A method for analyzing odd-chain fatty acids by reversed-phase chromatography-mass spectrometry

Abstract

The application discloses an odd carbon chain lipid analysis method combined with reversed phase chromatography-mass spectrometry, the method utilizes reversed phase liquid chromatography-mass spectrometry (RP-LC-MS) technology, is based on equivalent carbon number (ECN) retention time (RT) prediction model, is combined with lipidomics internal standard and standard product, constructs odd carbon chain lipid library through retention time and internal standard or standard product mass spectrum cracking behavior, and effectively identifies the lipid isomerism phenomenon existing in a large amount in odd carbon chain lipid, and it is proved that the method has good reproducibility and stability through methodological characterization, half-height peak width, coefficient of variation (CV) is within 20%, and there is no significant difference in peak area after 10 injections of internal standard in succession, so that high-throughput analysis of odd carbon chain lipid can be realized.

Description

Technical Field

[0001] This invention belongs to the field of odd-chain lipid analysis technology, specifically relating to a reversed-phase chromatography-mass spectrometry method for the analysis of odd-chain lipids. Background Technology

[0002] Odd-chain fatty acids (OCFAs) are widely present in natural lipids, but their content is low under normal conditions, accounting for approximately 1-5% of the fatty acid content in higher animals. In the past, due to limitations in research methods and detection techniques, most studies on lipid metabolism focused on even-chain fatty acids, while the precise detection and functional studies of OCFAs were largely neglected due to their low abundance. Although epidemiological studies have provided preliminary results on the association between OCFAs and disease risk, this contradicts the role of even-chain fatty acids. However, elucidating their potential functions requires precise detection, which depends on accurately quantifying the dynamic changes of OCFAs in disease development and their distribution in the formation of complex lipids. Furthermore, with the rise and development of lipidomics analysis technology, numerous studies have used this technology to discover the presence of C15:0, C17:0, C17:1, C19:1, and C23:0 in the human body, making it possible to quantitatively characterize lipids. However, the role of OCFAs as basic structural components in the synthesis of complex lipids (such as glycerides, phospholipids, sphingolipids, ceramides, and sterol esters) and their distribution remain unclear. Therefore, it is urgent to develop lipidomics analysis methods for odd-numbered carbon chain lipids for the quantitative analysis of this type of lipid in biological samples, thereby exploring the dynamic changes of OCFAs in diseases and mapping the distribution of OCFAs in the formation of complex lipids. Summary of the Invention

[0003] Therefore, the present invention aims to provide a reversed-phase chromatography-mass spectrometry (RP-MS) method for the analysis of odd-numbered carbon chain lipids. This method has good reproducibility and stability, a wide half-peak width, a coefficient of variation (CV) of less than 20%, and no significant difference in peak area after 10 consecutive injections of internal standard, enabling high-throughput analysis of odd-numbered carbon chain lipids.

[0004] To achieve the above objectives, the present invention provides a method for the analysis of odd-numbered carbon chain lipids using reversed-phase chromatography-mass spectrometry, comprising the following steps:

[0005] (1) Construction of an odd-numbered carbon chain lipid library: The lipids in the lipid library were downloaded from the Lipid Metabolism Pathway Research Project (LIPIDMAPS, https: / / www.lipidmaps.org / databases / lmsd / browse) structural database and computational simulation. The lipid library screening criteria were cholesterol ester CE, triglyceride TAG, diglyceride DG, monoglyceride MAG, sphingolipid SM, ceramide Cer, free fatty acid FFA, phosphatidic acid PA, phosphatidylcholine PC, phosphatidylethanolamine PE, phosphatidylglycerol PG, phosphatidylinositol PI, and phosphatidylserine PS;

[0006] (2) Construction of mass spectrometry multi-level reaction monitoring (MRM) ion pairs for lipids with odd carbon chains: MRM consists of a parent ion Q1 and a daughter ion Q3. In positive ion mode, MRM ion pairs for CE, glycerides TAG, DG, MAG, CER, DCER, HCER, LCER, and SM are constructed; in negative ion detection mode, MRM ion pairs for phospholipids containing PA, PC, PE, PG, PS, PI, and FFA are constructed.

[0007] (3) Lipid extraction: The internal standard Ultimate SPLASH is added during lipid extraction. TM ONE, after lipid extraction, the extract was dried with nitrogen flow and added to working solution to prepare the injection concentration. The working solution contained 10 mM ammonium acetate in DCM:MeOH, 1:1, v / v.

[0008] (4) Liquid chromatography-mass spectrometry separation:

[0009] Mass spectrometry conditions: The electrospray ionization source adopted a positive and negative ion switching scanning mode. In positive ion mode, the spray voltage was 5200V, the ion source temperature was 350℃, the nebulizer gas was 40psi, the curtain gas was 55psi, and the desolvator gas was 55psi. In negative ion mode, the spray voltage was -4500V, the ion source temperature was 600℃, the nebulizer gas was 35psi, the curtain gas was 60psi, and the desolvator gas was 60psi.

[0010] Chromatographic conditions: A BEH C18 column (2.1 mm × 50 mm, 1.8 μm, Waters) was used. Mobile phase A was methanol:acetonitrile:water 1:1:1, v / v / v, +5 mM ammonium acetate; mobile phase B was isopropanol +5 mM ammonium acetate. The elution gradient was: 0 min 20% B, 1 min 20% B, 2.5 min 40% B, 4 min 60% B, 14 min 90% B, 15 min 90% B, 15.1 min 20% B, 17 min 20% B. The flow rate was 0.2 mL / min, and the column temperature was set to 40 °C.

[0011] (5) Liquid phase separation and mass spectrometry determination of internal standard:

[0012] The lipidomics internal standard is Ultimate Splash. TM ONE (Cat#330820L, Avanti), the MRM ion pair of the internal standard was constructed according to the rules described in (2); the internal standard was separated by liquid chromatography-mass spectrometry using the separation conditions described in (4);

[0013] (6) Prediction of lipid retention time in odd-numbered carbon chain lipid libraries:

[0014] (6.1) Internal standard retention time acquisition: According to the separation conditions described in (4) and the internal standard mass spectrometry fragmentation law described in (5), the RT of the internal standard is obtained in the LC-MS system;

[0015] (6.2) Prediction of retention time of various lipids:

[0016] According to the ECN model, in reversed-phase liquid chromatography, the lipid retention time and the number of unsaturated double bonds or the carbon chain length conform to a binomial fitting distribution. Based on this theory, a binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x is constructed according to the retention time and carbon chain length in (6.1), where y is the lipid retention time / total elution time of 17 min, and the relative carbon chain length is the number of carbon atoms of the same type of lipid / the maximum number of carbon atoms. The lipid retention time of the odd-numbered carbon chain lipid library is predicted according to the constructed binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x.

[0017] (7) Based on the predicted lipid retention time of the odd-numbered carbon chain lipid library and the constructed MRM ion pairs, the odd-numbered carbon chain lipids in the test sample were confirmed and quantified.

[0018] Furthermore, the odd-numbered carbon chain lipid library constructed in (1) contains odd-numbered carbon chain lipids as shown in No.1 to No.678.

[0019] Furthermore, in (2), the Q1 construction of odd-carbon chain lipids under the positive ion mode of mass spectrometry multi-stage reaction monitoring (MRM) ion pairs follows the following characteristics: CE, MAG, SM, CER, DCER, HCER, and LCER are [M+H]. + DG and TAG are [M+NH4] + The construction of Q3 follows these characteristics: The Q3 mass-to-charge ratio m / z for CE is 369, for SM it is 184, for CER it is 264, for DCER it is 266, for HCER it is 264, and for LCER it is 264. For MAG, based on the neutral loss fragment NLS being a propylene glycol residue with a mass number of 74, the m / z of Q3 is determined to be Q1-74. For DG, the m / z of Q3 is the mass number of the odd-numbered carbon chain acyl group + the 74Da of the head group propylene glycol residue, i.e., the mass number of the odd-numbered carbon chain acyl group + 74. Furthermore, for every 1 increase in unsaturation, the m / z of Q3 decreases by 2. For TAG, the m / z of Q3 is Q1-NLS, where NLS is the mass number of the odd-numbered carbon chain acyl group + 34. The construction of Q1 in negative ion mode follows these characteristics: The Q1 mass-to-charge ratio m / z for PC is [M+acetyl AcO]. - The Q1 m / z for the remaining PA, PE, PG, PI, and PS is [MH]. - Based on the neutral loss fragment (NLS) characteristics of various lipids, the m / z of Q3 is Q1-NLS, where PA's NLS is 153+acyl group mass number, PC's NLS is 298+acyl group mass number, PE's NLS is 196+acyl group mass number, PG's NLS is 227+acyl group mass number, PI's NLS is 315+acyl group mass number, PS's NLS is 240+acyl group mass number, and FFA's Q1 = Q3, with m / z of [MH]. - .

[0020] Furthermore, in (3), the lipid extraction method includes the following:

[0021] (3.1) Lipid extraction method for tissue samples:

[0022] (3.1.1) Tissue sample preparation:

[0023] (3.1.1.1) Dry the tissue with filter paper, weigh 50.0 mg of tissue sample and put it into a 2 mL round-bottom centrifuge tube;

[0024] (3.1.1.2) Add 500 μL of physiological saline (0.85% physiological saline preparation: add 0.85 g NaCl to 100 mL of deionized water, then add 40 μL of glacial acetic acid);

[0025] (3.1.1.3) Grinding: Place 2 large magnetic beads (small magnetic beads are only used to homogenize liver, kidney, brain and mucosal tissues, generally two small magnetic beads are placed);

[0026] (3.1.1.4) Place the sample tube into the tissue homogenizer and homogenize twice, 30 seconds each time, with a 10-second interval, to avoid temperature rise;

[0027] (3.1.1.5) Centrifuge the homogenate at 12,000 rpm for 15 minutes and use it for further lipid extraction;

[0028] (3.1.2) Lipid extraction from tissue samples:

[0029] (3.1.2.1) Mix 50 μL of tissue homogenate supernatant with 5 μL of Ultimate SPLASH. TM Add the ONE internal standard to a glass tube, then add 950 μL H2O, 2 mL MeOH, and 0.9 mL DCM; gently shake the tube for 5 seconds. No separation should occur at this point. If separation does occur, add an additional 50 μL MeOH.

[0030] (3.1.2.2) Then place the tube at room temperature for 30 minutes, and then add 1 mL H2O and 0.9 mL DCM in sequence; gently shake the tube for 5 seconds, and then centrifuge at 2000 rpm for 15 minutes until the liquid separates into layers;

[0031] (3.1.2.3) Transfer the layered bottom liquid to a new glass tube using a pipette, and add 1.8 mL of DCM to the original tube for a second extraction; after the above steps, combine the second extract with the first phase; then dry the extract using a stable nitrogen stream.

[0032] (3.1.2.4) Then add 100 μL of working solution to the tube to dissolve it; then shake the sample on a benchtop shaker to mix it thoroughly, and transfer it to a 200 μL LC / MS insert tube; place the insert tube into a 1.5 ml plastic EP tube and centrifuge at 16000 rpm for 10 minutes. After centrifugation, transfer the insert tube to an LC-MS vial;

[0033] (3.1.2.5) Take 10 μL of each sample and put it into a QC tube, then use it as a control and mix thoroughly; finally, store the lipid extract in the LC-MS vial in a -80℃ freezer until LC-MS / MS analysis;

[0034] (3.2) Lipid extraction method from plasma samples:

[0035] (3.2.1) Plasma sample pretreatment:

[0036] (3.2.1.1) Use EDTA for anticoagulation and collect blood in a 5 mL purple-tipped tube;

[0037] (3.2.1.2) Centrifuge at 1500g for 5 min (if it cannot be processed immediately, place it at 4℃ and process it within 12 h);

[0038] (3.2.1.3) Take 2 mL of supernatant, aliquot it into 1.5 mL EP tubes, and store at -80℃; the sample volume used each time is 25 μL;

[0039] (3.2.2) Plasma lipid extraction:

[0040] (3.2.2.1) Reconstitute the serum or plasma to room temperature; add 25 μL of serum or cell supernatant to a glass tube;

[0041] (3.2.2.2) Add 975 μL H2O, 2 mL methanol (MeOH), and 0.9 mL dichloromethane (DCM);

[0042] (3.2.2.3) Gently shake for 5 seconds. If no layering occurs, add 50 μL of MeOH. Add an appropriate amount of internal standard and mix well.

[0043] (3.2.2.4) Let stand at room temperature for 30 min, then add 1 mL H2O and 0.9 mL DCM;

[0044] (3.2.2.5) Gently shake for 5 seconds; centrifuge at 2000 rpm for 15 minutes until the liquid separates into layers, then transfer the lower layer to a new glass centrifuge tube;

[0045] (3.2.2.6) Add 1.8 mL of DCM to the original tube and repeat the extraction once; gently shake for 5 seconds, centrifuge at 2000 rpm for 10 minutes until the liquid separates into layers;

[0046] (3.2.2.7) Dry the organic phase in the combined lower DCM with nitrogen, and then add 100 μL of working solution to dissolve it in a plastic EP tube;

[0047] (3.2.2.8) Shake the sample on a benchtop shaker to mix it well, then centrifuge at 16000 rpm for 10 min, and aspirate 90 μL of the sample to transfer it to a small glass sample tube.

[0048] (3.2.2.9) Take 20 μL from each of all samples and mix them into a QC tube as a control.

[0049] Further, the MRM (Q1>Q3) ion pairs of the internal standard in (5) are constructed as follows: 17:0-14:1DG-d5+NH4-IS (575.5>332.2), 17:0-16:1DG-d5+NH4-IS (603.5>332.2), 17:0-18:1DG-d5+NH4-IS (631.6>332.2), 17:0-20:3D G-d5+NH4-IS(655.6>332.2), 17:0-22:4DG-d5+NH4-IS(681.6>332.2), 14:0-13:0-14:0 TAG-d5+NH4-IS(731.7>486.4), 14:0-15:1-14:0TAG-d5+NH4-IS(757.7>512.4), 14:0-17 :1-14:0TAG-d5+NH4-IS(785.7>540.5), 16:0-15:1-16:0TAG-d5+NH4-IS(813.7>540.5) , 16:0-17:1-16:0TAG-d5+NH4-IS(841.8>568.5), 16:0-19:2-16:0TAG-d5+NH4-IS(867.8 >594.5), 18:1-17:1-18:1TAG-d5+NH4-IS(893.8>594.5), 18:1-19:2-18:1TAG-d5+NH4- IS(919.8>620.5), 18:1-21:2-18:1TAG-d5+NH4-IS(947.8>648.6), 14:1cholesteryl-d7 ester-IS(619.6>376.4), 16:1cholesteryl-d7 ester-IS(647.6>376.4), 18:1cholesteryl-d7 ester-IS(675.6>376.4), 20:3cholesteryl-d7 ester-IS(699.6>376.4), 22:4cholesteryl-d7 ester-IS(725.7>376.4), C16:1Ceramide-d7(d18:1-d7 / 16:1)-IS(543.5>271.4), C18:1Ceramide-d7(d18:1-d7 / 18:1)-IS( 571.6>271.4), C20:1Ceramide-d7(d18:1-d7 / 20:1)-IS(599.6>271.4), C22:1Ceramide-d7(d18:1-d7 / 22:1)-IS(627.6>271.4)、C24:1Ceramide-d7(d18:1-d7 / 24:1)-IS(655.7>271.4)、16:1SM(d18:1 / 16:1)-d9-IS(710.6>193.1)、18:1SM(d18:1 / 18:1)-d9-IS(738.6>193.1)、20:1SM(d18:1 / 20:1)-d9-IS(766.7>193.1)、22:1SM(d18:1 / 22:1)-d9-IS(794.7>193.1)、24:1SM(d18:1 / 24:1)-d9-IS(822.7>193.1)、17:0-14:1PC-d5+AcO.IS(781.6>225.3)、17:0-16:1PC-d5+AcO.IS(809.6>253.3)、17:0-18:1PC-d5+AcO.IS(837.6>281.3)、17:0-20:3PC-d5+AcO.IS(861.6>305.3)、17:0-22:4PC-d5+AcO.IS(887.6>331.3)、17:0-14:1PE-d5.IS(679.5>225.3)、17:0-16:1PE-d5.IS(707.5>253.3)、17:0-18:1PE-d5.IS(735.6>281.3)、17:0-20:3PE-d5.IS(759.6>305.3)、17:0-22:4PE-d5.IS(785.6>331.3)、17:0-14:1PG-d5.IS(710.5>225.3)、17:0-16:1PG-d5.IS(738.5>253.3)、17:0-18:1PG-d5.IS(766.6>281.3)、17:0-20:3PG-d5.IS(790.6>305.3)、17:0-22:4PG-d5.IS(816.6>331.3)、17:0-14:1PS-d5.IS(723.5>269.3)、17:0-16:1PS-d5.IS(751.5>269.3)、17:0-18:1PS-d5.IS(779.5>269.3)、17:0-20:3PS-d5.IS(803.5>269.3)、17:0-22:4PS-d5.IS(829.6>269.3)、17:0-14:1PI-d5.IS(798.6>269.3)、17:0-16:1PI-d5.IS(826.6>269.3)、17:0-18:1PI-d5.IS(854.6>269.3)、17:0-20:3PI-d5.IS(878.6>269.3), 17:0-22:4PI-d5.IS(904.6>269.3), 15:0Lyso PI-d5.IS(562.3>241.2), 17:0LysoPI-d5.IS(590.4>269.2), 19:0Lyso PI-d5.IS(618.4>297.2), 15:0Lyso PS-d5.IS(487.3>241.2), 17:0Lyso PS-d5.IS(515.3>269.2), 19:0Lyso PS-d5.IS(543.3>297.2), 15:0LysoPG-d5.IS(474.3>241.2), 17:0LysoPG-d5.IS(502.3>269.2), 19:0Lyso PG-d5.IS(530.3>297.2), 15:0Lyso PC-d5.IS(545.3>241.2), 17:0Lyso PC-d5.IS(573.4>269.2), 19:0LysoPC-d5.IS(601.4>297.2), 15:0LysoPE-d5.IS(443.3>241.2), 17:0Lyso PE-d5.IS(471.3>269.2)、19:0Lyso PE-d5.IS(499.4>297.2...)

[0050] Furthermore, in (6.2), the constructed binomial fitting distribution equation of the relative lipid retention time y and relative carbon chain length x is as follows: CE X:0: y=-1.899e-015+-2.581ex-015x 2 (R 2 =1), CE X:0:y=1.019-0.8169x+0.7936x 2 (R 2 =0.9922), SM(X:0):y=0.1993+0.109x+0.1609x 2 (R 2 =0.9989), SM(X:1):y=0.3041-0.1518x+0.2847x 2 (R 2 =0.9870), DG(X:0):y=-0.3931+1.54x-0.5524x 2 (R 2 =0.9943), DG(X:1):y=-0.115+0.75x+4.972e-014x 2 (R 2=1)、DG(X:2):y=-2.489+4.675x-1.691x 2 (R 2 =0.9997)、DG(X:3):y=-2.793+6..647x-3.386x 2 (R 2 =0.9881)、DG(X:4):y=-2.331+5.269x-2.407x 2 (R 2 =0.9967)、DG(X:5):y=0.1968-1.747x+2.31x 2 (R 2 =0.9989)、TAG(X:0):y=-0.4586+2.191x-0.8528x 2 (R 2 =0.9956)、TAG(X:1):y=-0.199+1.133x-1.138e-013x 2 (R 2 =0.9966)、TAG(X:2):y=-0.7236+2.574x-1.101x 2 (R 2 =0.9765)、TAG(X:3):y=0.3629-0.1095x+0.4762x 2 (R 2 =1)、TAG(X:4):y=-3.905+9.136x-4.545x 2 (R 2 =1)、TAG(X:5):y=-3.829+8.812x-4.287x 2 (R 2 =0.9662)、TAG(X:6):y=4.338-9.887x+6.478x 2 (R 2 =0.9942)、TAG(X:7):y=-1.571+3.571x-1.306x 2 (R 2 =0.9990)、TAG(X:8):y=2.011-3.939x+2.644x 2 (R 2 =1)、TAG(X:9):y=-0.0918+0.6308x+0.1302x 2 (R 2 =0.9995)、TAG(X:10):y=4.259-8.341x+4.762x 2 (R2 =0.9995), CER(X:0):y=0.1332+0.3386x+0.06678x 2 (R 2 =1), CER(X:1):y=0.1504+0.2218x+0.09077x 2 (R 2 =0.9997), DCER(X:0):y=0.3659-0.1912x+0.3676x 2 (R 2 =1), HCER(X:0):y=-0.8825+2.875x-1.562x 2 (R 2 =1), LCER(X:0):y=0.1636+0.9558x-0.4126x 2 (R 2 =0.9994), PC(X:0):y=0.1139+0.2522x+0.112x 2 (R 2 =0.9997), PC(X:1):y=-0.125+0.615x-1.292e-005x 2 (R 2 =1), PC(X:2):y=0.3871-0.398x+0.4326x 2 (R 2 =0.9997), PE(X:1):y=-0.125+0.645x-1.491e-005x 2 (R 2 =1), PS(X:1)y=-0.1610+0.7380x+4.767e-009x 2 (R 2 =0.9969), PG(X:1)y=1.424-2.925x+1.901x 2 (R 2 =1), PI(X:1)y=-0.02001+0.39x-7.415e-006x 2 (R 2 =1), FFA(X:0)y=-0.2621+0.99881x-0.3534x 2 (R 2 =0.9967), where X is the carbon chain length.

[0051] This invention also validated and applied the above-mentioned method for analyzing lipids with odd carbon chains. The validation and application results are as follows:

[0052] The lipids used in the library construction were validated using glycerol ester standards: Tritridecanoin (C13:0, CAS#26536-12-9), Tripentadecanoin (C15:0, CAS#7370-46-9), Triheptadecanoin (C17:0, CAS#2438-40-6), and Tritricosanoin (C23:0, CAS#86850-72-8). The prediction model was validated, and an MRM (Q1>Q3) detection ion pair was constructed based on the mass spectrometry fragmentation rules described in (2) of the above analytical method. The values ​​of Tritridecanoin (699.08>467.88), Tripentadecanoin (783.24>524.04), Triheptadecanoin (867.4>580.2), and Tritricosanoin (1119.88>748.68) were analyzed in LC-MS according to the separation conditions described in (4) of the above analysis method. The obtained RT values ​​were 8.58, 10.34, 11.88, and 15.03 min, respectively. These values ​​are basically consistent with the predicted values, thus proving the accuracy of the prediction model of the present invention.

[0053] Validation of lipids in the library using internal standard: Lipids were analyzed by LC-MS under the separation conditions described in (4) of the above analytical method. Representative lipids DG (17:0 / 16:1) and 17:0-16:1DG-d5, TAG43:1-FA15:0 and 14:0-15:1-14:1TAG-d5, LPE (19:0) and 19:0lyso PE-d5, PE (17:0 / 16:1) and 17:0-16:1PE-d5, PG (17:0 / 18:1) and 17:0-18:1PG-d5, and PI (17:0 / 16:1) and 17:0-16:1PI-d5 showed basically consistent retention times. This indicates that the RT prediction model based on ECN has good accuracy (validation results can be found in the appendix). Figure 2 ).

[0054] Biological sample response verification: This invention also utilizes colon cancer tissue samples to obtain total lipid extracts according to the lipid extraction process described in (3) of the above analytical method. After LC-MS separation under the conditions described in (4) of the above analytical method, total ion chromatograms of the tissue samples were obtained under positive and negative electrospray ionization modes (see appendix). Figure 3 In both modes, lipids with odd-numbered carbon chains showed good mass spectrometry signal responses.

[0055] The existence of lipid isomerism: This invention also uses colon cancer tissue samples to obtain total lipid extract according to the lipid extraction process described in (3) of the above analytical method. After separation by LC-MS under the conditions described in (4) of the above analytical method, the extracted ion chromatogram of lipids was obtained. Analysis showed that lipid isomerism is common. For example, under positive ion mode, TAG47:1, TAG49:1, TAG51:2 and TAG51:3, and under negative ion mode, PC (21:0 / 18:1), PE (21:0 / 18:1), PG (17:1 / 18:1), PA (17:0 / 18:1), PI (17:0 / 18:1) and PS (17:0 / 18:1) were observed. The existence of multiple chromatographic peaks was thus confirmed (see Appendix). Figure 4 ).

[0056] Evaluation between predicted and predicted values ​​of the RT model: LC-MS analysis of total lipids in colon cancer tissue yielded the actual RT values ​​for each lipid. The differences between these actual and predicted values ​​are shown in the statistical chart attached. Figure 5 A) As shown in the figure, under the positive and negative ion mode, the difference between the predicted value and the measured value is less than 0.4 min, which is within a reasonable range, indicating that the model has good predictive ability.

[0057] Stability assessment of the analytical method: After 10 consecutive LC-MS injections of total lipids from colon cancer tissue, the coefficient of variation (CV%) of various lipids was less than 20%, indicating that TAG showed better detection stability. This is consistent with the higher mass spectrometry response of the positive ion detection mode. (See appendix) Figure 5 B).

[0058] Statistical analysis of the peak widths at the half-peak showed that the half-peak widths of the lipids in the library were all within the range of 0.4 min, and that positive ion mode detection of lipids exhibited better peak shapes. These results indicate that the method possesses good analytical stability (see appendix). Figure 5 C).

[0059] The present invention also provides an application of the above-mentioned analytical method, wherein a total lipid extract is obtained from a tissue or plasma sample according to the lipid extraction process described in (3) of the above-mentioned analytical method, and the odd-numbered carbon chain fatty acids in the sample can be detected by LC-MS described in (4) of the above-mentioned analytical method.

[0060] The beneficial effects of this invention are as follows: Currently, there are no reported methodologies for odd-chain lipids. Here, this invention provides an analytical method for odd-chain lipids using reversed-phase chromatography-mass spectrometry. This method has good reproducibility and stability, a wide half-peak, a coefficient of variation (CV) of less than 20%, and no significant difference in peak area after 10 consecutive injections of the internal standard. It can achieve high-throughput analysis of odd-chain lipids. Attached Figure Description

[0061] Figure 1 A simplified flowchart for method establishment and application;

[0062] Figure 2 The retention time matching degree between the standard and the deuterated internal standard and the predicted lipid is given. Among them, A and B are the retention time matching degree between DG and TAG and the corresponding standard in the positive ion mode, and C, D, E and F are the retention time matching degree between each phospholipid molecule and its respective deuterated internal standard in the negative ion mode.

[0063] Figure 3 The total ion current chromatograms of the tissue samples under positive and negative electrospray ionization modes are shown, where A is the total ion current chromatogram under positive electrospray ionization mode and B is the total ion current chromatogram under negative electrospray ionization mode.

[0064] Figure 4 The presence of lipid isomers;

[0065] Figure 5 For model prediction and method stability evaluation, A is the difference between predicted and measured retention time, B is the CV% evaluation, and C is the peak height at half maximum.

[0066] Figure 6 The results of the method application are shown in the principal component analysis diagram of odd-numbered carbon chain lipids in colorectal cancer tissue, where CC-A: normal tissue, CC-B: adjacent normal tissue, and CC-C: cancer tissue.

[0067] Figure 7 The results of the method application - partial least squares regression analysis of odd-numbered carbon chain lipids in colorectal cancer tissue, where CC-A: normal tissue, CC-B: adjacent normal tissue, CC-C: cancer tissue;

[0068] Figure 8 The results of the method application are shown in the lipid volcano plot of odd-numbered carbon chains in colorectal cancer tissue. In the figure, Figure A is CC-A vs CC-B, and Figure B is CC-B vs CC-C. CC-A: normal tissue, CC-B: adjacent tissue, and CC-C: cancer tissue. Detailed Implementation

[0069] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. It should be understood that the following embodiments are merely illustrative and explanatory of the present invention and should not be construed as limiting the scope of protection of the present invention. All technologies implemented based on the above content of the present invention are covered within the scope of protection intended by the present invention. Unless otherwise stated, the raw materials and reagents used in the following embodiments are commercially available products or can be prepared by known methods.

[0070] Example 1

[0071] The inventors used the reversed-phase chromatography-mass spectrometry method disclosed in this invention to quantify odd-carbon chain fatty acyl lipids in cancerous tissues, adjacent tissues, and normal tissues of 12 colorectal cancer patients. The specific implementation method is as follows:

[0072] 1. Instruments: Waters Aquity UPLC liquid chromatography combined with AB Sciex QTRAP 6500 mass spectrometer, equipped with an electrospray ionization source.

[0073] 2. Preparation of internal standard: The lipidomics internal standard was Ultimate SPLASH. TM ONE (Cat#330820L, Avanti) was used directly after purchase without further verification or purification.

[0074] 3. Sample preparation:

[0075] Sample preparation includes two parts: tissue sample preparation after sample acquisition and lipid extraction.

[0076] Tissue sample preparation: 1> Dry the tissue with filter paper, weigh 50.0 mg of tissue sample and place it in a 2 mL round-bottom centrifuge tube; 2> Add 500 μL of physiological saline (0.85% physiological saline preparation: add 0.85 g NaCl to 100 mL of deionized water, then add 40 μL of glacial acetic acid); 3> Grind: Add 2 large magnetic beads (small magnetic beads are only used to homogenize liver, kidney, brain and mucosal tissues, generally two small magnetic beads are used); 4> Place the sample tube in a tissue homogenizer and homogenize twice, 30 seconds each time, with a 10-second interval to avoid temperature rise; 5> Centrifuge the homogenate at 12000 rpm for 15 minutes and use it for further lipid extraction.

[0077] Lipid extraction from tissue samples: 1> Mix 50 μL of tissue homogenate supernatant with 5 μL of UltimateSPLASH. TM Add the ONE internal standard to a glass tube, then add 950 μL H2O, 2 mL MeOH, and 0.9 mL DCM; gently shake the tube for 5 seconds. No separation should occur at this time. If separation occurs, add an additional 50 μL MeOH.

[0078] 2> Then place the tube at room temperature for 30 minutes, and then add 1 mL of H2O and 0.9 mL of DCM in sequence. Gently shake the tube for 5 seconds, and then centrifuge at 2000 rpm for 15 minutes until the liquid separates into layers (Note: excessive shaking may cause poor separation).

[0079] 3. Transfer the layered bottom liquid to a new glass tube using a pipette, and add 1.8 mL of DCM to the original tube for a second extraction. After the above steps, combine the second extract with the first phase. Then, dry the extract with a stable stream of nitrogen.

[0080] 4. Next, add 100 μL of working solution to the tube to dissolve it (working solution composition: DCM containing 10 mM ammonium acetate: MeOH, 1:1, v / v). Then, shake the sample on a benchtop shaker to mix thoroughly and transfer it to a 200 μL LC / MS insert tube. Place the insert tube into a 1.5 mL plastic EP tube and centrifuge at 16000 rpm for 10 minutes. After centrifugation, transfer the insert tube to an LC-MS vial.

[0081] 5. Take 10 μL of each sample and place it in a QC tube, then use it as a control and mix thoroughly. Finally, store the lipid extract in the LC-MS vial at -80°C until LC-MS / MS analysis.

[0082] 4. Test and analysis methods:

[0083] (1) Construction of an odd-numbered carbon chain lipid library: The lipids in the lipid library were downloaded from the Lipid Metabolism Pathway Research Project LIPIDMAP, S https: / / www.lipidmaps.org / databases / lmsd / browse structural database and computational simulation. The lipid library screening criteria were cholesterol ester CE, triglyceride TAG, diglyceride DG, monoglyceride MAG, sphingolipid SM, ceramide Cer, free fatty acid FFA, phosphatidic acid PA, phosphatidylcholine PC, phosphatidylethanolamine PE, phosphatidylglycerol PG, phosphatidylinositol PI, and phosphatidylserine PS. Lipids containing odd-numbered carbon chains were screened as shown in No.1-No.678.

[0084] (2) Construction of MRM ion pairs for mass spectrometry of lipids with odd carbon chains: The MRM detection ion pairs consist of a parent ion Q1 and a daughter ion Q3. In the positive ion mode, MRM ion pairs for CE, glycerides TAG, DG, MAG, CER, DCER, HCER, LCER, and SM are constructed. In the negative ion detection mode, MRM ion pairs for phospholipids containing PA, PC, PE, PG, PS, PI, and FFA are constructed.

[0085] In positive ion mode, Q1 construction follows these characteristics: CE, MAG, SM, CER, DCER, HCER, and LCER are [M+H]+, while DG and TAG are [M+NH]+. 4 ] + The construction of Q3 follows these characteristics: The Q3 mass-to-charge ratio m / z for CE is 369, for SM it is 184, for CER it is 264, for DCER it is 266, for HCER it is 264, and for LCER it is 264. For MAG, based on the neutral loss fragment NLS being a propylene glycol residue with a mass number of 74, the m / z of Q3 is determined to be Q1-74. For DG, the m / z of Q3 is the mass number of the odd-numbered carbon chain acyl group + the 74Da of the head group propylene glycol residue, i.e., the mass number of the odd-numbered carbon chain acyl group + 74. Furthermore, for every 1 increase in unsaturation, the m / z of Q3 decreases by 2. For TAG, the m / z of Q3 is Q1-NLS, where NLS is the mass number of the odd-numbered carbon chain acyl group + 34. The construction of Q1 in negative ion mode follows these characteristics: The Q1 mass-to-charge ratio m / z for PC is [M+acetyl AcO]. - The Q1 m / z for the remaining PA, PE, PG, PI, and PS is [MH]. - Based on the neutral loss fragment (NLS) characteristics of various lipids, the m / z of Q3 is Q1-NLS, where PA's NLS is 153+acyl group mass number, PC's NLS is 298+acyl group mass number, PE's NLS is 196+acyl group mass number, PG's NLS is 227+acyl group mass number, PI's NLS is 315+acyl group mass number, PS's NLS is 240+acyl group mass number, and FFA's Q1 = Q3, with m / z of [MH]. - .

[0086] (3) Lipid extraction: The internal standard Ultimate SPLASH is added during lipid extraction. TM ONE, after lipid extraction, the extract was dried with nitrogen flow and added to working solution to prepare the injection concentration. The working solution contained 10 mM ammonium acetate in DCM:MeOH, 1:1, v / v.

[0087] (4) Liquid chromatography-mass spectrometry separation:

[0088] Mass spectrometry conditions: The electrospray ionization source adopted a positive and negative ion switching scanning mode. In positive ion mode, the spray voltage was 5200V, the ion source temperature was 350℃, the nebulizer gas was 40psi, the curtain gas was 55psi, and the desolvator gas was 55psi. In negative ion mode, the spray voltage was -4500V, the ion source temperature was 600℃, the nebulizer gas was 35psi, the curtain gas was 60psi, and the desolvator gas was 60psi.

[0089] Chromatographic conditions: A BEH C18 column (2.1 mm × 50 mm, 1.8 μm, Waters) was used. Mobile phase A was methanol:acetonitrile:water 1:1:1, v / v / v, +5 mM ammonium acetate; mobile phase B was isopropanol +5 mM ammonium acetate. The elution gradient was: 0 min 20% B, 1 min 20% B, 2.5 min 40% B, 4 min 60% B, 14 min 90% B, 15 min 90% B, 15.1 min 20% B, 17 min 20% B. The flow rate was 0.2 mL / min, and the column temperature was set to 40 °C.

[0090] (5) Liquid phase separation and mass spectrometry determination of internal standard:

[0091] The lipidomics internal standard is Ultimate Splash. TM ONE (Cat#330820L, Avanti), the MRM ion pair of the internal standard was constructed according to the rules described in (2); the internal standard was separated by liquid chromatography-mass spectrometry under the separation conditions described in (4).

[0092] Specifically, the MRM (Q1>Q3) ion pairs of the internal standard are constructed as follows: 17:0-14:1 DG-d5+NH4-IS (575.5>332.2), 17:0-16:1 DG-d5+NH4-IS (603.5>332.2), 17:0-18:1 DG-d5+NH4-IS (631.6>332.2), 17:0-20:3 DG-d5+NH4-IS (655.6>332.2), 17:0-22:4 DG-d5+NH4-IS (681.6>332.2), 14:0-13:0-14:0 TAG-d5+NH4-IS (731.7>486.4), 14:0-15:1-14:0 TAG-d5+NH4-IS (757.7>512.4), 14:0-17:1-14:0 TAG-d5+NH4-IS (785.7>540.5), 16:0-15:1-16:0 TAG-d5+NH4-IS (813.7>540.5), 16:0-17:1-16:0 TAG-d5+NH4-IS (841.8>568.5), 16:0-19:2-16:0 TAG-d5+NH4-IS (867.8>594.5), 18:1-17:1-18:1 TAG-d5+NH4-IS (893.8>594.5), 18:1-19:2-18:1 TAG-d5+NH4-IS (919.8>620.5), 18:1-21:2-18:1 TAG-d5+NH4-IS (947.8>648.6), 14:1 cholesteryl-d7 ester-IS (619.6>376.4), 16:1 cholesteryl-d7 ester-IS (647.6>376.4), 18:1 cholesteryl-d7 ester-IS (675.6>376.4), 20:3 cholesteryl-d7 ester-IS (699.6>376.4), 22:4 cholesteryl-d7 ester-IS (725.7>376.4), C16:1 Ceramide-d7 (d18:1-d7 / 16:1)-IS (543.5>271.4), C18:1 Ceramide-d7 (d18:1-d7 / 18:1)-IS (571.6>271.4), C20:1 Ceramide-d7 (d18:1-d7 / 20:1)-IS (599.6>271.4), C22:1 Ceramide-d7 (d18:1-d7 / 22:1)-IS (627.6>271.4)、C24:1Ceramide-d7(d18:1-d7 / 24:1)-IS(655.7>271.4)、16:1SM(d18:1 / 16:1)-d9-IS(710.6>193.1)、18:1SM(d18:1 / 18:1)-d9-IS(738.6>193.1)、20:1 SM(d18:1 / 20:1)-d9-IS(766.7>193.1)、22:1SM(d18:1 / 22:1)-d9-IS(794.7>193.1)、24:1 SM(d18:1 / 24:1)-d9-IS(822.7>193.1)、17:0-14:1PC-d5+AcO.IS(781.6>225.3)、17:0-16:1 PC-d5+AcO.IS(809.6>253.3)、17:0-18:1PC-d5+AcO.IS(837.6>281.3)、17:0-20:3PC-d5+AcO.IS(861.6>305.3)、17:0-22:4PC-d5+AcO.IS(887.6>331.3)、17:0-14:1PE-d5.IS(679.5>225.3)、17:0-16:1PE-d5.IS(707.5>253.3)、17:0-18:1PE-d5.IS(735.6>281.3)、17:0-20:3PE-d5.IS(759.6>305.3)、17:0-22:4PE-d5.IS(785.6>331.3)、17:0-14:1PG-d5.IS(710.5>225.3)、17:0-16:1PG-d5.IS(738.5>253.3)、17:0-18:1PG-d5.IS(766.6>281.3)、17:0-20:3PG-d5.IS(790.6>305.3)、17:0-22:4PG-d5.IS(816.6>331.3)、17:0-14:1PS-d5.IS(723.5>269.3)、17:0-16:1PS-d5.IS(751.5>269.3)、17:0-18:1PS-d5.IS(779.5>269.3)、17:0-20:3PS-d5.IS(803.5>269.3)、17:0-22:4PS-d5.IS(829.6>269.3)、17:0-14:1PI-d5.IS(798.6>269.3)、17:0-16:1PI-d5.IS(826.6>269.3)、17:0-18:1PI-d5.IS(854.6>269.3)、17:0-20:3PI-d5.IS(878.6>269.3), 17:0-22:4PI-d5.IS(904.6>269.3), 15:0Lyso PI-d5.IS(562.3>241.2), 17:0LysoPI-d5.IS(590.4>269.2), 19:0Lyso PI-d5.IS(618.4>297.2), 15:0Lyso PS-d5.IS(487.3>241.2), 17:0LysoPS-d5.IS(515.3>269.2), 19:0Lyso PS-d5.IS(543.3>297.2), 15:0LysoPG-d5.IS(474.3>241.2), 17:0Lyso PG-d5.IS(502.3>269.2), 19:0Lyso PG-d5.IS(530.3>297.2), 15:0LysoPC-d5.IS(545.3>241.2), 17:0Lyso PC-d5.IS(573.4>269.2), 19:0LysoPC-d5.IS(601.4>297.2), 15:0Lyso PE-d5.IS(443.3>241.2), 17:0Lyso PE-d5.IS(471.3>269.2), 19:0LysoPE-d5.IS(499.4>297.2..

[0093] (6) Prediction of lipid retention time in odd-numbered carbon chain lipid libraries:

[0094] (6.1) Acquisition of internal standard retention time: Based on the separation conditions described in (4) and the internal standard mass spectrometry fragmentation rules described in (5), the RT of the internal standard is obtained in the LC-MS system.

[0095] (6.2) Prediction of retention time of various lipids:

[0096] According to the ECN model, in reversed-phase liquid chromatography, the lipid retention time and the number of unsaturated double bonds or the carbon chain length conform to a binomial fitting distribution. Based on this theory, a binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x is constructed according to the retention time and carbon chain length in (6.1), where y is the lipid retention time / total elution time of 17 min, and the relative carbon chain length is the number of carbon atoms of the same type of lipid / the maximum number of carbon atoms. The lipid retention time of odd-numbered carbon chain lipid libraries is predicted based on the constructed binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x.

[0097] Specifically, the binomial fitting distribution equation for the relative lipid retention time y and relative carbon chain length x constructed in (6.2) is as follows: CE X:0: y=-1.899e-015+-2.581ex-015x2 (R 2 =1)、CE X:0:y=1.019-0.8169x+0.7936x 2 (R 2 =0.9922)、SM(X:0):y=0.1993+0.109x+0.1609x 2 (R 2 =0.9989)、SM(X:1):y=0.3041-0.1518x+0.2847x 2 (R 2 =0.9870)、DG(X:0):y=-0.3931+1.54x-0.5524x 2 (R 2 =0.9943)、DG(X:1):y=-0.115+0.75x+4.972e-014x 2 (R 2 =1)、DG(X:2):y=-2.489+4.675x-1.691x 2 (R 2 =0.9997)、DG(X:3):y=-2.793+6..647x-3.386x 2 (R 2 =0.9881)、DG(X:4):y=-2.331+5.269x-2.407x 2 (R 2 =0.9967)、DG(X:5):y=0.1968-1.747x+2.31x 2 (R 2 =0.9989)、TAG(X:0):y=-0.4586+2.191x-0.8528x 2 (R 2 =0.9956)、TAG(X:1):y=-0.199+1.133x-1.138e-013x 2 (R 2 =0.9966)、TAG(X:2):y=-0.7236+2.574x-1.101x 2 (R 2 =0.9765)、TAG(X:3):y=0.3629-0.1095x+0.4762x 2 (R 2 =1)、TAG(X:4):y=-3.905+9.136x-4.545x 2 (R 2=1)、TAG(X:5):y=-3.829+8.812x-4.287x 2 (R 2 =0.9662)、TAG(X:6):y=4.338-9.887x+6.478x 2 (R 2 =0.9942)、TAG(X:7):y=-1.571+3.571x-1.306x 2 (R 2 =0.9990)、TAG(X:8):y=2.011-3.939x+2.644x 2 (R 2 =1)、TAG(X:9):y=-0.0918+0.6308x+0.1302x 2 (R 2 =0.9995)、TAG(X:10):y=4.259-8.341x+4.762x 2 (R 2 =0.9995)、CER(X:0):y=0.1332+0.3386x+0.06678x 2 (R 2 =1)、CER(X:1):y=0.1504+0.2218x+0.09077x 2 (R 2 =0.9997)、DCER(X:0):y=0.3659-0.1912x+0.3676x 2 (R 2 =1)、HCER(X:0):y=-0.8825+2.875x-1.562x 2 (R 2 =1)、LCER(X:0):y=0.1636+0.9558x-0.4126x 2 (R 2 =0.9994)、PC(X:0):y=0.1139+0.2522x+0.112x 2 (R 2 =0.9997)、PC(X:1):y=-0.125+0.615x-1.292e-005x 2 (R 2 =1)、PC(X:2):y=0.3871-0.398x+0.4326x 2 (R 2 =0.9997)、PE(X:1):y=-0.125+0.645x-1.491e-005x 2 (R2 =1), PS(X:1)y=-0.1610+0.7380x+4.767e-009x 2 (R 2 =0.9969), PG(X:1)y=1.424-2.925x+1.901x 2 (R 2 =1), PI(X:1)y=-0.02001+0.39x-7.415e-006x 2 (R 2 =1), FFA(X:0)y=-0.2621+0.99881x-0.3534x 2 (R 2 =0.9967), where X is the carbon chain length.

[0098] The predicted lipid retention times for odd-carbon chain lipid libraries are shown in the table below:

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] (7) Based on the predicted lipid retention time of the odd-numbered carbon chain lipid library and the constructed MRM ion pairs, the odd-numbered carbon chain lipids in the test sample were confirmed and quantified.

[0115] 5. Analysis Results:

[0116] Using the analytical method of this invention, odd-numbered carbon chain lipid LC-MS was performed on distal normal tissue (CC-A), adjacent normal tissue (CC-B), and cancerous tissue (CC-C) of 12 patients with colorectal cancer. Figure 6 As shown, principal component analysis (PCA) revealed no significant outliers in the detection data. Figure 7 As shown, partial least squares regression analysis (OPLS-DA) demonstrates that CC-A vs CC-B and CC-B vs CC-C can achieve good differentiation, such as... Figure 8 As shown, volcano plot analysis can effectively distinguish differential metabolites between CC-A and CC-B, and between CC-B and CC-C. This demonstrates that the method has good analytical capabilities and is suitable for omics analysis of lipids with odd-numbered carbon chains.

[0117] Example 2

[0118] The lipids used in the library construction were validated using glycerol ester standards: Tritridecanoin (C13:0, CAS#26536-12-9), Tripentadecanoin (C15:0, CAS#7370-46-9), Triheptadecanoin (C17:0, CAS#2438-40-6), and Tritricosanoin (C23:0, CAS#86850-72-8). The prediction model was validated, and an MRM (Q1>Q3) detection ion pair was constructed based on the mass spectrometry fragmentation rules described in (2) of the above analytical method. The values ​​of Tritridecanoin (699.08>467.88), Tripentadecanoin (783.24>524.04), Triheptadecanoin (867.4>580.2), and Tritricosanoin (1119.88>748.68) were analyzed in LC-MS according to the separation conditions described in (4) of the above analysis method. The obtained RT values ​​were 8.58, 10.34, 11.88, and 15.03 min, respectively. These values ​​are basically consistent with the predicted values, thus proving the accuracy of the prediction model of the present invention.

[0119] Example 3

[0120] Validation of lipids in the library using internal standard: Lipids were analyzed by LC-MS under the separation conditions described in (4) of the above analytical method. Representative lipids DG (17:0 / 16:1) and 17:0-16:1DG-d5, TAG43:1-FA15:0 and 14:0-15:1-14:1TAG-d5, LPE (19:0) and 19:0lyso PE-d5, PE (17:0 / 16:1) and 17:0-16:1PE-d5, PG (17:0 / 18:1) and 17:0-18:1PG-d5, and PI (17:0 / 16:1) and 17:0-16:1PI-d5 showed basically consistent retention times. This indicates that the RT prediction model based on ECN has good accuracy (validation results can be found in the appendix). Figure 2 ).

[0121] Example 4

[0122] Biological sample response verification: This invention also utilizes colon cancer tissue samples to obtain total lipid extracts according to the lipid extraction process described in (3) of the above analytical method. After LC-MS separation under the conditions described in (4) of the above analytical method, total ion chromatograms of the tissue samples were obtained under positive and negative electrospray ionization modes (see appendix). Figure 3 In both modes, lipids with odd-numbered carbon chains showed good mass spectrometry signal responses.

[0123] Example 5

[0124] The existence of lipid isomerism: This invention also uses colon cancer tissue samples to obtain total lipid extract according to the lipid extraction process described in (3) of the above analytical method. After separation by LC-MS under the conditions described in (4) of the above analytical method, the extracted ion chromatogram of lipids was obtained. Analysis showed that lipid isomerism is common. For example, under positive ion mode, TAG47:1, TAG49:1, TAG51:2 and TAG51:3, and under negative ion mode, PC (21:0 / 18:1), PE (21:0 / 18:1), PG (17:1 / 18:1), PA (17:0 / 18:1), PI (17:0 / 18:1) and PS (17:0 / 18:1) were observed. The existence of multiple chromatographic peaks was thus confirmed (see Appendix). Figure 4 ).

[0125] Example 6

[0126] Evaluation between predicted and predicted values ​​of the RT model: LC-MS analysis of total lipids in colon cancer tissue yielded the actual RT values ​​for each lipid. The differences between these actual and predicted values ​​are shown in the statistical chart attached. Figure 5A) As shown in the figure, under the positive and negative ion mode, the difference between the predicted value and the measured value is less than 0.4 min, which is within a reasonable range, indicating that the model has good predictive ability.

[0127] Example 7

[0128] Stability assessment of the analytical method: After 10 consecutive LC-MS injections of total lipids from colon cancer tissue, the coefficient of variation (CV%) of various lipids was less than 20%, indicating that TAG showed better detection stability. This is consistent with the higher mass spectrometry response of the positive ion detection mode. (See appendix) Figure 5 B).

[0129] Statistical analysis of the peak widths at the half-peak showed that the half-peak widths of the lipids in the library were all within the range of 0.4 min, and that positive ion mode detection of lipids exhibited better peak shapes. These results indicate that the method possesses good analytical stability (see appendix). Figure 5 C).

[0130] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for analyzing odd-numbered carbon chain lipids using reversed-phase chromatography-mass spectrometry, characterized in that, Includes the following steps: (1) Construction of an odd-numbered carbon chain lipid library: The lipids in the lipid library were all downloaded from the Lipid Metabolism Pathway Research Project (LIPIDMAPS, https: / / www.lipidmaps.org / databases / lmsd / browse) structural database and computational simulation. The lipid library screening criteria were cholesterol ester CE, triglyceride TAG, diglyceride DG, monoglyceride MAG, sphingolipid SM, ceramide Cer, free fatty acid FFA, phosphatidic acid PA, phosphatidylcholine PC, phosphatidylethanolamine PE, phosphatidylglycerol PG, phosphatidylinositol PI, and phosphatidylserine PS; (2) Construction of mass spectrometry multi-level reaction monitoring ion pairs (MRM) for lipids with odd-numbered carbon chains: The MRM consists of a parent ion Q1 and a daughter ion Q3. In positive ion mode, MRM ion pairs for CE, glycerides TAG, DG, MAG, CER, DCER, HCR, LCE, and SM are constructed. In negative ion detection mode, MRM ion pairs for phospholipids containing PA, PC, PE, PG, PS, PI, and FFA are constructed. In positive ion mode, Q1 is constructed according to the following characteristics: CE, MAG, SM, CER, DCER, HCR, and LCE are [M+H]. + DG and TAG are [M+NH4] + The construction of Q3 follows these characteristics: The Q3 mass-to-charge ratio (m / z) for CE is 369, for SM it is 184, for CER it is 264, for DCER it is 266, for HCER it is 264, and for LCER it is 264. For MAG, based on the neutral loss fragment NLS being a propylene glycol residue with a mass number of 74, the m / z of Q3 is determined to be Q1-74. For DG, the m / z of Q3 is the mass number of an odd-numbered carbon chain acyl group + the propylene glycol head group 74 Da, i.e., the mass number of an odd-numbered carbon chain acyl group + 74. Furthermore, for every 1 increase in unsaturation, the m / z of Q3 decreases by 2. For TAG, the m / z of Q3 is Q1-NLS, where NLS is the mass number of an odd-numbered carbon chain acyl group + 34. The construction of Q1 in negative ion mode follows these characteristics: The Q1 mass-to-charge ratio m / z for PC is [M+acetyl AcO]. - The Q1 m / z for the remaining PA, PE, PG, PI, and PS is [MH]. - Based on the neutral loss fragment (NLS) characteristics of various lipids, the m / z of Q3 is Q1-NLS, where PA's NLS is 153+acyl group mass number, PC's NLS is 298+acyl group mass number, PE's NLS is 196+acyl group mass number, PG's NLS is 227+acyl group mass number, PI's NLS is 315+acyl group mass number, PS's NLS is 240+acyl group mass number, and FFA's Q1=Q3, with m / z being [MH]. - ; (3) Lipid extraction: The internal standard Ultimate SPLASH™ ONE was added during lipid extraction. After lipid extraction, the extract was dried with nitrogen gas flow and added to the working solution to prepare the injection concentration. The working solution contained 10 mM ammonium acetate in dichloromethane (DCM): methanol (MeOH), 1:1, v / v. (4) Liquid chromatography-mass spectrometry separation: Mass spectrometry conditions: Electrospray ionization (ESI) source was used in positive and negative ion scanning modes. In positive ion mode, the spray voltage was 5200 V, the ion source temperature was 350 °C, the nebulizer gas was 40 psi, the curtain gas was 55 psi, and the desolvator gas was 55 psi. In negative ion mode, the spray voltage was -4500 V, the ion source temperature was 600 °C, the nebulizer gas was 35 psi, the curtain gas was 60 psi, and the desolvator gas was 60 psi. Chromatographic conditions: A BEH C18 column (2.1 mm × 50 mm, 1.8 μm, Waters) was used. Mobile phase A was methanol:acetonitrile:water 1:1:1, v / v / v, +5 mM ammonium acetate; mobile phase B was isopropanol +5 mM ammonium acetate. The elution gradient was: 0 min 20% B, 1 min 20% B, 2.5 min 40% B, 4 min 60% B, 14 min 90% B, 15 min 90% B, 15.1 min 20% B, 17 min 20% B. The flow rate was 0.2 mL / min, and the column temperature was set to 40℃. (5) Liquid phase separation and mass spectrometry determination of internal standard: The lipidomics internal standard was Ultimate SPLASH™ ONE (Cat#330820L, Avanti). The MRM ion pair of the internal standard was constructed according to the rules described in (2). The liquid chromatography-mass spectrometry separation of the internal standard was carried out under the separation conditions described in (4). (6) Prediction of lipid retention time in odd-numbered carbon chain lipid libraries: (6.1) Acquisition of internal standard retention time: Based on the separation conditions described in (4) and the internal standard mass spectrometry fragmentation rules described in (5), the RT of the internal standard is obtained in the LC-MS system; (6.2) Prediction of retention time for various lipids: According to the ECN model, in reversed-phase liquid chromatography, the lipid retention time and the number of unsaturated double bonds or the carbon chain length conform to a binomial fitting distribution. Based on this theory, a binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x is constructed according to the retention time and carbon chain length in (6.1), where y is the lipid retention time / total elution time of 17 min, and the relative carbon chain length is the number of carbon atoms of the same type of lipid / the maximum number of carbon atoms. The lipid retention time of the odd-numbered carbon chain lipid library is predicted according to the constructed binomial fitting distribution equation of the relative lipid retention time y and the relative carbon chain length x. (7) Based on the predicted lipid retention time of the odd-numbered carbon chain lipid library and the constructed MRM ion pairs, the odd-numbered carbon chain lipids in the test sample were confirmed and quantified.

2. The method for analyzing odd-numbered carbon chain lipids using reversed-phase chromatography-mass spectrometry according to claim 1, characterized in that, In (3), the lipid extraction method includes the following: (3.1) Lipid extraction method for tissue samples: (3.1.1) Tissue sample preparation: (3.1.1.1) Dry the tissue with filter paper, weigh 50.0 mg of tissue sample and put it into a 2 mL round-bottom centrifuge tube; (3.1.1.2) Add 500 μL of physiological saline. The physiological saline is prepared as follows: Add 0.85 g NaCl to 100 mL of deionized water, and then add 40 μL of glacial acetic acid. (3.1.1.3) Grinding: Place two large magnetic beads inside; (3.1.1.4) Place the sample tube into the tissue homogenizer and homogenize twice, 30 seconds each time, with a 10-second interval, to avoid temperature rise; (3.1.1.5) Centrifuge the homogenate at 12,000 rpm for 15 minutes and use it for further lipid extraction; (3.1.2) Lipid extraction from tissue samples: (3.1.2.1) Add 50 μL of tissue homogenate supernatant and 5 μL of Ultimate SPLASH™ ONE internal standard to a glass tube, then add 950 μL of H2O, 2 mL of MeOH and 0.9 mL of DCM; gently shake the tube for 5 seconds. No separation will occur at this time. If separation occurs, add an additional 50 μL of MeOH. (3.1.2.2) Then place the tube at room temperature for 30 minutes, and then add 1 mL H2O and 0.9 mL DCM in sequence; gently shake the tube for 5 seconds, and then centrifuge at 2000 rpm for 15 minutes until the liquid separates into layers; (3.1.2.3) Transfer the layered bottom liquid to a new glass tube using a pipette, and add 1.8 mL of DCM to the original tube for a second extraction; after the above steps, combine the second extract with the first phase; then, dry the extract with a stable stream of nitrogen. (3.1.2.4) Then add 100 μL of working solution to the tube to dissolve it; then shake the sample on a benchtop shaker to mix it thoroughly and transfer it to a 200 μL LC / MS insert tube; place the insert tube into a 1.5 mL plastic EP tube and centrifuge at 16000 rpm for 10 minutes; after centrifugation, transfer the insert tube to an LC-MS vial; (3.1.2.5) Take 10 μL of each sample and put it into a QC tube, then use it as a control and mix thoroughly; finally, store the lipid extract in the LC-MS vial in a -80°C freezer until LC-MS / MS analysis; (3.2) Lipid extraction method from plasma samples: (3.2.1) Plasma sample pretreatment: (3.2.1.1) Use EDTA for anticoagulation and collect blood using a 5 mL purple-tipped tube; (3.2.1.2) Centrifuge at 1500 g for 5 min; (3.2.1.3) Take 2 mL of supernatant, aliquot it into 1.5 mL EP tubes, and store at -80℃; the sample volume used each time is 25 μL; (3.2.2) Plasma lipid extraction: (3.2.2.1) Reconstitute serum or plasma to room temperature; add 25 μL of serum or cell supernatant to a glass tube; (3.2.2.2) Add 975 μL H2O, 2 mL MeOH, and 0.9 mL dichloromethane (DCM); (3.2.2.3) Gently shake for 5 seconds. If no layering occurs, add 50 μL of MeOH. Add an appropriate amount of internal standard and mix well. (3.2.2.4) Let stand at room temperature for 30 min; add 1 mL H2O and 0.9 mL DCM; (3.2.2.5) Gently shake for 5 s; centrifuge at 2000 rpm for 15 min until the liquid separates into layers, then transfer the lower layer to a new glass centrifuge tube; (3.2.2.6) Add 1.8 mL of DCM to the original tube and repeat the extraction once; gently shake for 5 s, centrifuge at 2000 rpm for 10 min until the liquid separates into layers; (3.2.2.7) Dry the organic phase in the combined lower DCM with nitrogen, and then add 100 μL of working solution to dissolve it in a plastic EP tube; (3.2.2.8) Shake the sample on a benchtop shaker to mix it, then centrifuge at 16,000 rpm for 10 min, and aspirate 90 μL of the sample to transfer it to a small glass sample tube; (3.2.2.9) Take 20 μL from each of all samples and mix them into a QC tube as a control.

3. The method for analyzing odd-numbered carbon chain lipids using reversed-phase chromatography-mass spectrometry according to claim 1, characterized in that, The MRM (Q1>Q3) ion pairs of the internal standard in (5) are constructed as follows: 17:0-14:1 DG-d5+NH4-IS(575.5>332.2), 17:0-16:1 DG-d5+NH4-IS(603.5>332.2), 17:0-18:1 DG-d5+NH4-IS(631.6>332.2), 17:0-20:3 DG-d5+NH4-IS(655.6>332.2), 17:0-22:4 DG-d5+NH4-IS(681.6>332.2), 14:0-13:0-14:0 TAG-d5+NH4-IS(731.7>486.4), 14:0-15:1-14:0 TAG-d5+NH4-IS(757.7>512.4), 14:0-17:1-14:0 TAG-d5+NH4-IS(785.7>540.5), 16:0-15:1-16:0 TAG-d5+NH4-IS(813.7>540.5), 16:0-17:1-16:0 TAG-d5+NH4-IS(841.8>568.5), 16:0-19:2-16:0 TAG-d5+NH4-IS(867.8>594.5), 18:1-17:1-18:1 TAG-d5+NH4-IS(893.8>594.5), 18:1-19:2-18:1 TAG-d5+NH4-IS(919.8>620.5), 18:1-21:2-18:1 TAG-d5+NH4-IS(947.8>648.6), 14:1cholesteryl-d7 ester-IS(619.6>376.4), 16:1 cholesteryl-d7 ester-IS(647.6>376.4), 18:1 cholesteryl-d7 ester-IS(675.6>376.4), 20:3 cholesteryl-d7 ester-IS(699.6>376.4), 22:4 cholesteryl-d7 ester-IS(725.7>376.4), C16:1 Ceramide-d7(d18:1-d7 / 16:1)-IS(543.5>271.4), C18:1 Ceramide-d7 (d18:1-d7 / 18:1)-IS(571.6>271.4), C20:1 Ceramide-d7 (d18:1-d7 / 20:1)-IS(599.6>271.4), C22:1 Ceramide-d7(d18:1-d7 / 22:1)-IS(627.6>271.4)、C24:1 Ceramide-d7 (d18:1-d7 / 24:1)-IS(655.7>271.4)、16:1 SM (d18:1 / 16:1)-d9-IS(710.6>193.1)、18:1 SM (d18:1 / 18:1)-d9-IS(738.6>193.1)、20:1 SM (d18:1 / 20:1)-d9-IS(766.7>193.1)、22:1 SM (d18:1 / 22:1)-d9-IS(794.7>193.1)、24:1 SM (d18:1 / 24:1)-d9-IS(822.7>193.1)、17:0-14:1 PC-d5+AcO.IS(781.6>225.3)、17:0-16:1 PC-d5+AcO.IS(809.6>253.3)、17:0-18:1 PC-d5+AcO.IS(837.6>281.3)、17:0-20:3 PC-d5+AcO.IS(861.6>305.3)、17:0-22:4 PC-d5+AcO.IS(887.6>331.3)、17:0-14:1 PE-d5.IS(679.5>225.3)、17:0-16:1 PE-d5.IS(707.5>253.3)、17:0-18:1 PE-d5.IS(735.6>281.3)、17:0-20:3 PE-d5.IS(759.6>305.3)、17:0-22:4 PE-d5.IS(785.6>331.3)、17:0-14:1 PG-d5.IS(710.5>225.3)、17:0-16:1 PG-d5.IS(738.5>253.3)、17:0-18:1 PG-d5.IS(766.6>281.3)、17:0-20:3 PG-d5.IS(790.6>305.3)、17:0-22:4 PG-d5.IS(816.6>331.3)、17:0-14:1 PS-d5.IS(723.5>269.3)、17:0-16:1 PS-d5.IS(751.5>269.3)、17:0-18:1 PS-d5.IS(779.5>269.3)、17:0-20:3 PS-d5.IS(803.5>269.3)、17:0-22:4 PS-d5.IS(829.6>269.3)、17:0-14:1 PI-d5.IS(798.6>269.3)、17:0-16:1 PI-d5.IS(826.6>269.3)、17:0-18:1 PI-d5.IS(854.6>269.3)、17:0-20:3 PI-d5.IS(878.6>269.3)、17:0-22:4 PI-d5.IS(904.6>269.3)、15:0 Lyso PI-d5.IS(562.3>241.2)、17:0 Lyso PI-d5.IS(590.4>269.2)、19:0 Lyso PI-d5.IS(618.4>297.2)、15:0 Lyso PS-d5.IS(487.3>241.2)、17:0 Lyso PS-d5.IS(515.3>269.2)、19:0Lyso PS-d5.IS(543.3>297.2)、15:0 Lyso PG-d5.IS(474.3>241.2)、17:0 Lyso PG-d5.IS(502.3>269.2)、19:0 Lyso PG-d5.IS(530.3>297.2)、15:0 Lyso PC-d5.IS(545.3>241.2)、17:0 Lyso PC-d5.IS(573.4>269.2)、19:0 Lyso PC-d5.IS(601.4>297.2)、15:0Lyso PE-d5.IS(443.3>241.2)、17:0 Lyso PE-d5.IS(471.3>269.2)、19:0 Lyso PE-d5.IS(499.4>297.2。.

4. The method for analyzing odd-numbered carbon chain lipids using reversed-phase chromatography-mass spectrometry according to claim 1, characterized in that, The binomial fitting distribution equation for the relative lipid retention time y and relative carbon chain length x constructed in (6.2) is as follows: CE X:0: y=-1.899e-015+ -2.581ex -015x 2 (R) 2 =1), CE X:0: y=1.019-0.8169x+0.7936x 2 (R) 2 =0.9922), SM (X:0):y=0.1993+0.109x+0.1609x 2 (R) 2 =0.9989), SM (X:1):y=0.3041-0.1518x+0.2847x 2 (R) 2 =0.9870), DG (X:0): y=-0.3931+1.54x-0.5524x 2 (R) 2 =0.9943), DG (X:1): y=-0.115+0.75x+4.972e-014x 2 (R) 2 =1), DG (X:2): y=-2.489+4.675x-1.691x 2 (R) 2 =0.9997), DG (X:3): y=-2.793+6..647x-3.386x 2 (R) 2 =0.9881), DG (X:4): y=-2.331+5.269x-2.407x 2 (R) 2 =0.9967), DG (X:5): y=0.1968-1.747x+2.31x 2 (R) 2 =0.9989), TAG (X:0): y=-0.4586+2.191x-0.8528x 2 (R) 2 =0.9956), TAG (X:1): y=-0.199+1.133x-1.138e-013x 2 (R) 2 =0.9966), TAG (X:2): y=-0.7236+2.574x-1.101x 2 (R) 2 =0.9765), TAG (X:3): y=0.3629-0.1095x+0.4762x 2 (R) 2 =1)、TAG(X:4): y=-3.905+9.136x-4.545x 2 (R 2 =1)、TAG(X:5): y=-3.829+8.812x-4.287x 2 (R 2 =0.9662)、TAG(X:6): y=4.338-9.887x+6.478x 2 (R 2 =0.9942)、TAG(X:7):y=-1.571+3.571x-1.306x 2 (R 2 =0.9990)、TAG(X:8): y=2.011-3.939x+2.644x 2 (R 2 =1)、TAG(X:9): y=-0.0918+0.6308x+0.1302x 2 (R 2 =0.9995)、TAG(X:10): y=4.259-8.341x+4.762x 2 (R 2 =0.9995)、CER(X:0): y=0.1332+0.3386x+0.06678x 2 (R 2 =1)、CER(X:1): y=0.1504+0.2218x+0.09077x 2 (R 2 =0.9997)、DCER(X:0): y=0.3659-0.1912x+0.3676x 2 (R 2 =1)、HCER(X:0): y=-0.8825+2.875x-1.562x 2 (R 2 =1)、LCER(X:0): y=0.1636+0.9558x-0.4126x 2 (R 2 =0.9994)、PC(X:0): y=0.1139+0.2522x+0.112x 2 (R 2 =0.9997)、PC(X:1): y=-0.125+0.615x-1.292e-005x 2 (R 2 =1)、PC(X:2): y=0.3871-0.398x+0.4326x 2 (R) 2 =0.9997), PE (X:1): y=-0.125+0.645x-1.491e-005x 2 (R) 2 =1), PS (X:1)y=-0.1610+0.7380x+4.767e-009x 2 (R) 2 =0.9969), PG (X:1)y=1.424-2.925x+1.901x 2 (R) 2 =1), PI (X:1)y=-0.02001+0.39x-7.415e-006x 2 (R) 2 =1), FFA (X:0)y=-0.2621+0.99881x-0.3534x 2 (R) 2 =0.9967), where X is the carbon chain length.

Images