Peng et al. reported that miR-134 and miR-1233 are upregulated 14.59 and 21.56 folds, respectively, in chronic obstructive pulmonary disease (COPD) patients with APE compared to levels in ordinary COPD patients. D-dimer is a classic biomarker for detecting and evaluating PE progression. Of interest, the areas under the curves (AUCs) for the correlation of miR-134 and miR-1233 with PE progression were 0.931 (95% confidence interval [CI] 0.863–0.999) and 0.884 (95% CI 0.79–0.978), respectively, which were higher than those of D-dimer (0.628) (95% CI 0.447–0.809) and Wells score (0.577) (95% CI 0.389–0.765) (Peng et al., 2020). This study provided evidence that the levels of serum miR-1233 and miR-134 are significantly upregulated in COPD patients with APE complication. For miR-134, a meta-analysis also reported that its level was increased in APE patients compared to that in the control group (SMD = 2.84, z = 3.69, p < 0.001). The sensitivity, specificity, and diagnostic odds ratio were 0.86 (0.72–0.94), 0.75 (0.66–0.82), and 19 (7–51) respectively, which indicated its great potential as a PE biomarker (Liu et al., 2021). Among the variety of miRNA-related PE studies, research on miR-134 is the most extensive and reliable up to date (Xiao et al., 2011; Xiang et al., 2019). For miR-1233, it was reported to be the first miRNA biomarker for distinguishing APE from acute non-ST elevated myocardial infarction and healthy individuals (Kessler et al., 2016). Considering the confusing and overlapping clinical symptoms between APE and myocardial infarction, we consider these findings as exciting and meaningful evidences that contribute to the field.

miR-483-3p and miR-532-5p were found to correlate with PE, with sensitivity, specificity, and AUC of (1.00, 1.00, and 1.00) and (1.00, 1.00, and 1.00), respectively (Xiang et al., 2019). From the mechanistic point of view, miR-532-5p and miR-483-3p were found to share the same target, NFATC2IP gene, which is expressed in T-helper two cells and regulates the transcription of cytokine genes, including IL-3, 4, 5 and 13. In patients with PE, cytokines are abnormally expressed in peripheral circulating mononuclear cells. Moreover, the cellular immune response and pathways are impaired in PE, which is highly relevant to cytokine expression and implies the correlation between PE/venous thrombosis embolism and cellular immune dysfunction (Lv et al., 2013). Further, the target genes of these two miRNAs include molecules in the MAPK and PI3K-Akt signaling pathways, which are well-known to be involved in macrophage inflammatory, oxidative stress response (Zhang et al., 2015) as well as the apoptosis process of endothelial cells (Lou et al., 2016).

Liu et al. reported, in a cohort of 60 PE patients and 50 healthy volunteers, a significant increase in miR-221 level in the plasma from APE patients compared to that in healthy individuals. Correlation analysis showed a positive correlation between this miRNA and BNP (r = 0.842, p < 0.05), troponin I (r = 0.853; p < 0.05), and D-dimer (r = 0.838; p < 0.05) (Liu et al., 2018). Wang et al. also found that plasma miR-27a/b levels were significantly higher and showed positive correlation in APE patients compared to in the control group. Receiver operating characteristic (ROC) curve analyses showed that plasma miR-27a was superior to miR-27b with regard to the diagnosis of APE (AUC = 0.784, AUC = 0.707, respectively). Combining miR-27a or miR-27b with D-dimer significantly increased the diagnostic capacity for APE (Wang et al., 2018). Furthermore, miR-27a/b is related to the proliferation of pulmonary artery smooth muscle cells (PASMC). Endothelin-1 (ET-1) can stimulate miR-27a/b expression by activating the NF-κB pathway, which subsequently leads to a decrease in peroxisome proliferator-activated receptor γ and contributes to ET-1-induced PASMC proliferation. These processes may have a correlation with pulmonary vascular remodeling and pulmonary hypertension (Xie et al., 2017).

This miRNA has been proved to play a role in virus-associated cancers, such as esophageal squamous cell carcinoma, and neuropathies, such as Alzheimer’s disease (Han et al., 2019; Qian et al., 2019). A recent publication by Zhang et al. reported that inhibition of miR-338-5p can increase the level of IL-6, which subsequently promotes the development and progression of DVT in a mouse study. In a cell model, it was shown that overexpression of miR-338-5p downregulated the expression of IL-6, while miR-338-5p inhibition upregulated IL-6 expression. An in vivo study found that mice treated with anti-IL-6 antibody or agomiR-338-5p delivery exhibited decreased expression of IL-6 and alleviation of DVT, whereas antagomiR-338-5p aggravated the progression of DVT (Zhang et al., 2020a).

Zhou et al. reported that among the twelve miRNAs they studied, miR-28-3p showed a significant increase in the plasma from PE patients. However, the increase in miR-28-3p level due to PE development was not significant compared to its level in the experimental dogs before inducing PE. The AUC of the ROC of plasma miR-28-3p was 0.792 (95% CI: 0.689–0.896). Moreover, pathway analysis revealed that the effect of miR-28-3p involves inositol phosphate metabolism and the phosphatidylinositol signaling system, which are highly relevant pathways to PE pathogenesis (Zhou et al., 2016).

The circulating miRNAs as biomarkers of PE introduced above may not be comprehensive, and there should be many other potential miRNAs for predicting PE (Suzuki et al., 2016). Using only one miRNA as a biomarker may have limitations. In the future, it is possible to use a group of miRNAs to achieve the ideal sensitivity and specificity for PE biomarker assays.

A recent study on APE-PAH found that in APE-PAH patients, miR-340-5p was lowly expressed, whereas IL-1β and IL-6 were highly expressed. In a large sample cohort, linear regression was found negatively correlated between miR-340-5p and IL-1β/IL-6. In cell culture, upon miR-340-5p overexpression, the levels of IL-1β and IL-6 were reduced, coupled with decreased proliferation and migration of PASMCs as well as ameliorated inflammatory response. As is well known, IL-1 has been characterized to significantly induce IL-6 expression (Tosato and Jones, 1990). In the dual-luciferase reporter assay regarding miR-340-5p, IL-1β and IL-6, both pro-inflammatory targets were confirmed to be downregulated by miR-340-5p directly, revealing a synergistic anti-inflammation effect of miR-340-5p. Therefore, miR-340-5p might be a promising therapeutic target as an anti-inflammatory agent (Ou et al., 2020).

In an APE mouse model, Chen et al. demonstrated miR-106-5p as a novel regulator of PASMC proliferation and pulmonary vascular remodeling, and it performed its functions via targeting neuron-derived orphan receptor-1 (NOR1). Several literatures have reported NOR1 to regulate vascular cell migration and proliferation, and these important cellular processes are highly correlated with inflammation, growth factors, lipoproteins, and thrombin in vascular diseases (Martínez-González et al., 2003; Rius et al., 2004). Importantly, these findings highlight a novel therapeutic mechanism and improve the biological understanding of the complex network modulating APE progression (Chen et al., 2020).

Yang et al. reported that resveratrol successfully prevented PE-induced cardiac injury, possibly by modulating metastasis-associated lung carcinoma transcript 1 (MALAT1) expression through promoter regulation. MALAT1 was found to directly target miR-22-3p, as revealed by luciferase assay. Furthermore, miR-22-3p can bind complementarily to the 3′ UTR of NLR family pyrin domain containing 3 (NLRP3), leading to downregulation of NLRP3. In PE-associated cardiac injury, it was found that the levels of MALAT1, NLRP3, caspase-1, IL-1β, and IL-18 were significantly elevated, while miR-22-3p level was downregulated (Yang et al., 2019).

A very recent study reported that curcumin can improve lung edema and outcomes of APE rats and that it decreased mPAP and RVSP levels, wet/dry weight ratio, thrombus size, and inflammatory markers in lung tissues. The possible mechanism is that curcumin modulated the miR-21-Sp1-PTEN axis, which reduced the NF-κB signaling pathway, thus suppressing lung injury and inflammation in APE (Liang et al., 2021).

As is known, pulmonary embolism is highly related to matrix metalloproteinases (MMPs) and venous thrombus (VT). miRNAs that modulate MMP biogenesis/degradation balance or VT pathophysiology have been demonstrated to indirectly modulate outcome of APE (Goldhaber and Bounameaux, 2012; Neto-Neves et al., 2013b).

Aberrant MMP activity has been demonstrated to be associated with APE-induced hemodynamic dysregulation. Neto-Neves et al. reported that MMPs were lowly expressed in healthy lung tissues and that the expression and activity of MMPs were augmented in lung injury (Neto-Neves et al., 2013a). Doxycycline, an inhibitor of MMP activity, was found to improve APE prognosis and protect against RV enlargement in a rat model, possibly via amelioration of APE-induced oxidative stress and inhibition of ventricular proteolytic activity (Cau et al., 2013). In addition, l-arginine also attenuated APE-induced pulmonary hypertension through other mechanisms, including the modulation of NO synthesis and downregulation of MMP-2 and MMP-9 activities (Souza-Costa et al., 2005). Therefore, specific miRNAs that are capable of modulating MMP expression also have potential treatment or prediction value in PE pathogenesis. Zhang et al. elucidated that miR-15a-5p expression was significantly increased in the lung tissue of PAH rats and that artificial overexpression of miR-15a-5p led to increased expression of MMP-2 (Zhang et al., 2020b). Some studies found that inhibition or degradation of the corresponding mRNA of MMP-12 could be a solution to treat pathological lung tissue remodeling (Garbacki et al., 2009).

VT is a complex pathologic condition with a highly heritable genetic component, which predisposes an individual to its development. VT often induces the occurrence of thrombosis or embolism, including PE, in multiple sites in both veins and arteries (Sun et al., 2021). There are four miRNAs (hsa-miR-126-3p, hsa-miR-885-5p, hsa-miR-194-5p, and hsa-miR-192-5p) that have been studied and proved to be correlated with VT. These showed significant correlations with intermediate phenotypes of VT. It was found that multiple miRNAs were correlated with several coagulation-related factors: miR-885-5p and miR-195-5p with the level of protein S and FVII; miR-192-5p with FVII and ADAMS13; and miR-126-3p with factor XI (Rodriguez-Rius et al., 2020). miR-494 has been shown to downregulate the expression of PROS1 and protein S in liver Huh cells (Tay et al., 2013). Further, factor X deficiency was found to be associated with aberrant expression of miR-24. In a cohort of 15 healthy adults and 36 severe trauma-induced coagulopathy patients, plasma miRNA screening showed that all coagulopathy patients had elevated levels of miR-24 and lower levels of factor X than did the healthy volunteers (Chen et al., 2017). It is believed that the expression pattern of miRNAs that target coagulation factors can result in thrombosis and coagulation-related diseases (Jankowska et al., 2020).