (If) in guinea-pig dissociated sinoatrial nodecells. Br. J. Pharmacol. 110, 343–349 (1993). doi: 10.1111/j.1476-5381.1993.tb13815.x; pmid: 7693281

35. K. S. Shin, B. S. Rothberg, G. Yellen, Blocker state dependenceand trapping in hyperpolarization-activated cation channels:Evidence for an intracellular activation gate. J. Gen. Physiol.117, 91–102 (2001). doi: 10.1085/jgp.117.2.91;pmid: 11158163

36. D. A. McCormick, H. C. Pape, Properties of a hyperpolarization-activated cation current and its role in rhythmic oscillation inthalamic relay neurones. J. Physiol. 431, 291–318 (1990).doi: 10.1113/jphysiol.1990.sp018331; pmid: 1712843

37. J. S. Solomon, J. M. Nerbonne, Hyperpolarization-activatedcurrents in isolated superior colliculus-projecting neurons fromrat visual cortex. J. Physiol. 462, 393–420 (1993). doi: 10.1113/jphysiol.1993.sp019561; pmid: 8331588

38. A. S. Lewis, D. M. Chetkovich, HCN channels in behavior andneurological disease: Too hyper or not active enough? Mol. Cell.Neurosci. 46, 357–367 (2011). doi: 10.1016/j.mcn.2010.11.007;pmid: 21130878

39. A. M. Phillips, T. Kim, E. Vargas, S. Petrou, C. A. Reid, Spike-and-wave discharge mediated reduction in hippocampal HCN1channel function associates with learning deficits in a geneticmouse model of epilepsy. Neurobiol. Dis. 64, 30–35 (2014).doi: 10.1016/j.nbd.2013.12.007; pmid: 24368169

40. S. P. Vaidya, D. Johnston, Temporal synchrony and gamma-to-theta power conversion in the dendrites of CA1 pyramidal

neurons. Nat. Neurosci. 16, 1812–1820 (2013). doi: 10.1038/nn.3562; pmid: 24185428

41. M. Wang et al., Neuronal basis of age-related working memorydecline. Nature 476, 210–213 (2011). doi: 10.1038/nature10243; pmid: 21796118

42. A. Maximov, T. C. Südhof, Autonomous function ofsynaptotagmin 1 in triggering synchronous releaseindependent of asynchronous release. Neuron 48, 547–554(2005). doi: 10.1016/j.neuron.2005.09.006; pmid: 16301172

43. A. Maximov, Z. P. Pang, D. G. Tervo, T. C. Südhof, Monitoringsynaptic transmission in primary neuronal cultures using localextracellular stimulation. J. Neurosci. Methods 161, 75–87(2007). doi: 10.1016/j.jneumeth.2006.10.009;pmid: 17118459

44. G. R. Anderson et al., b-Neurexins control neural circuits byregulating synaptic endocannabinoid signaling. Cell 162, 593–606(2015). doi: 10.1016/j.cell.2015.06.056; pmid: 26213384

ACKNOWLEDGMENTS

We thank Y. Zhang, S. Maxeiner, and S. J. Lee for advice,G. Feng (MIT) for reagents, and V. Sebastiano (Stanford) forsharing instruments. This work was supported by grants fromNIH (MH092931 to M.W.; NS077906 to T.C.S.; and U19MH104172to M.W. and T.C.S.) and a postdoctoral fellowship fromVetenskapsradet, Sweden (to S.C.B.). The data are includedin the main manuscript and in the supplementary materials. Authorcontributions: F.Y. performed the ES cell, molecular biology,expression, and cell-biology experiments, T.D. the electrophysiologicalexperiments, S.C.B. the protein chemistry experiments, C.P. theimmunoblotting experiments, and C.H.P. the gene-expressionexperiments. All authors planned the experiments, analyzed data, andedited the paper written by T.C.S.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/352/6286/aaf2669/suppl/DC1Materials and MethodsFigs. S1 to S14References (45–60)

15 January 2016; accepted 26 February 2016Published online 10 March 201610.1126/science.aaf2669

aaf2669-10 6 MAY 2016 • VOL 352 ISSUE 6286 sciencemag.org SCIENCE

RESEARCH | RESEARCH ARTICLEon F

ebruary 16, 2019

http://science.sciencemag.org/

Dow

nloaded from

channelopathy in human neuronshIAutism-associated SHANK3 haploinsufficiency causes Fei Yi, Tamas Danko, Salome Calado Botelho, Christopher Patzke, ChangHui Pak, Marius Wernig and Thomas C. Südhof

originally published online March 10, 2016DOI: 10.1126/science.aaf2669 (6286), aaf2669.352Science

, this issue p. 10.1126/science.aaf2669Sciencecognitive functions and the predisposition to epilepsy.through channelopathy could account for the phenotypes observed in Phelan-McDermid neurons, such as alterations inthe major phenotype consisting of a specific impairment of HCN channels. Chronic impairment of membrane currents Phelan-McDermid syndrome. Instead of affecting synapses, SHANK3 mutations primarily caused a channelopathy, withproduced human neurons lacking SHANK3 but not other genes that are also involved in the autism-like disease

et al.SHANK3 mutations cause autism-like behavioral changes and exhibit alterations in synaptic transmission. Yi SHANK3 is a widely expressed scaffolding protein that is enriched in postsynaptic specializations. In mutant mice,

Faulty channels, not faulty synapses

ARTICLE TOOLS http://science.sciencemag.org/content/352/6286/aaf2669

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2016/03/09/science.aaf2669.DC1

CONTENTRELATED

http://stke.sciencemag.org/content/sigtrans/11/513/eaan6500.fullhttp://stm.sciencemag.org/content/scitransmed/7/271/271ra8.fullhttp://stke.sciencemag.org/content/sigtrans/7/349/re10.fullhttp://stke.sciencemag.org/content/sigtrans/5/244/pe44.fullhttp://stm.sciencemag.org/content/scitransmed/3/103/103ra97.fullhttp://stm.sciencemag.org/content/scitransmed/4/128/128ra40.fullhttp://stm.sciencemag.org/content/scitransmed/4/131/131ra51.fullhttp://science.sciencemag.org/content/sci/351/6278/1153.fullhttp://science.sciencemag.org/content/sci/351/6278/1199.full

REFERENCES

http://science.sciencemag.org/content/352/6286/aaf2669#BIBLThis article cites 59 articles, 14 of which you can access for free

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

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.Sciencelicensee American Association for the Advancement of Science. No claim to original U.S. Government Works. The title Science, 1200 New York Avenue NW, Washington, DC 20005. 2017 © The Authors, some rights reserved; exclusive

(print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

on February 16, 2019

http://science.sciencem

ag.org/D

ownloaded from