Objective
The purpose of this study was to develop an animal model for intrapartum inflammation at term to investigate the interactions between maternal and fetal inflammatory responses and adverse neurologic outcome.
Study Design
Lipopolysaccharide (160, 320, or 640 μg/kg) was administered intraperitoneally to day 20 term-pregnant Sprague Dawley rat dams 2, 4, and 6 hours before sample collection. Maternal outcomes included dam core temperature and plasma interleukin 6 (IL-6). Fetal outcomes included plasma IL-6, brain IL-6 messenger RNA expression, and brain IL-6 protein expression. Primary cortical cell cultures were prepared to examine neuronal morphologic condition. Neurite counts were obtained with the use of automated Sholl analysis.
Results
Maternal plasma IL-6 levels peaked 2 hours after lipopolysaccharide stimulus and rapidly resolved, except for an observed low level persistence at 6 hours with 640 μg/kg. Fetal plasma and placental IL-6 expression also peaked rapidly but only persisted in placental samples. Fetal brain IL-6 RNA and protein expression was significantly higher than control litters at 6 hours after the exposure to both 320 μg/kg ( P ≤ .05) and 640 μg/kg ( P ≤ .01). Cortical cells from fetuses that were exposed for 6 hours to maternal systemic inflammation showed reduced neurite number and neurite length ( P < .001) with increasing lipopolysaccharide dose.
Conclusion
Our results demonstrate that fetal brain injury follows isolated systemic maternal inflammation and that fetal brain inflammation lags after maternal stimulus, which creates a potential 4-hour clinical window for therapeutic intervention.
Cerebral palsy affects 1-3 of every 1000 newborn infants after term birth, with distinct pathologic conditions that are related to the stage of brain development at the time of insult. In the immature human brain, the periventricular white matter of the motor cortex is the most susceptible to injury; the basal ganglia and gray matter are more likely to be affected at term. Intrapartum chorioamnionitis significantly increases the risk of neonatal encephalopathy and subsequent cerebral palsy; this association is most robust at term. A growing body of evidence suggests that clinical chorioamnionitis at term is largely attributable to maternally driven, noninfectious inflammatory causes rather than ascending infection. Fever at term is associated with low rates of placental infection by detection of bacterial RNA (5.4%). Further, 40% of patients with the diagnosis of clinical chorioamnionitis at term do not have evidence of bacteria in the amniotic cavity with the use of cultivation or molecular microbiologic techniques. Therefore, it is important that experimental models adequately reproduce the likely maternal-to-fetal transmission of inflammation that occurs with term chorioamnionitis.
Rodent models of fetal inflammation have been used widely to investigate the consequences of maternal inflammation on preterm delivery, fetal development and the mechanisms that are involved in preterm brain damage. In addition, 1 model of infectious term chorioamnionitis has been described previously. Because the clinical cause of inflammation varies with gestational age, the models that are used to study the mechanisms of perinatal brain injury should also vary with respect to inflammatory agent, dose, and delivery route. Lipopolysaccharide commonly is used as a nonspecific inducer of inflammation. In studies that are designed to mimic chorioamnionitis at term, it is critical to model maternal systemic inflammation with subsequent fetal inflammation, rather than the preterm mechanisms of ascending intrauterine infection. Therefore, we used an intraperitoneally administered lipopolysaccharide rodent model to induce maternal systemic to fetal inflammation.
The risk of fetal brain injury that is associated with inflammation at term increases synergistically with other adverse exposures, especially hypoxia and hyperthermia. Therefore, it is vital to understand the timing of maximal fetal brain inflammation after a maternal inflammatory stimulus so that secondary insults and interventions can be timed effectively in experimental investigations.
Materials and Methods
Animals
Twenty-four timed-pregnant Sprague Dawley rats (Harlan, Indianapolis, IN) were housed in our animal facility with a 12-hour light cycle (6 am to 6 pm ) and ad libitum access to food and water during both housing and experimental time. All experiments were performed on day 20 of pregnancy. To avoid any circadian effect, all experimental procedures were started at the same time of the day. The experimental protocol was reviewed and approved by the Institutional Animal Care and Use Committees at Medical University of South Carolina and Temple University. Before each experiment, a baseline sample of maternal blood (0.5 mL) was withdrawn from the tail vein and collected in a microcentrifuge tube that contained a solution of disodium dihydrate EDTA (1.8 mg/mL blood) and centrifuged (1300 g , 10 minutes, 4°C). A Lidoderm 5% lidocaine patch (2.5 cm 2 ; ENDO Pharmaceuticals, Malvern, PA) was applied on the dorsum between the shoulder blades 1 hour before the injection of an implantable glass-encapsulated passive transponder that was inserted in a dedicated stainless-steel syringe-type applicator (Implantable Programmable Temperature Transponder, IPTT-300; Bio Medic Data Systems Inc, Seaford, DE). Animals were then returned to their cages for temperature stabilization. After 20 minutes, temperature was recorded with a wireless receiver (DAS-6007; Bio Medic Data Systems Inc) every 2 minutes until at least 5 consecutive recordings were obtained that differed by no >0.3°C.
At this time, either 0.9% sodium chloride (Hospira, Inc, Lake Forest, IL) or increasing doses (0, 160, 320, and 640 μg/kg; n = 6 dams per dose) of Escherichia coli O55:B5 lipopolysaccharide (Calbiochem, San Diego, CA) resuspended and sonicated in normal saline (Hospira, Inc) was injected (500 μL) intraperitoneally. Injection time was counted as time 0 hour.
Core body temperature was then recorded every 2 minutes, and 0.5-mL blood samples were collected from the tail vein every 2 hours until the end of the experiment. Euthanasia was performed by decapitation at 2, 4, or 6 hours after intraperitoneal injection, and all the pups were collected by hysterectomy. From each litter, 3 pups were collected aseptically in ice-cold Ca 2+ and Mg 2+ free Hanks’ balanced salt solution (Gibco, Gaithersburg, MD) to prepare cortical cell cultures. From the remaining pups in each litter, brains and placentas were divided into 2 symmetrical halves; 1 was snap-frozen in liquid nitrogen, and the other was placed in RNAlater at 4°C for 24 hour and stored at –80°C until use. Blood was collected from the vessels of the neck after decapitation with an EDTA-Minivette POCT (Sarstedt, Nümbrecht, Germany).
Enzyme-linked immunosorbent assay
Quantification of interleukin-6 (IL-6) from plasma samples was performed with Quantikine rat IL-6 immunoassay kit (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. Because of the limited volumes that were obtained, plasma that was collected from the pups was pooled for IL-6 quantification. Sample absorbance was quantified by a microplate reader (Bio-Tek Instruments, Winooski, VT) at 450 nm with a correction wavelength set at 570 nm. Values were calculated by nonlinear regression with the use of the GraphPad Prism (version 6; GraphPad Software Inc, La Jolla, CA). The sensitivity of the assay was 21 pg/mL. The coefficient of variation was 8.1%.
Quantitative polymerase chain reaction
Total RNA was extracted from brain and placenta samples by homogenization in TRIzol reagent (1 mL/≤100 mg tissue; Ambion Life Technologies, Austin, TX) according to the manufacturer instructions. Extracted RNA was resuspended in 100 μL of nuclease-free water with the addition of RNasin ribonuclease inhibitor (Promega Corp, Madison, WI) and stored at –80°C. RNA was further purified with the RNeasy Mini Kit combined with the RNAse-Free DNase Set (Qiagen, Valencia, CA) according to manufacturer’s protocol. RNA was analyzed by spectrophotometry (NanoVue Plus; GE Healthcare, Piscataway, NJ) and by gel electrophoresis (6 V/cm). Each sample (2-4 μg) was loaded onto a 1.2% agarose gel with 0.7% formaldehyde added as a denaturing agent.
One-step quantitative polymerase chain reaction (qPCR) was performed with StepOne Plus (Applied Biosystems, Foster City, CA) reverse transcription PCR system with the GoTaq 1-Step RT-qPCR System (Promega Corp). One hundred nanograms of RNA and 500 nmol/L of carboxy-X-rhodamine reference dye (Promega Corp) were added to each 20 μL reaction. The housekeeping gene primer pair (β actin) was diluted to 300 nmol/L, and the IL-6 primer pair was added according to manufacturer instructions. QPCR cycles were set as recommended by the manufacturer, with minor modifications (15 minutes at 37.5°C; 10 minutes at 95°C, and 40 cycles of 15 seconds at 95°C and 1 minute at 60°C). Final melting curves were run by 0.5°C increments (15 seconds at 95°C; 1 minute at 60°C) to 95°C (15 seconds). Single peak melting curves were obtained from all qPCR reactions, which is indicative of specific product amplification (data not shown).
Primer sequences for β actin were obtained from NCBI Blast (F: 5′-AGC CAT GTA CGT AGC CAT CC -3′; R: 5′- CTC TCA GCT GTG GTG GTG AA -3′). The primer pair for rat IL-6 was a RT2 qPCR Primer Assay (Qiagen). For each experimental group, 8 samples from 2 litters were run in triplicate. Relative quantification of gene expression was calculated by 2 -ΔΔCt and are shown as folds of increased messenger RNA transcription after normalization over the control values.
Electrophoresis
QPCR product size was confirmed by gel electrophoresis on Criterion (Bio-Rad Laboratories, Hercules, CA) 15% tris/boric acid/ethylenediaminetetraacetic acid precast gels run for 3 hours in a Criterion Cell (Bio-Rad Laboratories), according to manufacturer instructions. Diluted β-actin samples (1:2) and undiluted IL-6 reactions were run together with a GeneRuler Ultra Low Range DNA ladder (Thermo Scientific, Waltham, MA) were used as reference, according to manufacturer protocol. Gels were poststained for 45 minutes on a rocker in a ×3 solution of RedGel (Biotium, Inc, Haywood, CA) diluted in water. Images were acquired with a Gel Doc EZ system (Bio-Rad Laboratories).
Quantitative western blot
Protein extracts were prepared from rat brain tissues. Cellular debris was removed by centrifugation for 5 minutes at 4°C; the supernatant was assayed for protein content by Bradford analysis (Bio-Rad Laboratories). Fifty micrograms of proteins were diluted with Laemmli sodium dodecylsulfate sample–reducing buffer, (6X), heated at 95°C for 10 minutes, and separated by gradient sodium dodecylsulfate sample–polyacrylamide gel electrophoresis in 1X Tris-Glycine-Sodium Dodecyl Sulfate buffer and transferred to reinforced supported nitrocellulose membranes (Whatman, Piscataway, NJ) for 2 hours at 4°C. Membranes were blocked with Odyssey Blocking Buffer (LI-COR, Inc, Lincoln, NE) for 1 hour at room temperature and incubated with specific primary antibodies for 2 hours in Odyssey Blocking Buffer with 0.1% Tween-20. The blots subsequently were washed 3 times and incubated with IRDye 800CW goat anti-mouse and IRDye 680RD goat anti-rabbit secondary antibodies and visualized with an Odyssey CLx Imaging System (LI-COR, Inc), with the use of Odyssey software (LI-COR Biosciences). Anti–IL-6 mouse monoclonal antibody (ab9324) was purchased from Abcam (Cambridge, MA). Anti-GAPDH rabbit polyclonal (# PA1-987), which was purchased from Pierce Chemical Co (Rockford, IL), served as a loading control.
Cortical cell cultures
Cultures were prepared as previously described. Briefly, brains were removed from the skull and sectioned into ice-cold Hank’s Balanced Salt Solution (Gibco) to discard olfactory bulbs, brain stem, cerebellum, and meninges. Cortices from each litter were stored separately until being processed in ice-cold Hybernate-E complete media (5 mL/pair; Gibco) supplemented as indicated by manufacturer with GlutaMAX-I and B27 supplement (Gibco). Tissues were trypsinized for 15 minutes in 0.05% Trypsin-EDTA (Gibco) diluted in Neurobasal media (Gibco) (1:80; 4 mL/pair of cortices) in a cell culture incubator (37°C; 5% CO 2 ; 95% humidity). Fragments of tissue were then added to neurobasal media that were enriched with 10% heat-inactivated fetal bovine serum (Sigma Chemical Company, St. Louis, MO). Samples were pooled together and further dissociated by mechanical trituration for 1 minute. The cell suspension was collected for cell counting and seeding.
Immunofluorescence
Cells were seeded (40,000 cells/well) into 12-well plates on 0.01% poly-L-lysine coated (Sigma Chemical Company) 18-mm glass coverslips in neurobasal media that were supplemented with GlutaMAX and B27 supplement (neuronal selective media; Gibco). On in vitro day 3, the cells were rinsed with warm phosphate-buffered saline solution (PBS) and fixed with ice-cold 4% paraformaldehyde for 30 minutes at room temperature.
Cultures were costained with DAPI (Sigma-Aldrich, St. Louis, MO) and anti-MAP2 antibody. Briefly, cell membranes were permeabilized with 0.5% Triton X-100 in PBS (Sigma) and nonspecific binding sites were saturated with 2% BSA (Fisher Scientific, Waltham, MA) in PBS. Samples were incubated overnight at 4°C with the primary antibody (ab11267, 1:500) in 1% goat serum (Sigma-Aldrich) in PBS. A secondary antibody solution that was prepared in the aforementioned manner (ab96882, 1:100) was incubated for 1.5 hours in the dark at room temperature. Nuclei were stained for 5 minutes at room temperature with 1 μg/mL DAPI in PBS. Coverslips were mounted with Prolong Gold anti-fade reagent (Invitrogen, Carlsbad, CA), according to manufacturer instructions. Images were acquired with a Nikon i80 microscope (Nikon Instruments Inc, Melville, NY) equipped with a Roper Scientific camera (Q Imaging, Surrey, Canada) with a ×100 objective.
Three brains from 2 litters were used to prepare cell cultures; from each culture, 3 coverslips were stained. From each coverslip, the images of at least 10 single nontouching neurons were acquired randomly (n = 60 per experimental condition). The total number of neurites and the length of each neurite were obtained with the use of automated software for Sholl analysis (ImageJ; National Institutes of Health, Bethesda, MD).
Statistical analysis
Ordinal-by-ordinal testing or 1-way analysis of variance with Least Significant Difference post-hoc testing was performed (SPSS Statistics for Windows, version 20.0; IBM, Armonk, NY). A probability value of < .05 was considered statistically significant. Forty neurons were counted from 6 pooled cortical cultures for each experimental condition (80% power to detect a 33% decrease in mean number of neurites compared with the control group in pairwise analysis). Eight brains were prepared for analysis of inflammatory cytokines for each experimental condition (80% power to detect a 60% increase in mean IL-6 compared with the control group in pairwise analysis).
Results
Lipopolysaccharide reduces maternal core temperature
The lowest dose of lipopolysaccharide that was administered (160 μg/kg) did not have any effect on maternal core temperature. At 320 μg/kg and 640 μg/kg, lipopolysaccharide significantly reduced body core temperature with dose-dependent latency, duration, and magnitude ( Figure 1 ). At the median dose, the effect lasted up to 4 hours after a latency of 1.5 hours; at the higher dose, the effect lasted up to 5.5 hours after a latency of 1 hour. The mean decrease in temperature was –1.0 ± 0.3°C (1.5 hours to 4 hours) after 320 μg/kg and –1.6 ± 0.3°C (1.5 hours to 5.5 hours) after 640 μg/kg.
Maternal and fetal plasma IL-6 concentrations after lipopolysaccharide
Peak maternal plasma IL-6 cytokine response occurred within the first 2 hours, and the magnitude of response did not differ with dose of lipopolysaccharide, except for a persistent low-level maternal response at 6 hours after the highest dose ( Figure 2 , A). Maternal plasma levels declined rapidly to baseline by 6 hours at lower doses. Control group dams had constant undetectable IL-6 plasma levels. Fetal plasma IL-6 levels also peaked at 2-4 hours, but the magnitude of the response did not appear to be dose dependent ( Figure 2 , B). Fetal plasma IL-6 levels returned to baseline by 6 hours at all doses.
Placental and fetal IL-6 messenger RNA expression
Placental IL-6 RNA expression peaked at 2 hours ( Figure 3 , A). At 6 hours, placental IL-6 RNA expression remained elevated compared with controls at all doses. Fetal brain IL-6 RNA expression was increased significantly only at the 6-hour time point and only at doses of 320 μg/kg and 640 μg/kg ( Figure 3 , B). Fetal brain IL-6 RNA expression for individual pups is shown in Figure 3 , C. These data indicate that fetal brain response is largely uniform in this intraperitoneal administration model.
IL-6 protein in fetal brain
Fetal brain IL-6 protein expression was quantified in fetal brain at 6 hours ( Figure 4 ). Given that IL-6 RNA expression was not increased at 160 μg/kg, only specimens from pups that were exposed to 320 or 640 μg/kg were assayed. Protein data confirmed our gene data. IL-6 protein expression was increased significantly at both 320 μg/kg and 640 μg/kg but was highest in the 640 μg/kg group. Further, the data spread again suggests a relatively uniform pup brain inflammatory response to maternal lipopolysaccharide treatment.
