Early Impairment of Synaptic and Intrinsic Excitability in Mice Expressing ALS/Dementia-Linked Mutant UBQLN2

Abstract

Frontotemporal dementia (FTD) and amyotrophic lateral sclerosis (ALS) are believed to represent the different outcomes of a common pathogenic mechanism. However, while researchers have intensely studied the involvement of motor neurons in the ALS/FTD syndrome, very little is known about the function of hippocampal neurons, although this area is critical for memory and other cognitive functions. We investigated the electrophysiological properties of CA1 pyramidal cells in slices from 1 month-old UBQLN2P497H mice, a recently generated model of ALS/FTD that shows heavy depositions of ubiquilin2-positive aggregates in this brain region. We found that, compared to wild-type mice, cells from UBQLN2P497H mice were hypo-excitable. The amplitude of the glutamatergic currents elicited by afferent fiber stimulation was reduced by ~50%, but no change was detected in paired-pulse plasticity. The maximum firing frequency in response to depolarizing current injection was reduced by ~30%; the fast afterhyperpolarization in response to a range of depolarizations was reduced by almost 10 mV; the maximum slow afterhyperpolarization (sAHP) was also significantly decreased, likely in consequence of the decreased number of spikes. Finally, the action potential (AP) upstroke was blunted and the threshold depolarized compared to controls. Thus, synaptic and intrinsic excitability are both impaired in CA1 pyramidal cells of UBQLN2P497H mice, likely constituting a cellular mechanism for the cognitive impairments. Because these alterations are detectable before the establishment of overt pathology, we hypothesize that they may affect the further course of the disease.

Keywords: ALS/dementia; CA1; glutamate; pyramidal cell.

Figures

Figure 1

Glutamatergic current is reduced in UBQLN2P497H mice. (A) Voltage clamp recordings (Vh = −70 mV) of hippocampal pyramidal cells in slice from a control (black traces) and an amyotrophic lateral sclerosis (ALS; red traces) mouse. (B) Summary of the synaptic currents recorded at −70 mV in 10 control and 16 UBQLN2P497H cells in response to extracellular stimulations of increasing magnitude. (C,D) No difference was detected in paired pulse ratio between cells from WT and UBQLN2P497H mice (Vh = −70 mV; 9 WT and 12 UBQLN2P497H cells).

Figure 2

Membrane properties of CA1 pyramidal cells of control and UBQLN2P497H mice. No differences were detectable between the basic membrane properties of UBQLN2P497H and WT pyramidal neurons. Bar charts show the summary data for resting membrane potential (A; measured upon breaking into the cell; 12 WT and 11 UBQLN2P497H cells), membrane capacitance (B, 12 and 13 cells, respectively) and input resistance (C, 12 and 13 cells, respectively). Internal solution for all these measurements was K-gluconate based.

Figure 3

Membrane responses to hyperpolarizing current injections are similar in UBQLN2P497H and control cells. (A) Voltage traces in response to negative current injections adjusted to produce the same peak hyperpolarization (−120 mV; for the traces shown here, the injected current was: −450 pA for WT and −250 pA for UBQLN2P497H, 1 s) in a cell from a WT (black trace) and one from a UBQLN2P497H mouse (red trace). (B) Cells from UBQLN2P497H show a small decrease in the peak to steady state hyperpolarization (the sag ratio was 0.88 ± 0.008 in UBQLN2P497H and 0.84 ± 0.01 in control mice, 14 and 12 cells, respectively, *p = 0.007, Wilcoxon-Mann-Whitney). (C) This difference, however did not significantly modify anode break excitability.

Figure 4

Firing frequency is reduced in UBQLN2P497H cells. (A,B) Voltage responses to 1 s long depolarizing current injections (200 pA) in a cell from control mouse (A) and one from a UBQL mouse. (B,C) Input/output function obtained in 12 cells from WT and 12 from UBQLN2P497H mice. Note the reduction in maximum firing frequency in UBQLN2P497H cells. (D) Instantaneous frequency plotted against spike number. For this plot the current injection in each cell was the largest that that did not induce depolarization block. Note the difference both in maximum frequency and in the steady-state values.

Figure 5

The fast and slow AHPs are reduced in cells from UBQLN2P497H mice. (A) The fast afterhyperpolarization (fAHP) amplitude (measured as the voltage change between the peak and trough of the first Action potential (AP), see inset) elicited with 400 pA (p = 0.0496, Wilcoxon-Mann-Whitney) or 700 pA (p = 0.01, Wilcoxon-Mann-Whitney) current injections was consistently smaller in UBQL recordings. (B) When comparing the first and last spike within an AP train, the fAHP reduction was considerably larger in UBQLN2P497H mice, further impairing the ability of these neurons to fire at high frequencies (*p = 0.03, Wilcoxon-Mann-Whitney). (C) The slow AHP recorded at the end of an AP train (inset; 700 pA current injection) was markedly reduced in UBQLN2P497H mice for any current injection ≥800 pA, as summarized in panel (D). Traces in (C) were obtained with 700 pA current injections.

Figure 6

AP upstroke is blunted in UBQLN2P497H cells. (A) APs recorded in cells from control (black trace) and UBQLN2P497H (red traces) mice. Resting potential was −70 mV, 200 pA current injections. (B) Phase plots of the APs in (A). Note the depolarized threshold (dotted lines) and smaller maximum rate of rise in UBQLN2P497H cells. (C,D) Summary bar charts of showing the differences in threshold and the maximum rate of rise (*p = 0.03, Wilcoxon-Mann-Whitney) in 12 controls and 14 UBQLN2P497H cells.