Polyamine metabolites provide pathophysiological information on disease or therapeutic efficacy, yet rapid screening methods for these biomarkers are lacking. Here, we developed high-throughput polyamine metabolite profiling based on multisegment injection capillary electrophoresis triple quadrupole tandem mass spectrometry (MSI-CE-MS/MS).
Journal of Chromatography A 1652 (2021) 462355 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma High-throughput screening of salivary polyamine markers for discrimination of colorectal cancer by multisegment injection capillary electrophoresis tandem mass spectrometry Kaori Igarashi a, Sana Ota a, Miku Kaneko a, Akiyoshi Hirayama a, Masanobu Enomoto b, Kenji Katumata b, Masahiro Sugimoto a,c, Tomoyoshi Soga a,d,∗ a Institute for Advanced Biosciences, Keio University, 246-2 Mizukami, Kakuganji, Tsuruoka 997-0052, Japan Department of Gastrointestinal and Pediatric Surgery, Tokyo Medical University, 6-7-1, Nishijinjuku, Shinjuku, Tokyo 160-0023, Japan c Research and Development Center for Minimally Invasive Therapies, Medical Research Institute, Tokyo Medical University, 6-1-1, Sinjuku, Tokyo 160-0022, Japan d Faculty of Environmental Information Studies, Keio University, 5322 Endo, Fujisawa 252-0882, Japan b a r t i c l e i n f o Article history: Received 13 April 2021 Revised June 2021 Accepted 15 June 2021 Available online 20 June 2021 Keywords: Multisegment injection capillary electrophoresis Mass spectrometry Saliva Polyamine Biomarker Colorectal cancer a b s t r a c t Polyamine metabolites provide pathophysiological information on disease or therapeutic efficacy, yet rapid screening methods for these biomarkers are lacking Here, we developed high-throughput polyamine metabolite profiling based on multisegment injection capillary electrophoresis triple quadrupole tandem mass spectrometry (MSI-CE-MS/MS), which allows sequential 40-sample injection followed by electrophoretic separation and specific mass detection To achieve consecutive analysis of polyamine samples, M formic acid was used as the background electrolyte (BGE) The BGE spacer volume had an apparent effect on peak resolution among samples, and 20 nL was selected as the optimal volume The use of polyamine isotopomers as the internal standard enabled the correction of matrix effects in MS detection This method is sensitive, selective and quantitative, and its utility was demonstrated by screening polyamines in 359 salivary samples within 360 min, resulting in discrimination of colorectal cancer patients from noncancer controls © 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) Introduction Recent analysis of the entire set of low-molecular-weight biological compounds (metabolome) has revealed new biomarkers that correlate with disease severity and that respond to therapeutic efficacy or toxicity [1–4] However, few metabolite biomarkers have been implemented in clinical practice because unlike proteins and peptides, it is difficult to produce monoclonal or polyclonal antibodies for low-molecular-weight molecules [5]; therefore, developing rapid screening methods such as the enzyme-linked immunosorbent assay (ELISA) commonly used in clinical laboratories is challenging ∗ Corresponding author at: Institute for Advanced Biosciences, Keio University, 246-2 Mizukami, Kakuganji, Tsuruoka, Yamagata 997-0052 Japan E-mail addresses: sugawara@ttck.keio.ac.jp (K Igarashi), sana.ota@ttck.keio.ac.jp (S Ota), KKK-miku@ttck.keio.ac.jp (M Kaneko), hirayama@ttck.keio.ac.jp (A Hirayama), enomoto@tokyo-med.ac.jp (M Enomoto), k-katsu@tokyo-med.ac.jp (K Katumata), msugi@sfc.keio.ac.jp (M Sugimoto), soga@sfc.keio.ac.jp (T Soga) GC/MS and LC-MS are commonly used for the analysis of lowmolecular-weight compounds However, these approaches require more than 10 per sample due to the low sample throughput associated with solute elution and column preconditioning; thus, the costs associated with these techniques preclude their clinical application Capillary electrophoresis mass spectrometry (CE-MS) is a powerful tool for the comprehensive analysis of charged metabolites [6,7] In this marriage of techniques, CE confers rapid analysis and efficient resolution, and MS provides high selectivity and sensitivity [6]; thus, CE-MS metabolomics has been widely applied in a variety of fields [4,8–10] Recently, Britz-McKibbin et al [11–16] reported the multisegment injection (MSI)-CE-MS method, which allows sequential multisample injection in series within a capillary tube using the sample stacking technique and enables many sample measurements within a single run Polyamines such as spermine, spermidine and their Nacetylated forms are low-molecular-weight cations Because of their positive charges, polyamines bind to DNA and RNA and are involved in a variety of biological processes, including gene expres- https://doi.org/10.1016/j.chroma.2021.462355 0021-9673/© 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 sion and translation, cell proliferation, membrane stabilization, and organ development [17–19] In cancer, polyamine metabolism is frequently dysregulated, indicating that elevated polyamine levels are necessary for transformation and tumor progression [20] Recently, polyamines have attracted much attention not only as targets for anticancer strategies but also as diagnostic tools [19,21,22] Thus, a high-throughput screening method for these biomarkers is urgently needed Here, we developed a high-throughput MSI-CE-tandem mass spectrometry (MS/MS) method that allowed 40 consecutive analyses of salivary polyamines within 40 min, resulting in the discrimination of colorectal cancer (CRC) patients from noncancer controls 2.5 Instrumentation All CE-MS experiments were performed using an Agilent G1600 CE system, an Agilent 6410 triple-quadrupole MS/MS system, an Agilent 6210 time-of-flight mass spectrometry (TOFMS) system, an Agilent 1100 series isocratic HPLC pump, a G1603A Agilent CE-MS adapter kit and a G1607A Agilent CE-ESI-MS sprayer kit (all Agilent Technologies, Santa Clara, CA) The CE-MS adapter kit included a capillary cassette that facilitates thermostating of the capillary, and the CE-ESI-MS sprayer kit, which simplifies coupling the CE system with MS systems, was equipped with an electrospray source LC-MS/MS was performed using an Agilent 1290 Infinity LC system comprised a HiP autosampler, quaternary pump and column compartment, and an Agilent 6460 triple-quadrupole MS/MS system Electrical conductivities were measured by a Yokogawa 73,301 Digital Multi Meter Materials and methods 2.1 Chemicals N1-Acetylspermidine (N1-Ac-Spd), N8-acetylspermidine (N8-AcSpd)-d3 and spermine-d20 were purchased from Toronto Research Chemicals (Toronto, Canada); N1-Ac-Spd-d6, N1,N8-diacetylspermidine (N1,N8-DiAc-Spd)-d6, N1-acetylspermine (N1-Ac-Spm)d3 and N1,N12-diacetylspermine (N1,N12-DiAc-Spm)-d6 were purchased from Santa Cruz Biotechnology (Dallas, TX); creatinined3 was purchased from Taiyo Nippon Sanso (Tokyo, Japan); and spermidine-d3 was purchased from IsoSciences (Ambler, PA) All other reagents were obtained from Sigma-Aldrich (St Louis, MO) or Wako (Osaka, Japan) Water was purified with a Milli-Q purification system (Millipore, Bedford, MA) 2.6 MSI-CE-MS/MS conditions for polyamine analysis Fused-silica capillaries with 50 μm i.d x 110 cm total length were used as the separation capillary The background electrolyte (BGE) for MSI-CE separation was a M formic acid solution Prior to first use, a new capillary was pretreated with BGE for 20 Before MSI injection, the capillary was preconditioned for by flushing with BGE Samples were injected with a pressure injection of 50 mbar, alternating between s (~1 nL) for each sample plug and 20 s (~20 nL) for the BGE spacer plug for a total of 40 discrete samples analyzed within a single run The applied voltage was set at 30 kV, the capillary temperature was thermostated to 20°C, and the sample tray was cooled below 5°C An Agilent 1100 series pump equipped with a 1:100 splitter was used to deliver 10 μL/min of mM ammonium acetate in 50% (v/v) methanolwater to the CE interface, where it was used as a sheath liquid around the outside of the CE capillary to provide a stable electrical connection between the tip of the capillary and the grounded electrospray needle The mass spectrometer was operated in multireaction monitoring (MRM) mode using positive ionization, and a 40 0 V ion spray voltage was applied The flow rate of nebulizer nitrogen gas and drying nitrogen gas (heater temperature 300°C) was maintained at 10 psig and 10 L/min, respectively The Q1 (protonated precursor ion), Q3 (production), fragmentor and collision energy for each polyamine and creatinine are listed in Table 2.2 Clinical samples This study was conducted according to the Declaration of Helsinki principles The study protocol was approved by the Ethics Committee of Tokyo Medical University (No 2346) Written informed consent was obtained from each subject before participating in the study The 359 collected saliva samples, including 57 healthy controls (HCs), 26 patients with benign colorectal tumors (BCTs) and 276 patients with CRC (Table S1), were divided into nine batches, and each batch comprising 39 or 40 saliva filtrates was applied to the MSI-CE-MS/MS system 2.3 Saliva collection Saliva providers were not allowed to take any food except water intake after 9:00 p.m on the previous day They were required to brush their teeth without toothpaste on the day of saliva collection and had to refrain from drinking water, smoking, toothbrushing, and intense exercise 1h before saliva collection Before saliva collection, they were required to gargle with water and then all saliva samples were collected from 8:00 to 11:00 a.m Approximately 400 μL of unstimulated saliva was collected in a 50 mL polypropylene tube using a polypropylene straw 1.1 cm in diameter to assist the saliva collection After collection, saliva samples were immediately stored at -80°C Creatinine normalization was used to correct for differences in salivary rate/hydration status that contributes to greater biological variability 2.7 CE-TOFMS and CE-MS/MS conditions for single-sample analysis Fused-silica capillaries with 50 μm i.d x110 cm total length were used as the separation capillary Samples were injected at 50 mbar for s (~5 nL) For CE-TOFMS, the fragmentor, skimmer, and Oct RFV voltages were set at 75 V, 50 V, and 125 V, respectively A flow rate of drying nitrogen gas was maintained at L/min Methanol-water (50% v/v) containing 0.1 μM hexakis(2,2difluoroethoxy)pho-sphazene was delivered as the sheath liquid at 10 μL/min Automatic recalibration of each acquired spectrum was performed using reference masses of reference standards ([13C isotopic ion of protonated methanol dimer (2CH3OH+H)]+, m/z 66.06306) and protonated hexakis+, m/z 622.02896) Exact mass data were acquired at a rate of 1.5 spectra/s over a 50–10 0 m/z range For CE-MS/MS, methanol-water (50% v/v) was delivered as the sheath liquid at 10 μL/min Others were identical to those used in MSI-CE-MS/MS conditions 2.4 Sample preparation Saliva (90 μL) samples were transferred to 1.5 mL polypropylene reaction tubes (Greiner Bio-One International, Tokyo, Japan), and 10 μL Milli-Q water containing 100 μM creatinine-d3 and 10 μM each polyamine isotopomer was added to the tube The solution was vortexed for 10 s and then centrifugally filtered through a PallNanosep-3-kDa Omega cutoff filter (Pall Corporation, Japan) to remove proteins and other macromolecules at 9100 × g for h at 4°C Then, the filtrate was injected into CE-MS/MS system 2.8 LC-MS/MS conditions for single-sample analysis The analytical conditions were identical to those of LC-triple quadrupole MS described by Tomita et al [22] K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Table Optimized MRM parameters for each polyamine and its isotopomer Compound Q1(m/z) Q3(m/z) Fragmentor (V) Collision Energy (V) Creatinine Creatinine-d3 Spermidine Spermidine-d8 N1-Ac-Spd N1-Ac-Spd-d6 N8-Ac-Spd N8-Ac-Spd-d3 Spermine Spermine-d20 N1,N8-DiAc-Spd N1,N8-DiAc-Spd-d6 N1-Ac-Spm N1-Ac-Spm-d3 N1,N12-DiAc-Spm N1,N12-DiAc-Spm-d6 114 117 146 154 188 194 188 191 203 223 230 236 245 248 287 293 86 89 72 80 100 106 114 117 112 126 100 103 100 103 100 103 100 100 90 90 105 105 110 110 90 90 125 125 120 120 125 125 8 17 17 12 12 16 16 17 17 16 16 20 20 24 24 2.9 Data analysis and electrophoresis occurs (Fig 1c) This phenomenon allows sequential multiple sample injection in series within a capillary, followed by electrophoretic separation and mass detection (Fig 1d) As we reported previously, formic acid showed excellent resolution capability for various cationic species [6,7,23] and was selected as the BGE The effect of its concentration on spermidine resolution among samples was examined, and the highest formic acid concentration (1 M) provided better resolution in 40 consecutive spermidine standard analyses (Fig 2S) Another key was the BGE spacer volume, which significantly affected spermidine resolution among the samples When the BGE spacer volume was less than 10 nL, poor resolution occurred (Fig 2) On the other hand, although better resolution was obtained, spermidine peaks from the 1st to 8th samples in the electrophero-gram were eluted from the separation capillary at 40 nL The use of 20 nL provided good resolution for spermidine peaks in all 40 samples (Fig 2); thus, we selected 20 nL as the BGE spacer volume Using the MSI-CE-MS/MS method, we demonstrated 40 consecutive analyses of a polyamine standard mixture containing spermiddine, N1-Acetylspermidine (N1-Ac-Spd), N8-Acetylspermidine (N8-Ac-Spd), spermine, N1,N8-diacetylspermidine (N1,N8DiAc-Spd), N1-acetylspermine (N1-Ac-Spm) and N1,N12diacetylspermine (N1,N12-DiAc-Spm) Unexpectedly, although their concentrations were identical, the peak area and height of each polyamine fluctuated (Fig in purple) We presumed that this phenomenon might be due to matrix effects, which are caused by the alteration of ionization efficiency of target analytes in the presence of coeluting compounds in MS To confirm this, we analyzed the standard mixture spiked with their isotope-labeled standards and found that the fluctuation patterns between each polyamine and its isotope-labeled standard were closely matched (Fig in grey), which suggested that matrix effects could be normalized by isotope-labeled standards Furthermore, the use of polyamine isotopomers as the internal standard solved another important peak identification problem in CE-MS data where migration time variations between samples are significant [7] In the MSI-CE-MS/MS method, polyamine identification was performed based on its accurate mass (m/z) and comigration with a matching isotope-labeled standard as reported [16] Validation of this MSI-CE-MS/MS method was performed, and the results are listed in Table Reproducibility corrected by isotope-labeled internal standards was practical; the relative peak areas for all polyamines with relative standard deviation (RSD) values (n = 40) were between 3.1 and 9.1% The calibration curves for all polyamines were linear in the range between 0.1 We developed a data analysis tool called MSI-MasterHands that can enable peak detection and integration of MSI-CE-MS/MS data The tool detected the peak top of each peak through the use of Python’s PeakUtils library, and determined the peak boundary by the minimum intensity between each peak top Then the peak area of every peak was calculated by integrating the intensity between the two boundaries Subsequently, the concentration in each sample was calculated by comparison of peak area of analyte with that of its corresponding isotope-labeled polyamine standard To evaluate the difference of the metabolite concentrations among three gropus, Kruskal-Wallis and Dunn’s test was used To evaluate the consistency among the three analytical methods, simple linear regressions were conducted These analyses were conducted using GraphPad (v8.4.3, GraphPad Software, San Diego, CA, USA) The receiver operating characteristic (ROC) curve analysis was conducted using MetaboAnalyst (ver 5.0, https://www.metaboanalyst.ca/) Results and discussion 3.1 Development of a high-throughput MSI-CE-MS/MS method for polyamine analysis Biological samples contain many isomers that are detected at the same m/z value; thus, in the case of the MSI-CE-MS system, these isomers are expected to overlap with others from different samples To confirm the presence of isomers, we analyzed polyamines in a saliva sample with normal single injection CETOFMS and CE-MS/MS (Fig 1S) As expected, several isomers, including N1- and N8-acetylspermidine, were detected at the same m/z value by CE-TOFMS However, CE-MS/MS with multireaction monitoring (MRM) provided sufficient selectivity, resulting in the detection of every polyamine at different m/z values Therefore, we selected CE-MS/MS for further experiments Then, we developed an MSI-CE-MS/MS method to achieve highthroughput screening of polyamine biomarkers One of the keys to success in expanding MSI to large-scale consecutive sample analysis was to amplify the sample stacking effect Theoretically, sample stacking effectively occurs when the sample plug exhibits lower electrical conductivity than the BGE spacer because this condition exhibits greater voltage in the sample zone (Fig 1a,b) Thus, polyamines migrate fast in the sample zone but slowly in the BGE spacer, which results in the concentration of polyamines at the sample and BGE boundary (Fig 1b) Afterwards, when the sample zone and the BGE are mixed, the voltage becomes constant, K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Fig Explanatory diagram of the MSI-CE-MS/MS method for polyamine analysis (a) The samples and BGE are alternately injected into a single capillary (b) Sample stacking of polyamines occurs when the sample plug exhibits lower electrical conductivity than the BGE (c) Electrophoresis starts when the voltage becomes constant by mixing the sample and BGE solutions (d) Polyamines in each sample are consecutively detected by tandem mass spectrometry at their specific m/z values E and the red line indicate the electromotive force and the voltage applied, respectively (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Fig Effect of BGE spacer volume on spermidine resolution among samples The spermidine standard (20 μM each) was injected 40 times consecutively and simultaneously analyzed by MSI-CE-MS/MS and 10 0 μmol/L with correlation coefficients above 0.9898 This method was considerably sensitive, and the concentration detection limits for the polyamines were between 2.9 and 21 nmol/L with a pressure injection of 50 mbar for s (1 nL); i.e., mass detection limits ranged from 2.9 to 21 amol at a signal-to-noise ratio of 3.2 High-throughput analysis of salivary polyamines from colorectal cancer patients Recently, salivary and urinary polyamines have been reported as promising biomarkers in various cancers, such as oral, breast, colorectal and pancreatic cancers [19,24–26] In many types of can4 K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Fig Fluctuation of peak heights and areas in 40 consecutive analyses of polyamine standards (1 μM each, purple) and their deuterium-labeled isotopomers (1 μM each, grey) with MSI-CE-MS/MS Abbreviations: DiAc, diacetyl; Spm, spermine; Ac, acetyl; and Spd, spermidine (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Table Reproducibility, linearity, and sensitivity for polyamine standard analysis Compound RSDa (n = 40) (%) Linearityb Correlation Detection Limit (nmol/L) Spermidine N8-Ac-Spd N1-Ac-Spd Spermine N1,N8-DiAc-Spd N1-Ac-Spm N1,N12-DiAc-Spm 5.0 9.1 6.4 8.4 3.1 7.6 4.5 0.9898 0.9946 0.9912 0.9926 0.9913 0.9971 0.9913 18 21 14 4.8 2.9 4.5 4.6 a RSD was calculated by the relative peak area, in which the peak area was divided by the peak area of its isotopomer b Calibration curves for all compounds were plotted at 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, 500 and 1000 μmol/L cer, the MYC oncogene is amplified or overexpressed due to several factors [9,27] MYC drives the transcription of the gene encoding ornithine decarboxylase (ODC), the rate-limiting enzyme in poly-amine biosynthesis, resulting in elevation of polyamine levels [20,24] To explore the possibilities of MSI-CE-MS/MS for highthroughput salivary polyamine analysis, we measured the electrical conductivities for the salivary samples and the BGE The conductivities of the saliva filtrates and the BGE (1 M formic acid) were 2.3 and 20.8 × 10−4 /Ωm, respectively, which was expected to allow effective sample stacking Then, we applied our MSI-CEMS/MS method for 40 consecutive analyses of a saliva sample obtained from patients with CRC Forty well-defined peaks of each polyamine were observed (Fig S3), and electric current was stable and constant during all the run (Fig S4), the amount of polyamines quantified using their isotopomers as internal standards and their %RSD values are listed in Table Overall, acceptable reproducibilities (4.4% - 14%, n = 40) were obtained except for N8-Ac-Spd, the less abundant polyamine (Fig S3) To investigate the quantification accuracy of this system, we analyzed saliva samples spiked with 10 nmol of each polyamine standard and calculated the recovery (Table 3) The recovery rates for spermine (61%) and spemidine (69%) were lower than others, which may be caused by partial peak overlap in the fast migrating peaks (Fig 3) Those for other polyamines were ranged from 73 to 89% These results indicate that most of the polyamines can be approximately precisely quantified by the proposed MSI-CE-MS/MS method Using the MSI-CE-MS/MS system, we performed polyamine analysis of a total of 359 saliva samples from 276 patients with colorectal cancer (CRC), 26 patients with benign colorectal tumors (BCTs) and 57 healthy controls (HCs) (Table S1) The 359 saliva K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Table Polyamine amount, reproducibility, and recovery in CRC patient saliva analysis Compound Amounta (nmol/L) RSD (n = 40) (%) Recovery Rate (%) Spermidine N8-Ac-Spd N1-Ac-Spd Spermine N1,N8-DiAc-Spd N1-Ac-Spm N1,N12–DiAc-Spm 3350 45 1480 170 127 44 387 4.4 45 7.6 5.8 7.2 14 6.7 69 89 82 61 88 73 86 a The amount of polyamines was quantified using their isotopomers as internal standards Fig MSI-CE-MS/MS analysis of polyamines in salivary samples obtained from HCs (n = 20, blue) and CRC patients (n = 20, magenta) For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article Table Median concentration (nmol/L) of polyamines in salivary samples from HCs (n = 57), BCTs (n = 26) and CRCs (n = 276) samples were divided into nine batches, and each batch containing 39 or 40 samples was successively determined within 360 For saliva analysis, normalization is an important issue because differences in salivary rate/hydration status contributes to greater biological variability To overcome this defect, we performed metabolome analysis of the 359 saliva samples with a single CE-TOFMS [7] and confirmed the positive correlation between overall metabolite concentrations and creatinine concentrations (Fig S5) Consequently, we used creatinine to correct for differences in salivary rate/hydration status in this study Fig shows representative MSI-CE-MS/MS electropherograms of salivary polyamines These seven compounds in 20 HCs were either slightly detected or not detected, whereas they were markedly increased in most CRC samples Overall, the polyamine levels in most of the HCs were low, whereas those in the patients with CRC or BCTs were significantly higher Among polyamines, N1-Ac-Spd, N1-Ac-Spm and N1,N12-DiAc-Spm showed a high ability to dis- compound HC BCT CRC N1-Ac-Spd N1-Ac-Spm N1,N12-DiAc-Spm 167 25.4 86.8 196 27.6 163 599 87.3 315 criminate CRC patients from non-CRC controls (Fig and Table 4) and the lower limits of quantification (at a signal-to-noise ratio of 10) for N1-Ac-Spd, N1-Ac-Spm and N1,N12-DiAc-Spm were 330, 12 and 79 nmol/L in saliva, respectively After normalization of salivary rate/hydration status by creatinine level, these results were compared with those obtained by a single-sample analysis method with CE-TOFMS and LC-MS/MS The Pearson correlation coefficients K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Fig MSI-CE-MS/MS analysis of N1-acetylspermidine, N1-acetyl-spermine and N1, N12-diacetylspermine in salivary samples from HCs (n = 57), BCTs (n = 26) and CRC (n = 276) To visualize individual polyamine levels, box and whiskers plots were used The horizontal bars represent the medians, quartiles, and 10% of both ends Outside data are depicted in plots The Kruskal-Wallis test and Dunn’s test as post-test were used to determine statistical significance ∗ ∗ ∗ p < 0.001, ∗ ∗ p < 0.01 and ∗ p < 0.05 Fig Linear regression of of polyamine levels in 359 salivary samples (A) between MSI-CE-MS/MS and normal single CE-TOFMS method, and (B) between MSI-CE-MS/MS and normal single LC-MS/MS method Both x and y-axis indicate log 10 of the polyamine concentration/creatinine concentration (no limit) The regressed lines with their coefficients and intercepts are shown Correlation coefficients (r) and p values are calculated by Pearson correlation Not detected peaks were excluded from this analysis demonstrated statistically significant relationships among the three methods (Fig 6), which indicates that the proposed MSI-CE-MS/MS method provides almost the same quantification accuracy as CETOFMS and LC-MS/MS The area under the receiver operating characteristic (ROC) curve (AUC) is used to assess the discrimination ability of biomarkers Commonly used serum markers for the diagnosis of CRC are neuron –specific enolase (NSE), carcinoembryonic antigen (CEA), cancer antigen (CA)19-9, CA125 and CA242 The AUC of NSE, CEA, CA19-9, CA 125 and CA242 are 0.766, 0.682, 0.560, 0.590 and 0.651, respectively [28] Among the salivary polyamines, N1-AcSpm showed the highest AUC of 0.834 (Fig 7), allowing for dis7 K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 Fig ROC curve analysis of the ability of N1-acetylspermidine, N1-acetylspermine, N1,N12-diacetylspermine and combined three polyamines to discriminate patients with CRC from noncancer controls ROC curves are black curves and 95% confidential intervals are shown in light purple (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) crimination of CRC from noncancer samples, i.e., samples from the patients with BCTs and the HCs, and the optimal cut off value in salivary samples determined by ROC analysis was 39.5 nmol/L Although our previous study reported that urinary N1,N12-DiAcSpm was the most useful biomarker for the discrimination of CRC [22], regarding salivary samples, N1-Ac-Spm showed the best discrimination ability between CRC patients and non-CRC controls Previous papers reported that MSI-CE-MS has been extensively validated to provide precise and accuracy metabolite measurements as compared to other analytical platforms, including colorimetric assays [29], GC-MS [13] and LC-UV [30] Our salivary polyamine analysis demonstrated that MSI-CE-MS/MS achieved the same quantification results with conventional CETOFMS and HPLC-MS/MS methods Taken together, this approach can be promising tools for the high-throughput screening of low molecular-weight species On the other hand, there is a limitation in this study As shown in Table S1, the number of healthy controls and patient groups including age, sex, was significantly different Moreover, salivary N1Ac-Spm might be elevated in other types of cancer Therefore, further studies should be necessary to transfer this approach to clinical applications Conclusions We present a high-throughput, selective and sensitive MSICE-MS/MS method for screening salivary polyamines Compared with other techniques, this method has several advantages: (1) the method is rapid; 40 salivary samples can be analyzed within 40 min; (2) polyamines are selectively determined without other matrix interference; (3) sensitivity is considerably high; and (4) sample preparation is minimal The methodology provides practical reproducibility, excellent linearity and quantification accuracy Its utility was demonstrated by analyzing polyamines in 359 salivary samples obtained from patients with CRC or benign polyps and HCs, and a couple polyamines (e.g., N1-acetylspermine) can potentially discriminate cancer patients from noncancer controls Therefore, it is expected that the proposed method can potentially be applied to screening CRC in laboratory tests, and this approach could lead to the practical use of many types of low-molecularweight biomarkers in a wide range of applications CRedit authorship contibution statement T.S led the entire project and wrote the manuscript K.I and A.H performed the MSI-CE-MS/MS and all CE-MS experiments S.O K Igarashi, S Ota, M Kaneko et al Journal of Chromatography A 1652 (2021) 462355 and M.K performed the LC-MS/MS experiments M.E., K.K and M.S collected and provided the human specimens T.S., K.I and M.S generated the figures [11] N.L Kuehnbaum, A Kormendi, P Britz-McKibbin, Multisegment injectioncapillary electrophoresis-mass spectrometry: a high-throughput platform for metabolomics with high data fidelity, Anal Chem 85 (22) (2013) 10664–10669 https://doi.org/, doi:10.1021/ac403171u [12] A DiBattista, N McIntosh, M Lamoureux, O.Y Al-Dirbashi, P Chakraborty, P Britz-McKibbin, Temporal signal pattern recognition in mass spectrometry: a method for rapid identification and accurate quantification of biomarkers for inborn errors of metabolism with quality assurance, Anal Chem 89 (15) (2017) 8112–8121 https://doi.org/, doi:10.1021/acs.analchem.7b01727 [13] S Azab, R Ly, P Britz-McKibbin, Robust method for high-throughput screening of fatty acids by multisegment injection-nonaqueous capillary electrophoresismass spectrometry with stringent quality control, Anal Chem 91 (3) (2019) 2329–2336 https://doi.org/, doi:10.1021/acs.analchem.8b05054 [14] R.J de Souza, M Shanmuganathan, A Lamri, S.A Atkinson, A Becker, D Desai, M Gupta, P.J Mandhane, T.J Moraes, K.M Morrison, P Subbarao, K.K Teo, S.E Turvey, N.C Williams, P Britz-McKibbin, S.S Anand, Maternal diet and the serum metabolome in pregnancy: robust dietary biomarkers generalizable to a multiethnic birth cohort, Curr Dev Nutr (10) (2020) https://doi.org/, doi:10.1093/cdn/nzaa144 [15] M Saoi, A Li, C McGlory, T Stokes, M.T von Allmen, S.M Phillips, P BritzMcKibbin, Metabolic perturbations from step reduction in older persons at risk for sarcopenia: plasma biomarkers of abrupt changes in physical activity, Metabolites (7) (2019) https://doi.org/, doi:10.3390/metabo9070134 [16] A DiBattista, D Rampersaud, H Lee, M Kim, P Britz-McKibbin, 3High throughput screening method for systematic surveillance of drugs of abuse by multisegment injection-capillary electrophoresis-mass spectrometry, Anal Chem 89 (21) (2017) 11853–11861 https://doi.org/, doi:10.1021/acs.analchem 7b03590 [17] A Gugliucci, Polyamines as clinical laboratory tools, Clin Chim Acta 344 (1-2) (2004) 23–35 https://doi.org/, doi:10.1016/j.cccn.2004.02.022 [18] T Thomas, T.J Thomas, Polyamines in cell growth and cell death: molecular mechanisms and therapeutic applications, Cell Mol Life Sci 58 (2) (2001) 244–258 https://doi.org/, doi:10.10 07/Pl0 0 0852 [19] Y Umemori, Y Ohe, K Kuribayashi, N Tsuji, T Nishidate, H Kameshima, K Hirata, N Watanabe, Evaluating the utility of N-1,N-12-diacetylspermine and N1,N-8-diacetylspermidine in urine as tumor markers for breast and colorectal cancers, Clin Chim Acta 411 (23-24) (2010) 1894–1899 https://doi.org/, doi:10.1016/j.cca.2010.07.018 [20] R.A Casero, T.M Stewart, A.E Pegg, Polyamine metabolism and cancer: treatments, challenges and opportunities, Nat Rev Cancer 18 (11) (2018) 681–695 https://doi.org/, doi:10.1038/s41568- 018- 0050- [21] M Sugimoto, D.T Wong, A Hirayama, T Soga, M Tomita, Capillary electrophoresis mass spectrometry-based saliva metabolomics identified oral, breast and pancreatic cancer-specific profiles, Metabolomics (1) (2010) 78– 95 https://doi.org/, doi:10.10 07/s11306-0 09-0178-y [22] A Tomita, M Mori, K Hiwatari, E Yamaguchi, T Itoi, M Sunamura, T Soga, M Tomita, M Sugimoto, Effect of storage conditions on salivary polyamines quantified via liquid chromatography-mass spectrometry, Sci Rep (2018) Ukhttps://doi.org/, doi:10.1038/s41598- 018- 30482- x [23] T Soga, D.N Heiger, Amino acid analysis by capillary electrophoresis electrospray ionization mass spectrometry, Anal Chem 72 (6) (20 0) 1236–1241 https://doi.org/, doi:10.1021/ac990976y [24] Y Asai, T Itoi, M Sugimoto, A Sofuni, T Tsuchiya, R Tanaka, R Tonozuka, M Honjo, S Mukai, M Fujita, K Yamamoto, Y Matsunami, T Kurosawa, Y Nagakawa, M Kaneko, S Ota, S Kawachi, M Shimazu, T Soga, M Tomita, M Sunamura, Elevated polyamines in saliva of pancreatic cancer, Cancers 10 (2) (2018) https://doi.org/, doi:10.3390/cancers10 020 043 [25] X.W Song, X.H Yang, R Narayanan, V Shankar, S Ethiraj, X Wang, N Duan, Y.H Ni, Q.G Hu, R.N Zare, Oral squamous cell carcinoma diagnosed from saliva metabolic profiling, Proc Natl Acad Sci USA 117 (28) (2020) 16167–16173 https://doi.org/, doi:10.1073/pnas.2001395117 [26] T Nakajima, K Katsumata, H Kuwabara, R Soya, M Enomoto, T Ishizaki, A Tsuchida, M Mori, K Hiwatari, T Soga, M Tomita, M Sugimoto, Urinary polyamine biomarker panels with machine-learning differentiated colorectal cancers, benign disease, and healthy controls, Int J Mol Sci 19 (3) (2018) https://doi.org/, doi:10.3390/ijms19030756 [27] C.V Dang, MYC on the path to cancer, Cell 149 (1) (2012) 22–35 https://doi.org/, doi:10.1016/j.cell.2012.03.003 [28] H Luo, K.X Shen, B Li, R.Q Li, Z.M Wang, Z.S Xie, Clinical significance and diagnostic value of serum NSE, CEA, CA19-9, CA125 and CA242 levels in colorectal cancer, Oncol Lett 20 (1) (2020) 742–750 https://doi.org/, doi:10.3892/ ol.2020.11633 [29] S.M Azab, A Zamzam, M.H Syed, R Abdin, M Qadura, P Britz-McKibbin, Serum metabolic signatures of chronic limb-threatening ischemia in patients with peripheral artery disease, J Clin Med (6) (2020) https://doi.org/, doi:10.3390/jcm9061877 [30] J Wild, M Shanmuganathan, M Hayashi, M Potter, P Britz-McKibbin, Metabolomics for improved treatment monitoring of phenylketonuria: urinary biomarkers for non-invasive assessment of dietary adherence and nutritional deficiencies, Analyst 144 (22) (2019) 6595–6608 https://doi.org/, doi:10.1039/ c9an01642b Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper Acknowledgements We thank Toru Akiba, Ayame Enomoto and Kumi Suzuki (Institute for Advanced Biosciences, Keio University) for technical assistance This work was supported by research funds from the Yamagata prefectural government and the city of Tsuruoka Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2021.462355 References [1] A Sreekumar, L.M Poisson, T.M Rajendiran, A.P Khan, Q Cao, J.D Yu, B Laxman, R Mehra, R.J Lonigro, Y Li, M.K Nyati, A Ahsan, S Kalyana-Sundaram, B Han, X.H Cao, J Byun, G.S Omenn, D Ghosh, S Pennathur, D.C Alexander, A Berger, J.R Shuster, J.T Wei, S Varambally, C Beecher, A.M Chinnaiyan, Metabolomic profiles delineate potential role for sarcosine in prostate cancer progression, Nature 457 (7231) (2009) 910–914 https://doi.org/, doi:10.1038/ nature07762 [2] T Soga, M Sugimoto, M Honma, M Mori, K Igarashi, K Kashikura, S Ikeda, A Hirayama, T Yamamoto, H Yoshida, M Otsuka, S Tsuji, Y Yatomi, T Sakuragawa, H Watanabe, K Nihei, T Saito, S Kawata, H Suzuki, M Tomita, M Suematsu, Serum metabolomics reveals gamma-glutamyl dipeptides as biomarkers for discrimination among different forms of liver disease, J Hepatol 55 (4) (2011) 896–905 https://doi.org/, doi:10.1016/j.jhep.2011.01.031 [3] E.A Mathe, A.D Patterson, M Haznadar, S.K Manna, K.W Krausz, E.D Bowman, P.G Shields, J.R Idle, P.B Smith, K Anami, D.G Kazandjian, E Hatzakis, F.J Gonzalez, C.C Harris, Noninvasive urinary metabolomic profiling identifies diagnostic and prognostic markers in lung cancer, Cancer Res 74 (12) (2014) 3259–3270 https://doi.org/, doi:10.1158/0 08- 5472.Can- 14- 0109 [4] Y Minami, T Kasukawa, Y Kakazu, M Iigo, M Sugimoto, S Ikeda, A Yasui, G.T.J van der Horst, T Soga, H.R Ueda, Measurement of internal body time by blood metabolomics, Proc Natl Acad Sci USA 106 (24) (2009) 9890–9895 https://doi.org/, doi:10.1073/pnas.0900617106 [5] Z.X Xu, H.J Gao, L.M Zhang, X.Q Chen, X.G Qiao, The biomimetic immunoassay based on molecularly imprinted polymer: a comprehensive review of recent progress and future prospects, J Food Sci 76 (2) (2011) R69–R75 https://doi.org/, doi:10.1111/j.1750-3841.2010.02020.x [6] T Soga, Y Ohashi, Y Ueno, H Naraoka, M Tomita, T Nishioka, Quantitative metabolome analysis using capillary electrophoresis mass spectrometry, J Proteom Res (5) (2003) 488–494 https://doi.org/, doi:10.1021/pr034020m [7] T Soga, R Baran, M Suematsu, Y Ueno, S Ikeda, T Sakurakawa, Y Kakazu, T Ishikawa, M Robert, T Nishioka, M Tomita, Differential metabolomics reveals ophthalmic acid as an oxidative stress biomarker indicating hepatic glutathione consumption, J Biol Chem 281 (24) (2006) 16768–16776 https://doi.org/, doi:10.1074/jbc.M601876200 [8] N Ishii, K Nakahigashi, T Baba, M Robert, T Soga, A Kanai, T Hirasawa, M Naba, K Hirai, A Hoque, P.Y Ho, Y Kakazu, K Sugawara, S Igarashi, S Harada, T Masuda, N Sugiyama, T Togashi, M Hasegawa, Y Takai, K Yugi, K Arakawa, N Iwata, Y Toya, Y Nakayama, T Nishioka, K Shimizu, H Mori, M Tomita, Multiple high-throughput analyses monitor the response of E-coli to perturbations, Science 316 (5824) (2007) 593–597 https://doi.org/, doi:10 1126/science.1132067 [9] K Satoh, S Yachida, M Sugimoto, M Oshima, T Nakagawa, S Akamoto, S Tabata, K Saitoh, K Kato, S Sato, K Igarashi, Y Aizawa, R Kajino-Sakamoto, Y Kojima, T Fujishita, A Enomoto, A Hirayama, T Ishikawa, M.M Taketo, Y Kushida, R Haba, K Okano, M Tomita, Y Suzuki, S Fukuda, M Aoki, T Soga, Global metabolic reprogramming of colorectal cancer occurs at adenoma stage and is induced by MYC, Proc Natl Acad Sci USA 114 (37) (2017) E7697–E7706 https://doi.org/, doi:10.1073/pnas.1710366114 [10] T Yoneshiro, Q Wang, K Tajima, M Matsushita, H Maki, K Igarashi, Z.P Dai, P.J White, R.W McGarrah, O.R Ilkayeva, Y Deleye, Y Oguri, M Kuroda, K Ikeda, H.X Li, A Ueno, M Ohishi, T Ishikawa, K Kim, Y Chen, C.H Sponton, R.N Pradhan, H Majd, V.J Greiner, M Yoneshiro, Z Brown, M Chondronikola, H Takahashi, T Goto, T Kawada, L Sidossis, F.C Szoka, M.T McManus, M Saito, T Soga, S Kajimura, BCAA catabolism in brown fat controls energy homeostasis through SLC25A44, Nature 572 (7771) (2019) 614 https://doi.org/, doi:10.1038/s41586- 019- 1503- x ... pressure injection of 50 mbar for s (1 nL); i.e., mass detection limits ranged from 2.9 to 21 amol at a signal-to-noise ratio of 3.2 High-throughput analysis of salivary polyamines from colorectal cancer. .. Azab, R Ly, P Britz-McKibbin, Robust method for high-throughput screening of fatty acids by multisegment injection- nonaqueous capillary electrophoresismass spectrometry with stringent quality control,... P Britz-McKibbin, 3High throughput screening method for systematic surveillance of drugs of abuse by multisegment injection- capillary electrophoresis- mass spectrometry, Anal Chem 89 (21) (2017)