Hypoxia LUAD H1975 cell-derived exosomal miR-671-3p promotes angiogenesis via regulating KLF2-VEGFR2 axis

Abstract

For solid tumors, hypoxia is associated with disease aggressiveness and poor outcomes. In addition to undergoing broad intracellular molecular and metabolic adaptations, hypoxic tumor cells extensively communicate with their microenvironments to facilitate conditions favorable for their survival, growth, and metastasis. This communication is mediated by diverse secretory factors, including exosomes (extracellular vesicles of endosomal origin). Exosomal cargo is altered considerably by hypoxia, with significant impacts on tumor-cell communication with both local and distant microenvironments. Exosomes released by cancer cells influence the tumor environment to accelerate metastasis. While tumor-derived exosomes have been identified as a major driver of premetastatic niche formation at distant sites, this mechanism in lung adenocarcinoma (LUAD) remains unclear. We found that miR-671-3p in exosomes derived from H1975 under hypoxic conditions target Krüppel-like factor 2 (KLF2) to regulate VEGFR2 expression in endothelial cells to promote angiogenesis. In addition, miR-671-3p is expressed at high levels in circulating exosomes isolated from patients with LUAD. Our study suggests that exosome miR-671-3p is involved in the formation of premetastatic niche and may serve as a blood-based biomarker for LUAD metastasis.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-97488-0.

Introduction

Premetastasis to niche formation facilitates the establishment of a supportive microenvironment for cancer cell dissemination through extracellular secretion. Its characteristic features include inflammation, immunosuppression, angiogenesis, vascular permeability, lymph angiogenesis, organotropism, and reprogramming of the primary tumor site prior to widespread metastasis^1–5. Several premetastatic niche biomarkers relevant to cancer diagnosis and prognosis prediction have been identified^6–8. It could be a promising strategy. Targeting the premetastatic niche is an attractive therapeutic tactic^9–11. Hypoxia is one of the most solid tumors dynamics and heterogeneity, associated with disease progression, overall survival and other pathogenic events¹². Solid tumor proliferation occurs in varying oxygen concentrations, due to the limitations of oxygen diffusion, resulting in the anatomical and physiological abnormalities associated with disordered tumor vasculature developing in normoxic and hypoxic regions¹³. In the absence of effective vasculature, tumor cells proliferate rapidly, leading to a depletion of nutrients and oxygen, producing a hypoxic zone (partial pressure of oxygen [PO₂] < 10 mmHg) around the tumor¹⁴. Cells supplied by normal and functional blood vessels are viable, while proliferating cells suited within 150 μm of the blood vessels may become necrotic; cells between these zones adjust to survive at different low O₂ levels (hypoxic regions)¹⁵. Therefore, by integrating existing blood vessels or inducing angiogenesis to form capillaries, tumor cells induce the formation and establishment of new blood vessels uptake oxygen and nutrients; invasion and metastasis are the primary causes of cancer-related death^16–18. Thus, increased angiogenesis in the hypoxic tumor microenvironment (TME) is a vital step toward metastasis. The identification of biomarkers associated with niche formation before metastasis, especially in hypoxic regions, is of immense value for early-stage lung cancer diagnosis, prognosis, and intervention. MicroRNAs (miRNAs) are small non-coding RNAs that have potential as such biomarkers.

Exosomes are non-replicative extracellular vesicles 30–150 nm in size that all cells can be found in the external secretion and release¹⁹. In vivo, exosomes have been isolated from body fluids, including blood, urine, saliva and breast milk; they also can be isolated from supernatants of cell cultures. Exosomes were initially regarded as waste material, but recent studies suggest that they play a key role in intercellular interactions^20,21. Exosomes can be transferred to receptor cells, genetic material from a donor cells and between cells also plays an important role in communication^22–25. Lipids, proteins, DNA, mRNA, miRNAs, and metabolites, perform both local and systemic intercellular transfer that a variety of contents what the exosomes could carry. By acting in both tumor cells and in the TME on fibroblasts, endothelium, leukocytes, and progenitor cells, tumor-secreted exosomes have emerged as important messengers in multiple tumor processes, including invasion, tumor-associated angiogenesis, tissue inflammation, and immunological remodeling²⁶. Angiogenesis provides cells with a controlled supply of oxygen and utilizes a complex control system of pro- and antiangiogenic factors²⁷. Exosomes have been proven to be reliable blood markers for cancer diagnosis²⁸. However, how such exosomes regulate the formation of premetastatic niches through the induction of angiogenesis requires further investigation.

Herein, we show that exosomal miR-671-3p derived from H1975 cells can be transferred to human umbilical vein endothelial cells (HUVECs) to promote angiogenesis by targeting Krüppel-like factor 2 (KLF2). Our results of clinical part suggest that miR-671-3p in the circulating exosomes of LUAD patients has utility as a blood biomarker for the prediction of metastasis.

Results

Hypoxia alters transcription in, and exosome composition of, H1975 cells

HIF-1 is a major transcription factor involved in hypoxic response; its expression was increased under hypoxic conditions. HIF-1α expression was increased in H1975 cells after 24 h of hypoxia (Fig. 1A). To identify other changes in gene expression, hypoxic and normoxic H1975 cells were subjected to transcriptome sequencing. GO enrichment analysis of the resulting differentially expressed genes (DEGs) is shown in Fig. 1B. We next focused on cellular components of exosomes. Hypoxic exosomes were isolated and characterized using TEM, NanoFC, and western blot detection of exosomal markers (Fig. 1C-E). HUVECs labeled with CCK-F were incubated with DiI-labeled exosomes derived from hypoxic H1975 cells, and DiI fluorescence was observed (uptake rate = 98.97%)(Fig. 1F). These results demonstrate that H1975-secreted exosomes can be transferred to HUVECs.

Fig. 1.

Hypoxia alters H1975 transcription. (A) Expression of HIF-1α in H1975 cells from normoxia and hypoxia condition (n = 3); (B) GO enrichment analysis of hypoxic vs. normoxic H1975 differentially expressed genes (DEGs); histograms refer to cellular component (CC), molecular function (MF), and biological process (BP); (C) Transmission electron micrographs of exosomes derived from hypoxic H1975 cells (n = 3); scale bar, 100 nm; (D) NanoFCM particle-size measurements of hypoxic H1975 exosomes (Exo^Hypoxia); mean diameter 82.60 nm; (E) Western blotting for CD9, CD81 and TSG101 in H1975 exosomes (n = 3); (F) Fluorescence micrographs of DiI-labeled hypoxic H1975 exosomes and CCK-F and Hoechst 33324-labeled HUVECs (n = 3).

Hypoxia LUAD exosomes primes premetastatic niche formation

Tumor metastasis is an important stage of cancer development, but it’s difficult to detect in the early onset. Hypoxia produces many secretory changes that facilitate metastasis, altering ECM, immune composition, and vascular permeability at future metastatic sites in a process known as premetastatic niche formation²⁹. To further explore the regulatory role of exosomes on HUVECs, we added H1975 exosomes from normoxia and hypoxia groups to HUVEC medium. The HUVEC proliferation, migration, and angiogenesis are signficantly accelerated when H1975 Exo^Hypoxia is added (Fig. 2A–C). As shown in Fig. 2, treatment with hypoxia exosomes dramatically promoted proliferation(Fig. 2A), migration(Fig. 2B) and angiogenesis(Fig. 2C) compared with normoxia. These results suggest that hypoxia LUAD-secreted exosomes transferred to HUVECs alter angiogenesis.

Fig. 2.

Hypoxic LUAD-secreted exosomes accelerate the premetastatic niche formation and angiogenesis. (A) HUVEC proliferation with H1975 Exo^Normoxia or Exo^Hypoxia. Data are shown as mean ± SEM are provided (n = 3); (B) The migration of HUVECs that were co-incubated with H1975 cells in Exo^Normoxia and Exo^Hypoxia conditions was assessed using a transwell assay; scale bar, 200 μm; (C) HUVEC tube-formation after treatment with H1975 Exo^Normoxia or Exo^Hypoxia; scale bar, 10 μm. Data shown as mean ± SEM are provided (n = 3), * compared with Normoxia, ^*P < 0.05, ^****P < 0.0001.

Exosomal miR-671-3p is upregulated in hypoxia

The Illumina test for small RNA sequencing of exosomes was used to generate a volcano plot of differentially expressed miRNAs (Fig. 3A). The upregulation of miR-671-3p was verified using qRT-PCR (Fig. 3B). Because it is well known that microRNAs can regulate gene expression by binding to the 3′-UTRs of specific mRNAs, we identified the potential targets of miR-671-3p using three online databases (Fig. 3C): TargetScan (http://www.targetscan.org/), miRDIP (http://ophid.utoronto.ca/mirDIP/index_precursor.jsp) and ENCORI (https://rnasysu.com/encori/) and. Among 36 overlapping predicted targets, we selected KLF2 for further analysis as it is significantly expressed in 4 LUAD databases (Fig. 3D). LUAD data from the UCSC Xena (https://xenabrowser.net/datapages/) were compiled for investigating the expression level of KLF2 associated with the malignant progression of lung carcinoma (Fig. 3E, F). ROC curve analysis was used to evaluate the ability of the model. ROC curve analysis (Fig. 3G) indicated that the AUC of the KLF2 based model was 0.926, suggesting that KLF2 has diagnostic potential. The StarBase database shows correlations between the expression levels of miR-671-3p and KLF2 (Fig. 3H).

Fig. 3.

Exosomal miR-671-3p is up-regulated in LUAD hypoxia exosomes. (A) Volcano plot of DEG (miRNAs) expressed in H1975 exosomes (red: increased; blue: decreased; gray: no significant difference); (B) RT-PCR quantification of miR-671-3p expression in H1975 exosomes. Data shown as mean ± SEM are provided (n = 3). * compared with Normoxia, ^***P < 0.001 according to two-tailed Student’s t-test; (C) Candidate target genes were predicted by intersecting outputs from prediction algorithms (TargetScan, mirDIP, and ENCORI); (D) Predicted genes expression in LUAD databases; (E) Decreased KLF2 is lower expression in LUAD tissues compared to normal tissue (LUAD, lung adenocarcinoma; red: tumor tissues; blue: normal tissues); (F) Relationship between KLF2 and overall survival of patient with LUAD; (G) ROC curve for KLF2, AUC (area under the curve) = 0.926. (H) A negative correlation between miR-671-3p and KLF2 levels is observed in LUAD tissues.

Exosomal miR-671-3p from hypoxic LUAD cells primes angiogenesis

After the addition of miR-671-3p mimic, the results of cell proliferation showed that compared with the NC mimic group, miR-671-3p mimic promoted cell proliferation; When the miR-671-3p inhibitor was added, cell proliferation was inhibited compared to the inhibitor NC group(Fig. 4A). Such as proliferation result, we test the cell migration assays and tube-formation assays, they both showed that exosomal miR-671-3p-induced angiogenesis was abolished by the miR-671-3p inhibitor (Fig. 4B-C). We transfected miR-671-3p mimics into HUVECs to enhance its function, while a miR-671-3p inhibitor was used to block its function in H1975/miR-671-3p exosomes. Western blotting showed that the expression of KLF2 was decreased by overexpression of miR-671-3p in HUVECs (Fig. 4D). To expel false positives from the experimental, Dual-Luciferase assay was needed. Luciferase reporter assays validated the interaction between KLF2 and miR-671-3p (Fig. 5A). The cell proliferation was inhibited compared with NC when the miR-671-3p inhibitor had effection; moreover, the rate of cell proliferation was limited in inhibition compared with the former condition as miR-671-3p inhibitor and siKLF2 were both added condition, and the function of cell proliferation could be clearly seen at 72 h, compared with NC. Knockdown of KLF2 reversed the inhibitory effects of miR-671-3p silencing on the overexpression of KLF2 (Fig. 5B). To ensure experimental rigor and reproducibility, we performed a series of rescue experiments. Furthermore, siKLF2 co-transfected with miR-671-3p inhibitor significantly promoted angiogenesis (Fig. 5B-D). As expected, subsequent Western blotting showed that KLF2 knockdown restored VEGFR2 expression after miR-671-3p was inhibited (Fig. 5E). CAM assays showed that inhibition of exosomal miR-671-3p retarded tumor progression (Fig. 5F). Taken together, these data indicate that exosomal miR-671-3p derived from LUAD cells regulates the expression of KLF2, which negatively regulates VEGFR2 expression and induces angiogenesis in HUVECs (Fig. 5G).

Fig. 4.

Exsomal miR-671-3p derived from hypoxic LUAD primes the premetastatic niche and induces angiogenesis. (A) CCK-8 analysis of HUVEC proliferation as a function of miR-671-3p treatment; (B) Transwell assay for effects of miR-671-3p mimics and inhibitor on migration; scale bar, 200 μm; (C) Effect of miR-671-3p mimics and inhibitor on the tube-formation by HUVECs; scale bar, 10 μm; (D) Western blot of KLF2 and VEGFR2 expression in HUVECs as a function of miR-671-3p treatment. Data shown as mean ± SEM are provided (n = 3), * compared with NC mimic, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; # compared with NC inhibitor, ^#P < 0.05, ^##P < 0.01, ^###P < 0.001.

Fig. 5.

KLF2 is a functional target of miR-671-3p in HUVECs. (A) Luciferase activities of the 3′-UTR KLF2-luc constructs in HEK-293T cells after transfection with miR-671-3p mimics; data shown as means ± SEM are provided (n = 3), * compared with KLF2-WT + NC mimic, ^***P < 0.001; # compared with KLF2-WT + miR-671-3p mimics, ^###P < 0.01; (B) Proliferation of HUVECs as a function of miR-671-3p treatment; (C) Transwell assay on the effect of miR-671-3p inhibitor with/without siKLF2; scale bar, 200 μm; (D) Tube-formation by HUVECs in the presence of miR-671-3p with/without siKLF2; scale bar, 10 μm; (E) Western blotting for KLF2 and VEGFR2 expression in HUVECs in the presence of an miR-671-3p inhibitor and siKLF2. Data shown as mean ± SEM are provided (n = 3), * compared with NC, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; # compared with NC inhibitor,^**P < 0.01, ^***P < 0.001, * is compared with NC, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; # is compared with miR-671-3p inhibitor, ^##P < 0.01, ^###P < 0.001, ^####P < 0.0001, for data shown in panels B, C and D; (F) NIH-H1975 Exo-miR-671-3p inhibitor and Exo-NC mixed with an equal volume of Matrigel were implanted onto the CAM of a 9-day old chicken embryo. After 3 days of implantation, the amount of blood vessels on the CAM induced by the H1975 was determined. (G) Mechanism diagram of the function of exosomal miR-671-3p in HUVEC.

Exosomal miR-671-3p is associated with the LUAD of premetastatic niche and angiogenesis

To measure exosomal miR-671-3p in plasma, circulating exosomes were extracted. The circulating exosomes of miR-671-3p was upregulated in patients with LUAD compared with healthy donors by RT-PCR (Fig. 6A, B). According to miR-671-3p expression levels (Table 1), increased expression of miR-671-3p was associated with pathological stage (P = 0.0081), tumor stage (P = 0.0049), lymph node infiltration (P = 0.0073), and tumor size (P = 0.0310) in patients with LUAD. These data suggest that miR-671-3p may be useful as a blood-based biomarker. Finally, our study suggests that HUVEC cells affected by hypoxic LUAD exosomal miR-671-3p may regulate KLF2 targets through miR-671-3p, thereby regulating downstream VEGFR2 (Fig. 6C).

Fig. 6.

Exosomal miR-671-3p is associated with metastatic progression (A) Violin plot of ΔCT miR-671-3p expression in patient plasma exosomes. * is compared with healthy control subjects, ^****P < 0.0001; (B) Diagram of relative miRNA expression levels measured using qRT-PCR in the plasma of healthy subjects (n = 34) and LUAD patients (n = 29). ^**P < 0.01; (C) Schematic representation of the model for the effect of hypoxic H1975 exosomal miR-671-3p on premetastatic niche formation and priming of angiogenesis.

Table 1.

The association of miR-671-3p expression with clinicopathological characteristics of LUAD patients.

Variables	Cases	miR-671-3p		P value
		High	Low
Total	29	22	7
Age(years)
≥ 60	15	9	6
< 60	14	13	1	0.2732
Gender
Male	11	9	2
Female	18	13	5	0.9965
Pathological stage
I/II	24	19	5
≥III	5	3	2	0.0081
T stage
T1/T2	23	18	5
≥ T3	3	2	1	0.0049
N stage
Negative	21	17	4
Positive	5	3	2	0.0073
Tumor volume(d/cm)
> 1.5	16	9	7
≤ 1.5	12	11	1	0.0310

Discussion

The suitable microenvironment for premetastatic niche is established before metastasis. Tumor-derived exosomes are considered to be a primary driver in the establishment, particularly by influencing angiogenesis. MiR-671-3p has been reported to be cancer-promoting and has been associated with poor prognoses in patients with lung cancer^29–31. However, the molecular mechanism of miR-671-3p regulation of LUAD transfer has not been clarified. In our study, we found that hypoxic LUAD cell-secreted miR-671-3p could be delivered to HUVECs via exosomes, thereby disrupting the integrity of endothelial barriers and inducing angiogenesis, contributing to niche formation. Overall, our study indicates a clinically significant role for miR-671-3p in LUAD progression.

Hypoxia is a common feature of solid tumors; it is a potent aspect of the TME promoting metastatic progression^32,33. Most solid tumors maintain hypoxic regions throughout progression, selecting aggressive malignant cells that withstand the ischemic stress of an unfavorable TME¹⁴. The hypoxia-inducible factor (HIF) family of transcription factors is central to the cellular response to hypoxia of both the tumor and stromal compartments³⁴. Under normoxic conditions, HIF-1α is rapidly degraded by the proteasomal system; this degradation is blocked under hypoxia, leading to its accumulation in the nucleus³⁴. Clinically, hypoxia and the expression of HIF-1 and HIF-2 are associated with increased distant metastasis and poor survival in multiple tumor types³². Our data show that hypoxic H1975 cells overexpress HIF-1α, which is consistent with what has been reported.

Exosomally transferred microRNAs (miRNAs) have been shown to alter the behavior of recipient tumor cells and stromal cells to promote oncogenesis, highlighting their role in cancer-cell communication³⁵. Although stromal content varies widely from tumor to the neoplasm, all solid tumors rely on the stroma for nutrients and waste disposal if they are to grow beyond the minimum size³⁶. Different from normal vasculature, tumor vessels were not arranged hierarchically, with irregular spaced and heterogeneous structure. In addition, they are hyperpermeable to plasma and plasma proteins and may lack pericytes; moreover, they are lined up by actively dividing endothelial cells³⁷. For example, HCC cell-derived exosomal miR-1247-3p can be delivered to cancer-associated fibroblasts, thereby priming cancer-associated fibroblast activation and promoting hepatocellular carcinoma metastasis³⁸. Exosome-mediated secretion of miR-1305 by multiple myeloma (MM) cells promotes the oncogenic activity of hypoxic MM cells. In addition, exosomal miR-1305 transferred from MM cells to macrophages and miR-1305 transferred macrophages show tumor-promoting M2-macrophage phenotypes³⁹. In summary, these studies suggest that exosome-mediated miRNA transfer between tumor cells and stromal cells is critical for cancer progression. The diagnostic value and prognostic effects of exosomes and their contents have been demonstrated in a variety of cancers⁴⁰. Further diagnostic studies using exosomes are also forthcoming⁴¹.

In our study, analysis of the Pearson product moment correlation coefficient t (Pearson’s r) showed that miR-671-3p and KLF2 levels in LUAD tissues were negatively correlated; moreover, endothelial cells (EC) took up exosomes, which suggests that miR-671-3p can be transferred via exosomes. In addition, we observed DiI fluorescence in HUVECs co-incubated with DiI-labeled LUAD exosomes. These results demonstrate that LUAD secreted exosomes, contains miR-671-3p, can be transferred into HUVECs.

The effect of miR-671-3p on tumors remains controversial and is primarily based on studies of cultured tumor cells^{29–31,42–47}. This particular miRNA has been reported to be involved in proliferation^42,48, apoptosis⁴⁹, migration⁴⁴, inflammation⁴⁹, metabolic⁴⁵, drug resistance⁴⁶, and tumor recurrence⁴⁷. According to recent reports, the abnormal expression of miR-671-3p in lung cancer (LC), especially in non-small cell lung cancer (NSCLC) is still controversial^32,44. By comparing the expression of existing clinical samples, we noticed that the expression of miR-671-3p was different in adjacent tissues and adjacent normal tissues compared with tumor lesion tissues^32,44. The low expression of miR-671-3p in adjacent tissues of NSCLC, combined with the clinicopathological characteristics of patients with tumors, suggests that miR-671-3p is highly expressed in tumor tissues at the early stage of the tumor for accumulation and release, thus promoting pre-metastatic angiogenesis⁴⁴. In the advanced stage of the tumor, the expression of miR-671-3p in tumor tissues was found to decrease due to the release, the reduction of the regulation mechanism of angiogenesis before metastasis, and the activation of other proliferation and metastasis mechanisms⁴⁴. Moreover, based on clinical data, it could be seen that the high expression of miR-671-3p predates lymphatic metastasis⁴⁴. However, the role of miR-671-3p in circulating exosomes in the TME has hardly been reported. We therefore focused on the study of angiogenesis before tumor metastasis to better understand the changes in the TME during that period.

Solid tumor cell-induced angiogenesis plays a vital role in cancer metastasis¹⁶.As we know that ECs form the inner layer of blood vessels; EC-enriched genes are the main determinants of EC-specific function, aligning the EC phenotype with the functional environment⁵⁰. The current study provides evidence that miR-671-3p in ECs can downregulate KLF2, thereby inducing angiogenesis via KLF2-VEGFR2 axis. The KLF family of zinc finger-containing transcription factors regulates many biological processes^51–56. Moreover, KLF2 by inhibiting vascular endothelial growth factor (VEGF) receptor 2 (VEGFR2) negative regulating angiogenesis promoter activity⁵⁷. Given the role of VEGF in pathological angiogenesis, continuous VEGFR2 activation could lead to excessive proliferation of EC⁵⁷. Analysis of the VEGF gene promoter sequence reveals the presence of several CACCC and GGGTG motifs, indicating that it could be another direct homeostatic target of KLF2⁵⁷. Furthermore, the expression of KLF2 is decreased NSCLC^58–60. Using receiver operating characteristic curves, we investigated the feasibility of the hub genes for prediction. We obtained an area under the curve value over 0.9, indicating significant diagnostic value. Our findings suggest that miR-671-3p in HUVECs downregulates KLF2 and subsequently increases VEGFR2 expression levels; in LUAD-secreted exosomes, miR-671-3p promotes angiogenesis by silencing KLF2.

Multiple studies have shown an essential role of premetastatic niche formation in LUAD metastasis. Lung-cancer mortality is higher than that of any other cancer, with metastasis as the primary cause (https://gco.iarc.fr/)⁶¹. The premetastatic niche is the microenvironment that prepares circulating tumor cells for colonization of specific organs, including angiogenesis, matrix remodeling, immune response, and biomarker expression⁸. In addition, premetastatic niche biomarkers have an emerging role in cancer diagnosis, intervention in metastasis, and prognostic prediction. However, the the primary generation of LUAD cells induced shift niche before formation of the mechanism is unclear^8,62. Tumor angiogenesis has recently been found to be involved in the regulation of vascular basement membrane components, a specialized form of extracellular matrix, suggesting attractive therapeutic targets⁶³. Notably, tumor-induced angiogenesis is a characteristic of the premetastatic niche.

In the study, we reveal how H1975 cell-derived exosomal miR-671-3p regulates the formation of the premetastatic niche. We found that miR-671-3p secreted by LUAD under hypoxic conditions induced niche formation at distant sites by promoting angiogenesis. Our data show that in the case of HUVEC, miR-671-3p downregulation leads to increased expression of the target gene KLF2, which in turn inhibits VEGFR2 expression in HUVEC, resulting in reduced angiogenesis.

This study has several limitations. First, we studied only a small number of clinical samples, particularly for metastasis. Second, we solely examined the transfer of exosomes to recipient cells as a strategy to initiate metastasis; however, there are additional processes to focus on, which may also be viable therapeutic targets, including the mechanisms of proinflammatory factors, inhibition of angiogenesis, destroying of the local matrix, and reactivation of antitumor immune responses. Third, we have studied the role of miR-671-3p in H1975, which does not fully represent LUAD. Fourth, as the study was in vivo, we have only a preliminary demonstration of the fact that the inhibition of miR-671-3p prevents angiogenesis using CAM; this implies that our use of nude mice, devoid of any functional immune system makes in vivo observations not fully representative of the clinical situation. Therefore, the function of exosomes^{miR−671−3p} involved in LUAD progression requires a deeper and more comprehensive study.

Therapies targeting premetastatic niches are promising strategies for intervention in cancer metastasis. This study shows a possible mechanism of premetastatic angiogenesis induced by hypoxia during solid tumor proliferation and suggests the miR-671-3p–KLF2–VEGFR2 axis as a target for LUAD therapy. We found a positive correlation in miR-671-3p levels between circulating exosomes and cancer tissues in patients with LUAD. Overall, our clinical samples’ data show that quantitative blood tests for miR-671-3p levels in circulating exosomes is a promising tool for the diagnosis of LUAD metastasis and for selecting patients with higher risk of metastasis to receive preventive treatment.

Methods

Study population

Plasma samples of 40 LUAD patients and 32 healthy subjects were analyzed in this study to detect miR-671-3p levels. All patients were diagnosed pathologically, based on the pathological classification and TNM staging criteria for LUAD implemented in accordance with the staging criteria of the International Union Against Cancer. Each LUAD plasma sample was collected from the patient after a pathological examination. All experiments were performed in accordance with relevant guidelines and regulations. Each participant in this study signed an informed consent form, and the Ethics Committee of the First Hospital of Jilin University approved this study.

Cell lines and reagents

The human LUAD cell line H1975 was provided by the Cell Bank, Chinese Academy of Sciences. The HUVEC and human embryonic kidney 293T cell line (HEK-293T) was provided by iCell bioscience (China). H1975 cells were cultured in 1640 medium (Gibco, USA) containing 10% exosome-depleted fetal bovine serum (Viva Cell, Uruguay), 100 U/mL penicillin, and 100 mg/mL streptomycin (ScienCell, USA). HEK-293T cells were cultured in DMEM (Gibco, USA) supplemented with 10% FBS (Gibco, USA), 100 U/mL penicillin, and 100 mg/mL streptomycin. HUVECs were cultured in extracellular matrix medium (ScienCell, USA) containing 5% FBS (ScienCell, USA), 1% ECGS (ScienCell, USA), 100 U/mL penicillin, and 100 mg/mL streptomycin. These cells were cultured in a humidified air atmosphere at 37 °C with 5% CO₂. Cell passage was conducted until cells were grown to 80% confluence. To induce hypoxia, cells were cultured under 1% O₂ for 24 h. Control cells were maintained in normoxic conditions and harvested at the specified times.

Transfection

Transfection plasmids were purchased from GenePharma, China. For in vitro experiments, the HUVECs were transfected for 48 h with miR-671-3p mimic and inhibitor or negative control (NC) and inhibitor negative control (NC-inhibitor) using the Lipofectamine RNAiMAX Reagent (Invitrogen, USA) according to the manufacturer’s protocol. For KLF2 knockdown (si-KLF2), siRNA targeting the coding region of KLF2 (GenePharma), and the siRNA transfection reagent (Invitrogen) was followed the manufacturer’s instructions. Non targeting siRNA (si-NC) as a negative control. The HUVECs were collected 24 h later to conduct cell-function experiments.

Proliferation assay

To obtain viability curves, proliferation assays were performed using the CCK-8 (TransGen, China). Pretreated cells were seeded into a 96-well plate at 3 × 10³ cells/well and cultured for 24, 48, and 72 h. After incubation at 37 °C for 3 h, the culture absorbance was measured at 450 nm using a microplate reader.

Transwell migration assay

Transwell chamber inserts (Costor, USA) without coated Matrigel were used for the migration assays. The upper chamber of the 6-well cell culture inserted with 1.5 mL serum-free medium seeded into transfected HUVECs. In addition, 2.6 mL of complete medium supplemented with 5% FBS was added to the bottom of the inserts. Cells were allowed to migrate for 24 h. After incubation, the HUVECs on the upper surface of the membrane were removed; the cells on the lower filter surface were fixed and stained with crystal violet. According to the manufacturer’s instructions, the number of migrated cells was counted in 3 random fields per sample under a microscope.

Quantitative Real-Time PCR

In vitro H1975 exosome miRNA was extracted using an EasyPure miRNA Kit (TransGen, China) according to the manufacturer’s protocol. For each plasma sample from the LUAD and healthy subjects, 500 µL of the sample were used to extract exosome miRNA using the TransExo Serum/Plasma Exosome Total RNA Extraction Kit (TransGen, China). Reverse transcription was performed using TransScript miRNA First-Strand cDNA Synthesis SuperMix for miRNAs (TransGen, China). Real-time PCR was conducted using the PerfectStart Green qPCR SuperMix (TransGen Biotech) according to the manufacturer’s instructions. U6 was used as an internal reference, and miR-671-3p levels were calculated using the 2^−ΔΔCt method. A Ct value > 40 was considered below the detection threshold of the array. The primer sequences used for the qRT-PCR reactions are: U6: 5′-CAAGGATGACACGCAAATTCG-3′; miR-671-3p: 5′-TCCGGTTCTCAGGGCTCCACC-3′.

Western analysis

For protein isolation, transfected HUVECs or H1975 cells were lysed using cell lysis buffer, shaken on ice for 30 min, and centrifuged at 15,000×g for 15 min at 4 °C. The total protein concentration was calculated using a BCA Protein Assay Kit (Beyotime, China). Anti-KLF2 monoclonal antibodies were purchased from Abcam (USA); anti-β-tubulin monoclonal antibodies were purchased from Beyotime (China); anti-VEGFR2, anti-HIF-1α, and anti-TSG101 monoclonal antibodies were purchased from ZENBIO (China); anti-CD9 and anti-CD63 were provided by Proteintech (USA). Horseradish peroxidase–labeled goat anti-rabbit or anti-mouse secondary antibody was used for detection. β-tubulin was used as an internal reference control. Using SDS-PAGE electrophoresis separation of protein samples, and then transferred to the PVDF membrane. The membrane were blocked in TBST, containing 5% no-fat milk for 1 h and incubated in primary antibodies at 4 ℃ overnight. And then the membranes were incubated with an HRP conjugated goat-anti rabbit (mouse) secondary antibody for 1 h and detected by ECL reagents. The image was semi-quantitatively analyzed using Fiji image analysis software.

Exosome isolation and characterization

Exosomes were purified from H1975-derived conditioned media by ultracentrifugation. Exosome miRNA was purified from the plasma of LUAD patients using the TransExo Serum/Plasma Exosome Total RNA Extraction Kit (TransGen, China). The H1975 normoxia- and hypoxia-conditioned media were collected after 24 h. The supernatant was passed through filter (0.22 μm, Millipore) and subjected to ultracentrifugation at 110,000×g for 70 min at 4 °C (himac). The exosome pellets were washed with D-PBS, followed by a second ultracentrifugation at 110,000×g for 70 min at 4 °C, and then resuspended in D-PBS. For transmission electron microscopy (TEM), 10 µL of hypoxic H1975 exosomes were fixed using 2% paraformaldehyde and placed on 200-mesh Formvar-coated grids. The grids were then stained using 2% phosphotungstic acid for 1 min and observed using TEM (Hitachi H-7700). The NanoFCM analyzes the size of exosomes. The exosomes (10–20 µL) were suspended in 1 mL PBS and vortexed for 1 min to distribute the exosomes evenly. NanoFCM (N30E) was then used to measure and observe the scale distribution of the exosomes.

Exosome labeling and uptake

For uptake analysis, LUAD exosomes were fluorescently labeled using the DiI membrane Staining Kit (Beyotime) according to the manufacturer’s instructions. The labeled exosomes were then co-cultured with HUVECs for 24 h at 37 °C. Exosomes that were absorbed by the HUVECs were observed under a fluorescence microscope. Hoechst 33324 solution (Beyotime) and a DiO membrane Staining Kit (Beyotime) were used to label the cell nucleus and membrane, respectively.

Luciferase assays

The 3′-UTR segment of the KLF2 gene was amplified using the polymerase chain reaction (PCR) and inserted into the vector. Co-transfections of the KLF2 3′-UTR plasmid with the miR-671-3p pmirGLO Dual-Luciferase miRNA Target Expression Vector into cells were performed using Lipofectamine 3000 (Invitrogen). Transfection after 48 h, the use of Dual-Luciferase Reporter Assay System (Promega) luciferase activity. All experiments in three times, each experiment repeated three times.

Migration assay

For transwell migration assays, LUAD exosome-treated HUVECs were suspended in serum-free medium and seeded into the transwell chambers with inserts of 8-µm pore size (Corning). Place the medium containing 10% FBS into the bottom chamber. After 12 h, the staining was done with hematoxylin and counted under 4 random fields of view (200×). Each experiment was repeated three times.

Angiogenesis assay

For tube formation assays, 24-well plate was added Matrigel matrix (Corning) and incubated at 37 °C for 30 min, in order to allow the Matrigel to polymerize. Seeded the treated HUVEC cells onto a Matrigel-coated well. At the 37 °C in a 5% CO₂ humidified atmosphere incubated the plate. At 12 h, tube formation was observed under a microscope. The tube formation ability was determined by measuring the number of tubes. Three times repeated of each experiment.

Chick Chorioallantoic membrane (CAM) assay

Fertilized chicken eggs were incubated at 37.8 °C and 60% humidity for 9 days. An artificial airbag and small window were created in each egg. NIH-H1975 Exo-miR-671-3p inhibitor and Exo-NC mixed with an equal volume of Matrigel were injected. The area around the implant was photographed after a 3-day incubation.

Statistical analysis

Using GraphPad Prism 8 software for statistical analysis. The quantitative values of all experiments were expressed as mean ± SEM. Differences among/between sample groups were analyzed using one-way ANOVA or the independent samples t-test. Pearson’s correlation coefficient was used to measure the degree of the linear relationship between miR-671-3p and KLF2. Survival analysis was performed using the Kaplan–Meier method. P < 0.05 was considered statistically significant. Adobe Photoshop CC and Fiji software were used for figure preparation.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

K.L., Y.W. and N.K.: conceptualization and methodology. K.L.: validation and investigation. K.L. and J.T.Z.: resources. K.L., Y.J.W., J.R.L. and Q.S.L.: data curation and visualization. K.L.: writing - original draft. K.L., Z.Y.C., F.Y.S. and Q.S.L.: writing - review & editing. K.L., Y.W., N.K. and Z.W.: Project administration.

Funding

Project supported by Basic scientific research project foundation of Beijing University of Chinese Medicine(2024-JYB-PY-021), Natural Science Foundation of Jilin Province (YDZJ202401313ZYTS) and the China Postdoctoral Science Foundation (2024M750262).

Data availability

The original datasets used in this study are available in the the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? acc=GSE285795). Further inquiries can be directed to the corresponding author.

Declarations

Competing interests

The authors declare no competing interests.