Agilent Lipidomics Discovery Profiling Targeted LC/MS Analysis in 3T3-L1 Differentiating Adipocytes Application Note
Contents
1. 79 Counts vs Acquisition Time min C O ESS 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 Counts vs Acquisition Time min Figure 10 A ESI positive ion mode TAG_C54 3 showing reduced signal for day 8 relative to day 2 and 0 B TAG_C48 1 showing increased signal for day 8 relative to day 2 and 0 10 Cpd 3 OddFFA_C17 1 ESI EIC 133 1128 169 0895 170 0882 179 1183 Scan Frag 100 0V Ovs2vs8 neg Day 8 3 1 9 9 29 557 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 70 72 74 76 78 80 82 84 86 88 90 92 94 Counts vs Acquisition Time min Figure 12 ESI negative ion mode EIC for m z 268 24023 corresponding to OddFFA_C17 1 showing signal only for day 8 no signal detected for day 2 and 0 Relative Fold Change of Free Fatty Acids In negative ion mode the rate of alpha oxidation increases during differentiation resulting in the appearance of odd chain free fatty acids Fig 11 14 i o S amp S N Day 8 Day 10 oOo N A O0 C15 0 C17 1 FFA FFA Figure 11 During the differentiation process alpha oxida tion of acyl coA is dramatically increased resulting in the elevation of odd chain free fatty acids C15 0 and C17 1 at day 8 compared to day 0 Samples were run in triplicate pvalue lt 0 01 pvalue lt 0 005 Indeed this pattern was observe2. 5 um 4 6 x 50 mm Guard C18 reversed phase 3 5 um 2 x 20 mm Mobile Phase A 95 5 water methanol 0 1 ammonium hydroxide Mobile Phase B 65 30 5 isopropanol methanol water 0 1 ammonium hydroxide Gradient 0 B at 0 min 20 B at 5 min 100 B at 65 min 0 B at 85 min MS stop time 95 min LC stop time 95 min Column temperature 30 C Flow rate 0 4 mL min 0 1 mL min 0 to 5 min Injection volume 30 uL Figure 1 The differentiation of 3T3 L1 cells is accomplished through the addition of a hormone cocktail to preadipocytes day 0 After 6 8 days the cells have changed mor phology and have begun to produce lipid droplets within their cell mem branes These lipid droplets can be visualized by staining the cells with the Oil Red O stain Positive ion mode runs Column Luna C5 reversed phase column 3 5 um 4 6 x 50 mm Guard C4 reversed phase 3 5 um 2 x 20 mm Mobile Phase A 95 5 water methanol 0 1 formic acid 5 mM ammonium formate Mobile Phase B 65 30 5 isopropanol methanol water 0 1 formic acid 5 mM ammonium formate Gradient same as negative mode MS conditions MS System Agilent 6520 Quadrupole Time of Flight Q TOF LC MS lonization mode ESI lonization Polarity Negative and Positive Drying gas flow 10 L min Drying gas temperature 350 C Nebulizer pressure 45 psi Data range 100 1 500 Acquisition Rate 1020 4 ms spectrum Fragmentor voltage 100 V Skimm
3. T on inj n ee 8 e e Lipidomics Discovery Profiling and a O Targeted LC MS Analysis in 3T3 L1 o o oy eo o o Differentiating Adipocytes Application Note Authors Abstract The mechanism s underlying the formation of adipose tissue is of tremendous scientific interest due to the potential to mitigate obesity The 3T3 L1 cell line has been a valuable model Agilent Technologies Inc for studying this process since many of the molecular processes that drive differentiation of this Santa Clara CA USA cell line in vitro is consistent with the processes of adipogenesis in vivo To gain greater insight into these lipid changes O TOF LC MS was used for lipidomics discovery profiling in both untargeted and targeted modes to find potential lipid biomarkers of differentiation in Alan Saghatelian and James Cardia adipocytes Harvard University Theodore Sana and Steve Fischer Introduction Department of Chemistry and Lipidomics is a branch of metabolomics and is a systems based study of all lipids non water soluble metabolites the molecules with which they interact and their function within the Cambridge MA USA cell Lipid abnormalities contribute to many diseases including atherosclerosis diabetes obesity Alzheimer s and metabolic disease Chemical Biology The differentiation of 3T3 L1 cells has previously been extensively studied using a variety of approaches including microarray and protein expression a
4. d for m z 268 24023 corresponding to the formula for odd chain Free Fatty Acid C17 1 OddFFA_C1 7 1 in day 8 samples only Fig 12 Conclusions We have demonstrated the capabilities of a complete suite of Agilent software for performing differential lipidomics profiling in the context of LC MS discovery workflows in both untargeted and targeted modes By incorporating both fast and robust peak finding algorithms differential analysis programs and visualization software we were able to show that the lipid profiles of these cells could be clearly distinguished between day 0 2 and 8 of differentiation Moreover many lipid compounds were found to have matches to the METLIN and LipidMaps databases GeneSpring MS analysis provided valuable information through PCA as well as the application of the K means algorithm to identify metabolites that respond in a similar fashion This was instrumental in identifying a number of important lipid classes and revealed underlying changes in metabolism related to fatty acid oxidation which impacted all acylated species In addition a series of transiently elevated metabolites were discovered and these lipids will provide a starting point to help understand the metabolic changes that occur in the transition from preadipocytes to adipocytes Lastly these same metabolites can be analyzed in a variety of different systems including human samples References 1 Pekala P H Moss J 373 L7 preadipocyt
5. des For example while the entire pool of triglycerides is greatly increased during differentiation specific species changes such as trioleoyl glycerol are actually lower at day 8 while species containing shorter acyl chains a product of fatty acyl CoA oxidation are strongly elevat ed Furthermore increases in odd chain acyl groups in glycerophospholipids such as C33 2 PC is expected due to an increase in the rate of alpha oxidation in differentiating 313 L1 cells 7 In positive ion mode decreased abundance of an m z 884 78329 species matching that of TAG_C54 3 was observed during differentiation Fig 10A This is consistent with generalized acyl chain shortening during differentiation due to increased pro tein levels and activity of stearoyl CoA desaturase 2 Furthermore a dramatic concomitant increase in the levels of m z 804 72069 matching the formula for TAG_ C48 1 Fig 10B was also observed and contains at least one degree of unsaturation x108 Cpd 98 TAG_C54 3 ESI E1C 443 3989 443 9006 460 4255 460 9271 Scan Frag 100 0V Ovs2vs8_pos_Day0_3 d 67 774 Cpd 85 TAG_C54 3 ESI E1C 443 3989 443 9006 460 4255 460 9271 Scan Frag 100 0V Ovs2vs8_pos_Day2_3 d 67812 ESI EIC 443 3989 443 9006 460 4255 460 9271 Scan Frag 100 0V Ovs2vs8_ pos Day8 3 d 67 892 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78
6. e differentiation and poly ADP ribose synthetase Mol Cell Biochem 1983 53 1 pp 221 232 2 Burton G R Nagarajan R Peterson C A McGehee R E Jr Microarray analysis of differentiation specitic gene expression during 313 L1 adipogenesis Gene 2004 329 pp 167 185 3 Ntambi J M Buhrow S A Kaestner K H Christy R J Sibley E Kelly T J Jr Lane M D Differentiation induced gene expression in 3T3 L1 preadipocytes Characterization of a differentially expressed gene encoding stearoyl CoA desaturase J Biol Chem 1988 263 33 pp 17291 17300 4 Molina H Yang Y Ruch T Kim J W Mortensen P Otto T Nalli A Tang Q Q Lane M D Chaerkady R Pandey A Temporal profiling of the adipocyte proteome during differentiation using a five plex SILAC based strategy J Proteome Res 2009 8 1 pp 48 58 5 Xiao Y Junfeng H Tianhong L Lu W Shulin C Yu Z Xiaohua L Weixia J Sheng Z Yanyun G Guo L Min L Cathepsin K in adipocyte differentiation and its potential role in the pathogenesis of obesity J Clin Endocrinol Metab 2006 91 11 pp 5420 5427 6 Folch J Lees M and Stanley GHS A simple method for the isolation and purification of total lipids from animal tissues J Biol Chem 1957 226 pp 497 509 7 Su X Han X Yang J Mancuso DJ Chen J Bickel PE and Gross RW Sequential Ordered Fatty Acid aOxidation and A9 D
7. er voltage 60 V Octopole RF voltage 250 V Capillary voltage 3 500 V Reference masses m z negative ion 119 0363 980 016375 positive ion 121 0509 922 0098 Results and Discussion Analysis of the 3T3 L1 cells lipidome by LC MS identified many metabolomic changes between pre adipocytes day 0 and mature adipocytes day 8 using both non targeted i e discovery profiling and targeted using a list of known compound formulas approaches The day 2 cells tran sitioning to become adipocytes could also be differentiated based on their abundance profile differences from the earlier and later time points In our untargeted analyses the data analysis workflow we employed Metabolite Extraction 3T3 L1 cells day 0 2 8 Separation 1200 Series LC Detection Data 6520 Q TOF Analysis Figure 2 Workflow for the analysis of the lipidome of 3T3 L1 cells at days 0 2 8 Lipids were isolated using the Folch 1 extraction method followed by LC MS analysis and data analysis using GeneSpring MS Fig 2 included pairwise comparisons of pre adipocytes day 0 and mature adipocytes day 8 in Mass Profiler software as well as PCA 1 way ANOVA and K means cluster analysis in GeneSpring MS software that differentiated the cells at various stages of adipogenesis The acquisition of spectral data on a Q TOF instead of targeted multiple reaction moni toring MRM data on a triple quadrupole Q0 Q permits both non targ
8. esaturation Are Major Determinants of Lipid Storage and Utilization in Differentiating Adipocytes Biochemistry 2004 43 17 pp 5033 5044 www agilent com chem metabolomics This item is intended for Research Use Only Not for use in diagnostic procedures Information descriptions and specifications in this publication are subject to change without notice Agilent Technologies shall not be liable for errors con tained herein or for incidental or consequential damages in connection with the furnishing performance or use of this material Agilent Technologies Inc 2009 Published in the U S A June 30 2009 5990 4212EN EE Agilent Technologies
9. eted and targeted data analysis from a single analytical run Thus for our targeted approach we con structed a METLIN Personal Metabolite database of 179 lipid compounds that are believed to be involved in adipocyte differ entiation We then queried each sample data file in MassHunter Qual by loading the lipid database containing the formula for each lipid species Each formula was sub sequently used to query the data files using a Find by Formula function A Mass vs Retention Time aaia Ie la ge ee Mass Da ammofos 70 b Retention Time min B Day 0 vs Day 8 Log2 Abundance Day 0 N N Log2 Abundance Day 8 Untargeted Metabolite Identification 1 Pairwise analysis of day 0 and day 8 data sets After performing Molecular Feature Extraction MFE of all data files in MassHunter Qualitative Analysis software the resulting mhd files were imported into Mass Profiler software for statistical analysis and identification The files were aligned and binned using specified mass and retention time RT tolerance windows A pairwise analysis of pre adipocytes day 0 and mature adipocytes day 8 was performed for the triplicate samples per group A plot of Mass and RT revealed several interesting features about the data acquired in ESI mode including a pattern consistent with a single CH subunit in growing or degrading polymeric chains spaced at approximately 0 5
10. including those that increased in day 8 cluster 4 decreased in day 8 cluster 6 or were transiently elevated in day 2 cluster 7 during differentiation Within these clusters each of the target masses identified in the profiling processes was searched against the METLIN database The database was searched over a 10 ppm mass window The empirical formula calculation was set to a mass error window of 5 ppm With this analysis many different classes of metabolites predominantly lipids were identified as significantly changing throughout the differentiation process For example Fig 7 shows a METLIN search result for cluster 4 revealing several lipids in day 8 adipocytes to be elevated GPE tn 16 0 1 8 4 12212 GPE tn AZ 12118 Z12215 T C3SH72NOEP C41H70NO08P SMId1 amp 1 200 Ghucosyices aerate 618 1 22 0 SM d1 amp 1 22 0 GPE tri 1 80 22 42 72 1021 32 162 1 GPSef 1 amp 0 SOYU Ghucosyiceraerede 01 1 24 0 Docetaxel M2 CA3HGON206P C4ASHEINOG C4SH32N206P C4SH78NOGP C43H84N010P CASHIN C43H53N015 SMid181 25 0 1 2 34n SZ_12Z hepladecadenoyl srrgpcerol 1 E heptadecenoyl 2 HZ 12Z heptadecadenoyl sn gycerol C4SHSBN206P CS4H9206 Figure 7 A summary of compound identification search results for cluster 4 using the METLIN database Links to the LipidMaps database are also provided Separately analysis of cluster 7 revealed several transiently elevated metabo
11. ipic acid Sulfolithecholic acid GPEtn 18 0 18 1 7Z 3 oxo tricosanoic acid GPEtn 16 0 18 2 9Z 122 GPEtn 18 0 17 0 U GPEtn 18 1 92Z 16 0 U GPEtnNMe 18 1 9Z 16 0 GPEtn 16 0 20 0 3alpha androstanediol glucuronide GPEtn 18 2 9Z 12Z 18 2 Z 12Z 9 10 16 trihydroxy palmitic acid ae ppm Score C25H4008 1 4 C39H78NO8P 1 5 C10H17NO0O3 4 7 C6H1005 4 6 C4H406 cC5H805 C41H78NO8P C41H80NO8P C6H1005 C41H74NO8P C24H4006S C41H80NO8P C23H4403 C3SH74NO8P C40H80NO8P C39H76NO8P C40H78NO8P C41H82NO8P C16H3205 oOo N 8 NNN WO WwW WwW RH Ww we ONNO mass MFG Abundance Replicates Abundance Replicates log2 O wo wm wwe WB WwW amp 8 Ww Ww WwW H yw ww WwW w Figure 4 Mass Profiler software analysis of day 0 and day 8 samples showing selected list of proposed differential metabolites after METLIN database matching within a 5 ppm mass tolerance window In order to identify the metabolites the METLIN Personal Metabolite Database was subsequently queried directly from within Mass Profiler software Fig 4 shows a selected summary of proposed compounds based on METLIN database matches for day 8 vs day 0 3T3 L1 cells Also included in the table is the log ratio and an associated p value Student s t test of significance based on triplicate samples Missing values were treated as being 0 2 Multi group analysis of day 0 day 2 and day 8 data sets GeneSpring MS software was used for the di
12. lites for day 2 such as C18 1 acyl carnitine Fig 8 that differentiates this time point from the other days C18 1 acyl carnitine 0 Figure 8 Temporally regulated metabolites were identi fied using GeneSpring MS software C18 1 acyl carnitine was found to be elevated by K Means analysis at day 2 compared to day 0 and day 8 The importance of these changes is not cur rently known but the ability of metabolomics approaches to discover such changes will likely lead to a better understanding of the differentiation process Finally the results from all METLIN searches for all clusters were incorporated into Excel spreadsheets for further interpretation This enabled the selection of individual lipids that were differential across the three days and to retrieve their profile information in GeneSpring MS software Fig 9 _ oS oe Normalized Intensity log scale So 103 Normalized Intensity log scale Free fatty acids Sphingomyelins Monoacylglycerols Diacylglycerols Triacylglycerols Cholesterol Cholesteryl esters Steroid hormones Retinoids Ceramides Targeted guided metabolite identification A list of known lipid compounds was created from several different lipid classes Table 1 and used to construct a custom METLIN Personal Metabolite Database of over 170 lipids The custom built lipid database was then used to interrogate the sample data files from each ESI acquired ion p
13. lot of all biological sample triplicates for day 0 2 and 8 based on the abundance profiles of 1 209 masses acquired in positive ESI mode The magnitude of the variances are explained by PCA component PC1 63 71 PCA component PC2 20 25 PCA component PC3 8 18 lormalized Intensity og scale Normalized Intensity 9 log scale Normalized Intensity 1 3 log scale 4 1 ee 0 1 Y Day 0 2 8 Figure 6 K Means cluster analysis in GeneSpring MS software Several clusters are highlighted by red boxes 4 6 and 7 Normalized Intensity og scale Normalized Intensity 1 1 og scale Normalized Intensity 15 log scale 1 Normalized Intensity log scale ier Intensity 12 og scale 0 2 8 Normalized Intensity log scale Day having different relative metabolite abundance profiles for day 8 adipocytes relative to day 2 and day 0 3T3 L1 cells Separation in negative ion mode was also observed not shown An ANOVA with a Tukey post hoc test provided a list of significantly differential compounds during differentiation of 3T3 L1 cells and they were identified by matching to the METLIN Name 1 2 h SZ heptadecenoy sn ghcerol database In addition due to the large number of differential metabolites we also performed K means cluster analysis that identified pools of co regulated compounds Fig 6 C37H6805 We analyzed several clusters
14. min intervals These polymers appeared only in day 8 adipocytes starting at approximately 55 min in the LC MS analysis Fig 3A A plot of the abundances of 1 532 features on the log scale revealed many differential metabolites with gt 4 fold change particularly in the direction of mature adipocytes Fig 3B Figure 3 A Mass vs RT plot displaying 1 532 features averaged for day 0 and day 8 data files in Mass Profiler software The data files were acquired using ESI analysis in positive ion mode The number of features were filtered based on the requirement that each feature must be present in 3 of 3 replicates in at least one group day The sloping trend in the features starting at 55 min is due to fatty acyl oxidation in day 8 adipocytes that causes shortening of polymeric chain in CH increments B A log log plot displaying fold change between averaged day 0 and day 8 data files in Mass Profiler software The sloping green lines indicate 1 2 and 4 fold difference on a log abundance scale A significant shift in fold abundances for many features can be observed for day 8 cells 468 2734 719 5448 199 1215 162 0518 148 0007 148 0363 743 5453 745 561 162 0517 739 5147 456 2523 745 5625 368 3309 715 5128 733 5605 717 5294 731 5505 747 5739 304 2265 GPEtn 14 0 20 0 U N Acetyltranexamic acid 3 Hydroxyadipic acid Dihydroxyfumarate Citramalic acid GPEtn 18 0 18 2 9Z 122 GPEtn 18 0 18 1 7Z 3 Hydroxyad
15. nalyses 1 5 These different approaches all provide important yet complementary information However there is very limited LC MS lipid profiling information available for this cell line that reveals the important changes that occur in some of the major lipid families during differentiation An LC MS system composed of an Agilent 1200 Series LC and Agilent 6520 Quadrupole Time of Flight Q TOF LC MS was selected for this study due to the high retention time reproducibility sub 2 ppm mass accuracy and outstanding abundance reproducibility necessary for successful profiling experiments Agilent MassHunter Qual MassHunter Mass Profiler and GeneSpring MS bioinformatics software was used to analyze the complex multi class data generated by this study Differentiation in 3T3 L1 cells is an 8 day process that Is initiated by the addition of a hormone cocktail to pre adipocytes This study focused on profiling lipids at days 0 2 and 8 during the differentiation process to find and identify metabolites whose abundances significantly changed over this time period For each time point three biological sample replicates were analyzed These results reveal that the metabolite profiles composed of many lipid species were able to clearly distinguish between the different time points In addition many of the changing lipids followed patterns expected for the formation of adipose cells whee Agilent Technologies Experimental Cell Growth and Differentia
16. olarity in a targeted or guided fashion in MassHunter Qual This was accomplished using the Find by Formula algorithm which finds and extracts all the co eluting isotopes salt adducts dimers etc belonging to each Figure 9 GeneSpring MS software results showing elevated normalized abundance profiles of day 8 samples for A m z 228 2078 and B m z 254 2233 corresponding to C H 0 myristic acid and palmitoleic acid C H 4 0 respectively Phosphatidylcholines Phosphatidylethanolamines Phosphatidylserines Phosphatidylinositols Phosphatidylglycerols Phosphatidic acids Lysophospholipids Table 1 List of lipid families identified in the 3T3 L1 cell line that was used to create a list of compounds in METLIN database for targeted profiling empirical formula Several classes of metabolites were identified this way These included lipids such as triglycerides odd chain free fatty acids and phosphatidyl choline Interestingly many of the changing species contained an acyl chain in their structure that is also consistent with the non targeted discovery analysis results that was observed in Fig 3A The importance of fatty acyl CoA oxidation in reshaping the acyl chain landscape throughout differentiation has previously been defined 7 Those observations are consistent with the results shown here where the changes in acyl chains cause a shifting distribution of acylated species such as triglyceri
17. scovery of metabolite biomarkers through the analysis and visualization of LC MS data A comprehensive suite of powerful statistical analysis tools were used to profile metabolites associated with changes in lipidogenesis function enabling the rapid discovery of these potential biomarkers The following workflow was used for untargeted metabolite identification in GeneSpring MS software Grouping and filtering of high quality features A Triplicate ESI processed mhd files for each day were imported into GeneSpring MS software Total number of binned and aligned features 10 011 C Group i e average the triplicate sample abundances by day D Apply a frequency filter retain only those features present in all triplicates in at least one group day 2 108 E Apply error filter on D above retain features having CV in the range from 0 to 1 000 1 209 oJ Statistical Analysis A Perform principle component analysis PCA on 1 209 features i e high quality set E B Perform analysis of variance ANOVA on E above p lt 0 05 with Tukey post hoc test for days 0 2 and 8 C K Means cluster analysis on E above to identify groups of co varying masses across the 3 days PCA analysis of the filtered data set i e 1 209 masses revealed patterns showing clear separation of the triplicate samples based on the abundance profiles of those metabolites across the three days Fig 5 Figure 5 PCA p
18. tion 1 8x10 3T3 L1 cells provided by Dr H Green Harvard Medical School Boston MA were seeded on 100 mm dishes and maintained in Dulbecco s Modified Eagle s medium DMEM supplemented with 10 bovine calf serum BCS Hyclone Two days after confluence the medium was changed to DMEM supplemented with 10 fetal bovine serum FBS and a hormone cocktail of 1 mM dexamethasone Sigma 0 5 mM 3 isobutyl 1 methylxanthine Sigma and 5 pg mL insulin Sigma This induces differentiation of 3T3 L1 cells which can be observed through the staining of adipocytes with Oil Red O stain Fig 1 Forty eight hours after induction the media was replaced with DMEM containing 10 FBS and 5 pg mL insulin Forty hours later the cells were fed maintenance media DMEM plus 10 FBS which was replaced every 2 days Lipid Sample Preparation Cell pellets were resuspended in 1 mL of PBS and were dounce homogenized in 2 1 1 CHCI MeOH PBS buffer 6 The homogenate was centrifuged at 2 000 x g for 5 min at 4 C to separate the organic and aqueous layers The organic layer was removed concentrated under N and then dissolved in 120 uL CHCI prior to analysis by LC MS Day 0 a Day 8 LC MS Analysis MS only data was acquired in both positive and negative ion modes using an Agilent 1200 Series LC and Agilent 6520 Quadrupole Time of Flight Q TOF Negative ion mode runs Column Gemini C18 reversed phase column 3
Download Pdf Manuals
Related Search
Agilent Lipidomics Discovery Profiling Targeted LC/MS Analysis in 3T3 L1 Differentiating Adipocytes