Abstract

Purpose

   The function of BZRAP1-AS1 is unknown in lung cancer. We evaluated the
   clinicopathologic significance of BZRAP1-AS1, and its role in
   non-small-cell lung cancer (NSCLC) progression.

Patient and methods

   Sixty-three NSCLC patients from Beijing Chest Hospital were included.
   The expression of BZRAP1-AS1 was detected by real-time quantitative
   polymerase chain reaction (RT-qPCR) in tumor tissues and adjacent
   normal tissues. Then, the clinicopathological significance and
   prognostic value of BZRAP1-AS1 were analyzed by using our cohort and
   TCGA cohort. Finally, the effect of BZRAP1-AS1 on proliferation and
   motility of NSCLC cell lines were evaluated by cell growth assay,
   colony formation assay, xenograft tumorigenesis experiment in nude mice
   and transwell assays respectively.

Results

   Compared with adjacent normal tissues, BZRAP1-AS1 showed lower
   expression in NSCLC tumor tissues. As for the relationship between
   BZRAP1-AS1 and clinical characteristics, our results were consistent
   with those of TCGA data. BZRAP1-AS1 was lower in T1 than T2–T4
   patients, N1-N3 than N0 patients. Low level BZRAP1-AS1 was related to
   shorter overall survival time (OS) in lung adenocarcinoma (LUAD), and
   poor first progression time (FP) in LUAD and lung squamous cell
   carcinoma (LUSC) patients. BZRAP1-AS1 was significantly associated with
   the prognosis of NSCLC patients. Overexpression of BZRAP1-AS1 inhibited
   proliferation and migration of H1299 and HCC827 cells.

Conclusion

   BZRAP1-AS1 expression decreases in tumor tissues with the increase of
   malignancy grades in NSCLC. BZRAP1-AS1 plays an anticancer role by
   inhibiting cell proliferation, invasion, and metastasis, and has a
   potential prognostic value in NSCLC. BZRAP1-AS1 may serve as a
   diagnostic marker and therapeutic target for NSCLC.

   Keywords: Non-small-cell lung cancer, Long non-coding RNA, BZRAP1-AS1,
   Prognosis, Proliferation, Migration

Introduction

   Lung cancer is one of the malignant tumors with the highest morbidity
   and mortality. There were 2.2 million new cases diagnosed and 1.8
   million deaths worldwide in 2020, as the latest Global Cancer
   Statistics has shown ([36]Sung et al., 2021). In lung cancer, NSCLC
   accounts for about 80%, and with 60% lung adenocarcinoma (LUAD) and 25%
   lung squamous cell carcinoma (LUSC) as the predominant pathologic
   subtypes ([37]Zheng, 2016). Although multiple therapy regimens
   including surgery, chemotherapy, radiotherapy, immunotherapy are used,
   the prognosis of NSCLC patients is still poor. Therefore, it is very
   urgent to explore the molecular mechanism of the occurrence and
   development of lung cancer and the search for new therapeutic targets.
   Recent evidences have shown that lncRNAs have important functions in
   regulation of gene expression, RNA splicing, and nucleation of
   subnuclear domains ([38]Statello et al., 2021). Many lncRNAs, such as
   HOTAIR, PVT1, LINC01123 have been implicated in development or
   progression of lung cancer ([39]Loewen et al., 2014; [40]Liu et al.,
   2018; [41]Zheng et al., 2020a; [42]Pan et al., 2020; [43]Xi et al.,
   2020; [44]Hua et al., 2019).

   As a novel lncRNA, benzodiazapine receptor associated protein 1
   antisense RNA 1 (BZRAP1-AS1), also named TSPOAP1-AS1, was mentioned
   firstly in Alzheimer’s disease ([45]Jun et al., 2017; [46]Witoelar et
   al., 2018; [47]Tan et al., 2021a). Subsequently, studies indicated that
   BZRAP1-AS expression was dysregulated in several cancers and played
   critical roles in tumor initiation and progression  ([48]Giulietti et
   al., 2018; [49]Tan, Jin & Wang, 2019; [50]Wang et al., 2019a; [51]Wang
   et al., 2019b; [52]Zheng et al., 2020b; [53]Tang et al., 2021).
   However, whether BZRAP1-AS1 as oncogene or tumor suppressor is still
   unclear in cancers. For example, knocking down BZRAP1-AS1 in
   hepatocellular carcinoma (HCC) cells inhibits HUVEC proliferation,
   migration, and angiogenesis, the lncRNA tends to be an
   oncogene ([54]Wang et al., 2019b); however, the conclusions are
   opposite in cervical cancer, prostate cancer, and lung
   adenocarcinoma ([55]Tan, Jin & Wang, 2019; [56]Zheng et al., 2020b;
   [57]Tang et al., 2021). The study of BZRAP1-AS1 in lung adenocarcinoma
   is only from prognostic bioinformatics analysis studies have been
   reported, the design is very limited only through bioinformatics to
   study lung adenocarcinoma ([58]Tang et al., 2021). In this study, 63
   NSCLC cases were collected and BZRAP1-AS1 levels in tumor and adjacent
   normal tissues were examined. Then the association between BZRAP1-AS1
   and clinical parameters was analyzed using 63 NSCLC cases cohort and
   TCGA database. Finally, the effect of BZRAP1-AS1 on cell proliferation
   and migration was investigated in vivo and in vitro.

Patients and Methods

Patient samples

   The samples were collected as previously described in [59]Zhang et al.
   (2021) (28 April 2021). After calculation, a sample size of at least 41
   cases is required based on the pre-experimental data (95% confidence
   interval, 80% statistical power). Specifically, 63 NSCLC patients (46
   males and 17 females; mean age of 61.83  ± 9.29 years) were enrolled
   between 2018-10-24 and 2020-12-22. All above patients received surgical
   treatment and no preoperative neoadjuvant therapy at Beijing Chest
   Hospital. Tumor tissues and adjacent normal tissues were obtained. All
   patients had no other accompanying malignancy. After resection, tissue
   samples were flash frozen in liquid nitrogen for 30 min within 2 h and
   then frozen at −80 °C for long-term storage. This study was approved by
   the Ethics Committee of Beijing Chest Hospital. Ethical Approval
   Number: 2019-71.

Cell lines and culture conditions

   Cell culture and part assay methods were performed as previously
   described ([60]Zhang et al., 2021). Human NSCLC cell lines H226 were
   obtained from the National Institutes of Health (NIH). The human NSCLC
   cell lines A549, H1395, H1299, HCC827 and H1703 were obtained from the
   National Infrastructure of Cell Line Resource (NICR). All cell lines
   were cultured in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA)
   supplemented with 10% fetal bovine serum (FBS; Gibco, Los Angeles, CA,
   USA). The cells were maintained at 37 °C in a humidified chamber
   containing 5% CO2.

RNA extraction and Quantitative Real-Time PCR (RT-qPCR)

   Total RNA was extracted from stored tissues and cells with the TRIzol
   reagent (Ambion, Carlsbad, CA, USA), according to the manufacturer’s
   instructions. NanoDrop 2000c (Thermo Scientific, Waltham, MA, USA) was
   used to measure the concentration of RNA. RNAs were reverse-transcribed
   into cDNA in a 20 µl system using the TransScrip First-strand cDNA
   Synthesis SuperMix (TRAN) according to the protocol. The quantitative
   PCR was performed according to the instructions of the SYBR®. Applied
   Biosystems 7500 Fast Real-Time PCR System (Thermo Fisher Scientific)
   was used for qPCR with a 15-µl reaction system, including 7.5 µl of
   PowerUp™ SYBR™ Green Master Mix (applied biosystems), 2.5 µl of
   RNase-free water, 1.25 µl of upstream and 1.25 µl of downstream
   primers, and 2.5 µl of cDNA template. GAPDH was used as an internal
   reference gene. The primer sequences involved as follows. BZRAP1-AS1
   (F): TGTCTGCATCCCACAACAGG, (R): GGACCAGCTTGGAGTTGTGT. GAPDH (F):
   ACTAGGCGCTCACTGTTCTC, (R): CGACCAAATCCGTTGACTCC (5′–3′).

Lentivirus construction and infection

   The lentiviruses for BZRAP1-AS1 overexpression (OE) (Lv-BZRAP1-AS1,
   66037-1) and negative control (NC) (LVCON238) both were purchased from
   the Shanghai Genechem Company. Lentivirus infection was performed in
   H1299 and HCC827 NSCLC cells following the manufacturer’s instruction.
   Firstly, H1299 and HCC827 cells were plated in 96-well culture plates
   at a density of 3–5 × 10^3/well. The next day, cells were infected with
   Lv-BZRAP1-AS1 and LVCON238 for 12 h. The infection efficiency was about
   80%, 48–72 h after infection. Finally, the infected cells were screened
   with Puromycin (2 µg/mL) for 2–3 days.

Online database analysis

   Gene Expression Profiling Interactive Analysis (GEPIA)
   ([61]http://gepia.cancer-pku.cn/index.html) is an online tool for gene
   expression analysis based on TCGA and GTEx data ([62]Tang et al.,
   2017). In this study, the expression of BZRAP1-AS1 in multiple tumor
   samples and paired normal tissues was evaluated using GEPIA.
   Transcriptome profiling HTSeq-Counts data and clinical information of
   NSCLC patients were downloaded from the TCGA database
   ([63]https://portal.gdc.cancer.gov). The expression of BZRAP1-AS in
   Lung normal tissues and cancer adjacent normal tissues were both
   downloaded from the GTEx database. Then the expression difference and
   OS survival analysis of BZRAP1-AS1 in NSCLC was analyzed by SPSS and
   GraphPad Prism. Additionally, FP and univariate survival analysis was
   made by KM Plotter, which is an online survival analysis tool based on
   data from GEO, EGA, and TCGA ([64]Győrffy et al., 2013; [65]Lánczky &
   Győrffy, 2021).

Cell growth curve assay

   H1299 cells (4 ×10^3/well) and HCC827 cell (3 ×10^3/well) were seeded
   in 96-well plates. Five repetitive wells were set for each experiment.
   Cell activity was tested by cell counting kit-8 (CCK-8) (Dojindo,
   Kyushu, Japan) for 4 days. 10 µl CCK-8 was added into each well and
   then incubated with the cells for 2 h at 37 °C. The absorbance was
   recorded at 450nm with Bio Tek Epoch Microplate Reader (Bio Tek
   Instruments, Inc). The experiments were repeated 3 times.

Colony formation assay

   300 cells/well were plated into 6-well plates and routinely cultured
   for 7–14 days. Subsequently the cells were washed quickly with PBS
   buffer solution and stained with 0.2% gentian violet. The number of
   colonies (≥50 cells) was recorded under an optical microscope. The
   experiments were repeated three times.

Tumor growth in nude mice

   All animal experiments were approved by the Ethics Committee of the
   Beijing Chest Hospital, Capital Medical University, Beijing, China (No.
   2021-059), in compliance with national or institutional guidelines for
   the care and use of animals. Female BALB/c-nu mice (aged 4–6 weeks)
   were purchased from Beijing Vital River Laboratory Animal Technology
   Co., Ltd. The average weight of mice was 17 g. All mice were housed in
   animal facility under pathogen-free conditions. The animal room was
   kept at 18–23 °C, 40%–60% humidity, and a 10 h–14 h light-dark cycle.
   In order to get the most scientific experimental results, at least 10
   mice in each group, but according to the 3R-reduction principle, so six
   mice in each group, a total of 24 mice were randomly divided into four
   groups. All food, cages, water, and other items that contact mice were
   sterile and handled using aseptic technique within a certified
   biosafety cabinet. Cells (H1299 NC vs H1299 OE, HCC827 NC vs HCC827 OE)
   were collected and resuspended in 10% RMPI-1640 at a density of 1
   × 10^7 cells/ml. 2 × 10^6 were injected into subcutaneously the right
   armpit of the mice, NC as control group. Then length (L) and width (W)
   sizes of tumors were measured regularly every 2–3 days by researchers
   who didn’t know the grouping. The mice were euthanized on the 30–40th
   day, and tumors were harvested and photographed. Tumor volume was
   analyzed by V = 1/2(L × W^2). On the 40th day following tumor injection
   and the average tumor diameter does not exceed 20 mm, or if ulceration,
   infection, or necrosis occurs, the experiment was terminated and the
   mice were euthanized using a standard carbon dioxide method. Finally,
   tumors were harvested and weighed. Data was analyzed using GraphPad
   Prism, the tumor volume data was analyzed using ANOVA and weight
   differences by unpaired t-test.

Cell migration and invasion assays

   Cell invasion was tested using Transwell chamber (8 µm in pore size,
   Costar, Beijing, China) pre-coated with Matrigel (Corning Matrigel
   Matrix, CA, USA). Cells were cultured in serum-free RMPI-1640 medium
   for 24 h. HCC827 cells (2  × 10^6) and H1299 cells (1.5  × 10^6) were
   seeded in the upper compartment with serum free medium. The lower
   chamber was filled with RMPI-1640 medium containing 10% FBS. After
   48 h, the chamber was taken out and stained with 0.2% gentian violet
   for 20min, then upper Matrigel was washed away the staining reagent and
   removed using cotton swab cautiously. Finally, metastatic cells were
   observed and pictures taken under a Nikon microscope. The experiments
   were repeated three times.

RNA-seq data analysis of mice tumor tissues and co-expression analysis

   Subcutaneous tumors were RNA sequenced in H1299 NC group and H1299
   BZRAP1-AS1 OE group. And different genes between the two groups were
   screened using EdgeR, then performed GO (Gene Ontology) enrichment
   analysis and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway
   enrichment analysis for up-regulated genes and down-regulated genes.
   Then, we selected genes with significant differences in expression and
   closely related to tumor proliferation, metastasis and signaling
   pathways, and predicted the genes affected by BZRAP1-AS1 based on the
   above results. Furthermore, the co-expression analysis between
   BZRAP1-AS1 and predicted genes was performed using Starbase Analysis
   platform ([66]https://starbase.sysu.edu.cn/) ([67]Li et al., 2014).

Statistical analysis

   The experiments were repeated three times. Normal distribution data
   were presented as Mean  ± SD, and compared using a two-tailed t-test.
   Abnormal distribution data were presented as median (25–75 percentile),
   and compared using the nonparametric test. The log-rank test was used
   to compare Kaplan–Meier survival curves. Statistical methods used for
   RNA-Seq analysis and TCGA data analysis were described above. P < 0.05
   was considered statistically significant.

Results

BZRAP1-AS1 expression reduced in NSCLC tumor tissues

   The results showed that the expression of BZRAP1-AS1 was significantly
   lower in tumor tissues than in adjacent normal tissues ([68]Table 1,
   [69]Fig. 1A). The nonparametric test was performed. The median was
   0.240 (25–75 percentile, 0.046–4.556) in tumor tissues and 1.076 (25–75
   percentile, 0.106–13.380) in adjacent normal tissues (P = 0.004). Then,
   further subgroup analysis exhibited that in >60 years old [0.228
   (0.487–2.486) vs 1.048 (0.143–14.076), P = 0.016], male [0.226
   (0.027–3.267) vs 0.821 (0.097–12.904), P = 0.013], smoker [0.235
   (0.025–2.695) vs 0.737(0.061–14.324), P = 0.007], LUSC [0.126
   (0.017–2.486) vs 0.556 (0.059–11.459), P = 0.002], and stage II–IV
   subgroup [0.205 (0.019–3.697) vs 0.905 (0.081–13.452), P = 0.012],
   BZRAP1-AS1 was lower in tumor tissues than adjacent normal tissues.
   There was no expression difference in ≤60 years old, female,
   non-smokers, LUAD, and stage I subgroup ([70]Table 1, [71]Figs.
   1B–[72]1D).

Table 1. Relative BZRAP1-AS1 expression in cancer and normal tissues.

   Group No. of cases BZRAP1-AS1 expression (Median, 25–75 Percentile) P
   value
   Tumor tissues Normal tissues
   Total 63 0.240,0.046–4.556 1.076,0.106–13.380 0.004
   Ages
   ≤60 31 0.293,0.271–5.645 1.076, 0.528–9.019 0.092
   >60 32 0.228,0.487–2.486 1.048, 0.143–14.076 0.016
   Sex
   Male 46 0.226, 0.027–3.267 0.821, 0.097–12.904 0.013
   Female 17 1.261, 0.100–5.245 1.763, 0.130–13.620 0.124
   Smoking history
   No 22 0.751,0.086–4.913 1.380,0.165–13.416 0.236
   Yes 41 0.235,0.025–2.695 0.737,0.061–14.324 0.007
   Histologic type
   LUAD 35 1.020,0.123–4.581 1.413,0.167–13.380 0.174
   LUSC 28 0.126,0.017–2.486 0.556,0.059–11.459 0.002
   TNM stage
   I 26 1.128,0.081–4.847 1.467,0.153–12.988 0.096
   II–IV 37 0.205,0.019–3.697 0.905,0.081–13.452 0.012
   T stage
   T1 29 1.528,0.117–12.300 1.763, 0.264–17.199 0.090
   T2–T4 34 0.169, 0.021–1.594 0.364, 0.066–6.737 0.007
   N stage
   N0 40 1.354, 0.094–5.843 2.346, 0.216–14.076 0.011
   N1 + N2 23 0.180, 0.001–1.020 0.359, 0.038–1.413 0.153
   [73]Open in a new tab

   Notes.

   Bold indicates that the difference is statistically significant
   (P < 0.05).

Figure 1. Relative BZRAP1-AS1 expression in cancer and normal tissues
detected by RT-qPCR.

   [74]Figure 1
   [75]Open in a new tab

   Relative BZRAP1-AS1 expression in 63 pairs NSCLC cancer (C) and normal
   (N) tissues from Beijing Chest Hospital detected by RT-qPCR. (A)
   Relative BZRAP1-AS1 in cancer and normal tissues. (B) Relative
   BZRAP1-AS1 expression in cancer and normal tissues of different
   histologic types. (C) Relative BZRAP1-AS1 expression in cancer and
   normal tissues of nonsmokers and smokers in own samples. (D) Relative
   BZRAP1-AS1 expression in the NSCLC TNM stage I and stage II-IV groups.
   (E) Relative BZRAP1-AS1 expression in the NSCLC stage T1 and T2-T4
   groups. (F–G) Relative BZRAP1-AS1 expression in stage N0 and N1 + N2
   groups. *Statistical significance P < 0.05, **Statistical significance
   P < 0.01 in own samples.

Relation between the expression of BZRAP1-AS1 and the clinical
characteristics of NSCLC

   To find the potential biological functions of BZRAP1-AS1, the relation
   between clinical characteristics and BZRAP1-AS1 in tumor and adjacent
   normal tissues was analyzed respectively ([76]Table 2). In tumor
   tissues, BZRAP1-AS1 was lower in T2–T4 than T1 patients [0.169
   (0.021–1.594) vs 1.528 (0.117–12.300), P = 0.040], and N1-N2 than N0
   patients [0.180 (0.001–1.020) vs 1.354 (0.094–5.843), P = 0.032]
   ([77]Table 2, [78]Figs. 1E–[79]1F). And in adjacent normal tissues,
   BZRAP1-AS1 was lower in N1+N2 than N0 patients (0.359 (0.038–1.413) vs
   2.346 (0.216–14.076), P = 0.018). Whether tumor tissues or adjacent
   normal tissues was no difference in subgroups divided by smoking
   history, histologic type, pleura invasion, and vessel carcinoma
   embolus. Additionally, the analysis results in subgroups may be biased
   because of the limitation of sample size, which still need to be
   validated.

Table 2. Clinicopathological characteristics of NSCLC patients and expression
of BZRAP1-AS1 in tumor tissues and normal tissues.

   Variables No. of Cases BZRAP1-AS1 expression (Median, 25–75 Percentile)
   Tumor tissues P value Normal tissues P value
   Smoking history 0.535 0.604
   No 22 0.751, 0.086–4.913 1.380, 0.165–13.416
   Yes 41 0.235, 0.025–2.695 0.737, 0.061–14.324
   Histologic type 0.103 0.326
   LUAD 35 1.020, 0.123–4.581 1.413, 0.167–13.380
   LUSC 28 0.126, 0.017–2.486 0.556, 0.059–11.459
   T stage 0.040 0.071
   T1 29 1.528, 0.117–12.300 1.763, 0.264–17.199
   T2–T4 34 0.169, 0.021–1.594 0.364, 0.066–6.737
   Lymphatic metastasis 0.032 0.018
   N0 40 1.354, 0.094–5.843 2.346, 0.216–14.076
   N1 + N2 23 0.180, 0.001–1.020 0.359, 0.038–1.413
   Pleura invasion 0.333 0.956
   Negative 35 0.185, 0.046–1.564 0.737, 0.135–13.525
   Positive 28 1.408, 0.048–5.379 1.145, 0.053–12.290
   Vessel carcinoma embolus 0.596 0.845
   Negative 37 0.235, 0.037–3.696 0.905, 0.120–10.858
   Positive 26 0.630, 0.082–6.116 1.133, 0.077–14.382
   [80]Open in a new tab

   Notes.

   Bold indicates that the difference is statistically significant
   (P < 0.05).

BZRAP1-AS1 expression and clinical parameters in TCGA

   To validate the clinicopathological value of BZRAP1-AS1 in NSCLC, we
   analyzed TCGA data using the SPSS, and Graphpad Prism. The results were
   in accordance with our RT-qPCR data. BZRAP1-AS1 was low expressed in
   NSCLC cancer tissue compared with normal tissue (41.00 (24.00–72.00) vs
   69.00 (37.00–95.00), P < 0.001) ([81]Fig. 2A). In different histologic
   types, BZRAP1-AS1 expression also was lower in LUSC cancer tissues than
   normal tissues [40.00 (24.00–69.00) vs 91.00 (76.00–114.00),
   P < 0.001], but no difference in LUAD ([82]Figs. 2B–[83]2C). In NSCLC,
   BZRAP1-AS1 was lower in T2–T4 than T1 patients (38.00 (22.00–63.00) vs
   58.00 (32.75–91.25), P < 0.001), N1-N3 than N0 patients (38.00
   (23.00–63.00) vs 43.00 (24.00–76.00), P = 0.018) ([84]Fig. 2D and
   [85]2G). In subgroup analysis, BZRAP1-AS1 was lower in T2–T4 than T1
   patients (38.50 (21.00–62.00) vs 58.50(32.00–95.00), P < 0.001), N1-N3
   than N0 patients (34.00 (22.00–57.00) vs 46.00 (27.00–83.00),
   P = 0.001) in LUAD ([86]Figs. 2E–[87]2F). BZRAP1-AS1 was lower in T2–T4
   than T1 patients (37.00 (22.00–63.00) vs 57.00 (33.75–82.00),
   P < 0.001), but no difference in N stages in LUSC ([88]Figs.
   2H–[89]2I). The expression of BZRAP1-AS1 in normal tissues was no
   difference in N stages ([90]Fig. 2J).

Figure 2. Bioinformatics analysis of the clinicopathological significance of
BZRAP1-AS1 in TCGA and GTEX database.

   [91]Figure 2
   [92]Open in a new tab

   The clinicopathological significance of BZRAP1-AS1 in TCGA database.
   (A–C) BZRAP1-AS1 expression level in cancer and adjacent normal tissues
   of NSCLC, LUAD and LUSC from TCGA data. (D–F) BZRAP1-AS1 expression
   level in stage T1–T4 cancer tissues of NSCLC, LUAD and LUSC from TCGA
   data. (G–I) BZRAP1-AS1 expression level in stage N0-N3 cancer tissues
   of NSCLC, LUAD and LUSC from TCGA data. (J) BZRAP1-AS1 expression level
   in stage N0-N2 normal tissues of NSCLC from TCGA and GTEX normal
   tissues. *Statistical significance P < 0.05, **Statistical significance
   P < 0.01.

BZRAP1-AS1 participated in the pathological progress of NSCLC

   Whereas BZRAP1-AS1 was lower in tumor tissues than adjacent normal
   tissues in smokers, smoking was more meticulously analyzed. Compared
   with the adjacent normal tissues of nonsmokers, BZRAP1-AS1 level
   reduced in tumor tissues of nonsmokers, lower in nonmalignant tissues
   of smokers, and the lowest in tumor tissues of smokers ([93]Fig. 1H).
   In other words, BZRAP1-AS1 levels decreased gradually with smoking from
   adjacent normal tissues to tumor tissues. Then, the relationship
   between TNM stage and BZRAP1-AS1 in tumor tissues was more meticulously
   analyzed. BZRAP1-AS1 levels decreased gradually from I to III–IV stage,
   from tumors less than three cm to tumors larger than five cm, and from
   N0 to N2 ([94]Table 3 and [95]Fig. 3).

Table 3. Combined analysis of BZRAP1-AS1 expression and smoking history,
tumor size and N stage.

   Variables No. of cases BZRAP1-AS1 expression (Median, 25–75 Percentile)
   Smoking Tumor (T)/Adjacent Normal (N)
   NO N 22 1.380, 0.165–13.416
   NO T 22 0.751, 0.086–4.913
   YES N 41 0.737, 0.061–14.324
   YES T 41 0.235, 0.025–2.695
   TNM stage
   I 26 1.128, 0.081–4.847
   II 14 0.237, 0.063–15.210
   III–IV 23 0.001, 0.205–2.552
   Tumor size(cm)
   ≤3 29 1.528, 0.117–12.300
   >3, ≤5 22 0.240, 0.078–1.973
   >5 12 0.025, 0.001–0.182
   N metastasis
   N0 40 1.354, 0.094–5.843
   N1 4 0.098, 0.004–98.284
   N2 19 0.187, 0.001–1.020
   [96]Open in a new tab

Figure 3. (A–D) The BZRAP1-AS1 expression level stratified with smoking, TNM
stage, tumor size, and N metastasis in 63 pairs NSCLC cancer and normal
tissues detected by RT-qPCR.

   [97]Figure 3
   [98]Open in a new tab

Survival analysis

   Overall survival (OS) and the first progression (FP) of NSCLC patients
   was analyzed by GraphPad Prism and Kaplan–Meier plotter in TCGA
   database. Quartile was selected as expression threshold to distinguish
   the high-expression and low-expression cohorts. Patients of low
   BZRAP1-AS1 level shown poor prognosis. Further subgroup analysis,
   compared with the high-expression group, LUAD patients of low
   BZRAP1-AS1 levels had shorter OS, but LUSC patients had no difference
   between two groups ([99]Figs. 4A–[100]4C). In KM Plotter, univariate
   analysis showed lower level BZRAP1-AS1 was related to poor FP in both
   NSCLC, and in LUAD and LUSC subgroup patients ([101]Figs. 4D–[102]4F).
   Next, multivariate survival analysis results also indicated that
   BZRAP1-AS1 was significantly associated with the prognosis of NSCLC
   patients ([103]Fig. 4G).

Figure 4. Bioinformatics analysis of the prognostic value of BZRAP1-AS1 in
TCGA database.

   [104]Figure 4
   [105]Open in a new tab

   (A–C) Kaplan–Meier survival analysis of overall survival of NSCLC based
   on TCGA cohort stratified by BZRAP1-AS1 expression (bottom 25% versus
   top 25%). (D–F) Kaplan–Meier survival analysis of first progression of
   NSCLC based on TCGA cohort by Kaplan-Meier plotter online database. (G)
   Multivariate survival analysis including histology, TNM stage, gender,
   smoking history and BZRAP1-AS1 expression (stratified by median) made
   by Kaplan-Meier plotter online database.

Overexpression of BZRAP1-AS1 suppressed proliferation, invasion, and
migration of NSCLC cells

   To validate the role of BZRAP1-AS1 in NSCLC, we firstly detected the
   level of BZRAP1-AS1 in NSCLC cell lines (A549, H1299, H226, H1395,
   H1703, HCC827) by RT-qPCR. Except for A549 cells, the expression of
   BZRAP1-AS1 was very low in NSCLC cells ([106]Fig. 5A). Second, H1299
   and HCC827 cells with low BZRAP1-AS1 levels were selected to
   overexpress BZRAP1-AS1. As the results of RT-qPCR, the relative
   expression of BZRAP1-AS1 in H1299 OE was more than 8000 times higher
   than H1299 NC, and HCC827 OE nearly 400 times higher than HCC827 NC
   ([107]Fig. 5B). The results of cell growth assay showed BZRAP1-AS1
   over-expressing cells were slower than control cells ([108]Figs.
   5C–[109]5D). The results of colony formation assay showed colony number
   of H1299 OE and HCC827 OE were less than respective controls (37.33 ±
   3.38 vs 53.33 ± 1.76, P = 0.014), less than HCC827 NC (42.00 ± 1.16 vs
   76.33 ± 3.48, P < 0.001) ([110]Figs. 5E–[111]5F). These results
   indicated that overexpression of BZRAP1-AS1 inhibited proliferation of
   H1299 and HCC827 cells in vitro. In vivo, the results of mice tumor
   growth curve showed H1299 OE were slower than H1299 NC groups
   (P < 0.001), and HCC827 OE than HCC827 NC group (P = 0.082) ([112]Figs.
   5G–[113]5H). The tumor weight of H1299 OE was less than H1299 NC (0.382
   ± 0.039 vs 0.970 ± 0.127, P = 0.001), and HCC827 OE less than HCC827 NC
   group (0.342 ± 0.025 vs 0.415 ± 0.019, P = 0.042) ([114]Fig. 5I).
   Additionally, the migration ability of H1299 OE cells was weaker than
   H1299 NC group (29.000 ± 2.160 vs 53.500 ± 2.533, P < 0.001), similar
   to HCC827 OE cells compared with HCC827 NC cells (16.250 ± 1.493 vs
   58.250 ± 1.797, P < 0.001). Thus, overexpression of BZRAP1-AS1
   inhibited migration of H1299 and HCC827 cells ([115]Figs. 5J–[116]5K).

Figure 5. BZRAP1-AS1 overexpressing suppresses H1299 and HCC827 proliferation
and migration.

   [117]Figure 5
   [118]Open in a new tab

   (A) BZRAP1-AS1 relative expression in NSCLC cell lines detected by
   RT-qPCR. (B) qPCR detection of BZRAP1-AS1 overexpression in H1299 and
   HCC827. OE, overexpression; NC, negative control (C–D) CCK-8 assays to
   measure the proliferation abilities of BZRAP1-AS1 overexpressing H1299
   and HCC827 cells. (E-F) Colony formation ability of BZRAP1-AS1
   overexpressing H1299 and HCC827 cells. (G–H) Mice Xenograft tumor
   growth of BZRAP1-AS1 overexpressing H1299 and HCC827 cells in BALB/c-nu
   mice, n = 6. (I) The tumor weight from BZRAP1-AS1 overexpressing and
   control cells. (J–K) Transwell assay of BZRAP1-AS1 overexpressing H1299
   and HCC827 cells. (×10). *Statistical significance P < 0.05,
   **Statistical significance P < 0.01, ***Statistical significance P <
   0.001.

Transcriptomic analysis of mice xenograft tumor tissues and co-expression
analysis

   RNA-seq analysis of mice tumor tissues derived from BZRAP1-AS1
   over-expressing H1299 cells and the control cells showed there were 405
   up-regulated genes and 376 down-regulated genes ([119]Fig. 6A). And GO
   analysis showed BZRAP1-AS1 correlated with metal ion binding,
   translation factor activity (RNA binding), translation repressor
   activity (nucleic acid binding), coreceptor activity involved in Wnt
   signaling pathway ([120]Figs. S1A–[121]S1B). The results of KEGG
   pathway enrichment analysis showed BZRAP1-AS1 correlated with
   tumor-associated signaling pathways including cell adhesion molecules
   (CAM), MAPK signaling pathway, mTOR signaling pathway, pathways in
   cancer ([122]Figs. 6B–[123]6C). According to the above results, we
   traced to the KEGG Pathway/GO annotation of differentially expressed
   genes, further predicted the genes affected by BZRAP1-AS1. Finally,
   twelve genes were predicted, including five down-regulated genes
   RPS6KA6, LUM, TWIST1, PRKACB, TWIST2, and seven up-regulated genes
   PPM1A, FGF14, CNTN1, L1CAM, VDR, DPYSL3, ARHGAP4 ([124]Table S2). Then
   co-expression analysis was performed to validate the correlation
   between BZRAP1-AS1 and twelve predicted genes in LUAD and LUSC
   preliminarily, and results showed BZRAP1-AS1 was positively correlated
   with RPS6KA6, LUM, PRKACB, FGF14, VDR, DPYSL3 and ARHGAP4 ([125]Figs.
   S2–[126]S3). In summary, FGF14, VDR, DPYSL3 and ARHGAP4 may be the
   genes affected by BZRAP1-AS1 to function in LUAD and LUSC ([127]Fig.
   6D).

Figure 6. (A) Heatmap of differentially expressed genes between H1299 NC and
H1299 OE group in RNAseq of mice Xenograft tumor tissues (405 up-regulated
genes and 376 down-regulated genes). (B–C) Statistics of KEGG pathway
enrichment analysis results for up-regulated and down-regulated genes with
overexpression of BZRAP1-AS1. (D) Venn diagram for twelve predicted genes
affected by BZRAP1-AS1 in RNA sequencing data and co-expression analysis in
LUAD and LUSC.

   [128]Figure 6
   [129]Open in a new tab

Discussion

   Increasing studies show lncRNA is closely correlated with diagnosis,
   prognosis, and drug resistance of tumors ([130]Goodall &
   Wickramasinghe, 2021; [131]Zhang et al., 2020; [132]Sun et al., 2020).
   LncRNA may be a next breakthrough in the fight against tumors ([133]Tan
   et al., 2021b; [134]Bhan, Soleimani & Mandal, 2017). BZRAP1-AS1 is a
   lncRNA being found in recent years. There are few studies on the
   relationship between BZRAP1-AS1 and tumors. In hepatocellular carcinoma
   (HCC), BZRAP1-AS1 is high expression and knockdown of BZRAP1-AS1
   inhibits cell proliferation, migration, and angiogenesis in
   HUVEC ([135]Wang et al., 2019b), which suggests BZRAP1-AS1 may play a
   role in promoting tumorigenesis. But in cervical cancer and pancreatic
   cancer, low level of BZRAP1-AS1 is correlated with poor
   OS ([136]Giulietti et al., 2018; [137]Zheng et al., 2020b), which
   suggests that BZRAP1-AS1 may play a role in inhibiting tumorigenesis.
   These inconsistent results indicate that BZRAP1-AS1 exerts different
   functions in the different tumors.

   For lung cancer, the relationship between BZRAP1-AS1 and prognosis is
   only studied by bioinformatics analysis in lung adenocarcinoma, and
   functional studies of BZRAP1-AS1 in lung cancer are very
   limited ([138]Wang et al., 2019a; [139]Tang et al., 2021). NSCLC
   patients account for 85% of lung cancer patients. Therefore, we focused
   on the relationship between BZRAP1-AS1 and NSCLC. Firstly, we detected
   the level of BZRAP1-AS1 in tumor tissues and adjacent normal tissues of
   63 NSCLC patients to investigate whether there is expression difference
   between them. We find BZRAP1-AS1 level is lower in tumor tissues than
   adjacent normal tissues. Further subgroup analysis exhibits that
   BZRAP1-AS1 is lower in tumor tissues than adjacent normal tissues in
   smoking group and not different in nonsmoking group, which suggests
   smoking may cause a decrease of BZRAP1-AS1 in tumor tissues. Then
   smoking is more meticulously divided and the effects of smoking on
   BZRAP1-AS1 expression are analyzed in tumor tissues and adjacent normal
   tissues. Compared with adjacent normal tissues of nonsmokers,
   BZRAP1-AS1 level reduced in tumor tissues of nonsmokers, lower in
   nonmalignant tissues of smokers, and the lowest in tumor tissues of
   smokers. Thus, we speculate BZRAP1-AS1 has a greater impact on in the
   process of smoking-related lung cancer. In addition, subgroup analysis
   shows BZRAP1-AS1 is lower in tumor tissues than adjacent normal tissues
   in male and LUSC groups and not different in female and LUAD groups. We
   think these differences of BZRAP1-AS1 in the male and LUSC groups have
   nothing to do with gender and pathological types, but because there is
   a higher proportion of smokers in male than female patients and in LUSC
   than LUAD groups (87.0% vs 5.9%, 56.5% vs 11.8%). Also, smoking is a
   major high-risk factor for LUSC ([140]Pesch et al., 2012). Then
   subgroup analysis in tumor tissues shows BZRAP1-AS1 is lower in T2–T4
   than in T1, and in N1-N2 than in N0. In other words, BZRAP1-AS1 level
   is high in small tumors and low in big tumors, and high in tumors
   without N2 lymph nodes metastasis and low in tumors with N2 lymph nodes
   metastasis. When TNM stage, tumor size, and N metastasis more
   meticulously analyzed, BZRAP1-AS1 expression reduces gradually from I
   to III-IV, from ≤ three cm to >5 cm, from N0 to N2. In other words, the
   levels of BZRAP1-AS1 associated with the malignancy of NSCLC. So, we
   deduce that BZRAP1-AS1 participates in the pathological progress of
   NSCLC. To validate the hypothesis, TCGA data is analyzed. The results
   are consistent with our data. In addition, the survival analysis
   results show BZRAP1-AS1 is related to OS in LUAD, and FP in LUAD and
   LUSC. Patients with high level of BZRAP1-AS1 have a better prognosis,
   which is consistent with the literatures ([141]Giulietti et al., 2018;
   [142]Zheng et al., 2020b).

   In addition, we study the expression of BZRAP1-AS1 in NSCLC cell lines.
   In six NSCLC cell lines, Except for A549 cells, the expression of
   BZRAP1-AS1 is very low in five cells and only high in A549 cells. This
   result is consistent with that of tumor tissues. That is to say that
   the expression of BZRAP1-AS1 is very low in tumors. As for A549 cells,
   we think it is just a one example. How does BZRAP1-AS1 affect
   pathological process of NSCLC? We overexpress BZRAP1-AS1 in NSCLC cells
   to study the effect of BZRAP1-AS1 on cell proliferation, invasion, and
   migration in vitro and in vivo. The data indicate that overexpressing
   BZRAP1-AS1 indeed suppresses proliferation, invasion, and metastasis of
   NSCLC cells. This further confirms BZRAP1-AS1 plays an anticancer role
   by inhibiting proliferation, invasion, and migration. This is
   consistent with the data in cervical cancer and pancreatic
   cancer ([143]Zheng et al., 2020b; [144]Tang et al., 2021).

   To further explore which genes are affected by BZRAP1-AS1 in NSCLC,
   transcriptome analysis was performed. By GO enrichment analysis and
   KEGG pathway enrichment analysis, we screened twelve potentially
   relevant genes: RPS6KA6, LUM, TWIST1, PRKACB, TWIST2, PPM1A, FGF14,
   CNTN1, L1CAM, VDR, DPYSL3, ARHGAP4. After co-expression verification,
   four genes were finally concerned, namely FGF14, VDR, DPYSL3 and
   ARHGAP4. According to available literatures, these four genes play a
   role in suppressing tumor ([145]Babina & Turner, 2017; [146]Turner &
   Grose, 2010; [147]Turkowski et al., 2020; [148]Campbell & Trump, 2017;
   [149]Yang et al., 2018; [150]Shen et al., 2019). So we guess BZRAP1-AS1
   exerts its tumor suppressive function by influencing above four genes.
   BZRAP1-AS1 is expected to become a new therapeutic target for NSCLC.

Conclusion

   BZRAP1-AS1 expression negatively correlates with malignancy grades of
   NSCLC. BZRAP1-AS1 plays an anticancer role by inhibiting cell
   proliferation, invasion, and metastasis, and has promising prognostic
   value in NSCLC. BZRAP1-AS1 may serve as a potentially diagnostic marker
   and therapeutic target for NSCLC.

Supplemental Information

   Figure S1. Statistics of GO molecular function (MF) enrichment results
   for up-regulated and down-regulated genes with overexpression of
   BZRAP1-AS1.
   [151]Click here for additional data file.^ (212.2KB, png)
   DOI: 10.7717/peerj.13871/supp-1
   Figure S2. Co-expression analysis in LUAD for BZRAP1-AS1 and twelve
   predicted target genes RPS6KA6, LUM, TWIST1, PRKACB, TWIST2, PPM1A,
   FGF14, CNTN1, L1CAM, VDR, DPYSL3, ARHGAP4.
   [152]Click here for additional data file.^ (2.3MB, png)
   DOI: 10.7717/peerj.13871/supp-2
   Figure S3. Co-expression analysis in LUSC for BZRAP1-AS1 and twelve
   predicted target genes RPS6KA6, LUM, TWIST1, PRKACB, TWIST2, PPM1A,
   FGF14, CNTN1, L1CAM, VDR, DPYSL3, ARHGAP4.
   [153]Click here for additional data file.^ (2.1MB, png)
   DOI: 10.7717/peerj.13871/supp-3
   Table S1. The proportion of smoking patients in sex and histologic
   type.
   [154]Click here for additional data file.^ (13.1KB, docx)
   DOI: 10.7717/peerj.13871/supp-4
   Table S2. Annotation information and expression changes of predicted
   target genes with the overexpression of BZRAP1-AS1 (Mouse tumor tissue
   sequencing data).
   [155]Click here for additional data file.^ (16.2KB, docx)
   DOI: 10.7717/peerj.13871/supp-5
   Supplemental Information 1. ARRIVE 2.0 checklist.
   [156]Click here for additional data file.^ (128.4KB, pdf)
   DOI: 10.7717/peerj.13871/supp-6
   Supplemental Information 2. Raw RNAseq data from mice tumor.

   Eexpression differential analysis for RNAseq of mice tumor.
   [157]Click here for additional data file.^ (6.1MB, zip)
   DOI: 10.7717/peerj.13871/supp-7
   Data S1. Raw data.
   [158]Click here for additional data file.^ (26.9MB, zip)
   DOI: 10.7717/peerj.13871/supp-8

Funding Statement

   This study was funded by the Beijing Municipal Natural Science
   Foundation (No. 4192062), and the National Natural Science Foundation
   of China (No. 61801477). The funders had no role in study design, data
   collection and analysis, decision to publish, or preparation of the
   manuscript.

Contributor Information

   Weiying Li, Email: li_weiying412@aliyun.com.

   Shaofa Xu, Email: xushaofa2017@yeah.net.

Additional Information and Declarations

Competing Interests

   The authors declare there are no competing interests.

Author Contributions

   Xuefeng Hao conceived and designed the experiments, performed the
   experiments, analyzed the data, prepared figures and/or tables,
   authored or reviewed drafts of the article, and approved the final
   draft.

   Minghang Zhang analyzed the data, prepared figures and/or tables, and
   approved the final draft.

   Meng Gu performed the experiments, analyzed the data, prepared figures
   and/or tables, and approved the final draft.

   Ziyu Wang conceived and designed the experiments, performed the
   experiments, prepared figures and/or tables, and approved the final
   draft.

   Shijie Zhou analyzed the data, prepared figures and/or tables, and
   approved the final draft.

   Weiying Li conceived and designed the experiments, prepared figures
   and/or tables, authored or reviewed drafts of the article, and approved
   the final draft.

   Shaofa Xu conceived and designed the experiments, authored or reviewed
   drafts of the article, and approved the final draft.

Human Ethics

   The following information was supplied relating to ethical approvals
   (i.e., approving body and any reference numbers):

   This study was approved by the Ethics Committee of Beijing Chest
   Hospital (Ethical Approval Number: 2019-71).

Animal Ethics

   The following information was supplied relating to ethical approvals
   (i.e., approving body and any reference numbers):

   This study was approved by the Ethics Committee of Beijing Chest
   Hospital (Ethical Approval Number: 2019-71).

Data Availability

   The following information was supplied regarding data availability:

   The raw data are available in the [159]Supplemental Files.

References