Life Sciences 149 (2016) 129–137 Contents lists available at ScienceDirect Life Sciences j ourna l homepage: www.e lsev ie r .com/ locate / l i fesc ie Prenatal lipopolysaccharide exposure affects sexual dimorphism in different germlines of mice with a depressive phenotype Thiago M. Reis-Silva a,b, Daniel W.H. Cohn b,c, Thaísa M. Sandini c, Mariana S.B. Udo c, Elizabeth Teodorov d, Maria Martha Bernardi d,e,⁎ a Department of Neuroscience and Behavior, Psychology Institute, University of São Paulo, Brazil b Neuroimmunomodulation Research Group, School of Veterinary Medicine, University of São Paulo, Brazil c Department of Pathology, School of Veterinary Medicine, University of São Paulo, Brazil d Mathematic, Computation and Cognition Center, Federal University of ABC, Brazil e Post-Graduate Program of Environmental and Experimental Pathology and Post-Graduate Program of Dentistry, Paulista University, Brazil ⁎ Corresponding author at: Rua Dr. Bacelar, 1212, Vila C 002, Brazil. E-mail address: marthabernardi@gmail.com (M.M. Be http://dx.doi.org/10.1016/j.lfs.2016.02.068 0024-3205/© 2016 Elsevier Inc. All rights reserved. a b s t r a c t a r t i c l e i n f o Article history: Received 7 August 2015 Received in revised form 14 February 2016 Accepted 16 February 2016 Available online 17 February 2016 The objective of the present study was to investigate whether prenatal lipopolysaccharide (LPS) administration modifies the expression of depressive and non-depressive-like behavior inmale and femalemice across two gen- erations. The sexual dimorphism of thesemicewas also examined in the open-field test. Male and femalemice of the parental (F0) generation were selected for depressive- or non-depressive-like behavioral profiles using the tail suspension test (TST). Animals with similar profiles were matched for further mating. On gestation day (GD) 15, pregnant F0 mice received LPS (100 μg/kg, i.p.) and were allowed to nurture their offspring freely. Adult male and female of the F1 generation were then selected according to behavioral profiles and observed in the open field. Male and female mice of the two behavioral profiles were then mated to obtain the F2 gener- ation. Adults from the F2 generation were also behaviorally phenotyped, and open field behavior was assessed. Male mice that were selected for depressive- and non-depressive-like behaviors and treated or not with LPS in the parental generation exhibited similar proportions of behavioral profiles in both filial lines, but LPS exposure increased the number of depressive-like behavior. An effect of genderwas observed in the F1 and F2 generations, inwhichmalemiceweremore sensitive to the intergenerational effects of LPS in the TST. These data indicate that prenatal LPS exposure on GD15 in the F0 generation influenced the transmission of depressive- and non- depressive-like behavior across filial lines, with sexual dimorphism between phenotypes. © 2016 Elsevier Inc. All rights reserved. Keywords: Endotoxin Lipopolysaccharide Tail suspension test Prenatal Generations 1. Introduction Major Depressive Disorder (MDD) is a worldwide disease that af- fects over 340 million people and generates a high burden to society. It is characterized by a wide range of symptoms, such as feelings of worthlessness and tiredness, sleep disturbances, thoughts of death, and inmany cases suicide [7]. TheWorld Health Organization estimated that depressionwill rank second in the next decade as the disease that is most responsible for premature life lost among all ages and sexes [43]. Environmental stressors and lipopolysaccharide (LPS) can trigger nonspecific immune events [13,16]. Lipopolysaccharide, also known as lipoglycans and endotoxin, is a large molecule that consists of a lipid and a polysaccharide that is composed of O-antigen and outer and inner cores that are joined by a covalent bond. Lipopolysaccharide is found in the outer membrane of Gram-negative bacteria and elicits strong immune responses in animals. By activating the immune system lementino, São Paulo, SP 04026- rnardi). similarly to infections, LPS can cause several behavioral changes (e.g., an increase in slow-wave sleep, anorexia, and depressive-like behavior) that may be associatedwithMDD and are referred to as sickness behav- ior [8,18]. From an immune-neuroendocrine perspective, LPS triggers the secretion of proinflammatory cytokines by the innate immune sys- tem and also increases the activity of the hypothalamic-adrenal- pituitary (HPA) axis, an important system that is known to be respon- sive to environmental stressors and infections [25,29,31,52] and has been reported to be related to MDD [40,59,61]. Depressive phenotypes can be studied in mice using different tests. Although they cannot precisely reproduce human psychopathology, much can be clarified using such tools [18,38]. DifferentMDDsymptoms can be assessed by the learned helplessness paradigm and different tests that employ uncontrollable and inescapable stress, such as behav- ioral despair in the tail suspension test (TST; [6,27]). Strain differences have been observed in the TST. High levels of immobility can be selected among animals, thus providing mice that are more sensitive for differ- ent studies. The results of the TST are easily reproducible, in contrast to othermodels that employ acute inescapable stress, such as the forced swim test (FST; [47]. Additionally, the TST avoids the possible water- http://crossmark.crossref.org/dialog/?doi=10.1016/j.lfs.2016.02.068&domain=pdf mailto:marthabernardi@gmail.com Journal logo http://dx.doi.org/10.1016/j.lfs.2016.02.068 Unlabelled image www.elsevier.com/locate/lifescie 130 T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 induced hypothermic conditions that are associated with the FST [9]. Because of its ability to detect numerous antidepressant-like effects, the TST has become popular for the rapid screening of antidepressant drugs [34]. Previous studies from our laboratory reported the intergenerational effects ofmaternal LPS exposure [22–24,45,51]. In the present study, we selected mice with different behavioral profiles (i.e., depressive- and non-depressive-like behavior) in the TST by considering the following factors: (1) Major Depressive Disorder may be related to the release of inflammatory mediators, (2) the TST is an adequate tool for assessing depressive-like behavior, and (3)maternal LPS exposure impacts subse- quent generations. The objective of the present studywas to investigate whether a single prenatal LPS administration in the parental generation affects these behaviors in subsequent generations of mice in the TST. The mice were also evaluated in the open field test (OFT) because LPS may also induce alterations in the sexual dimorphism of open field be- havior that was reported previously [3]. The dose of LPS used here was shown to induce sickness behavior [3], which is related to depres- sive illness [12]. 2. Methods 2.1. Ethics statement All of the animals that were used in the present study were main- tained in accordance with the Guide for the Care and Use of Laboratory Animals, National Research Council, USA. The protocols for the experi- mental studies were approved by Laboratory Animal Resources, School of Veterinary Medicine, and University of São Paulo, Brazil (protocol no. 2653/2012, FMVZ-USP). These guidelines are similar to those of the United States National Institutes of Health. The experiments were per- formed in accordancewith good laboratory practice protocols and qual- ity assurance methods. All efforts were made to minimize the suffering of the animals. 2.2. Animals A total of 40 male and 50 female Swiss mice, 6 weeks of age and weighing 25–30 g, were obtained from the Department of Pathology, School of the Veterinary Medicine, University of São Paulo, Brazil. Seven days before beginning the experiments, the mice were housed in groups of five in polypropylene cages (28 × 17 × 12 cm) under con- trolled room temperature (20–25 °C) and humidity (55–65%) with a 12 h/12 h light/dark cycle (lights on at 7:00 AM). The mice received standard rodent chow and water ad libitum. 2.3. Lipopolysaccharide Lipopolysaccharide (from Escherichia coli; Sigma, St. Louis, MO, USA; serotype 0127:B8) was dissolved in sterile saline (50 μg/ml LPS in 0.9% NaCl solution), and a single 100 μg/kg dose was administered intraper- itoneally exclusively to pregnant dams of the parental generation (F0) on gestational day 15 (GD15). This dose was chosen because it elicits sickness behavior, induces endocrine alterations in dams, and increases cytokines at the placental level [21]. The control damswere treatedwith 0.1 ml/100 g sterile saline solution (0.9% NaCl) according to the same treatment schedule as the LPS animals. 2.4. Behavioral testing 2.4.1. Tail suspension test The TST was performed as described previously [6,53]. Briefly, the mice were suspended by the tail using tape that was attached to a hook that was connected to a strain gauge. Immobility time, defined as the mouse not struggling, was recorded in a single 6 min trial using a video camera that was positioned in front of the apparatus. 2.4.2. General activity in the open field test The OFT was performed as described previously [63]. The open field device consisted of a round wooden arena (40 cm diameter with 25.5 cm high walls) that was painted black with an acrylic washable covering. For the observations, each mouse was individually placed in the center of the apparatus, and total locomotor activity (i.e., distance travelled, in centimeters) and mean velocity were automatically mea- sured over a 5 min period. A video camera that was mounted 100 cm above the arena was used to collect the data, which were analyzed using Ethovision 2.3 software (Noldus Information Technology, Lees- burg, VA, USA) that was installed on an IBM-compatible computer in an adjacent room. The time spent grooming, rearing frequency, freezing time, and freezing frequencyweremanually scored by an experimenter who was unaware of the pharmacological treatments. The device was washedwith a 5% alcohol/water solution before each animalwas placed in it to obviate possible bias that may be caused by odor cues that were left by previous animals. The control and experimental mice were intermixed for observations that were performed from 8:00 AM to 12:00 PM. 2.5. Experimental design 2.5.1. F0 generation groups: treatment and mating Male and femalemicewere tested in the TST at 7weeks age to assign them to different phenotypes. Higher immobility time (≥60 s) resulted in the selection of depressive-like behavior. Lower immobility time (≤30 s) resulted in the selection of non-depressive-like behavior. These cut-off points were chosen because they highlighted differences between both behaviors in the present study. After evaluation, the male and female mice were separated into a depressive-like group (DG; n = 12 males, n = 18 females) and non-depressive-like group (NDG; n=12males, n=18 females). During evaluation, an intermedi- ate group (IG) was also identified (immobility time N 30 and b60) and excluded from the experiment (n = 16 males, n = 14 females). At 8weeks of age, female andmalemice that presented the same behavior were mated. In this first step in order to avoid litter effects the animals usedwere originally fromdifferentmothers. Pregnancywasdetected by the presence of semen in vaginal smears, thus defining GD0. The fe- males were then further divided into four groups: LPSD (depressive- like behavior + LPS), SALD (depressive-like behavior + saline), LPSND (non-depressive-like behavior + LPS), and SALND (non-depres- sive-like behavior + saline). On GD15, females in the experimental groups received 100 μg/kg LPS (i.p.), and females in the control groups received 0.1 ml/100 g sterile saline solution. The GD15 time point was chosen because it is a critical period for central nervous system development. 2.5.2. F1 generation groups Female mice of the F0 generation were allowed to give birth and nurture their offspring (F1 generation) normally. At birth, the F1 gener- ation was culled to eight pups per litter when possible (four males and four females), yielding a total of 97male and 87 femalemice. At 3weeks of age, all of the pupswereweaned and allocated to the same conditions as their parents and kept undisturbed until 7 weeks age for the behav- ioral analyses in the TST and OFT. At 7 weeks of age, immobility time was tested in 97 males (LPSD n = 31, SALD n = 19, LPSND n = 23, SALND n = 24) and 87 females (LPSD n = 27, SALD n = 24, LPSND n = 18, and SALND n = 18) of the F1 generation divided into the four experimental groups according to the behavior in the previous generation. Similar to observations in the F0 generation, each group presented percentages of mice that ex- hibited different behavior. After the TST, animals that presented the highest immobility time in the LPSD group (n = 10 males, n = 10 fe- males) and SALD group (n = 10 males, n = 10 females) and animals that presented the lowest immobility time in the LPSND group (n = 9 131T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 males, n=10 females) and SALND group (n=8males, n=10 females) were selected for the evaluation of general activity in the OFT. 2.5.3. F2 generation groups To obtain the F2 generation, males and females of the F1 generation that exhibited the same behavioral profile were mated. The animals se- lected were the same ones that were previously evaluated in the OFT. No LPS treatmentwas given during the gestational period in the F2 gen- eration. The same procedure as in the F1 generation was employed in the F2 generation, yielding a total of 78 males and 110 females. Special carewas taken to preventmating between siblings in the F1 generation. However, although siblings were not crossed to obtain the F2 genera- tion, siblings were used as part of the n to achieve the appropriate num- ber of animals in each group. Even though, no more than one male and one female offspring from any given litter were added in our analyses to minimize the contribution of litter effects to the data [64]. Specifically, each group had two siblings to complete the n. At 7 weeks of age, immobility time was tested in 78 males (LPSD n = 20, SALD n = 22, LPSND n = 17, SALND n = 19) and 110 females (LPSD n = 28, SALD n = 34, LPSND n = 23, SALND n = 25) of the F2 generation divided into the four experimental groups, according to the behavior in the previous generation. Animals that presented the highest immobility time in the LPSD group (n = 10 males, n = 10 females), SALD group (n = 10 males, n = 10 females), LPSND group (n = 10 males, n = 10 females), and SALND group (n = 7 males, n = 10 fe- males) were selected for the evaluation of general activity in the OFT. The overall experimental design is shown in Fig. 1. Fig. 1. Experimental design for the different germ lines. LPSD, LPS-treatedmicewith depressive- mice with non-depressive-like behavior; SALND, saline-treated mice with non-depressive-like 2.6. Statistical analysis The results are expressed as mean ± SEM. Homoscedasticity was verified using Bartlett's test. Immobility time in the TST and behavior in the OFT were analyzed using three-way analysis of variance (ANOVA), with generation, behavior, and treatment as factors. When no interaction was found between factors, two-way ANOVA followed by the Holm-Sidak post hoc test was applied for data from experiments that consisted of four groups. Data from experiments that consisted of three groups were analyzed with two-way ANOVA followed by Tukey's post hoc test. The proportion of phenotypes was analyzed using Fisher's exact test. The minimum probability level of 5% was considered able to show significant differences between groups. Prism 6 software (GraphPad, La Jolla, CA, USA) was used to analyze the data. 3. Results 3.1. F0 generation 3.1.1. Immobility time in males and females Three different behaviors were identified among male and female mice. To determine differences in behavioral parameters and differ- ences between sexes, we performed two-wayANOVAs. Fig. 2 shows sig- nificant differences between behaviors (F2,62 = 114.7, p b 0.0001) but no differences between males and females (F2,62 = 0.2429, p = 0.6238) and no interaction between factors (F2,62 = 0.2551, p = 0.7756). We then performed Tukey's multiple-comparison test and de- tected a higher immobility time between the DG and NDG (p b 0.0001), like behavior; SALD, saline-treatedmicewith depressive-like behavior; LPSND, LPS-treated behavior. Image of Fig. 1 Fig. 2. Immobility time (in seconds) inmale F0mice (DG, n=12; IG, n=16, ND, n=12) and female F0mice (DG, n=18; IG, n=14; NDG, n=18) in the tail suspension test. The data are expressed as mean ± SEM. ****p b 0.001 (Kruskal-Wallis one-way analysis of variance on ranks with Dunn's all pairwise multiple-comparison test). Fig. 3. Immobility time in the TST (seconds) in F1 males (A) and F2 males (B) whose F0 dams received 100 μg/kg LPS on GD15. The data are expressed as mean ± SEM. ****p b 0.0001, ***p b 0.001, **p b 0.01, *p b 0.5 (two-way ANOVA followed by Holm- Sidak post hoc test). 132 T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 between the DG and IG (p b 0.0001), and between the IG and NDG (p b 0.05) for both genders. 3.2. Effects of lipopolysaccharide treatment in different germ lines 3.2.1. Immobility time in F1 and F2 males and females in the TST Our initial hypothesis was that prenatal LPS exposure would pro- mote behavioral differences that would be detectable through subse- quent generations. This effect was also hypothesized to differentially affectmales and females. To test this hypothesis,we conducted separate three-way ANOVAs in males and females, with generation, behavior, and treatment as factors. Considering the results of the TST, the three- way ANOVA did not reveal interactions between generation, the depressive-like behavior and the LPS treatment (F1,167 = 0.006, p = 0.939) or influence of the different generations over the LPS treatment (F1,167 = 1.124, p = 0.291) and the depressive-like behavior (F1,167 = 1.160, p = 0.689). Treatment did not affect the occurrence of depressive- and non-depressive-like behavior (F1,167 = 1.996, p = 0.160), but significant differences were found between depressive- and non-depressive-like animals (F1,167 = 57.50, p b 0.001). Once the generation factor was not significant in this test, we ana- lyzed males of the F1 and F2 generations separately. With regard to be- havior and treatment in F1 males, Fig. 3A shows significant differences in the immobility time between depressive- and non-depressive-like animals (F1,93 = 26.92, p b 0.0001) but no effect of the LPS treatment (F1,93 = 0.008, p = 0.927) and no interaction between factors (F1,93= 1.164, p=0.283). The Holm-Sidak post hoc test revealed signif- icant differences in the immobility time between the LPSD and LPSND groups (p b 0.0001) and between the SALD and SALND groups (p b 0.05). Fig. 3B shows the results of the two-way ANOVA for F2 males. The analysis revealed significant differences in the immobility time between depressive- and non-depressive-like animals (F1,74 = 31.50, p b 0.001) but no effect of the LPS treatment (F1,74 = 1.965, p=0.165) and no in- teraction between factors (F1,74 = 0.876, p = 0.352). The Holm-Sidak post hoc test revealed differences in the immobility time between the LPSD and LPSND groups (p b 0.001) and between the SALD and SALND groups (p b 0.01). With regard to behavior and treatment in females of the F1 genera- tion, Fig. 4A shows significant differences in the immobility time be- tween depressive- and non-depressive-like animals (F1,83 = 10.67, p = 0.001) but no effect of the LPS treatment (F1,83 = 0.141, p = 0.707) and no interaction between factors (F1,83 = 0.348, p = 0.556). However, the Holm-Sidak post hoc test did not reveal immobility time differences between groups. In F2 females, as shown in Fig. 4B, a signif- icant interaction between depressive-like behavior and the LPS treat- ment was observed (F1,106 = 4.934, p = 0.028). A significant difference in the immobility time was found between depressive- and non-depressive-like females (F1,106 = 22.98, p b 0.0001), with no effect of the LPS treatment (F1,106 = 1.428, p = 0.234). The Holm-Sidak post hoc test revealed significant differences in the immobility time between the LPSD and LPSND groups (p b 0.0001) and between the LPSD and SALD groups (p b 0.5). 3.2.2. Behavioral trend in male and female offspring We performed Fisher's exact test in F0 and F1 offspring to identify trends in the transmission of behavioral phenotypes. Fig. 5A shows a positive correlation in LPS-treated male offspring of the F0 generation (***p b 0.0001) and positive correlations both saline- and LPS-treated male offspring of the F1 generation (saline: **p = 0.0014; LPS: ***p = 0.0002). These data indicate that depressive- and non-depressive-like behavior was transmitted to the male filial germ lines. Fisher's exact test thatwas applied for F0 and F1 female offspring re- vealed a positive correlation in the saline-treated offspring of the F0 generation (*p = 0.0383) and positive correlations in both saline- and LPS-treated offspring of the F1 generation (saline: *p = 0.0394; LPS: Image of &INS id= Image of Fig. 3 Fig. 4. Immobility time (in seconds) in the TST in F1 females (A) and F2 females (B)whose F0 dams received 100 μg/kg LPS on GD15. The data are expressed as mean ± SEM. ****p b 0.0001, *p b 0.5 (two-way ANOVA followed by Tukey post hoc test). 133T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 **p = 0.0018; Fig. 5B). These data indicate that depressive- and non- depressive like behavior was transmitted to the female filial germ lines, similar to males. 3.2.3. General activity in males and females of the F1 and F2 generations We performed a three-way ANOVA of different parameters in the OFT for the selected depressive- and non-depressive-like male mice. The results indicated a significant interaction between generation, be- havior, and treatment for grooming frequency (F1,66 = 4.033, p = 0.049). Significant differences were found in mean velocity (F1,66 = 43.26, p b 0.001), rearing frequency (F1,66 = 20.67, p b 0.001), and dis- tance travelled (F1,66 = 40.25, p b 0.001) between F1 and F2 males. Dif- ferences were also found in immobility frequency (F1,66 = 12.81, p b 0.001) and immobility time (F1,66 = 16.91, p b 0.001) between gen- erations, with significant differences in the effects of LPS treatment on both parameters (F1,66 = 19.84, p b 0.001, and F1,66 = 12.22, p b 0.001, respectively; Table 1). In F1 females, the three-way ANOVA of the different parameters in the OFT in depressive- and non-depressive-like females revealed differ- ences in mean velocity (F1,70 = 37.12, p b 0.001) and distance travelled (F1,70= 34.87, p b 0.001) between the F1 and F2 generations and signif- icant interactions between generation and treatment for grooming fre- quency (F1,70 = 7.576, p = 0.008), rearing frequency (F1,70 = 17.59, p b 0.001), and immobility frequency (F1,70 = 15.48, p b 0.001). Signif- icant interactions between generation and behavior were also found for immobility frequency and immobility time in the OFT (F1,70 = 7.956, p = 0.006, and F1,70 = 6.991, p = 0.010, respectively; Table 2). 3.3. Sexual dimorphism in the TST and OF among males and females In the F1 generation, the three-way ANOVA revealed no differences in immobility time between males and females in the TST (F1,177 = 0.784, p=0.377; Fig. 4A). However, different parameters in the OFT in- dicated differences between males and females (Table 3). Significant differences in immobility time were found between depressive- and non-depressive-likemales and females (F1,67=7.487, p=0.008). A sig- nificant interactionwas found between gender, behavior, and treatment for immobility frequency (F1,67 = 50.69, p b 0.001). Similar to the F1 generation, the three-way ANOVA in the F2 gener- ation did not reveal significant differences in immobility time between males and females in the TST (F1,180 = 0.370, p = 0.544; Fig. 4B). In the OFT, significant differences were observed between males and fe- males (Table 4). Significant differences in rearing frequency (F1,69 = 8.984, p = 0.004), immobility time (F1,69 = 5.503, p = 0.022), and im- mobility frequency (F1,69 = 6.198, p = 0.015) were observed between males and females mice among the treatment (Table 4). 4. Discussion The TST is commonly used to screen the antidepressant effects of drugs or assess depressive-like behavior. It is also known as a test of be- havioral despair that consists of inescapable stress. We employed the TST because of its relatively easy applicability and advantages with re- gard to avoiding stress that is induced bywater exposure and hypother- mia, which differentiates it from the FST (i.e., another commonly used test that evaluates behavioral despair) [9]. In the present study, we found that males and females could be se- lected in different generations based on different immobility responses in the TST. High immobility times reflects depressive-like behavior. Low immobility times reflect non-depressive-like behavior. In addition to these two main phenotypes, we also identified an intermediate group, in which the mice did not present sufficient immobility time to be con- sidered depressed or non-depressed. Depressive-like females from the F0 generation, when mated with males of the same generation that exhibited the same behavior, had a higher percentage of offspring (i.e., the F1 generation) that also exhibit- ed this characteristic. This phenotype transmission was also evident in the F2 generation when the same mating procedure was applied to F1 mice. Similar phenotype transmission was also evident among non- depressive-like mice that were subjected to the same mating proce- dures. Male mice maintained the initial proportion of these behaviors through two generations (74% for non-depressive-like behavior and 82% for depressive-like behavior). Female mice also maintained a simi- lar proportion through two generations, despite being slightly less than males (71% for non-depressive-like behavior and 72% for depressive- like behavior), suggesting that genetic factors might be involved in this selection and its manifestation. Although these phenotype results were significant, the present study has limitations that need to be con- sidered. For example, although siblings were not crossed with each other to create the F2 generation, they were used as part of the n of the F1 generation groups. In a 20-year-long longitudinal study, Weissman et al. [60] followed three generations of patients with either a history or no history of MDD. These authors reported that the rates of psychopathology were the highest among grandchildren whose relatives had MDD. Gene- environment interactions are fundamental to this issue because rela- tives can provide information about both genetics and environmental risk. However, establishing the specific influences of genetics and the environment on the development of disease is difficult because both might modulate the response to the other. According to Caspi et al. [5], a functional polymorphism that is located in the promoter region of the serotonin transporter gene can act as a modulator of stressful events over an individual's lifespan, which can lead to high, low, or no expression of depressive symptoms. This may help explain why people Image of Fig. 4 Fig. 5. Behavioral phenotype transmission through depressive- and non-depressive germlines. (A)Male germline. The results show a Fisher's exact test of F0 offspring (Saline: F0 D:D n= 10, ND n=3and F0ND:D n=7,ND n=6; LPS: F0 D:D n=27, ND n=2and F0ND: D n=5, ND n=9) and F1males offspring (Saline: F1 D:D n=17, ND n=1and F1ND:D n=4,ND n=7; LPS: F1D:D n=12, ND n=3and F1ND: D n=1,ND n=12), F0 LPS ***p b 0.0001, F1 Saline: **p=0.0014; F1 LPS: ***p=0.0002. (B) Female germline. The results show a Fisher's exact test of F0 offspring (Saline: F0 D:D n=13, ND n=5and F0ND:D n=5, ND n=10; LPS: F0D:D n=14,ND n=7and F0ND:D n=4,ND n=8) and F1 female offspring (Saline: F1 D:D n=15, ND n=10 and F1ND:D n=6, ND n=16; LPS: F1D:D n=19, ND n=7and F1ND:D n=4,ND n=14), F0 Saline *p=0.0383, F1 Saline: *p=0.0394; F1 LPS: **p=0.0018. 134 T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 responddifferently to the same stressors but not how the stressormight modulate gene expression through generations. Stressful environmen- tal events might also produce epigenetic modifications, such as DNA methylation or histone alterations. These epigenetic alterations may af- fect the way in which the organism responds to MDD or may even alter a predetermined behavior-environment response [14]. To understand the biological mechanisms that underlie various dis- orders, valid animal models are indispensable, particularly in neuropsy- chiatric research. Mice that are genetically or selectively bred to model specific key symptoms of human depression can be successfully used to discover neurobiological endophenotypes that may bridge the gap between behavioral phenotypes and genotypes [10,19,20,55]. As an al- ternative to transgenicmice that carry specific genetic mutations, selec- tive breeding has been proven to be a powerful strategy for unraveling Table 1 Analysis of parameters in theopenfield test inmale F1 vs. F2mice. The F1 generationwasprenat on GD15. The F2 generation was not directly exposed to LPS or saline. The data were analyzed Source of variation of male F1 vs. F2 mice Velocity Rearing Grooming df F p df F p df F Generation 1 43.265 b0.001*** 1 20.675 b0.001*** 1 0.239 Behavior 1 0.859 0.357 1 3.199 0.078 1 4.271 Treatment 1 1.301 0.258 1 0.016 0.9 1 0.385 Generation × behavior 1 0.585 0.447 1 0.945 0.335 1 1.03 Generation × treatment 1 0.409 0.525 1 2.507 0.118 1 0.186 Behavior × treatment 1 1.09 0.3 1 1.105 0.297 1 0.915 Generation × behavior × treatment 1 0.0169 0.897 1 1.098 0.298 1 4.033 Residual 66 66 66 Total 73 73 73 df, degrees of freedom; F, F ratio; p, probability. the genetic basis of disorders, providing unique information about plei- otropy, epistasis, and gene-environment interactions [55]. With regard to gender differences, we found that the incidence of a depressive-like phenotypewasmore robust inmalemice than in female mice of the F1 generation that were born from dams that were treated with saline solution or LPS (i.e., there were more depressive-like males than females of the F1 generation, independent of treatment). Depression is a mental disease that affects complex cognitive and emotional functions. Depression has been reported to be twice more prevalent in women than in men [41,42,48]. The incidence of depres- sion in women also varies by life stage, with many interacting factors (e.g., childbearing and cyclic hormonal changes). This phenomenon may influence the response to various antidepressant therapies, and these differences are still underestimated in clinical treatment. ally exposed to LPS or saline, inwhichF0dams received100 μg/kg LPSor 0.9%NaCl solution using three-way ANOVA followed by the Holm-Sidak post hoc test. Freezing time Freezing frequency Distance p df F p df F p df F p 0.626 1 16.913 b0.001*** 1 12.811 b0.001*** 1 40.255 b0.001*** 0.043* 1 1.174 0.282 1 7.068 0.01 1 0.845 0.361 0.537 1 12.229 b0.001*** 1 19.847 b0.001*** 1 1.156 0.286 0.314 1 0.18 0.673 1 0.62 0.434 1 0.431 0.514 0.668 1 1.981 0.164 1 0.351 0.555 1 0.348 0.557 0.342 1 1.691 0.198 1 8.689 0.004*** 1 1.271 0.264 0.049* 1 1.181 0.281 1 0.0002 0.989 1 0.003 0.957 66 66 66 73 73 73 Image of Fig. 5 Table 2 Analyses of parameters observed in the open field test in female F1 vs. F2mice. The F1 generationwas prenatally exposed to LPS or saline, inwhich F0 dams received 100 μg/kg LPS or 0.9% NaCl solution on GD15. The F2 generation was not directly exposed to LPS or saline. The data were analyzed using three-way ANOVA followed by the Holm-Sidak post hoc test. Source of variation of female F1 vs. F2 mice Velocity Rearing Grooming Freezing time Freezing frequency Distance df F p df F p df F p df F p df F p df F p Generation 1 37.126 b0.001*** 1 1.54 0.219 1 5.345 0.024* 1 2.371 0.128 1 0.447 0.506 1 34.872 b0.001*** Behavior 1 1.693 0.197 1 1.224 0.272 1 0.594 0.443 1 0.533 0.468 1 0.269 0.606 1 1.949 0.167 Treatment 1 0.00527 0.942 1 2.214 0.141 1 0.594 0.443 1 0.134 0.716 1 1.384 0.243 1 0.0002 0.988 Generation × behavior 1 0.767 0.384 1 0.508 0.478 1 0.0121 0.913 1 6.991 0.01** 1 7.956 0.006*** 1 0.579 0.449 Generation × treatment 1 3.156 0.08 1 17.592 b0.001*** 1 7.576 0.008*** 1 2.389 0.127 1 15.488 b0.001*** 1 3.314 0.073 Behavior × treatment 1 0.451 0.504 1 2.248 0.138 1 0.303 0.584 1 0.214 0.645 1 3.426 0.068 1 0.556 0.459 Generation × behavior × treatment 1 0.243 0.624 1 0.0026 0.96 1 2.048 0.157 1 0.122 0.728 1 2.073 0.154 1 0.237 0.628 Residual 70 70 70 70 70 70 Total 77 77 77 77 77 77 df, degrees of freedom; F, F ratio; p, probability. 135T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 Data from animal models of depression have been contradictory. Previous studies that utilized the TST or FST found higher depressive- like behavior in females than inmales [11,33], lower depressive-like be- havior in females than inmales [44], or no differences between genders [4,46]. Both the TST and FST are based on rodents' innate behavior to devel- op an immobile posture after initial escape-oriented movements when placed in an inescapable stressful environment. In the TST, the stressful situation involves hemodynamic stress, in which the mouse is suspended by its tail in an uncontrollable fashion. In the FST, the mouse is placed in a cylinder that is filledwith water [54]. The presence of corticosterone has been suggested to be crucial for the incorporation of the learned response after swim stress. Therefore, corticosterone is implicated in helpless behavior in the FST and likely in the TST. Kokras et al. [26] found that manipulating corticosterone levels effectively re- versed the effects of adrenalectomy in male rats but not in female rats. These authors proposed the existence of two sex-oriented mechanisms of the stress response. The predominantlymale-type response is associ- ated with peripheral endogenous corticosterone. The female-type re- sponse is less dependent on this influence of corticosterone. Importantly, both depressive- and non-depressive mice were obtained by mating female and male mice that presented the same phenotype, suggesting a genetic influence on our results. Thus, differences between male and female mice that were caused by the genetic selection of depressive-like behavior, with consequently high corticosterone levels and responsiveness to stress, resulted in more vulnerability to depres- sion in male mice in the D group compared with female mice in the same group. Previous studies reported that the estrous cycle does not significant- ly modulate depressive- or anxiety-like behavior [1,49]. However, other studies reported a role for female gonadal hormones in the expression of both anxiety- and depressive-like behavior [28,32]. In the present study, we did not determine the estrous phase in female mice but Table 3 Analyses of parameters in the open field test inmale vs. female F1 mice. The F1 generation was solution on GD15. The F2 generation was not directly exposed to LPS or saline. The data were Source of variation of male vs. female F1 mice Velocity Rearing Grooming df F p df F p df F Generation 1 0.0101 0.92 1 0.173 0.679 1 1.948 Behavior 1 1.437 0.235 1 2.65 0.108 1 2.608 Treatment 1 4.493 0.038* 1 4.392 0.04* 1 0.825 Generation × behavior 1 0.381 0.539 1 1.534 0.22 1 0.916 Generation × treatment 1 0.0437 0.835 1 0.34 0.562 1 1.137 Behavior × treatment 1 0.136 0.714 1 3.265 0.075 1 4.948 Generation × behavior × treatment 1 0.376 0.542 1 0.139 0.71 1 0.062 Residual 67 67 67 Total 74 74 74 df, degrees of freedom; F, F ratio; p, probability. attempted to reduce estrous phase variations by housing female mice and their offspring in neighboring cages after weaning [37]. Prenatal exposure to LPS promoted significant differences between both phenotypes in male mice compared with saline-treated mice of the F1 and F2 generations. Prenatal LPS exposure increased the outcome of depressive-like behavior in male mice. The significant differences were greater in LPS-treated male mice compared with saline-treated male mice. Similar results were not observed among male F1 mice with non-depressive-like behavior. These resultsmight indicate sensiti- zation to LPS that is caused by parental exposure inmalemice. Such dif- ferences may be associated with the actions of various stress-related hormones, such as corticosterone, and epigenetic modifications that can contribute to sex-specific brain development and the susceptibility to environmental stress [17]. Stressful environmental situations, including psychological trauma during the early stages of neonatal development and prenatal exposure to various viral and bacterial infections, can cause short- and long-term changes in behavior and central nervous system activity [39,52]. Prena- tal infection that is mimicked by LPS exposure causes profound changes in fetal brain development, affecting such areas as the HPA axis and in- creasing the levels of glucocorticoids and proinflammatory cytokines within the maternal circulation and fetal brain [2,39,56]. Walker et al. [58] reported that neonatal LPS exposure increased anxiety-like behavior in mice in adulthood, and the anxiety-like pheno- type was transmitted to filial embryo lines when stressed females were bred with non-manipulated males. Neonatal stress can induce alter- ations that vary from impairments inmaternal care to phenotype trans- mission across filial lines. Neonatal and prenatal stress also affects behavior. Penteado et al. [45] reported that prenatal LPS exposure in- creased maternal care in the parental generation and promoted toler- ance to an LPS challenge during infancy in the F1 generation, leading to the hypothesis that maternal inflammation disrupts behavioral and immune responses. prenatally exposed to LPS or saline, in which F0 dams received 100 μg/kg LPS or 0.9% NaCl analyzed using three-way ANOVA followed by the Holm-Sidak post hoc test. Freezing time Freezing frequency Distance p df F p df F p df F p 0.167 1 3.693 0.059 1 49.718 b0.001*** 1 0.000618 0.98 0.111 1 3.636 0.061 1 5.074 0.028* 1 1.572 0.214 0.367 1 1.923 0.17 1 110.08 b0.001*** 1 4.448 0.039* 0.342 1 7.487 0.008*** 1 57.221 b0.001*** 1 0.159 0.692 0.29 1 0.0561 0.813 1 20.59 b0.001*** 1 0.132 0.717 0.029* 1 0.25 0.619 1 6.526 0.013 1 0.199 0.657 4 0.803 1 0.381 0.539 1 50.697 b0.001*** 1 0.63 0.43 67 67 67 74 74 74 Table 4 Analyses of parameters in the open field test in male vs. female F2 mice. The F2 generation was not directly exposed to LPS or saline treatment. The data were analyzed using three-way ANOVA followed by the Holm-Sidak post hoc test. Source of variation of male vs. female F2 mice Velocity Rearing Grooming Freezing time Freezing frequency Distance df F p df F p df F p df F p df F p df F p Generation 1 0.917 0.342 1 8.984 0.004*** 1 2.416 0.125 1 2.451 0.122 1 1.202 0.277 1 0.84 0.363 Behavior 1 1.214 0.274 1 1.664 0.201 1 1.074 0.304 1 1.741 0.191 1 2.287 0.135 1 1.305 0.257 Treatment 1 0.484 0.489 1 12.987 b0.001*** 1 6.71 0.012* 1 0.000893 0.976 1 0.00266 0.959 1 0.477 0.492 Generation × behavior 1 0.961 0.33 1 0.215 0.644 1 1.82E−32 1 1 1 0.321 1 0.37 0.545 1 0.892 0.348 Generation × treatment 1 1.15 0.287 1 3.514 0.065 1 2.416 0.125 1 5.503 0.022* 1 6.198 0.015* 1 1.115 0.295 Behavior × treatment 1 0.0135 0.908 1 0.56 0.457 1 1.074 0.304 1 0.26 0.612 1 0.27 0.605 1 0.0369 0.848 Generation × behavior × treatment 1 0.995 0.322 1 0.556 0.459 1 0 1 1 0.418 0.52 1 0.806 0.372 1 1.001 0.321 Residual 69 69 69 69 69 69 Total 76 76 76 76 76 76 df, degrees of freedom; F, F ratio; p, probability. 136 T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 Elevated levels of proinflammatory cytokines, such as interleukin-1, interleukin-6, and tumor necrosis factor-α, have been reported in pa- tients with MDD [35,36,50]. In fact, many of the symptoms of MDD can be understood as sickness behavior [30,50,62]. However, cytokines likely act in concert with other systems, such as monoamines (particu- larly norepinephrine and serotonin), to produce the characteristic symptoms of depression [57]. Lipopolysaccharide exposure affects HPA axis activation and in- creases cytokine and brain norepinephrine and serotonin levels [15]. Specific phenotypes that are initially caused by LPS-induced inflamma- tion can also lead to transmission along generations [58]. Interestingly, no differences in depressive- and non-depressive-like behavior, regardless of LPS treatment, were found in female mice of the F1 generation, indicating possible sexual dimorphism. In the F2 gen- eration, female mice in the LPS group exhibited an increase in depressive-like behavior not only compared with non-depressive fe- male mice in the LPS group but also compared with depressive-like fe- male mice in the saline group. Prenatal LPS exposure in the parental generation can readily repro- gram the HPA and immune axes in F2 females in adulthood. LPS expo- sure in the parental generation did not modify the proportion of depressive- and non-depressive-like phenotypes in the F1 generation, reinforcing the genetic nature of this phenomenon. However, the pro- portion of mice with the depressive-like phenotype was affected in the F2 generation by parental exposure to LPS. The reasons why these effects were not observed in the F1 generation remain to be explored. 5. Conclusion Male mice that were selected for depressive- and non-depressive- like behavior exhibited the same behavioral profile in the F1 and F2 gen- erations. Lipopolysaccharide exposure augmented these differences, suggesting that LPS reprograms the HPA and immune axes in both gen- erations. Female mice of the F1 generation did not present significant differences in the incidence of either the depressive- or non- depressive-like phenotype, suggesting sexual dimorphism. However, female F2 mice from the parental (F0) generation that was treated with LPS exhibited an increase in the incidence of depressive-like be- havior compared with non-depressive-like behavior. These results sug- gest that maternal exposure to LPS affects subsequent generations in a sexually dimorphic manner. We cannot state that prenatal inflammation did not promote other aspects of behavior in the filial lines. Lipopolysaccharide exposure could have promoted tolerance to its effects, thus masking possible changes. More studies will be necessary to elucidate other possible al- terations, including neurochemical and genetic effects that are caused by LPS exposure. Acknowledgements This research was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível superior (CAPES-DS), Programa Individual de pesquisa para docentes da Universidade Paulista (UNIP-7-02915- 2014) and the São Paulo Research Foundation (FAPESP grant nos. 2013/03912-0 and 09/51886-3) and is part of the Neuroimmunomodulation Research Group of the Veterinary Medical School of São Paulo University. References [1] S. Andrade, S. Silveira, B. Arbo, B. Batista, R. Gomez, H. Barros, et al., Sex-dependent antidepressant effects of lower doses of progesterone in rats, Physiol. Behav. 99 (5) (2010) 687–690. [2] H. Ashdown, Y. Dumont, M. Ng, S. Poole, P. Boksa, G.N. Luheshi, The role of cytokines in mediating effects of prenatal infection on the fetus: implications for schizophre- nia, Mol. Psychiatry 11 (1) (2006) 47–55. [3] M.M. Bernardi, L.P. Teixeira, A.P. Ligeiro-de-Oliveira, W. Tavares-de-Lima, J. Paler- mo-Neto, T.B. Kirsten, Neonatal lipopolysaccharide exposure induces sexually di- morphic sickness behavior in adult rats, Psychol. Neurosci. 7 (2) (2014) 113. [4] L.A. Brotto, A.M. Barr, B.B. Gorzalka, Sex differences in forced-swim and open-field test behaviours after chronic administration of melatonin [research support, non- U.S. Gov't], Eur. J. Pharmacol. 402 (1–2) (2000) 87–93. [5] A. Caspi, K. Sugden, T.E. Moffitt, A. Taylor, I.W. Craig, H. Harrington, et al., Influence of life stress on depression: moderation by a polymorphism in the 5-HTT gene, Sci- ence 301 (5631) (2003) 386–389. [6] Castagne, V., Moser, P., Roux, S., & Porsolt, R. D. (2011). Rodent models of depres- sion: forced swim and tail suspension behavioral despair tests in rats and mice. Curr Protoc Neurosci, Chapter 8, Unit 8 10A. [7] A.L. Chirita, V. Gheorman, D. Bondari, I. Rogoveanu, Current understanding of the neurobiology of major depressive disorder [research support, non-U.S. Gov't], Romanian journal of morphology and embryology = Revue roumaine de morphologie et embryologie 56 (2 Suppl) (2015) 651–658. [8] D.W. Cohn, D. Kinoshita, J. Palermo-Neto, Antidepressants prevent hierarchy desta- bilization induced by lipopolysaccharide administration in mice: a neurobiological approach to depression, Ann. N. Y. Acad. Sci. 1262 (2012) 67–73. [9] J.F. Cryan, C. Mombereau, A. Vassout, The tail suspension test as a model for assessing antidepressant activity: review of pharmacological and genetic studies in mice, Neurosci. Biobehav. Rev. 29 (4–5) (2005) 571–625. [10] J.F. Cryan, C. Mombereau, A. Vassout, The tail suspension test as a model for assessing antidepressant activity: review of pharmacological and genetic studies in mice, Neurosci. Biobehav. Rev. 29 (4–5) (2005) 571–625. [11] C. Dalla, K. Antoniou, N. Kokras, G. Drossopoulou, G. Papathanasiou, S. Bekris, et al., Sex differences in the effects of two stress paradigms on dopaminergic neurotrans- mission [research support, non-U.S. Gov't], Physiol. Behav. 93 (3) (2008) 595–605. [12] R. Dantzer, Cytokine, sickness behavior, and depression [research support, N.I.H., ex- tramural research support, non-U.S. Gov't review], Neurol. Clin. 24 (3) (2006) 441–460. [13] R. Dantzer, R.M. Bluthe, S. Laye, J.L. Bret-Dibat, P. Parnet, K.W. Kelley, Cytokines and sickness behavior, Ann. N. Y. Acad. Sci. 840 (1998) 586–590. [14] D.M. Dietz, Q. Laplant, E.L. Watts, G.E. Hodes, S.J. Russo, J. Feng, et al., Paternal trans- mission of stress-induced pathologies, Biol. Psychiatry 70 (5) (2011) 408–414. [15] A.J. Dunn, Effects of cytokines and infections on brain neurochemistry, Clin. Neurosci. Res. 6 (1–2) (2006) 52–68. [16] G. Hava, L. Vered, M. Yael, H. Mordechai, H. Mahoud, Alterations in behavior in adult offspring mice followingmaternal inflammation during pregnancy, Dev. Psychobiol. 48 (2) (2006) 162–168. http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0005 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0005 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0005 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0010 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0010 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0010 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0015 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0015 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0015 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0020 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0020 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0020 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0025 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0025 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0025 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0030 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0030 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0030 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0030 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0035 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0035 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0035 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0040 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0040 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0040 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0045 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0045 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0045 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0050 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0050 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0050 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0055 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0055 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0055 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0060 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0060 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0065 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0065 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0070 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0070 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0075 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0075 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0075 137T.M. Reis-Silva et al. / Life Sciences 149 (2016) 129–137 [17] G.E. Hodes, Sex, stress, and epigenetics: regulation of behavior in animal models of mood disorders, Biol. Sex Differ. 4 (1) (2013) 1. [18] G.E. Hodes, V. Kana, C. Menard, M. Merad, S.J. Russo, Neuroimmune mechanisms of depression [research support, N.I.H., extramural research support, non-U.S. Gov't re- view], Nat. Neurosci. 18 (10) (2015) 1386–1393. [19] L.H. Jacobson, J.F. Cryan, Feeling strained? Influence of genetic background on depression-related behavior in mice: a review, Behav. Genet. 37 (1) (2007) 171–213. [20] M.J. Kas, C. Fernandes, L.C. Schalkwyk, D.A. Collier, Genetics of behavioural domains across the neuropsychiatric spectrum; of mice and men, Mol. Psychiatry 12 (4) (2007) 324–330. [21] T.B. Kirsten, M. Taricano, P.C. Maiorka, J. Palermo-Neto, M.M. Bernardi, Prenatal lipo- polysaccharide reduces social behavior inmale offspring, Neuroimmunomodulation 17 (4) (2010) 240–251. [22] T.B. Kirsten, B.P. de Oliveira, A.P. de Oliveira, K. Kieling, W.T. de Lima, J. Palermo- Neto, et al., Single early prenatal lipopolysaccharide exposure prevents subsequent airway inflammation response in an experimental model of asthma, Life Sci. 89 (1– 2) (2011) 15–19. [23] T.B. Kirsten, G.P. Chaves, M. Taricano, D.O. Martins, J.C. Florio, L.R. Britto, et al., Prena- tal LPS exposure reduces olfactory perception in neonatal and adult rats, Physiol. Behav. 104 (3) (2011) 417–422. [24] T.B. Kirsten, G.P. Chaves-Kirsten, L.M. Chaible, A.C. Silva, D.O. Martins, L.R. Britto, et al., Hypoactivity of the central dopaminergic system and autistic-like behavior in- duced by a single early prenatal exposure to lipopolysaccharide, J. Neurosci. Res. 90 (10) (2012) 1903–1912. [25] T.B. Kirsten, G.P. Chaves-Kirsten, S. Bernardes, C. Scavone, J.E. Sarkis, M.M. Bernardi, et al., Lipopolysaccharide exposure induces maternal hypozincemia, and prenatal zinc treatment prevents autistic-like behaviors and disturbances in the striatal do- paminergic and mTOR systems of offspring [research support, non-U.S. Gov't], PloS one 10 (7) (2015) e0134565. [26] N. Kokras, C. Dalla, A.C. Sideris, A. Dendi, H.G. Mikail, K. Antoniou, et al., Behavioral sexual dimorphism in models of anxiety and depression due to changes in HPA axis activity, Neuropharmacology 62 (1) (2012) 436–445. [27] V. Krishnan, E.J. Nestler, Animal models of depression: molecular perspectives, Curr. Top. Behav. Neurosci. 7 (2011) 121–147. [28] N. Lagunas, I. Calmarza-Font, Y. Diz-Chaves, L.M. Garcia-Segura, Long-term ovariec- tomy enhances anxiety and depressive-like behaviors in mice submitted to chronic unpredictable stress, Horm. Behav. 58 (5) (2010) 786–791. [29] B.E. Leonard, Stress, norepinephrine and depression, J Psychiatry Neurosci S11-16 (26 Suppl) (2001). [30] B.E. Leonard, Psychopathology of depression, Drugs Today (Barc) 43 (10) (2007) 705–716. [31] B.E. Leonard, A. Myint, The psychoneuroimmunology of depression, Hum Psychopharmacol 24 (3) (2009) 165–175. [32] L.-H. Li, Z.-C. Wang, J. Yu, Y.-Q. Zhang, Ovariectomy results in variable changes in nociception, mood and depression in adult female rats, PLoS One 9 (4) (2014), e94312. [33] Y.L. Lin, S.Wang, Prenatal lipopolysaccharide exposure increases depression-like be- haviors and reduces hippocampal neurogenesis in adult rats [research support, non- U.S. Gov't], Behav. Brain Res. 259 (2014) 24–34. [34] X. Liu, D. Stancliffe, S. Lee, S. Mathur, H.K. Gershenfeld, Genetic dissection of the tail suspension test: a mouse model of stress vulnerability and antidepressant response, Biol. Psychiatry 62 (1) (2007) 81–91. [35] M. Maes, Evidence for an immune response in major depression: a review and hy- pothesis, Prog. Neuro-Psychopharmacol. Biol. Psychiatry 19 (1) (1995) 11–38. [36] M. Maes, Major depression and activation of the inflammatory response system, Adv. Exp. Med. Biol. 461 (1999) 25–46. [37] M.K. McClintock, Estrous synchrony: modulation of ovarian cycle length by female pheromones, Physiol. Behav. 32 (5) (1984) 701–705. [38] W.T. McKinney, Animal models of depression: An overview, Psychiatr Dev 2 (2) (1984) 77–96. [39] U. Meyer, B.K. Yee, J. Feldon, The neurodevelopmental impact of prenatal infections at different times of pregnancy: the earlier the worse? Neuroscientist 13 (3) (2007) 241–256. [40] R.F. Munoz, P. Cuijpers, F. Smit, A.Z. Barrera, Y. Leykin, Prevention of major depres- sion [research support, N.I.H., extramural research support, non-U.S. Gov't review], Annu. Rev. Clin. Psychol. 6 (2010) 181–212. [41] S. Nolen-Hoeksema, J. Larson, C. Grayson, Explaining the gender difference in de- pressive symptoms [research support, U.S. Gov't, P.H.S.], J. Pers. Soc. Psychol. 77 (5) (1999) 1061–1072. [42] S.J. Oliver, B.B. Toner, The influence of gender role typing on the expression of de- pressive symptoms, Sex Roles 22 (11−12) (1990) 775–790. [43] Organization, W. H., The World Health Report 2001: Mental Health: New Under- standing, New Hope, World Health Organization, 2001. [44] P. Palanza, Animal models of anxiety and depression: How are females different? Neurosci. Biobehav. Rev. 25 (3) (2001) 219–233. [45] S.H.N.W. Penteado, C.O.S.G. Massoco, T.B. Kirsten, T.M. Reis-Silva, R.C. Melo, M.K. Acenjo, et al., Prenatal lipopolysaccharide increases maternal behavior,decreases maternal odor preference, and induces lipopolysaccharide hyporesponsiviness, Psy- chology & Neurosciences 6 (2013) 31–38. [46] P.M. Pitychoutis, K. Nakamura, P.A. Tsonis, Z. Papadopoulou-Daifoti, Neurochemical and behavioral alterations in an inflammatory model of depression: sex differences exposed [research support, non-U.S. Gov't], Neuroscience 159 (4) (2009) 1216–1232. [47] Porsolt, R. D., Brossard, G., Hautbois, C., & Roux, S. (2001). Rodent models of depres- sion: forced swimming and tail suspension behavioral despair tests in rats andmice. Curr Protoc Neurosci, Chapter 8, Unit 8 10A. [48] R.K. Salokangas, K. Vaahtera, S. Pacriev, B. Sohlman, V. Lehtinen, Gender differences in depressive symptoms: an artefact caused by measurement instruments? J. Affect. Disord. 68 (2) (2002) 215–220. [49] S. Salomon, C. Bejar, D. Schorer-Apelbaum, M. Weinstock, Corticosterone mediates some but not other behavioural changes induced by prenatal stress in rats, J. Neuroendocrinol. 23 (2) (2011) 118–128. [50] O.J. Schiepers, M.C.Wichers, M. Maes, Cytokines and major depression, Prog. Neuro- Psychopharmacol. Biol. Psychiatry 29 (2) (2005) 201–217. [51] A.M. Soto, T.B. Kirsten, T.M. Reis-Silva, M.F. Martins, E. Teodorov, J.C. Florio, et al., Single early prenatal lipopolysaccharide exposure impairs striatal monoamines and maternal care in female rats, Life Sci. 92 (14–16) (2013) 852–858. [52] A.M. Soto, T.B. Kirsten, T.M. Reis-Silva, M.F. Martins, E. Teodorov, J.C. Florio, et al., Single early prenatal lipopolysaccharide exposure impairs striatal monoamines and maternal care in female rats, Life Sci. 92 (14–16) (2013) 852–858. [53] L. Steru, R. Chermat, B. Thierry, P. Simon, The tail suspension test: a newmethod for screening antidepressants in mice, Psychopharmacology 85 (3) (1985) 367–370. [54] B. Thierry, L. Steru, P. Simon, R.D. Porsolt, The tail suspension test: ethical consider- ations, Psychopharmacology 90 (2) (1986) 284–285. [55] C. Touma, M. Bunck, L. Glasl, M. Nussbaumer, R. Palme, H. Stein, et al., Mice selected for high versus low stress reactivity: a new animal model for affective disorders, Psychoneuroendocrinology 33 (6) (2008) 839–862. [56] A. Urakubo, L.F. Jarskog, J.A. Lieberman, J.H. Gilmore, Prenatal exposure to maternal infection alters cytokine expression in the placenta, amniotic fluid, and fetal brain, Schizophr. Res. 47 (1) (2001) 27–36. [57] L. Vismari, G.J. Alves, J. Palermo-Neto, Depression, antidepressants and immune sys- tem: a new look to an old problem, Revista De Psiquiatria Clinica 35 (5) (2008) 196–204. [58] A.K. Walker, G. Hawkins, L. Sominsky, D.M. Hodgson, Transgenerational transmis- sion of anxiety induced by neonatal exposure to lipopolysaccharide: implications for male and female germ lines, Psychoneuroendocrinology 37 (8) (2012) 1320–1335. [59] R.P. Waters, M. Rivalan, D.A. Bangasser, J.M. Deussing, M. Ising, S.K. Wood, et al., Ev- idence for the role of corticotropin-releasing factor inmajor depressive disorder [re- view], Neurosci. Biobehav. Rev. 58 (2015) 63–78. [60] M.M. Weissman, P. Wickramaratne, Y. Nomura, V. Warner, H. Verdeli, D.J. Pilowsky, et al., Families at high and low risk for depression: a 3-generation study, Arch. Gen. Psychiatry 62 (1) (2005) 29–36. [61] L. Yang, Y. Zhao, Y. Wang, L. Liu, X. Zhang, B. Li, et al., The effects of psychological stress on depression [research support, non-U.S. Gov't], Curr. Neuropharmacol. 13 (4) (2015) 494–504. [62] R. Yirmiya, Y. Pollak, M. Morag, A. Reichenberg, O. Barak, R. Avitsur, et al., Illness, cy- tokines, and depression, Ann. N. Y. Acad. Sci. 917 (2000) 478–487. [63] A. Zager, G. Mennecier, J. Palermo-Neto, Maternal immune activation in late gesta- tion enhances locomotor response to acute but not chronic amphetamine treatment in male mice offspring: role of the D1 receptor, Behav. Brain Res. 232 (1) (2012) 30–36. [64] J. Zhu, K.P. Lee, T.J. Spencer, J. Biederman, P.G. Bhide, Transgenerational transmission of hyperactivity in a mouse model of ADHD [research support, N.I.H., extramural], J. Neurosci. Off. J. Soc. Neurosci. 34 (8) (2014) 2768–2773. http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0080 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0080 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0085 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0085 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0085 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0090 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0090 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0090 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0095 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0095 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0095 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0100 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0100 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0100 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0105 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0105 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0105 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0105 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0110 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0110 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0110 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0115 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0115 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0115 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0115 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0120 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0120 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0120 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0120 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0120 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0125 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0125 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0125 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0130 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0130 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0135 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0135 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0135 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0140 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0140 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0145 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0145 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0150 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0150 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0155 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0155 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0155 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0160 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0160 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0160 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0165 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0165 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0165 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0170 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0170 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0175 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0175 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0180 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0180 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0185 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0185 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0190 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0190 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0190 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0195 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0195 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0195 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0200 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0200 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0200 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0205 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0205 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0205 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0205 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0210 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0210 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0215 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0215 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0220 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0220 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0220 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0220 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0225 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0225 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0225 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0225 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0230 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0230 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0230 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0235 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0235 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0235 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0240 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0240 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0245 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0245 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0245 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0250 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0250 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0250 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0255 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0255 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0260 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0260 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0265 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0265 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0265 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0270 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0270 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0270 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0275 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0275 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0275 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0280 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0280 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0280 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0280 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0285 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0285 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0285 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0290 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0290 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0290 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0295 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0295 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0295 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0300 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0300 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0305 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0305 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0305 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0305 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0310 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0310 http://refhub.elsevier.com/S0024-3205(16)30118-7/rf0310 Prenatal lipopolysaccharide exposure affects sexual dimorphism in different germlines of mice with a depressive phenotype 1. Introduction 2. Methods 2.1. Ethics statement 2.2. Animals 2.3. Lipopolysaccharide 2.4. Behavioral testing 2.4.1. Tail suspension test 2.4.2. General activity in the open field test 2.5. Experimental design 2.5.1. F0 generation groups: treatment and mating 2.5.2. F1 generation groups 2.5.3. F2 generation groups 2.6. Statistical analysis 3. Results 3.1. F0 generation 3.1.1. Immobility time in males and females 3.2. Effects of lipopolysaccharide treatment in different germ lines 3.2.1. Immobility time in F1 and F2 males and females in the TST 3.2.2. Behavioral trend in male and female offspring 3.2.3. General activity in males and females of the F1 and F2 generations 3.3. Sexual dimorphism in the TST and OF among males and females 4. Discussion 5. Conclusion Acknowledgements References