Epidemiologic studies have reported that about 4–23% of people with traumatic damages manifest post-traumatic stress disorder (PTSD) symptoms including re-experiencing, avoidance, and hyperarousal for at least 1 month (American Psychiatric Association, 2013; Arnberg et al., 2013; Zhou et al., 2013). Biological alterations may underline PTSD symptom onset and persistence (Clément et al., 2002; Ozer et al., 2003; Segman et al., 2005). The hypothalamic-pituitary-adrenal (HPA) axis is the main neuroendocrine arm of the stress response. It includes corticotrophin-releasing hormone (CRH) release from the paraventricular nucleus (PVN) of the hypothalamus into the hypophyseal-portal circulation, stimulating the anterior pituitary gland to release adrenocorticotropic hormone (ACTH) and ultimately glucocorticoids from the adrenal cortex. This mechanism maintains homeostasis in stressful environmental conditions (Jacobson, 2005). Neuroendocrinologic studies have reported three common features in patients with PTSD: (1) decreased plasma cortisol; (2) higher CRH in the cerebrospinal fluid (CSF) and plasma; and (3) increased inhibition of the hypophysis-adrenal system (HAS), which leads to reduced cortisol levels by enhancing negative feedback (Carvalho et al., 2008). CRH and vasopressin are the most important mediators of the HPA axis' effects on neuroendocrine signaling.

Studies have shown that AVP expression in paraventricular neurons and V1b receptor density in pituitary corticotroph significantly increased in response to chronic stress (Kovács and Sawchenko, 1996; Aguilera and Rabadan-Diehl, 2000). These findings indicate that AVP plays an important role in responding to stress by maintaining the ACTH sensitivity to stressors during chronic stress (Smith and Vale, 2006).

Considering the synergistic effect of corticotrophin and vasopressin in response to stress, as well as the role of stress in PTSD development, these neurohormones likely play a role in PTSD pathogenesis. ACTH is a key mediator in glucocorticoid secretion from the adrenal cortex. Glucocorticoid hormone release, (cortisol in humans and Corticosterone in rats), is the final effect of HPA axis activity, and they play a significant role in the stress response (Yehuda, 2009). Studies of patients with PTSD showed increases in the number and sensitivity of glucocorticoid receptors (Yehuda et al., 1995; Rohleder et al., 2004), as well as greater inhibition of cortisol and adrenocorticotrophin following dexamethasone treatment (McFarlane et al., 2011).

Crocin (Crocin di-gentiobiose ester) is an isolated chemical compound that belongs to a group of commercial carotenoid derived from the stigma branches of dried saffron. Crocin can improve learning and memory (Ebadi, 2006) and may prevent neurodegenerative disorders including Alzheimer's disease (Rohleder et al., 2004). It is not mutagenic (Papandreou et al., 2006) and prevents alcohol-induced disorders of memory and learning (Papandreou et al., 2006). Its mechanism is thought to be prevention of the inhibitory effect of ethanol on N-methyl-D-aspartate (NMDA) glutamate receptors in the hippocampus (Ebadi, 2006).

However, it is not clear that what cellular and molecular mechanisms are underlying the Crocin action on PTSD. In addition, it is not clear that what would happen if the stress paradigm became shorter in time duration. We aimed to evaluate the role of Crocin on HPA axis activity. To this end, we evaluated the effect of intracerebroventricular (ICV) infusion of Crocin on expression of genes involved in HPA pathway, plasma Corticosterone levels, and behavioral symptoms were evaluated in an electric foot shock rat model of PTSD (Langevin et al., 2010).

The aim of the present study was to test the efficacy of Crocin as a treatment for PTSD using an animal model. The system in our study re-exposed rats to stress conditions. On the day 21 post-shock, serum Corticosterone levels were reduced in rats with PTSD; gene expression of CRH and GluR increased in the hypothalamus and pituitary, respectively; and overt behavioral changes were observed. Our results indicated that left lateral ICV infusion of Crocin can reduce CRH and GluR expression in specific brain areas. It can also increase serum Corticosterone at different time points and reduce stress-induced behaviors such as freezing in rats with PTSD. Several studies conducted on acute and chronic stresses reported that stressing factors can affect nervous, endocrine, and behavioral systems and induce physiological adaptations to preserve homeostasis (Figueiredo et al., 2003).

CRH regulates stress responses through activation of the HPA axis, which results in glucocorticoid release. CRH is also secreted from hippocampus and amygdale, where it plays the role of a neurotransmitter agent and responds to behavioral and autonomic stresses (Aguilera, 1998; Chen et al., 2004). High levels of CRH in the anterior pituitary desensitize CRH receptors, decreasing their responses (Dautzenberg et al., 2001; de Kloet et al., 2006).

Increased CRH was also observed in our model of PTSD. Stress behaviors were observed, while there was a significant decrease in CRH expression in the treatment-Crocin group compared with those with no treatment. Therefore, it is the indication of Crocin effect through the HPA axis to reduce hypothalamic CRH. A study evaluating the effect of saffron extract and Crocin on PTSD in rats reported corroboratory results (Sahraei et al., 2012). A likely mechanism of Crocin in regulating CRH secretion might be through alteration of a glutamate-dependent mechanism. Researchers indicated that glutaminergic innervations of the PVN changes under conditions of chronic stress and may be involved in sensitizing HPA axis responses (Zelena et al., 1999; Pistovcakova et al., 2005; Popoli et al., 2012). On the other hand, inhibition of NMDA receptors (NMDARs) by some components in saffron extract has been shown (Lechtenberg et al., 2008). Moreover, it has been indicated that saffron extract can increase brain glutamate levels (Ettehadi et al., 2013). Other studies reported that glutamate receptors are in post-synaptic and presynaptic forms, with presynaptic inhibitory receptors playing an important role in regulating glutamate release (Yuen et al., 2012). Collectively, the data suggest that glutamate plays a complex role in exciting CRH neurons, acting at multiple levels to both drive HPA axis responses and limit over activation (Feldman and Weidenfeld, 1997; Ulrich-Lai et al., 2011). Considering the inhibition of CRH expression in rats treated with Crocin, it seems that the Crocin plays a genomic role, in addition to its inhibitory effects on NMDA receptors. The finding of reduced CRH is likely relevant to the mechanism of PTSD, but studies in primates or humans are needed to clarify this pathway.

Vasopressin is a neuropeptide expressed in different hypothalamic nuclei such as the PVN supraoptic nucleus (SO) and suprachiasmatic nucleus (SC). Chronic stress increases vasopressin synthesis, which is secreted in the pituitary portal system and stimulates the HPA axis, resulting in ACTH release (Dorsa et al., 1988; Hernando et al., 2001). A recent study showed decreased vasopressin in the magnocellular area of the PVN after a forced swimming stress (Ettehadi et al., 2013). Vasopressin can improve the inhibitory feedback of glucocorticoids on ACTH release and stimulate corticotrophin expression to respond to new stresses and reactivate the HPA axis (Hauger and Dautzenberg, 2000).

We did not find significant differences in AVP expression between the PTSD and non-PTSD groups, which may be due to the different mechanisms of parvocellular and magnocellular neurons in the stress response in rats with PTSD. It may also depend on the timing of stress exposure and re-exposure. However, further studies to evaluate the status of AVP gene expression under PTSD conditions are recommended to clarify this issue.

A major neuroendocrine finding is that PTSD patients exhibit abnormally high GluR sensitivity resulting in HPA over suppression due to corticosteroid negative feedback (Yehuda, 2009). We observed increased GluR expression in the pituitary of rats with PTSD compared with animals treated with Crocin. The results indicate that Crocin may recalibrate GluRs and change the mechanism of response to Corticosterone by decreasing GluR gene expression. These results hint at the mechanism of the brain's response to PTSD; however, human studies are recommended to precisely investigate the relationship between GluR gene expression and PTSD.

Studies have demonstrated decreased glucocorticoid levels in PTSD conditions, probably due to changes in the negative feedback activity of the associated gene receptors (Yehuda et al., 1991; Gold and Chrousos, 2002; Radley et al., 2011; Daskalakis et al., 2013). Our results are consistent with other studies regarding decreased Corticosterone in the PTSD rat model, which showed an increase at baseline, under stress, and during the return toward basal levels following Crocin treatment (Ettehadi et al., 2013). The ability of Crocin to inhibit Corticosterone changes in the current PTSD model indicate normalized HPA axis activity. It seems that the Corticosterone levels play an important role in the mechanism of PTSD.

To confirm the effect of Crocin in rats with PTSD, we evaluated changes in overt behaviors. Evaluating such behaviors in animal models may be used in therapeutic interventions. It has been reported that glucocorticoid in mesolimbic dopaminergic areas increases dopamine release and causes euphoric feelings and movements in rats (Howes et al., 2017). The increased freezing and decreased sniffing, rearing, and grooming behaviors in our study could be attributed to decreased Corticosterone levels in rats with PTSD.

We also observed a decreased incidence of freezing in crocin-treated rats, but no significant change was noted for other overt behaviors. Overt behaviors are those mediated by the dopaminergic system in the brain. Although physical movements (as opposed to freezing behavior) are directly controlled by the extra pyramidal dopamine system, they may be affected by glutamate, acetylcholine and gamma-aminobutryric acid neurotransmission systems (Shim et al., 2014). Since the anatomical locations of the behaviors in the brain are not similar, all behaviors may not be affected by the same intervention (medicine = Crocin, physical = stress) (Gottfried, 2006; Howes et al., 2017).

From cellular and molecular aspects, GluRs in the mesolimbic system play important roles in overt behaviors such as locomotion, rearing, and grooming (Howes et al., 2017). On the other hand, suppression of NMDARs receptors can decrease mesocorticolimbic system activity and ultimately inhibit brain dopamine and overt behaviors (Tzschentke, 2007). One study showed that suppression of GluR by MK-801 decreased movements in rats (Tzschentke, 2007). Similarly, suppression of dopamine receptors by sulpiride inhibits overt behaviors, indicating a close relationship between glutaminergic and dopamine pathways (Tzschentke, 2007). On the other hands, previous studies indicated that Crocin administration reduced the freezing time in the stress animals (Sahraei et al., 2012; Pitsikas, 2016). The reduced freezing time could be attributed to the reduced anxiety (Pitsikas, 2016). Interestingly, investigators insists that the Crocin can interact with NMDA receptors and act as it is antagonist (Hosseinzadeh et al., 2008; Berger et al., 2011). Hence, considering the inhibitory effect of Crocin on NMDARs, our results suggest a combinatorial mechanism by which Crocin decreases PTSD symptoms by influencing NMDAR activity and glucocorticoid levels.

In the present study, it has been demonstrated that crocin induces grooming (although it was not statistically significant), While in another study, it has been manifested that mCPP (a 5HT2c agonist) induced excessive self-grooming in the rat and crocin attenuated this effect of mCPP (Georgiadou et al., 2012); and in another study, crocin decreased grooming in mice (Hosseinzadeh and Noraei, 2009). These findings seem to be inconsistent with ours showing that crocin induces grooming. However, it should be noted that there is a remarkable difference between the procedures used in the mentioned studies. We have used the crocin alone as a treatment intervention in stressed animals, while the first study has used corcin against mCPP in animal model of obsessive–compulsive disorder (OCD) and, in the second study, crocin is used in intact animal.

Further advanced investigations into the relationships between different parts of the brain would be helpful to further elucidate this pathway.

We found no significant difference in the weights of the whole body or adrenal or thymus glands in PTSD rats with or without Crocin treatment compared with non-PTSD rats. Previous rodent studies reported that the chronic stress can cause weight loss and gain in males and females, respectively (Sato and Fahrenkamp, 2016). It seems that the present model, which had no significant effect on rat, is superior to those of the similar studies. Greater adrenal gland weight is one of the most important effects of chronic stress and results from hyperplasia in the fascicular zone of the adrenal cortex (Bali and Jaggi, 2015), which stimulates the HPA to have a more robust response to stress. Increased adrenal gland weight was not observed in the present study, possibly due to decreased sensitivity of the HPA axis. Others have reported smaller thymus size following chronic stress due to the inhibitory effect of glucocorticoids on T-cell differentiation (Pertsov et al., 2015). Again, these results were not observed in our acute stress model.

The question which might be asked is that if the animals' behavior and/or other experimental procedures were evaluated in different days (other than day 21), different results would be obtained. It is an important issue which needs to be further investigated in the future experiments. Indeed, considering various times of interventions and their results would be of importance, and it must be considered for future experiments. However, in the present study, the protocol used by was noticed (Mikics et al., 2008; Zoladz et al., 2012).

The psychoactive agent Crocin possesses glutamate receptor agonist properties that can regulate HPA axis activity and has the potential to reduce PTSD symptoms. The model employed in the current study is applicable for evaluating HPA axis disorders.

On behalf of the corresponding author I would like to undertake the responsibility that all authors have contributed in this study in terms of experimental work, statistical analysis, scientific writing and revision. SA, MT, HS, and GP contributed to the design of the study, experimental work, drafting of the paper, major revisions, and final approval for publication.