Targeted analysis of serum glycosphingolipids

Jinyong Kim¹, Seul Kee Byeon¹, Devin Oglesbee¹, Matthew J. Schultz¹, Dietrich Matern¹, Craig Butt², Paul RS Baker², and Akhilesh Pandey^1,3,4,

¹Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA; ²SCIEX, USA; ³Center for Individualized Medicine, Mayo Clinic, Rochester, MN, USA; ⁴Manipal Academy of Higher Education, Manipal Karnataka, India 
Published date: June 10, 2024

Abstract

In this technical note, the SCIEX 6500+ system was used to measure 84 molecular species of gangliosides and other glycosphingolipids in a targeted assay to analyze the ganglioside metabolome comprehensively. The detection and quantitation of glycosphingolipids has emerged as an important analytical technique due to the significant roles these lipids play in neuronal function, such as myelin maintenance and cell signaling^1-3. To better understand these compounds' physiological and pathophysiological roles, their metabolites must be accurately detected, quantitated, and put into metabolic context. As with all lipid classes, the sphingolipids comprise multiple molecular species that differ by carbon chain length and the position of double bonds. Glycosphingolipid analysis is further complicated by the addition of multiple sugar moieties attached to the sphingoid base4. Although LC-MS/MS methods have been developed to analyze gangliosides quantitatively, none can simultaneously target all glycosphingolipids and provide an integrated view of ganglioside metabolism.

The most efficient and accurate means of comprehensively analyzing glycosphingolipid molecular species is by LC-MS/MS using a targeted, scheduled multiple reaction monitoring (sMRM) scan mode. Here, the SCIEX 6500+ system was used to measure glycosphingolipids in human serum samples using an sMRM method covering 10 classes of gangliosides and 4 other classes of glycosphingolipids in a single run by leveraging the fast polarity switching of the instrument (Figure 1). These targets comprise the entire pathway of ganglioside metabolism, and the data generated can provide a metabolic “snapshot” of biological samples.

Key features of targeted analysis of glycosphingolipids
 

• The SCIEX 6500+ system, using a targeted sMRM method, a method was developed to detect and quantitate 84 glycosphingolipids in serum
  
  
• The rapid polarity switch time of the SCIEX 6500+ system (< 5 ms) allows for analysis in both the positive and negative ion modes in a single injection
  
  
• The comprehensive method uses a simple extraction protocol and a total analysis time of 15 min to facilitate high-throughput glycosphingolipid analysis

Introduction

Gangliosides are a sialylated glycosphingolipid comprising a hydrophobic ceramide lipid moiety and an oligosaccharide chain. The diverse sugar composition, the number of sialic acids, and the variations in the length of carbon chains within ceramide contribute to the formation of >200 distinct molecular species of gangliosides. These molecules form functional microdomains, known as lipid rafts, on cell membranes by associating laterally with sphingomyelin, cholesterol and proteins owing to the hydrophobic nature of their tails and their polar glycans. Gangliosides are predominantly found in human brain tissue, where they play vital roles in myelin maintenance, cell signaling and cell-cell interactions^1-3. 

Quantitative analysis of gangliosides and related molecules in blood, urine, cerebrospinal fluid, and other relevant samples can provide a sensitive and precise means to evaluate glycosphingolipid metabolism. Various LC-MS/MS methods have been developed to quantitatively determine ganglioside molecular species in various biological specimens, including plasma cells and tissues. . These methods often focus on measuring a limited number of specific ganglioside molecular species or are applied to matrices other than serum5-7. Furthermore, none of these LC-MS/MS methods were designed to target a full spectrum of glycosphingolipids simultaneously. To address these deficiencies, this study developed a multiplexed liquid chromatography-tandem mass spectrometry (LC-MS/MS) method targeting 84 individual species of gangliosides and other glycosphingolipids in serum. It targets 10 classes of gangliosides (GM1, GM2, GM3, GD1, GD2, GD3, GT1, GT2, GT3 and GQ1) and 4 classes of glycosphingolipids (GlcCer, LacCer, Gb3 and GA2). The presented method allows for monitoring the entirety of ganglioside metabolic pathways. 

Materials and methods

Materials: The following standards were purchased: GM3 (d18:1/18:0), GD3 (d18:1/18:0), GD1a (d18:1/18:0), GT1b (d18:1/18:0), Glucosyl ceramide (d18:1/16:0), Lactosyl ceramide (d18:1/18:0 and d18:1/24:0) and Gb3 (d18:1/18:0) from Avanti Polar Lipids (Alabaster, AL, USA); GM2 (d18:1/18:0) and GM1 (d18:1/18:0) gangliosides from Cayman Chemical (Ann Arbor, MI, USA); GD2 (d18:1/18:0) ganglioside from ChemCruz (Dallas, Texas, USA). Glycosphingolipids (Avanti Polar Lipids Inc., Alabaster, AL, USA) with odd-numbered or deuterated acyl chain were used as internal standards and were added to the serum lipid extract for MRM quantitation: GM3 (d18:1/18:0-d5), GM1 (d18:1/17:0), Glucosyl ceramide (d18:1/16:0-d3) and Lactosyl ceramide (d18:1/18:0-d3). High-performance liquid chromatography (HPLC)- grade solvents (water, acetonitrile, methanol, isopropanol and chloroform) were purchased from Fisher Scientific (Waltham, MA, USA).

Sample preparation: Human serum was obtained from apparently healthy human volunteers following the procedure outlined in the Mayo Clinic institutional review board (IRB) protocol #21-012890. For lipid extraction, 20 µl of serum was mixed with a deuterated standard mixture as an internal standard (IS; GM3-d3 d18:1/18:0, GlcCer-d3 d18:1/16:0 and Gb3-d3 d18:1/18:0), followed by addition of 200 µl of chloroform:methanol (1:2, v/v). The samples were vortexed briefly and incubated in a bath sonicator for 30 min. After centrifugation, the supernatant was collected and dried under pure nitrogen. The dried lipid extract was reconstituted in 20 µl of methanol and mixed before transferring to autosampler vials. A standard mixture was prepared by mixing 5 pmol/µl of the following standards - GM3 d18:1/18:0, GM2 d18:1/18:0, GM1 d18:1/18:0, GD3 d18:1/18:0, GD2 d18:1/18:0, GD1a d18:1/18:0, GT1b d18:1/18:0, GlcCer d18:1/16:0, LacCer d18:1/18:0, LacCer d18:1/24:0 and Gb3 d18:1/18:0. This standard mixture was diluted serially to generate a calibration curve by spiking into serum substitute matrix.

Chromatography: The targeted LC-MS/MS analysis was conducted on a SCIEX QTRAP 6500+ mass spectrometer connected to a SCIEX Eksigent Ekspert LC 400 with a microflow module (5 to 50 µl/min). Separation of gangliosides according to their number of sialic acid and carbon chain length in target lipids was performed using Waters BEH Phenyl-hexyl column (1.0 × 50 mm, 1.7 µm) with a flow rate of 20 µl/min for 15 min using a binary gradient of mobile phase A (water:acetonitrile (9:1, v/v) with 5 mM ammonium hydroxide and 1 mM ammonium formate) and mobile phase B (isopropanol:methanol:acetonitrile (7:1.5:1.5, v/v/v) with 5 mM ammonium hydroxide and 1 mM ammonium formate). Five microliters of samples were injected into the analytical column with 10% of mobile phase B for 1 min. After sample loading, mobile phase B was increased to 70% over 1 min, 100% over 4 min, and maintained at 100% for another 5 min. Thereafter, the mobile phase B was decreased to 10%, and the analytical column was reconditioned for 4 min., as shown in Table 1. 

Mass spectrometry: Glycosphingolipid detection was performed using a SCIEX QTRAP 6500+ system (SCIEX, USA). Multiple reaction monitoring (sMRM) analysis was performed in the positive and negative modes in a single experiment.. In all, 74 molecular ions were monitored for gangliosides (GM3, GM2, GM1, GD3, GD2, GD1, GT3, GT2, GT1 and GQ1) and other glycosphingolipid classes (GlcCer, LacCer, Gb3 and GA2). All targeted lipids were quantified by calculating the peak areas of extracted ion chromatograms, followed by normalization using the peak areas of internal standards. The compound parameter settings, including the precursor and product ion masses, the collision energy (CE), and the respective retention times (RTs), are listed in Table 2. The polarity in which each glycosphingolipid was measured is indicated by the sign associated with each CE value.

Data processing: LC-MS operation, data acquisition, and data processing were performed using SCIEX Analyst software (SCIEX, USA). The standard mixture spiked into a serum substitute was used to obtain compound-specific calibration curves from which the limits of detection (LOD) and quantification (LOQ) were calculated.

The linearity of the method was evaluated using calibrators across different concentration ranges. The linearity equation and coefficient of determination (r2) were determined based on a calibration curve. The limit of detection was calculated based on the standard deviation of the response (Sy) of the calibration curve and the slope of the calibration curve (S) at levels approximating the LOD according to the formula LOD = 3.3(Sy/S). The method reproducibility was assessed by spiking low and high concentration standards into 20 µl of serum substitute supplement. Extraction efficiency was evaluated by comparing the expected and measured MS response of the standards added to the serum substitute. Intra-assay and inter-assay reproducibility were evaluated by running the low and high-concentration samples five times daily over five days.

Results and Discussion

Development of multiplexed ganglioside and glycosphingolipid targeted method

Targeted analysis for serum gangliosides was performed in a 15 min LC-MS/MS method in the scheduled MRM (sMRM) mode. Figure 2 shows the principal biosynthetic pathways of gangliosides and other glycosphingolipids. This assay was developed to target all these components to comprehensively analyze the ganglioside metabolome. This multiplexed method applied to the standard mixture and to human serum quantified a total of 84 species of GM3, GM2, GM1, GD3, GD2, GD1, GT3, GT2, GT1 and GQ1 gangliosides and other glycosphingolipids including GlcCer, LacCer, Gb3 and GA2. Using a phenyl-hexyl HPLC column, the gangliosides were separated according to their number of sialic acids, and within a specific class of ganglioside, gangliosides possessing different ceramide carbon chain lengths were separated8. The performance of the phenyl-hexyl column is demonstrated by the separation of the mixture of six gangliosides and four glycosphingolipid standards (GM1, GM2, GM3, GD1, GD2, GD3, LacCer, Gb3 and GA2) as shown in Figure 1A, which were obtained with 1 pmol of each standard. Figure 1B shows the extracted ion chromatograms obtained from lipid extracts of human serum samples by targeting all species of gangliosides and glycosphingolipids, as depicted in Figure 2. The elution map of all targeted species detected in normal serum samples is shown in Figure 3A. Gangliosides belonging to classes with a higher sialic acid content, GQ1, eluted first, while gangliosides with a single sialic acid, such as GM1, GM2 and GM3, eluted the latest among the gangliosides. Glycosphingolipids exhibited minor variations in retention times based on the number of glycans bound to ceramide. The dehydrated sialic acid fragment (m/z 290.1) was observed as the most dominant product ion when gangliosides were fragmented in the negative ion mode (Figure 3B). Thus, the fragment ion at m/z 290 was used as the product and quantifier ion for all GM3, GM2, GM1, GD3, GD2, GD1, GT3, GT2, GT1 and GQ1 gangliosides detected in negative ion mode. Glycosphingolipids other than gangliosides were detected in the positive ion mode using the fragment ion corresponding to the dehydrated sphingosine backbone (m/z 264.3) as the product and quantifier ion. The standard mixture was spiked into a serum substitute matrix9 containing human serum albumin and globulins, but not lipids, to generate a calibration curve and determine the assay’s quantitative statistics (Table 3). The calibration standards' performance at various concentrations was assessed by calculating the coefficients of determination (r2) from three replicate runs. All r2 values were above 0.99, with a median of 0.994 (Table 3). Intra-assay precision showed <10% coefficient of variation for low and high-concentration mixtures. Inter-assay precision was assessed by measuring samples daily for 5 days and was determined to be <18% for both low and high QC mixtures. To evaluate the efficiency of a single-phase extraction using chloroform:methanol (1:2, v/v) as an extraction solvent, the peak areas of analytical standards were compared before and after extraction when added to a serum substitute. The average extraction efficiency for all standards was above 90%, with a median of 97.3% (data not shown).

Endogenous targets measured in samples from human serum samples

As shown in Figure 1B, the multiplexed method was used to target and analyze various ganglioside and glycosphingolipid species in lipid extracts from serum. Using the calibration curves derived from the standard mixture spiked into serum substitute (as indicated in Table 3), we quantified the endogenous targets detected in the lipid samples from the five individuals, as shown in Table 4. Quantitation of the total 7 classes of gangliosides (d36:1) revealed that GM3 ganglioside was the most abundant among the gangliosides present in the serum samples of five individuals (average 444.3 fmol). Following GM3, GD3 and GM2 were the subsequent most abundant gangliosides (average 241.2 and 150.4 fmol, respectively, across the five individuals). These data show that this targeted method for glycosphingolipid analysis is suitable for use with biological samples to generate quantitative data.

In summary, a robust multiplexed targeted LC-MS/MS method to simultaneously quantify the 84 molecular species of 10 ganglioside classes (GM1, GM2, GM3, GD1, GD2, GD3, GT1, GT2, GT3, and GQ1) and 4 classes of other glycosphingolipids (GlcCer, LacCer, Gb3, and GA2) in human serum has been developed. Using this multiplexed method, 61 species of gangliosides and glycosphingolipids species that were targeted were detected in human serum samples. Additionally, by testing serum samples, the total amounts of endogenous gangliosides from 7 different classes (GM1, GM2, GM3, GD1, GD2, GD3, GT1) were quantitated using the calibration curves generated from ganglioside standards spiked into a serum substitute matrix. 

Conclusion
 

• The SCIEX 6500+ system has the speed and sensitivity to quantitatively measure 84 glycosphingolipid molecular species in serum to provide a comprehensive understanding of ganglioside metabolism
  
  
• Taking advantage of the fast polarity switching of the instrument (< 5ms), all the molecules representing the ganglioside metabolome can be measured in a single run with a total run time of 15 min

References
 

1. Sipione S, Monyror J, Galleguillos D, Steinberg N, Kadam V. Gangliosides in the Brain: Physiology, Pathophysiology and Therapeutic Applications. Front Neurosci. 2020 Oct 6;14:572965. doi: 10.3389/fnins.2020.572965. PMID: 33117120
   
   
2. Yu RK, Tsai YT, Ariga T. Functional roles of gangliosides in neurodevelopment: an overview of recent advances. Neurochem Res. 2012 Jun;37(6):1230-44. doi: 10.1007/s11064-012-0744- y. Epub 2012 Mar 13. PMID: 22410735
   
   
3. Palmano K, Rowan A, Guillermo R, Guan J, McJarrow P. The role of gangliosides in neurodevelopment. Nutrients. 2015 May 22;7(5):3891-913. doi: 10.3390/nu7053891. PMID: 26007338
   
   
4. Ledeen R, Wu G. Gangliosides of the Nervous System. Methods Mol Biol. 2018;1804:19-55. doi: 10.1007/978-1-4939-8552- 4_2. PMID: 29926403
   
   
5. Gu J, Tifft CJ, Soldin SJ. Simultaneous quantification of GM1 and GM2 gangliosides by isotope dilution tandem mass spectrometry. Clin Biochem. 2008 Apr;41(6):413-7. doi: 10.1016/j.clinbiochem.2007.12.026. Epub 2008 Jan 17. PMID: 18241673
   
   
6. Fuller M, Duplock S, Hein LK, Rigat BA, Mahuran DJ. Liquid chromatography/electrospray ionization-tandem mass spectrometry quantification of GM2 gangliosides in human peripheral cells and plasma. Anal Biochem. 2014 Aug 1;458:20- 6. doi: 10.1016/j.ab.2014.04.018. Epub 2014 Apr 24. PMID: 24769373
   
   
7. Sarbu M, Raab S, Henderson L, Fabris D, Vukelić ZŽ , Clemmer DE, Zamfir AD. Cerebrospinal fluid: Profiling and fragmentation of gangliosides by ion mobility mass spectrometry. Biochimie. 2020 Mar;170:36-48. doi: 10.1016/j.biochi.2019.12.008. Epub 2019 Dec 17. PMID: 31857120
   
   
8. Ashta Lakshmi Prasad Gobburi, Eric Wekesa Kipruto, Denise M. Inman & David J. Anderson (2021): A new LC-MS/MS technique for separation of gangliosides using a phenyl-hexyl column: Systematic separation according to sialic acid class and ceramide subclass, Journal of Liquid Chromatography & Related Technologies, DOI: 10.1080/10826076.2020.1856136
   
   
9. Weathersbee PS, Pool TB, Ord T. Synthetic serum substitute (SSS): a globulin-enriched protein supplement for human embryo culture. J Assist Reprod Genet. 1995 Jul;12(6):354-60. doi: 10.1007/BF02215726. PMID: 8589555