Mid-pregnancy poly(I:C) viral mimic disrupts placental ABC transporter expression and leads to long-term offspring motor and cognitive dysfunction

Limited information is available about the effect of mid-pregnancy viral infections on the placental expression of efflux transporters and offspring behavior. We hypothesized that maternal exposure to polyinosinic-polycytidylic acid [poly(I:C)], a synthetic double-stranded RNA viral mimic, would impair placental cell turnover, the expression of selected ABC transporters and adult offspring behavior. C57BL/6 mice were administered poly(I:C) (10 mg/Kg;ip) or vehicle at gestational day (GD) 13.5 (mid-pregnancy). Dams were euthanized for blood collection 4 h after injection, fetal and placental collection at GD18.5 or allowed to deliver spontaneously at term. At GD 13.5, poly(I:C) induced an acute pro-inflammatory response characterized by an increase in maternal plasma levels of IL-6, CXCL-1 and CCL-2/MCP-1. At GD 18.5, poly(I:C) decreased cell proliferation/death in the labyrinthine and increased cell death in the junctional zones, characterizing a disruption of placental cell turnover. Abca1 and Abcg1 immunolabelling was decreased in the labyrinthine zone, whereas Abca1, Abcg1 and breast cancer resistance transporter (Bcrp) expression increased in the junctional zone. Moreover, adult offspring showed motor and cognitive impairments in the Rotarod and T-water maze tests. These results indicate that viral infection during mid-pregnancy may disrupt relevant placental efflux transporters, as well as placental cell turnover and offspring behavior in adult life.


Scientific Reports
| (2022) 12:10262 | https://doi.org/10.1038/s41598-022-14248-0 www.nature.com/scientificreports/ the tests were performed in the offspring at post-natal day 54 (PND54) and PND93 (Fig. 6A). Due to the fact that previous studies have identified sex differences in the impact of early adversity 21,40 , we determined whether the effects of poly(I:C) on offspring behavior were sex-specific. In the Rotarod test, both males and females born to poly(I:C) treated mothers spent less time on the rotating cylinder compared to their respective control groups at PND54 (Fig. 6B) and PND93 (Fig. 6C), indicating motor coordination deficits. Both male and female offspring from the poly(I:C) treated dams exhibited an increased latency to find the T-water maze platform, indicating deficits in spatial learning (Fig. 6D). No differences were observed in the weight of animals on test days (Supplementary Fig. S2).

Discussion
In this manuscript, we have showed that poly(I:C) exposure in mid-pregnancy impacts fetal and placental outcomes at term (GD18.5). The impairment of motor and neurological function of the offspring at PNDs 54 and 93 caused by poly(I:C) insult during pregnancy may be caused by maternal cytokine/chemokine responses associated with placental remodeling and expression changes of specific placental ABC transporters. The graphs were plotted in aligned dot plot, the gray circles represent the control group, and the black circles represent the poly(I:C) group. One outlier from control group (IL-1β), one outlier from control and poly(I:C) groups (IL-6) were removed according to Grubb's test. Values are expressed as mean ± SD. Mann-Whitney test was performed for fetal/placental weight ratio and IL-1β. Student's t test was used for the remaining analysis. www.nature.com/scientificreports/ Placental efficiency is defined as the relationship between the weight of the fetus and the weight of its respective placenta 41 . Previous studies have shown that poly(I:C) treatment on GD13.5 in the pregnant rats decreased placental weight, area and thickness 42 . Of all the studies involving gestational infections by different pathogens and viral mimics from our group [11][12][13] , poly(I:C) was not able to affect fetal weight and the fetal weight/placental weight ratio.
Few studies consider regional differences when investigating the rodent placenta. However, we previously demonstrated specific and independent responses of Lz and Jz compartments to maternal infective challenges [11][12][13] . , Jz (C) and total placental area (D). Representative images from Ki67 + nuclei and its quantification in Lz (E-H) and Jz (M-P). Representative images from apoptotic nuclei and its quantification in Lz (I-L) and Jz (Q-T). The graphs were plotted in aligned dot plot, the gray circles represent the control group, and the black circles represent the poly(I:C) group. The Lz's (G,K) and Jz's (O,S) negative controls were performed with the omission of the primary antibody. Values are expressed as mean ± SD. Student's t test was used. www.nature.com/scientificreports/ In the present study, we showed that poly(I:C) affects both the Lz and Jz placental compartments, causing region specific placental remodeling. Further, ABC transporter levels in these two placental regions were affected in an opposite way following poly(I:C) treatment, reinforcing the importance of investigating these two placental compartments separately. Considering the fact that the Lz and Jz have different physiological functions, these changes could impact different aspects of fetal development. Despite fetal weight and the fetal/placental weight ratio remaining unchanged, there was a decrease in cell proliferation in the Lz and an increase in cell death in the Jz. Considering that Lz is responsible for nutrient and gas exchanges 8,9 , a decrease in cell proliferation in this region could result in dysfunctional exchange between mother and fetus. With respect to the Jz, because it is primarily responsible for the endocrine function of the placenta 8,9 , an increase in cell death in this region may be affecting the hormonal output during pregnancy, a hypothesis that requires further investigation. Comparing with mid-pregnancy zika virus (ZIKV) infection 12 , Lz cell proliferation pattern in the poly(I:C) group was the opposite of both's high and low dose exposed dams, whereas cell death was similar to low dose ZIKV treatment. In the Jz, both poly(I:C) and ZIKV infection did not affect cell proliferation but did increase cell death in both poly(I:C) and high dose of ZIKV. These data show that poly(I:C) viral analogue and ZIKV infection affect Lz and Jz cell turnover in a specific manner 12 .
Abca1 which is present in the apical membrane of syncytiotrophoblasts, and Abcg1, present in the basolateral membrane, primarily transport cholesterol 4 . Altered levels of these transporters can cause an imbalance in cholesterol homeostasis at the maternal-fetal interface. A decrease in Abca1 causes cholesterol to not return efficiently to the maternal compartment, accumulating in the fetal compartment, whereas a decrease in Abcg1 results in an inefficient transport of cholesterol to the fetal compartment. Pathogens can disrupt placental ABC transporters, but in a pathogen-specific manner. In the present study, poly(I:C) did not modify Abca1 gene expression. This is similar to what we observed following maternal lipopolysaccharide (LPS, bacterial mimic) exposure 11 , but contrary to what we found in a gestational malaria model 13 . At the protein level, poly(I:C) decreased Abca1 staining in the Lz but increased staining in Jz, similar to what was found in malaria and ZIKV models 12,13 . Poly(I:C) decreased Abcg1 gene expression in the Jz of the placenta. However, malaria infection 13 and LPS challenge 11 did not affect Abcg1 mRNA levels in this region of the mouse placenta. Similar to the observed following LPS exposure 11 , poly(I:C) altered Abcg1 protein decreasing its expression in Lz and increasing it in the Jz. Given that cholesterol is essential for several development processes 43,44 , the present data would suggest that mid-pregnancy poly(I:C) treatment has the potential to impair fetal development.
P-gp (Abcb1a/b gene in mice) is responsible for the transport of drugs and cytokines, and Bcrp (Abcg2 gene) transports drugs, porphyrins, and prostaglandins. Both are present in the apical membrane of the syncytiotrophoblast 4 . Previous studies have shown that acute viral infection at GD17-18, modelled through use of poly(I:C), decreases the gene expression of placental Abcb1a/Abcb1b and Abcg2 in rats 15,16 . However, there was no effect on P-gp and Bcrp immunostaining in the placenta when the total placental area was considered 16 . www.nature.com/scientificreports/ Both malaria 13 and poly(I:C) but not LPS 11 decreased Abcb1b mRNA levels in the mouse placenta. In contrast to what was observed in LPS, ZIKV and malaria infections [11][12][13] , placental poly(I:C) exposure did not alter P-gp expression. Placental Abcg2 mRNA decreased following poly(I:C) exposure which was similar to the response to malaria infection 13 , but contrasted with LPS exposure where no change was observed 11 . Bcrp staining did not change in the Lz following poly(I:C), whereas it decreased in the same region after LPS, malaria and ZIKV infections [11][12][13] . In contrast, Jz Bcrp staining increased following poly(I:C) which was similar to that seen after maternal LPS exposure 11 . The fact that P-gp and BCRP staining in the different placental regions is impacted differently in infection models highlights the importance of assessing placental regions separately. Moreover, the decreased expression of Abcb1a/b (P-gp) and Abcg2 (Bcrp) in the mouse placenta following bacterial and viral challenges is consistent with findings of decreased P-gp and Bcrp expression in human placenta 10,14,29,45 . Ultimately, decreased levels of P-gp and Bcrp transporters in the placenta can result in the accumulation of substrates in the fetal compartment, which can be toxic to the fetus and impair fetal development. Therefore, although a causal relationship cannot be inferred from our experiments, it is possible that changes in the expression of efflux transporters in the placenta could also contribute to neurodevelopmental impairments, along with other factors such as the maternal immune response to viral infections. But it still needs to be studied. Another important point that should be taken in consideration when interpreting our data is the impact of the timing of poly(I:C) exposure, as well as of the administration route, on the expression of placental ABC transporters. Even though in our model we have assessed the effects of an acute ip administration of poly(I:C) on maternal cytokine/chemokine response, we have not evaluated the acute effects of poly(I:C) on ABC transporter expression at GD13.5 or GD14.5. Previously, we have demonstrated that acute poly(I:C) exposure does not alter www.nature.com/scientificreports/ placental P-gp function at GD15.5, but rather impairs P-gp function in the fetal blood-brain barrier, leading to greater accumulation of the P-gp drug substrate digoxin in the fetal brain 45 . Future studies should determine whether poly(I:C) (via ip or intravaginal administration) alters the placental expression of the ABC transporters and causes placental remodeling in earlier stages of pregnancy. The concept that maternal immune activation could affect fetal neurodevelopment was first proposed when epidemiological studies found an association between infection during pregnancy and increased rates of neuropsychiatric disorders 46 and impairment of cognitive and affective behaviors in offspring [46][47][48][49][50] . It is assumed that maternal immune activation is able to influence central nervous system development by increasing local production of inflammatory cytokines 39 , in particular interleukin (IL)-6 39,51 . The role of IL-6 has been particularly well studied, and this cytokine appears to be a key mediator of neurodevelopmental outcomes in the offspring 39 . The activation of IL-6 in the placenta may trigger an inflammatory response in the fetal brain and impact the behavior of the offspring 51 . Even though there is still debate as to whether there is unidirectional transfer of cytokines across the placental barrier 52 , it has been demonstrated that maternal IL-6 can cross the placenta 53 and in addition the placenta itself is able to produce this cytokine 39,51,53 and impact fetal brain development. In the current study, maternal mid-pregnancy poly(I:C) exposure induced an acute rise in IL-6 levels in the maternal plasma, which may potentially have contributed to the behavioral alterations in adult offspring herein observed.
In addition, several studies have described an increase in IL-6 in maternal plasma after poly(I:C) exposure, indicating a robust systemic inflammatory response 16,36,37,[54][55][56] . Poly(I:C) can also trigger the release of the chemokines CXCL-1 and CCL-2, possibly through the activation of TLR3 37 . TLR3 knockout mice did not show an increase in IL-6, CXCL-1 and CCL-2 in the placenta and plasma in response to poly(I:C) 55 . The release of www.nature.com/scientificreports/ cytokines activates the maternal immune system, which may culminate in behavioral changes. Therefore, the increase of IL-6, CXCL-1 and CCL-2 in maternal plasma could possibly underly the behavioral changes in the offspring found in the present study. With the exception of CCL-2 levels in female fetal brains ( Supplementary  Fig. S1C), we did not find changes in IL-6, IL-1β and CXCL-1 levels in fetal brains ( Supplementary Fig. S1) probably because the measurement was made five days after the poly(I:C) insult. According to other studies, changes in cytokines levels occur acutely, with peaks at 3 h, 6 h and 24 h after the insult, and then there is a return to baseline levels 36,37,46 . Given the sex-specificity of the effects of the early environment on offspring neurologic outcomes 21,40 , we investigated whether male and female offspring were impacted differentially following an infective challenge. Poly(I:C) led to impaired motor coordination in mice of both sexes at PND 54 and 96. Poly(I:C) also decreased cognitive abilities in both male and female offspring. Previous studies have shown that maternal immune activation can cause sensorimotor deficits 31,33,57 , as well as impairments in object recognition 32 and social interactions in the offspring 33 . It is possible that these deficits may be related to the release of the cytokine IL-6. Animals with deletion of the IL-6 receptor gene do not exhibit behavioral changes as severe as wild type animals 53 .
In summary, we provide evidence that an acute exposure to poly(I:C) during mid-pregnancy causes placental remodeling and changes the expression of ABC transporters in both Lz and Jz at term, and that these phenotypes may potentially be related to the long-lasting behavioral dysfunction observed in the offspring. Further studies are needed to elucidate whether the change in placental transporters would be correlated with behavioral changes, and which pathways or substrates would be involved in this process.

Experimental model. All the animal experimentation in this study was approved by the Animal Care
Committee of the Health Sciences Center from the Federal University of Rio de Janeiro (protocol number 036-16) and registered with the Brazilian National Council for Animal Experimentation Control. The humane animal care was in compliance with the "Principles of Laboratory Animal Care" formulated by the National Society C57Bl/6 J mice were housed in a temperature-controlled room (23 °C), under a 12/12 h light/dark cycle, and had free access to fresh food and water. Female mice aged between 8 and 10 weeks were mated overnight with male mice at a ratio of 2-3 females to each male. After 12 h, the males were removed from the cages, and if the vaginal plug was present, this day was considered as gestational day (GD) 0.5. The confirmation of pregnancy was performed by measuring the weight gain between GD0.5 and GD13.5. Weight gain of 3 g or more confirmed pregnancy. Once pregnant, dams were randomly distributed in experimental groups. For placental and fetal analysis (Fig. 1A), an intraperitoneal injection (ip) of poly(I:C) (Sigma Aldrich, Germany; Catalog Number P1530; lot #114M4028V), dissolved in 100 μL of sterile phosphate-buffered saline (PBS), was administered at a dose of 10 mg/kg at GD13.5 (n = 8). This gestational day was chosen since the murine placenta is fully formed on GD10.5. Furthermore, GD13.5 is somewhat equivalent to the first trimester of human pregnancy. Animals from the control group (n = 8) received an ip injection of the vehicle (100 μL of sterile PBS) at the same day. Maternal blood was collected through the caudal vein after 4 h of the treatment, centrifuged at 1500 G and the plasma was stored at − 20 °C. At GD 18.5 dams were euthanized with an ip injection of sodium pentobarbital at an overdose of 300 mg/kg. Each dam represents a litter (5-8 pups per litter) and the values of fetal and placental weights correspond to the litter mean per group. Placentas and fetuses were weighed, and the placentae closest to the mean weight in a litter were selected for further analysis and cut in half using umbilical cord insertion as reference 58,59 .
One-half of the placental disk was frozen in liquid nitrogen for qPCR (n = 8 per group), and the other half was fixed overnight in buffered paraformaldehyde 4% (n = 5 per group; Sigma-Aldrich, Brazil) for protein immunostaining analysis. For behavioral tests, a second cohort of 6 vehicles and 6 poly(I:C) treated dams were allowed to deliver spontaneously. Here, each dam also represents a litter (5-8 pups per litter) and all newborns in the litter were evaluated per group. The resulting offspring were nursed by their dams until postnatal day (PND) 21.
After weaning, they were housed under the same maternal conditions until the behavioral tests were performed at PNDs 54, 93 and 96.
qPCR. Total placental RNA was extracted from 8 placentas from each group, using the TRIzol method according to the manufacturer's instructions (TRIzol Reagent; Life Technologies, USA). The concentration of total RNA was assessed using NanoPhotometer (Implen, Munchen, Germany) and samples with RNA purity (260/280 absorbance) ratio ranging between 1.8 and 2.0 and with proven RNA integrity (confirmed through gel electrophoresis) were included in the study. Total RNA (1 μg) was used to synthesize cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, USA) according to the manufacturer's instructions. mRNA levels of selected ABC transporters (Table 1) were evaluated by qPCR following the manufacturer's recommendations (EVAGREEN; Solis Byodine, USA) and using the QuantStudio 3 Real-Time PCR System (Thermo Fisher, USA), with the following cycling conditions: combined initial denaturation at 50 °C (2 min) and 95 °C (10 min), followed by 40 cycles of denaturation at 95 °C (15 s), annealing at 60 °C (30 s) and extension at 72 °C (45 s). For quantification, the LinRegPCR program was used according to Ruijter et al. (2009) 60 . Tests with 95-105% efficiency were considered acceptable. Samples that did not pass in the quality checks of LinReg software (no amplification, no plateau, PCR efficiency outside 10%, excluded from mean Eff) were removed from the analysis. The reference genes were chosen according to their Cq values and variances between the groups. Gene expression was normalized to the geometric mean of reference genes, B2m and Pol2a, which exhibited stable expression levels following poly(I:C) insult. DNA contamination was ruled out using intron-spanning primers, reverse transcriptase-negative samples and melting curve analyses obtained from each qPCR reaction. All samples and standards were measured in duplicate.
Histological, immunohistochemical and TUNEL analysis. Five fixed placentas from 5 pregnancies of each group were processed following a protocol with increasing concentrations of ethanol (Isofar, Brazil), www.nature.com/scientificreports/ diaphanization (or clarification) with xylol (Isofar, Brazil) and inclusion in paraffin (Easypath, Brazil), followed by tissue sectioning in 5 μm thickness using a Rotatory Microtome CUT 5062 (Slee Medical GmbH, Germany). Slides were then subjected to Periodic Acid-Schiff (PAS) staining, immunohistochemistry or terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay, as described below. For PAS staining, following diaphanization with three xylol baths, and hydration with decreasing concentrations of ethanol (100%, 90%, and 70%), placental sections were oxidized with 0.5% periodic acid (Sigma-Aldrich, USA) for 20 min, washed in flowing water and incubated with Schiff 's reagent (Merck Millipore, Germany) for 10 min at room temperature.
For immunohistochemical analysis, after deparaffinization and rehydration, sections were exposed to hydrogen peroxide (3%) for 30 min and washed with PBS containing 0.2% Tween. Antigen recovery was achieved by immersing the slides in Tris-EDTA buffer (pH 9.0) followed by immersion in sodium citrate buffer (pH 6.0) as previously described 13 . The slides were then incubated in bovine serum albumin (3%) in PBS for 1 h to block non-specific antibody binding sites. Slides were then incubated overnight at 4 °C with one of the following primary antibodies: anti-Ki67 ( The TUNEL method was used for the detection of apoptotic nuclei, using the ApopTag® In Situ Peroxidase Detection Kit (Merck Millipore, USA), according to the manufacturer's recommendations.
After PAS staining, immunohistochemistry or TUNEL staining, the sections were counterstained with hematoxylin, dehydrated in increasing concentrations of ethanol (70%, 90%, and 100%), immersed in 3 baths of xylol and mounted with coverslips using Entellan (Merck, Germany). Image acquisition was performed using a highresolution Olympus DP72 (Olympus Corporation, Japan) camera coupled to an Olympus BX53 light microscope (Olympus Corporation, Japan). Fifteen digital images from different random fields were captured per tissue fragment of each placental zone (Lz and Jz) in 5 control mice and 5 poly(I:C)-treated mice, with a total of 300 digital images for each analysis. For the analysis of PAS-stained sections, the area of each region was measured using the free-drawing tool of the Image J software (National Institutes of Health, USA). Quantification of Ki-67 and TUNEL immunostained nuclei was performed using the STEPanizer software 61 . Quantification of the area stained by P-gp, BCRP, Abca1 and Abcg1 staining was performed with the mask tool present in the Image Pro Plus 5.0 software. The percentage of viable tissue area was considered upon exclusion of negative spaces. All negative controls were performed with the omission of the primary antibody. For all analysis, the examiner was blinded to group allocation. Offspring's behavioral analysis. The rotarod test was used to evaluate motor coordination, while the T-water maze was employed for the evaluation of cognitive function (spatial learning). All behavioral tests were carried out in a silent and dimly lit experimental room in the afternoon, and the order of mice was randomized. The examiner was blinded to treatment allocation and the results were analyzed considering the mean of each litter.

Measurement
Rotarod. The rotarod performance test was performed at PND54 and 93. The animals were placed in a neutral position and the rod was set to accelerate from 3 to 37 rpm in 5 min. Mice were subjected to three trials per session, with an interval of 10 min between each trial. The latency to fall was recorded automatically in each trial. The longest time spent on the rotarod was chosen for each animal.
T-water maze. The t-water maze task was performed at PND96. In this task, the ability of the mouse to remember the spatial location of a submerged platform was evaluated. The T-maze apparatus (length of stem, 60 cm; length of arms, 45 cm; width, 19 cm; height of walls, 20 cm) was made of clear fiberglass and filled with water (23 ± 1 °C) at a height of 15 cm. An escape platform (17.5 × 14 cm) was placed at the end of the target arm and was submerged 1 cm below the surface. The position of the platform was chosen randomly for each animal before testing. In this test, which allows the evaluation of left-right spatial learning, the mice were placed in the stem of the T-maze and swam freely until they found the submerged platform (located either in the right or in the left arm of the T-maze apparatus) and escaped to it. If the animals did not find the platform within 60 s, they were gently guided onto it. After reaching the platform, the mice remained on it for 20 s. The time to reach the platform was recorded in a training trial (trial 1) and in 3 testing trials (trials 2-4). The average of the three testing trials per litter was used for statistical analysis.