Fig. 4
Neuronal mechanisms underlying the activity-dependence of gal3-induced impairment of gamma oscillations. a Left, general schematic representation of the experimental set up. Right, representative power spectra of network activity (0–100 Hz) and representative FSN AP firing windows relative to concomitant gamma oscillation 30 min after KA application for the color-coded experimental conditions shown on the left. b–e correspond to recordings of FSN membrane potential performed in the quiescent network state for 15 min. Additional measurements for each condition (control, gal3 or gal3 + TD139 application) such as firing at basal membrane potential and firing threshold are provided in Fig. S3. Effects on EPSCs in basal state are provided in Additional file 1: Fig. S4. f–n correspond to FSN-gamma phase-lock analyzed from concomitant recordings of the same FSN and gamma network activity for 30 min after 100 nM KA application to induce stable gamma oscillations. Additional measurements for each condition in the activated state such as FSN EPSCs are provided in Additional file 1: Fig. S5. Note that both recording electrodes (whole cell FSN patch clamp and LFP recording) were positioned and left in place for the entire experiment (left hippocampal diagram), including gal3 or gal3 + TD139 applications during the quiescent state recordings (first 15 min) and during the subsequent gamma induction (following 30 min). b Time course of the effect of 15 min wash-in of 1 µM gal3 (red) or co-application of 1 µM gal3 + 10 µM TD139 (blue) on FSN membrane potential (Em) in the quiescent network state. Basal control condition (ACSF) is shown in gray. c–e show the quantification of FSN membrane potential over 15 min of control recordings (basal: − 60.2 ± 0.73 mV, ACSF: − 60.0 ± 0.92 mV, n = 18, N = 6; P = 0.7427), after gal3 (basal: − 60.4 ± 1.61 mV, gal3: − 62.3 ± 1.60 mV, n = 10, N = 7; P = 0.0346) or after gal3 + TD139 (basal: − 60.1 ± 0.82 mV, gal3: − 59.0 ± 1.14 mV, n = 10, N = 5; P = 0.1804). Statistics performed: two-tailed paired t-test. Inset: Example traces showing 1 min recorded in basal (control) condition and the last minute recorded for the quantification of the effect of gal3 (red), gal3 + TD139 (blue) as well as the control recorded just with ACSF (gray). Corresponding effects on AP firing for each condition when the FSN was firing at basal membrane potential are shown in Additional file 1: Fig. S5. f Representative traces of concomitant recordings (upper: APs, lower: gamma oscillations) in control conditions as well as in the presence of gal3 or gal3 + TD139. g Summary of gamma oscillation power calculated from 20 to 80 Hz for each condition showing that co-application of TD139 counteracts the gal3-induced decrease of gamma oscillation power (ACSF: 1.43 ± 0.21 × 10–09 V2, n = 15, N = 6; gal3: 0.52 ± 0.12 × 10–09 V2, n = 15, N = 7; P = 0.0050 vs control, P < 0.0001 vs gal3 + TD139, gal3 + TD139: 2 ± 0.31 × 10–09 V2, n = 11, N = 5; P = 0.0644 vs control). h Summary of gamma peak frequency showing that co-application of TD139 counteracts the gal3-induced slowing of the gamma rhythm (ACSF: 26.5 ± 0.71 Hz, n = 15, N = 6; gal3: 24.3 ± 0.63 Hz, n = 15, N = 7; P = 0.0372 vs control, P < 0.0099 vs gal3 + TD139, gal3 + TD139: 27.3 ± 0.59 Hz, n = 11, N = 5; P = 0.3867 vs control). i Summary of AP firing rate showing that gal3-induced decrease of AP rate is prevented by co-application of TD139 (ACSF: 7.36 ± 1.47 Hz, n = 12, N = 6; gal3: 0.79 ± 0.26 Hz, n = 9, N = 7; P = 0.0012 vs control, P < 0.0118 vs gal3 + TD139, gal3 + TD139: 5.88 ± 1 Hz, n = 10, N = 5; P = 0.3619 vs control). j Representative polar plots of the firing windows shown in a for each condition with the resultant vector (bottom right) showing the magnitude of the FSN-gamma phase-lock and the phase-angle preference for each condition. k Logarithmic distribution of the P values from the Rayleigh’s test for uniformity showing that in the presence of gal3 considerably fewer FSN are able to engage in a patterned firing locked to a specific gamma phase (ACSF: 1 out of 15 recorded cells, gal3: 3 out of 12 recorded cells, gal3 + TD139: all the recorded cells showed P < 0.05). Pink dashed line denotes P = 0.05. l Summary of the resultant vector length for each condition (ACSF: 0.55 ± 0.07, n = 12, N = 6; gal3: 0.33 ± 0.04, n = 9, N = 7; P = 0.0364 vs control, P < 0.0308 vs gal3 + TD139, gal3 + TD139: 0.59 ± 0.06, n = 10, N = 5; P = 0.7143 vs control). m Quantification of the phase-angle firing preference revealing that neither gal3 nor gal3 + TD139 induced significant changes (ACSF: 5.26 ± 0.11 radians, n = 12, N = 6; gal3: 4.81 ± 0.24 radians, n = 9, N = 7; P = 0.2328 vs control, P = 0.5881vs gal3 + TD139, gal3 + TD139: 5 ± 0.2 radians, n = 10, N = 5; P = 0.3921 vs control). n Summary bar graphs showing that 30 min of KA application did not differentially depolarize FSN either in control conditions, or in the presence of gal3 or gal3 + TD139 (ACSF: − 50 ± 1.17 mV, n = 12, N = 6; gal3: − 51.8 ± 1.32 mV, n = 9, N = 7; P = 0.5595 vs control, P = 0.2304 vs gal3 + TD139, gal3 + TD139: − 48.2 ± 1.54 mV, n = 10, N = 5; P = 0.5595 vs control). Data are presented as mean ± SE. Statistics performed: ordinary one-way ANOVA followed by Holm-Sidak’s multiple comparisons test. Significance levels are shown as *P < 0.05, **P < 0.01, ****P < 0.0001. n.s: no significant statistical difference; n: number of cells recorded in b–e, number of slices recorded in g–h and number of cells and slices recorded concomitantly in i, k–n; N: number of animals