In 1997, Lo et al. first discovered the presence of cfDNA in maternal plasma and serum, introducing the unique possibility of using non-invasive methods (i.e., routine blood tests) to obtain fetal material. Since then, numerous studies have focused on identifying the source, characteristics, and predictive potential of cfDNA, with particular emphasis on predicting pregnancy-related complications and prenatal diagnosis or screening of genetic fetal disorders. To date, cfDNA has been widely used in prenatal screening for aneuploidy, single-gene disorders, chromosomal abnormalities, placenta-associated diseases, and Rh factor assessment (7, 8). Some studies have found elevated levels of cfDNA in the blood of women with preeclampsia (9, 10), but the role of cfDNA as a reliable predictor of preeclampsia remains controversial (11). As such, the predictive value of cfDNA in preeclampsia is still under investigation. In addition to the well-established parameters such as fetal fraction and total cfDNA concentration, increasing attention has been directed toward cfDNA-associated mutational and fragmentation patterns. Mutations refer to specific genetic alterations present in fetal or maternal DNA. For instance, if DNA released from the placenta harbors mutations in certain genes, this may indicate placental dysfunction and has been associated with preeclampsia (PE). Some studies have reported that epigenetic markers, such as DNA methylation, can be utilized to differentiate fetal from maternal cfDNA or to detect alterations in specific methylation patterns. Placental dysfunction in preeclampsia has been linked to aberrant methylation in genes such as RASSF1A, HLTF, and TIMP3. Therefore, assessing methylation levels of placenta-derived cfDNA in maternal plasma may serve as an indirect indicator of placental health (12). Furthermore, the presence of other genetic variations, such as single nucleotide polymorphisms (SNPs) or copy number variations (CNVs), may also be associated with an increased risk of preeclampsia. Certain SNPs have been implicated in elevated PE susceptibility, and the detection of these variants via cfDNA analysis could aid in early risk prediction.

Fragmentation patterns, particularly the size distribution of cfDNA fragments, are another area of growing interest. In healthy pregnancies, fetal cfDNA fragments are typically shorter than maternal fragments, with a predominant size around 166 base pairs (bp). Preeclampsia may result in altered placental DNA release, leading to shifts in cfDNA fragment size distributions, such as an increased proportion of shorter fragments (< 150 bp) (13). This phenomenon is potentially linked to shifts in apoptotic or necrotic processes due to placental ischemia. Moreover, nucleosome positioning patterns may also be disrupted. Aberrant apoptosis in the preeclamptic placenta can modify nucleosome occupancy, leading to specific fragment end preferences, such as an increased presence of fragment ends in open chromatin or gene regulatory regions. These alterations are potentially associated with altered nuclease activity, particularly that of enzymes like DNASE1L3 (14). Additionally, genomic instability, including microdeletions or duplications in certain chromosomal regions, may be more prevalent in preeclampsia and can be detected through cfDNA sequencing. Finally, analysis of cfDNA end motifs—the specific nucleotide sequences at the fragment ends—may reveal disease-specific patterns. Differences in end motif distributions in PE patients could reflect alterations in nuclease activity or underlying cell death mechanisms.

Researchers have measured circulating DNA levels in samples from women with preeclampsia and those with normal pregnancies using real-time polymerase chain reaction (PCR), and the results suggest that elevated fetal DNA levels are associated with preeclampsia (9). Another study reported a fivefold increase in fetal cfDNA levels in symptomatic preeclamptic women compared to asymptomatic preeclamptic women (15). Similarly, it has been reported that fetal DNA levels rise in women who eventually develop preeclampsia, with the important finding that cfDNA levels increase prior to the onset of clinical symptoms (16).

Circulating cell-free DNA (cfDNA) refers to fragmented DNA molecules present in the bloodstream, with the fetal-derived fraction known as cell-free fetal DNA (cffDNA), which predominantly circulates in the maternal peripheral blood. cffDNA primarily originates from the programmed cell death (apoptosis) or necrosis of placental syncytiotrophoblasts and exhibits placental specificity. Its levels are closely associated with placental function, and significant elevations have been observed in pathological conditions such as placental abruption and preeclampsia (17). During its release, cffDNA exhibits characteristic fragmentation, with an average fragment size of ~143 base pairs (bp), which is markedly shorter than that of maternal cfDNA (~166 bp). This size difference may be attributed to placenta-specific nuclease activity. The release rate of cffDNA is influenced by changes in placental metabolism, and increased release is observed with placental aging during late pregnancy. cffDNA has a very short half-life in maternal circulation, ranging from ~15 min to 2 h, and is rapidly cleared through renal filtration, hepatic and splenic uptake, and nuclease-mediated degradation. This rapid turnover allows cffDNA levels to reflect the real-time status of the placenta; however, it also necessitates prompt sample processing after collection to ensure the accuracy of downstream analyses.

The proportion of cffDNA within the total cfDNA varies significantly with gestational age. In early pregnancy (10–20 weeks of gestation), cffDNA accounts for ~10%−15% of maternal plasma cfDNA, during which the total maternal cfDNA concentration is relatively low (around 50–100 ng/ml of plasma). During the mid-gestation period (20–30 weeks), the proportion of cffDNA increases to 15%−30%, accompanied by a slight rise in total maternal cfDNA due to increased blood volume (18). In late pregnancy (>30 weeks), the cffDNA fraction can exceed 30%, although substantial interindividual variability exists, influenced by factors such as maternal body mass index (BMI) and placental function. For example, in obese pregnant women, expanded blood volume may lead to a relative reduction in the proportion of cffDNA. In pathological conditions such as preeclampsia and placenta accreta, cffDNA levels can be abnormally elevated. Additionally, fetal chromosomal abnormalities, such as trisomy 21, may also result in increased cffDNA release.

In preeclampsia, the increase in circulating cell-free DNA (cfDNA) levels is associated with markers of disease severity, including preterm birth and worsening systolic blood pressure—two recognized key indicators of disease severity (19). These findings may suggest an increase in cfDNA release following maternal tissue injury (20). Given the close correlation between total cfDNA concentration and preeclampsia, one study analyzed the relationship between total cfDNA levels in preeclamptic patients and surrogate markers of disease severity. The results indicated that cfDNA concentration was negatively correlated with gestational age at delivery and moderately positively correlated with maximum systolic blood pressure (21). Furthermore, the study found that the fraction of placental-derived cfDNA was moderately negatively correlated with maximum systolic blood pressure, but showed no association with maximum diastolic blood pressure.

Fetal sex is considered an important risk factor for preeclampsia, with growing speculation that placental formation and maternal adaptation to pregnancy may be influenced by fetal sex. Therefore, it is essential to characterize fetal sex when studying the pathophysiological processes at the maternal-fetal interface (22, 23). Previous studies have indicated that the amount of placental-derived cfDNA in maternal plasma is higher in cases of preeclampsia; however, these studies were limited to male fetuses and employed alternative methods (such as PCR for Y-chromosome material), making it difficult to ascertain the specific contributions. In healthy non-pregnant individuals, cfDNA primarily originates from hematopoietic lineages, with smaller contributions from endothelial cells, neurons, and hepatocytes. Understanding the exact sources of maternal cfDNA can enhance insights into the pathophysiology of preeclampsia and determine whether cfDNA sources are related to disease phenotypes, given the particularly heterogeneous manifestations of preeclampsia (21). In addition to increased production, cfDNA levels are also influenced by clearance mechanisms found in the liver, spleen, and kidneys. It has been proposed that placental cfDNA increases 3 weeks prior to the clinical manifestations of preeclampsia, a rise attributed to accelerated apoptosis of trophoblasts.

In a large multicenter study involving 44 women with preeclampsia and 53 controls, it was concluded that not only are fetal cfDNA levels elevated in preeclampsia, but maternal cfDNA levels also exhibit a similar increase, both showing a tenfold rise compared to the control group. Moreover, this study's population was significantly larger than those previously described, allowing for stratification of preeclampsia cases based on the severity of symptoms. The analysis included three cases of HELLP syndrome (hemolysis, elevated liver enzymes, low platelet count) and four cases of eclampsia, revealing that the increases in cell-free fetal and maternal cfDNA corresponded with disease severity. Notably, cfDNA levels in women with severe preeclampsia increased by 3.5 times compared to those with mild preeclampsia, and were tenfold higher than in the control group. Furthermore, it was demonstrated that the levels of these two types of cfDNA are correlated in pregnancies affected by preeclampsia, unlike in normal pregnancies. Recent observational reports on cell-free fetal DNA and maternal DNA levels in pregnancies with preeclampsia and subsequent HELLP syndrome have yielded similar findings. These authors also noted that, compared to women with mild preeclampsia, levels of cell-free fetal and maternal DNA in pregnancies complicated by HELLP syndrome nearly quadrupled (24). Maternal cfDNA levels were similarly elevated. In summary, the increase in disease severity is associated with enhanced release of cell-free fetal and maternal DNA into the maternal circulation. These studies collectively suggest that quantitative and fragmentomic profiles of cfDNA are significantly associated with the clinical stratification of preeclampsia. Taken together, these findings indicate that cfDNA levels, particularly the placental-derived fraction, are closely associated with the severity of PE. This suggests that cfDNA may not only serve as a biomarker of disease severity but could also potentially participate directly in the pathophysiology of PE by mediating immune-inflammatory responses. Key evidence and correlations are summarized in Table 1.

Placentas in preeclampsia commonly exhibit inadequate perfusion and ischemia-hypoxia, leading to exacerbated trophoblast cell apoptosis and/or necrosis. This process releases substantial amounts of cfDNA/cffDNA exhibiting DAMP properties. Fragmentation profile: placenta-derived cfDNA in PE patients shows significantly shortened fragments (typically < 150 bp), resembling DNA released from apoptotic bodies and facilitating immune recognition. Hypomethylation status: in preeclampsia, the free circulating cell-free fetal DNA in maternal blood is hypomethylated, indicating that methyl groups are added to cytosine-guanine (CpG) residues in the DNA molecule. Hypomethylated CpG motifs structurally mimic bacterial DNA. Mitochondrial DNA (mtDNA) enrichment: in preeclamptic placentas associated with pregnancy, there is an increased apoptosis of trophoblasts, resulting in the release of circulating fetal DNA containing mitochondrial DNA (mtDNA) (25). MtDNA similarly contains abundant unmethylated CpG islands and demonstrates striking structural similarity to bacterial DNA.

Elevated levels of cfDNA/cffDNA/mtDNA in maternal circulation act as DAMPs, sensed by pattern recognition receptors (PRRs) that trigger a robust sterile inflammatory response—a hallmark of PE pathophysiology. The Toll-like receptor 9 (TLR9) pathway plays a central role: as the key receptor for unmethylated CpG DNA motifs, TLR9 demonstrates significantly upregulated expression in placental tissue and peripheral blood monocytes from preeclampsia patients. Serological analysis revealed that phosphorylation levels of MyD88 and NF-κB, key downstream molecules of the TLR9 signaling pathway, were significantly elevated in peripheral blood mononuclear cells (PBMCs) from preeclamptic patients (26). Activated NF-κB translocates to the nucleus, driving the transcription and expression of multiple pro-inflammatory cytokines (e.g., TNF-α, IL-6, IL-8) and type I interferons (e.g., IFN-β). In addition, the number of extracellular vesicles (EVs) in the placentas of preeclamptic patients was approximately twofold higher than that in normal pregnancies. These EVs were enriched in mitochondrial DNA (mtDNA) and nucleosomal fragments, and may exacerbate local placental inflammation via the TLR9/NF-κB axis (27). Key experimental evidence supports this mechanism that trophoblast cells treated with preeclamptic serum exhibited significantly increased mRNA expression of TLR9 and IFN-β (28). Treatment with genomic DNA removal agents significantly inhibited the secretion of IL-6 and IL-8 induced by both normal and preeclamptic serum. Beck et al. (29) proposed that cfDNA has pro-inflammatory effects, cfDNA injection provokes IL-6 production in mice via TLR9. When both cfDNA and the TLR9 inhibitor chloroquine were administered to mice, the pregnancy outcomes improved (30).

Cytosolic DNA-Sensing Receptors Pathway (AIM2/IFI16): at the placental level, protein expression of AIM2 and IFI16 is significantly elevated in placentas from preeclamptic patients (31). This recognition triggers inflammasome assembly, activating caspase-1. Active caspase-1 then cleaves pro-IL-1β and pro-IL-18 into their mature, highly pro-inflammatory forms, while concurrently suppressing pro-angiogenic factors like PlGF, thereby exacerbating PE symptoms.

CfDNA activates both TLR9 and cytosolic DNA-sensing pathways, collectively driving exaggerated inflammatory responses locally (placenta) and systemically. The resulting proinflammatory cytokine storm: Characterized by massive release of TNF-α, IL-6, IL-8, IL-1β, IL-18, IFN-β, etc.—inflicts direct cytotoxic effects on placental trophoblast cells and maternal vascular endothelial cells. Release of Anti-angiogenic Factors: at the RNA level, the mRNA expression of IL-6 and IL-8 is significantly elevated in trophoblast cells treated with preeclamptic serum compared to those treated with serum from normal pregnancies. However, there is no significant difference in mRNA expression of Eng and Flt1 between the two groups. At the protein level, trophoblast cells stimulated with preeclamptic serum exhibit significantly increased secretion of soluble endoglin (sEng) and soluble fms-like tyrosine kinase-1 (sFlt-1) (32, 33). However, the protein levels of IL-6 and IL-8 remained below the detection threshold in both preeclamptic and normal serum-treated groups. Notably, no significant differences were observed in the expression of Eng and Flt1 genes in placental tissues from preeclamptic patients compared to controls, despite markedly elevated serum levels of sEng and sFlt-1 (34, 35). These findings suggest that preeclamptic serum exacerbates placental dysfunction primarily by enhancing the secretion of anti-angiogenic factors (sEng and sFlt-1) from trophoblast cells, rather than by upregulating their gene expression. Endothelial Injury and Systemic Inflammation: proinflammatory cytokines and anti-angiogenic factors act synergistically to damage maternal vascular endothelium, leading to systemic vasoconstriction, hypertension, coagulation activation, and multi-organ hypoperfusion—the cardinal clinical manifestations of preeclampsia.

Research has reported that inflammatory cells in preeclamptic patients, including neutrophils and monocytes, are activated and secrete large amounts of inflammatory cytokines. Even in the absence of microbial infection, inflammation associated with preeclampsia occurs, representing a form of sterile inflammation. In preeclamptic patients, the number of trophoblast-derived extracellular vesicles entering the maternal bloodstream significantly increases. These extracellular vesicles contain various factors, including DNA, RNA, lipids, and proteins, acting as damage-associated molecular patterns (DAMPs). In fact, extracellular vesicles can induce sterile inflammation and preeclampsia-like features in mouse placentas. Recent studies indicate that cfDNA, considered a product of apoptosis and/or necrosis, functions as a DAMP and is associated with various inflammatory diseases (27). During pregnancy, the total amount of cfDNA in maternal blood significantly increases, and elevated cfDNA levels are significantly correlated with pregnancy complications. Empirical findings are detailed in Table 2.

The pathogenesis of preeclampsia further involves aberrations in metabolic pathways, in which the release and sensing of cfDNA may play interrelated roles: mitochondrial dysfunction and the release of mtDNA contribute to this process: placental ischemia and hypoxia not only promote cell death, leading to the release of cfDNA and mtDNA, but also reflect an underlying disruption in mitochondrial bioenergetics. Circulating cell-free mitochondrial DNA (ccf-mtDNA) is considered a DAMP molecule that triggers immune responses by activating the pattern recognition receptor Toll-like receptor 9 (TLR9). Compared to gestational age–matched healthy pregnant individuals, those with preeclampsia exhibit aberrations in circulating DNA dynamics, including elevated levels of ccf-mtDNA and compromised DNA clearance mechanisms (36).

Notably, this mechanism of immune activation mediated by mtDNA is not unique to PE. In sepsis, a systemic inflammatory response syndrome secondary to bacterial infection, mitochondrial dysfunction and the subsequent release of mtDNA are also established as central to its pathogenesis (37). Here, reactive oxygen species (ROS) and reactive nitrogen species can induce functional impairment in multiple organelles, including mitochondria, thereby amplifying the inflammatory cascade. Furthermore, studies suggest that variations in mitochondrial DNA haplogroups may influence the severity and progression of sepsis. This analogy provides strong supportive evidence for understanding the similar role of mtDNA in amplifying inflammation in preeclampsia.

Aberrant Expression of Metabolism-Associated Genes: canonical preeclampsia-associated genes—previously implicated in metabolic regulation, hypoxia, and angiogenesis—include LEP, HK2, FSTL3, FLT1, ENG, TMEM45A, ARHGEF4, and HTRA1. These genes have consistently been shown to be upregulated in preeclamptic placentas compared with normotensive controls across diverse ancestral backgrounds. In the present study, RNA sequencing of placental tissue from patients with severe PE (sPE) of African, Asian, and European ancestry confirmed the sustained upregulation of these canonical genes across all cohorts. Among them, HK2, FSTL3, LEP, and FLT1 exhibited the most pronounced increases. Hexokinase 2 (HK2), which plays a critical role in glucose metabolism, has been frequently reported to be dysregulated in PE (38). Leptin (LEP), a secreted adipokine, systemically regulates energy homeostasis, neuroendocrine signaling, and cytoplasmic metabolic processes. Elevated early-pregnancy leptin levels are a recurrent observation in PE patients, and exogenous leptin administration in murine models recapitulates key clinical features of preeclampsia (39).

The potential role of the cGAS-STING pathway: cytosolic DNA (including cfDNA/mtDNA) is recognized by cyclic GMP-AMP synthase (cGAS), catalyzing the production of the second messenger cyclic GMP-AMP (cGAMP). In pregnancies complicated by PE, increased cGAMP levels are associated with elevated circulating natriuretic peptides. Transfected cytosolic DNA triggers cGAMP production, which binds to the endoplasmic reticulum-resident protein STING and subsequently activates interferon regulatory factor 3 (IRF3) and interferon-beta (IFN-β). As a nucleotide second messenger belonging to the family of cyclic dinucleotides, cGAMP is capable of forming unique2′5′ phosphodiester linkages. Taken together, it is plausible that circulating fetal DNA in PE behaves in a manner similar to bacterial or mitochondrial DNA, potentially activating the cGAS–STING pathway and inducing inflammation (40). This process represents another potential pathway through which cfDNA drives inflammation, meriting further research. The interplay mechanisms between cfDNA and placental metabolic perturbations, including bioenergetic failure and oxidative stress, along with their clinical implications, are systematically collated in Table 3.