Initial values of sucrose test parameters were similar between all groups (Figure
3A) and stress exposure lowered sucrose preference in agreement with other reports
[49, 61, 65]. Stressed mice treated with DS showed no significant change in sucrose preference measured on the 7th and 10th day of stress as compared to control animals (Figure
3E). Administration of the classical antidepressant imipramine resulted in a similar effect. Earlier, we have shown in a model of stress-induced anhedonia that the stress-induced decrease in sucrose preference is paralleled by a reduction in sucrose intake
[46, 53]. In particular, a disruption of synaptic plasticity and pronounced behavioral despair in the forced swim test were observed in mice, which showed a decrease in both sucrose intake and preference but not a reduced sucrose preference alone, which occurs exclusively at the expense of high water intake
[45, 51, 53]. Thus, the partial preclusion of the stress-induced reduction for both sucrose preference and consumption by treatment with imipramine and DS manifests their antidepressant-like activity in our study. Importantly, administration of imipramine and DS did not alter sucrose test parameters in control animals ruling out any possible confounding artifacts for sucrose test measurements which could be related to treatment. Imipramine, used as a reference drug in this experiment, is well documented to counteract the stress-induced decrease in sucrose intake and preference seen in rodents
[51, 65]. Overall, our data suggest that chronic administration of imipramine and DS has the potential to counteract the development of stress-induced anhedonia, i.e. elicits an antidepressant-like activity in the mouse paradigm employed in the present study.
Described above evidences, for antidepressant-like effects of DS in the sucrose test, are in line with the outcome from the forced swim test. While the effects of chronic stress on the latency to the first episode of floating were not significant, DS and imipramine treated mice had significantly higher values of this parameter (Figure
4A). DS administration strongly reduced the duration of floating in the chronic stress experiment suggesting that treatment with DS prevents a stress-induced state of behavioral despair as measured by elevated floating behavior (Figure
[45, 46, 48, 66]. In the current work, these antidepressant like effects observed with DS administration were also demonstrated for treatment with imipramine; earlier, we reported congruent changes in the forced swim test following chronic administration of citalopram
. Coinciding with these results, another insulin sensitizer, rosiglitazone, was reported to reduce immobilization and floating behaviors in mouse tail suspension and forced swim tests respectively
Both treatment with imipramine and DS decreased anxiety scores as shown by increased time spent in the lit compartment in the dark/light box indicating their anxiolytic and anti-stress effects (Figure
4B). Such effects are well documented for imipramine and other tricyclics
. Elevated anxiety was found to parallel anhedonia induction in chronic stress models
[15, 45, 48, 49].
In the present study, the stress-induced loss in body weight does not correlate with an occurrence of anhedonia and depressive-like syndrome in mice. However, it has been shown that antidepressant effects of pharmacological treatment parallel a restoration of body weight
[48, 67]. We did not find such an effect with imipramine and DS treatment (Figure
4C) in the present study but it must be noted that a lack of positive effects on the restoration of body mass in depressed patients is well documented for many antidepressants including fluoxetine and other SSRIs
Treatment with imipramine and DS prevented stress-induced memory impairment in the step-down inhibitory avoidance task (Figure
5A). The latency of step down measured twenty four hours after training session was previously validated as a reliable measure of hippocampus-dependent performance in mice
[38, 39]. Deficits in hippocampus-dependent performance were earlier shown to be a specific feature of stressed anhedonic mice as compared to stressed individuals without a depressive syndrome in various chronic stress paradigms. Treatment with citalopram was shown to rescue contextual fear conditioning in a model of stress-induced anhedonia
[38, 45, 53]. Since imipramine administration also precluded deficits in the step-down avoidance test, beneficial effects of DS on learning can be accounted for by its antidepressant action which is generally accompanied by an improvement in cognitive function in clinical and pre-clinical studies. These data are in line with ameliorative effects of DS on hippocampus- and cortex-dependent learning in step-though and Morris water maze paradigms which this drug exerted under pathological conditions of diverse origins
Our study evidenced a sharp increase in hippocampal gene expression of IGF2, a member of the insulin gene family with known neurotrophic properties, after administration of DS, but not imipramine, in stressed mice (Figure
5B). Utilizing the hippocampi of mice from the same experimental groups (five animals per each group were analyzed), gene expression profiling was performed using Illumina technology (Integragen, Evry, France and Northwestern Chicago University, USA). This study revealed significant effects of DS on the expression of a number of functionally important genes (Strekalova and Malin, in preparation). Therefore, total RNA was isolated using RNeasy Mini Kit (Qiagen, Hilden, Germany). Total RNA samples were hybridized to IlluminaBeadChips (MouseRef-8 v2 Expression BeadChip; Illumina, Inc. San Diego, CA, USA) which were prepared using the IlluminaTotalPrep RNA Amplification kit (Applied Biosystems/Ambion, Carlsbad, CA, USA); the samples were assigned to the chips in random order with the constraint that no two samples from the same group were assigned to the same chip, to avoid confounding of experimental groups with the chips. Microarray data were analyzed using standard analysis procedures which included assessment of the overall quality of array data and statistical evaluation of differentially expressed genes (Integragen, Evry, France). Once the quality of array data was confirmed, the Gene Chip Operating System (Illumina, Inc. San Diego, CA, USA) was used to calculate signal intensities, detection calls, and their associated P values for each transcript on the array. Gene expression was normalized to the expression of the house gene beta-actin, due to its stable expression, and calculated as percent mean of the control group. Differences in gene expression between groups were evaluated using ANOVA followed by Fisher's Least Significant Difference test.
In line with the outcome from mRNA evaluation (Figure
5B), these data revealed a significant increase of IGF2 expression in stressed DS-treated mice (163.1 ± 30.17% from control) and its significant decrease in stressed mice that were not treated (83.0 ± 4.49% from control). This study indicated that stressed DS-treated mice showed a significant expression enhancement of Htra1, HtrA serine peptidase 1 which cleaves IGF-binding proteins (IGFBPs) from IGF1 and IGF2 and activates these factors, in comparison to non-treated mice (144.3 ± 9.16 vs. 119.2 ± 17.6% from control, respectively). Preliminary data showed no such changes of IGF2 in the non-stressed control group. Also, this experiment revealed significant effects of DS on the expression of other elements of the IGF1/IGF2 system in chronically stressed mice, including Htra1, as well as IGF1 and IGF1 receptor, the insulin receptor and several insulin-like growth factor binding proteins. Again, no such changes were detected in non-stressed mice treated with DS. While it is important to study above-mentioned findings with additional methods, they generally support our data on elevated mRNA of IGF2 in the hippocampus of DS-treated stressed mice and suggest this elevation to be a part of systemic changes in IGF1/IGF2 signaling in these animals. A lack of such molecular effects in naïve mice treated with DS might be due to distinct functional states of the IGF1/IGF2 system during stress and resting conditions; whereas, the activation of this signaling might occur as an adaptive mechanism in response to biological challenges.
A comparison of pharmacologically naïve anhedonic versus resilient animals in changes of the above-mentioned elements of IGF1/IGF2 signaling speaks in favor of the latter view. Our studies revealed an intriguing difference in the IGF2 expression between non-treated stressed anhedonic and resilient animals (58.8 ± 2.32% vs. 107.2 ± 8.7% from control, respectively), suggesting the elevated IGF2 to be a correlate of stress resilience while its decrease as a parallel of susceptibility to a depressive-like state. Moreover, the expression of Htra1 was reduced in non-treated anhedonic and essentially increased in the resilient group (91.08 ± 6.16% vs. 148.34 ± 11.16% from control, respectively). Anhedonic and non-anhedonic groups had differential expressions in most of the other above-listed elements of the IGF1/IGF2 signaling system (Strekalova and Malin, in preparation). Thus, the outcome from the gene expression profiling experiment is in line with a suggestion that enhanced expression of IGF2 can mediate resilience to stress-induced anhedonia induced by administration of DS in our study.
IGF2 is widely expressed throughout the brain and is abundant in the hippocampus
. Various challenges such as acute hypoxia, exposure to toxicity stress and cerebral ischemia were shown to induce long-lasting changes in IGF2 expression which is considered to have an important neuroprotective function
[69–71]. IGF2 was recently shown to be an important regulator of hipppocampal neurogenesis in the context of extinction in fear conditioning learning
. Moreover, IGF2 was shown to enhance adult neurogenesis
. Our results evidenced suppressive effects of stress on hippocampal levels of IGF2 demonstrating that chronic stress in mice has a tendency to decrease the content of this neurotrophic factor that might be associated with its above-mentioned role in the regulation of neurogenesis which is inhibited by stress
. In order to immediately address how crucial the role of IGF2 in the development of stress-induced depressive syndrome might be, a chronic intrahippocampal administration of this molecule could be applied with our model. Treatment with DS significantly elevated levels of IGF2 in stressed mice above that of controls. The mechanisms of this effect can be due to an earlier demonstrated DS-induced enhancement of choline content and acetylcholine function in the brain
 since a functional link between this neurotransmitter system and IGF2 is particularly evidenced by increased expression of this neurotrophic factor after choline administration in the hippocampus and frontal cortex
[40, 75, 76]. In these studies, choline supplementation increased levels of IGF2 in the hippocampus and changed expression of its receptors in the septum, it also enhanced IGF2-induced acetylcholine release and cholinergic neurontransmission
. Importantly, there was an increase of choline acetyltransferase activity after DS treatment in rats subjected to a toxic treatment with beta-amyloid peptide-(25–35)
 and elevated IGF2 content in our study on mice were observed two weeks after the termination of a chronic DS administration (Figure
Interestingly, recent results evidenced a critical role of IGF2 in inhibitory avoidance learning as shown in the fear conditioning paradigm
[72, 77] which can additionally explain the beneficial effects of DS on performance in chronically stressed mice in the step-down inhibitory avoidance task (Figure
5A). Taking these data into account and given the fact that hippocampal IGF2 signaling regulates adult neurogenesis in the context of fear extinction learning
 it would be of high interest to assess the expression of IGF2 in the step-down avoidance inhibitory task that is similar to the fear conditioning paradigm form of contextual learning. However, the fact that imipramine did not evoke any effect on IGF2 gene expression while exhibiting both prominent anti-depressant and memory-enhancing effects similar to DS suggest that elevated expression of this molecule cannot be the sole mechanism of antidepressant and memory-enhancing effects observed for DS in the current study.