Mental illness causes staggering burden to millions worldwide ensuing tremendous healthcare expenditures and hardship to patients and their families (1). Associated with distress and impairment of personal functioning, psychiatric disorders impact mood, cognition, behavior, and can lead to suicidal attempts and death (2, 3). Though plausibly multifactorial, the exact etiology of mental disorders and the underlying pathophysiological mechanisms are poorly understood.

More and more studies have examined immune disturbance, inflammatory processes and mental illnesses, particularly mood disorders (4–7). Though the direct intermediating immune-pathogenic mechanisms are still unofficially established, interactions between the immune system and the brain have attracted considerable attention in the field of neuropsychiatric diseases (8–10), and brain regions including the frontal cortex (FC), hippocampus (HC) and hypothalamus (HT) have been repetitively linked to such (11–13). Inflammatory mediators (such as prostaglandin [PG] E2, interleukin [IL]-6, and tumor necrosis factor [TNF]- α), which regulate brain function, proliferation, differentiation, and survival of brain cells, have also shown interconnection to psychiatric disorders (5, 14). To this end, pharmacotherapeutic strategies targeting neuroinflammatory components have been arduously explored in quest to further effective medicinal approaches to mental illness (15–26); this pursuit is of particular pertinence given the untoward side-effects, low adherence rates and limited positive outcomes associated with most available psychiatric medications today (2, 27).

Leukotrienes (LTs) are inflammatory mediators eventuating from the phospholipids-arachidonic acid (AA)-eicosanoids pathway in mammalian tissues. In this pathway, AA – a polyunsaturated fatty acid – is metabolized to eicosanoids, including prostanoids (PGs and thromboxanes) and LTs (see Figure 1 for illustration). AA is converted by the enzyme cyclooxygenase to PGH2 – the precursor from which all prostanoids are produced. Similarly, AA is transformed by the enzyme 5-lipoxygenase (5LOX) [and the enzyme 5LOX-activating protein (FLAP)] to LTA4 – the precursor from which all other LTs are produced. Several anti-inflammatory and allergy medications work by altering various junctions in the phospholipids-AA-eicosanoids cascade, such as nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids, inhibitors of 5LOX and FLAP ( Figure 1 ).

LTs strongly affect function of immune cells and play important roles in allergic, cancerous, respiratory, cardiovascular and inflammatory diseases (28, 29). Leukotrienes-modifying agents (LTMAs) are regarded as one of the most important drug families used clinically for the treatment of inflammation-and-allergy-related disorders (30). Among this drug class, montelukast (MTK) is a prime medication commonly prescribed for maintenance treatment of asthma, exercise induced bronchospasm, allergic rhinitis, and urticaria (28, 31, 32). MTK works mainly by blocking cysteinyl leukotriene (cys-LT) receptors in the lungs resulting in decreased inflammation and relaxation of smooth muscles (33–35). However, MTK purportedly exerts additional mechanisms of action. It alters several cellular signaling pathways [such as the cyclic adenosine monophosphate (cAMP)-extracellular signal-regulated kinase (ERK)] and possesses various immune-modulating and anti-inflammatory characteristics, including suppression of leukocyte proliferation and migration, attenuation of pro-inflammatory mediators (e.g., IL-6, IL-8, TNF-α and PGE2), inhibition of nuclear factor-κ B function, and more (36–44). As abovementioned, research shows an iterated correlation between inflammatory biomarkers and psychiatric disorders. For example, several studies demonstrated a distorted Th1/Th2 immune response in patients with major depression (45, 46). Thus, given the aforementioned secondary immune-modulating and anti-inflammatory mechanisms of MTK, it is logical that MTK would impact behavioral outcomes. Nonetheless, it is important to bear in mind that the alteration of the Th1/Th2 adaptive immune response by MTK (47, 48) may conceivably give way to adverse outcomes. Disbalance of the Th1/Th2 immune response may impair the defensive mechanisms against specific viral, bacterial and parasitic pathogens, thus, potentially increasing the incidence of opportunistic infections.

Reports show LT inhibitors ascertaining only modest therapeutic efficacy (49, 50). Suggested explanations for this include: individual differences in LT levels, heterogeneity of disease phenotypes, and, differences in drug pharmacokinetics, pharmacodynamics and pharmacogenomics (51–55). Another possible rationalization is sex-related differences in medication response, possibly relating to the impact of androgen levels on LTMA mediation (56), as well as the female sex hormones relating to allergic manifestations (57). In this context, a major limitation of many preclinical pharmacological studies is the exclusive inclusion of only male animals, neglecting half of the possible consumers of pharmacological treatments (51–54, 58–63). This underscores the necessity to evaluate effects of LTMAs (such as MTK) in both male and female subjects.

Almost nothing is known about the involvement of LTs in the pathogenesis of mental illness. Post-marketing reports and pharmacovigilance studies suggested that the use of LTMAs, and MTK, in particular, may be associated with the development of various adverse neuropsychiatric events (ANPEs), such as depression, aggression, suicidal ideation, anxiousness, hallucinations, sleep disturbances, irritability, tremors, and restlessness (32, 64–74). For example, a recent large retrospective, propensity score-matched cohort study by Paljarvi et al. (73) demonstrated that MTK treatment was associated with increased incidence of ANPEs among patients with asthma and allergic rhinitis. There was a notably high odds ratio for anxiety disorders among asthmatic patients treated with MTK, and a high odds ratio for insomnia in MTK-treated patients with allergic rhinitis (73). However, most of the evidence regarding ANPEs and MTK is circumstantial and not causative, based mainly on retrospective study models (65, 70, 71). Stratified analyses suggest that the reported increased incidence in ANPEs may instead relate to the underlying diseases for which MTK is being administered (e.g., asthma and allergic rhinitis), rather than MTK treatment itself (65, 71).

Though anecdotal and unestablished reports show that LTMAs (MTK in particular) may induce ANPEs, we hypothesized that given its immune-modulating and anti-inflammatory effects, MTK might potentiate beneficial behavioral effects in patients with psychiatric diseases (31, 32, 75–78). Taking into account the reports of adverse behavioral effects of MTK (70, 71, 79), we recognized that using MTK as treatment for psychiatric disorders may sound problematic. However, given the essence of the adverse data, and considering anteceding paradigms in medicine where an absolute contraindicated treatment transitioned to become mainstream (for example, β-adrenergic receptor antagonists for chronic heart failure) (80), we discretionarily concluded it was an advantageous undertaking.

This study aims to examine the behavioral effects of chronic MTK treatment in male and female rats using different animal behavioral models, as well as determine the effects of chronic MTK treatment on brain inflammatory mediator levels using pharmacological studies in male and female rats.

The main objective of the present study was to directly examine the effects of MTK on the behavioral phenotype of male and female rats. This is because several studies reported that MTK may induce ANPEs in treated patients (66–68, 79, 95, 96). Under the experimental conditions of the present study, we found that MTK does not cause prominent negative behavioral effects in rats. Table 1 summarizes the effects of MTK (20 mg/kg) on the behavior of male and female rats as evaluated in the utilized behavioral models. It is seen that, generally, MTK did not adversely influence animal behavior in either control (non-stressed) or CUMS-subjected rats ( Table 1 ). An exception was the aggression-inducing effect of MTK in male rats ( Figure 6A and Table 1 ). However, this negative effect of MTK is “counter-balanced” by its positive behavioral effects in males and females in other conditions. Together, within the scope of the particular experimental conditions applied, a generally safe behavioral profile of MTK in male and female rats was observed. This is consistent with the findings of several recent established studies that examined the association between MTK treatment and the occurrence of ANPEs in human subjects (67, 97, 98). Nonetheless, for the sake of scientific veracity, it has to be noted that other recent established studies revealed contradicting findings, i.e., that MTK increases the occurrence of ANPEs (73, 99, 100).

LTs are inflammatory mediators that strongly affect the function of mammalian tissues including the brain (28). Considering the pro-inflammatory effects of LTs and the large body of data which suggests that inflammation contributes to the pathophysiology of mental disorders (4, 6, 7, 15, 22), we hypothesized that treatment with MTK – a potent cys-LTs receptor antagonist – might exert beneficial behavioral effects in psychiatric disorders (31, 32, 75). Indeed, MTK incited some beneficial pre-clinical behavioral outcomes in rats ( Table 1 ): 1) MTK treatment increased sucrose consumption in stressed male rats, suggestive of an antidepressant-like effect. 2) MTK reduced the hyperactive/manic-like phenotype in CUMS-subjected males (as modeled in the OFT), including a reduction in mean velocity and reversal of the decreased time spent in the peripheral zone of the arena. In this regard, when placed into an open field, rats and mice tend to remain in the peripheral zone of the arena or against the walls (88, 89). To the best of our knowledge, this is the first study showing a significant anti-hyperactive/anti-manic-like effect of MTK in animals. 3) MTK mitigated aggressive-like behavior in female rats. These findings support our hypothesis that MTK might capacitate beneficial behavioral effects, similar to other typical anti-inflammatory medications such NSAIDs and corticosteroids (15–21). Recently, Tel et al. (101) showed that MTK exhibited antidepressant-like effects (assessed by measuring immobility time in the FST) in male and female mice that were subjected to an ovalbumin-induced asthma model. MTK significantly decreased immobility time in the ovalbumin-challenged, but not in control mice (101). These findings resemble the results that we obtained in the SCT, where MTK treatment did not alter sucrose consumption in control animals but reversed the reduction of sucrose consumption in CUMS-subjected male rats ( Figure 4C and Table 1 ). Another LTMA, the 5LOX inhibitor zileuton, was also found to decrease immobility time in a lipopolysaccharide induced depression-like model in mice (102), supporting the suggested possible anti-depressive effect of LTMAs. Furthermore, aggressiveness and self-harm were among the most commonly reported ANPEs of MTK (66–68, 95). In the present study, MTK increased aggressive-like behavior in male rats but decreased it in females ( Figure 6 and Table 1 ), suggestive of a sex-associated effect of the drug. These results are in line with those of a previous study which demonstrated that LTMAs were pharmacologically effective in female mice but not males (56), and may further reinforce conjectures of sex-related variance in pharmacotherapeutic responses among psychiatric patients (103). Of note, it is difficult to explain and categorize the MTK-induced reduction in distance traveled in control female rats ( Figure 3D and Table 1 ) as a therapeutic or harmful effect. This is because, on one hand, in this animal model this outcome is usually interpreted as an anti-manic/hyperactive-like effect of a given treatment intervention. However, on the other hand, since their “basal/normal” behavior was altered (compared to vehicle-treated control females, which is the “pure” control group in this setting), it can be argued that it represents an undesirable effect of MTK.

Furthermore, we tested the behavioral effects of MTK in the EPMT which models anxiety-like behavior. Rodents are inherently inclined to seek dark hiding spaces during light hours (in the present study the test was conducted during light hours) in attempt to evade being seen by predators. Therefore, remaining in the closed arms of the maze is considered characteristic normative behavior, if the time spent in these arms is proportionally comparable to that of the control animals. However, when the time rats’ dwell in the open arms significantly exceeds that of the closed arms, it is generally interpreted as non-anxious behavior. Contrastingly, if the time situated in the open arms markedly surpasses that observed among control rats, it may denote risk-taking/manic-like behavior. The results show that exposure to CUMS did not cause behavioral changes in the EPMT in male and female rats ( Figure 5 and Table 1 ). Moreover, treatment with MTK, both in control and CUMS-subjected rats, did not alter animals’ behavior in the EPMT. This clearly indicates that, under these experimental conditions, MTK did not induce anxious-like behavior.

Interestingly, we found that there were differences in the influential intensity of CUMS on the behavioral phenotypes of male vs. female rats ( Table 1 ). For example, in CUMS-subjected male rats there was a significant increase in mean velocity and a decrease in time spent the peripheral zone of the arena (both suggestive of manic-like behavior), while no such changes were observed in CUMS-subjected females. Additionally, the exposure to CUMS led to a significant decrease in sucrose consumption (indicative of depressive-like behavior) in males, while in females CUMS was associated with only a nearly significant decrease in sucrose consumption ( Figure 4 ). This is consistent with the results of previous studies which showed that stress protocols induce a reduction in sucrose preference/consumption only in male rats (104–106). However, other studies reported opposite results, i.e., that exposure to stress protocols induces a significant reduction in sucrose preference/consumption only in females (58, 107), or causes a decrease in males as well as females (108, 109). Furthermore, several previous studies reported that various stress protocols led to a significant increase in immobility time in the FST, suggestive of depressive-like behavior (81–84). Surprisingly, in the present study the utilized CUMS protocol did not cause a significant change in immobility time ( Figure 7 and Table 1 ). Our results are similar to those of previous studies (105, 110) which also demonstrated that some stress protocols do not induce significant changes in immobility time in male and female rats. It is known that the type, severity and duration of the stress protocol, and, the strain and sex of used animals all affect the measured behavioral outcomes (81–84, 106, 110–112).

The current study used subjects of the female sex heterogeneously spanning across different time points in the estrous cycle throughout experiments, with normal distribution applied similarly across all randomized female study cohorts. While we did not perform direct analytical or descriptive evaluation contingent upon the females’ hormonal cycles, we did ensure that sample sizes contained an ample number of subjects in order to neutralize potential deviations that may have related to such. Nevertheless, we acknowledge this as a limitation of our study, as research shows demonstrable correlations between inflammatory biomarkers, AA metabolites, the corpus luteum, and expressive ovulatory and implantation mechanisms (113–117). Pertinent to the behavioral experiments of the present study, Zhao et al. (118) investigated sex-specific depression and the LTA4 hydrolase haplotype gene (an aminopeptidase dictating the conversion of LTA4 to LTB4). Their findings show gene expression to confer with a significant protective effect in women in regard to depression, rationalizing further examination of cys-LTs and sex differences in mental illness. Importantly, MTK treatment has been reported to affect endometrial conditions (119) as well as other female hormonal aspects (57, 120, 121). For example, MTK was found to exert better control of asthmatic symptoms in women than men (57). Furthermore, Fujiwara et al. (121) found in a randomized, double-blind, placebo-controlled trial that MTK was effective in reducing dysmenorrheal pain in women. Numerous studies reported on the bidirectional association between LT function and the estrous cycle and reproductive system (116, 122–124). Therefore, it is feasible that the blatant sex-associated differences in our study are affiliated with the relating female hormonal processes involved, and stratified analyzation of this parameter may have yielded interesting outcomes.

As mentioned, a large body of data suggested that inflammation contributes to the pathophysiology of mental disorders (15, 22, 125–132). For example, several research papers reported increased IL-6 levels in patients with major depression (129, 130, 133), bipolar disorder (129, 131) and schizophrenia (129, 130). Moreover, multiple studies showed that IL-6 levels were prominently increased in the blood of subjects after suicidal attempts and in post-mortem brains of people after suicidal death (134–136). Similarly, many studies found that TNF-α levels are higher in patients with major depression (129, 133), bipolar disorder (129, 131, 137) and schizophrenia (129, 137) than in matched-controlled subjects. PGs (PGE2 in particular) have been recurrently observed as connected to the pathophysiology of psychiatric disorders (138, 139). These outcomes are highly relevant to the behavioral findings of the present work, because numerous studies have demonstrated that MTK decreases the levels of several pro-inflammatory mediators including IL-6, TNF-α and PGE2 under various experimental conditions (34, 37, 40, 42, 140, 141). Thus, we hypothesized that the behavioral effects of MTK may be influenced by and related to its effects on brain inflammation. In the present study we assessed brain inflammation by measuring inflammatory mediator levels in the FC, HT and HC (11–13). Table 2 summarizes the effects of MTK treatment and the exposure to CUMS on levels of IL-6, TNF-α and PGE2 in these brain regions. As seen, in control males, MTK increased IL-6 levels in the HT, and TNF-α levels it the FC, suggestive of a pro-inflammatory effect of the drug. In contrast, in control females, MTK treatment was associated with a robust anti-inflammatory effect; it significantly decreased IL-6 and TNF-α levels almost in all brain regions. On the other hand, MTK increased PGE2 levels in the FC and HC. These results are similar to those of previous studies which revealed that under certain conditions MTK may increase the production of PGE2 (44, 142). We speculate that the oppositional impact of MTK on IL-6 and TNF-α levels in male vs. female rats may contribute, at least in part, to its distinctive effect on aggressive-like behavior ( Table 1 ). MTK induced aggressive-like behavior in males, while it decreased this behavior in females. Numerous studies demonstrated that aggressive behavior in humans and rodents is affected by the function and structure of the FC, HT and HC (143–145). Importantly, MTK also attenuated aggressive-like behavior in CUMS-subjected females ( Table 1 ), and here too, there was a prominent reduction in TNF-α levels in the three brain regions and in IL-6 levels in the FC and HT ( Table 2 ). In CUMS-subjected male rats the effect of MTK on inflammatory mediator levels was inconsistent and difficult to interpret. As for the effect of CUMS (alone) on inflammatory mediator levels, we found that the stress protocol did not alter brain levels of either IL-6, TNF-α or PGE2 (except in the HT) in male rats. In contrast, in females, the exposure to CUMS was associated with a profound reduction in TNF-α levels in all brain regions but particularly in the HC ( Figure 9 ). A prominent reduction in TNF-α levels is usually considered an anti-inflammatory effect of a given intervention. However, again, since in this case it is a deviation from the basal/control condition, this interpretation may be disputable.

To the best of our knowledge, this is the first study that directly and thoroughly tested the behavioral effects of MTK in rats. Overall, the results indicate that MTK treatment does not induce prominent adverse behavioral effects and may instead be associated with select beneficial behavioral outcomes. Moreover, the results support our hypothesis that treatment with MTK differentially affects levels of brain inflammatory mediators in male vs. female rats, which plausibly explains the dissimilar behavioral phenotypes of the sexes. Randomized, double-blind clinical trials in human subjects are necessary to directly determine the behavioral effectual capacity of MTK.

The primary data used to support the findings of this study are available from the corresponding author upon reasonable request.

The animal study was reviewed and approved by Committee for the Use and Care of Laboratory Animals in Ben-Gurion University of the Negev, Israel.

Study concept and design – AA. Acquisition, analysis, or/and interpretation of the data – IR, BB-A, AN, ER, SU, JK, LB, AA. Statistical analysis – IR, SU, AA. Drafting of the manuscript – IR. Critical revision of the manuscript – All authors. Writing of the final version of the manuscript – AA. All authors approved the final version of the manuscript.

This study was supported by a grant from the Israel Science Foundation (Grant # 198/12) to AA.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.