Ru24969-treated mice showed repetitive circling, memory impairment, and anxiety

In previous studies, pharmacological OCD models were mainly derived from rats [10, 11]. We established OCD mouse models by administration of RU24969, 8-OH-DPAT, and MCPP to complement these rat models.

The effects of RU24969 treatment on mouse behavior were summarized in Fig. 1. We examined circling, self-grooming, spray-induced grooming, and MBT in RU24969-treated mice. Acute treatment with RU24969 increased both bouts (t(20) = 2.68, p < 0.05) and duration (t(20) = 2.69, p < 0.05) of circling, compared with saline-treated mice (Fig. 1a). RU24969-treated mice showed repeated circling around the edges of the open field, whereas saline-treated mice moved in random directions. These results suggest that RU24969 mice exhibited repetitive, stereotyped behavior. Probably due to the long-term repeated movement in circles in the cage, and thus ignoring the marbles and reducing the grooming, RU24969-treated mice did not show increased OCD-like behavior in MBT and grooming tests. Instead, the numbers of buried marbles and grooming bouts were decreased (Fig. 1b–d) (self-grooming, bouts: t(22) = 4.08, p < 0.001; self-grooming, duration: t(22) = 5.44, p < 0.0001; induced-grooming, bouts: t(22) = 3.84, p < 0.001; induced-grooming, duration: t(22) = 3.89, p < 0.001; MBT: t(22) = 4.82, p < 0.0001; Fig. 1b–d).

Fig. 1

Ru24969-treated mice showed repetitive circling behavior, memory impairments, and anxiety. Comparison between RU24969- and saline-treated mice regarding bouts and duration in the circling behavior (a), self-grooming (b), and spray-induced grooming (c) tests; the number of buried marbles in the MBT (d); bouts and duration in the OFT (e); and exploration time and discrimination index in the NOR (f). Data are expressed as the mean ± SEM. Unpaired t-test, except for the exploration time in the NOR using the paired t-test: **p < 0.01, ***p < 0.001, ****p < 0.0001

The OFT is frequently used to measure anxiety and locomotion levels in rodents [20]. In the OFT, the total time in the inner zoom was reduced in the RU24969-treated group (t(22) = 3.49, p < 0.01), and the total distance was increased (t(22) = 5.69, p < 0.0001; Fig. 1e), indicating that RU24969-treated mice exhibited anxiety and hyperlocomotion. In the NOR test, the control group mice showed a preference for novel objects in the test session (t(9) = 4.35, p < 0.01), whereas the RU24969-treated mice did not show this preference (t(11) = 0.99, p = 0.34; Fig. 1f). The discrimination index was significantly lower in the RU24969 group (t(20) = 4.01, p < 0.001; Fig. 1f). These results suggest that RU24969 administration led to the impairment of recognition memory.

MCPP induced repetitive self-grooming in mice

Previous studies reported that MCPP induced repetitive self-grooming in rats [16, 18]. Could the same or distinct behavioral phenotypes be induced in mice? To answer this question, we performed behavioral tests in MCPP-treated mice. Our results showed that, similar to the rats, MCPP-treated mice exhibited increased self-grooming (Fig. 2a–f). Both grooming bouts (t(22) = 3.33, p < 0.01) and duration (t(22) = 3.31, p < 0.01; Fig. 2b) were increased, suggesting that MCPP could induce over-grooming in mice. However, MCPP-treated mice did not show a difference in other behavioral tests than saline-treated mice (Fig. 2a, c–f).

Fig. 2

MCPP induced repetitive self-grooming behavior but no other behavioral abnormalities in mice. Bouts and duration in the circling behavior (a), self-grooming (b), and induced-grooming (c) tests; the number of buried marbles in the MBT (d), bouts and duration in the OFT (e), and exploration time and discrimination index in the NOR (f) comparing MCPP- with saline-treated control mice. Data are expressed as the mean ± SEM. Unpaired t-test, except for the exploration time in the NOR using the paired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

8-OH-DPAT-treated mice exhibited excessive grooming and memory impairment but not anxiety

8-OH-DPAT is a commonly used reagent to induce OCD-like behavior in the rat [10]. When we applied it to mice (Fig. 3a–f), we found that the duration of spray-induced grooming was significantly increased (t(23) = 2.22, p < 0.05), although the number of bouts was not different compared with the saline-treated group (t(23) = 1.19, p = 0.25; Fig. 3c). Surprisingly, in self-grooming test, 8-OH-DPAT-treated mice did not exhibit over-grooming (bouts: t(22) = 0.44, p = 0.67; duration: t(22) = 0.36, p = 0.72; Fig. 3b). Indeed, self-grooming and spray-induced grooming represent very different grooming forms; anomalies in one form were not necessarily accompanied by deficits in the other [25]. We found no significant alteration in circling, either (bouts: t(21) = 1.04, p = 0.31; duration: t(21) = 1.85, p = 0.08; Fig. 3a).

Fig. 3

Excessive grooming and memory impairment, but no anxiety, in 8-OH-DPAT-treated mice. Bouts and duration in the circling behavior (a), self-grooming (b), and induced-grooming (c) tests; the number of buried marbles in the MBT (d), bouts and duration in the OFT (e), and exploration time and discrimination index in the NOR (f) comparing 8-OH-DPAT mice with saline-treated mice. Data are expressed as the mean ± SEM. Unpaired t-test, except for the exploration time in the NOR using the paired t-test: *p < 0.05, **p < 0.01, ****p < 0.0001

Notably, the number of buried marbles was decreased in the 8-OH-DPAT-treated group (t(22) = 8.86, p < 0.0001; Fig. 3d). Besides, 8-OH-DPAT also caused reduced locomotion as shown in the OFT; the total distance was lower in the 8-OH-DPAT group than in the control group (t(21) = 6.67, p < 0.0001), whereas the total time in the inner zone was increased (t(21) = 2.86, p < 0.01; Fig. 3e). This suggests the motion inhibition and lack of anxiety in 8-OH-DPAT-treated mice. Noteworthy, 8-OH-DPAT-treated mice also presented memory impairments in the NOR; the discrimination index of 8-OH-DPAT-treated mice was significantly lower than that of the saline-treated group (t(10) = 2.42, p < 0.05; Fig. 3f).

Increased c-fos expression in the OFC, ACC, PrL, IL, CPu, NAc, hypothalamus, BSTLD, and IPACL in RU24969-treated mice

The distinct behavioral traits observed above suggest that each model might involve the activation of different brain regions. To validate this hypothesis, we used c-fos expression to indicate the neuronal activation in different brain regions. We found that RU24969 induced higher c-fos expression in many brain areas, including OFC, ACC, PrL, IL, CPu, nucleus NAc, hypothalamus, BSTLD, and IPACL (Fig. 4a, b, Additional file 1: Fig. S1). Noteworthy, dysfunctions of the OFC, including the lateral orbital cortex (LO), ventral orbital cortex (VO), and medial orbital cortex (MO), and ACC, have been implicated in the etiology of OCD in previous studies [11]. Consistently, all of these regions showed increased c-fos expression in our RU24969-treated mice (LO + VO: t(16) = 4.30, p < 0.001; MO: t(16) = 2.78, p < 0.05; Cg1: t(16) = 3.42, p < 0.01; Cg2: t(16) = 3.11, p < 0.01). The OFC and ACC are involved in many important neural functions, such as decision making, planning, inhibition of responses, and error monitoring. Indeed, in OCD patients, almost all of these functions are impaired [5, 11]. The PrL and IL, components of the medial prefrontal cortex (mPFC), showed increased c-fos expression in the RU24969-treated group (PrL: t(16) = 4.15, p < 0.001; IL: t(16) = 2.30, p < 0.05), which may suggest the dysfunction in recognition memory, attention, and decision making [26]. In the CPu, c-fos was also highly expressed in RU24969-treated mice, whereas hardly detectable in saline-treated mice (t(16) = 6.92, p < 0.0001). This was consistent with the findings that CPu’s functions in learning, memory, action selection, and goal-directed actions were impaired in OCD patients [5]. The accumbens nucleus, shell (AcbSh) and accumbens nucleus, core (AcbC) are two regions of the NAc and participate in the regulation of reinforcement learning [27]. They also exhibited higher c-fos expression in RU24969-treated mice (AcbSh: t(16) = 5.16, p < 0.0001; AcbC: t(16) = 3.94, p < 0.01). The BSTLD, a brain region related to the regulation of anxiety and acute stress [28], showed increased c-fos expression in RU24969-treated mice (t(16) = 5.73, p < 0.0001). Furthermore, we found that c-fos was elevated in the IPACL in RU24969-treated mice (IPACL: t(16) = 10.63, p < 0.0001; IL: t(16) = 3.94, p < 0.01). The IPACL receives projections from the amygdala [29], but its functions related to OCD have not been clarified. Both control and RU24969-treated mice showed robust c-fos expression in the hypothalamus, a brain region that responds to multiple stimulations including stress and fear [30, 31], but the c-fos level was significantly higher in the RU24969-treated group (t(16) = 6.93, p < 0.0001).

Fig. 4

Mapping c-fos expression in RU24969- and MCPP-treated mice. Representative immunofluorescence micrographs depicting c-fos expression in the OFC, PrL, IL, ACC, CPu, NAc, hypothalamus, BSTLD, and IPACL in RU24969-, MCPP- and saline-treated mice (Scale bars: 100 μm)

Neuronal activity is not significantly increased in MCPP-treated mice

Next, we investigated MCPP-induced c-foc expression in mice brains. Surprisingly, after MCPP injection, we found that limited brain areas expressed c-fos with a weak signal. When we compared it with the saline group, there were no statistically significant differences (Fig. 5b). No detectable changes were seen in the cortical brain areas that we examined.

Fig. 5

Quantification of the c-fos expression in RU24969- and MCPP-treated mice. a In RU24969-treated mice, c-fos expression levels were increased in the OFC, PrL, IL, CPu, NAc, hypothalamus, BSTLD, and IPACL. b There were no significant differences between MCPP- and saline-treated mice regarding the c-fos expression in the OFC, PrL, IL, CPu, NAc, hypothalamus, BSTLD, and IPACL. Data are expressed as the mean ± SEM. Unpaired t-test: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Increased c-fos expression in the ACC, PrL, IL, OFC, AcbSh, and hypothalamus in 8-OH-DPAT-treated mice

The brain regions expressing c-fos after 8-OH-DPAT administration were similar but not identical to those after RU24969 administration. In 8-OH-DPAT-treated mice, higher c-fos expression were detected in the ACC, PrL, IL, OFC, AcbSh, and hypothalamus (Cg1: t(16) = 5.99, p < 0.0001; Cg2: t(16) = 6.50, p < 0.0001; PrL: t(16) = 5.02, p < 0.0001; IL: t(16) = 5.25, p < 0.0001; LO + VO: t(16) = 5.13, p < 0.0001; MO: t(16) = 4.50, p < 0.001; AcbSh: t(16) = 6.32, p < 0.0001; hypothalamus: t(16) = 7.15, p < 0.0001; Fig. 6a, b). These results indicate that the excessive spay-induced grooming observed in 8-OH-DPAT-treated mice might be mediated by the activation of these brain areas.

Fig. 6

The c-fos expression in the brain of 8-OH-DPAT-treated mice. a Representative immunofluorescence micrographs depicting the c-fos expression levels in the ACC, PrL, IL, OFC, AcbSh, and hypothalamus. Expression levels were increased in 8-OH-DPAT-treated mice compared with saline-treated control mice, whereas c-fos expression levels in the Acbc and CPu were not significantly different (Scale bars: 100 μm). b Quantification of the c-fos expression in the OFC, PrL, IL, ACC, CPu, NAc, and hypothalamus in 8-OH-DPAT and saline-treated mice. Data are expressed as the mean ± SEM. Unpaired t-test: ***p < 0.001, ****p < 0.0001

OCD mouse models exhibited a distinct pattern of the activated brain regions

To more intuitively compare the changes in the activation of brain regions in the three OCD mouse models, we constructed a heatmap of c-fos expression (Fig. 7). The heatmap clearly illustrated the distinct pattern of the activated brain regions, in which increased c-fos were detected in the OFC, ACC, PrL, IL, CPu, NAc, and hypothalamus in RU24969-treated mice, and the OFC, ACC, PrL, IL, AcbSh, and hypothalamus in 8-OH-DPAT-treated mice. In contrast, no significant increase was detected in our interested brain regions in MCPP-treated mice.

Fig. 7

Heatmap of the c-fos expression in the OCD models. The color range represents a logarithm of 2 for the ratio of each mouse to the corresponding control average \(\left( \frac}\,}\,}\,}}}}\,}}}} \right)\). Red indicates high expression