Sage Journals: Discover world-class research

Abstract

BACKGROUND:

The potential of microRNAs (miRNAs) as molecular tumor biomarkers for early diagnosis and prognosis in lung cancer is still unclear.

OBJECTIVE:

To analyze expression of miRNAs in A549 lung adenocarcinoma (LUAD) cells and in primary, non-malignant bronchial epithelial (BE) cells from healthy donors. To analyze the most prominently deregulated miRNAs in plasma samples of LUAD patients and healthy donors.

MATERIALS AND METHODS:

The expression of 752 miRNAs in LUAD and BE cells was assessed by RT-qPCR with mean-centering restricted normalization. The relative plasma levels of 18 miRNAs in LUAD patients and healthy donors were analyzed using RT-qPCR and normalized to miR-191-5p and miR-16-3p. Putative interactions between miRNAs and their target genes were investigated in silico.

RESULTS:

Out of 752 miRNAs, 37 miRNAs were significantly deregulated in A549 cells compared to BE cells. MiR-15b-3p, miR-148a-3p, miR-193b-3p, and miR-195-5p were significantly deregulated in plasma samples of LUAD patients compared to donors. The target genes of those four miRNAs are involved in essential mechanisms in cancer development and progression.

CONCLUSIONS:

There are substantial differences between cancer and control miRNA expression in vitro and in plasma samples of LUAD patients compared to healthy donors. Four deregulated miRNAs are promising as a diagnostic biomarker for adenocarcinoma of the lung.

Keywords

Lung adenocarcinoma A549 cells bronchial epithelial cells microRNA plasma samples

1. Introduction

Lung cancer is a very common type of cancer. The number of deaths caused by lung cancer in 2020 in both sexes and all age groups was about 1.8 million, which is more than 18% of all cancer-related deaths, thus making lung cancer the leading cause of cancer death worldwide. Lung cancer can be divided into adenocarcinoma, squamous cell carcinoma, and neuroendocrine tumors, including small cell carcinoma and carcinoid [1]. The early detection of lung cancer plays an important role for prognosis and overall survival. Unfortunately, patients at the early stage are mostly asymptomatic; thus, lung cancer is frequently diagnosed at an advanced stage. Nowadays, for lung cancer diagnostics, different invasive methods such as bronchoscopy with endobronchial ultrasound (EBUS) or/and transbronchial needle aspiration (TBNA), CT-guided transthoracic percutaneous needle aspiration, thoracentesis of pleural effusion, or surgical procedures are used. The medical history and complaints of a patient also play an important role in this decision, since every invasive method has a risk of complications. Thus, the establishment of biomarkers for early detection of lung cancer could be crucial in diagnosing lung cancer at an earlier stage.

The known molecular tumor biomarkers have greatly changed the classification and management of lung cancer and improved the overall survival for lung cancer patients [2, 3]. New biomarkers could especially aid in the selection of individuals for low-dose computed tomography screening and management of small nodules with unclear dignity (benign/malignant) [4].

MicroRNAs (miRNAs) are small, typically 21–23 nucleotides long non-coding RNA molecules that are involved in many biological processes. MiRNAs are involved in gene regulation at post-transcriptional level. MiRNAs play an essential role in many biological processes of healthy, as well as malignant cells, via regulating mechanisms that ensure normal cell and organ activity, such as cell proliferation, differentiation, apoptosis and tumorigenesis [5]. Expression levels of miRNAs vary in different malignancies and depend on cancer type and stage [6]. Aberrant miRNA signatures in malignant cells are regulated by genetic alterations (e.g. chromosomal rearrangements, amplifications, deletions, and mutations), epigenetic changes, and transcriptional regulation [7]. Numerous studies have shown that functional miRNAs substantially contribute to the etiology and progression of different cancer entities, including lung cancer. MiRNAs exhibit tumor suppressive or promoting activities, thereby controlling initiation, progression, and metastasis of lung cancer [7, 8]. MiRNAs modulate various cellular mechanisms such as cell proliferation, motility, invasion, apoptosis, angiogenesis and immune escape, which frequently leads to development of resistance against current cancer therapies [7, 8, 9]. Several miRNAs deregulated in lung cancer have been found in extracellular vesicles (EV) secreted in body fluids, primarily in the circulation. MiRNAs are stable in plasma and serum, thus, circulating miRNA can be used as a minimally invasive biomarker for early lung cancer detection [7, 10]. Among others, Huang and coworkers have shown that EV miRNAs have a high diagnostic accuracy for non-small cell lung cancer (NSCLC) and suggested miR-21, miR-210, and miR-1290 as potential diagnostic biomarkers for NSCLC [10]. Regarding the clinical implications of aberrant miRNAs in lung cancer, miRNAs are promising as cancer subtype classifiers, diagnostic and prognostic markers, as drug-response predictor and, finally, as potential therapeutic targets [11].

In this study, we aimed to investigate in vitro expression profiles of miRNAs in A549 LUAD cells compared to primary, non-malignant bronchial epithelial (BE) cells. Furthermore, we compared the expression of miRNAs prominently deregulated in vitro in plasma samples of 18 LUAD patients and 18 healthy donors. We analyzed putative interactions and molecular functions of validated target genes of deregulated miRNAs. A detailed description and understanding of miRNA expression profiles and related molecular mechanisms can provide useful information for the search of future diagnostic and therapeutic targets in LUAD patients.

2. Materials and methods

2.1 Cell culture

A549 cells were purchased from ATCC (Rockville, Maryland, U.S.) and grown in DMEM/F12 medium supplemented with 10% fetal calf serum (FCS; Gibco, Paisley, UK). Primary BE cells were isolated from non-utilized human donor lungs, explanted at the Department of Surgery, Medical University of Graz, Austria. The investigations were conducted according to the Declaration of Helsinki principles and were approved by the local institutional review board. For BE cell isolation, a protocol modified after Fulscher et al. [12] was used. In brief, isolated proximal to segmental bronchi from explant lungs were cut in approximately 1–2 cm pieces, rinsed with PBS and added to a cell culture dish with DMEM/F12 medium containing 20% fetal calf serum (FCS), antibiotics (Pen/Strep), and antimycotic (Fungizone, 1:100; Gibco, Waltham, MA). Epithelial cells were scraped gently with a scalpel, resuspended, and transferred in a new cell culture dish coated with collagen type I (rat-tail collagen, 4.5 mg/ml, Sigma Aldrich, diluted 1:100 in PBS), containing DMEM/F12 and 20% FCS. After cell attachment, medium was replaced with BEGM (CC-3170, Lonza, Basel, Switzerland) containing growth factors, which was further used as a growth medium. BE cells were cultured on collagen-coated cell culture dishes and used in low passages ( $<$ 4). All cell lines were grown under standard conditions at 37 ${}^{\circ}$ C, 5% CO ${}_{2}$ and 95% humidity.

2.2 Patient cohort

This study confirms to the provisions of the Declaration of Helsinki (as revised in 2013) [13]. This study was approved by the Ethics Committee of the Medical University of Graz. All participants signed written informed consent and were randomly selected from those who presented in the Pulmonology Department. The inclusion criterion for LUAD patients was a completed histology showing lung adenocarcinoma. Exclusion criteria were: i) any other active malignant disease, ii) antitumor therapy for lung cancer before the histological material was retrieved, iii) acute or chronic pulmonary infection. In all patients, lung adenocarcinoma was diagnosed by experienced interdisciplinary team of lung pathologists, radiologists and pulmonologists from the University Hospital Graz, Austria. Classification and staging were done following the newest (8 ${}^{\text{th}}$ ) edition of TNM staging for lung cancer [14].

2.3 Cytokeratin immunofluorescence

BE cells were cultured in LabTek ${}^{\text{TM}}$ permanox chamber slides (Nunc, Rochester, NY). Upon washing with PBS, cells were fixed with 4% paraformaldehyde for 10 minutes at room temperature (RT) and permeabilized with 0.5% Triton-X100 in PBS (Sigma-Aldrich, St. Louis, MO). Undiluted monoclonal mouse cytokeratin antibody (clone MNF116, Dako, Glostrup, Denmark) was used for one hour staining at RT. Detection was performed using Alexa 594-labeled anti-IgG antibody (Invitrogen, Carlsbad, CA), diluted 1:600 for one hour at RT. As a negative control, an isotype control antibody was used as primary antibody.

2.4 RNA isolation from cells and quality check

The RNA isolation from cells was performed using the mirVana ${}^{\text{TM}}$ miRNA Isolation Kit (Invitrogen, by Thermo Fisher Scientific, Vienna, Austria) and protocol recommended by the producer. For quality check of isolated RNA, two different methods were used. First, concentration was measured by NanoDrop 200c UV-Vis Spectrophotometer and purity was assessed from the ratio A260/A280, being in the range of 1.8 to 2.0. Second, RNA quality check, including large and small RNA fragments and miRNAs, was performed by using the BioAnalyzer BA2100 (Agilent Technologies, Santa Clara, CA).

2.5 Reverse transcription

For cDNA synthesis, the miRCURY LNA ${}^{\text{TM}}$ Universal cDNA synthesis kit II was used (Qiagen, Venio, Netherlands), according to the producer’s recommendation. Briefly, each RNA sample was pre-diluted with nuclease-free water to a concentration of 5 ng/ $\mu$ l. The 5x reaction buffer, nuclease-free water, enzyme mix, and pre-diluted RNA were combined and the cDNA synthesis was performed on the peqSTAR Thermocycler (Peqlab, Erlangen, Germany) at 42 ${}^{\circ}$ C for 60 minutes, followed by heat inactivation of the enzyme at 95 ${}^{\circ}$ C for 5 minutes. Synthesized cDNA samples were stored in low nucleic acid binding tubes (Eppendorf, Vienna, Austria) at $-$ 20 ${}^{\circ}$ C and used within four weeks.

2.6 SYBR-Green based RT-qPCR

The expression levels of different miRNAs in the LUAD and BE cells were detected using the miRCURY LNA ${}^{\text{TM}}$ Universal RT microRNA PCR human panel I and II (Qiagen). The RT-qPCR was performed on the LC480 (Roche, Rotkreuz, Germany) by using following protocol: polymerase activation/denaturation, 95 ${}^{\circ}$ C for 10 min; amplification, 95 ${}^{\circ}$ C for 10 sec, 60 ${}^{\circ}$ C for 1 min, ramp-rate 2 ${}^{\circ}$ C/sec for 45 cycles in total. For further calculations and evaluations melting temperature-values and cycle threshold (Ct)-values were obtained using the software version LC480 1.5.0.39.

2.7 Normalization and statistical analysis of RT-qPCR data

Three replicates per sample and condition were analyzed and the results were represented as Ct-values determined with the 2 ${}^{\text{nd}}$ derivative method. In order to avoid plate-to-plate differences, three interpolate calibrators were used on every plate. Blank wells without any amplification reaction were also included as a negative control. To detect the set of the most stably expressed miRNAs used for further data-normalization, mean-centering restricted (MCR) normalization [15] was used. After MCR normalization, the expression levels of miRNAs were calculated using the $\Delta\Delta$ Ct comparative threshold method. By using the reference Ct-values, the $\Delta$ Ct of each sample was calculated according the formula: $\Delta\textit{Ct (sample)}=\textit{mean reference Ct}-\textit{mean target Ct}$ . Afterwards, the $\Delta\Delta$ Ct representing the difference between A549 and BE samples was calculated according the formula: $\Delta\Delta\textit{Ct}=\Delta\textit{Ct (A549)}-\Delta\textit{Ct (BE)}$ . Finally, the mean $\Delta\Delta$ Ct of each sample and the standard deviations were calculated.

2.8 Blood collection and miRNA isolation from plasma samples

Three milliliters of full peripheral venous blood were collected from each patient and healthy donor using EDTA-anticoagulant tubes (Thermo Fisher Scientific, Vienna Austria). Immediately upon collection, blood samples were centrifuged for 15 min at 1000 $\times$ g and 4 ${}^{\circ}$ C. Plasma was separated from the cellular part of the whole blood, collected in sterile DNA LoBind 1.5 ml tubes (Eppendorf, Vienna, Austria) and an additional centrifugation step (2500 $\times$ g, 15 min, 4 ${}^{\circ}$ C) was performed to eliminate any remaining blood cells that might influence miRNA isolation and interpretation of results. Finally, plasma samples were immediately frozen at $-$ 80 ${}^{\circ}$ C and used within two months of collection.

For miRNA isolation from plasma samples, miRNeasy Serum/Plasma Advanced Kit (Qiagen) was used. Isolation of miRNA was performed as recommended by producer, by using 200 $\mu$ L of plasma samples frozen at $-$ 80 ${}^{\circ}$ C and carefully defrosted on ice just before isolation. Final elution was performed in 20 $\mu$ L RNase-free water and miRNA aliquots were stored at $-$ 20 ${}^{\circ}$ C immediately after isolation. Each aliquot was defrosted just once for the cDNA synthesis in order to avoid possible miRNA degradation.

2.9 Mutational analyses

In all LUAD patients, the mutational status of tumor cells was determined routinely using the Ion Torrent S5 system for next generation sequencing (NGS) and two different panels. The first panel was the Ion AmpliSeq Colon and Lung Cancer Research Panel v2, containing 92 pairs of primers with an average amplicon length of 162 base pairs (bp). With this DNA-based panel following genes were analyzed: KRAS, EGFR, BRAF, PIK3CA, AKT1, ERBB2, PTEN, NRAS, STK11, MAP2K1, ALK, DDR2, CTNNB1, MET, TP53, SMAD4, FBX7, FGFR3, NOTCH1, ERBB4, FGFR1, and FGFR2. The RNA-based panel was the Archer Fusion Plex MUG Pan-TRK Lung panel v1 (custom modified for the Institute of Pathology, Medical University of Graz). This panel was used for analysis of following genes: ALK, BRAF, EGFR, FGFR1, FGFR2, FGFR3, MET, NRG1, NTRK1, NTRK2, NTRK3, RET, ROS1. PD-L1 expression was analyzed by immunohistochemistry and evaluated as percentage of positive tumor cells compared to all tumor cells.

2.10 Statistical calculations

All statistical analyses were performed by GraphPad Prism (v. 5.0). Data are represented as means $\pm$ SD. Values were tested for Gaussian distribution and two-sided Student’s $t$ -test was used to determine the level of statistical differences. A value of $p<$ 0.05 was considered statistically significant.

3. Results

3.1 miRNA expression profiles in cell samples

The experimental design of this study is shown in Fig. 1A. Isolated primary BE cells displayed a high degree of purity and nearly 100% positivity for cytokeratin (Suppl. Fig. 1A). Only RNA samples with an RNA integrity number (RIN) between 9.8–10.0 were used for further analyses (Suppl. Fig. 1B). Out of 752 miRNAs analyzed in our screening, 123 were detectable by RT-qPCR in both A549 and BE cells. The rest of analyzed miRNAs were either not expressed, or their expression level was below the detection range of our method. Based on the producer’s recommendation (Qiagen) and our own experience, all cycle threshold (Ct) values higher than 37 were considered negative.

Figure 1.

In vitro miRNA expression profiling. (A) Workflow of the present study. MiRNAs were isolated from A549 LUAD cells and primary, non-malignant bronchial epithelial (BE) cells from three healthy donors. MiRNA expression profiles were analyzed using RT-qPCR on panels containing 752 miRNAs in total. MiRNAs which were most prominently deregulated in vitro were selected for analyses in plasma samples of LUAD patients and healthy donors. (B) Expression levels of 18 miRNAs most prominently deregulated in A549 cells compared to non-malignant BE cells ( $\Delta\Delta$ Ct $>$ 2.5). Data are represented as means $\pm$ SD.

A reliable miRNA normalization is critical for data interpretation. Here, we used the mean-centering restricted (MCR) method [15]. Rather than focusing on all miRNAs, including miRNAs without detectable expression value, this method is restricted to all miRNAs that are detectable in all analyzed samples. MCR-based data normalization delivered very restrictive results. Without any cut-off, 58 miRNAs were differentially deregulated in A549 cells compared to BE cells. After setting the cut off to $\Delta\Delta$ Ct $\geqslant$ 1.5, a total of 37 miRNAs were differentially regulated, with 15 up- and 22 down-regulated miRNAs (Table 1). For further expression analysis in plasma samples of LUAD patients and healthy controls, we selected the 18 most prominently deregulated miRNAs with $\Delta\Delta$ Ct $\geqslant$ 2.5 (Fig. 1B).

Table 1

Deregulated miRNAs ( $\Delta\Delta$ Ct $>$ 1.5) in A549 LUAD cell line compared to the pool of primary, non-malignant bronchial epithelial cells isolated from three donors

Up		Down
miRNA	$\Delta\Delta$ Ct	miRNA	$\Delta\Delta$ Ct
miR-196a-5p	5.08	miR-205-5p	$-$ 14.58
miR-450a-5p	3.92	miR-141-3p	$-$ 8.22
miR-195-5p	3.29	miR-148a-3p	$-$ 5.96
miR-27b-5p	3.07	miR-135b-5p	$-$ 4.99
miR-193b-3p	2.79	miR-200b-3p	$-$ 4.34
miR-196b-5p	2.73	miR-34a-5p	$-$ 4.11
miR-10b-5p	2.56	miR-708-5p	$-$ 3.88
miR-15b-3p	2.54	miR-155-5p	$-$ 3.85
miR-126-3p	2.42	miR-429	$-$ 2.87
miR-582-3p	2.16	miR-200c-5p	$-$ 2.50
let-7i-5p	2.08	miR-19a-5p	$-$ 2.36
miR-589-3p	1.81	miR-210-3p	$-$ 2.34
miR-769-5p	1.77	miR-425-3p	$-$ 2.18
miR-93-3p	1.68	miR-200a-5p	$-$ 2.07
miR-15b-5p	1.59	miR-664a-3p	$-$ 2.05
		miR-1248	$-$ 1.93
		miR-31-5p	$-$ 1.88
		miR-181b-5p	$-$ 1.85
		mir-503-5p	$-$ 1.80
		miR-127-3p	$-$ 1.73
		miR-23a-3p	$-$ 1.65
		miR-944	$-$ 1.53

3.2 Patient characteristics

In this study, plasma samples of 18 LUAD patients and 18 healthy controls were included. All patients had a confirmed lung adenocarcinoma diagnosis and did not receive any anticancer therapy before sample collection. Donors were characterized as individuals without known malignancies or acute and/or chronic infections. Clinicopathological characteristics of LUAD patients are summarized in Table 2. Following data were collected: age, sex, cancer stage, anatomical cancer localization in the lung, smoking status, indicated mutations, PD-L1 expression, and metastases status. The stage classification was based on the last (8 ${}^{\text{th}}$ ) edition of TNM stage for lung cancer [14]. The median age of patients was 63 years (range from 52 to 88 years) and of donors 67 years (range from 31 to 84 years). The majority of patients had lung adenocarcinoma stage IV (44.5%), followed by stage I (33.3%), stage II (11.1%), and stage III (11.1%). All LUAD patients were either ex- or present-smokers.

Table 2
Clinicopathologic characteristics of LUAD patients

Patient	Age/sex	Stage	Localization	Smoking status	Mutations/PD-L1 status	Metastases
1	62/F	Ic	LLL	active	no	no
2	58/F	IV	LLL	active	PD-L1 5% pos.	multiple
3	88/F	IV	RUL	ex	EGFR positive (L858R)	pleura effusion
4	72/F	Ic	LLL	ex	no	no
5	57/M	IV	LLL	active	no	multiple
6	58/M	IIIa	RUL	ex	PD-L1 30% pos.	lymph nodes
7	82/F	Ia	LLL	ex	ALK positive	no
8	69/M	Ic	RUL	active	PD-L1 30% pos.	no
9	76/F	IIa	LLL	ex	EGFR positive	no
10	56/M	IV	R central	ex	PD-L1 5% pos.	multiple
11	52/M	IIb	RUL	ex	PD-L1 100% pos.	no
12	63/M	IV	RUL	ex	EGFR positive, PD-L1 5%	multiple
13	61/F	IIIa	L central	active	PD-L1 90% pos.	n. d.
14	61/M	IV	LLL	ex	no	multiple
15	74/M	Ic	LUL	ex	PD-L1 100% pos.	no
16	73/M	Ia	RUL	ex	no	no
17	80/M	IV	LUL	ex	PD-L1 100% pos.	bones
18	61/F	IV	L central	ex	no	multiple

F, female; M, male; LLL, left lower lobe; LUL, left upper lobe; RUL, right upper lobe; R, right, L, left; n. d., no data; neg., negative; pos., positive. Multiple, at least in two organs.

Table 3

Survival data of the patients

Patient #	First diagnosis	Survival (1 $=$ dead, 0 $=$ alive)	Time of death	Primary therapy	Overall survival (months)
1	Oct-19	0		S	25
2	Oct-19	1	Jan-20	N	4, †
3	Oct-19	1	Feb-20	N	5, †
4	Oct-19	0		S	25
5	Oct-19	1	Dec-19	N	3, †
6	Oct-19	0		C, R	25
7	Oct-19	0		S	25
8	Nov-19	0		S	24
9	Nov-19	0		S	24
10	Dec-19	0		C, I	23
11	Dec-19	0		S	23
12	Oct-19	0		T	25
13	Nov-19	0		C, I	24
14	Dec-19	1	Jan-20	C	2, †
15	Dec-19	1	Jan-21	R	14, †
16	Dec-19	0		R	23
17	Dec-19	1	Jun-20	I	7, †
18	Feb-20	1	Sep-20	C	8, †

S $=$ surgery, C $=$ chemotherapy, R $=$ radiation therapy, I $=$ immunotherapy, T $=$ target therapy, N $=$ no therapy, † $=$ dead.

We analyzed the survival data of patients from our cohort approximately two years after the first diagnosis (Table 3). Seven out of 18 (39%) patients died during this time. All patients who underwent surgery ( $n=$ 6; 33%) are still alive (October 2021). Patients who did not receive any therapy ( $n=$ 3; 17%) had the worst survival rate.

3.3 Normalization of circulating miRNAs

Finding stably expressed circulating miRNAs that can be used for normalization of expression data generated in biofluids (e.g. plasma) is even more challenging than the normalization of data collected under fully controlled in vitro conditions. This is partially due to biological diversity between different patients or donors, and can be further influenced by sample collection, as well as by miRNA isolation and storage. Therefore, stably expressed miRNA were investigated in plasma samples. These results demonstrated that miR-16-3p and miR-191-5p were expressed in all samples and the expression level of each of them was very consistent, both in plasma samples from patients and from healthy donors (Fig. 2A). MiR-191-5p was expressed at average Ct $=$ 29.4 and Ct $=$ 30.1 in patients and donors, respectively. MiR-16-3p was expressed at average Ct $=$ 22.7 and Ct $=$ 22.6 in patients and donors, respectively. These two miRNAs are therefore suitable for normalization of early- and late-expressed circulating miRNAs. In this study, we performed data normalization with a mean Ct-value of these two miRNAs in order to generate more robust data set.

Figure 2.

Expression profiling of selected miRNAs in plasma samples of LUAD patients and healthy donors. (A) Two control miRNAs, miR-16-3p and miR-191-5p, were chosen for data normalization based on their stable expression in all samples and in both cohorts. Expression levels of control miRNAs are presented as interleaved box and whiskers and as double gradient heat map. (B) Circulating miRNAs ( $n=$ 18) de-regulated in plasma samples of LUAD patients compared to tumor-free donors ( $n=$ 18 for each group). Ten out of eighteen miRNAs were detectable in plasma samples (upper panel); four were significantly deregulated in LUAD patients compared to healthy donors (rectangles and lower panel). Higher $\Delta$ Ct levels denote higher miRNA expression. ${}^{*}$ , $p<$ 0.05; ${}^{**}$ , $p<$ 0.01.

3.4 Expression profile of circulating miRNA in plasma samples of LUAD patients

Based on the miRNA expression data from our in vitro experiments, we chose the most prominently deregulated miRNAs ( $\Delta\Delta$ Ct $\geqslant$ 2.5; Table 1) and measured their expression in plasma samples of LUAD patients and healthy donors ( $n=$ 18 for each group). In addition to the two miRNAs used for normalization (miR-16-5p and miR-191-5p), we analyzed 18 miRNAs that were significantly deregulated in LUAD A549 cells (8 were up- and 10 down-regulated in A549 cells compared to BE cells). Eight of them (miR-27a-5p, miR-135b-5p, miR-196a-5p, miR-196b-5p, miR-200c-5p, miR-429, miR-450a-5p, and miR-708-5p) were not detectable in plasma samples in either LUAD patients nor in healthy donors. Out of ten detectable miRNAs, six miRNAs (miR-10b-5p, miR-34a-5p, miR-141-3p, miR-155-5p, miR200b-3p, and miR-205-5p) were expressed in both groups at a similar level (Suppl. Fig. 2), whereas four miRNAs (miR-15b-3p, miR-148a-3p, miR-193b-3p, and miR-195-5p) were significantly deregulated in plasma samples of LUAD patients compared to donors (Fig. 2B, Suppl. Table 1). Three miRNAs (miR-15b-3p, miR148a-3p, and miR-193b-3p) were down-regulated, whereas miR-195-5p was up-regulated.

Table 4
Annotation of published target genes by PANTHER overrepresentation test for pathway classification

PANTHER (16.0) pathways	H. sapiens (Ref.)	Input from this study
	#	#	Expected value	Fold enrichment	$\pm$	Raw $P$ value	FDR
Unclassified	17977	42	74.19	0.57	$-$	4.87E-17	8.13E-15
Cadherin signaling pathway	164	4	0.68	5.91	$+$	5.11E-03	3.55E-02
Inflammation mediated by chemokine and cytokine	255	7	1.05	6.65	$+$	1.03E-04	1.07E-03
signaling pathway
FGF signaling pathway	123	5	0.51	9.85	$+$	1.85E-04	1.72E-03
PDGF signaling pathway	147	6	0.61	9.89	$+$	4.02E-05	5.16E-04
Ras pathway	73	3	0.30	9.96	$+$	3.80E-03	2.88E-02
Apoptosis signaling pathway	118	5	0.49	10.27	$+$	1.54E-04	1.51E-03
B cell activation	70	3	0.29	10.38	$+$	3.39E-03	2.69E-02
Gonadotropin-releasing hormone receptor pathway	231	10	0.95	10.49	$+$	5.58E-08	1.86E-06
VEGF signaling pathway	68	3	0.28	10.69	$+$	3.13E-03	2.61E-02
TGF-beta signaling pathway	102	5	0.42	11.88	$+$	7.97E-05	8.87E-04
Alzheimer disease-presenilin pathway	127	7	0.52	13.35	$+$	1.33E-06	3.17E-05
p53 pathway	89	5	0.37	13.61	$+$	4.29E-05	5.12E-04
CCKR signaling map	172	10	0.71	14.09	$+$	3.85E-09	2.14E-07
T cell activation	86	6	0.35	16.90	$+$	2.17E-06	4.02E-05
Interleukin signaling pathway	86	6	0.35	16.90	$+$	2.17E-06	3.62E-05
Angiogenesis	175	13	0.72	18.00	$+$	7.96E-13	6.65E-11
Axon guidance mediated by Slit/Robo	25	2	0.10	19.38	$+$	5.48E-03	3.66E-02
PI3 kinase pathway	57	5	0.24	21.25	$+$	5.61E-06	8.51E-05
Cell cycle	22	2	0.09	22.03	$+$	4.35E-03	3.15E-02
Hypoxia response via HIF activation	30	3	0.12	24.23	$+$	3.34E-04	2.94E-03
Notch signaling pathway	44	5	0.18	27.53	$+$	1.72E-06	3-60E-05
p53 pathway feedback loops 2	51	6	0.21	28.51	$+$	1.24E-05	3.44E-06
Insulin/IGF pathway-mitogen activated protein	33	4	0.14	29.37	$+$	1.58E-05	2.20E-04
kinase kinase/MAP kinase cascade
Insulin/IGF pathway-protein kinase B signaling cascade	39	6	0.16	37.28	$+$	2.89E-08	1.21E-06

As a reference (Ref.), all genes in database for Homo sapiens were used and Fisher’s exact test with data correction for false discovery rate (FDR) was performed. Panther Classification System 16.0. (www.pantherdb.org; January 2021).

Figure 3.

Regulatory network and putative interactions of differentially expressed miRNAs and their target genes and pathways. (A) Regulatory network of four deregulated miRNAs and their direct target genes validated so far. Red diamonds, yellow diamonds, and green rectangles represent up-regulated miRNA, down-regulated miRNAs, and associated target genes, respectively. For data visualization, Cytoscape (version 3.8.2) was used. (B) Based on their direct target genes, deregulated miRNAs were annotated to particular pathways by performing Panther Overrepresentation Test. The down-regulated and up-regulated miRNAs are shown in green-line and red-lined rectangles, respectively. MiR-15b-3p was not annotated to any special pathway. Analysis was done by using the Panther Classification System 16.0. (www.pantherdb.org).

Figure 4.

Putative interactions and molecular functions of validated target genes for deregulated miRNAs. Analysis was performed by the Panther Classification System 16.0. (www.pantherdb.org).

3.5 Target genes of deregulated miRNAs

In order to shed light upon the target genes regulated by these four miRNAs (miR-15b-3p, miR148a-3p, miR-193b-3p, and miR-195-5p), we performed literature screening and regulatory network analyses for genes described so far as direct targets of these miRNAs. Upon screening (https://pubmed.ncbi.nlm.nih.gov; searching criteria: hsa-miR-X-Xp $+$ target gene; January 2021), only experimentally validated human target genes with direct binding site for the particular miRNA were taken into account. Based on the present literature and final confirmation of a direct binding site by using luciferase activity assay combined with appropriate mutagenesis, it was found that 107 validated genes are regulated by these four miRNAs (Fig. 3A). MiR-195-5p regulates the highest number of different target genes (61), followed by miR-148a-3p (42), miR-193-3p (3), and miR-15b-3p (1). Out of 107 validated target genes, five (PTEN, CXCR4, IGF1R, FGF2, and PD-L1) were described to be regulated with more than one of four deregulated miRNAs. After performing the gene annotation analysis by using the Panther Classification System (www.pantherdb.org), all target genes were assigned to 24 different pathways (Table 4, Fig. 3B). MiR-195-5p was involved in 23 of them, followed by miR-148a-3p (21), and miR-193-3p (8). The single target gene (DYNLT1, dynein light chain Tctex-type 1) of miR-15b-3p was not annotated to any specific pathway; however, it has been recently described to play a role in cell proliferation, migration, invasion and inhibition of apoptosis in vitro and in vivo [16]. Concerning their molecular function, 107 direct target genes were annotated to five main groups: binding, catalytic activity, molecular function regulator, molecular transducer activity, and transporter activity. Annotations of different subgroups are summarized in Fig. 4.

4. Discussion

Early detection of LUAD is still very challenging and the need to identify specific molecular tumor biomarkers, especially in serum, plasma, or in the whole blood is essential. MiRNAs have potential as molecular tumor biomarkers for early diagnosis and prognosis and for the evaluation of treatment efficacy. Such biomarkers might even help to establish personalized therapies for LUAD patients. In the study presented herein, expression levels of miRNAs in malignant A549 LUAD cells and primary, non-malignant BE cells were compared. Based on these in vitro data, the expression levels of the most deregulated miRNAs were analyzed in plasma samples of LUAD patients and healthy controls.

To determine the expression levels and deregulation of miRNAs in different samples, RT-qPCR data must be normalized to one or more miRNAs stably expressed over all samples. In the literature, different miRNA normalization methods have been described, but there is no agreement about the one single optimal strategy. Therefore, the normalization of miRNA qPCR data requires a well-considered decision to accurately handle the obtained data [17]. For in vitro data in this study, we used the MCR method for miRNA normalization, as this method has been described as the most reliable out of seven different methods [15]. The MCR method focuses on a restricted set of miRNAs, taking into account miRNAs that are fully detected in all analyzed samples, thereby excluding the miRNAs with non-determined values (which are e.g. included in the non-restricted mean-centering method).

Based on our in vitro data, we selected a panel of 18 miRNAs with $\Delta\Delta$ Ct $\geqslant$ 2.5 for analysis in plasma samples from LUAD patients. Plasma samples were used since it has been shown that in serum samples additional miRNAs are released from red and white blood cells due to haemolysis, as well as from platelets during the coagulation process [18]. Normalization of expression data for circulating miRNA is very challenging. Whereas in cell culture, biological variability is more controlled, comparing miRNA expression levels in samples from patients is prone to higher variability. To this end, miR-191-5p and miR-16-3p were selected for data normalization, as these two miRNAs showed stable expression levels over all individuals and in both cohorts. Stable expression levels of those two miRNAs in plasma suggest their potential as control- and normalization-miRNAs for different diseases. However, their expression levels in other biological fluids of LUAD patients (see review by Gayosso-Gomez et al. [19]), or of cancer patients in general, need to be further validated.

In this study, special care has been taken to avoid haemolysis and to remove all blood cells from plasma samples. These two factors frequently lead to increased concentration of miRNAs highly expressed in red and white blood cells or in platelets. Thus, upon blood collection, two consecutive centrifugation steps were performed in order to remove any residual blood cells. Sample collection and sample preparation are still two major drawbacks in performing miRNA expression profiling in biological samples, and their standardization is urgently needed. Another big challenge is the miRNA nomenclature. It is frequently not clear from the published studies which strand of particular miRNA was investigated. This problem should not be trivialized because it is well known that 5p and 3p miRNA molecules are biologically different in terms of stability, which has an impact on their function and ability to bind to the RISC complex and, finally, on the binding to their target genes.

In this study, we detected four miRNAs differentially expressed in plasma samples of LUAD patients compared to healthy donors. Human miR-15b-3p is located at chromosome 3q25.33. To our best knowledge, there are no data about the expression of miR-15b-3p in lung cancer. Several studies have looked at either the 5p strand of miR-15b or, mostly, the active strand was not specified at all. In one study, miR-15b-5p was described as down-regulated in exosomes from plasma of squamous cell carcinoma patients compared to healthy controls [20]. In another study, miR-15b-5p was up-regulated in the serum of non-small cell lung cancer (NSCLC) patients [21]. Shi et al. described down-regulated serum expression of miR-15b in NSCLC patients in comparison to healthy donors and benign lung disease, the miR-15b expression being lower in stage III–IV patients than in stage I–II [22]. Thus, the authors suggested miR-15b, together with miR-22, as reference markers for early NSCLC diagnosis. Furthermore, miR-15b appears to play a role in chemotherapy resistance of lung cancer, as the sensitivity to cisplatin of advanced NSCLC patients on cisplatin-based chemotherapy was higher in patients with high expression of miR-15b [23]. In advanced NSCLC, serum levels of miR-15b have been shown to change pre- and post-chemotherapy, these changes being associated with chemotherapeutic response [23]. This is further supported by studies showing that miR-15b is involved in DNA-repair regulation via Wip1/ATM/Chk1/p53 signaling [24], and via expression inhibition of PEBP4 [25]. On the other side, miR-15b was found to be more highly expressed in human NSCLC tissue and cells than in normal tissue/cells, thereby controlling the viability, cell cycle, and migration via targeting TIMP2 [26]. In summary, data about miR-15b-3p expression in lung cancer are lacking, and data regarding the role of miR-15b in cancer in general and in lung cancer in special are controversial. These discrepancies may be due to the fact that miR-15b is expressed in platelets and its level depends on sample preparation. This can lead to large differences in expression of circulating miR-15b in whole blood, serum, or in plasma.

Human miR-148a-3p is located at the 7p15.2 chromosomal region and belongs to the miR-148/152 family. In our study, miR-148a-3p was down-regulated in plasma samples of LUAD patients compared to healthy donors, consistent with published data. MiR-148a-3p has been described as down-regulated in tumor tissue, serum, and plasma of NSCLC patients, frequently showing a significant correlation with tumor stage but not with survival [27, 28, 29, 30]. Regarding the molecular mechanism, it has been shown that miR-148a-3p inhibits proliferation and migration, and induces apoptosis in NSCLC cell via the PI3K/AKT signalling pathway by targeting ERBB3 and ITGA5 genes [31]. Furthermore, miR-148a-3p binds to the lncRNA HOXA11-AS, which furthermore regulates DNMTA (a central DNA methyl transferase in mammalian cells) expression and facilitates NSCLC tumorigenesis in vitro and in vivo [32]. Together, all these data indicate a tumor suppressive role of miR-148a-3p and its downregulation in plasma samples of LUAD patients indicates its direct involvement in LUAD tumorigenesis.

Human miR-193b-3p is located on chromosome 16p13.12 and, based on literature, has tumor-suppressive function in human cancers, including lung cancer [33, 34, 35]. In our study, miR-193b-3p was the most down-regulated miRNA. Thus, our data are in line with data published, including that by Hu et al., who described lower level of hsa-miR-193b-3p in NSCLC cells compared to non-malignant cells [36]. They also showed that inhibition of mir-193b in A549 cells led to increased cell proliferation, migration, and invasion. These effects were based on repression of cyclin D1 and urokinase-type plasminogen activator, suggesting the tumor suppressive role of miR-193b in NSCLC [36]. Furthermore, it has been shown that both 5p and 3p strands of miR-193b were down-regulated in metastatic LUAD cells. Ectopic expression of both miR-193b strands, based on NSCLC cell transfection with miR-193b-5p and -3p precursors, inhibited proliferation and metastatic potential of lung cancer cells [34]. In their integrative analysis of miRNAs in NSCLC, based on two publicly available datasets and validated with RT-qPCR, Zhang et al. described miR-193b-3p as up-regulated in NSCLC to a significant degree [37]. However, this data describe the expression of miR-193b-3p in NSCLC tumor tissue and not of circulating miR-193b-3p, which may be the reason for this discrepancy. Recently, miR-193b-3p has been described in plasma exosomes as down-regulated in e-cigarette smokers compared to traditional smokers [38]. Although all patients in our cohort were active or ex-smokers, explicit information about e-cigarette smoking is lacking.

Located on chromosome 17p13.1, miR-195-5p has been shown to be aberrantly expressed in various cancers such as breast cancer [39], renal carcinoma [40], colorectal cancer [41], and lung cancer [42, 43, 44, 45, 46, 47]. In our study, we showed significant up-regulation of miR-195-5p in plasma samples of LUAD patients, with a very constant expression level both in patients and in healthy donors. This is in contrast to some studies reporting down-regulation of miR-195-5p in lung cancer [44, 45, 46, 47]. However, it should be stressed that these studies analyzed miR-195-5p expression in lung cancer tissue or cell lines, whereas the expression of miR-195-5p in plasma of NSCLC patients has not been analyzed thus far. In NSCLC cell lines A549 and H1299, miR-195-5p inhibited proliferation and induced apoptosis by targeting CEP55 [42]. As further targets of miR-195-5p in lung cancer cells, CHEK1 [43], CPNE1 [44], CIAPIN1 [45], GLUT3 [46], FOXK1 [47], and PDK4 [48] have been described. Together with five other miRNAs, miR-195-5p has been also identified as smoking-related miRNA, thus perhaps being at least partially responsible for cigarette smoke-induced malignant transformation of bronchial epithelial cells [49]. Expression of miR-195-5p in lung cells can be regulated by lncRNA CASC9 [48] and by circular RNA circMYLK [46], as recently described. Altogether, these data indicate the complexity of miR-195-5p regulation, both in cancerous and in non-cancerous tissues and cells, and elicit need for further investigations.

We found five overlapping target genes (PTEN, CXCR4, IGF1R, FGF2, and PD-L1) that are directly regulated by at least two of four deregulated miRNAs. These genes have different cellular functions and are involved in several signalling pathways in non-malignant and various malignant cells. Phosphatase and tensin homologue deleted on chromosome ten (PTEN) is a tumor suppressor gene frequently lost in variety of tumors (for review see [50]). PTEN inhibits cell proliferation, survival, growth, and migration by blocking the phosphatidylinositol 3 kinase (PI3K) signaling pathway and inhibiting inter alia, the membrane recruitment and activation of AKT. Tumor-suppressive function of PTEN can be regulated by its binding partners and by its localization [50]. In addition, PTEN can be exported from the cell and can be transferred from one cell to another in form of exosomes [51] and its translational variant (PTEN-Long) has also been detected in human serum and plasma [52].

C-X-C chemokine receptor 4 (CXCR4), a 7-trans- membrane G-protein coupled chemokine receptor, is known for its numerous pleiotropic functions under physiological and pathophysiological conditions. Through interaction with its ligand CXCL12, the CXCR4 plays a role in a number of cellular processes and is expressed in over 30 types of malignant tumors (for review see [53]). CXCR4 functions in autoimmunity, immunodeficiency, and cancer are well documented and it has been identified as prognostic marker for breast, ovarian, and pancreatic adenocarcinoma [54]. It is thought that the CXCL12/CXCR4 axis has a pro-tumorigenic role in NSCLC by supporting tumor growth, enhancing angiogenesis and metastases, and promoting chemo-resistance [55]. Although it has been shown that miR-195-5p directly regulates the CXCR4 in the heart failure mice model, thereby inactivating the JAK/STAT pathway [56], in lung cancer, CXCR4 regulation by miR-193b-3p or miR-195-5p has not been described.

Insulin like growth factor 1 receptor (IGF1R), a transmembrane tetrameric receptor containing two extracellular alpha and two transmembrane beta subunits, binds insulin-like growth factors (IGFs) and insulin in a variety of cells and tissues. IGF1R is overexpressed in many tumors and plays an important role in anti-apoptosis, thus enhancing cancer cell survival and proliferation [57]. Furthermore, IGF1R activation is connected to several metastatic features of cancer cells, including tumor angiogenesis, degradation of extracellular matrix, and loss of anchorage-dependent growth [58]. However, it is not clear how deregulation of miR-148a-3p and miR-195-5p might influence the expression of IGF1R and its potential role in NSCLC development.

Fibroblast growth factor 2 (FGF2) is frequently increased in different cancer entities, including NSCLC [59, 60], acting as regulator of cancer cell proliferation, invasion, angiogenesis, and metastases [61, 62, 63]. In NSCLC cells, correlations between FGF2 expression and different miRNAs including miR-182, miR-210, miR-497-5p, and miR-152 has been shown [63, 64, 65, 66]. Regulation of FGF2 by miR-148a-3p or miR-195-5p has not been put into context with NSCLC yet and should be further investigated.

Programmed cell death ligand 1 (PD-L1) is an approved predictive biomarker for lung cancer. PD-L1 binds to PD-1 expressed on T-cells, B-cells, dendritic cells and natural killer T-cells to suppress anti-cancer immunity. PD-L1 expression is regulated by different signaling pathways, as well as by transcriptional and epigenetic factors, and miRNAs (for review see [67]). Improved survival of a subset of lung cancer patients have been observed upon immunotherapy targeted against PD-L1 and its receptor PD-1 [68]. In colorectal cancer, direct regulation of PD-L1 by miR-148a-3p has been shown [69]. In our patient cohort, showing significant down-regulation of miR-148a-3p, PD-L1 was positive (30–100% of all tumor cells) in eight (44%) patients, however, direct regulation should be experimentally validated. Altogether, pathway analysis indicates that the target genes of our four deregulated miRNAs are involved in mechanisms and pathways essential in cancer development and progression.

As a limitation of this study, our preselection of deregulated miRNAs was performed in one single LUAD cell line (A549). This cell line is widely used as an in vitro model and was chosen to provide an overview of miRNAs deregulated in malignant cells in comparison to non-malignant BE cells. In our pilot experiments (data not shown), we also compared miRNA expression profiles of H460 LUAD cells in comparison to BE cells. However, none of 10 miRNAs deregulated in A549 cells was detectable in H460 cells. Although this is a strong indication that in H460 cells, these miRNAs might be also down-regulated, it is not certain whether it is a limitation of the methods or if they are not expressed at all. Variations in miRNA expression based on different genetic background of these two LUAD cell lines are expected, since A549 cell line was established from a lung carcinomatous tissue, whereas H460 cells represent a hypotriploid human cell line derived from pleural fluid of patient with large cell lung cancer. Our study found a considerable overlap between miRNAs deregulated in A549 cells and in LUAD patient’s plasma samples, but it also opened the door for discussion, especially regarding the variations in miRNA expression between different cell models and LUAD patients.

Next, this study was retrospective and the number of patients and controls was quite small. We relied on an explorative approach, based on in vitro experiments with one adenocarcinoma cell line, and this might not be representative for all adenocarcinoma cells. We did not investigate other types of lung tumors or other malignant diseases for our detected miRNAs and, therefore, cannot assess the specificity for LUAD. Thus, further studies are needed to clarify these important issues.

Early cancer detection is essential for reduction of cancer mortality worldwide. Based on benefits for patients, low-dose computed tomography (CT) scan is recommended by different international guidelines as a screening method for early lung cancer detection, especially for high risk patients [70]. However, the CT screening has some limitations and possible drawbacks, such as additional radiation exposure. Furthermore, overdiagnosis can lead to unnecessary invasive procedures, which are connected with increased risk for patients. There are still no well-established biomarkers for early lung cancer detection, but this is under intensive investigation, with a focus on miRNAs as promising candidates. As an advantage of our approach, we focused in patient samples on miRNAs pre-selected in vitro. Patient samples are unique; functional analyses of miRNAs and their target genes are very costly and time consuming. Thus, pre-selection and focusing on specific miRNAs and target genes is a benefit to address these challenges. MiRNAs described in this study might be important molecular markers for lung adenocarcinoma, both for early diagnosis and possibly for reexamination of patients after surgery or chemotherapy. However, this is still speculative and must be verified in future prospective studies.

In conclusion, the analysis of differentially expressed miRNAs in A549 cells compared to primary, non-malignant BE cells led to the identification of four de-regulated miRNAs in the plasma of LUAD patients. We suggest miR-15b-3p, miR-148a-3p, miR-193b-3p, and miR-195-5p as potential plasma markers for LUAD. Their direct target genes play an essential role in tumor development and growth. Improved understanding of molecular mechanisms and the role of specific molecular biomarkers will lead to improved detection of early stage tumors and, ultimately, to improved biomarker-directed therapies for LUAD patients.

Author contributions

Conception: Irina Robinson, Horst Olschewski, Andelko Hrzenjak.

Interpretation or analysis of data: Irina Robinson, Alexandra Bertsch, Andelko Hrzenjak.

Preparation of the manuscript: Irina Robinson, Andelko Hrzenjak, Katharina Leithner.

Revision for important intellectual content: Irina Robinson, Andelko Hrzenjak, Katharina Leithner, Philipp Stiegler, Horst Olschewski.

Supervision: Andelko Hrzenjak.

Funding

This study was supported by the Ingrid Shaker Nessmann Cancer Research Association (grant to A.H.). The founder did not have any role in study design, data collection and analysis, interpretation of results, decision to publish, or preparation of the manuscript.

Declaration of competing interest

The authors report no declarations of interest.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-210205.

sj-pdf-1-cbm-10.3233_CBM-210205.pdf - Supplemental material

Supplemental material, sj-pdf-1-cbm-10.3233_CBM-210205.pdf

Footnotes

Acknowledgments

We thank all patients and donors for participating in this study. We appreciate excellent technical assistance of Theodora Goessler and kind assistance of Core Facility for Molecular Biology, Center of Medical Research, Medical University of Graz.

References

Travis

W.D.

Brambilla

Nicholson

A.G.

et al., The 2015 world health organization classification of lung tumors: Impact of genetic, clinical and radiologic advances since the 2004 classification, Journal of Thoracic Oncology 10 (2015), 1243–1260.

Arbour

K.C.

and Riely

G.J.

, Systemic therapy for locally advanced and metastatic non-small cell lung cancer: A review, Journal of the American Medical Association 322 (2019), 764–774.

Calvayrac

Pradines

Pons

et al., Molecular biomarkers for lung adenocarcinoma, European Respiratory Journal 49 (2017).

Kalinke

Thakrar

and Janes

S.M.

, The promises and challenges of early non-small cell lung cancer detection: Patient perceptions, low-dose CT screening, bronchoscopy and biomarkers, Molecular Oncology (2020), 2544–2564.

Bartel

D.P.

, MicroRNAs: Genomics, biogenesis, mechanism, and function, Cell 116 (2004), 281–297.

Volinia

Calin

G.A.

Liu

C.G.

et al., A microRNA expression signature of human solid tumors defines cancer gene targets, Proceedings of the National Academy of Sciences U S A 103 (2006), 2257–2261.

Zhu

Kudo

Huang

et al., Frontiers of MicroRNA signature in non-small cell lung cancer, Frontiers in Cell and Developmental Biology 9 (2021), 643942.

Petrek

and Yu

A.M.

, MicroRNAs in non-small cell lung cancer: Gene regulation, impact on cancer cellular processes, and therapeutic potential, Pharmacology Research & Perspectives 7 (2019), e00528.

El Founini

Chaoui

Dehbi

et al., MicroRNAs: Key regulators in lung cancer, Microrna (2021).

10.

Huang

Zhu

Lin

et al., The potential diagnostic value of extracellular vesicle miRNA for human non-small cell lung cancer: A systematic review and meta-analysis, Expert Review of Molecular Diagnostics 21 (2021), 823–836.

11.

Ahn

Y.H.

and Ko

Y.H.

, Diagnostic and therapeutic implications of microRNAs in non-small cell lung cancer, International Journal of Molecular Sciences 21 (2020).

12.

Fulcher

M.L.

Gabriel

Burns

K.A.

et al., Well-differentiated human airway epithelial cell cultures, Methods in Molecular Medicine 107 (2005), 183–206.

13.

World Medical

, World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects, Journal of the American Medical Association 310 (2013), 2191–2194.

14.

Detterbeck

F.C.

Boffa

D.J.

Kim

A.W.

et al., The eighth edition lung cancer stage classification, Chest 151 (2017), 193–203.

15.

Wylie

Shelton

Choudhary

et al., A novel mean-centering method for normalizing microRNA expression from high-throughput RT-qPCR data, BMC Research Notes 4 (2011), 555.

16.

Wei

Peng

Yang

et al., Exosomal transfer of miR-15b-3p enhances tumorigenesis and malignant transformation through the DYNLT1/Caspase-3/Caspase-9 signaling pathway in gastric cancer, Journal of Experimental & Clinical Cancer Research 39 (2020), 32.

17.

Schwarzenbach

Nishida

Calin

G.A.

et al., Clinical relevance of circulating cell-free microRNAs in cancer, Nature Reviews Clinical Oncology 11 (2014), 145–156.

18.

Wang

Yuan

Cho

J.H.

et al., Comparing the MicroRNA spectrum between serum and plasma, PLoS One 7 (2012), e41561.

19.

Gayosso-Gomez

L.V.

and Ortiz-Quintero

, Circulating MicroRNAs in blood and other body fluids as biomarkers for diagnosis, prognosis, and therapy response in lung cancer, Diagnostics (Basel) 11 (2021).

20.

Jin

Chen

et al., Evaluation of tumor-derived exosomal miRNA as potential diagnostic biomarkers for early-stage non-small cell lung cancer using next-generation sequencing, Clinical Cancer Research 23 (2017), 5311–5319.

21.

Fan

Teng

et al., Identification of serum miRNAs by nano-quantum dots microarray as diagnostic biomarkers for early detection of non-small cell lung cancer, Tumour Biology 37 (2016), 7777–7784.

22.

Shi

G.L.

Chen

Sun

et al., Significance of serum MicroRNAs in the auxiliary diagnosis of non-small cell lung cancer, Clinical Laboratory 63 (2017), 133–140.

23.

Shi

G.L.

Zhang

X.Y.

Chen

et al., Prognostic Significance of Serum miR-22, miR-125b, and miR-15b in Non-Small Cell Lung Cancer Patients, Clinical Laboratory 66 (2020).

24.

Rahman

Lovat

Romano

et al., miR-15b/16-2 regulates factors that promote p53 phosphorylation and augments the DNA damage response following radiation in the lung, Journal of Biological Chemistry 289 (2014), 26406–26416.

25.

Zhao

Zhang

Yao

et al., miR-15b regulates cisplatin resistance and metastasis by targeting PEBP4 in human lung adenocarcinoma cells, Cancer Gene Therapy 22 (2015), 108–114.

26.

Wang

Zhan

Jin

et al., MicroRNA-15b promotes proliferation and invasion of nonsmall cell lung carcinoma cells by directly targeting TIMP2, Oncology Reports 37 (2017), 3305–3312.

27.

Chen

Min

Zhang

et al., Decreased miRNA-148a is associated with lymph node metastasis and poor clinical outcomes and functions as a suppressor of tumor metastasis in non-small cell lung cancer, Oncology Reports 30 (2013), 1832–1840.

28.

Song

Wang

et al., MicroRNA-148a suppresses epithelial-to-mesenchymal transition by targeting ROCK1 in non-small cell lung cancer cells, Molecular and Cellular Biochemistry 380 (2013), 277–282.

29.

Huang

M.X.

, Down-expression of circulating micro ribonucleic acid (miRNA), -148/152 family in plasma samples of non-small cell lung cancer patients, Journal of Cancer Research and Therapeutics 12 (2016), 671–675.

30.

and Xue

, MicroRNA-148a suppresses proliferation and invasion potential of non-small cell lung carcinomas via regulation of STAT3, OncoTargets and Therapy 10 (2017), 1353–1361.

31.

Zhang

Shen

et al., Cytological effects of honokiol treatment and its potential mechanism of action in non-small cell lung cancer, Biomedicine & Pharmacotherapy 117 (2019), 109058.

32.

Bai

Lang

Zhao

et al., Long non-coding RNA HOXA11-AS promotes non-small cell lung cancer tumorigenesis through microRNA-148a-3p/DNMT1 regulatory axis, OncoTargets and Therapy 12 (2019), 11195–11206.

33.

Bhayadia

Krowiorz

Haetscher

et al., Endogenous tumor suppressor microRNA-193b: Therapeutic and prognostic value in acute myeloid leukemia, Journal of Clinical Oncology 36 (2018), 1007–1016.

34.

Choi

K.H.

Shin

C.H.

Lee

W.J.

et al., Dual-strand tumor suppressor miR-193b-3p and -5p inhibit malignant phenotypes of lung cancer by suppressing their common targets, Bioscience Reports 39 (2019).

35.

Khordadmehr

Shahbazi

Sadreddini

et al., miR-193: A new weapon against cancer, Journal of Cellular Physiology 234 (2019), 16861–16872.

36.

Liu

et al., MicroRNA-193b modulates proliferation, migration, and invasion of non-small cell lung cancer cells, Acta Biochimica et Biophysica Sinica (Shanghai) 44 (2012), 424–430.

37.

Zhang

et al., Integrative analysis of mRNA and miRNA expression profiles reveals seven potential diagnostic biomarkers for nonsmall cell lung cancer, Oncology Reports 43 (2020), 99–112.

38.

Singh

K.P.

Maremanda

K.P.

et al., Exosomal microRNAs are novel circulating biomarkers in cigarette, waterpipe smokers, E-cigarette users and dual smokers, BMC Medical Genomics 13 (2020), 128.

39.

Yang

Xing

Zheng

et al., The circRNA circAGFG1 acts as a sponge of miR-195-5p to promote triple-negative breast cancer progression through regulating CCNE1 expression, Molecular Cancer 18 (2019), 4.

40.

Chen

Wang

Yao

et al., miR-195-5p is critical in REGgamma-mediated regulation of wnt/beta-catenin pathway in renal cell carcinoma, Oncotarget 8 (2017), 63986–64000.

41.

Feng

Zhang

Sun

et al., GDPD5, a target of miR-195-5p, is associated with metastasis and chemoresistance in colorectal cancer, Biomedicine & Pharmacotherapy 101 (2018), 945–952.

42.

Luo

Pan

Jin

et al., MiR-195-5p inhibits proliferation and induces apoptosis of non-small cell lung cancer cells by targeting CEP55, OncoTargets and Therapy 12 (2019), 11465–11474.

43.

Feng

Zhang

et al., MicroRNA biomarker hsa-miR-195-5p for detecting the risk of lung cancer, International Journal of Genomics 2020 (2020), 7415909.

44.

Liu

Zhang

et al., MiR-195-5p is a potential factor responsible for CPNE1 differential expression between subtypes of non-small cell lung cancer, Journal of Cancer 11 (2020), 2610–2620.

45.

Zheng

Chen

et al., MiRNA-195-5p functions as a tumor suppressor and a predictive of poor prognosis in non-small cell lung cancer by directly targeting CIAPIN1, Pathology and Oncology Research 25 (2019), 1181–1190.

46.

Xiong

Wang

et al., Circular RNA MYLK promotes glycolysis and proliferation of non-small cell lung cancer cells by sponging miR-195-5p and increasing glucose transporter member 3 expression, Cancer Management and Research 12 (2020), 5469–5478.

47.

Long

and Wang

, miR-195-5p suppresses lung cancer cell proliferation, migration, and invasion via FOXK1, Technology in Cancer Research & Treatment 19 (2020), 1533033820922587.

48.

Wang

H.R.

Guo

X.Y.

Liu

X.Y.

et al., Down-regulation of lncRNA CASC9 aggravates sepsis-induced acute lung injury by regulating miR-195-5p/PDK4 axis, Inflammation Research 69 (2020), 559–568.

49.

Wang

X.F.

Ouyang

et al., MicroRNA and mRNA interaction network regulates the malignant transformation of human bronchial epithelial cells induced by cigarette smoke, Frontiers in Oncology 9 (2019), 1029.

50.

Hopkins

B.D.

Hodakoski

Barrows

et al., PTEN function: The long and the short of it, Trends in Biochemical Sciences 39 (2014), 183–190.

51.

Putz

Howitt

Doan

et al., The tumor suppressor PTEN is exported in exosomes and has phosphatase activity in recipient cells, Science Signaling 5 (2012), ra70.

52.

Hopkins

B.D.

Fine

Steinbach

et al., A secreted PTEN phosphatase that enters cells to alter signaling and survival, Science 341 (2013), 399–402.

53.

Pozzobon

Goldoni

Viola

et al., CXCR4 signaling in health and disease, Immunology Letters 177 (2016), 6–15.

54.

Zhao

Guo

Zhao

et al., CXCR4 over-expression and survival in cancer: A system review and meta-analysis, Oncotarget 6 (2015), 5022–5040.

55.

Wald

, CXCR4 based therapeutics for non-small cell lung cancer (NSCLC), Journal of Clinical Medicine 7 (2018).

56.

Shen

Zhang

Lee

et al., Down-regulated microRNA-195-5p and up-regulated CXCR4 attenuates the heart function injury of heart failure mice via inactivating JAK/STAT pathway, International Immunopharmacology 82 (2020), 106225.

57.

Ouban

Muraca

Yeatman

et al., Expression and distribution of insulin-like growth factor-1 receptor in human carcinomas, Human Pathology 34 (2003), 803–808.

58.

Werner

and Bruchim

, The insulin-like growth factor-I receptor as an oncogene, Archives of Physiology and Biochemistry 115 (2009), 58–71.

59.

Farhat

F.S.

Tfayli

Fakhruddin

et al., Expression, prognostic and predictive impact of VEGF and bFGF in non-small cell lung cancer, Critical Reviews in Oncology/Hematology 84 (2012), 149–160.

60.

Berger

Setinek

Mohr

et al., Evidence for a role of FGF-2 and FGF receptors in the proliferation of non-small cell lung cancer cells, International Journal of Cancer 83 (1999), 415–423.

61.

Lau

M.T.

W.K.

and Leung

P.C.

, Fibroblast growth factor 2 induces E-cadherin down-regulation via PI3K/Akt/mTOR and MAPK/ERK signaling in ovarian cancer cells, PLoS One 8 (2013), e59083.

62.

Rades

Seibold

N.D.

Gebhard

M.P.

et al., Fibroblast growth factor 2 is of prognostic value for patients with locally advanced squamous cell carcinoma of the head and neck, Strahlentherapie und Onkologie 190 (2014), 68–74.

63.

Cheng

Tan

et al., MiR-152 suppresses the proliferation and invasion of NSCLC cells by inhibiting FGF2, Experimental & Molecular Medicine 46 (2014), e112.

64.

Stenvold

Donnem

Andersen

et al., Stage and tissue-specific prognostic impact of miR-182 in NSCLC, BMC Cancer 14 (2014), 138.

65.

Fan

Chang

et al., miR-210 transferred by lung cancer cell-derived exosomes may act as proangiogenic factor in cancer-associated fibroblasts by modulating JAK2/STAT3 pathway, Clinical Science (London) 134 (2020), 807–825.

66.

Huang

Wang

Liu

et al., MicroRNA-497-5p inhibits proliferation and invasion of non-small cell lung cancer by regulating FGF2, Oncology Letters 17 (2019), 3425–3431.

67.

Chen

Jiang

C.C.

Jin

et al., Regulation of PD-L1: A novel role of pro-survival signalling in cancer, Annals of Oncology 27 (2016), 409–416.

68.

Boyle

T.A.

Zhou

et al., PD-L1 expression in lung cancer, Journal of Thoracic Oncology 11 (2016), 964–975.

69.

Ashizawa

Okayama

Ishigame

et al., miRNA-148a-3p regulates immunosuppression in DNA mismatch repair-deficient colorectal cancer by targeting PD-L1, Molecular Cancer Research 17 (2019), 1403–1413.

70.

Jaklitsch

M.T.

Jacobson

F.L.

Austin

J.H.

et al., The American Association for Thoracic Surgery guidelines for lung cancer screening using low-dose computed tomography scans for lung cancer survivors and other high-risk groups, Journal of Thoracic and Cardiovascular Surgery 144 (2012), 33–38.

Circulating microRNAs as molecular biomarkers for lung adenocarcinoma

Abstract

BACKGROUND:

OBJECTIVE:

MATERIALS AND METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Cell culture

2.2 Patient cohort

2.3 Cytokeratin immunofluorescence

2.4 RNA isolation from cells and quality check

2.5 Reverse transcription

2.6 SYBR-Green based RT-qPCR

2.7 Normalization and statistical analysis of RT-qPCR data

2.8 Blood collection and miRNA isolation from plasma samples

2.9 Mutational analyses

2.10 Statistical calculations

3. Results

3.1 miRNA expression profiles in cell samples

Table 2 Clinicopathologic characteristics of LUAD patients

Table 4 Annotation of published target genes by PANTHER overrepresentation test for pathway classification

4. Discussion

Author contributions

Funding

Declaration of competing interest

Supplementary data

sj-pdf-1-cbm-10.3233_CBM-210205.pdf - Supplemental material

Footnotes

Acknowledgments

References

Table 2
Clinicopathologic characteristics of LUAD patients

Table 4
Annotation of published target genes by PANTHER overrepresentation test for pathway classification