list of probes can be found in Table S3. ..... Mushroom-shaped spine heads with an electron dense post-synaptic density (PSD) apposed to .... The pellets were washed twice with WB (4 mM .... Inputs and IPs were run on 4-15% gradient gels.
Cell Reports, Volume 24
Supplemental Information
Noonan Syndrome-Associated SHP2 Dephosphorylates GluN2B to Regulate NMDA Receptor Function Aaron D. Levy, Xiao Xiao, Juliana E. Shaw, Suma Priya Sudarsana Devi, Sara Marie Katrancha, Anton M. Bennett, Charles A. Greer, James R. Howe, Kazuya Machida, and Anthony J. Koleske
Figure S1. Recombinant SHP2 dephosphorylates recombinant GluN2B tail. Related to Figure 5. (A) Coomassie Blue-stained gel shows purified recombinant GST and GST-SHP2 PTP proteins used in Figure 5B. (B) Coomassie Blue-stained gel shows purified recombinant full-length SHP2 proteins used in Figures S1D-E and 5C-D. (C) (left) Coomassie Blue-stained gel shows purified recombinant GluN2B-tail +/- phosphorylation by Src used in Figures S1DE and 5C-D. (right) Western blot of GluN2B-tail for phosphotyrosine (pY) confirms tyrosine phosphorylation of GluN2B-tail by Src. (D) Representative western blots show time-dependent dephosphorylation of GluN2B-tail by active Noonan-associated SHP2 (DG), while auto-inhibited WT and phosphatase-dead (CS) SHP2 have no effect. (E) Quantification of GluN2B-tail tyrosine phosphorylation relative to no PTP is fitted with a one-phase decay curve, R2=0.98. Dephosphorylation plateaus at 42% of baseline.
A
GluN2B WT
C
20pA 200ms
GluN2B WT
5pA 100ms
B
— GluN2B WT 4pA 200ms
Figure S2. Raw traces of recombinant wild type NMDAR recordings. Related to Figure 6. (A) Examples of responses to 2 ms application of 1 mM glutamate (arrows) for GluN1:GluN2B wild type receptors in outsideout patches. (B) Mean currents evoked by 2 ms applications of 1 mM glutamate. (C) Examples of unitary currents recorded during continuous application of 1 mM glutamate for GluN1:GluN2B receptors from outside-out patches. All NMDA receptors currents were recorded at -100 mV in the continuous presence of 100 µM glycine.
Figure S3. Schematic of SH2-binding screen. Related to Figure 7. (A) Screen schematic. Phosphorylated and nonphosphorylated GluN2B Y1252 peptides were immobilized on nitrocellulose membranes (1=Y1252, 2=pY1252). The membranes were incubated with GST-tagged SH2 probes and binding detected with infrared dye-labeled secondary antibodies. (B) Summary of pY-binding domain probes used in the screen. A complete list of probes can be found in Table S3.
Figure S4. Quantified result for pY binding domain screen. Related to Figure 7. pY binding domain screening was performed as described in Methods. Background subtracted signal (∆ pY1252–Y1252) was measured. Mean values and SEM from eight experiments are shown.
Figure S5. Further analysis of GST-Nck2 SH2 domain infusion. Related to Figure 7. (A) Representative traces of NMDAR-EPSCs 0 minutes (black) and 30 minutes (red) after infusion of the indicated GST-Nck2 SH2 domain. The overlaps (right) have been scaled to match peak amplitude for visualization of current decay. (B) AMPAR-ESPC amplitudes are reduced slightly by infusion of GST-Nck2 SH2 domain (~10-20%), but WT GST-Nck2 SH2 does not reduce AMPAR-EPSCs by more than the R311K binding mutant (RK) or heat inactivated (HI) protein. Data were analyzed by one-way ANOVA (p > 0.05). Data are means + SEM, n = 8 WT, 6 RK, and 6 HI neurons. By contrast, WT GST-Nck2 SH2 infusion reduced NMDAR-EPSCs significantly more than RK or HI protein (see Fig. 7D). (C) Series resistance is unchanged across the course of protein infusion, indicating that the protein does not affect patch robustness (two-way repeated measures ANOVA, all effects p > 0.05). (D) WT GST-Nck2 SH2 masks the effect of ifenprodil on NMDAR-EPSC amplitude. Addition of 3 µM ifenprodil after 30 minutes of HI Nck2 infusion reduced NMDAR-EPSCs (black symbols) to a level similar to 5 µg/mL WT GST-Nck2 SH2 alone. By contrast, ifenprodil had no effect on neurons where 2.5 or 5 µg/mL WT
GST-Nck2 SH2 had been infused (blue symbols). The time course of EPSC amplitude loss was similar to neurons not treated with ifenprodil (red symbols, reproduced from Figure 7D). Arrow indicates time of ifenprodil addition. Data are presented as in B, n = 4 (WT 2.5) and (WT 5), 5 (WT 2.5+Ifen) and (WT 5+Ifen), and 3 (HI 5+Ifen) neurons. (E) WT GST-Nck2 SH2 partially masks the effect of ifenprodil on NMDAR-EPSC decay. While 3 µM ifenprodil applied 30 minutes after WT GST-Nck2 SH2 was still able to modestly reduce NMDAREPSC decay time 20 minutes later, ifenprodil had no effect when saturating 5 µg/mL WT GST-Nck2 SH2 was previously infused. By contrast, ifenprodil significantly reduced decay time when HI GST-Nck2 SH2 was infused (paired t-tests, WT 2.5 p = 0.02, WT 5 p = 0.69, HI 5 p = 0.03). Data points connect paired recordings beginning 30 minutes into GST-SH2 infusion at the start of ifenprodil application (Start) and 20 minutes later (+Ifen), and bars show group means, n = 5 (WT 2.5) and (WT 5), and 3 (HI 5) neurons.
Figure S6. Control IgG immunoprecipitation. Related to Figure 7. Representative western blot shows that mouse anti-GluN2B immunoprecipitates WT GluN2B and co-immunoprecipitates GluN1 and Myc-Nck2 from crosslinked HEK293 cells, while control mouse IgG does not. mw indicates location of molecular weight marker (black markings on blots).
GluN2B mutant
tfast (ms)
tslow (ms)
Fraction Islow
tweighted (ms)
WT
67.43 ± 21.23
323.30 ± 73.13
0.29 ± 0.06
139.30 ± 32.78
Y1252F
59.71 ± 6.93
300.40 ± 36.36
0.39 ± 0.08
152.60 ± 22.08
Y1252E
62.03 ± 11.4
370.40 ± 79.04
0.37 ± 0.08
148.40 ± 18.39
Table S1. Kinetic parameters of GluN1:GluN2B Y1252 NMDAR deactivation. Related to Figure 6. Deactivation kinetic parameters of GluN1:GluN2B diheteromers expressed in tsA201 cells. Recordings were made from outside-out patches stimulated 60 times at 6 s intervals with 2 ms pulses of 1 mM glutamate in the presence of 100 µM glycine. There are no significant difference deactivation kinetics between WT and either mutant receptor. Data were analyzed by one-way ANOVA (tfast p = 0.91, tslow p = 0.70, fraction Islow p = 0.74, tweighted p = 0.93). n = 4 WT, 9 YF, and 8 YE patches, respectively, and data are shown as means ± SEM.
GluN2B mutant
𝜏o1 (ms)
𝜏o2 (ms)
fraction 𝜏o2
𝜏o-weighted (ms)
Conductance (pS)
WT
0.29 ± 0.03
3.85 ± 0.02
0.76 ± 0.004
3.01 ± 0.04
45.0 ± 6.0
Y1252F
0.32 ± 0.03
4.16 ± 0.12
0.70 ± 0.04
3.03 ± 0.14
45.6 ± 3.4
Y1252E
0.29 ± 0.06
3.70 ± 0.55
0.66 ± 0.05
2.57 ± 0.42
47.9 ± 1.5
Table S2. Single channel properties for GluN1:GluN2B Y1252 mutant NMDARs. Related to Figure 6. Single-channel open time parameters and conductance of GluN1:GluN2B diheteromers expressed in tsA201 cells. Recordings were made from outside-out patches in the continuous presence of 1 mM glutamate and 100 µM glycine. There are no significant differences in the mean open time or conductance between WT and either mutant receptor. Data were analyzed by one-way ANOVA (weighted open time: p = 0.45; conductance p = 0.84). n = 2 WT, 4 YF, and 3 YE patches, respectively, and data are shown as means ± SEM.
#
Probe
Gene
GENE ID
1 2 3
Gst Abl Abl R161L
ABL1 ABL1
25 25
4
Abl(R>K)
ABL1
25
5 6 7 8 9 10 11 12
Arg Bks Blk Blnk Bmx Brdg1 Brk Btk
ABL2 BKS BLK BLNK BMX BRDG1 BRK BTK
27 55620 640 29760 660 26228 5753 695
13
Cbl 2x
CBL
867
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41
CblA CblB CblC Chimerin2 Cis1 Crk CrkL Csk Cten DAPP1 Dok1(PTB) Eat2 Emt/Itk Fer Fes Fgr Frk Frs2(PTB) FynA Gap(C) Gap(N) Gap(SH23) Grap Grap2 Grb10 Grb14 Grb2 Grb2 4x
CBL CBLB CBLC CHN2 CISH CRK CRKL CSK TNS4 DAPP1 DOK1 SH2D1B ITK FER FES FGR FRK FRS2 FYN RASA1 RASA1 RASA1 GRAP GADS GRB10 GRB14 GRB2 GRB2
867 868 23624 1124 1154 1398 1399 1445 84951 27071 1796 117157 3702 2241 2242 2268 2444 10818 2534 5921 5921 5921 10750 9402 2887 2888 2885 2885
42
Grb2 4x mut
GRB2
2885
43 44 45 46 47 48 49
Grb7 Hck HSh2 IRS1(PTB) Lck Lnk Lyn
GRB7 HCK HSH2D IRS1 LCK LNK LYN
2886 3055 84941 3667 3932 10019 4067
domain GST control SH2 GST-SH2 GST-SH2 control GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-PTB GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-PTB GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 control GST-SH2 GST-SH2 GST-SH2 GST-PTB GST-SH2 GST-SH2 GST-SH2
∆ pY1252– Y1252 Mean 825 30464 6550
Count
SEM
8 7 3
3658 7767 2504
-1925
2
1575
141650 -7350 -38745 -2306 9280 -15340 55065 -13840
8 9 10 8 5 5 13 5
51616 10584 38404 4926 27443 19064 15482 17682
41575
2
36925
79625 72481 30435 9040 -1467 -1081 -406 -3710 11900 7786 10450 400 28488 10450 -3088 9770 1588 3400 98343 8533 495 24857 7063 19050 -13157 -6300 -1100 1850
10 13 13 5 3 8 8 5 12 7 1 5 4 5 4 5 8 1 7 3 11 14 8 4 15 9 10 1
86665 28143 14982 18935 2967 16314 27351 1578 3100 4891
-1467
3
1213
14579 84700 10931 -16500 61700 -33619 7414
7 7 8 1 6 8 7
17953 60923 5228
1627 17625 8291 2782 30224 8569 42717 5172 11144 7265 3992 17524 14244 13551 4434
44103 37383 8234
50
Lyn 2x
LYN
4067
51
Lyn 3x
LYN
4067
52 53 54 55
Matk/Lsk Mist Nap4 Nck
MATK MIST SOCS7 NCK1
4145 116449 30837 4690
56
Nck 3x
NCK1
4690
57 58 59 60
Nck2 Nsp1 Nsp2 NSP3
NCK2 SH2D3A BCAR3 SH2D3C
8440 10045 8412 10044
61
p55g(NC)
PIK3R3
8503
62 63
p85a(C) rat or RC3 p85a(N)
PIK3R1 PIK3R1
5295 5295
64
p85a(NC)
PIK3R1
5295
65
p85b(NC)
PIK3R2
5296
66 67 68
p85bC p85bN PKCdelta(C2)
PIK3R2 PIK3R2 PRKCD
5296 5296 5580
69
Plcg1(NC)
PLCG1
5335
70
Plcg2(NC)
PLCG2
5336
71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90
Ptk70/Srms PTP-PEST DA/CS PTP1B D/A Q/A Sap Sck/ShcB SH2-B SH2A/SH2D4A SH2D4B SH2D5 Sh3bp2 Shb ShcA ShcA(PTB) ShcC ShcD SHD SHE SHF SHIP1 SHIP2
SRMS PTPN12 PTPN1 SH2D1A SHC2 SH2B SH2D4A SH2D4B SH2D5 SH3BP2 SHB SHC1 SHC1 SHC3 SHC4 SHD SHE SHF SHIP1 SHIP2
6725 5782 5770 4068 25759 25970 63898 387694 400745 6452 6461 6464 6464 53358 399694 56961 126669 90525 3635 3636
91
SHIP2 3x
SHIP2
3636
92
SHP-1 PTP DA/CS SHP-1 PTP FL DA/CS
PTPN6
5777
GST-SH2SH2 GST-SH2SH2-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2SH2 GST-SH2 GST-PTP GST-PTP GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-PTB GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2-SH2 GST-PTP
PTPN6
5777
GST-PTP
93
23350
5
18545
10750
1
4981 19742 33561 147400
8 6 9 10
2643 11565 19865 81202
970593
7
74609
284950 -8250 2860 -1793
4 1 5 7
84462
6625
10
3693
8500 25357
2 7
4950 18890
-15843
7
16078
8169
13
3024
-17070 16855 17500
5 10 1
31424 9334
5533
6
3987
-5786
7
16258
83140 7850 188905 10371 -4486 -2283 -1613 16508 17088 -4436 -11590 -12564 3775 -27750 12350 -24500 9375 6558 4550 28800
5 1 10 7 11 3 8 6 4 7 5 7 4 4 5 5 4 6 7 5
33980
6675
4
3513
1050
1
-14470
5
1673 12419
102487 6660 8680 2911 8713 10902 13796 4882 8177 8986 8230 17346 8054 15369 8818 2447 17793 20064
26481
94
SHP-1(NC)
PTPN6
5777
95
SHP-2 PTP DA/CS
PTPN11
5781
96
SHP-2(NC)
PTPN11
5781
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
Slap Slnk Slp76 Socs1 Socs2 Socs3 Socs4 Socs7 Src Stat1 Stat2 Stat4 SupT6h Syk(C) Syk(N)
SLAP SLNK SLP76 SOCS1 SOCS2 SOCS3 SOCS4 SOCS7 SRC STAT1 STAT2 STAT4 SUPT6H SYK SYK
6503 284948 3937 8651 8835 9021 122809 30837 6714 6772 6773 6775 6830 6850 6850
112
Syk(NC)
SYK
6850
113
Syk(NC) 2x 3x
SYK
6850
114 115 116 117 118 119 120
Tc-PTP DA Tec Tem6 Tenc1 Tensin Txk Vav1
PTPN2 TEC TNS3 TENC1 TNS1 TXK VAV1
5771 7006 64759 23371 7145 7294 7409
121
Vav1 2x 3x
VAV1
7409
122 123 124 125
Vav2 Vav3 Vrap Yes
VAV2 VAV3 SH2D2A YES
7410 10451 9047 7525
126
Zap70(NC)
ZAP70
7535
127
Zap70(NC)2x
ZAP70
7535
128
Zap70(NC)3x
ZAP70
7535
GST-SH2SH2 GST-PTP GST-SH2SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2SH2 GST-PTP GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2 GST-SH2SH2 GST-SH2SH2 GST-SH2SH2-SH2
-8836
7
17191
12500
1
-10900
7
10716
-5833 -1425 38293 2494 11833 2670 5300 2200 -58270 -28917 -12363 4775 -4181 11967 3033
6 4 7 8 6 5 13 1 5 6 4 2 8 3 3
7295 12628 32015 3716 4065 3198 6533
8344
8
5095
11393
7
8797
10300 23914 -6650 -21580 -1500 -151643 -4040
1 7 5 5 5 7 5
7344 14086 24480 1803 144581 3628
10488
4
4260
1120 -5790 35870 110000
5 10 5 5
1766 10073 26902 76099
28667
3
27449
3433
3
2767
-9700
1
59736 19874 11079 1525 18986 4116 2562
Table S3. Probe characteristics and quantified result for pY binding domain screen. Related to Figure 7. Probe name, gene symbol, NCBI Gene ID, mean ∆pY1252 –Y1252 value, spot replicates, and SEM are provided.
Supplemental Experimental Procedures Animals. All animal procedures were reviewed, approved, and monitored by the Yale University Institutional Animal Care and Use Committee. Ptpn11D61G/+ knock-in mice, hereafter referred to as NS mice (https://www.jax.org/strain/012593, (Araki et al., 2004)), were maintained in a mixed C57Bl/6J x 129/SvJ background. Some animals were crossed with Thy-1 GFP line M (https://www.jax.org/strain/007788, (Feng et al., 2000)). We used male mice for all experiments, as dendritic spine density fluctuates with estrus (Woolley et al., 1990), and made direct comparisons between paired littermates whenever possible. All animals were postnatal day (P) 40P46 at the time of experiment unless noted otherwise. NS mice were genotyped using primers 5’CAAGGTGAGTGGGCGTTTCATTTTAAC-3’, 5’-ACCTTTCAGAGGTAGGGTCTGCAC-3’, and 5’CCATAGAGGTCATAGTAGCCACCG-3’, which yield a 500 base pair product for the wild type allele and a 350 base pair product for the D61G allele. GFP genotyping is described by JAX (Bar Harbor, ME, https://www.jax.org/strain/007788). Novel object recognition. The test was performed essentially as previously described (Kerrisk et al., 2013). P42-56 WT and NS mice were habituated to the experimenter five minutes/day for five days, and to the testing environment (a large, empty cage in a quiet, dimly lit room) for one hour before testing. Mice were then allowed to explore the environment and two identical objects placed at either end of the long axis of the cage until they accumulated 30 seconds of object exploration time, defined as direct whisking or sniffing of the object. Mice were returned to their home cages and the test cages cleaned to remove scent cues. 48 hours later, mice were allowed to explore objects as on day one, except one object was replaced with a novel object. Mice were excluded if on either day, they failed to accumulate 30 seconds of object exploration time in six minutes of total exploration time or did not explore both objects. Exclusion rates were similar between genotypes. The object chosen as novel and its placement in the cage were randomized. Exploration was videotaped and scored manually by an experimenter blind to genotype and object novelty. Hippocampal slice preparation and electrophysiological recordings. Slices were prepared and recordings made as previously described (Xiao et al., 2016). Mice were deeply anesthetized with Nembutal and their brains rapidly removed and cooled in ice-cold sucrose-containing artificial cerebrospinal fluid (sucrose-ACSF, in mM): 2.5 KCl, 7 MgSO4•7H2O, 1.25 NaH2PO4, 0.5 CaCl2•2H2O, 28 NaHCO3, 7 dextrose, 205 sucrose saturated with 95% O25% O2 at 300-310 mOsmol/L, pH7.4. Transverse hippocampal slices were cut at 350 µm in ice-cold sucrose-ACSF on a vibratome (Leica, Wetzlar, Germany) and allowed to recover in a humidified, oxygenated environment at 33.5˚C for at least 30 minutes. Slices were perfused during recording at 1 mL/min with ACSF (in mM): 3 KCl, 1 MgSO4•7H2O, 1.25 NaH2PO4, 2 CaCl2•2H2O, 26 NaHCO3, 10 dextrose, 118 NaCl, 0.4 ascorbic acid, 4 Na-lactic acid, 2 Napyruvic acid saturated with 95% O2-5% O2 at 300-310 mOsmol/L, pH7.4. 10 µM bicuculline methiodide (BMI), 20 µM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 3 µM ifenprodil, and 1 µM tetrodotoxin (TTX) (MilliporeSigma, Darmstadt, Germany) were included in the ACSF where described. CA1 pyramidal neurons were identified using DIC and infrared optics with a water-immersion objective (NIR Apo 40x; BX51, Nikon, Tokyo, Japan) at room temperature (RT), and whole-cell recordings were obtained with an Axopatch 700B patch-clamp amplifier (Molecular Devices, Sunnyvale, CA) using recording electrodes pulled from 1.5 mm glass capillaries on a Flaming-Brown micropipette puller (Sutter Instruments, Novato, CA) filled with (in mM): 137 cesium gluconate, 10 NaCl, 0.2 EGTA, 4 Na2ATP, 0.3 Na3GTP, 5 Na2-phosphocreatine, 10 HEPES, 5 QX-314 chloride adjusted to 290-300 mOsmol/L pH7.4. Under these conditions, the recording electrodes had a resistance of 4-7 MΩ. Recombinant GST-Nck2 SH2 domain protein was included in the patch solution at 2.5 or 5 µg/mL where described. Currents were low pass filtered at 2 kHz and digitized at 5 kHz with a Digidata 1440 digitizer (Molecular Devices). Cells were excluded if series resistance deviated >20% or exceeded 20 MΩ. Data were collected in pClamp 10.3 and analyzed with Clampfit 10.3. For all data, n=number of neurons. Excitatory postsynaptic currents (EPSCs) were evoked with 0.05 Hz stimulation by a tungsten bipolar stimulating electrode placed ~150 µm away from the recording electrode in the Schaffer collateral using stimulus intensities that yielded 50-60% of the maximal response. AMPAR-EPSC amplitudes were measured at the peak of the EPSC recorded at -70 mV holding potential. NMDAR-EPSC amplitudes were recorded from the same neurons at +40 mV 40 ms after the stimulus artifact, after AMPAR-EPSCs have decayed (Choi et al., 2000). NMDAR-EPSC decay kinetics were measured using a nonlinear regression in pClamp to fit recordings with a double exponential decay 𝐼(𝑡) = 𝐼fast 𝑒 ()/+fast + 𝐼slow 𝑒 ()/+slow , where 𝐼(𝑡) is the amplitude of the EPSC at time t, Ifast and Islow are the peak amplitudes of the fast and slow components, respectively, and tfast and tslow are their respective time constants. Weighted time constants were calculated from the time constants and relative amplitudes of the two exponential
components. To determine GluN2B contribution, tweighted was calculated from NMDAR-EPSCs recorded under baseline conditions and from the same neurons after addition of 3 µM ifenprodil. Long-term potentiation of evoked EPSCs (recorded at -70 mV) was induced by theta-burst stimulation (five trains of four pulses at 100 Hz in 200 ms intervals, repeated four times with an interval of 10 s) in the presence of 10 µM BMI. Miniature EPSCs (mEPSCs) were recorded in whole-cell patch mode. AMPAR-mEPSCs were recorded at 70 mV with TTX and BMI in the external solution, and NMDAR-mEPSCs were recorded at -40 mV with TTX, BMI, and CNQX with low Mg2+ (0.1 mM) and high Ca2+ (3.8 mM) to partially remove the Mg2+ block. Amplitude, frequency, and decay time of miniature events were detected using Mini-Analysis software (Synaptosoft, Decatur, GA), and inspected visually to exclude overlapping events and events on a noisy baseline. NMDAR-mEPSC tweighted was calculated as described above. CA1 neuron anatomy. For all anatomical analyses, data were acquired and analyzed by an experimenter blind to genotype and region, and measured essentially as previously described (Kerrisk et al., 2013; Sfakianos et al., 2007; Warren et al., 2012). To measure dendrite morphology, acute 400 µm hippocampal slices were cut as described above and recovered for at least 1 hour in an interface chamber at 31˚C. Single CA1 pyramidal neurons were injected with 4% biocytin in 2 M potassium acetate pH7.5 with 200 ms current injections of 4 nA at 2 Hz for 20 minutes. Neurons were excluded if they did not maintain membrane potential or action potential firing during biocytin injection. Slices were fixed with 4% paraformaldehyde (PFA) in PBS, cryoprotected in 30% sucrose, resectioned to 40 µm on a sliding microtome, and stained with avidin-HRP and DAB (Vectastain Elite ABC, Vector Laboratories, Burlingame, CA). Filled neurons in serial sections were traced under 100x magnification on a light microscope outfitted with a zdrive and reconstructed using Neurolucida (MBF Bioscience, Williston, VT). Neurons that did not have a dendritefree section at both ends of the z-stack were excluded. Dendrite length and branch-point numbers were calculated using NeuroExplorer. n=number of neurons, with ≤ 5 neurons sampled from each of 5-8 animals. Outliers were identified as values greater than 1.5x the interquartile distance below the first or above the third quartile and excluded. To measure dendritic spine density, WT and NS littermates carrying the Thy1-GFP:M transgene were deeply anesthetized with Nembutal and transcardially perfused with 4% PFA/PBS. Brains were dissected and postfixed overnight. Vibratome-cut 40 µm tissue sections were imaged with a 100x oil immersion objective on an UltraVIEW VoX (Perkin Elmer, Waltham, MA) confocal microscope (Nikon Ti-E Eclipse) with Volocity software (Improvision, Coventry, UK). Spines were counted manually in the z-stack in Fiji (Schindelin et al., 2012), and 3D dendrite segment length was measured using the Simple Neurite Tracer Fiji plugin (Longair et al., 2011). All spines analyzed were on terminal oblique dendrites of the main apical dendrite of CA1 pyramidal neurons and located either proximal to the soma in stratum radiatum or distal to the soma just basal to stratum lacunosum. All dendrite segments were at least 30 µm long and of similar diameter. n=average spine density per mouse, with 5-8 dendrite segments analyzed per animal per region. To measure spine ultrastructure, WT and NS littermates were perfused with 4% PFA/2% glutaraldehyde and their brains postfixed and sectioned to 50 µm as above. Brain slices were processed in 4% OsO4 for 1 hour and dehydrated in a series of increasing ethanol dilutions and propylene oxide. Contrast was improved by staining with 1% uranyl acetate in 70% ethanol for 1 hour before resectioning, and by staining with lead citrate after resectioning. Tissue was resectioned with an ultramicrotome at 70 nm after embedding in Epon, and sections were collected on Formvar coated slot grids and viewed using a JEOL 1200 EX Electron Microscope at 12,000 X magnification. Images were analyzed in Fiji. Mushroom-shaped spine heads with an electron dense post-synaptic density (PSD) apposed to a presynaptic terminal containing synaptic vesicles were counted as synapses. PSD length was measured from these synapses. Spine head area was measured to the thinnest part of the spine neck only from spines with a clear neck. Synapse density n=average synapse density per mouse, averaged from 16-18 images per animal. Spine head areas and PSD lengths were quantified with n=number of spine heads/PSDs. DNA constructs. WT and D425A human SHP2 protein tyrosine phosphatase domain (PTP) (amino acids 218-528) in pGEX-2TK have been previously described (Kolli et al., 2004). The Q506A mutation was introduced into the D425A construct by site-directed mutagenesis. Mouse SHP2 cDNA was purchased from Open Biosystems (cat# MMM1013-9200724, GE Dharmacon, Lafayette, CO). To make full-length GST-mSHP2 pFastbac, mSHP2 cDNA was PCR amplified and ligated into GST-pFastBac1 (Bac-to-Bac, Thermo Fisher Scientific, Waltham, MA), a modified pFastBac1 with GST inserted between the BamHI and EcoRI sites, and the D61G and C459S mutations were introduced with site-directed mutagenesis. To make MBP-GluN2B C3-His pMAL-TEV, the sequence coding amino acids 1240-1482 of mouse GluN2B was PCR amplified from GluN2B cDNA (using a 3’ primer coding a 6xHis tag) and ligated into pMAL-TEV (NEB, Ipswich, MA). YFP-GluN2B pVivo2.5 and GluN1-1a GW were gifts of AnnMarie Craig (University of British Columbia). To make YFP-GluN2B pcDNA3.1+, the YFP-GluN2B coding
sequence was PCR amplified and ligated into pcDNA3.1+ (Thermo). To make untagged GluN2B pcDNA3.1+ used for single channel recordings, the GluN2B coding sequence was subcloned into pcDNA3.1+ from mouse GluN2B cDNA, and Y1252F and Y1252E mutations were introduced with site-directed mutagenesis. Mouse GluN1 pCMV (UniprotKB P35438) used for experiments on recombinant receptors was a gift of Gabriela Popescu (University at Buffalo, SUNY). To make GST-∆85-C-Src, the sequence coding amino acids 86-C terminus of Src kinase was PCR amplified from mouse Src pcDNA3.1+ and ligated into GST-pFastBac1. The ∆85C mutant lacks the first 85 amino acids and is constitutively active (Jeansonne et al., 2006). GST-Nck2 SH2 domain has been previously described (Machida et al., 2007), and the R311K mutation was introduced with site-directed mutagenesis. Myc-mouse Nck2 (accession NM_010879) pCMV-Entry was obtained from Origene, and eGFP-N1 was purchased from Clontech. All cloning was confirmed by sequencing. Western blotting. Rabbit anti pY1252, pY1336, and pY1472 (cat# p1516-1252, -1336, and -1472; each used at 1:1000) are from PhosphoSolutions (Aurora, CO), mouse anti GluN2B clone N59/20 (75-097; 1:2000) is from the UC Davis/NIH NeuroMab Facility (Davis, CA), mouse anti SHP2 clone 79 (610621; 1:5000) and mouse anti GluN2A clone 5 (612286; 1:1000) are from BD Biosciences (San Jose, CA), rabbit anti GluN2B c-terminus (06-600; 1:1500), rabbit anti GluA1 (ABN241; 1:1000), and mouse anti actin clone C4 (MAB1501; 1:100,000) are from MilliporeSigma, mouse anti HSP70 clone 3A3 (sc-32239; 1:10,000) is from Santa Cruz (Dallas, TX), rabbit anti GluN1 clone D65B7 (5704; 1:1000) and rabbit anti Myc-tag (2272; 1:1000) are from Cell Signaling Technologies (Danvers, MA), and mouse anti phosphotyrosine clone 4G10 was purified from a hybridoma to 1 mg/ml and diluted (1:2000). Secondary antibodies goat anti-rabbit IgG (H+L)-HRP conjugate (1721019) and goat anti-mouse IgG (H+L)-HRP conjugate (1721011) (1:5000 each) are from Biorad (Hercules, CA). For all western blots, samples were separated by 8% or 10% SDS-PAGE and transferred to 0.2 µm nitrocellulose (Biorad) at 35V for 720 minutes at 4˚C, and uniform transfer and equal loading were confirmed by Ponceau S stain. Blots were blocked at RT in 5% non-fat dry milk in Trisbuffered saline with 0.05% Tween-20 (TBST) for all antibodies except those against phosphotyrosines, which were blocked in 5% bovine serum albumin in TBST, then probed overnight at 4˚C with indicated primary antibodies at the dilutions above in 5% BSA/TBST. Blots were washed with TBST, probed at RT with the appropriate secondary antibody in 5% milk/TBST, washed again, and proteins detected with enhanced chemiluminescence. Bands on scanned films were quantified in Fiji. Surface biotinylation. All steps were carried out at 4˚C or on ice. WT and NS littermates were deeply anesthetized with Nembutal, decapitated, and their hippocampi rapidly dissected. Whole hippocampus was incubated 10 minutes in PBSCM (1x PBS, 1 mM MgCl2, 0.1 mM CaCl2 pH8.0) containing 0.5 mg/ml NHS-LC-biotin (Thermo), or in PBSCM without biotin as a control. Tissue was washed 5 minutes in PBSCM + 20 mM glycine to neutralize excess biotin and once with PBSCM alone, and finally homogenized in RIPA buffer (25 mM Tris pH7.5, 250 mM NaCl, 5% glycerol, 1 mM EDTA, 1% TX100, 0.1% SDS, 0.5% sodium deoxycholate, plus protease (1 mM PMSF, 20 µg/mL aprotinin, 1 mM benzamidine, 10 µg/mL each pepstatin A, leupeptin, and chymostatin) and phosphatase (1 mM sodium orthovanadate, 2 mM NaF) inhibitors), rotated for 10 minutes, and cleared at 14K RPM in a tabletop centrifuge. Protein concentration was determined by BCA assay (Thermo). 100 µg of protein was incubated overnight with 50 µl Neutravidin resin (Thermo), and 10 µg of protein reserved for input. The resin was washed 4 x 5 minutes with RIPA buffer then resuspended in 1x SDS-PAGE sample buffer (25 mM Tris pH6.8, 2% SDS, 5% glycerol, 2% beta-mercaptoethanol), and inputs were solubilized by addition of 4x sample buffer to 1x. Samples were eluted/solubilized by heating to 65˚C for 10 minutes and then western blotted as described above. All comparisons were made between paired littermates biotinylated in the same solutions. Data are presented as a surface/total ratio, and normalized to the WT average. Recombinant protein. For all proteins, concentration was determined by Bradford assay (Biorad) and purity by Coomassie Blue stained SDS-PAGE gels, and protein was aliquoted, snap frozen in liquid nitrogen, and stored at -80˚C until use. GST-SHP2 PTP domains were produced in Rosetta E. coli (Novagen, Madison, WI) by inducing protein expression with 0.3 mM IPTG overnight at 18˚C. Bacteria were collected by centrifugation, resuspended in lysis buffer (50 mM Tris pH8, 150 mM NaCl, 5% glycerol, 1 mM EDTA, 5 mM DTT, 1% TX100, protease inhibitors) and lysed with a probe sonicator. Insoluble material was removed by centrifugation at 150,000 x g for 45 minutes. The soluble lysate was incubated with glutathione agarose resin (Thermo) for 1 h, then washed with: buffer A (lysis buffer with 500 mM NaCl and no EDTA), buffer B (buffer A with 100 mM Tris pH8.0, 10 mM ATP, 5 mM MgCl2, and 0.1% TX100), buffer C (buffer B with 150 mM NaCl and no TX100), and buffer D (50 mM Tris pH8.0, 150 mM NaCl, 5% glycerol, 1 mM DTT, 1 mM PMSF). The resin was packed into a column and bound protein eluted with 10 mM glutathione in buffer D, and peak fractions were dialyzed into fresh buffer D to remove glutathione.
GST-mSHP2 baculovirus was generated per the manufacturer’s instructions (Bac-to-Bac, Thermo) and used to infect Hi5 cells, which were harvested 48 h after infection and processed as described above. Glutathione was removed from eluted fractions by buffer exchange using G-25 resin into assay buffer (25 mM HEPES pH7.25, 100 mM NaCl, 5% glycerol, 0.01% TX100, 1 mM DTT), and GST was cleaved overnight at 4˚C using GST-tagged PreScission protease (GE Healthcare, Little Chalfont, UK). Free GST and GST-PreScission protease were removed with fresh glutathione-agarose. MBP-GluN2B C3-His pMAL-TEV was purified similarly to GST-SHP2 PTP. Bacterial pellets were sonicated in lysis buffer (50 mM Tris pH7.4, 120 mM NaCl, 1 mM DTT, 20 mM imidazole, 5% glycerol, protease inhibitors). Soluble lysates were incubated 1 h with Ni-NTA agarose resin (Qiagen, Hilden, Germany), then washed with buffer A (lysis buffer with 500 mM NaCl and 0.1% TX100), buffer B (buffer A with 100 mM Tris pH7.4, 10 mM ATP, and 5 mM MgCl2), buffer C (buffer B with 120 mM NaCl), and buffer D (buffer A with 120 mM NaCl and 0.01% TX100). Resin was packed into a column and recombinant protein eluted with 250 mM imidazole in buffer D. A portion of the protein was then tyrosine phosphorylated by incubating 10 µM MBP-GluN2B C3-His with 100 nM Src kinase ∆85-C and 15 mM MgCl2, 5 mM MnCl2, 1 mM DTT, 1 mM Na3VO4, and 100 µM ATP for 2.5 h at room temperature, which was sufficient to saturate phosphorylation in assays using g-32P-ATP (data not shown). Proteins were then bound to amylose agarose (NEB) and washed with assay buffer + 500 mM NaCl, followed by a wash with assay buffer. Finally, the recombinant protein was eluted with 10 mM maltose, then buffer exchanged into fresh assay buffer. GST-Nck2 SH2 domain wild type and R311K (SH2 binding-dead mutant) were purified as previously described (Smith and Johnson, 1988) and dialyzed into PBS before use. Substrate trapping. 5x106 Phoenix-AMPHO cells (ATCC CRL-3213, Manassas, VA) were plated on 10 cm tissue culture dishes the day before transfection and maintained in DMEM plus 10% fetal bovine serum (FBS), 2 mM l-glutamine, and penicillin/streptomycin. Cells were transfected with 1 µg Src pcDNA3.1+, 2 µg GluN1-1a GW, and 6 µg YFP-GluN2B pcDNA3.1+ using calcium phosphate, then maintained in media supplemented with 150 µM APV (Cayman Chemicals, Ann Arbor, MI) and 11.25 µM MK801 (Tocris, Minneapolis, MN) to prevent excitotoxic cell death. Tyrosine phosphorylation was induced 48 h after transfection by treating cells for 15 minutes with 100 µM pervanadate before lysis in IAA buffer: 1% TX100, 50 mM HEPES pH7.4, 150 mM NaCl, 10% glycerol, 1.5 mM MgCl2, 1 mM EDTA, protease inhibitors, and 5 mM iodoacetic acid (IAA) to alkylate and inhibit endogenous PTPs. Protein concentration was determined by BCA assay and lysates aliquoted, snap frozen, and stored at -80˚C. For the assay, GST-SHP2 PTP domain was linked to glutathione agarose at 1 µg PTP / 1 µl resin at 4˚C for 30 minutes. For vanadate competition, PTP resin was incubated with 10 mM Na3VO4 for 10 minutes before addition of lysate. Excess IAA was neutralized with 10 mM DTT for 10 minutes, and equal amounts of lysate were incubated 3 h at 4˚C with 10 µl of GST-SHP2 PTP resin. The resin was washed 3x with lysis buffer without IAA, boiled 10 minutes in 1x sample buffer, and the eluted proteins western blotted as described above. Each assay was performed in triplicate. Dephosphorylation. MBP-GluN2B C3-His (140 nM) and full-length mSHP2 (14 nM) were mixed 1:1 in assay buffer + 5 mM DTT and incubated at 32˚C for the times indicated. Reactions were stopped by addition of 4x sample buffer to 1x. Samples were boiled 5 minutes and western blotted as described above. After exposure, antibodies were stripped for 45 minutes at 70˚C in stripping buffer (0.1% SDS, 1% beta-mercaptoethanol, 100 mM Tris pH6.8) and blots washed extensively with dH2O. Removal of antibody signal was confirmed by re-exposing blots before blocking and probing with the next primary antibody, in the order (1) Rb anti GluN2B pY-site (1:5000) or Ms anti phosphotyrosine, (2) Ms anti-SHP2, and (3) Rb anti GluN2B c-terminus. The assay was performed in triplicate, and in parallel for each phosphosite. Synaptic fractionation. Hippocampi were dissected as above, then snap frozen and stored at -80˚C. After thawing, tissue was homogenized using a pestle in HB (320 mM sucrose, 10 mM HEPES pH7.4, 1 mM EDTA, protease inhibitors, 1x HALT phosphatase inhibitor cocktail (Thermo)), spun 3 x 10 minutes at 900 x g to remove debris, then spun 20 minutes at 12,000 x g to pellet synaptosomes. The pellets were washed twice with WB (4 mM HEPES pH7.4, 1 mM EDTA, protease and phosphatase inhibitors), then resuspended in HB and solubilized by addition of 4x sample buffer to 1x and boiling 10 minutes. Protein concentrations were determined by BCA assay, and equal protein amounts were western blotted as described above. All comparisons were made between paired littermates sacrificed at the same time and processed with the same solutions. Data are presented as the ratio of pY site signal to total GluN2B, normalized to the WT mean, except for GluN2B, which is presented as the raw data normalized to the WT mean.
NMDAR kinetics. tsA201 cells were maintained in DMEM with 10% FBS. For patch-clamp experiments, cells were plated on poly-L-lysine coated glass coverslips and co-transfected with plasmids encoding GluN1 pCMV, GluN2B pcDNA3.1+ WT or phosphomutants and GFP (1:1:0.2) using X-tremeGENE 9 DNA transfection reagent (Roche, Basel, Switzerland) according to manufacturer’s instructions. Fresh medium with 0.5 mM APV and 1 mM MgCl2 was added 3-4 hours after transfection, and currents recorded 24-48 hours later. All recordings were made at room temperature with an EPC-9 amplifier (HEKA, Holliston, MA) as described (Robert and Howe, 2003), at a holding potential of -100 mV, using Patchmaster software (HEKA). Patch pipettes (open tip resistance 3-6 MW) were filled with a solution containing (in mM): 135 CsF, 33 CsOH, 2 MgCl2, 1 CaCl2, 10 HEPES and 11 EGTA adjusted to pH7.4 with CsOH. The external solution contained (in mM): 150 NaCl, 2.5 KCl, 1 CaCl2 and 0.1 glycine in 10 HEPES adjusted to pH8.0 with NaOH. The bath was superfused constantly with external solution at a rate of 1-2 ml/min. The external solution with and without glutamate (1 mM) was applied to outside-out patches using a theta glass application pipette mounted on a piezoelectric bimorph (Robert and Howe, 2003). Patches were positioned near the interface of the solutions flowing from the theta glass pipette. The speed of solution exchange (estimated from open tip potentials) was 300-600 µs. To measure deactivation decays, outside-out patches were subjected to fast application of glutamate (1 mM) for 2 ms and for each patch, 50 applications were repeated at 6 s intervals. The ensemble averages were constructed from these data and the decays of the currents were fitted with bi-exponential functions in Igor Pro software (Wavemetrics, Portland, OR). These fits were used to obtain weighted time constants of deactivation from the time constants and relative amplitudes of the two exponential components. Single channel data recorded during continuous glutamate application were sampled at 50 kHz and subjected to digital low-pass filtering at 3 kHz. The filtered data were exported to QuB acquisition software (www.qub.buffalo.edu) for analysis. The baseline was manually set to zero and the entire data set was idealized with the segmental k-means (SKM) algorithm of QuB to estimate conductance levels and event durations. The kinetic model used for SKM idealization of the results contained two closed states and two open states corresponding to the two major conductance levels of NMDA receptors. Conductance levels and open times were obtained from the MIL subroutine of QuB. Open durations were exported to ChanneLab (Synaptosoft, Inc.) and mean durations obtained from multi-exponential fits to open-time histograms. The open time and conductance estimates were obtained from fitting distributions of measurements of several hundred events in each patch analyzed. In addition, the conductance estimates for WT receptors have been made many times by different groups and the values we obtained are in excellent agreement with previous reports (Stern et al., 1992). SH2-domain screen. Peptides (tyrosine phosphorylated and unphosphorylated) surrounding Y1252 were synthesized by the Tufts University Core Facility (Boston, MA) with an N-terminal biotin-glycine-serine and an amidated C-terminus: biotin-GS-KKAGNL(pY)DISEDNS-NH2, biotin-GS-KKAGNL(Y)DISEDNS-NH2. Reversephase SH2 domain screening (Rosette assay) was performed as described previously (Ng and Machida, 2017). Peptides were diluted to approximately 10 µg/µl in spotting solution (50 mM Tris-HCl, pH6.8, 15% glycerol, 10% SDS) and immobilized in duplicate in a rosette pattern on gelatin-coated nitrocellulose membranes. The membranes were then blocked with 5% milk/TBST and incubated with 100 nM GST-tagged proteins (Table S3) for 2 h in a 96well plate, followed by incubation with Dylight 800-conjugated anti-GST antibody (Rockland Immunochemicals, Limerick, PA) and IRDye 680RD-conjugated streptavidin (LI-COR Biotechnology, Lincoln, NE). Probe binding was detected with an Odyssey IR scanner and the signal from 8 experiments was quantified using ImageStation software (LI-COR Biotechnology). Signal from binding to the unphosphorylated peptide was used for background subtraction. Co-immunoprecipitation. HEK293 cells were transfected with 1 µg Src pcDNA3.1+, 1 µg GluN1 pCMV, 3 µg GluN2B pcDNA3.1+, and 2 µg Myc-Nck2 pCMV-Entry, and maintained as described for substrate trapping. 48h after transfection, cells were washed with ice cold PBS before gently scraping off the plate and removing to an microcentrifuge tube, where they were crosslinked with cell-permeable 1 mM DSP (dithiobis(succinimidyl propionate; Thermo)) for 30 minutes before extra DSP was quenched with 20 mM Tris pH7.4. Cells were pelleted and washed again with ice cold PBS before being lysed in 2% TX100, PBS and protease and phosphatase inhibitors. Protein concentration was determined by BCA assay and 200 µg lysate was precleared with 20 µL protein A/G agarose (Thermo). 5% of pre-cleared lysate was reserved as input while the remaining pre-cleared lysate was incubated with 3 µg of either mouse anti-GluN2B (Neuromab 75-097) or mouse IgG (Rockland 010-0140) overnight at 4˚C. The next morning, antibody complexes were immobilized with 20µL of fresh protein A/G agarose for 1 h at 4˚C, and the agarose was washed 3 times with 2% TX100/PBS and once with PBS to remove nonspecific binding before bound proteins were eluted with sample buffer at room temperature for 15 minutes. Inputs and IPs were run on 4-15% gradient gels (Biorad) and western blotted as described.
Primary dissociated hippocampal culture and immunofluorescence. Primary neuronal cultures were prepared from littermate P0-P1 WT or NS hippocampus by papain digestion and subsequent dissociation as previously described (Lin et al., 2013). Neurons were plated in Neurobasal A (Invitrogen) supplemented with 2% Gem21 (Gemini) and 10% FBS on 12mm coverslips pre-coated with 20 µg/ml poly-(D)-lysine (BD Biosciences) and 1 µg/ml laminin (Corning). After 4 h the media was changed to serum-free Neurobasal A/Gem21 media containing 1% pen/strep and 2 mM L-glutamine. Glial growth was suppressed with 1 µM ara-C (Sigma) on DIV 3 and 50% removed on DIV 6. eGFP-N1 DNA was transfected by calcium-phosphate precipitation at DIV 12. On DIV 15 cells were fixed at room temperature for 20 minutes with 2% PFA in cytoskeleton buffer (10 mM MES pH 6.8, 138 mM KCl, 3 mM MgCl2, 2mM EGTA and 0.32 mM sucrose). Fixed cells were permeabilized for 15 minutes with 0.3% TX100/TBS (150 mM NaCl, 20mM Tris pH 7.4), washed 3 times with 0.1% TX100/TBS, and blocked for 30 minutes with 0.1% TX100/2% BSA/10% Normal Donkey Serum/TBS. Cells were then incubated at 4˚C overnight with rabbit anti-Nck2 (Invitrogen PA5-49726; 1:100) and GFP-booster_Atto488 (Chromotek gba-488; 1:500) in 0.1% TX100/2% BSA/TBS. The next morning, the cells were washed with 0.1% TX100/TBS and sequentially incubated with goat anti rabbit Atto594 (Sigma 77671; 1:1000) and Phalloidin-Atto647N (Sigma 65906; 1:40), both diluted in 0.1% TX100/2% BSA/TBS, with 0.1% TX100/TBS washes following each secondary. The cells were mounted with ProLong Diamond Antifade Mountant (Invitrogen P36970) after a final wash with TBS and imaged on a spinning disk confocal microscope as described above. Images were collected with identical acquisition parameters and analyzed by an experimenter blind to genotype. Phalloidin and Nck2 staining were quantified in FIJI from ROIs placed on dendritic spines where staining was unambiguously from the transfected neuron. Statistical analysis. All data are presented as means and SEM, unless otherwise noted. Most comparisons were made with unpaired two-tailed Student’s t-tests (or Welch’s test if F-test revealed different standard deviations) or with paired t-tests, and normality was confirmed with a D’Agostino and Pearson normality test where appropriate. For object recognition, we used one-sample t-tests of object 1 or novel object exploration time against a hypothetical mean of 15 s. PSD length, spine head area, and Nck2/phalloidin immunofluorescence were not normally distributed so we used Mann-Whitney tests. The frequency distributions of PSD length and spine head area were generated with bin widths of 40 nm and 0.04 µM2, respectively. We tested for group differences in LTP with two-way ANOVA. For evoked NMDAR-EPSC decay with ifenprodil, we used two-way repeated measures ANOVA with post-hoc HolmSidak test for comparisons between groups. Dephosphorylation time courses were fit with a one-phase exponential decay with Y0=1. Only total pY and pY1252 fit significantly better than a line (extra sum-of-squares F-test p
