Objective
We studied whether magnetic resonance imaging (MRI) could be used to detect placental inflammation before the detection of irreversible tissue damage. Next, we tested whether this early detection would enable the administration of treatment (ie, interleukin-1 receptor antagonist [IL-1Ra]) in a realistic clinical time after diagnosis.
Study Design
Pregnant rats were injected intraperitoneally with lipopolysaccharide with/without delayed IL-1Ra. MRI was performed at different time after the injection, and placentas were collected for comparison. Placental inflammation was assessed by determination of the levels of inflammatory cytokines.
Results
Placental inflammation was detected by MRI as early as 3 hours after maternal administration of lipopolysaccharide, concomitantly to IL-1β up-regulation. This was observed before any tissue damage, which appeared only at 24 hours after the administration of lipopolysaccharide. Delayed IL-1Ra administration (after MRI diagnosis) protected the placenta, as seen by the preserved tissue integrity and limited macrophages infiltration in the placental parenchyma.
Conclusion
These findings established a noninvasive diagnostic method for early in utero detection of placental inflammation that would allow the administration of placentoprotective intervention within a clinically relevant delay after diagnosis.
Gestational inflammation is recognized increasingly as a cause of fetal morbidity, which includes brain damage that results in several neurodevelopmental impairments such as learning disabilities, schizophrenia, autism, and cerebral palsy. Those conditions greatly affect patient cognition and/or motor functions and place a heavy burden on the families alongside important social costs. Fetal-threatening placental inflammation, especially in the case of maternal infection that is remote from the placenta, occurs in humans during the last trimester of gestation without any (or only discrete) clinical manifestations. Attempts to develop diagnostic biomarkers for perinatal (maternal and/or neonatal) inflammation, such as detection of inflammatory molecules in the amniotic fluid or neonatal blood, have remained disappointing so far. Those diagnostic tests have several practical limitations that are related to invasive sampling manipulation (eg, amniocentesis) or to delay postnatal results (eg, neonatal blood or placental sampling) and that are not optimal for early and safe in utero diagnosis that allows rapid and efficient therapeutic interventions. In addition, appropriate therapeutic interventions that are aimed at the reduction of placental inflammation are still poorly defined, because of the lack of knowledge about the early inflammatory events that link prenatal inflammation to fetal damage.
Human and experimental studies have implicated proinflammatory cytokines in inflammatory placental and fetal diseases. Among proinflammatory cytokines, we previously showed that interleukin-1β (IL-1β) is a key player in the causal pathway that links maternal inflammation to fetoplacental inflammation and subsequent brain damage in the offspring. However, early placental changes that are induced by maternal exposure to inflammation remain unknown as does our ability to detect and treat rapidly and efficiently before the effects become irreversible.
Using an established model of gestational inflammation that is induced by maternal lipopolysaccharide administration, we investigated whether magnetic resonance imaging (MRI) could be a translational noninvasive technique to detect in utero inflammatory–induced placental changes efficiently at an early stage, which would allow the establishment of a clinically relevant therapeutic window for antiinflammatory treatment.
Materials and Methods
Animals
Timed pregnant Lewis rats were obtained from Charles River Laboratories (Saint-Constant, Quebec, Canada, and Portage, MI) at gestational day (G) 16. They were allowed to acclimatize to our animal facility (20°C environmental temperature; 12-hour cycle; food and water ad libitum) before experimental manipulations. Experiments were approved by the Institutional Animal Research Ethics Review Board; the handling of animals was conducted in accordance with the Animal Care and Use Committee at the Université de Sherbrooke, Canada. In the first set of experiments ( Figure 1 , A) , pregnant rats were injected intraperitoneally with lipopolysaccharide (n = 24 rats; 200 μg/kg in 100 μL of pyrogen-free saline solution, from Escherichia coli , 0127:B8; Sigma, Ontario, Canada) for a short duration (ie, 3, 6, or 12 hours) before MRI or cesarean section delivery was performed at G20 as described later. Separately, another cohort of pregnant rats ( Figure 1 , B) were injected intraperitoneally from G18, every 12 hours, with lipopolysaccharide (n = 16 rats; 200 μg/kg in 100 μL of pyrogen-free saline solution, from E coli , 0127:B8; Sigma) or with saline solution (controls; n = 10 rats; 100 μL) with or without intraperitoneally recombinant human interleukin-1 receptor antagonist (IL-1Ra; 10 mg/kg/12 hours [Kineret Biovitrum, Stockholm, Sweden], starting 12 or 24 hours after the first lipopolysaccharide injection) until either MRI or cesarean section delivery at G20. The dose of IL-1Ra was selected based on our previous study.
MRI
MRI was performed at G20 with a small-animal 7T MRI system (Varian Inc, Palo Alto, CA) as previously described. T 2 -weighted respiration-gated images were acquired with the use of a fast spin-echo pulse sequence (repetion time/effective echo time: 2000/12 msec; 8 echoes; field of view: 6 × 6 cm 2 ; matrix: [256] 2 ; number of averages: 8; 20 slices of 1.5 mm) followed by a dynamic acquisition, with the use of T 1 -weighted images (repetion time/echo time: 197/2.5 msec; field of view: 6 × 6 cm 2 ; matrix: [128] 2 ; flip angle: 30 degrees; number of averages: 4; 20 slices of 1.5 mm) with contrast agent injection (gadolinium-diethylenetriaminopentaacetic acid, 0.143 mmol) by the tail vein during the acquisition. MRI analysis and normalization were performed as previously described with the use of Matlab (MathWorks, Natick, MA) and Microcal software (Origin version 8.0, Northampton, MA).
Placental cytokines quantification and tissue damage
Cesarean section delivery was performed for cytokine quantification and evaluation of tissue damage (3-6 placentas that were selected randomly per dam for each technique), as previously described. Briefly, placenta were homogenized manually in Tris buffer that contained 1% Triton and a protease inhibitor cocktail and centrifuged at 13000 rpm for 30 minutes at 4°C before the supernatant was added. We used Bio-Rad protein assay (Bio Rad Laboratories Canada Ltd, Mississauga, Ontario, Canada) to determine protein concentration . Cytokines were quantified with the rat enzyme-linked immunosorbent assay kit for IL-1β, tumor necrosis factor–α (TNF-α), IL-6 (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. The sample dilution for the enzyme-linked immunosorbent assay was chosen on the basis of previous experiments to be within the linear range that was given by the standard curve for each cytokine. Five micrometer–thick paraffin-embedded placental sections were processed for hematoxylin-eosin staining, immunofluorescence or immunohistochemistry, as previously described. Antibodies against CD68 (macrophage marker, 1:250; Millipore, Billerica, MA), polymorphonuclear leukocytes (1:500; Cedarlane, Burlington, Ontario, Canada) and IL-1β (1:500; AbD Serotec, Raleigh, NC) were used. Slides were mounted with the use of a DAPI-containing medium (Invitrogen Canada Inc, Burlington, Ontario, Canada) for immunofluorescence or counterstained with hematoxylin for immunohistochemistry.
Negative controls consisted in an additional set of sections that were treated the same way but without the primary antibody. The percentage of labeled cells, namely the number of labeled cells divided by the total number of cells, was determined in 3 areas of the placental labyrinth parenchyma at ×400 magnification and counted in triplicate (mean value was tabulated and used for data analysis). The levels of CD68+ cells that were detected in the vessels were based on a gradation system: grade 0, <10 positive cells; grade 1, 11-30 positive cells; grade 2, 31-50 positive cells; grade 3, >50 positive cells, in 1 area at ×400 magnification, that were counted in triplicate (mean value was tabulated and used for data analysis). Randomly selected walls from 2-3 vessels that arise from the fetal side of the placenta were counted per placenta. For both the percentage and levels of positive cells, the mean value from the 3 readings was tabulated and used for data analysis.
Data analysis
Data are presented as means ± SEM. Comparisons were performed with analysis of variance with the Newman-Keuls post-test throughout or unpaired t -test with Welch correction, when specified. The significance level was set at a probability value of .05 (2-tailed probability values).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
