Human placentas. All participants provided informed consent to the use of deidentified and discarded term placentas (37–41 weeks) after uncomplicated labor and delivery, under protocols approved by the Institutional Review Board at the University of Pittsburgh. This included the use of whole placentas for trophoblast cell dispersal (protocol STUDY20050066) or for placental biopsies (protocol STUDY19120076). These biopsies (5 mm3) were obtained immediately after delivery, using a region of the placenta that is midway between the cord insertion and the placental margin and between the chorionic and basal plates, as we previously detailed (99). All tissues were either snap-frozen in liquid nitrogen and stored at −80°C until processed or fixed with 4% paraformaldehyde (Thermo Fisher Scientific) in 0.1 M phosphate-buffered saline (PBS) and embedded in paraffin or in cryomold. Sections (5 μm) were used for hematoxylin/eosin staining and microscopic analysis.
Cell culture, ligands, and medium human chorionic gonadotropin and lactate dehydrogenase measurement. Term placental cells were dispersed, and PHT cells were isolated using a modification of previously published trypsin-DNase-dispase/Percoll protocols, as previously described (100, 101). PHT cells were cultured at a density of 350,000 cells/cm2 in DMEM (Sigma-Aldrich) supplemented with 10% fetal bovine serum (Sigma-Aldrich) and 1% antibiotics, as we previously detailed (102). The BeWo trophoblast cells (ATCC, CCL-98) were maintained in F12K Kaighn’s modified medium (Gibco) supplemented with 10% bovine growth serum (HyClone) and antibiotics. The HEK293T cell line (ATCC CRL-11268) was used for lentivirus transductions, as we previously described (102) (and see below). Some of the cells were cultured on coverslips in less than 1% O2 hypoxic atmosphere, using a hypoxia chamber (Thermo Electron), as we have described (66, 103). Where indicated, after 24 hours in culture, the cells were supplemented for 24 to 48 hours by a fatty acid mix of LA (stock 200 mM) and OA (stock 100 mM, Sigma-Aldrich or Cayman Chemical) or various other fatty acids (stock 100 mM, all from Cayman Chemical), as indicated, including palmitic acid, AA, docosatetraenoic acid (adrenic acid), docopentaenoic acid, or DHA. All fatty acids were preincubated with serum-free medium containing 0.5% fatty acid-free BSA (Sigma-Aldrich) for 20 minutes at 37°C to allow the formation of a fatty acid–albumin complex and used at a final concentration of 100 μM in 0.1% ethanol. Culture medium levels of human chorionic gonadotropin (hCG) and lactate dehydrogenase were measured using enzyme immunoassay, as we previously described (102).
Real-time PCR and quantitative PCR. Total RNA from placentas, cells, or human tissues (FirstChoice Survey Panel catalog 6000, Ambion) was extracted using TRI reagent (Thermo Fisher Scientific) according to the manufacturer’s protocol. RNA samples were further purified with on-column RNase-free DNase (QIAGEN). Real-time quantitative PCR (RT-qPCR) was performed as we previously described, in duplicate, with SYBR Select Master Mix (Thermo Fisher Scientific, catalog 4472908), according to standard procedures, as we previously described, using the ViiA 7 Real-Time PCR System (Thermo Fisher Scientific) (104, 105). The results were calculated using the 2−ΔCT method (106) and normalized to the expression of the housekeeping gene GAPDH or 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ). The specificity of amplification was confirmed using a dissociation curve of each PCR product. Control H2O samples were included in each qPCR experiment. All primer sequences are provided in Supplemental Table 3.
For regular PCR, RNA was extracted as above and processed with DNase I (Ambion). For reverse transcription, total RNA (1 μg) was added to 4 μL High-Capacity RNA-to-cDNA Master Mix (Applied Biosystems) according to the manufacturer’s instructions and diluted into 100 μL. For PCR, we mixed cDNA (2 μL) with a dNTP mixture (2.5 mM) and forward and reverse primers as noted in Supplemental Table 3, along with 2 μL of Ex Taq buffer and 0.1 μL of Ex Taq HS (Takara Bio) in a 20 μL reaction volume. PCR was performed using 30 to 35 cycles of 10 seconds at 98°C, 30 seconds at 55°C, 1 minute at 72°C. The PCR product (8 μL) was loaded and electrophoresed in 2% DNA gel and visualized using a UVP BioImaging System (Ultraviolet Products) or ChemiDoc MP Gel Imaging System (Bio-Rad).
Tissue immunofluorescence and laser-capture microdissection. Immunofluorescence of PNPLA9 in human or mouse placental sections was performed as previously described (102). A similar procedure was used for detection of CGI58, including blocking in donkey serum for 2 hours and incubation with primary anti-CGI58 rabbit monoclonal antibody and a secondary antibody, as detailed in Supplemental Table 4. For laser-capture microdissection, frozen sections (7 μm) of cryomold-embedded mouse placentas were transferred to polyethylene naphthalate membrane slides (Thermo Fisher Scientific) and stained with toluidine blue. The Leica LMD7000 laser-capture microdissection system was used to visualize, circumscribe, and collect 3 tissue types: labyrinth, junctional zone, and decidua. RNA was extracted from the microdissections, using the RNeasy FFPE Kit (QIAGEN, catalog 73504) according to the manufacturer’s instructions. The SuperScript VILO synthesis kit enzyme (Thermo Fisher Scientific, catalog 11755500) was used for the reverse transcription reaction, according to the manufacturer’s instructions. qPCR was performed in duplicate as detailed above.
Western immunoblotting. The source of all antibodies is provided in Supplemental Table 4. Cultured cells were washed with cold PBS, scraped, pelleted, and lysed in PBS containing 0.25 M sucrose and 0.05% NP-40 or in a 50 mmol/L Tris-HCl buffer containing 150 mmol/L NaCl, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (pH 7.5). Both lysis buffers contained protease inhibitor (Thermo Fisher Scientific, catalog 1861278). Following lysis, the mix was centrifuged at 12,000g for 15 minutes at 4°C and the supernatant was collected. Snap-frozen mouse placentas were thawed and homogenized or sonicated in the lysis buffer as above. Total protein concentration was measured using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Protein aliquots (20–30 μg) were separated by 8% to 10% SDS-PAGE and transferred to PVDF membranes (Bio-Rad). Blots were blocked with 5% nonfat milk in Tris-buffered saline with 0.1% Tween 20 (TBST) for 0.5 to 1 hour and incubated with the relevant primary antibody overnight at 4°C. Following TBST washes, membranes were incubated with the appropriate secondary antibody (Supplemental Table 4) for 1 hour at room temperature and enhanced using WesternBright Sirius HRP substrate (Advansta). Chemiluminescence signal was detected using the ChemiDoc MP Gel Imaging System.
Expression, purification, and cross-linking of CGI58 with Sulfo-SBED. Human CGI58 expression sequence was PCR-amplified with primers designed to introduce BamHI and NotI sites using human CGI58 expression plasmid as a template (GE Healthcare Life Sciences, MHS1010-74037 clone 3920855). The PCR product was cloned into the pET-Duet1 plasmid (Addgene plasmid 71146) for bacterial expression. Mouse Cgi58 was PCR-amplified with primers including a His tag and flanked by BamHI/XhoI sites for cloning into pcDNA3.1. Mouse Pnpla2 expression sequence was PCR-amplified with primers that include NotI/KpnI sites for cloning into pFlag-CMV2 (Sigma-Aldrich). Mouse Pnpla9 expression sequence was PCR-amplified with primers that include EcoRI/KpnI sites using mouse Pnpla9 expression plasmid as a template (GE Healthcare Life Sciences, MMM1013-202769194 clone 5329147). The PCR product was cloned into pFlag-CMV2. All constructs were verified by sequencing (GENEWIZ).
The plasmid pETDuet-1-His-CGI58 was transformed into Rosetta DE3 competent cells (Sigma-Aldrich), induced with 0.5 mM IPTG, and then grown for an additional 3 hours at 25°C. The cells were pelleted at 4,000g at room temperature and resuspended in PBS that contained 0.5% Triton X-100, 50 μg/mL lysozyme, and protease inhibitors (cOmplete ULTRA Tablets, Roche catalog 05892970001, Sigma-Aldrich). The mixture was sonicated and centrifuged at 100,000g for 1 hour at 4°C, the cell lysate was diluted to reduce the Triton X-100 concentration to 0.1%, and imidazole (5 mM) was added to the lysate. A tricorn column (Sigma-Aldrich) was packed with His-select beads (Sigma-Aldrich) and loaded with the lysate using an Äktaxpress system (GE Healthcare). CGI58 was then eluted with 300 mM imidazole after extensive washes with increasing concentrations of salt (20–80 mM).
Sulfo-SBED was used according to the manufacturer’s instructions (Thermo Fisher Scientific, catalog 33034). Briefly, Sulfo-SBED was dissolved in DMSO and added to purified CGI58 in a 5 M excess, then incubated at room temperature for 30 minutes, protected from light. The samples were desalted using 2 mL Zeba spin desalting columns (Thermo Fisher Scientific, catalog 89890), The protein was eluted and conjugation with Sulfo-SBED was confirmed by Western blot using streptavidin-HRP (Thermo Fisher Scientific, catalog 21130) and by quantifying biotin (Thermo Fisher Scientific, catalog 28005).
Co-precipitation of CGI58-interacting proteins. As a source of CGI58-interacting proteins, we homogenized WT C57BL/6 mouse placentas in a PBS buffer that contained 0.25 M sucrose, 0.05% NP-40 (Roche catalog 11332473001, Sigma-Aldrich), and protease inhibitors, as noted above. The homogenized tissue was incubated for 30 minutes on a rotating platform at 4°C and then centrifuged at 13,000g for 15 minutes. Labeled and cross-linked CGI58 protein was incubated with the placenta lysate overnight on a rocker at 4°C, protected from light. The mixture was transferred to a plastic well on ice and irradiated using a long-wave UV lamp (365 nm) for 10 minutes. The sample was transferred to a microcentrifuge tube, and 150 μL of Dynabeads MyOne Streptavidin T1 (Thermo Fisher Scientific, catalog 65601) was added and allowed to mix for 2 hours at 4°C. The beads were washed with a high-salt, 300 mM NaCl buffer and then with the homogenization buffer. The Dynabeads were resuspended in 1× SDS sample buffer and boiled at 95°C for 10 minutes before resolving on SDS-PAGE gel. Gel slices were cut and submitted to the University of Pittsburgh’s Biomedical Mass Spectrometry Center for processing and analysis. All proteins shown in Supplemental Table 2 were identified in the 65–80 kDa gel slice. A similar process was repeated for validating the interaction of CGI58 with PNPLA9, using 8% SDS-PAGE gel, followed by transfer to a PVDF membrane and probing for CGI58 and PNPLA9, using antibodies as noted earlier (Supplemental Table 4).
For co-precipitation of purified His-CGI58 and endogenous PNPLA9, we repeated the same procedure described above, using 2 WT C57BL/6 mouse placentas. After incubation, the lysate was precleared with Protein G Dynabeads (Thermo Fisher Scientific, catalog 10003) for 30 minutes at 4°C. The precleared lysate was mixed with 12 μg of purified CGI58 protein and 10 μg of anti-His antibody or mouse IgG as a negative control. The sample was mixed at 4°C overnight, and Protein G Dynabeads were then added for 2 hours at 4°C. The Dynabeads were washed and resuspended in 1× SDS sample buffer, then processed for Western immunoblotting with antibodies as above.
To assess the interaction of CGI58 with ectopically expressed tagged proteins, we used FuGENE 6 (Promega, catalog E2691) to transfect HEK293T cells with plasmids expressing mouse pcDNA3.1-CGI58, mouse pFlag-CMV2-PNPLA2, or mouse pFlag-CMV2-PNPLA9. Two days after transfection, the cells were collected and resuspended in a 100 mM NaCl buffer that contained 50 mM Tris-HCl pH 7.2, 1% Triton X-100, 2 mM β-mercaptoethanol, and protease inhibitors (as above). The samples were incubated on a rotating platform at 4°C and centrifuged at 13,000g for 15 minutes. The lysate was precleared with Protein G Dynabeads for 30 minutes on a mixer at 4°C. The precleared lysate was transferred to another tube, and 10 μg of anti-His or anti-FLAG antibody (Supplemental Table 4) was added per immunoprecipitation. The samples were mix-incubated at 4°C overnight. Protein G Dynabeads were mixed the next day for 2 hours at 4°C. The beads were washed, resuspended in SDS sample buffer, and processed for immunoblotting with 2 μg/mL anti-His antibody or 2 μg/mL anti-FLAG antibody (Supplemental Table 4).
KD or CRISPR/Cas9-mediated protein KO in BeWo cells. We have previously described the generation of stably expressing, doxycycline-inducible Cas9 in BeWo cells and their use to create PNPLA2-KO and PNPLA9-KO BeWo cells (102). The Cas9 plasmid (Addgene plasmid 50661) was generated by the Lander and Sabatini Labs (107). The plasmid pLKO5.sgRNA.EFS.GFP (Addgene plasmid 57822) was used for the PNPLA2 guide, and the plasmid pLKO5.sgRNA.EFS.tRFP657 (Addgene plasmid 57824) was used for the Pnpla9 guide. Both plasmids were generated by the Ebert Lab (108).
For KD of CGI58 we used shRNA lentiviruses that were produced using our published protocol (109). Briefly, we transiently transfected HEK293T cells with a combination of 3 lentiviral plasmids, including pLV-CGI58 shRNA (RHS4430-200174486, sequence 5′-GCATTAAAGCAGCGTATC-3′, Horizon Discovery, Dharmacon) alongside pMD2.G (Addgene plasmid 12259) and psPAX2 (Addgene plasmid 12260). After 60 hours, the lentiviruses in the conditioned media were pelleted by centrifugation at 18,000g for 2 hours at 4°C and resuspended in 100 μL of PBS. BeWo cells were transduced with concentrated lentiviruses in the presence of 8 μg/mL polybrene (Sigma-Aldrich; TR-1003-G). At 48 hours after transduction, BeWo cells were selected with 5 μg/mL puromycin (Invivogen; ant-pr-1) for 7 days to ensure the nontransduced WT cells were eliminated.
Mice and breeding. All mouse-related procedures and experimental protocols were approved by the IACUC of Magee-Womens Research Institute and the University of Pittsburgh (IACUC 19075519) and conducted in accordance with United States Public Health Service policy, as defined in the Guide for the Care and Use of Laboratory Animals (National Academies Press, 2011). Pnpla9-mutant mice (129S6/BL/6) were obtained from Jackson Laboratory as we previously detailed (102). Pnpla2- and Cgi58-mutant mice (129SV/BL/6) were provided by Rudolf Zechner (University of Graz, Graz, Austria) and Erin Kershaw (University of Pittsburgh, Pittsburgh, Pennsylvania, USA) and previously published (48). All data were validated using mice that were bred more than 6 times into the C57BL/6 strain. Timed breeding experiments were performed by pairing males and females overnight. Generation of Pnpla2/Pnpla9-DKO placentas was accomplished by mating Pnpla2KO Pnpla9Het males with Pnpla2Het Pnpla9Het females. The morning after mating was considered E0.5 after verifying the presence of a vaginal plug. All mice were kept under standard conditions of 12-hour light/12-hour dark cycle and fed a regular rodent chow diet and water ad libitum. Pregnancy was confirmed by a 10% weight gain at E13.5. Dams were euthanized by CO2 followed by cervical dislocation on E17.5. Fetuses and placentas were delivered transabdominally and weighed. One half of each placenta was fixed using 4% paraformaldehyde and the other half snap-frozen and stored at −80°C for Western blot, TG, or lipidomic analysis. Genomic DNA was isolated from fetal tails using the HotSHOT (hot sodium hydroxide and Tris) method, and genotyping was performed by standard PCR (Verity, Applied Biosystems). The list of all primers used for genotyping is provided in Supplemental Table 5.
For in vivo exposure to hypoxia, timed pregnant mice (E11.5–E12.5) were exposed to hypoxic conditions of 11% O2 for 5 to 6 days, using a hypoxia chamber designed specifically for mouse experiments (Coy Laboratory Products). Following removal from the hypoxia chamber, some of the mice were immediately sacrificed while others underwent a period of reoxygenation prior to sacrifice, as detailed in the Results section. After sacrifice, fetuses and placentas were delivered transabdominally for use in histological and biochemical analysis, as noted earlier.
Lentiviral transduction of blastocysts for CGI58 “rescue.” Cgi58-Het mice in C57BL/6 were crossbred for at least 6 generations with ICR mice, creating Cgi58-Het mice in the ICR (Jackson Lab) background that were used for blastocyst manipulations. Blastocyst-specific CGI58 overexpression was performed using the procedure we previously described in detail (70). Briefly, CGI58 overexpression was performed using the following 3 lentiviral DNA plasmids: expression plasmid FUGW (Addgene plasmid 14883) for overexpression of CGI58 (or no expression sequences, as control) driven by EF1a promoter, envelope plasmid pLTR-G (Addgene plasmid 17532), and packaging plasmid pCD/NL-BH*DDD (Addgene plasmid 17531). Lentiviruses were produced in HEK293T cells (70), and lentiviral titers were quantified using ELISA (Zeptometrix 0801111B). ICR females were injected with pregnant mare serum gonadotropin on day 1 and with hCG on day 3, when breeding took place. On day 7 of the experiment, or E3.5, blastocysts were flushed from the uterus, processed for lentiviral transduction, and transferred back to pseudo-pregnant females, now considered E2.5. Pregnancy was confirmed by 10% weight gain on E13.5, and delivery was at E18.5. Note that, on average, 18.2 (range 5–20) blastocysts were transferred, yielding a mean of 8.2 (range 1–16) live pups. Placental and fetal Cgi58 genotype was determined using the primers listed in Supplemental Table 5, which also confirmed that 25% of the newborns were Cgi58 KO.
Assessment of neutral lipids in tissues or cells. Oil Red O (Sigma-Aldrich) or LipidTox (Thermo Fisher Scientific) was used for staining of neutral lipids, as we previously described (66), for quantification of lipid accumulation in mouse placentas. Sections were counterstained with hematoxylin and mounted with aqueous mounting medium. Images were acquired with a 90i widefield microscope (Nikon), and quantification of lipid staining in the placental labyrinth was done using Nikon NIS-Elements software.
We used BODIPY, diluted to 10 μg/mL in PBS (Molecular Probes), to quantify neutral LDs in cells, which were cultured on coverslips coated with 0.01% poly-l-lysine (Sigma-Aldrich) in a 12-well plate, as we previously described (66). The cells were rinsed with cold PBS, fixed in 2% paraformaldehyde for 20 minutes, and washed with PBS. Nuclei were stained with DAPI (Sigma-Aldrich) as described (66). Images were acquired with a Nikon A1R confocal system, and quantification of LDs was done using Nikon NIS-Elements software or ImageJ software (NIH). The total number, area, and average size of LDs stained by BODIPY, along with the total number of nuclei for each field, were recorded.
For quantification of TGs in tissues or cells, whole mouse placentas were homogenized in a 5% NP-40 buffer in 2 mL/100 mg tissue or 500 μL/100 mg protein for cells. The samples were heated at 90°C for 2 minutes, then cooled to room temperature. The heating and cooling steps were repeated twice, and the samples were centrifuged at 8,000g for 15 minutes at room temperature. The clear lysate was used to measure TG content, using a colorimetric Triglyceride Quantification Kit (BioVision) according to the manufacturer’s instructions. Absorbance was determined at 570 nm using a VersaMax spectrometry microplate reader (Molecular Devices).
Reverse-phase column separation of TG. To analyze TG in cultured trophoblasts or in mouse placental tissue, total lipids were extracted by the Folch procedure (110) and analyzed by a Dionex Ultimate 3000 HPLC coupled online to a Q-Exactive hybrid quadrupole-orbitrap mass spectrometer (Thermo Fisher Scientific). TGs were separated on a reverse-phase column [Luna 3 μm C18 (2) 100A, 150 × 1.0 mm, Phenomenex]at a flow rate of 0.065 mL/min. The column was maintained at 35°C. The analysis was performed using gradient solvents (A and B) containing 0.1% NH4OH. Solvent A was methanol and solvent B was propanol. The column was eluted for 2 minutes from 0% B to 2% B (linear), from 3 to 6 minutes with a linear gradient from 2% solvent B to 3% solvent B, then isocratically from 3 to 18 minutes using 3% solvent B, 18 to 35 minutes with a linear gradient from 3% solvent B to 40% solvent B, 35 to 60 minutes using a linear gradient from 40% to 55% solvent B, then isocratically from 60 to 65 minutes at 55% solvent B, then from 65 to 80 minutes from 55% to 0% B (linear), followed by equilibration from 80 to 90 minutes at 0% B.
MS and MS/MS analysis of TGs. MS and MS/MS analysis of TGs were performed mostly on a Q-Exactive hybrid quadrupole-orbitrap mass spectrometer, and part of experiments were performed on the ion trap LXQ (Thermo Fisher Scientific). TG cations were formed through molecular ammonium adduction (TG + NH4). Positional analysis of acyl chains in TG species was performed after collision-induced dissociation fragmentation of TGs (111, 112). Analysis was performed in positive-ion mode at a resolution of 140,000 for the full MS scan and 17,500 for the MS2 scan in a data-dependent mode with an inclusion list for TGs. The scan range for MS analysis was m/z 300–1,200 with a maximum injection time of 128 ms using 1 microscan. A maximum injection time of 500 ms was used for MS2 (high-energy collisional dissociation) analysis with collision energy set to 24. An isolation window of 1.0 Da was set for the MS and MS2 scans. Capillary spray voltage was set at 4.5 kV, and capillary temperature was 320°C. Sheath gas was set to 8 arbitrary units and the S-lens Rf level was set to 60.
Statistics. The data were analyzed using 1-way ANOVA with Tukey’s post hoc method (for all pairwise comparisons) or with Dunnett’s post hoc test (for comparison with control). For 2-group comparisons, 2-tailed t test was used. Data are presented as means ± SD where relevant. P < 0.05 was determined to be significant. All analyses were performed using Prism, version 9.4.1 (GraphPad).
Study approval. All experiments involving mice were approved by the IACUC at the University of Pittsburgh (protocols IS00002512 and IS00015519). All experiments involving the use of human tissue were approved by the Institutional Review Board at the University of Pittsburgh. This included the use of whole placentas for trophoblast cell dispersal (exempt protocol STUDY20050066) or for placental biopsies (protocol STUDY19120076, with written informed consent received prior to participation).