Synthesis, Purification, and Mass Spectrometric Characterization of Stable Isotope-Labeled Amadori-Glycated Phospholipids.

Nonenzymatic glycation of lipids plays an important role in several physiological and pathological processes, such as normal aging and complications of diabetes mellitus. To develop liquid chromatography coupled with mass spectrometric (LC-MS) methods for accurate analysis of Amadori compound-glycated lipids from biological samples, it is essential to obtain isotope-labeled Amadori-lipid standards. Herein, we report optimized methods for the preparation of six stable isotope-labeled Amadori-glycated lipid standards covering four types of lipids, including [13C6]Amadori-phosphatidyl ethanolamine (PE), -phosphatidyl serine (PS), -LysoPE, and -LysoPS. Optimal conditions for the synthesis and purification of these four types of Amadori-glycated lipids were detailed in this study. LC-MS and LC-UV analyses showed that destination products were highly purified (>95%). Accurate mass and MS/MS fragmentation in both positive- and negative-ion modes further validated the identification of these six synthetic [13C6]Amadori-glycated lipid standards. Successful preparation of these highly purified isotope-labeled standards makes it possible to develop targeted LC-MS/MS methods for accurate analysis of Amadori-glycated phospholipids from biological samples.

Introduction

Nonenzymatic glycation, generally known as the Maillard reaction, is triggered by reaction of amino group in biomolecules and carbonyl function of reducing sugars (glucose, fructose, ribose, etc.) to form an unstable Schiff base, followed by the Amadori rearrangement reaction to form a more stable Amadori product.[1] Furthermore, Amadori compounds can undergo complex reactions to form advanced glycation end products (AGEs).[1] The Amadori-modified compounds and AGEs are implicated in the pathogenesis of age-related diseases and complications of mellitus.[1−4]

The Maillard reaction of proteins and peptides has attracted the most attention in glycation and advanced glycation studies.[1] However, aminophospholipids are also important targets of nonenzymatic glycation.[3] For example, the nonenzymatic glycation of membrane lipids can cause peroxidation of proteins and membrane lipids, inactivation of receptors, and other membrane dysfunctions, which are involved in various physiological and pathological processes, such as aging, diabetes, atherogenesis.[4−10]In vitro, Amadori products were reported with the ability to generate reactive oxygen species, which can further lead to lipid peroxidation involved in many physiological and pathological processes.[11]

To better understand the roles of Amadori-glycated lipids in physiological and pathological processes, it is essential to develop analytical methods for a comprehensive profiling of Amadori-glycated lipid species in biological samples. To date, reported methods using liquid chromatography coupled with mass spectrometric (LC-MS) technique have focused on Amadori-phosphatidyl ethanolamine (PE) with a few identified molecular species.[12−14] For lipidomic level investigation on Amadori-glycated lipids, LC-MS/MS methods with high sensitivity should be developed for measuring various types of Amadori lipids. To this end, it is necessary to obtain stable isotope-labeled Amadori-glycated lipid standards with high purity.

Several preparation methods for Amadori-PE were already reported. However, these methods are not well suited for preparing stable-isotope-labeled standards, in that either a complex process of six steps is used[15,16] or synthesis is carried out in methanol (MeOH) phosphate buffer (PB) medium, which requires a long reaction time (15 days) and a high amount of glucose (500 mM),[11,17] the latter is prohibitive for isotope-labeled Amadori compound synthesis due to the high cost of [13C6]glucose. A revised method using MeOH medium was also reported for the synthesis of Amadori-PE,[17−19] which requires a shorter reaction time (12 h) and less amount of glucose (100 mM). In spite of the methods mentioned above, no data are available about how yields change under various conditions, such as temperature, reaction duration, the ratio of reactants, and reaction medium. Considering the high cost of [13C6]glucose, optimal conditions are required for the preparation of [13C6]Amadori-PE standards. In addition, to our knowledge, preparation of Amadori-phosphatidyl serine (PS), -LysoPE, and -LysoPS has not been reported yet. Herein, we present optimized methods for the preparation of [13C6]Amadori-PE, -PS, -LysoPE, and -LysoPS with high purity. The identification of these synthetic [13C6]Amadori-lipid standards was performed by LC-MS/MS analysis in both positive- and negative-ion modes, and distinctive ions were identified for [13C6]Amadori-modified lipids.

Materials and Methods

Chemical and Solvent

1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (PE(16:0/18:1(9Z))), 1,2-dipentadecanoyl-sn-glycero-3-phosphoethanolamine (PE(15:0/15:0)), 1-tridecanoyl-sn-glycero-3-phosphoethanolamine (LysoPE(13:0)), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-l-serine (PS(16:0/18:1(9Z))), 1,2-diheptadecanoyl-sn-glycero-3-phospho-l-serine (PS(17:0/17:0)), and 1-tridecanoyl-sn-glycero-3-phospho-l-serine (LysoPS(13:0)) were purchased from Avanti Polar Lipids (Alabaster, AL). [U-13C6]-d-glucose (99%) was purchased from Cambridge Isotope Laboratories (Tewksbury, MA). Butylated hydroxytoluene 2,6-di-tert-butyl-p-cresol (BHT) and all solvents, including MeOH, acetonitrile (ACN), isopropanol (IPA), chloroform (CHCl3), and water (H2O), of LC-MS grade and high performance liquid chromatography (HPLC) grade were purchased from Fisher Scientific (Pittsburgh, PA).

Optimization of Synthetic Conditions

On the basis of the work of Miyazawa and co-workers,[11,17,18] we carried out a series of incubations to optimize conditions for the synthesis of [13C6]Amadori-glycated lipids.

To find which medium (MeOH or MeOH-PB) is better for the synthesis of [13C6]Amadori-lipids and whether BHT could increase the reaction yield, three incubations were performed as follows: (A) 7 μmol PE(16:0/18:1(9Z)) was mixed with 537 μmol [13C6]d-glucose in 5 mL of MeOH-0.1 M PB (1:2, v/v, pH 8) at 37 °C; (B) 7 μmol PE(16:0/18:1(9Z)) was mixed with 107 μmol [13C6]d-glucose in 1 mL of MeOH at 37 °C with and without 0.3 mg of BHT.

To further optimize the reaction temperature and time, 7 μmol PE(16:0/18:1(9Z)) was mixed with 107 μmol [13C6]d-glucose in 1 mL of MeOH and the mixture was stirred at 600 rpm in a Thermomix mixer (Eppendorf, Germany) at 37, 50, or 60 °C. At different time intervals (0–4 days), aliquots of reaction mixtures were collected for analysis. Incubations of [13C6]d-glucose with LysoPE(13:0), PS(16:0/18:1(9Z)), and LysoPS(13:0) were carried out similarly.

Optimization of reactant ratio was performed as follows: 7 μmol PE(16:0/18:1(9Z)) was incubated in 1 mL of MeOH at 60 °C with 54, 107, 161 and 215 μmol of [13C6]d-glucose.

Purification of Amadori-Glycated Lipids

Lipid Extraction

For [13C6]Amadori-PE/PS species, the Folch method was performed to remove [13C6]glucose and other water-soluble byproducts. After incubation, the reaction mixture in 1 mL of MeOH was mixed with 2 mL of CHCl3 and 0.6 mL of H2O. The mixture was vortexed for 20 s and then centrifuged at 3000 rpm for 15 min. The bottom layer was transferred out by a glass pipette and then dried with nitrogen.

C18 Solid-Phase Extraction (SPE)

Because of the high hydrophilicity of [13C6]Amadori-LysoPE/LysoPS species, the Folch method cannot be used for liquid–liquid extraction. Instead, C18 SPE was employed to separate synthetic products from [13C6]glucose or other water-soluble materials. In brief, incubation mixture was reconstituted in 10% MeOH and then loaded onto an ISOLUTE C18 cartridge (500 mg, Biotage, Sweden) conditioned with the same solvent. After rinsing with an additional 5 mL of 10% MeOH, Amadori lipids were eluted out with 10 mL of MeOH.

Phenylboronic Acid (PBA) SPE

Diol-containing Amadori-glycated lipids can bind to immobilized PBA, which enables isolation of Amadori-glycated lipids from their corresponding lipids. Bond Elut PBA cartridges (500 mg) purchased from Agilent (Palo Alto, CA) were adopted to perform SPE for cleanup. Briefly, PBA cartridges were wetted with 5 mL of MeOH and then conditioned with 3 mL of 150 mM ammonium formate (AF, pH 10), followed by 5 mL of 20% MeOH containing 100 mM AF (pH 8) (loading solvent). The dried resultant lipid extracts were redissolved in MeOH and diluted 10 times with loading solvent before loading onto equilibrated PBA cartridges. The cartridges were rinsed with 5 mL of MeOH containing 0.1% (v/v) ammonium hydroxide, then [13C6]Amadori-lipids were recovered by elution with 5 mL of 90% MeOH containing 1% formic acid. The elution effluent was nitrogen-dried and subjected to the following HPLC separation.

HPLC-UV Separation

Semipreparative separation was performed on a Shimadzu 20A HPLC equipped with LC-8A pumps. The samples were loaded onto a Kromasil C18 column (250 × 10 mm, 10 μm, AkzoNobel, Bohus, Sweden) using four isocratic mobile phases at a flow rate of 3 mL/min for different [13C6]Amadori-lipid species: [13C6]Amadori-PE: 100% MeOH containing 5 mM AF; [13C6]Amadori-PS: 100% MeOH containing 5 mM AF and 0.1% phosphoric acid; [13C6]Amadori-LysoPE: 80% MeOH containing 5 mM AF; and [13C6]Amadori-LysoPS: 75% MeOH containing 5 mM AF and 0.1% phosphoric acid. The effluent was monitored for UV absorbance at 220 nm, and the injection volume was set at 500 μL.

LC-MS Analysis of [13C6]Amadori-Lipids

A Vanquish UHPLC system coupled to a TSQ Quantiva triple quadrupole mass spectrometer (Thermo Fisher Scientific) was primarily used for the analysis of [13C6]Amadori-lipids. A core–shell Accucore C30 column (Thermo Fisher Scientific) was used with a column oven temperature of 40 °C and a flow rate of 350 μL/min. Two elution gradients were employed for the analysis of [13C6]Amadori-PE/PS and [13C6]Amadori-LysoPE/LysoPS, respectively, as follows:

Gradient 1: Identical to the method previously published by our laboratory for global lipidomics analysis,[20] the mobile phase was composed of solvent A (ACN/H2O, 60:40, v/v) and solvent B (IPA/ACN, 90:10, v/v), both containing 10 mM AF and 0.1% formic acid. The gradient was as follows: −3 to 0 min, 30% B for column equilibration; 0–5 min, 30–43% B; 5–5.1 min, 43–50% B; 5.1–14 min, 50–70% B; 14.1–21 min, 70–99% B; 21–24 min, 99% B; 24–24.1 min, 99–30% B; 24.1–28 min, 30% B for column reequilibration. The total analysis time including column reequilibration was 31 min.

Gradient 2: The mobile phase was composed of solvent A (H2O) and solvent B (MeOH), both containing 5 mM AF and 0.1% formic acid. The gradient was as follows: −3 to 0 min, isocratic elution with 60% B for the equilibration of the column; 0–3 min, 60% B; 3–13 min, 60–99% B; 13–15 min, 99% B; 15–15.1 min, 99–60% B; 15.1–19 min, 60% B for column reequilibration. The total analysis time including column reequilibration was 22 min.

The TSQ Quantiva mass spectrometer was equipped with a heated electrospray ionization source, with positive- and negative-ion spray voltages set at 3500 and 3000 V, respectively. For both ionization modes, nitrogen was used as the sheath, auxiliary, and sweep gases at flow rates of 20, 7, and 1 (arbitrary units), respectively. Vaporizer and ion-transfer tube temperatures were 400 and 350 °C, respectively. For MS analysis in full-scan mode, the scan range, scan rate, and resolution were m/z 200–1000, 1000 amu/s, and 0.7 (full width at half maximum), respectively. For MS/MS analysis in product ion scan mode, the argon collision gas pressure was 1.5 mTorr. The optimized collision energy was 20 eV for [13C6]Amadori-PE and 16 eV for [13C6]Amadori-PS/LysoPE/LysoPS in positive-ion mode, whereas it is 27 eV for [13C6]Amadori-PE/PS and 22 eV for [13C6]Amadori-LysoPE/LysoPS in negative-ion mode.

For high-resolution mass analysis, a Q Exactive HF hybrid quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific) was coupled to a Vanquish UPLC, as described above for TSQ Quantiva. Mass ranges for full-scan MS and mass resolution settings were m/z 100–1000 and 120 000, respectively.

Results and Discussion

Optimization of Synthetic Conditions

To determine the relative yield of [13C6]Amadori-glycated lipids, LC-MS analysis in full-scan mode was performed to quantify [13C6]Amadori-lipids synthesized under different conditions. An assumption was made that Amadori-glycated lipids have similar ionization efficiency to their corresponding unmodified lipids, so chromatographic peak areas were used to estimate the relative reaction yield.Reactions of PE(16:0/18:1(9Z)) with [13C6]d-glucose in MeOH-PB, MeOH, and MeOH with addition of BHT were carried out at 37 °C in parallel. After incubation for 0, 1, 2, 3, and 4 days, aliquots of reaction mixtures were dried and then subjected to Folch extraction. CHCl3 layer was transferred out, dried, reconstituted in MeOH, and then subjected to LC-MS analysis.

As shown in Figure , reaction yield in complete MeOH medium was 3.2, 4.9, 5.8, and 7.0 times higher than those in MeOH-PB after incubation for 1, 2, 3, and 4 days, respectively. On the other hand, yields in MeOH were 1.3, 1.4, 1.3, and 1.4 times higher than those in MeOH + BHT after 1, 2, 3, and 4 days, respectively. BHT, a synthetic antioxidant that can effectively inhibit the browning reaction and lipid peroxidation,[17,21,22] was expected to prevent the formation of secondary byproducts and thus increase the recovery of destination product; however, our results showed that addition of BHT did not help the production of [13C6]Amadori-PE.

Figure 1

Reaction yields of [13C6]Amadori-PE(16:0/18:1(9Z)) in the medium of MeOH-PB, MeOH, or MeOH in the presence of BHT.

In terms of effects of reaction temperature and time, as shown in Figure , yields peaked at 60 °C and 3 days for [13C6]Amadori-PE, 50 °C and 3 days for [13C6]Amadori-LysoPE, and 50 °C and 2 days for [13C6]Amadori-PS/LysoPS.

Figure 2

Effect of temperature and time on reaction yields of (A) [13C6]Amadori-PE(16:0/18:1(9Z)), (B) [13C6]Amadori-PS(16:0/18:1(9Z)), (C) [13C6]Amadori-LysoPE(13:0/0:0), and (D) [13C6]Amadori-LysoPS(13:0/0:0).

To optimize the ratio of reactants, incubations of 7 μmol PE(16:0/18:1(9Z)) in 1 mL of MeOH at 60 °C with 54, 107, 161, and 215 μmol [13C6]d-glucose were carried out in parallel, with the corresponding molar ratios of [13C6]glucose over PE(16:0/18:1(9Z)) being 7.7, 15.4, 23.1, and 30.9. After 3 days of incubation, the resulting yields were 19.9, 40.6, 49.9, and 53.0% for the molar ratios of 7.7, 15.4, 23.1, and 30.9, respectively. This result showed that yields increased with increasing molar ratio of [13C6]glucose over PE. However, the increase of yield was not significant when the molar ratio increased from 23.1 to 30.9. As such, molar ratios of [13C6]d-glucose over lipids were optimized to about 23 for the synthesis of all Amadori-glycated lipids. The optimized synthetic conditions for [13C6]Amadori-PE/PS/LysoPE/LysoPS are listed in Table .

Table 1

Optimized Synthetic Conditions for [13C6]Amadori-Lipids

	medium	temperature (°C)	incubation duration (days)	concentration of lipids (μmol/mL)	concentration of [13C6]d-glucose (μmol/mL)
[13C6]Amadori-PE	MeOH	60	3	7	161
[13C6]Amadori-PS	MeOH	50	2	7	161
[13C6]Amadori-LysoPE	MeOH	50	3	12	215
[13C6]Amadori-LysoPS	MeOH	50	2	10	215

Purification and Identification of [13C6]Amadori-Glycated Lipids

Purification methods differed between [13C6]Amadori-PE/PS, [13C6]Amadori-LysoPE, and [13C6]Amadori-LysoPS because of their different chemical properties. Figure shows the flowchart for purification of [13C6]Amadori-PE, -PS, -LysoPE, and -LysoPS, as synthesized in Section .

Figure 3

Outline of protocols for purification of [13C6]Amadori-lipids.

In terms of [13C6]Amadori-PE/PS species, we first removed [13C6]d-glucose from the incubation mixture by liquid–liquid extraction, for which the Folch method was employed. Diol-containing glucose has the ability to bind to immobilized PBA,[23,24] which will likely affect the binding between PBA and [13C6]Amadori-lipids. After lipid extraction, PBA SPE was performed to separate [13C6]Amadori-PEs from their corresponding substrates, according to the reversible covalent interaction between PBA and diol-containing [13C6]Amadori-PEs. [13C6]Amadori-PEs were finally purified by semipreparative HPLC. As for [13C6]Amadori-PS species, however, PBA SPE was not applicable because [13C6]Amadori-PSs could not be retained well on PBA SPE column. A similar poor retention was also observed for [13C6]Amadori-LysoPS. The possible reason is that [13C6]Amadori-PS/LysoPS carries negative charge under alkaline conditions, which results in electrostatic repulsion from immobilized boronate.[25] In contrast to [13C6]Amadori-PE, the separation performance of [13C6]Amadori-PS on C18 column was improved largely by adding acid to mobile phase. Consequently, [13C6]Amadori-PSs were purified on a semipreparative C18 column using isocratic elution with 100% MeOH containing 5 mM AF and 0.1% phosphoric acid, then one more Folch extraction was employed to remove phosphoric acid in eluent from purified compounds.

LC-MS full-scan chromatograms (Figure ) showed that [13C6]Amadori-PE(16:0/18:1(9Z)), -PE(15:0/15:0), -PS(16:0/18:1(9Z)), and -PS(17:0/17:0) were pure (>95%), with purification process recoveries of about 43, 51, 31, and 25%, respectively. LC-UV chromatograms (Supporting Information Figure S1) also validated the high purity of [13C6]Amadori-lipids.

Figure 4

LC-MS full-scan chromatograms of (A) blank control, and purified (B) [13C6]Amadori-PE(16:0/18:1(9Z)), (C) [13C6]Amadori-PE(15:0/15:0), (D) [13C6]Amadori-PS(16:0/18:1(9Z)), and (E) [13C6]Amadori-PS(17:0/17:0). The mass range for full-scan analysis was m/z 200–1000. The peak at 11.96 min appearing in all plots of this figure was from background.

Compared to [13C6]Amadori-PE/PS, [13C6]Amadori-LysoPE/LysoPS are more hydrophilic. Thus, different strategies were employed for purification of [13C6]Amadori-LysoPE/LysoPS. Instead of the Folch method, C18 SPE was performed to remove [13C6]d-glucose for [13C6]Amadori-LysoPE. In terms of [13C6]Amadori-LysoPS, besides its low recovery on PBA SPE, we found it was more unstable than the other species. To make the purification more efficient for [13C6]Amadori-LysoPS, direct purification using HPLC was employed, which utilizes isocratic elution with 75% MeOH containing 5 mM AF and 0.1% phosphoric acid. After HPLC preparation, phosphoric acid in the eluent was removed by C18 SPE. [13C6]Amadori-LysoPS was recovered in MeOH, nitrogen-dried, and kept at −80 °C. LC-MS full-scan chromatograms (Figure ) showed that [13C6]Amadori-LysoPE(13:0/0:0)/LysoPS(13:0/0:0) were pure (>95%), with purification process recoveries of about 48 and 19%, respectively. LC-UV chromatograms (Supporting Information Figure S2) also validated their high purity.

Figure 5

LC-MS full-scan chromatograms of (A) blank control, and purified (B) [13C6]Amadori-LysoPE(13:0/0:0) and (C) [13C6]Amadori-LysoPS(13:0/0:0). The mass range for full-scan analysis was m/z 200–1000.

The mass accuracy of the six purified [13C6]Amadori-lipids was confirmed by high-resolution mass spectrometry. A comparison between measured masses with calculated values exhibited a high mass accuracy (mass errors ranged from −2.6 to 2.4 ppm) (Supporting Information Table S1).

Isotopic purity of these six synthesized compounds was also investigated using high-mass-resolution LC-MS. Extracted ion chromatograms were generated and peak areas were integrated for each of the resolved isotopes related to these compounds (Supporting Information Figure S3). After subtracting the contribution from the non-fully labeled natural isotopes, peak area values of isotopes were used to calculate the relative percent of isotopic enrichment and the overall isotopic purity. As a result, these six synthesized compounds were found to be 93.67–94.3% fully enriched (containing six labeled carbons); and the overall isotopic purities ranged from 98.9 to 99.0%, showing that 98.9–99.0% of the potentially labeled positions in these compounds were occupied by 13C. This is in agreement with the 99% isotope purity of the starting material, [13C6]d-glucose.

Characteristic MS/MS Fragmentations of [13C6]Amadori-Lipids

To further verify the authenticity of products synthesized, MS/MS fragmentation in both positive- and negative-ionization modes was performed on a triple quadrupole mass spectrometer under collision-induced dissociation. First, Q1 mass spectrum confirmed the most abundant molecular ion peaks as [M + H]+ ions in positive-ion mode and [M – H]− ions in negative-ion mode (Supporting Information Figures S4 and S5). Next, product ion scanning was performed to investigate their fragmentation characteristics. In positive-ion mode, neutral losses of 309.1 Da (H2PO4C2H4NH13C6H11O5) and 353.1 Da (H2PO4C2H3COOHNH13C6H11O5) from [M + H]+ ion were observed for [13C6]Amadori-PE/LysoPE and [13C6]Amadori-PS/LysoPS, respectively, which were generated from the cleaved polar head groups (Figure ). It was notable that a neutral loss of 168.1 Da (isotope-labeled Amadori moiety, 13C6H10O5)—a characteristic neutral loss in MS/MS fragmentation of Amadori-peptides[1,26]—could also be observed but with a much lower intensity (Figure , inset). Meanwhile, neutral losses of H2O, 2H2O, 3H2O, and 3H2O + H13CHO could also be observed for all six synthetic [13C6]Amadori-lipids. In negative-ion mode, fragment ions generated from the neutral loss of 168.1 Da became dominant for [13C6]Amadori-PE/LysoPE species (Figure A,B,E), and could be clearly observed (10–12% relative intensities) for [13C6]Amadori-PS/LysoPS (Figure C,D,F). The base peaks in the MS/MS negative spectra of [13C6]Amadori-PS/LysoPS belong to fragment ions generated from the neutral loss of 255.1 Da ([PA/LPA – H]−). Furthermore, other characteristic ions related to fatty acyl chains of synthetic compounds and fragment ions of glycerophosphate groups could also be identified from the MS/MS negative spectra in Figure . These characteristic ions are summarized in Table , which are consistent with the fragmentation patterns of phospholipids previously reported.[27,28] Representative fragmentation schemes of these synthetic [13C6]Amadori-phospholipids are shown in Figure , taking [13C6]Amadori-PE(16:0/18:1(9Z)) and [13C6]Amadori-PS(16:0/18:1(9Z)) as examples.

Figure 6

MS/MS spectra for molecular ion peaks of (A) [13C6]Amadori-PE(16:0/18:1(9Z)), (B) [13C6]Amadori-PE(15:0/15:0), (C) [13C6]Amadori-PS(16:0/18:1(9Z)), (D) [13C6]Amadori-PS(17:0/17:0), (E) [13C6]Amadori-LysoPE(13:0/0:0), and (F) [13C6]Amadori-LysoPS(13:0/0:0) in positive-ion mode. The inset shows the magnified m/z 550–750 region of (A).

Figure 7

MS/MS spectra of molecular ion peaks of (A) [13C6]Amadori-PE(16:0/18:1(9Z)), (B) [13C6]Amadori-PE(15:0/15:0), (C) [13C6]Amadori-PS(16:0/18:1(9Z)), and (D) [13C6]Amadori-PS(17:0/17:0), (E) [13C6]Amadori-LysoPE(13:0/0:0), and (F) [13C6]Amadori-LysoPS(13:0/0:0) in negative-ion mode.

Table 2

MS/MS Fragmentation Analysis in Negative-Ionization Mode for [13C6]Amadori-PE(16:0/18:1(9Z)), -PE(15:0/15:0), -PS(16:0/18:1(9Z)), -PS(17:0/17:0), -LysoPE(13:0/0:0), and -LysoPS(13:0/0:0)—Product Ions (m/z) and Their Related Structures (Listed in the First Column) after Collision-Induced Dissociation Using a Triple Quadrupole Mass Spectrometera

proposed product ions	[13C6]Am-PE (16:0/18:1(9Z))	[13C6]Am-PE (15:0/15:0)	[13C6]Am-PS (16:0/18:1(9Z))	[13C6]Am-PS (17:0/17:0)	[13C6]Am-LysoPE(13:0)	[13C6]Am-LysoPS(13:0)
M – H	884.7	830.7	928.7	930.8	578.4	622.4
M – H-13C6H10O5	716.6	662.5	760.6	762.7	410.3	454.3
fatty acid-H (sn-1)	255.4	241.3	255.4	269.3	213.3	213.2
fatty acid-H (sn-2)	281.3	241.3	281.4	269.3	NA	NA
LPE-H (sn-1)	452.3	438.2	NA	NA	410.2	NA
LPE-H2O-H (sn-1)	434.4	420.3	NA	NA	392.5	NA
LPE-H (sn-2)	478.2	438.2	NA	NA	NA	NA
LPE-H2O-H (sn-2)	460.2	420.2	NA	NA	NA	NA
PA-H	673.4	619.6	673.5	675.5	NA	NA
LPA-H (sn-1)	409.5	395.6	409.3	423.4	367.2	367.3
LPA-H2O-H (sn-1)	391.3	377.2	391.3	405.3	349.3	349.1
LPA-H (sn-2)	435.5	395.6	435.4	423.4	NA	NA
LPA-H2O-H (sn-2)	417.4	377.4	417.3	405.3	NA	NA
glycero phosphoethanolamine-H2O-H	196.4	196.1	NA	NA	196.2	NA
glycerophosphate-H2O-H	153.1	153.2	153.2	153.2	153.2	153.1

PE: phosphatidyl ethanolamine; PS: phosphatidyl serine; PA: phosphatidic acid; LPE: LysoPE; LPA: LysoPA; NA: not applicable.

Figure 8

MS/MS fragmentation schemes of (A) [13C6]Amadori-PE(16:0/18:1(9Z)) in positive-ion mode and (B) negative-ion mode, and (C) [13C6]Amadori-PS(16:0/18:1(9Z)) in negative-ion mode. The MS/MS spectra corresponding to (A), (B), and (C) are shown in Figures A and 7A,C.

Unlabeled analogues corresponding to these six [13C6]-labeled compounds were also synthesized and subjected to LC-MS/MS analysis. Each unlabeled analogue had the same retention time as its corresponding labeled compound under the same LC condition (Supporting Information, Figure S6). From their MS/MS spectra in positive-ion mode, expected neutral losses of 303.1 Da (H2PO4C2H4NHC6H11O5) and 347.1 Da (H2PO4C2H3COOHNHC6H11O5) from [M + H]+ ion could be observed for Amadori-PE/LysoPE and Amadori-PS/LysoPS, respectively (Supporting Information Figure S7). From their MS/MS spectra in negative-ion mode, the expected neutral loss of 162.1 Da (Amadori moiety) could also be observed for all of these six unlabeled compounds (Supporting Information Figure S8). Collectively, the expected fragmentation patterns and the corresponding mass shifts observed for all synthetic [13C6]Amadori-lipids unambiguously validated the identification of these novel compounds.

Conclusions

In this study, six stable isotope-labeled [13C6]Amadori-glycated lipids were synthesized, purified, and structurally confirmed using accurate mass measurement and tandem mass spectrometry, covering four types of lipids, including Amadori-PE, -PS, -LysoPE, and -LysoPS. Optimal conditions for the synthesis and purification of these four types of stable isotope-labeled Amadori-glycated lipids were obtained. MS/MS fragmentation in both positive- and negative-ion modes validated the identification of these novel synthetic compounds. The high purity (>95%) of these synthetic compounds was verified by both LC-MS full-scan analysis and LC-UV analysis. As a result, successful preparation of highly purified stable isotope-labeled Amadori-lipid standards makes it possible to develop targeted LC-MS/MS methods for accurate analysis of endogenous Amadori-glycated phospholipids from biological samples.