Sex- and age- dependent effect of pre-gestational chronic stress and mirtazapine treatment on neurobehavioral development of offspring

Hormonal fluctuations, such as the perinatal period, may increase susceptibility of women to depression, which in turn exert a negative impact on child’s neurodevelopment, becoming a risk factor in development of neuropsychiatric disorders. Moreover, the use of antidepressants during this critical period presents a serious health concern for both the mother and the child, due to the consequences of treatment in terms of the reliability and safety for the proper neurodevelopment of the organism being not well known. Atypical antidepressants, such as mirtazapine, that targets both serotonergic and noradrenergic systems in the central nervous system (CNS), represent a novel focus of research due to its unique pharmacological profile. The aim of this work was to study the effects of maternal depression and/or perinatal antidepressant mirtazapine treatment on the neurobehavioral development of the offspring. Pre-gestationally chronically stressed or non-stressed Wistar rat dams were treated with either mirtazapine (10 mg/kg/day) or vehicle during pregnancy and lactation followed by analysis of offspring’s behavior at juvenile and adolescent age. We found mirtazapine induced alterations of nursing behavior. In offspring, pregestational stress (PS) had an anxiogenic effect on adolescent males and increased their active behavior in forced swim test. Interaction between pregestational stress and mirtazapine treatment variously induced anxiolytic changes of juvenile and adolescent females and impairment of spatial memory in adolescent females as well. Hippocampal density of synaptophysin, pre-synaptic protein marker, was decreased mainly by mirtazapine treatment. In conclusion, our results show mirtazapine induced alterations in maternal behavior and several sex- and age-dependent changes in neurobehavioral development of offspring caused by both prenatal mirtazapine treatment and/or chronic pregestational stress.


24
represent a novel focus of research due to its unique pharmacological profile. The aim of this work was to study 25 the effects of maternal depression and/or perinatal antidepressant mirtazapine treatment on the neurobehavioral 26 development of the offspring. Pre-gestationally chronically stressed or non-stressed Wistar rat dams were treated 27 with either mirtazapine (10 mg/kg/day) or vehicle during pregnancy and lactation followed by analysis of 28 offspring's behavior at juvenile and adolescent age. We found mirtazapine induced alterations of nursing behavior.

29
In offspring, pregestational stress (PS) had an anxiogenic effect on adolescent males and increased their active 30 behavior in forced swim test. Interaction between pregestational stress and mirtazapine treatment variously 31 induced anxiolytic changes of juvenile and adolescent females and impairment of spatial memory in adolescent 32 females as well. Hippocampal density of synaptophysin, pre-synaptic protein marker, was decreased mainly by 33 mirtazapine treatment. In conclusion, our results show mirtazapine induced alterations in maternal behavior and 34 several sex-and age-dependent changes in neurobehavioral development of offspring caused by both prenatal 35 mirtazapine treatment and/or chronic pregestational stress.

40
An estimation of 17% men and 25% women experience an episode of major depressive disorder (MDD) at least once 41 in their life (1). Higher susceptibility of women to depression may arise from the increased vulnerability caused by 42 periods of hormonal fluctuation, such as the perinatal period (2,3). Perinatal depression has been reported to exert 43 a negative impact in children neurodevelopment, becoming a risk factor in developing neuropsychiatric disorders 44 (4-6).

45
Second generation antidepressants (SGA), such as selective serotonin reuptake inhibitors (SSRIs) (fluoxetine, 46 sertraline) or serotonin-norepinephrine reuptake inhibitors (SNRIs) (venlafaxine) introduced in Europe in the 1980s, 47 are recorded as the first line antidepressant treatment during pregnancy by most medication guides (7,8). The World 48 Health Organization (WHO) estimates that depression is a leading cause of disability worldwide (9). However, the 49 main concern for specialists is the impact of the treatment on the developing fetus/child, namely risk of potential 50 congenital malformations, neonatal withdrawal symptoms or poor neonatal adaptation syndrome as well as long-51 term neurodevelopmental consequences (10)(11)(12)(13). In addition, the delay in treatment efficacy and the presence of 52 many side effects lead researchers to investigate alterative antidepressants with better efficacy, faster onset of 53 action and lesser counterproductive reactions. 54 Mirtazapine (MIR) (Fig 1) has a tetra-cyclic chemical structure with molecular weight of 265.36 and belongs to the 55 piperazino-azepine group of compounds. It is a new generation antidepressant that has a different mechanism of 56 action than SGAs, targeting both the serotonergic and noradrenergic systems in the CNS. It is a noradrenaline (NE) 57 and specific serotonin (5-HT) antidepressant (NaSSA) that acts as an antagonist at central α2-adrenergic inhibitory 58 autoreceptors and heteroreceptors, as well as at the 5-HT 2 and 5-HT 3 receptors. MIR enhances the release and 59 availability of NE by blocking presynaptic inhibitory α2-autoreceptors and enhances the 5-HT release by antagonism 60 of the α2-heteroreceptors in the serotonergic nerve terminals and simultaneously blockade of postsynaptic 5-HT 2 61 and 5-HT 3 receptors. Thanks to this double activity, MIR is suggested to induce earlier onset of antidepressant effects 62 avoiding the serotonergic related side effects such as high body temperature, agitation, increased reflexes, tremor, 63 sweating, dilated pupils, and diarrhea (14). However, it may induce an enhanced body weight gain and sleepiness 64 (15,16). Mirtazapine has low in vitro affinity for central and peripheral dopaminergic, cholinergic, and muscarinic 65 receptors, but high affinity for central and peripheral histamine H1 receptors. However, it appears that the 66 antihistaminergic effects of the drug are counteracted by noradrenergic transmission when the drug is commenced 67 at dosages ≥15 mg/day, i.e. within the recommended dosage range (17). Initial dose of MIR is 15 mg/day orally once 68 a day at bedtime and maintenance dose represents 15 to 45 mg orally once a day (18). Even though MIR seems to 69 represent a plausible alternative of antidepressant medication during gestation and there is no evidence that use of 70 mirtazapine in pregnancy causes birth defects, preterm birth, or low infant birth weight. While the evidence for 71 other pregnancy outcomes is also reassuring, only small numbers of women have been studied and number of 72 animal studies is not sufficient (19).

73
Despite lacking efficient translatability, several animal models, including chronic unpredictable stress, have been 74 investigated for decades to evaluate depression-like behavioral changes and their potential mechanisms of action.

75
Studies in rodents show that chronic pre-gestational and prenatal maternal stress, which disrupt the maternal 76 endocrine, nervous and immune systems, can induce long-term alterations in the synaptic structure and so impact 77 the behavioral outcomes in the offspring (6,20). Prior to regulatory roles of serotonin, norepinephrine and dopamine 78 neurotransmitters in adult brain, monoamines play an important role in the fetal maturation of the brain, such as -79 during neuronal proliferation, migration and differentiation, myelinization and synaptogenesis (21). The exposure 80 to stressful situations early in life may disable the optimal structural and functional development of hippocampus, 81 due to the damaging action of excessive corticosterone concentration and dysregulation of monoamines (11,22).

82
Persisting high levels of stress are thought to result in loss of synapses in circuits underlying affective and cognitive 83 processes. These reductions are presumed to contribute to the symptoms of depression associated with major 84 depressive disorder (23). Synaptophysin is a synaptic vesicle glycoprotein, which immunoreactivity is present in a 85 punctate pattern in the hippocampus and has been used as a presynaptic marker for quantification of synapses (24).

86
The hippocampal formation consists of several histologically distinguishable modules, such as Cornu Ammonis (CA) 87 regions (CA1, CA2, CA3, CA4), dentate gyrus (DG), presubiculum, and subiculum. These regions of the hippocampus 88 are associated with different functions (e.g. memory encoding and retrieval) and may be specifically disrupted in 89 various diseases (25). Granule cells, the main output cells of the DG, send axons (mossy fibers) through CA4 to CA3 90 and innervate a small number of pyramidal cells and a disproportionally large number of interneurons CA3 pyramidal 91 cells then form recurrent excitatory network and send axons to CA1. Theoretical work, as well as anatomical, 92 physiological, and behavioral experiments support the idea that the DG-CA3 system performs the pattern separation 93 and the pattern completion of the inputs to the hippocampus, operations needed for memory encoding and retrieval 94 (26). Purpose of the hippocampus happens to be severely affected by early life stress, predisposing the individual to 95 an impaired reactivity when exposed to adverse environmental stimuli (27). In our previous study, we have shown 96 that pre-gestational maternal stress may affect hippocampus at the time of birth by increasing the resting membrane 97 potential, suppressing depolarization-activated action potential firing, and increasing the spontaneous activity of 98 hippocampal cells from newborn rat offspring (28), but pre-gestational stress induced changes in different 99 subregions of hippocampus are not thoroughly known. However, antidepressant treatment may facilitate the 100 challenged neurogenesis by, upregulating the hippocampal concentration of glucocorticoid receptors, which help to 101 attenuate the hyperactivity of the HPA axis and inducing morphological changes in the neuronal network (11,27,29).

102
Some studies suggest that developmental fluoxetine (SSRI) exposure of prenatally stressed male and female 103 offspring reversed the effect of stress on the number of immature neurons in the dentate gyrus (DG), with effects 104 being more prevalent in adult male offspring (30,31). However, knowledge of mirtazapine treatment's impact on 105 pregnancy and lactation hippocampal neurogenesis of juvenile, adolescent and adult offspring during last days of 106 pregnancy and for following 2 weeks postpartum (PP), is very limited.

107
The aim of the present study was to determine the possible implications that pre-gestational chronic stress have on 108 the behavioral and neurodevelopmental outputs in the offspring of both sexes during juvenile and adolescent age 109 and to investigate the consequences of administration of the new generation antidepressant mirtazapine on these 110 variables aiming our focus on functional brain developments that starts day 10 PP (32-35). Slovak Republic. After 7 days of acclimatization, females were randomly assigned to stress or non-stress groups.

118
Animals in the stress group were exposed to unpredictable stressors of mild intensity for a total of 3 weeks. One 119 week after the end of the stress procedure, females were mated with males in the ratio 3:1. The presence of 120 spermatozoa in vaginal smears was considered day 0 of gestation. On day 15 of gestation, the females were 121 separated and housed individually. The animals had ad libitum access to food pellets and water and were kept in a 122 temperature and humidity-controlled room (20-24°C and relative humidity 50-60%) with 12/12 hours of light/dark

127
One day after birth, litters were culled to four males and four females (with eight offspring per cage). Reproductive 128 variables were recorded right after the birth. The offspring were weaned on post-partum day 21 and housed in litter 129 groups of same-sex four animals per cage. No more than two pups from the same mother per group (n= 6-8 130 animals/group) were used for behavioral testing.

149
Mothers were placed in individual cages for 10 min and each minute 4 pictures were taken from each rat from front 150 with a Nixon professional camera. Animals were tested 5 weeks after CUS. Each picture was evaluated by a score 151 from 0 to 2 according to the following four action units (37): 1. Orbital tightening: rats in pain display narrowing of the orbital area which manifests as partial or complete eye 153 closure or squeezing. 154 2. Nose/cheek flattening: less bulging of nose and cheek and absence of crease between cheek and whisker pads. 155 3. Ear changes: ears in pain tend to fold, curl and angle forwards or outwards, and the space between ears is wider.

166
The FST consists of a container filled with water where the animal is placed and cannot escape. The apparatus 167 consists of a vertical cylindrical glass container (height 45 cm, diameter 25cm) filled with tap water at 23±1°C. The 168 water volume is enough to ensure that the animals can not touch the bottom of the container with their hind paws.

169
The forced swim test was conducted over two days. On the first day, rats were introduced to the cylindrical glass 170 tank filled with water for 15 min (not videotaped), towel-dried and returned to their home cage. Twenty-four hours 171 later, the animals were exposed to the same experimental conditions for 5 min, dried and returned to their home 172 cage. Sessions were videotaped and scored using the software ANYMAZE™ (Stoelting Europa, Co., Ireland). All tests 173 were carried out between 8:00 a.m. to 12:00 p.m. The behavior scored in the forced swim test concerns: (1) 174 immobility-floating with the absence of any movement, (2) latency to be immobile-time duration, (3) swimming, 175 (4) climbing. Offspring was tested at the age of 48 days.  The Y-Maze Test is widely used to assess exploratory behaviors, learning, and memory function in rodents and short-187 term memory (38). The apparatus was made from black Plexiglas (50x16x32 cm) in which the arms were 188 symmetrically separated at 120°. No visual cues were placed inside the maze, but different extra-maze cues were 189 visible from all three arms to enable spatial orientation. During the first trial, each animal could freely explore two 190 arms of the maze for 15 min. and its behavior was recorded with a camera. Subsequently, animals were returned to 191 the home cage for one minute and the maze was cleaned with 70% ethanol. The second trial lasted 5 minutes with 192 animal having the access to all three arms. Sessions were videotaped and scored using the software ANYMAZE™ 193 (Stoelting Europa, Co., Ireland). All tests were carried out between 8:00 a.m. to 12:00 p.m. under dim light condition.

194
The animals were tested at the age of 43 days. Spontaneous alternation behavior (the alternation percentage is 195 calculated by dividing the number of alternations by number of possible triads x 100) is considered to reflect spatial 196 working memory, while the total number of arm entries was considered to reflect spontaneous locomotor activity 197 (39,40).

230
Rat grimace test scores has been used to evaluate spontaneous pain (41). We observed significant main effect of 231 stress on grimace test scores (F (1, 22) = 5.00; p≤0.05). Post-hoc analysis did not reveal any significant differences 232 (Fig 3).

234
There was no significant main effect of stress or mirtazapine on total time dams spent nursing (Fig 4A), however, we

247
In males, we observed marginally significant main effect of stress × mirtazapine interaction in total distance travelled 248 (F (1, 30) = 3.69; p=0.06). Post-hoc analysis showed significantly decreased total distance travelled in stress × vehicle 249 group (p≤0.05) compared to non-stress × vehicle group. Also, trend for decrease in non-stress × mirtazapine (p=0.07) 250 and stress × mirtazapine (p=0.07) groups compared to non-stress × vehicle group (Fig 6A) was present. In females, 251 we did not observe any differences in total distance travelled ( Fig 6A).

279
Post-hoc analysis revealed marginally significant decrease in non-stress × mirtazapine group compared to non-stress 280 × vehicle group (p=0.07) and significant decrease compared to stress × vehicle group (p≤0.05) (Fig 8A). We did not 281 observe any significant changes percentage of spontaneous alterations (Fig 8B).

282
In females, a significant main effect of stress was in total distance travelled (F (1, 25) = 4.95; p≤0.05). Post-hoc 283 analysis showed marginally significant increase of horizontal motor activity in stress × mirtazapine group compared 284 to non-stress × vehicle group (p=0.07) (Fig 8A). Further, we observed significant main effect of mirtazapine (F (1, 25)

302
We evaluated the spontaneous pain of the mothers by the Rat Grimace Scale (RGS) and we found increased facial 303 expression indicating pain in those mothers, which were exposed to stress schedule. RGS was developed to identify 304 acute and inflammatory pain but it can be applied to a wider range of pain types and chronicity (42,43) and this is 305 supported by our data that suggest ongoing influence of CUS even 5 weeks after the CUS procedure.

318
Offspring of mothers that have experienced stress during gestation may have impaired emotional development due 319 to a dysregulation of the HPA axis, leading to a higher risk of developing a cognitive and/or mood disorders in 320 adulthood (28,(51)(52)(53). We evaluated the effect of pregestational stress (PS) exposure on intensity of anxiety-like 321 behavior by the elevated plus-maze test, a validated behavioral paradigm based on rodent's preference of closed to 322 open spaces, with latter presenting a significant anxiety-like behavior inducing condition (54).

323
In this study, we didn't find significant sex-dependent differences in anxiety-like behavior of juvenile rats, what was 324 expected due to unmature reproductive system (55). However, female offspring of stressed untreated mothers 325 exhibited decreased anxiety-like behavior that was not present if mothers were treated with MIR. This was changed 326 in adolescence, when females showed decreased anxiety-like behavior if mothers were stressed and treated with 327 MIR. Different results were observed in adolescent male rats, who displayed decreased anxiety-like behavior due to 328 pregestational stress, but mirtazapine treatment of mothers ameliorated this effect. Changes in anxiety-like 329 behavior, due to maternal stress, were previously described by several research groups (28,56). Although there are 330 controversial results regarding effect of chronic pregestational stress on anxiety-like behavior of offspring, possibly 331 due to the different stress paradigm strategies, a great proportion of studies postulate that male offspring whose 332 mothers were submitted to stress before or during pregnancy are more reluctant to enter and spent time  of offspring whose mothers were exposed to high levels of corticosterone while pregnant (73). Also, Vázquez et al.

357
suggested that increased levels of serotonin in control animals after FST may be responsible for shorter time spent 358 immobile and increased time spent swimming, corroborating an effect of the serotonergic system, which in turn 359 modulates dopaminergic projections associated with the reward system and the coping with stress. In this terms, 360 lower levels of serotonin were linked to a stress-induced anhedonic state, while higher levels promote a non-361 anhedonic state (52). Accordingly, higher levels of serotonin may be a plausible explanation to our results.

363
We encountered sex-dependent differences in the Y-maze, a behavioral paradigm set to assess the spatial memory 364 performance building on the natural inclination of rodents to explore new environments. Male offspring from 365 mirtazapine treated mothers showed reduced locomotive activity, indicating an influence of treatment on executive 366 capacity. On the other hand, females seemed to be rather affected by the pregestational stress experience of the 367 mothers, which induced a hyperactivity-like response. Such a result may be reflective of hyperactivity or even ADHD 368 described in children of stressed mothers while pregnant (74,75). However, sex differences in reactivity to either 369 pre-gestational stress or perinatal mirtazapine treatment related to potentially hyperactive behavior require further 370 investigation. In children, PS is connected to cognitive, behavioral, and emotional problems such autism and ADHD 371 (76,77). The maternal stress, manifesting as increased cortisol and corticotropin-releasing factor (CRF) levels, affects 372 the fetal development by reprogramming the HPA axis, leading to impaired memory and learning due to the long-373 lasting effects in the hippocampus (78). Moreover, Deminière et al. had previously described an increased locomotor 374 reactivity to novelty in litters of pre-gestationally stressed mothers, which authors associated with a modified 375 dopaminergic activity in the prefrontal cortex and nucleus accumbens (79). Experience-dependent plasticity of 376 hippocampal neurons and adult neurogenesis, processes of importance for optimal learning or memory formation 377 and processing, are modulated through epigenetic mechanisms, which induce long-lasting changes determining the 378 ability of individuals to cope with adverse situations (80,81).

379
In our study, adolescent females had impaired spatial memory, represented by decreased percentage of 380 spontaneous alternations, as a result of maternal stress as well as mirtazapine treatment. However, this effect was 381 not present in males. Conrad et al. studied the effects of chronic stress on spatial memory performance of adult rats 382 concluding that stress impaired the spatial learning and memory, hippocampi-dependent spatial tasks. Authors also 383 resolved that females seem to be less vulnerable to these hippocampi-dependent memory deficits (38,82).

384
Nevertheless, they didn't study the effects in the offspring. Otherwise, female rats have been linked to more active 385 response in novel environments (30,83,84) and, in line with our results, pregestational stress as well antidepressant 386 treatment highlighted this response with effect of mirtazapine being more pronounced in animals without 387 pregestational stress exposure. Similar results have been postulated in association with other antidepressant such 388 as fluoxetine, bupropion or citalopram (74,85). Nevertheless, our results show a clear influence of pre-gestational 389 stress and antidepressant treatment in the onset of behavioral tasks, which may be driven by the differential 390 strategies used by each sex to cope with adversity. Moreover, we could take into account that we evaluated animals 391 during adolescence, which is known as a time of increased physiological and psychological changes, which make 392 them more vulnerable and unpredictable to external influences (86-88).

394
Synaptophysin is used as a marker of synaptic plasticity and synaptic nerve terminal density (89). We analyzed optical 395 density (OD) of synaptophysin, a calcium-binding glycoprotein widely distributed in the presynaptic vesicle 396 membrane, that is required for vesicle fusion and neurotransmitter release, in the CA3, CA4 and DG areas of the 397 hippocampus. Results of synaptophysin density in our study were age-dependent with no changes in juvenile 398 offspring. However, with onset of adolescence we observed prominent effect of sex with synaptophysin OD males 399 remaining intact under all conditions while females being influenced by maternal mirtazapine treatment. This effect 400 was regionally specific as changes were observed only in CA3 and CA4 areas but not in DG. Sex differences can be 401 seen in the neural plasticity at DG-CA3 synapses. In short, mossy fibers evoke larger population spikes in CA3 402 pyramidal neurons in females during proestrus and estrus relative to males, while mossy fibers in males have 403 stronger synaptic connections to CA3 neurons than females (90). Limitation of our study is that we didn't evaluate 404 estrus cycle phase before extraction of the brains. There is very limited research available on maternal 405 antidepressant treatment in association with synaptophysin OD. Fluoxetine, a SSRI, has been proven to impact 406 synaptophysin OD of CA3 in the same sex-and age-dependent manner as seen in our study (31). Previous studies 407 have encountered a reduced expression of synaptophysin in the hippocampus as whole or in DG, no research is done 408 on other specific regions, at different ages (PP7, PP14, adult) of pre-gestationally stressed pups (91-93), suggesting 409 the involvement of early-life stress in modulation of synaptic plasticity, which could lead to development of neuronal 410 deficits during adulthood (94,95). Previous studies suggested a regulatory role of ovarian hormones in the 411 developing brain (94) explaining the differential effect of stress on limbic synaptic plasticity in female rats. However, 412 there are differences between our study compared to studies previously mentioned. Our model comprises of CUS 413 prior to gestation and we tested the offspring in adolescent age, when the reproductive system of the offspring is 414 not fully developed, so we can't rule out other stress related changes in later life that were not seen in this study.