Introduction

miR-21-5p has emerged as a pivotal anti-apoptotic regulator implicated in a variety of pathological conditions, including cancer1, cardiovascular diseases2, and inflammatory disorders3. Its critical role lies in targeting pro-apoptotic pathways to enhance cellular survival and mitigate apoptosis induced by various stressors4. Extensive research has highlighted miR-21-5p’s protective functions in different tissues and contexts, particularly in the suppression of apoptosis and promotion of tissue repair5,6. In hyperoxia-induced acute lung injury (HALI), miR-21-5p has been identified as a significant protective factor, reducing alveolar epithelial cells (AECs) apoptosis and alleviating lung injury5,7. However, the mechanisms governing its expression, particularly during pathological conditions like HALI, remain poorly understood.

While miRNAs are well-known for their ability to regulate gene expression at the post-transcriptional level8, less is known about the upstream factors that regulate their transcription9,10. Research has primarily focused on the downstream effects of miRNAs, often overlooking the molecular mechanisms that control their expression in response to environmental or pathological stimuli11. This gap in understanding is significant, as changes in miRNA expression are often associated with disease progression and therapeutic resistance12. Exploring the upstream regulatory mechanisms, including factors that influence miRNA transcription, is crucial for understanding how their expression is modulated in disease contexts such as HALI. Identifying these mechanisms can provide insights into the differential expression of miRNAs and their roles in pathological conditions. A lack of understanding regarding the upstream regulatory mechanisms of miRNAs leaves our view of their roles in disease pathogenesis incomplete and limits their translational potential for therapeutic intervention. Therefore, elucidating the transcriptional regulation of disease-associated, differentially expressed miRNAs is essential to fully understand their biological function and clinical relevance. In recent years, increasing attention has been paid to this aspect, and several miRNA-transcription factor (TF) databases have been developed to systematically identify potential upstream regulators. In this study, we integrated multiple TF-miRNA databases to predict candidate regulators of miR-21-5p and identified Gata3 as the only conserved transcription factor shared across all databases and between humans and mice, highlighting its potential role as a key upstream regulator in the miR-21-5p pathway.

Transcription factors are among the key regulators of gene expression, including miRNAs, as they integrate environmental signals and cellular stress responses to modulate transcriptional activity. Among the many transcription factors studied, Gata3 has been well-documented for its roles in T-cell differentiation13, immune modulation14, and epithelial cell biology15. Gata3’s involvement in processes such as cellular differentiation16 and stress adaptation17 has made it a compelling candidate for investigating its potential regulatory effects on miRNAs. However, although Gata3 has been reported to regulate microRNAs in various contexts18, its regulatory role in miRNA expression under hyperoxic stress or lung injury conditions has remained unreported.

Based on this prediction, our study focused on the potential regulatory relationship between Gata3 and miR-21-5p. We conducted experimental investigations to determine whether Gata3 directly regulates miR-21-5p transcription and how this interaction influences epithelial cell survival and the outcomes of HALI. By exploring this regulatory axis, our research provides new insights into the upstream factors influencing miRNA expression and highlights the potential for targeting transcriptional regulators to modulate miRNA activity as a therapeutic strategy for HALI and other lung injuries.

Materials and methods

Materials and reagents

Cell culture materials were obtained from NEST, and the hyperoxia chamber was supplied by Heal Force. A custom-made animal hyperbaric chamber was provided by Puhe Biotechnology. 1.25% tribromoethanol, purchased from Nanjing Aibei Biotechnology Company. The Annexin V-FITC/PI kit was sourced from BD Corporation. The BCA protein quantification kit and hematoxylin-eosin (HE) staining reagents were purchased from Solarbio. Antibodies for Cleaved-caspase3, Caspase-3 were obtained from CST, while Bax, Bcl-2, and β-actin antibodies were supplied by Huabio. Chromatin immunoprecipitation (ChIP)-grade IgG was provided by Zenbio. Inhibitors and Gata3 plasmids/adeno-associated viruses were procured from GenePharma. Lipofectamine 3000 was acquired from Thermo Fisher Scientific. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) reagents were sourced from Takara, and dual-luciferase assay kits and Western Blotting (WB) gel kits were supplied by UE. The ChIP kit was obtained from ABclonal.

Cell culture and treatment

TC-1 mouse AECs, obtained from Shanghai Fuheng Biological, were cultured in 1640 medium supplemented with 10% serum. Transfections were performed following the Lipofectamine 3000 protocol, using 7.5 µL of inhibitor and 3 µg of plasmid per reaction. Hyperoxia exposure was conducted by incubating cells in a chamber maintained at 85% oxygen and 5% carbon dioxide at 37 °C.

Animal and animal model

Male C57 wild-type (WT) mice (6–8 weeks old) were purchased from Chongqing Ensiweier Company (Animal License No. SCXK (Xiang) 2021-0002). miR-21-5p KO mice were created by Cyagen Biosciences (Animal License No. SCXK(Su)2022-0016). Gata3 overexpression in lung tissue was induced via tracheal instillation of adeno-associated virus (SPC-GFP-Gata3-AAV, 2.85 × 1013 VG/ml), followed by a one-month feeding protocol. HALI was established by exposing mice to 90% oxygen in a hyperoxia chamber for 72 h. Mice were anesthetized with a mouse-specific anesthetic, tribromoethanol, by intraperitoneal injection of 400 µl prior to the tracheal drip, and the drip was administered after satisfactory anesthesia. At the end of modeling, lung tissues were extracted from mice euthanized by intraperitoneal injection of 800 µl of excess tribromoethanol. Animal procedures were approved by the Institutional Animal Care and Use Committee of Zunyi Medical University and carried out following institutional and national animal welfare regulations. Additionally, all animal experiments adhered to the ARRIVE guidelines (https://arriveguidelines.org) for reporting animal research.

WB

Cells or tissue were lysed with efficient RIPA buffer (20 mg, 200 µL), followed by centrifugation at 12,000 g for 15 min at 4 °C. The supernatant was collected for protein quantification using the BCA assay. The protein samples were then denatured by boiling at 100 °C for 7 min. SDS-PAGE was conducted to separate proteins (20–30 µg per well, 220 V for 45 min). The proteins were transferred onto a membrane (400 mA for 30 min). The membrane was incubated overnight at 4 °C with the primary antibody, followed by a 1-h incubation at room temperature with the secondary antibody. ECL chemiluminescence was used for protein visualization through exposure and development.

RT-qPCR

RT-qPCR was performed using SYBR Green Master Mix with gene-specific primers for cDNA amplification. The protocol included an initial denaturation at 95 °C for 30 s, followed by 40 cycles of 95 °C for 15 s, 65 °C for 10 s, and 72 °C for 30 s. Gene expression was quantified using the 2−ΔΔCt method.

ChIP

2 × 107 cells were crosslinked with 1% formaldehyde at room temperature for 15 min. Chromatin was fragmented by sonication (50 W for 5 min) and incubated with 5 µg of the target antibody at 4 °C for 3 h. Magnetic beads were added for immunoprecipitation at 4 °C for 2 h. The chromatin was then eluted, crosslinks were reversed, and DNA was purified using a centrifugation column. qPCR was performed to quantify the results.

Flow cytometry detection of apoptosis

Cells were cultured in 6-well plates and treated as required. After digestion with 0.25% trypsin (no EDTA), cells were centrifuged and the supernatant discarded. The cells were resuspended in 1×Binding Buffer at a concentration of approximately 1 × 105 cells/100 µL. To each suspension, 5 µL of FITC-Annexin V was added and incubated at room temperature in the dark for 15 min. After adding 200 µL of 1×Binding Buffer and 5 µL of PI, apoptosis was analyzed by flow cytometry.

Immunofluorescence

Paraffin sections were deparaffinized, rehydrated, and treated with 3% H₂O₂ (100 µL) at room temperature for 10 min to block endogenous peroxidase, followed by PBS washes. Antigen retrieval was performed with Tris-EDTA buffer (1 mM, pH 9.0) for 30 min. After blocking with 5% BSA for 20 min, sections were incubated with primary antibodies overnight at 4 °C and with secondary antibodies at 37 °C for 1 h, with PBS washes before and after each step. Tyramide signal amplification (Tyr-549, 100 µL) was applied for 10–30 min at room temperature. Finally, sections were counterstained with DAPI-containing antifade mounting medium and imaged.

HE

Paraffin-embedded sections were deparaffinized, rehydrated, and stained with hematoxylin for 3–5 min, followed by rinsing in running water. Sections were differentiated in 1% acid alcohol for 5 s, briefly rinsed, then blued in 1% ammonia water for 1 min. After a brief rinse, eosin staining was performed for 1–3 min. Following a final wash, staining was evaluated microscopically. Sections were then dehydrated through graded ethanol, cleared in xylene for 5 min, and mounted with neutral resin for light microscopy imaging. The assessment of tissue injury was based on previously published lung injury scoring systems19. Briefly, the score was calculated by evaluating the severity of specific histopathological features in each random microscopic field, including neutrophil infiltration in the alveolar space and interstitium, presence of hyaline membranes, proteinaceous debris, and thickening of the alveolar septa.

Rapid frozen sectioning

Tissue samples were embedded in OCT compound, rapidly frozen with liquid nitrogen, and sectioned at 5–10 μm using a cryostat at − 20 °C to − 25 °C. Sections were mounted on slides, air-dried, and stored at − 80 °C. For fluorescence detection, sections were stained with DAPI, incubated briefly in the dark, and imaged using 488 nm excitation and 510 nm emission filters. This method preserved tissue integrity for further analyses.

Statistical methods

Statistical analysis was performed using SPSS 29.0, image quantification with ImageJ 1.8.0, and graphing with GraphPad Prism 10.3.0. Data with normal distribution were assessed using the Kolmogorov-Smirnov test and are expressed as mean ± SD. Comparisons between two groups were made with the independent samples t-test, and for multiple groups, one-way ANOVA with post hoc tests (LSD, SNK, or Dunnett’s T3) was used. Statistical significance was defined as *P < 0.05, **P < 0.01, ***P < 0.001.

Results

Predicted upstream transcriptional regulator of miR-21-5p and its correlation with the expression of apoptosis-related genes

In previous studies, we observed that hyperoxia reduced miR-21-5p expression in AECs, contributing to apoptosis and lung injury7,20. However, the mechanisms behind the downregulation of miR-21-5p remained unclear. To investigate potential upstream regulatory factors, we examined miRNA transcription factor databases, which indicated that miR-21-5p might be regulated by the transcription factor Gata3 in both human and mouse (Fig. 1A). Analysis of the 3500 bp promoter region upstream of the transcription start site revealed the absence of CpG islands (Fig. 1B), suggesting that miR-21-5p expression was likely regulated by transcriptional, rather than epigenetic, mechanisms. Additionally, a search using the JASPAR database identified a binding site on the promoter with a high sequence similarity to the Gata3 motif (Fig. 1C), further supporting Gata3 as a potential transcriptional regulator of miR-21-5p. To investigate the relationship between Gata3 and apoptosis,

We queried the ChIPBase database and found that in human lung tissues, Gata3 expression was negatively correlated with the pro-apoptotic gene Caspase-3, and also showed a negative correlation with Bax, although the difference was not statistically significant, possibly due to limited sample size. In contrast, Gata3 expression was positively correlated with the anti-apoptotic gene Bcl-2. These results suggest that Gata3 may play a role in inhibiting apoptosis (Fig. 1D).

Fig. 1
figure 1

Prediction of upstream transcriptional regulatory mechanisms of miR-21-5p. (A) Venn diagram of overlapping transcription factors predicted by multiple databases. (B) Schematic of the miR‑21‑5p promoter showing the transcription start site, predicted transcription factor binding sites, and CpG island distribution based on MethPrimer analysis. (C) JASPAR-predicted Gata3 binding sites and motifs. (D) ChIPBase analysis of Gata3 expression correlation with apoptotic genes in human lung.

Full size image

HALI sequencing data confirmed Gata3 expression trends and functional enrichment

Using the GEO dataset GSE212284, we performed a differential expression analysis through the AnalystNetwork platform, which indicated a significant downregulation of Gata3 expression. This trend was consistent with the observed downregulation of miR-21-5p expression7 (Fig. 2A,B). GSEA enrichment analysis revealed that both upregulated and downregulated genes were involved in critical signaling pathways, including TNF-α, JAK-STAT, PI3K-Akt, P53, and microRNA regulation (Fig. 2C). GO enrichment analysis further identified that the downregulated genes were mainly associated with transcriptional repression of DNA, DNA catalytic activities, and binding to damaged DNA (Fig. 2D). Additionally, these downregulated genes were linked to biological processes such as microtubule dynamics, protein phosphorylation, and apoptosis (Fig. 2E). The GO enrichment analysis suggests that hyperoxia induces apoptosis in lung tissue, which is consistent with the findings of most previous studies21,22.

Fig. 2
figure 2

Differential analysis and enrichment analysis of sequencing data. (A) Principal component analysis of GSE212284. (B) Volcano plot depicting differential expression analysis. (C) GSEA enrichment analysis of differentially expressed genes. (D) GO enrichment analysis in the molecular function category. (E) GO enrichment analysis in the biological process category.

Full size image

Gata3 expression validated in HALI cellular and animal models

To confirm the differential expression of Gata3 in HALI, we established cellular and animal models, following previously reported methods23,24. RT-qPCR (Fig. 3A,C) and WB (Fig. 3B,D) analyses confirmed a significant reduction in Gata3 expression, consistent with GEO dataset sequencing results showing suppression and aligning with prior studies indicating a downregulation trend in miR-21-5p expression20.

Fig. 3
figure 3

Gata3 expression in HALI. (A) Bar graph showing differential Gata3 RNA expression in the cellular HALI model. (B) WB bands and bar graph showing differential expression of Gata3 in the cellular HALI model. (C) Bar graph showing differential Gata3 RNA expression in the HALI animal model. (D) WB bands and bar graph showing differential Gata3 protein expression in the HALI animal model. *P < 0.05, **P < 0.01, ***P < 0.001.

Full size image

Gata3 promoted miR-21-5p expression to inhibit apoptosis in AECs

After successful overexpression of Gata3 in TC-1 cells, as confirmed by RT-qPCR and WB (Fig. 4A–C), RT-qPCR analysis showed a significant upregulation of miR-21-5p expression (Fig. 4D). ChIP assays further demonstrated that Gata3 enhanced the transcription of miR-21-5p (Fig. 4E). Additional rescue experiments showed that Gata3 prevented hyperoxia-induced downregulation of miR-21-5p, and miR-21-5p knockdown partially restored its expression (Fig. 4F,G). WB and flow cytometry analyses revealed that Gata3 overexpression inhibited hyperoxia-induced apoptosis in AECs, with miR-21-5p knockdown partially reversing the anti-apoptotic effect (Fig. 4H–K).

Fig. 4
figure 4

Gata3 promotes miR-21-5p expression to reduce AECs apoptosis. (A) Bar graph of Gata3 RNA overexpression efficiency. (B) WB analysis of Gata3 protein bands. (C) Statistical analysis bar chart of protein band in (B). (D) Bar graph of miR-21-5p expression after Gata3 overexpression. (E) ChIP analysis bar graph. (F) Bar graph of miR-21-5p knockdown efficiency using inhibitor. (G) Statistical bar chart of miR-21-5p expression in Gata3 and inhibitor overexpression experiments. (H) WB bands of apoptosis-related proteins in Gata3 and inhibitor overexpression experiments. (I) Statistical analysis bar chart of protein band in (H). (J) Flow cytometry analysis of apoptosis rate in TC-1 cells. (K) Statistical bar chart of apoptosis rate from flow cytometry analysis. *P < 0.05, **P < 0.01, ***P < 0.001.

Full size image

Gata3 promoted miR-21-5p expression in WT mouse aecs, but not in KO mice, reducing apoptosis and alleviating HALI

WT mice were intratracheally administered AAV to overexpress Gata3 (Fig. 5A–D), leading to a significant increase in miR-21-5p expression (Fig. 5E). RT-qPCR analysis showed that Gata3 inhibited the hyperoxia-induced downregulation of miR-21-5p in WT mice AECs, but had no effect on miR-21-5p expression in KO mice (Fig. 5F). WB and immunofluorescence confirmed that Gata3 suppressed hyperoxia-induced apoptosis in AECs, although the effect was less pronounced in KO mice (Fig. 5G–J). HE staining showed that Gata3 overexpression improved lung injury in WT mice under hyperoxia, whereas KO mice exhibited severe damage, with no significant improvement in injury following Gata3 overexpression (Fig. 5I,K).

Fig. 5
figure 5figure 5

Gata3 alleviates AECs apoptosis and lung injury in WT but not in KO mice. (A) Fluorescence images showing SPC-GFP-Gata3 adenovirus infection efficiency via tracheal instillation. (B) Bar graph of Gata3 RNA overexpression. (C) WB showing Gata3 protein overexpression. (D) Statistical bar graph of protein band intensity in (C). (E) Bar graph of miR-21-5p expression after Gata3 overexpression. (F) miR-21-5p expression in WT and KO mice after Gata3 overexpression and HALI modeling. (G) WB showing apoptosis-related protein expression in WT and KO mice following Gata3 overexpression and HALI modeling. (H) Statistical analysis of apoptosis-related protein expression in (G). (I) Fluorescence images of Cleaved-caspase3 expression and HE staining in lung tissue from WT and KO mice after Gata3 overexpression and HALI modeling. (J) Statistical bar graph of Cleaved-caspase3 fluorescence intensity in (I). (K) Statistical bar graph of lung tissue injury in (I). (A) scale bar = 1000 μm; (I) scale bar = 100 μm; *P < 0.05, **P < 0.01, ***P < 0.001.

Full size image

Discussion

This study provided groundbreaking insights by establishing, for the first time, that Gata3 upregulates miR-21-5p expression, thereby playing a critical role in inhibiting AECs apoptosis and alleviating HALI. These findings significantly extended the known biological functions of Gata3 beyond its well-characterized roles in immune regulation, including T-cell differentiation and Th2 cytokine production14,25. While Gata3 has long been recognized for its ability to regulate immune responses, its involvement in epithelial cell survival and tissue repair, particularly under hyperoxic stress, remained poorly understood. This study bridged this gap by revealing a novel Gata3-mediated mechanism that enhances epithelial cell resilience and lung repair, which is consistent with previous studies15,26, and highlights its potential as a therapeutic target.

The identification of Gata3 as a regulator of miR-21-5p introduces a previously unrecognized transcriptional mechanism underlying miRNA modulation in HALI. miR-21-5p is well-documented for its anti-apoptotic and pro-survival roles in tissue injury and repair, and has already entered phase II clinical research12; yet, research has primarily focused on its downstream effects rather than the upstream regulators that control its expression27,28. This oversight has left a critical gap in our understanding of the transcriptional machinery governing miRNA dynamics. Our findings address this gap by demonstrating that Gata3 directly influences miR-21-5p transcription, providing evidence of a transcription factor’s pivotal role in modulating miRNA expression during pathological processes. Notably, ChIP assays confirmed that Gata3 directly binds to the miR-21-5p promoter to promote its transcription. The absence of a predicted CpG island in this region suggests that Gata3 activation occurs through a CpG-independent mechanism, potentially involving alternative transcriptional pathways29,30. This raises the intriguing possibility that Gata3 acts as a stress-responsive transcription factor, initiating miR-21-5p expression under specific pathological conditions, such as hyperoxia-induced lung injury. This aligns with previous findings indicating that miR-21-5p has its own promoter and that its expression may be subject to context-dependent regulation31,32. Alternatively, this regulation may be tissue-specific, with distinct epigenetic landscapes across cell types influencing transcription factor accessibility and activity. A deeper understanding of these mechanisms could provide valuable insights into how transcriptional control adapts to environmental stress and disease states, paving the way for novel therapeutic strategies. This insight not only expands our understanding of miRNA regulation but also underscores the necessity of exploring transcription factor-miRNA interactions as potential mediators of disease progression and recovery.

The Gata3/miR-21-5p axis has emerged as a promising area of investigation, with potential implications in various injury and disease contexts. Conditions such as chronic obstructive pulmonary disease33,34, acute respiratory distress syndrome35,36, and fibrotic lung diseases37,38 are all associated with dysregulated cell survival and tissue remodeling. Although our findings suggest that modulating this axis could provide dual benefits by attenuating apoptosis and promoting repair, further researches are needed to confirm its relevance and therapeutic potential in these specific diseases. Future studies should focus on validating the Gata3/miR-21-5p axis in these contexts and evaluating its efficacy in preclinical and clinical settings before definitive conclusions can be drawn.

However, this study also opens new questions that warrant further exploration. While we demonstrated the regulatory relationship between Gata3 and miR-21-5p, the precise molecular dynamics—including the binding sites and co-factors involved in Gata3-mediated transcriptional activation of miR-21-5p—remain to be elucidated. In this study, we confirmed that Gata3 can directly target the promoter region of miR-21-5p and markedly enhance its transcription. However, multiple potential binding sites were predicted, and mutation of one such site failed to fully abolish the transcriptional activation effect (data not shown). This is not unexpected, as the existence of multiple binding sites is a common regulatory feature and may serve as a protective mechanism to ensure transcriptional robustness. Further studies are required to identify which of these sites are most functionally relevant.

This study focused exclusively on the regulation of AECs injury and repair within the experimental model, without addressing therapeutic interventions targeting inflammatory cell infiltration. Clearly, given the complex pathogenesis of lung injury, a more comprehensive treatment strategy would be preferable and more aligned with clinical practice. Moreover, we did not specifically assess inflammatory responses but instead concentrated on the impact of AECs apoptosis on lung tissue damage. By investigating the transcriptional regulation of the anti-apoptotic gene miR-21-5p, we found that enhancing its expression could ameliorate structural lung injury. This protective effect may be partly attributed to miR-21-5p’s ability to target key pro-death mediators such as PDCD4 and PTEN, thereby suppressing the NF-κB signaling pathway and indirectly reducing inflammation39,40. However, whether Gata3 overexpression itself can attenuate pulmonary inflammation requires further investigation.

Additionally, the broader applicability of this mechanism across other tissues and disease models needs to be investigated to fully understand its therapeutic potential.

Overall, this research not only deepens the understanding of miRNA transcriptional regulation but also highlights the broader importance of transcription factors in shaping cellular responses to stress and injury. As a key anti-apoptotic gene, miR-21-5p plays a critical role in protecting against AECs apoptosis. Our findings demonstrate that Gata3 can reproduce this protective effect by promoting miR-21-5p expression, thereby reinforcing its potential as a therapeutic target. Enhancing endogenous miR-21-5p expression, possibly in combination with exogenous supplementation, may offer an even greater protective effect on lung tissue. By focusing on the upstream regulation of miRNAs, our study sets a precedent for future investigations into transcription factor–miRNA networks, potentially unlocking new therapeutic avenues for a range of pathological conditions.

Conclusion

In conclusion, our study reveals that Gata3 promotes the transcription of miR-21-5p, thereby inhibiting AECs apoptosis and alleviating HALI. This discovery uncovers a novel regulatory mechanism and underscores the essential role of transcription factors in regulating miRNA expression during pathological processes. While much research has focused on the downstream effects of miRNAs, the exploration of their upstream transcriptional regulators has been limited. Our findings bridge this gap by identifying the Gata3/miR-21-5p axis as a critical pathway in lung protection and repair, offering new insights into the interaction between transcription factors and miRNAs. These results not only advance our understanding of miRNA regulation but also suggest a promising therapeutic target for HALI and potentially other lung injuries. Further research is needed to assess the broader applicability and long-term clinical implications of targeting this pathway.