Study design
A total of 46 newly diagnosed pulmonary nodules patients aged 18–70 years old who were proven as LUAD by pathological biopsy later were included in our study at Guangdong Provincial People’s Hospital (Guangzhou, China) between October 2019 and October 2020. None of the patients received chemotherapy or radiotherapy before the biopsy. All patients recruited were evaluated the severity of hypoxia by overnight polysomnography (PSG). AHI ≥ 5 events/hour was defined as LUAD with OSA (Group LUAD + OSA). AHI < 5 events/hour was recruited into LUAD without OSA group (Group LUAD). According to PSG results, 23 patients were divided into Group LUAD + OSA and 23 patients were into Group LUAD. Among them, 10 patients (5vs5) were randomly enrolled in the sequencing set, and the remaining 36 patients were in the validation set. Blood samples were collected from all enrolled patients with EDTA vacutainer tubes (BD, Franklin Lakes, USA). All patients included have signed informed consent for blood sample collection, PSG, and the following research. The protocol was approved by the ethics committee of Guangdong Provincial People’s Hospital and the registration number is No. GDREC2017259H (R1).
Plasma Evs Isolation
We collected 5mL peripheral blood from the median cubital vein of patients at 6. am before breakfast. The blood samples were centrifuged at 3000×g for 15 min at 4°C within 30minutes and the supernatant was transferred into a 1mL fresh Eppendorf tube, storing at -80℃. The isolation of plasma exosomes was performed with a commercial kit (EXO-QUICK, SBI) according to the standard protocol[7, 8]. After thawing at room temperature, the plasma samples were centrifuged at 2000×g for 15min and the 4uL thrombin was added to each 500uL supernatant to remove the fibrin. The supernatant was centrifuged again at 10000×g for 5 min and was transferred to a fresh 1.5mL tube. One-fourth volume of Exo-Quick Reagents was mixed with supernatant and incubated overnight. The mixture was centrifuged at 1500×g for 30min and the precipitates were exosome pellets. Phosphate buffered saline (PBS) was added to the exosome pellet to resuspend the exosome pellet.
Exosomes Identification, Rna Isolation From Plasma Evs And High-throughput Sequencing
Transmission electron microscopy (TEM, Hitachi High-Technologies, Japan) was conducted to identify the shape and structures of EVs. The particle size distribution of EVs was analyzed by Zetasizer Ultra(Malvern, UK). Flow cytometry (BD accuri C6, BD Biosciences, USA) analyzed the expression of specific exosomal markers (CD81 and CD63) on the surface of the sample. Exosome thawed at 4℃ and 1000uL RNAiso plus reagent (Takara, Japan) was added into the resuspended EVs, vortexed, and incubated for 10seconds. Thereafter, 200µL chloroform was mixed with the sample and vortexed again, centrifuging at 12,000×g for 15minutes. Then, the upper aqueous phase (RNA) was carefully collected and transferred to a fresh 1.5ml tube, mixing with 400uL isopropanol and centrifuging at 12,000×g for 15minutes. The lower organic phase (protein and DNA) was discarded. The precipitates were washed by absolute ethyl alcohol twice and air-dried for 15minutes. Finally, the EVs RNA was dissolved in DEPC water, mixing thoroughly. Nanodrop ND measured the concentration and purity of the extracted RNA (A260/280 ratio was equal to or close to 2.0). Total RNA or purified small RNA fragments were subjected to RT-PCR first-strand synthesis followed by RNA 3’ and RNA 5’ adaptor ligation and later sequenced on Illumina HiSeqTM 2500 to obtain the differentially expressed genes. The process of constructing RNA library and sequencing was conducted by RiboBio (Guangzhou, China)
Real-time Pcr Analysis Of Evs Mirnas
Purified RNA was reversely transcribed into cDNA by constructing a reaction system (5uL mRQ Reaction Buffer, 3.75uL RNA Sample, 1.25uL mRQ Enzyme Mix). The real-time PCR for miRNA was performed using SYBR Premix Ex Taq Reagent Kit (Takara, Japan) with Bio-rad detection system.
Statistical analysis
SPSS software (version 24.0, Chicago, IL) and GraphPad Prism (La Jolla, CA) were used to perform the statistical analyses. The expression level of miRNAs relative to endogenous controls was calculated using the 2 -∆∆Ct method. Two samples t-test was used for measurement samples conforming to normal distribution. Mann–Whitney U test was used for comparison between groups conforming to normal distribution. p < 0.05 was considered to be significant.