Hypoxia‐Driven CA2 Promotes Trophoblast Dysfunction in Preeclampsia

ABSTRACT

Explore the potential function of Carbonic Anhydrase 2 (CA2) in the pathophysiology of preeclampsia. Peripheral blood and placental samples were collected from healthy pregnant women and patients with preeclampsia. Placental CA2 expression was validated in independent clinical cohorts and experimental mouse models. HTR‐8/SVneo cells were cultured and transfected with a CA2 overexpression plasmid to evaluate trophoblast dysfunction. Bulk RNA sequencing was performed to identify downstream genes associated with CA2 overexpression, followed by VEGFA rescue experiments. Hypoxia‐related regulation of CA2 was investigated. Lipid nanoparticle (LNP)‐encapsulated siRNA against CA2 was constructed in an L‐NAME‐induced PE model to evaluate therapeutic effects. CA2 expression was significantly increased in placental tissues and maternal serum from women with PE and was also elevated in experimental PE mouse models. CA2 predominantly localized in trophoblast cells, and correlated positively with serum creatinine and urea and correlated negatively with fetal birth weight. CA2 overexpression markedly suppressed trophoblast proliferation, migration, and invasion, without affecting apoptosis. Transcriptomic analysis suggested that vascular endothelial growth factor A (VEGFA) acted as a downstream mediator of CA2, and VEGFA overexpression partially rescued CA2‐induced trophoblast dysfunction. CA2 expression was upregulated under hypoxic conditions and regulated by hypoxia‐inducible factor‐1α (HIF‐1α). LNP‐mediated CA2 silencing partially improved placental structure and fetal outcomes in vivo. CA2 is upregulated in the preeclamptic placenta and is associated with placental dysfunction and adverse fetal outcomes. Mechanistically, CA2 may act downstream of HIF‐1α and contribute to trophoblast dysfunction partly through modulation of VEGFA signaling, and CA2 silencing partially improved maternal and fetal outcomes, highlighting CA2 as a potential therapeutic target for preeclampsia.

CA2 is upregulated in preeclampsia and impairs trophoblast function by suppressing VEGFA under hypoxia/HIF‐1α signaling. LNP‐mediated CA2 silencing improves placental structure and fetal outcomes, highlighting CA2 as a promising therapeutic target for PE.

Abbreviations

ACOG

America College of Obstetricians and Gynecologists

ALT

alanine aminotransferase

APTT

activated partial thromboplastin time

AST

alanine transaminase

ATCC

American type culture collection

BMI

body mass index

CA2

carbonic anhydrase 2

CCK‐8

cell counting Kit‐8

DBP

diastolic blood pressure

DEGs

differentially expressed genes

DLS

dynamic light scattering

ELISA

enzyme‐linked immunosorbent assay

FBS

fetal bovine serum

GD

gestational day

GDM

gestational diabetes mellitus

GEO

gene expression omnibus

HE

hematoxylin and eosin

HIF‐1α

hypoxia‐inducible factor‐1α

IF

immunofluorescence

IHC

immunohistochemistry

LNP

lipid nanoparticle

MWCO

molecular weight cutoff

PBS

phosphate buffered saline

PDI

polydispersity index

PE

preeclampsia

PT

prothrombin time

RT

room temperature

RT‐qPCR

real‐time quantitative polymerase chain reaction

RUPP

reduced uterine perfusion pressure

SBP

systolic blood pressure

SD

standard deviation

sFLT‐1

soluble fms‐like tyrosine kinase‐1

SPF

specific pathogen‐free

UPCR

urine protein‐to‐creatinine ratio

VEGFA

vascular endothelial growth factor A

1. Introduction

Preeclampsia (PE) is a serious complication of pregnancy, affecting approximately 2%–4% of pregnancies worldwide. It remains a leading cause of maternal and perinatal morbidity and mortality, accounting for nearly 46 000 maternal deaths and 500 000 fetal and neonatal deaths annually. Clinically, PE is defined by new‐onset hypertension after the 20th week of gestation (systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg), often accompanied by proteinuria and multi‐organ dysfunction [1]. Despite its substantial clinical burden, the pathogenesis of PE is incompletely understood, and effective preventive and therapeutic strategies remain suboptimal [2, 3, 4]. Therefore, elucidating the molecular mechanisms underlying PE is of critical importance.

During early pregnancy, trophoblasts proliferate and subsequently differentiate into an invasive phenotype that remodels maternal spiral arteries to establish adequate uteroplacental blood flow. This process normally occurs under physiologically low oxygen tension [5]. However, sustained or excessive hypoxic stress can impair trophoblast differentiation and invasion, leading to shallow placentation and insufficient spiral artery remodeling [6], which are key pathological features of PE [7]. Consistent with this concept, hypoxia‐inducible factor‐1α (HIF‐1α), a key regulator of cellular responses to hypoxia, is markedly elevated in placentas from women with PE, and experimental overexpression of HIF‐1α in the placenta induces PE‐like symptoms in animal models, including hypertension, proteinuria, and fetal growth restriction [8, 9]. Nevertheless, the downstream molecular mediators linking hypoxia to trophoblast dysfunction remain incompletely defined. Identifying hypoxia‐responsive molecules that mediate trophoblast dysfunction may therefore provide new insights into the molecular basis of PE.

Carbonic anhydrases (CAs, EC 4.2.1.1) are zinc‐containing metalloenzymes involved in CO₂ hydration and maintenance of cellular acid–base homeostasis [10]. Given the importance of metabolic adaptation in hypoxic microenvironments, dysregulation of CA family members has been implicated in several hypoxia‐associated diseases [11, 12, 13]. However, the involvement of individual CA isoenzymes in placental dysfunction remains poorly understood. CA2 is one of the most abundant CA isoforms and plays essential roles in pH regulation and tissue homeostasis. Previous studies have reported elevated levels of anti‐CA2 antibodies in women with PE [14], while whether placental CA2 expression is dysregulated and whether CA2 contributes to trophoblast dysfunction remain unclear.

In this study, we investigated the expression pattern of CA2 in preeclampsia and explored its potential role in trophoblast dysfunction. Transcriptomic analysis of placental datasets identified CA2 as a candidate gene of interest. We subsequently validated its expression in clinical placental specimens, serum samples, and experimental models relevant to PE. Furthermore, we examined the effects of CA2 modulation on key trophoblast functions associated with placental development according to VEGFA and evaluated its relationship with hypoxia‐related signaling pathways. Finally, LNP‐mediated CA2 silencing improved placental function and partially adverse pregnancy outcomes in a PE mouse model. Collectively, these findings provide new insights into the role of CA2 in hypoxia‐associated placental dysfunction and suggest CA2 as a potential therapeutic target for PE.

2. Materials and Methods

2.1. Patients and Sample Collection

Three independent clinical cohorts were included in this study. The first cohort (discovery cohort) consisted of placental tissues collected for evaluating CA2 expression in PE. The second cohort (independent serum validation cohort) was used to determine circulating CA2 levels by ELISA. The third cohort (independent gestational age‐matched validation cohort) was established to further validate placental CA2 expression after stratification by gestational age.

2.1.1. Discovery Cohort

Twenty pregnant women with severe PE and twenty normal pregnant women who delivered at the Department of Obstetrics, Women's Hospital, Zhejiang University School of Medicine between January and December 2021 were selected as the research subjects. Severe PE was diagnosed according to the America College of Obstetricians and Gynecologists (ACOG) criteria [1]. Normal pregnant women with uncomplicated singleton pregnancies served as controls. Women with chronic hypertension, pre‐existing diabetes mellitus, autoimmune diseases, renal disease, fetal congenital anomalies, multiple pregnancy, or other obstetric complications (including gestational diabetes mellitus and intrahepatic cholestasis of pregnancy) were excluded.

This study was approved by the Ethics Committee of the Women's Hospital, Zhejiang University School of Medicine (no. IRB‐20220245‐R), and informed consent was obtained from all patients.

Maternal demographic and clinical characteristics of the PE group and normal group were extracted from the electronic medical record system, including age, pregnancy body mass index (BMI), gestational age at delivery, systolic blood pressure (SBP) and diastolic blood pressure (DBP) at admission, and laboratory parameters before delivery including platelet count, alanine aminotransferase (ALT), aspartate aminotransferase (AST), urea, creatinine, activated partial thromboplastin time (APTT), and prothrombin time (PT). Due to normal pregnant did not undergoing quantitative detection of urinary protein, qualitative detection of urinary protein in urine routine was collected. Pregnancy outcomes, including mode of delivery, fetal weight, and APGAR score‐1/5 min were recorded after delivery.

Fresh placental tissue samples were obtained immediately after delivery. Approximately 2 × 2 × 2 cm placental tissue was excised from the central maternal side of the placenta, avoiding areas of infarction, calcification, and hemorrhage. The placental samples were washed with ice‐cold sterile phosphate‐buffered saline (PBS, pH 7.4). Then they were snap‐frozen in liquid nitrogen and stored at −80°C.

The placental samples were also dissected and immediately fixed in 4% paraformaldehyde in 0.1 M PBS at 4°C for 24 h. After fixation, tissues were dehydrated through a graded series of ethanol, cleared in xylene, and embedded in paraffin wax. Serial sections of 4 μm thickness were cut using a rotary microtome (Leica RM2235, Germany) and mounted on positively charged adhesive slides. Sections were dried at 60°C for 1 h prior to immunohistochemical staining.

2.1.2. Independent Serum Validation Cohort

To further validate the clinical relevance of CA2, an independent serum cohort comprising 16 women with severe PE and 16 normotensive pregnant women was additionally recruited at the Women's Hospital, Zhejiang University School of Medicine. The inclusion and exclusion criteria were identical to those described for the discovery cohort.

Peripheral blood samples were obtained from the pregnant women when informed consent was given before operation. The peripheral blood samples were collected in serum blood collection tubes with coagulant additive. Samples were left to stand for 30 min at room temperature to allow complete clot formation, followed by centrifugation at 3000 rpm, 4°C for 15 min to isolate serum. Then they were stored at −80°C.

2.1.3. Independent Gestational Age‐Matched Placental Cohort

To further minimize the potential influence of gestational age, an additional independent placental cohort was established. Placental tissues were collected from women with severe PE and normotensive controls who met the same inclusion and exclusion criteria as the discovery cohort at the Women's Hospital, Zhejiang University School of Medicine. PE and control cases were stratified into three gestational age groups (< 34, 34–37, and > 37 weeks), with six or seven placentas included in each subgroup.

The serum validation cohort and gestational age‐matched placental cohort were recruited independently from the discovery cohort. These cohorts were approved by the Ethics Committee of the Women's Hospital, Zhejiang University School of Medicine (no. IRB‐20260073‐R), and informed consent was obtained from all patients.

2.2. Public Transcriptomic Dataset Analysis

Publicly available placental transcriptomic data were obtained from the Gene Expression Omnibus (GEO) database. The dataset GSE279757, which contains placental samples from preeclamptic and normotensive pregnancies, was used for the discovery analysis. Gene expression data were downloaded in FPKM format and processed using R software (version 4.4.2).

Expression profiles of CA family members were extracted from the normalized expression matrix. Heatmaps were generated using the pheatmap package to visualize the overall expression patterns of CA family genes across samples. The expression level of CA family members between the PE and control groups was further compared and visualized using box plots.

These analyses were conducted to identify CA family genes potentially associated with preeclampsia and to guide subsequent experimental validation.

2.3. Cell Culture

The human trophoblastic cell line (HTR‐8/SVneo) was purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). Cells were cultured in Roswell Park Memorial Institute 1640 medium (RPMI‐1640, cat. no 30‐2001, ATCC, USA), supplemented with 10% fetal bovine serum (FBS, Gibco, Australia) and 1% penicillin and streptomycin (VivaCell, Shanghai, China). Routine incubation was performed at 37°C in a humidified incubator containing 21% O₂ and 5% CO₂. For hypoxic treatment, cells were cultured in a tri‐gas incubator (Thermo Fisher Scientific, USA) maintained at 1% O₂, 5% CO₂ and 94% N₂ for 24 h prior to subsequent experiments.

2.4. RNA Extraction and Real‐Time Quantitative PCR Analysis

Total RNA was extracted from the placenta tissue samples and HTR‐8/Svneo with the RNAiso Plus reagent (Cat. 9108, TakaRa, Japan) according to the manufacturer's protocol. Complementary DNA (cDNA) was reverse‐transcribed using the PrimeScript RT reagent Kit (Cat. RR037A, TakaRa, Japan) following the recommended protocol. Real‐time quantitative PCR (RT‐qPCR) was performed using TB Green Premix Ex Taq II (Cat. RR820A, TakaRa, Japan) on a LightCycler 480 II PCR System (Roche, Switzerland). Relative gene expression levels were quantified using the $2^{-\mathit{\Delta \Delta }{C}_{\mathrm{t}}}$ method, normalized to an internal reference gene. All experiments were performed with at least three biological replicates and three technical replicates. The detailed primer sequences are listed in Table S2.

2.5. Plasmid Transfection

HTR‐8/SVneo cells were used for plasmid‐mediated gene overexpression. The detailed information of plasmid was shown in Table S2. Cells were seeded in 6‐well plates at a density of 1.5 × 10⁵ cells per well in 2 mL of RPMI‐1640 complete medium supplemented with 10% FBS. Cells were cultured overnight at 37°C in a humidified incubator with 5% CO₂, reaching approximately 70%–80% confluence prior to transfection.

To overexpress the target gene, pEX‐3‐CA2, pEX‐3‐control, pEX‐3‐VEGFA, pcDNA3.1‐HIF‐1α, pcDNA3.1‐control plasmid was transfected into the cells using Lipofectamine 3000 reagent (Cat. L3000008, Thermo Fisher Scientific), following the manufacturer's protocol. Briefly, 2 μg plasmid DNA was diluted in 125 μL serum‐free medium OPTI‐MEM (Cat. 31985–070, Gibco, Australia), and 2 μL Lipofectamine 3000 and 4 μL P3000 were also diluted in 125 μL OPTI‐MEM. The two components were mixed and incubated at room temperature for 15 min to form 250 μL transfection complexes. These complexes were added dropwise to the cells in 1750 μL serum‐free RPMI 1640. After 6 h of incubation with transfection complexes, the medium was replaced with fresh complete RPMI‐1640 medium containing 10% FBS. Cells were harvested 48 h post‐transfection for downstream analyses. Overexpression efficiency was confirmed by RT‐qPCR and Western blot. The detailed sequences of the plasmids are listed in Table S2.

2.6. Western Blot

Total protein was extracted from HTR‐8/SVneo cells and placental tissues using the RIPA lysis buffer with a protease and phosphatase inhibitor cocktail (Cat. 5872, Cell Signaling Technology [CST], USA). The concentration of total protein was quantified via the BCA method (Cat. P0010, Beyotime Biotechnology, China), and equal amounts of protein samples were separated by SDS‐PAGE and transferred onto PVDF membranes. After blocking with 5% skim milk for 2 h at room temperature, the membranes were incubated overnight at 4°C with the following primary antibodies: anti‐CA2 (1:2000, Cat. GTX105562, GeneTex, USA), anti‐VEGFA (1:1000, Cat. 19003‐1‐AP, Proteintech, Wuhan, China), and anti‐β‐actin (1:10000, Cat. 66009‐1‐lg, Proteintech, Wuhan, China). Corresponding HRP‐conjugated secondary antibodies were incubated with membranes for 1 h at room temperature. Subsequently, PVDF membranes were incubated with ECL chemiluminescent substrate (Fude Bio, China) working solution for 1 min. Protein bands were visualized using an ImageQuant LAS 4000 mini chemiluminescence imaging system (GE Healthcare, Sweden). The grayscale value of each target band was quantified by ImageJ software, and the relative protein expression level was normalized to the internal reference β‐actin.

2.7. Immunohistochemistry (IHC)

Tissue sections were deparaffinized and rehydrated using standard protocols. Antigen retrieval was performed in 0.01 mol/L sodium citrate buffer (pH 6.0) at 60°C overnight. The sections were then cooled to room temperature and rinsed under running water three times for 5 min each. To eliminate endogenous peroxidase activity, sections were incubated in 3% (v/v) hydrogen peroxide solution for 20 min, followed by three washes with TBST buffer. Then nonspecific binding sites were blocked with 10% normal sheep serum at room temperature for 1 h. Subsequently, sections were incubated overnight at 4°C with a primary antibody CA2 (GeneTex GTX105562) diluted 1:200 in 5% BSA. The slides were washed three times with TBST (7–10 min each). Following another three TBST washes, a rabbit HRP‐conjugated secondary antibody (RGAR011, Proteintech, Wuhan, China) was added and incubated for 1 h at room temperature. The slides were washed and developed with DAB kit (Cat. PR30010, Proteintech, Wuhan, China) in a humidified chamber, protected from light. After rinsing with water, the sections were counterstained with hematoxylin, dehydrated through a graded ethanol series, cleared in xylene, and mounted with neutral resin. Once dried, the slides were imaged under a microscope.

2.8. Cell Viability Assay

Cell viability was assessed using the Cell Counting Kit‐8 (CCK‐8, Yeasen, Shanghai, China) according to the manufacturer's instructions. HTR‐8/SVneo cells were seeded in 96‐well plates at a density of 2 × 10³ cells per well in 100 μL of RPMI‐1640 medium supplemented with 10% FBS. Cells were allowed to attach overnight in a humidified incubator at 37°C with 5% CO₂. After 0/24/48/72/96 h pEX‐3‐CA2, pEX‐3‐control, pEX‐3‐VEGFA plasmid transfection, 10 μL of the CCK‐8 reagent was added to each well. The plates were incubated at 37°C for 2 h. The absorbance at 450 nm was measured using a microplate reader (Shimadzu, Japan). Each experimental condition was performed in triplicate wells, and a blank control (medium without cells) was included for background subtraction.

2.9. Transwell Migration and Invasion Assay

HTR‐8/SVneo cells were transfected with pEX‐3‐CA2, pEX‐3‐control, pEX‐3‐VEGFA plasmids, following the protocol above. After 48 h of transfection, cells were harvested, counted, and prepared for migration assay and invasion assay.

Cell migration was evaluated using 24‐well Transwell chambers (Cat. 3422, Corning, USA). Transfected cells were resuspended with RPMI‐1640 complete medium. And 2 × 10⁴ in 200 μL were seeded into the upper chamber. The lower chamber was filled with 500 μL of complete medium as a chemoattractant. After 24 h of incubation at 37°C in a humidified 5% CO₂ incubator, non‐migrated cells on the upper surface of the membrane were gently removed with a cotton swab. The migrated cells on the bottom surface were fixed with 4% paraformaldehyde for 15 min, stained with 0.1% crystal violet for 20 min, and washed with PBS. Images were captured using a microscope, and the number of migrated cells was counted in five randomly selected fields per membrane.

Cell invasion was assessed using Matrigel‐coated Transwell chambers. Matrigel (50 μL per insert) was diluted by serum‐free RPMI‐1640 medium with the ratio of 1:3. Then it was applied to the upper surface of the membrane and allowed to polymerize at 37°C for 30 min. Transfected cells (2 × 10⁴ in 200 μL of serum‐free RPMI‐1640 medium) were seeded into the upper chamber, while 500 μL of complete medium was added to the lower chamber. After 24 h of incubation, non‐invaded cells were removed from the upper surface, and invaded cells were fixed, stained, and imaged as described for the migration assay. The number of invaded cells was counted in five randomly selected fields.

2.10. Apoptosis Assay

HTR‐8/SVneo cells were transfected with pEX‐3‐CA2, pEX‐3‐control, and pEX‐3‐VEGFA plasmids, following the protocol above. After 48 h of transfection, cells were harvested, counted, and prepared for apoptosis assay. Cell apoptosis was assessed using the Annexin V‐APC/7‐AAD apoptosis kit (Cat. AP105, Lianke, Hangzhou, China) according to the manufacturer's instructions. Briefly, transfected cells were collected, washed twice with cold PBS, and resuspended in 1× Binding Buffer at a concentration of 1 × 10⁶ cells/mL. Then, 5 μL of Annexin V‐APC and 10 μL of 7‐AAD were added, and the mixture was gently vortexed and incubated for 5 min at room temperature in the dark. After incubation, samples were analyzed immediately using a flow cytometer (CytoFLEX S, Beckman, USA).

2.11. RNA Sequencing

Total RNA was extracted from HTR‐8/SVneo cells transfected with the empty vector or CA2 overexpression plasmid (n = 3 per group). RNA sequencing was performed by Lianchuan Biotechnology (Hangzhou, China) using an Illumina sequencing platform according to standard protocols. After quality control, clean reads were aligned to the human reference genome, and gene expression levels were quantified based on read counts. Differential expression analysis was performed using DESeq2 based on a negative binomial generalized linear model. Raw p values were adjusted for multiple testing using the Benjamini–Hochberg (BH) method to control the false discovery rate (FDR). Genes with |log₂ fold change| ≥ 1 and adjusted p value (FDR) < 0.05 were considered differentially expressed. Heatmaps and volcano plots were generated to visualize the expression patterns of differentially expressed genes.

2.12. PE Mouse Model

Specific pathogen‐free (SPF) adult C57BL/6 aged 6–8 weeks, were housed at room temperature of 18°C–28°C and relative humidity of 40%–70% without feed or water control. During the estrous period, female and male mice were kept in the same cage at a ratio of 2:1. Vaginal secretions were examined microscopically every morning, and the day a vaginal copulatory plug was detected was defined as gestational day 0.5 (GD0.5). Pregnant mice that died accidentally during the experiment were supplemented to maintain consistent sample size. The pregnant mice were randomly divided into two groups (n = 6 per group):Group 1 (normal pregnant control); Group 2 (pregnant mice with PE). The Zhejiang University's Animal Ethics Committee's guidelines were followed in all animal experiments.

2.12.1. L‐NAME Induced PE Model

The Nω‐nitro‐L‐arginine methyl ester (L‐NAME, MedChemExpress, USA) was used to construct the chemical preeclampsia model. From GD7.5 to GD17.5, mice in Group 2 received daily intraperitoneal injection of L‐NAME solution at a dosage of 125 mg/kg. Control mice (Group 1) were administered an equal volume of sterile PBS via intraperitoneal injection on the same schedule. Noninvasive tail‐cuff plethysmography (Softron Biotechnology, Beijing, China) was applied to measure systolic blood pressure between 9:00 and 11:00 a.m. at GD6.5, GD9.5, GD13.5 and GD17.5.

2.12.2. Reduced Uterine Perfusion Pressure (RUPP) Surgical PE Model

The RUPP surgery was performed on pregnant mice at GD14.5. Briefly, mice were anesthetized via intraperitoneal injection of tribromoethanol (Tigergene, USA; 0.2 mL/10 g body weight). Animals were fixed in a supine position, and abdominal hair from the xiphoid process to the last pair of nipples was removed along the abdominal midline. The surgical field was disinfected twice with povidone‐iodine solution.

A 2–3 cm longitudinal midline laparotomy was made to expose the abdominal cavity. Intestinal tissues were gently retracted with moistened cotton swabs to visualize the abdominal aorta and inferior vena cava. Approximately 0.5–1.0 cm proximal to the bifurcation of the common iliac arteries, a hooked glass probe was used to bluntly dissect a 1 cm segment of the abdominal aorta. A 6–0 surgical suture was looped around the isolated aorta and vena cava (≥ 0.5 cm above the iliac bifurcation) and tied to constrict aortic blood flow toward the uterus. Bilateral ovarian uterine arterial branches were further isolated and ligated with separate 6–0 sutures to sufficiently reduce uterine perfusion.

After ligation, visceral tissues were carefully repositioned into the abdominal cavity, and a small volume of sterile normal saline was instilled intraperitoneally to prevent tissue dehydration. The abdominal incision was sutured and disinfected daily postoperatively. Blood pressure was measured at GD13.5, GD15.5, GD16.5, GD17.5 and GD18.5 from 9:00 to 11:00 a.m. using the tail‐cuff system.

At GD17.5, all pregnant mice were housed individually in metabolic cages for 24 h urine collection with free access to drinking water (food was withheld during collection). Collected urine specimens were immediately stored at −80°C for subsequent creatinine detection using a Creatinine (Cr) Assay Kit (Jiancheng Bioengineering Institute, Nanjing, China).

On GD18.5, all gestational mice were anesthetized with tribromoethanol (0.2 mL/10 g). Peripheral venous blood was harvested to isolate serum. Fetal and placental tissues were separated and weighed immediately.

Dissected placentas and kidneys were processed in two ways: (1) snap‐frozen in liquid nitrogen and stored long‐term at −80°C; (2) fixed in 4% (v/v) paraformaldehyde and embedded in paraffin following standard histologic protocols. Paraffin blocks were sectioned into 4 μm‐thick slices with a rotary microtome, mounted on positively charged glass slides, and baked at 60°C for 1 h prior to Hematoxylin–Eosin (H&E) staining.

2.13. Hematoxylin and Eosin (HE) Staining

Paraffin sections were baked at 60°C for 1 h, followed by deparaffinization and rehydration, sections were stained with hematoxylin for 5 min, rinsed in running tap water, and differentiated in 1% acid alcohol. After bluing in ammonia water, sections were counterstained with eosin for 1 min. Subsequently, the sections were dehydrated through a graded ethanol series, cleared in xylene, and mounted with neutral balsam for microscopic examination.

2.14. Enzyme‐Linked Immunosorbent Assay (ELISA)

Serum CA2 from human samples, urinary albumin, and serum soluble FMS‐like tyrosine kinase‐1 (sFlt‐1) from mouse specimens were quantified using corresponding commercial ELISA kits (Cat. ml024091, Cat. ml037573, Cat. ml106745, MLBio, Shanghai, China), strictly following the manufacturer's protocols. The absorbance value of each well was detected at 450 nm via a microplate reader (Shimadzu, Japan). The concentrations of all target proteins were calculated based on the standard curve provided in the kit instructions.

2.15. Luciferase Reporter Assay

Luciferase reporter assays were performed to evaluate the transcriptional regulation of CA2 by HIF‐1α. The CA2 promoter fragments containing either the predicted HIF‐1α binding sites were cloned into luciferase reporter vectors (Genepharma Biotechnology, Shanghai, China). HTR‐8/SVneo cells were co‐transfected with 0.5 μg of CA2 promoter luciferase plasmids, HIF‐1α overexpression plasmids, or corresponding control plasmids, together with a Renilla luciferase plasmid as an internal control. The total amount of transfected DNA was kept constant across all groups. After 48 h of transfection, luciferase activities were measured using a dual‐luciferase reporter assay kit (Cat. RG027, Beyotime Biotechnology, Shanghai, China) according to the manufacturer's instructions. Firefly luciferase activity was normalized to Renilla luciferase activity. All experiments were independently repeated at least three times.

2.16. Hypoxia Treatment and Luciferase Reporter Assay

To further investigate the effect of hypoxia on CA2 transcriptional activity, HTR‐8/SVneo cells were co‐transfected with CA2 promoter luciferase reporter plasmids, HIF‐1α negative control plasmids, and Renilla luciferase plasmids. After transfection, cells were cultured in a tri‐gas incubator (1% O₂, 5% CO₂, 94% N₂) for 12, 24, or 36 h, followed by luciferase activity detection using the dual‐luciferase reporter assay system. Firefly luciferase activity was normalized to Renilla luciferase activity. All experiments were independently repeated at least three times.

2.17. Screening siRNA for Knockdown of CA2 in HTR‐8/SVneo Cells

HTR‐8/SVneo cells were seeded into 6‐well plate. Upon reaching 70% confluence, cells were transfected with 75 pmol siRNA against CA2 (Genepharma Biotechnology, Shanghai, China), which is a pool of 4 target‐specific siRNAs designed to knock down CA2 gene expression. The detailed sequences of the 4 siRNA are listed in Table S2. The negative control (NC) siRNA contains a scrambled sequence that does not specifically degrade any known cellular mRNA. After 48 h of transfection using Lipofectamine 3000 (Cat. L3000008, Thermo Fisher Scientific). following the manufacturer's protocol, HTR‐8/SVneo cells in each group were harvested for RT‐qPCR analyses. Among all candidate sequences, siCA2‐1208 exhibited the optimal silencing efficiency and was selected for subsequent functional assays. This siRNA was encapsulated into ionizable lipid nanoparticles to generate LNP‐siCA2 for in vitro and in vivo intervention experiments.

2.18. Preparation of CA2 siRNA‐Loaded Lipid Nanoparticles (LNPs)

CA2 siRNA‐loaded LNPs and NC‐loaded LNPs were prepared by microfluidic mixing. The lipid formulation consisted of Dlin‐MC3‐DMA (Cat. 1224606–06‐7, Sigma‐Aldrich), cholesterol (Cat. C8667‐5G, Sigma‐Aldrich), DSPC (Cat. S01005, AVT Pharmaceutical), and DMG‐PEG2000 (Cat. O02005, AVT Pharmaceutical). Lipids were dissolved in ethanol to form the organic phase. CA2 siRNA and NC were dissolved in 100 mM citrate buffer (pH 4.0) as the aqueous phase. The lipid ethanol phase (total lipid concentration 2 mg/mL) and the siRNA‐containing aqueous phase were mixed using a microfluidic mixer at a flow rate ratio of 3:1 (aqueous:organic) with a total flow rate of 2 mL/min, yielding an N/P ratio of 8 relative to the ionizable lipid. The resulting LNP suspension was dialyzed overnight against PBS (pH 7.4) using a 25 kDa Molecular Weight Cutoff (MWCO) dialysis membrane to remove ethanol and exchange the buffer. Particle size and polydispersity index (PDI) were measured by dynamic light scattering (DLS). The encapsulation efficiency of CA2 siRNA and NC was determined using a RiboGreen RNA assay.

2.19. LNP‐Mediated siRNA Delivery

To evaluate the efficiency of lipid nanoparticle (LNP)‐mediated siRNA delivery, HTR‐8/SVneo cells were treated with LNP‐encapsulated siCA2 (LNP‐siCA2), LNP‐empty, LNP‐encapsulated NC (LNP‐NC), siCA2, or NC. Cells were seeded in 6‐well plates at a density of 1.5 × 10⁵ cells per well in 2 mL of RPMI‐1640 medium supplemented with 10% FBS. Cells were cultured overnight at 37°C in a humidified incubator with 5% CO₂, reaching approximately 70%–80% confluence prior to transfection. After 48 h of treatment, total protein was extracted to evaluate the knockdown efficiency of CA2. The expression levels of CA2 were determined by Western blot analysis.

2.20. Therapeutic Effects of LNP‐siCA2 in Mice With PE

PE mouse models were constructed via L‐NAME administration following the protocol described in Section 2.12. All gestational mice received daily L‐NAME injection from GD7.5 to GD17.5 and were randomly assigned to two intervention groups (n = 6 per group): Group 1 (LNP‐NC negative control) and Group 2 (LNP‐siCA2 treatment). Mice received intravenous tail‐vein injections of LNP‐NC or LNP‐siCA2 on GD 10.5, 13.5, and 16.5, with a total siRNA dose of 200 ng per mouse, equally divided among the three administrations.

Systolic blood pressure was measured by noninvasive tail‐cuff plethysmography (Softron Biotechnology, Beijing, China) between 9:00 and 11:00 a.m. at GD6.5, GD9.5, GD13.5 and GD17.5. At GD17.5, each mouse was separately placed in a metabolic cage for 24 h urine collection; mice had free access to drinking water but were fasted during urine collection.

On GD18.5, mouse anesthesia, peripheral blood collection and isolation of placentas and kidneys were identical to Section 2.12. Serum biomarker detection, urinary albumin quantification and urinary creatinine were carried out using the same kits and protocols as Section 2.12 described. Placental tissues were used for subsequent RNA, protein, and histological analyses.

2.21. Paraffin‐Embedded Tissue Immunofluorescence (IF) Staining

Slices were baked at 60°C for 1 h, followed by sequential immersion in xylene I (10 min), xylene II (10 min), anhydrous ethanol (5 min), 95% ethanol (5 min), 80% ethanol (5 min), and double‐distilled water (3 min). Then sections were submerged in citrate antigen retrieval buffer (pH 6.0) and heated in a pressure cooker for 15 min. After natural cooling to room temperature, slides were rinsed 3 times with PBS for 3 min each. Tissue slices were incubated with 0.5% Triton X‐100 in PBS for 10 min to permeabilize cell membranes. After PBS washing, nonspecific binding sites were blocked with 5% bovine serum albumin (BSA) at room temperature for 1 h. Primary antibody incubation: Excess blocking solution was discarded, and sections were incubated with primary antibody targeting CA2 (Cat. 16961‐AP, Proteintech, Wuhan, China) overnight at 4°C in a humidified dark box. Slides were washed with PBS three times (5 min per wash) the next day. Secondary antibody incubation: Fluorescent‐labeled secondary antibody was diluted with 1% BSA‐PBS and applied to the sections. Incubation was performed at room temperature in the dark for 1 h, followed by three PBS washes (5 min each). Nuclear counterstaining: DAPI working solution was added dropwise to stain cell nuclei for 5 min at room temperature, protected from light. Excess DAPI was washed off with PBS. Mounting and imaging: Anti‐fade fluorescence mounting medium was used to seal the slices. Fluorescence signals were observed and captured under a laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany). Image fluorescence intensity was quantified using ImageJ software for statistical analysis.

2.22. Statistical Analysis

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). Data distribution and variance homogeneity were assessed before selecting the appropriate statistical test. Continuous variables were first assessed for normality using the Shapiro–Wilk test. Data with a normal distribution are presented as the mean ± standard deviation (SD). Comparisons between two groups were performed using an unpaired Student's t‐test or Welch's t‐test when variances were unequal. Comparisons among three or more groups were performed using one‐way analysis of variance (ANOVA) followed by Tukey's multiple‐comparison test. Categorical variables are presented as frequencies (percentages) and were analyzed using the chi‐square test or Fisher's exact test, as appropriate. Correlation analyses were performed using Pearson's correlation coefficient. All statistical tests were two‐sided, and p < 0.05 was considered statistically significant.

3. Results

3.1. CA2 Is Highly Expressed in PE Placental Syncytiotrophoblast Cells

Transcriptome analysis of the GEO dataset (GSE279757) including placental tissues from 6 women with PE and 6 normotensive pregnancies showed differential expression of several CA family members (Figure 1A). The box plot indicated the expression level of CA family members; CA2 exhibited significantly increased expression in PE placentas (Figure 1B). Among these candidates, CA2 was selected for further investigation because of its well‐established role in regulating cellular pH homeostasis and metabolic processes [15]. In addition, previous studies have suggested potential links between CA2 and preeclampsia [14], indicating that CA2 may participate in the pathophysiological processes of preeclampsia. Therefore, CA2 was chosen as the primary target for subsequent experimental validation. To validate these findings, placental tissues from the independent discovery cohort (20 women with severe PE and 20 normotensive pregnant women) were analyzed. The clinical characteristics of the participants are summarized in Table 1. Consistent with the transcriptomic findings, placental CA2 mRNA expression was significantly increased in the placental tissues of women with PE compared to those with normal pregnancies (p < 0.05, Figure 1C). Similarly, the protein level expression of CA2 was also markedly elevated in the placental tissues of women with PE (p < 0.05, Figure 1D,E). To further determine whether CA2 alterations could also be detected in the circulation, serum CA2 concentrations were measured in an independent validation cohort comprising 16 women with PE and 16 normotensive controls. Serum CA2 levels were significantly higher in women with PE than in controls (Figure 1F). The clinical characteristics of the participants are summarized in Table S1. Considering the significant difference in gestational age between the discovery cohort, we further validated placental CA2 expression using an independent gestational age‐stratified cohort. Placental samples were categorized into three gestational age groups (< 34, 34–37, and > 37 weeks). Overall, the increased expression of CA2 in PE placentas remained consistent after gestational age stratification. Significant upregulation of CA2 protein was observed in the < 34‐ and 34–37‐week subgroups (Figure S1A–D). In contrast, no significant difference was observed between PE and control placentas from term pregnancies (> 37 weeks) (Figure S1E,F). The clinical characteristics of the participants are summarized in Table S1. Finally, immunohistochemical staining was performed to examine the localization of CA2 in placental tissues. As illustrated in Figure 1G, CA2 was predominantly expressed in placental trophoblast cells, while it was scarcely detected in vascular endothelial cells. The above findings suggested that CA2 is highly expressed in PE placental syncytiotrophoblast cells.

FIGURE 1.

Expression and localization of CA2 in the human placentas and Serum. (A) Heatmap of carbonic anhydrase family expression in placental transcriptomes from normal and preeclamptic pregnancies. (B) Box plot analysis of carbonic anhydrase family expression in placental transcriptomes from normal and preeclamptic pregnancies. (C) MRNA expression level of CA2 in PE and normal placentas detected by RT‐qPCR in the discovery cohort. (D) Representative western blot bands for CA2. (E) Protein expression level of CA2 in PE and normal placentas detected by Western Blot in the discovery cohort. (F) Serum expression level of CA2 in PE and normal women detected by ELISA in the serum validation cohort. (G) Immunohistochemistry method to detect the expression localization of CA2 in placentas from normal and preeclamptic pregnancies. Scale bars: 3 mm (overview), 200 μm (magnified field). All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

TABLE 1.

Clinical characteristics of the discovery cohort.

	Control (N = 20)	PE (N = 20)	p
Age (year)	33.60 ± 2.96	34.00 ± 4.39	0.74	> 0.05
BMI (kg/m2)	25.63 ± 2.60	30.61 ± 3.13	< 0.01	***
Gestational week of labor (weeks)	38.40 ± 0.80	30.1 ± 2.00	< 0.01	***
SBP (mmHg)	116.75 ± 10.97	165.50 ± 16.28	< 0.01	***
DBP (mmHg)	70.75 ± 7.71	103.60 ± 10.46	< 0.01	***
Proteinuria	− (20)	+~++ (15)
		+++~++++ (5)
Neonatal birth weight (grams)	3405.50 ± 405.90	1283.5 ± 447.06	< 0.01	***
1‐min Apgar score	9.85 ± 0.48	8.05 ± 2.18	< 0.01	***
5‐min Apgar score	10.00 ± 0.00	9.2 ± 2.18	0.12	> 0.05
Platelet (109/L)	204.50 (60)	189.50 (62)	0.13	> 0.05
ALT (U/L)	15.50 (9)	18.50 (11)	0.02	*
AST (U/L)	21.50 (10)	22.50 (6)	0.15	> 0.05
Creatinine (mol/L)	43.50 (10)	58.00 (16)	< 0.01	***
Urea (mmol/L)	3.07 (0.82)	4.97 (2.13)	< 0.01	***
PT (s)	12.20 (1)	11.45 (0.80)	< 0.01	***
APTT(s)	31.06 ± 2.21	33.14 ± 3.11	0.02	*

Abbreviations: ALT, alanine aminotransferase; APTT, activated partial thromboplastin time; AST, aspartate aminotransferase; BMI, body mass index; DBP, diastolic blood pressure; PT, prothrombin time; SBP, systolic blood pressure.

^* p < 0.05.

^** p < 0.01.

^*** p < 0.001.

3.2. Placental CA2 Expression Levels Are Associated With Clinical Severity of Patients With PE

Pearson correlation analysis was used to analyze association between placental CA2 expression levels and maternal and fetal clinical indicators. Among maternal clinical indicators, the results demonstrated that the protein expression level of CA2 in placental tissues was positively correlated with maternal SBP (R ² = 0.3909, p < 0.05, Figure 2A), DBP (R ² = 0.3999, p < 0.05, Figure 2B), proteinuria (+) (R ² = 0.4952, p < 0.05, Figure 2F), creatinine (R ² = 0.4781, p < 0.05, Figure 2G) and urea (R ² = 0.2947, p < 0.05, Figure 2H). Among fetal indicators, it was negatively correlated with fetal weight (R ² = 0.6590, p < 0.05, Figure 2C) and APGAR score‐1 min (R ² = 0.1539, p < 0.05, Figure 2D). However, there were no significant correlations with the level of ALT, AST, platelet count and coagulation indices including APTT and PT.(Figure 2I–M, p > 0.05). Collectively, these correlation data demonstrate that elevated placental CA2 expression is tightly associated with aggravated maternal organ injury and impaired fetal outcomes, indicating a strong clinical relevance of CA2 to preeclampsia severity.

FIGURE 2.

Correlation analysis between CA2 expression and clinical indicators. (A) Correlation of CA2 levels with SBP. (B) Correlation of CA2 levels with DBP. (C) Correlation of CA2 levels with Fetal Weight. (D) Correlation of CA2 levels with APGAR‐1 min. (E) Correlation of CA2 levels with APGAR‐5 min. (F) Correlation of CA2 levels with Proteinuria. (G) Correlation of CA2 levels with maternal blood creatinine levels before delivery. (H) Correlation of CA2 levels with blood urea levels. (I) Correlation of CA2 levels with ALT. (J) Correlation of CA2 levels with AST. (K) Correlation of CA2 levels with PT levels. (L) Correlation of CA2 levels with APTT. (M) Correlation of CA2 levels with platelet levels. Correlation analyses were performed using Pearson's correlation coefficient. ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

3.3. Increased Expression of Placental CA2 in PE Models

To further verify the role of CA2 in PE, two classic PE mouse models (L‐NAME‐induced and RUPP) were constructed separately. First, the L‐NAME‐induced PE model was generated via continuous L‐NAME administration (Figure 3A). Consistent with PE pathological phenotypes, systolic blood pressure gradually increased in L‐NAME‐treated pregnant mice (Figure 3B). L‐NAME exposure triggered severe fetal growth restriction accompanied by reduced placental efficiency (Figure 3C–E). Moreover, L‐NAME dams exhibited an elevated urine protein‐to‐creatinine ratio (UPCR) and obvious renal histological damage (Figure 3F,G). Serum sFlt‐1 concentration was significantly upregulated in the L‐NAME group (Figure 3H). Histological quantification revealed a markedly decreased labyrinthine/junctional zone area ratio in placentas from L‐NAME‐treated dams (Figure 3I,J). Collectively, these phenotypic data validated the successful establishment of the L‐NAME PE model. Total RNA and total protein were extracted from placental tissues of control and L‐NAME groups. qPCR and western blot assays demonstrated that both mRNA and protein expression of placental CA2 were significantly increased in the L‐NAME PE group relative to normal control mice (Figure 3K–M).

FIGURE 3.

Preeclampsia modeling and placental CA2 expression in L‐NAME mouse model. (A) Experimental workflow of the L‐NAME‐induced preeclampsia model. (B) Changes in systolic blood pressure in L‐NAME and CONTROL groups. (C, E) Body weights of fetal mice in L‐NAME and CONTROL groups. (D) Placental efficiency of fetal mice in L‐NAME and CONTROL groups. (H) SFLT‐1 levels of mice in L‐NAME and CONTROL groups. (F) The ratio of urinary albumin and creatine of mice in L‐NAME and CONTROL groups. (G, I) Representative renal and placental stained with hematoxylin & eosin (H&E) in L‐NAME and CONTROL groups. Scale bars: 250 μm. (J) Ratio of palcental labyrinth and junction zone of mice in L‐NAME and CONTROL groups. (K) MRNA expression level of placental CA2 in L‐NAME and CONTROL groups detected by RT‐qPCR. (L) Representative western blot bands for CA2. (M) Protein expression level of placental CA2 in L‐NAME and CONTROL groups detected by Western Blot. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

In parallel, we established the well‐established RUPP PE model by partial ligation of the abdominal aorta and bilateral ovarian arteries to mimic placental hypoperfusion (Figure S2A,B). Consistent with the L‐NAME model, RUPP‐operated mice displayed typical PE manifestations: elevated blood pressure (Figure S2C), decreased fetal weight and placental efficiency (Figure S2D–F), increased urinary protein excretion and renal injury (Figure S2G,H), as well as higher circulating sFlt‐1 levels (Figure S2K). Placental morphological analysis also showed a reduced labyrinthine/junctional zone ratio in RUPP placentas (Figure S2I,J). Notably, placental CA2 mRNA and protein abundance were also significantly higher in the RUPP group compared with control animals (Figure S2L–N).

Taken together, these results from two independent PE mouse models strongly indicate that aberrantly elevated placental CA2 expression may exert a vital pathogenic role in the progression of preeclampsia.

3.4. Overexpression of CA2 Suppresses Proliferation, Migration and Invasion in HTR‐8/SVneo Trophoblast Cells

Given the elevated expression of CA2 observed in placental trophoblasts from PE placentas, we constructed a CA2 overexpression plasmid to explore its biological function in HTR‐8/SVneo cells.

HTR‐8/SVneo cells were transiently transfected with pEX‐3‐CA2 overexpression plasmid or empty pEX‐3 vector control. qPCR and western blot results confirmed that both CA2 mRNA and protein expression were markedly upregulated following transfection (p < 0.05, Figure 4A–C). The CCK‐8 assay was conducted to assess the relative cell viability between the two groups at various time points over 96 h post‐transfection. The results demonstrated that overexpression of CA2 significantly suppressed trophoblast proliferation compared to the control group at 72 h and 96 h (both p < 0.05, Figure 4D). Transwell assays were performed to examine the migration and invasion abilities of these cells. The results showed that CA2 overexpression significantly impaired cell migration and decreased invasion ability after 48 h compared to the control group (p < 0.05, Figure 4E,F). However, no significant difference in apoptosis levels was observed between the two groups after CA2 overexpression (p > 0.05, Figure 4G,H). Collectively, these functional assays demonstrate that high CA2 expression negatively regulates the proliferative, migratory and invasive behaviors of trophoblasts, without obvious influence on trophoblast apoptosis.

FIGURE 4.

Effects of CA2 on trophoblast cell function. (A) RT‐qPCR analysis to detect the expression of mRNA level after overexpression of CA2 in HTR‐8/SVneo cells. (B) Representative western blot bands for CA2. (C) Western Blot analysis to detect the expression of protein level after overexpression of CA2 in HTR‐8/SVneo cells. (D) Effect of CA2 overexpression on the proliferative capacity of HTR‐8/SVneo cells. (E) Impact of CA2 overexpression on the migration and invasion abilities of HTR‐8/SVneo cells. (F) Statistical analysis of the Transwell migration and invasion assays. (G) Effect of CA2 overexpression on the apoptotic properties of HTR‐8/SVneo cells. (H) Statistical analysis of the apoptosis assay. pEX‐3‐control: Empty vector control group; pEX‐3‐CA2: CA2 overexpression group. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

3.5. Overexpression of VEGFA Rescues Defective Proliferation, Migration and Invasion Triggered by CA2 Overexpression in HTR‐8/SVneo Trophoblasts

To investigate the potential mechanism underlying CA2‐mediated impairment of HTR‐8/SVneo cells function, RNA sequencing (RNA‐seq) was conducted in HTR‐8/SVneo cells with CA2‐overexpressing cells. Genes with |log₂ fold change| ≥ 1 and adjusted p value (FDR) < 0.05 were considered differentially expressed genes (DEGs). Hierarchical clustering heatmap clearly separated gene expression profiles between control and CA2‐overexpressing groups (Figure 5A), with 139 significantly upregulated DEGs and 71 downregulated DEGs identified. Among the top suppressed genes were BCL3, VEGFA and SOCS3 (Figure 5B). Given the well‐documented critical role of VEGFA in modulating trophoblast phenotypes, we prioritized VEGFA for subsequent mechanistic validation. Consistant with the RNA‐seq data, VEGFA expression levels was significantly reduced in PE placentas (p < 0.05, Figure S3A), reaching nearly half of the expression levels detected in normal control placentas. Rescue functional experiments were further performed to clarify whether CA2 impairs trophoblast behaviors via repressing VEGFA. Three experimental groups were established: the empty vector control group (pEX‐3‐control), the single CA2 overexpression group (pEX‐3‐CA2) and the CA2 and VEGFA co‐overexpression group (pEX‐3‐CA2 + pEX‐3‐VEGFA). We observed that VEGFA expression was decreased in the pEX‐3‐CA2 group and restored in the CA2 and VEGFA co‐overexpression group (Figure 5C). The CCK‐8 proliferation assays demonstrated that overexpression of VEGFA was able to rescue the inhibitory effect of CA2 on cell proliferation (p < 0.05, Figure 5D). Similarly, Transwell assays showed that VEGFA overexpression reversed the inhibition effects of cell migration and invasion by CA2 overexpression (p < 0.05, Figure 5E,F). These rescue experimental data indicate that CA2 negatively regulates the proliferative, migratory and invasive functions of placental trophoblasts through transcriptional downregulation of VEGFA.

FIGURE 5.

VEGFA overexpression rescues the impaired proliferation, migration and invasion of HTR‐8/SVneo cells induced by CA2 overexpression. (A) Heatmap showing hierarchical clustering of differentially expressed genes between the control and CA2 overexpression groups. (B) Volcano plot of differentially expressed genes. (C) Western Blot analysis of CA2 and VEGFA to detect the protein expression level on the of HTR‐8/SVneo cells in each expremental group. (D) Proliferative capacity analysis of HTR‐8/SVneo cells in the control group、CA2 overexpression group and CA2 and VEGFA co‐overexpression group. (E) Migration and invasion abilities of HTR‐228/SVneo cells in the control group, CA2 overexpression group and CA2 and VEGFA co‐overexpression group. (F) Statistical analysis of the Transwell migration and invasion abilities in the control group, CA2 overexpression group and CA2 and VEGFA co‐overexpression group. pEX‐3‐control 1/2/3: Empty vector control group; pEX‐3‐CA2 1/2/3: CA2 overexpression group; pEX‐3‐control: Empty vector control group; pEX‐3‐ CA2: CA2 overexpression group; pEX‐3‐ CA2+ pEX‐3‐ VEGFA: CA2 and VEGFA co‐overexpression group. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

3.6. HIF‐1α Activates the Transcription of CA2

Studies have shown that CA9 contains a HIF‐1‐dependent hypoxia response element in its minimal promoter and is tightly regulated by the HIF/pVHL pathway in tumors [16]. Therefore, we speculate that HIF‐1α may mediate the up‐regulation of CA2 expression. To mimic placental hypoperfusion in vitro, HTR‐8/SVneo trophoblasts were cultured in a tri‐gas incubator under hypoxic conditions (1% O₂, 5% CO₂, 94% N₂). Western blot and qPCR analysis showed that 1% oxygen exposure significantly increased the expressions of HIF‐1α and CA2 in HTR‐8/SVneo trophoblast cells (Figure 6A–E). Moreover, it was found that HIF‐1α over‐expression also could increase the expression of CA2 (Figure 6F–J). These data support the conclusion that HIF‐1α may mediate the up‐regulation of CA2 expression. To investigate whether HIF‐1α is involved in the transcriptional regulation of CA2, a luciferase reporter assay was performed using a CA2 promoter–driven reporter construct. As shown in Figure 6K, overexpression of HIF‐1α significantly increased CA2 promoter–driven luciferase activity in HTR‐8/SVneo cells compared with the control group. To further characterize the effect of hypoxia on CA2 transcriptional activity, CA2 promoter–driven luciferase activity was examined at different time points following hypoxic exposure. As shown in Figure 6L, hypoxia significantly increased CA2 promoter activity at 12 h. However, no significant difference was observed at 24 h, whereas prolonged hypoxic exposure for 36 h resulted in a marked reduction in CA2 promoter activity. These findings indicate that hypoxia regulates CA2 transcription in a time‐dependent manner.

FIGURE 6.

HIF‐1α activates the transcription of CA2. (A, B) RT‐qPCR analysis of CA2 and HIF‐1α to detect the expression of mRNA level under hypoxic or normoxia exposure in HTR‐8/Svneo cells. (C) Representative western blot bands. (D, E) Western Blot analysis of CA2 and HIF‐1α to detect the expression of protein level under hypoxic or normoxia exposure in HTR‐8/Svneo cells. (F, G) RT‐qPCR analysis of CA2 and HIF‐1α to detect the expression of mRNA level after overexpression HIF‐1α in HTR‐8/Svneo cells. (H) Representative western blot bands. (I, J) Western Blot analysis of CA2 and HIF‐1α to detect the expression of protein level after overexpression HIF‐1α in HTR‐8/Svneo cells. (K) Luciferase activity of HTR‐8/SVneo cells with or without overexpression HIF‐1α. (L) Luciferase activity of HTR‐8/SVneo cells with or without exposing hypoxic environment. pcDNA3.1‐control: Empty vector control group; pcDNA3.1‐HIF‐1α: HIF‐1α overexpression group; CA2 promoter + Vecter: Plasmid with CA2 promoter + empty vector control group; CA2 promoter + HIF‐1α OE: Plasmid with CA2 promoter + HIF‐1α overexpression group; CA2 promoter + Vecter + hypoxia 12 h: Plasmid with CA2 promoter + empty vector control group + exposing hypoxic environment 12 h; CA2 promoter + Vecter + hypoxia 24 h: Plasmid with CA2 promoter + empty vector control group + exposing hypoxic environment 24 h; CA2 promoter + Vecter + hypoxia 36 h: Plasmid with CA2 promoter + empty vector control group + exposing hypoxic environment 36 h. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

3.7. Preparation and Physicochemical Characterization of CA2 siRNA‐Loaded Lipid Nanoparticles (LNPs)

The above findings position CA2 as an important molecule in placental hypoxia and trophoblast dysfunction in PE. Based on this mechanistic evidence, we hypothesized that inhibition of CA2 may serve as a promising therapeutic strategy for PE. To verify this concept, multiple siRNA sequences targeting CA2 were designed and screened. Among all candidates, siCA2‐1208 exerted the strongest knockdown efficiency at both mRNA and protein levels (Figure S4A–C). Subsequently, blank LNPs (LNP‐NC) and siCA2‐encapsulated LNPs (LNP‐siCA2) were fabricated, and their physicochemical properties including particle size and polydispersity index (PDI) were comprehensively characterized via dynamic light scattering (DLS). DLS intensity profiles revealed a homogeneous unimodal particle size distribution for both formulations, without obvious impurity peaks or large aggregates across the detection range, demonstrating favorable dispersion of the prepared nanoparticles. The peak hydrodynamic diameter of LNP‐NC was 310.7 nm, while LNP‐siCA2 exhibited nearly identical particle size, indicating that siRNA loading exerted negligible influence on nanoparticle dimensions. Moreover, the PDI values of both LNP formulations were below 0.2, a threshold indicative of high monodispersity and stable colloidal properties that satisfy standard quality criteria for nucleic acid delivery carriers. Representative particle size distribution curves for LNP‐NC and LNP‐siCA2 are presented in Figure S4D,E. The siRNA encapsulation efficiency of LNP‐siCA2 was quantified using the RiboGreen fluorescence assay, yielding an encapsulation rate of 92.3% (well above the 80% quality threshold). This high encapsulation capacity enables LNPs to protect encapsulated siCA2 from nuclease degradation and facilitates efficient siRNA delivery both in vitro and in vivo. To further assess the efficiency of LNP‐mediated siRNA delivery, HTR‐8/SVneo cells were treated with LNP‐siCA2, LNP‐empty, LNP‐NC, siCA2, or NC. Western Blot analysis showed that CA2 expression was significantly reduced in both the LNP‐siCA2 and siCA2 groups compared with the control groups. Importantly, the knockdown efficiency of LNP‐siCA2 was comparable to that of direct siCA2 transfection. In contrast, no significant change in CA2 expression was observed in the LNP‐empty or LNP‐NC groups (Figure S4F,G). These results indicate that LNP‐mediated delivery effectively silences CA2 expression in vitro.

3.8. Silencing of CA2 Partially Rescues Adverse Phenotypes in an L‐NAME‐Induced PE Mouse Model

To investigate the potential role of CA2 in the pathogenesis of preeclampsia in vivo, LNP‐siCA2 was administered via tail vein injection in the L‐NAME–induced PE mouse model (Figure 7A). To verify the in vivo silencing efficiency of LNP‐siCA2, CA2 expression in placental tissues was first evaluated. RT‐qPCR demonstrated that placental CA2 mRNA expression was significantly reduced by approximately 50% in the LNP‐siCA2 group compared with the LNP‐NC group (Figure 7B). Immunofluorescence staining further showed a reduced CA2 fluorescence intensity by around 35% in placentas from LNP‐siCA2‐treated mice (Figure 7C,D). In contrast, Western blot analysis revealed no statistically significant difference in total placental CA2 protein levels between the two groups (Figure S5A,B). These findings indicate LNP‐siCA2 achieved partial suppression of placental CA2 expression at the transcript and tissue levels. We next evaluated its therapeutic effects in the L‐NAME‐induced PE mouse model. Compared with the LNP‐NC control group, CA2 knockdown did not produce a statistically significant reduction in maternal systolic blood pressure (Figure 7E). Likewise, the UPCR showed no obvious improvement after LNP‐siCA2 treatment (Figure 7F). Nevertheless, renal histological staining revealed that glomerular structural damage was markedly attenuated in LNP‐siCA2‐treated dams (Figure 7G). These results suggest that while CA2 knockdown partially ameliorates renal structural damage, its effects on renal functional parameters may be limited under the current experimental condition. And CA2 silencing reduced circulating sFlt‐1 levels compared with the LNP‐NC group, indicating that inhibition of CA2 may partially modulate the anti‐angiogenic state associated with preeclampsia (Figure 7H). Despite the limited effects on maternal symptoms, CA2 inhibition improved fetal and placental outcomes. Fetal birth weight was remarkably elevated in the LNP‐siCA2 group versus controls (Figure 7I,K). Consistently, placental efficiency was significantly restored following CA2 knockdown (Figure 7J). Placental histological quantification further demonstrated that the labyrinthine/junctional zone area ratio was substantially increased in LNP‐siCA2‐treated placentas (Figure 7L,M), suggesting improved placental development and enhanced maternal–fetal exchange capacity. Taken together, our in vivo animal experiments demonstrate that lipid nanoparticle‐mediated CA2 suppression partially mitigates placental structural dysfunction and rescues fetal growth restriction in PE model mice, despite limited therapeutic benefits for maternal hypertensive and renal functional phenotypes.

FIGURE 7.

Inhibition of CA2 partially improves outcomes in a L‐NAME‐induced PE mouse model. (A) Experimental workflow for LNP‐NC and LNP‐siCA2 treatment in L‐NAME PE mice. (B) MRNA expression level of placental CA2 in LNP‐NC and LNP‐siCA2 groups detected by RT‐qPCR. (C) Representative immunofluorescence staining of CA2 in LNP‐NC and LNP‐siCA2 groups. Scale bars: 500 μm. (D) Quantitative analysis of mean fluorescence intensity of CA2 immunofluorescence signals n LNP‐NC and LNP‐siCA2 groups. (E) Changes in systolic blood pressure in LNP‐NC and LNP‐siCA2 groups. (F) The ratio of urinary albumin and creatine of mice in LNP‐NC group and LNP‐siCA2 group. (G) Representative renal stained with hematoxylin & eosin (H&E) in LNP‐NC and LNP‐siCA2 groups. (H) SFLT‐1 levels of mice in LNP‐NC and LNP‐siCA2 groups. (I, K) Body weights of fetal mice in LNP‐NC and LNP‐siCA2 groups. (J) Placental efficiency of fetal mice in LNP‐NC group and LNP‐siCA2 group. (L) Representative placental stained with hematoxylin & eosin (H&E) in LNP‐NC and LNP‐siCA2 groups. Scale bars: 250 μm. (M) Ratio of palcental labyrinth and junction zone of mice in LNP‐NC and LNP‐siCA2 groups. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

4. Discussion

Preeclampsia is a complex pregnancy disorder characterized by placental dysfunction and systemic endothelial injury. While extensive studies have focused on angiogenic imbalance and hypoxia‐related pathways, the molecular mechanisms linking hypoxic stress to trophoblast dysfunction remain incompletely understood. In the present study, we identified CA2 as a molecule associated with PE by integrating public transcriptomic data, clinical sample validation, cellular functional assays, and animal experiments. These findings identify CA2 as a previously underappreciated molecule involved in placental dysfunction and expand the current understanding of hypoxia‐associated molecular alterations in PE.

CA2 is one of the most abundant and catalytically active isoforms and has been implicated in diverse biological processes, including cellular metabolism, vascular regulation, and proliferation [15]. Although CA2 has been extensively studied in other physiological and pathological contexts [16, 17], its role in placental biology and PE remains largely unexplored. Previous studies have reported alterations of carbonic anhydrase activity in pregnancy‐related disorders, including circulating CA isoenzyme‐related antibodies in hypertensive disorders and potential involvement of CA2 in metabolic disturbances such as gestational diabetes [18, 19, 20]. Together with our findings, these observations suggest that dysregulated CA2 may represent an underrecognized contributor to pregnancy‐associated pathological processes.

The clinical relevance of CA2 dysregulation was supported by several independent observations in this study. Elevated placental CA2 expression was associated with reduced fetal birth weight, suggesting a potential relationship between CA2 expression and impaired placental efficiency. In addition, increased circulating CA2 levels were detected in women with PE, suggesting that altered CA2 levels may also be detected at the systemic level in PE. Consistent with this possibility, previous studies have demonstrated that CA9 can be detected in maternal circulation and has been investigated as a potential biomarker in preeclamptic pregnancies, indicating that circulating carbonic anhydrase family members may have clinical relevance in PE and warrant further investigation as potential biomarkers [14]. Because gestational age is an important confounding factor in placental transcriptomic and protein studies, we further performed gestational age‐stratified validation. CA2 expression showed a consistent tendency toward higher levels in PE samples in the preterm gestational age groups, with statistically significant differences observed in the < 34 and 34–37 weeks groups, whereas no significant difference was observed in term pregnancies (> 37 weeks). These findings suggest that the increased CA2 expression observed in preeclamptic placentas, particularly in preterm PE, is unlikely to be solely attributable to gestational age differences. Importantly, preeclampsia is recognized as a heterogeneous disorder with distinct clinical subtypes. Early‐onset and preterm preeclampsia are more strongly associated with abnormal placentation, impaired trophoblast function, and placental hypoxia, whereas late‐onset preeclampsia is considered to involve a greater contribution from maternal cardiovascular and metabolic factors. Therefore, the preferential elevation of CA2 in preterm PE may reflect its involvement in the placental dysfunction component of the disease.

Consistent with the clinical observations, our in vitro experiments demonstrated that CA2 overexpression significantly impaired trophoblast proliferation, migration, and invasion while having minimal effects on apoptosis. These results suggest that CA2 primarily affects trophoblast functional capacity rather than cell survival. Proper trophoblast invasion and remodeling of maternal spiral arteries are essential for establishing adequate placental perfusion during early pregnancy [21]. Therefore, excessive CA2 expression may contribute to PE not by eliminating trophoblasts, but by preventing them from acquiring the functional characteristics required for normal placental development.

Further mechanistic analyses suggested that CA2 may be integrated into hypoxia‐related regulatory pathways in trophoblasts. Placental hypoxia and sustained activation of hypoxia‐inducible factor‐1α (HIF‐1α) are well‐recognized features of PE and contribute to abnormal gene expression in placental cells [9]. In our experimental models, CA2 expression was responsive to HIF‐1α signaling, suggesting that CA2 may represent a downstream component of hypoxia‐associated transcriptional responses in trophoblasts. This observation is consistent with previous findings that CA2 can be upregulated under hypoxic conditions [3]. Although hypoxia‐inducible regulation has been most clearly established for other carbonic anhydrase isoforms such as CA9 [16], our findings indicate that CA2 expression in trophoblasts may also be influenced by hypoxia‐related signaling pathways within the placental microenvironment.

Another important finding of this study is the potential involvement of VEGFA signaling in CA2‐mediated trophoblast dysfunction. VEGFA is a central regulator of placental angiogenesis and trophoblast invasion, and impaired VEGFA‐related pathways are strongly implicated in PE pathogenesis [22, 23, 24, 25, 26]. Our transcriptomic analysis identified VEGFA as a candidate downstream molecule affected by CA2 overexpression, and restoration of VEGFA expression partially rescued CA2‐induced trophoblast dysfunction. These findings support VEGFA as a functional mediator contributing, at least partially, to CA2‐induced trophoblast dysfunction. However, whether CA2 directly regulates VEGFA expression remains unknown. CA2 is a key carbonic anhydrase involved in maintaining intracellular acid–base homeostasis through catalyzing the reversible conversion of carbon dioxide and bicarbonate, thereby contributing to cellular metabolic adaptation [15]. Previous studies have demonstrated that extracellular acidosis can increase VEGF expression through enhanced promoter activity, increased transcription, and prolonged mRNA stability, indicating that pH homeostasis represents an important regulatory layer of VEGF signaling [27]. In addition, recent studies have highlighted the close interaction between metabolic remodeling and angiogenic regulation, showing that alterations in cellular metabolism can modulate VEGFA‐associated pathways [28]. Therefore, we speculate that CA2 may indirectly influence VEGFA signaling by altering the metabolic and intracellular pH environment of trophoblasts rather than acting as a direct transcriptional regulator. Further studies are required to elucidate the precise molecular mechanisms linking CA2 activity to VEGFA dysregulation.

The therapeutic relevance of CA2 was further supported by our in vivo experiments. Systemic inhibition of CA2 using LNP‐mediated siRNA partially improved placental structure, placental efficiency, and fetal growth in the PE mouse model. Interestingly, the beneficial effects of CA2 inhibition were more evident in the placental and fetal compartments than in maternal systemic manifestations. Although renal pathological changes and circulating sFLT‐1 levels were improved, blood pressure and urinary protein levels were not significantly altered. This observation suggests that CA2 may primarily participate in placental pathological processes rather than serving as a dominant regulator of maternal hypertension. The discrepancy between reduced CA2 mRNA/immunofluorescence signals and unchanged total protein abundance detected by Western blot may reflect the partial nature of in vivo siRNA‐mediated partial suppression and differences between transcript and protein regulation. Moreover, immunofluorescence analysis may better capture localized changes in CA2 expression within specific placental regions. Nevertheless, the consistent molecular and phenotypic changes support the biological relevance of CA2 inhibition in PE.

Several limitations should be acknowledged. First, expression patterns of CA2 were not entirely consistent across all publicly available transcriptomic datasets. This variability may reflect the inherent heterogeneity of placental tissue, differences in cohort characteristics, or variation in gestational age at delivery. Second, although multiple cohorts supported increased CA2 expression in PE, the clinical sample sizes, particularly in gestational age‐stratified analyses, remain limited. Third, while VEGFA rescue experiments support its role as a functional mediator downstream of CA2‐mediated effects, the molecular mechanism underlying this relationship requires further investigation. Lastly, systemic rather than placenta‐specific delivery was used for CA2 inhibition in vivo, which may have influenced the magnitude of therapeutic effects.

In conclusion, our study identifies CA2 as a potential metabolic regulator involved in PE‐associated placental dysfunction. We propose that chronic placental hypoxic stress may induce CA2 upregulation through HIF‐1α‐associated pathways, and that elevated CA2 may impair trophoblast function partly through VEGFA‐related signaling alterations, ultimately contributing to impaired placental development and adverse fetal outcomes. These findings expand our understanding of metabolic regulation in PE and highlight CA2 as a potential target for future therapeutic exploration.

5. Conclusion

In conclusion, this study identifies CA2 as a previously underrecognized molecule associated with placental dysfunction in preeclampsia. By integrating clinical validation, cellular functional studies, transcriptomic analysis, rescue experiments, and in vivo therapeutic evaluation, we demonstrate that increased CA2 expression is associated with impaired trophoblast function and adverse fetal outcomes. Our findings further support VEGFA as a functional downstream mediator contributing to the biological effects of CA2, although the precise molecular mechanism linking CA2 and VEGFA requires further investigation. Moreover, CA2 silencing partially improved placental development and fetal growth in experimental PE, suggesting that CA2 may represent a promising therapeutic target for placental dysfunction in preeclampsia.

Author Contributions

L.C., C.Z., and J.Z. contributed to complete the experiment and prepare the manuscript. H.L. and L.Y. contributed to collect, analyze, and interpret the experimental data. Y.J. and Q.L. contributed to study design, data interpretation, and the manuscript revision. Q.L. is the guarantor of this work, has full access to all the data in the study, and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

This work was supported by the National Key Research and Development Program of China (no. 2023YFC2705605).

Ethics Statement

This paper has received approval from the Medical Ethics Committee of Women's Hospital, Zhejiang University School of Medicine (ethics approval nos IRB‐20220245‐R, IRB‐20260073‐R).

Consent

Prior to commencing the questionnaire, written informed consent was obtained from each participant. Each questionnaire provided a detailed explanation of the study's purpose, methodologies, and potential risks associated with participation, ensuring informed consent. This information was presented on the first page of every questionnaire. Participants were afforded the freedom to decide whether or not to participate in the research.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: CA2 protein expression in an independent gestational age‐stratified validation cohort (< 34, 34–37, and > 37 weeks). (A) Representative western blot images of CA2 from placental specimens delivered before 34 weeks. (B) Quantitative analysis of CA2 protein levels in PE and control placentas within the < 34 weeks subgroup. (C) Representative western blot images of CA2 from placental specimens at 34–37 weeks of gestation. (D) Quantitative analysis of CA2 protein levels in PE and control placentas within the 34–37 weeks subgroup. (E) Representative western blot images of CA2 from placental specimens delivered after 37 weeks. (F) Quantitative analysis of CA2 protein levels in PE and control placentas within the > 37 weeks subgroup. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure S2: Preeclampsia Modeling and CA2 expression in the RUPP model. (A, B) Schematic illustration and experimental workflow of the reduced uterine perfusion pressure (RUPP) model. (C) Changes in systolic blood pressure in RUPP and SHAM groups. (D, F) Body weights of fetal mice in RUPP and SHAM groups. (E) Placental efficiency of fetal mice in RUPP and SHAM groups. (G) The ratio of urinary albumin and creatine of mice in RUPP and SHAM groups. (H, I) Representative renal and placental stained with hematoxylin & eosin (H&E) in RUPP and SHAM groups. Scale bars: 250 μm. (J) Ratio of placental labyrinth and junction zone of mice in RUPP and SHAM group. (K) sFLT‐1 levels of mice in RUPP and SHAM groups. (L) mRNA expression level of placental CA2 in RUPP and SHAM groups detected by RT‐qPCR. (M) Representative western blot bands for CA2. (N) Protein expression level of placental CA2 in RUPP and SHAM groups detected by Western Blot. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure S3: VEGFA expression levels in PE and normal placentas. A: RT‐qPCR analysis of VEGFA in PE and normal placentas. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure S4: Construction and characterization of liposomes encapsulating CA2 siRNA. (A) RT‐qPCR analysis of CA2 mRNA level after transfection with NC, siCA2‐1140, siCA2‐1208, siCA2‐982 and siCA2‐651. (B) Representative western blot bands. (C) Protein expression level of CA2 after transfection with NC, siCA2‐1140, siCA2‐1208, siCA2‐982 and siCA2‐651 detected by Western Blot. (D, E) Size distribution curves of LNP‐NC and LNP‐siCA2 by DLS analysis. (F) Representative western blot bands. (G) Protein expression level of CA2 after were treated with LNP‐siCA2, LNP‐empty, LNP‐NC, siCA2, or NC detected by Western Blot. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

Figure S5: Inhibition of CA2 partially improves outcomes in a L‐NAME‐induced PE mouse model. (A) Representative western blot bands for CA2 protein in LNP‐NC and LNP‐siCA2 groups. (B) Quantitative western blot analysis of CA2 protein levels in placental tissues in LNP‐NC and LNP‐siCA2 groups. All quantitative data are expressed as means ± SD; ns: p > 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

Table S1: Clinical characteristics of the two independent validation cohorts.

Table S2: Detailed information on the primers, plasmids, and siRNA sequences used in this study.

Data Availability Statement

The RNA‐seq data generated in this study have been deposited in the Gene Expression Omnibus (GEO) database under accession number GSE342874.