Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Mar;27(3):484-496.
doi: 10.1038/s41593-023-01552-9. Epub 2024 Jan 17.

The developmental timing of spinal touch processing alterations predicts behavioral changes in genetic mouse models of autism spectrum disorders

Aniqa Tasnim  1 , Ilayda Alkislar  1 , Richard Hakim  1 , Josef Turecek  1 , Amira Abdelaziz  1 , Lauren L Orefice  2   3 , David D Ginty  4
Affiliations

The developmental timing of spinal touch processing alterations predicts behavioral changes in genetic mouse models of autism spectrum disorders

Aniqa Tasnim et al. Nat Neurosci. 2024 Mar.

Abstract

Altered somatosensory reactivity is frequently observed among individuals with autism spectrum disorders (ASDs). Here, we report that although multiple mouse models of ASD exhibit aberrant somatosensory behaviors in adulthood, some models exhibit altered tactile reactivity as early as embryonic development, whereas in others, altered reactivity emerges later in life. Additionally, tactile overreactivity during neonatal development is associated with anxiety-like behaviors and social behavior deficits in adulthood, whereas tactile overreactivity that emerges later in life is not. The locus of circuit disruption dictates the timing of aberrant tactile behaviors, as altered feedback or presynaptic inhibition of peripheral mechanosensory neurons leads to abnormal tactile reactivity during neonatal development, whereas disruptions in feedforward inhibition in the spinal cord lead to touch reactivity alterations that manifest later in life. Thus, the developmental timing of aberrant touch processing can predict the manifestation of ASD-associated behaviors in mouse models, and differential timing of sensory disturbance onset may contribute to phenotypic diversity across individuals with ASD.

PubMed Disclaimer

Conflict of interest statement

L.L.O. and D.D.G. are consultants with Deerfield Management Company, and D.D.G. is also a consultant with Decibel Therapeutics, Inc., and Ono Pharma USA, Inc. The other authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Multiple global knockout models of ASD-associated genes exhibit tactile overreactivity in adulthood, with differences in ASD-associated behaviors.
a, Diagram showing tactile PPI of the startle reflex assay paradigm. The decrease in the startle response to a 120-dB acoustic pulse when preceded by a light air puff is defined as the effect of PPI (measured as a percentage). The interval between the prepulse and pulse was 250 ms; Stim, stimulus. b, Tactile PPI; n = 12 Gabrb3+/+ mice and n = 13 Gabrb3+/– mice, P = 0.0071; n = 14 Mecp2+/y mice and n = 5 Mecp2–/y mice, P = 0.0033; n = 14 Nlgn2+/+ mice and n = 18 Nlgn2+/– mice, P = 0.0043 for Nlgn2; n = 13 Rorb+/+ mice and n = 18 Rorbh1/+ mice, P = 0.0015. Data were analyzed by unpaired two-tailed t-tests. c, Response to the 0.9-psi air puff stimulus alone, expressed as a percentage of startle response to an acoustic startle stimulus; n = 12 Gabrb3+/+ mice and n = 13 Gabrb3+/– mice, P = 0.0071; n = 14 Mecp2+/y mice and n = 5 Mecp2–/y mice, P = 0.0033; n = 14 Nlgn2+/+ mice and n = 18 Nlgn2+/– mice, P = 0.0043 for Nlgn2; n = 13 Rorb+/+ mice and n = 18 Rorbh1/+ mice, P = 0.0457 for Gabrb3, P = 0.0003 for Mecp2, P = 0.0054 for Nlgn2 and P = 0.7374 for Rorb. Data were analyzed by two-tailed Mann–Whitney U-tests. d, Representative activity traces in the open field assay for mutant mice and control littermates. e, Fraction of time spent in the center of the open field chamber; n = 22 Gabrb3+/+ mice and n = 19 Gabrb3+/– mice, P < 0.0001; n = 15 Mecp2+/y mice and n = 8 Mecp2–/y mice, P = 0.0022; n = 23 Nlgn2+/+ mice and n = 30 Nlgn2+/– mice, P = 0.2050 for Nlgn2; n = 14 Rorb+/+ mice and n = 18 Rorbh1/+ mice, P = 0.5339. Data were analyzed by unpaired two-tailed t-tests. f, Representative heat maps showing time spent with a novel mouse or novel object (an empty cup) in the three-chamber social interaction assay for mutant mice and control littermates. g, Percent preference for a chamber containing a novel mouse versus a novel object in the three-chamber social interaction assay; n = 15 Gabrb3+/+ mice and n = 15 Gabrb3+/– mice, P = 0.0095; n = 9 Mecp2+/y mice and n = 8 Mecp2–/y mice, P = 0.0318; n = 10 Nlgn2+/+ mice and n = 14 Nlgn2+/– mice, P = 0.8603 for Nlgn2; n = 9 Rorb+/+ mice and n = 12 Rorbh1/+ mice, P = 0.9887. Data were analyzed by unpaired two-tailed t-tests. For b, c, e and g, data represent mean ± s.e.m. Statistical information is provided in Supplementary Table 1; NS, not significant; *P < 0.05; **P < 0.001; ***P < 0.0001. Source data
Fig. 2
Fig. 2. Neonatal tactile reactivity in global knockout ASD mouse models.
a, Diagram showing the experimental paradigm for the neonatal air puff responsivity assay. Air puff tubing was affixed 3 mm above the nape of the neck of neonatal mice. Maximal body displacements in the 500 ms following a 50-ms air puff were measured using optic flow analysis. Parts were created with BioRender.com. b, Example displacement trace of a control P4 mouse to a single 1.0-psi air puff stimulus. c, Displacement responses to a single presentation of 0.10-, 0.25-, 0.50-, 0.75- and 1.0-psi stimuli; n = 16 Gabrb3+/+ mice and n = 10 Gabrb3+/– mice; n = 12 Mecp2+/y mice and n = 11 Mecp2–/y mice; n = 17 Nlgn2+/+ mice and n = 15 Nlgn2+/– mice; n = 18 Rorb+/+ mice and n = 17 Rorbh1/+ mice. Data were analyzed by mixed-effects analyses, effect of genotype; Gabrb3: F1,24 = 9.060 and P = 0.0030; Mecp2: F1,21 = 9.253 and P = 0.0062; Nlgn2: F1,30 = 0.1071 and P = 0.7457; Rorb: F1,33 = 0.1204 and P = 0.7308. Data are shown as mean ± s.e.m. d, Average displacement responses to ten presentations of a 1.0-psi stimulus at 20- to 30-s interstimulus intervals; n = 16 Gabrb3+/+ mice and n = 10 Gabrb3+/– mice; n = 12 Mecp2+/y mice and n = 11 Mecp2–/y mice; n = 17 Nlgn2+/+ mice and n = 15 Nlgn2+/– mice; n = 18 Rorb+/+ mice and n = 17 Rorbh1/+ mice. P = 0.0003 for Gabrb3, P = 0.0018 for Mecp2, P = 0.2258 for Nlgn2 and P = 0.1970 for Rorb. Data were analyzed by unpaired one-tailed t-tests and are shown as mean ± s.e.m. e, Example traces showing displacements to ten presentations of 1.0-psi stimuli from a Gabrb3+/+ and Gabrb3+/– mouse. To calculate habituation, the average of the last three stimulus presentation responses was compared to the average of the first three responses. A >25% reduction in responsivity between blocks was considered to be a habituation response. f, Fraction of animals habituating to repeated presentations of 1.0-psi air puffs; P = 0.0244 for Gabrb3, P = 0.0317 for Mecp2; P = 0.7258 for Nlgn2 and P = 0.6017 for Rorb. Data were analyzed by one-sided Fisher’s exact tests. Statistical information is provided in Supplementary Table 1; *P < 0.05; **P < 0.001. Source data
Fig. 3
Fig. 3. Gabrb3-mutant mice are overreactive to activation of Aβ-LTMRs at birth and E18.5.
a, Experimental setup for the perinatal mechanoreceptor optical activation assay. P0 or E18.5 mice were placed on a clear acrylic stage, and LED illumination was directed to the paw or back hairy skin. Parts created with BioRender.com. b, Average displacement responses to five optical stimuli (20- to 30-s interstimulus intervals) delivered to mechanoreceptor subtypes in the forepaw, back hairy skin and hindpaw compared to control littermates that lack opsin expression (controls include opsin+Cre– and opsin–Cre+ animals). For the forepaw, n = 10 negative-control mice, n = 6 Ret+ mice, n = 8 TrkC+ mice and n = 5 TrkB+ mice. For back hairy skin, n = 9 negative-control mice, n = 8 Ret+ mice, n = 7 TrkC+ mice and n = 5 TrkB+ mice. For the hindpaw, n = 4 negative-control mice, n = 3 Ret+ mice, n = 4 TrkC+ mice and n = 5 TrkB+ mice. SA-LTMR, slowly adapting LTMR; RA-LTMR, rapidly adapting LTMR. c, Average displacement responses to five optical stimuli activating Ret+ Aβ-LTMRs to the forepaws of P0 RetcreER; AvilFlpO; Rosa26LSL-FSF-ReaChR-mCitrine; Gabrb3+/+ and Gabrb3+/– animals; n = 3 Gabrb3+/+ mice and n = 6 Gabrb3+/– mice; P = 0.0169. Data were analyzed by unpaired one-tailed Welch’s t-test. d, Example displacement traces from an E18.5 Gabrb3+/+ mouse and Gabrb3+/– mouse to optical activation of Ret+ Aβ-LTMRs in the forepaw. e, Average displacement responses to five optical stimuli activating Ret+ Aβ-LTMRs in the forepaws of E18.5 control and mutant animals; n = 11 Gabrb3+/+ mice and n = 9 Gabrb3+/– mice; P = 0.0029. Data were analyzed by unpaired one-tailed Welch’s t-test. For b, c and e, data are shown as mean ± s.e.m. Statistical information is provided in Supplementary Table 1; *P < 0.05. Source data
Fig. 4
Fig. 4. Differential cell-autonomous requirements for Gabrb3 and Nlgn2 in peripheral and spinal cord neuron types for tactile behaviors.
a,b, Schematics showing areas of Cre activity (shaded in red) in the spinal cord and/or DRG for the genetic strategies shown below (top row). Immunohistochemistry of the adult spinal cord shows the expression of VGLUT1-labeled synaptic terminals and NLGN2 in control, Cdx2Cre; Nlgn2fl/fl and Lbx1Cre; Nlgn2fl/fl mice (a) and expression of VGLUT and GABRB3 in control and Lbx1Cre; Gabrb3fl/fl mice (b). White dotted lines show the outline of the spinal cord gray matter, and yellow dotted boxes and insets show the dorsal horn; scale bars, 200 µm (full-size images) and 100 µm (insets). c, Tactile PPI; n = 17 control and n = 11 AvilCre; Gabrb3fl/fl mice, P = 0.0014; n = 12 control and n = 10 Lbx1Cre; Gabrb3fl/fl mice, P = 0.6746; n = 18 control and n = 17 Cdx2Cre; Nlgn2fl/fl mice, P = 0.0010; n = 22 control and n = 11 AvilCre; Nlgn2fl/fl mice, P = 0.7051; n = 13 control and n = 14 Lbx1Cre; Nlgn2fl/fl mice, P < 0.0001. Data were analyzed by unpaired two-tailed t-tests. d, Response to a 0.9-psi air puff stimulus alone; n = 17 control and n = 11 AvilCre; Gabrb3fl/fl mice, P = 0.0014; n = 12 control and n = 10 Lbx1Cre; Gabrb3fl/fl mice, P = 0.6746; n = 18 control and n = 17 Cdx2Cre; Nlgn2fl/fl mice, P = 0.0010; n = 22 control and n = 11 AvilCre; Nlgn2fl/fl mice, P = 0.7051; n = 13 control and n = 14 Lbx1Cre; Nlgn2fl/fl mice; P = 0.0011 for AvilCre; Gabrb3fl; P = 0.2543 for Lbx1Cre; Gabrb3fl; P = 0.0204 for Cdx2Cre; Nlgn2fl; P = 0.9919 for AvilCre; Nlgn2fl; P = 0.0107 for Lbx1Cre; Nlgn2fl. Data were analyzed by two-tailed Mann–Whitney U-tests. e, Fraction of time spent in the center of the open field chamber; n = 15 control and n = 21 AvilCre; Gabrb3fl/fl mice, P < 0.0001; n = 18 control and n = 13 Lbx1Cre; Gabrb3fl/fl mice, P = 0.4169; n = 15 control and n = 15 Cdx2Cre; Nlgn2fl/fl mice, P = 0.4524; n = 22 control and n = 11 AvilCre; Nlgn2fl/fl mice, P = 0.1019; n = 13 control and n = 13 Lbx1Cre; Nlgn2fl/fl mice, P = 0.4214. Data were analyzed by unpaired two-tailed t-tests. f, Displacement response to a single presentation of 0.10-, 0.25-, 0.50-, 0.75- and 1.0-psi stimuli at P4; n = 21 control and n = 19 AvilCre; Gabrb3fl/fl mice; n = 17 control and n = 21 Cdx2Cre; Nlgn2fl/fl mice. Data were analyzed by mixed-effects analyses, effect of genotype; AvilCre; Gabrb3fl: F1,38 = 36.38, P < 0.0001; Cdx2Cre; Nlgn2fl: F1,36 = 0.1169, P = 0.7344. g, Average displacement responses to ten presentations of a 1.0-psi stimulus at 20- to 30-s interstimulus intervals; n = 21 control and n = 19 AvilCre; Gabrb3fl/fl mice; n = 17 control and n = 21 Cdx2Cre; Nlgn2fl/fl mice. P < 0.0001 for AvilCre; Gabrb3fl; P = 0.2382 for Cdx2Cre; Nlgn2fl. Data were analyzed by unpaired one-tailed t-tests. h, Fraction of animals habituating to repeated presentations of 1.0-psi air puffs. For Gabrb3, P = 0.0171, and for Nlgn2, P = 0.9333. Data were analyzed by one-sided Fisher’s exact tests. For c–g, data are shown as mean ± s.e.m. Statistical information is provided in Supplementary Table 1; *P < 0.05; ***P < 0.0001. Source data
Fig. 5
Fig. 5. Machinery for presynaptic inhibition of sensory neurons develops neonatally, whereas spinal cord feedforward inhibition is weak and immature during early postnatal development.
a, Experimental setup and example image of a DRG whole-cell recording preparation for RuBi-GABA uncaging; scale bar, 25 µm. b, Example GABA-evoked depolarizations of medium- to large-diameter neurons to a 5-ms uncaging pulse (blue bar, not to scale) in current clamp configuration. c, Quantification of peak uncaged GABA-evoked depolarizations across time points and genotypes. Each dot represents a cell, and n = 5 for control P4, n = 5 for AvilCre; Gabrb3fl/fl P4, n = 6 for control P18–P30, n = 4 for AvilCre; Gabrb3fl/fl P18–P30 and n = 2 control P18–P30 in gabazine. N = 2 animals for each condition, except N = 3 for control P18–P30. P = 0.4592, as determined by unpaired two-tailed t-test. d, Recording schematic for measuring miniature postsynaptic currents from interneurons (both excitatory and inhibitory, denoted ex. and inh., respectively) in the LTMR-RZ (laminae III–IV, which resides underneath the IB4+ superficial lamina II) of sagittal spinal cord slices. Parts created with BioRender.com. e, Example mIPSCs in P4 and P19–P21 spinal cords. Recordings were performed in laminae III/IV of sagittal spinal cord sections. f,g, mIPSC frequencies (f) and amplitudes (g) at P4 and P19–P21, with or without the presence of blockers of synaptic inhibition, gabazine and strychnine. Each dot represents a cell, and n = 11 for P4, n = 8 for P4 (gabazine), n = 9 for P4 (strychnine), n = 9 for P19–P21, n = 7 for P19–P21 (gabazine) and n = 8 for P19–P21 (strychnine). N = 3 animals for each condition, except N = 2 for P4 (gabazine), P4 (strychnine) and P19–P21 (strychnine). For frequencies, P = 0.0075 for P4 versus P4 (gabazine), P = 0.9959 for P4 versus P4 (strychnine), P = 0.1404 for P4 (gabazine) versus P4 (strychnine), P = 0.0024 for P4 versus P19–P21, P = 0.9601 for P19–P21 versus P19–P21 (gabazine), P = 0.0011 for P19–P21 versus P19–P21 (strychnine) and P = 0.0134 for P19–P21 (gabazine) versus P19–P21 (strychnine). For amplitudes, P = 0.9994 for P4 versus P4 (gabazine), P = 0.5713 for P4 versus P4 (strychnine), P = 0.9673 for P4 (gabazine) versus P4 (strychnine), P = 0.0005 for P4 versus P19–P21, P = 0.2662 for P19–P21 versus P19–P21 (gabazine), P = 0.0002 for P19–P21 versus P19–P21 (strychnine) and P = 0.0577 for P19–P21 (gabazine) versus P19–P21 (strychnine). Data were analyzed by Welch’s analysis of variance with Dunnett’s T3 multiple comparisons tests. h, Cholera toxin subunit B (CTB) was injected into the dorsal column or dorsal column nuclei to retrogradely label PSDCs. Sagittal spinal cord slices were prepared, and PSDCs were targeted for whole-cell patch clamp recordings (left) and, in some cases, dialyzed with Alexa Fluor 488 dye for post hoc visualization (right); scale bars, 100 µm (full-size image) and 10 µm (inset). Parts created with BioRender.com. i, Example mIPSCs from CTB-labeled PSDCs in P4 and P19–P21 spinal cords. j, mIPSC frequencies (left) and amplitudes (right) at P4 and P19–P21 onto PSDCs. Each dot represents a cell, and n = 7 for P4 and n = 5 for P19–P21. N = 3 animals for both conditions. For frequencies, P = 0.0024. Data were analyzed by unpaired two-tailed Welch’s t-test. For amplitudes, P < 0.0001. Data were analyzed by unpaired two-tailed t-test. For c, f, g and j, data are shown as mean ± s.e.m. Statistical information is provided in Supplementary Table 1; #P < 0.1; *P < 0.05; **P < 0.001; ***P < 0.0001. Source data
Fig. 6
Fig. 6. NLGN2 is required for feedforward inhibitory neurotransmission in the mature spinal cord dorsal horn.
a, Example mIPSCs in spinal cord neurons of P19–P21 spinal cords in Cdx2Cre; Nlgn2fl/fl animals and control littermates. b, mIPSC frequencies (left) and amplitudes (right) of P19–P21 spinal cords in Cdx2Cre; Nlgn2fl/fl animals and control littermates. Each dot represents a cell, and n = 7 Nlgn2fl/fl cells and n = 7 Cdx2Cre; Nlgn2fl/fl cells. N = 2 animals for both conditions. For frequencies, P = 0.0003. Data were analyzed by unpaired two-tailed Welch’s t-test. For amplitudes, P = 0.0001. Data were analyzed by unpaired two-tailed t-test. c, Frequencies (left) and amplitudes (right) of mEPSCs in the spinal cords of Cdx2Cre; Nlgn2fl/fl animals and control littermates at P19–P21. Each dot represents a cell, and n = 15 Nlgn2fl/fl cells and n = 12 Cdx2Cre; Nlgn2fl/fl cells. N = 2 animals for both conditions. For frequencies, P = 0.6052. Data were analyzed by two-tailed Mann–Whitney U-test. For amplitudes, P = 0.6822. Data were analyzed by unpaired two-tailed t-test. d, Frequencies (left) and amplitudes (right) of mIPSCs in the spinal cords of Cdx2Cre; Nlgn2fl/fl animals and control littermates at P4–P5. Each dot represents a cell, and n = 8 Nlgn2fl/fl cells and n = 8 Cdx2Cre; Nlgn2fl/fl cells. N = 2 animals for both conditions. P = 0.9678 for frequencies, and P = 0.4385 for amplitudes. Data were analyzed by unpaired two-tailed t-tests. e, Proposed model describing the emergence of DRG presynaptic inhibition and spinal cord (SC) feedforward inhibition across development, with observed behavioral consequences of disrupting ASD-associated genes involved in either pathway. For b–d, data are shown as mean ± s.e.m. Statistical information is provided in Supplementary Table 1; **P < 0.001. Source data
Extended Data Fig. 1
Extended Data Fig. 1. Additional behavioral characterization of global knockout ASD models.
Related to Fig. 1. a, Distance traveled in the open field. n = 22 Gabrb3+/+ mice and n = 19 Gabrb3+/- mice, P = 0.7133; n = 15 Mecp2+/y mice and n = 8 Mecp2-/y mice, P = 0.0011; n = 14 Rorb+/+ mice and n = 18 Rorbh1/+ mice, P = 0.9394, unpaired two-tailed t-tests. n = 23 Nlgn2+/+ mice, n = 30 Nlgn2+/- mice, n = 14 Nlgn2-/- mice, P = 0.9832 for Nlgn2+/+ vs. Nlgn2+/-, P = 0.9755 for Nlgn2+/+ vs. Nlgn2-/-, one-way ANOVA with post hoc Dunnett’s tests. b, c, Discrimination indices for textured NORT assay, n = 13 Nlgn2+/+ mice and n = 19 Nlgn2+/- mice, P = 0.0376 (b) and shape/color NORT assay, n = 10 Nlgn2+/+ mice and n = 21 Nlgn2+/- mice, P = 0.9729 (c), with a positive value indicating a preference for a novel object versus a familiar object. Statistical analyses were performed using unpaired two-tailed t-tests. d, Fraction of time spent in the center of the open field chamber. n = 23 Nlgn2+/+ mice, n = 30 Nlgn2+/- mice, n = 14 Nlgn2-/- mice, P = 0.3289 for Nlgn2+/+ vs. Nlgn2+/-, P = 0.0020 for Nlgn2+/+ vs. Nlgn2-/-, one-way ANOVA with post hoc Dunnett’s tests. e, Fraction of time spent in the open arms of the elevated plus maze. n = 17 Nlgn2+/+ mice, n = 31 Nlgn2+/- mice, n = 12 Nlgn2-/- mice, P = 0.6302 for Nlgn2+/+ vs. Nlgn2+/-, P = 0.6886 for Nlgn2+/+ vs. Nlgn2-/-, one-way ANOVA with post hoc Dunnett’s tests. n = 14 Rorb+/+ mice and n = 18 Rorbh1/+ mice, P = 0.4630 for Rorb, unpaired two-tailed t-test. f, Tactile PPI. n = 15 Nlgn4+/+ mice, n = 9 Nlgn4+/- mice, n = 13 Nlgn4-/- mice, P = 0.8054 for Nlgn4+/+ vs. Nlgn4+/-, P = 0.8032 for Nlgn4+/+ vs. Nlgn4-/-. g, Response to a 0.9 PSI air puff stimulus alone. n = 15 Nlgn4+/+ mice, n = 9 Nlgn4+/- mice, n = 13 Nlgn4-/- mice, P = 0.8100 for Nlgn4+/+ vs. Nlgn4+/-, P = 0.6223 for Nlgn4+/+ vs. Nlgn4-/-. h, Fraction of time spent in the center of the open field chamber. n = 16 Nlgn4+/+ mice, n = 9 Nlgn4+/- mice, n = 15 Nlgn4-/- mice, P = 0.1075 for Nlgn4+/+ vs. Nlgn4+/-, P = 0.8535 for Nlgn4+/+ vs. Nlgn4-/-. i, Distance traveled in the open field. n = 16 Nlgn4+/+ mice, n = 9 Nlgn4+/- mice, n = 15 Nlgn4-/- mice, P = 0.9058 for Nlgn4+/+ vs. Nlgn4+/-, P = 0.8561 for Nlgn4+/+ vs. Nlgn4-/-. For a-i, data represent means ± s.e.m. For f-i, statistical analyses were performed using one-way ANOVAs with post hoc Dunnett’s tests. Statistical information provided in Supplementary Table 1.
Extended Data Fig. 2
Extended Data Fig. 2. Additional behavioral and molecular characterization of ASD mouse models during neonatal ages.
Related to Fig. 2. a, Displacement responses to single air puff stimuli of varying intensities delivered in sequence to the nape of the neck, in P4 animals treated with Nair on the back hairy skin and non-treated animals. At the end of back hairy skin trials, the 1.0 PSI stimulus was delivered to the whisker pad. N = 2 control mice and N = 2 Nair-treated mice. b, Displacement responses to a single 1.0 PSI air puff stimulus before and after topical lidocaine application to the nape of the neck. Each pair of dots represents the same P4 animal pre- and post-application. N = 3 mice. c, Dorsal root ganglia immunohistochemistry showing expression of NeuN and MeCP2 in control mice at P0-P1 and P4, and in Mecp2-/y mice to validate antibody specificity. Scale bar denotes 50 µm. d-e, Spinal cord immunohistochemistry showing expression of VGLUT1-labeled synaptic terminals and GABRB3 in control P0-P1 and adult mice. Scale bars denote 10 µm (d), 5 µm (e, left column), and 2 µm (e, right column). Red arrows denote co-labeled puncta. f, Quantification of VGLUT1 and GABRB3 (top) or VGAT and GABRB3 (bottom) co-labeled puncta, as a fraction of total VGLUT1+ or VGAT+ puncta in a field of view, at P0-P1 or adulthood. Red arrows denote examples of proximity between markers. Each dot represents the average value for one animal, with three spinal cord images taken per animal. N = 3 animals for both conditions. g, Displacement responses to a single presentation of 0.10, 0.25, 0.50, 0.75 and 1.0 PSI stimuli for global mutant animals. Each dot represents the displacement response of a single animal on an air puff trial, and dashes represent means. n = 16 Gabrb3+/+ mice and n = 10 Gabrb3+/- mice; n = 12 Mecp2+/y mice and n = 11 Mecp2-/y mice; n = 17 Nlgn2+/+ mice and n = 15 Nlgn2+/- mice; n = 18 Rorb+/+ mice and n = 17 Rorbh1/+ mice. h, Average displacement responses to 10 presentations of a 1.0 PSI stimulus at 20-30 second interstimulus intervals. n = 17 Nlgn2+/+ mice, n = 15 Nlgn2+/- mice, n = 8 Nlgn2-/- mice, P = 0.6848 for Nlgn2+/+ vs. Nlgn2+/-, P = 0.9749 for Nlgn2+/+ vs. Nlgn2-/-. n = 18 Rorb+/+ mice, n = 17 Rorbh1/+ mice, n = 7 Rorbh1/h1 mice, P = 0.6596 for Rorb+/+ vs. Rorbh1/+, P = 0.2211 for Rorb+/+ vs. Rorbh1/h1, one-way ANOVAs with post hoc Dunnett’s tests. i, Fraction of animals habituating to repeated presentations of 1.0 PSI air puffs. P = 0.7043 for Nlgn2+/+ vs. Nlgn2-/-, P = 0.6495 for Rorb+/+ vs. Rorbh1/h1, one-sided Fisher’s Exact tests. For a, f, and h, data represent means ± s.e.m. Statistical information provided in Supplementary Table 1.
Extended Data Fig. 3
Extended Data Fig. 3. Anatomical characterization of opsin expression in low-threshold mechanoreceptor labeling paradigm at birth.
Related to Fig. 3. a, Glabrous forepaw and back hairy skin immunohistochemistry showing expression of ReaChR::mCitrine (green), Neurofilament Heavy Chain (NFH; magenta), and DAPI (gray) from P0 animals with RetCreER, TrkCCreER, or TrkBCreER mechanosensory neuron labeling strategies. Scale bars denote 20 µm.
Extended Data Fig. 4
Extended Data Fig. 4. Immunohistochemical characterization of Nlgn2 sensory neuron knockout tissue.
Related to Fig. 4. a, Schematics showing imaging area (black dotted box) and areas of Cre activity, where relevant (shaded in red) (left). In situ hybridization labeling using RNAScope for Nefh (large-diameter DRG neurons), Mrgprd (small-diameter nonpeptidergic DRG neurons), and Nlgn2 transcripts in DRG cells of Nlgn2+/+, Nlgn2+/-, and AdvillinCre; Nlgn2flox/flox mice (right). Scale bar denotes 20 µm. b-e, Spinal cord immunohistochemistry showing expression of VGLUT1-labeled synaptic terminals and NLGN2 (b, c) and GABRB3 (d, e) in control and AdvillinCre; Nlgn2flox/flox mice. Scale bars denote 10 µm (b, d) and 5 µm (c, e). f, Quantification of VGLUT1 and NLGN2 (top) and VGLUT1 and GABRB3 (bottom) co-labeled puncta, as a fraction of total VGLUT1+ puncta in a field of view. Each dot represents the average value for one animal, with three spinal cord images taken per animal. N = 3 Nlgn2flox/flox animals and N = 2 AdvillinCre; Nlgn2flox/flox mice (top) and N = 8 Nlgn2flox/flox animals and N = 5 AdvillinCre; Nlgn2flox/flox mice (bottom). Data represent means ± s.e.m.
Extended Data Fig. 5
Extended Data Fig. 5. Additional behavioral characterization of conditional knockout ASD models.
Related to Fig. 4. a, Tactile PPI. n = 18 control mice, n = 12 Cdx2Cre; Nlgn2flox/+ mice, n = 17 Cdx2Cre; Nlgn2flox/flox mice, P = 0.0367 for Control vs. Cdx2Cre; Nlgn2flox/+, P = 0.0012 for Control vs. Cdx2Cre; Nlgn2flox/flox. n = 13 control mice, n = 14 Lbx1Cre; Nlgn2flox/+ mice, n = 14 Lbx1Cre; Nlgn2flox/flox mice, P < 0.0001 for Control vs. Lbx1Cre; Nlgn2flox/+, P < 0.0001 for Control vs. Lbx1Cre; Nlgn2flox/flox, one-way ANOVAs with post hoc Dunnett’s tests. b, Response to a 0.9 PSI air puff stimulus alone. n = 18 control mice, n = 12 Cdx2Cre; Nlgn2flox/+ mice, n = 17 Cdx2Cre; Nlgn2flox/flox mice, P = 0.0367 for Control vs. Cdx2Cre; Nlgn2flox/+, P = 0.0012 for Control vs. Cdx2Cre; Nlgn2flox/flox. n = 13 control mice, n = 14 Lbx1Cre; Nlgn2flox/+ mice, n = 14 Lbx1Cre; Nlgn2flox/flox mice, P = 0.0181 for Control vs. Cdx2Cre; Nlgn2flox/+, P = 0.0383 for Control vs. Cdx2Cre; Nlgn2flox/flox, P = 0.0014 for Control vs. Lbx1Cre; Nlgn2flox/+, P = 0.0174 for Control vs. Lbx1Cre; Nlgn2flox/flox, Kruskal-Wallis tests with post hoc Dunn’s tests. c, Discrimination indices for textured NORT. n = 15 control mice and n = 11 Cdx2Cre; Nlgn2flox/flox mice, P = 0.0357; n = 24 control mice and n = 12 Lbx1Cre; Nlgn2flox/flox mice, P = 0.0028, unpaired two-tailed t-tests. d, Distance traveled in the open field. n = 15 control and n = 21 AdvillinCre; Gabrb3flox/flox mice, P = 0.1222; n = 18 control and n = 13 Lbx1Cre; Gabrb3flox/flox mice, P = 0.4756;; n = 15 control and n = 15 Cdx2Cre; Nlgn2flox/flox mice, P = 0.2015; n = 22 control and n = 11 AdvillinCre; Nlgn2flox/flox mice, P = 0.1862; n = 13 control and n = 13 Lbx1Cre; Nlgn2flox/flox mice, P = 0.6495, unpaired two-tailed t-tests. e, Percent preference for a chamber containing a novel mouse versus a novel object in the 3-chamber social interaction assay. n = 14 control mice and n = 7 AdvillinCre; Gabrb3flox/flox mice, P = 0.0217; n = 16 control mice and n = 12 Lbx1Cre; Gabrb3flox/flox mice, P = 0.8816; n = 8 control mice and n = 11 Cdx2Cre; Nlgn2flox/flox mice, P = 0.7776, unpaired two-tailed t-tests. f, Displacement responses to a single presentation of 0.10, 0.25, 0.50, 0.75 and 1.0 PSI stimuli for P4 AdvillinCre; Gabrb3flox/flox (left) and Cdx2Cre; Nlgn2flox/flox (right) and control littermate animals. Each dot represents the displacement response of a single animal on an air puff trial, and dashes represent means. n = 21 control and n = 19 AdvillinCre; Gabrb3flox/flox mice; n = 17 control and n = 21 Cdx2Cre; Nlgn2flox/flox mice. For a-e, data represent means ± s.e.m. Statistical information provided in Supplementary Table 1.
Extended Data Fig. 6
Extended Data Fig. 6. Synaptic elements for spinal cord inhibition emerge at different times, and additional characterization of neurotransmission in developing spinal cords.
Related to Fig. 5. a, Anatomical labeling of the corticospinal tract in thoracic spinal cords of P0, P4, and adult animals labeled by in Emx1Cre; Rosa26LSL-synaptophysin-tdTomato. Scale bar denotes 200 µm. Parts created with BioRender.com. b, Example miniature excitatory post-synaptic currents (mEPSCs) from P4 and P19-21 spinal cord interneurons of lamina III/IV (top), and quantification of frequencies (bottom, left) and amplitudes (bottom, right). Each dot represents a cell, and n = 9 cells for P4 (N = 2 mice) and n = 15 cells for P19-P21 (N = 3 mice). For frequencies, P = 0.0929, two-tailed Mann-Whitney U test. For amplitudes, P = 0.6822, unpaired two-tailed t-test. c,d, Spinal cord immunohistochemistry showing expression of VGAT-labeled synaptic terminals and GLYRα1 in P3-P4 and adult mice. Scale bars denote 10 µm (c) and 2 µm (d). e, Quantification of VGAT and GLYRα1 co-labeled puncta, as a fraction of total VGAT+ puncta in a field of view, at P3-P4 or adulthood. Each dot represents the average value for one animal, with three spinal cord images taken per animal. N = 5 P3-P4 mice and N = 4 adult mice, P < 0.0001, unpaired two-tailed t-test. For b and e, data represent means ± s.e.m. Statistical information provided in Supplementary Table 1.
Extended Data Fig. 7
Extended Data Fig. 7. NLGN2 and gephyrin immunoreactivity are lacking at P4.
Related to Fig. 6. a, b, Spinal cord immunohistochemistry of VGAT and NLGN2 in P4 and adult animals. Scale bars denote 5 µm. c, Quantification of VGAT/NLGN2 co-labeling. d, e, Spinal cord immunohistochemistry of VGAT and gephyrin in P4 and adult animals. Scale bars denote 5 µm. f, Quantification of VGAT/gephyrin co-labeling. For c and f, Each dot represents the average value for one animal, with three spinal cord images taken per animal. N = 3 P3-P4 mice and N = 7 adult mice for NLGN2, and N = 3 P3-P4 mice and N = 4 adult mice for gephyrin. Data represent means ± s.e.m. g, Proposed model for the molecular and circuit development of presynaptic inhibition (PSI) and feedforward inhibition (FFI) in the spinal cord. Glu denotes glutamate, and GlyR denotes glycine receptor.
See this image and copyright information in PMC

Update of

Similar articles

See all similar articles

Cited by

References

    1. Crane L, Goddard L, Pring L. Sensory processing in adults with autism spectrum disorders. Autism. 2009;13:215–228. - PubMed
    1. Leekam SR, Nieto C, Libby SJ, Wing L, Gould J. Describing the sensory abnormalities of children and adults with autism. J. Autism Dev. Disord. 2007;37:894–910. - PubMed
    1. Marco EJ, Hinkley LB, Hill SS, Nagarajan SS. Sensory processing in autism: a review of neurophysiologic findings. Pediatr. Res. 2011;69:48R–54R. - PMC - PubMed
    1. Tomchek SD, Dunn W. Sensory processing in children with and without autism: a comparative study using the short sensory profile. Am. J. Occup. Ther. 2007;61:190–200. - PubMed
    1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders 5th edn (American Psychiatric Association Publishing, 2013).