Chemicals and Materials
MS-grade methanol, formic acid, acetic acid and acetonitrile were purchased from Fisher scientific. Ultra-pure water (18.2MΩ) was prepared with a Milli-Q water purification system (Millipore, Bedford, MA, USA). Glycine (Gly), β-Alanine (β-Ala), Sarcosine (Sar), L-Alanine (Ala), L-Serine (Ser), L-Proline (Pro), L-Valine (Val), L-Threonine (Thr), L-Cysteine (Cys), Taurine (Tau), L-Leucine (Leu), γ-aminobutyric acid (GABA), L-Isoleucine (Ile), trans-4-Hydroxy-L-Proline (t-Hyp), L-Asparagine (Asn), L-Aspartic Acid (Asp), L-Ornithine (Orn), L-Lysine (Lys), L-Glutamic Acid (Glu), L-Methionine (Met), 1-Methyl-L-histidine, 3-Methyl-L-histidine, L-2-Aminoadipic acid, L-Phenylalanine (Phe), L-Arginine (Arg), L-Citrulline (Cit), L-Tyrosine (Tyr), L-Tryptophan (Trp), L-Carnosine, Folic acid (FA) , Kynurenic acid, Pipecolinic acid were purchased from Sigma-Aldrich company (St. Louis, MO, USA); L-Histidine (His) and Glutamine (Gln) were purchased from TCI (Develoment Co., Ltd, Shanghai, China); Phenylalanine-d5 (Internal standard, IS) was purchased from Toronto Research Chemicals (YTO, Canada). The categories of lysophosphatidylcholine (LPC), phosphatidylcholine (PC), lysophosphatidylethanolamine (LPE), phosphatidylethanolamine (PE), sphingomyelin (SM), ceramide (Cer), were purchased from Avanti Polar Lipids company (Alabaster, AL, USA); triacylglycerol (TAG), free fatty Acid were purchased from Larodan company (Stockholm, Sweden). ISs of PC (19:0/19:0), LPC (19:0/0:0), PE (12:0/13:0), SM(d18:1/12:0), Cer (d18:1/17:0) were purchased from Avanti Polar Lipids company (Alabaster, AL, USA), TAG (15:0/15:0/15:0), FFA (C19:0), d3-FFA(C16:0) were purchased from Larodan company (Stockholm, Sweden). All other chemicals and reagents used were of analytical grade.
Study population
Patients that presented to the Liaocheng People’s Hospital LC clinic with suspected LC from September 2016 - January 2018 were included in the present study. A total of 51 paired pre- and post-operative serum samples were collected from patients that had provided written informed consent to participate (Table 1). Following tumor resection, histopathological confirmation of LC diagnosis was made, and tumor stage and clinicohistopathological characteristics were analyzed. Patients enrolled in the present analysis had not undergone any pre-operative radiotherapy or chemotherapy. As a control group, 51 age- and sex-matched healthy volunteers were recruited. The Ethics Committee of Liaocheng People's Hospital approved this study, which was conducted in a manner consistent with the Declaration of Helsinki.
Sample preparation
Preoperative serum samples were collected from LC patients on the morning following hospitalization, while postoperative serum samples were collected on the last morning prior to discharge for each patient (7 days post-operation). For all sample collections, peripheral blood (~3.2 mL) was collected in a Vacutainer tube (BD, NJ, USA), and samples were stored at room temperature (RT) prior to analysis for CTC levels. For metabolomics assays, an additional 2 mL of serum was collected for analysis, with control samples being collected contemporaneously.
Clinical effect criteria
RECIST (Response Evaluation Criteria in Solid Tumors) v1.1 was used for the evaluation of clinical responses. Responses were classified based upon tumor diameter and axillary lymph node status as follows: complete response (CR), stable disease (SD), partial response (PR), or progressive disease (PD). Clinical efficacy was considered to have been achieved when PR or CR was diagnosed. For analysis purposes, patients with CR or PR were incorporated into one analysis group (CR + PR) while patients with SD or PD were incorporated into a second group (SD + PD). Patients underwent follow-up through October 31, 2019, via phone call and outpatient visits. The time between surgery and the last diagnosis of recurrence was 21-37 months A total of 14 LC patients suffered from disease recurrence, while 31 did not suffer from recurrence and 6 did not have sufficient available clinical information to determine whether disease had recurred.
CTC detection
Negative enrichment fluorescence in situ hybridization (NE-FISH) was used to detect CTCs as in our previous studies [26, 27]. Briefly, patient blood samples were washed and RBCs were lysed with the CS1 and CS2 buffers, respectively, followed by resuspension in a solution containing magnetic particles bound to anti-CD45 (Cyttel Biosciences INC., Jiangsu, China). Gradient centrifugation was then performed using the CS3 buffer (Cyttel) to separate samples, and a magnetic stand was used to collect the CTC-containing sample. This sample was then fixed and spread onto a slide, which was then dried for analysis. FISH was then performed using the chromosome centromere probe (CEP) 8 + 17, with DAPI used to stain cell nuclei and AF594-conjugated anti-CD45 being used to stain cells. Slides were then imaged via microscopy (BX63, Olympus), and an IMSTAR high content screening device equipped with the PathfinderTM software (IMSTAR S.A., Paris, France) was used for image analysis. CTCs were cells that were found to be DAPI+/CD45−/CEP+ (Fig. 1).
UPLC-MS/MS
Amino acids, lipids, and fatty acids were analyzed with a UPLC system (Waters ACQUITY) that contained a quaternary pump, an autosampler, a degasser, and a Xevo TQ-S micro mass spectrometer with an ESI ionization source (Waters, MA, USA). The amino acid analysis was conducted following sample separation with a Waters ACQUITY UPLC BEH Amide column (2.1 mm × 100 mm, 1.7 μm; 0.3 mL/min flow rate). For this separation, the mobile phases used were 0.1% and 0.2% formic acid (A and B, respectively). A linear gradient was used for separation as follows: 0–0.5 min, 15 % B; 0.5-5.5 min, 15–20 % B; 5.5–12.5 min, 20–40 % B; 12.5–13 min, 40-15 % B; and 13-15min, 15% B. An initial 5 μL injection volume was used, and the column was warmed to 40 °C. Between each sample injection, the column was washed once twice using a weak 90% acetonitrile solution and a strong 10% acetonitrile solution.
A Waters ACQUITY UPLC BEH C8 column (2.1 × 100 mm, 1.7 μm; 0.26 mL/min flow rate) was used for analyses of fatty acids and lipids. For these separations, 60 % acetonitrile in water containing 5 mM ammonium formate and 90 % isopropanol in acetonitrile containing 5 mM ammonium formate served as the mobile phases (A and B, respectively). A linear gradient was used for separation as follows: 0–1.0min, 100% A; 1.0-2.0 min, 100–70% A; 2.0–12.0 min, 70–30% A; 12.0–12.5 min, 30-5 % A; 12.5-13.0 min, 5-0%A; 13.0-14.0 min, 0%A; 14.0-14.1 min, 0-100% A and 13-15 min, 100% A. Injection volumes in positive and negative ion mode were 1 μL and 2 μL, respectively, and the column was warmed to 55 °C.
Positive and negative ion modes and multiple reaction monitoring (MRM) were employed during micro mass spectrometer (Xevo TQ-S) operation, with data being acquired and processed using the Waters TargetLynx software.
Data Analysis
SPSS (v17, SPSS Inc, IL, USA) was used for all statistical testing. Receiver operating characteristic (ROC) curves were used to explore the diagnostic sensitivity and specificity of CTCs for LC diagnosis. Data were compared via Student’s t-test, with P < 0.05 as the significance threshold.
Metaboanalyst [28, 29] was used for multivariate statistical analyses of metabolomics data, including partial least squares discriminant analysis (PLS-DA), biomarker analyses, and pathway analyses. The R2 and Q2 statistics were utilized to assess PLS-DA model quality, while its reliability was assessed via a permutation test owing to the potential for this model to overestimate separation performance [28, 30].
For a pathway analysis of differentially expressed metabolites, pathway modules and pathway topology analyses were used to best identify key biological pathways linked to the observed changes in metabolite levels. The resultant data show both P-values in the pathway enrichment analysis (y-axis) and the pathway impact values from pathway topology analysis (x-axis), with red being used to identify the most affected pathways.