Application of hybrid surface technology for improving sensitivity and peak shape of phosphorylated lipids such as phosphatidic acid and phosphatidylserine

Journal of Chromatography A 1669 (2022) 462921

Contents lists available at ScienceDirect

Journal of Chromatography A
journal homepage: www.elsevier.com/locate/chroma

Application of hybrid surface technology for improving sensitivity and
peak shape of phosphorylated lipids such as phosphatidic acid and
phosphatidylserine
Giorgis Isaac a,∗, Ian D Wilson b, Robert S. Plumb a
a
b

Scientific Operations, Waters Corporation, Milford, MA, 01757, USA
Division of Systems Medicine, Dept Metabolism, Digestion and Reproduction, Imperial College, South Kensington, London, SW7 2AZ, UK

a r t i c l e

i n f o

Article history:
Received 8 November 2021
Revised 21 February 2022
Accepted 24 February 2022
Available online 25 February 2022
Keywords:
Lipidomics
Metal interaction
Surface treatment
Acidic phospholipids

Phosphatidic acid
Phosphatidylserine

a b s t r a c t
The use of hybrid surface technology (HST), applied to the metal surfaces of an ACQUITYTM UPLCTM system and column, designed to mitigate the chelation, poor peak shape and analyte loss seen with acidic
phospholipids was investigated. Compared to a conventional system significant improvements in both
sensitivity, recovery and peak shape were obtained following UPLC on a CSH C18 column when the HST
was used for the analysis of lysophosphatidic acid (LPA), phosphatidic acid (PA), lysophosphatidylserine
(LPS), phosphatidylserine (PS), phosphatidylinositol-monophosphates (PIP), ceramide phosphate (CerP)
and sphingoid base phosphate (SPBP). The benefits in chromatographic performance provided by the HST
were seen particularly at low concentrations of these analytes. The HST system and column reduced peak
tailing by 65–80% and peak width by 70–86% for LPA and PA. Moreover, increased signal intensities of up
to 12.7 times were observed for LPA with the HST approach compared to the equivalent untreated LC system and column. The application of this methodology to the analysis of chicken egg PA and brain porcine
PS extracts were accompanied by similar improvements in data quality.
© 2022 Waters Corporation. Published by Elsevier B.V.
This is an open access article under the CC BY license ( />
1. Introduction
Phospholipids (PLs) are one of the main components of cell
membranes and perform many important biological functions such
as cell signaling, regulation of apoptosis, lipid synthesis and transport [1–5]. The importance of PLs has led to an increased focus
on analytical methods for the characterization of both lipid phenotypes (lipidomics) and the individual molecular species. However, analysis of PLs is difficult because they exist in cells and
biological fluids as complex mixtures of lipids formed from different combinations of polar headgroups and hydrophobic fatty
acyl chains. The extreme structural diversity of lipids in real biological samples is challenging for analytical techniques due to
the large difference in physicochemical properties of the individual lipid species which may be present as acids or bases, and neutral, polar or non-polar forms. Currently the two main analytical
strategies used for lipidomics are direct infusion mass spectrometry (DIMS) and liquid chromatography coupled to mass spectrometry (LC/MS). DIMS can provide high-throughput analysis but suf-

∗

Corresponding author at: Waters Corporation, USA.
E-mail address: (G. Isaac).

fers from the inability to separate and resolve isobaric/isomeric
species and to identify low abundant species due to ion suppression [6,7]. For reasons of sample compatibility, specificity and selectivity LC/MS has become increasingly important for the identification, and/or quantification of lipids directly from biological
extracts. The chromatographic separation can be performed using normal-phase (NP), reversed-phase (RP), hydrophilic interaction chromatography (HILIC) [6], or via supercritical fluid chromatography (SFC); (these different analytical approaches are summarized in reference [8]). NP methods, employing solvents such
as hexane and chloroform are used for the separation of lipid
classes. However, NP separations often have long analysis times,
can demonstrate limited analyte solubility, use toxic solvents, and
lack solvent compatibility with mass spectrometry making the
technique increasingly less attractive. HILIC and SFC, have similar
characteristics to NPLC and can be used for the separation of lipid
classes with organic/water-based solvent systems for HILIC and liquid carbon dioxide as primary mobile phase with co-solvents for
SFC, overcoming many of the challenges of the NP mode [8–10].
The successful separation of phosphatidic acids (PA) and phosphatidylserines (PS) using HILIC has been previously reported [11–
18] and SFC has demonstrated advantages in short analysis time
for the separation of lipid classes. RPLC on the other hand is

/>0021-9673/© 2022 Waters Corporation. Published by Elsevier B.V. This is an open access article under the CC BY license ( />

G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

widely used for the separation of individual lipid molecular species
with separations based on their lipophilicity; a characteristic governed by the carbon chain length and number of double bonds
[19–21].
The PA and PS classes tend to elute as broad peaks under the
RPLC conditions generally used in lipidomics studies [19–22] and
this can, at least in part, be attributed to the metal-analyte interactions of these lipids, particularly the phosphate moieties, with
the metal surfaces of the LC system and the column. Thus, the
lipid classes that show these poor chromatographic properties typically contain functional groups such as phosphate or carboxylate,

groups that can form chelation complexes with iron and other
transition metal ions. Phosphorylated and carboxylate lipids encompass classes such as the PA, PS, phosphatidylinositols (PI), PI
with mono-, bis and tris-phosphates, phosphatidylglycerols, phosphorylated sphingolipids including ceramide-phosphate (CerP) and
sphingoid base-phosphates (SPBP). All these lipids are metal sensitive and readily adsorb to stainless steel surfaces present within
the flow path of the chromatographic system via a Lewis acidbase interaction, leading to the aforementioned poor peak shape
as well as low recovery and a reduced sensitivity compared to
non-interacting lipids. Mitigation of these interactions has, in the
past, involved the addition of metal chelators to the mobile phase
or sample. Volatile chelators such as acetylacetone, citric, phosphonic or medronic acids have been used for LC/MS analyses
[22–27]. However, the use of these chelators can negatively impact both chromatographic selectivity and MS sensitivity. Lee et al.
used carbamate-embedded C18 as the stationary phase to optimize
peak tailing and separation efficiencies of acidic lipids [28]. Alternatively, columns constructed from a metal free material such
as polyether ether ketone (PEEK) have been described for e.g.,
the determination of SPBP and CerP [29]. However, the application of PEEK is limited by its low-pressure tolerance and whilst
coating stainless steel LC tubing and column walls with PEEK
can provide a partial solution to this problem the issue of the
poor pressure tolerance of frits prepared from PEEK would remain a challenge. To address these issues, we have been encouraged by recent publications [30–32] to investigate the use
of a hybrid surface technology (HST) applied to the metal substrates in UltraPerformance LC (UPLCTM ) instruments and columns.
This HST is based on an ethylene-bridged siloxane polymer which
has been demonstrated to be well-suited for RP and HILIC. For
a detailed description and characteristics of the HST see reference [31]. Based on our previous experience of the analysis of
lipids [19] we have evaluated a UPLC system and ACQUITYTM CSH
C18 column incorporating the HST and compared the results obtained against those from a conventional UPLC system and ACQUITY CSH C18 column for the analysis of phosphorylated and
carboxylate lipids (mainly PA, PS, lysophosphatidic acids (LPA),
lysophosphatidylserines (LPS), CerP, SPBP and PI-monophosphate
(PIP)).

egg chicken PA and brain porcine PS extracts were purchased from
Avanti Polar Lipids (Birmingham, AL, USA).
2.2. Standard solutions

A stock solution was prepared for PA 16:0/18:1, LPS 18:1/0:0
and PS 16:0/18:1 in chloroform at 1 mg/mL. The stock solution
for LPA 16:0/0:0 was prepared at a concentration of 0.5 mg/mL in
–CHCl3 -MeOH-H2 O (80:20:2, v/v/v) with gentle heating at 40 °C
and sonication to facilitate complete dissolution. A working mixture dilution series of 5, 50 and 250 pg/μL was prepared in IPAACN (50:50, v/v) for the lipid standards. CerP 18:1;O2/16:0 and
SPBP 18:1;O2 were prepared in CHCl3 -MeOH (1:1, v/v) and PIP
(4 ) 18:1/18:1 in –CHCl3 -MeOH-H2 O (1:1:0.3, v/v/v). Powdered egg
chicken PA and brain porcine PS extracts from Avanti Polar Lipids
(Birmingham, AL, USA) were dissolved in CHCl3 at a concentration
of 1 mg/mL. Working solutions of the egg chicken PA and brain
porcine PS extracts were prepared in IPA-ACN (50:50, v/v) at a concentration of 10 ng/μL. The differential ion mobility system suitability Lipidomix lipid standards contained 1 mg/mL of each lipid
and (0.25 mg/mL PI) in CHCl3 -MeOH (1:1, v/v). It contained a certified mixture of 14 lipids comprising LPC 18:1/0:0, PC 14:1/14:1, PE
14:1/14:1, PS 14:1/14:1, PG 14:1/14:1, PI 14:1/14:1, PA 14:1/14:1, SM
d18:1/18:1, Cer d18:1/18:1, cholesterol, CE 19:0, DG 14:1/14:1, TG
(18:1/18:1/18:1) and CL 14:1/14:1/14:1/14:1. A stock solution was
prepared in IPA-ACN (50:50, v/v) at 0.05 mg/mL for each lipid and
0.0125 mg/mL for PI. A working Lipidomix lipid standard 100 pg/μL
solution (25 pg/μL PI) was prepared in IPA-ACN (50:50, v/v). All
operations involving preparing CHCl3 -based solutions were carried
out in a fume hood.
2.3. Chromatographic separation and mass spectrometry conditions
Chromatographic separations were performed using two separate, but similar, chromatography systems. Either a conventional
ACQUITY UPLC I-Class chromatography system coupled with ACQUITY CSH C18 column (2.1 × 100 mm, 1.7 μm, p/n 186005297)
or an ACQUITY Premier UPLC System with ACQUITY Premier CSH
C18 Column (2.1 × 100 mm, 1.7 μm, p/n 186009461) constructed
with HST (both columns packed with the same batch of stationary
phase) were employed. Both systems were equipped with a binary
solvent delivery system and flow through needle (FTN) autosampler (Waters Corporation, Milford, MA). The mobile phase consisted
of (A) –ACN-H2 O-1 M aqueous ammonium formate (600:390:10,
v/v/v) in 0.1% FA and (B) IPA-ACN-1 M aqueous ammonium formate

(900:90:10, v/v/v) in 0.1% FA. The chromatography columns were
operated at a flow rate of 0.4 mL/min and eluted using a multiphase gradient (Table S1) where the initial conditions were 1:1 A:B
the mobile phase composition B was then gradual increased over
11 min using the following profile; 50–53% B from 0 to 0.5 min,
53–55% B 0.5–4 min, 55–65% B 4–7 min, held at 80% B 7–7.5 min,
80–99% B 7.5–10 min, held at 99% B 10–11 min, followed by returning to the starting conditions for re-equilibration giving a total
run time of 12 min. Various injection volumes were used, with a
sample and column temperature of 10 °C and 55 °C respectively.
The column effluent was monitored using a SYNAPTTM XS mass
spectrometer (Waters Corporation, Wilmslow, UK) operated in “V”
mode and negative electrospray ionization (ESI) using the following instrument settings: capillary 2.5 kV, desolvation temperature
500 °C, desolvation gas 800 L/h, cone gas 50 L/h, cone voltage 30 V,
and source temperature 120 °C. The MS data were acquired using
data independent acquisition in resolution and continuum mode
with a mass range 10 0−120 0 m/z. The lock mass was acquired and
applied using a solution of 200 pg/μL leucine enkephalin. The low
collision energy was set to 6 eV, and for high energy, a transfer linear ramp of 25−45 eV was used. LC/MS data acquisition and pro-

2. Experimental
2.1. Chemicals and standards
Acetonitrile (ACN), isopropanol (IPA), formic acid (FA), chloroform (CHCl3 ), methanol (MeOH) and ammonium formate (LC/MS
grade) were purchased from Fisher Scientific (Pittsburgh, PA, USA).
Total recovery vials were supplied by Waters Corporation (Milford,
MA, USA). Ultrapure water was obtained, in-house, from an Elga
PURELABTM Flex water purification system (High Wycombe, England, U.K.). LPA 16:0/0:0, PA 16:0/18:1, LPS 18:1/0:0, PS 16:0/18:1,
CerP 18:1;O2/16:0, SPBP 18:1;O2, PIP (4 ) 18:1/18:1 and differential ion mobility system suitability LipidomixTM lipid standards,
2


G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

Fig. 1. Negative ion ESI mode base peak extracted ion chromatograms at a concentration of 50 pg/μL (A) LPA 16:0/0:0 m/z 409.2355 using the conventional system/CSH C18
column (B) LPA 16:0/0:0 m/z 409.2355 using the HST system/CSH C18 column (C) PA 16:0/18:1 m/z 673.4808 using the conventional system/CSH C18 column and (D) PA
16:0/18:1 m/z 673.4808 using the HST system/CSH C18 column. Inset figures show the chemical structure of LPA and PA.

cessing were performed using MassLynxTM V4.2 software (Waters
Corporation, Wilmslow, UK).

16:0/18:1, at a concentration of 50 pg/μL, are displayed in Figs. 1
and 2. As these figures show the HST system/column showed significant improvement in peak tailing, sensitivity and recovery of
these lipids compared to the conventional system/column. The average (n = 3) peak tailing factor (PTF) and full width at half maximum (FWHM) of LPA 16:0/0:0, PA 16:0/18:1, LPS 18:0/0:0 and PS
16:0/18:1 were calculated for the standard solution prepared at a
concentration of 250 pg/μL. These data are provided for both HST
and conventional systems in Table 1 showing that the HST system/column reduced peak tailing by 65–80% for LPA 16:0/0:0, PA
16:0/18:1, LPS 18:0/0:0 and PS 16:0/18:1 compared to the conventional configuration. The peak width at FWHM was reduced by 70–
86% for LPA and PA; and by 25–50% for LPS and PS using the HST
system/column combination compared to the conventional configuration. These improvements in chromatographic performance for
these acidic phospholipids are most probably due to the elimination of the interactions of phosphate and carboxylate groups with
the metal surfaces provided by the metal surface barrier [31]. Interestingly both LPA and LPS showed a significantly greater improvement in PTF and peak width compared to their corresponding PA
and PS respectively with the HST system/column. These improvements in chromatographic performance can be attributed to the
fact that the phosphate groups in LPA and the carboxylate group
in LPS are, due to their chemical structure, more exposed to metal
surfaces compared to PA and PS respectively. LPA and LPS eluted
earlier than PA and PS and, due to the shallow gradient in the
region where the LPA and LPS eluted, they would be expected to
have reduced PTF’s and peak widths at FWHM.

3. Results and discussion

As noted in the introduction, in order to investigate the potential benefits of the HST for the phosphorylated and carboxylated
lipid classes we took a previously optimized method developed using the ACQUITY CSH C18 phase [19] as our starting point. The
initial experiments were conducted to compare the conventional
and HST systems, using the differential ion mobility system suitability Lipidomix (described in the experimental section) as a standard mixture. This enabled the lipid separation to be evaluated and
provided a benchmark for instrument performance when changing
between conventional and HST systems. For the purposes of this
comparison all the samples were prepared in one vial and divided
into separate vials for the conventional and HST systems/columns.
In addition, the same batch of mobile phase and stationary phase
was employed for all of these analyses with the same chromatographic conditions employed for all instrumental configurations.
The three configurations evaluated were: (i) conventional LC system with conventional CSH C18 column, (ii) conventional LC system with HST CSH C18 column and (iii) HST LC system with HST
CSH C18 column.
3.1. Peak shape and tailing factor comparison
Under the RP LC conditions generally used in lipidomic studies the acid phospholipids LPA, PA, LPS and PS tend to elute as
quite broad peaks. However, with suitable modification of the mobile phase, using mobile phase additives such as phosphoric acid
[22], this peak tailing can be attenuated or eliminated. However,
these mobile phase additives such as phosphoric acid can compete
with analyte ionization, especially in negative ion mode thus the
approach of employing a metal surface barrier was an attractive
alternative. To evaluate whether or not the chromatographic performance was improved using the HST system and CSH C18 column we measured the lipid standards LPA 16:0/0:0, PA 16:0/18:1,
LPS 18:1/0:0 and PS 16:0/18:1 and compared the results with those
obtained using a conventional system and CSH C18 column. Representative negative ion ESI mode base peak extracted ion chromatograms of LPA 16:0/0:0, PA 16:0/18:1, LPS 18:1/0:0 and PS

3.2. Peak signal intensity and sensitivity comparison
The increase in sensitivity obtained with the HST system/column can be clearly seen for the analysis of LPA 16:0/0:0
shown in Figs. 1A and B at a concentration of 50 pg/μL. As these
figures clearly show there was a noticeable improvement in peak
intensity for these analytes when the HST system/column was used
compared to the conventional system. Similarly, Figs. 1C and D
(PA), 2A and B (LPS) and 2D and C (PS) also showed improved

sensitivity, peak shape and recovery obtained with the HST system/column compared to the conventional system/column. The
chromatographic performance of the two systems was compared
at 5, 50, and 250 pg/μL for LPA 16:0/0:0, PA 16:0/18:1, LPS 18:1/0:0
3


G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

Fig. 2. Negative ion ESI mode base peak extracted ion chromatograms at a concentration of 50 pg/μL (A) LPS 18:1/0:0 m/z 522.2832 using the conventional system/CSH
C18 column (B) LPS 18:1/0:0 m/z 522.2832 using the HST system/CSH C18 column (C) PS 16:0/18:1 m/z 760.5129 using the conventional system/CSH C18 column (D) PS
16:0/18:1 m/z 760.5129 using the HST system/CSH C18 column. Inset figures show the chemical structure of LPS and PS.
Table 1
Calculated average (n = 3) peak tailing factor (PTF) and full width at half maximum (FWHM) for LPA 16:0/0:0, PA 16:0/18:1, LPS 18:0/0:0 and PS 16:0/18:1 at a concentration
of 250 pg/μL using the conventional system and CSH C18 column compared to the hybrid surface technology (HST) system and CSH C18 column.
LPA 16:0/0:0
System

∗

PTF

∗

PA 16:0/18:1
FWHM

#

LPS 18:1/0:0

PS 16:0/18:1

PFT

FWHM

PFT

FWHM

PFT

FWHM

Conventional
system and
CSH C18
column

7.67

39.70

6.48

43.96

6.25

10.86

4.07

10.84

HST system
and CSH C18
column

1.82

5.3

2.25

12.78

1.23

4.94

1.25

8.08

The tailing factor was calculated as the ratio of the full width at 5% of peak height and two times of the front half width [33].
The full width at half maximum is the peak width, measured in seconds, at 50% of the maximum peak height.

#

and PS 16:0/18:1 molecular species (Figures S1-S4) and at 250
pg/μL for CerP 18:1;O2/16:0, SPBP 18:1;O2 and PIP (4 ) 18:1/18:1
(Figures S5). The LPA and PA were undetectable at 5 pg/μL with
the conventional system/column due to analyte loss caused by interaction with metal surfaces. This result contrasts with the results for the HST system/column where these low concentration
injections gave peaks that were clearly visible. Even though the increased sensitivity provided may not be required for quantification
at high concentrations of these analytes the improved peak shape
is still helpful in reducing peak overlap.
The peak intensities obtained for LPA 16:0/0:0 at a concentration of 250 pg/μL using either a (i) the conventional LC system/CSH
C18 column or (ii) the conventional LC system with HST CSH C18
column or, (iii) the HST LC system with HST CSH C18 column were
compared in negative ion ESI mode (m/z 409.235). The extracted
ion chromatograms obtained using these three separate configurations are displayed in Fig. 3. The increase in intensity obtained using the HST for LPA 16:0/0:0 is also given in Figure S6 as bar graph
with the average (n = 3) of the total ion intensity for each system
configuration used. Using the conventional LC system, a marked
increase in signal intensities was observed for the HST column
rather than the untreated column. However, the greatest increase
in response was obtained for the HST system/column configuration which showed a 12.7 fold increase over the conventional sys-

tem/column combination. Use of the HST CSH C18 column and the
untreated conventional LC system also provided increased intensities for the phosphorylated and carboxylate lipid standards evaluated. However, these increases were modest when compared to
the HST system/column configuration (Figure S6).
3.3. Application to PA and PS lipid extracts
This new HST technology was also applied to the analysis of
egg (chicken) PA and brain (porcine) PS extracts and the results
compared to those obtained with the conventional system/column
(Figs. 4A-D and 5A-D). The results of this comparison clearly show
that the PA lipid species at m/z 671.47 (16:1_18:1 and 16:0_18:2)
and 673.48 (16:0_18:1) eluted with very broad tailing peaks using

the conventional system and column (Fig. 4A). These results were
compared to the data acquired with HST system/column (Fig. 4B),
where the peaks showed significant improvements in shape and
intensity. The identities of the fatty acid constituents were confirmed via fragment ion information using the HST negative ion ESI
mode LC/MS data independent acquisition where lipid identification was performed based on low energy exact precursor ion mass
and corresponding high energy characteristic fragment ions. The
molecular species level information was provided only when the
presence of the fatty acyl chain was confirmed using fragment ion
4


G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

duced an average peak width of only 10.8 s (n = 3) (Fig. 4D). Similarly, the data displayed in Fig. 5 shows the results obtained for
the analysis of the porcine brain PS extract, analyzed using both
conventional and HST systems. In this application the HST system/column not only provided reduced peak tailing compared to
the conventional system/column but also allowed the visualization
of three additional peaks eluting with retention times of 5.3, 5.6
and 7.9 min. These additional peaks were identified based on exact mass, fragment ions and theoretical isotopic distribution as PS
38:2 m/z 814.55, PS 40:3 m/z 840.577 and PS 44:2 m/z 898.654 respectively. In contrast, these minor PS species were not detected
using the conventional system/column. The peak width at half
height from the two systems was calculated for PS 18:0_18:1 m/z
788.545 (Figs. 5C and D) with the conventional system/column
having produced a peak FWHM of 10.8 s (n = 3) (Fig. 5C) compared to value of 7.4 s (n = 3) for HST system/column (Fig. 5D).
As can be seen from these examples, and in contrast to previous methods using conventional RP columns [19–21], the HST system/column enabled the simultaneous analysis of phosphate and
carboxylate-containing lipid species in biological extracts without
evidence of significant binding or peak tailing and without compromising the determination of other lipid classes. The HST system/column increased the sensitivity, recovery, and coverage of
these biologically important lipids. Previous RP LC approaches have

not sufficiently addressed the simultaneous analyses of these acidic
lipids (LPA, PA, LPA, PS, CerP, SPBP and PIP) with other phospholipid classes. The application and performance of the HST system/column to other lipid classes was evaluated using the differential ion mobility system suitability Lipidomix that contains PA
and PS lipid species together with other lipid classes. The data
displayed in Fig. 6 shows a representative negative ion ESI-base
peak extracted ion chromatogram of LPC 18:1/0:0, PE 14:1/14:1, SM
d18:1/18:1 and Cer d18:1/18:1 analyzed using conventional system/column versus the HST system/column. These results showed
that, for non-phosphorylated lipid classes, the HST system/column
provided comparable retention behavior in terms of peak tailing,
shape, and signal intensity to the conventional system/column configuration (Table S2).
As we have noted, the complexity of the lipid profile in biological matrices makes the RPLC/MS analysis of phosphorylated and
carboxylated lipids challenging due to metal-analyte interactions.
However, the accurate measurement of these lipids may well be

Fig. 3. Overlayed negative ESI base peak extracted ion chromatogram of LPA
16:0/0:0 at a concentration of 250 pg/μL using the conventional system/CSH C18
column, the conventional system/HST CSH C18 column or the HST system/CSH C18
column.

information (Figure S7). The HST system/column provided four additional baseline separated peaks at retention times of 3.1, 3.3, 4.7
and 5.8 min respectively which were not observed with the conventional system. Based on exact mass and negative ion ESI mode
high energy fragment ion information these peaks were identified
as PA 38:6 m/z 719.46, PA 36:4 m/z 695.46, PA 36.2 m/z 699.49
and PA 18:0_18:1 m/z 701.51 lipids. In contrast, it was difficult
to make confident molecular species identification with the data
obtained using the conventional system/column (Fig. 4A) as there
was a large amount overlap of the fragment ions because of peak
tailing as well as very low MS signal intensity. The data displayed
in Figs. 4C and D show extracted ion chromatograms for the lipid
PA 16:0_18:1, from which it can be seen that an average peak
FWHM of 34.0 s (n = 3) was obtained with the conventional system/column (Fig. 4C) whereas that for the HST system/column pro-

Fig. 4. Avanti polar lipids egg chicken PA extract (10 ng/μL) measured using (A) the conventional system/CSH C18 column or (B) the HST system/CSH C18 column. Corresponding extracted ion chromatograms for PA 16:0_18:1 at m/z 673.481 using (C) the conventional system/CSH C18 column with average peak FWHM 34.0 s (n = 3) or (D)
the HST system/CSH C18 column with average peak FWHM 10.8 s (n = 3).
5


G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

Fig. 5. Avanti polar lipids porcine brain PS extract (10 ng/μL) measured using (A) the conventional system/CSH C18 column or (B) the HST system/CSH C18 column. Corresponding extracted ion chromatograms for PS 18:0_18:1 at m/z 788.545 using (C) the conventional system/CSH C18 column with average peak FWHM 10.8 s (n = 3) or (D)
the HST system/CSH C18 column with average peak FWHM 7.4 s (n = 3).

Fig. 6. The Avanti Differential Ion Mobility System Suitability standard Lipidomix Kit was used to evaluate the effect of the HST system/column on other lipid classes. Negative
ESI base peak extracted ion chromatograms of LPC 18:1/0:0, PE 14:1/14:1, SM d18:1/18:1 and Cer d18:1/18:1 at a concentration of 100 pg/μL using (A) the conventional
system/CSH C18 column (B) the HST system/CSH C18 column.

critical for the evaluation of their dysregulation in disease states.
Elimination of these unwanted interactions is especially important
for low abundant signaling lipid classes such as LPA, PA, LPS, PS,
CerP, SPBP and PIP where the systemic concentrations are in the
low ng/mL range and metal chelation can result in significant attenuation of the signal, or even its complete loss, as illustrated in
the examples shown here. As discussed, metal free materials have
been used to replace the metal components in chromatography
systems. Thus, PEEK components have been successfully employed
for the analysis of phosphate-containing sphingolipids [29] and
mobile phase additives are also an option [22–27]. However, the
limitations of these approaches are clear, and there are obvious
advantages to simply eliminating the problem from the system, removing the need for the analyst to investigate and mitigate these
unwanted analyte-surface interactions.

creased the recovery of phosphorylated and carboxylate lipids
compared to a conventional system/CSH C18 column. The HST system/column combination reduced peak tailing by 65–80% and peak
width by 70–86% for LPA and PA. Moreover, increased signal intensities of up to 12.7 times were observed for LPA with the HST approach compared to the untreated system. This methodology can
be applied to other phosphorylated and carboxylated lipid classes,
such as PIP with mono, di and tri phosphates, CerP and SPBP. In
contrast to most of the previous methodologies using conventional
RP columns the HST allowed for the simultaneous analysis of these
acidic phospholipid classes at the same time as other lipid classes
without adverse effects on the coverage of these biologically important lipids.

4. Conclusions

We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its
outcome.

Author declaration of interest statment

The HST system/column combination clearly provided a significant improvement in sensitivity, reduced peak tailing and in6


G. Isaac, I.D. Wilson and R.S. Plumb

Journal of Chromatography A 1669 (2022) 462921

We confirm that the manuscript has been read and approved by
all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been
approved by all of us.
We confirm that we have given due consideration to the protection of intellectual property associated with this work and that
there are no impediments to publication, including the timing of

publication, with respect to intellectual property. In so doing we
confirm that we have followed the regulations of our institutions
concerning intellectual property.
We further confirm that any aspect of the work covered in this
manuscript that has involved either experimental animals or human patients has been conducted with the ethical approval of all
relevant bodies and that such approvals are acknowledged within
the manuscript.

[12] E. Cífková, R. Hájek, M. Lísa, M. Holcapek, Hydrophilic interaction liquid chromatography-mass spectrometry of (lyso)phosphatidic acids,
(lyso)phosphatidylserines and other lipid classes, J Chromatogr A 1439
(2016) 65–73.
[13] S.L. Abidi, T.L. Mounts, High-performance liquid-chromatography of phosphatidic-acid, J. Chromatogr. B-Biomed. Appl. 671 (1995) 281–297.
[14] C. Bure, S. Ayciriex, E. Testet, J.M. Schmitter, A single run LC–MS/MS method
for phospholipidomics, Anal. Bioanal. Chem. 405 (2013) 203–213.
[15] A.A. Karlsson, P. Michelsen, A. Larsen, G. Odham, Normal-phase liquid chromatography class separation and species determination of phospholipids utilizing electrospray mass spectrometry tandem mass spectrometry, Rapid Commun. Mass Spectrom. 10 (1996) 775–780.
[16] M. Narvaez-Rivas, E. Gallardo, J.J. Rios, M. Leon-Camacho, A new high-performance liquid chromatographic method with evaporative light scattering detector for the analysis of phospholipids. Application to Iberian pig subcutaneous
fat, J. Chromatogr. A 1218 (2011) 3453–3458.
[17] L.Q. Pang, Q.L. Liang, Y.M. Wang, L. Ping, G.A. Luo, Simultaneous determination
and quantification of seven major phospholipid classes inhuman blood using
normal-phase liquid chromatography coupled with electrospray mass spectrometry and the application in diabetes nephropathy, J. Chromatogr. B 869
(2008) 118–125.
[18] S. Uran, A. Larsen, P.B. Jacobsen, T. Skotland, Analysis of phospholipid species
in human blood using normal-phase liquid chromatography coupled with electrospray ionization ion-trap tandem mass spectrometry, J. Chromatogr. B758
(2001) 265–275.
[19] C.W. Damen, G. Isaac, J. Langridge, T. Hankemeier, R.J. Vreeken, Enhanced lipid
isomer separation in human plasma using reversed-phase UPLC with high-resolution ion-mobility MS detection, J. Lipid. Res. 55 (2014) 1772–1783.
[20] P.A. Vorkas, G. Isaac, M.A. Anwar, A.H. Davies, E.J. Want, J.K. Nicholson, E. Holmes, Untargeted UPLC-MS profiling pipeline to expand tissue
metabolome coverage: application to cardiovascular disease, Anal Chem 87
(2015) 4184–4193.
[21] T. Cajka, O. Fiehn, Increasing lipidomic coverage by selecting optimal mobile-phase modifiers in LC–MS of blood plasma, Metabolomics 12 (2016) 1–11.

[22] H. Ogiso, T. Suzuki, R. Taguchi, Development of a reverse-phase liquid chromatography electrospray ionization mass spectrometry method for
lipidomics, improving detection of phosphatidic acid and phosphatidylserine,
Anal Biochem 375 (2008) 124–131 2008.
[23] J.A. Anspach, S. Rao, B. Rivera, Bioinert Versus Biocompatible: the Benefits of
Different Column Materials in Liquid Chromatography Separations, LCGC 36
(2018) 24–29.
[24] J.C. Heaton, D.V. McCalley, Some factors that can lead to poor peak shape in
hydrophilic interaction chromatography, and possibilities for their remediation,
J. Chromatogr. A 1427 (2016) 37–44 (put as 24).
[25] K.T. Myint, T. Uehara, K. Aoshima, Y. Oda, Polar Anionic Metabolome Analysis by Nano-LC/MS with a Metal Chelating Agent, Anal. Chem. 81 (2009)
7766–7772.
[26] D. Winter, J. Seidler, Y. Ziv, Y. Shiloh, W.D. Lehmann, Citrate Boosts the Performance of Phosphopeptide Analysis by UPLC-ESI-MS/MS, J. Proteome Res. 8
(2009) 418–424.
[27] O.L. Knittelfelder, B.P. Weberhofer, T.O. Eichmann, S.D. Kohlwein, G.N. Rechberger, A versatile ultra-high performance LC-MS method for lipid profiling, J
Chromatogr B Analyt Technol Biomed Life Sci 951–952 (2014) 119–128.
[28] J. Lee, Y. Kim, M. Moon, Enhancement of acidic lipid analysis by nanoflow
ultrahigh performance liquid chromatography-mass spectrometry, Anal Chim
Acta 1166 (2021) 338573.
[29] S.G..B. Gowda, K. Ikeda, M. Arita, Facile determination of sphingolipids under
alkali condition using metal-free column by LC-MS/MS, Anal. Bioanal Chem.
410 (2018) 4793–4803.
[30] R.S. Plumb, L.A. Gethings, A. King, L.G. Mullin, G. Maker, R. Trengove, I.D. Wilson, Hybrid organic/inorganic hybrid surface technology for increasing the performance of LC/MS(MS)-based drug metabolite identification studies: application to gefitinib and metabolites in mouse plasma and urine, J Pharm Biomed
Anal 200 (2021) 114076.
[31] M. DeLano, T.H. Walter, M.A. Lauber, M. Gilar, M.C. Jung, J.M. Nguyen, C. Boissel, A.V. Patel, A. Bates-Harrison, K.D. Wyndham, Using hybrid organic-inorganic surface technology to mitigate analyte interaction with metal surface in
UHPLC, Anal. Chem. 93 (2021) 5773–5781.
[32] N. Tanna, L.G. Mullin, P.D. Rainville, I.D. Wilson, R.S. Plumb, Improving
LC/MS/MS-based bioanalytical method performance and sensitivity via a hybrid surface barrier to mitigate analyte - Metal surface interactions, J Chromatogr. B 1179 (2021).
[33] U.D. Neue, Theory of peak capacity in gradient elution, J. Chromatogr. A 1079
(2005) 153–161.

Declaration of Competing Interest
None.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.chroma.2022.462921.
CRediT authorship contribution statement
Giorgis Isaac: Conceptualization, Methodology, Data curation,
Writing – original draft. Ian D Wilson: Writing – review & editing. Robert S. Plumb: Supervision, Writing – review & editing.
References
[1] C.L. Stace, N.T. Ktistakis, Phosphatidic acid- and phosphatidylserine-binding
proteins, Biochim. Biophys. Acta Mol. Cell Biol. Lipids 1761 (2006) 913–926.
[2] R. Welti, W. Li, M. Li, Y. Sang, H. Biesiada, H.-.E. Zhou, C.B. Rajashekar,
T.D. Williams, X. Wang, Profiling Membrane Lipids in Plant Stress Responses.
Role of Phospholipase D Alpha in Freezing-Induced Lipid Changes in Arabidopsis, J. Biol. Chem. 277 (2002) 31994.
[3] E. Cífková, R. Hájek, M. Lísa, M. Holcapek, Hydrophilic interaction liquid chromatography-mass spectrometry of lysophosphatidic acids, lysophosphatidylserines and other lipid classes, J. Chromatogr A. 1439 (2016) 65–73.
[4] A. Shevchenko, K. Simons, Lipidomics: coming to grips with lipid diversity, Nat.
Rev. Mol. Cell Biol. 11 (8) (2010) 593–598.
[5] X. Han, Lipidomics for studying metabolism, Nat. Rev. Endocrinol. 12 (11)
(2016) 668–679.
[6] T. Xu, C. Hu, Q. Xuan, G. Xu, Recent advances in analytical strategies for mass
spectrometry-based lipidomics, Anal. Chim. Acta 1137 (2020) 156–169.
[7] B. Brugger, G. Erben, R. Sandhoff, F.T. Wieland, W.D. Lehmann, Quantitative
analysis of biological membrane lipids at the low picomole level by nano-electrospray ionization tandem mass spectrometry, Proc. Natl. Acad. Sci. U.S.A. 94
(1997) 2339–2344.
[8] M. Holcapek, G. Liebisch, K. Ekroos, Lipidomic Analysis, Anal. Chem. 90 (2018)
4249–4257.
[9] E. Cífková, M. Holcˇ apek, M. Lísa, M. Ovcˇ acˇ íková, A. Lycˇ ka, F. Lynen, P. Sandra,
Nontargeted quantitation of lipid classes using hydrophilic interaction liquid
chromatography-electrospray ionization mass spectrometry with single internal standard and response factor approach, Anal Chem 84 (2012) 10064–10070
2012.

[10] E. Cífková, M. Holcˇ apek, M. Lísa, D. Vrána, B. Melichar, V. Študent, Lipidomic
differentiation between human kidney tumors and surrounding normal tissues
using HILIC-HPLC/ESI-MS and multivariate data analysis, J. Chromatogr B Analyt Technol Biomed Life Sci. 10 0 0 (2015) 144-21.
[11] A. Triebl, M. Trötzmüller, A. Eberl, P. Hanel, J. Hartler, H.C. Köfeler, Quantitation of phosphatidic acid and lysophosphatidic acid molecular species using
hydrophilic interaction liquid chromatography coupled to electrospray ionization high resolution mass spectrometry, J. Chromatogr A. 1347 (2014) 104–110.

7