2024 Volume 49 Issue 7 Pages 289-299
Background: Endothelial barrier dysfunction is critical for the pathogenesis of sepsis-induced acute lung injury (ALI). Lipopolysaccharide (LPS)-stimulated human pulmonary microvascular endothelial cells (HPMECs) are widely used as the cell model of sepsis-associated ALI for exploration of endothelial barrier dysfunction. Dickkopf (DKK) family proteins were reported to mediate endothelial functions in various diseases. The present study explored the effect of Dickkopf-3 (DKK3) on endothelial barrier permeability, angiogenesis, and tight junctions in LPS-stimulated HPMECs. Methods: RT-qPCR was required for detecting DKK3 and miR-98-3p expression. The angiogenesis of HPMECs was evaluated by tube formation assays. Monolayer permeability of HPMECs was examined by Transwell rhodamine assays. The protein expression of DKK3 and tight junctions in HPMECs was measured via western blotting. Luciferase reporter assay was used to verify the interaction between miR-98-3p and DKK3. Results: LPS treatment inhibited angiogenetic ability while increasing the permeability of HPMECs. DKK3 expression was upregulated while miR-98-3p level was reduced in LPS-treated HPMECs. DKK3 knockdown alleviated HPMEC injury triggered by LPS stimulation. MiR-98-3p targeted DKK3 in HPMECs. Overexpression of miR-98-3p protects HPMECs from the LPS-induced endothelial barrier dysfunction, and the protective effect was reversed by DKK3 overexpression. Conclusions: MiR-98-3p ameliorates LPS-evoked pulmonary microvascular endothelial barrier dysfunction in sepsis-associated ALI by targeting DKK3.
Acute lung injury (ALI) is characterized by reduced lung compliance, acute respiratory insufficiency with tachypnea, diffuse alveolar infiltration, and cyanosis refractory to oxygen (Mokrá, 2020). ALI is commonly defined as acute respiratory distress syndrome (ARDS) triggered by conditions such as pneumonia, trauma or sepsis (Millar et al., 2022). Dysfunction of the lung air-blood barrier in ALI/ARDS is critical for substantial respiratory function damage (Qiao et al., 2021). The blood-air barrier consists of the pulmonary microvascular endothelial barrier, the extracellular matrix and the pulmonary alveoli epithelial barrier, which is essential to ensure normal gas exchange between alveolar space and circulatory system (Qiao et al., 2021). According to previous studies, lipopolysaccharide (LPS)-stimulated human pulmonary microvascular endothelial cells (HPMECs) are commonly used as the ALI cell model (You and Zhang, 2022; Cen et al., 2021; Keskinidou et al., 2022; Xu and Zhou, 2020) since LPS can result in vascular endothelial barrier dysfunction and further lead to ALI (Zheng et al., 2018). Therefore, the cell model is also used in the present study. Endothelial hyper-permeability is a main feature in ALI (Zhang et al., 2018). The pulmonary endothelial permeability is mediated by intercellular junction proteins: gap junctions, tight junctions, and adherent junctions (Liu et al., 2019). Tight junctions help maintain normal functions of lung barrier by modulating signal transduction and intracellular fluid transportation (Englert et al., 2015). Therefore, exploring new methods to prevent endothelial hyper-permeability and loss of tight junctions in endothelial barrier dysfunction is necessary.
DKK (dickkopf) family proteins are important regulators of the Wnt/β-catenin signaling pathway, which play key roles in many essential biological processes (Baetta and Banfi, 2019). DKKs consist of four members, named DKK1, DKK2, DKK3 and DKK4 (Shao et al., 2017). DKKs have been reported to be related to endothelial cell activities. For example, DKK3 can directly differentiate human fibroblasts to functional endothelial cells (Chen et al., 2019). DKK1 leads to endothelial dysfunction by activating AKT (Dougherty et al., 2023); DKK1 knockdown in hypoxic-treated human pulmonary artery endothelial cells increases reactive oxygen species levels (Wang et al., 2022). DKK2 mediates juxtacrine signaling in pericyte-endothelial cell interactions, thereby promoting angiogenesis and rescuing erectile function of diabetic mice (Yin et al., 2018). Previously, DKK3 was verified to be upregulated in lung tissues of mice with sepsis-induced ALI (Zou et al., 2020), suggesting the potential role of DKK3 in ALI progression. In addition, DKK3 was reported to damage angiogenic ability of renal microvascular endothelial cells (Lipphardt et al., 2019) and disrupt blood-brain barrier in brain microvascular endothelial cells (You and Jiang, 2021). However, the role of DKK3 in LPS-stimulated HPMECs has not been reported yet.
MiRNAs are a type of small noncoding RNAs that can target and alter the expression of specific messenger RNAs (mRNAs) by promoting mRNA degradation or inhibiting mRNA translation (Mead and Tomarev, 2022). Evidence shows that miRNAs can function as mediators of endothelial regulation and damage by targeting mRNAs to modulate cell death, angiogenesis, inflammation, and autophagy, thereby participating in ALI development (Gu et al., 2024; Li et al., 2021a). For example, miR-450b-5p was downregulated in LPS-stimulated HPMECs and targeted HMGB1 to protect LPS-induced ALI by mitigating inflammation and apoptosis of HPMECs (Gong et al., 2021). MiR-539-5p exerts anti-inflammatory and anti-apoptotic properties in LPS-treated murine PMECs by targeting and restraining ROCK1 expression (Meng et al., 2019).
The current study explored the effect of DKK3 on endothelial barrier permeability, angiogenesis, and tight junctions in LPS-stimulated HPMECs and the underlying mechanism. This study may provide novel insight into the role of DKK3 and its upstream miRNA (miR)-98-3p in sepsis-induced ALI.
Potential miRNAs (miR-98-3p, miR-891b, miR-6873-5p, let-7a-3p, let-7f-1-3p and let-7b-3p) of DKK3 were identified by miRDB (http://www.mirdb.org/) with screening condition of target score ≥ 94. The binding site between miR-98-3p and DKK3 3’UTR was predicted by Targetscan (http://www.targetscan.org/).
Cell culture and treatmentHPMECs were purchased from ScienCell (San Diego, CA, USA) and incubated with Dulbecco's modified Eagle's medium (Gibco, Thermo Fisher Scientific, Waltham, USA) containing 10% fetal bovine serum (Gibco) and 1% streptomycin/penicillin (Gibco) at 37°C in 5% CO2.
LPS was dissolved in phosphate buffered saline (PBS) to prepare a stock solution (5 mg/mL) and stored at -20°C. For cell treatment, HPMECs were exposed to 10 μg/mL LPS (Sigma Aldrich, USA) for 24 hr (diluted in DMEM containing 10% fetal bovine saline) to establish the in vitro ALI models (Xu and Zhou, 2020).
Cell transfectionShort hairpin RNA targeting DKK3 (sh-DKK3) was used to silence DKK3, with sh-NC as a negative control. The miR-98-3p mimic was used to overexpress miR-98-3p with NC mimic as negative control. Full sequence of DKK3 was inserted into pcDNA3.1 vectors to amplify DKK3 expression, with empty vectors as the control. All plasmids were purchased from GenePharma (Shanghai, China). When the cell confluence reached 80%, a total of 40 nM sh-DKK3/sh-NC, 50 nM miR-98-3p mimic/NC mimic, and 30 nM pcDNA vectors were transfected into HPMECs at 37°C using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), and the transfection efficiency was examined using RT-qPCR after 48 hr.
Cell Counting Kit-8 (CCK-8) assayThe viability of HPMECs was evaluated using the CCK-8 assays. In brief, HPMECs (3,500 cells per well) were seeded in 96-well plates for 24 hr. Then, HPMECs with different treatments were incubated with 10 μL of CCK-8 reagent (Promega, USA) at 37°C for 2 hr. The optical density was measured with a microplate reader (Dynex Technologies, USA) at the wavelength of 450 nm.
Tube formation assayTube formation assay was performed to evaluate the angiogenesis of HPMECs. Briefly, HPMECs (1.5 × 105 cells/well) were seeded into 6-well plates and cultured at 37°C with 5% CO2 for 24 hr. The cells were then subjected to transfection and LPS or PBS treatment. After that, cells were cultured in 48-well plates (precoated with Matrigel, 3 × 104 cells/well) in serum-free medium (200 μL) at 37°C in 5% CO2 for 12 hr. Then, the formation of small tubes and nets were observed under a microscope (Olympus, Japan). To quantify the total tube length and the branch points, ImageJ software (Bethesda, MA, USA) was utilized.
Transwell rhodamine assayThe permeability of HPMECs with various treatments was measured using the Transwell rhodamine assay. HPMECs were cultured in Transwell inserts (Corning Incorporated, NY, USA; pore size: 0.4 μm) at the concentration of 1 × 104 cells/well in complete medium (100 μL) for 24 hr. The upper chamber was supplemented with rhodamine (40 μL; Sigma Aldrich) and 500 μL PBS was added to the lower chamber. After incubation at 37°C in 5% CO2 for 1 hr, fluids in the lower chamber were collected and a microplate reader (Thermo Fisher Scientific) was applied for measuring the absorbance at the wavelength of 554 nm.
RT-qPCRTotal RNA isolated from HPMECs using TRIzol reagent (Invitrogen) was reverse transcribed to complementary DNA using reverse transcription cDNA synthesis kit (Vazyme, Nanjing, China). SYBR Green PCR kit (Takara, Dalian, China) was then used for qPCR analysis on IQ5 real-time PCR system (Bio-Rad, MA, USA). The levels of miRNAs and DKK3 were calculated using the 2−ΔΔCt method (Lech and Anders, 2014) with normalization to U6 and GAPDH, respectively. The sequences of primers are presented in Table 1.
| Gene | Sequence (5’→3’) |
|---|---|
| miR-891b forward | ACACTCCAGCTGGGTGCAACTTACCTGAGT |
| miR-891b reverse | TGGTGTCGTGGAGTCG |
| miR-6873-5p forward | ACACTCCAGCTGGGCAGAGGGAATACAGAG |
| miR-6873-5p reverse | TGGTGTCGTGGAGTCG |
| let-7a-3p forward | ACACTCCAGCTGGGCTATACAATCTACTG |
| let-7a-3p reverse | TGGTGTCGTGGAGTCG |
| miR-98-3p forward | ACACTCCAGCTGGGCTATACAACTTACTAC |
| miR-98-3p reverse | TGGTGTCGTGGAGTCG |
| let-7b-3p forward | CCTCCACCCTATACAACCTACTGC |
| let-7b-3p reverse | CTCGACCCTGGCACCT |
| let-7f-1-3p forward | GCGCGCTATACAATCTATTGC |
| let-7f-1-3p reverse | AGTGCAGGGTCCGAGGTATT |
| DKK3 forward | ACACAGACACGAAGGTTGGA |
| DKK3 reverse | CGTCTCCCACAGATGTGATA |
| GAPDH forward | TGTTCGTCATGGGTGTGAAC |
| GAPDH reverse | ATGGCATGGACTGTGGTCAT |
| U6 forward | CTCGCTTCGGCAGCACA |
| U6 reverse | AACGCTTCACGAATTTGCGT |
Pierce BCA Assay Kit (ThermoFisher) was used to assess the concentration of total proteins isolated from HPMECs using RIPA lysis buffer (Beyotime). After being separated by 12% SDS-PAGE and transferred onto the PVDF membrane, the proteins were incubated with primary antibodies (Abcam, UK) at 4°C overnight and then with secondary antibody at room temperature for 2 hr. Primary antibodies include anti-DKK3 (ab187532; 1:1000), anti-Occludin (ab167161, 1:50,000 dilution), anti-ZO-1 (Abcam, ab96587, 1:2,000 dilution) and GAPDH (ab8245; 1:1000). Protein bands were visualized by BeyoECL Plus (Beyotime) and the signal intensity was analyzed by ImageJ software (NIH, USA).
Luciferase reporter assayThe binding site between miR-98-3p and DKK3 was predicted with Targetscan. Then, the binding sequence of DKK3 was mutated and inserted into the pmirGLO luciferase vectors (Promega), named as DKK3 3’UTR-Mut. Similarly, DKK3 3’UTR-Wt vectors were established by inserting the wild-type DKK3 sequence into pmirGLO vectors. Next, the established Mut or Wt vectors were transfected into HPMECs with miR-98-3p mimic/NC mimic using Lipofectamine 2000 (Invitrogen). After 48 hr, a Dual Luciferase Reporter Assay System (Lumi-glow, Suzhou, China) was utilized to assess the relative luciferase activity (firefly/Renilla).
Statistical analysisThe data of three independent experiments are presented as the mean ± standard deviation. Statistics were analyzed by GraphPad Prism software 6.0 (La Jolla, CA, USA). Student’s t-test was used for comparison between two groups, and one-way analysis of variance followed by Tukey post hoc analysis were utilized to compare significance among multiple groups. The value of p < 0.05 was considered statistically significant.
An in vitro cell model of ALI was established by stimulating HPMECs with LPS. The viability of HPMECs was markedly decreased after LPS treatment (52% ± 5.65) compared with that in the control group (100%), as suggested by CCK-8 assay (Fig. 1A). According to the results from tube formation assays, HPMECs treated with LPS showed fewer total branch points and shorter total tube length compared with the control group (Fig. 1B-D), indicating that LPS treatment impedes the angiogenesis of HPMECs. The dysfunction of endothelial barrier is an important hallmark of ALI progression and tight junction proteins (e.g. ZO-1 and Occludin) help maintain the integrity of air-blood barrier (Wang et al., 2021). As Fig. 1E shows, LPS stimulation decreased the protein expression of Occludin and ZO-1 in HPMECs. In addition, the endothelial cell monolayer permeability assay demonstrated that LPS treatment increased the monolayer permeability of HPMECs (Fig. 1F). All these data revealed that LPS triggers the damage of endothelial cell barrier and the sepsis-associated ALI cell model was established successfully. Moreover, DKK3 was also found to be highly expressed in LPS-stimulated HPMECs in contrast with its expression in the untreated HPMECs (Fig. 1G), which suggested that DKK3 may exert a promoting role in ALI development.
LPS induces endothelial barrier dysfunction in HPMECs and upregulates DKK3 expression. (A) The viability of HPMECs with or without LPS stimulation was evaluated using CCK-8 assay. (B-D) The angiogenesis of HPMECs with or without LPS stimulation was detected by tube formation assays (scar bar: 200 μm) with quantification of total branch points and total tube length. (E) Occludin and ZO-1 protein levels in HPMECs with or without LPS stimulation was detected by western blotting. (F) Transwell rhodamine assays were applied to measure the monolayer permeability of HPMECs in LPS or control group. (G) DKK3 level was examined by RT-qPCR in HPMECs with or without LPS stimulation. *p<0.05, **p<0.01.
To explore whether DKK3 contributes to LPS-induced endothelial cell dysfunction in HPMECs, the following functional experiments were performed. Transfection of sh-DKK3 into HPMECs caused significant decrease in mRNA and protein expression of DKK3 (Fig. 2A-B). CCK-8 assay revealed that DKK3 silencing markedly elevated HPMEC viability in the context of LPS (Fig. 2C). Endothelial barrier dysfunction can be caused by vascular endothelium damage in ALI, and the repair and regeneration of injured microvascular endothelial cells are essential for the maintenance of endothelial barrier function (Li et al., 2021b). Thus, we evaluated the influence of DKK3 on angiogenic ability of LPS-treated HPMECs via tube formation assay. Our data manifested that DKK3 depletion contributed to more total branch points and longer total tube length in LPS-stimulated HPMECs (Fig. 2D-F), suggesting that DKK3 deficiency promotes angiogenesis of HPMECs in ALI cell model. Additionally, DKK3 knockdown increased protein expression of tight junctions (Occludin and ZO-1) in LPS-stimulated HPMECs (Fig. 2G). DKK3 was then overexpressed in HPMECs via transfection of pcDNA3.1/DKK3 to explore the effect of DKK3 overexpression on protein levels of tight junctions. As shown by Figure 2H-I, DKK3 level was noticeably elevated in HPMECs of the DKK3 group, suggesting the successful transfection of pcDNA3.1/DKK3. Overexpressed DKK3 induced significant reduction of occludin and ZO-1 levels in LPS-treated endothelial cells (Fig. 2J). The above findings indicated that DKK3 expression was negatively related to protein levels of tight junctions in the ALI cell model. Moreover, DKK3 depletion mitigated the hyperpermeability of endothelial cells in response to LPS (Fig. 2K). The results demonstrate that DKK3 inhibition improves the endothelial cell function in the context of LPS by upregulating protein levels of tight junctions. In summary, DKK3 depletion is benefit to the alleviation of LPS-induced endothelial cell injury.
DKK3 knockdown alleviates LPS-induced endothelial cell dysfunction. (A-B) RT-qPCR and western blotting were conducted to evaluate the transfection efficiency of sh-DKK3. (C) CCK-8 assay was performed to measure the viability of LPS-treated HPMECs transfected with sh-DKK3 or sh-NC. (D) Tube formation assays were conducted to measure the angiogenesis of LPS-treated HPMECs with or without DKK3 inhibition. (E-F) Quantification of total branch points and total tube length in tube formation assays. (G) Western blotting was performed to measure Occludin and ZO-1 protein levels in LPS-treated HPMECs of sh-NC or sh-DKK3 group. (H-I) RT-qPCR and western blotting were performed to measure DKK3 expression in HPMECs transfected with pcDNA3.1/DKK3 or empty vector. (J) The influence of DKK3 overexpression on protein levels of tight junctions was evaluated by western blotting. (K) Transwell rhodamine assays were performed to measure monolayer permeability of LPS-treated HPMECs with or without DKK3 deficiency. *p<0.05, **p<0.01, ***p<0.001.
To figure out the underlying mechanism of DKK3 in mediating LPS-induced endothelial cell dysfunction, the upstream miRNA of DKK3 was investigated. Potential miRNAs were predicted with the bioinformatics tool miRDB, and the top six miRNAs (miR-891b, miR-6873-5p, let-7a-3p, miR-98-3p, let-7b-3p and let-7f-1-3p) with high target score were identified (Fig. 3A). Among these candidates, only miR-98-3p expression was significantly reduced by LPS stimulation in HPMECs, while the expression of other miRNAs was not altered by LPS (Fig. 3B). As revealed by RT-qPCR, transfection of miR-98-3p mimics into HPMECs resulted in marked elevation of miR-98-3p expression and significant reduction of DKK3 expression (Fig. 3C). The results indicated the negative correlation between miR-98-3p and DKK3 in HPMECs. Subsequently, the binding site between miR-98-3p and DKK3 3’UTR was predicted by Targetscan, and the binding sequence of DKK3 was mutated (Fig. 3D). Luciferase reporter assay confirmed the binding between miR-98-3p and DKK3, as evidenced by significant reduction of DKK3 3’UTR-Wt luciferase activity in response to miR-98-3p overexpression and no significant difference in the activity of DKK3 3’UTR-Mut between miR-NC and miR-98-3p mimics groups (Fig. 3E). Moreover, miR-98-3p overexpression decreased DKK3 protein level in HPMECs (Fig. 3F), further implying the negative correlation between DKK3 expression and miR-98-3p expression. In short, miR-98-3p targets DKK3 and inversely regulates DKK3 in HPMECs.
MiR-98-3p targets DKK3. (A) Potential upstream miRNAs were predicted with miRDB, and six top miRNAs (miR-98-3p miR-891b, miR-6873-5p, let-7a-3p, let-7f-1-3p and let-7b-3p) with high target score were identified and listed. (B) RT-qPCR was used to measure expression levels of the six miRNAs in HPMECs of control and LPS groups. (C) MiR-98-3p and DKK3 expression levels in HPMECs transfected with miR-98-3p mimics or NC mimics were quantified by RT-qPCR. (D) The binding site between miR-98-3p and DKK3 3’UTR was predicted from Targetscan, and the mutated DKK3 sequence was also provided. (E) Luciferase reporter assay was conducted to verify the interaction between miR-98-3p and DKK3. (F) DKK3 protein expression in HPMECs of miR-98-3p mimics group or miR-NC group was quantified by western blotting. **p<0.01.
In this section, the influence of miR-98-3p overexpression on angiogenesis and endothelial barrier breakdown in LPS-treated HPMECs was explored. Furthermore, rescue assays were carried out to verify whether miR-98-3p regulates endothelial cell dysfunction by inversely regulating DKK3. PCR and western blotting were conducted to measure the mRNA and protein levels of DKK3 in LPS-stimulated HPMECs transfected with miR-NC, miR-98-3p, or miR-98-3p + DKK3. As shown by Fig. 4A-B, DKK3 levels were greatly decreased in the miR-98-3p group, and the trend was successfully rescued by DKK3 overexpression. Tube formation assays revealed that miR-98-3p upregulation enhanced the angiogenetic ability of LPS-stimulated HPMECs, as evidenced by the increase in total branch point number and total tube length in response to miR-98-3p overexpression compared with those in the control group (Fig. 4C-E). However, the promoting role of miR-98-3p in angiogenesis was inhibited by DKK3 elevation (Fig. 4C-E). Moreover, the increase in Occludin and ZO-1 protein levels induced by overexpressed miR-98-3p in LPS-stimulated HPMECs was offset by DKK3 upregulation (Fig. 4F). Similarly, miR-98-3p elevation decreased the cell permeability in the context of LPS in contrast to the permeability in the miR-NC group, and the alteration was offset by DKK3 overexpression (Fig. 4G). The above results implied that miR-98-3p overexpression promotes angiogenesis and prevents endothelial barrier failure in LPS-treated HPMECs by downregulating DKK3 expression.
DKK3 overexpression reverses the protective effect of miR-98-3p upregulation on angiogenesis and endothelial barrier function in LPS-stimulated HPMECs. (A-B) RT-qPCR and western blotting were performed to measure mRNA and protein levels of DKK3 in LPS-treated HPMECs transfected with miR-NC, miR-98-3p, or miR-98-3p + DKK3. (C-E) Tube formation assays were conducted for measurement of the angiogenesis of LPS-treated HPMECs transfected with miR-NC, miR-98-3p or miR-98-3p+DKK3 (scar bar: 200 μm). (F) Western blotting was used to assess protein levels of tight junctions in miR-NC, miR-98-3p or miR-98-3p+DKK3 groups. (G) Transwell rhodamine assays were performed to determine the monolayer permeability of LPS-treated HPMECs in the abovementioned three groups. *p<0.05, **p<0.01, ***p<0.001, #p<0.05, ###p<0.001.
As a serious heterogenous pulmonary disorder, ALI is associated with high mortality (Mokra et al., 2019). Increased pulmonary vascular permeability induced by endothelial cell barrier dysfunction is a main pathological characteristic of ALI (Jiang et al., 2020). Tight junctions in endothelial cells can regulate intracellular signal transduction and are indispensable for normal pulmonary barrier (Smith et al., 2013). Destruction of tight junctions can lead to imbalance of lung permeability, subsequently resulting in lung injury (Yang et al., 2017). LPS, as a main component of the outer membrane of Gram-negative bacteria, can evoke inflammatory cascades, the necrosis, and apoptotic behaviors of endothelial cells (Violi et al., 2023). In this study, the viability and angiogenesis of HPMECs was inhibited in response to LPS stimulation. Additionally, protein levels of tight junctions were reduced in LPS-treated HPMECs, and the monolayer permeability of HPMECs was increased after LPS treatment. The findings reveal that LPS induces endothelial barrier dysfunction, suggesting the successful establishment of ALI cell model.
The study first revealed the role of DKK3 in LPS-induced endothelial barrier dysfunction. To be specific, DKK3 expression was elevated in LPS-treated HPMECs, and DKK3 depletion enhanced HPMEC viability and angiogenesis in the context of LPS. In line with our findings, Mark Lipphardt et al. reported that DKK3 could impair angiogenic ability of renal microvascular endothelial cells and induce endothelial-mesenchymal transition (Lipphardt et al., 2019). Moreover, DKK3 in this study was shown to increase the protein levels of tight junctions (occludin and ZO-1) while reducing the monolayer permeability of LPS-treated HPMECs. All the findings indicated that DKK3 downregulation is beneficial for maintaining endothelial barrier integrity in the sepsis-associated ALI cell model. Consistent with the results of our study, DKK3 expression is significantly increased in LPS-stimulated human brain microvascular endothelial cells and DKK3 disrupts the blood-brain barrier by suppressing the level of tight junction protein ZO-1 in LPS-induced encephalopathy (You and Jiang, 2021). In a recent study, LPS-stimulated WI-38 and MRC-5 cells were used to induce inflammatory injury of neonatal pneumonia in vitro. Different from the findings in this study, DKK3 has been shown to be downregulated in LPS-treated human WI-38 and MRC-5 cells and overexpression of DKK3 suppressed LPS-induced cell apoptosis and production of proinflammatory factors (Kang et al., 2023). In the abstract of the recent study, it states that DKK3 overexpression decreased LPS-induced inhibition of cell viability and reduced LPS-induced apoptosis of WI-38 and MRC-5 cells, and the sentence seems controversial, possibly because of a clerical error. The reason for the difference between our finding and theirs might be complex and requires more verification. The anti-inflammatory role of DKK3 is also mentioned in neuropathic pain (Zhang et al., 2022). The current study did not explore the impact of DKK3 on inflammatory response in LPS-stimulated ALI, which can become the direction for future work.
MiRNAs have 18-28 nucleotides in length and bind to the 3’UTR of their target mRNAs to negatively modulate gene levels (Correia de Sousa et al., 2019). In the present study, miR-98-3p expression was downregulated in LPS-stimulated HPMECs. Different from the present study, miR-98-3p has been reported to be upregulated in the plasma of patients with ARDS (n=8) compared with that in the healthy controls (n=10) (Parzibut et al., 2021). No more articles focusing on the role of miR-98-3p in ALI were reported. The current work revealed that miR-98-3p upregulation promoted angiogenesis of LPS-treated HPMECs. Additionally, the overexpression of miR-98-3p increased protein levels of tight junctions and decreased monolayer permeability of LPS-stimulated HPMECs. However, all these alterations exerted by miR-98-3p were reversed by DKK3 overexpression. Overall, miR-98-3p improves endothelial barrier dysfunction by negatively regulating DKK3 in LPS-induced ALI.
Despite DKK3, we also consider whether miR-98-3p can regulate other DKK genes in ALI. According to previous studies, DKK2 is downregulated in non-small cell lung carcinoma and lung adenocarcinoma (Zhang et al., 2023; Song et al., 2023). According to the bioinformatics database TargetScan, miR-98-3p also have potential binding site with DKK2 3’UTR. However, the binding area between miR-98-3p and DKK2 is not conserved among species such as mouse and rat. Therefore, it may bring trouble for deep investigation of miR-98-3p and DKK2 in animals. That is the reason why the present study did not include DKK2-related experiments.
The mechanisms underlying miR-98-3p downregulation in response to LPS stimulation were not explored using experiments in the present study, which is one of limitations of the study. Based on bioinformatics analysis and available literature, we suspected that four long noncoding RNAs (lncRNAs) might be upstream factors of miR-98-3p and may be responsible for the downregulation of miR-98-3p in LPS-stimulated HPMECs. According to the competing endogenous RNA (ceRNA) hypothesis, a certain lncRNA competes with mRNA for the chance of interacting with miRNA, and in this way the lncRNA upregulates mRNA expression by interacting with the miRNA (Shen et al., 2023). The four lncRNAs all have potential binding site with miR-98-3p according to the bioinformatics tool Targetscan. Specifically, lncRNA NORAD has been reported to be markedly induced in LPS-stimulated HPMECs (Zhou et al., 2022), which is highly consistent with the logic of the present study and NORAD might interact with miR-98-3p and thus upregulate the expression of DKK3. Moreover, lncRNA KCNQ1OT1 and NEAT1 were shown to promote LPS-induced ALI via the miR-7a-5p/Rtn3 axis (Yang et al., 2022; Chen et al., 2021), and whether KCNQ1OT1 or NEAT1 may mediate multiple ceRNA networks in ALI require more investigations. Similarly, lncRNA MALAT1 has been demonstrated to be a risk factor for sepsis-related ALI and lung ischemia/reperfusion injury (Li et al., 2023; Zhang et al., 2021). In the future, experiments focusing on these lncRNA candidates and competing endogenous RNA (ceRNA) shall be conducted.
Another limitation of the study is that in vivo experiments were not performed to further support the current findings. Since the binding site between miR-98-3p and DKK3 is conserved among species according to the Targetscan database, animal experiments can be conducted in the future.
In conclusion, the study reveals that miR-98-3p inhibits LPS-induced pulmonary microvascular endothelial cell dysfunction by targeting DKK3 in sepsis-associated acute lung injury. The findings reveal a novel miRNA/target axis in ALI, provide insight into the complex pathogenesis of ALI, and might be useful for the targeted therapy in sepsis-associated ALI.
Conflict of interestThe authors declare that there is no conflict of interest.
