Update on LIPID MAPS classification, nomenclature, and shorthand notation for MS-derived lipid structures

1Lipids have become increasingly recognized as the central metabolites affecting human physiology and pathophysiology, and LIPID MAPS has recently expanded its tools, resources, data, and training as a free resource dedicated to serving the lipid research community. Following development of the LIPID MAPS nomenclature, classification, and structural representation system, an initial shorthand nomenclature was proposed, which included a structural hierarchy as shown by others as well. These were the first attempts to provide rules for reporting mass spectrometric data dependent on the power for structural resolution of lipids by the instrumental set-ups in use at that time.

2Today, we recognize that the field has evolved in often diverging ways and that this has not enabled a unifying naming convention to be adopted throughout. For example, alternative shorthand notation has evolved for some lipid classes, a plethora of newly determined structures for lipids from various classes and phylogenetic kingdoms (higher plants and yeasts) have been described, and progress in the technological development of mass spectrometers with greater structural resolution as well as advances in automation in interpreting high-throughput data has occurred. To address this, it is the aim of this report to take into account these developments and to present an update on the LIPID MAPS classification and a pragmatic highly usable shorthand notation for those active in lipid research. This update will focus on five of the eight LIPID MAPS categories, namely Fatty Acyls (FA), Glycerolipids (GL), Glycerophospholipids (GP), Sphingolipids (SP), and Sterols (ST). Annotation is modified to permit annotation of oxygenated lipids and examples will be given for lipid classes occurring outside the mammalian kingdom.

3“Biological intelligence” has been considered as topical knowledge about a lipid molecule, such as its structural building blocks, enzymatic pathways for generation and metabolism, and biological functions. Interpretation by biological evidence in shorthand notation can be useful when mass spectra contain structural ambiguities or lack of clear structural evidence. Consequently, annotations with the help of biological evidence contain assumptions, and it must be recognized and recorded that this may lead to misinterpretations. Moreover, in the pragmatic approach presented in this work, we will make more use of common and/or trivial names for the shorthand notation. For example, the structures of sterols, prostaglandins, resolvins, etc. have been characterized by chemical and spectroscopic methods, including stereochemistry, and common names exist, as do shorthand notations in many cases. Their mass spectra are also known; however, their stereochemistry and isomerism and other structural information often cannot be deduced directly from the spectra when these lipids are measured in biological samples. Assignment of a common name or of shorthand notation to such chromatographic and MS/MS data is permissible, but it may be based on annotation that includes biological intelligence, and that needs to be clearly stated as well.

4In any case, assumptions made should be striking a unique balance between what we think we know about structure and function of a lipid molecule and what a specific MS-based analytical method definitively informs us about the lipid structure.

5Modification of Fatty Acyls by oxygen, either catalyzed enzymatically or by means of radical chemistry, is an important focus in biomedical research, due to the impressive biological activities of products thus obtained. Based on these two mechanisms, all compounds originating from polyunsaturated fatty acyls (PUFAs) having methylene-interrupted cis-double bonds (DBs) (also chemically referred to as allylic DBs) and being enzymatically or nonenzymatically oxygenated are grouped within the appropriate class in the Fatty Acyl category. Historically, the term “eicosanoid” has included “related oxygenated polyunsaturated fatty acids” with shorter or longer chain lengths, but in the LIPID MAPS classification, compounds are strictly assigned to a class based on their chain length (e.g., octadecanoids, eicosanoids, docosanoids). Recently, the common name “oxylipins”, standing for “oxygenated fatty acyls”, has come into widespread use. Similarly, in the Glycerophospholipids (GP) category, many newly described phospholipids contain oxygenated fatty acyls (or oxylipins) often termed “oxygenated phospholipids” (OxPLs). Those are produced by oxygenation of constituent fatty acyls enzymatically and nonenzymatically, or by chemical modification of polar head groups containing an amino function (PE and PS), i.e., N-modified phospholipids.

6In the following, we elaborate first on experimental prerequisites for correct annotation of lipid mass spectrometric data and, second, present the updates on rules for using shorthand notation. Finally, in order of categories, we present mostly in the form of easily readable tables, all updates on lipid nomenclature and classification including respective shorthand abbreviations according to the LIPID MAPS web resources and the updated shorthand notation for lipid species and lipid molecular species. To further enhance the understanding of shorthand notation, some chemical structures are presented in the tables. The updated shorthand notation schemes described herein have been incorporated into a number of key resources on the LIPID MAPS website, notably the LIPID MAPS Structure Database (LMSD) and the MS search tools (see the Hierarchical concept and application of shorthand notation section below), by generating level-specific abbreviations (e.g., sum-composition and chain-specific annotations) for lipid structures. This approach is important in terms of the development of MS search databases that are appropriate for the technique used (sum-composition databases for precursor ion data and chain-composition databases for MS/MS data).

7All lipid species and lipid molecular species data presented need information on levels of structural resolution attained by mass spectrometric analysis, and sufficient supplementary data to justify annotation by shorthand notation. At minimum, such data should contain the measured intact m/z value, the adduct ion used for identification, the retention time when chromatography is applied, and the measured fragment m/z values.

8Assignment and therefore use of specific shorthand nomenclature for defined functional groups (Table 1 A–C) requires additional techniques. An example is derivatization of hydroxyl groups by trimethylsilylation followed by GC/MS EI and analysis of fragment ions formed. In many cases ESI-MS/MS of underivatized constituent fatty acyls in general leads to specific product ions, if ESI populates a charge site near the functional group. Definition of DB positions can be determined by several techniques including ozonolysis during analysis (OzID) or specific adduct formation with acetone in photochemical Paterno-Büchi reaction. These reactions can be carried out in shotgun or LC-MS/MS experiments. High energy MS/MS has been used to assign DB position of fairly complex fatty acyls as well as methyl branching. Alternatively, GC/MS can be used including specific derivatization of the carboxylate group, to drive specific DB fragmentation in EI spectra. Chemical ionization techniques are also useful by application of specific chemical ionization reagent gases to define DB positions.

9Common names of lipid species, e.g., for certain fatty acids and for oxygenated fatty acids denote a chemically defined structure including stereochemistry. For proper annotations in these cases, the analytical method has to provide for chiral separation of known stereoisomeric compounds. This validation demands data on reproducibility and limit of quantification. Similarly, when novel structures are described, analytical details proving structural details need to accompany the data. Guidelines for method validation and reporting of novel lipid molecules are currently being developed within the Lipidomics Standard Initiative (https://lipidomics-standards-initiative.org) as community-wide effort.

10Here, we describe updates and rules applicable to all lipid categories described below. This includes rules on the hierarchical concept and application of the nomenclature and annotation of lipid structures as well as on annotation of stable isotope-labeled lipids. Three major updates are: The term “DBs” is replaced by “double bond equivalents” (DBEs), because removal of two hydrogen atoms from precursor lipid forms a double bond, an oxo group or a cyclic structure. Frequently, MS does not distinguish between these alternatives.Oxygen atoms represent not only the main component introduced during oxygenation, but occurs also in hydroxy groups as a principal structural feature in many lipid classes such as sphingoid bases. Because hydroxy, oxo or other oxygen functional groups may not be differentiated by high resolution/accurate mass analysis, annotation is done by the number of oxygens linked to the hydrocarbon chain.Use of parentheses and brackets is minimized. Parentheses indicate primarily positions and, with regard to functional groups only those with numbers behind them, like (OH)2, (NO2), (NH2). The use of square brackets is restricted to chemical configurations R and S, to stable isotopes, and to the frame of carbons in a ring structure.

11The term “DBs” is replaced by “double bond equivalents” (DBEs), because removal of two hydrogen atoms from precursor lipid forms a double bond, an oxo group or a cyclic structure. Frequently, MS does not distinguish between these alternatives.

12Oxygen atoms represent not only the main component introduced during oxygenation, but occurs also in hydroxy groups as a principal structural feature in many lipid classes such as sphingoid bases. Because hydroxy, oxo or other oxygen functional groups may not be differentiated by high resolution/accurate mass analysis, annotation is done by the number of oxygens linked to the hydrocarbon chain.

13Use of parentheses and brackets is minimized. Parentheses indicate primarily positions and, with regard to functional groups only those with numbers behind them, like (OH)2, (NO2), (NH2). The use of square brackets is restricted to chemical configurations R and S, to stable isotopes, and to the frame of carbons in a ring structure.

14Upon application of a validated MS-method, interpretation of mass spectra by “biological intelligence” and the use of common or trivial names, as alluded to specifically in the introduction, is permissible. Such annotations need to be clearly stated. Examples are ambiguities pertaining to bond type, oxygenated groups, and branched chains.“Species level” is now the lowest hierarchical level. It represents the sum composition, i.e., sum of carbon atoms, DBEs, and number of additional oxygen atoms, e.g., FA 18:1;O. It thus replaces former “Lipid class level” mass (i.e., lipid class and the – uncharged - molecular mass). Of note, for sterols, the ABCD ring system is assumed and not expressed as DBE.“Phosphate-position level” annotates positions of phosphate group(s), e.g., PIP(3′) or PIP2(4′,5′) at phosphatidylinositolphosphate.“Molecular species level” pertains to all categories addressed here and is reached as soon as constituent fatty acyl/alkyl-residues are identified, e.g., TG 16:0_18:1_18:1, a triglyceride.“sn-position level” is a more refined level in GL and GP categories, enabling annotation of the sn-position of fatty acyl/alkyl constituents at the glycerol backbone as indicated by a slash, e.g., TG 16:0/18:1/18:1.“DB-position level” or “DBE-position level” pertain to species having constituents with defined position of double bonds or double bond equivalents, e.g., FA 18:2 (9, 11);O.“Structure defined level” annotates molecular species composed of various constituents and functional groups, yet without positions and stereochemical details, e.g., FA 18:2;OH.“Full structure level” annotates molecular species composed of various constituents and functional groups including positions, yet without stereochemical details, e.g., FA 18:2(9Z,11E);13OH.“Complete structure level” defines detailed structures of all functional groups including stereochemistry as shown in the LMSD, e.g., 13R-HODE, 13S-HODE (= common name).

15Upon application of a validated MS-method, interpretation of mass spectra by “biological intelligence” and the use of common or trivial names, as alluded to specifically in the introduction, is permissible. Such annotations need to be clearly stated. Examples are ambiguities pertaining to bond type, oxygenated groups, and branched chains.

16“Species level” is now the lowest hierarchical level. It represents the sum composition, i.e., sum of carbon atoms, DBEs, and number of additional oxygen atoms, e.g., FA 18:1;O. It thus replaces former “Lipid class level” mass (i.e., lipid class and the – uncharged - molecular mass). Of note, for sterols, the ABCD ring system is assumed and not expressed as DBE.

17“Phosphate-position level” annotates positions of phosphate group(s), e.g., PIP(3′) or PIP2(4′,5′) at phosphatidylinositolphosphate.

18“Molecular species level” pertains to all categories addressed here and is reached as soon as constituent fatty acyl/alkyl-residues are identified, e.g., TG 16:0_18:1_18:1, a triglyceride.

19“sn-position level” is a more refined level in GL and GP categories, enabling annotation of the sn-position of fatty acyl/alkyl constituents at the glycerol backbone as indicated by a slash, e.g., TG 16:0/18:1/18:1.

20“DB-position level” or “DBE-position level” pertain to species having constituents with defined position of double bonds or double bond equivalents, e.g., FA 18:2 (9, 11);O.

21“Structure defined level” annotates molecular species composed of various constituents and functional groups, yet without positions and stereochemical details, e.g., FA 18:2;OH.

22“Full structure level” annotates molecular species composed of various constituents and functional groups including positions, yet without stereochemical details, e.g., FA 18:2(9Z,11E);13OH.

23“Complete structure level” defines detailed structures of all functional groups including stereochemistry as shown in the LMSD, e.g., 13R-HODE, 13S-HODE (= common name).

24Figure 1 presents such a hierarchical scheme, taking the example of glucosylceramide.

25A word of caution is appropriate here: Annotations based solely on m/z features and on returns from database retrieval are frequently incorrect due to over-interpretation of experimental data, i.e., returns of chemically defined lipid molecules at Complete structure level. It is therefore of major importance that database search tools return appropriate annotations based on sum composition, i.e., at Species level and Molecular species level. Such tools are, for example, the LIPID MAPS MS search tools (https://lipidmaps.org/resources/tools/bulk_structure_searches_overview.php) (see also comment in the discussion) or the “ALEX lipid calculator” (http://alex123.info/ALEX123/MS.php).

26Lipid species are annotated by class shorthand abbreviation (see Tables 2A-6A), followed by a space and C-atoms:DBE, e.g., TG 54:5, or C-atoms:DBE;O-atoms in fatty acyl/alkyl residues, e.g., FA 18:1;O or PC 38:3;O2.Variable constituents like fatty acyls/alkyls are assigned based on their mass as number of C-atoms and number of DBE (C-atoms:DBE), when experimental proof for DB is provided the annotation is C-atoms:DB. Where applicable, the number of oxygen-atoms is added, separated by a semi-colon, e.g., C-atoms:DBE;O-atoms.DB-position is indicated by a number according to D -nomenclature (geometry unknown) or a number followed by geometry (Z for cis, E for trans). Specific techniques are required for determination of DB-position (or geometry) to validly use this level of annotation, e.g., FA 18:2 (9, 12), FA 18:2(9Z,12Z).Positions for all functional groups are stated in front of functional group abbreviation, e.g., FA 20:4;12OH.Generally, all functional groups (see for abbreviations) are separated by a semicolon after the number of DBE. Functional groups are placed inside a separate pair of parentheses, only if more than one followed by the number of groups, e.g., FA 20:3;(OH)2;oxo. Moreover, functional groups containing numbers such as NO2 or NH2 are generally placed inside a separate pair of parentheses, e.g., FA 18:1;(NO2). The order of functional groups follows the IUPAC hierarchy (14).Except for DBE/DB-position, proven positions of all other functional groups are stated according to D -nomenclature in front of the functional group abbreviation that are separated by a comma if more than one, e.g., FA 20:3(5Z,13E);11OH,15OH;9oxoCyclic structures cyX (X = number of ring atoms, see for abbreviations) are presented in front of other functional groups. Their structural details are annotated within a pair of square brackets. Within the square brackets the positions of ring atoms, separated by hyphen, are placed in front of the cyX annotation. Other functional groups are placed after the ring structure of the cyX annotation, e.g., FA 20:2;[8-12cy5;11OH;9oxo];15OH = 8-iso-PGE2 or PGE2.Carbohydrate structures, e.g., in complex glycosphingolipids, are annotated as described for glycans (https://www.ncbi.nlm.nih.gov/glycans) (15). When the sequence of sugars components is known they are shown in this order separated by a hyphen, e.g., Gal-Glc-Cer 18:1;O2/16:0. In case the sequence is unknown the components (followed by their number if more than one) are shown in alphabetic order in front of the respective lipid backbone, e.g., Gal2GlcCer 18:1;O2/16:0.Acyl-linkages (N- and/or O-) are annotated by FA C-atoms:DBE inside a separate pair of parentheses with proven position in front, e.g., Cer 18:1;O2/26:0;26O(FA 18:2).Alkyl-linkages (N- and/or O-) are annotated by C-atoms:DBE inside a separate pair of parentheses with proven position in front, e.g., FA 18:1(12Z);9O(16:1) for an ether lipid.When functional groups are part of lipid class abbreviation, e.g., PIP2 or SPBP, their proven positions are shown inside parentheses, separated by a comma if more than one, e.g., PIP2(4′,5′) 38:4 or SPBP (1) 18:1;O2.Greek letters are transcribed to Latin letters as follows: α to a, β to b, γ to g, δ to d, ω to w.Proven stereochemistry is shown after the respective functional group/side chain in square brackets [R] or [S], e.g., FA 20:4(6Z,8E,10E,14Z);5OH[S],12OH[R] = LTB4.

27Lipid species are annotated by class shorthand abbreviation (see Tables 2A-6A), followed by a space and C-atoms:DBE, e.g., TG 54:5, or C-atoms:DBE;O-atoms in fatty acyl/alkyl residues, e.g., FA 18:1;O or PC 38:3;O2.

28Variable constituents like fatty acyls/alkyls are assigned based on their mass as number of C-atoms and number of DBE (C-atoms:DBE), when experimental proof for DB is provided the annotation is C-atoms:DB. Where applicable, the number of oxygen-atoms is added, separated by a semi-colon, e.g., C-atoms:DBE;O-atoms.

29DB-position is indicated by a number according to D -nomenclature (geometry unknown) or a number followed by geometry (Z for cis, E for trans). Specific techniques are required for determination of DB-position (or geometry) to validly use this level of annotation, e.g., FA 18:2 (9, 12), FA 18:2(9Z,12Z).

30Positions for all functional groups are stated in front of functional group abbreviation, e.g., FA 20:4;12OH.

31Generally, all functional groups (see for abbreviations) are separated by a semicolon after the number of DBE. Functional groups are placed inside a separate pair of parentheses, only if more than one followed by the number of groups, e.g., FA 20:3;(OH)2;oxo. Moreover, functional groups containing numbers such as NO2 or NH2 are generally placed inside a separate pair of parentheses, e.g., FA 18:1;(NO2). The order of functional groups follows the IUPAC hierarchy (14).

32Except for DBE/DB-position, proven positions of all other functional groups are stated according to D -nomenclature in front of the functional group abbreviation that are separated by a comma if more than one, e.g., FA 20:3(5Z,13E);11OH,15OH;9oxo

33Cyclic structures cyX (X = number of ring atoms, see for abbreviations) are presented in front of other functional groups. Their structural details are annotated within a pair of square brackets. Within the square brackets the positions of ring atoms, separated by hyphen, are placed in front of the cyX annotation. Other functional groups are placed after the ring structure of the cyX annotation, e.g., FA 20:2;[8-12cy5;11OH;9oxo];15OH = 8-iso-PGE2 or PGE2.

34Carbohydrate structures, e.g., in complex glycosphingolipids, are annotated as described for glycans (https://www.ncbi.nlm.nih.gov/glycans) (15). When the sequence of sugars components is known they are shown in this order separated by a hyphen, e.g., Gal-Glc-Cer 18:1;O2/16:0. In case the sequence is unknown the components (followed by their number if more than one) are shown in alphabetic order in front of the respective lipid backbone, e.g., Gal2GlcCer 18:1;O2/16:0.

35Acyl-linkages (N- and/or O-) are annotated by FA C-atoms:DBE inside a separate pair of parentheses with proven position in front, e.g., Cer 18:1;O2/26:0;26O(FA 18:2).

36Alkyl-linkages (N- and/or O-) are annotated by C-atoms:DBE inside a separate pair of parentheses with proven position in front, e.g., FA 18:1(12Z);9O(16:1) for an ether lipid.

37When functional groups are part of lipid class abbreviation, e.g., PIP2 or SPBP, their proven positions are shown inside parentheses, separated by a comma if more than one, e.g., PIP2(4′,5′) 38:4 or SPBP (1) 18:1;O2.

38Greek letters are transcribed to Latin letters as follows: α to a, β to b, γ to g, δ to d, ω to w.

39Proven stereochemistry is shown after the respective functional group/side chain in square brackets [R] or [S], e.g., FA 20:4(6Z,8E,10E,14Z);5OH[S],12OH[R] = LTB4.

40Isotope-containing lipid structures are indicated in square brackets annotating the isotope, followed by the number of isotopic atoms, e.g., FA 18:1[13C5].• Multiple isotopes are separated by a comma, e.g., FA 18:1[13C5,D4].When positions of isotopes are known, they are indicated in a separate pair of parentheses in front of the isotope number, e.g., FA 18:1[(14,15,16,17,18)13C5].Isotopes in fatty acyls or alkyls and in sphingoid bases are indicated in square brackets after the number of DBE, e.g., PC 34:1[D9] or PC O-16:0_18:1[13C5] and in Cer 34:1;O2[13C3], respectively. Isotopes in head groups of these structures are indicated in square brackets after class shorthand abbreviation, e.g., PC[D9] 34:1, TG[13C3] 54:3, SM[D9] 34:1;O2.When positions of isotopes in the lipid are not known, they are indicated in square brackets in front of class shorthand abbreviation, e.g., [D5]PC 34:1, [13C7]TG 54:3.

41Isotope-containing lipid structures are indicated in square brackets annotating the isotope, followed by the number of isotopic atoms, e.g., FA 18:1[13C5].• Multiple isotopes are separated by a comma, e.g., FA 18:1[13C5,D4].

42When positions of isotopes are known, they are indicated in a separate pair of parentheses in front of the isotope number, e.g., FA 18:1[(14,15,16,17,18)13C5].

43Isotopes in fatty acyls or alkyls and in sphingoid bases are indicated in square brackets after the number of DBE, e.g., PC 34:1[D9] or PC O-16:0_18:1[13C5] and in Cer 34:1;O2[13C3], respectively. Isotopes in head groups of these structures are indicated in square brackets after class shorthand abbreviation, e.g., PC[D9] 34:1, TG[13C3] 54:3, SM[D9] 34:1;O2.

44When positions of isotopes in the lipid are not known, they are indicated in square brackets in front of class shorthand abbreviation, e.g., [D5]PC 34:1, [13C7]TG 54:3.

45Shorthand abbreviations for Fatty Acyl classes are stated in Table 2A.

46Table 2B shows that lowest resolution level is based on m/z values, i.e., annotation at Species level (low mass resolution MS, e.g., carboxylate anion and oxygen atoms from functional groups). In addition, it is assumed that only a straight-chain fatty acid with or without DBE(s) is present. High mass resolution with accurate mass measurements may identify additional elements such as oxygen atoms of functional groups. Thus, a limited amount of structural information is provided at this level of analysis following the rules alluded to in the Annotation of lipid structures section, i.e., Species level. Annotation at DB-position level requires techniques such as ozonolysis or photochemical derivatization or GC-MS. The use of trivial or common names for even simple fatty acids implies that additional methods have been used to define the exact structure, such as a straight-chain, positions of DBs, or DB geometries. Chiral chromatography preceding MS/MS is required for respective stereochemistry. Because this is generally not routinely done, investigators should note in their reports when using a common name for a fatty acid that “The identity and stereochemistry of the fatty acid species reported using a common name (e.g., oleic acid, linolenic acid, arachidonic acid, etc.) is assumed based on biological intelligence”. This comment applies to simple as well as complex lipids that include fatty acids as part of the structure (e.g., glycerophospholipids, triacylglycerols, etc.). Examples for shorthand notation of fatty acids are presented in Table 2B.

47Fatty acyl esters, i.e., wax esters (WEs), wax diesters (WDs), fatty acyl estolides (FAHFAs, FA-EST), as well as N-acyl amines (NAs) and N-acyl ethanolamines (NAEs) are shown in Table 2C.

48Lipidomic studies of “oxygenated fatty acyls,” commonly referred to as “oxylipins” or “oxygenated PUFAs” in the literature, involves analysis of enzymatically and nonenzymatically generated lipids such as octadecanoids, eicosanoids, docosanoids, do- and tetratriacontanoids (Table 2D–F). Enzymatically generated isomers include prostaglandins, leukotrienes, and the various “specialized pro-resolving mediators,” i.e., lipoxins, protectins, maresins, and resolvin D/Es (Table 2F). Nonenzymatic oxygenation of polyunsaturated fatty acids leads to numerous cyclic structures with various stereochemistry, such as phytoprostanes, isoprostanes, neuroprostanes, and all families of furans. Some of these isoprostanoids were identified over 25 years ago, particularly those of mammalian origin and more recently also as components in foods of plant origin. The nomenclature for isoprostanoids is based on Taber, Morrow, and Roberts and Rokach et al., an update appeared in 2010. Table 2G presents the precursor-product relationships for the classes of phytoprostanes, isoprostanes, and neuroprostanes, for which abbreviations PhytoP, IsoP, and NeuroP, respectively, have been proposed.

49Standards for structural validation by MS-inspection of these oxygenated fatty acids are described by Galano et al. and are in agreement with those referred to for oxylipins. Specific shorthand nomenclature has been previously suggested and widely used for polyunsaturated oxygenated fatty acids.

50The use of a common name (Table 2B, D, E) for fatty acyls or in reporting lipidomic studies also requires a high level of validation, typically with a representative biological sample using, for example, stable isotope dilution and chiral LC-MS/MS or capillary GC/MS with highly reproducible retention times for authentic standards. Otherwise, assumptions made on the basis of biological intelligence must be clearly stated.

51See Table 3A and B for class abbreviations and examples, respectively. Lipid class abbreviation followed by number of C-atoms:number of DBE, for oxygenated lipids C-atoms:DBE;O-atoms, are as described in the Annotation of lipid structures section.

52Glycerolipids with known fatty acyl/alkyl constituents (molecular species): separator _: sn-position of acyl/alkyl constituents is not known. Constituents are presented in the order of increasing number of C-atoms, as are DB (DBE)-numbers for each C-atom number, e.g., TG 16:0_18:1_18:3.separator /: sn-position of acyl/alkyl constituents is proven (order sn-1/sn-2/sn-3; no FA linked 0:0), e.g., TG 16:0/18:3/18:1.When only one acyl chain of TG is known, it is presented in front of the sum of the remaining two acyl residues, e.g., TG 16:0_36:3.When only one of the sn-positions is defined, this is indicated inside a pair of parentheses, e.g., TG 16:0_18:1(sn-2)_18:0.

53separator _: sn-position of acyl/alkyl constituents is not known. Constituents are presented in the order of increasing number of C-atoms, as are DB (DBE)-numbers for each C-atom number, e.g., TG 16:0_18:1_18:3.

54separator /: sn-position of acyl/alkyl constituents is proven (order sn-1/sn-2/sn-3; no FA linked 0:0), e.g., TG 16:0/18:3/18:1.

55When only one acyl chain of TG is known, it is presented in front of the sum of the remaining two acyl residues, e.g., TG 16:0_36:3.

56When only one of the sn-positions is defined, this is indicated inside a pair of parentheses, e.g., TG 16:0_18:1(sn-2)_18:0.

57Other bond types than ester bonds are indicated as follows in front of the sum of C-atoms for acyl/alkyl constituents:O = alkyl, e.g., TG O-52:3P = proven O-alk-1-enyl-bond (acid-sensitive ether bond in “neutral plasmalogens” is not counted as a DB/DBE within the acyl-chain), e.g., TG P-52:3 or at higher resolution TG P-16:0/18:3/18:1.More than one “non”-ester bond is indicated in front of the bond type as d for di, t for tri, and e for tetra.

58O = alkyl, e.g., TG O-52:3

59P = proven O-alk-1-enyl-bond (acid-sensitive ether bond in “neutral plasmalogens” is not counted as a DB/DBE within the acyl-chain), e.g., TG P-52:3 or at higher resolution TG P-16:0/18:3/18:1.

60More than one “non”-ester bond is indicated in front of the bond type as d for di, t for tri, and e for tetra.

61See Table 4A–C for abbreviations and examples. Shorthand notation for phospholipid species contains abbreviation for phospholipid classes, followed by number of C-atoms:number of DBE, i.e., PS 36:4, for oxygenated lipids C-atoms:DBE;O-atoms, i.e., PS 36:3;O, as described in the Annotation of lipid structures section.

62Molecular species of phospholipids with known fatty acyl/alkyl constituents (Table 4B): separator _: sn-position of acyl/alkyl constituents is not known. Order of constituent presentation as described for glycerolipids, e.g., PC 16:0_18:2.separator /: sn-position of acyl/alkyl constituents is proven (sn-1/sn-2 or sn-2/sn-3); no constituent 0:0; e.g., PC 16:0/18:2.For BMP and CL classes sn-position order will be sn-2/sn-3/sn-2′/sn-3′ and sn-1/sn-2/sn-1′/sn-2′, respectively.When only one acyl chain or DG moieties of CL are known, sum of acyl residues are presented, e.g., CL 16:0_54:3 and CL 34:1_36:2, respectively.

63separator _: sn-position of acyl/alkyl constituents is not known. Order of constituent presentation as described for glycerolipids, e.g., PC 16:0_18:2.

64separator /: sn-position of acyl/alkyl constituents is proven (sn-1/sn-2 or sn-2/sn-3); no constituent 0:0; e.g., PC 16:0/18:2.

65For BMP and CL classes sn-position order will be sn-2/sn-3/sn-2′/sn-3′ and sn-1/sn-2/sn-1′/sn-2′, respectively.

66When only one acyl chain or DG moieties of CL are known, sum of acyl residues are presented, e.g., CL 16:0_54:3 and CL 34:1_36:2, respectively.

67Lysophospholipid classes are abbreviated as stated in LIPID MAPS nomenclature (Table 4A). Molecular species with unknown sn-position are presented as, e.g., LPE 18:1, with known sn-position as LPE 18:1/0:0 (Table 4B).

68Other bond types than ester bonds are indicated as described for Glycerolipids, e.g., for an ether phospholipid PE O-18:0/18:2, for a “plasmalogen” PE P-18:0/20:4.

69It is described in the Annotation of lipid structures section, when functional groups are part of lipid class abbreviation, their proven positions are shown directly at the abbreviation’s end inside parentheses, separated by a comma if more than one. A prominent example is PIP3(3′,4′,5′). Table 4C shows that “Phosphate position level” identifies phosphate position at inositol ring, i.e., PIP(3′) 38:4, otherwise it would be PIP 38:4. For ease of handling by databases, numbers of phosphates are not written in lower case.

70The amino function in PSs and PEs, including their lysoforms, is prone to react with a variety of electrophiles as has been shown in recent years. The products are generally termed N-mod PL and N-mod LPL in abbreviated form, common names and respective abbreviations are shown in Table 4A. Structures at Species-, Molecular species-, and sn-Position levels are presented in shorthand notation as described in the Annotation of lipid structures, Glycerolipids (GL), and Glycerophospholipids (GP) sections; specific examples are shown in Table 4D.

71Phospholipids containing PUFA-constituents having methylene-interrupted cis-DBs (allylic DBs) and/or polar headgroups having amino-residues are susceptible to oxidation with formation of OxPLs. OxPL, so far, is a general term for a class of lipids produced by several processes that most often cannot be distinguished by MS analysis of the products. In all these cases, the products are called OxPLs.

72Respective modes for production are the following: Oxygenation of PL to produce OxPL by direct action of lipoxygenases on PUFA constituents of PL gives rise to enzymatically produced specific oxPL. The stereochemistry of the resulting PUFA component usually reflects the specificity of the specific enzyme involved (26).The Land’s cycle is an alternative mechanism for enzymatic OxPL formation. Free, unesterified PUFAs liberated by phospholipase A2 and other enzymatic pathways from PL are first oxygenated by lipoxygenases, cyclooxygenases or CYP450 oxygenases. The resulting oxygenated PUFAs can then be reesterified into PLs resulting in the indirect enzymatic formation of specific oxPL.Nonenzymatic reactions are induced by free-radical oxygen/nitrogen species reacting directly with the PUFA constituents of PL or with free PUFAs which become incorporated into the PL by acyl transferases producing nonenzymatically derived oxPL. This oxygen transfer to PUFAs can further lead to DB rearrangement, cyclization and even truncation of such acyl-chains resulting in complex mixtures of oxPL (27).Nonradical reactive oxygen species like singlet oxygen or ozone can also contribute to PL oxidation with generation of full-chain or fragmented oxPL.PL having a polar head group with a modified amino-function (PE and PS) form a subclass named oxPL-Nmod.

73Oxygenation of PL to produce OxPL by direct action of lipoxygenases on PUFA constituents of PL gives rise to enzymatically produced specific oxPL. The stereochemistry of the resulting PUFA component usually reflects the specificity of the specific enzyme involved (26).

74The Land’s cycle is an alternative mechanism for enzymatic OxPL formation. Free, unesterified PUFAs liberated by phospholipase A2 and other enzymatic pathways from PL are first oxygenated by lipoxygenases, cyclooxygenases or CYP450 oxygenases. The resulting oxygenated PUFAs can then be reesterified into PLs resulting in the indirect enzymatic formation of specific oxPL.

75Nonenzymatic reactions are induced by free-radical oxygen/nitrogen species reacting directly with the PUFA constituents of PL or with free PUFAs which become incorporated into the PL by acyl transferases producing nonenzymatically derived oxPL. This oxygen transfer to PUFAs can further lead to DB rearrangement, cyclization and even truncation of such acyl-chains resulting in complex mixtures of oxPL (27).

76Nonradical reactive oxygen species like singlet oxygen or ozone can also contribute to PL oxidation with generation of full-chain or fragmented oxPL.

77PL having a polar head group with a modified amino-function (PE and PS) form a subclass named oxPL-Nmod.

78Shorthand notation for OxPLs in general are presented in Table 4E.

79Apart from sphingosine containing 18 C-atoms with two hydroxyl groups and one DB, other sphingoid bases reveal prominent backbones as well, particularly in brain or nonmammalian specimens. Consequently, the abbreviation SPB is strongly recommended as shorthand notation for the general term “sphingoid bases,” Cer for ceramides, and SM for sphingomyelins (Table 5A). Table 5B, C, and D define, in addition, shorthand notation according to structural resolution of sphingolipids. The updated rules for shorthand notation are the following:In case the long-chain base is not known, the sum composition of sphingoid base and fatty acid is shown as number of C-atoms:DBE;O-atoms, e.g., SPB 34:1;O2.In ceramides the sphingoid backbone is annotated C-atoms:DBE;O-atoms separated by a slash from the number of C-atoms:DBE;O-atoms of the N-linked fatty acid, e.g., Cer 18:1;O2/16:0.DB geometry and positions of hydroxyl groups (or other functional groups) are annotated as described for fatty-acyl-chains in Tab. 2B, e.g., Cer 18:1(4E);1OH,3OH/16:0.When the number of hydroxyl groups cannot be determined, numbers of C-atoms and DBE are assigned under the assumption of the number of hydroxyl groups in the major sphingoid base for that organism (e.g., dihydroxy in mammals).For further characterization of N-linked fatty acids, the rules as described in the Annotation of lipid structures section apply. The position of a fatty acid esterified to an N-linked hydroxy-fatty acyl is shown in a separate pair of parentheses xO(FA C-atoms:DBE) with x denoting the position of hydroxyl group (Δ nomenclature) in the N-linked fatty acids, e.g., Cer 18:1;O2/26:0;18O(FA 16:0).Any modification linked to a sphingoid base-OH is written in front of the (sub)class abbreviation with the integrated position number in parenthesis at the end of abbreviation, e.g., FA 24:1-ACer (1) 18:1;3OH/16:0 for an acylceramide, Gal-Cer (1) 18:0;3OH/16:0 for a galactosylceramide.Consequently, in shorthand notation from “Structure defined level” onwards only unmodified OH-groups of the sphingoid base are annotated.Shorthand notation for carbohydrate moieties is stated in Table 1C and examples are shown in Table 5D.For annotation of the sugar moiety in complex glycosphingolipids we refer to current practice in glycan science (https://www.ncbi.nlm.nih.gov/glycans) (15). When the sequence of sugars components is known, they are shown in this order separated by a hyphen. In case the sequence is unknown the components (followed by their number if more than one) are shown in alphabetic order in front of the respective lipid backbone. Annotation of the ceramide part follows the rules described above.Sphingoid base phosphates with unknown phosphate position are represented by SPBP, e.g., SPBP 18:1;(OH)2.Sphingoid base phosphates with known position of phosphate and of OH-positions is annotated by, e.g., SPBP (1) 18:1(4E);3OH.Ceramide phosphates with unknown phosphate position are represented by CerP, e.g., CerP 18:1;O2/16:0.Ceramide phosphates with known position of phosphate and of OH-positions are annotated by, e.g., CerP (1) 18:1(4E);3OH.Ceramide phosphates with 1,3 cyclic phosphate and known OH-positions are annotated by, e.g., CerP (1, 3) 18:1(4E).

80In case the long-chain base is not known, the sum composition of sphingoid base and fatty acid is shown as number of C-atoms:DBE;O-atoms, e.g., SPB 34:1;O2.

81In ceramides the sphingoid backbone is annotated C-atoms:DBE;O-atoms separated by a slash from the number of C-atoms:DBE;O-atoms of the N-linked fatty acid, e.g., Cer 18:1;O2/16:0.

82DB geometry and positions of hydroxyl groups (or other functional groups) are annotated as described for fatty-acyl-chains in Tab. 2B, e.g., Cer 18:1(4E);1OH,3OH/16:0.

83When the number of hydroxyl groups cannot be determined, numbers of C-atoms and DBE are assigned under the assumption of the number of hydroxyl groups in the major sphingoid base for that organism (e.g., dihydroxy in mammals).

84For further characterization of N-linked fatty acids, the rules as described in the Annotation of lipid structures section apply. The position of a fatty acid esterified to an N-linked hydroxy-fatty acyl is shown in a separate pair of parentheses xO(FA C-atoms:DBE) with x denoting the position of hydroxyl group (Δ nomenclature) in the N-linked fatty acids, e.g., Cer 18:1;O2/26:0;18O(FA 16:0).

85Any modification linked to a sphingoid base-OH is written in front of the (sub)class abbreviation with the integrated position number in parenthesis at the end of abbreviation, e.g., FA 24:1-ACer (1) 18:1;3OH/16:0 for an acylceramide, Gal-Cer (1) 18:0;3OH/16:0 for a galactosylceramide.

86Consequently, in shorthand notation from “Structure defined level” onwards only unmodified OH-groups of the sphingoid base are annotated.

87Shorthand notation for carbohydrate moieties is stated in Table 1C and examples are shown in Table 5D.

88For annotation of the sugar moiety in complex glycosphingolipids we refer to current practice in glycan science (https://www.ncbi.nlm.nih.gov/glycans) (15). When the sequence of sugars components is known, they are shown in this order separated by a hyphen. In case the sequence is unknown the components (followed by their number if more than one) are shown in alphabetic order in front of the respective lipid backbone. Annotation of the ceramide part follows the rules described above.

89Sphingoid base phosphates with unknown phosphate position are represented by SPBP, e.g., SPBP 18:1;(OH)2.

90Sphingoid base phosphates with known position of phosphate and of OH-positions is annotated by, e.g., SPBP (1) 18:1(4E);3OH.

91Ceramide phosphates with unknown phosphate position are represented by CerP, e.g., CerP 18:1;O2/16:0.

92Ceramide phosphates with known position of phosphate and of OH-positions are annotated by, e.g., CerP (1) 18:1(4E);3OH.

93Ceramide phosphates with 1,3 cyclic phosphate and known OH-positions are annotated by, e.g., CerP (1, 3) 18:1(4E).

94We use the term sterol to embrace all molecules based on the cyclopentanoperhydrophenanthrene skeleton. In the case of sterols, the ring system does not add to the number of DBE. Endogenously biosynthesized mammalian sterols are derived from cholesterol or its precursors, yet plant and yeast sterols can also be a source via the food chain. The stereochemistry of the cholesterol molecule is maintained to a large extent by mammalian sterols, which all contain at least one hydroxyl or oxo group attached to carbon 3. High resolution MS with accurate mass may identify other functional groups, as will MS/MS or MSn scans. Stereochemistry can often be defined by comparing the chromatographic retention time to authentic standards and, in some cases, by MS/MS or MSn. The class abbreviations within category ST are shown in Table 6A.

95The following rules for shorthand nomenclature have been adopted in the examples given in Table 6B.In shorthand notation the category abbreviation ST is used as class abbreviation. In some cases, other abbreviations e.g., FC, CE, BA, SE, SG and ASG can be used. In all cases, class abbreviation is followed by number of carbon atoms:number of DB, and separated by semicolon is the number of oxygens, e.g., ST 27:1;O for cholesterol and lathosterol (also zymostenol), or ST 24:1;O5 for an oxidized sterol and for cholic acid and ursocholic acid. The latter is an important point: Some bile acids have an identical mass and molecular formula to oxidized sterols lacking a carboxylic acid group. This must be considered, when class abbreviation “BA” is used.Shorthand notation of further functional groups are written, separated by a semicolon, after the number of oxygens, e.g., BA 24:1;O5;T for taurocholic acid (= common name, abbreviation TCA).Following the number of double bonds, proven position and stereochemistry is shown. R and S configurations are preferred for side-chain stereochemistry and are shown in square brackets. α (below ring/plane), written as a, and β (above ring/plane), written as b, are preferred for ring stereochemistry, e.g. 3aOH and 17bOH. Stereochemistry at C-5 introduced by reduction of the Δ5 bond is indicated by 5aH or 5bH. Replacing the number of oxygens, proven positions and stereochemistry of oxygen containing functional groups are shown. If such stereochemistry is known the common name of the compound can be used.The side-chain at carbon-17 of the cyclopentanoperhydrophenanthrene skeleton always has b-stereochemistry (17b) and consequently is not presented in the shorthand annotation.For structures fully proven or based on assumption by biological intelligence, such as e.g., cholesterol, cholesteryl esters, steryl esters, bile acids, sterylglycosides, and acylsterylglycosides abbreviations FC, CE, SE, BA, SG and ASG, respectively, can be used as shown in Table 6A. CE is followed by number of C-atoms:number of DBE of the fatty acid esterified to the hydroxyl group at position 3, e.g., CE 18:2 (Table 6B). Shorthand notation SE is used as above followed by slash (for monohydroxysterols) or underscore (for polyhydroxysterols) number of C-atoms:number DBE of the fatty acid esterified to the hydroxyl group (Table 6B).MS/MS scans reveal the presence of conjugates: Taurine (T) and glycine (G) each are conjugated through an amide bond to the carboxylic acid group of bile acids, respective amide bonds with conjugates are designated in shorthand notation “COT” and “COG” (Table 6B); sulfuric acid (S) is conjugated to a hydroxyl group through an ester bond; glucuronic acid (GlcA), N-acetylglucosamine (GlcNAc), and hexose (Hex) sugars are assumed to be linked to a hydroxyl group through an acetal linkage (Table 6B).In the case full stereochemistry is known the common names as presented in Table 6B can be used.

96In shorthand notation the category abbreviation ST is used as class abbreviation. In some cases, other abbreviations e.g., FC, CE, BA, SE, SG and ASG can be used. In all cases, class abbreviation is followed by number of carbon atoms:number of DB, and separated by semicolon is the number of oxygens, e.g., ST 27:1;O for cholesterol and lathosterol (also zymostenol), or ST 24:1;O5 for an oxidized sterol and for cholic acid and ursocholic acid. The latter is an important point: Some bile acids have an identical mass and molecular formula to oxidized sterols lacking a carboxylic acid group. This must be considered, when class abbreviation “BA” is used.

97Shorthand notation of further functional groups are written, separated by a semicolon, after the number of oxygens, e.g., BA 24:1;O5;T for taurocholic acid (= common name, abbreviation TCA).

98Following the number of double bonds, proven position and stereochemistry is shown. R and S configurations are preferred for side-chain stereochemistry and are shown in square brackets. α (below ring/plane), written as a, and β (above ring/plane), written as b, are preferred for ring stereochemistry, e.g. 3aOH and 17bOH. Stereochemistry at C-5 introduced by reduction of the Δ5 bond is indicated by 5aH or 5bH. Replacing the number of oxygens, proven positions and stereochemistry of oxygen containing functional groups are shown. If such stereochemistry is known the common name of the compound can be used.

99The side-chain at carbon-17 of the cyclopentanoperhydrophenanthrene skeleton always has b-stereochemistry (17b) and consequently is not presented in the shorthand annotation.

100For structures fully proven or based on assumption by biological intelligence, such as e.g., cholesterol, cholesteryl esters, steryl esters, bile acids, sterylglycosides, and acylsterylglycosides abbreviations FC, CE, SE, BA, SG and ASG, respectively, can be used as shown in Table 6A. CE is followed by number of C-atoms:number of DBE of the fatty acid esterified to the hydroxyl group at position 3, e.g., CE 18:2 (Table 6B). Shorthand notation SE is used as above followed by slash (for monohydroxysterols) or underscore (for polyhydroxysterols) number of C-atoms:number DBE of the fatty acid esterified to the hydroxyl group (Table 6B).

101MS/MS scans reveal the presence of conjugates: Taurine (T) and glycine (G) each are conjugated through an amide bond to the carboxylic acid group of bile acids, respective amide bonds with conjugates are designated in shorthand notation “COT” and “COG” (Table 6B); sulfuric acid (S) is conjugated to a hydroxyl group through an ester bond; glucuronic acid (GlcA), N-acetylglucosamine (GlcNAc), and hexose (Hex) sugars are assumed to be linked to a hydroxyl group through an acetal linkage (Table 6B).

102In the case full stereochemistry is known the common names as presented in Table 6B can be used.

103This publication updates both the classification and nomenclature and shorthand notation, and targets two goals. First, to emphasize and enable correct reporting of mass spectrometric data according to the resolving power of MS instrument platforms operating in high-resolution (and often high-throughput) mode. Second, to provide a comprehensible shorthand notation for the lipids commonly analyzed. Such common nomenclature is essential for standardized reporting of lipid species data and construction of data resources. Moreover, standardized data facilitate automated datamining and import into databases by script-based algorithms with only minimal data curation. Related data repositories require a hierarchical concept mirroring the structural resolution provided by mass spectrometric analysis reflected in the presented shorthand notation. To this end, the LMSD database, respective MS search tool, and, in particular, shorthand notations for all relevant lipids are now available on the LMSD detail view pages at “Species level” and “Molecular species level”, the latter embracing “Phosphate-”, “DB-”, and “sn-position level”. In a few instances, however, easy use of this shorthand notation by lipidomics experts has priority over its stringent use in a bioinformatics format.

104A standardized annotation for lipid species, as a common language, is a key component to promote and further advance this emerging omics discipline. Therefore, the Lipidomics Standards Initiative (LSI; https://lipidomics-standards-initiative.org/) has been recently introduced, pursuing development of guidelines and channeling community-wide efforts in close collaborations with LIPID MAPS (https://www.lipidmaps.org/) as has been emphasized recently. In addition, alignment with other initiatives, as for example, adaptation of mzTab-M, a data format developed for metabolomics, to the presented nomenclature is possible.

105In summary, the shorthand nomenclature presented here is viewed as a standard in lipidomics that can be updated periodically.

106All data are contained within this article.