Detection of miR-155-5p and imaging lung cancer for early diagnosis: in vitro and in vivo study

Human lung adenocarcinoma cells A549 and human small cell lung cancer cells (SCLC) H446 were purchased from American Type Culture Collection (Manassas, VA, USA). Human lung adenocarcinoma cells SPC-A1 were purchased from the Shanghai Institute of Cellular Biology. A549, SPC-A1, and H446 lung cancer cell lines were cultured in an incubator with 5% CO₂ at 37 °C using RPMI 1640 medium (Hyclone, Waltham, MA, USA) containing 10% fetal bovine serum (Haoyang, Tianjin, China). According to the previous study (Yao et al. 2014), TICs (CD133⁺CD338⁺ cells) were isolated from A549 cells as described previously and cultured in serum-free DMEM/F12 (Hyclone, Waltham, MA, USA) medium.

All in vivo experiments were approved by the Xinqiao Hospital Animal Care and Use Committee. Under aseptic experimental conditions, 1 × 10⁶ A549 and H446 cells were subcutaneously injected into the left back or front of anesthetized 6–8-week nude mice to establish subcutaneous xenograft models of lung adenocarcinoma and small cell lung cancer. 1 × 10⁶ A549 and H446 cells were injected via the tail veins to establish lung xenograft models. Thirty days following injection of lung cancer cells, subcutaneous transplanted tumors (Vol ~ 300 mm³) and lung transplanted tumors appeared in all mice. Cre adenovirus (Hanbio, Shenzhen, China), with the titer of 5 × 10⁹ PFU that can secrete Cre enzyme, was slowly instilled into the nasal cavity of the transgenic mice at the age of 8 weeks. These mice were sacrificed at weeks 4, 6, 8, and 12 after instilling adenovirus. One side of the lung tissues in the transgenic mice at different disease stages and the lung xenograft models were resected, fixed in 4% paraformaldehyde (Boster, Wuhan, China), and sliced as 4 μm paraffin sections for HE (hematoxylin–eosin) staining to observe the different pathological changes. The subcutaneous xenografts and lung xenografts from the other side were frozen in the liquid nitrogen for detecting miR-155-5p expression.

Based on a previous study (Zhu et al. 2014), CS nanoparticles (Guanghan Hengyu, Sichuan, China), hsa-miR-155-5p MB, mmu-miR-155-5p MB, or random sequence MB (RS MB) (Sangon, Shanghai, China) were mixed at the ratio of Wcs/W_MB = 7:1 to synthesize the CS-MB nanoparticles. Hsa-miR-155-5p MB (5′-Cy5-C + CAGCG-ACC + CCT + ATCA + CGAT + TAGCATTAA-CGCT + GG-BHQ3-3′) and Mmu-miR-155-5pMB (5′-Cy5-C + CAGCG-ACC + CCT + ATCA + CAAT + TAGCATTAA-CGCT + GG-BHQ3-3′) were designed according to the miR-155-5p sequences. A random sequence MB (RS MB: 5′-Cy5-C + CAGCG-AC + GCCA + ATG + ACC + TTA + AGCATTAA-CGCT + GG-BHQ3-3′) complementary to no known gene sequence was used as a negative control. The MB had a Cy5-molecule attached to the 5′-end and a black hole quencher-3(BHQ3) attached to the 3′-end. The underlining bases were the ones added to form a stem, and the +N represented the LNAs (locked nucleic acids) modified bases. All the synthetic oligodeoxynucleotides were purchased from Sangon Company (Shanghai, China). 1 × 10⁴ A549, SPC-A1, H446, and TIC lung cancer cells were seeded in culture dishes for 24 h. Subsequently, 200 μL of mixed CS-MB nanoparticles were added into the culture dishes, and 100 μL Opti-MEM^® I reduced serum medium (Life Technologies, Waltham, MA, USA) was added to adjust the concentration; and the final concentration of MB was 200 nmol/L. RS MB was used as a negative control. These cells were incubated in a CO₂ incubator at 37 °C for 2 h, followed by PBS washes. Then, the nuclei were stained with Hoechst 33342 (Beyotime, Shanghai, China) at 37 °C for 20 min, followed by imaging using confocal microscopy (Leica TCS Confocal Microscope, Leica, Wetzlar, Germany). An equivalent of 1 mL cell lysate (Beyotime, Shanghai, China) was added into each culture dish for cell lysis. A volume of 100 ml suspension was distributed into each of 6-wells of a 96-well plate, and the fluorescence intensity of Cy5 (excitation 649 nm/emission 670 nm) per well was detected by Varioskan Flash (Thermo, Waltham, MA, USA) in each group.

Total RNA of cells and tissues was extracted from A549, SPC-A1, H446 cells, and TICs, as well as subcutaneous xenografts, lung xenografts, and lung tissues of transgenic mice at different stages of the disease using RNAiso plus TRIzol reagent (TaKaRa, Japan). One microgram total RNA was used for cDNA synthesis,reverse transcription was performed with the PrimeScript RT reagent kit (Takara, Japan) at 37 °C for 15 min and 85 °C for 5 s. The changes in the expression of miR-155-5p in the four types of cells, xenografts, and transgenic mice lung tissues were detected using the SYBR PCR Master Mix reagent kits (TaKaRa, Japan). The PCR conditions for the miR-155 and U6 genes were: pre-denaturation at 95 °C for 30 s, followed by 40 cycles of denaturation 95 °C for 5 s and 60 °C for 34 s. For each sample, three duplicated wells were used for miRNA detection. U6 served as the internal reference gene and has-miR-155-5p or mmu-miR-155-5p as the target gene. The relative expression of miR-155-5p was evaluated using the ∆∆Ct method. All the premier sequences were synthesized by RiboBio Company (Riobio, Guangzhou, China). The hsa-miR-155-5p, mmu-miR-155-5p, and U6 Primer Set catalog numbers’ are MQPS0000685, MQPS0002476, and MQPS0000002, respectively.

After the mice were anesthetized by intraperitoneal injection of 4% chloral hydrate (Chron Chemicals, Chengdu, China), an equivalent of 100 μL of CS-miR-155-5p MB or CS-RS MB nanoparticles was injected via the tail veins. The final concentration of MB was 2 μM. There are two negative control groups, in the subcutaneous and lung xenograft nude mice models; A549 subcutaneous and lung xenografts models were used as the negative control groups after the injection of CS-RS MB; in the transgenic mice model, mice without intranasal inhalation of the adenovirus after the injection of CS-miR-155-5p MB were used as the negative control group. Depilatory cream was used for removing the black hair on the thorax of the transgenic mice for detection of Cy5 fluorescence signal. After 2 h, the mice were placed on the in vivo imaging system (IVIS) platform, and the fluorescence signals from subcutaneous xenografts, lung xenografts, and transgenic mice lung tissues were imaged and detected by the IVIS Spectrum Imaging System (Caliper Life Sciences, Boston, USA). After in vivo imaging, the mice were sacrificed by dislocating the neck. The xenografts and lung tissues were removed and imaged again by the IVIS Spectrum Imaging System. Next, the frozen sections were prepared and then fixed in paraformaldehyde, followed by DAPI (Beyotime, Shanghai, China) staining of the nucleus. After that, the fluorescent images were photographed by confocal microscopy.

This study was approved by the Medical Ethics Committee of Xinqiao Hospital (NO.2017004-011), and all participants were completely informed and signed the written informed consents. As described previously (Peng et al. 2005), three cases of fresh lung squamous cell carcinoma specimens and three cases of adenocarcinoma specimens were collected from the XinQiao Hospital of Army Medical University after confirmed by postoperative pathology. Frozen sections were prepared and fixed in ice acetone (Chron Chemicals, Chengdu, China) for 10 min. Then, 50 μL hsa-miR-155-5p MB or RS MB at 100 nmol/L was dropped to each section, followed by incubation at 37 °C for 60 min. Subsequently, the nuclei were stained by DAPI. The fluorescence signal of miR-155 in the lung tissues was detected by confocal microscopy; RS MB was used as the negative control.

An independent sample t test was performed for comparison between the two groups, and one-way ANOVA was performed for comparison among multiple groups. P < 0.05 was considered statistically significant. All data were represented as means ± standard deviations.

To determine whether the CS-MB probe could detect the level of miRNA expression in viable cells, their ability to detect target microRNAs was assessed. In vitro experiments showed strong red fluorescent signals in the cytoplasm in each group of cells in the presence of CS-miR-155 MB, while only a few signals were detected in the nuclei (Fig. 2a). However, no significant red fluorescent signals were detected in the CS-RS MB group (Fig. 2b). Furthermore, the average fluorescence intensity of the 6 wells in each group was calculated. The mean fluorescence intensity of TICs was the strongest, followed by H446, SPC-A1, and A549 cells, and the fluorescence intensity was significantly increased in the presence of the CS-miR-155-5p MB compare with the presence of the CS-RS MB (Fig. 2c). Moreover, the fluorescence intensity trend of the four cells, after miR-155-5p MB was transfected using CS nanoparticles, was similar to the expression of miR-155-5p detected by qRT-PCR (Fig. 2c). Therefore, CS-MB probe could be used for detecting miR-155-5p expression and imaging lung cancer cells.

HE staining was used to prove that lung xenograft model and different disease stages of transgenic mice were established, while qRT-PCR technology was used to detect the expression of miR-155-5p in lung cancer cells and animal models. HE staining revealed that lung adenocarcinoma cells A549 and small cell lung cancer cells H446 planting nodules were implanted in the lung tissues of nude mice (Fig. 3a), indicating successfully establishment of lung xenograft models. At 4, 6, 8, and 12 weeks after instillation of adenovirus, different disease processes of atypical hyperplasia, adenoma, carcinoma in situ, and lung adenocarcinoma could be observed in lung tissues of the transgenic mice (Fig. 3B), suggesting that lung cancer models of different disease stages were established successfully. qRT-PCR showed that miR-155-5p expressed in subcutaneous xenografts (SX) and lung xenografts (LX) of nude mice (Fig. 3c, d) as well as lung tissues of transgenic mice at different stages of the disease. In transgenic mice, miR-155-5p expression increased significantly with the progression of lung cancer as compared to that at 4, 6, 8, and 12 weeks (Fig. 3e), suggesting that miR-155-5p plays a major role in the occurrence and development of lung cancer, and can be used as a target for tracing, detection, and imaging.

The ability of CS-MB probe to detect miRNA in vivo was further investigated in xenografts models. The in vivo imaging showed fluorescent signals with varying intensities were detected in the subcutaneous xenograft and chests (Fig. 4a). After the tumor or lung tissues were removed, the imaging showed that the fluorescent signals were stronger in the H446 group than in the A546 group. The distribution of the fluorescent signals was found to be consistent with the size of subcutaneous xenografts (Fig. 4b). The fluorescence intensity was analyzed after injection using the Spectrum Living Image 4.0 software, and the results demonstrated that fluorescent signals in the H446 group with high miR-155-5p expression were stronger than those in the A549 group. However, no fluorescent signals were detected in the CS-RS MB negative control group of the A549 SX and LX models (Fig. 4c). Interestingly, the fluorescence intensity trend was similar to the expression of miR-155-5p detected by qRT-PCR (Fig. 3c, d). The frozen sections were prepared using tumor tissues and lung tissues and examined by confocal microscopy, which showed that the fluorescent signals were from some tumor cells in the subcutaneous xenografts model. In lung xenografts model, the fluorescent signals were from some tumor cells and some alveolar epithelial cells. Compared to the A549 group, the fluorescent signals were brighter in the H446 group (Fig. 4d). Moreover, the intense fluorescent signals in the cytoplasm suggested that CS nanoparticles could be used as carriers to transport MB in vivo, thereby facilitating its entry into the cytoplasm to specifically bind to miR-155 and produce the signals.

The ability of CS-MB nanoparticles to detect miR-155-5p at the different disease stages in the transgenic mice was further investigated. After CS-MB nanoparticles were injected via the tail veins using the same method, fluorescent signals of varying intensities could be detected in the lung tissues of transgenic mice (Fig. 5a). Also, the fluorescent signals were detected at different disease stages after removal of the lung tissue (Fig. 5b). Analysis of the fluorescence intensity showed an increasing trend with the progression of the disease (Fig. 5c). However, the fluorescence signals were not detected in the control group of mice without intranasal inhalation of the adenovirus. It was similar to the expression of miR-155 detected by qRT-PCR (Fig. 3e). Confocal microscopy of the frozen sections of the lung tissue showed that fluorescent signals were from the pleomorphic or cancer cells and some alveolar epithelial cells (Fig. 5d). The occurrence and development of lung cancer were dynamically monitored by detecting varying fluorescence intensities, which provided a novel technology for the early diagnosis of lung cancer.

The ability of miR-155-5p MB to detect miR-155-5p in human lung cancer tissues was investigated further. After miR-155-5p MB was added to frozen lung cancer tissues, red fluorescent signals with different intensities could be detected in squamous and adenocarcinoma tissues. The majority of the red fluorescent signals were detected in the cytoplasm of the tumor. However, no significant fluorescent signals were detected in the negative RS MB group (Fig. 6). These findings suggested that miR-155-5p MB can bind to miR-155-5p in the tumor tissues, leading to fluorescent signals, which laid a foundation for subsequent preclinical study.