Worsening of the Toxic Effects of (±)Cis-4,4′-DMAR Following Its Co-Administration with (±)Trans-4,4′-DMAR: Neuro-Behavioural, Physiological, Immunohistochemical and Metabolic Studies in Mice

4,4’-Dimethylaminorex (4,4’-DMAR) is a new synthetic stimulant, and only a little information has been made available so far regarding its pharmaco-toxicological effects. The aim of this study was to investigate the effects of the systemic administration of both the single (±)cis (0.1–60 mg/kg) and (±)trans (30 and 60 mg/kg) stereoisomers and their co-administration (e.g., (±)cis at 1, 10 or 60 mg/kg + (±)trans at 30 mg/kg) in mice. Moreover, we investigated the effect of 4,4′-DMAR on the expression of markers of oxidative/nitrosative stress (8-OHdG, iNOS, NT and NOX2), apoptosis (Smac/DIABLO and NF-κB), and heat shock proteins (HSP27, HSP70, HSP90) in the cerebral cortex. Our study demonstrated that the (±)cis stereoisomer dose-dependently induced psychomotor agitation, sweating, salivation, hyperthermia, stimulated aggression, convulsions and death. Conversely, the (±)trans stereoisomer was ineffective whilst the stereoisomers’ co-administration resulted in a worsening of the toxic (±)cis stereoisomer effects. This trend of responses was confirmed by immunohistochemical analysis on the cortex. Finally, we investigated the potentially toxic effects of stereoisomer co-administration by studying urinary excretion. The excretion study showed that the (±)trans stereoisomer reduced the metabolism of the (±)cis form and increased its amount in the urine, possibly reflecting its increased plasma levels and, therefore, the worsening of its toxicity.

4,4′-DMAR may be considered a methyl-derivative of 4-methylaminorex (4-MAR) and aminorex [6], two synthetic stimulants known for their anorexic properties [7]. Indeed, in 4-MAR the methyl group is in position 4 on the oxazoline ring, such as 4,4′-DMAR that has one more methyl group in para-position on the phenyl ring ( Figure 1). Aminorex, on the other hand, has no methyl group on the oxazoline and phenyl rings. The presence of two chiral centres on the oxazoline ring of 4-MAR and 4,4-DMAR gives rise to four enantiomers represented by two diastereomeric (±)-cis and (±)-trans forms [3].
4,4′-DMAR and 4-MAR are widely used as recreational substances, as was aminorex [8]; due to their toxicity, all three stimulants are already under national or international legislation control.
As with other NPS, despite legislative restrictions 4,4′-DMAR can be purchased through websites selling "research chemicals" where it is usually sold in different forms (e.g., as white powder or as different coloured tablets and pellets) and under different names, including: "Serotoni", "4-methyl-U4Euth" or "4-methyl-euphoria" [5,[9][10][11][12]. The routes of administration mostly used by consumers include nasal insufflation, inhalation and oral administration [12]. Nevertheless, according to a fatality report from Hungary, 4,4′-DMAR can also be injected [9]. Low dosages (e.g., 10-15 mg for insufflation or 10-25 mg for oral administration) are usually ingested to enjoy levels of hyper-arousal, euphoria, decreased appetite and motor stimulation. An intermediate dosage (e.g., 20-50 mg oral) can be associated with restlessness, agitation and insomnia; high dosages (e.g., above 100 mg) may present with severe anorexia, mild paranoia at times with hallucinations, hyperthermia, bruxism, facial spasms, an increase in aggression levels, seizures and increased heart rate levels, at times evolving in a cardiac arrest [6,10,13]. However, these dose-related pharmaco-toxicological reports need to be interpreted with caution, since deriving from subjective users' experiences and/or based on clinical descriptions of clients presenting to accident and emergency departments often referring to synthetic stimulant multi-drug intake episodes [10]. Indeed, in all the 27 4,4′-DMAR-related deaths reported to the EMCCDA, at least one further recreational drug and/or stimulant NPS was detected [9]. Aminorex, on the other hand, has no methyl group on the oxazoline and phenyl rings. The presence of two chiral centres on the oxazoline ring of 4-MAR and 4,4-DMAR gives rise to four enantiomers represented by two diastereomeric (±)-cis and (±)-trans forms [3]. 4,4 -DMAR and 4-MAR are widely used as recreational substances, as was aminorex [8]; due to their toxicity, all three stimulants are already under national or international legislation control.
As with other NPS, despite legislative restrictions 4,4 -DMAR can be purchased through websites selling "research chemicals" where it is usually sold in different forms (e.g., as white powder or as different coloured tablets and pellets) and under different names, including: "Serotoni", "4-methyl-U4Euth" or "4-methyl-euphoria" [5,[9][10][11][12]. The routes of administration mostly used by consumers include nasal insufflation, inhalation and oral administration [12]. Nevertheless, according to a fatality report from Hungary, 4,4 -DMAR can also be injected [9]. Low dosages (e.g., 10-15 mg for insufflation or 10-25 mg for oral administration) are usually ingested to enjoy levels of hyper-arousal, euphoria, decreased appetite and motor stimulation. An intermediate dosage (e.g., 20-50 mg oral) can be associated with restlessness, agitation and insomnia; high dosages (e.g., above 100 mg) may present with severe anorexia, mild paranoia at times with hallucinations, hyperthermia, bruxism, facial spasms, an increase in aggression levels, seizures and increased heart rate levels, at times evolving in a cardiac arrest [6,10,13]. However, these dose-related pharmaco-toxicological reports need to be interpreted with caution, since deriving from subjective users' experiences and/or based on clinical descriptions of clients presenting to accident and emergency departments often referring to synthetic stimulant multi-drug intake episodes [10]. Indeed, in all the 27 4,4 -DMAR-related deaths reported to the EMCCDA, at least one further recreational drug and/or stimulant NPS was detected [9].
Cardiac arrest, brain oedema, elevated body temperature, bleeding and seizures were the most common adverse events/autopsy findings in fatalities associated with 4,4 -DMAR [9]. The admission notes and autopsy reports of 4,4 -DMAR-related deaths [9] identified a range of clinical features consistent with serotonin toxicity [16]. Conversely, the 4,4 -DMAR-related cardiotoxicity may be related to its effects on both extracellular norepinephrine and 5-HT [3,17].
Nevertheless, there are no published pre-clinical or scientific safety data relating to the toxic/neurotoxic potential of 4,4 -DMAR in animals or humans. A further problem is represented by the fact that in the seized material, the exact concentration of either (±)cis-4,4 -DMAR or of the (±)trans-4,4 -DMAR stereoisomer, or indeed of their presence in a racemic mixture, is not known [18]. This issue, already previously described for stimulant compounds in racemic mixtures such as threo-methylphenidate [19], (±)-MDPV [20,21] or empathogenic drugs such as (±)-MDMA [22], could directly influence these molecules' pharmaco-toxicological profile, potentially representing a serious health problem for consumers but also complicating data interpretation in forensic toxicology analysis.
Therefore, the aim of the present study was to investigate the effects of the acute systemic administration of both the single (±)cis (0.1-60 mg/kg) and (±)trans (30 and 60 mg/kg) stereoisomers but also of their co-administration ((±)cis at 1, 10 or 60 mg/kg + (±)trans at 30 mg/kg) in mice. A range of physiological (e.g., sweating, salivation, hyperthermia) and neuro-behavioural (e.g., psychomotor agitation, aggressiveness, convulsion) parameters were here assessed. Moreover, to highlight possible neurotoxic mechanisms, and similar to what previously investigated in relation to MDMA [23] the effects of 4,4 -DMAR on the expression of key markers of oxidative/nitrosative stress (8-OHdG, iNOS, NT and NOX2) and apoptosis (Smac/DIABLO and NF-κB) were here analysed. Furthermore, in taking from some current preliminary data, which showed the emergence of a 4,4 -DMAR-related hyperthermia, the expression of heat shock proteins (HSP27, HSP70, HSP90), markers related to heat-induced response [24][25][26], was here assessed as well. Finally, in studying the excretion of the related urinary metabolites, the potential mechanism underlying the potentiation of the toxic effects of the (±)cis stereoisomer when co-administered with the (±)trans stereoisomer was here investigated.
All animals showed profuse sweating and salivation at the doses of 30 and 60 mg/kg, although at the highest dose of 60 mg/kg only 50% of the animals showed profuse sweating possibly because at this dose half of them died before clearly showing this physiological effect. (±)Cis-4,4 -DMAR dose-dependently induced hyperthermia at 30 mg/kg (+1.52 ± 0.11 • C) and at 60 mg/kg (+2.22 ± 0.12 • C), whilst lower doses (0.1-10 mg/kg) were ineffective (one-way ANOVA F (6.35) = 52.18, p < 0.0001). (±)Cis-4,4 -DMAR was not associated with increased levels of spontaneous aggressiveness in mice. Conversely, the molecule induced levels of stimulated aggressiveness in a dose-dependent manner at 10 and 30 mg/kg (t =12.30, df = 10; p < 0.0001). At the highest dose (60 mg/kg), the spontaneous and stimulated aggressiveness tests could not be performed as the animals were excessively agitated and convulsive. Systemic administration of (±)cis-4,4 -DMAR induced seizures in 50% and 100% of mice at the 30 and 60 mg/kg dosages, respectively; the duration of seizures was longer in those mice treated with the highest dose (t = 3.280, df = 10; p = 0.0083). (±)Cis-4,4 -DMAR at 30 and 60 mg/kg caused the death of 50% and 100% of mice, respectively, and at the highest dose the time to death was significantly shorter (t = 2.526, df = 10; p = 0.0301).

Physiological and Neuro-Behavioural Responses to (±)Trans-4,4 -DMAR
Systemic administration of (±)trans-4,4 -DMAR (30 and 60 mg/kg; Table 2) did not induce any physiological and neuro-behavioural changes at either of the dose tested. The (±)trans-4,4 -DMAR tested at the lower doses of 0.1 and 1 mg/kg was inactive on all parameters studied. Table 2. Effect of (±)trans-4,4 -DMAR (30 and 60 mg/kg i.p.) on neuro-behavioural changes (psychomotor agitation, aggressiveness, convulsion), physiological alterations (sweating, salivation, hyperthermia) and lethality in mice. Data expressed (see material and methods) represents the mean ± SEM of six animals for each treatment. The statistical analysis of the effects of the (±)trans-4,4 -DMAR in different concentrations were performed using a one-way ANOVA followed by a Bonferroni test for multiple comparisons. A Student's t-test was used to determine statistical significance (p < 0.05) between the two groups.

Histological Results
Histological analyses showed only a slight oedema in the cerebral cortex ( Figure 2).

Excretion Studies on Urine Samples
The principal markers of intake for (±)cis-4,4 -DMAR were the carboxylate (M2) and mono-hydroxylate (M1, M3) metabolites where M2 was the most excreted metabolite detected in mice urine samples, while the (±)trans-4,4 -DMAR did not exhibit phase I/II metabolism, see Figure 6 for the structures of metabolites.

Excretion Studies on Urine Samples
The principal markers of intake for (±)cis-4,4′-DMAR were the carboxylate (M2) and mono-hydroxylate (M1, M3) metabolites where M2 was the most excreted metabolite detected in mice urine samples, while the (±)trans-4,4′-DMAR did not exhibit phase I/II metabolism, see Figure 6 for the structures of metabolites. In the present study, the excretion levels of M1, M2 and M3 were compared in the two groups of mice. The first group was administered with (±)cis isomer and the second group was co-administered with both isomers at the same dose (10 mg/kg). The aim was to evaluate changes in the excretion profile of these metabolites (M1-M3) between the two groups. The excretion levels of the individual metabolites were compared as a percentage ratio. This ratio was calculated considering the excretion of the parent compound, assuming the percentage of (±)cis-4,4′-DMAR as 100%. The data were graphically reported as box plots, with the calculated minimum, maximum, median and average values ( Figure  7A).  In the present study, the excretion levels of M1, M2 and M3 were compared in the two groups of mice. The first group was administered with (±)cis isomer and the second group was co-administered with both isomers at the same dose (10 mg/kg). The aim was to evaluate changes in the excretion profile of these metabolites (M1-M3) between the two groups. The excretion levels of the individual metabolites were compared as a percentage ratio. This ratio was calculated considering the excretion of the parent compound, assuming the percentage of (±)cis-4,4 -DMAR as 100%. The data were graphically reported as box plots, with the calculated minimum, maximum, median and average values ( Figure 7A).

Excretion Studies on Urine Samples
The principal markers of intake for (±)cis-4,4′-DMAR were the carboxylate (M2) and mono-hydroxylate (M1, M3) metabolites where M2 was the most excreted metabolite detected in mice urine samples, while the (±)trans-4,4′-DMAR did not exhibit phase I/II metabolism, see Figure 6 for the structures of metabolites. In the present study, the excretion levels of M1, M2 and M3 were compared in the two groups of mice. The first group was administered with (±)cis isomer and the second group was co-administered with both isomers at the same dose (10 mg/kg). The aim was to evaluate changes in the excretion profile of these metabolites (M1-M3) between the two groups. The excretion levels of the individual metabolites were compared as a percentage ratio. This ratio was calculated considering the excretion of the parent compound, assuming the percentage of (±)cis-4,4′-DMAR as 100%. The data were graphically reported as box plots, with the calculated minimum, maximum, median and average values ( Figure  7A).  The values shown were the sum of the excretion of (±)cis-4,4 -DMAR and its metabolites in the range of hours considered. The results showed that the administration of (±)cis-4,4 -DMAR alone led to the excretion of M1 and in particular of M2 in greater quantities than the parent compound. Furthermore, a significant variability of the data was observed for all the metabolites, which could be attributed to the variable nature of the excretion among different mice and time ranges. The results were significantly different in the case of co-administration of both isomers. For all the metabolites there was a remarkable reduction in excretion values compared to the parent compound, mainly for the metabolites M1 and M2. The principal marker M2 showed maximum percentage values equal to half of the excretion of the parent compound whilst the average values of all three metabolites were significantly lower than this value. This observation could be explained by an inhibition of the metabolism of the (±)cis isomer caused by the co-administration with the (±)trans isomer. The relatively low variability of the data could be explained by the reduced metabolism of the parent compound, which entailed a lower excretion of the metabolites and consequently smaller levels of inter-individual variability. This was also highlighted by the comparison of excretion data showed in Figure 7B with data normalised to the sum of excretion of 4,4 -DMAR and its metabolites.

Discussion
The present study demonstrated for the first time the toxicity of the (±)cis-4,4 -DMAR in mice and the risk when taken together with the (±)trans-4,4 -DMAR form in racemic mixtures. (±)Cis-4,4 -DMAR dose-dependently induced psychomotor agitation, sweating, salivation, hyperthermia, stimulated aggression, convulsions and death in mice. Con-versely, the (±)trans-4,4 -DMAR was inactive. However, their co-administration resulted in a worsening of the toxic effects caused by (±)cis-4,4 -DMAR, inducing rapid and severe hyperthermia, convulsions and death of the animals. Immunohistochemical analysis showed that this acute intoxication caused high expression of markers of oxidative/nitrosative stress (8-OHdG, iNOS, NT and NOX2), apoptosis (Smac/DIABLO and NF-κB) and heat shock proteins (HSP27, HSP70, HSP90) in the frontal cortex, suggesting potential neurotoxic damage. The urinary excretion studies suggested that the worsening of physiological and neuro-behavioural parameters could be related to the inhibition of the metabolism of the (±)cis-4,4 -DMAR form by the (±)trans-4,4 -DMAR. Overall, the strength of the present study related here to its clinical-toxicological relevance of the use of doses of 4-4'-DMAR in mice, which are equivalent to those used in humans (HED; human equivalent dose); these were associated with mild, intermediate and strong responses ( Table 5). Although the current study was solely based on the preclinical mouse model, this may allow a translational evaluation of the pharmaco-toxicological effects that could be observed in humans.
Physiological and neuro-behavioural alterations observed in mice were broadly in agreement with the clinical scenario typically observed in consumers ingesting 4-4'-DMAR, with symptoms and signs including: psychomotor agitation, sweating, salivation, hyperthermia, aggression, convulsions and possible death [4,10,13]. This scenario is likely related to the increased release of catecholamines and facilitated serotonergic transmission [6]. In particular, the latter seems to be responsible for the more dangerous toxic effects found in 4-4'-DMAR intoxications, such as cardiac arrest, brain oedema, hyperthermia, bleeding and seizures [4]. Therefore, this study allows us to hypothesise that from a clinical point of view an acute 4-4'-DMAR intoxication can be treated, such as that caused by other stimulants already known. Furthermore, the metabolic profile of 4-4'-DMAR excretion can cover an important aspect in clinical-toxicological and forensic investigations. While changes in immunohistochemical markers (for example HSPs) are indicative of metabolic changes in the brain that may be of help in forensic investigations to clarify, for example, the presence of hyperthermia in the brain and its possible pathophysiological relapse. involve severe anorexia, mild paranoia (sometimes hallucinations), hyperthermia, bruxism, facial spasms, an increase in aggression and desire for violence, seizures, an increased heart rate that will be involved in a cardiac arrest

Psychomotor Agitation
Similarly to 4-MAR, psychomotor agitation was the first behavioural effect observed after administration of (±)cis-4,4 -DMAR; not only an increase in spontaneous locomotion activity [27], but also an "unusual hyperactivity" pattern (e.g., rapid and sudden horizontal displacements in all directions, sustained turning behaviour, involuntary falls from the high plate during the evaluation of the tests, stereotyped movement and rearing activity) were observed [28]. These enhanced behavioural responses were consistent with an increase in extracellular levels of monoamines [28][29][30], especially in the dorsal and ventral striatum [31][32][33].
Although a previous study showed that the motor stimulation and stereotypies caused by 4-MAR are serotonin-independent and prevented only by the blockade of dopaminergic D 1 and D 2 receptors [28], the motor effects induced by 4-4 -DMAR could also depend on the serotonergic signal, given its pharmacodynamic profile more similar to that of MDMA with respect to 4-MAR and aminorex [3,14].

Aggression
(±)Cis-4,4 -DMAR caused psychomotor agitation associated with stimulated aggressiveness in mice as typically reported for other psychostimulants such as cocaine, MDPV, methiopropamine, methamphetamine, amphetamine [42][43][44], which well identifies the adverse behavioural effects typically observed in users of synthetic stimulants and in particular of cathinone that gives rise to the so-called excited delirium syndrome (EDS) [45,46].
The EDS was first described as the sudden, unexpected death of people restrained by law enforcement officers [47] (in particular, positional holds are the restraints most frequently associated with sudden death in susceptible people with excited delirium) [48]. Frequently, the subjects tend to display signs of agitation, aggressiveness and hyperactivity prior to the exitus [49].
Hyperthermia, which is one of the symptoms of the serotonin syndrome [61], may be sustained by an increased serotonergic signal. In fact, a direct link between the serotonergic transmission and thermoregulation has been well established. (±)Cis-4,4 -DMAR, promoting serotonergic transmission and possibly activating 5HT 2A/C receptors [15], could induce hyperthermia by increasing metabolism whilst inducing hyperactivity coupled with hyperthermia [53,54] and by causing peripheral vasoconstriction [62,63]; this scenario has been typically reported as well for MDMA, cocaine, methylone, MDPV and 25B-NBOMe [51,56,57,64]. However, the 4,4 -DMAR-induced norepinephrine release may also account for its hyperthermic effect. Furthermore, increased norepinephrine plasma levels may lead to a loss of heat dissipation through α1AR-mediated vasoconstriction, while stimulation of α1 and β3 adrenergic receptors may regulate a mitochondrial protein in skeletal muscle, uncoupling protein-3 (UCP-3) inducing thermogenesis [65,66]. Overall, hyperthermia is considered a potential acute severe adverse effect and one of the primary causes of death [67,68]. According to the literature, this condition can be associated with a range of life-threatening complications, namely cerebral oedema, rhabdomyolysis, hyponatraemia, disseminated intravascular coagulation (DIC) and coma [69].
At present, knowledge of the effects of drug-induced hyperthermia on neurotoxicity is very limited. As reported by Bowyer at al., amphetamine-and methamphetamineinduced hyperthermia potentially enhances neurotoxicity through both the disruption of protein function, ion channels and enhanced ROS production and through its effects on the vasculature [70]. The induced hyperthermia may lead to transitory breakdowns in the blood-brain barrier, which result in neurodegeneration and neuroinflammation in laboratory animals and brain pathology [70,71]. In a study by Zhou et al., it was reported that the toxicity of 4-chloromethcathinone (4-CMC) and 4-methylmethcathinone (4-MMC) increased with changes in temperature, from 37 to 40.5 • C. According to the authors, the mechanism of mitochondrial toxicity of 4-CMC and 4-MMC, which is increased by hyperthermia, contributes to the neural toxicity of these compounds [72].
Surprisingly, in the present study, the (±)trans-4,4 -DMAR from a neuro-behavioural and physiological point of view was not active per se but was able to enhance the effects of (±)cis-4,4 -DMAR and increased its toxicity. At present, no in vivo pharmacological studies have been performed with the (±)trans-4,4 -DMAR stereoisomer whilst only one in vitro pharmacodynamic study has been published showing that (±)trans-4,4 -DMAR is a substrate-type releaser at the DAT, NET transporters and a SERT uptake blocker in rat brain tissue with a lower potency than the (±)cis-4,4 -DMAR stereoisomer [14]. However, in a recent pharmacokinetic study it was found that whilst the (±)cis-4,4 -DMAR compound is metabolised and excreted in the mouse urine, the (±)trans-4,4 -DMAR stereoisomer is not metabolised [84] suggesting a stereoselective metabolism and disposition of 4,4 -DMAR. This is common in stereoisomers of a drug that generally metabolised with different metabolic profile [85] both in pharmaceutical [86] and in abusing drugs [87,88].
Furthermore, the (±)trans-4,4 -DMAR may have a different binding to plasma or tissue proteins and transporters that is common in chiral drugs [89]. Its lack of pharmacological activity could, therefore, be due to the low availability of the compound at the central level. Further studies will need to be undertaken to investigate this aspect. However, the present study demonstrated that the (±)trans-4,4 -DMAR, although inactive from a neuro-behavioural and physiological point of view, caused a lower urinary excretion of metabolites (M1 and M2) of the (±)cis-4,4 -DMAR compound, causing an increase in the (±)cis-4,4 -DMAR itself ( Figure 7B). This results of inhibition after co-administration of a mixture of both stereoisomers was in accordance with those obtained in similar studies [88,90,91]. This may suggest that the (±)trans-4,4 -DMAR could inhibit the (±)cis-4,4 -DMAR metabolism, and thus increasing its bioavailability levels. This hypothesis is consistent with the worsening of the physiological and neuro-behavioural alterations observed in the same animals from which the urinary excretion studies were made ( Table 4). The worsening of the pharmaco-toxicological effects resulting from the co-administration of (±)cis-4,4 -DMAR and (±)trans-4,4 -DMAR was dramatic (Table 3) and demonstrated the great danger of taking these stereoisomers in mixture, as unfortunately could well occur with preparations purchased from rogue web sites. The availability of mixture of two chiral compounds to obtain an enhancement of the psychoactive effects has already been described and observed for other NPS, in particular with synthetic cannabinoids [92].

Immunoistochemical Studies
This was the first study to investigate the 4,4 -DMAR-induced brain damage. The results of in vivo studies suggested the occurrence of phenomena related to CNS alterations (e.g., hyperthermia; seizures). By analogy to other stimulant drugs (e.g., MDMA [23] and cocaine [93]) we hypothesised that the oxidative and/or nitrosative stress occurred in the brain following the intake of 4,4 -DMAR.
Hence, we investigated the expression of 8-OHdG, a marker of ROS-induced oxidative damage to DNA [94]. We observed that 8-OHdG was overexpressed in mice treated with 4-4 -DMAR, and especially so at higher dosages ((±)cis 30, (±)cis 60 and (±)trans 30 + (±)cis 60). These findings may well suggest that the administration of 4-4 -DMAR caused here levels of ROS-mediated DNA damage. We, therefore, evaluated the expression of NOX2, one of the main ROS-producing enzymes, which instead was not increased in treated mice compared to the controls. One could that conclude that in the mechanism of 4-4 -DMARinduced ROS-mediated DNA damage other pathways are involved, which will need to be investigated in future studies. Among these, there could be the production of superoxide by iNOS, a phenomenon that can occur in conditions of oxidative stress [95]. This enzyme, which in the neurons is responsible for the production of a share of NO, under stress conditions can produce NO at a high rate, leading to the production of peroxynitrite, a very toxic oxidant and nitrating agent. We observed that iNOS expression in treated mice is greater than in controls, in particular at the dose of (±)cis 60 and (±)trans 30 + (±)cis 60, there is a large increase in iNOS expression (p < 0.001).
It was, therefore, possible to expect damage derived from reactive nitrogen species (RNS). So, we investigated the expression of NT, known to be a direct marker of nitrosative stress, being the product of tyrosine nitration mediated by RNS such as peroxynitrite anion and nitrogen dioxide, formed in the presence of NO [96]. We observed a statistically significant increase in NT at the (±)cis 30 and (±)trans 30 + (±)cis 60 doses, but not at the (±)cis 60 dose. At the dose of (±)cis 10, NT is even reduced compared to controls. From the analysis of these data, it is possible to hypothesise the occurrence of nitrosative stress following the administration of 4-4 -DMAR, however, it is not possible to draw conclusions on the effect this has on cellular DNA.
The results reported in the study were related to the analysis of the frontal cortex, however, from a preliminary analysis we did not detect changes in specific areas, but instead changes were widespread in the CNS regions examined (e.g., cortex, striatum, hippocampus, cerebellum).
Having observed the occurrence of hyperthermia in mice treated with 4-4 -DMAR, we investigated some heat shock proteins (HSP27 [24], HSP70 [25], HSP90 [26]), which have been shown to be correlated with thermal damage response. HSP27 was overexpressed in mice treated at dosages of (±)cis 30, (±)cis 60 and (±)trans 30 + (±)cis 60. Interestingly, there was a significant increase in HSP27 expression in mice treated with (±)trans 30 + (±)cis 10 and (±)trans 30 + (±)cis 60 compared to (±)cis 10 and (±)cis 60. This trend, which was consistent with the respective increments in temperature recorded in vivo, may suggest a greater potential of the (±)cis-(±)trans mixture in inducing hyperthermia and the consequent adaptive cellular response. HSP70 showed a similar trend, with the exception that the administration of (±)trans 30 + (±)cis 60 was not associated with HSP70 expression compared to (±)cis 60. One possible explanation is that HSP27 showed an early increase upon thermal stimulus compared to HSP70, and the mice treated with the (±)cis 60 assays all died rapidly, not allowing full expression of HSP70 [25]. Regarding HSP90, we observed a reduced expression in mice treated with 4-4 -DMAR. This trend is difficult to explain, as an increase in expression could reasonably have been expected, similarly to other HSPs. HSP90 is physiologically abundant within the cell and is important in the formation and function of several protein complexes that maintain cell homeostasis. In addition, it plays an important role in the thermal stress response [97]. Its hypo-expression could, therefore, reduce the cell's ability to resist to 4-4 -DMAR-induced thermal and oxidative stress.
Thermic damage to cells can induce one of two opposing responses: apoptosis, that removes damaged cells to prevent inflammation and the heat shock response, to maintain cell survival. The modulation activity of apoptosis mechanisms exerted by HSPs is well known [98]. In particular, HSP70 acts on several targets, including JNK, preventing the release of cytochrome c from the mitochondria. JNK [99] is in turn involved in the release of Smac/DIABLO. HSP27 [100] also appeared to have an inhibitory role on the release of Smac/DIABLO. Indeed, Smac/DIABLO binds with the inhibitor of apoptosis proteins (IAPs), thus freeing caspases to activate apoptosis [101]. Our evaluation of Smac/DIABLO expression showed a significant increase at the doses of (±)cis 30, (±)cis 60 and (±)trans 30 + (±)cis 60. Furthermore, the immunoreactivity of Smac/DIABLO is greater in subjects treated with (±)trans 30 + (±)cis 60 compared to (±)cis 60. One could then conclude that high doses of 4-4 -DMAR have a pro-apoptotic effect. These data are consistent both with the trend of radical stress markers (in particular with 8-OHdG and iNOS), and with the expression of HSP27 and HSP70. It is not possible, with the current data, to determine whether 4-4 -DMAR induces oxidative/nitrosative stress per se or whether this was secondary to hyperthermia. These two phenomena are closely related considering, for example, that in conditions of hyperthermia, iNOS contributes to NO-dependent apoptosis [102].
ROS have been reported to both activate and repress NF-κB signalling. For instance, ROS often stimulates the NF-κB pathway in the cytoplasm but inhibits NF-κB activity in the nucleus [103]. In most cases the expression of NF-κB target genes typically promotes cellular survival, including those that antagonise the effects of ROS [104]. In light of this, our observation of NF-κB overexpression in 4-4 -DMAR-treated mice is not surprising. Regarding histological analyses, the finding of mild cerebral edema is consistent with what has been observed in fatalities associated with 4,4 -DMAR [9].

Behavioural Studies: Physiological and Neuro-Behavioural Responses
The effects of (±)cis-4,4 -DMAR, (±)trans-4,4 -DMAR and their co-administration were investigated by a protocol widely used in studies of "safety pharmacology" for the preclinical characterization of new molecules in rodents [106][107][108][109]. This protocol includes a series of observational behavioural tests carried out in a consecutive manner that monitor the animal responses up to 5 h after compound injections. In the present study, we reported the observation of physiological (sweating, salivation, core temperature) and neuro-behavioural (detection of psychomotor agitation, spontaneous and stimulated aggressiveness, convulsions) changes occurring in mice after injection of (±)cis-4,4 -DMAR, (±)trans-4,4 -DMAR and their co-administration. Before pharmacological treatment the mouse was placed on a square plate (30 cm× 30 cm) raised from the ground (20 cm) and left free to move on the plate for 10 min (habituation period). The animal was then injected with the vehicle, substances alone ((±)cis-4,4 -DMAR or (±)trans-4,4 -DMAR) or in cis + trans co-administration and placed on the plate. From this moment, the recording time of the physiological and neuro-behavioural responses started, and these were detected in 5-min sessions.
After the first 5 min, the animal was removed from the plate; the body core temperature was then measured, and the aggression tests were performed. The 5-min observation period was repeated at 15-30-60-90-120-180-240-300. Psychomotor agitation in the mouse was not simply related to an increase in spontaneous locomotion itself, but by an "unusual hyperactivity" which was characterised by rapid and sudden horizontal displacements in all directions, sustained turning behaviour, involuntary falls from the high plate during the evaluation of the tests, stereotyped movements and rearing activity. Mouse sweating and salivation were reported only if present or not. The core temperature was determined using a probe (1 mm diameter) that was gently inserted after lubrication with liquid Vaseline into the rectum of the mouse (to about 2 cm) and left in position until the temperature stabilised (about 10 s) [106]. The probe was connected to a digital thermometer. Spontaneous aggressive response was measured based on the number of times a mouse bit a grey cloth put in front of its snout. During the test, each mouse was free to move within the cage. In the case of stimulated aggressiveness, each mouse was manually restrained and held in a supine position following which an object was brought near the mouth. For both spontaneous and stimulated aggressive behaviour tests, a grey cloth was placed in front of the nose of each mouse 10 consecutive times (score 0/10 not aggressive, score 10/10 very aggressive). Convulsions were defined as a loss of righting reflexes for at least 5 s, combined with the presence of body tremors and clonic or tonic limb movements [110,111]. Deaths caused by 4-4'-DMAR injection occurring during the observation period (5 h) were also reported as frequency (%) and time of death (min). Experiments were conducted inside the LARP (Centralised Preclinical Research Laboratory, University of Ferrara, Ferrara, Italy) in a thermo-stated (temperature 20-22 • C, humidity about 45-55%) and light controlled (about 150 lux) room, in which there was a background noise of about 40 ± 4 dB. The experiments, conducted in blind by trained observers working together in pairs [92], were videotaped and eventually analysed off-line by a different trained operator. The experiments were performed between 8:30 a.m. and 2:00 p.m.

Collection of Tissue Samples
Following the acute death or the sacrifice of the animals (after 5-h observation period by dislocation of the spine), brains were taken. These were fixed in 4% buffered formalin and then sampled and incorporated into the paraffin (Leica ASP300 processor, Histoline TEC2900 incorporator; Leica Microsystems Srl, Buccinasco, Milano, Italy). Subsequently, the cut (performed with the Leica HistoCore autocut microtome; Leica Microsystems Srl, Buccinasco, Milano, Italy) was designed to create sections with a thickness of 5 µm for each sample.

Histological Procedure
For histological investigations, the sections were stained with haematoxylin/eosin and then observed under optical microscope (Nikon Eclipse E90i; Nikon, Roma, Italy).

Immunohistochemical Procedure
On 5-µm-thick paraffined lung sections, we evaluated the expression of a panel of markers: heat shock proteins (HSP27, HSP70, HSP90), SMAC/DIABLO, NF-kB, iNOS, NOX-2, NT, 8-OHDG. Dilution of antibodies and pre-treatments necessary for antigen retrieval are shown in Table 6. We used a detection system composed of a biotinylated secondary antibody and HRPconjugated streptavidin (4plus HRP Universal Detection, Biocare Medical). 3 -,3Diaminobenzidine (DAB) and H2O2 (Betazoid DAB Chromogen Kit, Biocare Medical, Concord, CA, USA) were used as chromogen/substrate. The subsequent counterstaining with haematoxylin-eosin allowed the visualization of cell morphology and nuclei. Once this procedure was completed, the slides were mounted and observed under optical microscope. In the pictures (sections of the frontal cortex;  to evaluate the variation in the expression of markers (Smac, NOX2, iNOS, 8-OHdG, NT, HSP27, HSP70, HSP90) following treatments, the change in intensity of brown colour (staining of the immunohistochemical technique) in the control and treated tissue was measured. Quantification of Smac, NOX2, iNOS, 8-OHdG, NT, HSP27, HSP70 and HSP90 positive-stained areas was performed by the ImageJ software (imagej.nih.gov/ij/ Accessed date 12 May 2020). One image for each animal of the different experimental groups was processed. Positivity was expressed as an extension of the stained analysed area.

Dose and Sample Collection
To study the possible metabolic interaction between cis and trans isomers, we chose to evaluate urinary excretion and metabolic profile of 4,4 -DMAR, as this allowed us to study and correlate physiological and neuro-behavioural responses in the same individual. For the in vivo studies, three different groups of mice were selected. To the first group, a dose of 10 mg/kg of cis-4,4 -DMAR was administered. To the second group, a dose of 10 mg/kg of both cis and trans-4,4 -DMAR was administered. The third group was selected for the collection of urine blank samples. Urine samples were collected in the range of 0-6 h after 4,4 -DMAR or vehicle were intraperitoneally injected.

Excretion Studies
Urine samples were treated and analysed through a protocol already used by our group for the analysis of the excretion of stimulant drugs in mice [84,112]. Briefly, our protocol allowed the conversion of conjugated metabolites (i.e., sulfo-and glucorono-conjugates) to phase I metabolites after two hydrolysis steps. The first employed β-glucuronidase for the hydrolysis of glucorono-conjugates (working pH of the enzyme mixture 7.4). The second used a mixture of β-glucuronidase/arylsulfatase for the hydrolysis of sulfo-conjugates (working pH of the enzyme mixture 5). Hydrolysed samples were extracted adding to the samples carbonate buffer (final pH 11) and 7 mL of ethyl acetate. The organic layer was next evaporated to dryness under nitrogen stream at a temperature of 30 • C. The residue was reconstituted with 50 µL of mobile phase and analysed with a targeted LC-MS/MS technique. In details, samples were analysed using an Agilent 1200 series HPLC instrument equipped with a SUPELCO Discovery C18 column (15 cm × 2.1 mm × 5 µm) coupled with an API4000 QqQ mass spectrometer (Sciex, 500 Old Connecticut Path, Framingham, MA, USA). The chromatographic and mass spectrometer parameters were described in the above cited articles.

Data and Statistical Analysis
Sweating and salivation were expressed as frequency (e.g., % of animals that developed symptoms); psychomotor agitation was expressed as frequency (% of animals) and duration (total time in mins). Core temperature values were expressed as the difference between control temperature (before injection) and temperature following drug administration (∆ • C). Aggressiveness was expressed as frequency (% of animals which become aggressive) and score (number of bites). Convulsions were expressed as frequency (% of animals that developed seizure), episodes (e.g., number of seizure events), latency of first episode (sec.) and duration of each episode (sec.). Lethality was expressed as frequency (% of animals that died during the observation period (5 h) or during the following 24 h). The statistical analysis of the effects of the individual substances in different concentrations were performed using a one-way ANOVA followed by a Bonferroni test for multiple comparisons. A Student's t-test was used to determine statistical significance (p < 0.05) between two groups. Regarding immunohistochemical analysis, positivity was expressed as the extension of stained analysed area. Results were expressed as means ± mean standard error (SEM). Data were analysed by one-way analysis of variance (ANOVA), followed by Tukey's post-hoc test. For all tests, a p value of <0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism 8 software for Windows (La Jolla, CA, USA).

Conclusions
Indeed, consistent with current findings one would conclude that some of the stimulant NPS such as the 4,4 -DMAR molecules here thoroughly investigated could putatively exhibit very severe clinical toxicity levels. The clinical-toxicological relevance of the study is due to the use of doses of 4-4 -DMAR in mice that are equivalent to the dose of 4-4'-DMAR in humans, evoking mild, intermediate and strong behavioural and physiological responses ( Table 5). The current drug scenario is changing very rapidly; the earliest of the possible appearance of a new substance might be evidenced on the deep, followed by a migration to the open, web. Eventually, the NPS would move into 'head shops' and/or the 'street' market, then reported by formal early warning systems, and new legislation would be implemented to counter the index NPS/substance [113]. Hence, an approach aiming at describing what is being discussed online by the web-based NPS enthusiasts 'e-psychonauts' [114] has been considered as potentially useful to identify in advance the NPS availability, market and diffusion [115]. To improve accuracy and provide a thorough evaluation of NPS pharmacology, further research should focus on an integrative model in which web-based analyses will be combined with more advanced research approaches. From this perspective, our currently ongoing related quantitative structure activity relationship (QSAR), docking and in silico studies will hopefully provide important findings in terms of which NPS, within a given class (e.g., stimulants; novel synthetic opioids; novel benzodiazepines) will present with higher levels of receptor affinities, and hence clinical potency. These data, taken from selected molecules, could then be used to plan further in vitro and in vivo/preclinical studies. Clinicians should be regularly informed about the range of NPS, their intake modalities, their psychoactive sought after-effects, the idiosyncratic psychotropics' combinations, and finally their medical, psychobiological and psychopathological risks. Funding: This research has been funded by the Drug Policies Department, Presidency of the Council of Ministers, Italy (project: "Effects of NPS: development of a multicentric research for the information enhancement of the Early Warning System" to MM), by local funds from the University of Ferrara (FAR 2019 and FAR 2020 to MM), by FIRB 2012 from the Italian Ministry of Education, University and Research (Grant no. RBFR12LDOW to F. De-Giorgio) and by local funds from the Catholic University of Rome (Linea D1 grants to F. De-Giorgio). All applicable international, national and/or institutional guidelines for the care and use of animals were followed. All procedures performed in the studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were conducted.

Institutional Review Board Statement:
The study was conducted according to the guidelines of the Declaration of Helsinki, and approved Ministry of Health (protocol code 335/2016-PR 1/4/2016).

Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the first (Micaela Tirri) and corresponding author (Matteo Marti) for researchers of academic institutes who meet the criteria for access to the confidential data.

Conflicts of Interest:
The authors declare no conflict of interest.
Ethical Statements: All applicable international, national and/or institutional guidelines for the care and use of animals were followed. All procedures performed in the studies involving animals were in accordance with the ethical standards of the institution or practice at which the studies were