Region-specific deletions of RIM1 reproduce a subset of global RIM1α(-/-) phenotypes

Abstract

The presynaptic protein RIM1α mediates multiple forms of presynaptic plasticity at both excitatory and inhibitory synapses. Previous studies of mice lacking RIM1α (RIM1α(-/-) throughout the brain showed that deletion of RIM1α results in multiple behavioral abnormalities. In an effort to begin to delineate the brain regions in which RIM1 deletion mediates these abnormal behaviors, we used conditional (floxed) RIM1 knockout mice (fRIM1). By crossing these fRIM1 mice to previously characterized transgenic cre lines, we aimed to delete RIM1 selectively in the dentate gyrus (DG), using a specific preproopiomelanocortin promoter driving cre recombinase (POMC-cre) line , and in pyramidal neurons of the CA3 region of hippocampus, using the kainate receptor subunit 1 promoter driving cre recombinase (KA-cre). Neither of these cre driver lines was uniquely selective to the targeted regions. In spite of this, we were able to reproduce a subset of the global RIM1α(-/-) behavioral abnormalities, thereby narrowing the brain regions in which loss of RIM1 is sufficient to produce these behavioral differences. Most interestingly, hypersensitivity to the pyschotomimetic MK-801 was shown in mice lacking RIM1 selectively in the DG, arcuate nucleus of the hypothalamus and select cerebellar neurons, implicating novel brain regions and neuronal subtypes in this behavior.

Figure 1. Immunofluorescent signal shows regional specificity of the fRIM1/POMC-cre+/R26R-YFP and fRIM1/KA-cre+/R26R-YFP reporter mice

YFP expression in fRIM1/POMC-cre+/R26R-YFP brain tissue from hippocampus (a), cerebellum (b) and hypothalamus (c). YFP and NeuN expression in a fRIM1/POMC-cre+/R26R-YFP hippocampus (d–f). FP expression in fRIM1/KA-cre+/R26R-YFP brain tissue from hippocampus (g), cerebelleum (h) and forebrain (i). YFP and NeuN expression in a fRIM1/KA/R26R-YFP hippocampus (j and k). f and l are merges of d and e and j and k, respectively (green – anti-YFP, red – anti-NeuN; a, h and i, scale bar = 1 mm; b, d–f, h, j and k, scale bar = 500 µm; c, scale bar = 250 µm; ML, molecular layer; GL, granule layer; Th, thalamus; St, striatum).

Figure 2. Cell-specific cre-mediated recombination in fRIM1/POMC-cre+/R26R-YFP and fRIM1/KA-cre+/R26R-YFP cerebellum

(a–c) High magnification of YFP and NeuN expression in the cerebellum of a fRIM1/POMC-cre+/R26R-YFP mouse. (d–f) High magnification of YFP and parvalbumin expression in the cerebellum of a fRIM1/POMC-cre+/R26R-YFP mouse. (g–i) High magnification of YFP and NeuN expression in the cerebellum of a fRIM1/KA-cre+/R26R-YFP mouse. c, f and i are merges of a and b, d and e and g and h, respectively (green – anti-YFP, red – anti-NeuN or anti-parvalbumin as labeled; all scale bars = 250 µm; GL, granule layer; PCL, purkinje cell layer; ML, molecular layer).

Figure 3. Regional specificity of fRIM1/POMC-cre+ and fRIM1/KA-cre+ conditional knockouts by qRT-PCR

Quantitative RT-PCR was used to measure levels of RIM1 transcript in the CA1, CA3 and dentate gyrus regions of the hippocampus as well as cortex and cerebellum in the fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice compared with their littermate controls. RIM1 transcript levels were internally normalized to MAP2 transcript levels and reported as a percentage of normalized RIM1 transcript levels observed in littermate controls (dashed line; *P < 0.05; **P < 0.01).

Figure 4. fRIM1/POMC-cre+ mice exhibit enhanced locomotor response to psychotomimetics

(a) Three-hour locomotor test in which fRIM1/POMC-cre+ mice received an injection of saline, 0.1 mg/kg MK-801 or 0.2 mg/kg MK-801 at the beginning of each hour as marked by the vertical dotted lines. (b) As fRIM1/KA-cre+ mice are hyperactive in novel environments, they were habituated to the 3-h locomotor test over 3 days, receiving only saline injections at the start of each hour (only the first hour is shown). (c) On day 4, fRIM1/KA-cre+ mice underwent the 3-h locomotor test, receiving saline, 0.1 mg/kg MK-801 or 0.2 mg/kg MK-801 at the start of each hour as marked by the vertical dotted line (*P < 0.05 using Student's t-test).

Figure 5. fRIM1/KA-cre+ mice display enhanced locomotion to novel stimuli

(a) fRIM1/POMC-cre+ mice underwent a 2-h test of locomotion in a novel home cage in which lengthwise movement was monitored using photobeams. (b) fRIM1/KA-cre+ mice underwent the same test. (c) Locomotor habituation of the fRIM1/KA-cre+ mice: Over 4 days, the first 10 min of locomotor activity was recorded (*P < 0.05 using Student's t-test).

Figure 6. fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice display normal startle response and prepulse inhibition

(a and d) The average startle response to a 120-dB tone of fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice, respectively, before the start of the prepulse inhibition test. (b and e) Prepulse inhibition test. The percent decrease in the startle response caused by a 4-, 8- or 16-dB prepulse tone given 100 milliseconds before the 120-dB stimulus tone. (c and f) Startle threshold. Average startle response to increasing stimulus intensity delivered in a pseudo random order.

Figure 7. fRIM1/KA-cre+ and fRIM1/POMC-cre+ display normal spatial learning in the MWM

(a) During 8 days of training in the Morris water maze, the distance traveled to find the hidden platform was measured in the fRIM1/POMC-cre+ mice. (b and c) fRIM1/POMC-cre+ probe trial. On day 9, the platform was removed from the pool and the amount of time spent in each quadrant (b) and the number of platform location crossings (c) was recorded. (d) Eight-day training period in Morris water maze of the fRIM1/KA-cre+ mice. (e and f) fRIM1/KA-cre+ probe trial. The platform was removed from the pool and the amount of time spent in each quadrant (e) and the number of platform location crossings (f) was recorded. (g and h) Representative tracings from the MWM probe trial of fRIM1/POMC-cre+ and fRIM1/POMC-cre− mice, respectively (quadrants and platform locations: T, target (west); N, north; S, south; E, east; *P < 0.05, *P < 0.01; n.s., not significant).

Figure 8. Contextual and cued fear conditioning as well as social interaction in the open field are normal in both the fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice

(a and c). Fear conditioning in fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice, respectively. Baseline freezing was measured during the first 2 min in the novel context before tone/footshock delivery. Contextual fear conditioning was measured during a 5-min exposure to the same context the following day. Cued fear conditioning was also measured the following day. Mice were exposed to a different context without the tone for 3 min (pretone), followed by 3 min with the tone playing (tone). (b and d) Social interaction in the open field was measured as the amount of time spent in an interaction zone with either an empty cage (object) or a cage containing a target mouse (social).

Figure 9. Three measures of anxiety are normal in both the fRIM1/POMC-cre+ and fRIM1/KA-cre+ mice

(a and b) Dark/light. The amount of time spent in the dark versus light zones is reported. (c and d) Elevated plus maze. The amount of time spent in the open versus closed arms is reported. (e and f) Open field. The amount of time spent in the center versus periphery of the open field box is reported.