Multiple stage linear ion-trap mass spectrometry toward characterization of native bacterial lipids-a critical review

Abstract

Great strides in the field of lipidomics driven by advances in mass spectrometry techniques in the last decade have moved lipid analysis to a new level and significantly improved our understanding of lipid biochemistry. Multiple stage mass spectrometry (MSⁿ) with high resolution mass spectrometry (HRMS) that allows sequential isolation, fragmentation, and recognition of ion structures, is a powerful tool for characterization of complex and diversified lipid in bacterial cells, in which lipids are often critical for cell aggregation and dissociation, and play important biological roles. In addition to common phospholipids, many bacteria contain unique lipids that are specific to the bacterium genus and even to the bacterium species. In this review, application of linear ion-trap (LIT) MSⁿ in the structural characterization of native bacterial lipids including (1) novel lipids consisting of many isomeric structures, (2) lipids with unique functional groups and modification, (3) complex sphingolipids, peptidolipids, and lipocyclopeptides from various bacteria are presented. LIT MSⁿ approach affords realization of the mechanisms underlying the fragmentation processes, resulting in identification of complex lipid structures that would be very difficult to define using other analytical methods.

Keywords: Bacterial lipid; Fragmentation mechanisms; Linear ion-trap multiple stage mass spectrometry; Lipidomics; Lipopeptide; Rearrangement; Sphingolipid.

Fig. 1.

The High resolution CID LIT MS² spectrum of the [M − H]⁻ ion of 1-hexadecanoyl, 2-β-tetradecanyloctadecanoyl-PG (16:0/16:0–16:0-MA PG) at m/z 961.7 (a), its MS³ spectrum of the ion of m/z 495 (961 → 495)(b) that defines the 16:0–16:0-mycolic acid substituent located at sn-2. The high resolution mass measurements of the fragment ions and the deduced elemental compositions, and ion structures are listed in the inset table of Panel a. See Scheme 1 for the fragmentation processes; and see text for assignment of the regiospecificity of the fatty acyl chains on the glycerol backbone [14].

Fig. 2.

The CID MS² spectrum of the [M − H]⁻ ion of lysyl-(17:0/15:0)(15:0/15:0)-CL at m/z 1451 (a), its MS³ spectra of the ion of m/z 1323(1451 → 1323) (b), of m/z 1249 (1451 → 1249) (c), deriving from 1451 by internal loss of lysylglycerol, and the MS⁴ spectrum of the ion of m/z 709 (1451 → 1249 → 799) (d) that supports the suggested internal loss of lysylglycerol pathway (Scheme 2b), and points to the identity and location of the 4 fatty acyl chains on the glycerol backbone. Fig. 2b is identical to the MS² spectrum of-(17:0/15:0)(15:0/15:0)-CL, indicating that the lysyl-CL possess the identical normal CL structure after cleavage of the lysine residue [20]. “Panel a” is a HR MS² spectrum that affords assessment of elemental compositions of all ions including m/z 1249, which contains a “lysyl-Gly” residue shorter (i.e., loss of lysyl-Gly) than m/z 14451. Please note: all ions denoted are nominal m/z (excepting precursor ions in Panel a–c).

Fig. 3.

The CID MS² spectrum of the [M − H]⁻ ion of a DHC-PIP-DHC lipid at m/z 1414.0 (a), its MS³ spectrum of the ion at m/z 876 (1414.0 → 876.5) (b), its MS⁴ spectra of the ions at m/z 634.5 (1414.0 → 876.5 → 634.5) (c), and at m/z 778 (1414.0 → 876.5 → 778.5) (d) that lead to identify d18:1/βh16:0-Cer-PIP-d18:1/βh16:0, d17:1/βh17:0-Cer-PIP-d17:1/βh17:0-Cer, and d18:1/βh16:0-Cer-PIP-d17:1/βh17:0-Cer isomers. The MS³ spectrum of the ion at m/z 890.5 (1414.0 → 890.5) (e), and its MS⁴ spectrum of the ion at m/z 648 (1414.0 → 890.5 → 648.5) (f) possess the similar profiles as Panel b and Panel c, respectively, supporting the presence of d18:1/βh17:0-PIP-d17:1/βh16:0-Cer isomer (see text and Scheme 3 for details). Please note: Panel a is a HR HCD MS² spectrum; and all ions are denoted as nominal mass.

Fig. 4.

The high-resolution CID MS² spectrum of the [M − H]⁻ ion of 15:0/βh16:0-GS lipid at m/z 639 (b), its HR HCD MS³ spectrum of the ion of m/z 397 (639 → 397) (b). The spectrum (Panel a) is dominated by the ion of m/z 397, arising from loss of the “piggyback” 15:0-fatty acyl chain (loss of 242 DA) to form a N-(α,β-unsaturated acyl)-GS (Scheme 4), which undergoes further dissociation (Panel b) to yield the signature amino acid anions at m/z 104 for serine (MW 105), at 74 for glycine (MW 75) and at m/z 131 for glycylserine (MW 132) (labeled in red) (Scheme 4). Panel a also contains minor ion at m/z 411 arising from analogous loss of the “piggyback” 14:0-fatty acyl chain (loss of 228 Da) on a βh17:0-FA substituent, indicating that a minor 14:0/βh17:0-GS isomer is also present. All ions are denoted as nominal m/z.

Fig. 5.

The MS² spectrum of the ion of m/z 1174(a), its MS³ spectrum of m/z 1027 (1174 → 1027)(b), MS⁴ spectrum of m/z 914 (1174 → 1027 → 914)(c), MS⁵ spectrum of m/z 813 (1174 → 1027 → 914 → 813)(d), MS⁶ spectrum of m/z 700 (1174 → 1027 → 914 → 813 → 700)(e), MS⁷ spectrum of m/z 587 (1174 → 1027 → 914 → 813 → 700 → 587)(f), and its MS⁸ spectrum of m/z 265 (1174 → 1027 → 914 → 813 → 700 → 587 → 265) (g) that reveal the peptide sequence. The structure of this cyclo-peptidolipid is recognized by first cleavage of the lactone bond to release an aldehyde (C₂₁H₄₁CHO) from the 3-hydroxy tetraeicosenoyl chain. In panel a, the ion of m/z 1027 arises from loss of Phe, and the ion of m/z 852 arises from β-cleavage of 3-hydroxy tetraeicosenoyl substituent to eliminate a C₂₁H₄₁CHO residue, reflexing the attachment of a β-OH 24:1-fatty acyl group to the N-terminal of peptide backbone (Scheme 5). These LIT MSⁿ (n = 2 to 8) spectra define the Phe-Ile-Thr-Ile-Leu-Ser-Leu peptide sequence by sequential cleavage of the amino acid residues. The MS⁷ spectrum of m/z 587 (Panel f) also contains the ion of m/z 265 arising from loss of C₂₁ H₄₁ CHO (an aldehyde) residue, further supporting the attachment of a β-OH 24:1-fatty acyl substituent to the N-terminal. Therefore, a complex cyclopeptidolipid structure (Scheme 5) can be determined simply applying LIT MSⁿ on the (M + Na]⁺ ion of a native lipid. Please note: Panel a is a HR HCD MS² spectrum, which confirms partial amino acid sequence and the attachment of lipid residue, while the complete amino acid sequence is deciphered by the unit-mass resolved sequential MSⁿ (n = 3 to 8) spectra (Panel 3–8).

Fig. 6.

The MS² spectrum of the [M − H]⁻ ion at m/z 728(a), and its MS³ spectrum of the ion of m/z 552 (728 → 552) (b), leading to define GlcA-d20:1/αh14:0-Cer structure, which consist of an unique LCB structure where the double bond is situated at C13 of the dS-20:1-LCB chain. This information is revealed by MS³ spectrum of the ion of m/z 566 (742 → 566) (c) arising from the MS² on the corresponding [M − H + 2Li]⁺ ion at m/z 742 (not shown), and gives rise to m/z 340 by loss of fatty acyl substituent (see Scheme 6). MS⁴ on m/z 340 (742 → 566 → 340) (Panel d) yields ion of m/z 308 by loss of (H₂ + HCHO), and the MS⁵ spectrum of m/z 308 (742 → 566 → 340 → 308) (Panel e) yields ions informative for locating the double bond on the LCB (Scheme 6). Panel f shows the MS⁴ spectrum of the ion of m/z 322 (756 → 580 → 356 → 322) from the [M − H + 2Li]⁺ ions of GalA-dS21:1/βh14:0-Cer at m/z 756, a homologous ion species in the same lipid family containing a cyclopropyl ring situated at C13 on the LCB (Panel f, inset), rather than a double bond. Panel a is a HR HCD MS² spectrum, from which the extracted elemental compositions lead to define GlcA-d20:1/αh14:0-Cer; while MSⁿ on the corresponding [M − H + 2Li]⁺ ions (Panel c–f) provide complementary information to achieve complete structure identification.

Scheme 1.

The fragmentation pathways proposed for the [M − H]⁻ ion of 1-hexadecanoyl, 2-(16:0/16:0) mycolyl PG. The mycolyl substituent at sn-2 is recognized by the fragmentation process unique to the mycolic acid as shown.

Scheme 2.

The fragmentation processes proposed for the [M − H]⁻ ion of (17:0/15:0)(15:0/15:0)-lysyl-CL at m/z 1451, which undergoes fragmentation to eliminate lysyl residue to yield (17:0/15:0)(15:0/15:0)-CL (a). An internal loss of lysylglycerol to form diphosphatidic acid at m/z 1249 is evidenced by the further fragmentation processes that form the diagnostic ions as shown in Scheme (b) (adopted from Ref. [20], with permission).

Scheme 3.

The fragmentation processes proposed for the [M − H]⁻ ion of d18:0/βh16:0-Cer-PIP-d18:0/βh16:0-Cer at m/z 1414, which also consists of d17:0/βh17:0-Cer-PIP-d18:0/βh16:0-Cer and possible d17:0/βh17:0-Cer-PIP-d17:0/βh17:0-Cer isomers. The latter isomers are recognized by the presence of d17:0/βh17:0-Cer anion at m/z 634, which forms ions of m/z 408 and 422 by loss of the βh17:0- and βh16:0-acyl residue as an aldehyde, respectively. Another d18:1/βh17:0-PIP-d17:1/βBh16:0-Cer isomer is realized by the observation of 890/862 pair (Fig. 3a), and their MS³ (Fig. 3e) and MS⁴ (Fig. 3f) spectra.

Scheme 4.

The fragmentation processes proposed for the [M − H]⁻ ion of 15:0-βh16:0-GS, where the fatty acyl chain is attached to the N-terminal of glycine (serine is at the C-terminus). The loss of Ser (loss as [Ser - H₂O) requires a prior rearrangement process as shown. The presence of Gly and Ser is recognized by observation of the individual amino acid substituent anions (illustrated in red). The piggyback fatty acid (15:0) is mainly in anteiso form (anteiso:iso, 2:1) (see ref. for details).

Scheme 5.

The fragmentation processes proposed for the [M + Na]⁺ ion of lipocyclopetide with the deduced structure. The peptide sequence is identified by the MSⁿ (n = 2 to 8) spectra which contain ions from sequential losses of amino acid residues (Fig. 4a–g); the β-OH24:1-FA residue linked to the Leu N-terminal is recognized by the facile loss of the FA chain as an aldehyde. This aldehyde loss is also seen in each of the MSⁿ spectra (label in purple in Fig. 4a–g).

Scheme 6.

(a). The fragmentation pathways proposed for the (a) the [M − H]⁻ ion of d20:1/βh14:0-GlcACer at m/z 728. The scheme shows the formation of the β-OH 14:0-carboxylate anion, diagnostic to the fatty acyl chain, involves a rearrangement process; the cleavages of the Glc and LCB that define the glycoside and LCB. Scheme (b) shows the fragmentation pathways of the corresponding [M − H + 2Li]⁺ ion that leads to locate the double bond position of the LCB via charge-remote fragmentation mechanism.