Placental Inflammation Leads to Abnormal Embryonic Heart Development

Background: Placental heart development and embryonic heart development occur in parallel, and these organs have been proposed to exert reciprocal regulation during gestation. Poor placentation has been associated with congenital heart disease, an important cause of infant mortality. However, the mechanisms by which altered placental development can lead to congenital heart disease remain unresolved. Methods: In this study, we use an in vivo neutrophil-driven placental inflammation model through antibody depletion of maternal circulating neutrophils at key stages during time-mated murine pregnancy: embryonic days 4.5 and 7.5. Pregnant mice were culled at embryonic day 14.5 to assess placental and embryonic heart development. A combination of flow cytometry, histology, and bulk RNA sequencing was used to assess placental immune cell composition and tissue architecture. We also used flow cytometry and single-cell sequencing to assess embryonic cardiac immune cells at embryonic day 14.5 and histology and gene analyses to investigate embryonic heart structure and development. In some cases, offspring were culled at postnatal days 5 and 28 to assess any postnatal cardiac changes in immune cells, structure, and cardiac function, as measured by echocardiography. Results: In the present study, we show that neutrophil-driven placental inflammation leads to inadequate placental development and loss of barrier function. Consequently, placental inflammatory monocytes of maternal origin become capable of migration to the embryonic heart and alter the normal composition of resident cardiac macrophages and cardiac tissue structure. This cardiac impairment continues into postnatal life, hindering normal tissue architecture and function. Last, we show that tempering placental inflammation can prevent this fetal cardiac defect and is sufficient to promote normal cardiac function in postnatal life. Conclusions: Taken together, these observations provide a mechanistic paradigm whereby neutrophil-driven inflammation in pregnancy can preclude normal embryonic heart development as a direct consequence of poor placental development, which has major implications on cardiac function into adult life.


Supplemental Materials
Extended Methods

Human Samples
This study was carried out following the Declaration of Helsinki, was approved by the Research Ethics Committee of the University of Southern Santa Catarina (UNISUL) under 34681920.8.0000.5369 (for Preeclampsia samples -approved in 20/08/2020) or CAAE: 36084720.9.0000.5369 (for cardiac defects -approved in 18/01/2021) and all participants signed written informed consent and collected samples have been anonymised.

Animals
Female C57/Bl6J mice and male Balb/C mice were purchased from Charles River UK.
CD45.1 (B6.SJL-Ptprca Pepcb/BoyJ) were originally purchased from Charles River Italy and C57BL/6-Tg(CAG GFP)131Osb/LeySopJ mice from The Jackson Laboratory, respectively. CCR2-deficient (B6.129S4-Ccr2tm1Ifc/J; JAX stock #004999) (Boring et al., 1997) were originally purchased from Jackson laboratory but were provided as a generous gift from Dr Gerard Graham, University of Glasgow. All mice were housed at the animal facility of Queen Mary University of London and were conducted with strict adherence to the Home Office guidelines (PPL P71E91C8E) following approval by the Animal Use and Care Committee of Queen Mary University of London (QMUL). Mice used in the experiments of this study, unless specified were used at 8-12 weeks old

In vivo BrdU uptake
Mice were injected with 100mg/kg 5-Bromo-2′-deoxyuridine (BrdU) i.v (Sigma; B5002-1G) and culled 1 hour later. Single cell suspension were prepared from the required organs and stained with FITC BrdU staining kit (Invitrogen; according to the manufacturer's instructions.

Transfer of GFP + leukocytes into pregnant females
5X10 6 Splenic leukocytes from GFP + age-matched females were injected intravenously into either isotype or NDPI pregnant females. 18 hours after injections, pregnant mice were culled and detection of GFP + leukocytes within the placentas and E14.5 embryonic hearts via flow cytometry and immunofluorescence.

Placenta permeability assay using FITC dextran
To test the permeability of placentas, 10mg/ml of FITC-dextran in PBS (20,000 mW, Sigma Aldrich) was injected intravenously in pregnant isotype control or NDPI mice for 15 minutes. Mice were then culled, and placentas were immediately fixed in 4% paraformaldehyde for two hours followed by 70%ethanol, dehydrated in 30% sucrose, and embedded in OCT for imaging. Wide field imaging was carried out using Nanozoomer S60 slide scanner (Hamamatsu).

Placental neutrophil isolation and splenic monocyte in vitro co-culture
Placentas obtained from E14.5 pregnant mice were cut into small fragments and digested in 1XPBS with 20mM HEPES, 60 U/ml DNase I and 450 U/ml Collagenase I at 37°C and 250rpm for 20 minutes. Suspensions were passed through 70µm filters, washed and red blood cells lysed with ACK lysis buffer. Placental neutrophils contained in the resulting single cell suspension were enriched via positive selection using 0.2µg anti-mouse Ly6G PE (Biolegend, clone 1A8) per 1x10 6 cells followed by anti-PE microbeads (Biolegend), used according to the manufacturer's instructions.
The enriched placental neutrophil population were then used in downstream applications.
Monocytes, from spleens of non-pregnant female C57BL/6J mice were isolated using a commercially available negative selection kit according to manufacturer's instructions (EasySep mouse monocyte isolation kit). 1x10 5 placental neutrophils were co-cultured with 1x10 5 splenic monocytes in DMEM (Gibco) growth media containing 10% FCS and 50 IU/mL penicillin, 50 μg/mL streptomycin and supplemented with 10ng/ml M-CSF (Biolegend; cat#576402) at time of culture and again at day 3 of culture. Cells were cultured for a total of 5 days.

Flow cytometry and antibodies
Single cell suspensions were prepared E14.5 embryo hearts, E14.5 embryo livers and P5 hearts through 70m cell strainers. For placentas (see neutrophil isolation methods) and P28 hearts, enzymatic digestion was used to obtain single cell suspensions. Briefly, P28 hearts were digested using 450U/ml Collagenase I, 125 U/ml Collagenase XI, 60U/ml Dnase1 in 1XPBS containing 20mM HEPES. In both cases, tissues were digested for 20 minutes, at 37C in a shaking incubator at 125 RPM.
Dead cells were excluded using a fixable viability dye LIVE/DEAD fixable aqua (Thermo Fisher Scientific, L34957). Cells were gated on singlets (See Figure S12).
Cells were then further phenotype as indicated in the results. Representative FACS plots show initial gating strategy for singlets and live cells.

Immunofluorescence microscopy
Tissues were fixed in 4% paraformaldehyde for 2 hours and paraffin or OCT embedded. Paraffin embedded sections were de-paraffinised with Histoclear and rehydrated prior to antigen-retrieval performed by heating sections to 95 o C for 10 min in Abcam antigen retrieval solution (ab64236). Sections were stained for hematoxylin/eosin or incubated with primary antibodies for immunofluorescence staining and mounted with Fluoromount G (Thermofisher 00-4958-02) Nuclei were visualized using DAPI. Wide field imaging (bright field and immunofluorescence) was carried out using Nanozoomer S210 or S60 slide scanner (Hamamatsu). Confocal microscopy was carried out on a Carl Zeiss LSM800. Images were analysed with ImageJ (NIH) and Volocity (Version 6.3, PerkinElmer).

High resolution episcopic microscopy (HREM)
Samples for high resolution episcopic microscopy (HREM) were fixed in Bouin's for a minimum of 12h followed by extensive washing in PBS, dehydration in graded methanol series, incubation in JB-4 (Sigma) /Eosine (Sigma)/Acridine orange (Sigma) mix overnight to ensure proper sample infiltration and then embedded in fresh mix by adding the JB4 accelerator. Once polymerized the block where imaged as previously described 46 . Samples were sectioned on a Leica sledge microtome at 1 or 2 µm or on commercial oHREM (Indigo Scientific) at 0.85 or 1.7 µm. An image of the surface of the block was then acquired under GFP excitation wavelength light using Olympus MVX10 microscope and High-resolution camera (Jenoptik). After acquisition the stacks were adjusted for gray level using Photoshop CS6 and then process for isotropic scaling, ortogonal re-sectioning, 25% downscaling, using a mixture of commercial and homemade software (see Wilson R et al. NAR 2016, Vol. 44 D855-D861). 3D volume rendering of the datasets were produced from the 25% downscale stack using Horos. Quantification of number and length of LV trabeculae were measured using Fiji imaging software 47 . Number of trabeculae were measured at the point of invagination within the endocardial wall and were made from 2D images obtained from the 3D rendering by two independent researchers.

In Vivo Cardiac Functional Assessments
Echocardiography was performed on offspring at postnatal day 28 to assess cardiac structure and function using a Vevo-3100 imagining system (VisualSonics, Toronto, Ontario, Canada). Mice were anaesthetized (3% isoflurane in 0.2L/min oxygen, maintained with 2% isoflurane and 0.4L/min) and body temperature maintained at 37 °C. Heart rate was monitored for the duration of the procedure. Images from two-dimensional brightness mode (B-mode) and one-dimensional motion mode (M-mode), were obtained. Left ventricular ejection fraction (EF%), fractional shortening (FS%), cardiac output and stroke volume measurements were obtained from short-axis Mmode. Values were averaged from three beats. Echocardiographic analyses were performed by two independent operators, using Vevo LAB dongle software V3.2.6.

Quantitative Real-time polymerase chain reaction
Placenta, fetal heart and fetal liver samples were stored in RNAlater (Invitrogen) at -80°C until processing. 30mg of placenta tissue was disrupted using a sample grinding kit (GE healthcare) and RLT buffer. Fetal heart and liver tissues were disrupted using a syringe and 25G needle and RLT buffer containing 2-Βme. RNA from the disrupted placental tissue and fetal heart and liver tissue was extracted using the RNeasy mini kit, according to the manufacturer's recommendations. RNA quality and quantity was assessed using absorption measurements (Roche, Nanodrop) prior to performing the reverse transcription reaction, as per the manufacturer's instruction (BioRad, iScript™ cDNA synthesis kit). qPCR was performed using 10ng cDNA, 1µM PCR primers (see table below for sequences) and 1x SYBR GreeN PCR mastermix (BioRad) in a CFX connect light cycler (BioRad). Expression was calculated according to the method whereby data is normalized to two housekeeping genes, HPRT and RPL32A. The thermal cycling conditions were 95°C for 30 seconds and 40 cycles of 95°C for 5 seconds followed by 60°C for 30 seconds and then melt curve analysis 95°C for 5 seconds, 65°C for 5 seconds and 95°C for 5 seconds. Experiments were performed in duplicate.

Bulk RNA sequencing
High-quality RNA was extracted from 3 individual placenta from. Each treatment group, isotype (control) and aLy6G (NDPI) using Qiagen RNA extraction kit (Qiagen, Germany) and submitted to Eurofins (Germany) for Illumina sequencing. All raw data has been deposited on Figshare (https://figshare.com/s/8b13463311cf442e9d15).
Transcript quantification was performed using Salmon (v.0.14.2) and the Mus musculus reference transcriptome (GRCm38, Gencode, release M23) 48 . Transcripts were summarized at the gene level, genes displaying expression levels below one count per million (CPM) in at least 3 individual libraries were filtered out. Differential expression (DE) analysis was performed between NDPI and control using EdgeR 49 .
A False-Discovery Rate (FDR) threshold (≤0.05, Benjamini-Hochberg) was set to define DE genes. Positive log2 fold change indicates that a gene is higher expressed in aLyG6-treated samples and vice versa. Cellular functions and pathways overrepresented for the DE gene list were assessed using the SIGORA in R 50 .

Fetal heart CD45 + cell isolations
Hearts from fetuses at E14.5 were dissected and passed through a 70µm cell strainer to obtain a single cell suspension. Heart CD45 + cells were enriched via positive selection using 0.2µg anti-mouse CD45 PE (Biolegend, clone 30-F11) per 1x10 6 cells followed by anti-PE microbeads (Biolegend), used according to the manufacturer's instructions. The enriched fetal heart CD45 + cells were subsequently used for single cell RNA sequencing.

Single cell sequencing
Fetal heart CD45 cells were fixed using 80% methanol and stored at -80°C according to a GENEWIZ-developed protocol. Upon shipping to the Genewiz facility cryopreserved cells were thawed, washed, and counted following the 10x Genomics protocol. Cell suspensions were loaded onto the 10x Genomics Chromium Controller targeting ~10,000 cells per sample for processing and microdroplet generation. The resulting libraries were sequenced with the Illumina® HiSeq and resulted in 52,874 reads on average per cell, per sample. Primary data analyses were conducted in R and KEGGprofile, respectively. Gene enrichment was also performed using enrichR.
Further plots were generated via scDataviz. Cluster annotations were identified using top expressed genes in each cluster as shown in heat map in figure S9 (see also table in Figure S10) All raw data has been deposited on Figshare     Each symbol represents an individual embryonic hearts from distinct pregnancies and statistical significance was tested by unpaired Student's t-test . ns = not significant, *p≤0.05, ** p≤0.01, ***p≤0.001 **** p≤0.0001. In all cases, data are mean ± SEM. of GFP + leukocytes in isotype and NDPI pregnancies at E14.5. Images taken at x20 magnification and x200 magnification (dotted lined box) (C) Immunofluoresnce images of placentas from In vivo FITC-dextran infusion for 1 hour prior to sacrifice from isotype and NDPI pregnancies at E14.5. Images taken at x20 magnification and x200 magnification (dotted lined box).

Relative gene expression normalized to Control
Cell number per embryo heart Placenta weight (g) Neutrophils were depleted at day 4.5 and 7.5 of pregnancy using αLy6G. Mice were 3 sacrificed at E14.5 of pregnancy and hearts dissected from harvested embryos from control 4 or neutrophil depleted placental inflammation (NDPI) pregnancies. CD45 + cells were isolated 5 from heart single cell suspensions using CD45 PE and anti-PE microbeads. Single cell 6 sequencing was performed on the isolated cells.

Supplementary Figure 10
Data in the table indicate top genes expressed in each cluster of the single-cell analyses, as described in Figure S9. These genes were used to identify the cell types for each of the clusters Supplementary Figure 11 (A) Neutrophils were depleted at day 4.5 and 7.5 of pregnancy using αLy6G. Offspring of these dams were sacrificed at post-natal day 5 (P5) or 28 as indicated and heart structure and immune composition assessed. Offspring from Control (white) and neutrophil depleted (NDPI) (pink). P5 offspring body weights in grams and heart: body weight ratio.