Neuronal deletion of CaV1.2 is associated with sex-specific behavioral phenotypes in mice

Abstract

The gene CACNA1C, which encodes the pore forming subunit of the L-type calcium channel Ca_V1.2, is associated with increased risk for neuropsychiatric disorders including schizophrenia, autism spectrum disorder, major depression, and bipolar disorder. Previous rodent work identified that loss or reduction of Ca_V1.2 results in cognitive, affective, and motor deficits. Most previous work has either included non-neuronal cell populations (haploinsufficient and Nestin-Cre) or investigated a discrete neuronal cell population (e.g. CaMKII-Cre, Drd1-Cre), but few studies have examined the effects of more broad neuron-specific deletion of Ca_V1.2. Additionally, most of these studies did not evaluate for sex-specific effects or used only male animals. Here, we sought to clarify whether there are sex-specific behavioral consequences of neuron-specific deletion of Ca_V1.2 (neuronal Ca_V1.2 cKO) using Syn1-Cre-mediated conditional deletion. We found that neuronal Ca_V1.2 cKO mice have normal baseline locomotor function but female cKO mice display impaired motor performance learning. Male neuronal Ca_V1.2 cKO display impaired startle response with intact pre-pulse inhibition. Male neuronal Ca_V1.2 cKO mice did not display normal social preference, whereas female neuronal Ca_V1.2 cKO mice did. Neuronal Ca_V1.2 cKO mice displayed impaired associative learning in both sexes, as well as normal anxiety-like behavior and hedonic capacity. We conclude that deletion of neuronal Ca_V1.2 alters motor performance, acoustic startle reflex, and social behaviors in a sex-specific manner, while associative learning deficits generalize across sexes. Our data provide evidence for both sex-specific and sex-independent phenotypes related to neuronal expression of Ca_V1.2.

Figure 1

Loss of neuronal Ca_V1.2 causes impaired motor performance learning in female but not male mice. (a) No differences in locomotor activity were detected in distance traveled in the open field test between neuronal Ca_V1.2 cKO mice and CTRL littermates. (b) Female neuronal Ca_V1.2 cKO mice display a shorter latency to fall than CTRL littermates on the accelerating rotarod. No differences in rotarod performance were detected between CTRL and cKO males. Data are expressed as mean ± s.e.m, using EthoVision 14, www.noldus.com/ethovision-xt.

Figure 2

Ca_V1.2 cKO mice display slower gait but no deficit in cerebellar-dependent associative motor learning on the Erasmus Ladder task. (a) Neuronal Ca_V1.2 cKO mice had fewer missteps than CTRL littermates on day 1 which resolved on days 2–4. (b) We observed a main effect of day in the use of long steps, with no sex or genotype-dependent effects. Therefore, on average all groups use more long steps over the course of training, representing normal gait adaptation. (c) Neuronal Ca_V1.2 cKO mice had a longer average step time compared to CTRL littermates. Males also had a longer average step time compared to females. (d) All groups learn to jump over a displaced peg in response to a tone cue, shown as a decrease in step time between days 5 and days 6–8. Post hoc testing revealed that CTRL mice had longer post-perturbation step times on day 5 compared to cKO mice, but this was not observed on days 6–8, suggesting that all groups successfully learn to associate the tone with the displaced peg, representative of cerebellum-dependent associative learning. No sex-dependent effects were detected. Data are expressed as mean ± s.e.m, using ErasmusLadder 2.0.214, www.noldus.com/erasmusladder.

Figure 3

Neuronal Ca_V1.2 cKO mice have a deficit in associative learning. (a) Neuronal Ca_V1.2 cKO mice do not learn to leave on the light cue whereas CTRL mice do. We found a day × genotype interaction, and post hoc testing showed that cKO mice differed from CTRL mice on day 4. There were no sex-dependent effects. (b) Neuronal Ca_V1.2 cKO mice do not learn to avoid the air cue with trial initiation but CTRL mice do. There was a day × genotype interaction, and post hoc testing showed that cKO mice differed from CTRL mice on day 4. (c) There were no differences in how frequently mice walked onto the ladder before any cue was given, suggesting that mice did not display impulsive initiation of the task. Data are expressed as mean ± s.e.m, using ErasmusLadder 2.0.214, www.noldus.com/erasmusladder.

Figure 4

Male neuronal Ca_V1.2 cKO mice display impaired startle response. (a) Male cKO mice have significantly lower acoustic startle response to the auditory startle pulse compared to CTRL littermates. Female cKO mice do not differ from female CTRL mice in acoustic startle response. (b) All groups display normal prepulse inhibition of the acoustic startle response, and no sex or genotype differences were observed. Data are expressed as mean ± s.e.m, using SR-Lab, www.sandiegoinstruments.com/product/sr-lab-startle-response.

Figure 5

Male Ca_V1.2 cKO mice display impaired social preference on the three-chamber social preference task. (a) Male cKO mice were more active than CTRL littermates during habituation but there were no sex or genotype differences in distance traveled during the preference phase. (b) Male Ca_V1.2 cKO mice do not spend more time with a conspecific mouse compared to a novel object, but all other groups display preference for spending time with a conspecific mouse over a novel object. Female cKO mice also spend less time in social interaction than CTRL female mice. Data are expressed as mean ± s.e.m, using EthoVision 14, www.noldus.com/ethovision-xt.

Figure 6

Neuronal Ca_V1.2 cKO mice exhibit longer latency to float but no other antidepressant-like behaviors. (a) Neuronal Ca_V1.2 cKO mice exhibit a longer latency to float in the forced swim test compared to CTRL littermates. (b) Time immobile in the forced swim test did not differ between groups. (c) Neuronal Ca_V1.2 cKO and CTRL mice both prefer sucrose over water. We observed no group differences in sucrose consumption. Data are expressed as mean ± s.e.m.

Figure 7

Neuronal Ca_V1.2 cKO mice display no differences in anxiety-like behaviors. (a) There were no sex- or genotype-dependent differences in time spent in the closed portions of the elevated zero maze. (b) We observed a sex-dependent difference where female mice spent more time in the periphery of the open field, also called thigmotaxis, compared to male mice. We did not observe any genotype-dependent differences in this task. Data are expressed as mean ± s.e.m, using EthoVision 14, www.noldus.com/ethovision-xt.