The Potential Role of Very Small Embryonic-Like Stem Cells in the Neuroinflammation Induced by Social Isolation Stress: Introduction of a New Paradigm

Background: Lack of social contacts could induce psychiatric features and lead to various behavioral and neurochemical abnormalities in rodents. Social isolation stress (SIS) is a valid paradigm of depressive- and anxiety-like behaviors in animals. It has demonstrated that psychiatric disorder could affect the peripheral blood population of very small embryonic-like stem cells (VSELs). The aim of the current study is to evaluate the role of VSELs in behavioral impairments induced by SIS through neuroinflammation in mice. Methods: Behavioral experiments were evaluated by using forced swimming test (FST), open field test (OFT), and splash test in male NMRI mice. In addition, plasma and bone marrow samples, as well as hippocampus, were collected to evaluate the population of VSELs, nitrite level, and inflammatory cytokines by using flow cytometry and ELISA. Results: Behavioral tasks showed that SIS could induce depressive- and anxiety-like behaviors in mice. Data obtained from flow cytometry showed that VSELs significantly increased in socially isolated animals in bone marrow, peripheral blood, and hippocampus. Also, TNF-α, IL-1β, and IL-6 significantly increased in hippocampal and plasma samples in socially isolated animals. Correlation analysis indicated that mice with higher VSELs counts have better results in behavioral tasks, and lower pro-inflammatory cytokines as well as nitrite level in mice. Conclusion: VSELs could be used as a biological marker to enhance diagnostic accuracy as well as predicting the prognosis. Also, increment in the VSELs counts might decrease the neuro-inflammation and subsequently improve the behavioral impairments induced by SIS. is involved


Introduction
It has been well documented that lack of social contacts could induce psychiatric features and lead to various behavioral and neurochemical abnormalities in rodents [1]. The social isolation stress (SIS), as a chronic stressor, is a valid paradigm, which has been used as an animal model to study neurobehavioral changes [2,3]. It has been reported that social isolation could induce depressive-and anxiety-like behavior as well as decreasing seizure threshold through various mechanism including opioid system, cannabinoid system, nitrergic system, and neuro-inflammation [3][4][5][6]. Based on clinical studies and psychiatric guidelines, the diagnosis of psychiatric disorders is limited to clinical criteria [7].
Biomarkers, as a measurable characteristic factor, have been found to increase the accuracy of both diagnosis and prognosis in various fields of medicine; however, few studies have been focused on biomarkers in the psychiatric disorders. Past studies have recommended using some biomarkers such as C-reactive protein, cytokines, neoprotein, serum S100B, and isoprostanes in the major depressive disorder and bipolar disorder [7].
Interestingly, recent studies have demonstrated that the population of stem cells in the peripheral blood could be affected in an individual diagnosed with a psychiatric disorder.
Over the past decade, Very Small Embryonic-Like Stem Cells (VSELs) have been discovered firstly in mouse bone marrow [8]. Recent clinical studies have shown that psychiatric disorder could affect the peripheral blood population of VSELs [9][10][11]. These studies suggested that the individuals diagnosed with the first psychotic episode had higher VSELs counts in their peripheral blood samples in comparison to control groups [12]. In addition, a related study has been shown that the population of VSELs in peripheral blood samples could be correlated with the duration of bipolar disorder [9].
Past reports demonstrated that VSEL isolated from murine are negative for lineage markers, negative for CD45, and positive for Sca-1 antigen in mice [13,14]. Although previous studies have shown that VSELs could be detected in various adult organs in mice, the maximum numbers of VSELs have been observed in the brain. It has been well documented that stem cells are expected to egress from the relatively quiescent state under stress conditions [15]. Activator signals such as cytokine release from injured tissues could induce gene activation in these cells and subsequently initiate the cell divisions to regenerate injured tissues and thereby maintain homeostasis in the body organs [15,16]. Interestingly, both clinical and experimental studies have reported that activated VSELs could be mobilized from the bone marrow into the peripheral blood under stress conditions such as stroke, spinal cord injury, myocardial infarction, post-cytotoxic treatments, etc. [17][18][19][20][21][22][23]. In addition, it has been suggested that chronic inflammation could significantly increase the process of mobilization of VSELs from the bone marrow to peripheral blood [11].
Although these studies suggested that VSELs could be used to evaluate the clinical progress of psychiatric diseases, no further studies have been established to investigate the exact role of VSELs in such diseases. In this regard, assessing the role of VSELs or its underlying mechanisms in psychiatric disorders is prior to use of these cell as a prognostic or as a diagnostic factor. Since understanding the role of VSELs in psychiatric diseases could present the potential application of VSELs as a new novel cell therapy for treating psychiatry disorders.
Based on previous publications, we hypothesized that chronic neuro-inflammation induced by SIS could be a trigger to provoke cell division of VSELs and increase the VSELs counts.
Also, in this study, we hypothesized that increasing VSELs counts could have a protective role against neuroinflammation and improve behavioral impairments induced by SIS. In the current project, we used SIS as a valid model of depressive-and anxiety-like behavior in mice to evaluate the alteration of VSELs levels in bone marrow, peripheral blood, and hippocampus samples in socially isolated animals compared to normal conditioned animals. Also, we assessed the levels of pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6 as well as nitrite levels in both plasma and hippocampus and their correlations with VSELs in socially isolated animals. Taken all together, the aim of the current study is to evaluate the role of VSELs in behavioral impairments induced by SIS through neuroinflammation in mice.

Animals and housing conditions
Total 60 male NMRI mice aged 21-25 days and weighing 10-12 g were used in this study. Animals were housed in two different conditions including social condition (SC, n = 30) and isolated condition (IC, n = 30). All animals were kept under standard laboratory conditions i.e. temperature: 22 ± 2 °C, humidity: 50 ± 10%, 12-h light-dark cycle, and ad-libitum access to food and water for a period of 5 weeks. Socially conditioned mice were placed in Plexiglas boxes (25 cm × 25 cm × 15 cm) (6 mice per cage) and IC animals were placed individually in Plexiglas boxes (24 cm × 17 cm × 12 cm) [4,24]. In order to diminish handling and social interaction cages of IC animals were cleaned weekly by the same experimenter. All behavioral tasks were carried out between 10:00 a.m. and 02:00 p.m. In this paradigm, grooming activity time, as an indirect measure of palatable solution intake, was evaluated. 10% sucrose solution was sprayed on the dorsal coat of animals.
The total grooming activity time was recorded during 5 min after the sucrose vaporization [28,29]. Grooming activity consists of nose/face grooming, head washing, and body grooming.

Open field test
Open-field test was done just before the FST to assess the locomotor activity of animals [2,30]. OFT was used to exclude the possibility that changes in immobility time are not the result of modifications in motor activity and also to interpret the locomotor activity in response to SIS [31]. The apparatus of OFT was made of white opaque Plexiglas (50 cm × 50 cm × 30 cm) and was softly illuminated. Each mouse was placed gently on the central area (30 cm × 30 cm) and behaviors were documented by a camera in a 5 min period and were analyzed by Ethovision software version 8 (Noldus, Netherlands). The apparatus was cleaned with 70% ethanol after each experiment. The distance moved (horizontal activity) and the numbers of rearing (vertical activity) and time spent in the central zone (central activity) were evaluated.

Hole-board test
Hole-board test is considered as a reliable trial to assess the anxiety-like behaviors in the rodents [32]. The apparatus was made of a white Plexiglas square (50 cm × 50 cm with 16 equally sized holes (3 cm in diameter) and was located 50 cm above the floor. Mice were placed in the center of the board, and the number of head-dips was calculated in a 5-min period by an experimenter.

Assessment of inflammatory cytokines
In this study, for evaluating the molecular assessment in target tissue we applied the following steps. In hippocampus samples, after sacrificing animals in all our experimental groups by decapitation under ether anesthesia, the hippocampal formation was separated and then homogenized based on our previous studies [33]. To measure circulating cytokine levels, the mixed venous-arterial blood was collected in heparinized containers and centrifuged. Pooled plasma from socially isolated and normal animals was stored at

Nitrite Assay
Nitrite levels of the hippocampus and plasma were measured according to our previous studies [35,36]. Based on the Griess reaction, the colorimetric assay was done to assess the nitrite concentration. Briefly, 100 µL of samples were mixed with 100 µL Griess reagent following 10 min of incubation at room temperature, absorbance was measured at 540 nm in an automated plate reader. Nitrite concentration was evaluated by reference to a standard curve of sodium nitrite (Sigma, USA) and normalized to the mg protein of each sample.

Preparation of tissues
In hippocampus samples, after sacrificing animals in all our experimental groups, the hippocampal formation was separated and then homogenized based on our previous studies [4,37]. The homogenized samples immediately were frozen by liquid nitrogen and stored in -80 degrees C. Samples for bone marrow cell analysis were obtained by removing the tibiae and femurs from mice and then flushing out from these bones with RPMI 1640 medium (serum-free) under sterile condition. Then, we crashed the bones, and the remaining bone tissue specimens were washed twice with the solution as mentioned earlier. All extracted solutions were labeled and collected in a falcon for each mouse. In this step, 70 µm nylon cell strainer were used to remove debris. Then the samples were centrifuged at 500 × g for 10 min at 4 °C, and the supernatants were discarded. Finally, the cells were resuspended in 3 ml RBC lysis buffer (for each mouse) and incubated for 10 min at room temperature, and then washed cells by adding 47 ml of RPMI 1640 medium with 2% FBS. In the final step, we centrifuged cells again at 500 × g for 10 min at 4 °C, and then discarded the supernatant. The cell pellet resuspended with 0.5 ml of RPMI 1640 medium with 2% FBS. Finally, the cells were resuspended in 1 mL of PBS-containing 2%

Statistical analysis
Correlation tests and T-test analyses were used to analyze the data in the current study (GraphPad Prism version 7). P value < 0.05 was the critical criterion for statistical significance. Also, the sample size was calculated with G*Power software version 3, considering the power of the study reaching 0.8 and α = 0.05.

Results
At the first part of this study, we evaluated the effect of SIS on some behavioral tasks including Forced swimming test (FST), splash test, and Open-field test (OFT), and Holeboard test (HBT). T-test analysis revealed that SIS could induce depressive-like behavior by increasing immobility time of FST (t = 18.74, df = 58, P < 0.001, Fig. 1A), and also decreasing grooming activity of splash test (t = 18.67, df = 58, P < 0.001, Fig. 1B) in comparison to social condition animals. Also, locomotor activity was assessed by using OFT. Results obtained from t-test showed that distance moved (t = 5.57, df = 58, P < 0.001, Fig. 1C), the number of rearings were increased (t = 8.07, df = 58, P < 0.001, Fig. 1D), and time spent in central zone of OFT was decreased in socially isolated animals in comparison to intact mice (t = 5.58, df = 58, P < 0.001, Fig. 1E). In addition, HBT showed that SIS could decrease the number of head-dips in comparison to normal animals (t = 8.20, df = 58, P < 0.001, Fig. 1D).
In the next step, we counted the population of very small embryonic-like stem cells (VSELs) in bone marrow, peripheral blood, and hippocampus samples in both socially isolated and normal animals. Based previous studies, VSELs were identified by flow cytometry as cells with the appropriate size of 2-10 µm (panels A and B, Fig. 2) and lineage negative, SCA-1 positive, and CD45 negative phenotype (panels C-H, Fig. 2). In this study, we estimated that the mean population of VSELs in bone marrow was 0.062% ±0.024 in normal animals and in socially isolated animals the mean population of VSELs in bone marrow was 0.191%±0.06. T-test analysis showed that socially isolated animals had a significantly higher level of VSELs in their bone marrow in comparison to normal mice (t = 10.83, df = 58, P < 0.001, Fig. 2I). In addition, data obtained from peripheral samples showed that the mean population of VSELs was 0.021%±0.014 in normal animals and was 0.034%±0.012 in socially isolated animals. As depicted in Fig. 2J, VSELs counts were significantly higher in socially isolated animals in peripheral blood samples in comparison to normal mice (t = 3.62, df = 58, P < 0.001, Fig. 2J). Also, in hippocampus samples, we observed that the mean population of VSELs was 2.05%±0.18 and 2.46%±0.4 in normal and socially isolated animals, respectively. T-test analysis showed that socially isolated animals had significantly higher levels of VSELs in the hippocampus in comparison to normal mice (t = 5.08, df = 58, P < 0.001, Fig. 2K). In addition, we used correlation study between VSELs counts in bone marrow samples and peripheral blood samples to clarify the mobilization of VSELs between bone marrow and peripheral blood. As illustrated in Fig. 3, the analysis revealed a direct correlation between the population of VSELs in the bone marrow and peripheral blood samples (P < 0.001, r = 0.8384, and 95% confidence interval = 0.6851 to 0.9206). These results strongly suggested the mobilization of VSELs between bone marrow and peripheral blood in SIS.
On the other hand, we evaluated the level of cytokines and nitrite level in both plasma and hippocampus in socially isolated and normal mice. Our results revealed that socially isolated animals had significant higher plasma levels of TNF-α (t = 14.10, df = 58, P < 0.001, Fig. 4A), IL-1β (t = 8.52, df = 58, P < 0.001, Fig. 4C), and IL-6 (t = 9.98, df = 58, P < 0.001, Fig. 4E) as well as nitrite levels (t = 18.74, df = 11.69, P < 0.001, Fig. 4G Table 1 shows the correlation analysis between the VSELs population of bone marrow, peripheral blood, as well as hippocampus and other behavioral or molecular assessments in both normal and socially isolated animals. Correlation test showed that data obtained from the behavioral tasks or molecular assessments had not any significant correlation with VSELs population of bone marrow in normal mice (P > 0.05, Table 1). In addition, the same analysis was used to clarify the correlations between behavioral tasks or molecular assessments and VSELs counts in peripheral blood samples. As previous results, the analysis failed to show any significant correlation between VSELs counts of peripheral blood samples and all behavioral tasks or molecular assessments (P > 0.05, Table 1). Also, we evaluated the correlation analysis between VSELs counts of the hippocampus and other factors. As shown in Table 1, correlation analysis failed to show any significant correlation between all behavioral tasks (P > 0.05) and hippocampal VSELs population in normal animals. Moreover, no significant correlation was observed between hippocampal levels of VSELs and TNF-α, IL-1β, IL-6, as well as nitrite (P > 0.05).
In the next step, we evaluated the correlation between VSELs population in the bone marrow and behavioral tasks as well as molecular assessments in socially isolated animals. Correlation test showed that the VSELs counts directly correlated with grooming activity of Splash test (P < 0.01), the number of head dips in Hole-board test (P < 0.001), and time in the central zone of OFT (P < 0.01). Also, we observed that VSELs population in bone marrow samples inversely correlated with immobility time of FST (P < 0.001), distance moved of OFT (P < 0.01) as well as the number of rearing of OFT (P < 0.001).
Interestingly, our results revealed that the plasma level of TNF-α significantly correlated with VSELs counts (P < 0.05). In addition to TNF-α, we observed that the plasma levels of IL-1β significantly correlated with VSELs counts in bone marrow samples (P < 0.05 and P < 0.01, respectively). Also, the analysis showed that nitrite levels of plasma samples inversely correlated with VSELs population in bone marrow samples of socially isolated animals (P < 0.05). In this study, we did not observe any significant correlation between plasma levels of IL-6 and VSELs population of bone marrow; however, this analysis suggested a marginally significant correlation (P = 0.056). All results and statistically reports were shown in Table 1.
On the other hand, we evaluated the correlations between VSELs counts of peripheral blood samples and behavioral tasks as well as molecular assessments in socially isolated animals. As depicted in Table 1, VSELs counts of peripheral blood samples had a significant inverse correlation with immobility time of FST (P < 0.05), distance moved of OFT (P < 0.05), and the number of rearing of OFT (P < 0.01). Also, the analysis showed a direct correlation between VSELs counts of peripheral blood samples and grooming activity of Splash test (P < 0.01), number of head dips in Hole-board test (P < 0.001), and time in the central zone of OFT (P < 0.001). In plasma samples, we observed that TNF-α, IL-1β, and nitrite level had a significant inverse correlation with VSELs counts of peripheral blood samples (P < 0.05, P < 0.01, and P < 0.001, respectively). Also, analysis revealed a marginally significant inverse correlation between IL-6 and VSELs counts of peripheral blood samples in socially isolated animals (P = 0.051).
In the final step, the correlation study demonstrated that the hippocampal VSELs levels inversely correlated with immobility time in FST (P < 0.001), distance moved and the number of rearings in OFT (P < 0.05 and P < 0.001, respectively). On the other hand, direct correlations were observed between VSELs population of hippocampus and grooming activity in Splash test (P < 0.05), number of head dips in HBT (P < 0.01), and distance moved in OFT (P < 0.05). In addition, we observed that VSELs inversely correlated with pro-inflammatory cytokines including TNF-α (P < 0.01), IL-1β (P < 0.01), and IL-6 (P < 0.05) as well as nitrite level (P < 0.01) in the hippocampus. All statistical analysis reports were mentioned in Table 1.

Discussion
The primary purpose of the current study was to evaluate the possible role of the VSELs in depressive-and anxiety-like behaviors and neuroinflammation induced by SIS. This study showed that the levels of the VSELs are correlated to mice's behavioral tasks as well as pro-inflammatory cytokines. Our results demonstrated that the mice with lower immobility time, lower locomotor activity, and higher grooming activity in the socially isolated group had higher amount of the VSELs in their bone marrow, peripheral blood, and hippocampus.
In addition, we observed that higher levels of the VSELs are correlated to lower proinflammatory cytokines levels in mice hippocampus and plasma. Based on these findings, we speculate the proliferation of VSELs in response to chronic stressor might act as an anti-inflammatory agent in mouse brain.
It has been well-documented that exposure to SIS could induce depressive-and anxiety- to the control group. Also, they found a correlation with the duration of the disorder with the number of these cell in circulation. In addition, the circulating VSELs was similar to their number in healthy subjects in individuals treated with long-term lithium; and finally, authors observed a negative correlation between the duration of lithium treatment and serum lithium concentrations [9]. In contrast to recently mentioned reports, the results of a study on VSELs in individuals with panic disorder were also presented by Jabłoński et al.
This study mentioned a lower number of VSELs in individuals with panic disorder prior to initiation of the treatment in comparison to post-treatment conditions [59]. Also, it has been reported that chemoattractants such as sphingosine-1-phosphate (S1P), stromal cellderived factor 1 (SDF-1) and the elements of the immune system, which belong to innate immunity, such Toll-like receptors could induce the mobilization of VSELs to circulation [60]; and this subject might be a connector between these cells and neuroinflammation.
Nonetheless, it was suggested that an increased number of VSELs might be used as a biological marker of psychiatric disorders. In the current study, we observed that socially isolated animals had a significantly higher number of VSELs in bone marrow, peripheral blood, and hippocampus in comparison to normal mice. This observation might be due to the proliferation of VSELs after exposing to chronic stress condition. In addition, the direct correlation between the number of circulating VSELs and bone marrow in socially isolated animals suggests an active mobilization of these cells into the peripheral blood. On the other hand, we evaluated the pro-inflammatory cytokines both in hippocampus and plasma. In both hippocampus and plasma samples, higher levels of pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6 as well as nitrite level were observed in socially isolated animals in comparison to normal mice. Interestingly, we observed the population of VSLEs in both circulation and bone marrow was negatively correlated with nitrite levels, TNF-α, and IL-1β. Finally, our results demonstrated that hippocampal VSELs not only negatively correlated with hippocampal pro-inflammatory cytokines and nitrite levels but also negatively correlated with the depressive-and anxiety-like behavior in socially isolated animals. Briefly, our correlation analysis showed that higher VSELs counts might decrease the pro-inflammatory cytokines in both plasma and hippocampus. In this regard, further studies should be established to evaluate the exact underlying mechanisms of a correlation mentioned above.
Based on previously mentioned data and our correlation analysis, we hypothesized that VSELs not only could be used as a biological marker but also it could be used as diagnostic and prognostic factors; in addition, VSELs could shed more light on stem cellbased therapeutic strategies in psychiatric disorders. In this regard, further studies should be established to evaluate the effects of stereotaxic administration of cultured VSELs on behavioral outcomes in animals. Finally, we hypothesize that VSELs have high potential to be an alternative treatment in wide ranges of neurodegenerative diseases.

Conclusion
In the current study, we concluded that VSELs could be proliferated in response to a chronic stressor such as SIS. Also, we observed that socially isolated animals with the higher hippocampal number of VSELs had lower levels of pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6 as well as nitrite levels in hippocampus. Finally, our results showed that the hippocampal number of VSELs is negatively correlated with depressive-and anxiety-like behavior in socially isolated animals. These results suggested that VSELs could be used as a biological marker to enhance diagnostic accuracy as well as predicting the prognosis. Also, based on correlation analysis, increment in the VSELs counts might decrease the neuro-inflammation and subsequently improve the behavioral impairments induced by SIS.

Conflicts of interest
The authors have indicated that they have no conflicts of interests regarding the content of this article.

Availability of supporting data
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.       Correlation analysis between the VSELs counts of bone marrow and peripheral blood for clarifying the mobilization of VSELs from bone marrow to peripheral blood.

Figure 3
Correlation analysis between the VSELs counts of bone marrow and peripheral blood for clarifying the mobilization of VSELs from bone marrow to peripheral blood.  Values are expressed as the mean ± SD from 30 animals in each group and were analyzed using the unpaired t-test. *** P<0.001, ** P<0.01 compared with normal animals.