Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology 56
Development of Enhanced
Analytical Methodology for
Lipid Analysis from Sampling
to Detection
A Targeted Lipidomics Approach
GIORGIS ISAAC
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2005
ISSN 1651-6214
ISBN 91-554-6259-6
urn:nbn:se:uu:diva-5810
!
"# "##$ %&$ ' ( ) ' ' *((+ ,( - .)(+
/
0+ "##$+ ' .(
1
() ' 2 1
'
3
) + 1 ,
) 2 1
(+ 1
+
$4+ 5 + + /36 78$$84"5984+
,( ( - ) ' ( ' )
9 ' ) : ' *!.
-( (
; (
)
( *82 :
.3/<3 3+
,( ; ' '
( -( ) '
(
( ' - (;+ *!. ' (
- '
'' ; (; -(( -( ; + / ( ( ( '' ' (
*!. (; -
)
' ( ' ' ()
(
- ' ' '
'''+ / '' - ( -( (+
1 *82 .3/<3 ' ( ' ((
( * () 3 -
' ( ' ' )( ) ' -( ( + ,( '' '
(
-
)
+ =(( ( '' ( ' ( :
(
- + ,( ( (( (
- (
(
(
+
!
3 -
: ( ' '
'
'
1 131 ' 131 8<8 -
131 ><> + ,( 131 8<8 - ' ( )
(
?( -( ( ' ' ( : 131+ ,( (- (
( ' '
-
- ( 131 8<8 131 ><> +
,( ( (- (
( - )( (
' ' )
+
:
; (
)
( ((
( : ' () '
!" " # " $% &''" "
()*&+,- " @ 0) /
"##$
/336 4$84"
/36 78$$84"5984
&&&&
8$5# (&<<+?+<AB&&&&
8$5#
This thesis is dedicated in memory of my beloved father Isaac (January 19,
1938-December 19, 1994) a man of endless energy and talent who used to
guide, encourage and explain to me the value of education during my childhood. He was a hard working man always striving to be better and still a
family man. He was a proud man and he would do anything possible to take
care of his family. With out his dedication I would never be where I am today.
Giorgis Isaac
May 2005
List of Papers
The following Papers are discussed in this thesis and they are referred to in
the text by their Roman numerals:
I
Total Lipid Extraction of Homogenized and Intact Lean Fish
Muscles Using Pressurized Fluid Extraction and Batch Extraction Technique
Giorgis Isaac, Monica Waldebäck, Ulla Eriksson, Karin E.
Markides and Göran Odham, Accepted for publication in
J. Agric. Food Chem., 2005.
II
Pressurized Fluid Extraction (PFE) as an Alternative General Method for the Determination of Pesticide Residues in
Rape Seed
Tuija Pihlström, Giorgis Isaac, Monica Waldebäck, Bengt-Göran
Österdahl and Karin E. Markides, The Analyst, 2002, 70, 554559.
III
Analysis of Phosphatidylcholine and Sphingomyelin Molecular Species From Brain Extracts Using Capillary Liquid
Chromatography Electrospray Ionization Mass Spectrometry
Giorgis Isaac, Dan Bylund, Jan-Eric Månsson, Karin E. Markides
and Jonas Bergquist, J. Neurosci. Methods, 2003, 128, 111-119.
IV
Brain Lipid Composition in Postnatal Iron-Induced Motor
Behavior Alterations Following Chronic Neuroleptic Administration in Mice
Giorgis Isaac, Anders Fredriksson, Rolf Danielsson, Per Eriksson
and Jonas Bergquist, Submitted to Neurosci. Lett.
V
Characterization of Brain Sulfatide Molecular Species in
Arylsulfatase A Deficient Mice Using Electrospray Ionization
Tandem Mass Spectrometry
Giorgis Isaac, Jan-Eric Månsson, Volkmar Gieselmann, Elisabeth Hansson, Zarah Pernber and Jonas Bergquist, Manuscript.
Reprints of the publications were made with kind permission of the publishers.
The author’s contribution to the Papers:
x
x
x
x
x
Paper I: I was responsible for practical work except the NMR part. I
wrote the paper in close consultation with the co-authors.
Paper II: The experiments were planed by Tuija Pihlström and Monica
Waldebäck. I carried out the PFE extraction and contributed to the paper
writing.
Paper III: Planning the experiments together with Jonas Bergquist and I
was responsible for experimental work of the LC/MS experiments. I
wrote the paper in close consultation with the co-authors.
Paper IV: I was responsible for experimental work and planning of the
LC/MS experiments. Paper writing and data analyses were made in collaboration with co-authors.
Paper V: I was responsible for all experimental part in ESI/MS. Paper
writing and data analysis was made in collaboration with co-authors.
Contents
1. Introduction.................................................................................................9
2. Lipids and Lipidomics ..............................................................................11
2.1 Lipids: Definition ...............................................................................11
2.2. General Classification of Lipids........................................................11
2.3. Lipidomics: The Next Frontier..........................................................13
2.4. Membrane Structure and Occurrence of Phospholipids,
Sphingomyelin and Sulfatides..................................................................14
2.4.1. Phospholipids.............................................................................16
2.4.2. Sphingolipids .............................................................................17
3. Pressurized Fluid Extraction .....................................................................19
3.1. Extraction Using High Diffusion Fluids ...........................................19
3.2. Theory and Concept of PFE Technique ............................................20
3.3. PFE Instrumentation..........................................................................21
3.4. Selected Application and Comparison of the PFE Technique with
Conventional Methods .............................................................................23
3.4.1. Total Lipid Extraction Using PFE (Paper I) ..............................23
3.4.2. PFE as an Alternative Method for Extraction of Pesticide
Residues from Fatty Matrix (Paper II).................................................24
4. Analysis of Intact Phosphatidylcholine, Sphingomyelin and Sulfatide
Molecular Species Using LC/ESI/MS and ESI/MS/MS...............................26
4.1. Reversed Phase LC of Intact PC and SM..........................................27
4.2. Mass Spectrometry ............................................................................29
4.2.1. The Mechanism of ESI/MS .......................................................29
4.2.2. The Principles of ESI/MS of Lipids ..........................................31
4.2.3. MS Scan Modes.........................................................................31
4.3. Reversed Phase LC/ESI/MS..............................................................35
5. Data Handling ...........................................................................................38
6. Conclusions and Future Aspects ...............................................................39
7. Acknowledgements...................................................................................41
8. Summary in Swedish ................................................................................43
9. References.................................................................................................45
Abbreviations
Ag-LC
ASA
CID
ESI
FA
GC
GPC
LC
LLE
MAE
MP
MS
m/z
PA
PC
PE
PFE
PI
PIS
PL
PS
RP
SFE
SM
SP
TLC
Silver ion Liquid Chromatography
Arylsulfatase A
Collision Induced Dissociation
Electrospray Ionization
Fatty Acid
Gas Chromatography
Gel Permeation Chromatography
Liquid Chromatography
Liquid-Liquid Extraction
Microwave Assisted Extraction
Mobile Phase
Mass Spectrometry
Charge-to-Mass Ratio
Phosphatidic Acid
Phosphatidylcholine
Phosphatidylethanolamoine
Pressurized Fluid Extraction
Phosphatidylinositol
Precursor Ion Scan
Phospholipid
Phosphatidylserine
Reversed Phase
Supercritical Fluid Extraction
Sphingomyelin
Stationary Phase
Thin Layer Chromatography
1. Introduction
Many of the traditional sample extraction methods have become the weakest
link in modern chemical analytical methods. Some of these methods might
be effective but slow, require large amounts of organic solvents and be difficult to automate and hence extremely labour intensive. During the past decade, driving forces such as the need for faster analyses, increased productivity, concern for cost of analysis, safety of laboratory personnel and the reduction of environmental pollution sources have encouraged chemists to
pursue modern sample extraction methods 1, 2.
Pressurized fluid extraction (PFE) is a relatively new automated liquid extraction technique which uses elevated temperature and pressure with standard liquid solvents to achieve fast and efficient sample extractions. In this
thesis the reliability and efficiency of the PFE technique was utilized for
novel application in the extraction of total lipid content from cod and herring
muscles (Paper I), pesticides from rape seed (Paper II), and phosphatidylcholine (PC) and sphingomyelin (SM) molecular species from human brain (Paper III). The developed methodology was described and the results were
compared to different traditional extraction methods. Furthermore, in Paper
III a method was developed for the separation and characterization of the PC
and SM molecular species using reversed phase capillary liquid chromatography (RP-LC) coupled online to electrospray ionization mass spectrometry
(ESI/MS). The developed method was applied to investigate lipid compositions in mouse brain (Paper IV). In Paper V a tandem MS in precursor ion scan
(PIS) mode was used to characterize the sulfatide content in brain tissue.
From an ecological point of view, it is more relevant to express the residue level of lipophilic persistent pollutants on fat weight basis than on traditional fresh weight basis 3. Numerous intercalibration laboratories have
shown that the original Jensen extraction method for total lipids and persistent organic pollutants, gave satisfactory yields when applied to fatty aquatic
organisms 4. However, in case of lean fish muscle (fat content below 1%) the
lipid yields with that method is about 25% too low and consequently residue
levels on lipid bases are correspondingly too high 4. In Paper I, the reliability
and efficiency of the PFE technique for the extraction of total lipid content
from cod and herring muscle tissue was evaluated and the results were compared to both the original and the later modified Jensen extraction methods.
The influence of several experimental variables affecting the extraction effi9
ciency of the total lipid content with PFE was studied using experimental
design.
The determination of pesticide residues in fatty matrices such as rape seed
is associated with various analytical difficulties. Since the studied pesticides
cover a wide range of polarity (Paper II, Table 1) the recovery of very nonpolar pesticides from the lipid rich-matrix makes the extraction challenging.
Therefore there is a need of developing extraction techniques, which efficiently recovers the different classes of pesticides, over the polarity range,
with low solvent consumption and rapid extraction time from lipid-rich matrices. In Paper II, a PFE multi-method was developed for the determination
of pesticide residues in rape seed and the results were compared to the traditional liquid-liquid extraction (LLE) method. The analysis was performed
with gas chromatography (GC) using different detectors after clean-up by
gel permeation chromatography (GPC).
Disturbances in lipid membrane cause various disorders and therefore the
development of methods for the analysis of lipids is essential. Paper III describes a method based on RP capillary LC coupled online to ESI/MS for the
analysis of PC and SM molecular species. Furthermore, the extraction efficiencies of the traditional Folch method and PFE were compared using human brain samples. In a subsequent study (Paper IV), the developed method
was applied for the analysis of PC and SM molecular species in brains from
eight groups of mice treated with vehicle, low dose clozapine, high dose
clozapine and haloperidol. Both the effects of postnatal iron administration
in the lipid composition and behavior; and whether or not the established
effects may be altered by subchronic administration of the neuroleptic compounds, clozapine and haloperidol were investigated.
In Paper V a tandem MS in PIS mode was used to compare brain sulfatide
molecular species accumulation in arylsulfatase A (ASA) knockout (ASA -/-)
and wild type control mice (ASA +/+) at different ages. ASA -/- mice were developed as a model of monogenic disease metachromatic leukodystrophy (MLD)
with deficiency of the lysosomal enzyme ASA. MLD is a human disease in
which sulfatide is accumulated within the lysosomes due to defective enzyme activity of ASA. As a comparison, the sulfatide molecular species in
cultured astrocytes were also characterized.
The aim of this thesis is to describe and discuss method development using the
recent extraction technique PFE and analysis of intact PC, SM and sulfatide molecular species using LC/ESI/MS and tandem MS from complex biological samples.
10
2. Lipids and Lipidomics
2.1 Lipids: Definition
Before going into details of the different lipid classes (phospholipids (PLs),
SM and sulfatide) and their molecular species considered in this thesis, it
may be useful to define the term “lipid”. Lipids can not be defined, as it is
often done based on their physical properties, as a class of natural compounds that share a high degree of solubility in organic solvents. This designation is at the same time both too broad and to narrow because it encompasses a diversity of structurally and functionally unrelated molecules and
excludes partly water soluble lipids such as gangliosides and phosphoinositides 5, 6. The alternative definition proposed by W. Christie 6 based on their
chemical structure and function seems more adequate i.e. “Lipids are fatty
acids (FAs) and their derivatives, and substances related biosynthetically or
functionally to these compounds”. By this definition, every biomolecule
derived from the condensation of malonyl coenzyme A units and implicated
in cell structure and function should be considered as a lipid. Thus FAs are
defined as compounds synthesized in nature via condensation of malonyl
coenzyme A units by a FA synthase complex. The FAs contain even number
of carbon atoms (usually C14-C24), although the synthase can also produce
odd and branched chain FAs to some extent when supplied with appropriate
precursors. Other substituent groups, including double bonds are normally
incorporated into the aliphatic chains later by different enzyme systems 6.
2.2. General Classification of Lipids
It is not the aim of this thesis work to give detailed classification of the different lipid compounds but rather to show the complexity and structural varieties of the different lipids as a base for their diverse characteristics. Unlike
proteins and nucleic acids that are linear polymers of amino acids and nucleotides respectively, with linkages at only one position, lipids can adopt
complex structures. The main lipid classes of animal origin can be classified
as shown in Figure 1. This global classification is fairly comprehensive and
remains only as a guide. Some lipid compounds classified in one class by
some may be classified in another class by others.
11
Figure 1. Lipid classification scheme. This global classification is fairly comprehensive but remains only as a guide.
Christie in his recent book 6 classified the lipids into two broad classes
which is convenient especially for chromatographic purposes. Simple lipids
(glycerols, waxes, free FAs and sterols) are defined as those that, on hydrolysis, yield at the most two types of primary product per mole; complex
lipids (PLs, sphingolipids and glycoglycerolipids) yield three or more primary hydrolysis products per mole. Alternatively, the terms "neutral" and
"polar" lipids respectively are used to define these groups, but are at the
same time less exact terms.
In a recent review Han and Gross 7 roughly classified the cellular lipids
into three large groups including non-polar lipids, polar lipids and metabolites based on their relative polarities of the head group regions. Non-polar
lipids include predominantly cholesterol and cholesterol esters as well as
triacylglycerols. All these classes of lipids consist of a small or weak polar
part and a dominant hydrophobic region. Polar lipids are predominately
comprised of PLs, sphingolipids, and glycolipids. The last groups of lipids
are lipid metabolites, which are derived by enzymatic action on parent lipids
or their precursors. Common metabolites include long chain acylcarnitines,
nonesterified FA, FA esters, acyl anamide, ceramides, lysolipids, eicosanoids, diacylglycerols, sphingosine-3-phosphate, etc. Many of the metabolites are biologically active second messengers.
12
2.3. Lipidomics: The Next Frontier
The term “lipidomics” has just emerged in the literature in the context of
genomics and proteomics 8-12. However lipidomics will certainly pick up
momentum by bringing in sophisticated technologies (e.g. recent advance
and novel application of LC/ESI/MS and MS/MS) to combat lipid associated
diseases such as coronary heart disease, stroke, cancer, diabetes and Alzheimer’s disease to mention a few disorders. In a sense while genes can be
considered to be the blue prints and proteins are the structures built from
these blue prints, the lipids provide a functional family that makes the house
a home 13. According to Spenner et al. 10 if the entire spectrum of lipids in
the biological system is called “lipidome” then mapping this spectrum can be
called lipidomics. Thus, lipidomics can be defined as the full characterization of lipid molecular species and of their biological roles with respect to
expression of proteins involved in lipid metabolism and function, including
gene regulation. Therefore the first tasks in characterizing the global change
in lipids (lipidomics) can be described by the following 10:
x
x
x
New analytical approaches for mapping the lipidome: Global characterization of lipid molecular species by LC/MS and development of
novel detection techniques.
Application of biophysical methods to understand lipid-protein interactions: This mainly focus in the interaction of lipid micro- and
nanodomains of biological membranes and will be benefited by advanced synthesis of specific lipids, analogs and probes.
Identification of lipid network: Identifying the lipid network including
lipid mediators, for metabolic and gene regulation and its integration
with non-lipid signaling. The challenge here is the integration of lipidomics with proteomics to study and characterize homeostatic and aberrant cellular states.
Recently, in the United States the National Institute of Health funded a $
35 million five year collaborative project between 18 laboratories, entitled
“Lipid metabolites and pathway strategy” (www.lipidmaps.org) to determine
the lipidome (all the lipids) in the mouse macrophage. These researchers will
now work together to analyze the structure and function of lipids 14. As illustrated in Figure 2, Prestwich 13 suggested that with the solution of the human
genome sequence and the growing ability to conduct massive proteomic
analyses, lipid constitute the research frontier for this decade.
13
Figure 2. The importance of functional lipidomics 13.Targeted lipidomics focuses on
specific lipids rather than the indiscriminate identification of all endogenous lipids.
The first step in global lipidomics is to measure the amount of each distinct chemical entity present in a cells lipidome and identify those alterations
in chemical constituents that are associated with disorders 7. In this thesis a
targeted lipidomics approach was used to characterize and profile PC and
SM molecular species (Papers III and IV) and sulfatide molecular species
(Paper V) by LC/MS and MS/MS using different scan modes from complex
biological lipid extracts. It is considered a targeted lipidomics approach because it focuses only on PC, SM and sulfatide molecular species rather than
the indiscriminate identification of all endogenous lipids.
2.4. Membrane Structure and Occurrence of
Phospholipids, Sphingomyelin and Sulfatides
Contrary to the classical view of the plasma membrane as a homogenous PL
assembly that is loaded with proteins, proposed by Singer and Nicholson 15,
it is now accepted that the plasma membrane is heterogeneous. Biological
membranes are structurally and compositionally complex. The biological
complexity is the result of the physicochemical properties of the individual
membrane components, which is governed by its composition and chemical
environment. Compositional diversity arises through chemical diversity of
the membrane constituents. The requirement for a cell membrane to be
formed is the existence of lipids. In general, a cell membrane is mainly composed of PLs, sphingolipids and sterols. Sterols show little structural variation compared to PLs and sphingolipids, in which structural variation originates through variability of the polar head groups and the non-polar hydrocarbon chains (e.g. carbon chain length, degree of saturation, hydroxylation,
branching). This structural variation yields a multitude of similar built molecules with diverse physical properties, which together with membrane proteins; regulate membrane properties in a dynamic fashion 16. The lipid bi14
layer illustrated as a sea of PLs 17, provides the basic structure of the membrane and serves as a relatively impermeable barrier to the passage of most
water-soluble molecules 18 and it contributes to the structural organization,
stability and fluidity of the membrane. The structure of a cell membrane is
shown in Figure 3.
Figure 3. Structure of cell membrane. Phospholipids, glycosphingolipids and cholesterol are the major constituents of the lipid bilayer in which a variety of membrane proteins, such as enzymes, receptors and channels are arranged 19.
Lipids comprise about half of the brain tissue dry weight. Brain contains
many complex lipids some of which (gangliosides, cerebrosides, sulfatides,
phosphoinositides) were first discovered in the brain, where they are highly
enriched in comparison to other tissues. Brain lipids have many physiological functions. They participate in the function as well as the structure of cellular membranes. The biomessenger function of non-membrane lipids (steroid hormones and eicosanoids) has long been known. However, in recent
years some membrane lipids, such as inositides and PC, which were previously believed to have only a structural role, have also been shown to have
important function in the signal transduction process across biological membranes 20. These recent discoveries reveal the increasing importance of lipids
not only in maintaining membrane structure, but also in the regulation of
cellular metabolism and growth, pathological state and neurodegenerative
disorders, signal transduction and endomembrane transport 5, 21.
In the next section a detailed description and biological importance of the
lipid molecular species considered in this thesis will be provided.
15
2.4.1. Phospholipids
Glycerophospholipids or simply phospholipids (PLs) are major constituents
of cellular membranes in animal and plant tissues. A typical structure of PL
consists of three parts: a glycerol backbone, a polar head group with a phosphate moiety and two FA chains esterified at the sn-1 and sn-2 positions of
the glycerol backbone, Figure 4. Structural and functional diversity are
therefore possible within each class of lipids since the head group can be
combined with a large pool of FAs that vary in both chain length and degree
of unsaturation. This yields similarly built molecules having a large diversity
of physical prosperities, which allows the living cell to regulate the intrinsic
heterogeneity of membranes in a dynamic fashion. Such heterogeneities are
dynamically regulated and can play striking functional roles.
Figure 4. The general structure and classes of phospholipids.
A first level of structural diversity is determined by the presence of specific hydrophilic head groups. Based on the differences in the chemical
structures that are linked to the phosphate, PLs can be classified into different classes. For example, if the specific head group is choline or ethanolamine the corresponding class of the PL is called PC or phosphatidylethanolamine (PE) respectively. Other classes of PL include phosphatidic acid
(PA), phosphatidylserine (PS), phosphatidylinositol (PI) and phosphatidylglycerol or diphosphatidylglycerol etc., Figure 4. Each of these classes has a
16
specific distribution, not only among different cell types, but also among
different organelles of the same cell 5-7.
The complexity of PL molecular species is not only because of variation
of polar head groups but also because of the nature of the chain where the
FA residues are linked to the sn-1 and sn-2 positions of the glycerol backbone. These residues vary in chain length as well as the number and location
of double bonds. The FA residues determine the shape and flexibility of
individual PC molecular species and affect the special properties of membrane bilayer. The double bonds introduce kinks into the carbon backbone of
the FA; the more double bonds there are, the more curved is the molecule. In
general, saturated FAs make membranes more stable and rigid, while the
more unsaturated is the FA composition the greater is the degree of flexibility introduced into the membrane 22.
This thesis will focus on the subclass diversity in PC molecular species.
PC (commonly termed as lecithin) is usually the most abundant lipid in the
membranes of animal tissues and is often a major lipid component of plant
membranes and sometimes of microorganisms.
2.4.2. Sphingolipids
The second major category of polar lipids are sphingolipids that contain a
long chain base (most frequently sphingosine) backbone or its analogs and
represent approximately 5-10% of total lipid mass in most brain cells 7, 23-26.
Based on the nature of the covalent linkage of the polar moieties to ceramide
(i.e N-acylsphingosine), sphingolipids can be classified into SM, sulfatide,
cerebroside, ganglioside, glucosylceramide, lactosylceramide and other glycosphingolipids that contain multiple sugar rings, Figure 5. Abnormal metabolism of sphingolipids could lead to their deposition in neural tissues and
results in conditions of sever mental retardation known as sphingolipidoses
26. Therefore, the biochemistry of sphingolipids such as SM (Paper IV) and
sulfatide (Paper V) in normal and pathological states has evoked a great deal
of interest.
SM is comprised of a choline head group as PC and resembles PC in
many of its physical properties and can apparently substitute in part for this
in membrane. Both PC and SM are located in the outer leaflet of plasma
membranes 27, although they are also found in various other biological structures such as lipoproteins 28. SM consists of a ceramide unit linked at position 1 to phosphoryl choline and it is found as a major component of the
complex lipids of all animal tissues but is not present in plants or microorganisms 6. Since different FAs that vary in chain length and number of double bond may be linked to a long chain base, a great number of molecular
species are found in natural extracts, Figure 5. SM is an important constituent in nervous tissue and plasma membrane of higher animals and accounts
for about 4.2-12.5% of the PL content in the brain of various species 26. It is
17
actively involved in various cellular processes such as protein trafficking and
signal transduction 29.
O
HN
X
C
R
CH3
O
OH
Class of Sphingolipids
Ceramide
X
H
O
+
(H 3C) 3N H2CH2C
O
Sphingomyelin
P
O-
CH2OH
O
HOSO H
OH
3
H
H
Sulfatide
H
OH
CH2OH
O
H
OH
H
OH
Cerebrocide
H
H
H
OH
Figure 5. The general structure and classes of sphingolipids.
Sulfatide is a major constituent of brain lipids and is found in trace
amounts in other tissues. It is an essential glycosphingolipid in myelin in the
peripheral as well as the central nervous system 30, constituting 4-6 mol% of
the total lipids in adult brain myelin 26. The structure of sulfatide consists of
two parts: a polar galactosyl-3-O-sulfate head group facing outward from the plasma
membrane and a lipid core, ceramide, composed of a FA and a long chain base, sphingosine, Figure 5. Since the FA chain can vary in length, degree of unsaturation and hydroxylation, each natural sulfatide contains numerous molecular
species. These species determine the interaction of the sulfatide molecule in
the cell membrane. Sulfatide mediate diverse biological process including
the regulation of cell growth, protein trafficking, signal transduction, cell
adhesion, neuronal plasticity and morphogenesis 23-25.
18
3. Pressurized Fluid Extraction
3.1. Extraction Using High Diffusion Fluids
The extraction process is one of the most important steps in pretreatment for
solid and liquid samples. It is therefore not surprising that a wide range of
techniques has been developed for the extraction of these samples. The most
widely used method for the extraction of solid samples is Soxhlet extraction.
However in recent years various high diffusion fluid techniques such as supercritical fluid extraction (SFE) 31-33, microwave assisted extraction (MAE)
34, 35 and PFE 36 have been developed as an alternative to other conventional
techniques. High diffusion fluids are being created by raising the temperature and controlling the pressure of the liquid. Raising temperature results in
increased diffusion coefficients and therefore increased extraction rates resulting in shorter extraction times. By extracting the target analyte at elevated temperature and pressure the rates of diffusion and desorption can be
significantly increased.
SFE is a selective extraction method that incorporates thermal and solubility driven mechanism using supercritical fluids such as carbon dioxide
(CO2), dinitrogen oxide, sulfurhexaflouride, methanol, freons and water. In
practice, more than 90% of all analytical SFE is performed with CO2 for
several practical reasons 32, 37. Apart from having low critical pressure (74
bar) and temperature (32oC), CO2 is chemically inert, available in high purity
at relatively low cost, nonflammable, is easily removed from the extract and
environment and analyst friendly when used for analytical purpose. However, SFE has not become the “universal technique” due to some inherent
disadvantages. Method development is not straight forward, the polarity of
supercritical CO2 is poor for the extraction of polar analytes and the technique is highly matrix dependent. Despite its shortcomings, SFE has successfully been applied for the determination of fat content in foods 38-42,
pesticide residue 43-45 and total poly aromatic hydrocarbons (PAH) and
polychlorinated biphenyls (PCB) 44, 46-51.
MAE is another high diffusion fluid extraction technique which uses microwave energy to heat solvents in contact with a sample in order to partition
analytes from the sample matrix into the solvent. To be able to absorb microwave, either the solvent or the matrix has to be dipole with a high dielectric constant 52-54. It is also believed that high efficiency is obtained due to
the destruction of the matrix macrostructure 55. MAE could be performed in
19
a closed or open vessel heating system. The major advantages achieved using the MAE technique are reduced extraction times, improved extraction
efficiencies and reduced solvent consumption.
PFE, also been known as pressurized liquid extraction (PLE), pressurized
solvent extraction (PSE), accelerated solvent extraction (ASE) or enhanced
solvent extraction (ESE), is considered to be an extension of the SFE technique in which CO2 is replaced by organic solvents to solve potential polarity troubles 56. In the following section detailed discussions about the theory
of the PFE technique, instrumentation, capabilities and limitations as well as
selected applications from Papers I and II will be provided.
3.2. Theory and Concept of PFE Technique
PFE is a relatively new automated liquid extraction technique which uses
elevated temperature and pressure with standard liquid solvents to achieve
fast and efficient sample extractions 36, 57. The technique has been developed in an effort to reduce the large solvent consumption, long extraction
times and solvent exposure of operator compared to traditional liquid extraction techniques.
Liquid solvents at elevated temperature and pressure provide enhanced
extraction capabilities compared to their use at room temperature and atmospheric pressure. This is because increased temperature can disrupt the strong
solute-matrix interactions caused by van der Waal´s forces, hydrogen bonding and dipole attractions of the solute molecules and active sites on the matrix 36, 56. Higher temperature also decreases the viscosity and surface tension of solvents, allowing better penetration of solvents into the matrix particles. Both lower viscosity and lower surface tension facilitate better contact
of the analytes with the solvent and hence enhance the extraction 36, 56, 58.
Thus, the elevated temperature used in PFE increases the capacity of solvents to solubilize analytes, improves mass transfer and hence increases
extraction rates.
The purpose of higher pressure in PFE is mainly to keep the solvents as
liquid even at a temperature above their atmospheric boiling points. Pressurized flow would also aid in the solubilization of air bubbles so that the solvent penetrates the sample matrix more easily 36. Pressure has minor importance for the resulting recovery in PFE as has been shown in several studies
58-62. However improved extraction might be achieved for analytes trapped
in water blocked pores, as has been demonstrated by increased recoveries of
polycyclic aromatic hydrocarbons from a model matrix consisted of wet
silica 63. This could be explained by the fact that the organic solvents are
being forced into the pores at high pressure.
As a result of these combined effects, time and solvent consumption are
significantly reduced (Papers I and II) and improved (Paper I) or comparable
20
(Paper II) efficiencies are achieved when using PFE technique compared to
classical solvent extraction methods. Method development is easy with PFE
compared for example to SFE. Temperature and solvent are the two most
important parameters that affect extraction recovery of target analytes. In
Paper I, the extraction temperature, solvent and time were optimized using
experimental design. Co-extracted materials like lipids are the main problem
to be considered when using high temperature to extract lipophilic target
analytes. In Paper II, the temperature was optimized to reduce the coextracted lipids. However, even at the optimized temperature an additional
clean-up step was still required. If a particular extraction solvent works well
for the traditional methods, in most cases that solvent is also a good choice
for PFE 58-62. In Paper II, the same solvent as in the LLE method was used
which simplified method development.
Other parameters that can affect recovery include static time, particle size,
the flush volume and the number of static cycles used in the extraction. The
influence of particle size on the recovery of analytes using PFE has been
reported by Björklund et al. 64. They found that smaller particle size results
in increased extraction efficiency. Also in Paper I homogenized cod muscle
resulted in higher total lipid recovery compared to intact muscle. This is
because finely divided solids are more easily dissolved and extracted due to
their large surface area to volume ratio.
3.3. PFE Instrumentation
The commercial PFE instrument used in this study is a fully automated sequential extraction system. The system (ASE® 200, Dionex Co., Sunnyvalle,
CA, USA) can operate with up to 24 sample containing extraction cells and
up to 26 collection vials plus an additional 4 vial positions for rinse 65. Figure 6 shows a schematic diagram of PFE. To perform an extraction, a filter
paper is inserted into the extraction cell, followed by the sample. The filled
extraction cells are loaded on a carrousel, as well as collection vials are
loaded in the lower collection vial carrousel. A robotic arm transfers selected
cell separately into the oven for extraction. The oven is maintained at chosen
operating temperature, ranging from room temperature to 200oC. After the
cell is in place in the oven, the pump immediately begins to deliver solvent
to the extraction cell. The static valve closes to allow pressurization of the
cell. Because the solvent expands as it heats, the pressure in the cell will
increase when the static valve closes. When the pressure reaches 1.5 MPa
above the set point, the static valve opens rapidly to relieve the pressure and
then closes again. The pump also delivers fresh solvent to the cell to return
the pressure to the set point value. This addition of fresh solvent during PFE
is analogous to fresh solvent dripping down from the condenser onto the
extraction thimble during Soxhlet extraction 58. The first period of extraction
21
is called the heat up time, as the cell contents are heated by the oven to the
chosen operating temperature to reach thermal equilibrium. Compared to
MAE in which the solvent is heated directly by microwave radiation, heating
in PFE is a convection process in which the extraction cell is heated and heat
is transferred to the enclosed solvent. Once thermal equilibrium has been
reached the extraction enters a static period with a user-chosen duration.
Typical static times are 5-10 min but they can vary from 1 to 99 min depending on the optimum extraction time for the particular sample. After the static
time, fresh solvent is flushed through the cell to remove extracted analytes.
The flush amount can vary from 5 to 150% of the extraction cell volume
(with 60% as most common).
Figure 6. Schematic diagram of Pressurized Fluid Extraction system, Dionex ASE“
200.
If required, multiple extractions can be performed by programming from
one to five static cycles. The flush volume is divided by the number of static
cycles so that fresh solvent is present at the beginning of each static cycle.
Following the final solvent flush, compressed nitrogen is used to force all
solvent and remaining target analytes from the lines and cell, into the collection vial. After the completion of the purge step, the cell is returned to the
carrousel and the next sample is taken to the oven to begin the next extraction. Current instrumentation enables unattended operation of extraction
systems without user intervention if the sample matrix is simple and does not
block the solvent line (Paper II).
22
3.4. Selected Application and Comparison of the PFE
Technique with Conventional Methods
3.4.1. Total Lipid Extraction Using PFE (Paper I)
There is some uncertainty regarding the effectiveness of total lipid content
extraction using the traditional methods in lean fish muscle (fat content below 1%) where polar PLs dominate. Since the concentration of organic pollutants should be reported on both fresh and lipid weight basis, it is crucial
that the used extraction procedures lead to correct results for the pollutants
as well as for the lipids. In this study, the efficiency of the PFE technique
was investigated for the extraction of total lipid content from cod and herring
muscle tissue. The results were compared to original and modified Jensen
extraction methods.
The variables extraction time, temperature and extraction solvent were
optimized and optimum conditions were found to be 115oC with 2propanol/n-hexane (65/35, v/v) as a first and n-hexane/diethyl ether (90/10,
v/v) as a second solvent. Extraction time had no significant effect on total
lipid recovery and was kept at 10 min. PFE gave significantly higher lipid
yields (0.62%, RSD=3.1, n=5) compared to original (0.38%, RSD=6.3, n=5)
and modified (0.55%, RSD=2.8, n=5) Jensen methods from homogenized
wet cod muscle. The efficiency of modified Jensen was also further confirmed in comparison to original Jensen by replacing acetone with isopropanol and n-hexane with diethyl ether earlier reported by Jensen et al. 66. A
sub-portion of the freeze dried samples did not show channeling and gave
clear extracts however the low lipid yield (0.41%, RSD=2.5, n=5) could be
possibly due to difference in the extraction solvent composition as the wet
sample contains water as indigenous solvent and loss of lipids during freeze
drying process. PFE and original Jensen were also compared for the extraction of total lipid content from herring. PFE gave significantly higher yield
(7.8%, RSD=8.5, n=5) compared to original Jensen method (6.9%,
RSD=2.2, n=5).
The amount of total lipid extracted depends on the way the sample is
treated prior to extraction. To investigate this effect on the total lipid yield
homogenized and intact tissues were used and the results were compared to
the modified Jensen method. Significantly higher total lipid yields were
achieved with PFE [0.68%, RSD=3.7, n=5 using IPR/n-Hx (65/35, v/v) and
n-Hx/DEE (90/10, v/v)] and [0.68%, RSD=8.5, n=5 using IPR/DEE (25/10,
v/v) and n-Hx/DEE (90/10, v/v)] than the modified Jensen method (0.62%,
RSD=1.4, n=5) from a homogenized wet cod muscle.
Using PFE, the solvent mixture IPR/DEE (25/10, v/v) used by Jensen et
al. 66 together with n-Hx/DEE (90/10, v/v) gave significantly higher yield
than IPR/n-Hx (65/35, v/v) along with n-Hx/DEE (90/10, v/v) for intact cod
muscle (0.51%, RSD=18 n=5) compared to (0.36%, RSD=2.9 n=5) respec23
tively. However, when using homogenized tissue the statistical t-test showed
no significant difference in yields between the two extraction solvents (Paper
I, Table 5).
In conclusion, PFE gave about 10% higher yield in comparison to the
conventional batch extraction methods. Infrared and NMR spectroscopy
study suggested that the additional 10% lipid-like material is rather homogenous and is PL origin with head groups of inositidic and/or glycosidic nature.
It is difficult to find comparable data on similar biological tissue samples
with low lipid content as most PFE extractions were performed on high fat
tissues 67-69.
Despite the clean-up step, using PFE technique both the extraction time
and solvent consumption were reduced by about 50% compared to original
and modified Jensen methods.
3.4.2. PFE as an Alternative Method for Extraction of Pesticide
Residues from Fatty Matrix (Paper II)
In Paper II, a PFE multi-residue method has been validated for the determination of 25 pesticides and metabolites that cover a wide range of polarity
from rape seed. The reliability and efficiency of the PFE technique was investigated comparing the recoveries with conventional LLE technique. Finally incurred samples containing vinclozolin and iprodione were extracted
and compared using LLE, PFE and SFE. The extracts obtained were analyzed
on a GC using different detectors after clean-up by GPC.
When pesticide residues have to be determined from fatty matrices the extraction is associated with several analytical difficulties. Since the matrix is
complex, like in the case of rape seed with a high lipid content of 43%, the
recovery of very nonpolar fat soluble pesticides from the lipid-rich matrix is
troublesome. Clean-up of the extracts using GPC has shown to be necessary
in order to remove the co-extracted lipids. Purification of the extracts by
GPC resulted in a considerable decrease in the lipid content, Figure 7. Thus,
the interpretation of the chromatograms was simplified and the lifetime of
the GC column could be extended. Alumina was tested as an adsorbent to
retain the co-extracted lipids by directly adding alumina inside the cell. Despite the decrease in lipid content in the extracts, alumina as a fat retainer
was abandoned since most of the target analytes were also strongly adsorbed.
24
500
Residual lipids (mg g
-1
)
450
400
350
300
250
200
150
100
50
0
Before PFE
After PFE
After GPC
Figure 7. Residual lipids in rape seed (mg g-1) after PFE extraction and GPC purification (n=3) at 600C, 10 min, hexane saturated with acetonitrile,10 MPa, 60%
flush volume and 1 min N2 purge. The error bars represent RSD value.
The quantification was performed using standards dissolved in residue free
rape seed extract to eliminate the possible matrix effect on the recovery. Most
recoveries were within the acceptable range, 70-110% (Paper II, Table 2).
A comparison of extraction efficiency between the PFE and LLE was
conducted. The comparison of the extraction scheme is shown in Paper II,
Figure 3. Both methods recover the studied pesticides within acceptable
levels, with overall recoveries of 96% for LLE and 91% for PFE. The LLE
technique gave more precise results with lower RSD values. When using
fortified samples, there is a risk that the method does not imply the true recovery since the spiking on samples is performed with small amounts and
the penetration and binding of analytes into the matrix is not fully controlled.
Therefore, the PFE method was tested using residues from agriculturally
incurred samples and the results were compared to the LLE and SFE methods. The concentrations obtained are in good agreement with low RSD values of 1-4%, Table 1. Using PFE, both the extraction time and solvent consumption were substantially reduced by 90% compared to LLE. PFE also
allows multiple samples to be extracted sequentially in an automated system.
However the main disadvantage of the current PFE method is that a sample
clean-up is still required as an additional step.
Table 1. Extraction of pesticide residues from incurred samples of rape seed with
LLE, SFE and PFE determined by using GC/ECD and GC/MS/ITD.
Found concentration/mg kg-1 (RSD, %)
Sample
1 (n=4)
2 (n=2)
Pesticide
Vinclozolin
Iprodione
Vinclozolin
LLE
0.02 (1.1)
0.21 (2.5)
0.10
SFE
0.02 (2.7)
0.15 (1.7)
0.10
PFE
0.02 (4.3)
0.14 (3.2)
0.10
25
4. Analysis of Intact Phosphatidylcholine,
Sphingomyelin and Sulfatide Molecular
Species Using LC/ESI/MS and ESI/MS/MS
The interest in the analysis of lipids in general and phosphatidylcholine,
sphingomyelin and sulfatides in particular is continuously increasing due to
the importance of these molecules related to various disorders. Phosphatidylcholine, sphingomyelin and sulfatide molecular species exist in nature as
a complex mixture of closely related components which differ in the fatty
acid chain length, degree of unsaturation and hydroxylation. Based on the
fatty acid composition of the molecules each class can be separated into molecular species 6. These species differ greatly in their chemical and biological properties. They determine the interaction of the molecule in the cell
membrane and their identification and characterization has been of great
interest. Various analytical methods have been employed to separate and
analyze individual molecular species from its “intact” (underivatized or unmodified) form. The general advantages analyzing a lipid in its intact form
have been described as 6, 70, 71:
x
x
x
x
Elimination of the time consuming derivatization step
Possibility to study and follow biosynthesis, metabolism, turnover and
transport of the molecular species
Avoiding the danger of rearrangement of the FA chains during derivatization
Usefulness in the study of the immunogenicity of sphingolipid (sulfatides)
This section of the thesis is intended to give an overview of the analytical
methods such as RP-LC, ESI/MS and ESI/MS/MS (Paper V) and RPLC/ESI/MS and RP-LC/ESI/MS/MS (Papers III and IV) used for the analysis of molecular species of intact PLs and sphingolipids (mainly PC, SM and
sulfatide molecular species). The advantages and drawbacks of each analytical method will be discussed.
26
4.1. Reversed Phase LC of Intact PC and SM
Thin layer chromatography (TLC), LC, and GC are common techniques
used for separation of intact or derivatized PL molecular species. A review
of derivatization methods for PL and SM molecular species analysis using
these methods was recently presented by Wang et al. 72 and Toyo'oka 73.
Although most chemical derivatizations improve selectivity, resolution and
sensitivity, they are often very time and reagent consuming.
PL and sphingolipid analysis typically falls into two parts. The more general analysis determines each PL classes (e.g. PC, PE, PS or PI) including
SM based on differences on the hydrophilic portion of the lipid. The order in
which the lipid classes elute and separate depends primarily on the polarity
of the head group adsorption to the normal phase material such as silica gel,
74-77, Figure 8A. More specific analysis characterize the membrane of one
class, which share the common hydrophilic head group but differ in their
aliphatic residue (e.g. PC, SM and sulfatides, Papers III-V). Most PL molecular species analyses are carried by RP-LC of intact or after derivatization. In this section only analysis of intact PC and SM molecular species will
be discussed.
One or two dimensional silver ion TLC has been used in the past to separate PC and SM molecular species. However, further analysis by LC and/or
MS is required for specific molecular species analysis because of its low
resolution and poor reproducibility. Silver ion liquid chromatography (AgLC) either silica gel impregnated with silver nitrate or by binding silver ion
to an ion exchange phase has been used for lipid molecular species analysis
6, 78. The principle of the method is that silver ions interact reversibly with
the pi electrons of the double bonds (cis more strongly than trans) to form
polar complexes. The greater the number of double bonds in the FA chain,
the stronger the complex formation and the longer it is retained 6, 78-80. It
enables the separation of the molecular species according to the number,
geometry and position of the double bonds in FA chain 6, 78. The mechanisms of the interaction between Ag-LC and unsaturated FA have been studied in detail by Nikolova-Damyanova et al. 79. They concluded that silver
ion simultaneously interacts with the double bonds of the FA and other electron rich moiety such as the ester group. In general Christie and his coworkers made the greatest progress in developing Ag-chromatographic systems 6, 81-83.
27
(A) Normal phase
Hydrophilic head group
R1
(B) Reversed phase
R2
Hydrophobic FA chain
Figure 8. Separation of PLs using liquid chromatography A) PL class separation
using normal phase. B) PL molecular species separation using RP. The separation
is mainly due to specific hydrophobic interaction of the FA chain with the RP.
RP-LC is widely used for the separation of lipid molecular species. In Papers III and IV using a RP column 200x0.2 mm (Paper III) and 200x0.5 mm
(Paper IV) packed with YMC-GEL ODS A C18 120 Å, 5 µm, the PC and
SM were separated into their molecular species. The separation is mainly
based on the hydrophobic interaction of the FA tail group with the RP packing material, Figure 8B. PC and SM have higher polarity of the head group
in comparison for e.g. to PE. As the polarity of the head group increases the
efficiency of the chromatography decreases due to interactions between polar groups of molecules immobilized in the stationary phase (SP) and those
dissolved in the mobile phase (MP) as well as between the polar groups of
the molecular species and any exposed silanols on the RP column 84. This
problem can be reduced by the appropriate choice of SP and MP. Among the
different MPs studied a mixture of methanol/tetrahydrofuran/water (80:15:5)
in 0.1% formic acid at a flow rate of 2 µL/min provided the best separation
of the molecular species. It has the advantage that isocratic elution and therefore a simple and inexpensive LC pump can be used. Furthermore, the absence of non-volatile components (like salts) in the MP allows subsequent
identification by ESI/MS and isocratic elution reduces the ESI fluctuation
considerably that could be caused by changes in gradient elution. It was
found that the elution order of the molecular species within each class depends mainly on the length and the number of double bond on the FA chain.
Thus the longer and the more saturated the FA chain, the longer the retention
time of molecular species. Each double bond reduces the apparent FA chain
length by approximately less than two carbon atoms 85, 86.
28
The main problem of the developed method was a shift in retention time
of the molecular species after about 40-50 injections of the real sample extract. This could possibly be due to the lipid materials and artifacts built up
in the column. It was thus necessary to wash and equilibrate the column for 5
min with the MP before the next injection could be performed. The shift in
retention time may not be considered as a main problem if MS is used as a
detector. In Papers III and IV, MS was used as a detector and the identity of
the chromatographic peaks could be easily confirmed even if there would be
a shift in retention time. For the real sample analysis in Paper IV the column
was changed to a larger internal diameter (0.5 mm, at a flow rate of 15
µL/min) in order to avoid frequent plugging of the column, variation in retention time from run to run and to accommodate the large number of runs
from the crude lipid extracts. Moreover, switching the MS between positive
and negative mode for a while after a certain number of runs helped to stabilize the electrospray and improve the chromatographic peak shape by dumping any material build-up on the tip of the capillary transfer line (i.d. 50 µm).
All the columns used in this thesis were slurry packed in our laboratory at a
fraction of the cost of commercial columns and hence very cost effective.
4.2. Mass Spectrometry
Mass spectrometry is a powerful detection technique that enables separation
and characterization of substances according to their mass-to-charge ratio
(m/z). Its essential components include a sample inlet, ion source, mass analyzer, detector and data handling system. The combination of sensitivity,
selectivity, speed and its ability for structural information of unknown lipid
compound makes MS an ideal method for intact lipid molecular species
analysis. It can be performed by direct infusion (Paper V) or coupled to a
separation technique such as RP-LC (Papers III-IV).
4.2.1. The Mechanism of ESI/MS
ESI/MS, first conceived in the 1960s by chemistry professor Malcolm Dole
but was put into practice by Fenn et al. 87, 88, is a technique for transformation of ions from a liquid to a gas phase. It is a simple method that operates
at atmospheric pressure and ambient temperature. The principal mechanism
of ESI is under investigation and still not fully understood. Many theories of
the physical processes involved in ion generation during ESI have been proposed and validated in some detail 88-91. The principles of the ESI process
can be schematically represented as shown in Figure 9. Initially, the LC effluent (Papers III and IV) or direct infusion (Paper V) containing the analytes of interest is pumped into the ESI ion source through a capillary tubing.
The narrowed orifice at the end of the capillary tubing and the mechanical
29
forces imparted as the solution passes through the narrow orifice facilitate
the formation of sprayed small droplets in the ionization chamber. If an electric potential (approximately 2-5 kV) is applied between the end of the capillary tube and the entry into the mass analyzer, because of oxidation/reduction processes, the droplets can carry net charges and are directed
into the mass analyzer by the applied electric field. This applied potential
could be either positive or negative depending on the substance to be analyzed. As shown in Figure 9, if a positive electric potential is applied to the
end of a capillary tube and a negative electric potential is present at the entrance of the mass analyzer in the positive-ion mode, droplets carry net positive charges. The resulting electric field causes charge separation of ions in
the solution, with positive ions moving towards the liquid surface and negative ions migrating back into the liquid, Figure 9. During flight, the droplets
are desolvated, by passage through a curtain gas. Thus, during desolvation,
the coulombic force between ions is dramatically increased. Once this force
exceeds the surface tension of the solvent, the droplets explode to form
smaller droplets. This cycle is iteratively repeated until molecular ions are
generated prior to their entrance into the mass analyzer 7, 89. The exact
mechanism of ion formation, whether it is by ion evaporation (ionevaporation model) or by complete solvent removal (charge-residue model)
from the charged droplet is under debate, and probably different mechanisms
may apply in different situations.
Figure 9. Schematic diagram of the principles of electrospray ionization in positive
ion mode.
30
4.2.2. The Principles of ESI/MS of Lipids
The soft ionization MS techniques such as fast atom bombardment (FAB) 9294, field desorption (FD) 95, thermospray (TS) 96-99, matrix assisted laser
desorption ionization (MALDI) 100, 101 and ESI (Papers III-V) have the ability to ionize lipid molecules with out causing extensive fragmentation. Recently ESI has become the preferred and widely used method to ionize biologically important lipid molecules. In a recent review Han and Gross 7,
classified the lipid classes into four groups based on their electrical propensities which have been exploited for ESI/MS analyses of lipids. These include:
x
x
x
x
Anionic lipids: containing one or more net negative charge(s) at neutral
pH such as sulfatide, PS, PI, phosphatidylglycerol, etc.
Weak anionic lipids: carrying a net negative charge at alkaline pH such
as PE, lysoPE, ceramide etc.
Neutral polar lipids: are polar but electrically neutral lipids at both neutral and alkaline pH. This group of lipids includes choline containing lipids (PC, lysoPC and SM), diacylglycerol, triacylglycerol etc.
Special lipids: varying in their electrical propensity. Includes acylcarnitines, cholesterol and its derivatives including cholesterol ester and
sterols, etc.
This part will only focus on the ESI/MS behavior of the choline containing
lipids PC and SM (Papers III and IV) and sulfatide (Paper V).
Both PC and SM are choline containing species and are closely related in
their ESI properties. They are characterized by the presence of quaternary
nitrogen atom whose positive charge is neutralized by the negative charge of
the phosphate group. The quaternary nitrogen atom readily forms an abundant [M+H]+ molecular ion by ESI in positive ion mode because the phosphate anion can be protonated during the electrospray process. Furthermore,
when various alkali metal ions such as Na+ are present in the electrospray
solvent sodiated species [M+Na]+ are observed in a positive mode. In Paper
III both the [M+H]+ accompanied by sodium adduct [M+Na]+ molecular ions
were observed.
Sulfatide contains a polar galactosyl-3-sulfate that carries a negative
charge and can be detected in a negative mode ESI. Abundant negative ion
[M+H]- molecular species were produced by ESI/MS (Paper V).
4.2.3. MS Scan Modes
A number of mass analyzers are available, e.g. quadrupole, ion trap, time of
flight, ion cyclotron resonance and sector instruments, which separate
charged species in vacuum depending on their m/z ratio. Apart from separation, the mass analyzers can also be used for fragmentation of the charged
31
species. The most widely used are the triple quadrupole instrument (Papers
III-V). In a triple quadrupole instrument the middle field free quadrupole
focuses and transmits almost all ions and may be used as a collision cell for
fragmentation reaction. The type of fragmentation occurring in quadrupole
instruments is referred to as collision induced dissociation (CID). As a result
of the collision, the internal energy of the ion may be increased by conversion of kinetic energy into internal energy 102, 103. The choice of collision
gas, its pressure and the collision energy affect the degree of fragmentation.
The fragment ions are then analyzed in the second mass analyzer. In addition
to structural information tandem MS provides higher sensitivity and reduced
chemical background by selecting different scan modes. In Paper III, in order to obtain structural information of the PC and SM molecular species,
tandem MS was performed at optimized collision energies on the triple
quadrupole instrument. Product ion scans of standard PC and SM of the
[M+H]+ and [M+Na]+ ions were performed, where the first quadrupole is
locked to a selected m/z ratio and the third is scanning the daughter ions produced in the collision cell. Figures 10 and 11 show the product ion mass
spectra for the standard PC16:0/18:1 species of the [M+H]+ ion at m/z 760.6
and the [M+Na]+ ion at m/z 782.6, respectively. The fragment ions are consistent with the fragment structures proposed in the literature 104, 105. The
CID product ion spectrum differs depending on the choice of the parent ion.
The CID product ion spectrum of [M+H]+ produced few and low abundant
fragment ions in comparison to the [M+Na]+. However, both tandem spectra
of [M+H]+ and [M+Na]+ ions contain a characteristic intense ion at m/z 184,
corresponding to the head group, phosphocholine [H2PO4CH2CH2N(CH3)3]+.
CID of the [M+H]+ ion produced low intensity fragment ions at m/z 577, 504
and 478 due to loss of [HPO4CH2CH2N(CH3)3], sn-1 FA (16:0) and sn-2 FA
(18:1), respectively. The fragment ions at m/z 522 and 496 resulted from the
and
loss
of
alkyl
ketenes
[CH3(CH2)13CH=C=O]
[CH3(CH2)7CH=CH(CH2)6CH=C=O], respectively.
The CID spectra of sodiated standard PC16:0/18:1 species is shown in
Figure 11. Neutral loss of trimethylamine, N(CH3)3 [M+Na-59]+ yielded an
ion at m/z 723. Ions at m/z 599 [M+Na-183]+ and 577 [M+Na-204]+ are
through the loss of [HPO4CH2CH2N(CH3)3] and [NaPO4CH2CH2N(CH3)3]
respectively. Neutral loss of the sn-1 FA substituents as free FA [M+Na256]+ or as sodium salt [M+Na-278]+ are observed at m/z 526 and 504, respectively. In addition, neutral loss of the sn-2 FA substituents as free FA
[M+Na-282]+ or as sodium salt [M+Na-304]+ are observed at m/z 500 (weak
signal) and 478, respectively. The fragment ions at m/z 239 and 265 represent
the
acylium
ions
of
[CH3(CH2)14C=O+]
and
+
[CH3(CH2)7CH=CH(CH2)7C=O ], respectively. Losses of trimethylamine
plus palmitic acid or oleic acid are observed at m/z 441 and 467 respectively.
As confirmed by Hsu et al. 104 using lithiated adducts of standard
PC16:0/18:1 and PC18:1/16:0 comparison of the relative abundance of the
32
ions at m/z 441 and 467 in Figure 11 indicates that the abundance of the ions
reflecting loss of trimethylamine plus the sn-1 substituent exceeds that of the
ion reflecting loss of trimethylamine plus the sn-2 substituent. This permits
the assignment of FA moieties as sn-1 or sn-2. However Ho and Huang 105
suggest that an average of over 40 spectra is required to produce a consistent
abundance ratio that lead to correct assignment of the FA moieties. It can be
concluded that the sodium adduct greatly improved the information content
of the CID spectra by yielding diagnostic fragment ions that provide more
detailed information about the structure of the PC in comparison to [M+H]+.
PIS can also allow the selective detection of PC and SM molecular species among other PL classes by head group scanning (by scanning for precursor of m/z 184). However this scan mode is non specific between PC and
SM and do not yield information about FA substituent, hence it is difficult to
assign the spectra unless a separate CID product ion scan is performed in
parallel.
496
184
100
504
478
577
522
50
400
450
500
550
600
760.6
(M+H)+
0
0
100
200
300
400
500
m/z
600
700
800
+
Figure 10. Product ion mass spectra of the [M+H] from PC16:0/18:1 at m/z 760.6
using nitrogen as a collision gas.
467
526
184
100
441
239
478
504
782.6
(M+Na)+
265
200
250
300
350
400
450
599
50
500
550
723
146
577
0
0
100
200
300
400
500
600
700
800
m/z
Figure 11. Product ion mass spectra of [M+Na]+ from PC16:0/18:1 at m/z 782.6
using nitrogen as a collision gas.
33
When single quadrupole instrumentation is used, fragmentation reaction
can be induced in the source region by elevating the cone-to-skimmer potential difference. Protonated molecules desorbed from the electrosprayed droplets are accelerated between the cone and skimmer and can undergo CID in
collision reaction with residual gas molecules 106. This process is called upfront fragmentation or nozzle-skimmer fragmentation. In Papers III and IV
up-front fragmentation was performed in order to confirm the assignment of
the chromatographic peaks from the biological sample. The orifice potential
was varied between the optimal settings for quantitation (30 V) and up-front
fragmentation (120 V) for structural confirmation. Figures 12A and B represent the up-front fragmentation mass spectra of the deuterium labeled 1,2Dipalmitoyl-sn-Glycero-3-Phosphocholine (d4-16:0/16:0-PC) standard at
orifice potentials of 30 V and 120 V respectively. The 1,2-Dipalmitoyl-snGlycero-3-Phosphocholine (d4-16:0/16:0-PC) was deuterium labeled at its
head group. When a low orifice potential was applied the [M+H]+ ion m/z
738.6 was dominant and used for quantification. The spectrum obtained at
the higher orifice potential showed substantial fragmentation as indicated by
the presence of the dominant phosphocholine ion at m/z 188.6 (a fragment of
the deuterium labeled head group IS). The results obtained from the real
sample extracts also showed a dominant [M+H]+ ion (in this case m/z 760.6,
Figure 12C) at low orifice potential, and at high orifice potential the phosphocholine fragment ion m/z 184.6, becomes the dominant peak in the mass
spectra, Figure 12D.
(M+H)+
738.6
A
Intensity (cps)
5000000
Intensity (cps)
Phosphocholine ion
188.6
0
150
250
350
Phosphocholine ion
184.6
450
550
650
750
0
160
850
260
360
m/z
Intensity (cps)
5000000
(M+H)+
760.6
C
4500000
460
560
660
760
B
Phosphocholine ion
188.6
Intensity (cps)
3000000
250
350
450
550
m/z
960
D
Phosphocholine ion
184.6
(M+H)+
760.6
(M+H)+
738.6
0
150
860
m/z
650
750
850
0
150
250
350
450
550
650
750
850
m/z
Figure 12. Up-front fragmentation of the deuterium labeled 1,2-Dipalmitoyl-snGlycero-3-Phosphocholine (d4-16:0/16:0) standard and 16:0/18:1 from human
brain extract. (A) d4-16:0/16:0 at low orifice potential (30 V) (B) d4-16:0/16:0 at
high orifice potential (120 V), (C) 16:0/18:1 from human brain extract at low orifice
potential (30 V), and (D) 16:0/18:1 from human brain extract at high orifice potential (120 V).
34
In Paper V, sulfatide molecular species were analyzed by PIS. PIS is
based on independent operation of two mass analyzers positioned inside a
tandem MS at both sides of the collision cell. The second analyzer (quadrupole Q3) is usually set to transmit ions with a certain constant m/z, while the
first mass analyzer (quadrupole Q1) is scanning. Ions passing through the
first mass analyzer Q1 fragment in the collision cell are recorded by the detector only if they produce a fragment with specific m/z at which the second
analyzer has been set. Therefore, PIS selectively detects precursor ions that
produce the same characteristic fragment ion upon their CID even if a very
complex mixture is injected into the mass spectrometer. In Paper V, the acquisition of precursor ion spectra of the sulfate fragment ion allowed detection of major sulfatide molecular species. CID was performed with nitrogen
as a collision gas. The intense signal at m/z 97 corresponds to the HSO4- ion.
Thus, sulfatide can be detected by scanning for the precursors of m/z 97 in
the negative ion mode 107-109. First, synthetic IS 14:0-sulfatide was used to
optimize the instrument operational parameters and intense [M+H]- ion was
obtained at m/z 750.6 that corresponds to the 14:0-sulfatide IS.
The developed method was applied to compare endogenous sulfatide molecular species accumulation in arylsulfatase A knockout (ASA -/-) developed as
a model of metachromatic leukodystrophy and wild type control mice (ASA +/+)
at different ages. An alteration in the composition of sulfatide molecular species was observed between the ASA -/- and ASA +/+ mice with accretion of
sulfatide mainly with stearic acid in the ASA -/- mice (15% higher at the
older ages). In cultivated astrocytes the proportion of short chain FAs (C16C18) was 60%, with C18 FAs constituting 40%.
4.3. Reversed Phase LC/ESI/MS
RP-LC is widely used for the separation of intact lipid molecular species
using different detectors. Phospholipids and sphingolipids lack chromophores that permit direct specific spectrophotometric detection. Some of the
common detectors used for intact molecular species analysis are ultra violet
(UV) 85, 86, 110-112, evaporative light scattering detector (ELSD) 81, 113, 114
and MS (Papers III-V). LC with UV detection offers a fast and convenient
method however sever limitation are imposed on selecting the MP since UV
absorption of underivatized PL molecules only occur near 200 nm depending
on the degree of unsaturation of the FA moieties. UV detection in the 200210 nm range creates problem for quantification because the detector response is not the same for various molecular species and most of the commonly used MPs absorb strongly in this region 112, 115. In ELSD detectors
the column effluent is nebulized to an aerosol which is subsequently evaporated, leaving the solute components as fine droplets. These droplets are
illuminated by a laser and the amount of scattered light is measured. Such
35
detectors are universal in their applicability, because they respond to any
lipid molecular species that does not evaporate 115. Moreover both UV and
ELSD do not give structural information of the molecular species and confirmation of each peak identity must be established using other techniques
such as GC/MS or tandem MS. The advantages of UV and ELSD are that
they are simple, inexpensive and versatile. The ELSD has an advantage over
UV in that the response is largely independent of the number and configuration of double bonds in the FA chains, thus making quantification more
straight forward 70, 115.
In order to analyze molecular species of lipids from complex biological
samples, development of a more efficient technique which can simultaneously provide unambiguous structural information and sensitivity is required.
An alternative method to overcome some of the above difficulties in the
analysis of intact lipid molecular species is to use LC coupled with ESI/MS.
It is one of the powerful techniques used today in the analysis of intact polar
lipid molecular species. The separation power of LC into either different
lipid classes or molecular species within a class together with selective MS
detection makes the simultaneous determination of protonated molecules and
the identification of several structural elements possible 116. The LC/MS
system used in Papers III and IV is shown in Figure 13.
Figure 13. Experimental set-up of the LC/MS and LC/MS/MS system used in Papers
III and IV. Pump: Jasco PU-980 operated at constant pressure, Injector: a 60 nL
and 200 nL internal loop, Column: 200x0.2 mm (Paper III) and 200x0.5 mm (Paper
IV) packed with 5 µm particles C18 (ODS A, YMC-GEL), capillary transfer line 50
µm i.d.
LC/MS is still limited to conditions that are suitable for ESI/MS conditions. There are restrictions on pH, choice of solvent, solvent additives and
flow rates for LC in order to achieve optimal MS sensitivity 90. It is important to select a solvent system that renders chromatographic resolution and is
compatible with ESI/MS performance. In Papers III and IV a mixture of
methanol/tetrahydrofuran/water (80:15:5) in 0.1% formic acid was found to
be a suitable MP at a flow rate of 2 µL/min during isocratic elution. By reducing the capillary LC and using lower flow rates the sensitivity can be
36
enhanced due to lower flow rates and the smaller i.d. capillaries resulting in
smaller droplets and increased charge to volume ratio (increased analyte
concentration). In Papers III and IV, using a capillary column, the PC and
SM molecular species were separated successfully, Figure 14. Reconstructed
ion chromatogram using standards of PC and SM at 60 nL injection volume
in single ion monitoring mode indicated a base line separation of the species.
The single ion monitoring mode showed minimum detectable quantity as
low as 8 fmol using deuterium labeled 1,2-Dipalmitoyl-sn-Glycero-3Phosphocholine (d4-16:0/16:0) IS. It also showed good linearity (r2=0.9991)
over two orders of magnitude in concentration range. More over, as the column i.d. is decreased, the lower volumetric flow rate through a capillary
column would minimize consumption of chemicals in the MP and amount of
sample consumed to perform an analysis. In Papers III and IV, 60 and 200
nL volume was injected respectively, using an internal loop injector, which
gave an added advantage when working with a limited amount of sample. In
Paper IV, the developed method was applied to investigate the effect of
postnatal iron administration in lipid composition (PC and SM molecular
species) in brains of eight groups of mice containing six individuals in each
group treated with vehicle, low dose clozapine, high dose clozapine and
haloperidol.
Peak identification
1. SM 16:0
2. PC 16:0/18:2
3. SM 18:0
4. PC 16:0/18:1
5. PC 18:0/18:2
6. SM 20:0
7. PC 18:0/18:1
8. SM 24:1
4
250000
Intensity (cps)
2
3
1
5
8
6
7
0
5
10
15
20
25
30
35
Time (min)
Figure 14. Reconstructed ion chromatogram for the [M+H]+ ions in a standard
phosphatidylcholine and sphingomyelin mixture in single ion monitoring mode.
Approximately 4 pmol of the PC and SM were injected into the system.
37
5. Data Handling
Nowadays, modern analytical instrumentations make it possible to record
very large data sets in short period of time in chemical laboratories and there
is a need for efficient data handling. Chemometrics involves experimental
design and multivariate analysis methods that include mathematical and
statistical methods for modeling and data handling.
Multivariate methods such as principal component analysis (PCA) and
projections to latent structure (PLS) make up a set of mathematical tools that
may be applied to the analysis and interpretation of spectroscopic data. The
data structure can be represented in 1- and 2-dimentional plots 117, 118. A
more detail descriptions of these methods can be found elsewhere 119, 120.
PCA is a technique used to reduce the number of variables needed to describe the relevant variations. It is useful for data compression and information extraction. The results must be presented so that the desired information
can be easily understood (e.g. in a graph form, like in Paper III) or expressed
in a few information rich parameters. In Paper III, PCA and PLS approach
was used for the lipid concentration and behavioral variables to elucidate the
variations between the various treatment groups.
Experimental design is useful to find out the optimal operating conditions
and factors that give the best recovery with a small number of experiments
121. In Paper I, an optimization study was carried out to investigate the optimum values of the different variables (temperature, time and extraction solvent) expected to have the largest effect on the extraction efficiency of total
lipid content from cod muscle.
38
6. Conclusions and Future Aspects
This thesis covers a wide range of analytical method development for lipid
analysis in complex biological samples; from sample preparation using PFE
and separation with RP-LC to detection by ESI/MS and tandem MS. It
shows that modern analytical methods can provide new insights in the extraction and analysis of lipids from complex biological samples.
In Papers I, II and III, PFE was used as an alternative extraction technique
in comparison to different traditional liquid extraction methods. Recoveries
and precision are equivalent or better than those achieved by traditional extraction methods. The PFE technique has the additional advantage in terms
of small solvent volume consumption, rapid extraction times and automation. Using PFE, both the extraction time and solvent consumption was substantially reduced by 90% (Paper I) and 50% (Paper II) compared to the
traditional extraction methods. The main drawback of the PFE technique is
the lack of selectivity and hence further sample clean-up steps were found to
be needed (Papers I-III).
To use environmentally friendly solvents such as water, PFE in combination with pressurized hot water as an extraction solvent is suggested to be a
favored approach in the future. When heated above its boiling point in a
closed container, water becomes a more powerful solvent 122. Ease of coupling to other analytical instrumentation could be an additional advantage
offered by this technique in the future. Further automation and integration of
extraction, clean-up and detection steps in single reliable automated process
appears to be the next challenge for PFE. However it seems unlikely that
future generations of the PFE technique will incorporate higher temperature
capabilities (> 200oC) because most reported work lies in the normal operating range of 75-150oC. This is limited by reported problems with analyte and
matrix stability 1.
Functional lipidomics is an emerging field of life science and is fast becoming important in the understanding of lipid related diseases. In Papers
III-V a targeted lipidomics approach was presented to characterize and profile PC, SM and sulfatide molecular species using LC/MS and MS/MS. A
successful separation of PC and SM molecular species was obtained with a
minimum detectable quantity in the low fmol range injected on column (Papers III and IV). Although some of the need for LC separation is eliminated
by using advanced MS instruments which can utilize different scan modes
for separation and also handle lower detection limits, the coupling of LC is
39
still the most promising approach since it can handle total lipid extracts of
lower concentration and of more complex nature as an added advantage.
Moreover, mass spectrometric injection through LC will often reduce ion
suppression effects that are introduced by other components in the crude
lipid extract or competition between the analytes (lipids). In Paper V, a tandem MS in PIS mode was used to analyze the developmental profile of molecular species brain sulfatide accumulation in ASA deficient (ASA -/-) and wild type
control (ASA +/+) mice.
The term “lipidomics” has just emerged in the literature due to recent
technological advances in LC/MS and MS/MS and rising concerns over lipid
associated disorders. One strategy to combat these diseases is to understand
how the composition of cellular lipids and lipid membrane changes. However analysis and quantification of these changes is complicated by the structural diversity, instability, the wide range of lipid concentration and the specialized biochemical systems for their biosynthesis and metabolism 123.
Thus, it seems that lipidomics might benefit from a more targeted approach
that will focus on a specific group of lipids. Combining data from the analysis of various lipids is likely to produce a complete lipidomic map, possibly
with the help of bioinformatics methodology that will be developed in the
future for lipid profiling.
40
7. Acknowledgements
I would like to express my sincere gratitude to:
My supervisors Assoc. Prof. Jonas Bergquist, Prof. Karin Markides and
Monica Waldebäck (senior lecturer) for their limitless effort and guidance,
helpful and interesting discussions, encouragement and making me see the
positive side of the world throughout my graduate study period. Karin
Markides for always making me concentrate on my studies whenever there
have been ups and downs in the source of funding. Jonas Bergquist for introducing me to neuroscience and lipid mass spectrometry and for always
being there to listen to me even if I knock at his office door late after 18:00.
Monica Waldebäck for introducing me to the research world in general and
PFE technique in particular and for all the information and social life she
provided me during my first arrival to Sweden.
Dr. Dan Bylund, Assoc. Prof. Jan-Eric Månsson and Assoc. Prof. Anders
Fredriksson for being generous with their time, knowledge and valuable
discussions in all lipid/MS projects. Prof. Göran Odham and Ulla Eriksson
from Stockholm University for all valuable discussions and meetings in lipid
extraction. Assoc. Prof. Rolf Danielson for helpful discussions about
chemometrics and data handling. Andreas Dahlin for his help with the
graphics. All my co-authors for their interesting and fruitful discussions.
Dr. Ardeshir Amirkhani and Sara Bergström for introducing me to the practical lab work in mass spectrometry and column packing respectively. Daniel
Bäckström for his kind assistance in all computer related problems.
All staff members, present and past graduate students at the department of
Analytical Chemistry who always remained positive to give any assistance
and for providing a friendly and comfortable atmosphere to work in. I really
enjoyed the social life and extracurricular activities at the department such as
Friday evening Spray Bar, innebandy club (even if I am not as good as….),
summer soccer game, hiking in the north of Sweden and the recently introduced “Film Kväll”.
Barbro Nelson and Yngve Öst for all administrative and technical assistance.
41
My home University, Asmara University in Eritrea for giving me the opportunity to study in Sweden. SIDA/SAREC for the financial assistance up to
my Licentiate degree and Uppsala University for finding financial means to
complete my studies.
All staff members of the International Science Programme (ISP) for their
assistance in all technical and administrative matters in Sweden.
All my friends, especially the SAUA-group and all Eritrean friends residing
in Sweden for their continuous encouragement, advice, motivation, social
life and for making me not miss home. Especially Asmerom with his wife
Selamawit for the hospitality and warm reception whenever I stay at their
home in Stockholm, Husby. All my best friends back home in Asmara, especially Zekarias E., Semere Tekle and Semere T/mariam.
My Swedish family Mr. Börje Nordh for introducing me to the Swedish
culture, for always being there to drive me anywhere at my convenience and
for the wonderful “Fika” company and refreshment at times my head was
overwhelmed with a lot of Chemistry. It was fun every year helping you
decorate your Christmas tree. One thing though, would you please stop beating me in “Mini Golf” and let me win only for a single time!!
Finally, my mother Emuna as well as my brothers and sisters for their continuous motivation, inspiration and moral support throughout my graduate
study period.
42
8. Summary in Swedish
Lipider är inblandade i många vitala biologiska processer i djur, växter och
mikroorganismer. Störningar i lipidmetabolismen kan medföra en mängd
olika problem och därför är utvecklandet av metoder för extraktion och analys av lipider avgörande för vår förståelse av dessa system.
Många av de traditionella provextraktionsmetoderna har blivit den svagaste länken i modern kemisk analysmetodik. Traditionella metoder kan vara
tillförlitliga men långsamma, kräva stora mängder organiska lösningsmedel,
vara svåra att automatisera och därför mycket arbetskrävande. Under de
senaste decennierna har drivande krafter, såsom behovet av snabbare analyser, ökad produktivitet, hänsynstagande till kostnader för analyser, säkerhet
för laboratoriepersonal och reduktion av miljöbelastande utsläpp, uppmuntrat kemister att utveckla nya extraktionstekniker. Superkritisk vätskeextraktion, SFE, mikrovågsextraktion, MAE, och trycksatt lösningsmedelsextraktion (PFE) är moderna tekniker, där tryck och temperatur kontrolleras och på
så sätt förändrar diffusionsegenskaperna hos lösningsmedlen och massöverföringen under själva extraktionen. Dessa tekniker ger överlägset snabbare
extraktioner och system som är enklare att automatisera. De ger dessutom
bättre eller likvärdiga utbyten, möjlighet till selektivitet samt eliminerar eller
kraftigt minskar volymen av hälsovådliga och miljöbelastande lösningsmedel. I denna avhandling studeras tillförlitlighet och effektivitet av PFE för
extraktion av totalt lipidinnehåll från torsk- och sillmuskel (Artikel I), pesticider i rapsolja (Artikel II), och fosfatidylkolin (PC) och sphingomyelin
(SM) från human hjärnvävnad (Artikel III).
Ur ett ekologisk perspektiv är det mer relevant att uttrycka den kvarvarande mängden lipofila miljögifter i mängden fett snarare än som total våt
vikt. Åtskilliga jämförande laboratorier har visat att den ursprungliga Jensenextraktionsmetoden för totala lipider och organiska föroreningar ger tillfredsställande utbyten när den används för fettrik vävnad från marina organismer. I mager fiskmuskel (där fettinnehållet är <1%) är däremot utbytet
med denna metod ca 25% under det egentliga värdet, vilket i sin tur leder till
att mängden miljögifter per lipidinnehåll blir falskt förhöjt. En PFE-metod
har utvecklats, där olika extraktionsvariabler (temperatur, tid och sammansättning av en blandning av lösningsmedel) har undersökts i ett flerfaktor
försök. Den utvecklade PFE-metoden har beskrivits (Artikel I) och resultaten
har jämförts med olika traditionella extraktionsmetoder. Den utvecklade
43
metoden sparade tid (50%) och lösningsmedelsvolymen minskade med,
50%.
I Artikel II beskrivs en multimetod för att extrahera pesticider, allt från
polära till helt opolära, i rapsfrön. Svårigheten är att extrahera de opolära
pesticiderna ur de fettrika fröna utan att få med fett från rapsfröna, då detta
fett stör den slutliga analysen. Den utvecklade metoden har jämförts med
den metod Livsmedelsverket använder idag. Utbytena är likvärdiga men den
utvecklade metoden är väsentligt snabbare samt minskar lösningsmedelsanvändningen med 90 %.
Vidare har i Artikel III en metod utvecklats för separation och karakterisering av PC och SM genom användandet av reversed phase kapillärvätskekromatografi (RP-LC) kopplat till elektrosprayjonisationsmasspektrometri
(ESI/MS). Därutöver jämfördes extraktionseffektiviteten av traditionell
Folch-metod och PFE för human hjärnvävnad.
I Artikel IV användes den utvecklade metoden för att undersöka lipidsammansättningen i hjärnvävnad från möss som behandlats med placebo
eller neuroleptika (låg eller hög dos clozapine eller haloperidol). Effekten av
postnatal järnbehandling på lipidsammansättning och beteende undersöktes i
djuren. Därefter studerades om dessa effekter påverkades av subkronisk neuroleptikabehandling.
I Artikel V användes tandemmasspektrometri för s.k. precursor ion scan
(PIS) för att karakterisera sulfatidinnehållet i transgena möss (knock-out)
som saknar arylsulfatas A (ASA). Denna musstam användes som en sjukdomsmodell för den humana monogena sjukdomen metakromatisk leukodystrofi (MLD) där funktionen av det lysosomala enzymet ASA är nedsatt.
Våra analyser påvisar en förändrad sulfatidsammansättning hos bristdjur
jämfört med kontrolldjur.
44
9. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Handley, A. J. Extraction Methods in Organic Analysis; Sheffield Academic
Press Ltd: Sheffield, 1999.
Majors, R. E. LCGC 1999, 17, S8-S13.
Bignert, A.; Göthberg, A.; Jensen, S.; Litzen, K.; Odsjö, T.; Olsson, M.; Reutergårdh, L. Sci. Total Environ. 1993, 128, 121-139.
Quasimeme ; Quasimeme Project Office, Marine Laboratory: Aberdeen, UK,
1995.
Piomelli, D. Prostaglandins 2004, In Press.
Christie, W. W. Lipid Analysis: Isolation, Separation, Identification and Structural Analysis of Lipids, Third ed.; The Oily Press: Dundee, 2003.
Han, X.; Gross, R. W. Mass Spectrom. Rev. 2005, 24, 367-412.
Han, X.; Gross, R. W. J. Lipid Res. 2003, 44, 1071-1079.
Han, X.; Yang, J.; Chang, H.; Ye, H.; Gross, R. W. Anal. Biochem. 2004, 330,
317-331.
Spener, F.; Lagarde, M.; Record, M. Eur. J. Lipid Sci. Technol. 2003, 105, 481482.
Lagarde, M.; Geloen, A.; Record, M.; Vance, D.; Spener, F. Biochim. Biophys.
Acta 2003, 1634, 61-61.
Wilson, J. F. The Scientist 2003, 17, 34-36.
Prestwich, G. D. Prostaglandins 2004, In Press.
Martindale, D. The Scientist 2003, 17.
Singer, S. J.; Nicolson, G. L. Science 1972, 175, 720-731.
Ekroos, K. Ph.D. Dissertation, Max Planck Institute of Molecular Cell Biology
and Genetics, Dresden, 2003.
Serhan, C. N. Prostaglandins 2004, In Press.
Albert, B.; Bray, D.; Leis, J.; Raff, M.; Robert, K.; Watson, J. D. Molecular
Biology of the Cell, Third ed.; Garland: New York, 1994.
http://cellbio.utmb.edu/cellbio/membrane_intro.htm .
Siegel, G. J.; Agranoff, B. W.; Albert, R. W.; Molinoff, P. B. Basic Neurochemistry Molecular, Cellular and Medical Aspects, Fifth ed.; Raven Press:
New York, 1994.
Brugger, B.; Erben, G.; Sandhoff, R.; Wieland, F. T.; Lehmann, W. D. Proc.
Natl. Acad. Sci. U.S.A. 1997, 94, 2339-2344.
Horrobin, D. F.; Glen, A. I. M.; Vaddadi, K. Schizophr. Res. 1994, 13, 195-207.
Ishizuka, I. Prog. Lipid Res. 1997, 36, 245-319.
Merrill, A. H., Jr.; Schmelz, E.-M.; Dillehay, D. L.; Spiegel, S.; Shayman, J.
A.; Schroeder, J. J.; Riley, R. T.; Voss, K. A.; Wang, E. Toxicol. Appl. Pharmacol. 1997, 142, 208-225.
Vos, J. P.; Lopes-Cardozo, M.; Gadella, B. M. Biochim. Biophys. Acta 1994,
1211, 125-149.
Sastry, P. S. Prog. Lipid Res. 1985, 24, 69-176.
Barenholholz, Y.; Thompson, T. E. Chem. Phys. Lipids 1999, 102, 29-34.
45
28. Hevonoja, T.; Pentikäinen, M. O.; Hyvönen, M. T.; Kovanen, P. T.; AlaKorpela, M. Biochim. Biophys. Acta 2000, 1488, 189-210.
29. Niemeä, P.; Hyvönen, H. T.; Vattulainen, I. Biophys. J. 2004, 87, 2976-2989.
30. Norton, W.; Auttilio, L. J. Neurochem. 1966, 13, 213-222.
31. Hawthorne, S. B.; Miller, D. J. J. Chromatogr. A 1987, 403, 63-76.
32. Bøwadt, S.; Hawthorne, S. B. J. Chromatogr. A 1995, 703, 549-571.
33. Chester, T. L.; Pinkston, J. D.; Raynie, D. E. Anal. Chem. 1992, 64, 153R170R.
34. Gazler, K.; Bati, J.; Valko, K. J. Chromatogr. 1986, 371, 299-306.
35. Lopez-Avilla, V.; Young, R.; Kim, R.; Beckert, W. F. J. Chromatogr. Sci.
1995, 33, 481-484.
36. Richer, B. E.; Jones, B. A.; Ezzell, J. L.; Porter, N. L. Anal. Chem. 1996, 68,
1033-1039.
37. Turner, C.; King, J. W.; L., M. J. Chromatogr. A 2001, 936, 215-237.
38. Berg, H.; Mågärd, M.; Johansson, G.; Mathiasson, L. J. Chromatogr. A 1997,
785, 345-352.
39. Johnson, R. B.; Barnett, H. J. Aquaculture 2003, 216, 263-282.
40. Snyder, J. M.; King, J. W.; Jackson, M. A. J. Chromatogr. A 1996, 750, 201207.
41. King, J. W.; Johnson, J. H.; Friedrich, J. P. J. Agric. Food Chem. 1989, 37,
951-954.
42. Lembke, P.; Engelhardt, H. Chromatographia 1993, 35, 509-516.
43. Motohashi, N.; Nagashima, H.; Párkányi, C. J. Biochem. Bioph. Meth. 2000,
43, 313-328.
44. Bartle, V. J. D.; Clifford, A. A. J. Chromatogr. A 1993, 642, 283-299.
45. Rissato, S. R.; Galhiane, M. S.; Knoll, F. R. N.; Apon, B. M. J. Chromatogr. A
2004, 1048, 153-159.
46. Hartonen, K.; Bøwadt, S.; Dybdahl, H. P.; Nylund, K.; Sporring, S.; Lund, H.;
Oreld, F. J. Chromatogr. A 2002, 958, 239-248.
47. Berset, J. D.; Holzer, R. J. Chromatogr. A 852, 852, 545-558.
48. Berg, B. E.; Lund, H. S.; Kringstad, A.; Kvernheim, A. L. Chemosphere 1999,
38, 587-599.
49. Bøwadt, S.; Mazeas, L.; Miller, D. J.; Hawthorne, S. B. J. Chromatogr. A 1997,
785, 205-217.
50. Fernández, I.; Dachs, J.; Bayona, J. M. J. Chromatogr. A 1996, 719, 77-85.
51. Tilio, R.; Krishnan, K.; Kapila, S.; Nam, K. S.; Facchetti, S. Chemosphere
1994, 29, 1849-1858.
52. Paré, J. R. J.; Jacqueline, M. R. B.; Stafford, S. S. TrAC-Trend Anal. Chem.
1994, 13, 176-184.
53. Eskilsson, C. S.; Björklund, E. J. Chromatogr. A 2000, 902, 227-250.
54. LeBlanc, G. LCGC 1999, 17, S30-S37.
55. Focant, J. F.; Pirard, C.; Pauw, E. D. Talanta 2004, 63, 1101-1113.
56. Fitzpatrick, L. J.; Zuloaga, O.; Etxebarria, N.; Dean, J. R. Rev. Anal. Chem.
2000, 19, 75-122.
57. Waldebäck, M. Licentiate Thesis, Uppsala University, Uppsala, 1999.
58. Richter, B. E. LCGC 1999, 17, S22-S28.
59. Lundstedt, S.; Bavel, B. V.; Haglund, P.; Tysklind, M.; Öberg, L. J. Chromatogr. A 2000, 883, 151-162.
60. Schantz, M. M.; Nichols, J. J.; Wise, S. A. Anal. Chem. 1997, 69, 4210-4219.
61. Saim, N.; Dean, J. R.; Abdullah, M. P.; Zakaria, Z. Anal. Chem. 1998, 70, 420424.
62. Lou, X.; Janssen, H.-G.; Cramers, C. A. Anal. Chem. 1997, 69, 1598-1603.
46
63. Björklund, E.; Nilsson, T. Trends Anal. Chem. 2000, 19, 434-445.
64. Björklund, E.; Bøwadt, S.; Nilsson, T.; Mathiasson, L. J. Chromatogr. A 1999,
836, 285-293.
65. Dionex. ASE 200 Accelerated Solvent Extractor Operator´s Manual, Dionex
Corporation: United States of America, 1996.
66. Jensen, S.; Häggberg, L.; Jorundsdottir, H.; Odham, G. J. Agric. Food Chem.
2003, 51, 5607-5611.
67. Schäfer, K. Anal. Chim. Acta. 1998, 358, 69-77.
68. Boselli, E.; Velazco, V.; Caboni, M. F.; Lercker, G. J. Chromatogr., A 2001,
917, 239-244.
69. Richer, B. E.; Covino, L. LCGC 2000, 18, 1068-1073.
70. Olsson, N. U.; Salem, N., Jr. J. Chromatogr. B 1997, 692, 245-256.
71. Robins, S. J.; Patton, G. M. J. Lipid Res. 1986, 27, 131-139.
72. Wang, Y.; Krull, I. S.; Liu, C.; Orr, J. D. J. Chromatogr. B 2003, 793, 3-14.
73. Toyo'oka, T. J. Chromatogr. B 1995, 671, 91-112.
74. Wang, F.; Zhhao, Y.; Wang, P. J. Chromatogr. Sci. 2003, 41, 142-144.
75. Christie, W. W.; Urwin, R. A. J. High Resol. Chromatogr. 1995, 18, 97-100.
76. Christie, W. W.; Gill, S.; Nordbäck, J.; Itabashi, Y.; Sanada, S.; Slabas, A. R.
Phytochem. Analysis 1998, 9, 53-57.
77. Kaduce, T. L.; Norton, K. C.; Spector, A. A. J. Lipid Res. 1983, 24, 1398-1403.
78. Dobson, G.; Christie, W. W.; Nikolova-Damyanova, B. J. Chromatogr. B 1995,
671, 197-222.
79. Nikolova-Damyanova, B.; Christie, W. W.; Herslöf, B. J. Chromatogr. A 1996,
749, 47-54.
80. Christie, W. W. Analusis 1998, 26, M34-M40.
81. Christie, W. W. J. Chromatogr. A 1988, 454, 273-284.
82. Christie, W. W. J. Lipid Res. 1994, 33, 9-18.
83. Christie, W. W.; Breckenridge, G. H. M. J. Chromatogr. B 1989, 469, 261-269.
84. Patton, G. M.; Robins, S. J. Method Enzymol. 1990, 187, 195-215.
85. Jungalwala, F. B.; Hayssen, V.; Pasquini, J. M.; McCluer, R. H. J. Lipid Res.
1979, 20, 579-587.
86. Smith, M.; Jungalwala, F. B. J. Lipid Res. 1981, 22, 697-704.
87. Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Science
1989, 246, 64-71.
88. Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass
Spectrom. Rev. 1990, 9, 37-70.
89. Cech, N. B.; Enke, C. G. Mass Spectrom. Rev. 2001, 20, 362-387.
90. Cole, R. B. Electrospray Ionization Mass Spectrometry; Wiley: New York,
1997.
91. de la Mora, J. F.; van Berkel, G.; Enke, C. G.; Cole, R. B.; Martinez-Sanchez,
M.; Fenn, J. B. J. Mass Spectrom. 2000, 35, 939-952.
92. Griffiths, W. J.; Brown, A.; Reimendal, R.; Yang, Y.; Zhang, J.; Sjövall, J.
Rapid. Commun. Mass. Spectrom. 1996, 10, 1169-1174.
93. Korachi, M.; Blinkhorn, A. S.; Drucker, D. B. Eur. J. Lipid Sci. Technol. 2002,
104, 50-56.
94. Tavana, A. M.; Drucker, D. B.; Hull, P. S.; Boote, V. FEMS Immunol. Med.
Microbiol. 1998, 21, 57-64.
95. Murray, H. E.; Schulten, H. R. Chem. Phys. Lipids 1981, 29, 11-21.
96. Ma, Y. C.; Kim, H. Y. Anal. Biochem. 1995, 226, 293-301.
97. Kim, H. Y.; Yergey, J. A.; Salem, N., Jr. J. Chromatogr. 1987, 394, 155-170.
98. Kim, H. Y.; Salem, N., Jr. Anal. Chem. 1987, 59, 722-726.
99. Kim, H. Y.; Salem, N., Jr. Anal. Chem. 1986, 58, 9-14.
47
100. Schiller, J.; Süß, R.; Arnhold, J.; Fuchs, B.; Leßig, J.; Müller, M.; Petkovic, M.;
Spalteholz, H.; Zschörnig, O.; Arnold, K. Prog. Lipid Res. 2004, 43, 449-488.
101. Jackson, S. N.; Wang, H. J.; Woods, A. S. J. Am. Soc. Mass Spectrom. 2005,
16, 133-138.
102. Hoffmann, E. d. J. Mass Spectrom. 1996, 31, 129-137.
103. Kuksis, A.; Myher, J. J. J. Chromatogr. B 1995, 671, 35-70.
104. Hsu, F. F.; Bohrer, A.; Turk, J. J. Am. Soc. Mass Spectrom. 1998, 9, 516-526.
105. Ho, Y.; Huang, P. Rapid Commun. Mass Spectrom. 2002, 16, 1582-1589.
106. Griffiths, W. J. Mass Spectrom. Rev. 2003, 22, 81-152.
107. Hsu, F. F.; Bohrer, A.; Turk, J. Biochim. Biophys. Acta 1998, 1392, 202-216.
108. Marbois, B. N.; Faull, K. F.; Fluharty, A. L.; Raval-Fernandes, S.; Rome, L. H.
Biochim. Biophys. Acta 2000, 1484, 59-70.
109. Whitfield, P. D.; Sharp, P. C.; Johnson, D. W.; Nelson, P.; Meikle, P. J. Mol.
Genet. Metab. 2001, 73, 30-37.
110. Abidi, S. L.; Mounts, T. L. J. Chromatogr. A 1992, 598, 209-218.
111. Patton, G. M.; Fasulo, J. M.; Robins, S. J. J. Lipid Res. 1982, 23, 190-196.
112. McHowat, J.; JH Jones, J. H.; Creer, M. H. J. Lipid Res. 1996, 37, 2450-2460.
113. Brouwers, J. F. H. M.; Gadella, B. M.; van Golde, L. M. G.; Tielens, A. G. M.
J. Lipid Res. 1998, 39, 344-353.
114. Brouwers, J. F.; Vernooij, E. A.; Tielens, A. G.; van Golde, L. M. J. Lipid Res.
1999, 40, 164-169.
115. Christie, W. W. In Advance in Lipid Methodology-One; The Oily Press: Dundee, 1992, pp 239-271.
116. Karlsson, A. A.; Michelsen, P.; Odham, G. J. Mass. Spectrom. 1998, 33, 11921198.
117. Wold, S.; Berglund, A.; Kettaneh, N. J. Chemometrics 2002, 16, 377-386.
118. Jaumot, J.; Vives, M.; Gargallo, R. Anal. Biochem. 2004, 327, 1-13.
119. Esbensen, K. H. Multivariate Data Analysis - in practice, 4th ed.; CAMO ASA:
Oslo, 2000.
120. Eriksson, L.; Johansson, E.; Kettaneh-Wold, N.; Wold, S. Introduction to multiand megavariate data analysis using projection methods (PCA & PLS);
UMETRICS AB: Umeå, 1999.
121. Eriksson, L.; Johansson, E.; Kettaneh-Wold, N.; Wikström, C.; Wold, S. Design of Experiments; UMETRICS AB: Umeå, 2000.
122. Majors, R. E. LCGC 2002, 20, 1098-1113.
123. Balazy, M. Prostagladins 2004, 73, 173-180.
48
Acta Universitatis Upsaliensis
Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology 56
Editor: The Dean of the Faculty of Science and Technology
A doctoral dissertation from the Faculty of Science and
Technology, Uppsala University, is usually a summary of a
number of papers. A few copies of the complete dissertation
are kept at major Swedish research libraries, while the
summary alone is distributed internationally through the series
Digital Comprehensive Summaries of Uppsala Dissertations
from the Faculty of Science and Technology. (Prior to January,
2005, the series was published under the title "Comprehensive
Summaries of Uppsala Dissertations from the Faculty of
Science and Technology".)
Distribution: publications.uu.se
urn:nbn:se:uu:diva-5810
ACTA
UNIVERSITATIS
UPSALIENSIS
UPPSALA
2005