Tailored Extraction Procedure Is Required To Ensure Recovery of the

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Technical Note

A tailored extraction procedure is required to ensure recovery of the main lipid classes in whole blood when profiling the lipidome of dried blood spots. Juan Jose Aristizabal Henao, Adam H Metherel, Richard W. Smith, and Ken D Stark Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03030 • Publication Date (Web): 30 Aug 2016 Downloaded from http://pubs.acs.org on September 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

A tailored extraction procedure is required to ensure recovery of the main lipid classes in whole blood when profiling the lipidome of dried blood spots.

Juan J. Aristizabal Henao1, Adam H. Metherel1, Richard W. Smith2, Ken D. Stark1* 1

Department of Kinesiology, 2University of Waterloo Mass Spectrometry Facility, Department of

Chemistry, University of Waterloo, Waterloo, ON, Canada

*Correspondence to: Dr. Ken D. Stark Canada Research Chair in Nutritional Lipidomics Associate Professor Department of Kinesiology University of Waterloo Waterloo, ON, Canada N2L 3G1 +1 (519) 888 4567 ext. 37738 email: [email protected]

Abbreviations: Dried blood spot (DBS), Whole blood (WB), Ultra-High Performance Liquid Chromatography (UHPLC), Tandem Mass Spectrometry (MS/MS).

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT The use of dried blood spots has increased in research and clinical settings recently, particularly in field studies and screening, but comprehensive acyl-specific lipidomic profiling of dried blood spots has yet to be examined. An untargeted ultra-high performance liquid chromatographytandem mass spectrometry method was adapted for the analysis of lipid extracts from human whole blood samples and dried blood spots collected on chromatography paper. Lipid recoveries were examined after; different durations of exposure to extraction solvents (chloroform/methanol), physical disruption (homogenization or sonication) of the paper containing the dried blood spots, and acidification of extraction solvents. We demonstrated that comprehensive untargeted profiles can be obtained from dried blood spot samples that are comparable with whole blood for several species of lipids including phosphatidylcholine, lysophosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, triacylglycerol and cholesteryl ester. However, homogenization of the dried blood spots, followed by a 24h exposure to solvents, and extraction with an acidic buffer (0.2M NaHPO4 + 0.1M hydrochloric acid) was required. Dried blood spots can be used for comprehensive, untargeted lipidomics of the most abundant lipid species in whole blood, but additional sample processing steps are required.


Page 2 of 20

Page 3 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. INTRODUCTION Traditional fatty acid profiling techniques have provided insights with regards to lipid metabolism in vivo and in vitro, and have been used to support relationships between dietary fat intakes and disease risk1. Blood biomarkers such as the Omega-3 Index2 are commonly used to assess the dietary intake of omega-3 polyunsaturated fatty acids, but these biomarkers are based on fatty acyl chains that have been isolated from their complex “parent” lipids3. Lipidomic analysis would preserve the information about the original acyl-containing parent lipid and potentially enhance and expand the utility of fatty acyl-based lipid biomarkers for assessing diet and health. However, high throughput strategies that have been employed for fatty acid analysis need to be applied to lipidomics. Dried blood spot (DBS) sampling is used regularly to screen for congenital diseases such as phenylketonuria, cystic fibrosis and sickle cell disease4-6. The convenient collection process has led to DBS sampling being used increasingly in field studies for various blood measurements7,8 and the use of DBS in lipid research is increasing9,10. While there is a substantial body of research adapting DBS for fatty acid profiling using gas chromatography3,7,8,11-14, we are aware of only two reports of the use of DBS for lipidomic profiling by high-performance liquid chromatography and mass spectrometry15,16. These previous investigations report only selected complex lipids and other complex lipids known to be in whole blood have been ignored. In fact, lipidomic profiling of whole blood in general is understudied as most of the available literature on lipidomic profiling of human blood examines plasma17-20, erythrocytes20-22 or leukocytes/platelets22-24. Whole blood analysis is challenging as the lipidome is a complex mixture of both nonpolar and polar lipids with ranges of abundance and a comprehensive comparison of the lipidome of whole blood and DBS is required.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In order to comprehensively compare the whole blood and DBS lipidomes, we developed an untargeted lipidomic profiling method to separate several of the main lipid classes in blood including phosphatidylcholines, phosphatidylethanolamines, phosphatidylserines, triacylglycerol and cholesteryl esters using UHPLC programming and detection by MS/MS. This included assessing the linear dynamic range using serial dilutions of lipid extracts from whole blood25. With this method, the extraction yield of lipids from DBS were compared to yields from whole blood and extraction protocols were modified by increasing the time of exposure to extraction solvents (chloroform/methanol), physically disrupting the paper containing the dried blood spots by sonication or homogenization, and altering the pH of the of extraction procedure. This work indicates comprehensive, untargeted lipidomic profiling of DBS is possible in a single UHPLCMS/MS run in positive mode, but extra sample processing steps beyond simple exposure to chloroform/methanol are required to recover all lipid classes.


Page 4 of 20

Page 5 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2. EXPERIMENTAL SECTION 2.1 Chemicals Acetonitrile (HPLC grade), isopropanol (HPLC grade), chloroform (HPLC grade), methanol (HPLC grade), hexane (HPLC grade), monobasic sodium phosphate, hydrochloric acid (optima grade), formic acid, ammonium formate and Whatman cellulose chromatography paper were purchased from Thermo-Fisher Scientific (Napean, ON, Canada). Phospholipid standards were purchased from Avanti Polar Lipids (Alabaster, AL, USA).

2.2 Serial Dilutions of Whole Blood Lipid Extracts Whole blood of a single 23-year-old male participant was collected to examine the relationship between increasing blood lipid concentrations and instrument response. A series of dilutions were prepared by extracting lipids from seven 25µL-aliquots of whole blood and diluting in chloroform to represent 25.0µL, 20.0µL, 15.0µL, 10.0µL, 5.0µL, 2.0µL and 1.0µL of blood. All samples were spiked with 1,2-diheptadecanoyl-sn-glycero-3-phosphatidylcholine (17:0/17:0 PC) as the internal standard to account for inter-sample variability. Samples were dried under a stream of nitrogen gas, re-suspended into 100µL 65:35:5 acetonitrile:isopropanol:water + 0.1% formic acid and stored at 4ºC until analysis. Since different lipid species exhibit different ionization efficiencies in electrospray ionization mass spectrometry25, the data generated from these analyses which used a single internal standard is not quantitative (i.e. cannot compare abundances of phosphatidylcholines vs. phosphatidylethanolamines). However, the objective here was to compare the abundances of specific lipid species across dilution levels within lipid classes, and not between them. We therefore consider our data semi-quantitative, and values are presented as peak area ratios of analyte/internal standard. Collection of blood from human



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

participants in these studies received clearance from the University of Waterloo Clinical Research Ethics Committee.

2.3 Dried Blood Spotting compared with Whole Blood To compare DBS with whole blood, venous blood was collected from 4 volunteer participants into ethylene diamine tetraacetic acid-lined vacutainers (BD Canada, Mississauga, ON). Whole blood was stored in 1.0mL aliquots in cryovials at -80ºC until analysis and used to prepare DBS samples. For DBS, 25µL of fresh whole blood was applied onto 1.0cm2 strips of Whatman cellulose chromatography paper and air-dried for 30 minutes11 before storing at -80ºC until analysis. Technical replicates from one participant were prepared for the initial extraction assessment experiments, and findings were then confirmed in a small group of biological replicates (n=3). Ad hoc experiments to increase recoveries of specific complex lipid classes were completed using technical blood replicates from a single participant.

2.4 Lipid Extraction Experiments Total lipid extracts were prepared using a modified Folch protocol26. In brief, 25µL aliquots of whole blood and DBS were each extracted with 3mL of 2:1 chloroform:methanol (v/v). To examine the effect of extraction time, a sample set (n=3 each) were processed immediately (hereafter referred to as WBi and DBSi), while another set (n=3 each) were left on chloroform/methanol for 24h (WB24 and DBS24). Extractions for additional experiments would use the 24h incubation. Samples were extracted by vortexed vigorously for 1 minute, adding 500µL of 0.2M NaHPO4 in ddH2O (pH = 4.4), inverting twice and then centrifuging at 1734 rcf for 5 minutes to induce phase separation. The organic layer was collected and an additional 2 mL


Page 6 of 20

Page 7 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of chloroform was added to the aqueous portion for a second round of extraction. The second organic collection was added to the first, dried down under nitrogen gas and reconstituted into 0.25 mL of chloroform. A 100 µL aliquot of total lipid extracts (representing 10 µL of blood) were spiked with 500 pmol 17:0/17:0 PC, dried under nitrogen gas and re-suspended into 100 µL 65:35:5 acetonitrile:isopropanol:water + 0.1% formic acid. Samples were then stored at 4ºC until analysis. Preliminary results (in detail below) indicated additional experiments were required to improve the recovery of phosphatidylserines from DBS. Acid-treatment of chromatography paper prior to the application of blood, acidification of extraction solvents, and homogenization or sonication of the paper were examined. Chromatography paper strips were acidified with 50 µL of 0.1M HCl, with 1h air-drying prior to blood application. The extraction buffer was acidified with 0.1M HCl in 0.2M NaHPO4 in ddH2O (pH = 2.3). Homogenization of DBS on paper was performed using a Kinematica PT 1200E Polytron (Kinematica Inc., Bohemia, NY, USA) at power level 75% continuously in the 2:1 chloroform:methanol. Sonication of DBS on paper was performed using a Qsonica Q55 Sonicator (Qsonica LLC., Newtown, CT, USA) at a frequency of 20 kHz and amplitude of 75% continuously for 60 seconds in the 2:1 chloroform:methanol while in an ice bath.

2.5 Ultra-High Performance Liquid Chromatography and Mass Spectrometry Analyses were completed on a Thermo Q-Exactive Quadrupole-Orbitrap Mass Spectrometer (Thermo-Fisher Scientific, Waltham, MA, USA) coupled to a Dionex UltiMate 3000 UPLC System (Dionex Corporation, Bannockburn, IL, USA). A reversed-phase, binary multi-step HPLC protocol was used27-29, with modifications to increase the number of lipid classes



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

identified in whole blood. Specifically, a 15cm x 2.1mm x 2.0µm column (C18 Ascentis Express, Sigma-Aldrich, St. Louis, MO) was adapted to allow UHPLC. Flow was kept at 0.26mL/min, which resulted in instrument pressures to rise to 550 bar. The mobile phase consisted of A: 60:40 acetonitrile:water +0.1% formic acid +10 mM ammonium formate, and B: 90:10 isopropanol:acetonitrile +0.1% formic acid +10 mM ammonium formate. for better chromatographic separation of phospholipids, the polar mobile phase step was extended. The final gradient protocol was: 32% B for 0 – 1.5 minutes, 45% B for 1.5 – 4 min, 50% B for 4 – 8 min, 55% B for 8 – 18 min, 60% B for 18 – 20 min, 70% B for 20 – 35 min, 95% B for 35 – 40 min, 95% B for 40 – 45 min, decreased to 32% B for 45 – 47 and allowed to equilibrate until the 48-minute mark. Column temperature was set at 45°C, and tray temperature at 4°C. The mass spectrometer was operated in positive electrospray ionization mode, 35,000 resolution, with scan range m/z 200 – 2000, spray voltage +3.0kV, sheath gas flow rate 35 units, and capillary temperature 300ºC. MS/MS experiments were done under data-dependent conditions with top 5 ions, 17,500 resolution, and the normalized collision energy was 17.5. Di-isooctyl phthalate (m/z 391.28429) was used as the lock mass. The Thermo Xcalibur QualBrowser software (Version 2.1; Thermo-Fisher Scientific, Waltham, MA, USA) was used for extracting ion chromatograms, integrating peak areas and exporting MS/MS spectra. Chromeleon Xpress (Version 7.2; Thermo-Fisher Scientific, Waltham, MA, USA) was used to monitor and control the Dionex UHPLC settings. MS/MS spectra were analyzed using the NIST MS Search Program (Version 2.0; National Institute of Standards and Technology, Gaithersburg, MD, USA) and LipidBlast libraries 30.

2.6 Data Normalization and Statistical Analyses


Page 8 of 20

Page 9 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Data are expressed as peak area ratios of analyte/internal standard. Comparisons of whole blood and DBS lipidomics for technical replicate extraction experiments and phosphatidylserine recovery experiments were assessed by one-way ANOVA with Tukey post-hoc test (significance was inferred as p < 0.05). Biological replicate experiments were assessed using two-tailed Student’s t-tests (significance was inferred as p < 0.05). SPSS for Windows statistical software (release 11.5.1; SPSS Inc., Chicago, IL, USA) was used. Data are presented as the mean ± standard deviation as indicated.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

3. RESULTS AND DISCUSSION 3.1 Serial Dilutions of Whole Blood Lipid Extracts The UHPLC-MS/MS method with an extended gradient protocol separated the major lipid classes in whole blood including phosphatidylcholines, phosphatidylethanolamines, phosphatidylserines, triacylglycerol and cholesteryl esters between 13-37 minutes (data not shown). Highly abundant phosphatidylcholine, triacylglycerol and cholesteryl ester species in blood were identified, with the predominant species being 16:0/18:2 phosphatidylcholine (m/z 758.5694), 16:0/18:1/18:2 triacylglycerol (m/z 874.7851), and 18:2 cholesteryl ester (m/z 666.6183). Positive linear relationships between increasing lipid concentrations and instrument response were observed from 1.0uL up to 25uL of blood (Figure 1; R2 > 0.95 for all). Subsequent experiments extracted 25µL of blood either in liquid form or as DBS, but only 40% of the lipid extracts (representing 10µL blood) were aliquoted for UHPLC-MS/MS analyses to stay well within the linear dynamic range of the instrument.

3.2 Lipidomics of Whole Blood and Dried Blood Spots from Technical Replicates and the Effect of Extraction Time Eleven representative acyl species of lipids were examined that included phosphatidylcholine (16:0 lyso, 16:0/18:2, 16:0/20:5, 16:0/22:6, 18:0/20:5, 18:0/22:6), phosphatidylethanolamine (16:0/18:2, 16:0/22:6), phosphatidylserine (18:0/22:6), triacylglycerol (16:0/18:1/18:1) and cholesteryl ester (18:2). There were no significant differences in recoveries between WBi, WB24 and DBS24, but DBSi recoveries were approximately 50% lower (Table 1). The larger contact area of the extraction solvents with whole blood compared with DBS may explain this observation as whole blood was added to chloroform/methanol as liquid the dispersed throughout


Page 10 of 20

Page 11 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the solvents while DBS samples remain constrained within a ~1 cm2 piece of chromatography paper. Increasing solvent exposure to DBS samples to 24h increased the recoveries of 10 of the representative lipids, but interestingly 18:0/22:6 phosphatidylserine recoveries in DBS24 remained significantly lower as compared with recoveries from WBi and WB24 (Table 1, p < 0.05). Since these results were based on technical replicates of whole blood from a single volunteer, we sought to confirm these observations in different individuals (biological replicates) and examine additional acyl species of lipids.

3.3 Lipidomics of Whole Blood and Dried Blood Spots from Biological Replicates The lipidomic profiles of whole blood and DBS from three individuals were generated following the 24h extraction. The 11 representative acyl lipid species and 14 additional acyl lipid species were examined (Table 1). Notably, there were no significant differences found between the WB24 and DBS24 for the biological replicates except for approximately 55-65% lower recoveries of the 4 phosphatidylserine species in DBS24. This supports the observations from the technical replicate experiments in section 3.2 and indicated this was common to phosphatidylserine species in general and independent of fatty acyl chain composition. This behavior can be explained by the acidic nature of phosphatidylserine headgroup31. While glycerophospholipids share common features such as a glycerol backbone, two esterified fatty acyl constituents at the sn-1 and sn-2 positions, and a phosphate group at the sn-3 position with at least one free hydroxyl group, residues such as choline, ethanolamine, serine, inositol and glycerol on the phosphate group result in different physical and chemical properties32. Serine residues contain a carboxylic acid group that makes phosphatidylserine species uniquely acidic at physiological pH31. Hydrogen bonding between acidic compounds and the cellulose fibers within the



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

chromatography paper used for DBS is known to occur33, and therefore may limit the ability of the solvents to extract phosphatidylserine species. We examined several approaches in order to address this issue, including acidifying the lipid extraction protocol which has been shown to result in higher recoveries of polar lipids34, as well as physical disruption of the paper using homogenization35 or sonication36.

3.4 Optimizing the Extraction Protocol to Increase Phosphatidylserine Recoveries from Dried Blood Spots All the phosphatidylserine acyl species examined behaved similarly. Combining homogenization with acidified extraction of DBS resulted in recoveries of phosphatidylserine species similar to whole blood extraction and significantly higher than simple DBS24 (Figure 2, p < 0.05). Homogenization of the DBS paper increased recoveries above the simple 24h extraction of DBS, but levels were lower than WB24 and the homogenization combined with acidified solvents. The use of acidified solvents alone (data not shown) resulted is a slight increase in recoveries over simple DBS24 but were much lower than homogenized DBS24. Acid-treating the paper prior to collecting the DBS did increase phosphatidylserine recoveries but this approach was not as effective as the acidification of the solvents (data not shown). The acidification of solvents also has a significant practical advantage of not interfering sample collection protocols in clinical settings where the DBS samples are being used for multiple biological markers and metabolites. Sonication was also examined as a alternate for homogenization as it has been shown to be effective in increasing the extraction of lipids from solid matrices36 and it can be applied remotely or with simple probes. Sonication alone gave similar recoveries to homogenization alone, but sonication in combination with acid solvents


Page 12 of 20

Page 13 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


resulted in lower recoveries than homogenization with acid solvents. Sonicating the samples for longer than 60 seconds or at a higher frequency may help in further increasing lipid yields and should be explored further, but sample cooling may be necessary.

5. CONCLUSION We demonstrated that DBS sampling has the potential to be used in comprehensive lipidomics screening but lipid extraction methods should be considered. Although extending a simple Folch extraction to 24h can provide accurate lipidomic profiles from DBS for phosphatidylcholine, phosphatidylethanolamine, triacylglycerol and cholesteryl ester species, additional steps must be taken in order to efficiently extract the phosphatidylserine species. Specifically, samples must be homogenized and the extraction buffer must be acidified. Further work is necessary in examining the behavior of lipid molecules that are found in low abundance in blood but are highly bioactive and of clinical interests, such as free fatty acids, eicosanoids, resolvins and steroid hormones 37,38

. The use of appropriate internal standards such as stable isotopes or standard dilution curves

may enable the absolute quantitation of acyl-specific lipids in DBS with the extraction and UHPLC-MS/MS conditions proposed herein. In conclusion, lipidomic screening to identify the fatty acyl species of the major lipid classes in blood can be completed with DBS however, extra sample preparation steps are necessary for comprehensive recoveries across all lipid classes.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACKNOWLEDGEMENTS The authors acknowledge support for this work by the Natural Sciences and Engineering Research Council of Canada and the Canada Foundation for Innovation. JJAH is supported by an NSERC Post-Graduate Doctoral Scholarship. AHM was supported by a Mitacs Elevate Industrial Postdoctoral Fellowship. KDS is supported by a Canada Research Chair in Nutritional Lipidomics. The authors do not report any conflicts of interest, and thank Sigma-Aldrich for providing a sample UHPLC column.

CONFLICT OF INTEREST DISCLOSURE The authors declare no conflicts of interest.


Page 14 of 20

Page 15 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


REFERENCES (1) Mozaffarian, D.; Lemaitre, R. N.; King, I. B.; Song, X.; Huang, H.; Sacks, F. M.; Rimm, E. B.; Wang, M.; Siscovick, D. S. Ann Intern Med 2013, 158, 515-525. (2) Harris, W. S.; Dayspring, T. D.; Moran, T. J. Postgrad Med 2013, 125, 100-113. (3) Metherel, A. H.; Aristizabal Henao, J. J.; Ciobanu, F.; Taha, A. Y.; Stark, K. D. Lipids 2015, 50, 895-905. (4) Green, A. Rinsho Byori 1998, 46, 211-216. (5) Health Quality, O. Ont Health Technol Assess Ser 2003, 3, 1-36. (6) Haynes, C. A.; De Jesus, V. R. Clin Biochem 2016, 49, 161-165. (7) Nurhasan, M.; Roos, N.; Henao, J. A.; Chamnan, C.; Stark, K.; Lauritzen, L. Prostaglandins, Leukotrienes and Essential Fatty Acids (PLEFA) 2015, 96, 57-61. (8) Lin, Y. H.; Hanson, J. A.; Strandjord, S. E.; Salem, N. M.; Dretsch, M. N.; Haub, M. D.; Hibbeln, J. R. Lipids 2014, 49, 839-851. (9) Lacher, D. A.; Berman, L. E.; Chen, T. C.; Porter, K. S. Clin Chim Acta 2013, 422, 54-58. (10) Turgeon, C. T.; Moser, A. B.; Morkrid, L.; Magera, M. J.; Gavrilov, D. K.; Oglesbee, D.; Raymond, K.; Rinaldo, P.; Matern, D.; Tortorelli, S. Mol Genet Metab 2015, 114, 46-50. (11) Metherel, A. H.; Aristizabal Henao, J. J.; Stark, K. D. Lipids 2013, 48, 1079-1091. (12) Metherel, A. H.; Stark, K. D. J Nutr 2015, 145, 654-660. (13) Marangoni, F.; Colombo, C.; Galli, C. Anal Biochem 2004, 326, 267-272. (14) Bailey-Hall, E.; Nelson, E. B.; Ryan, A. S. Lipids 2008, 43, 181-186. (15) Koulman, A.; Prentice, P.; Wong, M. C.; Matthews, L.; Bond, N. J.; Eiden, M.; Griffin, J. L.; Dunger, D. B. Metabolomics 2014, 10, 1018-1025. (16) Prentice, P.; Koulman, A.; Matthews, L.; Acerini, C. L.; Ong, K. K.; Dunger, D. B. J Pediatr 2015, 166, 276-281 e276. (17) Jove, M.; Naudi, A.; Portero-Otin, M.; Cabre, R.; Rovira-Llopis, S.; Banuls, C.; Rocha, M.; Hernandez-Mijares, A.; Victor, V. M.; Pamplona, R. FASEB J 2014, 28, 5163-5171. (18) Graessler, J.; Bornstein, T. D.; Goel, D.; Bhalla, V. P.; Lohmann, T.; Wolf, T.; Koch, M.; Qin, Y.; Licinio, J.; Wong, M. L.; Chavakis, T.; Xu, A.; Shevchenko, A.; Schuhmann, K.; Schwarz, P. E.; Schulte, K. M.; Patel, A.; Bornstein, S. R. Pharmacogenomics J 2014, 14, 201207. (19) Grassler, J.; Kopprasch, S.; Passauer, J.; Fischer, S.; Schuhmann, K.; Bergmann, S.; Siegert, G.; Shevchenko, A.; Bornstein, S. R.; Julius, U. Atheroscler Suppl 2013, 14, 151-155. (20) Holcapek, M.; Cervena, B.; Cifkova, E.; Lisa, M.; Chagovets, V.; Vostalova, J.; Bancirova, M.; Galuszka, J.; Hill, M. J Chromatogr B Analyt Technol Biomed Life Sci 2015, 990, 52-63. (21) Koehrer, P.; Saab, S.; Berdeaux, O.; Isaico, R.; Gregoire, S.; Cabaret, S.; Bron, A. M.; Creuzot-Garcher, C. P.; Bretillon, L.; Acar, N. PLoS One 2014, 9, e106912. (22) Leidl, K.; Liebisch, G.; Richter, D.; Schmitz, G. Biochim Biophys Acta 2008, 1781, 655664. (23) Duvernay, M. T.; Matafonov, A.; Lindsley, C. W.; Hamm, H. E. Biochemistry 2015, 54, 5578-5588. (24) Balgoma, D.; Astudillo, A. M.; Perez-Chacon, G.; Montero, O.; Balboa, M. A.; Balsinde, J. J Immunol 2010, 184, 3857-3865. (25) Koivusalo, M.; Haimi, P.; Heikinheimo, L.; Kostiainen, R.; Somerharju, P. J Lipid Res 2001, 42, 663-672. (26) Folch, J.; Lees, M.; Sloane Stanley, G. H. J Biol Chem 1957, 226, 497-509.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(27) Bird, S. S.; Stavrovskaya, I. G.; Gathungu, R. M.; Tousi, F.; Kristal, B. S. Methods Mol Biol 2015, 1264, 441-452. (28) Bird, S. S.; Marur, V. R.; Sniatynski, M. J.; Greenberg, H. K.; Kristal, B. S. Anal Chem 2011, 83, 940-949. (29) Hu, C.; van Dommelen, J.; van der Heijden, R.; Spijksma, G.; Reijmers, T. H.; Wang, M.; Slee, E.; Lu, X.; Xu, G.; van der Greef, J.; Hankemeier, T. J Proteome Res 2008, 7, 4982-4991. (30) Kind, T.; Liu, K. H.; Lee do, Y.; DeFelice, B.; Meissen, J. K.; Fiehn, O. Nat Methods 2013, 10, 755-758. (31) Folch, J. J Biol Chem 1948, 174, 439-450. (32) Christie, W. W. Nat Prod Rep 1984, 1, 499-511. (33) Sahin, H. T.; Arslan, M. B. Int J Mol Sci 2008, 9, 78-88. (34) Bollinger, J. G.; Ii, H.; Sadilek, M.; Gelb, M. H. J Lipid Res 2010, 51, 440-447. (35) Kitson, A. P.; Marks, K. A.; Aristizabal Henao, J. J.; Tupling, A. R.; Stark, K. D. Nutr Res 2015. (36) Metherel, A. H.; Taha, A. Y.; Izadi, H.; Stark, K. D. Prostaglandins Leukot Essent Fatty Acids 2009, 81, 417-423. (37) Ji, R. R.; Xu, Z. Z.; Strichartz, G.; Serhan, C. N. Trends Neurosci 2011, 34, 599-609. (38) Bradley, R. M.; Marvyn, P. M.; Henao, J. J. A.; Mardian, E. B.; George, S.; Aucoin, M. G.; Stark, K. D.; Duncan, R. E. Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids 2015, 1851, 1566-1576.


Page 16 of 20

Page 17 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Whole blood lipid extract dilution curves for A. 16:0/18:2 phosphatidylcholine, B. 16:0/18:1/18:2 triacylglycerol, and C. 18:2 cholesteryl ester.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1 2 3 4 5 6

Figure 2. Improving recoveries of phosphatidylserine (PS) species from DBS following a 24h Folch based extraction. Different letters denote significant differences (p < 0.05) after one-way ANOVA with Tukey post-hoc test across each and every PS species.


Page 18 of 20

Page 19 of 20

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

7

8 9 10 11 12 13 14 15


Table 1. Lipidomic profiles of whole blood and dried blood spot biological replicates Technical replicates* Lipid WBi WB24 DBSi DBS24 Analyte response/internal standard response 16:0 lyso-PC 0.98 ± 0.04a 0.97 ± 0.04a 0.43 ± 0.05b 1.00 ± 0.04a a a b 10.84 ± 0.17 3.82 ± 0.57 10.73 ± 0.22a 16:0/18:2 PC 10.47 ± 0.69 16:0/20:4 PC 13.37 ± 0.45a 12.93 ± 0.03a 7.50 ± 0.97b 13.33 ± 0.29a 16:0/20:5 PC 0.59 ± 0.02a 0.56 ± 0.02a 0.29 ± 0.04b 0.59 ± 0.02a a a b 16:0/22:6 PC 1.32 ± 0.09 1.38 ± 0.03 0.52 ± 0.08 1.33 ± 0.03a a a b 18:0/22:6 PC 0.28 ± 0.02 0.30 ± 0.01 0.13 ± 0.04 0.29 ± 0.01a a a b 16:0/18:2 PE 0.12 ± 0.01 0.13 ± 0.01 0.06 ± 0.01 0.13 ± 0.01a a a b 16:0/22:6 PE 0.21 ± 0.01 0.22 ± 0.01 0.12 ± 0.02 0.22 ± 0.01a 16:0/18:1/18:1 TAG 2.16 ± 0.08a 2.26 ± 0.06a 0.97 ± 0.29b 2.32 ± 0.06a 18:2 CE 1.12 ± 0.04a 1.14 ± 0.06a 0.63 ± 0.11b 1.15 ± 0.10a a a b 0.57 ± 0.01 0.59 ± 0.03 0.37 ± 0.04 0.40 ± 0.01b 18:0/22:6 PS

Biological replicates† WB24 DBS24

0.90 ± 0.11 0.81 ± 0.16 11.48 ± 2.73 10.88 ± 3.13 6.89 ± 1.34 7.05 ± 0.85 0.29 ± 0.18 0.30 ± 0.19 0.68 ± 0.36 0.60 ± 0.26 0.20 ± 0.10 0.17 ± 0.06 0.08 ± 0.01 0.07 ± 0.01 0.18 ± 0.05 0.17 ± 0.06 2.32 ± 0.16 2.46 ± 0.17 1.14 ± 0.14 1.29 ± 0.19 0.37 ± 0.08 0.16 ± 0.07‡ Additional acyl species 18:0/20:5 PC 0.14 ± 0.06 0.13 ± 0.05 18:1/22:6 PC 0.03 ± 0.01 0.02 ± 0.01 p16:0/20:5 PE 0.02 ± 0.01 0.02 ± 0.01 p16:0/22:6 PE 0.05 ± 0.03 0.05 ± 0.02 p18:0/20:4 PE 0.31 ± 0.03 0.27 ± 0.02 16:0/18:2/20:5 TAG 0.01 ± 0.01 0.01 ± 0.01 16:0/18:2/22:6 TAG 0.04 ± 0.03 0.04 ± 0.03 18:1/18:2/20:5 TAG 0.04 ± 0.03 0.04 ± 0.03 16:0/18:1/22:6 TAG 0.07 ± 0.05 0.07 ± 0.05 20:5 CE 0.03 ± 0.03 0.03 ± 0.03 22:6 CE 0.05 ± 0.02 0.04 ± 0.02 0.06 ± 0.01 0.02 ± 0.01‡ 18:0/18:1 PS 18:0/20:3 PS 0.03 ± 0.01 0.01 ± 0.01‡ 18:0/20:4 PS 0.56 ± 0.07 0.20 ± 0.05‡ Data is mean ± SD. *n=3 single participant analyzed in triplicate across treatments, †n=3 samples from different participants across treatments. Whole blood immediate (WBi); whole blood 24h (WB24); dried blood spot immediate (DBSi); dried blood spot 24h (DBS24); phosphatidylcholine (PC); phosphatidylethanolamine (PE); plasmenyl (p); phosphatidylserine (PS); triacylglycerol (TAG); cholesteryl ester (CE). Letter superscripts indicate statistically significant differences after one-way ANOVA with Tukey post-hoc test and p-value

Tailored Extraction Procedure Is Required To Ensure Recovery of the

Recommend Documents