The clinical significance and mechanism of microRNA-22-3p targeting TP53 in lung adenocarcinoma

Abstract

BACKGROUND:

At present, studies on MircoRNA-22-3p (miR-22-3p) in lung adenocarcinoma use a single method, lack multi-center validation and multi-method validation, and there is no big data concept to predict and validate target genes.

OBJECTIVE:

To investigate the expression, potential targets and clinicopathological significance of miR-22-3p in lung adenocarcinoma (LUAD) tissues.

METHODS:

LUAD formalin-fixed paraffin-embedded (FFPE) tumors and adjacent normal lung tissues were collected for real-time quantitative polymerase chain reaction (RT-qPCR). Collect miR-22-3p in LUAD and non-cancer lung tissue from high-throughput datasets, standardized mean difference (SMD) and area under the curve (AUC) of the comprehensive receiver operating curve (summary receiver operating characteristic cure, sROC curve) were calculated. Cell function experiments on A549 cells transfected with LV-hsa-miR-22-3p. Target genes were predicted by the miRwalk2.0 website and the resulting target genes were subjected to Gene Ontology (GO) pathway enrichment analysis and constructed to protein-protein interaction network. Finally, the protein expression level of the key gene TP53 was validated by searching The Human Protein Atlas (THPA) database to incorporate TP53 immunohistochemical results in LUAD.

RESULTS:

RT-qPCR result from 41 pairs of LUAD and adjacent lung tissues showed that miR-22-3p was downregulated in LUAD (AUC $=$ 0.6597, $p=$ 0.0128). Globally, a total of 838 LUADs and 494 non-cancerous lung tissues were included, and were finally combined into 14 platforms. Compared with noncancerous tissue, miR-22-3p expression level was significantly reduced in LUAD tissue (SMD $=-$ 0.32, AUC $=$ 0.72l); cell function experiments showed that miR-22-3p has inhibitory effects on cell proliferation, migration and invasion, and has promotion effect on apoptosis. Moreover, target genes prediction, GO pathway enrichment analysis and PPI network exhibited TP53 as a key gene of target gene of miR-22-3p; at last, a total of 114 high-throughput datasets were included, including 3897 LUADs and 2993 non-cancerous lung tissues, and were finally combined into 37 platforms. Compared with noncancerous tissue, TP53 expression level was significantly increased in LUAD (SMD $=$ 0.39, $p<$ 0.01) and it was verified by the protein expression data from THPA.

CONCLUSION:

Overexpression of miR-22-3p may inhibit LUAD cell proliferation, migration and invasion through TP53, and promote cell apoptosis.

Keywords

miR-22-3p lung adenocarcinoma target gene TP53

1. Introduction

Primary lung cancer is the deadliest cancer in the world [1, 2]. As the most common histological type of lung cancer, lung adenocarcinoma (LUAD) has become more prevalent over the past two decades [3]. Due to the development of precision medicine and non-small-cell lung cancer (NSCLC) therapy for genetic changes, many microRNAs (miRNAs) can be used as ideal markers for early diagnosis and prognosis prediction of lung cancer [4, 5], and are also potential therapeutic targets for lung cancer [6]. However, resistance to current targeted therapies is inevitable, and immune-based therapies still lack robust predictive biomarkers [7]. Therefore, it is still imperative to find effective therapeutic targets for LUAD patients.

MiRNAs are non-coding RNAs that are thought to play crucial roles in human physiology and lesion motility. Benefiting from high-throughput sequencing technologies, more and more non-coding RNAs are being discovered and defined as key regulators of tumorigenesis, aging and immune defense [8]. Dysregulated miRNAs cause various human diseases, including miR-22-3p. Previous studies have shown that upregulated miR-22-3p in extracellular vesicles (EVs) derived from human adipose tissue stem cells (AT-MSCs) can regulate osteogenesis [9]. The expression levels of miR-22-3p in serum may predict breast cancer at an early stage [10]. MiR-22-3p directly targets TCF7L2 in osteosarcoma and ENO1 in glioblastoma [11, 12]. Therefore, detecting the expression levels of miR-22-3p helps us to predict the risk of disease, diagnose disease at an early stage and evaluate the effect of treatment.

However, in the current studies of miR-22-3p, it has been found that the expression of miR-22-3p in LUAD is lower than that in normal tissue by PCR detection of clinically collected lung tissue [13, 14]. There are also studies using NSCLC lung tissue or blood PCR to learn that the expression of miR-22-3p in NSCLC is lower than that in normal lung tissue [15]. Some studies have verified the expression of miR-22-3p by cell function experiments [16]. Another study has concluded that the expression level of miR-22-3p is related to the prognosis of patients by extracting miRNA from patients’ blood and analyzing clinical information [17]. These research methods are relatively simple, lack multi-center validation and multi-method validation, and none of them have big data concepts to predict and validate target genes. Therefore, this study comprehensively analyzed an online database integrating in-house real-time quantitative polymerase chain reaction (RT-qPCR) data to validate the expression pattern of miR-22-3p, overexpressed miR-22-3p. In vitro experiments were carried out to detect the cell function after Lv-has-mir-22-3p lentiviral vector transfection, and computational biology was performed to predict the potential regulatory mechanism of miR-22-3p in LUAD.

2. Materials and methods

2.1 Clinical samples

41 pairs of formalin-fixed paraffin-embedded (FFPE) tumors and adjacent normal lung tissues were collected from LUAD patients from January 2014 to February 2016 from the First Affiliated Hospital of Guangxi Medical University. All involved patients received no treatment before surgery and were diagnosed postoperatively by two pathologists. All experiments were approved by the Medical Ethics Committee of the First Affiliated Hospital of Guangxi Medical University.

2.2 RT-qPCR

The extraction, reversion transcribed into cDNA and correction of total RNA were processed according to previous study [18]. RT-qPCR assay was performed on the Applied Biosystems PCR 7900. The expression of miR-22-3p was detected by miRcute Plus miRNA qPCR Kit (SYBR Green) (TIANGEN, Beijing, China). The expression of miR-22-3p was calculated by formula 2- $\Delta$ cq. The primer sequence of miR-22-3p was as follows:

Forward GGCTGAGCCGCAGTAGTTCT, Reverse GTGCAGGGTCCGAGGT; and the primer sequence of the endogenous control RNU6B was as follows: Forward CTCGCTTCGGCAGCACA, Reverse GTGCAGGGTCCGAGGT.

2.3 Data source and processing of miR-22-3p

Microarray datasets containing miR-22-3p expression were included by searching public databases such as ArrayExpress, Gene Expression Omnibus (GEO), Oncomine, and The Cancer Genome Atlas (TCGA). The inclusion criteria for the dataset were: (1) the study subjects were humans; (2) included miR-22-3p expression data. The exclusion criteria are: (1) duplicate or incomplete sample data; (2) the number of samples in the LUAD group and the control group in the combined data set is less than 3 in any one.

Log ${}_{2}$ ( $x+1$ ) transformation was performed on miR-22-3p expression data. In order to reduce the impact of batch effects between different datasets, datasets under the same platform (e.g. GPL11154 and GPL8179) are merged. During this process, the “SVA” package was used to eliminate batch effects, and the “limma” [19] and “edgeR” [20] packages were used to normalize the combined high-throughput gene expression matrix. Standardized mean difference (SMD) forest plots, funnel plots, summary receiver operating curve (sROC cure), and integrated sensitivity and specificity forest plots were then constructed by R software.

2.4 Cell culture, transfection and functional experiments

The LUAD cell line A549 was purchased from the Shanghai Cell Bank of the Chinese Academy of Sciences. A549 cells were cultured in RPMI 1640 medium containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (Hyclone, Logan, UT) and maintained at 5% carbon dioxide, 37.0 ${}^{\circ}$ C and 95% humidity in the incubator. A549 cells (Genechem, Shanghai, China) were divided into three groups: blank control group (untreated cells), negative control group (transfected with no-load virus) and experimental group (transfected with LV-has-miR-22-3p). Stably transfected cells were selected with 5 $\mu$ g/ml puromycin. The proliferation of A549 cells was analyzed using the MTS assay: the cell culture and transfection were as described above. After 0, 5, and 10 days of culture, 20 $\mu$ L of MTS solution was added after medium change and incubated for 4 h. The OD values were measured at 490 nm. CCK-8 assay followed the protocol of Xiong et al. [21]. The methods of in vitro cell migration and invasion experiments were carried out according to the methods of existing research [22]. Caspase-Glo 3/7 assay system (Promega, Madison, Wisconsin, USA) was used to assay for apoptosis of untransfected and transfected A549 cells. Apoptosis assays were performed according to the manufacturer’s instructions.

2.5 Prediction of candidate target genes

“Limma” package from R software was used to screen for differentially expressed genes in microarray and sequencing data from LUAD. Compared with normal lung tissue, genes with SMD $>$ 0 were selected to be more highly expressed in LUAD and used as candidate differentially expressed molecules (up-molecules) for further analysis. MiR-22-3p target genes were predicted by the miRwalk2.0 website. All predicted target genes intersect with up-molecules, and then the overlapping genes are candidate target genes, which will be used for the next potential mechanistic analysis.

2.6 Analysis of potential molecular mechanisms of candidate target genes

“Clusterprofiler” is a powerful R package [23] that supports enrichment analysis of Gene Ontology (GO). GO enrichment analysis included biological process (BP), cellular component (CC) and molecular function (MF), mainly showing the biological pathways involved in candidate target genes. Protein-protein interaction (PPI) network analysis can study the molecular mechanism of LUAD from a systems perspective, which is helpful to discover the hub of miR-22-3p. We input candidate target genes into the STRING (V11.0) website (https://string-db.org/) to draw PPI networks to analyze the interactions between these genes effect.

2.7 Data source and processing of TP53

The PPI network constructed from the candidate target genes showed that TP53 was a key gene of the miR-22-3p pathway (see 3.3.2 for details). TP53 multi-center sample information collection and data processing methods are the same as the above (see 2.3). The protein expression level of TP53 was verified by searching The Human Protein Atlas (THPA) database to incorporate TP53 immunohistochemical results in LUAD.

2.8 Statistical methods

Wilcoxon rank sum test and SMD were used to determine differences in gene expression between the LUAD and control groups. A $p$ value $>$ 0.1 in Egger’s test indicated no significant publication bias in the SMD. The ability of miR-22-3p to discriminate between LUAD and control samples was assessed using the area under the curve (AUC) of the receiver operating characteristic curve. A $p$ -value $<$ 0.05 suggested statistical significance in all the statistical analyses without specific identification.

All the experiments were independently conducted at least three times. Experiment data were analyzed by SPSS 22.0 statistical software package. Normal distribution data were presented as Mean $\pm$ SD. The differences between two groups or more groups were analyzed with the employment of Student’s $t$ test or nonparametric test and ANOVA. Visualization steps of experiment data were performed in GraphPad Prism 7.0.

All calculation and visualization steps of data from public databases were performed in R software (v4.1.0) except that the integrated receiver operating characteristic curve was drawn in Stata 15.0.

3. Results

3.1 Expression of miR-22-3p in LUAD

3.1.1 RT-qPCR of samples from our hospital

In the 41 pairs of lung tissues collected in our hospital, the expression of miR-22-3p in LUAD showed a downward trend compared with the adjacent normal tissues (Fig. 1A). In addition, miR-22-3p expression levels were not significantly different from clinicopathological parameters ( $p>$ 0.05, results not shown). However, the AUC value of the reference receiver operating curve (ROC) was between 0.5 and 1, and the closer the AUC was to 1, the better the recognition performance of miR-22-3p for LUAD. Therefore, miR-22-3p had a certain effect on the recognition of LUAD (AUC $=$ 0.6597, $p=$ 0.0128) (Fig. 1B), that is, the expression level of miR-22-3p in LUAD showed a downward trend.

3.1.2 MiRNA-sequencing and microarray data

Table 1
Basic information about the microarray chips included in this study

Database	Year	Source	N1	M1		SD1		N2	M2		SD2
GPL11154	2010	Tissue	111	12.	01221612	1.	150983226	125	11.	91540756	0.	82793353
GPL16770	2013	Tissue	157	1.	813463603	0.	586544346	32	2.	060948606	0.	616813626
GPL8179	2009	Tissue	315	13.	40004758	0.	363921101	134	13.	31687952	0.	40882334
GSE135918	2020	Tissue	5	3.	481171087	0.	79999622	5	3.	752047389	0.	678704232
GSE152702	2020	Blood	21	$-$ 0.	515622557	1.	444821203	7	2.	11397497	1.	356342419
GSE171517	2021	Blood	3	1.	916660691	0.	391355946	6	2.	16147649	0.	716897118
GSE19945	2013	Tissue	4	6.	666173668	0.	967883704	8	7.	031602718	0.	261035785
GSE25508	2011	Tissue	5	9.	971797944	0.	788209247	13	10.	20252204	0.	686419623
GSE27486	2011	Blood	22	2.	547359464	0.	345562811	13	2.	626456482	0.	441714624
GSE40738	2012	Blood	45	1.	114726734	0.	154514453	58	1.	172466884	0.	191776573
GSE47525	2013	Tissue	6	5.	315933908	0.	484801661	14	6.	092698355	0.	338974982
GSE51853	2014	Tissue	76	4.	189457805	0.	554250899	5	4.	400898087	0.	30952221
GSE63805	2015	Tissue	32	7.	338482634	0.	548185371	30	7.	582069205	0.	544942827
GSE74190	2015	Tissue	36	8.	667971885	0.	888325906	44	8.	679402706	0.	380123421
TCGA	2019	Tissue	521	16.	15272707	0.	541273059	46	15.	2956407	0.	788364537

N1: Number tumor; M1: Mean tumor; SD1: Standard deviation tumor; N2: Number control; M2: Mean control; SD2: Standard deviation control.

Figure 1.

Analysis of RT-qPCR data indicates that miR-22-3p expression was down-regulated in LUAD. A. violin diagram. B. ROC chart. LUAD: lung adenocarcinoma. AUC: area under the curve. $p<$ 0.05 ${}^{*}$ , the results are statistically significant.

Figure 2.

Integrated analysis of RT-qPCR and all microarray chips of miR-22-3p got from public databases. A. Standardized mean difference (SMD) forest map. B. Funnel chart. C. Summary receiver operating curve (sROC) curve.

The data included in this study were shown in Table 1. We obtained RNA-sequencing data for 521 LUAD tissues and 46 normal lung tissues from the TCGA database which composed with very few Asian patients. Compared with the normal lung tissues, miR-22-3p was down-regulated in LUAD tissues ( $p<$ 0.0001, 15.2956 $\pm$ 0.7884 vs 16.1527 $\pm$ 0.5413). A random effect model was further constructed to calculate SMD, and the expression levels of miR-22-3p in 14 platform datasets were integrated and analyzed. The results showed that the expression of miR-22-3p was down-regulated in the LUAD group compared with the control group (SMD $=-$ 0.32 [95% CI: $-$ 0.58– $-$ 0.06]; Fig. 2A), and no significant publication bias was found in the SMD results ( $p=$ 0.0002; Fig. 2B). AUC was 0.72 (Fig. 2C), which means that miR-22-3p has a strong ability to distinguish LUAD tissue from normal lung tissue. The combined sensitivity was 0.59 [0.44, 0.72] and the combined specificity was 0.72 [0.68, 0.75] (Fig. 2C).

3.2 Cell biological function

3.2.1 LV-miR-22-3p transfection

To explore the function of miR-22-3p in LUAD, A549 cell lines were transfected with LV-miR-22-3p and no-load virus, respectively, and a blank control group was set. The interference efficiency was verified by RT-qPCR. The results showed that A549 was successfully transfected into LV-miR-22-3p, and compared with the control group transfected with no-load virus, the relative expression of LV-miR-22-3p in lung cancer cells was significantly higher decreased ( $p=$ 0.014) (Table 2).

Table 2
Over expression efficiency of LV-miR-22 in lung cancer cell

	Expression of miR-22-3p			Mean $\pm$ SD	$t$	$p$
LV-miR-22	0.3076	0.4236	0.2990	0.3434 $\pm$ 0.0696	8.326	0.014
No-load virus	0.0093	0.0123	0.0035	0.0084 $\pm$ 0.0045

3.2.2 MiR-22-3p inhibits A549 cells proliferation in vitro

In this study, the effect of miR-22-3p on the proliferation of lung cancer cell A549 was detected by MTS kit. The established experimental group and control group were the same as the above experiments. It can be concluded that the cell proliferation ability of the experimental group added with miR-22-3p inhibitor was lower than that of the control group on the 5th day, and was stronger than that of the control group on the tenth day, but there was no significant difference. Compared with the control group, the cell proliferation ability of the experimental group of the mock gradually decreased on the fifth day and the tenth day, and the difference was statistically significant (Fig. 3, Table 3).

Table 3
The effect of transfection of miR-22-3p on the proliferation of A549 cells was detected by MTS

Time	Negative control of inhibitor	miR-22-3p inhibitor	$t$	$P$	Negative control of mimic	miR-22-3p mimic	$t$	$p$
	Mean $\pm$ SD	Mean $\pm$ SD			Mean $\pm$ SD	Mean $\pm$ SD
0D	102 $\pm$ 1.000	101 $\pm$ 3.606	0.463	0.667	100.667 $\pm$ 4.726	88.667 $\pm$ 5.508	2.864	0.046
5D	102 $\pm$ 1.732	101 $\pm$ 2.646	0.548	0.613	101.333 $\pm$ 0.577	103 $\pm$ 2.000	$-$ 1.387	0.238
10D	99.667 $\pm$ 1.158	100.667 $\pm$ 1.528	$-$ 0.905	0.417	102 $\pm$ 1.000	79 $\pm$ 1.732	19.919	0.000

Table 4

CCK-8 kit was used to detect the effect of transfected LV-miR-22 on the proliferation of A549 cells

Time	LV-miR-22	No-load virus	$t$	$P$
	Mean $\pm$ SD	Mean $\pm$ SD
0 h	0.186 $\pm$ 0.0236	0.1587 $\pm$ 0.01343	1.741	0.157
12 h	0.1947 $\pm$ 0.03530	0.1687 $\pm$ 0.01290	1.198	0.297
24 h	0.2050 $\pm$ 0.02152	0.2353 $\pm$ 0.01882	$-$ 1.838	0.140
36 h	0.2010 $\pm$ 0.01353	0.2057 $\pm$ 0.00503	$-$ 0.560	0.605
48 h	0.3493 $\pm$ 0.01528	0.3750 $\pm$ 0.03051	$-$ 1.303	0.263
60 h	0.4057 $\pm$ 0.02850	0.4753 $\pm$ 0.02354	$-$ 3.264	0.031
72 h	0.5050 $\pm$ 0.03568	0.6743 $\pm$ 0.12987	$-$ 2.178	0.095

To further validate the ability of transfected miR-22-3p to inhibit cell proliferation, CCK-8 cell proliferation assay was used. The growth curve of the transfected LV-miR-22-3p group was lower than that of the empty virus group. And at 60 hours, the overexpression of miR-22-3p group compared with the empty group had a significant difference in the effect on cell proliferation ( $p<$ 0.05) (Fig. 4, Table 4).

3.2.3 Overexpression of miR-22-3p inhibits the migration and invasion of A549 cells

The scratch assay was used to detect the effect of overexpression of miR-22-3p on the migration ability of A549 cells in vitro. Figures 5 and 6 show that after the use of lentiviral vector to interfere with miR-22-3p expression, the in vitro migration ability of A549 cell line was consistently inhibited, and was statistically different from the corresponding empty group at 18 hours and 24 hours (Table 5).

Table 5
Overexpression of miR-22-3p inhibits the migration of A549 lung cancer cells

Time	No-load virus	LV-miR-22	$t$	$P$
	Mean $\pm$ SD	Mean $\pm$ SD
0 h	396865 $\pm$ 12927.1558	383477 $\pm$ 18073.0798	1.044	0.356
6 h	351612 $\pm$ 18785.0420	354939 $\pm$ 6213.2173	$-$ 0.291	0.785
12 h	323065 $\pm$ 38464.0356	349772 $\pm$ 11703.9304	$-$ 1.151	0.314
18 h	298006 $\pm$ 11338.4334	336747 $\pm$ 15647.7708	$-$ 3.472	0.030
24 h	293344 $\pm$ 6311.4349	330794 $\pm$ 13071.5368	$-$ 4.469	0.023

Figure 3.

The effect of transfection of miR-22-3p on the proliferation of A549 cells was detected by MTS ( ${}^{*}<$ 0.05, ${}^{***}<$ 0.01).

Figure 4.

The growth curve of A549 cells transfected with LV-miR-22 was detected by CCK-8 kit ( ${}^{*}<$ 0.05).

Figure 5.

Migration ability of A549 lung cancer cells overexpressed with miR-22-3p ( ${}^{*}<$ 0.05).

Figure 6.

Over-expression of miR-22-3p inhibits the migration of A549 lung cancer cells. A. 18 hours later no-load virus group. B. 18 hours after transfection with LV-miR-22 group. C. 24 hours after no-load virus group. D. 24 hours later, the cells were transfected into LV-miR-22 group.

Figure 7.

The effect of over-expression of miR-22-3p on the migration of lung cancer cells was detected by Transwell assay. A549 cells were (A) transfected with LV-miR-22 and (B) without virus transfected for 24 hours and then stained with crystal violet. The number of migration cells was observed under microscope (200 $\times$ ).

To further verify the scratch test results, Transwell assay was used to detect the effect of miR-22-3p on the migration ability of A549 cells. In the A549 cell line, the number of cells transfected with LV-miR-22-3p was less (Fig. 7), which was statistically different from the control group ( $t=-$ 6.353, $p=$ 0.003, Table 6). Similar to scratch results. After overexpression of miR-22-3p, the migration ability of cells decreased ( $p<$ 0.01). Both scratch assay and Transwell migration assay showed that overexpression of miR-22-3p significantly inhibited the migration ability of cells.

Table 6

The effect of over-expression of miR-22-3p on the migration of A549 cells was detected by Transwell assay

	Average number of migration cells			Mean $\pm$ SD	$t$	$P$
LV-miR-22	35	29	31	31.6667 $\pm$ 3.0551	$-$ 6.353	0.003
No virus	53	48	46	49.0000 $\pm$ 3.6056

Table 7

The effect of over-expression of miR-22-3p on the invasion ability of A549 cells was detected by Matrigel Transwell assay

	Average number of invading cells			Mean $\pm$ SD	$t$	$P$
LV-miR-22	35	34	33	34.0000 $\pm$ 1.0000	$-$ 16.280	0.003
No virus	120	116	134	123.3333 $\pm$ 9.4516

Table 8

Caspase-3/7 kit was used to detect the effect of miR-22-3p on apoptosis of lung cancer cells

Time	Negative control of inhibitor	miR-22-3p inhibitor	$t$	$P$	Negative control of mimic	miR-22-3p mimic	$t$	$p$
	Mean $\pm$ SD	Mean $\pm$ SD			Mean $\pm$ SD	Mean $\pm$ SD
0D	1.0000 $\pm$ 0.0721	1.0200 $\pm$ 0.0436	$-$ 0.411	0.702	1.0033 $\pm$ 0.0252	1.0067 $\pm$ 0.0764	$-$ 0.072	0.946
5D	0.9633 $\pm$ 0.0513	1.0100 $\pm$ 0.0361	$-$ 1.289	0.267	1.0300 $\pm$ 0.0265	1.4667 $\pm$ 0.2558	$-$ 2.941	0.096
10D	1.0000 $\pm$ 0.0265	1.0067 $\pm$ 0.0802	$-$ 0.137	0.898	1.0300 $\pm$ 0.0361	1.7467 $\pm$ 0.3137	$-$ 3.931	0.057

Figure 8.

The effect of over-expression of miR-22-3p on the migration of A549 cells was detected by Matrigel transwell assay. A549 cells (A) transfected with LV-miR-22 and (B) without a virus were stained with crystal violet 48 hours. The number of invading cells was observed under microscope (200 $\times$ ).

Figure 9.

Caspase-3/7 kit was used to detect the effect of miR-22-3p on apoptosis of A549 cells.

Figure 10.

Potential mechanism of miR-22-3p in lung adenocarcinoma. A. The Venn diagram of predicted target genes and DEGs. B. Bar chart of candidate target genes enriched in cellular component.

Figure 11.

PPI analysis of candidate target genes in the network that interact most with other genes. TP53 is the determination gene within the whole PPI network.

Figure 12.

Integrated analysis of expression levels of TP53 in LUAD got from public databases. A. SMD forest map. B. Funnel chart.

Figure 13.

Protein expression of TP53 in LUAD and control tissues. N: Normal tissue; T: Tumor tissue.

Matrigel Transwell assay was used to verify the regulatory effect of overexpression of miR-22-3p on A549 cell invasion ability. The results showed that after A549 cells were transfected with LV-miR-22, the invasive ability of A549 cells was weakened (Fig. 8), and there was a statistical difference ( $t=-$ 16.280, $p=$ 0.033, Table 7). The results of this experiment showed that overexpression of miR-22-3p in lung cancer cell lines could significantly inhibit the invasive ability of cells.

3.2.4 Effects of miR-22-3p on A549 cells apoptosis

The effect of miR-22-3p on apoptosis of A549 cells was detected by Caspase-3/7 kit. After transfection of miR-22-3p inhibitors and mimics into A549 cells, apoptosis was promoted on both the fifth and tenth days, but the changes were not obvious (Fig. 9, Table 8).

3.3 Analysis of pathway and potential molecular mechanisms of miR-22-3p

3.3.1 Candidate target gene prediction

According to the screening criteria, 8589 up-molecules were selected from the microarray chip and sequencing dataset. 4540 related target genes were predicted on the miRwalk2.0 website. 620 relevant target genes were predicted on the TargetScan website. We intersected these ascending genes with related target genes, resulting in 56 candidate target genes (Fig. 10A).

3.3.2 Analysis of potential molecular mechanisms of candidate target genes

The possible mechanisms of 56 candidate target genes were analyzed using the “clusterProfiler” R package and STRING. The GO analysis results mainly focused on the RNA polymerase II transcriptional regulatory complex, the transcriptional regulatory complex, the ficolin-1-rich granule lumen, the $\beta$ -catenin-TCF complex, and the mitochondrial intermembrane space (Fig. 10B). The 56 candidate target genes were drawn into the PPI network and 10 proteins had a close relationship with TP53, as shown in Fig. 11. TP53 was the most critical gene among the 56 candidate genes. Therefore, this study decided to investigate TP53 as a possible target gene for further exploration.

3.3.3 Integrated analysis of TP53 public database

Globally, a total of 114 high-throughput datasets were included, including 3897 LUADs and 2993 noncancerous lung tissues. After removing batch effects, 37 platforms were finally obtained. A random-effects model was further constructed to calculate SMD, and the expression level of TP53 (the gene expression value included in the 35 platform datasets) was integrated for analysis. The results showed that TP53 expression was up-regulated in the LUAD group compared with the control group (SMD $=$ 0.39 [95% CI: 0.23–0.55]; Fig. 12A), and no significant publication bias was found in the SMD results ( $p=$ 0.0002; Fig. 12B).

The expression of TP53 in LUAD was verified from the protein level. In this study, a total of 23 LUAD tissues and 12 control tissues were retrieved through the THPA database to obtain immunohistochemical pictures. As shown in Fig. 13, the TP53 protein expression level in the LUAD group was higher than that in the control group ( $p<$ 0.05), which was consistent with the results of the mRNA levels.

4. Discussion

In recent years, thanks to significant advances in molecular oncology, there have been more effective approaches to the treatment of NSCLC, improving survival for patients with advanced stages. However, the drug resistance of targeted therapy is still a problem of precision therapy [7]. As the most common NSCLC, LUAD has poor survival and high recurrence rate [2]. is the focus of current LUAD treatment research to find effective therapeutic targets and avoid drug resistance. Despite tremendous progress in lung cancer treatment, the 5-year survival rate for patients with stage remains around 55% [24]. Therefore, finding effective therapeutic targets and avoiding drug resistance is the focus of current LUAD treatment research.

In this study, the expression patterns of miR-22-3p and overexpressed miR-22-3p were verified by RT-qPCR data of our hospital and cell function detection after Lv-has-mir-22-3p lentiviral vector transfection. And computational biology was performed to predict the potential regulatory mechanism of miR-22-3p in LUAD.

Previous studies on the expression of miR-22-3 in LUAD have either single-center data or only a single method. Ma et al. collected 62 LUAD patients’ tumor tissues and adjacent normal tissues from Peking Union Medical College Hospital for PCR detection of miR-22-3p expression levels [13]. Yang et al. collected 91 LUAD lung tissues and 10 normal lung tissues from the Department of Thoracic Surgery of the University of Michigan for RT-qPCR to detect the expression levels of miR-22-3p [16]. Zhenan Xu et al. collected blood samples from 2 cases of LUAD and 2 cases of healthy people in the Second Hospital of Jilin University to detect the expression of miR-22-3p by exosome microarray [15]. Zhou et al. confirmed that ENO1 is one of the targets of miR-22-3p by RT-qPCR and western blot analysis of A539 cell line overexpressing ENO1 and miR-22-3p [14]. Ulivi et al. collected 99 LUAD blood samples and matched clinical information from Medical Oncology Department of Perugia Hospital and by the Thoracic Surgery Department of Morgagni-Pierantoni Hospital of Forlì, and performed univariate analysis after detecting miR-22-3p expression. The results showed that miR-22-3p was significantly associated with disease-free survival and overall survival of patients [17]. Dong et al. used ChipBase, LncRNAdb and StarBase to identify that miR-22-3p was regulated by DGCR5, and verified by cellular luciferase experiments [25]. However, the above studies are only validated with single-center data or a single method.

In this study, RT-qPCR was performed on 41 pairs of LUAD and matched adjacent normal tissues collected in our hospital, and the results showed that miR-22-3p was expressed lower in LUAD than in normal tissues. At the same time, by incorporating high-throughput data from around the world, 838 cases of LUAD and 494 cases of non-cancer lung tissue were comprehensively analyzed, and the results were consistent with the RT-qPCR results. We also performed functional experiments with A539 cells transfected LV-has-miR-22-3p, showing that miR-22-3p inhibited cell proliferation, migration and invasion, and promoted cell apoptosis, which was consistent with the low expression of miR-22-3p in LUAD.

In order to understand the tumor suppressor effect mechanism of miR-22-3p, this study collected miR-22-3p-related genes for gene enrichment and signaling pathway analysis, and constructed PPI network. The results showed that TP53 is a key gene of the miR-22-3p-related signaling pathways. The integrated analysis of TP53 expression levels showed that TP53 is up-regulated in LUAD. At the same time, we collected the LUAD tissue immunohistochemical data of TP53 from THPA, which also showed that the protein expressed by TP53 was increased in LUAD. These indicated that miR-22-3p was inversely correlated with TP53, suggesting that miR-22-3p might inhibit TP53 expression. Research has shown that, mutated or unregulated forms of TP53 exert novel oncogenic activities [26], further verifying that miR-22-3p inhibits the development of LUAD. Qian et al. showed that TP53 and STAG2 were associated with overall survival of NSCLC patients. It may be the manifestation of PPI interaction on LUAD in this study [27].

5. Conclusion

miR-22-3p is underexpressed in lung adenocarcinoma and it plays a role in inhibiting tumor proliferation, migration and invasion, and promoting apoptosis in LUAD, which may be achieved through pathways related to cell cycle and cell migration. In addition, TP53 is a key gene in these pathways and its expression is up-regulated in LUAD, suggesting that miR-22-3p inhibits TP53 expression and thus acts as a tumor suppressor. The research on miR-22-3p and its related pathways needs to be verified by more in-depth in vivo studies.

Consent for publication

All authors have read the final draft of the manuscript and agreed to its publication.

Funding

This work was supported by the Guangxi Zhuang Autonomous Region Medical Health Appropriate; Technology Development and Application Promotion Project (S2020031); Guangxi Medical High-level Key Talents Training “139” Program (2020); Guangxi Medical University Undergraduate Innovation and Entrepreneurship Training Program (X202210598227); and Future Academic Star of Guangxi Medical University (WLXSZX22112).

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Footnotes

Acknowledgments

The authors greatly acknowledge Guangxi Medical University and other institutions for funding the study.

Conflict of interest

None to report.

References

Al-Yozbaki

Jabre

Syed

Wilson

. Targeting DNA methyltransferases in non-small-cell lung cancer. Semin Cancer Biol. 2022; 83: 77-87. doi: 10.1016/j.semcancer.2021.01.005.

Xiao

Geng

Zhou

Wang

Huang

, et al. POU6F1 cooperates with RORA to suppress the proliferation of lung adenocarcinoma by downregulation HIF1A signaling pathway. Cell Death Dis. 2022; 13(5): 427. doi: 10.1038/s41419-022-04857-y.

Sun

Luo

Zheng

Zhang

, et al. A novel immune checkpoints-based signature to predict prognosis and response to immunotherapy in lung adenocarcinoma. J Transl Med. 2022; 20(1): 332. doi: 10.1186/s12967-022-03520-6.

Shen

Zhang

Liu

. Diagnostic and prognostic value of microRNA-486 in patients with lung cancer: A systematic review and meta-analysis. Int J Biol Markers. 2022; 3936155221115750. doi: 10.1177/03936155221115750.

Xie

Zhong

Duan

. miR-1269a and miR-1269b: Emerging Carcinogenic Genes of the miR-1269 Family. Front Cell Dev Biol. 2022; 10: 809132. doi: 10.3389/fcell.2022.809132.

Kalhori

Soleimani

Arefian

Alizadeh

Mansouri

Echeverria

. The potential role of miR-1290 in cancer progression, diagnosis, prognosis, and treatment: An oncomiR or onco-suppressor microRNA? J Cell Biochem. 2022; 123(3): 506-31. doi: 10.1002/jcb.30191.

Yang

Schultheis

Mandelker

Ladanyi

Buttner

. Precision medicine in non-small cell lung cancer: Current applications and future directions. Semin Cancer Biol. 2022; 84: 184-98. doi: 10.1016/j.semcancer.2020.07.009.

Elrebehy

Al-Saeed

Gamal

El-Sayed

Ahmed

Waheed

, et al. miRNAs as cornerstones in colorectal cancer pathogenesis and resistance to therapy: A spotlight on signaling pathways interplay – A review. Int J Biol Macromol. 2022; 214: 583-600. doi: 10.1016/j.ijbiomac.2022.06.134.

Kaur

Abu-Shahba

Paananen

Hongisto

Hiidenmaa

Skottman

, et al. Small non-coding RNA landscape of extracellular vesicles from human stem cells. Sci Rep. 2018; 8(1): 15503. doi: 10.1038/s41598-018-33899-6.

10.

Fang

Zhu

Khadka

Zou

, et al. Plasma MicroRNA Pair Panels as Novel Biomarkers for Detection of Early Stage Breast Cancer. Front Physiol. 2018; 9: 1879. doi: 10.3389/fphys.2018.01879.

11.

Xue

Guo

Liu

Meng

. MicroRNA-22-3p targeted regulating transcription factor 7-like 2 (TCF7L2) constrains the Wnt/beta-catenin pathway and malignant behavior in osteosarcoma. Bioengineered. 2022; 13(4): 9135-47. doi: 10.1080/21655979.2021.2003942.

12.

Wang

Zong

Wang

Yang

, et al. EGR1 modulated LncRNA HNF1A-AS1 drives glioblastoma progression via miR-22-3p/ENO1 axis. Cell Death Discov. 2021; 7(1): 350. doi: 10.1038/s41420-021-00734-3.

13.

Zhou

Qin

Tian

Liu

. MiR-22-3p Expression is down-regulated in lung adenocarcinoma. Acta Biochim Pol. 2021; 68(4): 667-72. doi: 10.18388/abp.2020_5540.

14.

Zhou

Zhang

Chen

. CircRNA-ENO1 promoted glycolysis and tumor progression in lung adenocarcinoma through upregulating its host gene ENO1. Cell Death Dis. 2019; 10(12): 885. doi: 10.1038/s41419-019-2127-7.

15.

Wang

Sun

Xin

. Evaluation of exosomal miRNA in blood as a potential diagnostic biomarker for human non-small cell lung cancer. Med Sci Monit. 2020; 26: e924721. doi: 10.12659/MSM.924721.

16.

Yang

Zhou

Shan

, et al. MiR-22-3p suppresses cell growth via MET/STAT3 signaling in lung cancer. Am J Transl Res. 2021; 13(3): 1221-32.

17.

Ulivi

Petracci

Marisi

Baglivo

Chiari

Billi

, et al. Prognostic role of circulating miRNAs in early-stage non-small cell lung cancer. J Clin Med. 2019; 8(2). doi: 10.3390/jcm8020131.

18.

Kotorashvili

Ramnauth

Liu

Lin

Kim

, et al. Effective DNA/RNA co-extraction for analysis of microRNAs, mRNAs, and genomic DNA from formalin-fixed paraffin-embedded specimens. PLoS One. 2012; 7(4): e34683. doi: 10.1371/journal.pone.0034683.

19.

Ritchie

Phipson

Law

Shi

, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015; 43(7): e47. doi: 10.1093/nar/gkv007.

20.

Robinson

McCarthy

Smyth

. edgeR: A Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010; 26(1): 139-40. doi: 10.1093/bioinformatics/btp616.

21.

Xiong

Feng

Lai

Qin

Liu

, et al. High throughput circRNA sequencing analysis reveals novel insights into the mechanism of nitidine chloride against hepatocellular carcinoma. Cell Death Dis. 2019; 10(9): 658. doi: 10.1038/s41419-019-1890-9.

22.

Liang

Yang

Zhang

Zhu

, et al. A PTAL-miR-101-FN1 axis promotes EMT and invasion-metastasis in serous ovarian cancer. Mol Ther Oncolytics. 2020; 16: 53-62. doi: 10.1016/j.omto.2019.12.002.

23.

Wang

Han

. clusterProfiler: An R package for comparing biological themes among gene clusters. OMICS. 2012; 16(5): 284-7. doi: 10.1089/omi.2011.0118.

24.

Zhang

Wang

Chen

Wang

, et al. Ferroptosis-related genes are potential therapeutic targets and the model of these genes influences overall survival of NSCLC patients. Cells. 2022; 11(14). doi: 10.3390/cells11142207.

25.

Dong

Wang

Jin

Zeng

Pan

. LncRNA DGCR5 promotes lung adenocarcinoma (LUAD) progression via inhibiting hsa-mir-22-3p. J Cell Physiol. 2018; 233(5): 4126-36. doi: 10.1002/jcp.26215.

26.

Parvin

Sedighian

Sohrabi

Mahboobi

Rezaei

Ghasemi

, et al. Prediction of genes involved in lung cancer with a systems biology approach based on comprehensive gene information. Biochem Genet. 2022; 60(4): 1253-73. doi: 10.1007/s10528-021-10163-7.

27.

Qian

Nie

Zhang

Wang

Lou

Dong

Niu

Yan

Zhong

Zhang

Chu

Zhong

Han

. Racial differences in characteristics and prognoses between Asian and white patients with nonsmall cell lung cancer receiving atezolizumab: An ancillary analysis of the POPLAR and OAK studies. Int J Cancer. 2022; 146(11): 3124-3133.

The clinical significance and mechanism of microRNA-22-3p targeting TP53 in lung adenocarcinoma

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Clinical samples

2.2 RT-qPCR

2.3 Data source and processing of miR-22-3p

2.4 Cell culture, transfection and functional experiments

2.5 Prediction of candidate target genes

2.6 Analysis of potential molecular mechanisms of candidate target genes

2.7 Data source and processing of TP53

2.8 Statistical methods

3. Results

3.1 Expression of miR-22-3p in LUAD

3.1.1 RT-qPCR of samples from our hospital

3.1.2 MiRNA-sequencing and microarray data

Table 1 Basic information about the microarray chips included in this study

3.2.1 LV-miR-22-3p transfection

Table 2 Over expression efficiency of LV-miR-22 in lung cancer cell

Table 3 The effect of transfection of miR-22-3p on the proliferation of A549 cells was detected by MTS

Table 5 Overexpression of miR-22-3p inhibits the migration of A549 lung cancer cells

3.3 Analysis of pathway and potential molecular mechanisms of miR-22-3p

3.3.1 Candidate target gene prediction

3.3.2 Analysis of potential molecular mechanisms of candidate target genes

3.3.3 Integrated analysis of TP53 public database

4. Discussion

5. Conclusion

Consent for publication

Funding

Data availability

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Basic information about the microarray chips included in this study

Table 2
Over expression efficiency of LV-miR-22 in lung cancer cell

Table 3
The effect of transfection of miR-22-3p on the proliferation of A549 cells was detected by MTS

Table 5
Overexpression of miR-22-3p inhibits the migration of A549 lung cancer cells